JPH07260450A - Shape measuring device - Google Patents

Abstract

PURPOSE: To measure the shape with high precision by regulation the zero point of an angle detecting means autonomously only by means of data processing during the measurement. CONSTITUTION: A shape measuring device is provided with a sensor head 14 by which angles to an object 15 to be measured are detected at the same time in two points separated from each other by a predetermined distance (d), moving means 12, 13 which relatively moves the object 15 to be measured and the sensor head 14 in the parallel direction mutually for detecting an angle between two points in each of a plurality of positions in the object 15 to be measured, and a microcomputer 20 finding a shape of the object 15 to be measured from the angle data of the two points outputted from the sensor head 14 in each of the plural positions, and the microcomputer 29 finds values obtained by dividing an angle data difference between the two points by the predetermined distance (d) in the plural positions, and these values ane subjected to second order numerical integration so as to find shape data.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a shape measuring device,
More specifically, the present invention relates to a shape measuring device capable of autonomously performing zero point adjustment (parabolic error) of two light beams corresponding to a tangential angle detection probe of an object to be measured by only data processing.

[0002]

2. Description of the Related Art As a scanning type profile measuring device, one using a differential laser autocollimation method is known. In this shape measuring apparatus, the influence of both translational displacement and pitching in the height direction of the movement error during scanning, that is, the z direction, is separated from the shape by integrating the difference between the outputs of the two displacement gauges. Can be removed. Also, in this shape measuring device, due to the zero point adjustment error of the plurality of sensors (light beams), the shape measurement result has a parabolic error, that is, the zero point error (angle of the other light beam with respect to one light beam as a reference). There is a problem in that an error component which is proportional to the square of the measured distance is introduced.

Therefore, in the prior art, for example, the reference plane mirror is scanned with two light beams to measure the shape, and the angular difference is obtained from the output difference of the sensors by both beams, and the zero point is irregular. Was adjusted in terms of hardware or software.

[0004]

However, in the above-described conventional zero-point adjusting method, since there is no perfect plane mirror, the flatness of the reference plane mirror is, for example, a displacement of λ / 20 level. However, when viewed as an angle, it is often large, and there is a problem that the influence of the zero point error in shape measurement is still large.

The present invention solves the above-mentioned conventional problems, and an object thereof is to measure the zero point adjustment of an angle detecting means for simultaneously detecting the tangent angle to the object to be measured at two points. It is an object of the present invention to provide a shape measuring device capable of performing highly accurate shape measurement by autonomously performing only the middle data processing.

[0006]

In order to achieve the above object, the invention of claim 1 is an angle detecting means for simultaneously detecting an angle with respect to an object to be measured at two points separated from each other by a predetermined distance, and the object to be measured. Moving means for relatively moving the object to be measured and the angle detecting means in a direction parallel to each other so as to detect angles at two points at a plurality of points, and angles of two points output from the angle detecting means at the plurality of points. A shape measuring device having a calculation means for calculating the shape of an object to be measured from data, wherein the calculation means obtains a value obtained by dividing the difference between the angle data of the two points by the predetermined distance for a plurality of points, It has a function to obtain shape data by numerically integrating the values by the second order.

According to a second aspect of the present invention, there is provided an angle detecting means for simultaneously detecting angles with respect to the object to be measured at two points which are separated from each other by a predetermined distance d, and to detect the angles of the two points at a plurality of points of the object to be measured. Moving means for relatively moving the object to be measured and the angle detecting means in a direction parallel to each other, and calculating means for obtaining the shape of the object to be measured from angle data of two points output from the angle detecting means at the plurality of locations. A shape measuring apparatus having: and the calculating means obtains a value obtained by dividing the difference between the angle data of the two points by the predetermined distance d for a plurality of locations,
The first shape data from the measurement range 0 to L and the second shape data from the measurement range d to L + d are respectively obtained by numerically integrating these values by the second order, and the one angle data is further calculated as the first order numerical value. Fourth shape data obtained by first-order numerical integration of the integrated third shape data and the other angle data is obtained, and the sum of the difference between the second shape data and the first shape data and the fourth shape data are obtained. Shape data and the third
It has a function of obtaining the angle difference between two points due to the drift of the angle detecting means from the sum of the difference of the shape data.

According to a third aspect of the present invention, there is provided an angle detecting means for simultaneously detecting angles with respect to the object to be measured at two points which are separated from each other by a predetermined distance d, and to detect the angles of the two points at a plurality of points of the object to be measured. Moving means for relatively moving the object to be measured and the angle detecting means in a direction parallel to each other, and calculating means for obtaining the shape of the object to be measured from angle data of two points output from the angle detecting means at the plurality of locations. A shape measuring apparatus having: and the calculating means obtains a value obtained by dividing the difference between the angle data of the two points by the predetermined distance d for a plurality of locations,
The first shape data from the measurement distance 0 to L and the second shape data from the measurement range d to L + d are respectively obtained by numerically integrating these values with the second order numerical value, and further, the one angle data is the first-order numerical value. Fourth shape data obtained by first-order numerical integration of the integrated third shape data and the other angle data is obtained, and the sum of the difference between the second shape data and the first shape data and the fourth shape data are obtained. Shape data and the third
From the sum of the difference of the shape data, the angle difference between the two points due to the drift of the angle detecting means, the parabolic error is calculated from the angle difference between the two points, and the parabolic error is subtracted from the first shape data. Is provided with a function of obtaining true shape data.

According to a fourth aspect of the present invention, the calculating means obtains the first angle data by subtracting the angle difference between the two points obtained in the second aspect from the other angle data, and the angle data and the one The value obtained by dividing the difference from the angle data of 1 by the predetermined distance d is obtained at a plurality of points, the values are first-order numerically integrated to obtain the scanning error, and the scanning error is further subtracted from the one angle data to obtain the first error. It is provided with a function of obtaining the true shape data by obtaining the second angle data and performing the first-order numerical integration of the second angle data.

[0010]

In the present invention, according to the structure of claim 1, the calculating means obtains a value obtained by dividing the difference between the angle data of two points by the predetermined distance for a plurality of places, and secondarily numerically integrates these values to obtain the shape. Since the data is obtained, the angle difference α between the two points
It enables highly accurate shape measurement without the influence of.

Further, according to the present invention, by the constitution of claim 2,
The calculating means calculates the sum of the differences between the second shape data and the first shape data and the difference between the fourth shape data and the third shape data to obtain a difference between the two points due to the drift of the angle detecting means. By obtaining the angle difference α of and correcting with this angle difference α, it is possible to obtain the correct shape that is not affected by the angle difference α.

Still further, according to the present invention, according to the configuration of claim 3, the calculation means includes the sum of the differences between the second shape data and the first shape data, the fourth shape data and the third shape. The difference between the two points due to the drift of the angle detecting means is obtained from the sum of the differences of the data, the parabolic error is obtained from the difference between the two points, and the parabolic error is subtracted from the first shape data to obtain the true value. Since it is configured to obtain the shape data of, the zero point adjustment can be autonomously performed only by data processing, and highly accurate shape measurement can be performed.

Further, according to the present invention, in accordance with the constitution of claim 4,
The calculation means obtains a value obtained by dividing the difference between the first angle data and one angle data by a predetermined distance d at a plurality of positions, first-order numerically integrates these values to obtain a scanning error, and then the scanning error. Is subtracted from one of the angle data to obtain the second angle data, and the true shape data is obtained by first-order numerical integration of the second angle data. Therefore, zero point adjustment is autonomously performed only by data processing. It is possible to obtain the shape measurement with high accuracy.

[0014]

Embodiments of the present invention will be described below with reference to the drawings. FIG. 1 is a schematic diagram of a scanning system (moving system) and a processing system used for shape measurement of a sample (measurement object) to which the present invention is applied, and FIG. 2 is a principle diagram for explaining the present invention. An example of straightness measurement by the dynamic autocollimation method is shown.

In FIG. 1, 10 is a base and 11 is a base 1.
0 is a table provided so as to be movable in the longitudinal direction thereof. The table 11 includes a feed screw 12 and the feed screw 12.
A motor 13 that rotates the base 10 can move in the longitudinal direction of the base 10. A sensor head 14 including a semiconductor laser and an optical system is fixed to the table 11. The two light beams 14a and 14b emitted from the sensor head 14 are applied to the object to be measured (plane mirror sample) 15. The DUT 15 is fixed to a manual X-axis stage 16 that moves in a direction parallel to the table 11.

Reference numeral 17 is a table 11 and an X-axis stage 1.
A displacement gauge for detecting the displacement of 6 and a measuring circuit 18 for measuring the reflection angle of the light beam reflected from the surface of the object to be measured 15 (corresponding to the tangent angle of the beam spot point of the object to be measured 15). The angle signal output from the measuring circuit 18 and the position signal output from the displacement meter 17 are converted into digital values by the A / D converter 19 and then output to the microcomputer 20. The microcomputer 20 calculates the shape of the DUT 15 and adjusts the zero point of the light beam.

In the scanning / processing system having the above structure, the motor 13 is driven to move the sensor head 14 including the table 11 in the X-axis direction from the measuring range x = 0 to x = L, and the measuring circuit 18 to A The angle data corresponding to the two light beams input to the microcomputer 20 through the / D converter 19 is fetched at each sampling interval and stored in the memory built in the microcomputer 20.

In the microcomputer 20, each angle data corresponding to two light beams is first-order integrated, a difference δ between the average values of both integrated values in the measurement range and a value Δf ₁ which will be described later are obtained, and the light is calculated from this difference. Light beam 1 based on beam 14a
An angle difference α between 4a and 14b is obtained. Then, using this angle difference α, there is no angle difference α, that is, the light beams 14a, 1
The true shape data in which the zero point irregularity of 4b is corrected is obtained.

Next, the process of shape measurement will be described with reference to the principle diagram shown in FIG. In FIG. 2, the DUT 1
The sensor head 14 that moves in the X-axis direction with respect to 5 includes two probes 14a and 14b corresponding to a laser beam. Each of the probes 14a and 14b irradiates the surface of the DUT 15 with the laser beam emitted from the probe 14a and captures the laser beam reflected by the surface of the DUT 15, thereby obtaining a tangential angle of the laser beam irradiation point. The angle to be detected is detected, and the shape data that is not affected by the zero point imbalance of the probe is obtained based on the difference between the first-order differential coefficients of the shape obtained from this angle data.

In FIG. 2, C ₀ is a representative point of the sensor head 14, and here, it is taken at the position A of the probe 14a. Further, L is the basic measurement length, d is the distance between the probes 14a and 14b, φ _z (x) is the translation error of the sensor head 14, φ _p (x)
Is pitching of the sensor head 14. Furthermore, f
(X) is the shape of the DUT 15 to be measured. In this case, each probe 14a, 14b output m _a of '(x),
m _b ′ (x) is given by the following equation.

[0021] _{m a '(x) = f} ' (x) + φ p (x) ··············· (1) m b '(x) = f' (x + d) + Φ _p (x) + α (2) where α is the deviation of the zero point at the position B of the probe 14b with respect to the position A of the probe 14a, and is an unknown amount. is there. However, in angle measurement, translation error φ _z (x)
Therefore, the translation error φ _z (x) cannot be expressed in the above equations (1) and (2). The differential output m ″ (x) of the probes 14a and 14b shown in the above equations (1) and (2) is given by the following equation.

M ″ (x) = {m _b ′ (x) −m _a ′ (x)} / d (3) This equation (3) is used for the probes 14a and 14b. It is a value obtained by dividing the difference in the angle data of 2 by the inter-probe distance d.
By performing the second-order numerical integration on (x), it is possible to obtain the shape data f ₁ (x) that is not affected by the scanning error (translation error and pitching). f ₁ (x) is given by the following equation.

F ₁ (x) = f (x) + α (x−L / 2) ² / 2d ... (4) However, x = 0 to L. In the obtained shape f ₁ (x),
Includes parabolic error (zero point error). Strictly speaking, f (x) in the equation (4) includes an error caused by treating m ″ (x) obtained by the difference as a differential value and an error caused by numerical integration. It has almost no effect on the calculation accuracy of the angle difference α.

In order to obtain the angle difference α, the following processing is performed. First, the outputs of the probes 14a and 14b represented by the equations (1) and (2) are integrated by the first order. _{m a (x) = f (} x) + φ p (x) ················· (5) m b (x) = f (x + d) + φ p (x) + αd ············ (6) the difference between the average value in the range of the obtained m _a (x) and x = 0 to a _{m b (x) (L-} d) δ And

Here, when the range of the sum of the differences δ is shown by [], δ when the number of data N _L in the measurement range L and the number of data between probes is N _d is given by the following equation. δ = Δf−α (L + d) / 2 (7) Note that Δf in the equation (7) is given by the following equation. Delta] f = {f [L-d, L] -f [0, _{d]} / (N L -N d)} ··· (8)

If this Δf is known, α can be obtained. here,
Substituting f ₁ (x) of equation (4) into equation (8) and letting Δf ₁ be the result of calculation, Δf ₁ = Δf. (9), and it can be seen that Δf ₁ does not include the influence of α. Using Δf ₁ and δ, α can be calculated by the following equation. α = 2 (Δf ₁ −δ) / L−d (10)

If corrected with the obtained α, the data of the correct shape f (x) which is not affected by α from the equation (4) is x = 0 to 0.
It can be obtained in the range of L. Further, the correct f ′ (x) data is obtained from this, and the pitching φ _p (x) can be known using the equations (1) and (2). Further, by integrating m _b ′ (x) of the equation (1), the data of f (x) can be obtained within the range of x = d to L + d.

On the other hand, creating a m _b 'angle data minus α from (x) (measurement data) m _b' (x), Formula (2)
Becomes m _b ′ (x) = f ′ (x + d) + φ _p (x) (11). Here, when x = x + d, equation (1) _{is, m a '(x + d} ) = f' become _{(x + d) + φ p} (x + d) ········· (12). Equation (11) from equation _{(12), m a '(} x + d) -m b' (x) = φ p (x + d) -φ
_p (x), _{i.e., {m a '(x +} d) -m b' (x)} / d = {φ p (x
+ D) -φ _p ) x)} / d (= φ _p ′ (x)), and ma _a ′
The data of φ _p ′ (x) can be obtained from the (x) and m _b ′ (x) angle data. By integrating this, the data of φ _p (x) can be obtained.
Then, by substituting the data of φ _p (x) into the equation (1), f ′
The data of (x) is obtained. From this, correct data of f (x) can be obtained in the range of x = d to L + d. By substituting the data of φ _p (x) and α into the equation (2), the data of f ′ (x + d) is obtained. Therefore, from both f (x), x = 0 to L +
f (x) in the range of d is obtained.

In the present embodiment as described above, in the software datum for removing the scanning error components of translation and pitching during scanning, the difference between the zero points of the two probes is calculated without using the reference sample. Can only be determined. As a result, the shape such as straightness can be measured with high accuracy on an actual device without using an external reference such as a metrology frame.

Further, since the actual angle difference α during the measurement can be calculated, the influence of the time-dependent change of the difference between the zero points of the two probes such as the thermal displacement of the sensor is reduced by averaging, and the two points are reduced. Since it is not necessary to make a preliminary measurement for calibrating the difference between the zero points of the probe, the measurement time can be shortened.

Although the differential autocollimation method is used in the above embodiment, the present invention is not limited to the above, and can be applied to shape measurement by the three-point method and the mixing method. Here, the mixing method outputs the displacement with respect to the measured object, and the difference calculation of the output values twice is
This corresponds to the differential output m ″ (x) in the above embodiment, and one difference calculation of the displacement output value corresponds to the output m _a ′ (x), m _b ′ (x) of the probe in the above embodiment. .

[0032]

As described above, according to the first aspect of the present invention.
According to the calculation means, the difference between the angle data of two points is divided by the predetermined distance to obtain a plurality of values, and these values are secondarily numerically integrated to obtain the shape data. It is possible to perform highly accurate shape measurement without being affected by the angle difference α.

Further, according to a second aspect of the present invention, the calculation means includes a sum of differences between the second shape data and the first shape data, the fourth shape data and the third shape data. By calculating the angle difference α between the two points due to the drift of the angle detecting means from the sum of the differences and correcting with this angle difference α, a correct shape that is not affected by the angle difference α can be obtained.

Furthermore, according to claim 3 of the present invention,
The calculating means calculates the sum of the differences between the second shape data and the first shape data and the difference between the fourth shape data and the third shape data to obtain a difference between the two points due to the drift of the angle detecting means. Is calculated, the parabolic error is calculated from the angular difference between the two points, and the true shape data is calculated by subtracting the parabolic error from the first shape data. It can be performed autonomously only by data processing, and highly accurate shape measurement becomes possible.

According to a fourth aspect of the present invention, the calculating means obtains a plurality of values obtained by dividing the difference between the first angle data and one angle data by the predetermined distance d, and calculates these values as 1
After obtaining the scanning error by numerically integrating the first order, this scanning error is subtracted from one of the angle data to obtain the second angle data, and the second-order angle data is first-order numerically integrated to obtain the true shape data. Since it is obtained, zero point adjustment can be autonomously performed only by data processing, and highly accurate shape measurement becomes possible.

Therefore, in the present invention, the difference between the zero points of the plurality of angle detecting means which are separated by the predetermined distance can be autonomously determined only by the data processing during the measurement without using any reference. Accordingly, the shape of the object to be measured can be measured with high accuracy on an actual device without using an external reference such as a metrology frame, and the measurement time can be shortened.

[Brief description of drawings]

FIG. 1 is a schematic configuration diagram of a scanning system and a processing system used for measuring the shape of an object to be measured to which the present invention is applied.

FIG. 2 is a principle diagram for explaining the present invention.

[Explanation of symbols]

 11 table 12 feed screw (moving means) 13 motor (moving means) 14 sensor head (angle detecting means) 14a, 14b light beam (laser beam, probe) 15 object to be measured 18 measuring circuit 19 A / D converter 20 microcomputer ( (Calculation means)

─────────────────────────────────────────────────── ───

[Procedure amendment]

[Submission date] March 22, 1994

[Procedure Amendment 1]

[Document name to be corrected] Drawing

[Correction target item name] All drawings

[Correction method] Change

[Correction content]

[Figure 1]

[Fig. 2]

 ─────────────────────────────────────────────────── ─── Continuation of the front page (71) Applicant 594046961 Okiyama Eiki 59-59 Nakadai, Akita-shi, Akita-shi, Akita (72) Inventor Kei Kei 2-35-403, Katahira, Aoba-ku, Sendai-shi, Miyagi (72) ) The inventor, Gao Wei, No.7, Mao-Shin-mura, Daoukou-gu, Chongqing, Sichuan Province, China (72) Inventor, Eiki Okuyama 59-57, Nakadai, Akita, Akita, Japan

Claims (4)

[Claims] 1. An angle detecting means for simultaneously detecting angles with respect to an object to be measured at two points separated from each other by a predetermined distance, and the object to be measured and the angle for detecting angles at two points at a plurality of points of the object to be measured. A shape measuring device having a moving means for relatively moving the detecting means in a direction parallel to each other, and a calculating means for obtaining the shape of the object to be measured from angle data of two points output from the angle detecting means at the plurality of locations. The calculation means has a function of calculating a value obtained by dividing the difference between the angle data of the two points by the predetermined distance for a plurality of positions, and numerically integrating the values to obtain the shape data. Shape measuring device. 2. An angle detecting means for simultaneously detecting angles with respect to an object to be measured at two points separated from each other by a predetermined distance d, and the object to be measured so as to detect angles at two points at a plurality of points of the object to be measured. Shape measurement having moving means for relatively moving the angle detecting means in a direction parallel to each other, and calculating means for obtaining the shape of the object to be measured from angle data of two points output from the angle detecting means at the plurality of locations. In the device, the calculating means obtains a value obtained by dividing a difference between the angle data of the two points by the predetermined distance d for a plurality of places, and performs a second-order numerical integration of these values to measure ranges from 0 to L. And the second shape data from the measurement range d to L + d are obtained.
Third shape data obtained by numerically integrating the floor and fourth shape data obtained by numerically integrating the other angle data by the first floor are respectively obtained, and the sum of the difference between the second shape data and the first shape data and the first shape data are obtained. 4. A shape measuring device having a function of obtaining an angle difference between two points due to a drift of the angle detecting means from a sum of differences between the shape data No. 4 and the third shape data. 3. An angle detecting means for simultaneously detecting angles with respect to an object to be measured at two points separated from each other by a predetermined distance d, and the object to be measured so as to detect angles at two points at a plurality of points of the object to be measured. Shape measurement having moving means for relatively moving the angle detecting means in a direction parallel to each other, and calculating means for obtaining the shape of the object to be measured from angle data of two points output from the angle detecting means at the plurality of locations. In the device, the calculating means obtains a value obtained by dividing the difference between the angle data of the two points by the predetermined distance d for a plurality of places, and numerically integrates these values from the measurement distance 0 to L. And the second shape data from the measurement range d to L + d are obtained.
Third shape data obtained by numerically integrating the floor and fourth shape data obtained by numerically integrating the other angle data by the first order are respectively obtained, and the sum of the difference between the second shape data and the first shape data and the first shape data are obtained. From the sum of the difference between the shape data of No. 4 and the third shape data, the angle difference between the two points due to the drift of the angle detecting means is obtained, and the parabolic error is obtained from the angle difference between the two points, and the first shape is obtained. A shape measuring apparatus having a function of obtaining true shape data by subtracting the parabolic error from the data. 4. The calculating means obtains first angle data by subtracting the angle difference between the two points obtained in claim 2 or 3 from the other angle data, and obtains this angle data and the one of the one. A value obtained by dividing the difference from the angle data by the predetermined distance d is obtained at a plurality of points, a first-order numerical integration of these values is performed to obtain a scanning error, and the scanning error is subtracted from the one angle data to obtain a second value. Angle data of
3. The shape measuring device according to claim 2, further comprising a function of obtaining true shape data by first-order numerically integrating the second angle data.