JPH11281306A - Calibrated-value detection method for coordinate-measuring machine and calibration method for shape data using the same calibrated data - Google Patents

Abstract

PROBLEM TO BE SOLVED: To enable to detect squareness calibrated value and contactor calibrated value by alternatevely estimating squareness calibrated value from shape data of a reference spherical surface and contactor calibrated value of shape data of other reference spherical surface. SOLUTION: Two reference spherical surface with different curvature-radiuses are shape-measured, and then squareness calibrated value is estimated by shape data of one of the reference spherical surface and contactor calibrated value is estimated by shape data of other reference spherical surface. By repeating these procedures alternatively, the squareness calibrated value calibrating a squareness error caused from squareness shift by a coordinate axis of a coordinate-measuring machine, and the contactor calibrated value calibrating a contactor error caused from distortion to true spherical shape of the contactor are separated to detect. Thereby, a reference spherical surface is easily produced because of no necessity to accurately consist both curvature radiuses of two reference spherical surface.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method of detecting a calibration value of a squareness error and a contact error of a coordinate measuring machine which scans the surface of a measurement object with a contact and outputs the shape data of the measurement object. The present invention relates to a shape data calibration method using both calibration values.

[0002]

2. Description of the Related Art Conventionally, in order to evaluate the shape of a fine optical member such as an aspherical lens, the surface of the object to be measured is scanned with a contact to form a two-dimensional or three-dimensional unevenness on the surface of the object. A coordinate measuring machine that can be output as a sequence of coordinate data points and evaluated is used. Such a coordinate measuring machine includes a measurement error (squareness error) caused by a deviation of a squareness between coordinate axes and a measurement error (contact error) caused by distortion of a contact from a true spherical shape. There is a possibility that the shape evaluation including such an error will have low accuracy. Therefore, it is necessary to detect the calibration values of these errors in advance, and to calibrate the obtained shape data with these calibration values for evaluation.

[0003] Since the squareness error and the contact error generally appear in the measured shape data in a combined form, it is preferable that the calibration values of these two errors are obtained separately. As described above, as a method of separately detecting the calibration value of the squareness error and the calibration value of the contact error, the shape data is obtained by measuring the shapes of the convex and concave reference spherical surfaces having the same radius of curvature, and obtaining these. There is known a method of detecting an error by detecting an error obtained in comparison with both ideal spherical surfaces of irregularities.

[0004]

However, according to the above method, the sphericity of the reference spherical surface of both the concave and the convex is extremely accurately determined.
In addition, it is necessary to make both radii of curvature precisely coincident with each other, and in order to calibrate the squareness error over the entire measurement range of the measuring machine, both reference spherical surfaces can cover the entire measurement range. It must be large. Since it is extremely difficult to create a large reference spherical surface with high accuracy and with the curvature radii of both concave and convex surfaces precisely matched, calibration accuracy is limited, for example, when the measurement range exceeds several hundred mm. Had been lost.

The present invention has been made in view of such a problem, and separates and detects a squareness calibration value for calibrating a squareness error of a coordinate measuring machine and a contactor calibration value for calibrating a contactor error. It is an object of the present invention to provide a method for detecting a calibration value of a coordinate measuring machine, which is capable of easily creating two reference spheres used for detection, and a method for correcting shape data using the calibration value.

[0006]

In order to achieve the above object, in a method of detecting a calibration value of a coordinate measuring machine according to the present invention, a surface of a measuring object is scanned with a contact to output shape data of the measuring object. By measuring the shape of two reference spheres whose curvature radii are different from each other,
Detects the squareness calibration value for calibrating the squareness error caused by the deviation of the squareness between the coordinate axes of the coordinate measuring machine and the contactor calibration value for calibrating the contactor error caused by distortion from the true spherical shape of the contactor A method for detecting a calibration value of a coordinate measuring machine, which alternates between estimating a squareness calibration value using shape data of one reference sphere and estimating a contactor calibration value using shape data of the other reference sphere. , The squareness calibration value and the contactor calibration value are detected.

For this purpose, the shape data of one reference spherical surface (for example, the first reference spherical surface i in the embodiment) is calibrated with the estimated value of the contactor calibration value (for example, the estimated contactor calibration value in the embodiment). The first shape determined from the calibrated shape data and the theoretical shape data of one of the reference spheres.
A first step of obtaining an estimated value of the squareness calibration value based on the error data (for example, an estimated squareness calibration value in the embodiment);
The shape data of the other reference spherical surface (for example, the second reference spherical surface j in the embodiment) is calibrated with the estimated value of the squareness calibration value,
A second step of obtaining an estimated value of a contactor calibration value based on the calibrated shape data and second error data obtained from the theoretical shape data of the other reference spherical surface, wherein a first step and a second step Are alternately repeated, and when the obtained change amount of the estimated value of the squareness calibration value and the obtained change amount of the estimated value of the contactor calibration value are each equal to or less than a predetermined reference amount, these two estimated values are directly corrected. Detected as the angle calibration value and contact calibration value.

According to such a method, the estimated value of the squareness calibration value and the estimated value of the contactor calibration value converge to constant values, that is, the squareness calibration value and the contactor calibration value, respectively. The calibration value and the contactor calibration value can be detected separately. Further, since the two reference spheres used at this time have different radii of curvature, it is not necessary to make the two radii of curvature precisely coincide with each other, so that the reference sphere can be easily formed.

It is preferable that the estimated value of the squareness calibration value is obtained by using the least square method for the first error data.
According to this, the squareness calibration value can be efficiently estimated, and the estimation values of both calibration values can be quickly converged. If the estimated value of the contactor calibration value represents the second error data as a function of the contact angle of the contactor, when the estimated value of the squareness calibration value converges on the squareness calibration value, the contactor calibration value is obtained. The estimated value also converges to the contact calibration value.
Here, the contact angle is an angle formed by a line segment connecting the center of the contact and the contact point between the contact and the object to be measured, with a vertical line lowered from the center of the contact.

Further, it is preferable that at least one of the two reference spherical surfaces has a size that covers the entire measurement range of the coordinate measuring machine. If a calibration value for calibrating the squareness error is obtained based on the error data obtained from the reference sphere having such a size, an accurate calibration value for the entire measurement range of the measuring machine can be obtained. You can ask.
Further, since at least one of the large reference spheres and the other one may be small, the efficiency of forming the reference sphere is improved.

Further, in the shape data calibration method of the present invention, the squareness calibration value and the contactor calibration value detected by the above-described method for detecting the calibration value of the coordinate measuring machine are stored in a storage means of the coordinate measuring machine. The shape data of the measurement object obtained by the measurement is calibrated using the squareness calibration value and the contactor calibration value.
As a result, the squareness error and the contact error can be removed from the obtained shape data, and highly accurate shape evaluation can be performed.

[0012]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, a preferred embodiment of the present invention will be described with reference to the drawings. The coordinate measuring device 11 shown in FIG. 2A is capable of three-dimensional measurement of the surface shape of the measuring object M, and has a portal-type support frame 13 provided on a horizontal table 12, and the horizontal member 13a has YZ. A stage 14 is mounted movably in the Y and Z directions. An X stage 16 is attached to an X slider 15 provided on the horizontal table 12 so as to be movable in the X direction.
The measuring object M is mounted on the stage 16. The contact 17 of the probe mechanism built in the YZ stage 14 is YZ
It protrudes from below the stage 14 and can be displaced only in the Z direction. Note that this Z-direction displacement is measured by the coordinate measuring device 1
1 is measured by a length measuring scale (not shown) built in the apparatus. Each of the stages 14 and 16 has a built-in position detecting means (not shown), and the amount of movement of each of the stages 14 and 16 in the axial direction is detected.

The coordinate measuring machine 1 shown in FIG. 2 (b) is an example capable of measuring the two-dimensional shape of the measuring object M, and its configuration, measuring method and procedure are almost the same as those of the coordinate measuring device 11 described above. is there. That is, the Z stage 4 can move in the Z direction along the Z slider 3 a provided on the vertical column 3 on the horizontal table 2, and the X stage 4 moves along the X slider 5 mounted on the upper surface of the horizontal table 2. The stage 6 can move in the X direction. Then, the measurement target M is placed on the X stage 6.

A contact 7 of a probe mechanism built in the Z stage 4 projects from below the Z stage 4, and the contact 7 can move only in the Z direction. The movement amount of the Z stage 4 in the Z direction and the movement amount of the X stage 6 in the X direction are read by the displacement sensor 9 with reference to the straight edges 8 provided in parallel with the respective movement axes. Therefore, in the coordinate measuring machine 1, the longitudinal direction of each straight edge 8 corresponds to the X and Z axes of the measuring coordinate system. Hereinafter, for simplicity, the calibration value detection method and calibration method of the present invention will be described based on two-dimensional measurement using the coordinate measuring machine 1.

When measuring the shape of the measuring object M by the coordinate measuring device 1, first, the measuring object M is fixed on the X stage 6, the X stage 6 and the Z stage 4 are moved, and the contact 7 is It is assumed that the measurement start point on M is brought into contact with an appropriate pressing force. Then, by moving the X stage 6 in the X direction and continuously reading the Z-axis coordinates of the contact 7 with respect to the X-axis coordinates, the unevenness data of the surface of the measurement target M can be measured.

In order to separately detect the calibration value for the squareness error θ and the calibration value for the contact error Ep of the coordinate measuring machine 1 shown here, two concave reference spheres i having different radii of curvature are used. First, the squareness error and the contact error of the coordinate measuring machine 1 will be described.

Now, as shown in FIG.
It is assumed that the angle between the axis and the Z axis deviates from a right angle, and the X axis is inclined from the right angle by θ with respect to the Z axis to face the direction of X ′. In this case, the measured shape data includes the squareness error θ, and the shape data (measurement result) indicated by the solid line 22 with respect to the original shape of the measurement target M indicated by the broken line 21 is obtained. This shape data includes a measurement error ΔZc shown in the following equation (1). Here, L is the distance in the X direction from the origin O of the coordinate system of the coordinates.

[0018]

(1) ΔZc = L · sin θ (1)

FIG. 3 shows, for example, ΔZ1 when L = Lx.
(= Lx · sin θ). Usually, since the squareness error θ of the coordinate system is a small angle of about several seconds, Equation (1) can be approximated as Equation (2).

[0020]

(2) ΔZc ≒ L · θ (2)

Equation (2) shows that a measurement error ΔZc occurs in proportion to the distance L from the origin O in the X direction. As is well known, this error causes an S-shaped error when the spherical surface is measured and the optimum fitting is performed. By performing an analysis using a least square method or the like, this error is generated. The squareness error θ can be obtained.

Next, consider a case where the contact 7 is distorted from a true spherical shape. The shape data obtained at this time includes a contact error. This contact error ΔZp is a function depending on the contact angle φ, and the contact angle φ is, as shown in FIG. 4, the center A1 of the contact 7, the contact 7 and the measuring object M
A line segment connecting the contact point A2 with the vertical line AX lowered from the center A1 of the contact 7 is an angle formed by the line segment. This contact error ΔZp is a function of only the contact angle φ when the measurement object M has an arbitrary shape and is independent of the measurement length L in the X direction (however, as will be described later, the reference spherical surface i, j Is not irrelevant since L is a function of φ in the measurement of).
Accordingly, the error ΔZp can be expressed as a function of the contact angle φ as shown in Expression (3).

[0023]

## EQU3 ## ΔZp = Ep (φ) (3)

As shown in FIG. 4A, when there is a portion 7a protruding from a true spherical shape in a part of the contact 7, the shape data obtained by measuring the measurement object M is shown in FIG.
As shown in (b), the actual shape indicated by the broken line 23 becomes the shape data indicated by the solid line 24, and a contact error ΔZ2 (= Ep (φ1)) occurs when φ = φ1. Note that the contact angle φ is usually recognized by analyzing the gradient of the obtained shape data. Ep (φ) is hereinafter abbreviated as Ep.

From the above, the measurement error ΔZ when the squareness error θ and the contact error Ep both exist in the obtained shape data is expressed by equation (4).

[0026]

(Equation 4)

Next, a procedure for detecting a calibration value using two reference spherical surfaces i and j having different radii of curvature will be described. The reference spheres i and j have different radii of curvature as described above, and the reference sphere having the larger radius of curvature (radius of curvature R1).
Is defined as a first reference spherical surface i, and a reference spherical surface having a smaller radius of curvature (radius of curvature R2) is defined as a second reference spherical surface j. The first reference spherical surface i has a size that covers the entire measurement range of the coordinate measuring machine 1. As shown in FIG. 5, the measurement unit (the spherical portion) of the first reference spherical surface i and the measurement of the second reference spherical surface j Both parts are the deflection angles from the line (vertical line) drawn from the center of curvature (point P) to the lowest part (points B1, B2) of the spherical surface.
It is between φ0 and + φ0. Further, the curvature radius ratio is R1: R
When 2 = A: 1 (A> 1), the ratio of the X-direction distances La and Lb therebetween is La: Lb = A: 1. Also,
Note that the deflection angle φ in this case matches the above-mentioned contact angle φ.

First, the shape of the first reference spherical surface i is measured (the range from the point C1 'to the point C1 in FIG. 5 and the range from -φ0 to + φ0) to obtain the shape data. Then, the error data is obtained by fitting the shape data to the theoretical shape data of the first reference spherical surface i. This is defined as first error data ΔZa1. This first error data ΔZa1
Is the error data whose length in the X direction is La, and the following equation (5)
The error component proportional to the length La in the X direction (= R1 · sin φ) and the contact error Ep (function of φ) component as shown in FIG.

[0029]

ΔZa1 = La · θ + Ep (5)

The first error data ΔZa1 has an S-shape as described above, and an estimated squareness calibration value θ1 for correcting the squareness error is obtained by analysis using the least squares method. The estimated squareness calibration value θ1 is stored as calibration data in the storage unit of the coordinate measuring machine 1.

Here, obtaining the estimated squareness calibration value θ1 is based on the fact that all of the first error data ΔZa1 are squareness errors θ1.
This is equivalent to estimating the calibration value by assuming that it has occurred. Therefore, in equation (5), ΔZa1 = La
・ By setting θ1, the calibration error Δθ1 at this time
(= Θ1−θ) is expressed as in equation (6). this is,
The graph shows that the larger the contact error Ep is, the larger the portion is, and the larger the measured length La is, the smaller the error is.

[0032]

[Equation 6] Δθ1 ≒ Ep / La (6)

Next, the shape of the second reference spherical surface j is measured (the range of points C2 'to C2 in FIG. 5 and the range of -φ0 to + φ0) to obtain shape data (this shape data is estimated. The error of the squareness calibration value θ1 is excluded). Then, error data is obtained by fitting the theoretical shape data of the second reference spherical surface j to the shape data. This error data is referred to as second error data ΔZb1.
The second error data ΔZb1 is calculated based on the estimated squareness calibration value θ1.
Although the minute error has been calibrated (removed) but includes the calibration error Δθ1, it is represented by the following equation (7).

[0034]

(Equation 7)

The calibration value for calibrating the second error data ΔZb1 shown in the equation (7) is expressed by a function of φ, and the estimated contactor calibration value E
Let it be p1. Then, the estimated contactor calibration value Ep1 is replaced with the estimated squareness calibration value θ1, and stored in the storage unit of the coordinate measuring machine 1 as calibration data.

At this time, a calibration error ΔEp1 (= Ep1−Ep) expressed by the following equation (8) occurs between the actual contact error Ep and the estimated contact calibration value Ep1. This is obtained by setting ΔZb1 = Ep1 in equation (7).

[0037]

ΔEp1 = Ep / A (8)

Next, the shape of the first reference spherical surface i is measured again to obtain shape data (from this shape data, the error of the estimated contactor calibration value Ep1 has been removed). Then, the first error data ΔZa2 is obtained by fitting the ideal shape data of the first reference spherical surface i to the shape data. In the first error data ΔZa2, an error corresponding to the second calibration value Ep1 has been calibrated (removed) from the first error data ΔZa1.
And written as in equation (9). This is obtained by replacing Ep with Ep−Ep1 (= ΔEp1) in equation (5).

[0039]

(Equation 9)

By analyzing the first error data ΔZa2, an estimated squareness calibration value θ2 is obtained, replaced with the estimated contactor calibration value Ep1, and stored in the storage unit of the coordinate measuring machine 1 as calibration data.

Obtaining the first calibration value θ2 is the first calibration value θ2.
This is equivalent to estimating the calibration value on the assumption that all the error data ΔZa2 is caused by the squareness error θ. Therefore, by setting ΔZa2 = La · θ2 in equation (9), the calibration error Δθ2 (= θ2−θ) at this time is expressed as in equation (10). Or E in the formula (6)
It is obtained by replacing p with ΔEp1.

[0042]

(Equation 10)

Next, shape measurement is again performed on the second reference spherical surface j to obtain shape data (this shape data excludes an error corresponding to the estimated squareness calibration value θ2). Then, the second error data ΔZb2 is obtained by fitting the ideal shape data of the second reference spherical surface to the shape data. The second error data ΔZb2 is obtained by calibrating (removing) the error of the estimated squareness calibration value θ2 from the second error data ΔZb1, and is expressed as in equation (11). This is obtained by replacing Δθ1 with Δθ2 in equation (7).

[0044]

[Equation 11]

The second error data ΔZb2 shown in equation (11)
Is represented by a function of φ and is set as an estimated contactor calibration value Ep2. And this estimated contactor calibration value Ep2 is
It is replaced with the estimated squareness calibration value θ2 and stored in the storage unit of the coordinate measuring machine 1 as calibration data.

At this time, a calibration error ΔEp2 (= Ep2−Ep) shown in the equation (12) occurs between the actual contact error Ep and the estimated contact calibration value Ep2. This is given by equation (1)
It is obtained by setting ΔZb2 = Ep2 in 1).

[0047]

ΔEp2 = Ep / (A · A) (12)

When the procedure as described above is repeated, n
General formulas of the first error data ΔZan and the second error data ΔZbn for ≧ 2 are as shown in Expressions (13) and (13) ′.

[0049]

ΔZan = La · θ + ΔEp (n−1) = La · θ + Ep / A ⁽ⁿ⁻¹⁾ (13) ΔZbn = Ep / A ⁿ + Ep (13) ′

Here, as can be seen from equations (13) and (13) ′, as n increases, the first error data ΔZan
From this, the calibration error corresponding to the contact error Ep is gradually removed to obtain error data La · θ based on the pure squareness error θ. Therefore, the estimated squareness calibration value θn converges to a constant value (ie, squareness error θ), and similarly, the estimated contactor calibration value Ep
n also converges to a constant value (that is, contact error Ep). As described above, when both the estimated squareness calibration value θn and the estimated contactor calibration value Epn converge, the repetition of the procedure is stopped. At this time, if n = N, the estimated squareness calibration value θN and the estimated contactor calibration The value EpN is detected as a squareness calibration value and a contactor calibration value to be obtained.

The amount of change in the estimated squareness calibration value θn (n ≧ 2), that is, the difference between the estimated squareness calibration value θn and the estimated squareness calibration value θ (n−1) is a predetermined reference amount (small). Value) is equal to or smaller than Δθ (in this case, n = N1), it is considered that the estimated squareness calibration value θN1 has converged,
The change amount of the estimated contact calibration value Epn (n ≧ 2), that is, the estimated contact calibration value Epn and the estimated contact calibration value Ep
The difference of (n−1) is a preset reference amount (small value) Δ
When it is less than Ep (at this time, n = N2), it is considered that the estimated contactor calibration value EpN2 has converged. Further, in the present embodiment, the estimated contact error Ep is the second error data ΔZbn sequentially obtained as it is (as a function of φ), so that the estimated squareness calibration value θn converges to a constant value. In this case, the estimated contact calibration value Epn automatically becomes a constant value. Therefore, it is sufficient to detect the change amount of the estimated value only on one side of the estimated squareness calibration value θn.

According to the method described above, the required squareness calibration value and contactor calibration value can be detected separately. Then, for example, the ratio of the measured lengths La and Lb is 10: 1.
Is set to three times, the expression (1)
The calibration error ΔEpn shown in 3) is (1/10) ³ = 1/1
000 to a practically negligible level.

FIG. 1 shows the flow of the procedure described above. First, since the estimated squareness calibration value and the estimated contactor calibration value have no value, these values are both set to zero.
For convenience, if the estimated squareness calibration value at this time is θ0 and the estimated contactor calibration value is Ep0, θ0 = 0 and Ep0 = 0 (step S1). In this state, the first reference spherical surface i is measured to obtain shape data, and is calibrated with the estimated contactor calibration value Ep0 (equivalent to not being calibrated because Ep0 = 0). Then, based on the calibrated shape data, the estimated squareness calibration value θ1
Is obtained (step S2). And this squareness calibration value θ
1 is used as calibration data to measure the second reference spherical surface j to obtain shape data. And, based on this, the estimated contact calibration value Ep
1 is obtained (step S3). And the estimated squareness calibration value θ
1 are calculated, and these are compared with the reference amounts Δθ and ΔEp (Step S).
4). Here, θ1−θ0 <Δθ, and Ep1−Ep0
<ΔEp, squareness calibration value = θ1, contactor calibration value =
Ep1 is set (step S5), but otherwise, feedback is made to step S2. And step S again
Proceeding to 3, set n = n + 1, and repeat these procedures.
Then, θn−θ (n−1) <Δθ, and Epn−Ep
When (n−1) <ΔEp (n = N), the squareness calibration value = θN and the contactor calibration value = EpN (step S5), and the calibration is completed.

In this embodiment, two reference spherical surfaces i,
Both j are concave, but this may be any combination of concave and convex. It is preferable that the reference spherical surface having a larger radius of curvature (for example, the first reference spherical surface i) has a size that covers the entire measurement range of the measuring instrument. As long as the value can be detected, there is no problem even if the curved surface shape has a somewhat higher-order waviness shape. On the other hand, the smaller reference sphere (for example,
Since the second reference spherical surface j) needs to detect a contact error caused by the shape of the contact 7, its curved surface shape needs to have high accuracy, but its size can be reduced. For this reason, both reference spherical surfaces i and j can be manufactured relatively easily, and 10 nm which is difficult by the conventional method.
Calibration with a high accuracy lower than is possible.

If the first measurement results of the shape data of the first reference sphere i and the shape data of the second reference sphere j are stored in the storage unit of the coordinate measuring machine 1, the subsequent first reference sphere i and The measurement of the second reference spherical surface j can be omitted.
That is, the first error data ΔZan and the second error data ΔZbn sequentially obtained as n = 2, 3,.
The stored shape data of the first first reference spherical surface i and the stored shape data of the first second reference spherical surface j may be obtained by calibrating with the estimated squareness calibration value θn or estimated contactor calibration value Epn which can be sequentially obtained. Alternatively, the first error data ΔZa1 and the second error data ΔZb1 are stored, and these are sequentially calibrated and obtained.

The estimated squareness calibration values θ1, θ2,.
In the case of using the least square method for the S-shaped error as shown in the embodiment, the square calibration value can be efficiently estimated, and the estimated values of both calibration values converge quickly. You can do that, but you can ask for it in other ways.

Further, as in the above embodiment, at least one of the two reference spheres i and j has a size covering the entire measurement range of the coordinate measuring machine 1 and is obtained from the reference sphere having such a size. If a calibration value for calibrating the squareness error is obtained based on the obtained error data, an accurate calibration value for the entire measurement range of the coordinate measuring machine 1 can be obtained. In addition, since such a large reference spherical surface may be at least one and the other may be small,
The efficiency of forming the reference sphere is improved.

The squareness calibration value (function of the X coordinate) and the contactor calibration value (function of the contact angle φ) obtained in this manner are stored in the storage device of the coordinate measuring machine 1, and are measured. If the shape data obtained by measuring the shape of M is obtained by calling the squareness calibration value and the contact error after obtaining the shape data, the shape data obtained by the measurement will be the squareness error. θ and the contact error Ep are not included, and highly accurate shape evaluation can be performed.

[0059]

As described above, according to the calibration value detecting method of the coordinate measuring machine of the present invention, the shape of two reference spheres having different radii of curvature is measured, and the squareness is measured using the shape data of one of the reference spheres. By alternately repeating the estimation of the calibration value and the estimation of the contactor calibration value using the shape data of the other reference sphere, the straightness error caused by the deviation of the squareness between the coordinate axes of the coordinate measuring machine is corrected. The angle calibration value and the contactor calibration value for calibrating the contact error caused by the distortion of the contact from the true spherical shape of the contact can be separately detected. Further, since the two reference spheres used at this time have different radii of curvature, it is not necessary to make the two radii of curvature precisely coincide with each other, so that the reference sphere can be easily formed.

According to the shape data calibration method of the present invention, the squareness calibration value and the contactor calibration value detected by the above-described method for detecting the calibration value of the coordinate measuring machine are stored in the storage means of the coordinate measuring machine. Since the shape data of the measurement object obtained by the measurement is calibrated using the squareness calibration value and the contactor calibration value, the squareness error and the contactor error can be removed from the obtained shape data, High shape evaluation can be performed.

[Brief description of the drawings]

FIG. 1 is a diagram showing a flow of a procedure of a calibration value detection method of the present invention.

FIG. 2A is a diagram illustrating an example of a three-dimensional coordinate measuring machine, and FIG. 2B is a diagram illustrating an example of a two-dimensional coordinate measuring machine.

FIG. 3 is a conceptual diagram illustrating a measurement error caused by a squareness error.

4A is a diagram illustrating a concept in which a contact error occurs, and FIG. 4B is a diagram illustrating an example of shape data in which a contact error occurs.

FIG. 5 is a diagram showing a relationship between two reference spherical surfaces.

[Explanation of symbols]

 1 coordinate measuring machine 7 contact i first reference spherical surface j second reference spherical surface φ contact angle M measurement object

Claims (4)

[Claims] 1. A coordinate measuring machine which scans a surface of a measuring object with a contact and outputs shape data of the measuring object,
By performing shape measurement of two reference spheres having different radii of curvature, a squareness calibration value for calibrating a squareness error caused by a difference in squareness between coordinate axes of the coordinate measuring machine and a true spherical surface of the contactor. A calibration value detection method for a coordinate measuring machine for detecting a contactor calibration value for calibrating a contactor error caused by distortion from a shape, wherein the squareness calibration value is estimated using shape data of one reference spherical surface. And the orthogonality calibration value and the contactor calibration value are detected by alternately repeating the estimation of the contactor calibration value performed using the shape data of the other reference spherical surface. Detection method. 2. The shape data of the one reference spherical surface is calibrated with the estimated value of the contactor calibration value, and the first error data obtained from the calibrated shape data and the theoretical shape data of the one reference spherical surface is calculated. A first step of obtaining an estimated value of the squareness calibration value based on the first step; and calibrating the shape data of the other reference spherical surface with the estimated value of the squareness calibration value, and correcting the calibrated shape data and the other reference value. And a second step of obtaining an estimated value of the contactor calibration value based on second error data obtained from the theoretical shape data of the spherical surface. The first step and the second step are alternately repeated, 2. The method according to claim 1, wherein the squareness calibration value and the contact calibration value are detected. 3. The method according to claim 1, wherein the first step and the second step are alternately repeated, and a change amount of the obtained estimated value of the squareness calibration value and a change amount of the obtained estimated value of the contactor calibration value are preset. The calibration value detection of the coordinate measuring machine according to claim 1 or 2, wherein both of the estimated values when the measured values are equal to or less than the reference amount are detected as the squareness calibration value and the contactor calibration value. Method. 4. The squareness calibration value and the contactor calibration value detected by the method for detecting a calibration value of a coordinate measuring machine according to any one of claims 1 to 3, in a storage means of the coordinate measuring machine. A shape data calibrating method, comprising memorizing and calibrating shape data of the measurement object obtained by measurement using the squareness calibration value and the contactor calibration value.