TW201326741A - Method of compensating probe measurement - Google Patents

Abstract

The invention relates to a method of compensating probe measurement. The method includes establishing an error function of a probe tip, controlling the probe tip to contact with an object to be measured, recording the coordinate of the contact point, calculating a normal vector of each contact point, obtaining a position angle based on the normal vector, taking the position angle in the error function to calculate the errors of the probe tip with respect to each contact point along the normal direction for compensating, and adding the compensating value to the coordinate of the contact point of the object to be measured. Because the error has been added with the coordinate of the contact point of the object to be measured, the error caused by the shape of the probe tip itself is eliminated. Therefore, the contour of the object to be measured can be more precisely measured.

Description

Probe measurement compensation method

The invention relates to a probe measurement and compensation method, in particular to a compensation method for compensating the shape error of the probe itself.

In the three-dimensional measuring machine, a probe is clamped by a probe chuck, and the machine is wired to a microprocessor. When measuring the surface contour of the object to be tested, the object to be tested is first placed on a moving platform, and the microprocessor controls the moving platform to make the probe contact the surface of the object to be tested, and records the coordinates of the contact point, according to all The contact point coordinates plot the surface profile of the object to be tested.

Since the machine itself may cause errors in measurement due to factors such as manufacturing and operating environment, various methods for improving the error have been proposed.

For example, the Republic of China Patent No. I258566 "Three-coordinate Measuring Machine Error Compensation System and Method" includes an input/output module, a computing module, a data security module and a data storage module, the input/output The module is used for data acquisition and error compensation result output; the calculation module is used for error compensation calculation and returning front coordinate value calculation; the data security module is used for measuring and encrypting and decrypting documents in the calculation process; The data storage module is used to measure the temporary storage of various result documents generated during the calculation process. Thereby, the error of the three-coordinate measuring machine tool can be compensated quickly and widely, and the processing cost of the mechanism is reduced, and the measuring accuracy of the machine is improved.

For example, the Republic of China Patent Publication No. 201037268 "Method for Compensation of Thermal Deformation Errors for Three-Dimensional Beds", which is to establish thermal deformation geometric error data at different ambient temperatures, including thermal deformation geometric coordinate error parameters and mechanical parameters, The thermal deformation geometric error model is obtained, and the central control unit of the cubic element bed is input, and then the three-dimensional error compensation amount is converted to obtain the thermal deformation geometric error compensation model, thereby compensating the thermal deformation geometric error compensation model to complete the three-dimensional quantity. The thermal deformation geometric compensation of the bed.

For example, the Republic of China Patent Publication No. 200912242 "Coordinate Measuring Machine and Corresponding Compensation Method" includes a moving unit having a plurality of members movable along a coordinate axis to drive a measuring sensor to move in the measuring space. The moving members are linked together and provide a magnitude associated with the dynamic deformation of the unit; the magnitude is processed to compensate for measurement errors of the machine born with the dynamic deformation.

The above techniques are all for error compensation of the machine itself, but the shape error of the probe is not taken into consideration. When the size of the analyte is smaller, the precision of the probe is more important. Due to the process technology of the probe, the shape accuracy of the probe may not meet the requirements of the small object to be tested, so the error of the shape of the probe itself will affect the measurement result.

SUMMARY OF THE INVENTION It is therefore a primary object of the present invention to provide a probe measurement compensation method for measuring the surface profile of a test object in an equivalent manner to reduce errors due to the probe itself.

For the purpose of the prior art, the technical means adopted by the present invention is that the probe measurement compensation method comprises the steps of: establishing an error function of a probe needle; controlling the probe needle to contact an object to be tested, and recording the contact point thereof; Coordinates, and calculate the normal vector of each contact point, obtain the positioning angle from the normal vector; substitute the positioning angle into the error function to calculate the error of the probe needle corresponding to the normal direction of each contact point as the compensation value; The compensation value is added to the contact point coordinates of the object to be tested.

Therefore, the present invention first establishes an error function of a probe needle, wherein all the contact point coordinates recorded represent the surface contour of the object to be tested, and after obtaining the surface contour of the object to be tested, further according to the error function at each contact point method The error in the line direction to eliminate the error of the probe needle itself, thereby obtaining a more accurate surface profile of the object to be tested.

Referring to FIG. 1, the machine 10 performs related operations and control through a microprocessor 20, and the plane formed by the xy axis represents a horizontal plane, and the z-axis is a vertical horizontal plane, wherein a straight line passing through the origin is projected onto the xy plane. Thereafter, the angle with respect to the x-axis is defined as an angle θ, and the angle between the origin line and the z-axis is defined as a Φ angle. Please refer to FIG. 2, which is a flowchart of the present invention.

The invention first establishes a circular contour error function and a spherical contour error function (100) of a probe. Referring to the schematic diagram shown in FIG. 1, a probe 11 is mounted on the probe collet 12 of the machine table 10, and the machine 10 is wired to the microprocessor 20. As shown in FIG. 3, the tip of the probe 11 has a needle 110 which is divided by a dividing line 111 into a lower half 112 and an upper half 113, wherein the dividing line 111 is the position of the pointer head 110 having the largest outer diameter. .

Referring to FIG. 4, a ring gauge 30 is disposed on the machine table 10. The ring gauge 30 has a cylindrical inner surface 31 and a circular contour 32 of the inner surface 31 as a standard circle.

A. Calculation of the circular contour error function E _c (θ):

The circular contour error function is applied to the two-dimensional object to be measured, and the microprocessor 20 controls the probe chuck 12 to move the probe 11 so that the probe 11 extends vertically into the ring gauge 30. The microprocessor 20 controls the probe 11 to contact the separation line 111 of the needle 110 to the inner surface 31 of the ring gauge 30 at the same height and to contact the inner surface 31 of the ring gauge 30 at a fixed angular angle, for example The inner surface 31 is contacted once every 5° with the axis of the ring gauge 30 as the center. When the probe 11 touches the inner surface 31 once, the microprocessor 20 records the coordinates (x _i , y _i ) of a contact point; that is, after the probe 11 contacts the inner surface 31 of the ring gauge 30 for one turn, The microprocessor 20 records the coordinates of all contact points of the needle 110.

Since the method uses the inner surface 31 of the ring gauge 30 as a standard circle, after the respective contact points are connected, the reflected contour is the actual contour of the needle 110 dividing line 111, as shown in FIG. Contour 114. However, each contact point is discontinuous and discontinuous. In order to obtain a continuous contour, linear, second-order or third-order internal difference calculation can be used, that is, multiple internal differences are calculated between adjacent contact points, and all will be The inner point difference is connected to the contact point to represent the contour of the needle 110 at the position of the separation line 111.

With reference to FIG. 5, the dividing line 114 for comparison with the contour to a standard circle 115, 114 and calculate the standard round contour parting line 115 in the radial direction of the circular contour of the error function E _c (θ). Please refer to FIG. 6 , for example, the contact error of each contact point coordinate (x _i , y _i ) relative to a point (x _s , y _s ) of the standard circle is (Δx, Δy), where Δx=E _c (θ Cos θ, Δy = E _c (θ) sin θ.

B. Calculation of the spherical contour error function:

The spherical contour error function is applied to the three-dimensional object to be measured. Referring to FIG. 7, a standard sphere 40 is disposed on the machine table. The standard sphere 40 is divided into an upper hemisphere 411 and a lower hemisphere by a dividing line 41. The face 412 has the upper hemispherical surface 411 as a standard spherical surface.

The microprocessor 20 controls the probe collet 12 to vertically move the probe 11 with the position of the standard sphere 40 separating the line 41 as the initial height, so that the dividing line 111 of the needle 110 is along the upper hemisphere at a fixed interval angle at the initial height. The 411 contacts, when the probe 11 contacts a circle along the upper hemispherical surface 411, the microprocessor 20 records the coordinates of the contact point of the probe 11 at the initial height.

After the initial height recording is completed, referring to FIG. 8, the microprocessor 20 controls the probe 11 to vertically shift upward by an interval, and horizontally moves to the lower half 112 of the probe 11 to contact the upper hemispherical surface 411, that is, The probe 11 is raised by a Φ angle with respect to the upper hemispherical surface 411, and then the needle 110 is controlled to contact along the upper hemispherical surface 411 at a fixed interval angle. When the probe 11 is contacted one turn along the upper hemispherical surface 411, the microprocessor 20 records The coordinates of the contact point of the probe 11 at this height. Therefore, after the probe 11 is displaced upward several times, the microprocessor 20 can obtain a plurality of contact point coordinates, and construct a curved surface according to all the contact point coordinates. Since the present invention uses the above hemispherical surface 411 as a standard spherical surface, The surface thus reflects the actual surface of the lower half 112 of the needle 110.

Then, the surface is compared with the spherical surface of a standard sphere, for example, the distance from each contact point to the center of the standard sphere is calculated, and the obtained distance is subtracted from the radius value of the standard sphere to obtain an error value, and all the contacts are integrated. After the point is compared with the error value of the standard spherical surface, the spherical contour error function E _s (θ, Φ) can be obtained. In summary, after performing the above steps, the microprocessor 20 obtains the circular contour error E _c (θ) of the probe 11 and the spherical contour error function E _s (θ, Φ) of the lower half 112 of the needle 110.

Then, the contour measurement of the object to be tested is performed, and a curve fitting operation or a surface fitting operation is performed to obtain a normal vector of each contact point, and then the positioning angle of each contact point is estimated according to the normal vector (200). When the probe 11 needle 110 measures the shape of a test object, it is assumed that the object to be tested is a concave hole, and the needle 110 is inserted into the concave hole, and the inner surface of the concave hole is measured by the position of the separation line 111 of the needle 110. . When the needle 110 contacts a circle along the inner surface of the recess, the microprocessor 20 can obtain coordinates of a plurality of contact points 50 as shown in FIG.

Next, an ith contact point coordinate is selected, and a curve fitting operation is performed according to the coordinates of the i, i-1, i-2, ..., i+1, i+2, ..., and several contact points. , that is, the contact point coordinates are substituted into the quadratic curve equation ax ² +by ² +dxy+gx+hy+n=0 for quadratic curve fitting operation, and the equation coefficients a, b, d, g, h, and n are solved. Therefore, according to the calculated coefficient, the quadratic curve equation ax ² +by ² +dxy+gx+hy+n=0 is substituted, and the normal vector A (p, at the ith contact point of the object to be tested is calculated. q), the positioning angle θ _{i of} the ith contact point, θ _i =tan ^-1 (q/p) can be estimated from the normal vector A (p, q).

Similarly, assuming that the object to be tested is measured as the outer surface of the curved surface, the needle 110 of the probe 11 can be circumferentially contacted with the outer surface of the object to be tested at a plurality of fixed heights to obtain all contact point coordinates. Next, selecting an ith contact point coordinate (x _i , y _i , z _i ), and a plurality of jth, k... contact point coordinates (x _j , y _j , z _j ) of the adjacent i-th contact point, (x _k , y _k , z _k )..., substituting a number of contact point coordinates such as i, j, k... into the quadric equation: ax ² + by ² + cz ² + dxy + eyz + fxz +gx+hy+mz+n=0, to perform quadratic surface fitting operation, and then solve the equation coefficients a~h, m, n; yes, according to the coefficients a~h, m, n and then return the above two In the subsurface equation, calculate the normal vector A (p, q, r) of the object to be measured at the ith contact point, and then calculate the positioning angle of the ith contact point according to the normal vector A (p, q, r) θ _i and Φ _i , wherein

After obtaining the positioning angle of each contact point, the positioning angle is substituted into the circular contour error function E _c (θ) or the spherical contour error function E _s (θ, Φ), and the circular contour error of the needle 110 in the direction corresponding to the normal is calculated or The spherical contour error is taken as the compensation value (300). In this step, by taking the positioning angle θ _i into the circular contour error function E _c (θ) as an example of the circular contour error, the error of the needle 110 relative to the standard circle relative to the positioning angle θ _i (Δx, Δy) is obtained. ). Similarly, taking the spherical profile error as an example, the positioning angles θ _i and Φ _{i are} substituted into the spherical contour error function E _s (θ, Φ) to obtain the error of the needle 110 relative to the standard spherical surface with respect to the positioning angles θ _i and Φ _i . (Δx, Δy, Δz).

The obtained compensation value is added to the contact point coordinates of the object to be tested (400). For example, in the case of the circular contour error, after the error (Δx, Δy) is obtained, the needle 110 can be excluded by adding the error (Δx, Δy) to the coordinate (x _i , y _i ) of the corresponding contact point of the object to be tested. The shape error itself, and the contour of the object to be tested is more accurate. For example, the spherical profile error is obtained by adding the error (Δx, Δy, Δz) to the coordinates (x _i , y _i , z _i ) of the corresponding contact point of the object to be tested.

Therefore, according to the method of the present invention, after measuring the contour of the object to be tested, the microprocessor further considers the contour error of the probe itself to compensate the error of the probe itself to the result measured by the object to be tested; Therefore, the compensated contact point coordinates of the object to be tested have eliminated the error of the probe itself, and have better accuracy.

10. . . Machine

11. . . Probe

110. . . Needle

111. . . Separation line

112. . . Lower half

113. . . Upper half

114. . . Divider outline

115. . . Standard circle

12. . . Probe chuck

20. . . microprocessor

30. . . Ring gauge

31. . . The inner surface

32. . . Round outline

40. . . Standard sphere

41. . . Separation line

411. . . Upper hemisphere

412. . . Lower hemisphere

50. . . Contact point

Figure 1: Schematic diagram of the probe measuring device of the present invention.

Figure 2: Schematic diagram of the process of the present invention.

Figure 3 is a perspective view of the probe of the present invention.

Figure 4 is a schematic illustration of the relative position of the probe of the present invention and a ring gauge.

Fig. 5 is a schematic view showing the alignment of the dividing line and the standard circle of the present invention.

Figure 6 is a schematic diagram showing the error of the contact point coordinates of the present invention and its corresponding standard coordinates.

Figures 7 and 8 are schematic views showing the relative positions of the probe of the present invention and a standard sphere.

Fig. 9 is a schematic view showing the coordinates of contact points of the object to be tested of the present invention.

Claims (6)

A probe measurement compensation method comprises the steps of: establishing an error function of a probe needle; controlling the probe needle to contact an object to be tested, recording a contact point coordinate thereof, and calculating a normal vector of each contact point, The normal vector obtains the positioning angle; the positioning angle is substituted into the error function to calculate the error of the probe needle corresponding to the normal direction of each contact point as a compensation value; and the obtained compensation value is added to the contact point coordinate of the object to be tested. . The probe measurement compensation method according to claim 1, wherein the error function of the probe needle comprises a circular contour error function of a probe and a spherical contour error function. The probe measurement compensation method according to claim 2, wherein the obtaining the circular contour error function comprises the steps of: controlling the probe needles to contact the inner surface of a ring gauge at a constant height with the separation line at a fixed interval The angle is contacted along the inner surface of the ring gauge, and the coordinates of all the contact points are recorded; the internal difference calculation is performed to calculate a plurality of internal points between adjacent contact points, and all the internal points are connected to the contact points to Obtain the outline of the dividing line; compare the outline of the dividing line with a standard circle, and use the error of the dividing line outline and the standard circle in the radial direction as the circular contour error function. The probe measurement compensation method according to claim 3, wherein in the step of calculating a normal vector of each contact point, an ith contact point coordinate is selected, and according to the i, i-1, i-2, . .., i+1, i+2..., coordinates of several contact points are subjected to curve fitting operation, and the normal vector at the ith contact point is calculated according to the fitting operation result. The probe measurement compensation method according to claim 2, wherein the obtaining the spherical contour error function comprises the steps of: controlling the probe needle to contact the upper hemisphere of a standard sphere with its separation line at an initial height, and The hemisphere is in contact with one turn to record the coordinates of the contact point of the probe needle at the initial height; several times the control probe needle is vertically displaced for a period of time, and horizontally moved to the lower half of the probe needle to contact the upper hemisphere In which each probe needle surrounds the upper hemisphere, the coordinates of each contact point are recorded; and the distance from each contact point to the center of a standard sphere is calculated, and the distance obtained is subtracted from the radius of the standard sphere. Obtain an error value and integrate the error values of all contact points with respect to the standard sphere to obtain a spherical contour error function. The probe measurement compensation method according to claim 5, in the step of calculating a normal vector of each contact point, selecting an ith contact point coordinate, and an adjacent plurality of contact point coordinates, the number The coordinates of the contact points are subjected to a surface fitting operation, and the normal vector of the object to be tested at the ith contact point is calculated according to the result of the surface fitting operation.