Method for identifying errors in 5-axis controlled machine tools, program for identifying errors, 5-axis controlled machine tools

Abstract

To make it possible to identify an inter-shaft geometrical error with high accuracy without an influence of a shaft single body geometrical error of a linearly advancing shaft.SOLUTION: A five-axes control machine tool error identification method comprises: a preparation step (S1) of mounting a touch probe to a main shaft, and fixing a ball array to a table; an initial position measuring step (S2) of measuring an initial position of the ball array by the touch probe; a reference angle calculation step (S3) of calculating a reference angle of each rotation axis for locating the ball array at the reference position from a measurement value in the initial position measuring step; a calibrator measuring step (S5 to S7) of indexing each rotation axis to a plurality of indexed angles with respect to the reference angle respectively, and measuring a center position of a sphere of the ball array fixed to the table at each indexed angle by the touch probe respectively; and an error identification step (S8 to S9) of identifying a positioning error of the linearly advancing axis and straightness error thereof, a position error of each rotation axis and inclination error thereof on the basis of the measurement value in the calibrator measuring step.SELECTED DRAWING: Figure 2

Description

[Technical field]

[0001] The present disclosure relates to a method for identifying geometric errors in a 5-axis controlled machine tool such as a 5-axis controlled machining center, an error identification program for executing the method, and a 5-axis controlled machine tool capable of executing the method. [Background technology]

[0002] FIG. 1 is a schematic diagram of a five-axis controlled machining center having three linear axes and two rotary axes. The spindle head 2 is capable of translational movement with two degrees of freedom relative to the bed 1 by the X-axis and Z-axis, which are linear axes and perpendicular to each other. The table 3 is capable of rotational movement with one degree of freedom relative to the cradle 4 by the C-axis, which is a rotary axis. The cradle 4 is capable of rotational movement with one degree of freedom relative to the trunnion 5 by the A-axis, which is a rotary axis. The A-axis and C-axis are perpendicular to each other. Furthermore, the trunnion 5 is capable of translational movement with one degree of freedom relative to the bed 1 by the Y-axis, which is a linear axis and perpendicular to the X-axis and Z-axis. Therefore, the spindle head 2 is capable of translational movement with three degrees of freedom and rotational movement with two degrees of freedom relative to the table 3. Each feed axis is driven by a servo motor controlled by a numerical control device (not shown). The workpiece is fixed to the table 3, a tool is attached to the spindle head 2 and rotated, and machining can be performed by controlling the relative position and relative attitude of the workpiece and the tool.

[0003] Major factors affecting the motion accuracy of a 5-axis controlled machining center include geometric errors between each axis, such as error in the central position of the rotary axis (deviation from the expected position) and tilt error of the rotary axis (perpendicularity between axes, parallelism).For example, in the 5-axis controlled machining center in Figure 1, there are three inter-axis geometric errors related to the linear axes: perpendicularity between the X and Y axes, perpendicularity between the Y and Z axes, and perpendicularity between the Z and X axes.There are two inter-axis geometric errors related to the main axis: perpendicularity between the tool and Y axis, and perpendicularity between the tool and X axis.There are eight inter-axis geometric errors related to the rotary axes: X-direction error of the C-axis central position, offset error between the CA axes, angular offset error of the A-axis, perpendicularity between the CA axes, Y-direction error of the A-axis central position, Z-direction error of the A-axis central position, perpendicularity between the AZ axes, and perpendicularity between the AY axes. Furthermore, even for linear axes alone, there are positioning errors for each axis and geometric errors for each axis, such as straightness, which are factors that affect the motion accuracy of the machine. If these geometric errors exist, the motion accuracy of the machine will deteriorate, and the machining accuracy of the workpiece will deteriorate. To address this, the general methods are to manufacture or adjust the machine to reduce the geometric errors, or to compensate for the errors in the tool position relative to the workpiece caused by the geometric errors. To perform such compensation control, it is necessary to measure or identify the geometric errors inherent in the machine.

[0004] As a method for identifying inter-axis geometric errors, the present inventor has proposed a method as disclosed in Patent Document 1. In this method, the table is rotated and tilted at a plurality of angles by the rotary axis, and a touch probe attached to the spindle is used to measure the center positions of balls fixed on the table, and the inter-axis geometric errors can be identified from the obtained measurement values. By simply placing a ball on the table, a series of measurements can be performed automatically, and it is also possible to automatically identify geometric errors, which has the advantage that geometric errors can be identified even if the machine operator does not have knowledge or skills in measurement. On the other hand, as a method for identifying the geometric error of a single axis, Patent Document 2 discloses a method in which the orientation of a step gauge having multiple blocks in a three-dimensional measuring machine is changed in multiple directions, the distance between the blocks is measured, and the overall expansion / contraction component (first-order component) of the positioning error of three linear axes and the perpendicularity between the three linear axes are identified. Furthermore, Patent Document 3 discloses a method for identifying the first-order component of the positioning error of three linear axes and the perpendicularity between the three linear axes by changing the orientation of a ball array having a plurality of balls in a machine tool in a plurality of directions and measuring the distances between the plurality of balls in the ball array using a touch probe attached to the spindle. [Prior art documents] [Patent documents]

[0005] [Patent Document 1] JP 2011-38902 A [Patent Document 2] JP 2007-101279 A [Patent Document 3] JP 2020-46301 A Summary of the Invention [Problem to be solved by the invention]

[0006] The touch probe mainly used in Patent Documents 1 to 3 is a device that outputs a signal when it comes into contact with an object, and the control device detects the signal and obtains the detected position of the linear axis position detector at the time the signal is received or at a time taking into account a delay, thereby making it possible to measure the position of the object. For this reason, the measurement result of the ball using the touch probe is affected by the axis-only geometric error (positioning error, straightness) of each linear axis. Therefore, the method of Patent Document 1 has a problem in that when the individual axis geometric error of each linear axis is large, the inter-axis geometric error cannot be identified with high accuracy. On the other hand, the methods of Patent Documents 2 and 3 make it possible to identify the first-order component (overall expansion / contraction component) of the positioning error of each linear axis and the squareness between each axis, but there is a problem in that the identification of the squareness is affected by the straightness of the linear axis. In addition, in order to change the orientation of the step gauge or ball array, the angle of the jig used to install them must be changed, which is troublesome for workers and may result in a collision if the jig is turned in the wrong direction. If a drive unit is provided for the rotation mechanism of the jig, the jig becomes expensive.

[0007] Therefore, the present disclosure has an object to provide an error identification method and error identification program for a 5-axis controlled machine tool, and a 5-axis controlled machine tool, which enable accurate identification of inter-axis geometric errors without being affected by the axis-specific geometric errors of linear axes, and which also enable automatic identification of the axis-specific geometric errors and inter-axis geometric errors of linear axes without the knowledge or measurement skills of a machine operator. [Means for solving the problem]

[0008] In order to achieve the above object, a first configuration of the present disclosure is a method for identifying an error in a five-axis controlled machine tool having a rotatable main spindle on which a tool is attached, a table on which a workpiece and / or a jig can be fixed, three linear axes that are mutually perpendicular and allow the main spindle to move relatively with respect to the table, and two rotary axes that allow the table to be turned and / or tilted, the method comprising: A position measuring sensor tool is attached to the spindle; Arranged on a straight line a preparatory step of fixing a calibrator having three or more spheres to said table; an initial position measuring step of measuring an initial position of the calibrator by the position measuring sensor tool; a reference angle calculation step of calculating a reference angle of each of the rotation shafts for positioning the calibrator at a predetermined reference position from a measurement value in the initial position measurement step; a calibrator measurement step of indexing each of the rotation axes to a plurality of index angles with respect to the reference angle and measuring the center position of the sphere of the calibrator fixed to the table at each of the index angles by the position measurement sensor tool; and an error identification step of identifying positioning errors and straightness errors of the linear axes and squareness errors between the linear axes, and position errors and inclination errors of the rotation axes, based on the measurement values ​​in the calibration measurement step. Another aspect of the first configuration of the present disclosure, in the above configuration, in the error identification step, The present invention is characterized in that, using the measurement values ​​in the calibration measurement stage, the positioning errors and straightness errors of the two linear axes, the three squareness errors of the linear axes, the expansion / contraction components of the positioning error of one linear axis other than the two linear axes, and the position errors and inclination errors of each of the rotation axes are identified. Another aspect of the first configuration of the present disclosure is, in the above configuration, the calibration device measurement step includes: The present invention is characterized in that an arc measurement step is performed in which, while one of the rotation axes is indexed to any one of the index angles so that the ball of the calibrator describes an arc trajectory, the other rotation axis is sequentially indexed to a plurality of the index angles, and the center position of at least one ball of the calibrator is measured. Another aspect of the first configuration of the present disclosure is, in the above configuration, the calibration device measurement step includes: The calibrator The straight line on which the balls are arranged A parallel measurement step of indexing each of the rotation axes so that their orientations are parallel to the axes of the two linear axes, respectively, and measuring the centers of three or more spheres of the calibrator using the position measurement sensor tool; The calibrator The straight line and a plane diagonal measurement step of indexing each of the rotation axes so that their orientation is in the diagonal direction of the plane formed by the axes of the two linear axes, and measuring the center positions of three or more spheres of the calibrator using the position measurement sensor tool. Another aspect of the first configuration of the present disclosure is, in the above configuration, the calibration device measurement step includes: The calibrator The straight line on which the balls are arranged A parallel measurement step of indexing each of the rotation axes so that their orientations are parallel to the axes of the two linear axes, respectively, and measuring the center positions of three or more spheres of the calibrator using the position measurement sensor tool; The calibrator The straight line The direction is the space formed by the axes of the three linear axes. Between each of the axes and a spatial diagonal measurement step of measuring the center positions of three or more spheres of the calibrator using the position measurement sensor tool by indexing each of the rotation axes so that they are in the diagonal directions of the respective spheres. In order to achieve the above object, a second configuration of the present disclosure is an error identification program for a five-axis controlled machine tool, The method is characterized in that the error identification method for an axis-controlled machine tool disclosed herein is executed by a numerical control device of a five-axis controlled machine tool having a rotatable main spindle on which a tool can be attached, a table on which a workpiece and / or a jig can be fixed, three linear axes which are mutually perpendicular and allow the main spindle to move relative to the table, and two rotary axes which allow the table to be rotated and / or tilted. In order to achieve the above object, a third configuration of the present disclosure is a five-axis controlled machine tool having a rotatable main spindle on which a tool is attached, a table on which a workpiece and / or a jig can be fixed, three linear axes that are mutually perpendicular and allow the main spindle to move relatively with respect to the table, and two rotary axes that allow the table to turn and / or tilt, A position measuring sensor tool attached to the spindle, Arranged on a straight line an initial position measuring means for measuring an initial position of a calibrator having three or more spheres and fixed to the table; a reference angle calculation means for calculating a reference angle of each of the rotation shafts for positioning the calibrator at a predetermined reference position from a measurement value by the initial position measurement means; a calibrator measuring means for indexing each of the rotation axes to a plurality of index angles with respect to the reference angle, and measuring the center position of the sphere of the calibrator fixed to the table at each of the index angles by the position measuring sensor tool; The present invention is characterized in that it comprises an error identification means which identifies the positioning error and straightness error of the linear axes and the squareness error between the linear axes, and the position error and inclination error of each of the rotation axes, based on the measurement values ​​by the calibrator measurement means. Effect of the Invention

[0009] According to the present disclosure, it is possible to simultaneously identify the axis-specific geometric errors (positioning errors, straightness) of the linear axes of a five-axis controlled machine tool and the inter-axis geometric errors (squareness between linear axes, center error of a rotation axis, and tilt error of a rotation axis). Therefore, it is possible to accurately identify the inter-axis geometric errors without being affected by the axis-specific geometric errors of the linear axes. Furthermore, by using the rotary axis of a 5-axis controlled machine tool, a series of measurements can be performed automatically, and the axis-specific geometric error and inter-axis geometric error of linear axes can be identified automatically without the knowledge or measurement skills of the machine operator. [Brief description of the drawings]

[0010] [Figure 1] This is a schematic diagram of a 5-axis controlled machining center. [Diagram 2] 1 is a flow chart of an error identification method. [Diagram 3] FIG. 1 is a schematic diagram of a touch probe and a ball array. [Figure 4] FIG. 2 is a schematic diagram of the initial installation position of the ball array. [Diagram 5] FIG. 13 is a schematic diagram of the position of the ball array in the X-axis parallel direction measurement. [Figure 6] FIG. 13 is a schematic diagram of the position of the ball array in the Y-axis parallel direction measurement. [Figure 7] FIG. 1 is a schematic diagram of the position of the ball array in the diagonal direction measurement of the XY axes. [Figure 8] FIG. 1 is a schematic diagram of the position of the ball array in the YZ axis diagonal direction measurement. [Figure 9] FIG. 13 is a schematic diagram of the position of the ball array in the ZX axis diagonal direction measurement. [Figure 10] FIG. 13 is a schematic diagram of the position of the ball array in a spatial diagonal measurement. [Figure 11] FIG. 13 is a schematic diagram showing the position of a ball array in a circular arc measurement centered on the C-axis. [Figure 12] FIG. 13 is a schematic diagram showing the position of a ball array in an arc measurement centered on the A-axis. [Figure 13] 13 is a detailed flowchart of S8. [Figure 14] 13 is a detailed flowchart of S9. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0011] Hereinafter, an embodiment of the present disclosure will be described with reference to the drawings. In this embodiment, an example of a 5-axis controlled machine tool that implements the error identification method of the present disclosure is the 5-axis controlled machining center of FIG. 1 described above. Therefore, duplicated explanations will be omitted. This 5-axis controlled machining center includes the initial position measurement means, reference angle calculation means, calibrator measurement means, and error identification means of the present disclosure, and executes the error identification method of the present disclosure in accordance with an error identification program stored in a numerical control device. The flow of the error identification method will be described below with reference to the flowchart of FIG. 2. First, in step (simply referred to as "S"; the same applies below) 1, preparations for measurement are made. Specifically, as shown in FIG. 3, a touch probe 10 is attached to the spindle head 2. Also, a ball array 11, which is a calibrator to which spheres Q1 to Q5 are fixed, is attached to the table 3. Furthermore, the feed axis is operated so that the tip of the touch probe 10 is positioned approximately directly above sphere Q1 (preparation stage). Next, in S2, the initial position of the ball array 11 is measured (initial position measurement step). The center positions of the spheres Q1 and Q2 are measured by the touch probe 10. Here, the touch probe 10 is a detector that outputs a trigger signal when the ball at the tip touches an object, and the numerical control device that receives the trigger signal can measure the position of the object by acquiring the current positions of the X, Y, and Z axes at the time of receiving the signal. The center position of the ball can be measured by touching the ball at four or more points (three or more points if the diameter of the ball is known).

[0012] Next, in S3, a reference angle of the rotation axis where the ball array 11 is in the reference position is calculated from the measurement value of the initial position (reference angle calculation step). Here, the reference position of the ball array 11 is set to a state where the ball array 11 is parallel to the X-axis as shown in FIG. 5. Therefore, the reference angle A of the A-axis is O is 0° (the C-axis is parallel to the Z-axis), and the reference angle of the C-axis is the C-axis angle at which the ball array 11 is parallel to the X-axis. When the A-axis is at 0° and the ball array 11 is placed on the table 3 as shown in FIG. 4, the installation angle θ0 of the ball array 11 can be calculated by the following formula 1, assuming that the measured central coordinate value of the sphere Q1 is PI1=(XI1, YI1, ZI1), the central coordinate value of the sphere Q2 is PI2=(XI2, YI2, ZI2), and the unit vector parallel to the X-axis is VX=(1, 0, 0). In addition, if the C-axis angle at the time of measuring the initial position is Cc, then the C-axis reference angle C O can be calculated from the following equation 2.

[0013] [Number 1] θ0=tan -1 {(VP × VX) / (VP VX)} Here, VP=(PI2-PI1) / |PI2-PI1| [Number 2] C O =C C +θ0

[0014] Next, in S4, a command value list for each axis is created based on the calibration value of the center position of each ball in the ball array 11, the ball array installation angle θ0, the reference angle of each rotation axis, and the angle of each rotation axis for achieving a preset measurement position. Next, in S5 to S7, each rotation axis is indexed according to the created command value list, and the center position of the ball on the ball array 11 is measured at each position (calibrator measurement stage). It is also possible to check whether each command position in the command value list is within the operating range of each axis, and if it is outside the range, to delete or omit that command position, or to output an alarm. In S5, the rotation axis is calculated so that the orientation of ball array 11 is parallel to the X-axis and Y-axis, respectively, and the central positions of spheres Q1 to Q5 of ball array 11 are measured (parallel measurement stage). When the X-axis is parallel, the A-axis angle is A as shown in Figure 5. O , C-axis angle is C O The measurement is then carried out. When the Y axis is parallel, the A axis angle is A as shown in Figure 6. O , C-axis angle is C O Index to -90° and take the measurement.

[0015] In S6, the rotation axis is determined so that the orientation of ball array 11 is parallel to the diagonal direction of the two linear axes, and the center positions of spheres Q1 to Q5 of ball array 11 are measured (plane diagonal measurement stage). In the case of the diagonal direction of the X and Y axes, the A axis angle is A as shown in Figure 7. O , C-axis angle is C O Index to -θxy. θxy is ±45° and measurements are performed in two directions. Measurement in either direction is also acceptable. In the case of the diagonal direction of the Y and Z axes, the A axis angle is A as shown in Figure 8. O -θyz, C-axis angle is C O The angle is calculated to -90°. Measurements are performed in two directions with θyz at ±45°. Measurement in one direction is also possible. However, if it is not possible to measure all of the balls in the ball array 11 due to the relationship between the movable range of the Z-axis and the central position of the A-axis, the absolute value of θyz is set to a smaller value. In this case, the closer it is to 0°, the greater the identification error in S8, so it is desirable to set the value as large as possible. If two or more balls can be measured, identification calculations in S8 are possible, but the smaller the maximum value of the measured inter-ball distance, the greater the identification error.

[0016] In the case of the diagonal direction of the Z and X axes, the A axis angle is set to A as shown in Figure 9. O -90°, C-axis angle is C O -θzx. Measurements are performed in two directions with θzx set to ±45°. However, with a touch probe 10 as shown in Figure 3, the touch probe 10 and spindle head 2 interfere with the ball array 11, and only some of the balls can be measured. This results in a large identification error in S8. Also, some balls cannot be measured due to the relationship between the movable range of the Z-axis and the central position of the A-axis, in which case the absolute value of θzx is set to a smaller value. If two or more balls can be measured, identification calculations in S8 are possible, but the smaller the maximum value of the measured inter-ball distance, the larger the identification error. Instead of measuring the diagonal directions of the Z and X axes, measurements in the diagonal directions of the X, Y, and Z axes in space may be performed as shown in FIG. 10 (space diagonal measurement step). In this case, the A-axis angle is set to A O -φsd, C-axis angle is C O-θsd. Measurements are performed in two directions with φsd at -45° and θsd at ±45°. If there is a ball that cannot be measured due to the relationship between the Z-axis movable range and the A-axis central position, the absolute value of at least one of φsd and θsd is set to a smaller value. In this case, the closer it is to 0°, the greater the identification error in S8, so it is desirable to set the value as large as possible. If two or more balls can be measured, identification calculations in S8 are possible, but the smaller the maximum value of the measured inter-ball distance, the greater the identification error. Although measurements are made in two diagonal directions, measurements may be made in only one of the diagonal directions.

[0017] On the other hand, measurements may be made in four diagonal directions in space without measuring the diagonal directions of the X and Y axes, the Y and Z axes, and the Z and X axes. In this case, the A-axis angle is set to A O -φsd, C-axis angle is C O If -θsd, measurements are taken at φsd of ±45° and θsd of ±45°. In S7, while one of the two rotation axes is indexed to an arbitrary angle so that the balls on ball array 11 trace an arc trajectory, the other rotation axis is sequentially indexed to multiple angles to measure the center position of ball Q5 on ball array 11 (arc measurement stage). In the case of arc measurement centered on the C axis, the A axis angle is set to A as shown in Figure 11. O and the C-axis angle is C O Then, multiple angles are calculated and the center position of sphere Q5 is measured at each angle. In the circular measurement centered on the A axis, the C axis angle is set to C as shown in Figure 12. O Index the A-axis angle to -90°. O The center position of each sphere Q5 is measured by dividing the pitch from -φra by Δφra. αf shown in FIG. 12 is the offset angle of the center position of the sphere Q5 on the YZ plane. The ball to be measured may be any one of balls Q1 to Q4 instead of Q5. Also, multiple balls may be measured. In S8, the geometric error of the linear axis is identified by calculation using the measurement values ​​in S5 and S6 (error identification stage). Details will be described later. In S9, the geometric error of the rotary axis is identified by using the measurement value in S7 and the geometric error of the linear axis identified in S8 (error identification step). Details will be described later.

[0018] Details of S8 will be described with reference to the flowchart of FIG. In S8-1, the X-axis error (positioning error, straightness) is identified. The central position of each ball of the ball array 11 is assumed to be calibrated by a three-dimensional measuring machine or the like, and the vector of the calibration value of the central position of the ball Qi (i=1 to 5) (the three-dimensional coordinate value of each ball center in the coordinate system of the ball array 11 with the center of the sphere Q1 as the origin) is assumed to be MOi=(XMi, YMi, ZMi). Here, the calibration value may be corrected according to the temperature of the ball array 11. If the center position measurement value vector of sphere Qi in the X-axis parallel measurement in S5 is PPXi=(XPXi, YPXi, ZPXi), the X-axis parallel measurement error vector DPXi=(dXPXi, dYPXi, dZPXi) of each sphere is expressed by the following equation. [Number 3] DPXi=PPXi-PPX1-MOi

[0019] If the n-th order approximation curve of dXPXi is FXX(x), the X-axis positioning error EXX(x) can be calculated using the following formula. [Number 4] EXX(x)=FXX(x)

[0020] If the least squares straight line of dYPXi is SYX(x) and the n-th order approximation curve (n≧2) is FYX(x), the Y component of straightness on the X-axis, EYX(x), can be found by the following equation. [Number 5] EYX(x)=FYX(x)-SYX(x)

[0021] If the least squares line of dZPXi is SZX(x) and the n-th order approximation curve (n ≥ 2) is FZX(x), the Z component of straightness of the X-axis, EZX(x), can be found by the following equation. [Number 6] EZX(x) = FZX(x) - SZX(x)

[0022] Next, in S8-2, the Y-axis error (positioning error, straightness) is identified. The calibration value vector MOi at the center position of each sphere is rotated 90° around the Z axis with the center of the sphere Q1 as the center, and the calibration value vector after this rotation is defined as MYi. If the center position measurement value vector of sphere Qi in the Y-axis parallel measurement in S5 is PPYi = (XPYi, YPYi, ZPYi), the Y-axis parallel measurement error vector DPYi = (dXPYi, dYPYi, dZPYi) of each sphere is expressed by the following equation. [Number 7] DPYi = PPYi - PPY1 - MYi

[0023] If the n-th order approximation curve of dYPYi is FYY(y), the Y-axis positioning error EYY(y) can be calculated using the following equation. [Number 8] EYY(y)=FYY(y)

[0024] If the least squares line of dZPYi is SZY(y) and the n-th order approximation curve (n≧2) is FZY(y), the Z component of straightness of the Y-axis, EZY(y), can be found by the following equation. [Number 9] EZY(y) = FZY(y) - SZY(y)

[0025] If the least squares straight line of dXPYi is SXY(y) and the n-th order approximation curve (n≧2) is FXY(y), the X-component of straightness of the Y-axis EXY(y) can be found by the following equation. [Number 10] EXY(y)=FXY(y)-SXY(y)

[0026] Next, in S8-3, the effects of the X- and Y-axis errors identified in S8-1 and S8-2 are removed from the diagonal measurement values. If the center position measurement value vector of the sphere Qi (i = 1 to 5) in the j-th diagonal measurement in S6 is PDji = (XDji, YDji, ZDji), the center position corrected measurement value vector CDji = (XCji, YCji, ZCji) of each sphere after correcting the positioning errors of the X and Y axes and straightness can be calculated by the following equation. [Number 11] XCji=XDji-{EXX(XDji)+EXY(YDji)} YCji=YDji-{EYX(XDji)+EYY(YDji)} ZCji=ZDji-{EZX(XDji)+EZY(YDji)}

[0027] Next, in S8-4, the squareness between the axes and the Z-axis error (positioning error expansion / contraction (gradient) component) are calculated. The calibration value vector of the center position of sphere Qi (i = 1 to 5) in the j-th diagonal measurement in S6 is rotated around the C and A axes around the center of sphere Q1, and the calibration value vector after this rotation is MDji = (XMDji, YMDji, ZMDji). The error vector DDji = (dXDji, dYDji, dZDji) of the diagonal measurement is expressed by the following equation. [Number 12] DDji=CDji-CDj1-MDji

[0028] The A-axis angle in the j-th diagonal measurement is A O +αj, C-axis angle is C O In this case, the unit vector in the length direction of ball array 11 is VDj=(cosγj,-cosαj·sinγj,sinαj·sinγj), and the gradient component dLji of the error in the center-to-center distance between the center of sphere Q1 and the center of each sphere is calculated by the following equation, where i=2 to 5. [Number 13] dLji=(DDji-DDj1)·VDj / (XMi-XM1)

[0029] The relationship between the YZ squareness EAYZ, the ZX squareness EBZX, the XY squareness ECXY, the Z-axis positioning error expansion / contraction component KZ, and dLji is as follows: [Number 14] dLji=Sj1*KZ+Sj2*EAYZ+Sj3*EBZX+Sj4*ECXY Sj1=-sin 2 αj*sin 2 γj Sj2=-sin2αj*sin 2 γj / 2 Sj3=-sinαj*sin2γj / 2 Sj4=-cosαj*sin2γj / 2

[0030] By solving the equation 14 as a system of simultaneous equations, EAYZ, EBZX, ECXY, and KZ can be obtained. For example, if a vector having KZ, EAYZ, EBZX, and ECXY as components is XX, a vector having dLji as components is BB, and a matrix having Sj1 to Sj4 as components is AA, then EAYZ, EBZX, ECXY, and KZ can be obtained from the following equations. [Number 15] XX=(AA T *AA) -1 *AA T *BB

[0031] The errors can also be calculated separately, in which case the effects of other errors or errors not to be identified can be eliminated. Only the XY perpendicularity is identified from the measurement results of the XY diagonal direction measurement. When θxy = ±45°, that is, α1 = 0°, γ1 = -45°, and α2 = 0°, γ2 = -45°, it can be found from the following formula. Here, mean() is a function that calculates the average value. This calculation is not affected by errors in the Z axis or errors in the Z direction of the X and Y axes. [Number 16] ECXY = mean(-dL1i + dL2i)

[0032] Next, the YZ squareness EAYZ and the Z-axis positioning error expansion / contraction component KZ are identified from the measurement results of the YZ diagonal direction measurement. When θyz = ±45°, that is, α3 = -45°, γ3 = -90°, and α4 = 45°, γ4 = -90°, they can be obtained from the following formula. In this case, there is no effect from the X-axis error or the X-direction errors of the Y and Z axes. [Number 17] KZ=mean(-dL3i-dL4i) EAYZ = mean(dL3i-dL4i)

[0033] Next, the ZX squareness EBZX is identified from the results of the spatial diagonal measurement. When φsd=45°, θsd=±45°, that is, α5=-45°, γ5=-45°, and α6=-45°, γ6=45°, it can be obtained from the following formula. [Number 18] EBZX=(2^0.5)*mean(-dL5i+dL6i)

[0034] Next, details of S9 will be described with reference to the flowchart of FIG. In S9-1, the influence of the positioning errors of the X-, Y-, and Z-axes, the straightness, and the squareness between the axes (straight axis geometric errors) identified in S8 are removed from the ball center measurement position measured in S7. If the center position measurement value vector of sphere Q5 at the kth index angle in the jth arc measurement (j=1 is an arc measurement centered on the C-axis, j=2 is an arc measurement centered on the A-axis) is PRjk = (XRjk, YRjk, ZRjk), then the corrected measurement value vector CLjk = (XLjk, YLjk, ZLjk) of the center position of each sphere after removing the positioning errors of the X-, Y-, and Z-axes, the straightness of the X- and Y-axes, and the perpendicularity between the axes can be calculated using the following formula. [Number 19] XLjk=XRjk-{EXX(XRjk)+EXY(YRjk)-ECXY*YRjk+EBZX*ZRjk} YLjk=YRjk-{EYX(XRjk)+EYY(YRjk)-EAYZ*ZRjk} ZLjk=ZRjk-{EZX(XRjk)+EZY(YRjk)+KZ*ZRjk}

[0035] In S9-2, the arc error of each ball center position measurement value group measured by indexing the rotation axis to draw an arc trajectory in S7 is calculated. The arc error is the radius error, and the horizontal and vertical directions of the center offset, which can be expressed by the 0th order coefficient, the 1st order cosine coefficient, and the 1st order sine coefficient of the Fourier coefficient, respectively. In the circular measurement centered on the C axis in S7, the A axis angle is A O and set the C-axis angle to C O +γck (k=1~nc) and measure it, where γck=Δθra*(k-1). If the command position of the center of sphere Q5 at the kth index angle of the jth measurement condition is PO1k = (XO1k, YO1k, ZO1k), the axial vector of the C axis is VA1k, the radial vector is VR1k, and the tangential vector is VT1k, the axial error EA1k, radial error ER1k, and tangential error ET1k in the C-axis arc measurement can be calculated from the following equations. [Number 20] EA1k=(CL1k-PO1k)·VA1k ER1k = (CL1k - PO1k) · VR1k ET1k=(CL1k-PO1k) VT1k

[0036] If the Fourier coefficients of the axial components are A0A1, A1A1, B1A1, the Fourier coefficients of the radial components are A0R1, A1R1, B1R1, and the Fourier coefficients of the tangential components are A0T1, A1T1, B1T1, the relationship between EA1k, ER1k, ET1k and each Fourier coefficient can be expressed by the following equation. By solving nc simultaneous equations for each direction, the arc error in each direction in C-axis arc measurement can be found. [Number 21] EA1k=A0A1+A1A1*cosγck+B1A1*sinγck ER1k=A0R1+A1R1*cosγck+B1R1*sinγck ET1k=A0T1+A1T1*cosγck+B1T1*sinγck

[0037] In the circular measurement centered on the A axis in S7, the C axis angle is C O Fix the A axis angle at -90° and set the A axis angle to A O +αck (k=1 to na), where αck=-φra-Δφra(k-1)-αf (αf is the offset angle of the center position of the sphere Q5 on the YZ plane). If the command position of the centre of sphere Q5 at the kth index angle for the jth measurement condition is PO2k = (XO2k, YO2k, ZO2k), the axial vector of the A-axis is VA2k, the radial vector is VR2k and the tangential vector is VT2k, the axial error EA2k, radial error ER2k and tangential error ET2k in the A-axis arc measurement can be calculated from the following equations. [Number 22] EA2k=(CL2k-PO2k)·VA2k ER2k = (CL2k - PO2k) · VR2k ET2k=(CL2k-PO2k) VT2k

[0038] If the Fourier coefficients of the axial components are A0A2, A1A2, B1A2, the Fourier coefficients of the radial components are A0R2, A1R2, B1R2, and the Fourier coefficients of the tangential components are A0T2, A1T2, B1T2, then, as with the C-axis arc measurement, the component errors in each direction can be expressed by the following equation using the respective Fourier coefficients, and by solving this as a system of na equations for each direction, the arc errors in each direction in the A-axis arc measurement can be found. [Number 23] EA2k=A0A2+A1A2*cosαck+B1A2*sinαck ER2k=A0R2+A1R2*cosαck+B1R2*sinαck ET2k=A0T2+A1T2*cosαck+B1T2*sinαck

[0039] In S9-3, the rotation axis error (center error (position error) and inclination error of the rotation axis) is calculated from the circular error in each direction calculated in S9-2. The relationship between the circular error in each direction found in S9-2 and the C-axis central position X-direction error dXca, the CA-axis offset error dYca, the A-axis central position Y-direction error dYax, the A-axis central position Z-direction error dZax, the A-axis angle offset error dAca, the CA-axis perpendicularity dBca, the AZ-axis perpendicularity dBax, and the AY-axis perpendicularity dCax is given by the following equation: Here, Zc is the center Z position of sphere Q5 relative to the A-axis center in the C-axis circular measurement, Rc is the radius of gyration of sphere Q5 in the C-axis circular measurement, and Ra is the radius of gyration of sphere Q5 in the A-axis circular measurement. [Number 24] A1R1=-dXca-Zc*dBax B1R1=-dYca-dYax+Zc*dAca A1R2=-dYax B1R2=-dZax A1A1=Rc*(dBca+dBax) B1A1=-Rc*dAca A1A2=-Ra*dCax B1A2=-Rc*dBax

[0040] The geometric error of the rotary axis can be obtained by solving Equation 24. The identified geometric error of the rotary axis can be identified with higher accuracy than before because it is identified after removing the effect of the geometric error of the linear axis alone.

[0041] In this manner, the error identification method and error identification program, and the five-axis controlled machining center of the above-described embodiment include a preparation stage S1 in which a touch probe 10 (position measurement sensor tool) is attached to a spindle head 2 (spindle) and a ball array 11 (calibrator) having five balls Q1 to Q5 is fixed to a table 3, an initial position measurement stage S2 in which an initial position of the ball array 11 is measured by the touch probe 10, and a reference angle of the A-axis and C-axis (rotation axes) for positioning the ball array 11 at a reference position is calculated based on the measurement value in the initial position measurement stage. The process includes a reference angle calculation step S3 in which the A-axis and C-axis are respectively indexed to a plurality of index angles relative to the reference angle, calibrator measurement steps S5-S7 in which the A-axis and C-axis are respectively indexed to a plurality of index angles relative to the reference angle and the center positions of balls Q1-Q5 of a ball array 11 fixed to a table 3 at each index angle are measured using a touch probe 10, and error identification steps S8 and S9 in which the positioning errors and straightness errors of the X-, Y-, and Z-axes and the squareness errors between the axes, and the position errors and inclination errors of the A-axis and C-axis are identified based on the measurement values ​​in the calibrator measurement step.

[0042] This configuration makes it possible to simultaneously identify the geometric errors of the X-, Y-, and Z-axes (positioning errors, straightness) and the inter-axis geometric errors (squareness between the X-, Y-, and Z-axes, center errors of the C- and A-axes, and tilt errors of the C- and A-axes). This makes it possible to accurately identify the inter-axis geometric errors without being affected by the geometric errors of the linear axes. Furthermore, by using the C-axis and A-axis of a 5-axis controlled machining center, a series of measurements can be performed automatically, making it possible to automatically identify the axis-specific geometric error and inter-axis geometric error of linear axes without the machine operator's knowledge or measurement skills.

[0043] It should be noted that the position measurement sensor tool can be anything other than a touch probe. The calibration tool can also be anything other than the ball array described above. For example, the number of balls can be three or more. The 5-axis controlled machine tool is not limited to the above-mentioned type of machining center. [Explanation of symbols]

[0044] 1··bed, 2··spindle head, 3··table, 4··cradle, 5··trunnion, 10··touch probe, 11··ball array, Q1~Q5··balls.

Claims

1. A method for identifying an error in a five-axis controlled machine tool having a rotatable main spindle on which a tool is attached, a table on which a workpiece and / or a jig can be fixed, three linear axes which are mutually perpendicular and which allow the main spindle to move relatively with respect to the table, and two rotary axes which allow the table to be turned and / or tilted, comprising: a preparation step of mounting a position measurement sensor tool on the spindle and fixing a calibrator having three or more collinearly arranged spheres on the table; an initial position measuring step of measuring an initial position of the calibrator by the position measuring sensor tool; a reference angle calculation step of calculating a reference angle of each of the rotation shafts for positioning the calibrator at a predetermined reference position from a measurement value in the initial position measurement step; a calibrator measurement step of indexing each of the rotation axes to a plurality of index angles with respect to the reference angle and measuring the center position of the sphere of the calibrator fixed to the table at each of the index angles by the position measurement sensor tool; an error identification step of identifying a positioning error and a straightness error of the linear axis, a squareness error between the linear axes, and a position error and an inclination error of each of the rotation axes based on the measurement values ​​in the calibration step; 1. A method for identifying errors in a five-axis controlled machine tool, comprising:

2. In the error identification step, 2. The error identification method for a five-axis controlled machine tool according to claim 1, characterized in that the measurement values ​​in the calibration measurement stage are used to identify the positioning errors and straightness errors of the two linear axes, the three squareness errors of the linear axes, the expansion and contraction components of the positioning error of one linear axis other than the two linear axes, and the position errors and inclination errors of each of the rotation axes.

3. In the calibrator measurement step, 3. The method for identifying errors in a five-axis controlled machine tool according to claim 1 or 2, further comprising the step of performing an arc measurement step in which, while one of the rotation axes is indexed to any one of the index angles so that the ball of the calibrator describes an arc trajectory, the other rotation axis is sequentially indexed to a plurality of the index angles, thereby measuring the center position of at least one of the balls of the calibrator.

4. In the calibrator measurement step, a parallel measurement step of indexing each of the rotation axes so that the orientation of the straight line on which the balls of the calibrator are arranged is parallel to the axis lines of the two linear axes, and measuring the centers of three or more balls of the calibrator using the position measurement sensor tool; A plane diagonal measurement step of indexing each of the rotation axes so that the direction of the straight line of the calibrator is a diagonal direction of a plane formed by the axes of the two linear axes, and measuring the center positions of three or more spheres of the calibrator with the position measurement sensor tool; 4. The method for identifying errors in a five-axis controlled machine tool according to claim 1, further comprising the steps of:

5. In the calibrator measurement step, a parallel measurement step of indexing each of the rotation axes so that the orientation of the straight line on which the balls of the calibrator are arranged is parallel to the axis lines of the two linear axes, and measuring the center positions of three or more balls of the calibrator using the position measurement sensor tool; a spatial diagonal measurement step of determining each of the rotation axes so that the direction of the straight line of the calibrator is a diagonal direction between each of the axes in a space formed by the axes of the three linear axes, and measuring the center positions of three or more spheres of the calibrator using the position measurement sensor tool; 4. The method for identifying errors in a five-axis controlled machine tool according to claim 1, further comprising the steps of:

6. 6. An error identification program for a five-axis controlled machine tool, for causing a numerical control device of a five-axis controlled machine tool having a rotatable main spindle on which a tool can be attached, a table on which a workpiece and / or a jig can be fixed, three linear axes which are mutually perpendicular and which allow the main spindle to move relatively to the table, and two rotary axes which allow the table to be turned and / or tilted, to execute the error identification method for a five-axis controlled machine tool according to any one of claims 1 to 5.

7. A five-axis controlled machine tool having a rotatable main spindle on which a tool is attached, a table on which a workpiece and / or a jig can be fixed, three linear axes which are mutually perpendicular and which allow the main spindle to move relatively with respect to the table, and two rotary axes which allow the table to turn and / or tilt, an initial position measuring means for measuring an initial position of a calibrator fixed to the table, the calibrator having three or more balls arranged on a same straight line, by using a position measuring sensor tool attached to the spindle; a reference angle calculation means for calculating a reference angle of each of the rotation shafts for positioning the calibrator at a predetermined reference position from a measurement value by the initial position measurement means; a calibrator measuring means for indexing each of the rotation axes to a plurality of index angles with respect to the reference angle, and measuring the center position of the sphere of the calibrator fixed to the table at each of the index angles by the position measuring sensor tool; an error identifying means for identifying a positioning error and a straightness error of the linear axis, a squareness error between the linear axes, and a position error and an inclination error of each of the rotation axes based on the measurement values ​​obtained by the calibration measuring means; A five-axis controlled machine tool comprising:

Images