Geometric error identification method for machine tool

Abstract

To provide a geometric error identification method capable of easy identification of a geometric error between shafts in a machine tool with a narrow movable range.SOLUTION: A geometric error identification method for a machine tool includes the following steps: (a1) positioning a reference sphere the center of which aligns with a tool spindle; (a2) mounting a position measurement device onto a rotating part such that a probe of the position measurement device faces a rotation center line; (a3) measuring a position of the reference sphere with the probe while rotating either one of rotation objects of a slewing part and the rotating part in a state in which the probe faces the center of the reference sphere, and thereby, acquiring displacement data corresponding to a rotation angle of the rotation object with reference to the center of the reference sphere; and (a4) identifying the geometric error by means of a least square method, using a mathematical expression including a term representing the geometric error and representing the displacement according to the rotation angle, and the displacement data.SELECTED DRAWING: Figure 11

Description

[Technical Field]

[0001] The present invention relates to a method for identifying geometric errors in a machine tool. [Background technology]

[0002] Sliding-spindle lathes, which use tools to machine a workpiece held by the spindle, are well known as machine tools. Sliding-spindle lathes are capable of performing complex machining using multiple tools while continuously supplying bar material from the back of the spindle, making them suitable for mass production of small precision parts. To perform complex machining, sliding-spindle lathes are sometimes equipped with a swivel unit that changes the orientation of the tool around the swivel centerline. Because sliding-spindle lathes equipped with a swivel unit have a large number of axes, it is thought that there will be many geometric errors between the axes.

[0003] Previous research has focused on methods for evaluating geometric errors in large multi-task machine tools that combine the functions of both NC (numerically controlled) lathes and machining centers. The main methods used to evaluate the accuracy of multi-task machine tools are ball bars and R-tests, which measure the relative positional relationship between any two points on the machine tool, as specified in ISO (International Organization for Standardization) 10791-6.

[0004] Patent Document 1 discloses a method for identifying geometric errors in a five-axis controlled turning center, which has three linear axes controlled for translation in the X, Y, and Z directions and two rotary axes controlled for rotation along the B and C axes. This geometric error identification method includes the steps of: installing a reference sphere on the C-axis side and attaching a touch probe to the B-axis side; measuring the center position of the reference sphere while indexing the B-axis angle at predetermined intervals with the C-axis fixed; measuring the center position of the reference sphere as in the first measurement step by changing the turning radius of the measurement point of the touch probe; measuring the center position of the reference sphere while indexing the C-axis angle at predetermined intervals with the B-axis fixed; and measuring the center position of the reference sphere while changing the initial position of the reference sphere in the Z direction as in the third measurement step. Because the reference sphere is positioned at a position offset from the center of rotation of the C-axis, it moves significantly as the C-axis rotates. [Prior art documents] [Patent documents]

[0005] [Patent Document 1] Japanese Patent Application Publication No. 2018-142064 Summary of the Invention [Problem to be solved by the invention]

[0006] Since sliding-spindle lathes are designed to process small products, their range of motion is extremely small, making it impossible to evaluate them using ball bars or R-tests, which are commonly used to evaluate the accuracy of large multi-task machine tools. Furthermore, no research has been conducted on methods other than ball bars and R-tests to evaluate the accuracy of geometric errors in automatic lathes. Therefore, for machine tools with a narrow range of motion like sliding-spindle lathes, the types of geometric errors that exist have not been clarified, and the impact of these geometric errors on machining accuracy and evaluation methods have not been clarified.

[0007] The present invention discloses a method for identifying geometric errors that can easily identify inter-axis geometric errors in machine tools with narrow movable ranges. [Means for solving the problem]

[0008] The geometric error identification method for a machine tool of the present invention is a geometric error identification method for identifying a geometric error between axes from a workpiece to the tool spindle in a machine tool including: a rotating unit that can rotate together with a workpiece about a rotation center line; a swivel unit that has a tool spindle and can swivel about a swivel center line that is oriented differently from the tool spindle; and a control unit that controls movement of the rotation axis about the rotation center line, movement of the swivel axis about the swivel center line, and movement of a plurality of linear axes, the method comprising: (a1) placing a reference sphere so that its center is aligned with the tool spindle; (a2) attaching the position measurement device to the rotating part so that a probe of the position measurement device faces the rotation center line; (a3) measuring the position of the reference sphere with the probe while rotating one of the rotation objects of the swivel unit and the rotating unit with the probe facing the center of the reference sphere, thereby obtaining displacement data according to the rotation angle of the rotation object with the center of the reference sphere as a reference; (a4) identifying the geometric error by a least squares method using a mathematical expression that has a term including the geometric error and that represents a displacement according to the rotation angle, and the displacement data; The present invention has an aspect including the following. [Effects of the Invention]

[0009] According to the present invention, it is possible to provide a geometric error identification method that can easily identify inter-axis geometric errors in a machine tool with a narrow range of motion. [Brief explanation of the drawings]

[0010] [Figure 1] FIG. 1 is a perspective view schematically showing an example of a moving spindle type lathe. [Figure 2] FIG. 10 is a diagram schematically illustrating a geometric error between axes. [Figure 3]FIG. 10 is a diagram schematically showing an example in which a reference sphere and a position measuring device are provided on a moving-spindle type lathe. [Figure 4] FIG. 10 is a diagram schematically showing an example of a tool-side jig. [Figure 5] FIG. 1 is a diagram schematically illustrating an example of a position measurement device. [Figure 6] FIG. 10 is a diagram showing a schematic example of how the X and Z directions of the center of the reference sphere change when the swivel unit rotates -90°. [Figure 7] FIG. 10 is a diagram schematically showing an example of measuring the displacement of a reference sphere by a probe while rotating a rotating part. [Figure 8] FIG. 10 is a diagram showing a schematic example of measuring the displacement of a reference sphere in the X direction with a measuring probe by performing circular interpolation on the Z and X axes while rotating the swivel unit. [Figure 9] This is a diagram showing a schematic example of measuring the displacement of a reference sphere in the Y direction using a measuring probe by performing circular interpolation on the Z and X axes while rotating the swivel unit. [Figure 10] FIG. 10 is a diagram schematically illustrating an example of a procedure for identifying geometric errors from displacement data in the radial direction of the C-axis. [Figure 11] FIG. 1 is a diagram showing a schematic example of the procedure for identifying geometric errors from displacement data when the B, Z, and X axes are controlled simultaneously. [Figure 12] FIG. 10 is a diagram showing an example of displacement data when the B, Z, and X axes are controlled simultaneously. DETAILED DESCRIPTION OF THE INVENTION

[0011] The following describes embodiments of the present invention. Of course, the following embodiments are merely examples of the present invention, and not all of the features shown in the embodiments are necessarily essential to the solution of the invention.

[0012] (1) Summary of the aspects included in the present invention: First, an overview of the embodiments included in the present invention will be described with reference to the examples shown in Figures 1 to 12. Note that the figures in this application are diagrams showing examples in a schematic manner, and the magnifications in the directions shown in these figures may differ, and the figures may not be consistent with each other. Of course, each element of this embodiment is not limited to the specific example indicated by the symbol. In addition, in the present application, a numerical range "Min to Max" means a value equal to or greater than the minimum value Min and equal to or less than the maximum value Max.

[0013] [Aspect 1] A geometric error identification method for a machine tool (e.g., lathe 1) according to one embodiment is directed to a machine tool (1) including: a rotating unit (e.g., front spindle 10) that can rotate together with a workpiece W1 about a rotation center line AX1; a swivel unit 20 that has a tool spindle (e.g., tool spindle S2) and can swivel about a rotation center line AX2 that is oriented differently from the tool spindle (S2); and a control unit 30 that controls the movement of the rotation axes about the rotation center line AX1, the movement of the swivel axes about the swivel center line AX2, and the movement of multiple linear axes. This geometric error identification method identifies geometric errors between axes from the workpiece W1 to the tool spindle (S2), and includes the following steps. (a1) Step of arranging a reference sphere 51 so that its center 52 is aligned with the tool spindle (S2). (a2) Step of attaching the position measuring device 60 to the rotating part (10) so that the probe 62 of the position measuring device 60 faces the rotation center line AX1. (a3) A step of measuring the position of the reference sphere 51 with the probe 62 facing the center 52 of the reference sphere 51 while rotating one of the rotating objects, the swivel unit 20 and the rotating unit (10), thereby obtaining displacement data corresponding to the rotation angle of the rotating object with the center 52 of the reference sphere 51 as a reference. (a4) identifying the geometric error by a least squares method using a mathematical expression that has a term including the geometric error and that expresses a displacement according to the rotation angle, and the displacement data;

[0014] The above formula can be derived from a mathematical model constructed in consideration of the geometric error existing between the axes. The formula has terms including the geometric error, as in formulas (6) to (9) described below, and expresses the displacement according to the rotation angle of the rotation object. By preparing the formula and then performing steps (a1) to (a4), the geometric error between the axes can be identified. First, the reference sphere 51 is placed so that its center 52 is aligned with the tool spindle (S2), and the position measuring device 60 is attached to the rotating unit (10) so that the probe 62 of the position measuring device 60 faces the rotation center line AX1. By measuring the position of the reference sphere 51 with the probe 62 facing the center 52 of the reference sphere 51 while rotating the rotation object, displacement data corresponding to the rotation angle of the rotation object with the center 52 of the reference sphere 51 as the reference is obtained. The geometric error is identified by the least squares method using the above formula and the displacement data.

[0015] By arranging the reference sphere 51, to which the stylus 62 of the position measuring device 60 is directed, on the tool spindle (S2), it is possible to reduce the movement of the reference sphere 51 compared to when the reference sphere 51 is installed at a position shifted from the rotation center line AX1. This reduces the space required for measurement, making measurement possible even when the position measuring device 60 cannot be attached to the tool spindle (S2). Then, using the acquired displacement data and the above formula, the geometric error is identified with high accuracy by the least squares method. Therefore, the above aspect can provide a geometric error identification method that can easily identify inter-axis geometric errors in machine tools with a narrow range of motion.

[0016] There are various examples of the above-described aspects. The multiple linear axes may be, for example, three axes: an X-axis, a Y-axis, and a Z-axis, but are not limited to this example and may be two or more axes. Furthermore, the linear axes may be oriented in different directions, and may or may not be perpendicular to each other. The rotation axis may be, for example, the C axis around the Z axis, but is not limited to this example and may be the A axis around the X axis, the B axis around the Y axis, or the like. The rotation axis may be, for example, the B axis around the Y axis, but is not limited to this example and may be the A axis around the X axis, the C axis around the Z axis, or the like. The tool spindle refers to a part of the swivel section to which a tool is attached, and is not limited to being rotatable, and may be non-rotatable. When the geometric error included in the terms of a mathematical formula is expressed by a coefficient of the mathematical formula, the mathematical formula can be said to have a coefficient representing the geometric error. In this case, the geometric error can be identified by finding the coefficient using the mathematical formula and displacement data by the least squares method. The mathematical formula may also include a constant representing the geometric error. Of course, the above remarks also apply to the following aspects.

[0017] [Aspect 2] As illustrated in FIG. 8 etc., in step (a3), at least some of the plurality of rectilinear axes may be controlled so that the probe 62 is maintained facing the center 52 of the reference sphere 51 regardless of the rotation of the rotation object. In the above case, since no apparent relative linear movement occurs in the reference sphere 51 when acquiring displacement data according to the rotation angle, it is possible to more easily identify inter-axis geometric errors in machine tools with narrow movable ranges.

[0018] [Aspect 3] The plurality of linear axes may include a first linear axis along the rotation center line AX1 and a second linear axis oriented in a direction different from both the rotation center line AX1 and the turning center line AX2. Step (a3) ​​may include the following steps. (b1) A step of rotating the tool spindle (S2) at a first rotation angle (e.g., B-axis 0°) around the rotation center line AX2, bringing the probe 62 into contact with the reference sphere 51, and determining a first center position (e.g., Z-direction position Z0) of the reference sphere 51 on the first translational axis based on the position of the probe 62 which changes in accordance with the movement of the first translational axis. (b2) The tool spindle (S2) is rotated around the rotation center line AX2 at a second rotation angle (for example, B-axis -90°) different from the first rotation angle, and the probe 62 is brought into contact with the reference sphere 51. A second center position (for example, Z-direction position Z) of the reference sphere 51 on the first translational axis is determined based on the position of the probe 62 which changes in accordance with the movement of the first translational axis. -90 ) step. (b3) Based on the first center position and the second center position, the distance between the swivel center 21 of the swivel unit 20 and the center 52 of the reference sphere 51 (for example, an estimated value of the tool length) B l n ) step. (b4) A step of making the first linear axis and the second linear axis move in an arc with the distance as a radius so that the probe 62 is maintained facing the center 52 of the reference sphere 51 when the rotation object is the swivel part 20.

[0019] In the above case, the relative position of the reference sphere 51 can be easily controlled to a constant value based on the distance between the center of rotation 21 of the swivel unit 20 and the center 52 of the reference sphere 51. Therefore, the above embodiment can provide a suitable example in which the reference sphere does not appear to undergo relative rectilinear motion.

[0020] Here, the terms "first," "second," etc. in this application are terms for distinguishing between elements among a plurality of elements having similarities, and do not indicate an order. Which elements among a plurality of elements fall under the categories of "first," "second," etc. is determined relatively. The above remarks also apply to the following aspects.

[0021] [Aspect 4] The first translational axis may be the Z axis, and the second translational axis may be the X axis. The orientation of the tool spindle (S2) at the first pivot angle may be along the X axis, and the orientation of the tool spindle (S2) at the second pivot angle may be along the Z axis. In the above case, the distance between the swivel center 21 of the swivel unit 20 and the center 52 of the reference sphere 51 can be easily calculated from the difference between the first center position and the second center position on the Z axis. Therefore, the above embodiment can provide a suitable example that does not cause rectilinear motion in the reference sphere.

[0022] [Aspect 5] In step (a3), as illustrated in Figures 6 to 8, the displacement data may be acquired by measuring the position of the reference sphere 51 with the probe 62 while rotating the rotational object within a predetermined angle range of 180° or less. By using a formula that expresses displacement according to the rotation angle to identify the geometric error, displacement data according to the rotation angle can be obtained with a rotation of 180° or less. Therefore, the above-mentioned aspect makes it possible to more easily identify inter-axis geometric errors in machine tools with a narrow range of motion. When the linear axes are perpendicular to each other, the angle range is preferably 90° or greater. The reason is that the direction of the geometric error is defined as parallel to the directions of the two perpendicular linear axes, and if measurements are performed at angles of 90° or greater, the displacement data will contain measurement values ​​in the same direction as the direction of the geometric error to be identified.

[0023] Although not included in the above-mentioned embodiment 5, the disclosure of the present application also includes acquiring displacement data while rotating the rotation target by an angle greater than 180° but less than or equal to 360°, such as rotating the rotation target by 360°.

[0024] [Aspect 6] 10 and 11 , in step (a3), the rotational object may be rotated in a first direction within the predetermined angle range while measuring the position of the reference sphere 51 with the probe 62, thereby obtaining outward displacement data as the displacement data. Alternatively, in step (a3), the rotational object may be rotated in a second direction opposite to the first direction within the predetermined angle range while measuring the position of the reference sphere 51 with the probe 62, thereby obtaining backward displacement data as the displacement data. In step (a4), the forward displacement data and the backward displacement data may be averaged to identify the geometric error by the least squares method. The measurement results for the position of the reference sphere 51 often vary slightly depending on the direction of motion due to the influence of backlash and frictional forces present in the drive mechanism. By averaging the forward displacement data and the backward displacement data, the influence on displacement due to differences in the direction of motion of the rotating object is suppressed. Therefore, the above embodiment can improve the accuracy of identifying geometric errors.

[0025] [Aspect 7] In step (a3), first displacement data may be acquired as the displacement data by measuring the position of the reference sphere 51 under first measurement conditions while rotating the rotational object. Alternatively, in step (a3), second displacement data may be acquired as the displacement data by measuring the position of the reference sphere 51 under second measurement conditions different from the first measurement conditions while rotating the rotational object. In step (a4), the geometric error may be identified by determining the first coefficient by a least squares method using a first mathematical formula having a first coefficient representing the geometric error and the first displacement data, and determining the second coefficient by a least squares method using a second mathematical formula having a second coefficient representing the geometric error and the second displacement data. In the above case, the coefficients of a plurality of mathematical formulas are calculated based on displacement data acquired under a plurality of measurement conditions, so that a larger number of geometric errors can be identified.

[0026] [Aspect 8] As illustrated in FIG. 1, the multiple linear axes may include an X-axis, a Y-axis, and a Z-axis. The Y-axis may be along the pivot center line AX2, and the Z-axis may be along the rotation center line AX1. The rotating unit (10) may be a spindle (e.g., a front spindle 10) capable of gripping the workpiece W1 and moving along the rotation center line AX1. The rotating unit 20 may be a B-axis unit to which a tool TO1 for machining the workpiece W1 is attached along the tool spindle (S2). In the above case, it is possible to provide a geometric error identification method that can easily identify inter-axis geometric errors in a moving-spindle type lathe with a narrow movable range.

[0027] (2) Example of axis configuration for a moving spindle lathe: FIG. 1 shows a schematic configuration example of a lathe 1 as an example of a machine tool. 1 is a moving-spindle type lathe as an example of an automatic lathe, and is also an NC (numerically controlled) lathe equipped with a control unit 30, also known as an NC device. The lathe 1 is equipped with a front spindle 10, a guide bush 11, a back spindle 12, a gang tool rest 15 having a swivel unit 20, a back machining tool rest 25, the control unit 30, and an operation unit 31.

[0028] When the workpiece W1 is long, it is supplied to the face spindle 10 from the rear end thereof. The face spindle 10 releasably holds the workpiece W1 and is rotatable together with the workpiece W1 about the rotation center line AX1. In other words, the workpiece W1 is rotatable about the rotation center line AX1 by being held by the face spindle 10. The face spindle 10 is also movable along the rotation center line AX1. The control unit 30 controls the translational movement of the face spindle 10 along the Z1 axis and controls the rotational movement of the face spindle 10 about the rotation center line AX1 as the C1 axis around the Z1 axis. The guide bush 11 supports the workpiece W1, which protrudes forward from the face spindle 10, so that it can slide in the Z1 direction. Note that the guide bush 11 may not be provided in the lathe 1, the workpiece W1 may be short, and the workpiece W1 may be supplied to the face spindle 10 from the front end thereof. If the workpiece W1 is short, it may be fed from the rear end of the front spindle 10 or from the front end of the front spindle 10 using a loader. Therefore, even if the workpiece W1 is short, the geometric error identification method can be applied. The back spindle 12 releasably holds the workpiece W1 after front machining and is rotatable together with the workpiece W1 around the rotation center line AX1. In other words, the workpiece W1 is rotatable around the rotation center line AX1 by being held by the back spindle 12. The back spindle 12 is movable along the rotation center line AX1 and also along the X2 axis. The control unit 30 controls the linear motion of the back spindle 12 along the Z2 axis, controls the linear motion of the back spindle 12 along the X2 axis, and controls the rotational motion of the back spindle 12 around the rotation center line AX1 as the C2 axis around the Z2 axis.

[0029] A plurality of tools TO1 for machining a workpiece W1 are directly or indirectly attached to the gang tool rest 15 and the back-machining tool rest 25. The control unit 30 controls the translational motion of the gang tool rest 15 having the swivel unit 20 along the X1 axis, controls the translational motion of the gang tool rest 15 having the swivel unit 20 along the Y1 axis, and controls the translational motion of the back-machining tool rest 25 along the Y2 axis. The swivel unit 20 has tool spindles S1, S2, ... and can rotate around a rotation center line AX2 that is oriented differently from the tool spindles. FIG. 1 shows a tool spindle center line AX3, which represents the center line of the tool spindle S2. The control unit 30 controls the rotational motion of the swivel unit 20 around the rotation center line AX2 as the B axis around the Y1 axis. Therefore, the swivel unit 20 can be said to be a B-axis unit to which the tool TO1 for machining the workpiece W1 is attached along the tool spindles S1, S2, . . .

[0030] The control unit 30 includes a CPU (Central Processing Unit) which is a processor, a ROM (Read Only Memory) in which a control program is written, a RAM (Random Access Memory) in which a machining program is stored, a clock circuit, an interface, etc. The CPU uses the RAM as a work area and executes the control program recorded in the ROM to realize the functions of the NC device. Of course, some or all of the functions realized by the control program may be realized by other means such as an ASIC (Application Specific Integrated Circuit). An operation unit 31 having an input unit and a display unit is connected to the control unit 30. The operation unit 31 can receive an operation from an operator to manually operate each axis.

[0031] Here, the X1-axis and X2-axis are collectively referred to as the X-axis, the Y1-axis and Y2-axis are collectively referred to as the Y-axis, the Z1-axis and Z2-axis are collectively referred to as the Z-axis, and the C1-axis and C2-axis are collectively referred to as the C-axis. The X-axis, Y-axis, Z-axis, B-axis, and C-axis can be considered control axes controlled by the control unit 30. In this specific example, the front spindle 10 is used as an example of a rotating unit, and the tool spindle S2, the second-lowest one in the swivel unit 20, is used as an example of a tool spindle. The geometric error between the axes from the workpiece W1 gripped by the front spindle 10 to the tool spindle S2 is the target of identification. Therefore, in this specific example, the multiple linear axes include the X-axis, Y-axis, and Z-axis, with the Y-axis extending along the swivel center line AX2 and the Z-axis extending along the rotation center line AX1. The Z-axis is a first linear axis along the rotation center line AX1, the X-axis is a second linear axis oriented in a direction different from both the rotation center line AX1 and the pivot center line AX2, the B-axis is a rotation axis about the pivot center line AX2, and the C-axis is a rotation axis about the rotation center line AX1. The control unit 30 controls the five-axis motion, i.e., the rotational motion of the C-axis as a rotation axis, the rotational motion of the B-axis as a pivot axis, and the linear motion of the X-, Y-, and Z-axes as multiple linear axes.

[0032] The orientations of the X, Y, and Z axes are not particularly limited. In the example shown in FIG. 1, the X and Z axes are horizontal, and the Y axis is vertical. That is, the rotation center line AX1 is horizontal, and the pivot center line AX2 is vertical. The tool spindle center line AX3 is horizontal. The X and Y axes are orthogonal to each other, but they do not have to be orthogonal as long as they are oriented in different directions. The X and Z axes are orthogonal to each other, but they do not have to be orthogonal as long as they are oriented in different directions. The Y and Z axes are orthogonal to each other, but they do not have to be orthogonal as long as they are oriented in different directions. Note that translatory axes being oriented in different directions can be rephrased as translatory axes intersecting each other. The tool spindle S2 is orthogonal to the Y axis, but it does not have to be orthogonal as long as it is oriented in a different direction. Furthermore, the explanation of the positional relationships of each part is merely an example, and therefore, the present application also includes changing the left-right direction to the up-down direction or the front-back direction, changing the up-down direction to the left-right direction or the front-back direction, changing the front-back direction to the left-right direction or the up-down direction, changing the rotation direction to the opposite direction, etc.

[0033] (3) Mathematical model considering geometric errors: For convenience, the tool spindle S2 will also be called the S2 axis. From the workpiece W1 held by the front spindle 10 to the tool spindle S2, there are the C1 axis, Z1 axis, X1 axis, Y1 axis, B axis, and S2 axis. Based on shape generation theory, each of the six axes is defined as having its own coordinate system. Each coordinate system has a translation error δ in the X, Y, and Z directions. x ,δ y ,δ z , and rotation errors α, β, and γ around the X, Y, and Z axes, respectively. However, by expressing the geometric errors as relative errors between adjacent coordinate axes, it is not necessary to consider all six geometric errors.

[0034] Figure 2 shows a schematic diagram of the geometric error between adjacent coordinate axes. For example, between the C1 and Z1 axes, there is a negligible error in the C1-axis motion, γ, and a negligible error in the Z1-axis motion that does not overlap with the C1-axis motion, that is, δ. x ,δ y ,δ z Therefore, the geometric error to be considered between the C1 axis and the Z1 axis is α C1Z ,β C1Z In the notation of each rotation error, the subscript "C1" means the C1 axis, and the subscript "Z" means the Z axis as the Z1 axis. That is, the rotation error α C1Z ,β C1Z indicates the geometric error between the axis indicated by the first subscript and the axis indicated by the second subscript. The geometric errors shown in Figure 2 should be considered between the Z1 and X1 axes, between the X1 and Y1 axes, between the Y1 and B axes, and between the B and S2 axes.

[0035] First, we will build a mathematical model that does not take geometric error into consideration. To do this, we will define the workpiece coordinate system fixed to the workpiece origin, the B-axis coordinate system fixed to the B-axis rotation center, and the tool coordinate system fixed to the tool center point. Then, we will define the vector representation of the tool center point position in the workpiece coordinate system as W XT The vector notation of the B-axis rotation center position in the workpiece coordinate system is W X B The vector representation of the tool center point position in the B-axis coordinate system is B X T The subscripts of the vector symbols indicate that the upper left is the coordinate system and the lower right is the target coordinate. C-axis direction rotational angular displacement D C , a position vector representing the B-axis rotation center position in the workpiece coordinate system W X B , and a position vector representing the tool center point coordinates in the B-axis coordinate system B X T Using the above, the position vector representing the tool center point position in the workpiece coordinate system is W X T is expressed by the following formula:

number

[0036] Here, the vector representing the x, y, and z direction displacement of the machine tool is D xyz , the position vector representing the initial position of the B-axis rotation center in the workpiece coordinate system is W X B0 Then, W X B ,D xyz , W X B0 The relationship is expressed by the following formula:

number

[0037] Furthermore, the position vector representing the tool center point coordinates in the B-axis coordinate system when the B-axis rotation angle is 0 degrees is B If we assume L, B XT , B The relationship of L is expressed by the following formula.

number

[0038] From equations (1), (2), and (3), the mathematical model constructed from the axis configuration of the lathe 1 shown in Figure 1 without considering geometric errors is expressed by the following equation.

number

[0039] Geometric errors are expressed as positions and angles between coordinate systems fixed to each axis. The geometric errors that exist between the coordinate systems are shown in Figure 2. By using a homogeneous transformation matrix that takes these geometric errors into account, the tool motion in the workpiece coordinate system can be expressed. The constructed mathematical model is expressed by the following equation.

number

[0040] The homogeneous transformation matrix corresponding to each geometric error is expressed as follows:

number

number

number

number

number

number

number

number

number

number

number

number

number

[0041] (4) Specific examples of geometric error identification methods: Fig. 3 shows a schematic example of a lathe 1 having a moving spindle, in which a reference sphere 51 and a position measuring device 60 are provided to identify geometric errors between axes. Fig. 4 shows a schematic example of a tool-side jig 50 including the reference sphere 51. Fig. 5 shows a schematic example of a position measuring device 60. The geometric error identification method of this specific example includes the above-mentioned steps (a1) to (a4).

[0042] In step (a1), a reference sphere 51 is positioned so that its center 52 is aligned with the tool spindle S2. The tool-side jig 50 includes the reference sphere 51, a base 53 connected to the reference sphere 51, a socket 54 that holds the base 53 so that the position of the center 52 of the reference sphere 51 can be adjusted, and a cylinder 55 connected to the socket 54. The cylinder 55 is attached to the tool spindle S2 along the tool spindle center line AX3. The socket 54 has a recess into which the base 53 is inserted and is threadedly engaged with multiple adjustment screws 56. Each adjustment screw 56 penetrates the socket 54 from its outer periphery to the recess, and the position of the center 52 of the reference sphere 51 can be adjusted via the base 53 inserted in the recess. For example, an operator of the lathe 1 places the probe of a dial gauge on the reference sphere 51 in a direction perpendicular to the tool spindle center line AX3 and operates the adjustment screws 56 so that fluctuations in the dial gauge reading when the tool spindle S2 is rotated are minimized. At this time, the lathe 1 can be placed in manual mode by manually operating each axis using the operation unit 31. When the variation in the dial gauge reading becomes zero, it can be considered that the tool spindle center line AX3 and the center 52 of the reference sphere 51 coincide with each other. In step (a1), it is not necessary to align the center 52 of the reference sphere 51 with the rotation center 21 (see FIG. 6) of the swivel unit 20. Therefore, the attachment work of the tool-side jig 50 including the reference sphere 51 can be performed quickly.

[0043] In step (a2), the position measuring device 60 is attached to the front spindle 10 so that the probe 62 of the position measuring device 60 faces the rotation center line AX1. The position measuring device 60 includes a device main body 61, a probe 62 at the tip of the device main body 61, and a workpiece-side jig 63. The workpiece-side jig 63 includes a cylinder 64 that is held by the front spindle 10 via a guide bush 11 along the rotation center line AX1, and a holder 65 that is fixed to the front end of the cylinder 64 and holds the device main body 61. The device main body 61 is held by the holder 65 so that the probe 62 faces the rotation center line AX1. In manual mode, an operator can attach the position measuring device 60 to the front spindle 10 so that the probe 62 faces the rotation center line AX1 by holding the cylinder 64 on the front spindle 10 via the guide bush 11. The position measurement device 60 in this example is an electric micrometer that measures the position of a probe 62 in contact with an object as an electric signal, but it is also possible to use a non-contact sensor or the like that measures the position of an object without contacting the object as the position measurement device. It should be noted that step (a2) may be performed after step (a1) or before step (a1).

[0044] In step (a3), the position of the reference sphere 51 is measured by the probe 62 while the rotation object is rotated with the probe 62 facing the center 52 of the reference sphere 51, thereby obtaining displacement data according to the rotation angle of the rotation object with the center 52 of the reference sphere 51 as the reference. The rotation object is the swivel unit 20 or the front spindle 10. The measurement conditions are set according to a formula derived from the mathematical model described above. The formula has a term including the geometric error of the object to be identified, and represents the displacement according to the rotation angle of the rotation object. Details of step (a3) ​​will be described later.

[0045] In step (a4), the geometric error is identified by the least squares method using the equation derived from the mathematical model and the displacement data. An example of the equation used for identifying the geometric error will be described below.

[0046] When the C axis, i.e., the front spindle 10, is rotated, the X-direction displacement of the center 52 of the reference sphere 51 in the C-axis coordinate system C Δx can be expressed by the following equation based on the above-mentioned mathematical model.

number

[0047] When the B axis, i.e., the swivel unit 20, is rotated while simultaneously controlling the three axes B, Z, and X, the X-direction displacement Δx of the center 52 of the reference sphere 51 is expressed by the following equation, based on the above-mentioned mathematical model.

number

[0048] Once again, the X-direction displacement of the center 52 of the reference sphere 51 in the C-axis coordinate system during C-axis rotation C The formula for Δx is shown below.

number

[0049] When the B axis, i.e., the swivel unit 20, is rotated while simultaneously controlling the three axes B, Z, and X, the Y-direction displacement Δy of the center 52 of the reference sphere 51 is expressed by the following equation, based on the mathematical model described above.

number

[0050] Furthermore, by deriving different mathematical formulas based on the above-described mathematical model and setting measurement conditions according to the mathematical formulas, it is possible to identify more geometric errors.

[0051] The coefficients (including constants) in the above formulas (6) to (9) can be determined by the least squares method. Here, the above formula (7) will be used as an example for explanation. First, the coefficients in the above equation (7) are replaced as follows:

number

[0052] When the above equation (10) is solved, the following is obtained:

number

[0053] For the above equation (9), the coefficients can be calculated based on the displacement data by using the above equation (11) with the sine coefficient set to α, the cosine coefficient set to β, and the constant set to γ.For the above equations (6) and (8), the coefficients can be calculated based on the displacement data by setting the sine coefficient to α, the cosine coefficient set to β, the constant set to 0, and replacing the rotational angle displacement c in the above equation (11) with the rotational angle displacement b.

[0054] By focusing on the fluctuation components of the measurement waveform and using the least squares method to calculate the sine and cosine coefficients and use them to identify geometric errors, it is possible to measure over an angular range of, for example, 90°. Geometric errors can be identified even when measuring an angular range smaller than 90°, but they will be affected by measurement errors and motion error factors other than geometric errors. Since the direction of geometric errors is defined as parallel to the directions of two orthogonal linear axes, if measurements are performed over an angular range of 90° or more, the displacement data will contain measurement values ​​in the same direction as the geometric error being identified. Therefore, it is preferable for the measurement angular range to be 90° or more.

[0055] Note that components other than the sine and cosine coefficients in the above equations (7) and (9), i.e., the constant term corresponding to γ ​​in the above equation (10), are affected by the mounting position of the position measurement device in the measurement sensitivity direction and the deviation of the zero point. In this specific example, the constant terms in the above equations (7) and (9) are not used to identify the geometric error, so the geometric error can be identified without being affected by the mounting position of the position measurement device in the measurement sensitivity direction or the deviation of the zero point. By not using the constant terms to identify the geometric error, it is no longer necessary to consider the mounting position of the position measurement device in the measurement sensitivity direction in step (a2). This means that the mounting position does not affect the results of identifying the geometric error, and it also has the advantage of shortening the time required for mounting adjustment.

[0056] Next, the step (a3) ​​of acquiring displacement data will be described in detail with reference to FIGS. 6 is a schematic diagram showing an example in which the X- and Z-direction positions of the center 52 of the reference sphere 51 change when the swivel unit 20 rotates -90°. In FIG. 6, the swivel unit 20 at a first swivel angle of the B-axis 0° is shown by a solid line, and the swivel unit 20 at a second swivel angle of the B-axis -90° is shown by a two-dot chain line. Since the tool spindle center line AX3 at the B-axis 0° is aligned along the X-direction, the orientation of the tool spindle S2 at the B-axis 0° is aligned along the X-axis. Since the tool spindle center line AX3 at the B-axis -90° is aligned along the Z-direction, the orientation of the tool spindle S2 at the B-axis -90° is aligned along the Z-axis.

[0057] As described above, center 52 of reference sphere 51 is not aligned with center of rotation 21 of swivel unit 20. Therefore, if no linear movement is performed in the X and Z axes, when swivel unit 20 rotates by -90°, the X and Z direction positions of center 52 of reference sphere 51 change. In order to simultaneously control the three axes, B, Z, and X, steps (b1) to (b4) described above are performed in step (a3).

[0058] In step (b1), the operator first places the tool spindle S2 at 0° on the B axis, centered on the pivot center line AX2 (the position of the pivot center 21 in Figure 6), and brings the probe 62 into contact with the reference sphere 51. In this state, the front spindle 10 is moved in the Z direction, and the Z-direction position Z0 of the center 52 of the reference sphere 51 on the Z axis is found based on the position of the probe 62, which changes in response to the movement of the Z axis. The Z-direction position Z0 can be obtained by moving the front spindle 10 in the Z direction and finding the Z-direction position at which the reading of the position measuring device 60 is smallest. The Z-direction position Z0 is an example of the first center position of the reference sphere 51, and can also be said to be the Z-direction position of the vertex of the reference sphere 51 at the portion where the probe 62 comes into contact with the reference sphere 51 at 0° on the B axis.

[0059] In step (b2), the operator first sets the tool spindle S2 at -90° to the B axis around the turning center line AX2 and brings the probe 62 into contact with the reference sphere 51. In this state, the front spindle 10 is moved in the Z direction, and the Z-direction position Z of the center 52 of the reference sphere 51 on the Z axis is calculated based on the position of the probe 62 which changes in accordance with the movement of the Z axis. -90 Calculate the Z-direction position Z -90 is obtained by moving the front spindle 10 in the Z direction and searching for the Z direction position at which the reading of the position measuring device 60 is smallest. -90 is an example of a second center position of the reference sphere 51, and can also be said to be the Z-direction position of the vertex of the reference sphere 51 at the portion where the tracing stylus 62 contacts the reference sphere 51 at −90° on the B axis.

[0060] In step (b3), the Z-direction position Z0 and the Z-direction position Z -90 Based on this, the distance ( B ln ) is calculated. The estimated value of the distance between the center of rotation 21 and the center 52 of the reference sphere 51 is B l n is Z -90 -Z0. Estimated value B l n corresponds to the tool shape offset radius value. In order to simultaneously control the three axes B, Z, and X, the operator must B l n The tool shape offset value corresponding to the above is set in the control unit 30. Although it is preferable that the difference between the first and second rotation angles be 90°, it is possible to calculate an estimate of the distance between the rotation center 21 and the center 52 of the reference sphere 51 even if the difference between the first and second rotation angles is other than 90°.

[0061] In step (b4), the control unit 30 calculates the distance ( B l n ) to move in an arc with a radius of 0.05 mm. This prevents any apparent relative linear motion from occurring on the reference sphere.

[0062] After the above-described measurement preparation, the displacement of the reference sphere 51 is measured over time when the rotation object rotates within a predetermined angle range as illustrated in Figures 7 to 9. First, referring to Figure 7, the X-direction displacement for applying the least squares method to the above equations (6) and (8) is calculated. C An example of measuring Δx will be described. In Fig. 7, the position measurement device 60 at C-axis 0° is shown by a solid line, and the position measurement device 60 at C-axis 90° is shown by a two-dot chain line. The B-axis is set to 0°. The front main spindle 10 is controlled to rotate from C-axis 0° to C-axis 90° as a first orientation within a predetermined angle range between C-axis 0° and C-axis 90°, and then controlled to rotate from C-axis 90° to C-axis 0° as a second orientation.

[0063] X-direction displacement CWhen measuring Δx, the operator first adjusts the Z-direction position of the probe 62 of the position measurement device 60, which is at 0° on the C-axis, to Z0 so that the probe 62 faces the center 52 of the reference sphere 51. Then, by operating the operation unit 31, the front spindle 10, on which the position measurement device 60 is mounted, is rotated in a first direction from 0° on the C-axis to 90° on the C-axis, while measuring the position of the reference sphere 51 with the probe 62. This ensures that the probe 62 remains facing the center 52 of the reference sphere 51 regardless of the rotation of the front spindle 10. Measurement data for the position of the reference sphere 51 may be collected, for example, by a data logger at predetermined time intervals. In this case, the measurement data is a collection of discontinuous measurement values ​​at each time point microscopically, but can be considered continuous data over time macroscopically. By measuring the position of the reference sphere 51 from 0° on the C-axis to 90° on the C-axis, forward displacement data ( C Δx g ) is acquired. Thereafter, the position of the reference sphere 51 is measured by the probe 62 while the front spindle 10 on which the position measurement device 60 is mounted is rotated in the second direction from C-axis 90° to C-axis 0°. By measuring the position of the reference sphere 51 from C-axis 90° to C-axis 0°, return displacement data ( C Δx r ) is obtained.

[0064] Next, referring to FIG. 8 , an example of measuring the X-direction displacement Δx for applying the least squares method to the above equation (7) will be described. FIG. 8 schematically illustrates an example of measuring the X-direction displacement Δx of the reference sphere 51 using the probe 62 by performing circular interpolation on the Z and X axes while rotating the swivel unit 20. In FIG. 8 , the swivel unit 20 and the position measurement device 60 at 0° on the B axis are indicated by solid lines, and the swivel unit 20 and the position measurement device 60 at −90° on the B axis are indicated by two-dot chain lines. The position measurement device 60 is located at 0° on the C axis, and the probe 62 is oriented along the X axis. The swivel unit 20 is controlled to rotate from 0° on the B axis to −90° on the B axis as a first direction within a predetermined angular range between 0° and −90° on the B axis, and then controlled to rotate from −90° on the B axis to 0° on the B axis as a second direction.

[0065] The probe 62 of the position measuring device 60, which is located at 0° on the C axis, faces the center 52 of the reference sphere 51. When measuring the X-direction displacement Δx, first, the swivel unit 20, on which the reference sphere 51 is mounted, is rotated in a first direction from 0° on the B axis to -90° on the B axis by operating the operation unit 31, while measuring the position of the reference sphere 51 with the probe 62. When the swivel unit 20 rotates, the control unit 30 simultaneously controls the three axes of the B, Z, and X axes. In the lathe 1 shown in FIG. 1, the Z-direction position of the position measuring device 60 is controlled together with the front spindle 10, and the X-direction position of the swivel unit 20 is controlled together with the rotation angle. As a result, the probe 62 remains facing the center 52 of the reference sphere 51 regardless of the rotation of the swivel unit 20, and no relative linear movement appears to occur on the reference sphere 51. By measuring the position of the reference sphere 51 from 0° on the B axis to -90° on the B axis, forward displacement data (Δx g ) is acquired. Thereafter, the position of the reference sphere 51 is measured by the probe 62 while rotating the swivel unit 20 on which the reference sphere 51 is mounted in the second direction from B-axis -90° to B-axis 0°. Here again, the control unit 30 simultaneously controls the three axes of B, Z, and X. By measuring the position of the reference sphere 51 from B-axis -90° to B-axis 0°, return displacement data (Δx r ) is obtained.

[0066] Furthermore, referring to FIG. 9, an example of measuring the Y-direction displacement Δy for applying the least squares method to the above equation (9) will be described. FIG. 9 schematically shows an example of measuring the Y-direction displacement of the reference sphere 51 with the probe 62 by performing circular interpolation on the Z and X axes while rotating the swivel unit 20. FIG. 9 shows the swivel unit 20 at 0° on the B axis. The position measurement device 60 is located at 90° on the C axis, and the probe 62 is oriented along the Y axis. As shown in FIG. 6, the swivel unit 20 is controlled to rotate from 0° on the B axis to −90° on the B axis as a first direction within a predetermined angle range between 0° on the B axis and −90° on the B axis, and then controlled to rotate from −90° on the B axis to 0° on the B axis as a second direction.

[0067] The probe 62 of the position measuring device 60, which is positioned at 90° on the C axis, is directed toward the center 52 of the reference sphere 51. When measuring the Y-direction displacement Δy, first, the position of the reference sphere 51 is measured by the probe 62 while the swivel unit 20, on which the reference sphere 51 is mounted, is rotated in a first direction from 0° on the B axis to -90° on the B axis by operating the operation unit 31. When the swivel unit 20 rotates, the control unit 30 simultaneously controls the three axes of the B, Z, and X axes. This keeps the probe 62 directed toward the center 52 of the reference sphere 51 regardless of the rotation of the swivel unit 20, and no apparent relative linear movement occurs on the reference sphere 51. By measuring the position of the reference sphere 51 from 0° on the B axis to -90° on the B axis, forward displacement data (Δy g ) is acquired. Thereafter, the position of the reference sphere 51 is measured by the probe 62 while rotating the swivel unit 20 on which the reference sphere 51 is mounted in the second direction from B-axis -90° to B-axis 0°. Here again, the control unit 30 simultaneously controls the three axes of B, Z, and X. By measuring the position of the reference sphere 51 from B-axis -90° to B-axis 0°, return displacement data (Δy r ) is obtained.

[0068] The predetermined angle range for rotating the rotation target is not limited to 90°, but may be 180° or the like.

[0069] Figure 10 shows a schematic example of a procedure for identifying geometric errors from displacement data in the radial direction of the C axis. In this procedure, the X-direction displacement data shown in Figure 7 is used. C The geometric error is identified from the measurement of Δx. Steps S10 and S12 shown in Fig. 10 correspond to the above-mentioned step (a3), and steps S14 and S16 shown in Fig. 10 correspond to the above-mentioned step (a4). Steps S10 and S12 can be performed by causing the lathe 1 to execute a machining program that realizes the following operations. First, the position of the reference sphere 51 is measured while rotating the front spindle 10 from C-axis 0° to C-axis 90°, thereby obtaining forward displacement data in the C-axis radial direction. C Δx gNext, the position of the reference sphere 51 is measured while rotating the front spindle 10 from 90° to 0° along the C axis, thereby obtaining the return displacement data in the radial direction of the C axis. C Δx r is acquired (step S12).

[0070] Furthermore, forward displacement data C Δx g and return displacement data C Δx r and the displacement data by averaging C Δx is acquired (step S14), thereby suppressing the influence on displacement caused by differences in the direction of motion of the front spindle 10 due to the influence of backlash and frictional force present in the drive mechanism. Finally, the above equation (6) and the displacement data C The geometric error is identified by finding a coefficient by the least squares method using Δx (step S16). As a result, the geometric error δ, which is the coefficient of sine in the above equation (6), is yB(S2) is identified.

[0071] FIG. 11 shows a schematic example of a procedure for identifying geometric errors from displacement data when the B, Z, and X axes are simultaneously controlled. In this procedure, geometric errors are identified from the measurement of the X-direction displacement Δx shown in FIG. 8. Steps S20 and S22 shown in FIG. 11 correspond to the above-mentioned step (a3), and steps S24 and S26 shown in FIG. 11 correspond to the above-mentioned step (a4). Steps S20 and S22 can be performed by having the lathe 1 execute a machining program that realizes the following operations. First, the forward displacement data Δx is measured while rotating the swivel unit 20 from 0° on the B axis to −90° on the B axis. g (Step S20). At this time, the control unit 30 simultaneously controls the three axes B, Z, and X so that the probe 62 is maintained facing the center 52 of the reference sphere 51 regardless of the rotation of the swivel unit 20. Next, the return displacement data Δx is acquired while rotating the swivel unit 20 from -90° on the B axis to 0° on the B axis. r is acquired (step S22).

[0072] Furthermore, the forward displacement data Δx g and return displacement data Δx r and the displacement data Δx is obtained by averaging them (step S24). This reduces the influence on displacement caused by differences in the direction of movement of the swivel part 20 due to the influence of backlash and frictional force present in the drive mechanism. Finally, the geometric error is identified by finding coefficients by the least squares method using the above equation (7) and the displacement data Δx (step S26). As a result, the geometric error ( b l r - b l n ) is identified. In addition, the geometric error δ contained in the cosine coefficient of the above equation (8), which is the same as the above equation (6), is xYB is the geometric error ( b l r - b l n ) is identified, and thus is identified.

[0073] As described above, by using multiple formulas, more geometric errors can be identified. Here, steps S10 and S12 in FIG. 10 and the measurement conditions shown in FIG. 7 are applied to the first measurement conditions as an example, and the acquired displacement data C Δx corresponds to the first displacement data, for example. Steps S10 and S12 in Fig. 10 can be said to be steps of acquiring the first displacement data by measuring the position of the reference sphere 51 under the first measurement conditions while rotating the rotational object. Steps S20 and S22 in Fig. 11 and the measurement conditions shown in Fig. 8 correspond to the second measurement conditions, for example, and the acquired displacement data Δx corresponds to the second displacement data, for example. Steps S20 and S22 in Fig. 11 can be said to be steps of acquiring the second displacement data by measuring the position of the reference sphere 51 under the second measurement conditions while rotating the rotational object.

[0074] Furthermore, the above equations (6) and (8) are applied to a first equation having a first coefficient representing the geometric error, and the above equation (7) is applied to a second equation having a second coefficient representing the geometric error, for example. Steps S14 and S16 in FIG. 10 can be considered to be steps of determining the first coefficient by the least squares method using the first equation and the first displacement data. Steps S24 and S26 in FIG. 11 can be considered to be steps of determining the second coefficient by the least squares method using the second equation and the second displacement data. By determining both the first coefficient and the second coefficient, more geometric errors can be identified.

[0075] The procedure for identifying the geometric error from the displacement data when the B, Z, and X axes are simultaneously controlled using the above equation (9) that represents the Y-direction displacement Δy is the same as the procedure shown in Figure 11. In the procedure using the above equation (9), the geometric error is identified from the measurement of the Y-direction displacement Δy shown in Figure 9. First, the forward displacement data Δy g (corresponding to step S20). At this time, the control unit 30 simultaneously controls the three axes B, Z, and X so that the probe 62 is maintained facing the center 52 of the reference sphere 51 regardless of the rotation of the swivel unit 20. Next, the return displacement data Δy r (corresponding to step S22). Furthermore, the forward displacement data Δy g and return displacement data Δy r The displacement data Δy is obtained by averaging these (corresponding to step S24). Finally, the geometric error is identified by finding a coefficient by the least squares method using the above formula (9) and the displacement data Δy (corresponding to step S26). Tool length b l n Since is known, the geometric error α contained in the sin coefficient of the above equation (9) YB is identified, and the geometric error γ YB is identified.

[0076] FIG. 12 shows an example of steps S20 and S22 in FIG. 11 and the forward displacement data Δx when the B, Z, and X axes are simultaneously controlled as shown in FIG. 8.g and return displacement data Δx r 12. In FIG. 12, the horizontal axis represents time and the vertical axis represents displacement. To clarify the start and end points of the forward and backward movements, an operation of moving the swivel unit 20 along the X-axis is added. Specifically, before the start of the forward movement, the X-direction position of the swivel unit 20 is set to +0.1 mm, and the swivel unit 20 is first moved -0.1 mm in the X direction and stopped for one second. Then, the B-axis is rotated from 0° to -90° at a rotational speed of 200 deg / min while performing circular interpolation motion on the Z and X-axes, stopping for one second at B = -90°, and the swivel unit 20 is moved +0.1 mm in the X direction. Then, the swivel unit 20 is moved -0.1 mm in the X direction and stopped for one second. Then, the B-axis is rotated from -90° to 0° at a rotational speed of 200 deg / min while performing circular interpolation motion on the Z and X-axes, stopping for one second at B = 0°, and the swivel unit 20 is moved +0.1 mm in the X direction. An electric micrometer MCH-331 (measurement range: ±0.5 mm) manufactured by Mitutoyo Corporation was used as the position measuring device 60. An A-8014-0141-07 Center Ball Ballbar Center Bar Assembly (sphere diameter: 12.7 mm) manufactured by Renishaw was used as the reference sphere 51.

[0077] As shown in FIG. 12, the displacement fluctuates when the B axis rotates because there is a geometric error between the axes. By identifying the geometric error using the geometric error identification method described above and correcting the parameters of the control unit 30 based on the geometric error, it is possible to reduce the fluctuation in displacement during B-axis rotation.

[0078] As described above, by arranging the reference sphere 51 on the tool spindle S2, it is possible to reduce the movement of the reference sphere 51 compared to when the reference sphere 51 is installed at a position offset from the rotation center line AX1. This reduces the space required for measurement, making it possible to perform measurements even when the position measurement device 60 cannot be attached to the tool spindle S2. Furthermore, by attaching the position measurement device 60 to the front spindle 10, it is also easy to route wiring and the like. Then, the geometric error is accurately identified by the least squares method using the acquired displacement data and a formula derived from a mathematical model. Therefore, it is possible to easily identify inter-axis geometric errors in machine tools with a narrow range of motion, such as a moving-spindle lathe. Furthermore, by deriving a formula different from the above formulas (6) to (9) based on a mathematical model and acquiring displacement data under measurement conditions according to that formula, it is possible to identify more geometric errors.

[0079] (5) Variation: In the above-described embodiment, the second-lowest tool spindle S2 of the multiple tool spindles in the swivel unit 20 was the subject of explanation, but geometric errors can also be identified in a similar manner for the remaining tool spindles. In the above-described embodiment, the front spindle 10 is described as the rotating part, but the rear spindle 12 may also be the rotating part. The machine tool to which the above-described geometric error identification method can be applied is not limited to a moving spindle type lathe, but may also be a fixed spindle type lathe in which the front spindle does not move in the spindle center line direction.

[0080] (6) Conclusion: As described above, the present invention provides, through various aspects, a geometric error identification method that can easily identify inter-axis geometric errors in machine tools with narrow movement ranges. Of course, even in an aspect that consists only of the elements set forth in the independent claims, the basic functions and effects described above can be obtained. Furthermore, it is possible to implement configurations in which the components disclosed in the above examples are substituted with each other or the combination is changed, or configurations in which the components disclosed in the publicly known techniques and the above examples are substituted with each other or the combination is changed, etc. The present invention also includes these configurations. [Explanation of symbols]

[0081] 1...Lathe (example of machine tool), 10...Front spindle (example of rotating part), 11...Guide bush, 15...Gang tool post, 20...swivel portion, 21...swivel center, 30...control unit, 31...operation unit, 50...Tool side jig, 51...Reference sphere, 52...Center, 60... position measuring device, 61... device body, 62... measuring element, 63... workpiece side jig, AX1...rotation center line, AX2...swivel center line, AX3...tool spindle center line, S1, S2…Tool spindle, TO1...tool, W1...workpiece.

Claims

1. A geometric error identification method for identifying a geometric error between axes from the workpiece to the tool spindle in a machine tool including: a rotating unit that can rotate together with a workpiece about a rotation center line; a swivel unit that has a tool spindle and can swivel about a swivel center line that is oriented differently from the tool spindle; and a control unit that controls movement of the rotation axis about the rotation center line, movement of the swivel axis about the swivel center line, and movement of a plurality of linear axes, the method comprising: (a1) arranging a reference sphere so that its center is aligned with the tool spindle; (a2) attaching the position measurement device to the rotating part so that a probe of the position measurement device faces the rotation center line; (a3) measuring the position of the reference sphere with the probe while rotating one of the rotational objects of the swivel unit and the rotating unit with the probe facing the center of the reference sphere, thereby acquiring displacement data according to the rotation angle of the rotational object with the center of the reference sphere as a reference; (a4) identifying the geometric error by a least squares method using a mathematical expression that has a term including the geometric error and that represents a displacement according to the rotation angle, and the displacement data; A method for identifying geometric errors in a machine tool, comprising:

2. 2. The method for identifying a geometric error of a machine tool according to claim 1, wherein in said step (a3), at least some of said plurality of translatory axes are controlled so that a state in which said stylus is directed toward the center of said reference sphere is maintained regardless of rotation of said rotation object.

3. the plurality of linear axes include a first linear axis along the rotation center line and a second linear axis oriented in a direction different from both the rotation center line and the turning center line, The step (a3) (b1) a step of bringing the measuring piece into contact with the reference sphere while rotating the tool spindle at a first rotation angle around the rotation center line, and determining a first center position of the reference sphere on the first translational axis based on a position of the measuring piece that changes in accordance with the movement of the first translational axis; (b2) rotating the tool spindle at a second rotation angle different from the first rotation angle about the rotation center line, bringing the probe into contact with the reference sphere, and determining a second center position of the reference sphere on the first translational axis based on a position of the probe that changes in accordance with the movement of the first translational axis; (b3) calculating a distance between the center of rotation of the swivel unit and the center of the reference sphere based on the first center position and the second center position; (b4) when the rotation object is the swivel portion, causing the first rectilinear axis and the second rectilinear axis to perform an arc motion with the distance as a radius so that the probe is maintained facing the center of the reference sphere; 3. The method for identifying a geometric error of a machine tool according to claim 2, comprising:

4. the first linear axis is the Z axis, the second translational axis is the X-axis, the orientation of the tool spindle at the first pivot angle is along the X-axis; 4. The method for identifying a geometric error of a machine tool according to claim 3, wherein the orientation of the tool spindle at the second rotation angle is along the Z axis.

5. 5. The method for identifying a geometric error of a machine tool according to claim 1, wherein in the step (a3), the displacement data is acquired by measuring the position of the reference sphere with the measuring element while rotating the rotational object within a predetermined angle range of 180° or less.

6. In the step (a3), the rotational object is rotated in a first direction within the predetermined angle range, while measuring the position of the reference sphere with the probe, thereby obtaining outward displacement data as the displacement data; and the rotational object is rotated in a second direction opposite to the first direction within the predetermined angle range, while measuring the position of the reference sphere with the probe, thereby obtaining backward displacement data as the displacement data; 6. A method for identifying a geometric error of a machine tool according to claim 5, wherein in said step (a4), said forward displacement data and said backward displacement data are averaged and said geometric error is identified by said least squares method.

7. In the step (a3), first displacement data is acquired as the displacement data by measuring the position of the reference sphere under first measurement conditions while rotating the rotational object, and second displacement data is acquired as the displacement data by measuring the position of the reference sphere under second measurement conditions different from the first measurement conditions while rotating the rotational object, 5. The method for identifying a geometric error of a machine tool according to claim 1, wherein in step (a4), the geometric error is identified by determining a first coefficient by a least squares method using a first mathematical expression having a first coefficient representing the geometric error and the first displacement data, and determining a second coefficient by a least squares method using a second mathematical expression having a second coefficient representing the geometric error and the second displacement data.

8. the plurality of linear axes include an X-axis, a Y-axis, and a Z-axis; The Y axis is along the turning centerline, the Z axis is along the rotation center line, the rotating part is a main shaft capable of gripping the workpiece and moving along the rotation center line, 4. The method for identifying a geometric error of a machine tool according to claim 1, wherein the swivel unit is a B-axis unit to which a tool for machining the workpiece is attached along the tool spindle.

Images