Patent Grant
10946491
The present invention relates to a method of identifying a relative motion error between a spindle for holding a tool and a table for mounting a workpiece thereon in a machine tool configured to move the spindle and the table relative to each other in directions of three orthogonal axes, an X-axis, a Y-axis, and a Z-axis.
Conventionally, positioning errors in X-axis, Y-axis, and Z-axis feed axes (i.e., an X-axis feed mechanism, a Y-axis feed mechanism, and a Z-axis feed mechanism) and straightness in the feed axes are taken into account as factors contributing to a motion error in a machine tool. In order to compensate for such a motion error, a numerical controller as disclosed in Japanese Unexamined Patent Application Publication No. H8-152909 (Patent Literature 1 listed below) has been proposed.
This numerical controller, as disclosed in Patent Literature 1, includes grid-point-compensation-vector storing means that stores therein grid-point compensatio vectors which are measured in advance at grid points of a grid area defined by dividing a coordinate system at certain intervals in each coordinate-axis direction, interpolating means that outputs an interpolation pulse for each feed axis in accordance with a movement command, current-position recognizing means that recognizes a current position in each feed axis by adding the interpolation pulse, current-position-compensation-vector calculating means that calculates a current-position compensation vector at the current position based on the grid-point compensation vectors, compensation-pulse outputting means that compares the current-position compensation vector with a start-point compensation vector at the previous current position before interpolation and outputs an amount of change as a compensation pulse, and adding means that adds the compensation pulse to the interpolation pulse.
With this numerical controller, each time an interpolation pulse is output, a three-dimensional compensation vector at a current position is calculated and the calculated three-dimensional compensation vector is added as a compensation pulse to the interpolation pulse. Therefore, a positional error that is caused by a mechanical system in a three-dimensional space can be compensated for by a single interpolation-type error compensation function.
Note that the grid-point compensation vector at each grid point of the grid area is obtained by measuring a positioning error of a reference point in a three-dimensional space occurring in controlling positioning in the feed axes with a certain interval, the reference point being set as appropriate on the axis of the spindle. Further, the measurement is typically carried out with a laser interferometer, a laser length measuring device, an auto-collimator, or the like. Furthermore, the reference point is typically set at, for example, a position at which the axis of the spindle intersects with a front end surface of the spindle or a position which is located forward away from the front end surface of the spindle by a predetermined distance on the axis of the spindle; the reference point is determined as appropriate depending on the measurement method.
Recently, a motion error (positioning error) in a three-dimensional space in a machine tool are considered to occur with errors in translational motions in the feed axes, angular errors in the feed axes, and errors concerning perpendicularities between the feed axes affecting one another, as shown in
As for a method for measuring these errors, a measurement method using a measurement apparatus as shown in
The above-mentioned errors are measured with a laser length measuring device 101 disposed on the bed 51 as well as a mirror 102 attached to the spindle 54. Specifically, first, the laser length measuring device 101 is arranged at a predetermined position, for example, the position indicated by the solid line in
Subsequently, the laser length measuring device 101 is sequentially arranged at three other different positions (for example, the positions indicated by the broken lines in
Based on the measurement data obtained in the above-described manner, the position of the mirror 102 at each grid point of the three-dimensional space is calculated in accordance with the principle of triangulation. Further, based on the calculated position data and analysis of the position data, the above-mentioned errors are obtained.
However, the measurement method using the laser length measuring device 101 has a problem that the laser length measuring device 101 per se is expensive, and further has a problem that the measurement requires long time and is complicated and troublesome in operation because it is necessary to sequentially arrange the laser length measurement device 101 at four positions, and at each of the four positions, sequentially position the mirror 102 at each grid point of the three-dimensional space to carry out the measurement.
At the same time, the translational motion errors in the feed axes and the angular errors in the feed axes can be measured in accordance with an already established measurement method, as provided in JIS B 6190-2, JIS B 6336-1, and JIS B 6336-2. Further, as for the errors A0Z, B0Z, C0Y, etc. regarding the perpendicularities between the X-axis, the Y-axis, and the Z-axis, a measurement method using a double ball bar as disclosed in Non-Patent Literature 1 listed below has been proposed.
Accordingly, the above-mentioned errors can be measured by means of these methods without depending on the above-described measurement method using the laser length measuring device 101 and the mirror 102.
Patent Literature 1: Japanese Unexamined Patent Application Publication No. H8-152909
Non-Patent Literature 1: Yoshiaki Kakino, Yukitoshi Ihara, and Yoshio Nakatsu: “A Study on the Motion Accuracy of NC Machine Tools (2nd Report)”, Journal of the Japan Society for Precision Engineering, 52/10/1986 pp. 73-79
The above-described motion error of the spindle (specifically, the reference point) in a three-dimensional space has to be compensated for eventually. Therefore, typically, for control reasons, a motion error in a three-dimensional space of a machine coordinate system that is defined with respect to the so-called machine zero has to be identified.
However, where the perpendicularity errors between the X-axis, the Y-axis, and the Z-axis are measured with a double ball bar as mentioned above, there is a problem that it is not possible to measure the errors with respect to the machine zero. That is to say, in order to measure the perpendicularity errors with respect to the machine zero with a double ball bar, it is necessary to turn the spindle which has the double ball bar attached thereto about the machine zero with the length of the bar as a turning radius. However, the feed axes do not allow movement in the negative direction beyond the machine zero; therefore, such a turning operation cannot be realized.
The above-mentioned errors EXX, EYY, EZZ, EYX, EZX, EXY, EZY, EXZ, EYZ, EAX, EAY, EAZ, EBX, EBY, EBZ, ECX, ECY, and ECZ are theoretically considered to be affected by the perpendicularity errors A0Z, B0Z, C0Y, etc. Therefore, it is conceivable that these errors also cannot be identified with respect to the machine zero.
Thus, when the measurement methods provided in the JIS and the method disclosed in Non-Patent Literature 1 are used, the motion error in the three-dimensional space of the machine coordinate system cannot be identified immediately based on measurement values of the methods. However, if it is possible to identify the motion errors in the three-dimensional space of the machine coordinate system based on measurement values measured by these methods, there is an advantage in cost because the laser length measuring device 101 as shown in
Further, if it is possible to identify the motion error in a three-dimensional space of a coordinate system having its origin at an arbitrary reference position based on measurement values measured by the measurement methods provided in the JIS and the method disclosed in Non-Patent Literature 1, the degree of freedom of usage of the data is increased, which is convenient.
The present invention has been achieved in view of the above-described circumstances, and an object thereof is to provide a method of identifying a motion error of a machine tool in a coordinate system having its origin at an arbitrary position in the machine tool based on error data measured by a conventional commonly-used measurement method.
The present invention, for solving the above-described problems, relates to a method of identifying a relative motion error between a spindle holding a tool and a table for mounting a workpiece thereon in a three-dimensional space in a machine tool,
the machine tool including the spindle and the table and including a Z-axis feed mechanism, an X-axis feed mechanism, and a Z-axis feed mechanism respectively corresponding to a Z-axis, an X-axis, and a Y-axis as reference axes, the Z-axis extending along an axis of the spindle, the X-axis and the Y-axis being orthogonal to the Z-axis and orthogonal to each other,
the machine tool being configured such that the spindle and the table are moved relative to each other in the three-dimensional space by the X-axis feed mechanism, the Y-axis feed mechanism, and the Z-axis feed mechanism,
the method including:
operating the X-axis feed mechanism, the Y-axis feed mechanism, and the Z-axis feed mechanism in a three-dimensional space of a machine coordinate system defined with respect to a machine zero X0, Y0, Z0 respectively set for the X-axis feed mechanism, the Y-axis feed mechanism, and the Z-axis feed mechanism, and measuring the following errors with respect to an arbitrary coordinate position in the machine coordinate system:
deriving, based on the measured actual error data, the following errors in a three-dimensional space of a set coordinate system having its origin at a reference position Xa, Ya, Za preset in the machine coordinate system:
deriving, based on the derived error data, the relative motion error between the spindle and the table in the three-dimensional space of the set coordinate system.
In the present invention, the X-axis feed mechanism, the Y-axis feed mechanism, and the Z-axis feed mechanism are operated in a three-dimensional space of a machine coordinate system defined with respect to a machine zero X0, Y0, Z0 respectively set for the X-axis feed mechanism, the Y-axis feed mechanism, and the Z-axis feed mechanism, and a positioning error in the X-axis direction, a positioning error in the Y-axis direction, a positioning error in the Z-axis direction, straightness errors in the X-axis, the Y-axis, and the Z-axis, angular errors around the X-axis, the Y-axis, and the Z-axis in the X-axis, angular errors around the X-axis, the Y-axis, and the Z-axis in the Y-axis, angular errors around the X-axis, the Y-axis, and the Z-axis in the Z-axis, and perpendicularity errors between the X-axis, the Y-axis, and the Z-axis are measured with respect to an arbitrary coordinate position in the machine coordinate system.
The positioning error in the X-axis direction, the positioning error in the Y-axis direction, the positioning error in the Z-axis direction, the straightness errors in the X-axis, the Y-axis, and the Z-axis, the angular errors around the X-axis, the Y-axis, and the Z-axis in the X-axis, the angular errors around the X-axis, the Y-axis, and the Z-axis in the Y-axis, and the angular errors around the X-axis, the Y-axis, and the Z-axis in the Z-axis can be measured, for example, in conformity with the regulations of JIS B 6190-2, JIS B 6336-1, and JIS B 6336-2. The perpendicularity errors between the X-axis, the Y-axis, and the Z-axis can be measured, for example, by means of the double ball bar method disclosed in Non-Patent Literature 1 listed above.
Based on the measured actual error data, a positioning error in the X-axis direction in the X-axis feed mechanism, a positioning error in the Y-axis direction in the Y-axis feed mechanism, a positioning error in the Z-axis direction in the Z-axis feed mechanism, straightness errors in the X-axis feed mechanism, the Y-axis feed mechanism, and the Z-axis feed mechanism, angular errors around the X-axis, the Y-axis, and the Z-axis in the X-axis feed mechanism, angular errors around the X-axis, the Y-axis, and the Z-axis in the Y-axis feed mechanism, angular errors around the X-axis, the Y-axis, and the Z-axis in the Z-axis feed mechanism, and perpendicularity errors between the X-axis feed mechanism, the Y-axis feed mechanism, and the Z-axis feed mechanism in a three-dimensional space of a set coordinate system having its origin at a reference position Xa, Ya, Za preset in the machine coordinate system are derived.
Subsequently, based on the derived error data, a relative motion error between the spindle and the table, i.e., a positioning error of the spindle with respect to the table, in the three-dimensional space of the set coordinate system is derived.
As described above, in the present invention, error data relating to the X-axis feed mechanism, the Y-axis feed mechanism, and the Z-axis feed mechanism in the three-dimensional space of the set coordinate system having its origin at the reference position Xa, Ya, Za preset in the machine coordinate system is derived based on measured actual error data that is measured by an existing commonly-used measurement method, and a motion error of the machine tool in the three-dimensional space of the set coordinate system is derived based on the derived error data. Note that the reference position Xa, Ya, Za can be set at any positon; for example, it may be set at the machine zero X0, Y0, Z0.
Accordingly, with the present invention, a motion error of a machine tool in a three-dimensional space of a machine coordinate system can be identified based on actual error data that is measured by an existing commonly-used measurement method which does not use an expensive laser length measuring device as mentioned above and which can be carried out more simply in operation than a measurement using such a laser length measuring device. Accordingly, identification of the motion error can be carried out inexpensively and simply in operation.
Further, identifying the motion error in a set coordinate system, the reference position Xa, Ya, Za of which is set at an arbitrary position, can increase the degree of freedom of usage of the error data.
Note that the error data derived may relate to a spindle center position on a front end of the spindle. With this configuration, a variable element, such as protrusion of a measuring device, in the measurement of the actual errors can be canceled.
Further, the relative motion error between the spindle and the table in the three-dimensional space of the set coordinate system may relate to a tip of a tool to be attached to the spindle. Using a motion error amount obtained with this configuration enables appropriate motion error compensation which is consistent with actual machining using the tool.
As described above, with the present invention, a motion error of a machine tool in a three-dimensional space of a machine coordinate system can be identified based on measured actual error data that is measured by an existing commonly-used measurement method which does not use an expensive laser length measuring device as mentioned above and which can be carried out more simply in operation than a measurement using such a laser length measuring device. Accordingly, identification of the motion error can be carried out inexpensively and simply in operation.
Hereinafter, a specific embodiment of the present invention will be described with reference to the drawings.
In this embodiment, a method for identifying a motion error of a horizontal machining center 1 as shown in
The column 3 is moved in the X-axis direction by an X-axis feed mechanism (not shown), the spindle head 4 is moved in the Y-axis direction by a Y-axis feed mechanism (not shown), and the table 6 is moved in the Z-axis direction by a Z-axis feed mechanism (not shown). Thus, the X-axis feed mechanism, the Y-axis feed mechanism, and the Z-axis feed mechanism cause the spindle 5 and the table 6 to move relative to each other in a three-dimensional space formed by the three orthogonal axes, i.e., the X-axis, the Y-axis, and the Z-axis.
1. Motion Error Calculation Equations
As for the thus-configured machining center 1, it is known that a motion error (positioning error) of a front-end center position (reference point) of the spindle 5 in a three-dimensional space of a machine coordinate system can be calculated by the equations below. Note that α, β, and γ are respectively command values for the X-coordinate, the Y-coordinate, and the Z-coordinate, EX (α, β, γ) is a positioning error in the X-axis direction, EY (α, β, γ) is a positioning error in the Y-axis direction, and EZ (α, β, γ) is a positioning error in the Z-axis direction.
EX(α,βγ)=EXX(α)+EXY(β)+EXZ(γ)−(ECX(α)+ECZ(γ)+C0Y)×β (Equation 1)
EY(α,β,γ)=EYX(α)+EYY(β)+EYZ(γ)+ECZ(γ)×α (Equation 2)
EZ(α,β,γ)=EZX(α)+EZY(β)+EZZ(γ)+(EAX(α)+EAZ(γ)+A0Y)×β−(EBZ(γ)+B0X)×α (Equation 3)
Note that other conceivable error parameters are as follows:
Further, a positioning error of a tool attached to the spindle 5 can be calculated by the equations below. Note that TX, TY, TZ are a deviation in the X-axis direction, a deviation in the Y-axis direction, and a deviation in the Z-axis direction of a tip of the tool with respect to the front-end center position (reference point) of the spindle 5, respectively.
EX(α,β,γ, TX, TY, TZ)=EXX(α)+EXY(β)+EXZ(γ)−(ECX(α)+ECZ(γ)+C0Y)×β+(EBX(α)+EBY(β)+EBZ(γ))×TZ−(ECX(α)+ECY(β)+ECZ(γ))×TY (Equation 4)
EY(α,β, γ, TX, TY, TZ)=EYX(α)+EYY(β)+EYZ(γ)+ECZ(γ)×α+(ECX(α)+ECY(β)+ECZ(γ))×TX−(EAX(α)+EAY((β)+EAZ(γ))×TZ (Equation 5)
EZ(α,β, γ, TX, TY, TZ)=EZX(α)+EZY(β)+EZZ(γ)+(EAX(α)+EAZ(γ)+A0Y)×β−(EBZ(γ)+B0X)×α+(EAX(α)+EAY(β)+EAZ(γ))×TY−(EBX(α)+EBY(β)+EBZ(γ))×TX (Equation 6)
Further, the equations below can calculate the positioning errors in a set coordinate system having its origin at a position Xa, Ya, Za that is an arbitrary position in the machine coordinate system.
EX(α,β,γ)=EXX(α)+EXY(β)+EXZ(γ)−(ECX(α)+ECZ(γ)+C0Y)×(β−Ya) (Equation 7)
EY(α,β,γ)=EYX(α)+EYY(β)+EYZ(γ)+(ECZ(γ)×(α−Xa) (Equation 8)
EZ(α,β,γ)=EZX(α)+EZY(β)+EZZ(γ)+(EAX(α)+EAZ(γ)+A0Y)×(β−Ya)−(EBZ(γ)+B0X)×(α−Xa) (Equation 9)
EX(α,β, γ, TX, TY, TZ)=EXX(α)+EXY(β)+EXZ(γ)−(ECX(α)+ECZ(γ)+C0Y)×(β−Ya)+
(EBX(α)+EBY(β)+EBZ(γ))×TZ−(ECX(α)+ECY(β)+ECZ(γ))×TY (Equation 10)
EY(α,β, γ, TX, TY, TZ)=EYX(α)+EYY(β)+EYZ(γ)+ECZ(γ)×(α−Xa)+
(ECX(α)+ECY(β)+ECZ(γ))×TX−(EAX(α)+EAY(β)+EAZ(γ))×TZ (Equation 11)
EZ(α,β,γ, TX, TY, TZ)=EZX(α)+EZY(β)+EZZ(γ)+(EAX(α)+EAZ(γ)+A0Y)×(β−Ya)−(EBZ(γ)+B0X)×(α−Xa)+(EAX(α)+EAY(β)+EAZ(γ))×TY−(EBX(α)+EBY(β)+EBZ(γ))×TX (Equation 12)
2. Motion Error Measurement
In this embodiment, firstly, errors are measured for the items below in conformity with JIS B 6190-2 and JIS B 6336-1. Note that, in the description below, X, Y, and Z representing a position represent the position of the front-end center (reference point) of the spindle 5 in the machine coordinate system, X, Y, and Z representing the position of the reference point with respect to a machine zero in the X-axis feed mechanism, the Y-axis feed mechanism, and the Z-axis feed mechanism, respectively.
[X-Axis]
The X-axis feed mechanism (not shown) is actuated and the reference point is sequentially moved to command positions X1, X2, . . . and Xn at predetermined pitch intervals, during which errors MI1 (Xk) through MI6(Xk) for the measurement items 11 through I6 below are measured. Note that k is an integer from 1 to n. Note further that command positions in the Y-axis feed mechanism (not shown) and Z-axis feed mechanism (not shown) during measurement of each measurement item are respectively arbitrary positions YIm and ZIm, m corresponding to the suffix of the measurement item.
[Y-Axis]
The Y-axis feed mechanism (not shown) is actuated and the reference point is subsequently moved to command positions Y1, Y2, . . . and Yn at predetermined pitch intervals, during which errors MI7(Yk) through MI12(Yk) for the measurement items I7 through I12 below are measured. Note that k is an integer from 1 to n. Note further that command positions in the X-axis feed mechanism (not shown) and the Z-axis feed mechanism (not shown) during measurement of each measurement item are respectively arbitrary positions XIm and ZIm, m corresponding to the suffix of the measurement item.
[Z-Axis]
The Z-axis feed mechanism (not shown) is actuated and the reference point is sequentially moved to command positions Z1, Z2, . . . and Zn at predetermined pitch intervals, during which errors MI13(Zk) through MI18(Zk) for the measurement items 113 through I18 below are measured. Note that k is an integer from 1 to n. Note further that command positions in the X-axis feed mechanism (not shown) and the Y-axis feed mechanism (not shown) during measurement of each measurement item are respectively arbitrary positions XIm and YIm, m corresponding to the suffix of the measurement item.
[Perpendicularity]
Measurement is performed with a double ball bar In accordance with Non-Patent Literature 1, wherein the center position of the table-side ball is set at an arbitrary position Xi, Yi, Zi , the reference point of the spindle 5 is circularly moved with the length of the bar as a rotation radius in an X-Y plane, in an X-Z plane, and in a Y-Z plane, and lengths MAij (in the Y-Z plane), MBij (in the X-Z plane), and MCij (in the X-Y plane) of the bar are measured based on the amount of expansion/contraction of the bar. MAij is a length of the bar at a position YAij, ZAij in the circular movement of the reference point of the spindle 5 in the Y-Z plane that is defined with Xi fixed, MBij is a length of the bar at a position XBij, ZBij in the circular movement of the reference point of the spindle 5 in the X-Z plane that is defined with Yi fixed, and MCij is a length of the bar at a position XCij, YCij in the circular movement of the reference point of the spindle 5 in the X-Y plane that is defined with Zi fixed. Note that i is an integer from 1 to g and means the number of times of perpendicularity measurement, and j is an integer from 1 to h and means a sampling number of the position of the spindle 5.
Based on the obtained measurement values MAij, MBij and MCij, perpendicularities PAi, PBi, and PCi and perpendicularity errors A0Yi, B0Xi, and C0Yi for the X-axis feed mechanism, the Y-axis feed mechanism, and the Z-axis feed mechanism when the center position of the table-side ball is positioned at theposition Xi, Yi, Zi are calculated in accordance with Non-Patent Literature 1.
Note that the perpendicularities PAi, PBi, and PCi are respectively represented as functions of the measurement values MAij, MBij, and Mg as follows:
3. Identification of Error Parameters the X-Axis Feed Mechanism, Y-Zxis Feed Mechanism, and Z-Axis Feed Mechanism
Subsequently, the above-mentioned error parameters EXX, EYY, EZZ, EYX, EZX, EXY, EZY, EXZ, EYZ, EAX, EAY, EAZ, EBX, EBY, EBZ, ECX, ECY, and ECZ in the X-axis feed mechanism, the Y-axis feed mechanism, and the Z-axis feed mechanism are identified based on the error data MI1(Xk) through MI6(Xk), MI7(Yk) through MI12(Yk), and MI13(Zk) through MI18(Zk) measured in the above-described manner.
By way of example, the X-axis straightness error MI3(Xk) (in the Z-axis direction) is examined. As shown in
MI3(Xk)=EZ(Xk, YI3,ZI3,LI3X,LI3Y,LI3Z)+ConstI3.
When EZ(Xk, YI3,ZI3,LX,LY,LZ) in the above equation is expanded as an error in the set coordinate system having its origin at the position Xa, Ya, Za that is an arbitrary position in the machine coordinate system, the equation below is obtained according to Equation 12.
MI3(Xk)=EZX(Xk)+EZY(YI3)+EZZ(ZI3)+(EAX(Xk)+EAZ(ZI3)+A0Y)×(YI3−Ya)−(EBZ(ZI3)+B0X)×(Xk−Xa)+(EAX(Xk)+EAY(YI3)+EAZ(ZI3))×LI3Y−(EBX(Xk)+EBY(YI3)+EBZ(ZI3))×LI3×+ConstI3
Further, the X-axis angular error MI6(Xk) around the Z-axis is examined. Since the angular errors involve no other error factors, the angular error MI6(Xk) can be represented by the following equation:
MI6(Xk)=ECX(Xk)+ConstI6.
On the basis of the above examination, the above-mentioned errors can be represented by equations which do not use the perpendicularity (B0X) in the X-axis feed mechanism, the perpendicularity (A0Y) in the Y-axis feed mechanism, and the perpendicularity (C0Y) in the Z-axis feed mechanism as follows:
MI1(Xk)=EXX(Xk)−ECX(Xk)×(YI1−Ya)+EBX(Xk)×LI1Z−ECX(Xk)×LI1Y+ConstI1;
MI2(Xk)=EYX(Xk)+ECX(Xk)×LI2X−EAX(Xk)>LI2Z+ConstI2;
MI3(Xk)=EZX(Xk)+EAX(Xk)×(YI3−Ya)+EAX(Xk)×LI3Y−EBX(Xk)×LI3X+ConstI3;
MI4(Xk)=EAX(Xk)+ConstI4;
MI5(Xk)=EBX(Xk)+ConstI5;
MI6(Xk)=ECX(Xk)+ConstI6;
MI7(Yk)=EYY(Yk)+ECY(Yk)×LI7X−EAY(Yk)×LI7Z+ConstI7;
MI8(Yk)=EXY(Yk)+EBY(Yk)×LI8X−EAY(Yk)×LI8Z+ConstI8;
MI9(Yk)=EZY(Yk)+EAY(Yk)×LI9X−EAY(Yk)×LI9Z+ConstI9;
MI10(Yk)=EAY(Yk)+ConstI10;
MI11(Yk)=EBY(Yk)+ConstI11;
MI12(Yk)=ECY(Yk)+ConstI12;
MI13(Zk)=EZZ(Zk)+EAZ(Zk)×(YI13−Ya)−EBZ(Zk)×(XI13−Xa)I+EAZ(Zk)×LI13Y−EBZ(Zk)×LI3X+ConstI3;
MI14(Zk)=EXZ(Zk)−ECZ(Zk)×(YI14−Ya)+EBZ(Zk)×LI14Z−ECZ(Zk)×LI14Y+ConstI2,
MI15(Zk)=EYZ(Zk)+ECZ(Zk)×(XI13−Xa)+ECZ(Zk)×LI15X−EAZ(Zk)×LI15Z+ConstI15,
MI16(Zk)=EAZ(Zk)+ConstI16;
MI17(Zk)=EBY(Zk)+ConstI17; and
MI18(Zk)=ECY(Zk)+ConstI18.
Based on these equations, the error parameters are as follows:
EXX(Xk)=MI1(Xk)+ECX(Xk)×(YI1−Ya)−EBX(Xk)×LI1Z+ECX(Xk)×LI1Y−ConstI1;
EYX(Xk)=MI2(Xk)−ECX(Xk)×LI2X+EAX(Xk)×LI2Z−ConstI2;
EZX(Xk)=MI3(Xk)−EAX(Xk)×(YI3−Ya)−EAX(Xk)×LI3Y+EBX(Xk)×LI3X−ConstI3;
EAX(Xk)=MI4(Xk)−ConstI4;
EBX(Xk)=MI5(Xk)−ConstI5;
ECX(Xk)=MI6(Xk)−ConstI6;
EYY(Yk)=MI7(Yk)−ECY(Yk)×LI7X+EAY(Yk)×LI7Z+ConstI7;
EXY(Yk)=MI8(Yk)−EBY(Yk)×LI8X+ECY(Yk)×LI8Y+ConstI8;
EZY(Yk)=MI9(Yk)−EAY(Yk)×LI9Y+EBY(Yk)×LI9X+ConstI9;
EAY(Yk)=MI10(Yk)−ConstI10;
EBY(Yk)=MI11(Yk)−ConstI11;
ECY(Yk)=MI12(Yk)−ConstI12;
EZZ(Zk)=MI13(Zk)−EAZ(Zk)×(YI13−Ya)+EBZ(Zk)×(XI13−Xa)I−EAZ(Zk)×LI13Y+EBZ(Zk)×LI13X−ConstI13;
EXZ(Zk)=MI14(Zk)+ECZ(Zk)×(YI14−Ya)−EBZ(Zk)×LI14Z+ECZ(Zk)×LI14Y−ConstI2;
EYZ(Zk)=MI15(Zk)−ECZ(Zk)×(XI15−Xa)−ECZ(Zk)×LI15X+EAZ(Zk)×LI15Z−ConstI15;
EAZ(Zk)=MI16(Zk)−ConstI16;
EBY(Zk)=MI17(Zk)−ConstI17; and
ECY(Zk)=MI18(Zk)−ConstI18.
Thus, the error parameters in the set coordinate system having its origin at the position Xa, Ya, Za that is an arbitrary position in the machine coordinate system can be identified by the above equations. Note that the constant terms ConstI1 through ConstI18 can be each considered as the degree of freedom for changing setting of the zero point for the respective error.
4. Identification of Perpendicularity Error Parameters
Subsequently, based on the perpendicularity measurement values MAij, MBij, and MCij measured in the above-described manner, the perpendicularities PAi, PBi, and PCi which are calculated below, and the perpendicularity errors A0Yi, B0Xi, and C0Yi, the perpendicularity errors A0Y, B0X, and C0Y in the set coordinate system having its origin at the position Xa, Ya, Za that is an arbitrary position in the machine coordinate system are identified.
Prior to identification of the perpendicularity errors A0Y, B0X, and C0Y, the calculation basis for the identification is described. As described above, the perpendicularities PAi, PBi, and PCi are respectively represented as functions of the measurement values MAij, MBij, and MCij as follows:
fA(RAij)=PAi; (Equation 13)
fB(RBij)=PBi; and (Equation 14)
fC(RCij)=PCi, (Equation 15)
where MAi is total data of the measurement value MAij, MBij is total data of the measurement value MBij, and MCi is total data of the measurement value MCij.
On the other hand, when the spindle 5 is circularly moved with a double ball bar, the positioning errors of the spindle 5 with respect to the command values in the set coordinate system having its origin at the position Xa, Ya, Za that is an arbitrary position in the machine coordinate system can be calculated by Equations 10 to 12 above. Therefore, where the position of the table-side sphere in the machine coordinate system is represented by Xi, Yi, Zi and the position of the reference point of the spindle 5 circularly moved in the X-Y plane with respect to the position Xi, Yi, Zi is represented by Xik, Yik, Zi , the length SCik of the bar can be calculated by the following equation:
SCik=((Xik+EXik−Xi)2+(Yik+EYik−Yi)2+(Zi+EZik−Zi)2)1/2. (Equation 16)
In Equation 16 above, EXik, EYik, and EZik are infinitesimal values; therefore, where the squared terms thereof are approximated by zero, Scik can be represented by the following equation:
SCik=((Xik−Xi)2+(Yik−Yi)2+2EXik(Xik−Xi)+2EYik(Yik−Yi))1/2. (Equation 17)
Further, where total data of the length SCik of the bar calculated is represented by SCi, a perpendicularity P′Ci calculated from SCi can be represented by the following relational equation:
f(SCi)=P′Ci. (Equation 18)
Accordingly, by using Equation 19 above, the true perpendicularity error C0Y in the set coordinate system having its origin at the position Xa, Ya, Za that is an arbitrary position in the machine coordinate system can be identified based on the temporary perpendicularity error C0Y′, the perpendicularity PCi calculated by Equation 15, and the perpendicularity P′Ci calculated by Equation 18.
Similarly, as for the perpendicularity error B0X, where the reference point of the spindle 5 circularly moved in the X-Z plane is represented by Xik, Yi, Zik, the length SBik of the bar is as follows:
SBik=((Xik+EXik−Xi)2+(Yi+EYik−Yi)2+(Zik+EZik−Zi)2)1/2.
Accordingly, the true perpendicularity error B0X in the coordinate system having it origin at the position Xa, Ya, Za that is an arbitrary position in the machine coordinate system can be identified by Equation 22 below based on the temporary perpendicularity error B0X′, the perpendicularity PBi calculated by Equation 14 above, and the perpendicularity P′Bi calculated by Equation 21 above.
B0X=B0X′−P′Bi+PBi (Equation 22)
Further, as for the perpendicularity error A0Y, where the position of the spindle 5 circularly moved in the Y-Z plane is represented by Xi, Yik, Zik, the length SAik of the bar is as follows:
SAik=((Xi+EXik−Xi)2+(Yik+EYik−Yi)2+(Zik+EZik−Zi)2)1/2.
Where total data of the length SAik of the bar calculated is represented by SAi, a perpendicularity P′Ai calculated from SAi can be represented by the following relational equation:
f(SAi)=P′Ai. (Equation 24)
Accordingly, the true perpendicularity error A0Y in the coordinate system having it origin at the position Xa, Ya, Za that is an arbitrary position in the machine coordinate system can be identified by Equation 25 below based on the temporary perpendicularity error A0Y′, the perpendicularity PAi calculated by Equation 13 above, and the perpendicularity P′Ai calculated by Equation 24 above.
A0Y=A0Y′−P′Ai+PAi. (Equation 25)
In the above-described manner, the perpendicularly errors A0Y, B0X, and C0Y in the set coordinate system having its origin at the position Xa, Ya, Za that is an arbitrary position in the machine coordinate system are identified. Note that, in the case where the perpendicularity errors A0Y, B0X, and C0Y in the machine coordinate system defined with respect to the machine zero are identified, the perpendicularity errors A0Y, B0X, and C0Y are identified with values of EXik, EYik, and EZik which are calculated under the conditions: Xa=0, Ya=0, and Za=0.
5. Identification of Motion Error
Based on the error parameters identified in the above-described manners, the positioning errors EX(α,β,γ), EY(α, β, γ), and EZ(α,β,γ) of the reference point of the spindle 5 in the three-dimensional space of the machine coordinate system are identified by Equations 1 to 3 above, and the positioning errors EX(α,β,γ, TX, TY, TZ), EY(α,β,γ, TX, TY, TZ), and EZ(α,β,γ, TX, TY, TZ) of the tip of the tool attached to the spindle 5 in the three-dimensional space of the machine coordinate system are identified by Equations 4 to 6 above.
Further, the positioning errors EX(α,β,γ), EY(α, β, γ), and EZ(α,β,γ) of the reference point of the spindle 5 in the set coordinate system having its origin at the poaition Xa, Ya, Za that is an arbitrary position in the machine coordinate system are identified by Equations 7 to 9 above, and the positioning errors EX(α,β,γ, TX, TY, TZ), EY(α,β,γ, TX, TY, TZ), and EZ(α,β,γ, TX, TY, TZ) of the tip of the tool attached to the spindle 5 in the set coordinate system having its origin at the position Xa, Ya, Za that is an arbitrary position in the machine coordinate system are identified by Equations 10 to 12 above.
Thus, in this embodiment, the motion error (positioning errors) EX(α, β, γ), EY(α, β, γ), EZ(α, β, γ) of the reference point of the spindle 5 and the positioning errors EX(α,β,γ, TX, TY, TZ), EY(α,β,γ, TX, TY, TZ), and EZ(α,β,γ, TX, TY, TZ) of the tool tip in the set coordinate system having its origin at the position Xa, Ya, Za that is an arbitrary position in the machine coordinate system can be calculated in the above-described manner. Further, the motion error EX(α,β,γ), EY(α, β, γ), EZ(α,β,γ) and errors EX(α,β, γ, TX, TY, TZ), EY(α,β,γ, TX, TY, TZ), and EZ(α,β,γ, TX, TY, TZ) in the machine coordinate system can be calculated by using the conditions: Xa=0, Ya=0, and Za=0.
As described above, in this embodiment, the motion errors of the reference point and the motion errors of the tool tip in the three-dimensional space of the machine coordinate system and in the three-dimensional space of the set coordinate system having its origin at the arbitrary reference position Xa, Ya, Za can be identified based on actual error data that is measured by means of the commonly-used measurement methods complying with the regulations of JIS. Therefore, identification of the motion errors can be carried out inexpensively and simply in operation.
Further, since the motion errors in the set coordinate system having its origin at the arbitrary reference position Xa, Ya, Za can be identified, the degree of freedom of usage of the error data is increased.
Hereinbefore, one embodiment of the present invention has been described. However, the present invention is not limited thereto and can be implemented in other modes.
For example, the measurement of the errors in the above embodiment complies with the regulations of JIS; however, the errors may be measured by means of any other method which is able to measure the errors as accurately and easily as or more accurately and easily than the regulations of JIS.
The equations above for calculating the error parameters EXX, EYY, EZZ, EYX, EZX, EXY, EZY, EXZ, EYZ, EAX, EAY, EAZ, EBX, EBY, EBZ, ECX, ECY, and ECZ are shown by way of example only; the present invention is not limited thereto and these error parameters may be calculated by other equations. The identification of the perpendicularity errors A0Y, B0X, and C0Y is not limited to the above example as well, and these perpendicularity errors may be identified by means of any other method.
1 Machine tool
2 Bed
3 Colum
4 Spindle head
5 Spindle
6 Table
| Number | Date | Country | Kind |
|---|---|---|---|
| JP2016-249163 | Dec 2016 | JP | national |
| Filing Document | Filing Date | Country | Kind |
|---|---|---|---|
| PCT/JP2017/038842 | 10/27/2017 | WO | 00 |
| Publishing Document | Publishing Date | Country | Kind |
|---|---|---|---|
| WO2018/116635 | 6/28/2018 | WO | A |
| Number | Name | Date | Kind |
|---|---|---|---|
| 5678964 | Dashevsky | Oct 1997 | A |
| 20080114485 | Katoh | May 2008 | A1 |
| 20140297022 | Uenishi | Oct 2014 | A1 |
| 20150160049 | Oki | Jun 2015 | A1 |
| 20150277424 | Kondo | Oct 2015 | A1 |
| 20180079009 | Fujii | Mar 2018 | A1 |
| 20180133860 | Fujita | May 2018 | A1 |
| 20180150049 | Schranz | May 2018 | A1 |
| Number | Date | Country |
|---|---|---|
| S61209857 | Sep 1986 | JP |
| H08152909 | Jun 1996 | JP |
| 2008269316 | Nov 2008 | JP |
| 2015069355 | Apr 2015 | JP |
| 2012101788 | Aug 2012 | WO |
| Number | Date | Country | |
|---|---|---|---|
| 20200086444 A1 | Mar 2020 | US |
