JP2012040645A - Processing device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To increase the processing accuracy of shape processing which makes a processing tool scan a workpiece.SOLUTION: A processing device performs shape processing by making the processing tool 23 scan the workpiece W. The processing device detects each posture error of X, Y, Z stages 4, 5, 6 and obtains an amount of the displacement of a processing point W1 based on a relative position and the posture error of the processing point W1 relative to a reference point of each stage. The processing device corrects a target position of each stage stored in a numerical number calculator 12 based on the amount of the displacement of a processing point W1 on the workpiece to prevent the positional deviation of the processing point W1 caused by the posture error of the stage, thereby enhancing processing accuracy.

Description

  The present invention relates to a processing apparatus that performs shape processing or the like with high processing accuracy.

  Conventionally, in a processing apparatus that performs mechanical processing such as cutting and grinding, time axis data such as a target position and a moving speed of a moving stage of the processing apparatus is used as processing data. The machining data is calculated in advance from set values relating to the shape of the machining surface, machining tools, and the like, and is stored in the numerical arithmetic unit of the machining apparatus. When processing the processed surface of the workpiece into a curved surface, the processing tool is moved following the normal line of the curved surface. As a method of following the normal, a turning R (radius) that can be obtained by turning a machining tool or a tool R (radius) having a tool tip shape is used. In this case, the relative position between the machining tool and the workpiece is controlled, and machining is performed by matching the trajectory of the turning tool and the curved surface normal of the tool tip shape with the normal of the machining surface. Also in this method, the processing data given to the processing apparatus is data on the time axis such as the target position and moving speed of the moving stage.

  In such a processing apparatus, it is premised that a plurality of moving stages accurately follow the target position of the processing data, and even if a posture error occurs due to insufficient rigidity or guide distortion in the bearing mechanism of the moving stage. It becomes. In other words, when tilt occurs due to a change in posture on a moving stage on which a machining tool or workpiece is mounted, the machining tool or workpiece also tilts and the relative positional relationship changes, so that the machining tool does not contact a predetermined machining point. Processing errors. When machining by following the normal using the turning R or tool R, the distance from the rotation center of the tilt generated by the posture change on the stage to the machining point cannot be estimated from the target position of the stage, and the posture change Even if the amount was measured, the machining error amount could not be corrected.

  Conventionally, the influence of posture change has been reduced by high precision and high rigidity, such as a static pressure bearing mechanism as the bearing mechanism of the moving stage. However, there is a limit to increasing the accuracy and rigidity of the bearing mechanism. Therefore, several methods have been proposed as a method for correcting an error due to a posture change.

  The first method is to increase the attitude error of the moving stage beyond the accuracy of the bearing. FIG. 5 shows a conventional example. The guide 102 is a static pressure guide, and supports the Y stage 105 through a fluid gap of air and is movable in the Y direction. The Y stage 105 is driven by two linear motors 115a and 115b. In such a stage mechanism, the laser length measuring devices 111a and 111c using the laser beams 110a and 110c detect the position in the Y direction and yawing that is tilted around the Z direction. By adjusting the thrusts of the two linear motors 115a and 115b, the posture of the Y stage 105 in the yawing direction can be controlled although it is within the range of the fluid gap described above.

  The second method is a method of correcting the change in posture of the moving stage using another actuator, as disclosed in Patent Document 1. In this case, when the guide surface is tilted at the bearing of the moving stage, a mechanism for correcting the tilt of the guide surface is incorporated in the bearing mechanism to correct the pitching and yawing components of the moving stage caused by the influence of the guide surface. However, this configuration has a small range that can be corrected mechanically. Although the posture correction in the orthogonal direction of the moving stage is possible, the stroke in the mechanism is small and the correction in the orthogonal direction can be performed only within a minute range.

JP 2002-154028 A

  In a processing apparatus for processing a mold for optical components, it is required to process a mold having a complicated shape with very high accuracy (on the order of several tens of nanometers). Deterioration of machining accuracy due to stage attitude error is also regarded as a problem.

  As means for solving such a problem, a control axis for controlling the rotation direction of the moving stage is provided for correcting the change in posture of the moving stage as in the conventional example of FIG. However, since the bearing rigidity for high accuracy is increased, the range in which the attitude of the moving stage can be controlled by the actuator is limited.

  In addition, since a mechanism for controlling the rotational direction of posture changes is provided, the number of axes to be controlled increases, the moving stage mechanism becomes complicated, and on the contrary, the rigidity of the entire apparatus decreases, causing resonance, etc. The processing accuracy may be reduced.

  An object of this invention is to provide the processing apparatus which can shape-process the metal mold | die etc. of an optical component precisely.

  The processing apparatus of the present invention is a processing apparatus for processing a workpiece by scanning a workpiece with a plurality of moving stages, a stage driving means for driving each of the plurality of moving stages, and the plurality of moving stages. A stage attitude error detecting means for detecting each of the attitude errors, a relative position of the machining point on the workpiece with respect to a reference point of each moving stage, and the displacement of the machining point from the relative position and the attitude error Means for calculating a quantity, wherein the stage driving means corrects a target position of each moving stage based on a displacement amount of the machining point.

  The processing point coordinates as well as the target position of the moving stage as processing data of the processing apparatus are given in advance to the numerical calculation device, and the displacement amount with respect to the posture error is obtained from the posture error when moving the moving stage and the coordinates of the processing point. At this time, the calculation load is reduced by calculating only the displacement in the three orthogonal directions generated by the change in the posture of the moving stage with respect to the coordinates of the machining points registered in advance, and in real time on the numerical calculation device. It becomes possible to calculate. Machining accuracy can be improved by calculating the displacement amount of the machining point due to the change in the posture of the moving stage using only the orthogonal three-axis positional deviation components and correcting the target position of each moving stage.

It is a figure which shows the processing apparatus and processing process by Example 1. FIG. It is a figure explaining the method of making the normal of a process point and the normal of a process tool correspond. It is a figure which shows the processing apparatus and processing process by Example 2. FIG. It is a figure explaining the measurement part which detects a stage error. It is a figure which shows one prior art example.

Example 1
FIG. 1 shows a processing apparatus according to the first embodiment. A plurality of moving stages are moved in X, Y, and Z directions by guides 1, 2, and 3, respectively. Stage 6. Each guide 1, 2, and 3 is a linear guide using a static pressure bearing, and the drive source (stage drive means) of each stage 4, 5, 6 has a linear motor. The X stage detector 7, the Y stage detector 8, and the Z stage detector 9 constituting the stage attitude error detecting means measure the attitudes of the X, Y, and Z stages 4, 5, and 6, respectively. 10 is used to detect the posture error of each stage.

  The numerical arithmetic unit 12 has machining data 13, which is obtained from stage target position data 14 that is a target position of each stage and machining point coordinate data 15 that is coordinates indicating the position of the machining point. Composed. The stage target position data 14 and the processing point coordinate data 15 are each formed as an enumeration of data with the position and coordinates in the elapsed time from the processing start time as one block. As for the stage target position at a certain time, the target position in the driving direction for each of the X stage 4, the Y stage 5, and the Z stage 6 is one block. This target value, for example, defines a target position and a moving speed (time) to the target position as G01X0.0Y0.1Z0.2F1.0 when expressed in a general G code. Further, the processing point coordinate data 15 at this time is one block with coordinates such as coordinates (Xt, Yt, Zt) of the apparatus coordinate system with an arbitrary point on the processing apparatus as the origin. The stage target value and the coordinates of the machining point when performing a certain machining are equal, and the block of the stage target position at a certain point during machining and the block indicating the coordinates of the machining point at that point are one to one. Correspond.

  The reading step S1 is a step of reading data corresponding to one block of the stage target position data 14 and the machining point coordinate data 15 from the machining data 13 as data corresponding to the elapsed time from the machining start at the time of machining. The stage posture error detection step S2 is a step in which the posture of each stage is detected, and the numerical calculation device 12 reads the posture error of each stage. That is, the X stage detector 7, the Y stage detector 8, and the Z stage detector 9 detect the attitude, and read the attitude of each stage. The Y stage 4 carries the workpiece W, and the Y stage 5 supports the spindle motor 22 via the support 21 and scans the workpiece W in the Y direction while rotating the machining tool 23. Let

  Here, a method for detecting an attitude error will be described taking the Y stage 5 as an example. The Y stage detector 8 detects the posture of the Y stage 5 using a plurality of laser beams 10. For example, the position and pitching of the Y stage in the movable direction can be detected from the average and difference between the two laser beams 10. Position displacement in the movable direction is detected at multiple points at different positions on the stage, and the posture error detects the moving distance, pitching, yawing, and rolling in the movable direction by the Y stage detector 8 that measures the measurement point on the stage. To do. Pitching, yawing, and rolling of the Y stage 5 correspond to an inclination around the X direction, an inclination around the Z direction, and an inclination around the Y direction, respectively. The X stage 4 and the Z stage 6 other than the Y stage 5 are also measured by the X stage detector 7 and the Z stage detector 9 having the same configuration.

  In a high-precision processing apparatus, the center of gravity is supported by a guide mechanism and driven so as not to lower the stage movement accuracy. If the guide mechanism is a static pressure bearing, the position of the center of gravity of the stage can be made substantially coincident with the center of rotation. Therefore, the posture error is detected on the assumption that the position of the center of gravity of each stage coincides with the center of rotation that is the reference point.

  The first calculation step S3 is a step of obtaining a relative position from the reference point of each stage to the processing point W1 on the workpiece. In order to calculate the relative position up to the processing point W1, it is necessary to determine which position on the stage is the reference for the relative position. Here, the relative position is calculated using the position of the center of gravity of each stage as a reference point.

  In FIG. 1, the Y stage 5 is tilted around the X direction, which is a pitching error. The Y stage 5 supported and guided by the guide 2 is movable only in the Y direction which is a movable direction. The coordinates of the Y stage 5 when the target position of the Y stage 5 changes to y0, y1, y2,. It is assumed that the coordinates of the center of gravity of the stage, which is the reference point of the Y stage 5, in the X direction and the Z direction are YGy and YGz. The target position of the Y stage 5 changes as (YGx, y0, YGz), (YGx, y1, YGz), (YGx, y2, YGz),. Similarly, the machining point positions change as (Kx0, Ky0, Kz0), (Kx1, Ky1, Kz1), (Kx2, Ky2, Kz2). The relative position of the Y stage 5 from the position of the center of gravity of the stage to the machining point can be calculated from the difference in position in the three orthogonal axes. Therefore, (YGx-Kx0, y0-Ky0, YGz-Kz0), (YGx-Kx1, y1-Ky1, YGz-Kz1), (YGx-Kx2, y2-Ky2, YGz-Kz2), and so on. Perform the operation. The same calculation is performed for the other X stage 4 and Z stage 6.

  In the second calculation step S4, the displacement amount of the processing point W1 due to the posture error from the relative position from the center of gravity of each stage calculated in the first calculation step S3 and the posture error calculated in the stage posture error detection step S2. Is calculated for each stage. Here, the coordinates are calculated from the pitching, yawing and rolling angles of each stage. This is a simple coordinate transformation performed once for each block of the machining data 13 for each time to calculate the displacement amount in the three orthogonal directions. This is calculated for each stage in the second calculation step S4.

  The calculation method can be easily calculated by general formulas such as the following (Formula 1) and (Formula 2). In the calculation of this embodiment, it is only necessary to consider the displacement amount of the machining point, but the rotations around the X, Y, and Z directions are calculated as α, φ, and θ for each stage. As a calculation reference, the origin of the coordinate system is calculated as the stage barycentric position as the reference point. Here, UVW in (Expression 2) substitutes the relative position calculated in the first calculation step S3.

  The amount of displacement in the three orthogonal directions is obtained from the difference between the coordinates calculated using such a general formula and the original coordinates when there is no posture error. In FIG. 1, the Y stage 5 changes its posture and inclines in the ωx direction, which is an inclination around the X direction. Due to this influence, the processing tool 23 that should originally be in contact with the processing point W1 on the workpiece W has moved to the coordinates shifted in position from the processing point W1. The amount of deviation of this position is obtained from the relative distance from the center of gravity of the stage, which is the rotation center of the posture change of the Y stage 5, to the processing point W1. If this relative distance is a radius and an angle inclined around the X direction due to a change in posture is α, it can be calculated from (Equation 2). The coordinates of the center of gravity position, the coordinates of the machining point W1 and the relative distance thereof are obtained from the stage target position 14 and the machining point coordinates 15 in the first calculation step S3.

  The third calculation step S5 is a step of synthesizing the displacement amounts in the three orthogonal directions calculated for each stage in the second calculation step S4. Up to the second calculation step S4, the displacement amount is calculated as the displacement in the orthogonal triaxial direction for each stage, and the direction in which correction is performed is whether the orthogonal triaxial stage is mounted on the machining tool side or the workpiece side. Determine the positive and negative directions. The method for determining the positive and negative directions will be described. The posture of the Y stage 5 changes around the X direction, and the processing tool 23 is displaced from the processing point W1. The distance and direction in which the position is shifted are obtained in the second calculation step S4. The distance and direction in which the position is shifted were obtained from the two coordinates of the stage barycentric position and the processing point W1 and the direction of the inclination angle around the X direction. In FIG. 1, since the Y stage 5 is equipped with the processing tool 23, the processing tool 23 is displaced from its original coordinates. As for the direction of displacement, the point that should be in contact with the machining point W1 on the machining tool 23 is deviated from the machining point W1 in the positive Y direction and the negative Z direction. The correction of the target position to the Y stage 5 on which the processing tool 23 is mounted returns the corrected processing tool 23 to the original position, so that the correction value is opposite to the shifted direction. That is, the target position of the Y stage 5 is corrected in the negative direction in the Y direction. In the Z stage 6 on which the workpiece W is mounted, the contact point on the processing tool 23 moves in the negative direction in the Z direction. In order for the machining point W1 to come into contact with this point, the target position of the Z stage is corrected in the negative direction in the Z direction so as to follow the machining tool 23 that has moved. When the X stage 4 or the Z stage 6 has a posture change, the correction direction is determined in the same manner. Thereafter, a composite value of the displacement is calculated by simple four arithmetic operations for each of the X, Y, and Z directions.

  The correction control step S6 is a step of controlling the stage driving means for driving each stage by putting a correction value in the target position of each stage given in the reading step S1. The stage target position data 14 includes stage target values in the three orthogonal directions of the X stage 4, the Y stage 5, and the Z stage 6. A result calculated in the third calculation step S5 is added as a correction value to the stage target position data 14, and the stage position is controlled at the corrected target position.

  FIG. 2 illustrates an operation of causing the machining tool 23 to follow the machining point W1, and the machining tool 23 is installed on the spindle motor 22 in a disk shape. The spindle motor 22 rotates the processing tool 23 at a constant rotational speed during processing. The rotation center T0 of the spindle motor 22 is installed so as to be parallel to the Y direction. The processing tool 23 is disk-shaped but has a thickness in the Y direction, and the shape of the outer peripheral portion 24 has a constant radius R1. The radius R1 is called a tool R. The tool shape is a shape in which the outer peripheral portion 24 of the processing tool 23 has a constant radius around the X direction when the disk-shaped processing tool 23 is viewed from the X direction orthogonal to the central axis. A distance R2 from the rotation center T0 when the machining tool 23 is rotated by the spindle motor 22 is a turning R.

  The workpiece W to be processed has a shape having a curvature also in the X direction. The shape having such a curvature is processed by bringing one point of the outer peripheral portion 24 of the processing tool 23 into contact with the processing point W1. At this time, machining is performed by matching the normal line A2 on the workpiece side with the normal line A1 on the workpiece side at the machining point W1 of the workpiece W. This processing apparatus does not have an indexing shaft for causing the processing tool 23 to follow the rotation direction around the X direction. When the entire surface of the workpiece W is processed by the processing tool 23 with normal tracking, the point where the normal line A2 on the processing tool side coincides with the processing point W1 on the outer peripheral portion 24 of the processing tool 23.

  The target position of each stage is calculated from the relative position between the stages using parameters such as the tool R and the turning R of the machining tool 23 at this time. In the processing data calculation step, the position of the processing point W1, such as acceleration and speed in relative movement of a plurality of stages, is also calculated using parameters for scanning the surface of the workpiece W. In order to smooth the machining shape and increase the accuracy, it is necessary to shorten the distance interval for calculating the machining point W1. This calculation has a heavy calculation load and is difficult to calculate in real time when machining. Therefore, in the calculation of machining data, the target position of the stage and the coordinates of the machining point are created as machining data before the machining is started.

(Example 2)
FIG. 3 shows a second embodiment. In Example 1, corrections were made for pitching, yawing, and rolling components, which are stage posture changes. However, in addition to the rotational component, which is a change in posture, the stage error may move the stage in a direction other than the movable direction among the three orthogonal axes, and the stage causes a position error due to this parallel translation component. Of course, it becomes a processing error. Therefore, in this embodiment, the stage detectors 7, 8, and 9 constituting the stage error detecting means also detect the parallel translation error of each stage after the stage attitude error detecting step S12 for detecting the attitude error of each stage. To do. Other points are the same as in the first embodiment.

  The parallel translation will be described in detail. The Y stage 5 moving in the Y direction includes pitching and yawing ωx and ωy. When a rotation error around the X axis, which is pitching, occurs, the position of the center of gravity of the Y stage 5 may move in the X direction or the Z direction, which is a direction other than the movable direction in the Y direction, due to this change in posture. The amount of movement is detected as a parallel translation component of the stage error in the stage parallel translation error detection step S13. In the Y stage 5, the amount of movement of the center of gravity of the Y stage 5 in the X direction and the Y direction is detected as a parallel translation component, and this error is also corrected.

  The displacement of the machining point generated by the posture error and the parallel translation error for each stage detected in the stage posture error detection step S12 and the stage parallel translation error detection step S13 is calculated in the calculation step S14. In the calculation step S14, the calculation for the posture error detected in the stage posture error detection step S13 is the same as in the second calculation step S4 of the first embodiment, but the parallel translation component detected in the stage parallel translation error detection step S13 is described. Processing has been added. Then, with respect to the displacement due to the influence of the attitude error calculated for each stage, the displacement generated in parallel translation is synthesized for each of the three orthogonal directions of the X, Y, and Z directions. In this step, a calculation result is obtained by synthesizing both the influence of the posture error for each stage and the influence of the parallel translation error.

  The calculation in the calculation step S15 is a step of combining the displacement amount for each stage calculated in the calculation step S14 as a correction value. In the synthesis, the displacement amount for each stage is synthesized for each of the three orthogonal directions of the X, Y, and Z directions. The combined position is used as a correction value for the X stage 4, Y stage 5, and Z stage 6 to correct the target position of each moving stage.

  FIG. 4 shows a measurement unit constituting the stage attitude error detecting means of the first embodiment and the stage error detecting means of the second embodiment. The Y stage 5 is taken as an example to detect the stage position, attitude error, and parallel translation error. The Y stage detector 8 will be described. The position and orientation of the Y stage 5 are detected by the six laser length measuring devices 11a to 11f using the laser beams 10a to 10f, respectively. The laser length measuring devices 11a, 11b, and 11c detect the position of the Y stage 5 in the Y direction. From the average and difference of the values detected by these three laser length measuring devices, the Y-direction position and posture change pitching, the rotation around the X direction and the rotation around the Z direction as yawing components are measured. The laser length measuring device 11d detects parallel translation in the X direction, and the laser length measuring devices 11e and 11f detect parallel translation in the Z direction and rolling of the Y stage 5. The same applies to the X stage detector 7 and the Z stage detector 9 mounted on the X stage 4 and the Z stage 6.

(Example 3)
In the first embodiment, all attitude errors of the X and YZ stages are detected and corrected. In the second embodiment, all the position changes in the XYZ directions including the six axis directions, pitching, yawing, rolling and movable directions are detected and corrected by detecting the position and orientation of each stage. However, it is not essential to detect in real time the posture and parallel translation of all the X, Y, and Z stages. If the stage movement accuracy is high, or if the degree of influence due to the attitude error is small compared to the required machining accuracy, it is possible to ignore the attitude error with a small influence and calculate and correct only the influence attitude error. It is. Further, when the reproducibility is high, it is possible to previously store the posture error and the parallel translation error with respect to the target position of the stage in a database.

  For example, if the reproducibility of the posture error in the yawing direction with respect to the position in the movable direction is very high due to the stage configuration, the posture error in the yawing direction with respect to the position in the movable direction is detected in advance, and this detection result is compiled into a database. Keep it.

  At the time of machining, at the time of error detection shown in the first embodiment or the second embodiment, the posture error stored in the database is referred to. The database reference refers to the posture error corresponding to the target position of the stage from the database. Other than that is the same as Example 1 and Example 2.

Example 4
In this embodiment, when the reproducibility of each stage with respect to the target position is high, posture errors and parallel translation errors are detected in advance and stored in a database, and correction values are set in advance in the machining data generation process without correction in real time. Generate machining data including. In this case, in the CAM that generates the machining data of the numerical arithmetic device, the posture error is obtained by referring to the posture error with respect to the position in the movable direction for each stage in the database. The position displacement of the machining point caused by the influence is calculated from the attitude error and the parallel translation error obtained from the database.

  Here, the posture change and the parallel translation are detected for each stage using an arbitrary point on the stage as a rotation center and a movement point. Assuming that this arbitrary point moves in the movable direction, the target coordinates of the stage are also calculated. At this time, the target position of each stage is defined in a coordinate system having an arbitrary point in space in the processing apparatus as the origin, and the target coordinates are calculated.

  Next, machining data to be used at the time of machining is calculated in the numerical operation device of the machining apparatus. The machining point is obtained from the target shape of the workpiece. At this time, the machining point is defined in the same coordinate system as the target value, and the coordinates of the machining point are calculated from the scanning pattern between the machining points and the moving speed. The machining points are calculated at every elapsed time interval or distance from the machining start, and the target coordinates of each stage are calculated from the machining points. At this time, the relative position from the target coordinate of each stage to the machining point is calculated for each stage. The database is referred to at the calculated target coordinates. The position displacement of the processing point coordinates accompanying the posture change is calculated from the posture change of each stage and the above-mentioned relative distance, and the displacement amount due to parallel translation is also synthesized with this. The combined displacement amount is further combined as the displacement amount of all the stages to obtain a correction value. The correction value is calculated as a displacement amount in the orthogonal triaxial direction, and is added as a correction value to the target position with respect to the stage in the orthogonal triaxial direction.

1, 2, 3 Guide 4 X stage 5 Y stage 6 Z stage 7 X stage detector 8 Y stage detector 9 Z stage detector 11a to 11f Laser length measuring device 12 Numerical calculation device 22 Spindle motor 23 Processing tool

Claims (2)

In a processing apparatus that scans and processes a processing tool with respect to a workpiece by a plurality of moving stages,
Stage driving means for driving each of the plurality of moving stages;
Stage posture error detecting means for detecting posture errors of each of the plurality of moving stages;
Means for calculating a relative position of a processing point on a workpiece with respect to a reference point of each moving stage, and calculating a displacement amount of the processing point from the relative position and the posture error;
The stage driving means corrects the target position of each moving stage based on the displacement amount of the machining point. In a processing apparatus that scans and processes a processing tool with respect to a workpiece by a plurality of moving stages,
Stage driving means for driving each of the plurality of moving stages;
Stage error detection means for detecting each posture error and parallel translation error of the plurality of moving stages;
Means for calculating a relative position of a processing point on a workpiece with respect to a reference point of each moving stage, and calculating a displacement amount of the processing point from the relative position, the posture error, and a parallel translation error;
The stage driving means corrects the target position of each moving stage based on the displacement amount of the machining point.

Images