Method for determining the position of at least one axis of rotation of a rotary table and measuring system

Abstract

Method for determining the position of at least one axis of rotation (DA) of a rotary table (2) on which a workpiece (8) is arranged, comprising the steps: a. Determining a preliminary position of the axis of rotation (DA) of the rotary table (2), b. Generating at least three sets of measuring points at a first probing height (z1), wherein a set of measuring points comprises measuring points that are detected by probing the workpiece (8) at a starting point (P1, P2, P3) and during a subsequent rotation of the workpiece (8) by a sensor of a coordinate measuring machine (3), wherein the starting points (P1, P2, P3) for generating the sets of measuring points are different, wherein the coordinates of the measuring points are determined depending on the position of the axis of rotation (DA) of the rotary table (2), c. Determine at least one dimension piece of information for a probing height-specific workpiece profile for each of the three measurement point sets, d. Generating at least three sets of measurement points on a second probing height (z2), wherein the starting points (P1, P2, P3) for generating the sets of measurement points are different, e. Determining at least one dimension information of a probing height-specific workpiece profile for each of the three measurement point sets, f. Determining the position of the axis of rotation (DA) of the rotary table (2) as a function of the probing height-specific size information.

Description

[0001] The invention relates to a method for determining the position of at least one axis of rotation of a rotary table and a measuring system.

[0002] When measuring workpieces with a coordinate measuring machine, a workpiece coordinate system is typically first determined or defined through a process called calibration. During the subsequent measurement of the workpiece, the coordinates of the measured values ​​are determined in a machine coordinate system. For the subsequent evaluation of inspection features, the measured values ​​must then be transformed into the previously defined workpiece coordinate system. This allows, for example, the location (and orientation) of a hole, or the diameter of a hole, to be determined with respect to the workpiece coordinate system. These dimensions then usually have the reference specified on the corresponding manufacturing drawing.

[0003] When a rotary table is used to hold the workpiece in a coordinate measuring machine (CMM), particularly for a so-called 4-axis measurement operation, the measured values ​​obtained in the machine coordinate system must first be transformed into the rotary table-rotor coordinate system and then into the workpiece coordinate system. This requires that the exact position of the rotary table-rotor coordinate system relative to the machine coordinate system be known. The precision of this position determination has a significant influence on the measurement uncertainty of the measured elements and the derived inspection characteristics. For this reason, the position and orientation of the rotary table, which may be determined by the position and orientation of the rotary table axis, are measured before the measurement begins. This process is called "rotary table axis calibration."

[0004] There are several methods for measuring the rotary table axis. The preferred method for determining the rotary table axis, or the method that guarantees results with the lowest measurement uncertainty, depends, among other things, on the type of coordinate measuring machine used, its individual device-related residual errors, the stylus orientations used for measurement, and the geometric dimensions of the workpiece.

[0005] One possibility is to perform a probe of a calibration element while the rotary table is at rest, which is mounted on the rotary table for calibration purposes.

[0006] In particular, such single-point probing can be performed in four or more different angular positions of the rotary table and thus also of the calibration element. This procedure results in a relatively high time expenditure due to the single-point probing and the necessary repositioning of the coordinate measuring machine to the calibration element. The calibration element can be, for example, a sphere, a gauge ring, a triple element (ball or roller triplet) designed especially for self-centering probing, a cylinder, or even the geometry of the workpiece to be measured.

[0007] For example, the position of the center point of a spherical calibration element can be determined in four different rotational positions of the rotary table. A circle is then fitted to each of these four centers, and its center point defines the position of the rotary table axis, or its position vector, in the machine coordinate system. The normal perpendicular to the circle's surface, passing through the circle's center point, defines the orientation of the rotary table axis, or its direction vector, in the machine coordinate system.

[0008] For example, the position of the center point of a first calibration element, as well as of a second calibration element, can be determined in four different rotational positions of the rotary table, with the calibration elements arranged at different heights above the rotary table's faceplate. A circle can then be fitted into each of the four height-specific center points, with the center point of one of these circles defining the position of the rotary table axis, or its position vector, in the machine coordinate system. The line connecting the centers of both circles defines the orientation of the rotary table axis, or its direction vector, in the machine coordinate system.

[0009] This method also allows for the measurement of a cylindrical calibration element, which can also be the object being measured, at different heights. This can be performed, for example, as a so-called inversion measurement on a cylinder. This typically requires complex, laterally projecting probe configurations, which are time-consuming to change and may need to be calibrated. An example probe configuration is a T-shaped probe with two stylus balls. Often, the achievable accuracy with these probes is reduced, or depending on the measuring head system used, long-projecting probe configurations are not possible.

[0010] When measuring at two heights, especially with long probes extending laterally, an exact determination of the position of the rotary table axis is not possible or difficult due to residual errors, e.g., a so-called zRz error, where the zRz error denotes a rotational error around a usually vertically oriented z-axis of the machine coordinate system in different z-positions of the sensor, e.g., a probe.

[0011] In such methods, the rotary table is used only as a means of transport and not as a measuring instrument to determine the rotary table axis, whereby the workpiece to be measured, i.e. the object being measured, is moved around the mechanical rotary table axis, which is not yet known in terms of software, and possibly also around the workpiece axis, which may already be known, but the position of the rotary table axis is not taken into account to determine the coordinates of measuring points.

[0012] When a rotary table is used as a measuring instrument, the workpiece to be measured, i.e., the object being measured, is moved around the (already) mechanically known rotary table axis and, if necessary, also around the possibly already known workpiece axis. The position of the rotary table axis is taken into account to determine the coordinates of measuring points. For example, deviations between a rotary table axis and a workpiece axis caused by wobble or eccentricity can be corrected when determining the coordinates.

[0013] Furthermore, a common problem is that a suitable calibration element must be positioned on the rotary table's faceplate to calibrate the rotary table axis. This may require removing the workpiece from the faceplate first, which increases the time required and also reduces the accuracy of the rotary table axis calibration, as the weight forces and moments exerted by the workpiece can affect the position of the rotary table axis.

[0014] Similar problems arise when determining the workpiece coordinate system.

[0015] DE 10 2013 219 389 A1 discloses a method for reducing errors of a rotary device, wherein the errors occur or become effective in determining the coordinates of a workpiece or in machining a workpiece.

[0016] CN 107 560 583 A discloses a method for axially correcting a cylindrical workpiece and a method for measuring the diameter of each segmented cylindrical section of a cylindrical workpiece.

[0017] From WO 2013 / 164344 A1, a method for repeatedly determining the position of a rotary table axis in the device coordinate system of a coordinate measuring machine is known. This method describes probing a test specimen located on the rotary table in several rotary table positions using the measuring system of the coordinate measuring machine and determining the X-coordinate and Y-coordinate of the rotary table axis in the device coordinate system from the probing points. The test specimen is only probed when stationary, i.e., not during rotation.

[0018] WO 02 / 090879 A2 discloses a method for determining the properties of a coordinate measuring machine and a test object for this purpose. The document discloses that it is important to know the position of the axis of rotation in the coordinate system of the base.

[0019] DE 198 15 098 B4 discloses a method for measuring rotary table deviations.

[0020] The technical problem is to create a method for determining the position of at least one axis of rotation of a rotary table and a measuring system that enables a simple and accurate determination of the position.

[0021] The solution to the technical problem is provided by the articles with the features of the independent claims. Further advantageous embodiments of the invention are described in the dependent claims.

[0022] A method is proposed for determining the position of at least one axis of rotation of a rotary table on which a workpiece is arranged, comprising the following steps: a. Determining a preliminary position of the rotary table's axis of rotation, b. Generating at least three sets of measuring points at a first probing height, wherein a set of measuring points comprises measuring points that are detected by probing a workpiece at a starting point and during a subsequent rotation of the workpiece by a sensor of a coordinate measuring machine, wherein the starting points for generating the sets of measuring points are different, and wherein the coordinates of the measuring points are determined depending on the position of the axis of rotation. c. Determine at least one dimension piece of information for a probing height-specific workpiece profile for each of the three measurement point sets, d. Generating at least three sets of measurement points at a second probing height, wherein the starting points for generating the sets of measurement points are different, e. Determining at least one dimension information of a probing height-specific workpiece profile for each of the three measurement point sets, f. Determining the position of the rotary table's axis of rotation depending on the size information.

[0023] The position of the rotary table axis can be determined in one of the previously described coordinate systems, in particular the machine coordinate system. This can be a Cartesian coordinate system. Preferably, the machine coordinate system comprises a vertical axis (z-axis), a longitudinal axis (x-axis), and a transverse axis (y-axis), each oriented perpendicular to the others. The rotary axis of the rotary table can be oriented essentially parallel to one of these axes, preferably the z-axis, which is hereinafter referred to as the reference axis. A probing height can then be defined as a height along this reference axis.

[0024] The position of the rotary table axis can denote a position and / or an orientation of the rotary table axis, in particular in the form of a position vector or a direction vector in the machine coordinate system. The position vector can be the position vector of a reference point of the rotary table axis, in particular an intersection point with a reference plane, e.g., with a plane of the surface of the rotary table or with a workpiece profile surface, wherein the workpiece profile surface is a cross-sectional area through the workpiece in a cross-sectional plane that is oriented perpendicular to the rotary table axis.

[0025] The preliminary position of the rotary table axis can be determined, for example, as explained in more detail below. Alternatively, the preliminary position can be predetermined, i.e., fixed. Information about such a preliminary position can be stored in a memory device in a retrievable format.

[0026] When rotating the workpiece to generate a set of measurement points, the workpiece can be rotated by exactly 360° or at least 360°. However, this is not mandatory. The workpiece can also be rotated by less than 360° to generate a set of measurement points. A set of measurement points can comprise exactly three, but preferably more than three, measurement points, these measurement points being preferably evenly distributed along the circumference of the workpiece at the respective probing height.

[0027] The measurement points of a measurement point set are measurement points along a profile line on the workpiece surface at a specific probing height. Depending on the coordinate values ​​of measurement points within a measurement point set, information about the probing-height-specific workpiece profile can be determined, e.g., the shape of the workpiece profile, its size, or the location of a reference point of the workpiece profile. The reference point can be, for example, a center point, in particular a geometric center, a centroid, or another uniquely definable point of a profile surface bounded by the profile line. Size information can be, for example, an area, geometric information of a substitute element fitted into the profile line, such as the diameter of a circle fitted into the profile line, or, in particular, a maximum diameter of the profile surface.The coordinates of the measuring points can first be determined in the machine coordinate system and then transformed into the workpiece coordinate system depending on the preliminary position of the rotary table axis, in particular via the rotary table-rotor coordinate system. Thus, the size information and / or the position information can be referenced to the workpiece coordinate system.

[0028] For each probing height, at least three different profile lines, and thus three different workpiece profiles and their profile-specific size information, are determined. These can differ from one another, particularly the profile-specific dimensions and / or the profile-specific positions. The deviations of these dimensions and / or positions can also be referred to as variations in probing-height-specific dimension information or variations in probing-height-specific position information.

[0029] The starting points can have the same coordinates with respect to the reference axis, but differ from each other in at least one coordinate with respect to the remaining axes. For example, a starting point could be a point on the workpiece surface that has the maximum or minimum coordinate value along one of the remaining axes at the probing height.

[0030] After completing steps a. to e., at least three touch-height-specific dimension pieces of information are available for at least two touch heights. These pieces of information depend on the preliminary position of the rotary table axis determined in step a., since the coordinates of the measuring points were determined based on this preliminary position of the rotary table axis, particularly when transformed into the workpiece coordinate system. In particular, two touch-height-specific positions of the rotary table axis can thus be determined, depending on the respective touch-height-specific dimension pieces of information.

[0031] As explained in more detail below, the position of the rotary table axis, in particular its probing height-specific position, is determined based on the size information, specifically the variations in that probing height-specific size information. It was thus recognized that the size information, especially the aforementioned variations, contains information about a rotary table axis error, i.e., a deviation of the true rotary table axis position from the preliminary rotary table axis position. Therefore, the position of the rotary table's axis of rotation can be determined simply and accurately in a particularly advantageous manner.

[0032] If, for example, the workpiece has a circular or substantially circular cross-sectional profile at the first and second probing heights, the starting points for a probing height can be positioned as follows: The starting points at each probing height are positioned such that a connecting line between a first and a second starting point intersects the preliminary rotary table axis or the reference axis, and the connecting line of the third starting point is perpendicular to the connecting line of the other two starting points.

[0033] A circle is fitted as a substitute element into each of the measurement points of the first set of measurement points, which are recorded starting from the first initial point; into the measurement points of the second set of measurement points, which are recorded starting from the second initial point; and into the measurement points of the third set of measurement points, which are recorded starting from the third initial point. This is done, for example, using a best-fit method. That is, a first circle is fitted into the measurement points of the first set of measurement points, a second circle into the measurement points of the second set of measurement points, and a third circle into the measurement points of the third set of measurement points. Then, an average diameter can be determined from the measured diameters of the fitted first circle and the fitted second circle.

[0034] Depending on this mean diameter, a first correction factor is calculated. This first correction factor indicates how far the position of the provisional rotary table axis must be shifted at the respective probing height along or parallel to the line connecting the first and second starting points. The first correction factor is determined to be half the deviation of the diameter of the first circle from the mean diameter. Alternatively, the first correction factor can also be determined as half the deviation of the diameter of the second circle from the mean diameter. In this case, the position of the provisional rotary table axis can be shifted by the correction factor along or parallel to the line connecting the first and second starting points.

[0035] The direction in which the position of the temporary rotary table axis is shifted depends on whether the diameter of the respective circle (i.e., the first circle or the second circle) is smaller or larger than the average diameter. If the diameter of the respective circle (i.e., the first circle or the second circle) is smaller than the average diameter, the position of the temporary rotary table axis is shifted away from its corresponding starting point (i.e., from the first starting point if the first circle is used, or from the second starting point if the second circle is used). Conversely, if the diameter of the respective circle is larger than the average diameter, the position of the temporary rotary table axis is shifted toward its corresponding starting point (i.e., toward the first starting point if the first circle is used, or toward the second starting point if the second circle is used).The sign of the first correction factor can therefore be determined accordingly.

[0036] The position of the temporary rotary table axis can also be shifted along or parallel to the line connecting the third starting point and the temporary rotary table axis. The amount of this shift can be chosen as half the deviation of the third circle's diameter from the mean diameter and the amount of a second correction factor. The direction in which the position of the temporary rotary table axis is shifted along or parallel to this line depends, in turn, on whether the diameter of the third circle is smaller or larger than the mean diameter. If the diameter of the third circle is smaller than the mean diameter, the position of the temporary rotary table axis is shifted away from the corresponding third starting point.If, however, the diameter of the circle in question is larger than the mean diameter, the position of the provisional rotary table axis is shifted towards the corresponding third starting point. The sign of the second correction factor can then be determined accordingly.

[0037] The position of the preliminary rotary table axis, corrected depending on the two correction factors, is then the (corrected) probing height-specific position of the rotary table axis.

[0038] If the position of the starting points at a given probing height is not chosen such that a connecting line between the first and second starting points exactly intersects the preliminary rotary table axis or the reference axis, and / or the connecting line of the third starting point with the preliminary rotary table axis or the reference axis is perpendicular to the connecting line of the other two starting points, then, depending on geometric relationships, especially when considering a circle, a corrected diameter for the circles in question can be determined. That these positional conditions for the starting points are not met, which can be the case, for example, with eccentric clamping of the workpiece, can be detected if the center coordinates of the circles deviate from the coordinates of the zero position in one of the coordinate systems described, especially in the preliminary workpiece coordinate system, at the respective probing height.

[0039] The position of the starting points can also be converted into a corrected position based on geometric relationships, fulfilling the previously explained conditions for the position of the starting points. Then, the correction factors can be determined as previously explained, depending on the corrected diameter. The corrected positions of the preliminary rotary table axis can also be determined as previously explained.

[0040] In a further embodiment, the position of the rotary table axis, in particular a probing height-specific position, is calculated as a function of the variation of the probing height-specific dimension information. It was recognized that an analytical relationship exists between the variation of the probing height-specific dimension information and the position of the rotary table axis, especially for a workpiece with a circular cross-sectional profile. This analytical relationship can be evaluated to determine the position of the rotary table axis. For example, the deviation between the preliminary position of the rotary table axis and the true position of the rotary table axis can be determined using this analytical relationship. The true position of the rotary table axis is then determined as a function of the preliminary position determined in step a. and the deviation. A possible determination of the deviation in the form of correction factors was explained above.This allows the position of the rotary axis of the turntable to be determined easily and accurately in an advantageous manner.

[0041] Alternatively, the position of the rotary table axis, in particular a probing height-specific position, is determined such that the variation of the probing height-specific parameter information is zero or less than a predetermined threshold. This can be achieved using an optimization procedure, where the at least one parameter to be optimized represents the true position of the rotary table axis or the deviation between the preliminary position of the rotary table axis and the true position, and the cost function depends on the variation. In particular, the parameter to be optimized can be determined such that the probing height-specific variation is minimized. This can be done iteratively, for example, by repeatedly changing the true position or the deviation, e.g., in the form of at least one correction factor, and then checking whether the variation of the probing height-specific parameter information is zero or less than a predetermined threshold.During the optimization process, after each change to the parameter being optimized (i.e., after each iteration step), and to determine the cost function, the orientation of the rotary table axis can be determined as the orientation of the connecting line between the probing height-specific positions of the rotary table axis determined (corrected) in the iteration step. These positions can be determined as probing height-specific corrected positions. It is possible that after each iteration step, at least three sets of measurement points, or at least one of these sets, are regenerated by probing with the sensor. Based on this regenerated set of measurement points, at least one piece of information about the quantity, and consequently the position of the rotary table axis, is determined. However, this is not mandatory.Thus, after each iteration step, the coordinates of the previously generated measurement points of the three sets of measurement points can be recalculated. Based on these recalculated coordinates, at least one piece of information about the size of the data, and consequently the position of the rotary table axis, can be determined. This also advantageously results in a simple and accurate determination of the position of the rotary table's axis of rotation.

[0042] In a further embodiment, at least one correction factor is determined for a preliminary position of the rotary table's axis of rotation specific to the touchpoint height, and the position of the rotary table's axis of rotation is then determined as a function of the preliminary position and the at least one correction factor. In particular, a corrected position of the rotary table's axis can be determined by shifting the preliminary position by the at least one correction factor. The correction factor can encode or represent an amount and a direction of the shift. This advantageously enables a simple and easily implemented determination of the rotary table's axis of rotation.

[0043] In another embodiment, the orientation of the rotary table axis is determined as the orientation of a connecting line between touch-height-specific positions of the rotary table axis, i.e., between the previously explained corrected positions.

[0044] As explained above, at least one correction factor can be calculated to correct a workpiece profile position specific to the probing height, depending on the variation in the probing height-specific dimension information. The probing height-specific position of the rotary table axis can, for example, correspond to a preliminary position corrected by the at least one correction factor. Since the probing height can be fixed along the reference axis, preferably one correction factor can be determined for each remaining axis of the machine coordinate system, e.g., an x-axis correction factor as the first correction factor and a y-axis correction factor as the second correction factor.

[0045] The orientation of the rotary table axis can then be determined as the orientation of the connecting line between the positions of the preliminary rotary table axis, corrected by at least one correction factor. The position of the rotary table axis can be determined as one of the probing-height-specific corrected positions. Thus, the position of the rotary table's axis of rotation can be determined simply and accurately in an advantageous manner.

[0046] In a further embodiment, the method additionally comprises the following steps: g. Determining at least one positional information of a probing height-specific workpiece profile for at least one, preferably for each, of the at least three probing height-specific measurement point sets and at least one correction factor, h. Determining the position of the workpiece coordinate system depending on the probing height-specific position information and at least one correction factor.

[0047] Before step g., a preliminary position of a workpiece coordinate system can be determined.

[0048] Regarding the position of the workpiece coordinate system, reference can be made to the explanations concerning the position of the rotary table axis. The position information can be information about the position of a reference point of the workpiece profile. This has already been explained previously. For each probing height, at least one, preferably at least three, profile-specific position information pieces are determined, whereby the profile-specific position information pieces for a probing height can be different from each other, the same, or differ from each other by no more than a predetermined amount.

[0049] After performing steps g. to h., at least one probing height-specific position information can be available for each probing height, which may depend on a previously determined preliminary position of the workpiece coordinate system, since the coordinates of the measuring points can be determined as a function of this preliminary position of the workpiece coordinate system, in particular if they have been transformed into the workpiece coordinate system.

[0050] Furthermore, at least one correction factor is determined. This correction factor can specify how far the position of the reference point must be shifted at the respective probing height, in particular along or parallel to one of the previously explained connecting lines between starting points.

[0051] As explained in more detail below, the position of the workpiece coordinate system is determined based on the position information and at least one correction factor, particularly based on the variations in the probing height-specific position information. It was thus recognized that the position information, especially the aforementioned variations, contains information about a workpiece coordinate system error, i.e., a deviation of the true position of the workpiece coordinate system from its preliminary position. Therefore, the position of the workpiece coordinate system can be determined simply and accurately in a particularly advantageous manner.

[0052] Referring to the previously explained example of the method for a workpiece that has a circular or substantially circular cross-sectional profile at the first and second probing heights, the position of the center point averaged from the centers of at least two, preferably at least three, circles of the set first circle, second circle, third circle can be determined as position information for the probing height-specific workpiece profile. However, it is also conceivable that the center point of one of the circles from this set is determined as position information. Then, the (corrected) position of the workpiece axis at the respective probing height can be determined as the position of the center point shifted by the at least one correction factor. Regarding the magnitude and direction of the shift of the center point position encoded by a correction factor, reference can be made to the preceding explanations.

[0053] The orientation of the workpiece axis can then be determined as the orientation of the connecting line between the positions of the probing-height-specific center points, each corrected by at least one correction factor. The (corrected) position of the workpiece axis can be determined as one of the probing-height-specific corrected positions. Thus, the position of the workpiece axis can be determined simply and accurately in an advantageous manner.

[0054] In accordance with the explanations for determining the position of the rotary table axis, the position of the workpiece coordinate system can also be calculated as a function of the variation of the probing height-specific position information, or the position of the workpiece coordinate system can be determined in such a way that the variation of the probing height-specific position information is zero or less than a predetermined threshold value.The orientation of a reference axis of the workpiece coordinate system can be determined as the orientation of a connecting line between (corrected) reference points of the probing height-specific workpiece profiles. This involves calculating at least one correction factor for correcting a probing height-specific workpiece profile position as a function of the variation in the probing height-specific position information, or determining at least one correction factor for correcting a probing height-specific workpiece profile position such that the variation in the probing height-specific position information is zero or less than a predetermined threshold. Reference can be made to the explanations for determining the position of the rotary table axis, which apply in the same or a similar way to determining the position of the workpiece coordinate system.

[0055] Thus, at least two correction factors can be determined for each probing height: at least one correction factor for determining the position of the rotary table axis and at least one correction factor for determining the position of the workpiece coordinate system. Alternatively, at least four correction factors can be determined for each probing height, for example, one correction factor per of the previously explained connecting lines or per remaining axis of the machine coordinate system for determining the position of the rotary table axis, and one correction factor per remaining axis of the machine coordinate system for determining the position of the workpiece coordinate system. However, the correction factors for determining the position of the workpiece coordinate system can be determined as a function of, or as, the correction factors for determining the position of the rotary table axis.

[0056] In a further embodiment, a set of measuring points is determined at a third probing height, wherein the preliminary position of the rotary table axis and / or the preliminary position of the workpiece coordinate system is determined as a function of the measuring points of this set of measuring points. The set of measuring points can preferably comprise exactly three or more than three measuring points. The set of measuring points can include measuring points that are detected by probing the workpiece at a starting point and during a subsequent rotation of the workpiece by the sensor of a coordinate measuring machine. However, the set of measuring points can also include measuring points that are detected without rotation of the workpiece.

[0057] The measuring points of the third probing height are also measuring points along a profile line on the workpiece surface. The preliminary position can be determined based on the coordinate values ​​of these measuring points. For example, a direction vector of the preliminary position of the rotary table axis and / or a reference axis of the workpiece coordinate system can represent the orientation of the normal to the profile surface bounded by the profile line. The profile line could, for example, be the boundary line of a substitute element fitted into the measuring points. The profile surface can then be the area bounded by this boundary line.

[0058] A position vector representing the preliminary position of the rotary table axis and / or a reference axis of the workpiece coordinate system can represent the position of a reference point on the workpiece profile. In particular, a preliminary position of the rotary table axis can correspond to a preliminary position of a reference axis of the workpiece coordinate system. This results in a simple and reliable determination of the preliminary position.

[0059] In a further embodiment, the workpiece has at least a section with a circular cross-sectional profile, wherein the diameter of a circular profile is determined as size information and / or the position of the center point of a circular profile is determined as position information. In this embodiment, the profile lines along which the measuring points were generated are circles, and the area enclosed by them is a circle. This results in a simple and reliable determination of the preliminary position.

[0060] In another embodiment, the workpiece is probed tactilely or optically. This advantageously results in a simple and reliable generation of measuring points.

[0061] A further proposed measuring system comprises at least one rotary table, at least one coordinate measuring machine, and at least one evaluation unit, wherein the measuring system is configured such that a method according to one of the embodiments described in this disclosure, i.e., at least with steps a. to e., can be carried out with the measuring system. In particular, the measuring points acquired in steps b. and d. can be acquired by the coordinate measuring machine. At least steps c., e., and f. can be carried out by the evaluation unit. Thus, the implementation of a method for determining the position of the rotary axis of the rotary table is advantageously enabled, with the advantages explained above.

[0062] The invention is explained in more detail using exemplary embodiments. The figures show: Fig. 1 a schematic representation of a measuring system according to the invention, Fig. 2 a schematic flowchart of a method according to the invention and Fig. 3 A schematic representation of different starting points.

[0063] In the following, identical reference symbols denote elements with the same or similar technical characteristics.

[0064] Fig. Figure 1 shows a schematic representation of a measuring system 1 according to the invention, which comprises a rotary table 2, a coordinate measuring machine 3 and an evaluation unit 4. Fig. Figure 1 shows only a vertically oriented quill 5 of the coordinate measuring machine 3, with a tactile sensor with a stylus 6 and a probe ball 7 for tactile probing of a workpiece 8 attached to a free end of the quill 5. The sensor with the attached stylus 6 and probe ball 7 can be moved along three spatial directions in a machine coordinate system using the coordinate measuring machine 3. This system has a longitudinal axis xK, a transverse axis yK (see Figure 1). Fig. 3) and a vertical axis zK, which are arranged perpendicular to each other. The rotary table 2 is arranged on a surface of a measuring table 9 and comprises a stationary part 10 and a rotatable part 11. The workpiece 8, for example a shaft comprising several cylindrical sections with different diameters and gear sections, is arranged on a surface of the rotatable part 11. A rotary table axis DA or rotation axis DA of the rotary table 2 and a reference axis zW of a workpiece coordinate system are shown. The workpiece coordinate system can also be designed as a Cartesian coordinate system and include two further axes (not shown). An origin of the workpiece coordinate system can be arranged along the reference axis zW, for example at an intersection of this reference axis with the base surface of the workpiece 8, which is arranged on the surface of the rotatable part 11.

[0065] The vertical axis zK of the machine coordinate system can be a reference axis for determining / specifying a probing height in relation to the workpiece 8.

[0066] A method for determining the position of the rotation axis DA of the rotary table 2 is now described with respect to Fig. Section 2 explains. In a first step S1, a preliminary position of the rotary axis DA of the rotary table 2 and a preliminary position of the workpiece coordinate system are determined. For this purpose, the workpiece 8 can be probed at a predetermined probing height z0 along the vertical axis zK to generate a set of measuring points comprising exactly or at least three measuring points. The workpiece 8 can be probed at a starting point and during a subsequent rotation of the workpiece 8 about the rotary table axis DA by the probe ball 7 to generate measuring points. However, rotating the workpiece to generate the set of measuring points in the first step S1 is not strictly necessary. Thus, coordinates of measuring points can be acquired along a dotted profile line 12 of the workpiece 8, where this profile line 12 outlines a profile surface.Profile line 12 can be the boundary line of a substitute element fitted into the set of measuring points, in this exemplary embodiment a circle. The preliminary orientation of the rotary table axis DA and the preliminary orientation of the reference axis zW of the workpiece coordinate system are then determined as the orientation of a normal to the profile surface. The preliminary position is determined as the center point of the profile surface.

[0067] Of course, the preliminary position of the rotary table axis DA and / or the workpiece coordinate system can also be determined in other ways; for example, it can be predetermined.

[0068] In a second step S2, three sets of measuring points are generated at a first probing height z1, one of these sets of measuring points comprising measuring points that are obtained by probing the workpiece 8 at a starting point P1, P2, P3 (see Fig. 3) and are generated during a subsequent rotation of the workpiece 8 around the rotary table axis DA by the probe ball 7, wherein the starting points P1, P2, P3 for generating the measurement point sets are different. The coordinates of the measurement points are determined depending on the preliminary position of the rotary axis DA, in particular since the coordinates are transformed into the workpiece coordinate system.

[0069] The coordinates of measuring points from a set of measuring points are thus recorded along a dotted profile line 12 of the workpiece 8, where this profile line 12 outlines a profile surface. In a third step S3, the diameter of each of these three profile surfaces is determined. For this purpose, a circle is fitted into the measuring points of each of these three profile lines using a fitting procedure, for example, a best-fit procedure, and the diameter of the fitted circle is determined. The position of the center point of at least one of these three profile surfaces is also determined. As a result, for this first probing height z1, a first, a second, and a third circle, as well as their respective diameters and center point positions, were determined.

[0070] In a fourth and a fifth step S4, S5, steps S2, S3 are repeated, but three sets of measurement points are generated at a further probing height z2.

[0071] In a sixth step S6, the position of the rotary table axis DA is determined as a function of the diameters and the position of the workpiece coordinate system, in particular the reference axis zW, is determined as a function of the center positions, in particular as a function of the variation of the probing height-specific diameters and the probing height-specific center positions.

[0072] For this purpose, a mean diameter of the first and second circles and a first correction factor can be determined for each probing height z1, z2, where the first correction factor indicates how far the position of the preliminary rotary table axis DA in the respective probing height z1, z2 must be shifted along or parallel to a connecting line between the starting points P1, P2 of the first and second set of measuring points.

[0073] The first correction factor is determined, for example, as half the deviation of the diameter of the first circle from the mean diameter. Then, the position of the temporary rotary table axis DA can be shifted by this correction factor along, or parallel to, the line connecting the starting point P1 of the first set of measurement points (first starting point) and the starting point P2 of the second set of measurement points (second starting point). If the diameter of the first circle is smaller than the mean diameter, the position of the temporary rotary table axis DA is shifted away from the first starting point P1. Conversely, if the diameter of the first circle is larger than the mean diameter, the position of the temporary rotary table axis DA is shifted towards the first starting point P1.

[0074] A second correction factor is then determined to shift the position of the preliminary rotary table axis DA along or parallel to a connecting line between a starting point P3 of the third set of measurement points (third starting point P3) and the preliminary rotary table axis DA. The magnitude of this second correction factor can be chosen as half the deviation of the diameter of the third circle from the mean diameter. The direction in which the position of the preliminary rotary table axis DA is shifted along or parallel to this connecting line depends, in turn, on whether the diameter of the third circle is smaller or larger than the mean diameter. If the diameter of the third circle is smaller than the mean diameter, the position of the preliminary rotary table axis DA is shifted away from the third starting point P3.If, however, the diameter of the third circle is larger than the middle diameter, the position of the temporary rotary table axis DA is shifted towards the third starting point P3.

[0075] The position of the preliminary rotary table axis DA, corrected according to the two correction factors, is then the (corrected) probing-height-specific position of the rotary table axis DA. The corrected position of the rotary table axis DA can then be determined as one of these probing-height-specific (corrected) positions. The orientation of the rotary table axis DA can be determined as the orientation of a connecting line between these (corrected) probing-height-specific positions of the rotary table axis DA.

[0076] Furthermore, for each probing height z1, the position of an averaged center point of all three circles can be determined as positional information of the probing height-specific workpiece profile, whereby the averaged center point is determined as the mean of the centers of the three circles. Then, the (corrected) probing height-specific position of the workpiece axis at the respective probing height z1, z2 can be determined as the position of the averaged center point, which is shifted by the explained correction factors along or parallel to the explained connecting lines.

[0077] If the diameter of the first circle is smaller than the average diameter, the position of the averaged center point is shifted away from the first starting point P1. Conversely, if the diameter of the first circle is larger than the average diameter, the position of the averaged center point is shifted towards the first starting point P1. This shift is made by the amount of the first correction factor and along, or parallel to, the line connecting the first starting point P1 and the second starting point P2.

[0078] If the diameter of the third circle is smaller than the mean diameter, the position of the averaged center point is shifted away from the third starting point P3. Conversely, if the diameter of the third circle is larger than the mean diameter, the position of the averaged center point is shifted towards the third starting point P3. This shift is made by the amount of the second correction factor and along, or parallel to, the line connecting the third starting point P3 and the temporary rotary table axis DA.

[0079] The position of the averaged center points, corrected according to the two correction factors, is then the (corrected) probing-height-specific position of the workpiece axis. The corrected position of the workpiece axis can then be determined as one of these probing-height-specific (corrected) positions. The orientation of the workpiece axis can be determined as the orientation of a connecting line between these (corrected) probing-height-specific positions of the workpiece axis.

[0080] The position of the rotary table axis DA can also be calculated as a function of the variation in the probing height-specific diameters, and the position of the workpiece axis can be calculated as a function of the variation in the probing height-specific center point positions. Thus, as explained above, at least one correction factor, in particular an x-axis correction factor and a y-axis correction factor, can be calculated for each probing height z1, z2, depending on the variation in the probing height-specific diameters. The x-coordinate of the corrected position of the rotary table axis DA, specific to the probing height, is determined as the sum of the x-coordinate of the preliminary position of the rotary table axis DA and the x-correction factor. The y-coordinate of the corrected position of the rotary table axis DA, specific to the probing height, is determined as the sum of the y-coordinate of the preliminary position of the rotary table axis DA and the y-correction factor.The orientation of the rotary table axis DA is then determined as the orientation of a connecting line between the touch-height-specific, corrected positions. The position of the rotary axis DA can then be determined as the position of one of the touch-height-specific, corrected positions.

[0081] For each probing height z1, z2, at least one correction factor, in particular an x-axis correction factor and a y-axis correction factor, can be calculated depending on the variation of the probing height-specific center point positions. The x-coordinate of the probing height-specific, corrected workpiece profile position is determined as the sum of the x-coordinate of the preliminary position of the reference axis zW, which can be the same as the center point position, and the x-correction factor. The y-coordinate of the probing height-specific, corrected workpiece profile position is determined as the sum of the y-coordinate of the preliminary position of the reference axis zW and the y-correction factor. The orientation of the reference axis is then determined as the orientation of a connecting line between reference points of the probing height-specific, corrected workpiece profiles, in particular the position-corrected positions of the center points.The position of the reference axis zW can then be determined as the position of one of the touch-height-specific, corrected workpiece profile positions.

[0082] Alternatively, the position of the rotary table axis DA can be determined such that the variation in the probing height-specific diameters is zero or less than a predetermined threshold. Similarly, the position of the workpiece coordinate system can be determined such that the variation in the probing height-specific center point positions is zero or less than a predetermined threshold. This can be achieved through an optimization procedure, where the parameters to be optimized are the correction factors explained above, and the cost function to be minimized represents the variation in the probing height-specific diameters or center point positions. For each step of the optimization procedure, an updated position of both the rotary table axis DA and the reference axis zW is determined, which in turn affects the variation in the probing height-specific diameters or center point positions.Using the correction factors that minimize the cost function, the position of the rotary table axis DA and the reference axis zW of the workpiece coordinate system can then be determined, as explained.

[0083] Fig.Figure 3 shows a schematic representation of different starting points P1, P2, P3 for generating the three sets of measuring points in the machine coordinate system. A first probing height-specific set of measuring points can be generated, for example, with a first starting point P1, which is a point on the workpiece surface that has the maximum coordinate value along the y-axis at the respective probing height z1, z2. A second probing height-specific set of measuring points can be generated, for example, with a second starting point P2, which is a point on the workpiece surface that has the minimum coordinate value along the y-axis at the respective probing height z1, z2. A third probing height-specific set of measuring points can be generated, for example, with a third starting point P3, which is a point on the workpiece surface that has the maximum coordinate value along the x-axis at the respective probing height z1, z2.This choice of starting point positions P1, P2, P3 is merely an example. Naturally, other positions relative to the coordinate axes xK, yK can also be chosen. The position of the rotary table axis DA is also shown.

Claims

[1] Method for determining the position of at least one axis of rotation (DA) of a rotary table (2) on which a workpiece (8) is arranged, comprising the steps: a. Determining a preliminary position of the axis of rotation (DA) of the rotary table (2), b. Generating at least three sets of measuring points at a first probing height (z1), wherein a set of measuring points comprises measuring points that are detected by probing the workpiece (8) at a starting point (P1, P2, P3) and during a subsequent rotation of the workpiece (8) by a sensor of a coordinate measuring machine (3), wherein the starting points (P1, P2, P3) for generating the sets of measuring points are different, wherein the coordinates of the measuring points are determined depending on the position of the axis of rotation (DA) of the rotary table (2), c. Determine at least one dimension piece of information for a probing height-specific workpiece profile for each of the three measurement point sets, d. Generating at least three sets of measurement points on a second probing height (z2), wherein the starting points (P1, P2, P3) for generating the sets of measurement points are different, e. Determining at least one dimension information of a probing height-specific workpiece profile for each of the three measurement point sets, f. Determining the position of the axis of rotation (DA) of the rotary table (2) as a function of the probing height-specific size information. [2] Method according to claim 1, characterized by , that the position of the axis of rotation (DA) of the rotary table (2) is calculated as a function of the variation of the probing height-specific size information. [3] Method according to claim 1, characterized by , that the position of the axis of rotation (DA) of the rotary table (2) is determined such that the variation of the probing height-specific size information is zero or less than a predetermined threshold. [4] Method according to any of the preceding claims, characterized by, that at least one correction factor for a probing height-specific preliminary position of the axis of rotation (DA) of the rotary table (2) is determined, wherein the position of the axis of rotation (DA) of the rotary table (2) is then determined as a function of the preliminary position and the at least one correction factor. [5] Method according to any of the preceding claims, characterized by , that an orientation of the axis of rotation (DA) of the rotary table (2) is determined as the orientation of a connecting line between touch-height-specific positions of the axis of rotation (DA) of the rotary table (2). [6] Method according to any of the preceding claims, characterized by that the procedure additionally includes the following steps: g. Determining at least one positional information of a probing height-specific workpiece profile for at least one of the at least three probing height-specific measurement point sets and determining at least one correction factor, h. Determining the position of the workpiece coordinate system depending on the position information and at least one correction factor. [7] Method according to any of the preceding claims, characterized by , that a set of measuring points is determined at a third probing height (z0), whereby the preliminary position of the axis of rotation (DA) of the rotary table (2) and / or the preliminary orientation of the workpiece coordinate system is determined as a function of the measuring points of this set of measuring points. [8] Method according to any of the preceding claims, characterized by , that the workpiece (8) has at least sectionally a circular cross-sectional profile, wherein as size information a diameter of a circular workpiece profile and / or as position information a position of a center point of a circular workpiece profile is determined. [9] Method according to any of the preceding claims, characterized by, that the probing of the workpiece (8) is done tactilely or optically. [10] Measuring system (1) comprising at least one rotary table (2), at least one coordinate measuring machine (3) and at least one evaluation device (4), wherein the measuring system (1) is configured such that a method according to one of claims 1 to 9 can be carried out with the measuring system (1).

Images