Localisation of a calibration object using a touch probe

Abstract

The invention relates to a method for localising a calibration object (1, 1') having a curved surface using a touch probe, the touch probe comprising a touch probe base, a deflectable arm having a shaft (52), and a probing element (53), the probing element (53) having a rest position with respect to the touch probe base, the touch probe having a reference axis (S) which is fixed with respect to the touch probe base and along which the shaft (52) extends when in the rest position, and the touch probe being designed to output a deflection signal as a function of a mechanical deflection of the probing element (53) from the rest position. The method is based on determining at least one apex (SPi) of the curved surface, wherein, at each apex (SPi), the reference axis of the touch probe is oriented parallel to a tangent which extends through the apex (SPi) on the curved surface. The invention also relates to a machine tool comprising a calibration object (1, 1') and a touch probe, the machine tool being designed to carry out said method.

Description

[0001] TITLE

[0002] LOCALIZATION OF A CALIBRATION OBJECT USING A MEASURING PROBE

[0003] TECHNICAL AREA

[0004] The present invention relates to a method for locating a calibration object with a curved surface in a coordinate system using a measuring probe, and to a machine tool comprising a calibration object, a measuring probe and a control system configured to perform the method.

[0005] STATE OF THE ART

[0006] Tactile measuring systems, i.e., those based on physical contact, hereinafter referred to as "probes," are used in a variety of ways in manufacturing technology, for example, to locate and / or measure workpieces in a machine tool. To obtain reliable measurements, probes are often calibrated using a calibration object. Such a calibration object can be positioned in various locations, for example, within the machine tool. Before the probe is calibrated, it may be necessary to locate the calibration object. In this context, the term "localization" means determining the position and / or orientation of the calibration object within an arbitrary, but predefined, coordinate system.

[0007] Typical multidirectional probes feature a deflectable arm with a shaft connected to a probe base. A stylus is located at a free, deflectable end of the arm. When no external forces act on the stylus, it is in a rest position relative to the probe base. In an ideal, error-free state, the shaft in this rest position runs along a reference axis, which forms the probe's axis of symmetry. The stylus can be deflected from its rest position by the object being measured. The probe then outputs a deflection signal, which is ideally proportional to the mechanical deflection of the stylus from its rest position.

[0008] In DE10241087A1, an example of a multidirectional measuring probe with a unidirectional sensor is disclosed: A ring-shaped support bearing is formed in a housing, defining a bearing plane and a central axis of the measuring probe normal to it. A spring is clamped between the housing and a support body, which has a ring-shaped counter bearing and defines a longitudinal axis of the support body through this counter bearing. The spring tends to hold the support body in a rest position in which the counter bearing rests against the support bearing and the longitudinal axis of the support body coincides with the central axis of the measuring probe. A stylus receptacle is centrally arranged on the support body and is designed to receive a stylus.A transmission element, in the form of a straight bending rod (in the rest position of the support body), is centrally attached at one end to the stylus receptacle and guided along the central axis with a section adjacent to its other end. This section converts any deflections of the support body from its rest position into linear movements. A sensor is provided to convert these movements of the transmission element into measurement signals. The transmission element is attached to a point on the support body that, viewed from the sensor, lies beyond the bearing plane.

[0009] WO 2016 / 188842 A2 discloses a further example of a multidirectional measuring probe and its application for scanning workpiece contours in a machine tool. The measuring probe has a support body on which a stylus holder is arranged for receiving a stylus. Furthermore, the measuring probe has an analog sensor that converts movements of the support body into output signals characteristic of the contour of a workpiece surface. The measuring probe has a stylus that is coupled to the support body and comprises a rod-shaped shaft and a probe head attached to the shaft. A processing unit is coupled to the analog sensor to determine at least one measurement characteristic of the contour of the workpiece surface from the output signals.

[0010] The measuring probes described above exhibit anisotropic sensitivity due to their rotational symmetry around the reference axis; that is, the slope of a characteristic curve, which specifies the output voltage per unit of displacement, differs when probing an object in the direction of the reference axis from the slope of the characteristic curve when probing the object perpendicular to the reference axis.

[0011] If a calibration object is to be located using such a measuring probe, this anisotropic sensitivity can affect the position and / or orientation of the calibration object determined by the measuring probe, which can make the localization faulty or inaccurate.

[0012] PRESENTATION OF THE INVENTION

[0013] In a first aspect, it is therefore an object of the present invention to provide a method which enables the localization of a calibration object with a curved surface using a measuring probe, wherein the method is particularly suitable for minimizing errors in the localization of the calibration object which may occur due to anisotropy of the measuring probe.

[0014] The problem is solved by a method according to claim 1. Further embodiments are specified in the dependent claims.

[0015] A method for locating a calibration object with a curved surface in a coordinate system using a probe is described, wherein the probe comprises: a probe base; a deflectable arm, the arm comprising a shaft connected to the probe base; and a stylus element, in particular in the form of a stylus ball, the stylus element being arranged at a free end of the arm, the stylus element having a rest position with respect to the probe base; the probe having a reference axis fixed with respect to the probe base and along which the shaft extends in the rest position; and the probe being configured to output a deflection signal as a function of a mechanical deflection of the stylus element from the rest position, the method comprising:

[0016] Determining at least one vertex of the curved surface of the calibration object, wherein at each vertex the reference axis of the measuring probe is aligned parallel to a tangent which passes through the vertex on the curved surface, and locating the calibration object in the coordinate system based on the at least one determined vertex, wherein determining each vertex comprises:

[0017] Moving the stylus relative to the curved surface along at least one stylus path;

[0018] Capturing the deflection signal of the measuring probe along the stylus path to obtain a profile measurement curve, and

[0019] Determining the vertex in the coordinate system using the profile measurement curve.

[0020] In this context, the term "vertex" refers to a point on the curved surface which, when the probe is at rest in space, with its reference axis parallel to a tangent to the curved surface and the probe element not deflected, has the shortest distance perpendicular to the reference axis. Which point on the curved surface is considered the vertex therefore depends on the position and orientation of the reference axis in space.

[0021] Because the localization is based on at least one vertex, which is determined such that the probe is oriented tangentially to the curved surface during the determination process—that is, the probe's reference axis is aligned parallel to a tangent that passes through the respective vertex on the curved surface—the probe's deflection at that vertex exhibits only components perpendicular to the reference axis and no components parallel to it. This takes into account any anisotropy of the probe caused by rotational symmetry around the reference axis during measurement, thereby minimizing errors in the localization of the calibration object that could arise from this anisotropy. Ideally, if the localization is performed exclusively using such vertices, the accuracy of the calibration object's localization can be increased.

[0022] The procedure may include an initialization, which may comprise the following steps: specifying a number of vertices to be determined, and / or

[0023] Specifying a tolerance threshold, which serves to evaluate a quality parameter, and / or procedures and / or swiveling the measuring probe to a defined basic position, and / or

[0024] Setting the coordinate system, and / or

[0025] Setting one or more settings, in particular a sampling rate and / or a maximum permissible deflection of the stylus element, and / or

[0026] Reading data, in particular parameters for the key element, and / or calculating or loading transformation rules and / or safety points and / or approach points, and / or

[0027] Specifying an approximate location and / or orientation of the calibration object.

[0028] The approximate position and / or orientation of the calibration object may have been determined prior to initialization, particularly via manual or automatic single-point probing using the measuring probe. In single-point probing, the probe element is brought into contact with the curved surface of the calibration object; however, no scan is performed along the curved surface, meaning the probe element is not moved relative to the curved surface along a probing path while maintaining continuous contact. The approximate position and / or orientation of the calibration object may be stored in a data memory and can be read from this memory during initialization.

[0029] Alternatively, determining the approximate location and / or orientation of the calibration object can be part of the initialization.

[0030] Determining the vertex using the profile measurement curve can include: determining an extremum, in particular a maximum, of the profile measurement curve, and / or calculating a measurement adjustment circle using the profile measurement curve, and / or calculating a regression parabola using the profile measurement curve.

[0031] To achieve higher accuracy, the vertex can be determined using several profile measurement curves, whereby the measuring probe is preloaded to a different degree for each profile measurement curve, i.e., the probe element has a different initial mechanical deflection.

[0032] The probe is preferably a multidirectional probe, i.e., a probe in which the sensing element can be deflected in two or, preferably, in all three spatial dimensions. The probe arm can be straight. In such a case, the shaft forms the arm; that is, the terms "arm" and "shaft" can be used synonymously. Alternatively, the probe arm can also be L-shaped or T-shaped and thus comprise several arm sections. In such a case, the term "shaft" refers to a first arm section, which is connected to the probe base and extends along the reference axis. With L-shaped or T-shaped probes, the sensing element can be located at a free end of a second arm section arranged perpendicular to the shaft.

[0033] The reference axis is preferably perpendicular to the probe base; that is, the shank extends preferably perpendicularly from the probe base in its rest position. If the probe is deflected from its rest position, the shank no longer extends along the reference axis but forms a non-zero angle with it. The probe may exhibit rotational symmetry with respect to the reference axis. The probe may, for example, be spherical, hemispherical, ellipsoidal, cylindrical, or disc-shaped.

[0034] The deflection signal of the measuring probe can include a voltage signal, i.e., an output voltage that depends on the deflection of the probe element.

[0035] The measuring probe may include a signal generation unit which has a multidirectional sensor, i.e., which is designed to output a deflection signal that contains information about the direction of the deflection in space.

[0036] Alternatively, the measuring probe can include a signal generation unit which has a unidirectional sensor, i.e., which is designed to output a deflection signal that does not contain any information about the direction of the deflection in space.

[0037] If the measuring probe has a unidirectional sensor, this can, in particular, be a photoelectric sensor. To generate the photoelectric sensor, the signal generation unit can include a light transmitter configured to emit light and a light receiver configured to receive the emitted light. The photoelectric sensor can, in particular, be perpendicular to the reference axis. The signal generation unit can also include a light-blocking element, for example, in the form of a pin, which is guided in the measuring probe base in such a way that it can block the light from the photoelectric sensor depending on a deflection of the probe element from its rest position.

[0038] The probe base can include a spring mechanism by which the shaft is resiliently mounted. In particular, the probe base can include a tilting element, wherein the end of the shaft opposite the free end can be rigidly attached to the tilting element. The spring mechanism can connect the tilting element to the signal generation unit. If a force is exerted on the probe by an object to be detected, the probe can be deflected from its rest position. If the force acts on the probe exactly along the reference axis, a linear displacement of the tilting element within the probe base can occur, and the spring mechanism can be compressed symmetrically with respect to the reference axis. If the force on the probe has a component perpendicular to the reference axis, this can lead to a tilting of the tilting element, causing the spring mechanism to be compressed asymmetrically with respect to the reference axis.

[0039] The light-blocking element can be connected to the tilting element, for example via a joint. The light-blocking element can be guided mechanically in a linear fashion along the reference axis such that any movement of the probe element in three-dimensional space is translated into a linear movement of the light-blocking element along the reference axis. The deflection signal, particularly in the form of an output voltage from the probe, can be generated based on the degree of dimming measured by the light receiver. This deflection signal depends on the magnitude of the mechanical deflection of the probe element but contains no information about its direction. For a given magnitude of mechanical deflection, the deflection signal can depend on whether the spring mechanism is compressed symmetrically or asymmetrically with respect to the reference axis, thus enabling the probe to exhibit anisotropic sensitivity.

[0040] The coordinate system used to determine the vertices can be chosen arbitrarily. In particular, the coordinate system can be that of the measuring probe, especially if the probe has a multidirectional sensor.

[0041] Alternatively, the coordinate system can be a coordinate system of a device used to move the probe in space. In particular, the coordinate system can be a coordinate system of a machine tool (machine tool coordinate system) in which the probe is located. Specifically, the machine tool can include movable components such as slides and / or spindles that define the axes of the machine tool's coordinate system. Alternatively, the coordinate system used to determine the vertices can be arbitrarily shifted relative to the machine tool's coordinate system.

[0042] Alternatively, the coordinate system can be the coordinate system of a workpiece, which may be located in the machine tool in which the measuring probe is also located.

[0043] Preferably, the probe element is moved relative to the curved surface while maintaining continuous contact with it. The probe element moves along the scanning path. The straight line that the probe element would travel in its rest position if the calibration object were not present is referred to in this context as the corresponding "scanning path." Such continuous scanning of the curved surface enables rapid acquisition of the profile measurement curve and thus leads to efficient localization of the calibration object.

[0044] As an alternative to continuous contact, it is conceivable to individually probe a multiple of contact points along the stylus path, i.e., in such a way that no contact occurs between the curved surface and the probe element as long as the probe base is moved parallel to the stylus path. In such a case, the profile measurement curve can be obtained by interpolation between the contact points.

[0045] To eliminate interference signals in the profile measurement curve, which can make it difficult to determine the vertex, the method can also include a smoothing process after obtaining the profile measurement curve, wherein the smoothing process includes:

[0046] Applying a filter to the profile measurement curve before determining the vertex from the profile measurement curve.

[0047] The filter can be applied in the time and / or frequency domain. Alternatively or additionally, the filter can be causal or non-causal. Alternatively or additionally, the filter can be linear or non-linear. In the frequency domain, the filter can, for example, be a bandpass filter centered around a frequency corresponding to a ripple in the scanning path, i.e., chosen so that the desired contour of the curved surface is not distorted.

[0048] Alternatively, the filter can be a low-pass filter, which eliminates interference signals with interference frequencies that are higher than a frequency of the ripple of the sampling path.

[0049] Alternatively, the filter can be a high-pass filter, which eliminates interference signals with frequencies lower than the ripple frequency of the sampling path. If interference signals occur with significantly lower frequencies than the ripple frequency of the sampling path, this can manifest as a tilt in the profile measurement curve. Such a tilt can be corrected, for example, by a filter that performs a slope correction, e.g., using linear regression.

[0050] In order to perform the movement along the probe path in a desired working area of ​​the measuring probe, the determination of each vertex before the movement along the respective probe path can include a surface detection process, wherein the surface detection process includes:

[0051] Positioning the measuring probe on a starting point where the probe element is in its rest position;

[0052] Calculating a lead point in the coordinate system, wherein the lead point is chosen such that the probe element comes into contact with the calibration object when the probe is moved in the direction of the lead point;

[0053] Moving the measuring probe perpendicular to the reference axis in the direction of the lead point;

[0054] Capturing a trigger measurement point at which the probe element is deflected such that the deflection signal corresponds to a predetermined trigger value, where the trigger measurement point is located between the starting point and the lead point.

[0055] The probe can have an analog signal output and / or a digital signal output. If the probe has a digital signal output, the trigger measurement point can be detected, in particular, via the digital signal output. The digital signal output can be configured such that the output signal has a continuous character when probed accordingly. The deflection signal can be linearly proportional to the deflection of the probe element, at least over a certain operating range. To verify this and / or to determine a dependence of the deflection signal on the deflection of the probe element, the determination of each vertex after the surface detection process can also include a characteristic curve acquisition process, wherein the characteristic curve acquisition process comprises: positioning the probe on the starting point;

[0056] Calculating a limit point based on the detected trigger measurement point, wherein the probe element has a maximum permissible deflection when the probe is at the limit point, the limit point lying on a calibration line defined by the starting point and the trigger measurement point;

[0057] Moving the measuring probe from the starting point to the limit point along the calibration line, whereby the deflection signal of the measuring probe is recorded along the calibration line in order to obtain a characteristic curve.

[0058] The characteristic curve acquisition process can take place before or after obtaining the profile measurement curve.

[0059] The calibration object can have a singly curved or a doubly curved surface. The calibration object can be, for example, a calibration sphere, a physical standard, an etalon, a measurement standard, a calibration standard, or a gear standard.

[0060] The number of vertices identified may depend on the calibration object.

[0061] Examples of localizing the calibration object using a vertex: Determining a starting position on a curved surface, for example a glass hemisphere as a calibration object for calibrating and / or checking the measuring probe.

[0062] Determining the highest point of a convex tooth flank as a calibration object

[0063] Examples of locating the calibration object using two or more vertices:

[0064] Localization of a circle on a cross-section of a cylinder, cone, or sphere as a calibration object by determining a circle center point. Examples of localization of the calibration object using three or more vertices:

[0065] Localization of a sphere as a calibration object

[0066] Examples of locating the calibration object using four or more vertices:

[0067] Localization of a cylinder as a calibration object by determining the position and orientation of an axis of the cylinder.

[0068] Examples of locating the calibration object using five or more vertices:

[0069] Localization of a cone as a calibration object

[0070] Examples of locating the calibration object using six or more vertices:

[0071] Localization of a torus as a calibration object

[0072] If the calibration object has a doubly curved surface, determining each vertex can include:

[0073] Moving the probe element relative to the doubly curved surface along a first probe path and capturing the deflection signal of the probe along the first probe path to obtain a first profile measurement curve;

[0074] Moving the probe element relative to the doubly curved surface along a second probe path and capturing the deflection signal of the probe along the second probe path to obtain a second profile measurement curve, wherein the second probe path is chosen such that the second probe path intersects the first probe path at right angles;

[0075] Determining a first pre-vertex in the coordinate system using the first profile measurement curve either before or after moving the probe element along the second probe path;

[0076] Determining a second pre-vertex in the coordinate system using the second profile measurement curve, and

[0077] Determining the vertex in the coordinate system using the first and second pre-vertex points. In particular, the second probe path can be chosen such that it intersects the first probe path at a right angle at the first pre-vertex point, making the second pre-vertex point the vertex.

[0078] In the present context, the term "pre-vertex" refers to a point on the doubly curved surface which, in a plane containing the first probe path and the reference axis, has the shortest distance perpendicular to the reference axis when the probe is at rest in space and the probe element is not deflected.

[0079] To achieve higher accuracy, the pre-vertex can be determined using several profile measurement curves, with the probe being pre-tensioned to a different degree for each profile measurement curve, i.e., the probe element exhibiting a different initial mechanical deflection. In particular, the probe element can be moved along the first and / or second sensing path two or more times, with the sensing path being shifted parallel to itself each time, such that the probe element is deflected to a different degree each time.

[0080] As an example of a calibration object with a doubly curved surface, the calibration object can be a calibration sphere which has a spherical center. In such a case, preferably at least three vertices of the calibration sphere are determined, and the localization of the calibration object preferably includes:

[0081] Determining the center of the sphere in the coordinate system based on the at least three determined vertices.

[0082] To increase the robustness of the method and / or to allow conclusions to be drawn about the quality of the result, at least four vertices of the calibration sphere can be determined, in particular, including determining the center of the sphere:

[0083] Performing a regression analysis to determine a regression plane for the at least four vertices;

[0084] Projecting the at least four vertices onto the regression plane, and

[0085] Calculating a projection adjustment circle based on the four projected vertices.

[0086] In the case of a calibration sphere as the calibration object, the first probe path or the second probe path can correspond to a great circle of the calibration sphere, where the great circle defines a great circle plane which is oriented perpendicular to the reference axis.

[0087] One advantage of locating a calibration sphere by determining vertices as described above is that the precisely calibrated diameter of the calibration sphere does not need to be known.

[0088] Alternatively, the calibration object can have a cylindrical part which is substantially rotationally symmetrical with respect to a central axis, in particular where the calibration object is a gear standard. In such a case, the method can include:

[0089] Determining at least one upper vertex pair and one lower vertex pair on a lateral surface of the cylindrical part, wherein the vertex pairs each consist of a first vertex and a second vertex diametrically opposite each other with respect to the central axis, and wherein the upper vertex pair is offset relative to the lower vertex pair along the central axis, and wherein determining the position of the calibration object comprises:

[0090] Determining the position of the central axis based on at least two pairs of vertices.

[0091] The calibration object can be mounted in such a way as to be rotatable about an axis of rotation, and determining the vertices can include:

[0092] Determining the upper vertex pair and the lower vertex pair in a first rotational angular position of the calibration object;

[0093] Rotate the calibration object 180° around the axis of rotation to bring the calibration object into a second rotational angular position;

[0094] Determining the upper vertex pair and the lower vertex pair in the second rotational angular position of the calibration object; wherein the central axis is defined as a straight line through an effective upper center and an effective lower center, and wherein determining the position of the central axis comprises:

[0095] Determining the effective upper center point and the effective lower center point based on the determined vertex pairs in the first and second rotational angular positions. Determining the effective upper center point and the effective lower center point may include:

[0096] Calculating a first upper midpoint, which is located midway between the first vertex and the second vertex of the upper pair of vertices determined in the first rotation angle position;

[0097] Calculating a second upper center point, which is located midway between the first vertex and the second vertex of the upper pair of vertices determined in the second rotation angle position;

[0098] Calculating the effective upper midpoint, which is located midway between the first upper midpoint and the second upper midpoint;

[0099] Calculating a first lower midpoint located between the first vertex and the second vertex of the lower pair of vertices determined in the first rotation angle position;

[0100] Calculating a second lower midpoint located between the first vertex and the second vertex of the lower vertex pair determined in the second rotational angular position;

[0101] Calculating the effective lower midpoint that lies midway between the first lower midpoint and the second lower midpoint.

[0102] Determining the location of the central axis can include:

[0103] Calculating at least one positional deviation of the central axis, in particular a positional deviation and / or an orientational deviation, from a given ideal axis, wherein the given ideal axis preferably coincides with the axis of rotation.

[0104] The procedure for locating the calibration object can be part of a probe calibration process, which may include:

[0105] Mount the calibration object in an area accessible to the measuring probe;

[0106] Locate the calibration object according to the procedure described above;

[0107] Measuring the calibration object;

[0108] Calculation of touch deviations and optionally of corresponding correction fields.

[0109] Alternatively, the method for locating the calibration object can be part of a method for commissioning a machine tool with moving components, wherein the method for commissioning the machine tool may include:

[0110] Mounting the calibration object on a moving component, in particular a spindle;

[0111] Locating the calibration object according to the procedure described above, in particular determining a positional and / or orientation deviation of an axis of the calibration object;

[0112] Determining geometric deviations, in particular

[0113] Parallelism deviations of the moving component.

[0114] In a second aspect, it is an object of the present invention to provide a machine tool with a calibration object having a curved surface and a measuring probe, wherein the machine tool is designed to be able to locate the calibration object, in particular in such a way that errors in the localization which occur due to anisotropy of the measuring probe are minimized.

[0115] A machine tool, especially for machining pre-toothed workpieces, is specified, wherein the machine tool comprises: a calibration object with a curved surface;a measuring probe, wherein the measuring probe comprises: a measuring probe base, a deflectable arm, wherein the arm comprises a shaft which is connected to the measuring probe base, and a sensing element, in particular in the form of a sensing ball, wherein the sensing element is arranged at a free end of the arm, wherein the sensing element has a rest position with respect to the measuring probe base, wherein the measuring probe has a reference axis which is fixed with respect to the measuring probe base and along which the shaft extends in the rest position, and wherein the measuring probe is configured to output a deflection signal as a function of a mechanical deflection of the sensing element from the rest position, and a control which is configured to perform the above-described method for locating the calibration object.

[0116] The calibration object can be installed in the machine tool in a fixed or removable manner.

[0117] The machine tool can, in particular, comprise: a tool holder and a workpiece carrier, wherein the tool holder and the workpiece carrier are displaceable relative to each other, wherein the measuring probe is arranged on the tool holder and wherein the calibration object is arranged on the workpiece carrier. In particular, the measuring probe can be attached to the tool holder by means of a pivotable measuring probe holder.

[0118] The workpiece carrier can be designed in particular as a rotating tower.

[0119] The machine tool may also include: a spindle designed to hold a tool or workpiece and drive it to rotate around a spindle axis, with the calibration object optionally designed to be held in the spindle and rotated around the spindle axis.

[0120] The spindle can in particular be a tool spindle, which is designed to hold a tool and drive it to rotate around a tool spindle axis.

[0121] Alternatively, the spindle can be a workpiece spindle which is arranged on the workpiece carrier and which is designed to hold a workpiece, in particular a pre-toothed workpiece, and to drive it to a rotation around a workpiece spindle axis.

[0122] Alternatively, all linearly moving components of the machine tool can be arranged either only on the workpiece side, i.e., between the machine bed and a workpiece holder, or only on the tool side, i.e., between the machine bed and a tool holder. If rotation of the calibration object is not required, the calibration object can be mounted on the machine bed, and the probe can be positioned on one of the machine tool components that moves relative to the machine bed.

[0123] The control system can be a machine control system, which is also designed to control one or more moving components of the machine tool, such as slides or spindles arranged on the workpiece carrier and / or on the tool carrier. The machine tool can have serial and / or parallel kinematics.

[0124] Alternatively, the control can be a touch probe control separately from a machine control.

[0125] In particular, the machine tool could be a gear grinding machine, in which the tool is preferably a grinding worm. Alternatively, the machine tool could be a peeling machine.

[0126] BRIEF DESCRIPTION OF THE DRAWINGS

[0127] Preferred embodiments of the invention are described below with reference to the drawings, which serve only for illustration and are not to be interpreted restrictively. The drawings show:

[0128] Fig. 1A schematically shows an example of a measuring probe in its rest position;

[0129] Fig. 1B schematically shows a calibration object and the measuring probe from Fig. 1A in a deflected state;

[0130] Fig. 2A schematically shows a general deflection signal of the measuring probe as a function of a travel distance;

[0131] Fig. 2B schematically shows a subdivision of the deflection of the measuring probe into angular ranges;

[0132] Fig. 3 shows a schematic perspective view of a first embodiment of a machine tool according to the second aspect of the present invention;

[0133] Fig. 4 shows a simplified top view of a second embodiment of a machine tool according to the second aspect of the present invention;

[0134] Fig. 5 is a simplified top view of a third embodiment of a

[0135] machine tool according to the second aspect of the present invention;

[0136] Fig. 6 is a simplified top view of a fourth embodiment of a

[0137] machine tool according to the second aspect of the present invention;

[0138] Fig. 7 is a simplified top view of a fifth embodiment of a

[0139] machine tool according to the second aspect of the present invention;

[0140] Fig. 8 shows a simplified top view of a sixth embodiment of a

[0141] machine tool according to the second aspect of the present invention;

[0142] Fig. 9 shows a simplified top view of a seventh embodiment of a

[0143] machine tool according to the second aspect of the present invention;

[0144] Fig. 10 shows a simplified top view of an eighth embodiment of a

[0145] machine tool according to the second aspect of the present invention;

[0146] Fig. 11 is a simplified top view of a ninth embodiment of a

[0147] machine tool according to the second aspect of the present invention;

[0148] Fig. 12 shows an embodiment of a method for locating a

[0149] Calibration object according to the first aspect of the present invention using a schematic flowchart;

[0150] Fig. 13 schematically shows the position of the measuring probe at four vertices of a calibration sphere during an embodiment of the method for locating the calibration sphere according to the first aspect of the present invention;

[0151] Fig. 14 shows an enlarged section of the calibration sphere from Fig. 13 and illustrates the travel paths of the measuring probe during a surface detection process, a characteristic curve acquisition process, and the acquisition of the profile measurement curves to determine a vertex;

[0152] Fig. 15 shows an embodiment of determining a vertex with determining a pre-vertex;

[0153] Fig. 16 shows an embodiment of a surface detection process in the

[0154] Determining a pre-vertex;

[0155] Fig. 17 shows an embodiment of a characteristic curve acquisition process in the

[0156] Determining a pre-vertex;

[0157] Fig. 18 shows a schematic top view of the calibration sphere from Fig. 13 and the

[0158] Measuring probe during the surface detection process from Fig. 16;

[0159] Fig. 19 schematically shows a deflection signal as a function of a travel path along the calibration line shown in Fig. 18;

[0160] Fig. 20 shows an embodiment of a profile acquisition system for determining a

[0161] Vertex;

[0162] Fig. 21 shows a schematic top view of the calibration sphere from Fig. 13 and the

[0163] Measuring probe during profile acquisition from Fig. 20;

[0164] Fig. 22 schematically shows a profile measurement curve recorded along the probe section shown in Fig. 21;

[0165] Fig. 23 shows an embodiment of a profile evaluation for determining the pre-vertex based on the profile measurement curve shown in Fig. 22;

[0166] Fig. 24 shows an embodiment of a surface detection process for determining a vertex;

[0167] Fig. 25 shows an embodiment of a characteristic curve acquisition process for determining a vertex;

[0168] Fig. 26 schematically shows a top view of the calibration sphere from Fig. 13 and the measuring probe during the surface detection process from Fig. 24;

[0169] Fig. 27 schematically shows a deflection signal as a function of a travel path along the calibration line shown in Fig. 26;

[0170] Fig. 28 shows an embodiment of a profile acquisition system for determining a vertex;

[0171] Fig. 29 schematically shows a top view of the calibration sphere from Fig. 13 and the measuring probe during profile acquisition from Fig. 28; Fig. 30 schematically shows a profile measurement curve acquired along the probe path shown in Fig. 29;

[0172] Fig. 31 shows an embodiment of a profile evaluation for determining the vertex based on the profile measurement curve shown in Fig. 30;

[0173] Fig. 32 schematically shows a profile measurement curve before the application of a filter;

[0174] Fig. 33 schematically shows a profile measurement curve before and after the application of a filter;

[0175] Fig. 34 shows an embodiment for determining the center of a sphere from the determined vertices;

[0176] Fig. 35 shows a schematic side view of a cylindrical calibration object in an initial rotational angular position;

[0177] Fig. 36 shows a schematic top view of the cylindrical calibration object from Fig. 35 together with a rotary table and a probe element with shaft;

[0178] Fig. 37 schematically shows the cylindrical calibration object of Fig. 35 in a first rotational angular position relative to the initial rotational angular position, and

[0179] Fig. 38 schematically shows the cylindrical calibration object of Fig. 35 of a second rotational angular position relative to the initial rotational angular position.

[0180] DESCRIPTION OF PREFERRED EXECUTION FORMS

[0181] Operating principle of the measuring probe

[0182] Figures 1A and 1B show a highly schematic example of a measuring probe 50. Such measuring probes are known in the prior art. For a detailed exemplary description, reference is made to document DE10241087A1 and to the description of Figures 2 and 3 in document WO 2016 / 188842 A2.

[0183] The probe has a sensing element, in the example shown in Fig. 1A and Fig. 1B, a sensing ball 53, which is arranged at a free end of a shaft 52. The shaft 52 is connected to the probe base 51 at an end opposite the free end; in particular, the end of the shaft 52 opposite the free end is rigidly attached to a tilting element 59. The probe 50 has a probe base 51 which includes a signal generation unit 510. The sensing element 53 has a rest position with respect to the probe base 51. In this rest position, the shaft 52 extends along a reference axis S. The shaft 52 is resiliently mounted in the probe base 51 and extends away from the probe base 51 such that the sensing element 53 can be deflected from the rest position in all three spatial directions. Figures 1A and 1B schematically depict a spring mechanism 58.The spring mechanism 58 can be designed in various ways, particularly analogous to the embodiments mentioned in DE10241087A1, but is preferably rotationally symmetric with respect to the reference axis S. The spring mechanism 58 connects the tilting element 59 to the signal generation unit 510. If a force is exerted on the probe element 53 by an object to be detected, the probe element 53 is deflected from its rest position. If the force on the probe element 53 acts exactly along the reference axis S, a linear displacement of the tilting element 59 occurs within the probe base 51, and the spring mechanism 58 is compressed symmetrically with respect to the reference axis S. If the force on the probe element 53 has a component perpendicular to the reference axis S, this leads to a tilting of the tilting element 59, causing the spring mechanism 58 to be compressed asymmetrically with respect to the reference axis S.

[0184] In the example shown in Fig. 1A and Fig. 1B, the signal generation unit 510 comprises a unidirectional sensor in the form of a photoelectric sensor 56. The photoelectric sensor 56 is generated by light being emitted from a light transmitter 54 and received by a light receiver 55. In this example, the photoelectric sensor 56 is perpendicular to the reference axis S. A light-blocking element 57, here in the form of a pin, is guided in the probe base 51 such that it can block the light from the photoelectric sensor 56 depending on the deflection of the probe element 53 from its rest position. For this purpose, the light-blocking element 57 is connected to the tilting element, for example via a joint 571. In this example, the light-blocking element 57 is guided mechanically in a linear direction along the reference axis S, so that any movement of the probe element 53 in space is translated into a linear movement of the light-blocking element 57 along the reference axis S.A deflection signal, particularly in the form of an output voltage of the light receiver, can be generated based on the degree of dimming measured by the light receiver 55. This deflection signal depends on the magnitude of the mechanical deflection of the sensor element 53 but contains no information about the direction of the mechanical deflection. The less light reaching the light receiver, the greater the deflection magnitude. Due to the rotational symmetry of the spring mechanism, a deflection of the sensor element parallel to the reference axis S results in a different displacement of the light-blocking element 57 for the same deflection magnitude than a deflection perpendicular to the reference axis S. Therefore, such measuring sensors exhibit anisotropic sensitivity; that is, a deflection in the direction of the reference axis S results in a different deflection signal magnitude than a deflection perpendicular to the reference axis S.If, as intended, the mechanical deflection amount is deduced from the direction-independent deflection signal without directional information, this can lead to corresponding errors.

[0185] Signal output of the measuring probe

[0186] Figures 2A and 2B schematically illustrate the signal output of a measuring probe, in particular a measuring probe 50 according to Figures 1A and 1B. Figure 2A shows a deflection signal U as a function of a travel distance g, where the travel distance g is perpendicular to the reference axis S. The deflection signal can be divided into five sections:

[0187] Section I: Non-contact area ( o, s sqso, e )

[0188] Section II: Contact area ( o, e < g < m eas,s)

[0189] Section III: Measuring range ( mea s,ssg < g meas , e )

[0190] Section IV: Transitional Area ( me as,esqsg op , e )

[0191] Section V: Crossover area (deflection signal saturated) (q o t, s sqsg o t, e ) Operating range for voltage output (includes measuring range): (g op , s < g < g op , e )

[0192] In Fig. 2A, Uo denotes a rest position signal, U me as,s the starting value of the measuring range, Umeas.e the ending value of the measuring range, Ut a trigger value which corresponds to a trigger point g t corresponds to U m a limit value which corresponds to a maximum deflection signal.

[0193] Fig. 2B shows a subdivision of the deflection into angular ranges. In Fig. 2B, (p) is denoted. op an angular range for operation, while (p ot denotes an angle range for a crossing.

[0194] Exemplary machine tool setup

[0195] Figure 3 shows an example of a machine tool, a gear grinding machine 10, which will also be referred to as "machine" in the following. The machine has a machine bed 11 on which a tool carrier 12 is slidably guided along a radial feed direction X. The tool carrier 12 carries an axial slide 13, which is slidably guided along a feed direction Z relative to the tool carrier 12. A grinding head is mounted on the axial slide 13, which can be pivoted about a pivot axis (the so-called A-axis) parallel to the X-direction to adapt to the helix angle of the gear teeth to be machined. The grinding head, in turn, carries a shift slide 14 on which a tool spindle 15 can be slid along a shift direction Y relative to the grinding head.The tool spindle 15 comprises a drive motor housing 151, in which a drive motor is arranged, and a tool spindle shaft, which is rotatable about the tool spindle axis B within the drive motor housing 151 and is driven by the drive motor. A machining tool in the form of a helically profiled grinding wheel (grinding worm) 16 is mounted on the tool spindle 15. The grinding worm 16 is driven by the drive motor to rotate about a tool spindle axis B. The tool spindle 15 with the mounted grinding worm 16 is displaceable along a shift direction Y relative to the tool holder 12.

[0196] To allow adjustment to the helix angle of the gear being machined, the shift slide 14, and thus the tool spindle 15 mounted on it, can be pivoted about a swivel axis (the so-called A-axis) running parallel to the X-axis. This is achieved using a swivel body that can be pivoted about the A-axis relative to the axial slide 13, and on which the shift slide 14 is linearly displaceable.

[0197] The machine bed 11 further supports a workpiece carrier 20, which can be designed as a rotary turret that can be pivoted about a rotary axis C3 between two or more positions. Two identical workpiece spindles 21, 22 are mounted diametrically opposite each other on the workpiece carrier 20, of which only one workpiece spindle 21 with its associated tailstock is visible in Fig. 3. A workpiece can be clamped onto each of the workpiece spindles 21, 22 and driven to rotate about a workpiece axis C1 or C2, respectively. The workpiece spindle 21 visible in Fig. 3 is in a machining position in which the workpiece 23 can be machined with the grinding screw 16. To clamp an unmachined workpiece (blank), preferably by an automatic workpiece changer (not shown), the workpiece carrier 20 can be pivoted about the rotary axis C3 into a workpiece change position.The other workpiece spindle, offset by 180° and not visible in Fig. 3, is in a workpiece change position, in which a finished workpiece can be removed from this spindle and a new blank can be clamped on. A dressing unit 30 is mounted offset by 90° to the workpiece spindles. In Fig. 3, the probe 50 is attached to a pivotable probe holder. The probe 50 is preferably a probe as described in Figures 1 and 2. The probe holder is attached to the drive motor housing 151 of the grinding screw 16. The probe 50 can thus be moved along the respective machine axes in the feed direction X, in the shift direction Y, and in the axial direction Z. The calibration object 1, here a calibration sphere, is arranged on the workpiece carrier 20 in the area of ​​the dressing unit 30 and can be rotated with the workpiece carrier 20 about the axis of rotation C3.

[0198] Machine 10 has a large number of moving components, such as slides or spindles, which are controlled by corresponding drives. These drives are often referred to in the industry as "NC axes," "machine axes," or simply "axes." In some cases, this term also includes the components driven by the drives, such as slides or spindles.

[0199] All driven moving components of machine 10 are digitally controlled by a machine control unit 40. The machine control unit 40 comprises several axis modules 41, a control computer 42, and an operator panel 43. The control computer 42 receives operator commands from the operator panel 43 as well as sensor signals from various sensors of machine 10 and calculates control commands for the axis modules 41 from these. It also outputs operating parameters to the operator panel 43 for display. The axis modules 41 provide control signals for one machine axis each at their outputs.

[0200] A monitoring device 44 is connected to the control computer 42.

[0201] The monitoring device 44 can be a separate hardware unit assigned to the machine 10. It can be connected to the control computer 42 via a known interface, e.g., via the known Profinet standard, or via a network, e.g., via the Internet. It can be physically part of the machine 10, or it can be located physically separate from the machine 10.

[0202] During machine operation, the monitoring device 44 receives a variety of different measurement data from the control computer 42. This measurement data can include sensor data acquired directly by the control computer 42 and data read by the control computer 42 from the axis modules 41, such as data describing the target positions of the various machine axes and the target current consumption in the axis modules. In particular, the deflection signal of the measuring probe 50 can be transmitted via the axis modules 41 and read from them by the control computer 42.

[0203] The monitoring device 44 can optionally have its own analog and / or digital sensor inputs to directly receive sensor data as measurement data. In particular, this sensor data can be the deflection signal of the measuring probe 50, if it is not routed via the axis modules 41.

[0204] The monitoring device 44 can alternatively be implemented as a software component of the machine control 40, which is executed, for example, on a processor of the control computer 42, or it can be designed as a software component of a server (not shown).

[0205] Arrangement of the calibration object and the measuring probe in the machine tool

[0206] Figures 4-11 schematically show different embodiments of a machine tool 10 in which a calibration object 1 and a measuring probe 50 are arranged in different ways, wherein the measuring probe 50 is preferably a measuring probe as described in Figures 1 and 2, and the machine tool 10 is a gear grinding machine as described in Fig. 3.

[0207] The calibration object 1 shown in Figures 4-11 as a calibration sphere, but can also have other shapes.

[0208] In Figures 4-11, the measuring probe 50 is attached to a pivotable probe holder, which allows the probe 50 to be pivoted from a parked position (shown with dashed lines) to a measuring position (shown with solid lines). The probe holder is attached to the drive motor housing 151 of the grinding screw 16. In the examples shown here, the reference axis S of the measuring probe 50 is arranged in the measuring position with respect to the feed direction X at a mounting angle α of 20°. However, other angles for the mounting angle α in the range of 0° to 90° are also conceivable. In Figure 4, the calibration sphere 1 is located next to the dressing device 30, with the rotatable workpiece carrier 20, on which the dressing device 30 is arranged, being in a dressing position in which the grinding screw 16 can be dressed by the dressing device 30.

[0209] In Fig. 5, the calibration sphere 1 is located next to the dressing device 30, with the rotatable workpiece carrier 20, on which the dressing device 30 is arranged, being in a measuring position which differs from the dressing position in which the grinding screw 16 can be dressed by the dressing device 30.

[0210] In Fig. 6, the calibration sphere 1 is located in the area of ​​the first workpiece spindle 21, with the rotatable workpiece carrier 20, on which the first workpiece spindle 21 is arranged, being in a first grinding position in which the grinding screw 16 could machine a workpiece clamped on the first workpiece spindle 21. The calibration sphere 1 can be arranged on a clamped workpiece, directly on the first workpiece spindle 21, or on a device that is attached to or next to the workpiece spindle.

[0211] In Fig. 7, the calibration sphere 1 is located in the area of ​​the first workpiece spindle 21. The rotatable workpiece carrier 20, on which the first workpiece spindle 21 is arranged, is in an additional measuring position, which differs from the first grinding position in which the grinding screw 16 could machine a workpiece clamped on the first workpiece spindle 21. The calibration sphere 1 can be arranged on a clamped workpiece, directly on the first workpiece spindle 21, or on a device that is attached to or next to the first workpiece spindle 21.

[0212] In Fig. 8, the calibration sphere 1 is located in the area of ​​the second workpiece spindle 22, with the rotatable workpiece carrier 20, on which the second workpiece spindle 22 is arranged, being in a second grinding position in which the grinding screw 16 could machine a workpiece clamped on the second workpiece spindle 22. The calibration sphere 1 can be arranged on a clamped workpiece, directly on the second workpiece spindle 22, or on a device attached to or next to the second workpiece spindle 22. In Fig. 9, the calibration sphere 1 is located in the area of ​​the second workpiece spindle 22, with the rotatable workpiece carrier 20, on which the second workpiece spindle 22 is arranged, being in an additional measuring position, which differs from the first grinding position in which the grinding screw 16 could machine a workpiece clamped on the second workpiece spindle 22.The calibration sphere 1 can be arranged on a clamped workpiece, directly on the second workpiece spindle 22, or on a device which is attached to or next to the second workpiece spindle 22.

[0213] In Fig. 10, the calibration sphere 1 is located in an area of ​​the tool carrier 20 which is diametrically opposite the dressing device 30, wherein the rotatable workpiece carrier 20, on which the calibration sphere 1 is arranged, is in a position which is pivoted by 180° relative to the dressing position shown in Fig. 4.

[0214] In Fig. 11, the calibration sphere 1 is located in an area of ​​the tool carrier 20 which is diametrically opposite the dressing device 30, wherein the rotatable workpiece carrier 20, on which the calibration sphere 1 is arranged, is in a position which differs from the position shown in Fig. 10.

[0215] Method for locating a calibration object

[0216] Fig. 12 shows an embodiment of a method for locating a calibration object according to the present invention, illustrated by a schematic flowchart. After starting, an initialization 100 can be performed, which may comprise the following steps:

[0217] Specifying a number N of vertices SPj to be determined, and / or

[0218] Specifying a tolerance threshold E, which serves to evaluate a quality parameter, and / or procedure and / or swiveling the measuring probe to a defined basic position, and / or

[0219] Setting the coordinate system, and / or

[0220] Setting one or more settings, in particular a sampling rate and / or a maximum permissible deflection of the stylus element, and / or

[0221] Reading data, in particular parameters for the key element, and / or calculating or loading transformation rules and / or safety points and / or approach points, and / or

[0222] Specifying an approximate location and / or orientation of the calibration object.

[0223] The approximate position and / or orientation of the calibration object may have been determined beforehand, in particular by manual or automatic single-point probing using the measuring probe. The approximate position and / or orientation of the calibration object may be stored in a data memory, for example in the control computer 42 or in the monitoring device 44, and read from this data memory during initialization 100.

[0224] In a second step (200), an i-th vertex SP; of the curved surface of the calibration object 1 is determined. This second step (200) is repeated until the specified number N of vertices SP; has been determined. Subsequently, in a third step (300), the calibration object is located based on the N determined vertices. The number N can be greater than or equal to 1. To evaluate the quality of the localization, a quality parameter ö can be calculated, which represents a normalized vectorial difference of the deviations between two iteration stages. If the quality parameter ö is below the specified tolerance threshold E, the procedure is terminated; otherwise, steps 200 and 300 are performed again.

[0225] Example: Calibration sphere

[0226] Figures 13-34 illustrate an embodiment of a method for locating a calibration object according to the present invention, wherein the calibration object is a calibration sphere 1. To locate the calibration sphere 1, the exact position of the sphere's center M is determined. The spherical surface of the calibration sphere 1 is an example of a doubly curved surface.

[0227] The approximate position of the sphere's center M may have been determined beforehand. For example, the approximate position of the sphere's center M may be known with an accuracy in the tenths of a millimeter range. If the calibration sphere 1 is permanently installed in the machine tool, for example on the workpiece carrier, the sphere's center M can be measured after installation during commissioning or servicing, e.g., by single-point probing with a measuring probe, manually with a dial gauge and / or a gauge block, and stored in a data memory. During initialization 100, the approximate sphere's center M can be read from the data memory. If the calibration sphere 1 is removable and not permanently installed in the machine tool, initialization 100 may involve determining the approximate position of the sphere's center M by single-point probing with a measuring probe.

[0228] In one embodiment, the calibration sphere 1 can be arranged on a device that can be mounted on the workpiece spindle of a machine tool. The device can comprise several reference surfaces. To determine the approximate position of the sphere's center M, the reference surfaces can be probed and a relative position to the reference axis of the probe determined. The orientation can then be adjusted by rotating the device about the workpiece spindle axis until the device and the probe's reference axis are oriented relative to each other as desired. Using the known geometry of the device and machine information regarding the position of the machine tool's moving components, an approximate position of the sphere's center M can be determined in this way.

[0229] Fig. 13 shows the calibration sphere 1 in a coordinate system K' of the measuring probe, wherein the coordinate system K' of the measuring probe is tilted relative to a coordinate system K of the machine tool, in particular pivoted about the Z-axis. The method according to the invention can include determining the center point M of the sphere in the coordinate system K' of the measuring probe and / or the coordinate system K of the machine tool. It is also conceivable that the coordinate system K' of the measuring probe and the coordinate system K of the machine tool are identical. In the example described below, N=4 vertices SPi, SP2, SP3, SP4 are determined, which lie on a great circle of the calibration sphere 1. In the coordinate system K' of the measuring probe, the reference axis S of the measuring probe runs parallel to the X' direction. The great circle is chosen such that it lies in the Y'-Z' plane.Thus, the reference axis S of the probe is aligned parallel to a tangent at each vertex SP1, SP2, SP3, SP4, which runs along the curved surface through the respective vertex SP1, SP2, SP3, SP4. This means that the probe element 53 is deflected only perpendicular to the reference axis S to determine the vertices SP1, SP2, SP3, SP4. This prevents position determination errors that would occur due to the anisotropy of the probe if deflection components of the probe element 53 in the direction of the reference axis S were used for position determination. Fig. 14 illustrates an example of the movements of the probe 50 that can be performed to determine a vertex SP; using an enlarged section of the calibration sphere 1 from Fig. 13 around one of the vertices SPj.In this example, the sphere's surface is scanned with the probe element, here a probe sphere 53, along two probe paths to determine the vertex SPj, with the two probe paths forming a right-angled cross. The detailed procedure is illustrated in Figures 15-30.

[0230] Fig. 15 illustrates substeps of step 200 from Fig. 12. In the first part of step 200, a so-called pre-vertex SPj' is determined. In this example, this pre-vertex SPj' corresponds to the maximum deflection of the probe element 53 along a first probe path, which is traversed parallel to the reference axis S. In the second part of step 200, the vertex SPj itself is then determined. The vertex SPj then corresponds to the maximum deflection of the probe element 53 along a second probe path, which is traversed perpendicular to the reference axis S. The sequence in which the probe paths are traversed parallel and perpendicular to the reference axis S can, of course, be reversed. In this example, a radial surface detection process 20T, 201 is performed both when determining the pre-vertex SPj' and when determining the vertex SPj.In radial surface detection, the probe element 53 is brought into contact with the calibration sphere 1, thus positioning the calibration sphere 1 within the working range of the probe 50. Following the surface detection process 20T, 201, a characteristic curve acquisition step 202', 202 is preferably performed to obtain a characteristic curve that includes the deflection signal, preferably an output voltage, as a function of the deflection magnitude. However, the characteristic curve acquisition step 202', 202 can also be omitted or performed at a different point in the process. In a further step, the tangential profile acquisition 203', 203 of the sphere's surface is carried out. For this purpose, the probe 50 is moved along the first or second sensing path, whereby the probe element 53 is deflected along the respective sensing path and a corresponding profile measurement curve is acquired.This profile measurement curve is then evaluated in a profile evaluation step 204', 204 to determine the pre-vertex SPj' and the vertex SPj, respectively. The surface detection process 201, 201 makes it possible to select the probe path such that the deflection of the probe element 53 during profile acquisition remains within the probe's working range, thus ensuring a meaningful deflection signal is captured. Exemplary substeps of the surface detection process 20T, 201 are described in more detail as flowcharts in Figures 16 and 24, and schematically illustrated in Figures 18 and 26.

[0231] The following refers to Figures 16, 18, and 19. In the surface detection process 20T prior to profile acquisition for the vertex SPj', the probe 50 is moved to a starting position in which the center of the stylus ball 53 is located at a starting point Pi in the coordinate system K' of the probe 50. In the starting position, the stylus ball 53 is ideally already close to the surface of the sphere, but preferably not yet touching it. Thus, the stylus ball 53 is not deflected, i.e., it is in its rest position and therefore outputs a rest position signal, preferably a rest position voltage Uo (see Fig. 19). The rest position voltage Uo can be zero or have a constant value other than zero. The probe 50 can be moved to the starting position automatically or manually. Subsequently, a lead point P2, p , calculated, where this lead point P2, pis chosen such that the probe element 53 comes into contact with the sphere surface when the measuring probe 50 moves in the direction of the lead point P2, p is shifted and thus a deflection signal U2, P outputs a signal that differs from the rest position signal Uo. The measuring probe 50 is now moved perpendicular to the reference axis S in the direction of the lead point P2. p shifted. The measuring probe 50 can, but does not have to, extend to the lead point P2. p be moved. As soon as the key element 53 is deflected such that the deflection signal corresponds to a predefined trigger value Ut, the corresponding position is designated as trigger measurement point P2. t This trigger measurement point P2, t is located between the starting point Pi and the lead point P2, p . The key element 53 can then be moved back to the starting point Pi.

[0232] Figure 17 describes the characteristic curve acquisition process in more detail. Based on the trigger measurement point P2, t and knowledge of the permissible working range of the measuring probe, in particular its maximum permissible deflection, a limit point P2 can be defined. m to determine at which the key element 53 is maximally deflected and a limit value U m for the output signal, where the limit point P2, m The measuring probe 50 lies on a calibration line L', which is defined by the starting point Pi and the trigger measuring point P2,t. To acquire the characteristic curve, which is shown schematically in Fig. 19, the measuring probe 50 is moved from the starting point P1 to the limit point P2. mThe probe is shifted along the calibration line L', whereby the deflection signal of the measuring probe 50 is recorded along the calibration line L' to obtain a characteristic curve. In Fig. 19, the characteristic curve between the point where the probe ball 53 first comes into contact with the calibration sphere surface and the limit point P2, m , at which the maximum permissible and / or achievable deflection signal U mOnce a linear section is reached, the output signal U of the probe 50 is linearly proportional to the deflection of the stylus 53 in this section. In practice, however, the relationship between the output signal and the deflection can be non-linear, at least section by section. Especially with non-linear behavior, capturing the characteristic curve can be important to minimize errors when converting the deflection signal back into a geometric distance for further use in determining the position of the calibration object. Particularly for automated identification of a probe's working range, it can be useful to capture both a resting region before material contact and a travel range. The travel range is defined as the region in which the deflection signal saturates, although mechanical deflection continues and is permissible.Typically, this override area can serve to protect the measuring probe from damage. It can be helpful to define the limit point P2. m to place in this crossing area.

[0233] In the example shown, the lead point P2 corresponds to p , the trigger measurement point P2, t and the border point P2, mEach point corresponds to the point in coordinate system K' of the measuring probe 50 where the center of the stylus ball 53 would be located if the calibration object 1 were not in the way and the stylus ball 53 were in its rest position. Of course, it is conceivable to calculate and record these points alternatively or additionally in a different way than in the coordinate system K' of the measuring probe, for example, by reading the machine coordinate values ​​of a machine tool, if the measuring probe 50 is attached to the machine tool as shown, for example, in Figures 4-11, and is moved relative to the calibration object 1 by means of its movable components, such as the tool carrier 12 (feed direction X in the coordinate system of the machine tool), and / or the axial slide 13 (axial direction Z in the coordinate system of the machine tool), and / or the shift slide 14 (shift direction Y in the coordinate system of the machine tool).

[0234] Exemplary substeps of the profile acquisition 203', 203 are described in more detail as flowcharts in Figures 20 and 28, and schematically illustrated in Figures 21 and 29. Reference is made below to Figures 20, 21, and 22. Figure 21 shows a cross-section through the calibration sphere 1 in the X'-Y' plane. To perform the profile acquisition for the purpose of determining the pre-vertex SP;', points P2, P3, and P4 are calculated in addition to the starting point Pi. At point P2, the probe sphere 53 is deflected such that the output voltage U2 is generated. Point P2 is located on the calibration line L' between the trigger measurement point P2, t and the border point P 2.mThe output voltage U2 is thus located between the trigger value Ut and the limit value Um of the maximum permissible deflection. Points Pa and P4 are chosen such that they lie on opposite sides of point P2 and, together with point P2, on a profile measurement line that runs parallel to the reference axis S and, in this example, even coincides with the reference axis S, as shown in Fig. 21. The output voltage Us at point Pa and the output voltage U4 at point P4 are each lower than the output voltage U2, since the probe ball 53 is deflected less at points P3 and P4 than at point P2. The distance between point Pa and point P4 is referred to in this context as the first sensing path. The corresponding path that the probe ball 53 travels on the spherical surface of the calibration sphere 1 is referred to in this context as the first sensing path.The measuring probe 50 is first moved from point Pi to point P2 and then to point P3. The acquisition of the displacement signal is then started, and the measuring probe 50 is moved from point P3 to point P4. During this movement, an initial profile measurement curve, i.e., the displacement signal as a function of the displacement, is recorded. Upon reaching point P4, the acquisition of the displacement signal is stopped, and the measuring probe 50 is moved perpendicular to the reference axis S away from the calibration sphere 1 to reach point P5, where the probe sphere 53 is again in its rest position. Alternatively, point P5 can also be chosen (not shown in the figures) such that the measuring probe 50 is moved radially away from the calibration sphere 1 to reach point P5.

[0235] Exemplary substeps of the profile evaluation 204', 204 are described in more detail as a flowchart in Figures 23 and 31 respectively.

[0236] The following refers to Fig. 23. To determine the pre-vertex SP;', the measurement data of the deflection signal, which was acquired during the movement along the probe path between points P3 and P4, are first imported. In addition, the travel distance q of the probe 50, i.e., the probe path between points P3 and P4, is calculated as a scalar value. In Figures 21 and 22, the travel distance q corresponds to the distance traveled from the probe base 51 along the X' axis. If the probe 50 is mounted on a machine tool as described above and is moved along the machine axes X, Y, Z of the machine tool, the travel distance q can be determined using the Euclidean norm of the acquired travel distances along the X, Y, and Z axes of the coordinate system K of the machine tool. where , Yi and

[0237] The initial position values ​​along the respective machine axes are represented by Zj. Subsequently, a profile measurement curve is generated. The profile measurement curve can represent the displacement signal, in particular the output voltage U, as a function of the travel distance q. Alternatively, it is also possible to calculate a plane in which the travel distance q lies and to reconstruct the effective geometric displacement d in this plane from the output voltage U. In this case, the profile measurement curve can represent the effective geometric displacement d in this plane as a function of the travel distance q.

[0238] The profile measurement curve generated from the measurement data can typically exhibit noise or interference signals. Therefore, the profile measurement curve is preferably smoothed and / or aligned using a filter in a smoothing process 204 T.

[0239] To determine the pre-vertex SPj' from the smoothed profile measurement curve, various methods can be applied, such as 2042' determining an extremum, in particular the maximum, and / or 2043' calculating a measurement adjustment circle through the profile measurement curve and / or 2044' calculating a regression parabola through the profile measurement curve. If the pre-vertex SPj' has been determined in the coordinate system K' and / or in the coordinate system K, the procedure is continued as shown in Figures 24-31 with the determination of the vertex SPj.

[0240] The steps shown in Fig. 24 and Fig. 25 correspond to the steps shown in Fig. 16 and Fig. 17, respectively, where the points Pi, P2, p , P2,t and P2, m each through the points Pe, P?,p, P?,t and Py, m were replaced.

[0241] Figures 26 and 27 are analogous to Figures 18 and 19 respectively, where the points Pi, P2, p, P2, t and P2 m each through the points Pe, Py, p , Py.t and Py, m were replaced. The starting point, i.e., point Pe, for determining the vertex SPj is chosen such that when the probe ball 53 is moved perpendicular to the reference axis S, the probe ball 53 comes into contact with the spherical surface of the calibration sphere 1 at the already determined pre-vertex SPj'. The preceding explanations regarding the lead point P2, p or border point P2, m This also applies, mutatis mutandis, to the holding point P7. p or the border point P7, m The points Pe, P7, p , P7,t and P7, m lie on a calibration line L, which is perpendicular to the reference axis S and which passes through the pre-vertex SPj'.

[0242] The following refers to Figures 28, 29, and 30. Figure 29 shows a cross-section through the calibration sphere 1 in the Y'-Z' plane. To perform profile acquisition for the purpose of determining the pre-vertex SPj', points P7, Ps, and Pg are calculated in addition to the starting point Pe. At point P7, the probe sphere 53 is deflected such that the output voltage U7 is generated. Point P7 is located on the calibration line L between the trigger measurement point P7, t and the border point P 7.mThe output voltage U7 is positioned between the trigger value Ut and the limit value Um of the maximum permissible deflection. Points Ps and Pg are chosen such that they lie on opposite sides of point P7 and, together with point P7, on a profile measurement line that is perpendicular to the reference axis S and perpendicular to the calibration line L. The output voltage Us at point Ps and the output voltage Ug at point Pg are each lower than the output voltage U7, since the probe ball 53 is deflected less at points Ps and Pg than at point P7. The distance between point Ps and point Pg is referred to in this context as the second probe path. The corresponding path that the probe ball 53 travels on the surface of the calibration sphere 1 is referred to in this context as the second probe path. The first probe path P3-P4 and the second probe path Ps-Pg are perpendicular to each other.The measuring probe 50 is first moved from point Ps to point P7 and then back to point Ps. The acquisition of the displacement signal is then started, and the measuring probe 50 is moved from point Ps to point Pg. During this movement, a second profile measurement curve, i.e., the displacement signal as a function of the movement, is recorded. Upon reaching point Pg, the acquisition of the displacement signal is stopped, and the measuring probe 50 is moved perpendicular to the reference axis S away from the calibration sphere 1 to reach point P10, where the probe sphere 53 is again in its rest position. Alternatively, point Pg can also be chosen (not shown in the figures) such that the measuring probe 50 is moved radially away from the calibration sphere 1 to reach point P10.

[0243] Fig. 31 shows the substeps of the profile evaluation 204 for determining the vertex.

[0244] SPj. These substeps are analogous to the substeps of the profile evaluation 204' for determining the pre-vertex SPj' and thus the description for Fig. 23 also applies mutatis mutandis to Fig. 31 .

[0245] To increase measurement accuracy, the profile acquisition 203' and profile evaluation 204' for determining the pre-vertex SPj' and / or the profile acquisition 203 and profile evaluation 204 for determining the vertex SPj can be performed multiple times. In particular, the profile measurement curve can be performed at different deflections of the probe. Specifically, during profile acquisition 203', two or more measurements can be performed in which the position of point P2 in the Y' direction, and thus also the positions of points Pa and P4 in the Y' direction, are selected differently. This allows inaccuracies to be minimized, in particular, which can arise during movement along the first sensing path due to a residual force that—despite the essentially perpendicular deflection of the probe ball 53 to the reference axis—is exerted on the probe 50 parallel to the reference axis, i.e., in the X' direction in the example shown here.

[0246] Fig. 32 schematically shows a raw data profile measurement curve, which exhibits noise in the form of an almost pure sine wave, as well as data segments recorded before and after the actual sampling point. Fig. 33 shows the smoothed profile measurement curve (solid line) after applying a filter to the raw data profile measurement curve (dashed line) and after the irrelevant data segments before and after the desired sampling point were removed. The profile measurement curve shown here could be the first profile measurement curve for determining the pre-vertex SPj' and / or the second profile measurement curve for determining the vertex SPj.

[0247] In the present embodiment, the four vertices SP1, SP2, SP3 and SP4 shown in Fig. 13 are determined in the manner described above.

[0248] Exemplary steps for locating the calibration object 300 from Fig. 12, based on the determined vertices, are described in more detail in Fig. 34 as a flowchart for the case of the illustrated example of the calibration sphere 1. To determine the position of the calibration sphere, the sphere's center M is determined. For this purpose, a least square calculation 301 is first performed to determine a plane of least squares for the four vertices SP1, SP2, SP3, and SP4. Subsequently, in step 302, the four vertices SP1, SP2, SP3, and SP4 are projected onto the plane of least squares. In step 303, a projection circle of least squares is then calculated based on the four projected vertices. The center of the projection circle of least squares then corresponds to the desired center of the sphere M.

[0249] Example: Calibration object with a cylindrical part

[0250] Figures 35-38 illustrate a further embodiment of a method for locating a calibration object according to the present invention, using the example of a calibration object T having a cylindrical part 63 which is substantially rotationally symmetric with respect to a central axis CBA. The calibration object T shown in Figures 35-38 is a simplified schematic representation of a gear standard, which is clamped between two centering points 60 and is located on a rotary table 61, which is rotatable about an axis of rotation R. Each centering point 60 engages in a centering bore 62, with a first centering bore 62 being located centrally in an upper end face of the calibration object T and a second centering bore 62 being located centrally in a lower end face of the calibration object T.Alternatively, the calibration object T could be clamped on a spindle of the machine tool 10 described above instead of on a rotary table 61, for example on the first workpiece spindle 21 or the second workpiece spindle 22, in particular with a counter support, preferably with a tailstock.

[0251] To determine the position of the calibration object T, the position of the central axis CBA is determined with respect to a Cartesian coordinate system with coordinate axes X, Y, Z based on two pairs of vertices: An upper pair of vertices and a lower pair of vertices are arranged on a surface of the cylindrical part 63, each pair consisting of a first vertex and a second vertex that are diametrically opposed to each other with respect to the central axis CBA, with the upper pair of vertices being offset relative to the lower pair of vertices along the central axis CBA. The surface is an example of a simply curved surface. The vertices are determined using the measuring probe 50. For clarity, only the shaft 52 and the probe element, here a stylus 53, are shown in Fig. 36. The axis of rotation R of the rotary table 61 is aligned parallel to the Z' axis.The reference axis S is parallel to the X' axis. To acquire the profile measurement curves for determining the vertices, the probe sphere 53 is moved only along the X' axis in this example. To perform a surface detection process and / or a characteristic curve acquisition process to acquire a characteristic curve analogous to the sphere example described above, the probe sphere 53 is moved perpendicular to the reference axis S, i.e., along the Y' axis.

[0252] In the example shown in Figures 35-38, the upper vertex pair is located on an upper measuring flange OM and the lower vertex pair on a lower measuring flange UM of the calibration object T. The upper measuring flange OM and the lower measuring flange UM can each be sections of the cylindrical surface of part 63, which, for example, belong to a continuous cylindrical surface or have been specifically ground as surface segments. Depending on the grinding process, the two measuring flanges may unintentionally have different diameters and / or a coaxiality deviation. The method described below takes such differences and deviations into account.

[0253] By rotating the rotary table 61 about the axis of rotation R, the calibration object T can be brought into different angular positions. In Fig. 35, the calibration object T is in angular position C. s , i.e., by an angle C swith respect to the zero angles of the rotation axis R shown in Fig. 36, or rotated by an angle CN with respect to a zero mark N of the rotary table. In Fig. 37, the calibration object T is in a first rotational angular position, which is +90° relative to the rotational angular position C. s The calibration object T is rotated. In Fig. 38, the calibration object T is in a second rotational angular position, which is -90° relative to the rotational angular position C. s is rotated. To move the calibration object T from the first rotational angular position to the second rotational angular position, the calibration object T is rotated 180° around the rotational axis R.

[0254] The vertices are determined as follows based on their position vectors in the coordinate system K' (X 1 , Y', Z') of the measuring probe denotes:

[0255] • r io,+90° : first vertex of the upper vertex pair in the first rotational angular position (upper measuring collar, left side, C) s + 90°)

[0256] • r 20,+90° : second vertex of the upper vertex pair in the first rotational angular position (upper measuring collar, right side, C) s + 90°)

[0257] • r iu,+9o° : first vertex of the lower vertex pair in the first rotational angular position (lower measuring collar, left side, C) s + 90°)

[0258] • r 2u,+9o° : second vertex of the lower vertex pair in the first rotational angular position (lower measuring collar, right side, C) s + 90°) • r io,-90° : first vertex of the upper vertex pair in the second rotational angular position (upper measuring collar, right side, C) s - 90°)

[0259] • r 20,-90° : second vertex of the upper vertex pair in the second rotational angular position (upper measuring collar, left side, C) s - 90°)

[0260] • r iu,-9o° :first vertex of the lower vertex pair in the second rotational angular position (lower measuring collar, right side, C) s - 90°)

[0261] • r 2u - 9o° ■ second vertex of the lower vertex pair in the second rotational angular position (lower measuring collar, left side, C s - 90°)

[0262] The following vertex points correspond geometrically to each other:

[0263] • Upper measuring band: rio,+90° un dr lo ,-90° r2o,+90° and r2o,-90°

[0264] • Lower measuring band: riu,+90° and riu,-90° r2u,+90° and r 2u -90°

[0265] The central axis CBA is defined as a straight line through an effective upper center point r. Mo and an effective lower midpoint r Mu is defined. Determining the effective upper midpoint r Mo and the effective lower midpoint r Mubased on the determined vertex pairs in the first and second rotation angle positions.

[0266] Determining the effective upper midpoint r Mo and the effective lower midpoint r Mu This initially includes a first averaging for each rotation angle position in order to eliminate diameter deviations of the measuring bundles ("conicity"), i.e.:

[0267] Calculating a first upper midpoint r Mo +90 =, which is located midway between the first vertex r lo +90 = and the second vertex (r 2o +90 =) of the upper vertex pair, which was determined in the first rotational angular position,

[0268] Calculating a second upper midpoint (r) Mo -90 =), which is located midway between the first vertex (r lo _ 90 =) and the second vertex (r 2o _ 90=) of the upper vertex pair, which was determined in the second rotational angular position,

[0269] Calculating a first lower midpoint (r) Mu +90 =), which lies between the first vertex (r lu +90 =) and the second vertex (r 2u +90 °) of the lower pair of vertices, which was determined in the first rotational angular position,

[0270] Calculating a second lower midpoint (r) Mu -90 =) between the first vertex (r lu -90 =) and the second vertex (r 2u -90 °) of the lower pair of vertices, which was determined in the second rotational angular position,

[0271] The axes RA shown in Fig. 37 and Fig. 38 for the respective rotation angles C + 90° and C - 90° pass through the first upper center point r. Mo +90 = and through the first lower midpoint r Mu +90= (first rotation angle position, Fig. 37) or by second upper center point r Mo -90 ° and through the second lower midpoint r Mu -90 ° (second rotation angle position, Fig. 38).

[0272] Determining the effective upper midpoint r Mo and the effective lower midpoint r Mu This then includes a second averaging (reversal) to eliminate the influence of the axis through the centers of the measurement bundles, i.e.

[0273] Calculating the effective upper midpoint (r) Mo ), which is located midway between the first upper midpoint (r Mo +90 =) and the second upper midpoint (r Mo -90 =) is located

[0274] Calculating the effective lower midpoint (r) Mu ), which is located midway between the first lower midpoint (r Mu +90 =) and the second lower midpoint (r Mu -90 =) is located

[0275] The central axis CBA, which runs through the centering bores in which the centering tips are received, is determined by r Mu and r Mo .

[0276] Determining the position of the central axis CBA can also include calculating the positional deviation of the central axis CBA from a given ideal axis, here the rotational axis R of the rotary table 61. In particular, the following positional deviation can be determined: EXOCBA = X o : Position deviation of the CBA axis in the X direction

[0277] E Y0CBA = ^o : Position deviation of the CBA axis in the Y direction Perpendicularity deviation of CBA to the Y direction

[0278] (Orientation deviation)

[0279] E B0CBA Perpendicularity deviation of CBA to X-direction

[0280] (Orientation deviation)

[0281] The coordinates X o , Y o , Z ocorrespond to the intersection point of the central axis CBA with the X-Y plane at Z = 0.

[0282] The coordinates X1, Y1,Z1 correspond to the intersection point of the central axis CBA with a first cutting plane, which runs parallel to the XY plane at the height Z = Z.

[0283] The coordinates X2, Y2, Z2 correspond to the intersection of the central axis CBA with a first cutting plane, which runs parallel to the XY plane at the height Z = Z2. The height difference AZ is defined as AZ = Z2 - Z.

[0284] Further variations

[0285] The examples described above can, of course, be adapted to calibration objects that have different shapes and / or are attached in a different way. Furthermore, the stylus element does not have to be spherical, but can have another suitable shape.

[0286] REFERENCE SYMBOL LIST ,1' Calibration object 59 Tilting element 0 Gear grinding machine 60 Centering tip 1 Machine bed 61 Rotary table 2 Tool carrier 62 Centering bore 3 Axial slide 63 Cylindrical part 4 Shift slide S Reference axis 5 Tool spindle K Coordinate system of the 51 Drive motor housing Machine tool 6 Grinding worm K' Coordinate system of the 0 Workpiece carrier Measuring probe 1 First workpiece spindle a Mounting bracket 2 Second workpiece spindle r Position vector in 11 Workpiece spindle drive three-dimensional space 3 Workpiece X Feed direction 0 Dressing device Y Shift direction 1 Swivel device Z Axial direction 2 Dressing spindle A Swivel axis 3 Dressing tool B Tool spindle axis 0 Machine control C1, C2 Workpiece spindle axis 1 Axis modules C3 Rotary axis 2 Control computer Cs Rotation angle position 3 Operating panel CN ​​Angular position with respect to4 Monitoring device zero mark 0 Measuring probe 1 Measuring probe base L, L' Calibration line 10 Signal generation unit CBA Central axis 2 Shaft R Rotation axis 3 Sensor element OM Upper measuring collar 4 Light transmitter UM Lower measuring collar 5 Light receiver N Zero mark Rotary table 6 Light barrier SPj Vertex 7 Light blocking element SPj', SP" Pre-vertex 71 Joint 8 Spring mechanism.

Claims

PATENT CLAIMS 1. Method for locating a calibration object (1) with a curved surface in a coordinate system (K, K') using a measuring probe (50), wherein the measuring probe (50) comprises: a measuring probe base (51); a deflectable arm, wherein the arm comprises a shaft (52) which is connected to the probe base (51), and a probe element (53), in particular in the form of a probe ball, wherein the probe element (53) is arranged at a free end of the arm, wherein the probe element (53) has a rest position with respect to the probe base (51), wherein the probe has a reference axis (S) which is fixed with respect to the probe base (51) and along which the shaft (52) extends in the rest position, wherein the probe (50) is configured to output a deflection signal as a function of a mechanical deflection of the probe element (53) from the rest position, and wherein the method comprises: (200) Determining at least one vertex (SPj) of the curved surface of the calibration object (1), wherein at each vertex (SPj) the reference axis (S) of the measuring probe (50) is aligned parallel to a tangent which passes through the vertex (SPj) on the curved surface, and (300) Locating the calibration object (1) in the coordinate system (K, K') based on the at least one determined vertex (SPj), wherein determining each vertex (200) comprises: (203', 203) Moving the stylus element (53) relative to the curved surface along at least one stylus path; Capturing the deflection signal of the measuring probe (50) along the probe path to obtain a profile measurement curve, and (204', 204) Determining the vertex (SPj) in the coordinate system (K, K') using the profile measurement curve.

2. Method according to claim 1, wherein the determination (204', 204) of the The vertex, based on the profile measurement curve, includes: (2042', 2042) Determining an extreme value of the profile measurement curve; or (2043', 2043) Calculating a measurement compensation circuit based on the profile measurement curve; or (2044', 2044) Calculating a regression parabola using the profile measurement curve.

3. Method according to one of the preceding claims, wherein the displacement (203,203') of the sensing element (53) relative to the curved surface is carried out under continuous contact of the sensing element (53) with the curved surface.

4. Method according to one of the preceding claims, wherein the method after obtaining the profile measurement curve further comprises a smoothing process (204T, 2041), wherein the smoothing process (204T, 2041) comprises: Applying a filter to the airfoil measurement curve before determining the vertex (Spi) from the airfoil measurement curve.

5. A method according to any of the preceding claims, wherein determining each vertex (200) before moving along the respective probe path comprises a surface detection process (201, 20T), wherein the surface detection process (201, 20T) comprises: Positioning the measuring probe (50) on a starting point (Pi, Pe) where the probe element (53) is in its rest position; Calculating a lead point (P2, p , P7, p ) in the coordinate system (K, K'), where the lead point (P2, p , P7, p ) is selected such that the probe element (53) comes into contact with the calibration object (1) when the measuring probe (50) is moved in the direction of the lead point (P2, p , P7, p ) is postponed; Moving the measuring probe (50) perpendicular to the reference axis (S) in the direction of the lead point (P2, p , P7, p ); Capturing a trigger measurement point (P2,t , P 7.t ), at which the sensing element (53) is deflected such that the deflection signal corresponds to a predetermined trigger value (Ut), wherein the trigger measurement point (P2, t , P7, t ) between the starting point (Pi, Pe) and the lead point (P2, p , P7, p ) is located.

6. Method according to claim 5, wherein determining each vertex (200) Each surface detection process (201 , 20T) also includes a characteristic curve acquisition process (202', 202), wherein the characteristic curve acquisition process (202', 202) includes: Positioning the measuring probe (50) on the starting point (Pi, Pe); Calculating a boundary point (P2, m , P?,m) based on the detected trigger measurement point (P 2.t , Pz.t), wherein the probe element (53) has a maximum permissible deflection when the measuring probe (50) is at the limit point (P 2.m , P7,m) is located, where the boundary point (P2.m , P7,m) on a calibration line (L 1 , L) lies, which is defined by the starting point (Pi, Pe) and the trigger measurement point (P2, t , P7, t ) is defined; Moving the measuring probe (50) from the starting point (Pi, Pe) to the boundary point (P) 2.m , P7, m ) along the calibration line (L',L), wherein the deflection signal of the measuring probe (50) is recorded along the calibration line (L',L) to obtain a characteristic curve.

7. Method according to any of the preceding claims, wherein the calibration object has a doubly curved surface, and wherein determining (200) each vertex (SPj) comprises: (203') Moving the probe element (53) relative to the doubly curved surface along a first probe path and capturing the deflection signal of the probe (50) along the first probe path to obtain a first profile measurement curve; (203) Moving the probe element (53) relative to the doubly curved surface along a second probe path and capturing the deflection signal of the probe (50) along the second probe path to obtain a second profile measurement curve, wherein the second probe path is chosen such that the second probe path intersects the first probe path at right angles; (204') Determining a first pre-vertex (SPj') in the coordinate system (K, K') using the first profile measurement curve either before or after moving the probe element along the second probe path; Determining a second pre-vertex (SP") in the coordinate system (K, K') using the second profile measurement curve, and (204) Determining the vertex (SPj) in the coordinate system (K, K') using the first pre-vertex (SPj) 1 ) and the second pre-vertex (SP").

8. Method according to claim 7, wherein the second probe path is selected such that the second probe path overlaps the first probe path at the first pre-vertex (SPj). 1 ) intersects at right angles, whereby the second pre-vertex (SP") corresponds to the vertex (SPj).

9. Method according to any of the preceding claims, wherein the calibration object (1) is a calibration sphere having a sphere center (M), wherein at least three vertices (SPi, SP2, SP3) of the calibration sphere are determined, and wherein the localization of the calibration object (300) comprises: (304) Determining the center of the sphere (M) in the coordinate system (K, K') based on the at least three vertices (SP) determined.

10. The method of claim 9, wherein at least four vertices (SP1, SP2, SP3, SP4) of the calibration sphere are determined, and wherein determining the center of the sphere (M) comprises: (301) Performing a regression analysis to determine a regression level for the at least four vertices (SP1, SP2, SP3, SP4); (302) Projecting the at least four vertices (SP1, SP2, SP3, SP4) onto the regression plane, and (303) Calculating a projection adjustment circle based on the four projected vertices.

11. Method according to any one of claims 1-6, wherein the calibration object (T) has a cylindrical part (63) which is substantially rotationally symmetric with respect to a central axis (CBA), in particular wherein the calibration object (T) is a gear standard, and wherein the method comprises: Determining at least one upper vertex pair and one lower vertex pair on a lateral surface of the cylindrical part (63), wherein the vertex pairs each consist of a first vertex and a second vertex diametrically opposite each other with respect to the central axis (CBA), and wherein the upper vertex pair is offset relative to the lower vertex pair along the central axis (CBA), and wherein determining the position of the calibration object (1) comprises: Determining the position of the central axis (CBA) based on at least two pairs of vertices.

12. Method according to claim 11, wherein the calibration object (T) is enlarged by a The axis of rotation (R) is rotatably mounted, and the determination of the vertices includes: Determining the upper vertex pair (r lo +90 =, r 2o +90 °) and the lower vertex pair (r lu ,+90°, r 2u ,+9 o°) in a first rotational angular position of the calibration object (1'); Rotating the calibration object (T) 180° around the axis of rotation to bring the calibration object (T) into a second rotational angular position; Determining the upper vertex pair (r lo -90 =, r 2o -90 =) and the lower vertex pair (r lu _ 90 °, r 2u _ 90 =) in the second rotational angular position of the calibration object (1'); wherein the central axis (CBA) is a straight line through an effective upper center point (r) Mo ) and an effective lower midpoint (r Mu ) is defined, and includes determining the location of the central axis (CBA): Determining the effective upper midpoint (r) Mo ) and the effective lower midpoint (r Mu ) based on the determined vertex pairs in the first and second rotation angle positions.

13. Method according to claim 12, wherein the determination of the effective upper midpoint (r) Mo ) and the effective lower midpoint (r Mu ) includes: Calculating a first upper midpoint (r) Mo +90 =), which is located midway between the first vertex (r lo +90 =) and the second vertex (r 2o +90 =) of the upper vertex pair, which was determined in the first rotational angular position; Calculating a second upper midpoint (r) Mo -90 =), which is located midway between the first vertex (r lo _ 90 =) and the second vertex (r 2o _ 90 =) of the upper vertex pair, which was determined in the second rotational angular position; Calculating the effective upper midpoint (r) Mo ), which is located midway between the first upper midpoint (r Mo +90 =) and the second upper midpoint (r Mo -90 =) is located; Calculating a first lower midpoint (r) Mu +90 =), which lies between the first vertex (r lu +90 =) and the second vertex (r 2u +90 =) of the lower vertex pair, which was determined in the first rotational angular position; Calculating a second lower midpoint (r) Mu -90 =) between the first vertex (r lu -90 =) and the second vertex (r 2u -90 =) of the lower vertex pair, which was determined in the second rotational angular position; Calculating the effective lower midpoint (f Mu ), which is located midway between the first lower midpoint (f Mu +90 =) and the second lower midpoint (f Mu -90 =) is located.

14. Method according to any one of claims 11-13, comprising determining the position of the central axis (CBA): Calculate at least one positional deviation of the central axis (CBA), in particular a positional deviation (XOCBA) and / or an orientational deviation (EAOCBA). von a given ideal axis, wherein the given ideal axis preferably coincides with the axis of rotation (R).

15. Machine tool (10), in particular for machining pre-toothed workpieces (23), comprising: a calibration object (1) with a curved surface; a measuring probe (50), wherein the measuring probe (50) comprises: a measuring probe base (51), a deflectable arm, wherein the arm comprises a shaft (52) which is connected to the measuring probe base (51), and a sensing element (53), in particular in the form of a sensing ball, wherein the sensing element (53) is arranged at a free end of the arm (52), wherein the sensing element (53) has a rest position with respect to the measuring probe base (51), wherein the measuring probe has a reference axis (S) which is fixed with respect to the measuring probe base (51) and along which the shaft (52) extends in the rest position, and wherein the measuring probe (50) is configured to output a deflection signal as a function of a mechanical deflection of the sensing element (53) from the rest position;and a control system designed to perform the method according to any of the preceding claims.

16. Machine tool (10) according to claim 15, further comprising a tool carrier (12) and a workpiece carrier (20), wherein the tool carrier (12) and the workpiece carrier (20) are displaceable relative to each other, wherein the measuring probe (50) is arranged on the tool carrier (12) and wherein the calibration object (1) is arranged on the workpiece carrier (20), and wherein the machine tool (10) optionally further comprises: a spindle (15, 21, 22) which is configured to receive a tool (16) or a workpiece (23) and to drive it to a rotation about a spindle axis (B, C1, C2), wherein the calibration object (1) is optionally configured to be received in the spindle (15, 21, 22) and rotated about the spindle axis (B, C1, C2).

Images