DE102018208189B4 - Method and device for determining the torsional errors of a machine axis - Google Patents

Abstract

Method for determining torsional error values (zRz) in a device (10), preferably designed as a coordinate measuring machine, which comprises at least one controllable machine axis (X, Y, Z), comprising:
- Determination of translation error values (zTy _1,2 ) for the machine axis (X, Y, Z) based on measured values that were obtained on the basis of moving the machine axis (X, Y, Z) along a measuring section (M);
- Determination of first torsional error values (zRz) for the machine axis (X, Y, Z) based on measured values that were obtained along a short measuring section (A), the short measuring section (A) being shorter than the measuring section (M);
- Determination of an error size (Δs̃) based on the first torsion error values (zRz);
- Calculation of second torsion error values (zRz) for the machine axis (X, Y, Z) on the basis of the error size (Δs̃) and the translation error values (zTy _1,2 ).

Description

The invention relates to a method and a device for determining the torsion errors for a controllable machine axis, in particular in relation to coordinate measuring machines and / or to a device designed as a coordinate measuring machine.

Coordinate measuring devices, machine tools, machine kinematics or general technical systems can include controllable machine axes in a manner known per se in order to carry out predetermined movements. These can be translation or rotation axes, for example. The movements can be implemented by a drive, preferably individually assigned to the machine axis, which can include an electric motor, for example. This can be controlled in a manner known per se in order to carry out desired movements, for example by means of an electronically and / or digitally operated control device. The control signals used for actuation can cause the axis to perform a desired movement and, for example, specify a drive power to be applied for this.

A coordinate measuring device, with which object coordinates can be recorded in a known manner, for example from an industrially manufactured workpiece, typically comprises at least three controllable machine axes which are arranged in such a way that a measuring device moved by them (e.g. a measuring sensor that scans the object in a tactile or non-contact manner) can be connected to any Positions in a Cartesian coordinate system is movable. For this purpose, the machine axes themselves can be aligned along the corresponding axes of the coordinate system and / or each enable a movement along one of these axes (ie one machine axis can be assigned to a coordinate axis). As a result, a coordinate measuring machine typically has two machine axes extending horizontally and at right angles to one another (typically referred to as the X and Y axes). Furthermore, a vertically extending machine axis is often provided, which can also be referred to as the Z-axis. A corresponding structure of a coordinate measuring machine is explained, for example, in the following DE 10 2008 024 444 A1 shown.

It is also known that such machine axes can have so-called guidance errors. These are, for example, translation or rotation errors which, when moving along the machine axis, superimpose the actually desired movement and in particular result in movement components that are present along or around coordinate axes that deviate from the machine axis (i.e. that deviate from a target direction of movement of the machine axis Directions are available). In addition, when moving along a machine axis, torsional errors can also occur, which relate to a rotation of the machine axis about itself or about its own longitudinal axis.

• - Ein Positionsfehler zTz(z_i) betrifft die Verschiebung der Maschinenachse Z in Richtung der z-Achse am Ort z_i;
• - Ein Translationsfehler zTy(z_i) betrifft die Verschiebung der Maschinenachse Z in Richtung der y-Achse am Ort z_i;
• - Ein Translationsfehler zTx(z_i) betrifft die Verschiebung der Maschinenachse Z in Richtung der x- Achse am Ort z_i;
• - Ein Torsionsfehler zRz(z_i) betrifft die Verdrehung der Maschinenachse Z um die z-Achse am Ort z_i;
• - Ein Rotationsfehler zRy(z_i) betrifft die Verdrehung der Maschinenachse Z um die y-Achse am Ort z_i;
• - Ein Rotationsfehler zRx(z_i) betrifft die Verdrehung der Maschinenachse Z um die x-Achse am Ort z_i.

Such guide errors can have various causes and result, for example, from manufacturing inaccuracies in the guides, the scales and / or the bearings of a machine axis. The following lists the possible guidance errors when moving a machine axis designated with Z. The Z machine axis (or its axis of movement) extends along (or parallel to) the z axis of a Cartesian coordinate system, which also includes a y and an x axis. The possible guidance errors that occur when moving the Z machine axis along the z coordinate axis are as follows:

• - A position error zTz (z _i ) relates to the displacement of the machine axis Z in the direction of the z axis at location z _i ;
• - A translation error zTy (z _i ) relates to the displacement of the machine axis Z in the direction of the y-axis at the location z _i ;
• - A translation error zTx (z _i ) relates to the displacement of the machine axis Z in the direction of the x axis at location z _i ;
• - A torsion error zRz (z _i ) relates to the rotation of the machine axis Z about the z axis at the location z _i ;
• - A rotation error zRy (z _i ) relates to the rotation of the machine axis Z about the y-axis at the location z _i ;
• - A rotation error zRx (z _i ) concerns the rotation of the machine axis Z about the x-axis at the location z _i .

The locations z _i extend over the entire movement path that can be traversed by the Z machine axis, which also corresponds to a measurement path explained below.

It goes without saying that the nomenclature can also be applied analogously to an X machine axis extending along (or parallel to) the x axis and a Y machine axis extending along (or parallel to) the y axis . Instead of z _i , a location x _i or y _{i would then be} considered. Furthermore, the torsion error then relates in each case to a rotation of the Y, X machine axis about the coordinate axes parallel thereto or, in other words, about itself. Rotation errors exist about the remaining coordinate axes. The person skilled in the art is familiar with the corresponding nomenclature, which is why a complete reproduction of all 18 possible guide errors about all three X, Y, Z machine axes is not provided at this point.

It is also known to detect such guidance errors in the form of specific guidance error values at the individual x _i , y _i and z _i positions and to store them in a machine control system. The guidance errors can be used to mathematically correct position values and / or control signals of the machine axis (s) in the sense of calibration values. If, for example, position values of the machine axis (s) are used to derive measured values from them, as is common with a coordinate measuring machine, for example, the previously determined guide error values can be used to correct the measured values accordingly (i.e. to calculate the guide errors from the determined values). Further background information on the determination of guidance errors and the calibration of coordinate measuring machines based on them can be found in DE 10 2008 024 444 A1 and the DE 44 21 302 C1 .

The invention specifically relates to ways of determining such guidance errors. These can then then be used in a manner known per se for calibrating the machine axis and / or a device comprising a plurality of machine axes. Various approaches are known from the prior art for determining guidance errors.

In the DE 10 2008 024 444 A1 For example, a solution is disclosed in which a scanning device is moved along a reference body in order to determine guidance error values based thereon (referred to there as axis error values). Since the reference body is measured in advance with a calibrated coordinate measuring machine and body error values are determined for this purpose, imperfections of the reference body can also be taken into account. In other words, guidance error values adjusted for imperfections in the reference body can be determined.

A disadvantage of this solution is that the reference body, especially for torsion error measurements, must be at least as long as a measurement path along which axis error values are to be determined (i.e. the reference body must generally be as long as a maximum movement path of a single machine axis). The length and weight of such a reference body are therefore correspondingly high. If a coordinate measuring machine is to be calibrated after it has already been delivered to the customer, the customer must either have his own correspondingly large and therefore expensive reference body or the reference body must be delivered to the customer with a correspondingly high transport effort.

The DE 44 21 302 C1 however, relates to the optical detection of guidance error values by means of a laser interferometer. In this case, laser beams are directed in a manner known per se onto a reflector that is moved together with the axis in order to obtain a back-reflected radiation component. A radiation component decoupled from the originally emitted laser beam and the back-reflected radiation component can then be superimposed in order to determine transit time differences between them and, based thereon, to calculate, for example, distance values. Such laser interferometers are, however, often characterized by the disadvantage that they require a high outlay on equipment. In order to determine precise measured values, a particularly precise beam alignment is typically required, which is correspondingly time-consuming and requires experience.

Further background information on determining guide errors using laser interferometers can also be found in DE 33 34 460 A1 .

Furthermore, from the DE 37 81 674 T2 a position determination method within a measuring space of a coordinate measuring machine is known. Geometrical errors of the coordinate measuring machine are determined and error data to compensate for these geometrical errors are stored for the travel of each axis. A special measuring body is used here.

The DE 198 30 646 A1 discloses a method for correcting geometric process errors of a coordinate measuring machine, in which at least one recording of the process errors is performed with a measuring device an axis over the entire measurement volume with a grid of interpolation points, and in which the process errors of at least one axis are also recorded in at least one sub-volume of the entire measurement volume with an increased interpolation point density, the measuring device at least approximately at a virtual zero mark of at least one axis of the sub-volume is arranged and then the recording of the process errors is carried out in the partial volume.

In practice, expensive inclination trolleys are often used to determine torsional errors that cannot be detected directly with lasers. These also require a certain set-up effort when attached to a machine axis. For vertical machine axes in which such inclination carriages cannot be used, torsion errors must be determined in a different way. In this case, for example, the measurement along a reference body to be formed correspondingly long can be used again, as in FIG DE 10 2008 024 444 A1 explained.

In summary, the previous approaches each have disadvantages in the form of costly and / or complex measurement setups to be set, in particular when torsional errors about a machine axis are to be determined.

One object of the present invention is therefore to enable a precise determination of torsion errors of a machine axis with a reasonable cost and outlay on equipment.

This object is achieved by a method and a device according to the attached independent claims. Advantageous further developments are given in the dependent claims. Furthermore, it goes without saying that the features mentioned in the introductory description can also be provided individually or in any combination in the presently disclosed solution, unless otherwise stated or evident.

The invention makes it possible to calculate the torsional error for an axis without having to record torsional error measured values along the entire machine axis (i.e. along an entire measuring or movement path of this machine axis). Instead, an error size can be determined, which is determined on the basis of the measurement along a comparatively short distance (hereinafter short measuring distance), and the torsional error values can be calculated based on this, preferably along the entire measuring distance, but without necessarily being measured along this entire measuring distance.

Determining the size of the defect can, for example, help to reduce the set-up effort, as this does not have to be eliminated by complex and, above all, time-consuming manual settings. Instead, it can be determined and therefore taken into account in the calculation. At the same time, however, this takes place along a comparatively short short measuring section, which in particular can be shorter than a measuring section considered overall for the machine axis (and / or maximum movement distance of the machine axis). If a reference body is used for this, it can also be correspondingly short and thus also lighter, whereby the costs associated with the use of a reference body and / or the associated transport costs can also be reduced. If the measured values are obtained in a different way along the short measuring section (for example by optical measurements), the measuring device used only needs to be set up and / or calibrated along a correspondingly short section, which reduces the effort required. However, since the mentioned error size is taken into account, sufficient precision of the determined guide error values is guaranteed. Furthermore, through the only short and low-cost measurement and subsequent calculation of the torsion error values, it is possible to dispense with cumbersome torsion measurement methods previously carried out, in particular for measurements on vertical axes. On the other hand, reference devices previously required, such as expensive inclination scales, can be dispensed with on horizontal axes.

• - Ermitteln von Translationsfehlerwerten für die Maschinenachse basierend auf Messwerten, die durch (oder, mit anderen Worten, auf Basis von oder bei) ein(em) Bewegen der Maschinenachse entlang einer Messtrecke gewonnen wurden;
• - Ermitteln von ersten (per Messung ermittelten) Torsionsfehlerwerten für die Maschinenachse basierend auf Messwerten, die entlang einer Kurzmessstrecke gewonnen wurden, wobei die Kurzmessstrecke kürzer als die Messstrecke ist;
• - Ermitteln einer Fehlergröße basierend auf den ersten Torsionsfehlerwerten;
• - Berechnen von zweiten (per Rechnung ermittelten) Torsionsfehlerwerten für die Maschinenachse auf Basis der Fehlergröße und der Translationsfehlerwerte.

In detail, a method is proposed for determining torsional error values in a device, preferably designed as a coordinate measuring machine, which comprises at least one controllable machine axis, comprising:

• - Determining translation error values for the machine axis based on measured values obtained by (or, in other words, based on or during) moving the machine axis along a measuring section;
• Determination of first torsional error values (determined by measurement) for the machine axis based on measured values that were obtained along a short measuring section, the short measuring section being shorter than the measuring section;
• Determining an error size based on the first torsion error values;
• - Calculation of second torsion error values (determined by calculation) for the machine axis on the basis of the error size and the translation error values.

As described, the device can comprise a plurality of controllable machine axes, in particular three machine axes, which enable movements along all axes of a Cartesian coordinate system and are in particular aligned along one axis of such a coordinate system. In particular, two of the machine axes can extend horizontally and one vertically to this. If the device is designed as a coordinate measuring device, it can be implemented according to any design, e.g. according to a so-called portal construction or as a horizontal arm measuring device.

The determination of the translation and torsion errors for the machine axis can be understood to mean that corresponding error values are determined when moving along the machine axis and / or in different positions along this machine axis. In particular, the error values can be determined based on measured values determined at the various positions along the machine axis. In general, the translation error values can relate to displacements of the machine axis under consideration along a coordinate axis (and / or a further machine axis) that deviates from this. The translation error values preferably relate to displacements of the machine axis under consideration along (or in the direction of) a common coordinate axis, this coordinate axis preferably running at an angle to the machine axis under consideration. In the following with reference to the nomenclature listed in the introduction, zTy (z _i ) values or zTx (z _i ) values can be considered as translation error values in order to determine a torsion error along a Z machine axis, for example. All of the above or below statements relating to the Z machine axis apply mutatis mutandis to the X and Y machine axes.

The measuring section can be traversed by appropriate control or driving of the machine axis and preferably without additional control and / or driving of further machine axes. The measuring distance can correspond to the maximum movement distance of the machine axis or take at least part of this movement distance.

The measured values can be determined in a manner known per se on the basis of the laser interferometry described in the introduction. For this purpose, a laser interferometer arrangement can be provided and, for example, encompassed by the device described below (and / or, if necessary, can be arranged therein). The laser interferometer arrangement can comprise a beam source which, for example, can be positioned or positionable in the measuring volume of the device (preferably stationary). Furthermore, the laser interferometer arrangement can comprise at least one optical unit which is irradiated thereby and which can preferably be moved together with the machine axis. The optical unit can, for example, be arranged on a quill of the coordinate measuring machine and / or comprise at least one reflector or be implemented as such. In general, the optical unit can be set up to reflect the incoming radiation, in particular in the direction of a beam evaluation unit, which can include, for example, at least one interferometer head and / or at least one prism unit (e.g. with a Wollaston prism).

The translation error values can be detected essentially simultaneously and / or as part of a common movement process along the machine axis. The translation error values can each be determined based on measured values that are recorded by different measuring devices (for example different photodetectors or interferometer heads). For example, first measured values can be acquired and viewed for the first translation error values explained below and second measured values for the second translation error values explained below. According to one variant, measured values for determining the first translation error values are generated and / or recorded with the aid of a first optical unit, and measured values for determining the second translation error values are generated and / or recorded with the aid of a second optical unit. The first and second optical units are preferably in different positions. An optical unit with, for example, two reflectors at correspondingly different positions can also be provided. The different positions can relate to relate to a coordinate axis and / or machine axis that deviates from the machine axis under consideration or be arranged along this axis. This can in particular be the remaining coordinate axis and / or machine axis to which the translation error values do not relate (ie the axis for which neither the shift or the guidance error is considered, nor the direction in which this shift takes place). The different positions can therefore also be referred to as transverse positions in relation to the machine axis under consideration. If, for example, translation error values zTy (z _i ) are considered, the different positions can be arranged along the x-axis.

In principle, the different positions for acquiring the measured values can also be provided by using only one measuring unit and / or optical unit (e.g. comprising only one reflector), but moving them to different positions for acquiring the individual measured values or moving them into new positions is mounted.

The translation error values determined by measurement can, in addition to a component that goes back to the guidance error actually sought, also include other translation error components that can result, for example, from measurement inaccuracies and / or the general measurement setup. As explained in more detail below, according to certain embodiments it can be assumed (that is, within the framework of the method and in particular for determining and / or calculating the values sought), it can be assumed that the translation error values determined can include several translation error components that differ from one another at least one, for example, can be traced back to measurement inaccuracies and / or the measurement setup. In addition, however, the assumption can be made that the translation error values (e.g. if they include first and translation error values) also include at least one identical or the same translation error component, which e.g. relates to the actual translation guidance error, which is independent of the measurement method used.

A possible translation error component relating to the measurement setup can result from the fact that the measurement values are at least indirectly generated and / or recorded in a position spaced from the machine axis under consideration, for example because an optical unit of the type explained above is located there. In interaction with a torsion around the machine axis under consideration, measuring units for recording the measured values in such spaced-apart positions can be tilted and / or swiveled in a plane running orthogonally to the machine axis, which results in corresponding translation error components (see also the following 3 ). Further possible translation error components are explained in more detail below.

In contrast to the second torsion error values, the first torsion error values can be values determined by measurement technology or at least indirectly on the basis of an actual measurement process. In other words, the first torsional error values can represent the at least indirect result of a measurement and, in particular, scanning of a reference body, whereas the second torsional error values can represent the preferably direct result of a calculation explained below. Scanning can be tactile (e.g. by means of a measuring probe) or contactless (e.g. by means of optical or capacitive sensors). The short measuring section can extend along the measuring section or, in other words, coincide with a section thereof (i.e. the short measuring section lies within the measuring section). As a result, the short measuring section can in turn extend along the machine axis and / or a movement axis thereof.

The measured values along the short measuring section can in principle be obtained using any measuring method, e.g. optical measurements.

A further development of the method and the device provides that the short measurement section extends along a reference body and the measured values were obtained by scanning the reference body. The scanning can be done tactilely, capacitively or optically, for example.

The reference body can (at least in one dimension) have essentially the same length as the short measuring section and preferably no significantly larger dimension than the short measuring section. For example, a length of the reference body that essentially corresponds to the short measurement path can be greater than its height and / or width. The reference body preferably has at least one flat surface, along which the short measuring section extends. Alternatively, the reference body may comprise a double row (i.e. rows extending parallel to one another) of predetermined geometric elements with known (i.e. calibrated) positions. The elements can be spheres, cylinders or holes, for example.

The short measurement path, which can correspond to a length of the reference body, preferably comprises only part of a maximum possible movement path (hereinafter also maximum movement path) of the machine axis. In a variant, the short measuring section is shorter than 1 m and preferably shorter than 0.5 m (for example 30 cm). In contrast, the maximum movement distance can be at least 1.5 m, at least 2 m or even up to 2.5 m or have any length that is greater than that of the short measurement distance.

A spherical plate, a plane surface or a block (for example a granite block) or cuboid can be considered as a reference body. Background information on determining sample values with the aid of a reference body can be found in the above explained DE 10 2008 024 444 A1 .

In general, the scanning of the reference body can take place after (or also in advance) to the preferably optical determination of the translation error values. Depending on the specific design of the device, the reference body can also remain on or in the device when the translation error values are determined (or the measured values required for this purpose) so that all measured values can in principle be recorded without additional conversion or setup measures.

The error size can comprise at least one, but preferably at least two, translation error components (or, in other words, be formed by these). In particular, the error variable can include translation error components which can be traced back to measurement-related inaccuracies and / or which are not generated due to actual guidance errors of the machine axis under consideration. In general, translation error components can be error components that could be avoided when setting up a measuring system, for example by making precise settings and / or by using other (e.g. neighboring) machine axes with ideal properties. In the context of the method, however, it is generally provided not to avoid such error components by setting up the measuring system and / or the device as comprehensively and precisely as possible (which in principle can also be provided as an additional measure), but rather to determine or at least approximate the error size, for example by calculation in order to be able to compensate for these mathematically and / or take them into account.

This can be done in particular in such a way that the error variable is used to calculate the second torsional error values. As described, the second torsion error values can be determined mathematically and, in particular, exclusively mathematically. In other words, only values that have been determined and are available on the basis of measurements that have already been completed can be used for this purpose, but not additional measured values that still have to be recorded (e.g. by means of separate axis movements). The second torsion error values can additionally be determined and in particular calculated on the basis of the first translation error values and / or the second translation error values. The error magnitude (and / or the translation error values) can be used, for example, as parameter values in an equation used to determine the second torsional error values.

As explained below, however, the translation error values can also include further translation error components and, in particular, translation error components that are independent of the error size, in particular those that can be traced back to actual guidance errors of the machine axis under consideration.

A further development of the method and the device provides that the second torsional error values are calculated for a range that lies outside the short measurement section. The area can be, for example, a section of the measuring section (or also the entire measuring section) along which the translation error values were recorded. While the first torsional error values were only determined on the basis of measured values along the short measuring section, this also enables torsional error values to be determined outside the short measuring section.

In particular, provision can be made to calculate second torsional error values along the entire measuring section (i.e. to determine the torsional error values within the entire measuring section) or at least along half or at least two thirds of the length of the measuring section along which the translation error values were also recorded. For this purpose, the assumption can be made, for example, that the error size determined with reference to the short measuring section is essentially also present outside the short measuring section and / or is valid (or, in other words, is constant along the measuring section). This justifies using or taking into account the size of the error for calculating torsional error values outside of the short measurement path. The other variables used to calculate torsional error values, e.g. in the form of the (first and second) translation error values, can already be present outside of the short measuring path by acquiring corresponding measured values along the measuring section.

In a further variant of the method and the device, the translation error values include first and second translation error values for the machine axis, which are each determined based on measured values obtained on the basis of moving the machine axis along the measuring path (e.g. first measured values for determining the first translation error values and second measured values for determining the second translation error values).

It is preferred that the first and second translation error values are determined on the basis of measured values that are determined in measuring positions spaced apart from the machine axis, e.g. first measured values (for first translation error values) in first measuring positions and second measured values (for second translation error values) in second measuring positions, wherein the first and second measurement positions are different from each other). In other words, the measured values can relate to a position spaced apart from the machine axis (and in particular from a longitudinal axis, central center line and / or movement axis of the machine axis) and / or the measurement positions can be at a distance from or next to the machine axis. Consequently, the measuring positions can also be referred to as transverse positions in relation to the machine axis under consideration. In particular, the measured values can be recorded and / or generated at least indirectly in these positions. This can be done, for example, in that optical units and in particular reflectors for interacting with an optical measuring beam are arranged in the corresponding positions. Axes along which the correspondingly spaced positions are present and / or can be arranged can be present, for example, in the form of coordinate axes extending at an angle to the machine axis and / or longitudinal or movement axes of further machine axes. Preferably, the first and second translation error values relate to mutually different positions, for example along such an axis.

In summary, the first translation error values and the second translation error values can thus be determined in mutually different transverse positions in relation to a machine axis under consideration. In general, the first and second translation error values can each be detected in a plurality of measurement positions, with a measurement position of the second translation error values with two identical and one different coordinate values existing for each measurement position of the first translation error values. The different coordinate values are preferably present along a coordinate axis that extends at an angle to the machine axis under consideration (e.g. different x coordinates when looking at the Z machine axis).

In a further embodiment of the apparatus and method, the machine axis is a machine axis extending at an angle to the horizontal (i.e., the horizontal in a spatial coordinate system). In particular, the machine axis can extend along a vertical and / or along an axis which is oriented parallel to the direction of the gravitational force and preferably opposite to it. The machine axis can also be referred to as the Z axis, whereas coordinate and / or machine axes extending in the horizontal can be referred to as x and y axes (or X and Y machine axes). As described, in the case of vertical axes of this type, comparatively extensive additional measurements are often used to determine torsion errors, which e.g. require reference bodies that extend along the entire maximum movement path of the machine axis. With the method presented here, on the other hand, it is possible to reduce the measurement effort by using comparatively short reference bodies. In general, however, it is understood that the method also offers advantages when non-vertical machine axes and in particular horizontal machine axes are considered. In this case, with previous solutions for determining torsion error values, it is often necessary to use comparatively expensive inclination carriages. With the method described, such inclination carriages can be dispensed with and instead comparatively short reference bodies can be used.

According to a further development of the device and method, the (e.g. first and / or second) translation error values are determined based on measured values acquired by means of at least one optical measuring device. The optical measuring device can record the measured values on the basis of laser radiation and / or laser interferometry. In particular, the measuring device can be a laser interferometer arrangement described above. This offers the advantage that at least the translation error values can be determined by means of a comparatively precise optical measuring method and can be determined without the aid of reference bodies to be provided (and / or manufactured) specifically for this purpose.

The error magnitude can generally include translation error components related to both the first and the second translation error values. In particular, it can be provided that the error size for at least one of the translation error components described herein (ie for a type of Translation error component) comprises both a translation error component relating to the first and a translation error component relating to the second translation error values, provided that corresponding first and second translation error values are present. An example is given in equation (5) below.

As described, the error size can include at least one translation error component that cannot be traced back to an actual guidance error of the machine axis. An example of a translation error component, which is largely due to the measurement method used and / or the measurement setup used, is a first translation error component which is optionally comprised by the error size and which results from a non-ideal course of an optical measurement beam. This can also be referred to as an optical translation error component. The optical measuring beam can be a laser beam that is used to acquire and / or generate the measured values and in particular the translation error values. In particular, it can be a measuring beam or radiation component which is emitted and preferably not yet reflected back in the direction of an optical unit (e.g. a reflector) that is moved together with the machine axis. The ideal course of the optical measuring beam can generally relate to a course extending parallel to the machine axis and / or measuring section, whereby a back-reflected radiation component used to determine the translation error components can preferably run parallel to the measuring section and / or machine axis. Such a course is only achieved with a correspondingly high outlay on equipment and, in particular, beam alignment. Deviations from the ideal course of the optical measuring beam and in particular the described parallel course can generally be referred to as an oblique course of the optical measuring beam and / or in the form of such an oblique course (i.e. obliquely based on the axis to which the beam is ideally parallel should extend). This inclined course can result in an associated translation error component or, in other words, contribute to the determined translation error value, for example because the measured values determined based thereon are recorded and / or generated in a position shifted according to the inclination.

In summary, the ideal course can relate to a desired alignment and / or a desired course of the measuring beam, whereas the non-ideal course can deviate from this, e.g. in the form of an inclined position.

According to a further embodiment of the method and device, the error variable comprises at least one (second) translation error component, which results at least partially from a guidance error along a further machine axis. This can also be referred to as an additional guidance translation error component. The guidance error can be a translation, rotation or torsion error and / or the further machine axis can be a machine axis different from the machine axis under consideration. In particular, it can be a machine axis along which neither the (e.g. first or second) translation error values are considered, nor the coordinate and / or machine axis along which the machine axis under consideration is shifted due to these translation errors. If, for example, a shift of a Z machine axis in a y coordinate axis is considered, a guidance error along the X machine axis can cause the translation error component. A torsion error about this further machine axis is preferably considered as a guidance error.

The inventors have recognized that by taking this further translation error component into account, a particularly precise determination of the error size or the torsion error values ultimately calculated therefrom can be achieved. This is due, for example, to the fact that, on the basis of the error size and thus within the scope of the calculation, it is possible to differentiate between error values actually actually caused by guidance errors along the machine axis under consideration and guidance errors that are independent of this along the other machine axis and / or that these additional guidance errors can even be taken into account.

The method and the device can furthermore provide that (for example, to calculate the second torsional error values) it is assumed or the assumption is made that the error variable is at least partially constant along the measurement path. The at least partial constancy can be understood to mean that at least one of the translation error components explained above or below, but preferably all the translation error components and thus the error size per se, is constant along the measurement path. This makes it possible, essentially without additional calculation and / or measurement effort, to also calculate torsional error values outside the short measuring section (within which actual measured values for the error size were determined). This contributes to the fact that the short measuring section and thus the reference body can be correspondingly short.

In a further development of the device and the method, it is assumed or assumed that the first and second translation error values at a (and preferably any) given position along the measurement path (and / or along a movement path of the measurement axis) include at least one identical translation error component . This can also be referred to as a lead translation error component. This translation error component can be a translation error component different from the first and second translation error component described above. In particular, it can be a translation error component that can be traced back to an actual guidance error about the machine axis under consideration. The assumption of an identical translation error component can, within the scope of the method, enable advantages with regard to the determination of the second torsional error values in particular, since, for example, certain relationships can be suitably combined and / or equated with one another in view of the identical translation error components.

• - Translationsfehlerwerte für die Maschinenachse basierend auf Messwerten zu ermitteln, die durch ein Bewegen der Maschinenachse entlang einer Messstrecke gewinnbar sind (oder gewonnen wurden);
• - erste Torsionsfehlerwerte für die Maschinenachse basierend auf Messwerten zu ermitteln, die durch ein Abtasten eines Referenzkörpers entlang einer Kurzmessstrecke gewinnbar sind (oder gewonnen wurden), wobei die Kurzmessstrecke kürzer als die Messstrecke ist; und
• - eine Fehlergröße basierend auf den ersten Torsionsfehlerwerten zu ermitteln;

und wobei die Berechnungseinrichtung dazu eingerichtet ist, zweite Torsionsfehlerwerte für die Maschinenachse auf Basis der Fehlergröße und der Translationsfehlerwerte zu berechnen.The invention also relates to a device, in particular a coordinate measuring machine, with at least one controllable machine axis, an evaluation device and a calculation device, the evaluation device being set up to:

• To determine translation error values for the machine axis based on measured values that can be (or have been) obtained by moving the machine axis along a measurement path;
• to determine first torsional error values for the machine axis based on measured values which can be (or have been) obtained by scanning a reference body along a short measuring section, the short measuring section being shorter than the measuring section; and
• to determine an error size based on the first torsion error values;

and wherein the calculation device is set up to calculate second torsion error values for the machine axis on the basis of the error magnitude and the translation error values.

The device can comprise any further development and any further feature in order to provide or carry out all of the above or below steps, operating states and functions. In particular, the device can be set up to carry out a method in accordance with any of the above or below aspects. In particular, the device can include a measuring device for determining at least the first and second translation error values, wherein the measuring device can optionally also include a unit that can be moved together with the machine axis. This unit can comprise the optical unit described above, for example in the form of a reflector. Furthermore, the measuring device can comprise a beam source and / or generally a laser interferometer arrangement or can be implemented as the latter. Furthermore, the device can also comprise the reference body, which can preferably only be arranged within the coordinate measuring machine as required (e.g. within the scope of the described method, but not during an actual workpiece measurement).

The evaluation device and calculation device can each be operated electronically and / or digitally. They can each be implemented by a common computing device or be comprised of such a device. The computing device can be implemented as a conventional PC, for example. The computing device can have a computer program which, when executed on a processor unit, provides the evaluation device and calculation device functions described herein.

The device can also be set up to determine first and second translation error values for the machine axis based on measured values that can be (or have been) obtained by moving the machine axis along the measurement path. The first and second translation error values can be used to determine the second torsion error values.

• 1 eine Ansicht einer erfindungsgemäßen Vorrichtung, mit der ein erfindungsgemäßes Verfahren ausführbar ist;
• 2 eine Detaildarstellung einer gemeinsam mit einer Maschinenachse der Anordnung aus 1 bewegbaren optischen Einheit;
• 3 eine Darstellung eines Translationsfehleranteils, der auf einen nicht-idealen Verlauf eines Messstrahls zurückzuführen ist;
• 4 eine Darstellung eines Translationsfehleranteils, der auf einen Führungsfehler entlang einer weiteren Maschinenachse zurückzuführen ist;
• 5 ein Ablaufschema eines mit der Vorrichtung aus 1 ausführbaren Verfahrens.

In the following, an embodiment of the invention is explained with reference to the attached schematic figures. Features that match in terms of their type and / or function can be provided with the same reference symbols across the figures. They represent:

• 1 a view of a device according to the invention with which a method according to the invention can be carried out;
• 2 a detailed representation of a joint with a machine axis of the arrangement 1 movable optical unit;
• 3 a representation of a translation error component that can be traced back to a non-ideal course of a measuring beam;
• 4th a representation of a translation error component that can be traced back to a guidance error along a further machine axis;
• 5 a flow chart of one with the device 1 executable procedure.

In 1 is a device designed as a coordinate measuring machine 10 shown according to an embodiment of the invention. The measurement setup and the process sequence explained below are based, for example, on the determination of guide errors along a vertical Z machine axis and in particular a torsion error about this axis. However, guidance errors and in particular torsion errors about any other machine axis described below can also be determined, for which purpose the measurement setup and process sequence can be adapted accordingly. This concerns, for example, a corresponding adaptation of the axis movements to be carried out and / or the alignment of the reference body explained below 31 . However, based on the principles of the procedure according to the invention explained below, such adaptations are unproblematic for the person skilled in the art.

The coordinate measuring machine (or the device) 10 has a base 12th , on which, for example, a portal 14th is arranged. The portal 14th can in a longitudinal direction by means of a first machine axis, which is referred to here as the X machine axis, on the base 12th be proceeded. At the traverse of the portal 14th is a sled 16 arranged, which can be moved by means of a second machine axis, which is referred to here as the Y machine axis. On the sledge 16 there is a quill that can be moved by means of a third machine axis, which is referred to here as the Z machine axis 18th .

At the lower free end of the quill 18th is a measuring head 19th arranged. At the measuring head 19th is optionally a scanning device 26th or an optical unit explained below 50 can be arranged. The scanning device is in the state shown 26th on the measuring head 19th arranged. This is analogous to the scanning device 26th in the 2 and 3 the DE 10 2008 024 444 A1 formed, but aligned upright or vertically in the present case.

The machine axes X , Y , Z extend and each allow movements along and parallel to the axes of a Cartesian coordinate system. These axes are in 1 referred to as x, y and z coordinate axes, the Z machine axis extending along the z coordinate axis, the Y machine axis extending along the y coordinate axis and the X machine axis extending along the x coordinate axis. It can be seen that the x and y coordinate axes run in a horizontal spatial plane and the z coordinate axis is vertical to this. For the machine axes assigned to the x, y, z coordinate axes X , Y , Z the same applies.

Each machine axis is conventional X , Y , Z individually controllable and movable and has its own in 1 unrecognizable drive unit. More precisely, the coordinate measuring machine has 10 Drives (not shown here) through which the portal 14th , the sled 16 and the quill 18th can be moved or moved by motor in the three spatial directions along the x, y and z coordinate axes. This allows the probe 19th within a measurement volume of the coordinate measuring machine 10 proceed three-dimensionally.

With the reference numbers 20th , 22nd , 24 are scales with the help of which the respective axis position of the measuring head 19th can be determined. With the standards 20th , 22nd , 24 it is typically a glass scale with a scale that is optically scanned. Alternatively, the coordinate measuring machine 10 however, they also have other position measuring devices, for example inductive measuring devices.

In addition, for the sake of completeness, it should be pointed out that the present invention can be used not only with coordinate measuring machines with a portal design, but also with other coordinate measuring machines, for example with a horizontal arm or bridge design.

On the base 12th of the coordinate measuring machine 10 is a reference body 31 on which a scale 30th is arranged, shown in simplified form. By means of the scanning device 26th becomes the reference body 31 sampled at a variety of measuring points. The measuring head can do this 19th and thus the scanning device arranged on it 26th are moved in the x, y and z directions via the drives not shown in such a way that the respective measuring points of the reference body 31 are approachable. Thus, the spatial coordinates of a measuring point are through the x, y, z axis positions of the measuring head 19th who are beyond the standards 20th , 22nd , 24 are readable, determinable.

In the context of the method explained below, the reference body 31 used to obtain first torsional error values for the Z machine axis along a short measurement section A. to determine which extends in the z-direction. This is explained in more detail below. In addition (as will also be explained in more detail below), translation error values for (or around the) Z machine axis are set along a measurement section M. calculated, which includes the entire maximum (or, in other words, the entire possible) movement distance of the Z machine axis. One recognizes from 1 immediately that the short measuring distance A. is less than the measuring distance M. , where in 1 but no exact length ratio of these routes A. , M. is shown because it is only a schematic illustration. The reference body can accordingly 31 be dimensioned comparatively small and to provide the short measurement section A. for example have a length of no more than 30 cm (see corresponding length dimension A. in 1 ). The measuring section M. can, however, have a length of more than 1 m and for example 2 m.

Furthermore, one recognizes one at least partially on the basis 12th arranged optical measuring device in the form of an interferometer arrangement 40 conventional design. The interferometer arrangement 40 comprises a laser beam source fixedly arranged on the base 42 , which are two parallel laser beams 44 sends out and also on the machine basis 12th Fixed prism units 46 directed, each of which can be designed as a Wollaston prism, for example. The prism units 46 are therefore in mutually different positions along the x coordinate axes. Above the prism units 46 (ie opposite one another along the z coordinate axes) can be an optical unit explained below 50 be positioned, which instead of the scanning device 26th to the measuring head 19th is attached. Then the Z machine axis is driven and the optical unit 50 Moved along the z coordinate axis, guidance errors and in particular translation errors for the Z axis can be determined in a manner known per se. More background information on such interferometer arrangements 40 can e.g. B. the introductory explained DE 44 21 302 C1 can be removed. In principle, any interferometer arrangement suitable for the following determination of measured values can be used 40 be used.

The interferometer arrangement 40 and the reference body 31 do not have to be on the base at the same time 12th be arranged. It may also be desirable that only one of the interferometer arrangement 40 and the reference body 30th on the base 12th is arranged and / or that these are optionally arranged thereon (ie either the interferometer arrangement 40 or the reference body 30th ), depending on which of the following measured values are currently to be determined.

The coordinate measuring machine 10 also has a computing device for determining the guidance error values 32 , which includes an evaluation, control and calculation device 33 , 34 , 35 includes or, in other words, provides. The control device 34 controls the movement of the coordinate measuring machine 10 by means of the X, Y, Z machine axes and in particular along the measuring section M. and the short measuring distance A. . The evaluation device 33 performs the computational evaluation of the position meter 20th , 22nd , 24 recorded position values and is set up to determine the guide error values explained below as well as the measured values required for this. Samples of these measured values are those of probe elements that are on the scanning device 26th are arranged, and determined by scale values obtained using the scale 30th be determined. In addition, in the manner explained below, the in 1 optical unit not yet shown 50 Measured values are determined in particular for determining translation error values. By means of the calculation device 35 can finally, based on predetermined input variables and / or input values, torsional error values along (or around the) Z-axis over the entire measuring section M. be calculated.

In a manner known per se, the determined guide error values can be used to calibrate one or all of the X, Y, Z machine axes. This means that you can use the coordinate measuring machine later 10 measurements carried out, the corresponding guidance errors can be eliminated, for example by means of a mathematical compensation. Possibilities for calibrating the X, Y, Z machine axes on the basis of the guidance error values and in particular the mathematical approaches to be used for this are known from the prior art and are not explained in more detail here.

In 2 is the optical unit 50 shown as an alternative to the one in 1 scanning device shown 26th on the measuring head 19th of the coordinate measuring machine 10 is attachable. The other details of the coordinate measuring machine 10 are in 2 Not shown. One also recognizes the coordinate system, which is analogous to that with regard to the vertical and horizontal 1 is oriented. However, the point of view is opposite 1 slightly changed, as can be seen from the corresponding orientation of the x and y coordinate axes. The optical unit 50 includes two reflectors 52 which are at a distance h ₂ or h ₁ from a longitudinal axis L. of the measuring head 19th are positioned. The longitudinal axis L. runs parallel to the Z Machine axis and thus also the z coordinate axis. In particular, a longitudinal axis and a movement axis of the Z machine axis coincide with that in 2 shown longitudinal axis L. together. The distance values h ₂ and h ₁ thus represent distance values along the x coordinate axis or, in other words, relate to the positions of the reflectors 52 along the x-coordinate axis. Hence the in 2 shown left reflector 52 in a position x ₂ and the in 2 right reflector 52 positioned in a position x ₁ . The means of the reflectors 52 _{Acquired measured values therefore relate to positions x 1} , x ₂ spaced apart from the actually considered Z machine axis (or its longitudinal and / or movement axis) around which the torsional error is to be determined.

Furthermore, one recognizes in 2 the laser beams 44 from the beam source not shown there 42 emitted and by means of the prism units 46 towards the reflectors 52 be redirected. In a manner known per se, the reflectors can thus be used 52 back-reflected laser beams are generated, which with a decoupled portion from the originally emitted laser beams 44 are superimposed in order to determine measured values, for example in the form of distance values, on the basis of the interferences that arise.

Alternatively, just a reflector could be used 52 be provided. In this case, the reflector could be used for detecting the first and second translation error values 52 Move to the corresponding x ₁ , x ₂ positions or be re-assembled in these positions and the translation error values could, so to speak, be recorded one after the other. A single prism unit can also be used for this purpose 46 used analogous to the reflector 52 moved or re-assembled, whereupon a laser beam 44 on this prism unit 46 to be led.

From the prior art explained in the introduction in the form of DE 44 21 302 C1 and also the DE 10 2008 024 444 A1 (See in particular [0077] there) it is known that with such a distance value detection in a position spaced apart from the machine axis under consideration, torsional errors about a corresponding axis can also be calculated.

However, the inventors have recognized that this is only possible with a considerable outlay on equipment with regard to the measurement systems used, or that assumptions that are simplistic and falsify the accuracy of the measurement results have to be made for this. The solution shown provides instead, by means of the optical unit 50 first to detect translation errors that arise when moving along the Z machine axis around which the torsion error is to be determined, along at least one of the other X and Y machine axes or in the corresponding x and y coordinate directions. In this case, translation error components are also taken into account, which can arise independently of the actual guidance error to be determined along the Z machine axis.

Details are given with each of the reflectors 52 out 2 and thus a corresponding translation error is determined in both the x ₁ and the x _{2 position.} In the following, the translation error when moving along the Z machine axis in the y direction or along the Y machine axis is considered as an example. However, it would also be possible, for example, to consider the displacement of the Z machine axis in the x direction or along the X machine axis. In general, the corresponding error values are recorded on individual z-axis values according to a selected measurement resolution (ie individual z-positions are considered). Due to the selected measurement setup, the translation error values determined are also _{dependent on the respective detection location in the form of the x 2} , x ₁ position to which they refer and to which the reflectors 52 are arranged.

At the beginning of the procedure are with the optical unit 50 Translation error values along the entire measuring section M. by moving the Z machine axis along this route and recorded in the computing device 32 deposited. These values can then be used for the following calculations.

$${\text{z}\tilde{\text{T}}\text{y}}_2\left({\text{x}}_2,\text{z}\right)=\text{zTy}\left(\text{z}\right)+{\text{s}}_2\ast \text{z}+{\text{h}}_2\ast \text{zRz}\left(\text{z}\right)$$

The details are with the optical unit shown 50 the following translation errors can be recorded, whereby reference is made to the nomenclature chosen in the introduction: $${\text{z}\tilde{\text{T}}\text{y}}_1\left({\text{x}}_1,\text{z}\right)=\text{zTy}\left(\text{z}\right)+{\text{s}}_1\ast \text{z}+{\text{H}}_1\ast \text{zRz}\left(\text{z}\right)$$

$${\text{z}\tilde{\text{T}}\text{y}}_2\left({\text{x}}_2,\text{z}\right)=\text{zTy}\left(\text{z}\right)+{\text{<figure-callout id="2" label="s" filenames="DE102018208189B4\_0010.png,DE102018208189B4\_0011.png" state="{{state}}">s</figure-callout>}}_2\ast \text{z}+{\text{<figure-callout id="2" label="H" filenames="DE102018208189B4\_0010.png,DE102018208189B4\_0011.png" state="{{state}}">H</figure-callout>}}_2\ast \text{zRz}\left(\text{z}\right)$$

In these equations (1) and (2) defined for the method according to the invention, it is assumed that the translation error values comprise an identical translation component zTy (z). This corresponds to a translation error component that can be attributed solely to the actual guidance error (or translation guidance error) when moving along the Z machine axis. The further translation error components in the form of the further summands of equations (1) and (2), on the other hand, are due to other causes and each individually to one of the reflectors 52 related (or to an x-position and an h-distance).

A first translation error component is due, for example, to a non-ideal course of the laser beams 44 due, which occurs at the latest when this by means of the prism units 46 towards the reflectors 52 be redirected. More precisely, the corresponding beams can deviate from an actually desired (ie ideal) course extending parallel to the considered Z machine axis along which the optical unit is located 50 is moved to detect the zTy translation error. In particular, the laser beams can 44 at an angle α to the Z machine axis (and / or the z coordinate axis running parallel thereto), as in FIG 2 indicated accordingly. This can also be done in a portion extending in the y-coordinate direction y ₂ and y ₁ result, which results in a translation error component that cannot be traced back to a translation error of the Z machine axis. The corresponding y ₁ and y ₂ - Proportions can be determined using trigonometric considerations. As already applied in the above equations (1) and (2), the current z value (ie position value along the Z machine axis and / or z coordinate axis) can be multiplied by _{a coefficient s 1} and s _{2 for this purpose.} The coefficient S ₁ relates to the tangent of the angle α ₁ and the coefficient s _{2 is} the tangent of the angle α ₂ .

Another error component already taken into account in equations (1) and (2) results from the selected measurement setup and specifically the arrangement of the reflectors 52 at the distances h ₁ and h ₂ to the longitudinal axis L. or. Z -Machine axis. This is explained below with the aid of the merely schematic 3 explained. In 3 the coordinate system is off 2 shown in plan view, the individual coordinate axes in turn extending parallel to the X, Y and Z machine axes and / or coinciding with these. Furthermore, the coordinate system is oriented in such a way that the z-coordinate axis is viewed from above and this extends perpendicularly out of the plane of the sheet. In the following, only with reference to one of the reflectors 52 it can be seen that, as a result of a torsion error, it can be moved around the Z machine _{axes at a distance y β} from the x axis. More specifically, the torsion around the Z machine axes can cause the reflector 52 is pivoted by the angle β from its actual position x _1. The one with the reflector 52 The measured value recorded can thus be superimposed by a y component, which forms a corresponding translation error component that cannot be traced back to an actual translation guidance error of the Z machine axis.

Due to trigonometric considerations, this error component (or the one in 3 y _β component shown) _{can be calculated from the distance value h 1} multiplied by the tangent of the angle β. For the case of guidance errors in a coordinate measuring machine 10 Considered very small angular amounts, the value of the angle β can be equated with the tangent of this angle, so that the torsional error value zRz (z) corresponding to the angle β can be viewed directly and multiplied by _{the distance h 1.} It goes without saying that also for the second reflector 52 an analogous consideration applies at position x ₂ , as also taken into account accordingly in the above equation (2).

Finally, there is one based on the following 4th explained further translation error component, which is also not due to the actually considered Z machine axis or its guiding errors, but to guiding errors along a further machine axis. For the structure given here, a translation error component is specifically considered, which can be traced back to torsion errors around the X machine axis. Specifically, it is taken into account here that when the optical unit is arranged 50 and thus moving the reflectors 52 in their individual x ₁ and x ₂ positions, translation error portions or y components arise, which can be traced back to a torsion error about the X machine axis. In the 4th the corresponding torsional error xRx is expressed at positions x ₁ and x _{2 in the angular errors φ 1} and φ ₂ that arise at these positions and are only shown as examples of different sizes. It can be seen that these angle errors φ ₁ and φ ₂ also have corresponding y components y ₁ and y ₂ accompanied. The translation error component effectively resulting from these angle errors when moving along the Z machine axis in the y direction corresponds to the difference between the in 4th shown y ₁ - and y ₂ Components.

Based on analogous trigonometric considerations, as above with reference to 3 explained, the corresponding resulting translation error component can be multiplied as the product of the current z-value by the difference between the angle errors φ ₂ and φ _{1 shown.} This can also be taken into account in one of the above equations (1) or (2). On the basis of equation (2), the following equation (3) results, for example, with which the corresponding extended translation error can be determined. The resulting equation (3) is as follows: $${\text{z}\widehat{\tilde{\text{T}}}\text{y}}_2\left({\text{x}}_2,\text{z}\right)=\text{zTy}\left(\text{z}\right)+{\text{<figure-callout id="2" label="s" filenames="DE102018208189B4\_0010.png,DE102018208189B4\_0011.png" state="{{state}}">s</figure-callout>}}_2\ast \text{z}+{\text{<figure-callout id="2" label="H" filenames="DE102018208189B4\_0010.png,DE102018208189B4\_0011.png" state="{{state}}">H</figure-callout>}}_2\ast \text{zRz}\left(\text{z}\right)+\left({\mathrm{\phi }}_2-{\mathrm{\phi }}_1\right)\ast \text{z}$$

If you convert equations (1) and (3) according to the common term zTy (z) they contain and equate the corresponding expressions, the following relationship can be obtained: $${\text{z}\widehat{\tilde{\text{T}}}\text{y}}_2\left({\text{x}}_2,\text{z}\right)\text{-z}\tilde{\text{T}}\text{y}\left({\text{x}}_1\text{, e.g.}\right)=\left({\text{s}}_2-{\text{s}}_1\right)\ast \text{z}+\left({\mathrm{\phi }}_2-{\mathrm{\phi }}_1\right)\ast \text{z +}\left({\text{H}}_2-H_1\right)\ast \text{zRz}\left(\text{z}\right)$$

This expression can be simplified by considering an error size Δs̃ which, for each of the translation errors of equations (1) and (2), includes the translation error components explained above due to the oblique or non-ideal course of the laser beams and due to the angular errors around the X -Machine axis.

This error size is as follows: $$\Delta \tilde{\text{s}}=\left({\text{s}}_2-{\text{s}}_1\right)+\left({\mathrm{\phi }}_2-{\mathrm{\phi }}_1\right)$$

Furthermore, from the above equation (4), the difference between the h ₂ and h ₁ values can be designated as Δx. Based on these simplifications, equation (4) can be formulated as the following equation (6): $${\text{z}\widehat{\tilde{\text{T}}}\text{y}}_2\left({\text{x}}_2,\text{z}\right)-{\text{z}\tilde{\text{T}}\text{y}}_1\left({\text{x}}_1,\text{z}\right)=\Delta \tilde{\text{s}}\ast \text{z}+\Delta \text{x}\ast \text{zRz}\left(\text{z}\right)$$

A rearrangement of the equation according to the torsion error values zRz (z) that are actually of interest around the Z machine axis results in the following: $$\text{zRz}\left(z\right)=\frac{{\text{z}\widehat{\tilde{\text{T}}}\text{y}}_2\left({\text{x}}_2,\text{z}\right)-{\text{z}\tilde{\text{T}}\text{y}}_1\left({\text{x}}_1,\text{z}\right)-\Delta \tilde{\text{s}}\ast z}{\Delta \text{x}}$$

In this equation, the torsional error values zRz (z) are unknown. The same also applies to the error size Δs̃ or the product thereof with the current z-value. The latter unknown would be irrelevant if the slopes s ₁ and s _{2 of} the translation measurement and also the angle errors φ ₁ and φ _{2 were the} same, since the error size Δs̃ would then assume the value 0. However, such conditions can only be achieved with a considerable outlay on setting up and adjusting the measurement setup and the device. The example shown therefore provides for this error variable Δs̃ to be determined and taken into account computationally, instead of physically eliminating it, so to speak, by means of the outlay on equipment.

In this context, it was initially recognized that the assumption can be made that the error size Δs̃ along the entire measuring section can be made without significant loss of accuracy M. is constant. In other words, the inventors have recognized that when moving along the Z machine axis, the bevels s ₁ and s ₂ and the angle errors φ ₁ and φ ₂ are essentially unchangeable, since the Z machine axis or its position has little effect on these errors or has no influence at all. This makes it possible to detect the corresponding error size Δs̃ along only a comparatively short route and to assume the validity of this measurement also for areas outside this route. The short distance also lies in the movement spectrum (or along the general movement axis and / or the maximum movement distance) of the Z machine axis.

The short route is the one already based on FIG 1 explained short measuring section A. . Along this short measuring section A. in a separate process step, the in Torsional error values zRz contained in equation (7) are determined, specifically for the z-values along this short measuring section A. . The optical unit is used for this 50 on the measuring head 19th by the scanning device 26th replaced. Possibilities to eliminate torsion errors zRz around the Z-machine axis by means of a corresponding scanning device 26th and a reference body scanned by it (albeit a longer one) 31 are to be determined from the above-mentioned DE 10 2008 024 444 A1 are known and are explained there in particular in paragraph [0077]. Alternatively, the torsion error values can be zRz along the short measuring section A. can be determined by any other method, for example by optical methods.

Since at least along the short measurement section A. Torsional error values around the Z machine axis zRz (z) can be determined, equation (7) only contains one unknown in the form of the error size Δs̃. If equation (7) is rearranged according to this remaining unknown, precisely this error quantity Δs̃ can be calculated using the following equation (8): $$\Delta \tilde{\text{s}}=\frac{{\text{z}\widehat{\tilde{\text{T}}}\text{y}}_2\left({\text{x}}_2,\text{z}\right)-{\text{z}\tilde{\text{T}}\text{y}}_1\left({\text{x}}_1,\text{z}\right)-\text{zRz}\left(\text{z}\right)\ast \Delta x}{\text{z}}$$

In principle, any desired z value (ie every z _i location) within the short measurement section can be used as the z value A. be used. Different z-values are preferably used in equation (8) and the results obtained are averaged, from which the value of the error quantity Δs̃ results.

The zTy translation error values contained in equation (8) were determined by means of the optical unit 50 by moving the Z machine axis along the entire measuring path M. recorded at the beginning of the method, these being the respective total values which also contain the further translation error components (see equations (1) and (2)). It goes without saying that in equation (8) only zTy translation error values at z positions along the short measurement section A. to be viewed as. Furthermore, the z-values (along the short measurement section A. ) as well as the Δx-value known from the measurement setup. Finally, the zRz (z) torsional error values used in equation (8) are those along the short measurement section A. metrologically determined error values. Consequently, the error size Δs̃ can at least along the short measurement section A. can be calculated, with the result based on the constancy assumption explained above also for the entire measuring section M. applies.

In the manner explained above, the error size Δs̃ was thus determined based on measured values and, more precisely, calculated on the basis of these measured values. Based on the assumption outlined above, according to which this error variable Δs̃ along the entire measuring section M. is assumed to be constant, the correspondingly calculated value can be used, for example, in equation (7) described above. Then z-values can be used in this equation along the entire measuring section M. and also the metrologically determined translation error values zTy along the entire measuring section M. can be used. As a result, torsional error values zRz (z) around the Z machine axis at every z position along the measurement path M. be calculated.

The result is thus obtained on the basis only along a comparatively short short measuring section A. Metrologically recorded torsional error values zRz around the Z machine axis along the entire measuring section M. calculated zRz torsional error values about this Z machine axis. In contrast to previous solutions, due to the reduced short measuring distance A. on a comparatively small and therefore inexpensive reference body 31 can be used, which in particular does not extend along the entire measuring section M. , but only along the short short measuring distance A. must extend.

Finally, a flowchart based on the 5 explained, in which the essential steps of the method explained above are summarized. In step S1, the zTy translation error values are determined along the Z machine axis in the first position x ₁ . In step S2, the zTy translation error values are determined along the Z machine axis in the second position x ₂ . Steps S1 and S2 can also be carried out partially or completely parallel to one another, for example by moving the reflectors together 52 by means of the optical unit 50 out 2 .

The measured values are taken along the entire measuring section M. out 1 recorded and thus at all the locations considered along this measuring section M. (ie at all z positions that are considered according to the selected measurement resolution along the Z machine axis). Values for the expressions on the left-hand sides of the above-explained equations (1) and (2) are thus obtained, where the individual error components that make up these values are not yet known on the right-hand side of these equations.

In step S3, to determine the unknown error variable Δs̃, the torsion error zRz about the Z machine axis is used by scanning the reference body 31 determined. One thus obtains for the z-values along the short measurement section A. out 1 First torsional error values measured by measurement around the Z machine axis.

In step S4, the error size Δs̃ is then determined arithmetically on the basis of the measured values determined using the previous equation (8).

In the final step S5, the torsion error zRz about the Z machine axis along the entire measuring section is calculated using the previous equation (7) by inserting the error magnitude Δs̃ \ and the translation error values recorded by measurement M. , which is the short measurement distance A. exceeds in length, calculated.

Within the scope of the present disclosure, provision can also generally be made for determining the torsional error values to be the first (along the short measurement section A. measured) and second (calculated) torsional error values. Especially when the measuring section M. and short measuring distance A. as in 1 can overlap shown along the short measurement section A. the measured and in the remaining part of the measuring section M. that does not deal with the short measuring distance A. overlaps, the calculated torsional error values are considered and used, for example, to calibrate the Z machine axis there.

List of reference symbols

10

Device / coordinate measuring machine

12th

Base

14th

portal

16

Sledge

18th

Quill

19th

Measuring head

20-24

scale

26th

Scanning device

30th

scale

31

Reference body

32

Computing device

33

Evaluation device

34

Control device

35

Calculation device

40

Measuring device

42

Laser beam source

44

Laser beams

46

Prism unit

50

optical unit

52

reflector

M.

Measuring section

A.

Short measuring distance

L.

Longitudinal axis

X, Y, Z

Machine axis

h1, h2

Distance value

y1, y2

y- (error) component

x1, x2

x (error) component

φ

Angle error

zTx (zi), zTy (zi)

Translation error

zTz (zi)

Position error

zRz (zi)

Torsion failure

zRy (zi), zRx (zi)

Rotation error

∆s

Error size

Claims (11)

Method for determining torsional error values (zRz) in a device (10), preferably designed as a coordinate measuring machine, which comprises at least one controllable machine axis (X, Y, Z), comprising: determining translation error values (zTy _1,2 ) for the machine axis (X , Y, Z) based on measured values obtained on the basis of moving the machine axis (X, Y, Z) along a measuring section (M); - Determination of first torsional error values (zRz) for the machine axis (X, Y, Z) based on measured values that were obtained along a short measuring section (A), the short measuring section (A) being shorter than the measuring section (M); - Determination of an error size (Δs̃) based on the first torsion error values (zRz); - Calculation of second torsion error values (zRz) for the machine axis (X, Y, Z) on the basis of the error size (Δs̃) and the translation error values (zTy _1,2 ). Procedure according to Claim 1 , the second torsional error values (zRz) being calculated for a range that lies outside the short measuring distance (A). Procedure according to Claim 1 or 2 , wherein the short measurement section (A) extends along a reference body (31) and the measured values were obtained by scanning the reference body (31). Method according to one of the preceding claims, wherein the translation error values (zTy _1,2 ) comprise first and second translation error values (zTy ₁ , zTy ₂ ) for the machine axis (X, Y, Z), which are each determined based on measured values based on a movement of the machine axis (X, Y, Z) along the measuring section (M) were obtained; and wherein the first and second translation error values (zTy ₁ , zTy ₂ ) are preferably determined on the basis of measurement values that are determined in measurement positions (x ₁ , x ₂ ) spaced from the machine axis (X, Y, Z). Method according to one of the preceding claims, wherein the machine axis (X, Y, Z) extends at an angle to the horizontal. Method according to one of the preceding claims, wherein the translation error values (zTy _1,2 ) are determined based on measured values acquired by means of at least one optical measuring device (40). Procedure according to Claim 6 , wherein the error size (Δs̃) comprises at least a first translation error component (s ₁ * z, s ₂ * z) which results at least in part from a non-ideal course of an optical measuring beam (44). Method according to one of the preceding claims, wherein the error size (Δs̃) _{comprises at least a second translation error component (φ 1} * z, φ ₂ * z), which results at least partially from a guidance error along a further machine axis (X, Y, Z). Method according to one of the preceding claims, wherein at least a portion of the error variable (Δs̃) is constant along the measuring section (M). Procedure according to Claim 4 , wherein the first and second translation error values (zTy ₁ , zTy ₂ ) at a given position along the measurement path (M) comprise at least one equal translation error component (zTy). Device (10), in particular coordinate measuring machine, with at least one controllable machine axis (X, Y, Z), an evaluation device (33) and a calculation device (35), the evaluation device (33) being set up for: - Translation error values (zTy _1,2 ) to determine for the machine axis (X, Y, Z) based on measured values that can be obtained on the basis of moving the machine axis (X, Y, Z) along a measuring section (M); - to determine first torsional error values (zRz) for the machine axis (X, Y, Z) based on measured values which can be obtained along a short measuring section (A), the short measuring section (A) being shorter than the measuring section (M); and - to determine an error size (Δs̃) based on the first torsion error values (zRz); and wherein the calculation device (35) is set up to calculate second torsion error values (zRz) for the machine axis (X, Y, Z) on the basis of the error size (Δs̃) and the translation error values (zTy _1,2 ).

Images