METHOD FOR TESTING COMPONENT SURFACES FOR DEFECTS AND / OR DETERMINING PROPERTIES OF COMPONENT COATINGS

Description

[0001] The invention relates to methods for testing component surfaces for defects and / or determining the properties of component coatings. These are, in particular, curved surfaces with a three-dimensional shape. Coatings specifically include paints.

[0002] The invention relates to an inspection device in a defined orientation and / or distance relative to the at least one testing surface area on a surface of a component, which should generally be defined in relation to this surface in order to detect defects or determine properties.

[0003] Surface defects pose a significant technical and economic risk to manufacturers if they are not detected early and reliably, and if necessary, defective parts are not rejected or reworked. This is particularly problematic for coated, especially painted, components, where deviations in layer thickness, gloss level, color, roughness, texture from the target state, or delamination are common defect types. Furthermore, localized defects such as cracks, blisters, inclusions, uneven coating application, or defects in the coated substrate are also frequent types of defects.

[0004] Numerous systems for detecting surface defects are known in the prior art. Many of these systems require precise alignment of the sensor system with respect to distance and angle, for example, centered and perpendicular, relative to the surface area to be inspected.

[0005] An example of such a test is the measurement of the gloss or reflectance of coatings. Here, directed light is shone onto the sample surface at a defined angle of incidence. At the same angle of reflection, a detector measures the light radiation reflected from the sample surface. The gloss is calculated as the integral over the reflectance indicatrix of the sample surface for a solid angle range determined by the apertures of the illumination and detector optics. The values ​​for the angular ratios and the characteristics of the light source and the receiving unit are specified in several standards, including ISO 2813. Typical angles are 20°, 60°, and 85°. Similar arrangements with defined angles of incidence and reflection, and properties of the light source and receiving unit, are known and standardized for measuring hue, saturation, and brightness.

[0006] Another typical example is the non-destructive measurement of layer thickness using magnetic-inductive methods for ferromagnetic substrates and eddy current methods for substrates made of other metals. In these methods, the sensor heads must be brought into perpendicular contact with the surface being tested.

[0007] All these methods, or related methods, for testing coatings and surfaces require precise alignment of the measuring device with respect to the respective surface. This means that such measurements are usually limited to flat surfaces, or, in the case of curved surfaces, especially at bends or freeform geometries, the probe head must be manually aligned correctly. Without feedback regarding the correct alignment, this represents a potential source of error in data analysis.

[0008] For automated inspection, if the component geometry and position are known, the probe head can be positioned accordingly by a robot or other suitable kinematic system. This method is well-established, for example, in fringe light inspection during the final inspection of vehicle bodies. Due to the significant setup effort required, this is only economically viable in series production with correspondingly high production volumes. For unknown components or those with high tolerances, the problem of automated alignment remains unresolved or requires prior geometric measurement.

[0009] JP 2008046103 relates to a device for checking surfaces.

[0010] An optical inspection unit is known from JP S60 154145 A.

[0011] CN 111257342 A discloses a system for determining the position and positioning of a camera.

[0012] JP H02 10250 A contains details of a device and a method which demonstrate possibilities for checking the size of defects on surfaces.

[0013] It is therefore an object of the invention to provide possibilities for a defined optical determination of defects and / or the determination of properties of component coatings on components in single production or in small quantities using hand-held or manipulator-guided (robot) probes, in which the respective positioning and orientation of the probe and the respective surface in relation to each other can also be taken into account or corrected.

[0014] According to the invention, this problem is solved by the features specified in claim 1. Advantageous embodiments and further developments of the invention can be realized by the features specified in dependent claims.

[0015] In the method according to the invention, at least one pattern of known geometry and dimensions is projected onto at least one surface area to be inspected and / or its surroundings using at least one projection light source. The image of the pattern on the respective surface is then captured by at least one optical detector unit, in particular a camera or optical detector, and a comparison is made with the original geometry and dimensions of the pattern based on the captured image. Using mathematical relationships suitable for the geometry and / or dimensions of the pattern, the alignment of the probe head with the optical detector unit in at least one solid angle with respect to the respective surface normals, and / or the distance of the probe head relative to the surface areas to be inspected, and / or the alignment of the probe head's measuring point with the desired position on the component are determined.The orientation and / or distance of the probe head, as determined in this way, are corrected relative to the respective surface if necessary. If a correction of the probe head's orientation and / or distance relative to the respective surface has been achieved or is not required, further measurements are evaluated. These measurements are used to determine defects and / or flaws or properties on component surfaces or coatings. Additional measurements can be recorded for each type of defect or property to be determined. For example, in the case of gloss determination, this could be the light radiation reflected from the respective surface, as already mentioned.

[0016] The at least one probe head can be used to determine, in particular, the gloss level, the color tone and / or the coating thickness.

[0017] As a pattern, one can project at least two light lines intersecting at a known angle, at least one circle or at least one ellipse, concentric arrangements of these, or a dot and / or line pattern with known distances between dots or lines onto a respective surface.

[0018] The projection light source and the data acquisition of the detector unit and / or the measurement process of the probe can be triggered or at least partially synchronized by a control unit in a temporal sequence. In particular, pulsed short-term activation with high illuminance and short exposure times should be used to reduce the effects of stray or ambient light on the actual correction determination and the identification of defects or properties. This can involve single or multiple pulsed irradiation and / or projection of a pattern.

[0019] The orientation, distance and / or alignment of the at least one probe head can be determined in relation to the respective surface on which the at least one pattern is projected by measuring the angles of intersection, line lengths, line spacings, position of the intersection points of projected light lines or points and / or the curvature of the image of the respective pattern and can be taken into account when correcting the orientation and / or distance of the probe head in relation to the respective surface and / or when evaluating further measured values ​​that are used to determine defects and / or flaws or properties on component surfaces or component coatings.

[0020] Correcting the position and orientation of at least one probe head relative to the respective surface can be achieved using an axis kinematic system, particularly a robot. The distance and / or alignment of the probe head's measuring point with the desired position can be fed back into the axis kinematic control system to execute a targeted, automated correction movement, thereby achieving correct alignment and / or distance of the probe head relative to the component surface.

[0021] With at least one hand-held probe, the determined orientation, distance and / or alignment in relation to the respective surface can be used to indicate to an operator in which direction or to which position the probe should be moved in order to facilitate the correct manual alignment of the probe and / or to achieve a tolerance window around the desired target distance and / or alignment.

[0022] During and / or after the forming of a component, the determination of defects and / or flaws or properties on component surfaces or component coatings can be carried out in order to be able to determine defects and / or specific properties of a coating, a sample material and / or flaws on the formed surface areas of the respective component with at least one probe.

[0023] When determining measured values ​​and / or surface properties that are not measured at nearly ideal points but rather over an extended area (a so-called measurement spot), particularly when integrating or averaging measurement acquisition is used for an area, the distortion of the projected pattern is analyzed to calculate an idealized, effective, planar measurement spot. For this purpose, the curvature of the component's surface within or at the measurement spot is determined based on the distortion of the captured pattern, and the distortion, area, and / or surface normal distribution of the measurement spot on the curved component surface are deduced. By subsequently mathematically projecting the local measurement distribution within the measurement spot, as determined by at least one probe, onto the plane of the idealized, effective measurement spot, the corresponding RMS values ​​of the measurement results can be derived.Suitable mathematical methods can be based on a scalar product of the determined set of surface normal vectors with the surface normal vector of the idealized measurement spot, or on methods from the field of projective geometry or affine / homographic mappings.

[0024] The invention will be explained in more detail below by way of example.

[0025] This shows: Fig. 1: A schematic representation of a measurement setup with which the method according to the invention can be carried out; Fig. 2: An exemplary measurement setup in which a pattern can be projected onto the surface of a component by means of cross-line projection; Fig. 3: A schematic representation of the image of a cross-shaped projection pattern on the surface of the component to be tested under two exemplary orientations (solid and dashed line); Fig. 4: A schematic representation of the image of a cross-shaped projection pattern on the surface of a component to be tested and the geometric-trigonometric relationships under two exemplary orientations, wherein the solid line cross shown can be detected by the detector unit in a non-perpendicular arrangement and the dashed line cross shown in a perpendicular arrangement; Fig.Fig. 5: Schematic representation of the image of a cross-shaped projection pattern on a highly curved surface of a component to be tested with freeform geometry and Fig. 6: a schematic representation of the image of a circularly projected pattern on a highly curved surface of a component to be tested with freeform geometry, whereby the image on the component surface appears distorted and the centroid of this image is not perpendicular to the centroid of the pattern.

[0026] In Figure 1 is shown by way of example how a pattern 1 of known geometry and dimensioning is projected onto the surface area 2.1 of a component 2 to be tested and / or its surroundings with a projection light source 6 and its Figure 3 can be detected with at least one detector unit 4, in particular a camera or optical detector. Based on how this Figure 3Compared to the original geometry and dimensions of pattern 1, the orientation relative to the respective surface normals / perpendiculars 2a and the distance 5b of the probe head (5) relative to the surface 2 to be inspected can be determined by the detector unit 4. Based on this evaluation, characteristic values ​​for the orientation 5a and / or the distance 5b of the probe head 5 relative to the surface of the surface area 2.1 can be determined. These characteristic values ​​can, for example, be directly derived as measured values ​​for the angular orientation 5a and the metric distance 5b to the surface 2.1. An alternative approach does not provide direct measured values ​​for the orientation angles 5a and distance 5b, but rather a distance metric of the detected Figure 3of the sample 1 in its current actual position compared to the theoretical appearance of the sample 1 in its target position, for example, with a central and perpendicular alignment to the surface area 2.1 to be inspected. It is particularly advantageous to define a tolerance window for the target range for this distance metric and to calculate a correction measure in order to generate at least the direction of a position correction of the probe 5 in order to successively bring it into accordance with the target position and alignment. Such correction measures and characteristic values ​​can be displayed to the operator via a display unit to facilitate the correct manual alignment of the probe 5. In a further embodiment according to the invention, the characteristic values ​​are fed back into the control system of a kinematic device, in particular a robot, in order to perform a targeted automated correction movement to achieve correct alignment of the probe 5.

[0027] A probe head 5 is designed to detect defects and / or properties on the surface 2.1 of the respective component 1; for this purpose, knowledge of its position and orientation relative to the surface 2.1 is required. The position and orientation can be influenced by the invention so that they are at least within a tolerance range, or a probe head 5 can be moved to such a position with a corresponding angular orientation 5a.

[0028] In one embodiment of the invention, a simple pattern 1 consisting of at least two light lines 1a intersecting at a known angle is projected onto the surface, e.g., by cross-line lasers or grating projectors. The orientation 5a, the distance 5b, and / or the alignment of the measuring point of the probe 5 with the desired position on the surface 2.1 can then be very advantageously determined by measuring the intersection angles 1b, the position of the intersection points 1c, the distances 1d between elements of the pattern (1), the line lengths 1e, and / or the curvature of the pattern 1, as can be seen in Figure 3 can recognize.

[0029] In Fig. 2 This is illustrated using a simple example with a cross-line laser as the projection light source 6, where the pattern projection and the detection of the Figure 3The pattern 1 is measured coaxially. It is conceivable that, with very large inclinations, the intersection points 1c may be mapped outside the detection range of detector unit 4 and must be mathematically reconstructed by extrapolation.

[0030] In Fig. 4 This simple example of a cross-line projection illustrates the geometric relationships between a probe arrangement tilted in two spatial directions α and β, or not aligned centrally (displacement Δx and Δy), relative to a perpendicular arrangement, which is particularly advantageous for numerous measurement tasks. In this example, a tilt leads to a shortening and lengthening of distances (for example, the distances shown in the diagram). Fig. 4The distances shown (a₁, a₂, b₁, b₂) within the projected pattern 1 perceived by the detector unit 4 are analyzed. Furthermore, the shifts in the intersection angles of intersecting lines of pattern 1 can also be included in the evaluation. Based on the information provided in Fig. 4Based on guided trigonometric formulas, the tilt angles α and β relative to a perpendicular arrangement can be calculated. The lateral displacement Δx and Δy relative to the perpendicular can be calculated by measuring the intersection point. In this example, for a perpendicular, centrally aligned arrangement, a1 and a2 as well as b1 and b2 would be equal in pairs, and the intersection point of the cross-shaped pattern 1 would be positioned in the center of the image of the detector unit 4. For other shapes of patterns 1 and / or non-coaxial arrangements of projected patterns 1 and their detection, similar geometric and trigonometric relationships for calculating the tilt α and β and displacement Δx and Δy can be established or supplemented with corresponding correction terms.

[0031] Patterns 1, including at least one circle or ellipse and concentric arrangements thereof, are also advantageously applicable. The orientation can be determined by evaluating the patterns 1 projected onto the surface 2.1 by measuring their centroid and their deviation from circularity or eccentricity and comparing them with the specified original pattern 1. Another advantageous embodiment provides for a pattern 1 formed with points in the form of a grid with known point-to-point distances. Instead of points, lines with known lengths and spacing between the lines can also be used as a suitable pattern. Dash-dot patterns can also be used.

[0032] There are no restrictions regarding the spectral composition of the light radiation emitted by the projection light source 6; in particular, wavelengths outside the visible range can be very advantageous for specific applications. Laser light sources are especially advantageous for this purpose due to the low divergence of the emitted light. Furthermore, the projection light source 6 and the data acquisition of the detector unit 4 and / or the measurement process of the probe head 5 can be operated or activated sequentially or at least partially synchronized by an electronic control unit (not shown). In particular, pulsed short-term activation with high illuminance and short exposure times is very advantageous in order to reduce interference effects from ambient light and motion blur.In a further embodiment, it may be advantageous to use a beam splitter 6a to project the pattern 1 coaxially to the detection direction of the detector unit 4 onto the surface 2.1 of the component, as shown in . Figure 2 shown.

[0033] Another application area of ​​the invention is the characterization and inspection of components 2 or test specimens in connection with uniaxial or multiaxial forming processes, in particular cupping or warping tests, such as the Erichsen cupping test. In this process, a sample or component 2 is clamped with a blank holder and deformed by a working medium, test tool, or indenter. Of particular interest is how surface properties or coatings behave depending on the deformation. In this context, the correct alignment and / or distance of the probe 5 in relation to the respective surface 2.1 is of particular importance in the formed area and can be advantageously taken into account by the invention.

[0034] One embodiment of the invention relates to measuring principles of the probe head 5 that do not acquire at nearly ideal point measurements, but rather in a planar area (a so-called measuring spot). This is particularly the case when an integrating or averaging measurement method is used for an area. The curvature of the surface 2.1 of the component 2 in or at the measuring spot is determined based on the distortion of the detected pattern 1 and calculated back to the distortion, area, and / or surface normal distribution of the effective measuring spot. This enables a mathematical correction of the measurement result to an effective, planar measuring spot.For example, in gloss level measurement, the reduction in the intensity of the reflected beam due to the radius deviation over the curved surface at the measuring point can be mathematically corrected to its equivalent RMS value on a flat sample surface.

Claims

1. A method for inspecting component surfaces for defects and / or determining component coating properties, particularly paintwork, with at least one probe (5), in which at least one pattern (1) of generally known geometry and dimensions is projected for testing onto at least one surface (2.1) and / or its surroundings using at least one projection light source (6), and the image (3) of the pattern (1) on the respective surface is detected using at least one optical detector unit (4) and, based on the captured image (3), a comparison is made with the original geometry of the pattern (1) and dimensions using mathematical relationships appropriate to the geometry and / or dimensions of the pattern (1) to determine the alignment (5a) of the probe (5) with the optical detector unit (4) in at least one spatial angle in terms of the respective surface normal (2a) and / or the distance (5b) of the probe (5) relative to the surface (2.1) for testing and / or the alignment of the measurement point of the probe (5) with the desired position on the component (2) and / or the alignment (5a) and / or the distance (5b) of the probe (5) determined in this way is corrected in relation to the respective surface or is / are taken into account when evaluating further measured values which are evaluated to determine any defects and / or faults or properties on component surfaces or component coatings, and for correcting measured values for measuring principles of the probe (5), which is not nearly ideally punctual, but rather in an extended area as a measuring spot, particularly by integrating or averaging measured values for an area, and which record at least one measured value, determining the curvature of the respective surface (2.1) of the component (2) in or near the measuring spot is based on the distortion of the captured pattern (1) projected onto the respective surface (2.1) and the distortion, surface area and / or its surface normal distribution of the effective measuring spot are calculated back to be able to perform a mathematical correction of the measurement result to an effective, flat measuring spot.

2. A method according to claim 1, characterised in that the gloss level, colour tone and / or coating thickness are determined by using the at least one probe (5).

3. A method according to at least one of the preceding claims, characterised in that, as a pattern (1), at least two light lines intersect at a generally known angle, at least one circle or at least one ellipse, concentric arrangements thereof or a dot or line pattern with generally known distances between dots and / or lines are projected onto a respective surface as a pattern (1).

4. A method according to at least one of the preceding claims, characterised in that the projection light source (6) and the data acquisition of the detector unit (4) and / or the measuring process of the probe (5) are triggered in a temporal sequence or at least partially synchronised with a control unit, whereby particularly a pulse-like short-term activation with high illuminance and short exposure times is performed.

5. A method according to one of the preceding claims, characterised in that the orientation (5a), the distance (5b) and / or the alignment of the probe (5) in relation to the respective surface (2.1) on which the at least one pattern (1) is projected is determined by measuring the intersection angles (1b), line lengths (1c), line distances (1d), position of the intersections (1c) of projected light lines or points, and / or the curvature (1d) of the image (3) of the respective pattern (1) and, when correcting the alignment (5a) and / or the distance (5b) of the probe (5) in relation to the respective surface (2.1) and / or when evaluating further measured values that are evaluated to determine any defects and / or faults or properties on component surfaces or component coatings.

6. A method according to one of the preceding claims, characterised in that a correction of the position and alignment (5a) of the at least one probe (5) in relation to the respective surface (2.1) is performed via an axis kinematic, particularly a robot.

7. A method according to one of the preceding claims, characterised in that, in the case of at least one hand-held probe (5), the determined alignment (5a), distance (5b) and / or alignment in relation to the respective surface (2.1), an operator is indicated in which direction or to which position the respective probe (5) is to be moved in order to facilitate the correct manual alignment of the probe (5) and / or to achieve a tolerance window around the desired target distance and / or alignment.

8. A method according to one of the preceding claims, characterised in that during and / or after re-shaping a component (2), the determination of any defects and / or faults or properties on component surfaces or component coatings is carried out to determine with at least one probe (5) at the reshaped areas of the surface (2.1) of the respective component faults and / or specific properties of a coating, of a sample material and / or defects .