Method for additive manufacturing of three-dimensional objects by means of arc fuse surfacing

Abstract

The invention relates to a method for the additive production of three-dimensional objects (2) by means of arc fuse surfacing, in particular by melting and solidifying a building material (4), which is in the form of a metal wire in the initial state thereof, layer by layer, during the production process, at least two different surfaces (11, 11 ') of at least one partially produced three-dimensional object (2) are detected by means of at least two different detection devices (6-10) and at least one construction quality parameter describing the construction quality of the object (2) is determined.

Description

Technical Field

[0001] The invention relates to a method for additively producing a three-dimensional object by means of arc-wire build-up welding, in particular by layer-by-layer melting and solidification of a building material which is in the form of a metal wire in its initial state. Background Art

[0002] Methods for additively manufacturing three-dimensional objects are known in principle from the prior art, in which arc-fused wire cladding is used to build up the object layer by layer. In this method, a metal wire is melted by means of an electric arc so that in the cooled state of the metal wire the desired cross section of the three-dimensional object to be manufactured is formed in each layer.

[0003] As in all additive manufacturing methods, process errors or anomalies may occur which negatively affect the quality of the component (e.g. the mechanical properties of the object). For example, cracks or voids may appear in the component, or oxidation of the building material may occur. In order to prevent this, the process can be monitored, for example, by employees visually monitoring the additive manufacturing of the object. In order to be able to carry out a visual inspection on the part of the employee, the production process usually has to be interrupted so that the employee can observe all relevant surfaces. Furthermore, it should be taken into account that the visual inspection by the employee is a subjective assessment or evaluation of the quality of the component.

[0004] Due to the visual inspection by employees and the interruption of the process, on the one hand, personnel costs are incurred, and on the other hand, the automation of the process is reduced due to the interruption and subjective evaluation of the employees. In particular, process errors can only be identified if they are already clearly present in the object so that the employees can visually recognize them. Summary of the invention

[0005] The object of the present invention is to specify a comparatively improved method for the additive production of three-dimensional objects by means of arc-wire deposition welding, in which method in particular the monitoring of the production process is improved.

[0006] This object is achieved by a method according to claim 1. The dependent claims relate to possible embodiments.

[0007] As described, the invention relates to a method for additively manufacturing a three-dimensional object by means of arc wire cladding, in particular by melting and solidifying a building material which is in its initial state in the form of a wire layer by layer. In other words, the wire can be fed at a defined feed rate and melted by means of an electric arc in order to form the layers of the three-dimensional object after the molten building material has cooled.

[0008] The invention is based on the knowledge that at least two different surfaces of at least one partially manufactured three-dimensional object are detected during the manufacturing process by means of at least two different detection devices and at least one build quality parameter describing the build quality of the object is determined. Advantageously, the method according to the invention allows monitoring of different surfaces, for example all surfaces, of a three-dimensional object that is partially manufactured in an additive manufacturing process. For example, images can be taken from different directions by means of different detection devices so that individual surfaces or sides of the three-dimensional object being manufactured can be detected. This allows, for example, automatic detection of whether the additive manufacturing process has been correctly carried out or whether process errors or anomalies have occurred during manufacturing, which can be corrected or remedied as early as possible if necessary.

[0009] Therefore, compared with the prior art described above, visual monitoring by employees is not required, but the at least two different detection devices can automatically monitor the compliance with the quality requirements of the components and, if necessary, can achieve early identification of process errors or abnormal situations. To this end, the object surface assigned to them is detected by the detection device and a construction quality parameter describing the construction quality of the object is determined based on the detection results. The construction quality can be defined, for example, in the following way, that is, whether the object is built without errors or whether there are errors in the object, in particular in the current or last manufactured layer, which require measures to be taken. If the construction quality parameter indicates, for example, that the construction quality of the object is fully met, the additive manufacturing process can be continued. On the contrary, if the construction quality parameter indicates that there is at least one influence related to the construction quality of the object, such as oxidation, cracks or gaps, then as described below, measures can be taken to ensure that a defined construction quality of the object is maintained or achieved.

[0010] The construction quality parameters for at least two surfaces, in particular all surfaces, of an object can be determined by means of detection devices that are oriented differently, in particular oriented at least 90° to each other. As described, at least two detection devices can be provided, which detect the object being manufactured from different directions. For example, the orientation of the central axis of the detection area can be understood as the orientation of the detection device. A plurality of detection devices can be provided, whose orientations or detection directions are oriented at 90° to each other. In particular, each side or surface of the object can be assigned a separate detection device. For example, in the case where the structural space or the construction volume is substantially rectangular or square, a detection device can be arranged on each surface. The detection device can be oriented along the surface normal of the side surface of the construction volume. For example, the method can be performed by means of four detection devices assigned to the side surfaces and one detection device assigned to the top surface. If a defect or a process error is identified on one of the surfaces, corresponding measures can be taken to eliminate or remedy the process error or reduce the occurrence of the process error.

[0011] According to another embodiment of the method, at least one defect in at least one surface of the object can be detected, and a build quality parameter can be determined based on the detected defect. As already described, the build quality parameter can in principle describe the build quality of the object. In particular, the build quality parameter can indicate whether a defect has occurred or is occurring during the additive manufacturing process. The build quality parameter can indicate, for example, where the defect occurred, or in which side surface of the object it occurred. Similarly, the build quality parameter can indicate the quality or type of the defect, for example, whether the defect can be corrected, whether measures need to be taken, or whether it will lead to an interruption of the construction process.

[0012] In principle, the build quality parameter can be determined based on raw data detected by a plurality of detection devices or a plurality of detection devices. According to one embodiment of the method, the build quality parameter can be determined based on at least one defect detected by a detection algorithm, in particular artificial intelligence ("AI") and / or computer vision and / or pattern recognition. The described detection algorithm can be applied, for example, to image data detected by means of a plurality of detection devices. In principle, the detection algorithm can be used to determine whether a defect is present in the corresponding surface of the object in the recorded or detected image data.

[0013] For example, raw data or image data detected by means of a detection device can be used as input data for the detection algorithm. The detection algorithm then detects whether a defect is present in the captured image data. If no defect is present, a build quality parameter can be output as output, which indicates that the build quality of the object being manufactured is sufficiently good. If a defect is identified, a corresponding build quality parameter can be output. Subsequently, as will be described further below, measures can be taken to correct the reduced build quality caused by the defect.

[0014] In the simplest case, in addition to the raw data recorded by the inspection device, the inspection algorithm can also be fed with a target state for the respective surface or the surface structure at the surface, so that the algorithm can determine deviations from the target state, for example, whether a defect is present. This can be performed, for example, with the aid of artificial intelligence, computer vision or pattern recognition. For example, a corresponding pattern that can be assigned to a defect can be fed to the inspection algorithm, which can detect the defect relative to the target state and thus output a correct build quality parameter.

[0015] In other words, the at least one defect can be detected based on a deviation of the inspected surface from a target state. For this purpose, it is in particular possible to feed the inspection algorithm with a target state of the surface, for example which shape the surface should have in its ideal state. The inspection algorithm can then be based on a comparison of the target state with the raw data recorded by the inspection device or the inspection device responsible for the inspected surface.

[0016] Before the detection algorithm is implemented, a learning process can be provided in advance, or the detection algorithm can be taught pattern errors in advance. For example, common or expected defects can be marked, that is, they can be fed to the detection algorithm in a classified manner, so that the detection algorithm can then independently detect and identify these defects. In addition to teaching the detection algorithm, a continuous learning process can also be provided, for example, if deviations from the target state cannot be clearly detected. By corresponding input of whether such ambiguous detections are defects, the detection algorithm can be continuously improved.

[0017] The detection algorithm described can in principle be applied to all raw data, which can be detected independently of one another for each surface of an object or each side of a build volume, as described above. This makes it possible, in particular, to automatically perform the detection and output of build quality parameters without interrupting the manufacturing process.

[0018] As described above, different measures can be taken according to specific construction quality parameters in order to continue the further manufacturing process. One possibility is to adjust at least one process parameter according to the construction quality parameter. In principle, any variable parameter in the additive manufacturing process can be understood as or adjusted as a process parameter. For example, the gas flow of the process gas used in the manufacturing process can be influenced, which means in particular the flow velocity or volume flow. Another possibility is to adjust the so-called trajectory planning in the additive manufacturing process. For example, trajectory planning is understood as, in the additive manufacturing process, at which positions and in what order a certain layer is selectively coated with material, or how to implement or select the movement speed of the processing head. Another possibility is to adjust the wire feed rate or the feed rate of the processing head. Thus, in particular, it can be set at which positions and how fast the building material is coated.

[0019] Furthermore, the waiting time or cooling time and the process temperature can be varied, for example, in order to prevent or reduce various process errors in the future. Another possibility can provide for adjusting the component properties, for example, changing the geometric parameters of the component or at least one process parameter as a function of the component properties. If, for example, it can be detected by a test in one of the test devices that the weld seam is implemented lower than it is defined in the target state, it can be determined from this that the process temperature is too high. In this case, in order to improve the quality of the build, more heat can be dissipated, or a longer waiting time can be planned between two layers in order to increase the dissipated heat and reduce the temperature.

[0020] According to a further embodiment of the method, provision can be made for the abort of the building process or the correction of certain defects to be carried out as a function of the build quality parameter. For example, if the build quality parameter indicates that a process error or defect is so severe that it cannot be corrected in the further course of the manufacturing process, an automatic abort of the building process can be achieved. If the build quality parameter indicates that a correction of a certain defect can be carried out, this correction can then be carried out. In principle, there is thus the possibility of remedying or repairing a detected defect.

[0021] In order to correct a specific defect, a correction of the defect can be carried out in at least one subsequent layer (particularly by changing at least one process parameter) and / or removal of at least one layer having the defect and application of a new layer (particularly by changing at least one process parameter) can be carried out. Thus, depending on a certain build quality parameter, for example in the case of minor process errors, an adjustment of the process parameters can be carried out so that the process errors that have occurred no longer occur in subsequent layers or can be corrected.

[0022] In contrast, if, for example, a more serious process error occurs, this can be corrected by removing the defective layer and applying a new layer. To this end, the defective layer is removed (for example by means of milling). Subsequently, in order to prevent the defect from reappearing in the newly applied layer, at least one process parameter can be changed. Then, a new layer can be applied with a new set of process parameters, thereby preventing the defect from reappearing. The device on which the method is implemented can, for example, have a device for tool replacement, by which a replacement to a removal device, such as a milling head, can be achieved. Alternatively, a corresponding tool can be provided in addition.

[0023] As described above, in principle any suitable detection device can be used to detect the surface of the object. According to a specific design, at least two different surfaces of the object can be detected by means of CCD and / or CMOS and / or thermal images (in particular based on color and / or based on temperature and / or based on depth information). Here, a basic combination of different detection devices is also possible. In particular, at least two different detection devices based on different detection mechanisms can be assigned to the same surface. In principle, it is also possible to assign different detection devices with different mechanisms to different surfaces.

[0024] In addition to the method, the present invention relates to a device for additively manufacturing three-dimensional objects by means of arc-fusible wire cladding, in particular by layer-by-layer melting and solidification of a building material which is in the form of a metal wire in its initial state, the device comprising a detection device with at least two detection devices with different orientations, the detection devices being constructed for detecting at least two different surfaces of at least one partially manufactured three-dimensional object during the manufacturing process, wherein the detection device is constructed for determining at least one build quality parameter describing the build quality of the object.

[0025] The test device thus comprises a plurality of test devices or at least two test devices, which can be assigned to different surfaces of the object, i.e. different sides of the build volume. This makes it possible to simultaneously detect process errors occurring on different surfaces of the object and to determine build quality parameters. To determine the build quality parameters, the device can, for example, have a separate control device, or the test device can have a control device, for example a processor, on which the above-described test algorithm can be implemented.

[0026] All advantages, details, embodiments and / or features described with regard to the method are fully transferable to the device and vice versa. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] The present invention is explained according to embodiments with reference to the accompanying drawings. The accompanying drawings are schematic diagrams and show:

[0028] Figure 1 A schematic diagram of a device for additively manufacturing a three-dimensional object by means of arc fuse deposition welding according to a first embodiment is shown;

[0029] Figure 2a-2c shows a schematic flow chart of a method according to a second embodiment; and

[0030] Figure 3a-3c A schematic flow chart of a method according to a third exemplary embodiment is shown. DETAILED DESCRIPTION

[0031] Figure 1A device 1 for additively manufacturing a three-dimensional object 2 is schematically shown, wherein the three-dimensional object 2 is manufactured layer by layer by means of arc fuse buildup welding. The device 1 has a processing head 3, for example a coating device, which is configured to melt a building material 4 which is in the form of a metal wire in its initial state when an electric arc is generated. In other words, the device 1 is configured to position the processing head 3 along the object 2 or relative to the object 2 at different positions and to apply the building material 4 layer by layer in order to additively manufacture the object 2.

[0032] The device 1 has a detection device 5 with a plurality of detection devices 6-10, which are configured to detect different surfaces 11 of a partially completed object 2 during the manufacturing process. The arrangement of the detection devices 6-10 is only exemplary and schematic and can be adjusted arbitrarily with respect to a specific object 2 to be manufactured or a specific device 1. In this embodiment, the detection devices 6-10 are assigned to different surfaces 11, 11' of the partially completed object 2 or are oriented toward these surfaces. In the specifically shown embodiment, the surfaces 11, 11' of the object 2 are respectively assigned to a detection device 6-10, and the detection device is configured to detect the surfaces 11, 11'. In the shown embodiment, four detection devices 6-9 for the side surfaces 11 and one detection device 10 for the top surface 11' are shown. For example, the detection devices 6-10 are oriented along the surface normal of the side surface 11 of the object 2 or the building volume (which is implemented as a square in this case).

[0033] A plurality of defects 12 that can be detected by one of the detection devices 6-10 are shown only by way of example. This illustration is to be understood only as an example, since, as described below in FIGS. 2 and 3 , when a defect 12 is detected, measures are taken so that the defect is corrected or remedied. To this end, the detection devices 6-10 can detect images of the surfaces 11 , 11 ′ of the objects 2 assigned to them and provide these images, for example, as raw data. The detection device 5 can have a control device, not shown in detail, for example in the form of a processor, on which a detection algorithm can be implemented. For example, the detected raw data can be provided to the detection algorithm. Subsequently, the detection algorithm can detect the defect 12. In particular, a construction quality parameter indicating the construction quality of the object 2 can be determined.

[0034] In this case, the inspection device 8 can output a construction quality parameter, which indicates that the surface 11, 11' assigned to the inspection device is produced without defects, so that the construction quality of the object 2 meets the quality requirements set for the surface assigned to the inspection device 8. However, the inspection devices 7, 10 detect defects 12 in this example, so that a construction quality parameter can be output accordingly, indicating that the quality requirements for the object 2 are not met. The defects 12 can be found, for example, by comparison with target states of the respective surfaces 11, 11', for example, the target states for the respective surfaces 11, 11' of the object 2 can be provided to the inspection device 5. Deviations from the target states can then be detected or determined as defects 12.

[0035] For example, the detection algorithm described above can be based on pattern recognition, AI or computer vision. In particular, various common forms of defects 12 can be taught during the teaching process, for example based on color or based on depth data or based on shape. As described, the manufacturing process can be continued as long as a certain build quality parameter does not output a defect 12, or as long as the build quality parameter confirms compliance with the required quality in terms of the partially manufactured object 2. If a defect 12 is detected, the measures described below with reference to Figures 2-3 can be taken by the device 1.

[0036] Figure 2a An object 2 is shown by way of example, in which a defect 12 in a currently produced layer is detected by means of one of the detection devices 6-10, for example the detection device 7, 10. The defect 12 may be, for example, an oxidation, a void or a crack in the object 2. Based on the detected defect 12, which is determined, for example, by pattern recognition in the raw data of the image recorded by the detection device 7, 10, a build quality parameter is output, which prevents further continuation of the additive manufacturing process and prompts correction of the defect 12.

[0037] like Figure 2a As shown, the previously produced layer with defects 12 is removed, in particular by means of a milling device 13. In this case, the device 1 can control the milling device 13 so that the layer carrying the previously detected defects 12 is removed and the additive manufacturing process can then be continued. The use of the milling device 13 should be understood as an example. The layer with defects 12 can also be removed in any other way.

[0038] exist Figure 2b, the state of a partially manufactured object 2 after the layer carrying the defect 12 has been removed. Based on the previously outputted and in particular determined build quality parameters based on the detected defect 12, changes can be made to the process parameters of the device 1. For example, the trajectory planning that determines the material application and the movement of the processing head 3 can be changed. In this case, in particular, the cooling time or waiting time between the material application at different locations or in different layers can be increased in order to ensure sufficient heat dissipation. If the defect 12 is caused, for example, by a seam in an area of ​​the object 2 that is implemented flatter than it should be in the target state, it can be determined on the part of the detection device 5 that the processing temperature of the building material 4 in this area is too high, so that the trajectory planning can be adjusted accordingly in order to increase the cooling time or waiting time.

[0039] Based on a changed process parameter or a changed set of process parameters, the previously removed layer can then be reapplied. Figure 2c The object shown in FIG. 1 is a 3D image of ...

[0040] As another alternative to suspending the building process or removing the layer having the defect 12, Figure 3a-3c It is shown how a defect 12 can be remedied in an upper layer. Figure 3a Only a state is shown by way of example, in which, for example, a defect 12 is again detected by the detection device 7, 10. Accordingly, a construction quality parameter which characterizes the defect 12 can be determined.

[0041] If this is, for example, a gap that can be corrected in a subsequent layer, this can be done as in Figure 3b For example, the defect 12 can be closed by filling the gap and another layer can be applied thereon. In addition, the process parameters can be adjusted based on the specified build quality parameters, which describe the manufacturing process performed by the device 1. For example, the wire feed rate can be adjusted. Figure 3c The manufactured object 2 is shown, wherein Figure 3a The defects shown in 12 have been corrected by adjusting the further manufacturing process (as in Figure 3b ) is remedied.

[0042] The advantages, details and features shown in the individual exemplary embodiments can be combined with one another, exchanged with one another and transferred to one another as desired. The device 1 is designed in particular to carry out the method described here. Reference numerals list

[0043] 1 device

[0044] 2 Objects

[0045] 3 processing head

[0046] 4. Construction Materials

[0047] 5. Testing Equipment

[0048] 6-10 Detection device

[0049] 11, 11' surface

[0050] 12 Defects

[0051] 13 Milling device

Claims

1. A method for additively producing a three-dimensional object (2) by means of arc wire deposition welding, in particular by melting and solidifying a building material (4) which is in the form of a metal wire in its initial state layer by layer, characterized in that During the manufacturing process, at least two different surfaces (11, 11') of at least one partially manufactured three-dimensional object (2) are inspected by means of at least two different inspection devices (6-10) and at least one build quality parameter describing the build quality of the object (2) is determined.

2. The method according to claim 1, characterized in that Build quality parameters for at least two surfaces (11, 11'), in particular all surfaces (11, 11'), of an object (2) are determined by means of differently oriented detection devices (6-10), in particular oriented at least 90° to one another.

3. The method according to claim 1 or 2, characterized in that: At least one defect (12) in at least one surface (11, 11') of the object (2) is detected and a build quality parameter is determined based on the detected defect (12).

4. The method according to any one of the preceding claims, characterized in that A build quality parameter is determined based on at least one defect (12) detected by a detection algorithm, in particular AI and / or computer vision and / or pattern recognition.

5. The method according to claim 4, characterized in that The at least one defect (12) is detected based on a deviation of the detected surface (11, 11') from a target state.

6. The method according to any one of the preceding claims, characterized in that At least one process parameter is adjusted as a function of the build quality parameter, in particular the gas flow and / or the trajectory planning and / or the wire feed rate and / or the feed rate of the processing head (3) and / or the waiting time between two applied layers and / or the cooling time and / or the process temperature and / or the component properties.

7. The method according to any one of the preceding claims, characterized in that Depending on the build quality parameters, the build process is stopped or specific defects (12) are corrected.

8. The method according to any one of the preceding claims, characterized in that In order to correct a specific defect (12), the defect (12) is remedied in at least one subsequent layer, in particular by changing at least one process parameter, and / or at least one layer having the defect (12) is removed and a new layer is applied, in particular by changing at least one process parameter.

9. The method according to any one of the preceding claims, characterized in that The at least two different surfaces (11, 11') of the object (2) are detected by means of a CCD and / or a CMOS and / or a thermal image, in particular based on color and / or based on temperature and / or based on depth information.

10. Device for additively manufacturing a three-dimensional object (2) by means of arc wire deposition welding, in particular by layer-by-layer melting and solidification of a building material (4) which is in the form of a metal wire in its initial state, characterized in that The device is provided with a detection device (5) having at least two detection devices (6-10) with different orientations, wherein the detection devices are configured to detect at least two different surfaces (11, 11') of at least one partially manufactured three-dimensional object (2) during a manufacturing process, wherein the detection device (5) is configured to determine at least one build quality parameter describing the build quality of the object (2).

Images