Method and device for calibrating a manufacturing apparatus, and system

Abstract

The invention relates to a method for calibrating a manufacturing apparatus. The manufacturing apparatus is configured for the additive manufacture of three-dimensional workpieces by layer-by-layer solidification of raw material by means of location-selective irradiation. The method comprises using a multiplicity of measurement results of optical measurements carried out with a time offset to estimate at least one property of an irradiation point or / and irradiation pattern to be provided, with each measurement result specifying at least one property of an irradiation point or / and irradiation pattern at the respective measurement time. The method also comprises a calibration of the manufacturing apparatus on the basis of the estimated at least one property. The invention also relates to a device and a system.

Description

[0001] Method and device for calibrating a manufacturing apparatus, as well as system

[0002] The present invention relates to a device for calibrating a manufacturing apparatus which is designed for the additive manufacture of three-dimensional workpieces by layer-by-layer solidification of raw material by means of location-selective irradiation, a method for calibrating such a manufacturing apparatus, and a system.

[0003] Additive manufacturing, which involves solidifying successive layers of raw material, allows three-dimensional workpieces to be produced layer by layer by creating a sequence of solidified and interconnected layers. Such processes differ, among other things, in the type of raw material used and / or the method of solidifying the raw material to produce the workpiece.

[0004] One example of such a layer-based additive manufacturing process is powder bed consolidation. This involves processing layers of raw material, particularly powdered, metallic and / or ceramic raw materials, into three-dimensional workpieces. For this purpose, a raw material powder layer is applied to a build platform or carrier and solidified in a location-selective manner depending on the desired geometry of the workpiece to be produced. For location-selective solidification, the powder layer can in particular be irradiated, e.g. with laser radiation or an electron beam. The radiation striking the powder layer causes it to heat up and thus melt or sinter the powder particles. After the desired areas of the powder layer have solidified, further raw material powder layers are then successively applied and location-selectively solidified until the workpiece has the desired shape and size.Selective laser melting or laser sintering can be used in particular for the production of prototypes, tools, spare parts or medical prostheses, such as dental or orthopedic prostheses, based on CAD data.

[0005] In contrast to powder bed consolidation, other additive manufacturing processes do not involve layer-by-layer consolidation of raw material using site-selective irradiation. For this purpose, for example, an unconsolidated raw material can be fed to a print head, which applies the material to a carrier, where it solidifies. An important parameter in additive layered manufacturing processes is the quality of the manufactured workpieces. Therefore, one aim of the present invention is to improve, in particular, the quality of workpieces manufactured using an additive manufacturing apparatus configured for the additive manufacturing of three-dimensional workpieces by layer-by-layer consolidation of raw material using site-selective irradiation.

[0006] According to a first aspect of the present disclosure, a device for calibrating a manufacturing apparatus is provided, wherein the manufacturing apparatus is configured for the additive manufacture of three-dimensional workpieces by layer-by-layer solidification of raw material by means of location-selective irradiation.

[0007] The manufacturing apparatus is particularly designed for powder bed consolidation and comprises an irradiation system for providing an irradiation point and / or irradiation pattern. The irradiation system is particularly designed to irradiate the raw material in a location-selective manner in order to solidify it. The manufacturing apparatus can be designed for additive manufacturing by means of powder bed melting and / or sintering. The irradiation system comprises, for example, one or more irradiation sources, in particular one or more lasers. The irradiation system can be designed to provide a laser irradiation point and / or a laser irradiation pattern, in particular in a construction area of ​​the manufacturing apparatus, for example on a powder layer of raw material. The irradiation system is particularly designed to provide one or more beams which, upon impact with a surface (e.g.on the surface of a powder layer) form the laser irradiation point and / or the laser irradiation pattern.

[0008] The manufacturing device may include one or more sensors designed to collect status data. The status data may indicate a status and / or a change in status of the manufacturing device.

[0009] A state can correspond to a manufacturing process or a specific part of a manufacturing process that is carried out by the manufacturing apparatus. A state can correspond to a planned process break during which no irradiation is carried out. Such a process break can serve, for example, for a cleaning process (e.g. of a process gas filter) of the manufacturing apparatus or to ensure predetermined process parameters (e.g. protective gas flow and / or temperature). However, a state can also correspond to an unplanned process break. Such a process break can, for example, be caused by a malfunction of a manufacturing process carried out by the manufacturing apparatus. The manufacturing apparatus can comprise a powder layer monitoring device that is configured to detect such a malfunction if a powder layer is or has been applied incorrectly.

[0010] It is conceivable that the status data indicate a change in an irradiation process, a change in a process gas atmosphere, a coating process with raw material, and / or a change in the position of a build platform of the production apparatus. The irradiation process can be defined by an orientation of exposure vectors and / or an exposure vector feed direction in which exposure vectors are arranged and exposed for exposing a powder layer. Such sensors include, in particular, optical sensors such as light detectors and cameras, gas pressure sensors, gas sensors for detecting the composition of a gas mixture, position sensors, motion sensors, and temperature sensors.

[0011] The production apparatus may include a signal interface via which such status data is transmitted to a control unit. It is also conceivable that the status data is provided by a control unit of the production apparatus that controls components of the production apparatus.

[0012] The manufacturing apparatus may comprise an optical sensor, in particular a spatially resolving light sensor such as a camera. The optical sensor is in particular configured to detect, in a spatially resolved manner, radiation which (a) is emitted by the irradiation system and reflected by a surface (e.g., a surface of an irradiated powder layer), (b) is emitted by the irradiation system and scattered at a surface (e.g., a surface of an irradiated powder layer), (c) comprises thermally induced heat radiation induced by radiation emitted by the irradiation system impinging on a surface (e.g., a powder layer surface), and / or (d) comprises process radiation originating in particular from a melt pool caused by radiation emitted by the irradiation system impinging on a surface (e.g., a powder layer surface).The optical sensor is in particular designed to be sensitive to radiation that falls within a predetermined wavelength range. The predetermined wavelength range can be selected to comprise one or more wavelengths of radiation, wherein this radiation (i) can be emitted by the irradiation system, (ii) is emitted by the irradiation system and reflected by a surface (e.g., a surface of an irradiated powder layer), (iii) is emitted by the irradiation system and scattered at a surface (e.g., a surface of an irradiated powder layer), (iv) comprises thermally induced heat radiation that is generated by radiation emitted by the irradiation system striking a surface (e.g.,a powder layer surface) is induced, and / or (v) comprises process radiation which originates in particular from a melt pool which is caused by radiation emitted by the irradiation system impinging on a surface (e.g. a powder layer surface). In particular, the optical sensor is designed to detect an irradiation point provided by the irradiation system (e.g. on the powder layer) and / or an irradiation pattern provided by the irradiation system. For this purpose, the optical sensor can be designed and arranged such that it can record an image of a section of the powder layer or of the entire powder layer which depicts the irradiation point and / or the irradiation pattern. At least one property of a provided irradiation point and / or irradiation pattern can be detected by the optical sensor.The optical sensor can be controlled by a control unit that controls the irradiation system. The sensor can be provided as part of the irradiation system.

[0013] The manufacturing apparatus, in particular the irradiation system, may comprise a lens, in particular a dynamic lens. The lens may be designed such that the focus provided by the lens is controllable (e.g., adjustable by an electronic control signal). The lens may be a liquid lens, for example, a focus-tunable liquid lens. The lens may be associated with the optical sensor. In particular, the lens is arranged such that the optical sensor can detect an irradiation point provided by the irradiation system (e.g., on the powder layer) and / or an irradiation pattern provided by the irradiation system through the lens. In other words, the lens may be arranged in an optical path between the optical sensor and the build area and / or the powder layer. For example, the optical sensor can capture an image through the lens.The lens can be configured and / or controlled so that the focus of the optical sensor lies in a processing plane, in particular on the surface of a powder layer currently to be irradiated.

[0014] Instead of a single optical sensor, it is also conceivable to provide multiple optical sensors, each configured as described above. In this case, multiple lenses can also be provided, each lens being assigned to a different one of the optical sensors.

[0015] The manufacturing apparatus may further comprise at least one reference element. The at least one reference element may be arranged such that it lies outside the powder layer and / or outside the build area and / or outside the processing plane. The at least one reference element may be installed at a predetermined position. The predetermined position may be fixed relative to a surface of a powder layer to be irradiated. The optical sensor may be configured to capture an image depicting one or more of the at least one reference element. It is conceivable that such an image may additionally include an image of a provided irradiation point and / or irradiation pattern.

[0016] The manufacturing apparatus comprises, in particular, a first deflection unit. Alternatively or additionally, the manufacturing apparatus can comprise a second deflection unit and / or a third deflection unit. Such a deflection unit can each comprise a (e.g., 2-axis) scanning system which comprises one or more controllably movable mirrors. The deflection units can be designed such that they can be aligned to a common (e.g., overlapping) processing area. Thus, it is possible to direct an energy beam into the processing area and / or to direct a field of view of an associated optical sensor into the processing area via each deflection unit. The processing area can comprise a processing plane, e.g., a surface of at least a part or section of a powder layer.

[0017] In a first variant, the optical sensor is designed such that it always detects the same spatial area and / or the same part of a powder layer. In this variant, it can be provided that the optical sensor receives radiation emanating from the powder layer directly, i.e. without this radiation being directed to the optical sensor via one of the deflection units. In a second variant, the first deflection unit is assigned to the optical sensor. The first deflection unit can be designed to controllably define a field of view of the optical sensor, in particular an image recording range of the sensor relative to the powder layer and / or a reference element. This makes it possible to control the field of view of the optical sensor such that the irradiation point and / or the irradiation pattern and / or the reference element or one or more of the reference elements can be seen in the recorded image.In this variant, the optical sensor receives radiation emanating from the powder layer indirectly, i.e. only after it has been directed to the optical sensor via the first deflection unit.

[0018] The optical sensor can be arranged off-axis with respect to a beam provided for irradiating a powder layer. If multiple beams are provided, the optical sensor can be arranged off-axis with respect to all or only a first subset of the multiple beams. In the latter case, it is conceivable for the optical sensor to be arranged on-axis with respect to a second subset of the multiple beams. The optical sensor can, in particular, be arranged off-axis with respect to a beam that can be provided via the second or third deflection unit, and can optionally be arranged on the axis of a beam that can be provided by the first deflection unit.

[0019] The at least one property of a provided irradiation point and / or irradiation pattern can be detected by the optical sensor, in particular via the first deflection unit. Likewise, a position of the reference element can be detected by the optical sensor, in particular based on an image recorded by the sensor containing a depiction of the reference element. It is also conceivable for the first deflection unit to be configured to position an irradiation point and / or irradiation pattern on the powder bed. In this case, an irradiation source can be assigned to the first deflection unit, and the first deflection unit fulfills a dual function, so to speak.

[0020] The corresponding irradiation point and / or the corresponding irradiation pattern, which can be detected by the optical sensor (e.g. via the first deflection unit), can be positioned via the second deflection unit. In other words, the second deflection unit can be designed to define a direction in which (laser) radiation is emitted in the direction of a powder layer. The second deflection unit can be designed to generate a corresponding irradiation pattern by deliberately changing an impact position of the radiation on the powder layer, which can be depicted in an image recorded by the optical sensor. Alternatively, it is conceivable for an irradiation pattern to be directed to the second deflection unit, and for the second deflection unit to merely define its position on the powder layer.

[0021] The third deflection unit can be configured like the second deflection unit. For example, the second and third deflection units are supplied with different energy beams, in particular with respective laser beams, which can be provided by the laser source(s) of the irradiation system. It is conceivable that the corresponding irradiation point and / or the corresponding irradiation pattern can be positioned selectively or via the second deflection unit or the third deflection unit. A corresponding irradiation point and / or a corresponding irradiation pattern can also be provided simultaneously via the second and third deflection units and / or imaged by the optical sensor.

[0022] The manufacturing apparatus can comprise additional appropriately designed deflection units to enable workpiece production by simultaneous irradiation with multiple energy or laser beams. As mentioned above, it is conceivable for the manufacturing apparatus to comprise not just one, but multiple optical sensors. In this case, each optical sensor can be assigned to a different one of the deflection units and detect an irradiation point provided (e.g., via a deflection unit not assigned to the respective sensor) and / or irradiation pattern via the deflection unit assigned to the respective sensor. It is possible to provide one, two, three, or even each of the deflection units with a corresponding optical sensor and, optionally, with a respective lens.

[0023] The device includes a control unit for controlling the irradiation system.

[0024] The control unit is configured to estimate at least one property of an irradiation point and / or irradiation pattern to be provided based on a plurality of measurement results of optical measurements performed at different times (e.g., using the optical sensor), wherein each measurement result indicates at least one property of an irradiation point and / or irradiation pattern at the respective measurement time. The control unit is further configured to calibrate the additive manufacturing apparatus based on the estimated at least one property. The control unit is configured, in particular, to carry out the method according to the third aspect described below.

[0025] The control unit can be configured to receive the status data and / or control the optical sensor, for example, to capture one or more images. The control unit can be configured to control the deflection unit(s), for example, to specify and / or adjust a position of an irradiation point and / or irradiation pattern provided via a controlled deflection unit. The control unit can also be configured to control the deflection unit(s) to specify and / or adjust a field of view of the optical sensor associated with this controlled deflection unit(s). The control unit can further be configured to control the lens(es), in particular to adjust the focus of the lens(es). The control unit can be configured to control further components of the irradiation system and / or the manufacturing apparatus.

[0026] According to a second aspect of the present disclosure, a system is provided. The system comprises the device according to the first aspect and the manufacturing apparatus.

[0027] According to a third aspect of the present disclosure, a method for calibrating a manufacturing apparatus configured for the additive manufacturing of three-dimensional workpieces by layer-by-layer solidification of raw material using location-selective irradiation is provided. For example, the method is performed by the device according to the first aspect, in particular by the control unit. The device, the control unit, and / or the manufacturing apparatus, as well as its components, can be configured as described above for the first aspect.

[0028] The method comprises estimating at least one property of an irradiation point and / or irradiation pattern to be provided based on a plurality of measurement results from optical measurements performed at different times. Each measurement result indicates at least one property of an irradiation point and / or irradiation pattern at the respective measurement time. Since a plurality of measurement results are used, an accurate and reliable estimation of the at least one property can be ensured, in contrast to a solution that only considers a single measurement.

[0029] The estimation can be performed by a computational neural network (CNN). The estimation can be based on an algorithm trained by machine learning.

[0030] At least one of the optical measurements can comprise a recorded image, wherein the irradiation point and / or the irradiation pattern at the time the image is recorded is depicted in the image. It is conceivable to record images at different measurement times. These images can be recorded by the optical sensor, in the first variant by directly imaging radiation reflected from the powder bed or, according to the second variant, by imaging radiation reflected from the powder bed and guided to the optical sensor via the first deflection unit. If several optical sensors are present, it is conceivable to record images at different times by different optical sensors and to use this series of images as (e.g. part of) the measurement(s) carried out at different times.

[0031] An irradiation point can generally be understood as a point on a surface onto which an energy beam impinges, in particular a point in the processing plane that is irradiated by a laser beam. Typically, an irradiation point is formed by exactly one energy beam. Thermally induced radiation reflected, scattered, or emitted at the irradiation point, or radiation emitted by a melt pool located at the irradiation point, can be detected by the optical sensor. An irradiation pattern can be understood as a pattern consisting of multiple irradiation points.

[0032] The at least one property of the irradiation point and / or irradiation pattern may comprise a property in the processing plane. For example, the at least one property comprises a geometric feature or a geometric property such as a shape (e.g., contour) and / or a position (e.g., in the processing plane and / or in the field of view of the optical sensor) and / or an intensity distribution and / or a spectrum. The position may be defined relative to another irradiation point and / or another irradiation pattern and / or a reference element.

[0033] The method further comprises calibrating the production apparatus on the basis of the estimated at least one property. During calibration, the production apparatus can be absolutely calibrated or adjusted (e.g., based on a gold standard). Alternatively, the production apparatus can be calibrated by a relative calibration of multiple components of the production apparatus (e.g., by adjusting multiple deflection units and / or radiation sources). In this case, the production apparatus can be adjusted on the basis of the estimated at least one property such that the irradiation point and / or irradiation pattern to be provided by the irradiation system has a predetermined or desired property. In particular, if the estimated at least one property of the irradiation point and / or irradiation pattern to be provided by the irradiation system is determined by one or more desired (e.g.,If the manufacturing process deviates from the predefined properties, the manufacturing apparatus can be calibrated to minimize or completely prevent this deviation. Calibration can involve determining and / or storing one or more calibration parameters that are used in the subsequent provision of an irradiation point and / or irradiation pattern to ensure its desired property(ies).

[0034] Calibration parameters can influence the control of the deflection unit(s). In particular, calibration parameters are determined and / or stored, which can be used for subsequent control of the deflection unit(s). Calibration parameters can minimize a deviation of a position of an irradiation point and / or irradiation pattern provided via the correspondingly controlled deflection unit from a desired position. The calibration parameters can specify deviation values ​​and / or correction values ​​for one or more of the deflection units to compensate for these deviation values, wherein the deviation values ​​indicate deviations of planned irradiation positions from irradiation positions actually irradiated via the corresponding deflection unit. The deviation values ​​and / or correction values ​​can indicate a distortion of a scan field of the corresponding deflection unit.The deviation values ​​and / or correction values ​​can alternatively or additionally specify a translation and / or rotation of the scan field of the corresponding deflection unit. The translation and / or rotation can be defined relative to an absolute reference, for example, relative to a predefined point in the production apparatus (e.g., a position defined relative to the calibration element or a position defined relative to the build platform). It is also conceivable that the translation and / or rotation is defined relative to another of the deflection units, in particular relative to a coordinate system or a scan field of another of the deflection units.

[0035] Other properties of the irradiation point and / or irradiation pattern to be provided, such as a shape, a spectrum, or an intensity, can also be adjusted using calibration parameters. Calibrating and / or recalibrating the production apparatus can include adjusting at least one irradiation power, adjusting at least one beam shape (e.g., beam cross-section) of an energy or laser beam, and / or adjusting an intensity distribution in at least one energy or laser beam.

[0036] The method can comprise selecting and / or weighting measurement results from the plurality of measurement results. For example, the at least one property is estimated based on the selected and / or weighted measurement results. Thus, to estimate the at least one property, individual measurement results can be neglected and / or the measurement results can be prioritized differently. The selection can comprise filtering the measurement results. In this case, (e.g., statistical) outliers can be filtered out and disregarded in the subsequent estimation. It is also conceivable to take such outliers into account for the estimation, but to give them a lower weighting than other measurement results. The selection and / or weighting, in particular the filtering, can be based on a recursive least squares algorithm (RLS algorithm), a Wiener filter and / or a Kalman filter.

[0037] The selection and / or weighting can be performed by a computational neural network (CNN). The selection and / or weighting can be based on an algorithm trained by machine learning or an algorithm trained by machine learning.

[0038] The measurement results can be selected (e.g., filtered) and / or weighted based on information that is at least partially predetermined. Alternatively or additionally, the information can be obtained at least partially from a manufacturing process currently being performed by the manufacturing apparatus. The information can include the status data. The information can be assigned selection criteria, weighting rules, weighting factors, and / or filter criteria that are used for selection and / or weighting. The selection criteria, weighting rules, weighting factors, and / or filter criteria can be predetermined or determined by a recursive least squares algorithm (RLS algorithm), a Wiener filter, and / or a Kalman filter.

[0039] The information can be associated with a workpiece to be manufactured by the manufacturing apparatus. The information can comprise at least one property of the workpiece to be manufactured. This at least one property can comprise a (e.g., minimum, maximum, or average) layer thickness and / or a material and / or a shape and / or a volume of at least one part (e.g., a layer) of the workpiece to be manufactured.

[0040] The layer thickness can be a determined or actual layer thickness of a powder layer or of a powder layer section of the workpiece currently being manufactured. To determine the actual powder layer thickness, a light projection system (e.g., a strip light projection system) can be provided (e.g., as part of the manufacturing apparatus) that is designed to emit light onto a surface of a powder layer so that, based on one or more images of the powder layer illuminated in this way, surface information (e.g., a surface topography) of the raw material layer or a section thereof can be acquired (e.g., by the projection system and / or the control unit described herein). This surface information can, for example, be recorded after the powder has been applied and / or after exposure of the powder layer or raw material layer.For example, surface information of a first powder layer onto which a second powder layer is to be applied and / or surface information of the second powder layer can be acquired. The difference between the surface planes of the first and second powder layers indicated by this surface information can then be interpreted as the layer thickness of the second powder layer. The same approach is of course also possible for two layer sections lying one above the other. Typically, the build platform of the production apparatus is lowered before the second powder layer is applied. Such a height shift of the surface of the first powder layer can be taken into account when determining the layer thickness of the second powder layer, particularly if the surface information of the first powder layer is acquired before it is lowered.More generally, the surface information acquired using the structured light projection system enables the determination of an actual powder layer thickness (e.g., the second powder layer). The resulting powder layer thickness can be used to select suitable measurement results from previous measurements or to give them a higher weighting.

[0041] The information and / or the at least one property can comprise a layer thickness of an unirradiated powder layer section, a layer thickness of an irradiated powder layer section, and / or a difference between these two layer thicknesses. In this case, the information can be obtained based on the surface information of the (un)irradiated powder layer, which was determined, for example, using the stripe light projection system. In other words, the difference can indicate a height offset between (e.g., a section) of the raw material powder layer and a solidification plane of the irradiated raw material powder (e.g., located directly below the section).

[0042] The information can indicate the current number of layers to be irradiated, the total number of layers (e.g., of the workpiece to be manufactured), and / or a ratio of these two values. The information can indicate a predicted or recorded raw material powder layer thickness. In particular, the information can indicate the current settling phase of the layer thicknesses.

[0043] The information may specify an area of ​​a layer of the workpiece to be manufactured that is to be irradiated. The information may specify a change in this area between two or more consecutive layers of the workpiece to be manufactured, in particular if the change is above a predefined limit. The information may specify wall thicknesses of the workpiece to be manufactured, for example, a proportion of wall thicknesses that are below a predetermined threshold and / or a proportion of wall thicknesses that are above one or more of the predetermined thresholds. The wall thicknesses may be specified for the workpiece to be manufactured or for one or more layers thereof.

[0044] Alternatively or additionally, the information may comprise at least one process parameter of the manufacturing apparatus. The information may comprise (e.g., at least) one process parameter of a manufacturing process, in particular a currently being performed one, for the additive manufacturing of the workpiece to be manufactured. The (e.g., at least) one process parameter may comprise an irradiation intensity, a laser power, a process gas temperature, a powder temperature, an irradiation duration, and / or a scanning speed. The information may comprise other process parameters of the irradiation system and / or the manufacturing apparatus that can be adjusted during the manufacture of the workpiece.

[0045] The information may alternatively or additionally specify a scanning angle, a laser incidence angle, a lateral offset of a laser position between a surface plane of the unirradiated powder material and a surface of irradiated, solidified powder material, a melt hole shape and / or a melt pool shape.

[0046] For example, the at least one property is estimated based on a temporal progression of the at least one property (e.g. position) that is specified by the measurement results used for the estimation. The at least one property can be estimated based on a temporal progression of the same property (e.g. position) or the same type of property. Alternatively or additionally, the measurement results can be selected and / or weighted based on the temporal progression of the at least one property that is specified by the measurement results used for the estimation. The at least one property can be selected and / or weighted based on the same property or the same type of property. The at least one property specified by the respective measurement result is assigned to the corresponding measurement time. A time series of the at least one property can be created based on the plurality of measurement results.A function can also be determined that approximates the at least one property specified by the measurement results, e.g., depending on time or one or more function parameters (e.g., the information mentioned above). The at least one property of a still-to-be-provided irradiation point and / or irradiation pattern can then be estimated based on the time series and / or the function.

[0047] In one example, estimating the at least one property corresponds to forecasting. The estimation can comprise averaging, intrapolating, and / or extrapolating the properties specified by the measurement results used for estimation. In the case of a time series, the at least one property can be estimated for future points in time, in particular by extrapolation. In the case of a function, a value of the function (e.g., a position of the irradiation point to be provided, specified by the function) can be estimated for a future intended value of the function parameter (e.g., a laser power intended for providing the irradiation point during production of the next layer) by averaging or interpolating the remaining function values.Other approaches to estimating the at least one property by averaging, intrapolating and / or extrapolating the properties indicated by the measurement results used for estimation are also possible.

[0048] For example, a change in the state of the production equipment occurs between two or more optical measurements performed at different times. In one example, the change in the state of the production equipment is caused by the completion or start of a manufacturing step performed by the production equipment to manufacture a workpiece (e.g., the workpiece to be manufactured in the currently performed manufacturing process). The change in state may be accompanied by a change in one or more process parameters. The change in state may be a change in an irradiation process, a change in a process gas atmosphere, a coating process with raw material, and / or a change in the position of a build platform of the production equipment. The change in state may be indicated by the state data.The method may further comprise providing the irradiation point and / or irradiation pattern to be provided by the calibrated manufacturing apparatus. This comprises, in particular, irradiating the surface of a powder layer in the irradiation point and / or irradiation pattern. During the provision, in particular, the irradiation point and / or the irradiation pattern are provided with the desired property. Without prior calibration, the irradiation point and / or the irradiation pattern could instead have the estimated property. The irradiation point and / or the corresponding irradiation pattern provided by the calibrated manufacturing apparatus can be used to manufacture the workpiece, in particular for the location-selective solidification of raw material of a powder layer.

[0049] The method may comprise performing an optical measurement to acquire a measurement result that indicates at least one property of the irradiation point and / or irradiation pattern provided (e.g., by the calibrated manufacturing apparatus). The optical sensor may capture an image depicting the provided irradiation point and / or irradiation pattern. As mentioned above, this image may be captured via the first deflection unit, while the irradiation point and / or irradiation pattern is directed via another (e.g., the second or third) deflection unit, in particular projected into a processing area.

[0050] Based on this optical measurement carried out after calibration, the production apparatus can be recalibrated. The method can therefore comprise recalibrating the production apparatus based on the at least one property of the provided irradiation point and / or irradiation pattern. The measurement can be carried out during a manufacturing process for producing a workpiece. It is conceivable that the production apparatus is calibrated (e.g. only) based on previously known measurement results (e.g. from a reference construction process), in particular if no other measurement results are used to estimate the at least one property. However, the calibration can also be carried out based on measurement results of optical measurements that were carried out during a manufacturing process for producing the workpiece. The recalibration can take place during the manufacturing process for producing the workpiece.When recalibrating, it is possible to only take into account the last optical measurement from the previous optical measurements (e.g. including the previously known measurement results from a reference construction process).

[0051] The acquired measurement result (e.g., after calibration) can be stored as part of the plurality of measurement results. This allows this measurement result to be considered for future estimations of the at least one property. In other words, the number of measurement results used for estimation can be gradually increased, whereby the manufacturing equipment can be (re-)calibrated between the respective optical measurements.

[0052] Based on the estimated at least one property and / or based on the calibration of the manufacturing apparatus, at least one of the following can be specified: a desired property of the irradiation point to be provided (e.g., in the future or subsequently) and / or of the irradiation pattern to be provided (e.g., in the future or subsequently); a temporal offset and / or a number of optical measurements to be performed, in particular for recalibration; a spatial distribution of optical measurements to be performed, in particular for recalibration. The spatial distribution can comprise one or more fields of view of the optical sensor and / or one or more positions of the irradiation point and / or irradiation pattern to be provided.The spatial distribution can be defined by deflections of the first, second and / or third deflection unit, which are intended for the optical measurements to be carried out for recalibration.

[0053] As already mentioned in the first aspect, the manufacturing apparatus can comprise an optical sensor, a first deflection unit, and a second deflection unit and / or a third deflection unit. The at least one property of an irradiation point and / or irradiation pattern can be detected by the optical sensor via the first deflection unit, and the corresponding irradiation point and / or the corresponding irradiation pattern can be positioned via the second deflection unit, in particular at a position in the processing area and / or on a powder surface. Alternatively or additionally, the corresponding irradiation point and / or the corresponding irradiation pattern can be positioned selectively via the second deflection unit or the third deflection unit.

[0054] Calibrating and / or recalibrating the production apparatus can include aligning the first deflection unit with the second deflection unit and / or aligning the second deflection unit with the third deflection unit and / or aligning the first deflection unit with the third deflection unit. In this process, a coordinate transformation between projection coordinate systems of the aligned deflection units can be determined. The alignment can minimize or completely avoid an undesired offset between positions of irradiation points and / or irradiation patterns to be provided via the two deflection units. The alignment of two deflection units can be performed such that the aligned deflection units are calibrated relative to one another.However, the adjustment can also be carried out in such a way that both deflection units are absolutely calibrated, for example relative to at least one point in the processing area, relative to at least one reference element and / or relative to at least one point on the powder layer to be solidified.

[0055] These and other aspects of the invention are described in more detail below with reference to the figures, in which:

[0056] Fig. 1 shows a schematic representation of a manufacturing apparatus according to the present disclosure;

[0057] Fig. 2 shows a flowchart of a method according to the present disclosure;

[0058] Fig. 3 shows a schematic representation for explaining a height offset;

[0059] Fig. 4 shows a schematic representation to explain a transient behavior of layer thicknesses;

[0060] Fig. 5 shows a schematic representation for explaining a lateral offset as a function of a laser incidence angle;

[0061] Fig. 6 shows a schematic diagram to explain different melt hole and melt pool shapes; and

[0062] Fig. 7 shows a timeline with exemplary measurement times.

[0063] The system 2 of Fig. 1 comprises an additive manufacturing apparatus 3 which is used for

[0064] Powder bed consolidation is set up and an irradiation system 4 for providing an irradiation point and / or irradiation pattern in a processing area 6. If a powder layer 8 made of raw material lies on a carrier 10 of the production apparatus, the irradiation point and / or the irradiation pattern can be provided on the surface 12 of the powder layer 8, which in this case lies in the processing plane 13 of the production apparatus 2. To provide the irradiation point and / or irradiation pattern, the irradiation system 4 comprises a plurality of irradiation sources 14, 16, 18, the emitted radiation of which can be guided into the processing area 6 via respective deflection units 20, 22, 24 as directed beams 21, 23, 25.The deflection units 20, 22, 24 serve in particular to position a laser beam on the surface 12 of the powder layer 8 for the location-selective melting or sintering of the powder material and can each comprise a 2-axis scanning system with one or more controllably movable mirrors. The manufacturing apparatus 3 further comprises optical sensors 26, 28, each of which can capture at least part of the processing area using a dynamic liquid lens 30, 32 with adjustable focus. The field of view of the optical sensors 26, 28 can be shifted by the deflection units 20, 24 assigned to them. Fig. 1 shows, by way of example, a central viewing axis 34 of the optical sensor 26. Depending on the arrangement of the optical sensor and the design of the deflection unit 20, the central viewing axis 34 can deviate from or coincide with the path of the directed beam 21.A reference element 36 is installed at a predetermined position next to the carrier 10. The system 2 further comprises a device 38 for calibrating the production apparatus 3. The device 38 comprises a control unit 40 for controlling the irradiation system 4.

[0065] Fig. 2 shows a flowchart of a method according to the present disclosure. The method can be executed on the system 2, in particular by the control unit 40 of the device 38.

[0066] The method is based on the use of a plurality of measurement results from optical measurements performed at different times. These measurement results include, in particular, images recorded of the processing area by the optical sensors 30, 32 through the dynamic lenses 30, 32. These images each comprise an image or reproduction of an irradiation point and / or irradiation pattern provided by the irradiation system 4 at the time of image capture and can further comprise an image of the reference element 36. Consequently, the images indicate at least one property of the provided irradiation point and / or irradiation pattern. The at least one property can be specified by the irradiation system 4, in particular by the irradiation system 4.Examples of the at least one property include a position, shape, intensity (distribution), and a spectrum, with the position being considered particularly advantageous as a property. Therefore, although the present disclosure is not limited thereto, the position will be referred to as the at least one property in the following.

[0067] In an optional step 202, measurement results are selected and / or weighted from the plurality of measurement results. The selection may include filtering the plurality of measurement results. Measurement results that are particularly meaningful for a subsequent estimation of at least one property of an irradiation point and / or irradiation pattern to be provided are selected or given a higher weighting.

[0068] The measurement results can be selected and / or weighted based on information that is at least partially predetermined. Alternatively or additionally, the information can be obtained at least partially from a manufacturing process currently being performed by the manufacturing apparatus 3. If, for example, individual measurement results (e.g. from a previous reference manufacturing process or from the currently performed manufacturing process) are assigned to a laser power of 1000 W and others to a laser power of 500 W, then in the case of an irradiation point to be provided subsequently (e.g. in the currently performed manufacturing process) with a laser power of 500 W, the measurement results assigned to a laser power of 500 W can be selected and / or weighted higher. Instead of or in addition to the laser power, other (process) parameters assigned to the measurement results can also be taken into account.It is also conceivable to use statistical criteria to select or weight the measurement results. This allows statistical outliers to be filtered out, in particular.

[0069] The information used for selection and / or weighting can be associated with a workpiece to be manufactured by the manufacturing apparatus, and in particular can include a shape, a material, a volume, and / or a layer thickness of the workpiece. For example, the workpiece to be manufactured can have layer thicknesses of 200 μm. In this case, measurement results obtained from a powder layer with a thickness of 50 μm can be assigned a lower weighting or not selected for the subsequent estimation. In contrast, measurement results associated with a layer thickness of 150 μm can be selected and / or assigned a higher weighting. The information can indicate a height offset between the raw material powder layer and a solidification plane of the irradiated raw material powder. This height offset arises in particular from a compaction of the powder layer during exposure of the powder when the powder layer is remelted into a denser layer.

[0070] This effect is illustrated in Fig. 3. As shown there, the unirradiated raw material powder layer 326 (which corresponds, for example, to layer 8) has powder grains 302 of various sizes. The powder layer 326 is applied to an underlying, already solidified layer 324. The powder grains 302 form a powder layer surface 304, the surface roughness of which is determined by the grain size and grain size distribution of the raw material powder. In Fig. 3, the surface roughness is indicated by the applied contour line 306. The laser beam 21 has a focal plane 308, which here lies in the plane of the surface 304. However, as the powder layer melts, its volume decreases. The surface 310 of the irradiated and thereby melted or solidified powder layer 311 is therefore located below or deeper than the surface 304 in the height / Z direction. This height offset is designated by the reference numeral 312. The height difference orHeight offset depends on the properties of the raw material powder (e.g. material type, grain size distribution, bulk density, etc.) and on the powder particle properties (e.g. shape of the grains or

[0071] Powder particles 302, etc.). For a powder with a low bulk density, the solidification plane 310 will be at a greater distance from the unirradiated powder bed plane 304 than for a powder with a high bulk density.

[0072] The height offset can also be influenced by material losses, as not all of the irradiated raw powder material forms the solidified layer. Thus, splashes of molten powder material often occur during irradiation, which then do not form part of the solidified layer 311, and a portion of the irradiated powder material passes into the gas phase. This material loss inevitably leads to a further decrease in the layer height compared to the unirradiated powder layer 304, thus increasing the height offset 312. The material losses depend, among other things, on the laser power, the laser spot size, the laser scanning speed, the process gas flow in the area of ​​the irradiated powder surface, and possibly other parameters.

[0073] It is conceivable to select measurement results for the estimation and / or to give them a higher weighting which are assigned to a height difference 312 and / or parameter values ​​influencing the height difference 312 (e.g. material type, bulk density, laser spot size, etc.) which fit or fit particularly well to a predicted height difference or parameter values ​​of an upcoming exposure.

[0074] It has been found that a certain material-dependent powder layer thickness of the applied powder only develops after a large number of powder layers have been applied. This behavior is illustrated in Fig. 4. When a first powder layer is applied to a base, which can also be referred to as a "base" (e.g., a previously solidified layer 324 or a build platform of the additive manufacturing apparatus), the powder layer 326 has a height hl. This height hl can correspond to the height shift caused by lowering the build platform before applying the powder layer 326. During irradiation, the volume of the powder layer 326 decreases, resulting in the height offset 312 or Ah explained above. If, after the build platform has been lowered again by the same height shift hl, a second layer is applied to the thus solidified first layer 311, this second powder layer typically has a layer thickness of hl + Ahi.The second applied raw material powder layer is therefore Ahi thicker than the first applied raw material powder layer 326. This second powder layer shrinks during irradiation by a greater amount than was the case with the first powder layer, because the second powder layer is comparatively thicker. This applies accordingly to the third layer, the fourth layer, etc. Only after a large number of layers (e.g., 5, 6, 7, 8, 9, 10, 15 or more) have been applied does a height offset 312, essentially constant from layer to layer, arise between the surface of a newly applied powder layer and its irradiated surface.

[0075] To account for this transient response, the measurement results can be selected and / or weighted based on information that specifies the current number of layers to be irradiated, the total number of layers, and / or a ratio of these two values. This information can indicate a predicted or recorded raw material powder layer thickness. In particular, measurement results that match the current transient phase of the layer thicknesses can be selected and / or weighted more highly.

[0076] A sharp or abrupt change in the irradiation area from a first layer to a subsequent layer can have a significant impact on the location of the irradiation point. Therefore, the information used for one or more powder layers can indicate a shape or area of ​​a section of the corresponding powder layer that is to be solidified to produce the workpiece, and in particular, indicate a sudden change in the irradiation area.

[0077] Powder layer regions to be solidified with a beam are typically distributed across a scan field. A large-area layer region in a first scan field region (e.g., in an edge region of the scan field, i.e., at maximum deflection of the beam provided by the corresponding deflection unit) of a powder layer to be exposed can result in a longer residence time of the laser beam in this first scan field region, whereas a small-area layer region in a different, second scan field region (e.g., in a central region of the scan field, i.e., at minimal deflection of the same beam) requires a shorter residence time of the laser beam. These different residence times can result in different thermal loads on the corresponding deflection unit, which affect the irradiation point and / or the irradiation pattern.To take this into account, the information used can also specify a scan duration, a scan speed, a scan pattern, or irradiation vectors. In particular, the information can specify an orientation of exposure vectors and / or an exposure vector feed direction in which the exposure vectors are arranged and exposed for exposing the powder layer.

[0078] The height difference 312 described above between the unirradiated and the irradiated powder layer surface 304 or 310 can also result in a lateral offset of an incident laser beam 21, in particular if the latter strikes the surface to be irradiated at an acute angle of < 90°, so to speak flat. This offset can in particular lead to the laser beam 21 not striking the (irradiated or unirradiated) powder material at the desired position. The lateral offset is greater the larger the deflection angle of the laser beam 21 is or the flatter the laser beam 21 strikes the surface. With large deflection angles of the laser beam 21, in particular with large powder layer thicknesses (e.g.300 pm), a large offset is to be expected between the impact position on the upper side 304 of the unsolidified powder layer 326 and the upper side 310 of the melted and thus compacted layer 311 of the component (particularly in the XY direction). This relationship is shown in Fig. 5. As indicated there, a deflection unit 20 or a laser scanner can be controlled such that the laser beam 21 strikes the powder surface 304 at an angle 314 at a point PL. However, this surface 304 is higher than the surface 310 of the solidified powder layer by the height offset 312. Therefore, the point PL deviates from the location 316 at which the laser beam would or will strike the solidified powder layer. The offset between the points PL and 316 can be calculated from the height offset and the laser angle 314. This offset can be taken into account as a correction value during calibration, especially if (e.g.During workpiece production, the position of the laser beam relative to a surface 310 of a solidified powder layer (e.g., 311) is controlled. The following applies: tan(laser angle) = h / d = (height offset 312) / (offset Ad in the XY direction). The lateral offset is also indicated in Fig. 3 ("X / Y offset").

[0079] When the laser beam strikes the powder surface, a so-called "keyhole" or melt hole is usually formed in the direct area of ​​impact. This is largely free of powder material, as it would otherwise evaporate there. The shape of the melt hole in the powder layer varies depending on the angle of impact of the laser beam. This influences the shape of the melt pool, and ultimately also the properties of the irradiated contour or the finished workpiece. This effect is indicated in Fig. 6.

[0080] Fig. 6 shows three side views of a melt hole 318 and the associated melt pool 320, depending on the laser incidence angle of the laser beam 21 on the powder layer surface 304. The lower images show corresponding top views. As can be seen, the outer contour of the melt pool 320 changes both in the side view and in the top view depending on the laser incidence angle. The deviation of the melt hole 318 can be defined, for example, by a melt hole depth, the (e.g. tangent of the) deflection angle(s), and a material factor. In the case of multiple deflection units 20, 22, 24, it can happen that immediately adjacent regions of a powder layer 326 irradiated via the different deflection units 20, 22, 24 have different melt hole and / or melt pool shapes during irradiation.This can lead to transitions between adjacent areas being visible after the workpiece has been manufactured and / or to the formation of mechanically unstable boundary zones. To avoid this effect, it is possible to consider only the measurement results associated with relevant laser scanning angles, laser incidence angles, melt hole shapes, and / or melt pool shapes.

[0081] In order to take the aforementioned effects into account, the information used can therefore include, in particular in addition to or alternatively to the height offset, a scan angle, a laser incidence angle, the lateral offset, a melt hole shape and / or a melt pool shape. The rules used for selection and / or weighting (e.g. selection criteria, filter parameters, weighting rules and / or weighting factors) can be designed variably, i.e. they can change over time, in particular depending on the large number of measurement results. If, for example, additional measurement results are provided, the rules can be adapted. The rules can be provided using a trained machine learning system and / or adapted from time to time. Measurement results from one or more reference manufacturing processes can be used as training data for the machine learning system, the selection and / or weighting of which is specified by a user.The machine learning system can then consider additional measurement results acquired during a currently performed manufacturing process to adapt the rules. It is also conceivable for the machine learning system to not only manage the rules but also implement step 202. The machine learning system can be configured as part of the control unit 40.

[0082] In step 204, the at least one property of the irradiation point and / or irradiation pattern to be provided is estimated based on the plurality of measurement results. If step 202 was performed previously, in step 204 (e.g., exclusively) the selected and / or weighted measurement results are used to estimate the at least one property.

[0083] Estimating the at least one property comprises, in particular, averaging, intrapolating, and / or extrapolating the properties specified by the (e.g., selected and / or weighted) measurement results used for the estimation. If, for example, the selected images specify respective imaged positions of an irradiation point as the at least one property, wherein the images are each assigned to a planned position of the irradiation point, then a predicted position supplied by the corresponding deflection unit can be assigned to an irradiation point to be provided at a planned position in the future. A probable position can therefore be estimated, in particular, by averaging, intrapolating, and / or extrapolating the positions specified by the measurement results used for the estimation. Other properties, such as, for example, an intensity of an irradiation pattern, can also be estimated in this way.

[0084] Step 202 and / or step 204 can be performed based on a temporal progression of the same property. In other words, not only the measurement results as such, but in particular the temporal progression of the property(ies) indicated by the measurement results can be taken into account in order to select measurement results and / or weight them and / or estimate the at least one property. For example, the temporal progression can indicate a temporal change in the distance of a measured position of an irradiation point from a planned position. In this way, a future position of the irradiation point can be estimated. As a further example, a temporal decrease in intensity of an irradiation pattern can be analyzed in order to estimate a future intensity of an irradiation pattern to be provided. Thus, not only are multiple measurement results taken into account, but also their temporal relationship to one another.

[0085] Since the measurement results are based on optical measurements performed at different times, various events can occur between the individual measurements. In particular, a change in the state of the production apparatus 3 can occur between two or more of the optical measurements performed at different times. Examples include a change in an irradiation process, a change in a process gas atmosphere, a coating process with raw material to form a powder layer 8, and / or a change in the position of the carrier 10. In particular, the measurement results associated with the same state of the production apparatus can be used for the estimation (e.g., only measurement results recorded at the same powder temperature and / or gas temperature).

[0086] In step 206, the manufacturing apparatus 3 is calibrated based on the estimated at least one property. Calibration can ensure that the at least one property (e.g., position) of an irradiation point and / or irradiation pattern provided after calibration corresponds to a desired property. Calibration can include calibrating the irradiation system 3, in particular calibrating the lasers 14, 16, 18 and / or the deflection unit(s) 20, 22, 24. For this purpose, calibration parameters can be determined that are used for later control of the irradiation system 4. The calibration parameters can specify an offset or a position correction function and serve to ensure that an irradiation point is provided at a planned position after calibration. Calibration can include aligning several of the deflection units 20, 22, 24 with one another.If the measurement results indicate positions of irradiation points that were positioned, for example, via the deflection unit 22 on the one hand and the deflection unit 24 on the other, then the deflection units 22, 24 can be calibrated relative to one another. If the measurement results are based on images recorded by the sensor 30 via the deflection unit 20, then this deflection unit can be calibrated relative to one or both of the deflection units 22, 24 used for irradiation during image recording. It is also possible, instead of a relative calibration of the deflection units 20, 22, 24, to calibrate one or more deflection units absolutely, in particular based on a position of the reference element 36 in an image recorded by the sensor 30 and / or 32.As mentioned above, calibration can also ensure other properties of an irradiation point to be provided, such as the intensity or contour of an irradiation point, instead of or in addition to a position correction. Such properties can also be adjusted by calibration relative to properties of an irradiation point provided via another deflection unit or relative to an absolute reference.

[0087] After calibration, the optional step 208 can be carried out, in particular during a manufacturing process for producing a workpiece, e.g., during, before, or after solidifying a region of the powder layer 8 to produce the workpiece. In step 208, an irradiation point and / or irradiation pattern is provided by the manufacturing apparatus 3, in particular by the calibrated irradiation system 4. If, for example, the deflection unit 22 has been calibrated, then this irradiation point or irradiation pattern is directed onto the powder layer 8 via this deflection unit 22. If the further deflection units 20, 24 are calibrated relative to the deflection unit 22 or absolutely like it, the irradiation can alternatively be carried out by one of these deflection units 20, 24.

[0088] In the subsequent optional step 210, an optical measurement is performed to capture a measurement result. In particular, an image can be captured depicting the irradiation point or irradiation pattern provided in step 208. This measurement result also indicates a property of the provided irradiation point and / or irradiation pattern, e.g., an intensity, position, or spectrum.

[0089] In the optional subsequent step 212, the production apparatus is recalibrated based on the measurement result acquired from step 210. It is conceivable that, of the available measurement results, only the measurement result acquired in step 210 is used for the recalibration. Alternatively or additionally, the measurement result acquired in step 210 can be saved, in particular as part of the plurality of measurement results. One or more of steps 202-206 can now be performed again, whereby the plurality of measurement results also includes the measurement result from step 210.

[0090] It is conceivable that the estimated at least one property (e.g., the estimated position or an estimated position deviation from a reference position) is used to influence steps 208 and / or 210. In particular, a temporal offset and / or a number of optical measurements to be performed for recalibration can be specified or determined based on the estimated property. If the estimated position deviation is, for example, greater than a specific threshold value, multiple iterations of steps 208, 210 can be provided in order to subsequently recalibrate the manufacturing apparatus 3 in step 212. Alternatively or additionally, a spatial distribution of the optical measurements to be performed for recalibration can also be specified by the estimated at least one property.For example, the spatial distribution can define the planned positions of the irradiation points to be provided during measurements and / or the viewing areas to be used for image acquisition. The spatial distribution can, for example, be selected such that predicted position deviations for the planned positions exceed a predetermined threshold.

[0091] Fig. 7 shows a timeline over time t with exemplary measurement points. Measurement results from optical measurements selected for estimation are indicated as exemplary measurement groups Ci and C2. In the example shown, the measurement results are assigned to different workpiece information Si, S2 (e.g., different layers of the component), which are used to select the measurement results used for estimation. It can be seen that not all measurement results based on optical measurements I nare used for the estimation, but rather a targeted subselection I1-I3 or I3-I5. These measurement results are based on optical measurements taken at different measurement times. Other measurement data (e.g., status data or data from other sensors) at different times are indicated as M1-M3. Measurement data from temperature sensors (e.g., powder temperature, process gas temperature) are indicated by Ti, T2. The present disclosure is not limited to the example according to Fig. 7. Instead of the workpiece information, the measurement results I to be used for the estimation can be n e.g. based on other information, e.g. based on the measurement results T n or / and M n selected and / or weighted. The technique described above and further details are explained below in other words.

[0092] It is possible that positions irradiated via the deflection units 20-24 may vary over time if the deflection units 20-24 are controlled unchanged. Calibration of the production apparatus 3 can result in an adjustment of the scanning angles of the scanning systems of the deflection units 20-24 so that planned irradiation positions in the processing plane 13 are actually irradiated. Since the production apparatus 3 comprises several deflection units 20-24, the calibration ensures, in particular, that the deflection units 20-24 are calibrated relative to one another and / or absolutely.Thus, after calibration, positions and / or focuses of the irradiation points and / or irradiation patterns positioned via the involved deflection units 20-24 can coincide at any positions of the processing area 6 common to the deflection units 20-24, even if each deflection unit is controlled with respect to an absolute machine coordinate system (MCS).

[0093] During the manufacturing process, thermal and other physical influences on the irradiation system 4 may cause the positions and / or focuses of the irradiation points and / or irradiation patterns to no longer align. This may result in seamless production of a component no longer being possible. While such misalignments can be compensated for during the manufacturing process by calibration based on a single measurement, this approach, which relies exclusively on single measurements, is subject to interference such as influences from the imaging process, the material, or the beam parameters at the time the image is acquired. Furthermore, the acquisition of numerous individual measurements may cause undesirable interruptions to the manufacturing process.

[0094] In particular, to reduce the influence of noise and to increase the accuracy of the calibration (e.g. determining corrections of mappings of scan angles to positions in the machine coordinate system), the properties of the irradiation point and / or irradiation pattern determined from the recorded images, for example positions of irradiation points, distances between irradiation points or between external reference points (e.g. the reference element 36), are not interpreted individually, but in combination, preferably as a time series. The influence of noise can be reduced by averaging and filtering. Transitions in the machine state, such as exposure, coating processes and adjustment processes of the carrier 10 (also referred to as component carrier structure, substrate plate or build platform), can take place between the image recordings, but do not have to.The filtered measurement results can be located in an original image or back-projected scanner or machine coordinate systems. Therefore, not only a single measured actual position should be used for calibration, but rather an estimate of at least one property (e.g., position) based on multiple measurements.

[0095] The filtering can be adaptive in time and / or space. With adaptive filtering, filter parameters can be adjusted according to the concepts of machine learning. For example, the filter parameters can be adjusted based on changes in the exposed area or the number of deflection units used. Changes in the measurements can therefore adjust the adjustments, so one could speak of self-regulating adjustment.

[0096] To capture the measurement results from the optical measurements (e.g., images), a brightness-based adaptive method can be used that does not employ fixed thresholds. Such brightness-based adaptive methods differ from thresholding methods with a fixed threshold or methods with multiple thresholds. The intensities of the exposure points can vary greatly, resulting in a wide range of brightness values ​​for the imaged points. Fixed thresholding methods for detecting the imaged points or methods with multiple fixed thresholds are disadvantageous compared to such adaptive methods.

[0097] Parameters used for laser control (e.g. power, interaction time, etc.) can be individually adjusted to the laser and the powder material.

[0098] Information (e.g., regarding energy per shift, power, areas, geometries, volumes) of the components to be manufactured, which is known before the start of the production process, can be incorporated to adjust filter parameters. This information can also be used to regulate the frequency of measurements or the number and spatial distribution of measurements.

[0099] In another example, measurement data M n of the manufacturing process (e.g.

[0100] Laser power, exposure time and distances) are included to

[0101] Adjust filter parameters. Here, too, the information can be used to regulate the frequency of measurements, the number and / or spatial distribution and / or adaptation rate of the filter methods used.

[0102] In the processing area 6, which forms the common overlap area of ​​the deflection units 20-24, irradiation points and / or irradiation patterns are generated by a subset of the deflection units 20-24 and the energy beams 21-25 of the energy sources 14-18 deflected by them. These points or patterns are imaged by the optical sensor 26 (e.g., a camera) via a deflection unit 20 that is not involved in providing the points or patterns. In order to also compare focus positions of the deflection unit 20, which previously defined the field of view of the optical sensor 26, with other deflection units 22, 24, this image acquisition process can be repeated, using a different optical sensor 28 and the deflection unit 24 assigned to it.

[0103] The focusing of the image recorded by the respective optical sensor 26, 28 onto the processing plane 13 can be done by means of the dynamic lens 30, 32 assigned to the respective sensor, preferably a liquid lens (e.g. Focus Tunable Liquid Lens).

[0104] In order to calibrate the deflection units 20-24, which are compared relative to one another using the method described above, with respect to an external absolute reference, the calibration element 36 can be provided in or adjacent to the processing area. Such a calibration element 36 can be physically marked at the predetermined absolute position or projected as a light pattern.

[0105] It is understood that the system 2 described here, as well as the method described with reference to Fig. 2, can be modified. For example, additional deflection units with corresponding radiation sources can be provided, or only two deflection systems in total. Further modifications to the system and method, as well as further technical advantages of the technology described here, will become apparent to those skilled in the art from the present disclosure.

Claims

Claims 1. A method for calibrating a manufacturing apparatus (3) which is designed for the additive manufacture of three-dimensional workpieces by layer-by-layer solidification of raw material by means of location-selective irradiation, the method comprising: Estimating (204) at least one property of an irradiation point and / or irradiation pattern to be provided on the basis of a plurality of measurement results (I1-I5) of optical measurements carried out at different times, each measurement result indicating at least one property of an irradiation point and / or irradiation pattern at the respective measurement time; and Calibrating (206) the manufacturing apparatus (3) based on the estimated at least one property.

2. The method of claim 1, further comprising: Selecting and / or weighting (202) measurement results (I1-I3; I3-I5) of the plurality of measurement results (I1-I5), wherein the at least one property is estimated on the basis of the selected and / or weighted measurement results.

3. The method according to claim 2, wherein the measurement results (I1-I3; I3-I5) are selected and / or weighted based on information that is at least partially predetermined and / or at least partially obtained from a manufacturing process currently being carried out by the manufacturing apparatus (3).

4. The method according to claim 3, wherein the information is associated with a workpiece to be manufactured by the manufacturing apparatus (3) and / or comprises at least one process parameter of the manufacturing apparatus (3).

5. The method according to claim 4, wherein the information comprises at least one property of the workpiece to be manufactured and / or at least one process parameter of a particularly currently carried out manufacturing process for the additive manufacturing of the workpiece to be manufactured.

6. Method according to one of claims 1 to 5, wherein the at least one property is estimated based on a temporal course of the same property and / or the measurement results (I1-I3; I3-I5) are selected and / or weighted based on a temporal progression of the same property.

7. The method according to any one of claims 1 to 6, wherein the estimation of the at least one property comprises averaging, intrapolating and / or extrapolating the properties indicated by the measurement results used for the estimation.

8. Method according to one of claims 1 to 7, wherein between two or more of the optical measurements carried out at different times there is a change in the state of the production apparatus (3), in particular a change in an irradiation process, a change in a process gas atmosphere, a coating process with raw material and / or a change in a position of a construction platform (10) of the production apparatus (3).

9. Method according to one of claims 1 to 8, further comprising: providing (208) the irradiation point to be provided and / or Irradiation pattern by the calibrated manufacturing apparatus (3); Carrying out (210) an optical measurement to acquire a measurement result that indicates at least one property of the provided irradiation point and / or irradiation pattern; and Recalibrating the manufacturing apparatus (3) based on the at least one property of the provided irradiation point and / or irradiation pattern, and / or storing the acquired measurement result as part of the plurality of measurement results (I1-I5).

10. The method according to claim 9, wherein based on the estimated at least one property and / or based on the calibration of the manufacturing apparatus (3), at least one of the following is specified: a desired property of the irradiation point to be provided and / or the irradiation pattern to be provided; a temporal offset and / or a number of optical measurements to be performed, in particular for the recalibration; a spatial distribution of optical measurements to be performed, in particular for the recalibration.

11. The method according to any one of claims 1 to 10, wherein the manufacturing apparatus (3) comprises an optical sensor (26), a first deflection unit (20), and a second deflection unit (22) and / or a third deflection unit (24), wherein: (i) the at least one property of an irradiation point and / or irradiation pattern can be detected by the optical sensor (26) via the first deflection unit (20) and the corresponding irradiation point and / or the corresponding irradiation pattern can be positioned via the second deflection unit (22), or / and (ii) the at least one property of an irradiation point and / or irradiation pattern can be detected by the optical sensor (26), and the corresponding irradiation point and / or the corresponding irradiation pattern can be positioned selectively via the second deflection unit (22) or the third deflection unit (24); wherein the calibration (206) and / or recalibration (212) of the production apparatus (3) comprises aligning the first deflection unit (20) with the second deflection unit (22) and / or aligning the second deflection unit (22) with the third deflection unit (24) and / or aligning the first deflection unit (20) with the third deflection unit (24).

12. Device (38) for calibrating a manufacturing apparatus (3) which is designed for the additive manufacture of three-dimensional workpieces by layer-by-layer solidification of raw material by means of location-selective irradiation, wherein the manufacturing apparatus comprises an irradiation system (4) for providing an irradiation point and / or irradiation pattern, wherein the device (38) comprises a control unit (40) for controlling the irradiation system (4), which is designed to: Estimating (204) at least one property of an irradiation point and / or irradiation pattern to be provided on the basis of a plurality of measurement results (I1-I5) of optical measurements carried out at different times, each measurement result indicating at least one property of an irradiation point and / or irradiation pattern at the respective measurement time; and Calibrating (206) the additive manufacturing apparatus (3) based on the estimated at least one property.

13. Device (38) according to claim 12, wherein the control unit (40) is configured to carry out the method according to one of claims 2 to 11; and / or wherein the manufacturing apparatus (3) comprises an optical sensor (26), a first deflection unit (20), a second deflection unit (22) and / or a third deflection unit (24), which are designed such that: (i) the at least one property of an irradiation point and / or irradiation pattern can be detected by the optical sensor (26) via the first deflection unit (20) and the corresponding irradiation point and / or the corresponding irradiation pattern can be positioned via the second deflection unit (22), or / and (ii) the at least one property of an irradiation point and / or irradiation pattern can be detected by the optical sensor (26) and the corresponding irradiation point and / or the corresponding irradiation pattern can be positioned selectively via the second deflection unit (22) or the third deflection unit (24).

14. System (2) comprising: the device (38) according to one of claims 12 or 13; and the manufacturing apparatus (3).

Images