A removable calibration plate (10) comprises a sheet (20) comprising an upper face (21) intended to face towards the powerful incident-radiation beam, and bearing a reference marking (30) and being intended to receive a test marking (40), and a lower face (23). The plate (10) comprises an etching layer (22) to be etched by a powerful incident-radiation beam (F),this layer being secured to the upper face (21) of the sheet (20) and opaque to visible light, and being able to be destroyed locally by the powerful incident-radiation beam (F) in order to form the at least one test marking (40), the sheet (20) being transparent to visible light, the lower face (23) of the sheet (20) being frosted.

A removable image-capture apparatus (200) comprises an opening (202) intended to receive a calibration plate (10) bearing a reference marking (30) and possibly a test marking (40). The apparatus (200) comprises a source (204) of backlighting visible light situated beneath the opening (202), a sensor (205) for acquiring an image, in the backlighting visible light, of the plate (10), a guiding and supporting device (206) for positioning the sensor (205) above the opening (202) relative to the surround (201), a calculation device (207) configured to analyze the image, recognize the marking (30) and possibly the marking (40) in the image, and calculate aiming-command corrections intended for a firing system firing a powerful incident-radiation beam, which system belongs to an additive manufacturing apparatus, distinct and separate from the apparatus (200).

A removable image-capture apparatus (200) comprises an opening (202) intended to receive a calibration plate (10) bearing a reference marking (30) and possibly a test marking (40). The apparatus (200) comprises a source (204) of backlighting visible light situated beneath the opening (202), a sensor (205) for acquiring an image, in the backlighting visible light, of the plate (10), a guiding and supporting device (206) for positioning the sensor (205) above the opening (202) relative to the surround (201), a calculation device (207) configured to analyze the image, recognize the marking (30) and possibly the marking (40) in the image, and calculate aiming-command corrections intended for a firing system firing a powerful incident-radiation beam, which system belongs to an additive manufacturing apparatus, distinct and separate from the apparatus (200).

A calibration method serves for calibrating a manufacturing device for additively producing a three-dimensional object by applying layer by layer and selectively solidifying a building material. The manufacturing device comprises at least two scanning units, each of which is capable of directing a beam to different target points in the working plane, which are located within a scanning region assigned to the respective scanning unit, wherein the scanning regions region of the at least two scanning units overlap in an overlap area. At least a first of the at least two scanning units is assigned a first monitoring unit whose monitoring region extends to a target point of the first scanning unit and its proximity, wherein a change of a position of the monitoring region is carried out as a function of a change of a position of the target point.

A planning device for planning locally selective irradiation of a work region using an energy beam in order to produce a component from a powder material arranged in the work region is provided. The planning device is configured to obtain a plurality of irradiation vectors for irradiating a powder material layer arranged in the work region with the energy beam. The planning device is further configured to determine a vector alignment in a coordinate system on the work region for at least one irradiation vector of the plurality of irradiation vectors, and to specify, for the at least one irradiation vector, a beam alignment for a non-circular beam shape of the energy beam on the work region relative to the vector alignment of the at least one irradiation vector.

A planning device for planning locally selective irradiation of a work region using an energy beam in order to produce a component from a powder material arranged in the work region is provided. The planning device is configured to obtain a plurality of irradiation vectors for irradiating a powder material layer arranged in the work region with the energy beam. The planning device is further configured to determine a vector alignment in a coordinate system on the work region for at least one irradiation vector of the plurality of irradiation vectors, and to specify, for the at least one irradiation vector, a beam alignment for a non-circular beam shape of the energy beam on the work region relative to the vector alignment of the at least one irradiation vector.

Embodiments relate to in-situ process monitoring of a part being made via additive manufacturing. The process can involve capturing computed tomography (CT) scans of a post-built part. A neural network (NN) can be used during the build of a new part to process multi-modal sensor data. Spatial and temporal registration techniques can be used to align the data to x,y,z coordinates on the build plate. During the build of the part, the multi-modal sensor data can be superimposed on the build plate. Machine learning can be used to train the NN to correlate the sensor data to a defect label or a non-defect label by looking to certain patterns in the sensor data at the x,y,z location to identify a defect in the CT scan at x,y,z. The NN can then be used to predict where defects are or will occur during an actual build of a part.

Embodiments relate to in-situ process monitoring of a part being made via additive manufacturing. The process can involve capturing computed tomography (CT) scans of a post-built part. A neural network (NN) can be used during the build of a new part to process multi-modal sensor data. Spatial and temporal registration techniques can be used to align the data to x,y,z coordinates on the build plate. During the build of the part, the multi-modal sensor data can be superimposed on the build plate. Machine learning can be used to train the NN to correlate the sensor data to a defect label or a non-defect label by looking to certain patterns in the sensor data at the x,y,z location to identify a defect in the CT scan at x,y,z. The NN can then be used to predict where defects are or will occur during an actual build of a part.

Selective laser solidification apparatus is described that includes a powder bed onto which a powder layer can be deposited and a gas flow unit for passing a flow of gas over the powder bed along a predefined gas flow direction. A laser scanning unit is provided for scanning a laser beam over the powder layer to selectively solidify at least part of the powder layer to form a required pattern. The required pattern is formed from a plurality of stripes or stripe segments that are formed by advancing the laser beam along the stripe or stripe segment in a stripe formation direction. The stripe formation direction is arranged so that it always at least partially opposes the predefined gas flow direction. A corresponding method is also described.

Selective laser solidification apparatus is described that includes a powder bed onto which a powder layer can be deposited and a gas flow unit for passing a flow of gas over the powder bed along a predefined gas flow direction. A laser scanning unit is provided for scanning a laser beam over the powder layer to selectively solidify at least part of the powder layer to form a required pattern. The required pattern is formed from a plurality of stripes or stripe segments that are formed by advancing the laser beam along the stripe or stripe segment in a stripe formation direction. The stripe formation direction is arranged so that it always at least partially opposes the predefined gas flow direction. A corresponding method is also described.