In one example, a decaking system for 3D printing includes a platform to support multiple green parts in unbound powder surrounding the green parts, a decaking tool to remove unbound powder from around the green parts, a camera to photograph green parts on the platform as unbound powder is removed from around the green parts, and a controller operatively connected to the camera. The controller is programmed to detect a pattern of light intensity in the photographs and, in response to a determination a detected pattern matches a reference pattern, modulate or stop the decaking tool.

In one example, a decaking system for 3D printing includes a platform to support multiple green parts in unbound powder surrounding the green parts, a decaking tool to remove unbound powder from around the green parts, a camera to photograph green parts on the platform as unbound powder is removed from around the green parts, and a controller operatively connected to the camera. The controller is programmed to detect a pattern of light intensity in the photographs and, in response to a determination a detected pattern matches a reference pattern, modulate or stop the decaking tool.

A three-dimensional (3D) metal object manufacturing apparatus has a plurality of thermally insulative members that float in a volume of heat transfer lubricating fluid in which a X-Y translation mechanism moves to position a platform opposite an ejector. The apparatus also includes a housing having an internal volume in which the platform and X-Y translation mechanism are located. The heat transfer lubricating fluid can be a molten salt, such as a molten fluoride, chloride, or nitrate molten salt. The thermally insulative members can be spheres made of zirconium oxide or zirconium dioxide. The thermally insulative layer formed by the members floating in the fluid protects the X-Y mechanism while the housing helps keep the surface temperature of the object being formed on the platform in an optimal range for bonding of melted metal drops ejected from the ejector to a surface of a metal object being formed on the platform.

A three-dimensional (3D) metal object manufacturing apparatus has a plurality of thermally insulative members that float in a volume of heat transfer lubricating fluid in which a X-Y translation mechanism moves to position a platform opposite an ejector. The apparatus also includes a housing having an internal volume in which the platform and X-Y translation mechanism are located. The heat transfer lubricating fluid can be a molten salt, such as a molten fluoride, chloride, or nitrate molten salt. The thermally insulative members can be spheres made of zirconium oxide or zirconium dioxide. The thermally insulative layer formed by the members floating in the fluid protects the X-Y mechanism while the housing helps keep the surface temperature of the object being formed on the platform in an optimal range for bonding of melted metal drops ejected from the ejector to a surface of a metal object being formed on the platform.

The invention relates to an apparatus for producing an object by means of additive manufacturing, comprising a process chamber for receiving a bath of material which can be solidified by exposure to electromagnetic radiation; a support for positioning the object in relation to the surface level of the bath of material; and a solidifying device for solidifying a selective layer-part of the material on the surface level by means of electromagnetic radiation. Furthermore optical control device is provided with a focus unit in an optical pathway of the electromagnetic radiation of the solidifying device, and arranged for controlling at least the focus of the electromagnetic radiation emitted by the solidifying device on the surface level. According to the invention, the optical control device comprises a sensor element arranged for detecting a measure for the accuracy of the focus of the electromagnetic radiation and a focus correction lens element that is arranged to be movable. By moving said focus correction lens element, focus may be corrected, for example due to thermal behaviour of the optical system.

Examples of a method of operating an additive manufacturing system, a three-dimensional (3D) printing system and a non-transitory machine-readable medium are described. In an example, a build material is supplied to a print region of an additive manufacturing system. A temperature distribution, corresponding to a pattern, of at least a surface of the build material is generated. An image of the pattern is captured using a thermal sensor. Image data representative of the image of the pattern is compared with data representative of an expected position of the pattern. On the basis of the comparing, difference data indicative of a difference between a position of the thermal sensor during capture of the image and an expected position of the thermal sensor associated with the expected position of the pattern is generated. Operation of the additive manufacturing system is controlled at least in dependence on the difference data.

The invention relates to a device (100) for the additive manufacture of three-dimensional workpieces, in particular a 3D metal printer, comprising a print head (1) and a device (40) for generating an inert atmosphere (22) within the print head (1) by means of a gas (55), in particular inert gas, wherein the print head (1) comprises a housing (3), a device (28) for feeding a metal (14), a piston (5), a reservoir (7) with an outlet opening (10) and an actuator device (12) for displacing the piston (5), wherein the reservoir (7) has a melt region (20) and a displacement body chamber (21) for a liquid phase (8) of the metal (14), wherein the melt region (20) adjoins the inert atmosphere (22) and is connected to the displacement body chamber (21) such that, as a result of the displacement of the piston (5), the liquid phase (8) of the metal (14) can be caused to pass through the outlet opening (10). The invention is distinguished by the fact that the device (40) for generating the inert atmosphere (22) is arranged outside the print head (1), wherein said device comprises an accumulator (41), at least one means (42, 43) for pressure control, and a gas line (50, 51, 52). The invention furthermore relates to methods for operating the device (100).

This disclosure describes an additive manufacturing method that includes monitoring a temperature of a portion of a build plane during an additive manufacturing operation using a temperature sensor as a heat source passes through the portion of the build plane; detecting a peak temperature associated with one or more passes of the heat source through the portion of the build plane; determining a threshold temperature by reducing the peak temperature by a predetermined amount; identifying a time interval during which the monitored temperature exceeds the threshold temperature; identifying, using the time interval, a change in manufacturing conditions likely to result in a manufacturing defect; and changing a process parameter of the heat source in response to the change in manufacturing conditions.

An electron beam source including a cathode, an anode, a means for deflecting an electron beam over a target surface and at least one vacuum pump, the electron beam source further including a contraction area arranged between the anode and the means for deflecting the electron beam where a hole in the contraction area is aligned with a hole in the anode with respect to the cathode, a first vacuum pump is arranged between the contraction area and the anode and a second vacuum pump is arranged above the anode, a gas inlet is provided between the contraction area and the means for deflecting the electron beam, wherein a first crossover of the electron beam is arranged between the cathode and the anode and a second crossover is arranged at or in close proximity to the contraction area.

Techniques for manufacturing optimization using a multi-tenant machine learning platform are disclosed. A method for manufacturing optimization includes: obtaining physical sensor data, by a manufacturing device associated with a tenant of a multi-tenant machine learning platform; determining, by a machine learning spoke system associated with the tenant, a machine learning parameter based on at least the physical sensor data; preventing exposure of the first physical sensor data of the first manufacturing device to any other tenant of the multi-tenant machine learning platform; transmitting the machine learning parameter from the machine learning spoke system to a machine learning hub system of the multi-tenant machine learning platform; and updating, by the machine learning hub system, a multi-tenant machine learning model based at least on the machine learning parameter.