Method and apparatus for drop-based additive manufacturing with in situ inspection

Abstract

For better process quality control, the invention provides a method and an apparatus (10) for drop-based additive manufacturing of a component (12). The method comprises: a) applying a liquid material drop by drop by means of a printhead (20) in a working chamber (22) in order to build up a substrate (24) to form the component (12), and b) detecting at least one drop (18) in flight after said drop is ejected from the printhead (20) and before said drop hits the substrate (24), by means of at least one camera (26, 26.1, 26.2) arranged outside and / or inside the working chamber (22) and at least one exposure device (28, 28.1, 28.2), wherein the position of the at least one camera (26, 26.1, 26.2) and the position of the at least one exposure device (28, 28.1, 28.2) are coupled to the position of the printhead (20), and comparing at least one physical parameter of the recorded drop (18) with a reference value in order to detect a change in the at least one physical parameter, wherein step b) is repeated at predefined intervals during step a).

Description

[0001] GROB-WERKE GmbH & Co. KG 1523 0142 P-WO

[0002] Industriestraße 4 256A-PCT

[0003] D-87719 Mindelheim

[0004] Method and device for drop-based additive manufacturing with in-situ testing

[0005] The invention relates to a method for drop-based additive manufacturing of a component, comprising the drop-by-drop application of a liquid material by means of a printhead in a working chamber to build up a substrate for forming the component, in particular layer by layer. The invention further relates to a device for drop-based additive manufacturing of a component, comprising a manufacturing unit with a base arranged in a working chamber, a printhead for drop-by-drop application of liquid material onto the base, a movement device for moving the printhead and base relative to each other, and a computer-implemented control system, and configured for drop-by-drop application of the liquid material by means of the printhead in the working chamber to build up a substrate on the base for forming the component.

[0006] The following literature references are made for the technological background:

[0007] [1] DE 102021 117 285 A1

[0008] [2] Tianjiao Wang et al. “In-situ Droplet Inspection and Control System for Liquid Metal Jet 3D Printing Process”; 45th SME North American Manufacturing Research Conference, NAMRC 45, LA, USA; Procedia Manufacturing 10 (2017) 968 - 981

[0009] [3] Tammy Chang et al. “ln-situ monitoring for liquid metal jetting using a millimeter-wave impedance diagnostic”; Nature Research Scientific Repots (2020) 10:22325, https: / / doi.Org / 10.1038 / S41598-020-79266-2

[0010] [4] CN 104 493 186 A

[0011] [5] US 2022 / 379564 A1

[0012] [6] WO 2020 / 061544 A1

[0013] [7] Grob website “Additive Manufacturing”; downloaded on 27 October 2025 from https: / / www.grobgroup.com / produkte / produktbereiche / additive-fertigung /

[0014] [8] Wikipedia: Feret diameter, downloaded on 03.11.2025 from https: / / en.wikipedia.org / wiki / Feret_diameter

[0015] The invention generally lies in the field of additive manufacturing, often also referred to as 3D printing or by the English term "additive manufacturing." Additive manufacturing technology has long experienced strong growth in all areas. In addition to the steadily increasing number of suppliers, vendors, and service providers, the fields of application are expanding due to continuously improving process quality and thus increasing component complexity and size. Each of these various 3D printing processes is supported by measurement technology that measures the parameters responsible for achieving the component requirements. This measurement technology is individually tailored to the respective 3D printing process. One application for measurement technology is the monitoring of the properties of the material being applied. The shape of the material and the sometimes extreme environmental conditions determine which sensors can be used effectively.The temperatures prevailing in the environment at such pressures alone preclude most tactile and thermosensitive measurement techniques. Therefore, optical sensors are predominantly used in most 3D printing processes. Examples include laser point control in laser powder bed fusion or online measurement of component geometry using an optical line sensor, not only in molten metal jetting.

[0016] The measurement data obtained through metrology are documented, controlled, or further processed. 3D printing processes, for example, utilize software that calculates a process-specific axis movement from the 3D model of an additively manufactured component and the results of measurements of the material's properties before the printing process begins. This so-called "slicer" calculates, for instance, in the "Laser Powder Bed Fusion" process, how the mirror must be moved to deflect the laser, which in turn is responsible for melting the powder. The slicer of a "Metal Fused Deposition Modeling" system calculates the path the nozzle travels to build the intended component with the filament fed at a defined feed rate.The slicer is therefore a software solution that, taking into account the process characteristics of a 3D printing process, is responsible for calculating multiple axis movements for the additive manufacturing of components of varying complexity.

[0017] A key focus of 3D printing processes is the production of components of the highest possible quality, defined by criteria such as surface finish, microstructure homogeneity, density, mechanical properties, etc. This component quality can only be achieved if the measured properties provided to the slicer remain stable or are kept stable throughout the entire process.

[0018] In droplet-based additive manufacturing processes, instabilities are defined as changes in droplet diameter, velocity, or angle, or when a change in the nozzle configuration at the droplet exit point alters the dynamics of the droplet detachment process and affects the droplet shape, most often resulting in the formation of satellite droplets. A satellite droplet is a second droplet that detaches from the excited melt during detachment processes that are too energetically intense and deviate from the target values ​​for diameter, velocity, and direction. If such or similar process instabilities are not detected, or not detected in time, the requirements for the printed component generally cannot be met.If the process cannot be brought into the desired state through parameter variations, the alternative is to terminate the printing process, replace the printhead hardware including the worn nozzle, and continue printing the incomplete part from the interrupted layer. For further information on this topic, please refer to reference [1]. During part production, a layer offset may occur at the part height where the hardware is changed, as the droplet angles can vary between the different setups.

[0019] If it were possible to measure and compare the droplet angle of the old and new setups, the position of the print head relative to the component could be compensated for to prevent the offset described above. The following description presents a solution that uses a novel droplet measurement method to measure the process characteristics during the printing process of a droplet-based 3D printing process and to identify when a parameter adjustment of the process is necessary.

[0020] For several years, GROB-WERKE GmbH & Co. KG has been developing a liquid metal printing process called "Grob Metal Printing," abbreviated "GMP," see [1], [7]. This wire-based 3D printing process generates liquid droplets with defined properties from a nozzle, which are deposited onto a build platform. The distance between the print head and the component surface is currently between 5 and 10 millimeters. The system incorporates a camera system with image processing that performs measurements, among other things, on the liquid aluminum droplets. The sensors used are calibrated during commissioning. This allows pixels or electrical signals to be converted into metric units. Immediately after the highly dynamic ejection, the droplets have a very irregular shape, making it impossible to reliably assess the required properties.To obtain meaningful results from the droplet measurements, a larger image area is therefore necessary. For this, the printhead must interrupt the manufacturing process and move to a specific test position. Among other conditions, it must be ensured that the material ejected during the measurements is collected. In this test position, a camera, in conjunction with a conventional image processing system, determines the droplet diameter, droplet velocity, and droplet angle (from the camera's perspective). Grayscale analysis allows the droplets to be identified in the image. Based on the droplet edges, cross-sectional area, and / or circumference, a theoretical volume or diameter of an ideal sphere can be approximated.The droplet velocity can be calculated from the distance traveled in a certain time or the droplet ejection frequency together with the distance between two successive drops.

[0021] The third dimension, invisible to a camera, is disregarded in this setup. If multiple drops are captured in a single image, the angle at which the drops exit the nozzle, relative to the camera, can be calculated by averaging the drop centers. If the measuring system is referenced to the nozzle, the component, or the build platform, the drop angle can also be calculated relative to these references. In-situ measurement, i.e., drop control during component production, is not currently feasible due to factors such as limited workspace, adverse environmental conditions, and the printhead's varying height. Therefore, regular monitoring of droplet properties, for example, after 20 minutes, 100,000 drops, or 10 layers, is advisable. The interval is adjusted depending on the component geometry and process quality.

[0022] Besides area scan cameras, the measurement tasks can also be performed by other sensor types, such as line scan cameras or electromagnetic measurement principles.

[0023] References [2] to [6] describe various known measurement principles. Reference [2] describes an in-situ measurement for a liquid metal jetting process. However, no components are printed in this setup; a functioning droplet-based additive manufacturing process with in-situ measurement is not disclosed. Reference [3] describes an electromagnetic measurement concept that can detect aluminum droplets with diameters between 0.8 and 1.6 millimeters.

[0024] The invention aims to improve methods and devices for droplet-based additive manufacturing of the type mentioned above, in particular with regard to quality control, and preferably also with regard to process duration and process quality.

[0025] To solve this problem, the invention provides a method according to claim 1 and a device according to the further independent claim.

[0026] Advantageous embodiments are the subject of the dependent claims.

[0027] According to one aspect of the invention, a method for drop-based additive manufacturing of a component is provided, comprising: a) applying a liquid material drop by drop using a print head in a working chamber to build up a substrate for forming the component, in particular layer by layer; and b) capturing at least one drop in flight after ejection from the print head before impact on the substrate using at least one camera arranged outside and / or inside the working chamber and at least one exposure device, wherein the position of the at least one camera and the position of the at least one exposure device are coupled to the position of the print head; and comparing at least one physical parameter of the captured drop with a reference value to detect a change in the at least one physical parameter, wherein step b) is repeated at predetermined intervals during step a).

[0028] The term "workspace" refers to the area required to create the component using a drop-by-drop build-up process. This can be defined, for example, by the necessary travel distance of the base and / or the print head for component creation. The workspace can be open to the environment or it can be an area at least partially, and especially completely, separated from the environment. In a workspace that is partially, and especially completely, separated, it is possible to create a protective gas atmosphere within this area.

[0029] In some embodiments, the camera and / or the exposure unit are arranged outside the workspace. If the camera and / or the exposure unit are insensitive to the conditions prevailing in the workspace – especially heat – or if measures have been taken to protect the components against the effects of heat, the camera and / or the exposure unit can also be arranged inside the workspace.

[0030] According to the invention, the position of the at least one camera and the position of the at least one exposure device are coupled to the position of the print head.

[0031] This coupling means that the camera's field of view (e.g. 6.5 mm in one embodiment) remains essentially constant throughout the entire component creation process and moves "upwards" with the print head.

[0032] Such coupling is lacking in the solutions described in the prior art. Without such coupling, the camera eventually (depending on the size of the component) only sees the component and no longer the droplet, or the camera and lighting are arranged in such a way that they capture the entire area required for component production. In the latter case, the detection area is very large, especially at the beginning, which leads to significant distortions in the camera images. Furthermore, in most cases, the camera is not oriented perpendicular to the direction of droplet fall, which also results in distortions in the images.

[0033] The term "reference value" can refer to both a comparative value (e.g., from a reference measurement) and a (predefined) target value / setpoint (or setpoint limit, in the sense of less than / greater than).

[0034] In some embodiments, the comparison with the reference value involves a simple comparison of parameters, for example, from a reference image. If, for instance, the droplet in the current image is lower than in the reference image, this indicates that the droplet velocity is too high. In some embodiments, the comparison with the reference value also includes measuring absolute values ​​(e.g., the droplet velocity is measured to be 2.5 mm / s) and comparing them to the target value (e.g., 2.0 mm / s).

[0035] In the solution according to the invention, droplet testing during component manufacturing is much more accurate than in the prior art. Changes exceeding a predetermined limit are detected more precisely and effectively. Thus, droplet testing is significantly improved, which contributes to improvements in process quality and process duration.

[0036] In some embodiments, step b) includes step: b1) capturing the at least one drop in the working space during step a) using the at least one camera in transmitted light.

[0037] The transmitted light method produces images in which the droplet edges have very high contrast. This allows the droplet outlines and centers to be captured with low computing power, and the target values ​​can then be easily calculated.

[0038] In particular, step b) or b1) includes the step: bla) capturing the at least one drop in the working space during step a) by means of the at least one camera in transmitted light, wherein the position of the at least one camera and the position of the exposure device are outside the working space and / or are coupled to the position of the printhead.

[0039] In some embodiments, the camera used in step b), or at least one of several cameras used in step b), is an area scan camera. In some embodiments, the camera used in step b), or at least one of several cameras used in step b), is a line scan camera. Combinations of area scan cameras and line scan cameras are also possible.

[0040] In some embodiments, step b) includes step b2) moving the at least one camera and the exposure device with the print head, in particular along the height direction (component height direction) Z-direction).

[0041] In some embodiments, step b) includes step: b3) providing the at least one camera behind protective glass to separate it from the workspace.

[0042] In some embodiments, step b) includes step b3a) providing the at least one camera with a predetermined minimum distance between the camera lens and the print head. The minimum distance depends on the base's range of motion (travel path) to prevent component collisions during operation. In particular, the minimum distance is kept as small as possible, as a smaller distance allows for more accurate measurements. For example, the minimum distance is 300 mm.

[0043] In some embodiments, step b) includes step: b4) providing the at least one exposure device behind protective glass to separate it from the work area.

[0044] In some embodiments, step b) includes step: b4a) providing at least one area light, in particular with an area greater than 1000 mm² 2 , especially between 1000 mm 2 and 50,000 mm 2, as an exposure device.

[0045] In some embodiments, step b) includes step: b4b) providing at least one area light of such a size that the entire image background of an associated camera is illuminated when recording a drop in transmitted light, as an exposure device.

[0046] In other words, the size of the area of ​​light depends on the distance from the camera to the droplet.

[0047] In embodiments of the invention, no stroboscopic lamp is used, which is only suitable for incident light illumination. With incident light illumination, the light would be reflected within the droplet, making the droplet's contour barely discernible to the human eye. The reflections within the droplet would hardly allow for precise edge detection.

[0048] In one specific configuration, a surface light source (80x80mm) is used, emitting collimated (not a laser, but approximately directional) light. The light source is large enough to illuminate the entire background of the image, making the contours of the drops visible. It also preferably has a very short flash duration, e.g., approximately 10p, to minimize motion blur from rapidly falling drops at 2m / s and thus also the measurement inaccuracy. Furthermore, the light intensity is chosen to be high enough to ensure that sufficient light reaches the camera sensor within the flash duration.

[0049] In some embodiments, step b) comprises step b5) capturing the at least one droplet with a camera beam path directed from the camera through an area between the printhead and the substrate to the exposure device. In some embodiments, step b) comprises step b6) capturing the at least one droplet with a first camera and a second camera arranged at an angle of 1° to 175°, in particular 30° to 120°, more particularly 45° to 100°, and more particularly 60° to 90° to each other.

[0050] As mentioned above, the cameras can be, for example, area scan cameras or line scan cameras, or one of the cameras is an area scan camera and the other is a line scan camera.

[0051] In some embodiments, step b) includes step: b7) capturing at least one contour of the at least one droplet.

[0052] In some embodiments, step b) includes step: b8) recording the drops with the at least one camera having a pixel size of 5 pm or less.

[0053] In some embodiments, step b) includes step: b9) capturing at least one contour line of the at least one droplet with subpixel-accurate edge finding.

[0054] In some embodiments, step b) includes step: b10) flash-like exposure of the at least one droplet with a flash time of 1 ps to 30 ps, ​​in particular 5 ps to 20 ps, ​​more particularly 10 ps to 15 ps.

[0055] In some embodiments, step b) includes step: b11) acquiring the physical parameter from the group which includes a drop diameter, an averaged drop diameter, a drop velocity, a drop center position, a drop cross-sectional area, a Feret diameter, a deviation of drop edges from a reference shape and a drop compactness.

[0056] In some embodiments, a standard deviation is also determined for each recorded physical parameter. The standard deviation is advantageous for evaluation in order to take the variation into account, as it allows one to determine how stable the droplet diameter is. One goal of process control, for example, is to achieve the most stable droplet diameter possible; thus, the lowest possible variation is advantageous. Therefore, determining the variation via the standard deviation is also advantageous for droplet testing.

[0057] Accordingly, in some embodiments, step b) includes step: b12) acquiring the physical parameter and determining a standard deviation thereof. In some embodiments, step b) includes step: b13) acquiring a change in the physical parameter from the group which includes a drop diameter, an averaged drop diameter, a drop velocity, a drop center position, a drop cross-sectional area, a Feret diameter, a deviation of drop edges from a reference shape, and a drop compactness.

[0058] In some embodiments, step b) includes step: b14) detecting a change in the physical parameter and determining a standard deviation of the change in the physical parameter.

[0059] In some embodiments, step b) includes step: b15) comparison of a current image of the at least one drop with a reference image.

[0060] In some embodiments, step b) includes step: b16) Recording the at least one drop with a first and a second camera with a predetermined delay between the recordings and determining a drop velocity by comparing the recordings.

[0061] In some embodiments, step b) includes step: b17) Adjusting the reference value depending on previous passes of step b).

[0062] Particularly preferred embodiments of the method include the further step: c) continuing step a) if no change or only a change up to a predetermined level is detected in step b), and, if a change of the at least one physical parameter is detected beyond a predetermined level, stopping step a) and moving the printhead to a drop control position located away from the substrate and / or the working space, and measuring at least one drop parameter of drops ejected by the printhead in the drop control position.

[0063] In some embodiments, the drop control position is located outside the working area. However, the drop control position can also be located within the working area. The drop control position is specifically positioned further away from the substrate to provide more space for measurement and / or more sophisticated measurement technology.

[0064] In some embodiments, step b) is performed using an in-situ testing device, and step c) is performed using a droplet measuring device that differs from the in-situ testing device. In particular, when step b) is performed using an in-situ testing device, it is possible to change the at least one physical parameter without completely stopping step a). To change the at least one physical parameter, it is only necessary to extend the time interval between two drops ejected by the printhead such that the at least one physical parameter can be adjusted during this time interval. In other words, droplet ejection is delayed or briefly interrupted until the at least one physical parameter has been adjusted.

[0065] In some embodiments, step a) is carried out such that the print head is located in a predetermined distance range, in particular from 5 mm to 10 mm, from the substrate surface, wherein in step c) in the drop control position a measurement of the drops is carried out by means of camera recordings with a larger image section compared to this distance range and by means of image evaluation on these camera recordings.

[0066] In some embodiments, depending on the measurement in step c), step a) is continued without or with process adjustment, or the production of the component is aborted.

[0067] Some embodiments include the step:

[0068] Adjusting process parameters in step a) depending on the comparison in step b) and / or the measurement in step c).

[0069] In some embodiments, step b) includes step: b18) performing an image evaluation in real time, in particular such that the results for one image are output before the next image is taken.

[0070] In some embodiments, step b) includes step: b19) Performing the following sequence of steps at each predetermined interval: image acquisition, image transmission, image processing and result output.

[0071] In some embodiments, step b) includes step: b20) capturing images with a predetermined time interval between image captures, wherein in each time interval the steps image capture, image transmission, image processing and result output are performed.

[0072] According to a further aspect, the invention provides a device for drop-based additive manufacturing of a component, comprising: a manufacturing device comprising a base arranged in a working space, a printhead for dropwise application of liquid material onto the base, a motion device for relative movement of the printhead and base, and a computer-implemented control system, and configured for dropwise application of the liquid material by means of the printhead in the working space in order to build up a substrate on the base for forming the component, and an in-situ inspection device comprising at least one camera and at least one exposure device, wherein the position of the at least one camera and the position of the at least one exposure device are coupled to the position of the printhead.wherein the in-situ testing device is configured to capture at least one droplet in flight after ejection from the printhead before impact on the substrate and to compare at least one physical parameter of the captured droplet with a reference value in order to detect a change in the at least one physical parameter, and to repeat the capture and comparison at predetermined intervals during the application of the liquid material in the build-up of the substrate.

[0073] In some embodiments of the device, the camera and / or the exposure unit are arranged outside the workspace. If the camera and / or the exposure unit are insensitive to the conditions prevailing in the workspace – especially heat – or if measures have been taken to protect the components from the effects of heat, the camera and / or the exposure unit can also be arranged inside the workspace.

[0074] In the device according to the invention, the position of the at least one camera and the position of the at least one exposure unit are coupled to the position of the print head. The camera's field of view (e.g., 6.5 mm in one embodiment) thus remains essentially constant throughout the entire component production process and moves "upwards" with the print head. The definitions of "reference value" and "working area" have already been explained above and apply equally to the device.

[0075] In the device according to the invention, droplet testing during component manufacturing is much more accurate than in the prior art. Changes exceeding a predetermined limit are detected more precisely and effectively. Thus, droplet testing is significantly improved, which contributes to improvements in process quality and process duration.

[0076] Some embodiments of the device further include: a droplet measuring device for measuring at least one droplet parameter of drops ejected by the printhead when the printhead is moved from a printing position to a droplet control position remote from it, wherein the control is configured, depending on an output from the in-situ testing device, to continue the substrate build-up without interruption or to interrupt or stop the build-up and to initiate a measurement of the at least one droplet parameter by the droplet measuring device with the printhead moved to the droplet control position. That is, depending on the output of the in-situ testing device:

[0077] • the substrate build-up continued without interruption (-> no or only minor, negligible deviations that do not require correction), or

[0078] • substrate build-up is interrupted (-> drop output is either delayed to adjust the parameters accordingly between two drops, or substrate build-up is interrupted for measurement in the drop control position, the parameter is adjusted if necessary, and build-up is then resumed), or

[0079] • the substrate build-up was stopped (-> parameter deviation so large that the required quality of the component being built can no longer be guaranteed).

[0080] In some embodiments, the droplet measuring device is arranged outside the working area. In other embodiments, the droplet measuring device is arranged inside the working area; this has advantages because a better protective atmosphere generally prevails in the working area and / or because the droplet measurement can be carried out under more comparable conditions than outside the working area.

[0081] In some embodiments of the device, the control system is designed to cause the device to carry out the method according to one of the above-mentioned configurations.

[0082] The camera used can be, for example, an area scan camera or a line scan camera. It is also possible to arrange multiple area scan cameras and / or multiple line scan cameras.

[0083] In some embodiments, the in-situ testing device includes at least a first and a second camera.

[0084] In some embodiments, the in-situ testing device is provided to have an exposure device arranged on the opposite side of the working space for area-wide exposure of the at least one droplet to be recorded in transmitted light for each camera.

[0085] In some embodiments, the in-situ testing device is provided to have at least one protective glass between the camera and the working area.

[0086] In some embodiments, the in-situ testing device is provided to have at least one protective glass between the exposure device and the working area.

[0087] In some embodiments, the in-situ testing device is provided to have at least two cameras arranged at an angle of 1° to 175°, in particular 30° to 120°, more particularly 45° to 100°, and more particularly 60° to 90° to each other. In some embodiments, the in-situ testing device is provided to have at least one area light, in particular with an area greater than 1000 mm². 2 , especially between 1000 mm 2 and 50,000 mm 2 , as an exposure device.

[0088] In some embodiments, the in-situ testing device is provided to have at least one camera with a lens at a minimum distance of 300 mm from the print head.

[0089] In some embodiments, the in-situ testing device is provided to have at least one area light of such a size that the entire image background of an associated camera is illuminated when a drop is recorded in transmitted light.

[0090] Some embodiments of the device include a user interface.

[0091] In some embodiments, the user interface is configured to display a current image of the at least one droplet captured by the in-situ testing device and a reference image.

[0092] In some embodiments, the user interface is configured to select at least one physical parameter of the at least one droplet to be detected.

[0093] In some embodiments, the user interface is configured to display a current image of the at least one drop recorded by the in-situ testing device and at least one limit value for a parameter to be detected.

[0094] In some embodiments, the user interface is configured to set a delay time between recordings from multiple cameras of the in-situ testing device.

[0095] In some embodiments, the user interface is configured to start the measurement using the drop measuring device.

[0096] In some embodiments, the in-situ testing device is configured to perform image evaluation in real time, in particular such that the results for one image are output before the next image is taken.

[0097] In some embodiments, the in-situ test device is configured to perform the following sequence of steps at each predetermined interval: image acquisition, image transmission, image processing, and result output. In some embodiments, the in-situ test device is configured to acquire images with a predetermined time interval between image acquisitions, with the steps of image acquisition, image transmission, image processing, and result output being performed at each time interval.

[0098] Some embodiments of the invention relate to a method and a device for measuring drops in a drop-based 3D printing process.

[0099] In some embodiments, it is possible to measure the process properties during the printing process of a drop-based 3D printing process using a novel droplet measurement method and to recognize at what point a parameter adjustment of the process is necessary.

[0100] In particular, embodiments of the invention provide a method and a device for analyzing the process quality of a drop-based 3D printing process using optical sensors.

[0101] Exemplary embodiments of the invention are explained in more detail below with reference to the accompanying drawings. These show:

[0102] Fig. 1 shows a schematic side view of an embodiment of a device for drop-based additive manufacturing of a component with an in-situ testing device during the execution of an embodiment of a method for drop-based additive manufacturing of a component with in-situ testing;

[0103] Fig. 2 shows a schematic side view of an embodiment of the device during the execution of an embodiment of the method, showing a coupling of the movements of a print head and an image acquisition device and an exposure device of the in-situ testing device as well as the effects of the coupling;

[0104] Fig. 3 shows a schematic top view of an embodiment of the device in which the in-situ testing device has several cameras, during the execution of the corresponding method using several cameras;

[0105] Fig. 4 is a schematic top view of an embodiment of the device with a droplet measuring device; Fig. 5 is a schematic top view of another embodiment of the device with a droplet measuring device;

[0106] Fig. 6 shows a schematic flowchart of an embodiment of the method with drop measurement at a drop measurement position;

[0107] Fig. 7 is a photograph showing an example image taken by a camera during in-situ testing or in-situ measurement during drop-based additive manufacturing;

[0108] Fig. 8 is a photograph showing a first embodiment of a user interface for operating the device and / or controlling the method; and

[0109] Fig. 9 is a photograph showing a second embodiment of the user interface.

[0110] Figures 1 to 3 show different views of embodiments of a device 10 for drop-based additive manufacturing of a component 12 with an in-situ testing device 14. The device 10 is shown in each of these views performing embodiments of a method for drop-based additive manufacturing of the component 12 with in-situ testing. Figures 4 and 5 show further embodiments of the device 10, which, in addition to the in-situ testing device 14, is equipped with a drop measurement device 16. A flowchart illustrating embodiments of the method, in which drop measurement is performed as needed in addition to the in-situ testing, is shown in Figure 6.

[0111] According to Figures 1 to 6, a method for drop-based additive manufacturing of a component 12 is proposed, comprising the steps: a) drop-wise application of a liquid material by means of a print head 20 in a working chamber 22 to build up a substrate 24 for forming the component 12, and b) capturing at least one drop 18 in flight after ejection from the print head 20 before impact on the substrate 24 by means of at least one camera 26, 26.1, 26.2 arranged outside and / or inside the working chamber 22 and at least one exposure device 28, 28.1, 28.2, wherein the position of the at least one camera 26, 26.1, 26.2 and the position of the at least one exposure device 28, 28.1, 28.2 are determined.2 are coupled to the position of the printhead 20, and compare at least one physical parameter of the captured drop 18 with a reference value to detect a change in the at least one physical parameter, wherein step b) is repeated at predetermined intervals during step a).

[0112] In some embodiments, an in-situ measurement of the drop 18 is performed during the in-situ test. In some embodiments of the method, an example of which is shown as a flowchart in Fig. 6 and which can be carried out, for example, with the embodiments of the device 10 shown in Figs. 4 and 5, the method includes the additional step: c) continuing step a) if no change or only a change up to a predetermined level is detected in step b), and, if a change of the at least one physical parameter is detected beyond a predetermined level, stopping step a) and moving the printhead 20 to a drop control position 30, which is arranged away from the substrate 24 and / or the working space 22, and measuring at least one drop parameter of droplets 18 ejected by the printhead in the drop control position 30.

[0113] According to Figures 1 to 5, the device 10 for drop-based additive manufacturing of a component 12 is equipped with: a manufacturing unit 32, which has a base 34 arranged in the working space 22 – for example, configured as a build plate 36 –, the print head 20 for applying liquid material dropwise to the base 34, and a movement device 38 for moving the print head 20 and the base 34 relative to each other, and a computer-implemented control 40, and is configured for applying the liquid material dropwise by means of the print head 20 in the working space 22 in order to build up a substrate 24 on the base 34 for forming the component 12, and the in-situ inspection device 14, which has at least one camera 26, 26.1, 26.2 arranged outside and / or inside the working space 22 and at least one exposure device 28, 28.1, 28.2, wherein the position of at least one camera 26, 26.1 , 26.2 and the position of the at least one exposure device 28, 28.1, 28.2 are coupled to the position of the printhead 20, wherein the in-situ testing device 14 is configured to detect at least one droplet 18 in flight after ejection from the printhead 20 before impact on the substrate 24 and to compare at least one physical parameter of the detected droplet 18 with a reference value in order to detect a change in the at least one physical parameter, and to repeat the detection and comparison at predetermined intervals during the application of the liquid material during the build-up of the substrate 24.

[0114] In particular, the base 34 is stationary and the movement device 38 is designed to move the printhead 20 in the x, y and z directions. The at least one camera 26, 26.1 , 26.2 and the at least one exposure device 28, 28.1 , 28.2 are moved accordingly.

[0115] As illustrated in particular in Fig. 2, this coupling of the movements of print head 20, camera 26, 26.1 , 26.2 and exposure unit 28, 28.1 , 28.2 enables the viewing area 42 of the camera 28, 28.1 , 28.2 (in one embodiment e.g. 6.5mm) to remain essentially constant throughout the entire component production process and to move “upwards” with the print head 20.

[0116] In some embodiments of the device 10, examples of which are shown in Figures 4 and 5, the device 10 further comprises a drop measuring device 44 for measuring at least one drop parameter of drops 18 ejected by the printhead 20 when the printhead 20 has been moved from a printing position 46 to a drop control position 30 located away from it. In these embodiments, the control unit 40 is specifically configured, depending on an output from the in-situ test device 14, to continue the build-up of the substrate 24 without interruption, or to interrupt or stop the build-up and to initiate a measurement of the at least one drop parameter by the drop measuring device 44 with the printhead moved to the drop control position 30.

[0117] Thus, step a) can be carried out such that the printhead is located at a predetermined small distance range, in particular from 5 mm to 10 mm, from the substrate surface, wherein the field of view of the at least one camera 28, 28.1, 28.2 of the in-situ inspection device 14 is directed towards this smaller distance range and is kept constant due to the coupling. In step c), a measurement of the drops can then be carried out in the drop control position by means of camera images with a larger field of view compared to this distance range and by means of image evaluation of these camera images.

[0118] Exemplary embodiments of the device 10 and the method are described in more detail below with reference to the accompanying figures.

[0119] Fig. 1 shows a schematic side view of a first embodiment of the device 10 for drop-based additive manufacturing of a component 12 with the in-situ testing device 14 equipped with a camera 26 during the execution of a first embodiment of the method for drop-based additive manufacturing of a component with in-situ testing.

[0120] In particular, Fig. 1 shows the manufacturing device 32 with printhead 20, motion device 38, control unit 40, a build plate 36 as a base 34, and the component 12 – substrate 24 – currently being assembled. The in-situ testing device 14 has the camera 26 on a first side 48.1 of the work area 22 and the exposure device 28 for illuminating the camera image in transmitted light on the opposite third side 48.3 of the work area 22. The camera 26 and the exposure device 28 are each separated from the hot area of ​​the work area 22 by a heat-resistant disc 50. In some embodiments, other measures for heat protection are provided, such as a greater distance or heat-resistant components for the camera 26 and exposure device 28, so that the heat-resistant discs 50 can then be omitted. In particular, the beam path 52 of a camera 26, 26.1 , 26 is shown in Fig. 1.2 of the in-situ test facility shown.

[0121] Embodiments of the method and the device 10 are configured for a novel measurement method of a droplet-based 3D printing process. Using one or more cameras 26, 26.1, 26.2 – in particular area scan cameras (or in some embodiments also line scan cameras) – with a transmitted light illumination concept, the droplet 18 is captured and evaluated in flight after ejection from the nozzle 54 and before impact on the component 12.

[0122] Since the position of the printhead 20 changes above the height of the component 12, the cameras 26, 26.1, 26.2 and the associated exposures - exposure devices 28, 28.1, 28.2 - are on controllable axes 56 (in the sense of motion mechanisms) which are coupled, e.g. in terms of control technology, to the position of the printhead 20, or, if possible, are attached directly to the printhead 20 (i.e., physically coupled to the printhead 20).

[0123] In some embodiments, the cameras 26, 26.1, 26.2 and lighting devices - exposure unit 28, 28.1, 28.2 - are located outside the work area 22 (i.e., an area in which the workpiece is manufactured) to avoid being exposed to the high temperatures.

[0124] For this purpose, the cameras 26, 26.1, 26.2 and lighting devices - exposure unit 28, 28.1, 28.2 - can be arranged behind protective glass - temperature-resistant disc 50 - or similar and / or positioned with sufficient distance (= distance so that the temperature at the position of camera and lighting does not exceed a limit) to the working space 22.

[0125] According to Fig. 1, the optical beam path 52 passes through two temperature-resistant, transparent disks 50 (refraction of light may be taken into account) and centrally through the area between print head 20 and component 12.

[0126] Fig. 2 illustrates the technical effect of the described control-related or physical coupling. The component 12 is built up layer by layer during 3D printing. During the build of each layer, the print head 20 is moved in the x and y directions according to the component plan stored in the memory of the controller 40. The successively produced component layers 58 are shown on the build plate 36. After a component layer is completed, the print head 20 is moved upwards – in the Z direction. The camera 26 and the exposure unit 28 move with the print head 20. The beam path 52 is thus always directed to the appropriate position below the nozzle 54 of the print head 20. When the print head 20 is moved in the Z direction, the camera 26 and exposure unit 28 also move accordingly. In Fig.Figure 2 shows the arrangement of printhead 20, camera 26, and exposure unit 28, and the field of view 42 of the camera 26 at the beginning of component production, with solid lines. Dashed lines show the corresponding arrangement of printhead 20, camera 26, and exposure unit 28, and the field of view 42, during the production of a higher component layer 58. As can be seen in Figure 2, the field of view 42 remains essentially constant throughout the entire component production process.

[0127] In some embodiments of the device, examples of which are shown in Figs. 3 to 5, the in-situ testing device 14 has not just one camera 26, but several cameras, for example a first camera 26.1 and a second camera 26.2, with associated exposure devices 28.1 , 28.2 for transmitted light exposure.

[0128] Since the characteristics of the droplets 18 can only be analyzed from one perspective, i.e., in two dimensions, with a single camera 28 – first camera 28.1 – and the ejected droplets 18 can only be approximated to the shape of a triaxial ellipsoid, three-dimensional acquisition of the droplet 18 is advantageous to increase the measurement accuracy (compared to two-dimensional acquisition of the droplet 18). For this purpose, at least one additional camera – second camera 28.2 – is provided for measuring the third dimension, as shown in Figures 3 to 5. The measurement of the angle in space at which the droplet 18 is ejected from the nozzle can also be performed with an additional camera 28.2. This at least second (=additional) camera 28.2 can be positioned relative to the first camera 28.1 (around the Z-axis) at an angle of approximately 1° to approximately 175°, in particular from approximately 30° to approximately 120°, further in particular from approximately 45° to approximately 100°, and further in particular from approximately 60° to approximately 90°. The best results with regard to high measurement accuracy can be achieved in particular with an angle of approximately 90° to the first camera 28.1 (rotated around the Z-axis).

[0129] Fig. 3 schematically and by way of example shows the camera setup in a second embodiment of the device 10. In particular, Fig. 3 shows a top view of the device 10, with the printhead 20 omitted for illustrative purposes in order to show the position of the drop 18 at the printing position 46. The first camera 26.1 is located on the first side 48.1 of the working chamber 22, the associated first exposure unit 28.1 is located on the third side 48.3 of the working chamber 22, the second camera 26.2 is located on the second side 48.2 of the working chamber 22, and the associated second exposure unit 28.2 is located on the fourth side 48.4 of the working chamber, opposite the second side 48.2. In this example, the beam paths 52.1, 52.2 of the first and second cameras 26.1, 26.2 are rotated approximately 90° relative to each other with respect to the Z-axis. Also in this example, the cameras 26.1, 26.2 and the exposure devices 28.1, 28.2 are each separated from the hot area of ​​the working chamber 22 at the printing position 46 by a temperature-resistant disc 50. The control unit 40 (shown only schematically in Fig. 1 and not shown in the other figures for illustrative purposes) is computer-implemented and includes a processor and memory. An evaluation unit 60 is implemented in the control unit 40, which evaluates the images from the cameras 26, 26.1, 26.2 in real time. In particular, image processing software is loaded into the control unit 40.

[0130] For testing / measurement, at least one physical parameter of drop 18 is recorded and compared with a reference value. The reference value can, for example, be predefined or taken from a reference image taken with the device optimally adjusted.

[0131] For example, the parameter is determined by edge detection on droplet 18.

[0132] Fig. 7 shows an example of an image taken during in-situ testing, in particular during in-situ measurement. The printhead 20, the illumination by the exposure unit 28 as background 60, the drop 18, and the component 12 to be built up – substrate 24 – are shown.

[0133] In particular, the measurement accuracy can be further improved by combining a pixel size of camera 26, 26.1, 26.2 (or the camera sensor) of 5 pm or less with subpixel-accurate edge detection. Accordingly, some embodiments provide for image evaluation with subpixel-accurate edge detection.

[0134] It should be noted that the term "pixel" or "pixel size" can refer to both a "sensor pixel" or "sensor pixel size" and an "image pixel" or "image pixel size." The "sensor pixel" or "sensor pixel size" describes the pixel or pixel size as it exists on the camera sensor. The "image pixel" or "image pixel size" is a value derived from the ratio of the image area to be captured to the sensor size and is usually specified in [length / pixel]. The image pixel or image pixel size can, for example, depend on the lens used in conjunction with camera 26, 26.1, 26.2.

[0135] The term "subpixel-accurate edge detection" means that an edge, i.e., the transition between the drop (dark) and the background (light) in the image – see Fig. 7 – can be detected even if it lies within a single pixel (not just between two pixels). This further improves the measurement accuracy for detecting a drop 18, since the drop size – an example of the physical parameter – can be calculated not only based on whole pixels but also on fractional pixels.For example, if the drop 18 on a captured image starts at about half of a first pixel, extends over the entire second pixel and ends at about a quarter of the size of the third pixel; then, using subpixel-accurate edge detection, these edges (each the transition from the light background to the dark drop 18) can be detected within the pixels, and thus the size of the drop 18 can be determined to be about one and three-quarters the size of a pixel.

[0136] Without subpixel-accurate edge detection, brightness values ​​are only considered per pixel. For each pixel, an average brightness value is determined and compared to a predefined brightness threshold. If the average brightness value of the pixel is brighter than the threshold, the entire pixel is assigned to background 62. Similarly, the entire pixel is assigned to droplet 18 if the average brightness value of the pixel is darker than the threshold. In the example above, it is possible (depending on the brightness threshold) that the average brightness value of the third pixel is brighter than the brightness threshold and the average brightness value of the first pixel is darker than the brightness threshold.In this case, the size of drop 18 would be determined to be approximately the size of two pixels (larger than actual), or, if the average brightness value of the first pixel is brighter than the brightness threshold, the size of drop 18 would be determined to be approximately the size of one pixel (smaller than actual).

[0137] Furthermore, in some embodiments, it is provided that either the exposure time of the camera and / or the flash duration of the illumination are so short that the droplet velocity of approximately 1.5 to 3.0 m / s does not result in motion blur that affects the measurement accuracy.

[0138] The lighting devices 28, 28.1, 28.2 have area lamps with corresponding controls (implemented in the control unit 40).

[0139] The flash duration of the area lighting with collimated light is, for example, 10 ps and has sufficient power (e.g. over 300 W / m²).2 at a distance of 0.2 m) to display the droplet 18 on the camera sensor of camera 26, 26.1, 26.2 within this short time. The resulting images – see Fig. 7 – exhibit sharp edges at the transition between the dark droplet 18 and a bright background 62 of the area illumination resulting from the lighting and are therefore suitable for precise image processing.

[0140] Generally speaking, the shorter the flash duration, the less motion blur. Specifically, for this application, flash durations of approximately 1 ps to approximately 30 ps, ​​and furthermore, particularly from approximately 5 ps to approximately 20 ps, ​​and furthermore, particularly from approximately 10 ps to approximately 15 ps, are conceivable. Motion blur resulting from the duration of the flash duration can be at least partially reduced or even eliminated by subpixel-accurate edge detection.

[0141] The image processing software detects the droplets 18 and can evaluate them in various ways. To minimize the measurement error of the results, the data from all available cameras 26, 26.2, 26.1 are collected and combined. For example, the droplet diameter – an example of a physical parameter – can be determined using two cameras as shown in Figures 3 to 5 from the general formula of the triaxial ellipsoid. derived and in our case using the formulas r z = (Drop height camera 1 + Drop height camera 2') / 2 calculated / approximated.

[0142] To meet the geometric accuracy requirements of the printed components, the measured, averaged droplet diameter should deviate from the actual diameter by only a few micrometers. With currently used droplet diameters, an error of approximately 1 pm already corresponds to a volume change of about 1%. Errors of approximately less than 10% in volume can be compensated for by integrated compensation using a line sensor. This line sensor scans the surface of the component to be printed 12 at regular intervals, calculates for each area of ​​the surface how much the actual geometry deviates from the target geometry, and transmits this information directly to the axis controller (=example for motion device 38).

[0143] Due to the sensitivity described above for the precise calculation of the diameter, the image processing software should preferably enable the function of subpixel-accurate edge detection.

[0144] The measurement of droplet velocity – another example of a physical parameter – can be enhanced by the additional information from the third dimension / second (third, fourth...) camera.

[0145] 26.2 also yields results with reduced measurement uncertainty (i.e., with higher measurement accuracy). Since the time of droplet initiation in the printhead 20, for example, the time of an electrical trigger of a piezoelectric element (= actuator in the printhead 20 for initiating the droplet),

[0146] 18), does not have to coincide with the time of droplet emission, it is for the

[0147] Reliability of the speed calculation is advantageous when performed on an image depicting two or more drops (18). The calculation uses the...

[0148] Information from at least two cameras 26.1, 26.2 forms a three-dimensional vector (X, Y, Z) between the center point of the first ejected drop 18 and the center point of the second ejected drop 18. The magnitude of this vector multiplied by the predetermined / predefined drop frequency f in Hz yields the drop velocity Vdrop, as in the following formula. The following is a selection of further possibilities for evaluating the images - with further examples for at least one physical parameter to be recorded - which does not claim to be exhaustive and is therefore not to be understood as limiting:

[0149] In some embodiments of the method, the droplet center is calculated using blob analysis during image evaluation. In blob analysis, all pixels that are above or below a grayscale threshold—that is, brighter or darker than a brightness limit—are selected and evaluated. Specifically, the evaluation outputs the positions of the droplet centers in the X and Y directions of the specified coordinate system as a function of the printhead's position in the X and Y directions of the specified coordinate system. Unlike subpixel-accurate edge detection, blob analysis can only detect pixels as a whole. That is, if a pixel is brighter than the brightness limit, the entire pixel is assigned to the background (62), and similarly, if a pixel is darker than the brightness limit, the entire pixel is assigned to the drop (18).

[0150] In some embodiments of the method, the evaluation of the images involves calculating the cross-sectional area of ​​a droplet using blob analysis and processing the pixel count or area (in pixels) into a metric unit (for example, a length in [pm], [mm], [cm], etc., or an area [pm 2 ], [mm 2 ], [cm 2 ], etc.). carried out.

[0151] In some embodiments of the method, during the evaluation of the images, a calculation of Feret diameters, see [8], is performed in a variable angular interval using blob analysis, and the resulting data is processed to obtain a representative diameter for the drop.

[0152] In some embodiments of the method, the evaluation of the images involves calculating the deviation of the droplet edges from a perfect, round droplet of the same cross-sectional area, analogous to measuring a roundness tolerance.

[0153] In some embodiments of the method, a compactness K is calculated during the evaluation of the images, which is derived from the circumference U and the area A of a drop using the formula K=U 2 / (4*TT*A) calculated.

[0154] In some embodiments of the method, the image evaluation includes calculating a droplet diameter based on two edge detections each in the X and Y directions (two edges in the X direction and two edges in the Y direction) and calculating a theoretical diameter based on the distance between the previously determined edges. In some embodiments of the method, the image evaluation also includes calculating a droplet velocity using the droplet center point and the time between droplet ejection and camera trigger (the time the image was captured).

[0155] In some embodiments of the method, the evaluation of the images involves calculating a droplet velocity from a droplet center point of a first image with a defined image capture time and a droplet center point from a second image with a capture time a certain time later.

[0156] In particular, a change in at least one of the physical parameters, examples of which were mentioned previously, is detected by comparison with a reference value. In some embodiments, the in-situ test detects the at least one physical parameter or its change and a standard deviation thereof.

[0157] According to Figures 8 and 9, which show different embodiments of operator interfaces 64 of an operator interface 66 indicated in Figure 1, in some embodiments of the device 10 and the method, the process operator also has the option of saving a reference image 68 after successfully setting the process parameters to achieve the target values ​​for drop diameter, drop velocity, and / or drop angle. Using this reference image 68, it is possible to compare a current image 70 of the drops 18 with the reference image 68 of the well-adjusted drop generation process, whose drop size, drop velocity, and drop angle correspond to the input parameters of the slicer.

[0158] The following explains the labels used in the photo of Figs. 8 and 9 on the respective user interface 68 for actuation elements (e.g. fields on a touch screen), input and output fields:

[0159] 70 Current Picture - current picture

[0160] 68 Reference Picture - Reference picture 68

[0161] 72 Acquisition In Situ - In-Situ Acquisition

[0162] 74 Camera Delay Time - Camera delay time or

[0163] Recording time / recording interval

[0164] 76 Start Acquisition - Start data collection

[0165] 78 Acquisition - Recording

[0166] 80 Reference - Reference

[0167] 82 Set New Reference - Set a new reference

[0168] 84 Drop Control - drop control

[0169] 86 Drop Area - Drop surface [pm] 2 ]

[0170] 88 Max. Feret - maximum Feret diameter [pm]

[0171] 90 Max. Feret Angle - maximum Feret angle [°] 92 Compactness - compactness K

[0172] 94 Set - Adjust (preset Camera Delay Times / camera delay times - are configurable)

[0173] 96 AutoSet - automatic setting (program sets the Camera Delay Time automatically)

[0174] 100 Camera Delay Time can be entered manually - accurate to one decimal place.

[0175] 102 Distance Drop-Nozzle - Distance T drop - nozzle [mm]

[0176] 104 Number of Drops - Number of drops (reference is usually 1; if >1 is recorded, this usually indicates an error)

[0177] In the embodiment according to Fig. 8, even if depicted differently therein, all the aforementioned evaluation options of the in-situ measurement can be displayed on the user interface 68. Whether the droplet velocity has decreased or increased compared to the reference can be seen at the X-position of the droplet's center when the recording time (Camera Delay Time) is set to a fixed value. Similarly, the droplet's exit angle can be assessed based on the Y-position of the droplet's center. The results of the in-situ measurement also gain in reliability if at least one additional camera 28.2 is used, which is arranged at a 90° angle to the first camera 28.1.A possible change in the nozzle situation in the droplet exit area can alter the dynamics of the droplet detachment process and influence the droplet shape or lead to the formation of satellites, i.e., a second droplet that has different values ​​in the properties of diameter, speed and direction.

[0178] Alternatively, a person trained in this process can visually compare the actual images with the reference image 68 and then decide whether a drop measurement needs to be initiated for potential parameter adjustment. The drop measurement—which will be explained in more detail below with reference to Figures 4 to 6—can be initiated via the actuating element 84 labeled "Drop Control".

[0179] In the embodiment shown in Fig. 9, two lines 106 and 107 have been added to the current picture 70 to assist the operator in estimating the deviation of the drop velocity. The two additional horizontal lines 106 and 107, located below and above the reference line 108, indicate the vertical drop position for a velocity change of ± 0.2 m / s. They adjust dynamically according to the velocity last determined in the drop measurement. If the current velocity has increased by 0.2 m / s compared to the velocity at the time of the last reference measurement, the drop center on horizontal line 107 must lie below the horizontal reference line 108.If the current velocity has decreased by 0.2 m / s compared to the velocity at the time of the last reference measurement, the center of the drop must be on the horizontal line 106 above the horizontal reference line 108.

[0180] The measurement of the droplet velocity can also yield results with reduced measurement uncertainty (i.e., higher measurement accuracy) thanks to the additional information from the third dimension / second (third, fourth...) camera 28.2. Since the time of droplet initiation in the printhead 20, for example, the time of the electrical trigger of the piezoelectric sensor (the actuator in the printhead for initiating the droplet), does not necessarily coincide with the time of droplet ejection, it is advantageous for the reliability of the velocity calculation if it is performed on an image depicting two or more drops. For the calculation, a three-dimensional vector (X, Y, Z) is generated between the center point of the first ejected droplet and the center point of the second ejected droplet using the information from at least two cameras.

[0181] The evaluable image area between printhead 20 and component 12 during in-situ measurement is too small (5-10 mm in height) for typical printing parameters to depict two drops and calculate a drop velocity using the three-dimensional vector described above. To ensure velocity measurement despite the small evaluation area between printhead and component 12, some embodiments calculate the three-dimensional vector between the drop centers (X, Y, Z) of two different "Camera Delay Times" (abbreviated CDT in the formula). The length of the vector is then divided by the difference between the Camera Delay Times to obtain the drop velocity.

[0182] Therefore, if the center of the drop has traveled a total distance of 2.0 millimeters in the X, Y and Z directions with a camera delay time of 6 milliseconds, compared to the center of the drop with a camera delay time of 5 milliseconds, the resulting speed is 2.0 m / s.

[0183] Ideally, the same droplet would be captured 18 times consecutively, once with a first camera delay time of, for example, 5 ms, and once with a second camera delay time of, for example, 6 ms, where the first and second camera delay times are different. In practice, this is often not possible, for example, due to a limited maximum frame rate combined with a limited frame acquisition window. To still obtain meaningful results, a large number of droplet images, for example, 100, can be captured with the first camera delay time of, for example, 5 ms, and a large number of droplet images, for example, 100, with the second camera delay time, where the number of droplet images captured is identical for both camera delay times.From the multitude of droplet images with the first camera delay time and the multitude of droplet images with the second camera delay time, an averaged droplet center is calculated in each case, and the vector is calculated based on the averaged values.

[0184] In this embodiment, the camera delay time can be set manually, see reference numeral 94 in Fig. 9. However, it is also conceivable to have this type of speed calculation of the drops performed automatically, see reference numeral 96 in Fig. 9.

[0185] This provides a second way to detect a change in droplet velocity in in-situ measurement.

[0186] As mentioned above, according to Figures 4 and 5, a drop control position 30 can be provided at a distance from the substrate 24 or at a pressure position 46 during the setup of the substrate 24, where a more precise drop measurement can be carried out by the drop measuring device 16. The drop control position 30 can be provided inside the working space 22, as shown in Figure 4, or outside the working space 22, as shown in Figure 5. In some embodiments, the drop measuring device 16 has corresponding cameras 27, 27.1 for this purpose.

[0187] 27.2 and exposure devices 29, 29.1, 29.2 and an evaluation device (not shown, e.g. implemented in the controller 40) comparable to the in-situ test device 14, but with a larger field of view and / or greater accuracy. In some (not shown) embodiments, the same cameras 26.1, 26.2 and / or exposure devices are used.

[0188] 28.1. 28.2 As with the in-situ test device 14, the drop measuring device 16 is also used and, together with the printhead 20, is moved to the drop control position 30, where it can be operated with a larger field of view and / or greater accuracy. The same physical parameters as in the in-situ test can be recorded during drop measurement, or different physical parameters or more physical parameters than in the in-situ test can be recorded.

[0189] As shown in Fig. 6, in some embodiments of the method with a higher image acquisition frequency during manufacturing (step a)), an in-situ inspection (step b)) is performed. During image evaluation BA, the physical parameter is compared with a reference value.

[0190] Image analysis is performed in real time, such that the results for one image are displayed before the next image is captured. In other words, the steps of image acquisition, image transmission, image processing, and result output are carried out at each predetermined interval between in-situ tests.

[0191] If, in step b), the image evaluation BA shows no change or only a minor change 110, production continues. If, in step c), the image evaluation shows a change exceeding a predetermined level 112, production (step a)) is stopped, and the printhead 20 is moved to the drop control position 30, where the drop measuring device 44 performs the drop measurement. If this shows that the drop quality is still acceptable 114, production – step a) continues. If the drop quality is no longer acceptable 116, a decision is made at 118 as to whether production can be continued with a parameter adjustment 120. If so, parameters are adjusted 120, and then production – step a) continues with the adjusted parameters. If not, production is aborted 122.

[0192] The following explains some advantages of in-situ testing in advantageous configurations.

[0193] Several advantages of this optimized design arise from the fact that continuous feedback on droplet quality is obtained. Previously, droplet measurements were taken every XX drops / XX layers / XX minutes. This could result in the droplet quality deteriorating so significantly since the last measurement interval that the printed component 12 was no longer acceptable (OK) even after process adjustments and therefore had to be rejected (rejected). Furthermore, the printing process had to be interrupted for this droplet measurement. It was also possible that the printing process would be interrupted for the droplet measurement, only to find that no adjustments were necessary. This unnecessarily increased the overall printing time for component 12.

[0194] Now, by analyzing the droplet quality with an image acquisition frequency of, for example, approximately 3-5 Hz, changes, especially deteriorations, in droplet quality 112 can be detected early. In such a case 112, the printing process is interrupted for a droplet measurement – ​​step c) – and, based on this measurement, appropriate process adjustments 120 are made if necessary. This ensures that the printed component 12 has a sufficiently good / high quality at the end of the printing process (and therefore does not have to be rejected). Consequently, the amount of scrap in the production of components 12 is reduced.

[0195] Since droplet measurement is only performed when there is a need for process intervention, e.g. because the current image 70 shows conspicuous or detectable deviations from the reference image 68, non-productive times (=times that are not attributed to the actual printing process of component 12) can be reduced and material waste for unnecessary measurements, as in the case of regular droplet checks, can be avoided.

[0196] Assuming a printing process time of 4 hours and a regular control interval of 30 minutes (old, known procedure), the printhead 20 is moved to the drop control position 30 seven times to perform the drop measurement. There, a decision is made as to whether a parameter adjustment is necessary before printing continues, or even whether the print must be aborted because the component 12 can no longer achieve the required quality.

[0197] With in-situ measurement, the printhead 20 is moved to the drop control position 30 for drop measurement (step c) only if it is determined (e.g., by comparing the current drop image 70 with the reference image 68) that the process has changed. Ideally, this might not occur at all during the assumed printing process time of four hours, or it might only occur after two hours, for example. Furthermore, it is conceivable that during the assumed printing process time of four hours, a process change might only be detected once, twice, or three times, and thus the drop measurement would only be performed once, twice, or three times, respectively.For example, if a process change is detected three times, and the drop measurement is therefore carried out three times, four drop measurements and the associated printhead travel, time and material used for a drop check are saved over a printing process time of four hours compared to the "regular control intervals".

[0198] If the component 12 can still be salvaged despite the droplet 18 temporarily deteriorating, by adjusting a parameter 120 during droplet generation, the direct feedback and request for a process adjustment 120 ensures that the component 12 meets the required quality.

[0199] In-situ testing / measurement also ensures complete process documentation for components with increased safety requirements, such as in the aerospace sector.

[0200] In some embodiments, the in-situ testing device 14 has at least one first and one second camera 26.1, 26.2. The implementation of a second camera 26.2 in the drop measurement provides information about the third dimension, which is not visible when using only one camera 26. This improves the measurement accuracy of certain properties—physical parameters—such as drop diameter, drop velocity, and drop angle. Beyond more reliable results, this also makes it possible to compensate for drop misalignment caused by hardware wear or replacement during a printing process, thus preventing layer offsets in the component height.

[0201] This can also reduce the amount of waste during component manufacturing.

[0202] Furthermore, the presence of satellites (see above) can be detected in any case. Using one camera 26, the satellite may be hidden in the area of ​​the droplet 18, either in front of or behind the image plane. In this case, the presence of a satellite can only be detected by the second camera 26.2.

[0203] Embodiments of a method and device 10 for droplet measurement in a droplet-based 3D printing process have been described, wherein the method and the device are set up for analyzing the process quality of a droplet-based 3D printing process using optical sensors.

[0204] Reference symbol list:

[0205] 10 Device

[0206] 12 components

[0207] 14 In-situ test facility

[0208] 16 drop measuring device

[0209] 18 drops

[0210] 20 printhead

[0211] 22 Workroom

[0212] 24 substrate

[0213] 26 Camera

[0214] 26.1 first camera

[0215] 26.2 second camera

[0216] 27, 27.1, 27.2 Camera (droplet measuring device)

[0217] 28 Exposure equipment

[0218] 28.1 First exposure unit

[0219] 28.2 second exposure unit

[0220] 29, 29.1, 29.2 Exposure device (drop measuring device)

[0221] 30 Drop Control Position

[0222] 32 Manufacturing facility

[0223] 34 base

[0224] 36 building plates

[0225] 38 Movement device

[0226] 40 Control

[0227] 42 field of view

[0228] 44 Drop measuring device

[0229] 46 Print position

[0230] 48.1 first page

[0231] 48.2 second page

[0232] 48.3 third page

[0233] 48.4 fourth page

[0234] 50 temperature-resistant pane (protective glass)

[0235] 52, 52.1, 52.2 Beam path

[0236] 54 nozzle

[0237] 56 controllable axes (motion mechanism for moving camera or

[0238] Exposure equipment

[0239] 58 component layers

[0240] 60 Evaluation unit

[0241] 62 Background

[0242] 64 User interface

[0243] 66 User interface 68 Reference image

[0244] 70 current images

[0245] 72 In-situ recording

[0246] 74 Camera Delay Time (CDT)

[0247] 76 Start recording

[0248] 78 Recording

[0249] 80 Reference

[0250] 82 Set new reference

[0251] 84 Drop control / drop measurement

[0252] 86 droplet area [pm] 2 ]

[0253] 88 maximum Feret diameter [pm]

[0254] 90 maximum Feret angle [°]

[0255] 92 Compactness K

[0256] 94 Set - Adjust (preset Camera Delay Times / camera delay times - are configurable)

[0257] 96 AutoSet - automatic setting (program sets the Camera Delay Time automatically)

[0258] 100 Camera Delay Time can be entered manually - accurate to one decimal place.

[0259] 102 Distance droplet - nozzle [mm]

[0260] 104 Number of drops (reference is usually 1; if >1 is detected, this usually indicates an error)

[0261] 106 top line

[0262] 107 lower line

[0263] 108 Reference line

[0264] 110 no or only slight change

[0265] 112 Change beyond the predetermined extent

[0266] 114 drops, quality OK

[0267] 116 Drop quality not OK

[0268] 118 Parameter adjustment possible?

[0269] Adjust 120 parameters

[0270] 122 Demolition

[0271] BA Image Analysis

Claims

GROB-WERKE GmbH & Co. KG 1523 0142 P-WO Industriestraße 4 256A-PCT D-87719 Mindelheim Claims:

1. A method for drop-based additive manufacturing of a component (12), comprising a) drop-wise application of a liquid material by means of a print head (20) in a working chamber (22) to build up a substrate (24) for forming the component (12), and b) capturing at least one drop (18) in flight after ejection from the print head (20) before impact on the substrate (24) by means of at least one camera (26, 26.1, 26.2) and at least one exposure device (28, 28.1, 28.2), wherein the position of the at least one camera (26, 26.1, 26.2) and the position of the at least one exposure device (28, 28.1 . 28.2) are coupled to the position of the printhead (20), and compare at least one physical parameter of the ingested drop (18) with a reference value to detect a change in the at least one physical parameter, wherein step b) is repeated at predetermined intervals during step a).

2. The method according to claim 1, characterized in that step b) comprises at least one or more of the following steps: b1) capturing the at least one drop (18) in the working chamber (22) during step a) by means of the at least one camera (26, 26.1, 26.2) in transmitted light; b1) capturing the at least one drop (18) in the working chamber (22) during step a) by means of the at least one camera (26, 26.1, 26.2) in transmitted light, wherein the position of the at least one camera (26, 26.1, 26.2) and the position of the exposure device (28, 28.1, 28.2) are determined. 28.2) are located outside the working area (22) and / or are coupled to the position of the printhead (20); b2) Moving the at least one camera (26, 26.1, 26.2) and the exposure unit (28, 28.1, 28.2) with the printhead (20); b3) Providing the at least one camera (26, 26.1, 26.2) behind protective glass (50) to separate it from the working area (22); b3a) Providing the at least one camera (26, 26.1, 26.2) with a minimum distance of 300 mm between the lens of the camera (26, 26.1, 26.2) and the printhead (20); b4) Providing the at least one exposure unit (28, 28.1, 28.2) behind protective glass (50) to separate it from the working area (22); b4a) Provision of at least one area light, in particular with an area greater than 1000 mm² 2 , especially between 1000 mm 2 and 50,000 mm 2 , as an exposure device (28, 28.

1. 28.2); b4b) Providing at least one area light of such a size that the entire image background (62) of an associated camera (26, 26.1, 26.2) is illuminated by transmitted light when a drop (18) is captured; as an exposure device (28, 28.1, 28.2); b5) Capturing the at least one drop (18) with a beam path (52, 52.1, 52.2) of the camera (26, 26.1, 26.2) which is directed from the camera (26, 26.1, 26.2) through an area between print head (20) and substrate (24) to the exposure device (28, 28.1, 28.2); b6) Recording the at least one drop (18) with a first camera (26.1) and a second camera (26.2) arranged at an angle of 1° to 175°, in particular 30° to 120°, more particularly 45° to 100°, and more particularly 60° to 90° to each other.

3. A method according to any one of the preceding claims, characterized in that step b) comprises at least one or more of the following steps: b7) capturing at least one contour of the at least one drop (18); b8) recording the drop (18) with the at least one camera (26, 26.1, 26.2) with a pixel size of 5 pm or less; b9) capturing at least one contour line of the at least one drop (18) with subpixel-accurate edge detection; b10) flash-like exposure of the at least one drop (18) with a flash time of 1 ps to 30 ps, ​​in particular 5 ps to 20 ps, ​​more particularly 10 ps to 15 ps; b11) Identifying the physical parameter from the group comprising a drop diameter, an averaged drop diameter, a drop velocity, a drop center position, a drop cross-sectional area, a Feret diameter, a drop edge deviation from a reference shape, and a drop compactness;b12) Recording the physical parameter and determining a standard deviation thereof; b13) Recording a change in the physical parameter from the group comprising a drop diameter, an averaged drop diameter, a drop velocity, a drop center position, a drop cross-sectional area, a Feret diameter, a deviation of drop edges from a reference shape, and a drop compactness; b14) Recording a change in the physical parameter and determining a standard deviation of the change in the physical parameter; b15) Comparing a current image (70) of the at least one drop (18) with a reference image (68).

4. Method according to one of the preceding claims, characterized in that step b) comprises at least one of the following steps: b16) Recording the at least one drop (18) with a first and a second camera (26.1 , 26.2) with a predetermined delay between the recordings and determining a drop velocity by comparing the recordings; b17) Adjusting the reference value depending on previous iterations of step b).

5. Method according to one of the preceding claims, characterized by the further step: c) continuing step a) if no change or only a change up to a predetermined extent is detected in step b), and, if a change of the at least one physical parameter is detected beyond a predetermined extent, stopping step a) and moving the printhead (20) to a drop control position (30) located away from the substrate (24) and / or the working space (22), and measuring at least one drop parameter of drops ejected by the printhead (20) in the drop control position.

6. Method according to claim 5, characterized in that step a) is carried out such that the print head (20) is located in a predetermined distance range, in particular from 5 mm to 10 mm, from the substrate surface, and that in step c) in the drop control position (30) a measurement of the drops (18) is carried out by means of camera recordings with a larger image section compared to this distance range and by means of image evaluation on these camera recordings.

7. Method according to claim 5 or according to claim 6, wherein, depending on the measurement in step c), step a) is continued without or with process adjustment (120) or the production of the component (12) is aborted (122).

8. Method according to one of the preceding claims, characterized by adjusting process parameters (120) in step a) depending on the comparison in step b) and / or the measurement in step c).

9. A method according to one of the preceding claims, characterized in that step b) comprises at least one or more of the following steps: b18) performing image evaluation in real time, in particular such that the results for one image are output before the next image is captured; b19) performing the following sequence of steps at each predetermined interval: image capture, image transmission, image processing and result output; b20) capturing images with a predetermined time interval between image captures, wherein the steps image capture, image transmission, image processing and result output are performed at each time interval.

10. Device (10) for drop-based additive manufacturing of a component (12), comprising: a manufacturing unit (32) comprising a base (34) arranged in a working space (22), a print head (20) for drop-wise application of liquid material onto the base (34), and a motion device (38) for relative movement of printhead (20) and base (34) and a computer-implemented control (40) and is configured for dropwise application of the liquid material by means of the printhead (20) in the working space (22) in order to build up a substrate (24) on the base for forming the component (12), and an in-situ testing device (14) comprising at least one camera (26, 26.1, 26.2) and at least one exposure device (28, 28.1, 28.2), wherein the position of the at least one camera (26, 26.1, 26.2) and the position of the at least one exposure device (28, 28.1, 28.1) are determined by the position of the at least one camera (26, 26.1, 26.2).2) are coupled to the position of the printhead (20), wherein the in-situ testing device (14) is configured to detect at least one droplet (18) in flight after ejection from the printhead (20) before impact on the substrate (24) and to compare at least one physical parameter of the detected droplet (18) with a reference value in order to detect a change in the at least one physical parameter, and to repeat the detection and comparison at predetermined intervals during the application of the liquid material in the build-up of the substrate (24).

11. Device (10) according to claim 10, characterized by a drop measuring device (44) for measuring at least one drop parameter of drops ejected by the printhead (20) when the printhead (20) is moved from a printing position (46) to a drop control position (30) located away from it, wherein the control (40) is configured to continue the build-up of the substrate (24) without interruption, or to interrupt or stop the build-up, depending on an output from the in-situ test device (14), and to initiate a measurement of the at least one drop parameter by the drop measuring device (44) with the printhead (20) moved to the drop control position (30).

12. Device (10) according to claim 10 or 11, characterized in that the control (40) is configured to cause the device (10) to carry out the method according to one of claims 1 to 9.

13. Device (10) according to one of claims 10 to 12, characterized in that the in-situ testing device (14) 13.1 includes at least one first and one second camera (26, 26.1, 26.2) and / or 13.2 Each camera (26, 26.1, 26.2) has an exposure device (28, 28.1, 28.2) arranged on the opposite side of the working space (22) for area exposure of the at least one droplet (18) to be recorded in transmitted light; and / or 13.3 has at least one protective glass (50) between the camera (26, 26.1, 26.2) and the working space (22); and / or 13.4 has at least one protective glass (50) between exposure device (28, 28.1, 28.2) and working space (22); and / or 13.5 comprising at least two cameras (26, 26.1, 26.2) arranged at an angle of 1° to 175°, in particular 30° to 120°, more particularly 45° to 100°, and more particularly 60° to 90° to each other; and / or 13.6 at least one surface light, in particular with an area greater than 1000 mm² 2 , especially between 1000 mm 2 and 50,000 mm 2 , as an exposure device (28, 28.1, 28.2); and / or 13.7 which has at least one camera (26, 26.1, 26.2) with a lens at a minimum distance of 300 mm from the print head (20); and / or 13.8 at least one area light of such a size that the entire image background (62) of an associated camera (26, 26.1 , 26.2) is illuminated in transmitted light when recording a drop (18), as an exposure device (28, 28.1 , 28.2).

14. Device (10) according to one of claims 10 to 13, comprising a user interface (64) which is configured to: 14.1 to display a current image (70) of the at least one drop (18) recorded by the in-situ testing device (14) and a reference image (68); and / or 14.2 to select at least one physical parameter (86, 88, 90, 92, 102, 104) of the at least one drop (18) to be recorded; and / or 14.3 to display a current image (70) of the at least one drop (18) recorded by the in-situ testing device (14) and at least one limit value (106, 107) for a parameter to be measured therein; and / or 14.4 to set a delay time (74) between recordings of several cameras of the in-situ testing device (14); and / or 14.5 to start the measurement by the drop measuring device (84).

15. Device (10) according to one of claims 10 to 14, characterized in that the in-situ test device (14) is configured for: 15.1 Performing real-time image analysis, in particular by displaying the results for one image before the next image is taken; and / or 15.2 Performing the following sequence of steps at each predetermined interval: image acquisition, image transmission, image processing, and result output; and / or 15.3 Capturing images with a predetermined time interval between image captures, wherein in each time interval the steps of image capture, image transmission, image processing and result output are performed.

Images