Additive Manufacturing Control System

Abstract

Additive manufacturing control systems are provided, particularly systems and methods for controlling additive manufacturing systems. A powder bed melting apparatus may include an energy beam source that generates an energy beam and a deflector that applies the energy beam to melt powder material to create a 3D object based on an object model. The system may also include a characterizer that acquires information related to melting the powder material. The characterizer may be a sensor that measures the shape of the object, a processor that determines a physically-based model of the object, etc. The system may also include a comparator that determines a difference from the object model based on the information, and a compensator that modifies the energy applied to the powder material based on the difference. For example, the applied energy may be increased in regions requiring higher energy to completely melt the powder material (such as regions with thicker powder layers).

Description

[0001] This application is a divisional application of application number 201810399070.7, filed on April 28, 2018, entitled "Additive Manufacturing Control System".

[0002] Cross-reference to related applications

[0003] This application claims the benefit of U.S. Patent Application No. 15 / 582,457, entitled “ADDITIVE MANUFACTURING CONTROL SYSTEMS”, filed April 28, 2017, which is expressly incorporated herein by reference in its entirety. Technical Field

[0004] This disclosure generally relates to additive manufacturing systems, and more specifically to control systems in additive manufacturing. Background Art

[0005] Additive manufacturing (“AM”) systems, also described as 3D printing systems, can produce structures (called build pieces) with geometrically complex shapes, including some shapes that are difficult or impossible to create using conventional manufacturing processes. AM systems (such as powder bed fusion (PBF) systems) create build pieces layer by layer. Each layer, or each 'piece', is formed by depositing powder layers and exposing portions of the powder to an energy beam. The energy beam is applied to the molten region of the powder layer that coincides with the cross-section of the build piece in that layer. The molten powder cools and melts to form a piece of the build piece. This process can be repeated to form the next piece of the build piece, and so on. Each layer is deposited on top of the previous one. The resulting structure is a build piece configured piece by piece from scratch.

[0006] The intended build parts are to conform to the desired printing parameters, such as the desired shape, desired material density, and desired mechanical properties. However, build parts often do not perfectly conform to the desired printing parameters. In some cases, this lack of conformity may necessitate post-processing techniques, such as sanding and filing, to correct the shape of the build parts, which increases production costs. In other cases, build parts cannot be secured and must be discarded, which reduces yield and significantly increases production costs. Summary of the Invention

[0007] The following sections will describe in more detail several aspects of the apparatus and methods used in AM control systems.

[0008] In various aspects, the apparatus for powder bed melting may include: a powder bed melting system comprising an energy beam source that generates an energy beam and a deflector that applies the energy beam to melt powder material to create a three-dimensional (3-D) object based on an object model, a characterizer that obtains information related to the melting of the powder material, a comparator that determines the variation with the object model based on the information, and a compensator that modifies the energy applied to the powder material based on the variation.

[0009] In various aspects, the apparatus for powder bed melting may include an adaptive controller that provides instructions for printing a 3-D object (the instructions being based on a data model of the 3-D object); a powder bed melting system that prints the 3-D object based on the instructions; and a feedback system configured to sense the shape of at least a portion of the printed 3-D object, compare the sensed shape with a reference shape to determine a difference parameter, and update the instructions based on the difference parameter.

[0010] In various aspects, the powder bed melting method may include generating an energy beam, applying the energy beam to melt the powder material to create a (3-D) object based on an object model, obtaining information related to the melting of the powder material, determining the differences from the object model based on the information, and modifying the energy applied to the powder material based on the information.

[0011] In various aspects, the powder bed melting method may include providing instructions for printing a 3-D object (the instructions being based on a data model of the 3-D object), printing the 3-D object based on the instructions, sensing the shape of at least a portion of the printed 3-D object, comparing the sensed shape with a reference shape to determine a difference parameter, and updating the instructions based on the difference parameter.

[0012] Other aspects will become readily apparent to those skilled in the art from the following detailed description, wherein only a few embodiments are illustrated and described by way of illustration. As those skilled in the art will recognize, the concepts herein are capable of other and different embodiments, and many details can be modified in various other ways without departing from this disclosure. Therefore, the accompanying drawings and detailed description are to be considered illustrative rather than restrictive in nature. Attached Figure Description

[0013] The various aspects of the accompanying figures will now be presented in a detailed description by way of example rather than limitation, in which:

[0014] Figure 1A -D illustrates an example PBF system during different operational phases.

[0015] Figure 2The illustration shows a side view of an exemplary sagging deformation in a PBF system that can cause an overhang area.

[0016] Figure 3 An exemplary PBF device including closed-loop control is illustrated.

[0017] Figure 4 An exemplary PBF device including feedforward control is illustrated.

[0018] Figure 5 An exemplary operation of a comparator is illustrated.

[0019] Figure 6A -C illustrates an example of applying energy to a powder layer using modified printing instructions.

[0020] Figure 7 This is a flowchart illustrating an exemplary method for closed-loop compensation in a PBF system.

[0021] Figure 8A -C illustrates another exemplary application of energy to a powder layer using modified printing instructions.

[0022] Figure 9 This is a flowchart illustrating an exemplary method for feedforward compensation in a PBF system.

[0023] Figure 10A -E illustrates an exemplary PBF device with post-processing closed-loop control.

[0024] Figure 11 This is a flowchart illustrating another exemplary method for compensation in a PBF system. Detailed Implementation

[0025] The following detailed description with reference to the accompanying drawings is intended to provide a description of various exemplary embodiments of the concepts disclosed herein, and is not intended to represent only embodiments that may be practiced with respect to this disclosure. The term “exemplary” as used in this disclosure means “serving as an example, instance, or illustration” and should not necessarily be construed as being preferred or advantageous relative to other embodiments set forth in this disclosure. The detailed description includes specific details in order to provide those skilled in the art with a full and complete disclosure that adequately conveys the scope of the concepts. However, this disclosure may be practiced without these specific details. In some cases, well-known structures and components may be shown in block diagram form or omitted entirely to avoid obscuring the various concepts presented throughout this disclosure.

[0026] This disclosure relates to control systems in AM, such as powder bed melting (PBF). Current PBF systems can achieve component geometric accuracy between ±20 μm and ±130 μm, with R a=25μm surface roughness. The minimum wall thickness achievable by PBF is 150μm. On the other hand, electron beam melting (EBM) systems can achieve R... a Minimum wall thicknesses of 40 μm and 700 μm are required. This presents a challenge when smooth surfaces or sub-millimeter features are needed. Furthermore, the internal features of some 3D-printed parts may be non-uniform due to phenomena such as discontinuous melt vectors, "spheroidization" effects, uneven powder distribution, and incomplete melting. These phenomena can limit dimensional accuracy, speed, and throughput.

[0027] One cause of inhomogeneity is that the areas of powder material exposed to the energy beam can undergo volume shrinkage as the material melts, solidifies, and solidifies into a solid mass. In this respect, the height of the molten region can be lower than the remainder of the powder bed, resulting in a thicker layer of powder material deposited on top of these regions during the deposition of the next powder layer. In various embodiments, an additional thickness of powder can be determined, and the energy applied to the powder material can be increased to compensate for the increased energy required to melt the additional thickness of powder. For example, this approach can ensure that each layer is completely melted and reduce porosity in 3D constructs.

[0028] In various embodiments, the building block can be constructed based on an object model that specifies the desired shape of the building block. The object model may also include other desired properties of the building block, such as density, internal stress, and completeness of melt flow. Differences from the object model can be determined before, during, and / or after the printing process. For example, shrinkage of the actual building block can be determined by comparing the actual building block with the object model of the building block. Shrinkage can result in an additional thickness of powder deposited in the next powder layer. For example, the additional thickness of the powder layer on the shrinkage area can be determined based on the determined shrinkage. In some embodiments, shrinkage can be determined in real time by sensing the shape of the actual building block (e.g., by optical measurement). In some embodiments, shrinkage can be determined before the printing process based on, for example, a physics-based model that predicts the shape of the actual building block by calculating the thermal factor, gravitational factor, etc.

[0029] In various embodiments, improvements in 3D printer accuracy and throughput can be achieved by compensating for variations in ambient temperature, humidity, material chemistry and particle size, laser intensity, layer thickness, and differences in the geometry of nearby components.

[0030] In various embodiments, the 3D printer can print standardized test parts / patterns, which can then be scanned for comparison with an object model. Deviations and variances from the geometric data can be measured and calculated. These differences in printer performance can then be compensated for before printing the product assembly begins.

[0031] In various embodiments, optical scanning of the build piece can be performed before and after each layer is printed. A monitoring system can be established to scan the powder bed and determine the distribution of powder material after the powder coating process. If areas are not covered by powder, the coating mechanism can be reactivated to coat the layer. Improvements can be applied after scanning each layer using the monitoring system. If areas are missed or where only partial melting of powder material occurs during the energy beam scan, the energy beam can be activated to rescan these areas.

[0032] In various embodiments, a high-resolution thermal imaging system can be used to create a closed-loop feedback loop for self-calibration accuracy. The high-resolution thermal imaging device can monitor the formation of the molten carrier during energy beam discontinuities or misaligned exposures. The feedback loop can compensate for drift and width of the molten carrier, thereby maintaining geometric accuracy and print quality. A calibration coupon can be permanently fixed within the build chamber to ensure the accuracy of the imaging system.

[0033] Figure 1A -D illustrates corresponding side views of the exemplary PBF system 100 during different operational phases. As described above, Figure 1A The specific embodiment illustrated in -D is one example among many suitable examples of PBF systems employing the principles of this disclosure. It should also be noted that... Figure 1A Elements in the figures and other accompanying drawings of this disclosure are not necessarily drawn to scale, but may be drawn larger or smaller for better illustration of the concepts described herein. The PBF system 100 may include a depositor 101 capable of depositing each layer of metal powder, an energy beam source 103 capable of generating an energy beam, a deflector 105 capable of applying the energy beam to fuse the powder material, and a build plate 107 capable of supporting one or more components (such as build member 109). The PBF system 100 may also include a build base plate 111 placed within a powder bed container. The walls 112 of the powder bed container generally define the boundaries of the powder bed container, which is laterally sandwiched between the walls 112 and adjacent to a portion of the build base plate 111 below. The build base plate 111 may be gradually lowered to allow the depositor 101 to deposit the next layer. The entire structure may reside within a chamber 113 that can enclose other components, thereby protecting the equipment, enabling atmospheric and temperature regulation, and mitigating the risk of contamination. The depositor 101 may include a funnel 115 for containing powder 117 (such as metal powder) and a leveler 119 for leveling the top of each layer of deposited powder.

[0034] Specific reference Figure 1AThe figure shows a PBF system 100 after one component 109 has melted but before the next layer of powder has been deposited. In fact, Figure 1A The illustration shows the current state of the PBF system 100 at which sheets in multiple layers (e.g., 150 layers) have been deposited and melted to form (e.g., 150 sheets formed) a building block 109. The multiple layers that have been deposited have created a powder bed 121, which comprises deposited but unmelted powder.

[0035] Figure 1B A PBF system 100 is shown at a stage where the build base plate 111 can be lowered by a powder layer thickness 123. This lowering of the build base plate 111 causes the build member 109 and the powder bed 121 to descend by the powder layer thickness 123, such that the tops of the build member and the powder bed are lowered by an amount equal to the powder layer thickness than the top of the powder bed container wall 112. In this way, for example, a space with a consistent thickness equal to the powder layer thickness 123 can be created on top of the build member 109 and the powder bed 121.

[0036] Figure 1C A PBF system 100 is shown in the following stage: a depositor 101 is positioned to deposit powder 117 in a space created on the top surface of a component 109 and a powder bed 121, defined by a powder bed container wall 112. In this example, the depositor 101 moves gradually within the defined space while releasing powder 117 from a funnel 115. A leveler 119 can level the released powder to form a layer with a thickness substantially equal to the powder layer thickness 123 (see [link to documentation]). Figure 1B The powder layer 125 has a thickness of 123. Therefore, the powder in the PBF system can be supported by a powder material support structure, which may include, for example, a building plate 107, a building base plate 111, a building element 109, a wall 112, etc. It should be noted that the thickness of the powder layer 125 shown in the figure (i.e., powder layer thickness 123) is... Figure 1B ()) Greater than the references above Figure 1A The actual thickness used in the 150 examples of previous deposited layers discussed.

[0037] Figure 1D The PBF system 100 is shown in the following stage: wherein in the powder layer 125 ( Figure 1CFollowing deposition, energy beam source 103 generates energy beam 127 and deflector 105 applies the energy beam to melt the next piece in the building block 109. In various exemplary embodiments, energy beam source 103 may be an electron beam source, in which case energy beam 127 constitutes an electron beam. Deflector 105 may include deflection plates that can generate an electric or magnetic field that selectively deflects the electron beam to cause it to sweep across a designated area to be melted. In various embodiments, energy beam source 103 may be a laser, in which case energy beam 127 is a laser beam. Deflector 105 may include an optical system that uses reflection and / or refraction to manipulate the laser beam to scan the selected area to be melted.

[0038] In various embodiments, deflector 105 may include one or more balance rings and actuators that can rotate and / or translate the energy beam source to position the energy beam in place. In various embodiments, energy beam source 103 and / or deflector 105 may modulate the energy beam, for example, by turning the energy beam on and off as the deflector scans, such that the energy beam is applied only to appropriate regions of the powder layer. For example, in various embodiments, the energy beam may be modulated by a digital signal processor (DSP).

[0039] Figure 2 The illustration shows a side view that can cause an exemplary sagging deformation in a PBF system with a hanging area. Figure 2 A build plate 201 and a powder bed 203 are shown. Build 205 is located within the powder bed 203. For comparison, an object model 207 is illustrated with dashed lines. In one embodiment, object model 207 includes data from a data model created in CAD and used as input to an AM processor to render the build. Object model 207 shows the desired shape of the build. In most locations, i.e., in locations without deformation, build 205 overlaps with object model 207. Therefore, in the region to the right of the overhang boundary 210, the solid line characterizing build 205 overlaps with the dashed line defined in object model 207. However, sagging deformation occurs in the overhang region 209. In this example, the overhang region 209 consists of multiple sheets fused to each other on top. In this case, the deformation is more severe as the overhang region 209 extends from the body of build 205.

[0040] It should be noted that although melting does not occur directly on the loose powder, certain problems (such as deformation, high residual stress, etc.) can also occur in areas where the powder in one layer melts near the edge of the sheet in the lower layer. For example, melting powder near the edge of the lower sheet can lead to unexpectedly high temperatures because there is less molten material below and it cannot conduct the heat away. These problems are particularly severe where the lower sheet forms sharp edges.

[0041] Figure 3 An exemplary PBF device 300 including closed-loop control is illustrated. Figure 3 A build plate 301, a powder bed 303, and a build component 305 are shown. An energy application system 309 can apply energy to melt the powder material in the deposited powder layers. For illustration purposes, the powder depositor is not shown in this figure. The energy application system 309 may include an energy applicator 310, which may include an energy beam source 311 and a deflector 313. The energy application system may also include a processor 314 and a computer memory 315 (such as random access memory (RAM), a computer storage disk (e.g., a hard disk drive, a solid-state drive), etc.). The memory 315 may store an object model 316 and print instructions 317. The print instructions 317 may include instructions for each powder layer during the printing process, and these instructions may control how the energy beam source 311 and the deflector 313 scan each powder layer. For example, the print instructions 317 may control printing parameters such as scan rate, beam power, beam melting position, etc. The print instructions 317 may be determined by the processor 314 based on the object model 316. In other words, processor 314 can generate print instructions 317 by determining scan rate, beam power, beam melting position, etc., to form each piece of building block 305 based on object model 316. Energy applicator 310 can receive print instructions 317 from memory 315 and can apply energy beams to melt powder material to create building block 305 based on print instructions.

[0042] The PBF device 300 may include a characterizer 319 that acquires information related to the melting of the powder material. In this example, the characterizer 319 may be a sensor 321 capable of sensing information about the shape of the build piece 305. For example, the sensor 321 may include an optical sensor, such as a camera. The sensor 321 may sense shape information 323 (e.g., dimensional measurements) of the build piece 305 and may send the shape information to a comparator 325. For example, after each piece of the build piece 305 has been melted by the energy application system 309, the sensor 321 may sense the shape of the build piece before the next layer of powder material is deposited and send the sensed shape as shape information 323 to the comparator 325.

[0043] Comparator 325 can retrieve object model 316 from memory 315 and perform a comparison between object model and shape information 323 to determine differences from the object model. For example, some portions of building block 305 may be drooping compared to object model 316. Comparator 325 can send this difference information to compensator 327. Compensator 327 can modify print instruction 317 based on this difference. For example, based on this difference, compensator 327 can determine the region where the next powder layer will be thicker than the remaining layers. Compensator 327 can modify print instruction 317 to increase energy application to the thicker powder region in the next scan to ensure that the powder material in the thicker region is properly melted. For example, compensator 327 can modify print instruction 317 to increase beam power and / or decrease scan rate in the thicker region to apply more energy to these regions.

[0044] In various embodiments, characterizer 319 may include an edge sensor that senses edge information of the molten powder material. For example, melting problems often occur at or near the edges of the sheet. In these cases, the edge sensor can provide useful information about the edge shape of the sheet. In various embodiments, the edge sensor can sense information about the edges of the molten powder material, such as shape, position, height, etc.

[0045] In various embodiments, the characterizer may include a thermal sensor that senses thermal information, such as a thermocouple, an infrared sensor, etc. In various embodiments, the characterizer may include an optical sensor, such as a camera.

[0046] Figure 4 An exemplary PBF device 400 including feedforward control is illustrated. Figure 4 A build plate 401, a powder bed 403, and a build component 405 are shown. An energy application system 409 can apply energy to melt the powder material in the deposited powder layer. For illustration purposes, the powder depositor is not shown in this figure. The energy application system 409 may include an energy applicator 410, which may include an energy beam source 411 and a deflector 413. The energy application system may also include a processor 414 and a computer memory 415 (such as RAM, a computer storage disk, etc.). The memory 415 may store an object model 416 and print instructions 417. The print instructions 417 may include instructions for each powder layer during the printing process, and these instructions may control how the energy beam source 411 and the deflector 413 scan each powder layer. For example, the print instructions 417 may control printing parameters such as scan rate, beam power, and the location of beam melting.

[0047] In this example, print instruction 417 can be determined by processor 414 based on object model 416 and physical model 418. Specifically, processor 414 may include characterizer 419, which can acquire information related to the melting of the powder material. More specifically, characterizer 419 can receive print instruction 417 from memory 415 and can determine the physical model 418 of the component 405 by applying physical modeling to the print instruction. For example, characterizer 419 can execute software stored in memory 415 that can use physical modeling to predict the shape of the component based on print instruction 417. The predicted shape of the component is based on physical model 418, which is stored in memory 415. In this example, Figure 4 The illustration shows the shape of the physics-based model 418, including the sag at the edge region of the component. The sag can be determined by modeling the behavior of the heated powder material based on fluid dynamics, determining the effectiveness of beam heating based on thermodynamics, and determining the forces generated by the deposition of the powder material based on physics-mechanical modeling.

[0048] Therefore, according to the physics-based model 418, if the construct is printed using the print instruction 417 currently stored in memory 415, the construct will have drooping portions. However, the print instruction 417 can be modified before the printing process to eliminate or reduce the drooping. Specifically, the comparator 425 of the processor 414 can receive the object model 416 and the physics-based model 418 from memory 415 and can perform a comparison between the object model and the physics-based model to determine the differences with the object model. In this way, the comparator 425 can determine that certain portions of the physics-based model 418 are drooping compared to the object model 416. The comparator 425 can send the difference information to the compensator 427 of the processor 414. The compensator 427 can modify the print instruction 417 based on this difference. For example, based on this difference, the compensator 427 can determine that less energy should be applied in the regions that will droop according to the physics-based model 418. The compensator 427 can modify the print instruction 417 to reduce the energy applied in these regions in order to prevent or reduce the drooping. For example, compensator 427 can modify print instruction 417 to apply less energy to areas that will sag by reducing beam power in these areas and / or increasing scan rate in these areas. In this way, print instruction 417 can be modified, for example, based on a physics-based model before the print operation.

[0049] In various embodiments, the above process can be iterated multiple times. For example, the modified print instruction 417 can be fed back to the characterizer 419, which can determine the updated physics-based model. The comparator 425 can compare the updated physics-based model with the object model 416 and send the updated difference to the compensator 427, which can then update the modified print instruction. For example, the iteration can continue until the change is less than a threshold tolerance. At this point, the modified print instruction 417 can be used for printing.

[0050] The energy applicator 410 can receive the modified print command 417 from the memory 415 and can apply an energy beam based on the modified print command to melt the powder material to create the building block 405. In this example of feedforward control, the building block 405 has the correct shape because the print command is modified before printing.

[0051] Therefore, in various embodiments utilizing a physics-based model, a set of print instructions can be created prior to the printing process. A characterizer can determine the physics-based model based on the original set of print instructions before the printing process begins. A comparator can compare the physics-based model with the object model to determine differences between the models. A compensator can modify the print instructions to compensate for these differences, such that the actual building blocks are printed according to the object model. Furthermore, in various embodiments, the process of modifying the print instructions can be an iterative process, wherein a first modified set of print instructions can be generated, the physics-based model can be updated based on the first modified set of print instructions, the updated physics-based model can be compared with the object model, and if any difference is greater than a threshold tolerance, a second modified set of print instructions can be determined, and this process can be repeated until no difference is greater than the threshold tolerance.

[0052] Figure 5 An exemplary operation of comparator 500 is illustrated. Comparator 500 can receive object model 501 from memory 502. Comparator 500 can also receive construction information 503 from construction information source 504 (such as memory, sensors, etc.). For example, construction information 503 can be information about the construction obtained by sensors, such as information from... Figure 3 The shape information of sensor 321. For example, construction information 503 can be based on a physical model (such as... Figure 4The information is from the physics-based model 418. Comparator 500 can perform comparison operation 505 to determine the differences between object model 501 and build information 503. In this example, comparison operation 505 determines differences 507 and 509. Difference 507 is the space for missing parts of the build, that is, the space that does not include a part of the build, even though the space should include a part of the build. Difference 509 is the space that includes additional parts of the build, that is, the space that includes a part of the build, even though the space should not include a part of the build. Differences 507 and 509 can be sent to compensator 511 to determine modifications to the print instructions.

[0053] In various embodiments, differences may include size, shape (e.g., deformation), melt integrity, location, etc. In various embodiments, the characterizer may sense whether the powder material in a region of the powder layer has completely melted within a predetermined time after an energy beam is applied to the powder material in that region, and if the powder material is not completely melted after the predetermined time, the compensator may modify the printing instruction to apply additional energy to the powder material in that region. For example, the modified printing instruction may include the additional application of energy, such as returning the energy beam to the incompletely melted region after the sheet has been scanned.

[0054] In various embodiments, for example, the build information may include sensor information about the position of the molten powder material within a sagging layer. In this case, as the next layer of powder is deposited, the powder layer on the sagging region will be thicker than other areas of the powder layer. A compensator can increase the energy applied to the areas of powder material deposited on the sagging region of the previous layer to ensure that the thicker layer of powder will completely melt. In this way, for example, the sagging region can be filled with the molten powder to build up the height to a desired level.

[0055] In various embodiments, for example, the build information may include physically based model information that can predict sagging areas before they occur. In this case, print instructions can be modified to prevent or reduce sagging. For example, a compensator can reduce the energy applied to areas of powder material that would sag if higher energy were applied. In this way, sagging compensation can be performed, for example, before sagging occurs.

[0056] In various embodiments, a physics-based model can characterize (e.g., due to evaporation) the loss of the molten powder material. In various embodiments, a physics-based model can characterize the melt pool viscosity of the molten powder material.

[0057] In various embodiments, the printing instructions can be modified to compensate only for differences that are additional parts of the component, such as difference 509 above. For example, if a portion of the component protrudes upwards into a space that does not include the component, the printing instructions can be modified to melt less powder on the protruding portion when forming the next piece.

[0058] In various embodiments, the print instructions can be modified to compensate only for discrepancies, such as missing portions of the build piece, as shown in discrepancy 507 above. For example, if sagging occurs and real-time compensation is being used, it may be impossible to correct portions of the build piece that have sagged to the point of not including the space below. In this case, the print instructions can be modified to melt more powder in the space above the sagging area when forming the next piece, as illustrated in the example of Figure 6 below. In this way, for example, missing portions of the build piece can be corrected, but the sagging portions below can be retained. The sagging portions can be removed after the printing process by, for example, filing, sanding, etc.

[0059] Figure 6A -C illustrates an exemplary application of energy to a powder layer using the modified printing instructions. (As shown) Figure 6A As shown, the PBF device 600 includes a build plate 601 on which a build member 603 is formed in a powder bed 605. The powder bed 605 includes a powder layer 607 having a desired powder layer thickness 609. A portion of the powder layer 607 has a thicker powder layer thickness 611 on the drooping portion of the build member 603, and is therefore thicker than the desired powder layer thickness 609. The PBF device 600 also includes an energy beam source 613 and a deflector 615. A modified print instruction 617 has been generated to compensate for the increased thickness of the powder layer 607 on the drooping portion of the build member 603. In this example, the modified print instruction 617 modifies the beam power of the energy beam source 613.

[0060] Figure 6B The illustration shows the use of modified beam power to melt powder in a portion of a powder layer 607 having a thicker powder layer thickness 611. Specifically, to melt the portion of the powder layer 607 with the thicker powder layer thickness 611, the modified print instruction 617 commands the energy beam source 613 to increase the beam power to achieve a higher power energy beam 619 when scanning over the thicker portion of the powder layer. In this way, for example, more energy can be applied to the portion of the powder layer 607 with the thicker powder layer thickness 611, allowing the powder to be completely melted.

[0061] Figure 6CThe illustration depicts melting powder in a portion of a powder layer 607 having a desired powder layer thickness 609. In this case, the modified print instruction 617 can command the energy beam source 613 to reduce the beam power to achieve a lower power energy beam 621, which can be the beam power used to completely melt the powder having the desired powder layer thickness 609.

[0062] Figure 7 This is a flowchart illustrating an exemplary method for closed-loop compensation for a PBF system. The PBF system can generate (701) an energy beam and can apply (702) the energy beam to melt powder material to create a three-dimensional (3-D) object with an object model. The PBF system can obtain (703) information related to the melting of the powder material. For example, the information may include sensor information about the shape of the component (e.g., deformation, sagging, etc.), the completeness of melting, etc. The PBF system can determine (704) discrepancies with the object model based on the information. For example, if the information indicates sagging in a specific area, the system can determine the amount of sagging. The PBF system can modify (705) the energy application to the powder material based on this information. For example, the system can increase the beam power of the energy beam based on information about the amount of sagging to completely melt areas of thicker powder. In various embodiments, modifying the energy application may include modifying printing instructions. In various embodiments, modifying the energy application may include modifying beam power, scan rate, etc. in real time based on feedback from one or more sensors. For example, a temperature sensor can sense the temperature at the beam location that is too low for melting powder, and the beam power can be increased based on the sensed temperature. In various embodiments, modifying the application of energy can be accomplished by modifying the printing instructions for the next layer; for example, when sagging is detected in the previous layer, the beam power can be increased to melt the powder in the next layer on the sagging portion of the previous layer.

[0063] Figure 8A -C illustrates another exemplary application of energy to a powder layer using modified printing instructions. (As shown) Figure 8A As shown, the PBF device 800 includes a build plate 801 on which a build member 803 is formed in a powder bed 805. The powder bed 805 includes a powder layer 807. A portion of the powder layer 807 is in a hanging region 809. The PBF device 800 also includes an energy beam source 813 and a deflector 815.

[0064] In this example, (as in the above reference) has already been executed. Figure 4 The feedforward process determines a modified print instruction 817 to compensate for sagging that will occur when the powder layer 807 in the overhang region 809 is melted. In this example, the modified print instruction 817 modifies the beam scan rate of the deflector 815.

[0065] Figure 8B The illustration shows the melting of powder in a portion of powder layer 807 in overhang region 809 using a modified beam scan rate. Specifically, to melt the portion of powder layer 807 in overhang region 809 without causing sagging, the modified print command 817 instructs deflector 815 to increase the beam scan rate to achieve a faster scan energy beam 819 when scanning in overhang region 809. In this way, for example, less energy can be applied to the portion of powder layer 807 in overhang region 809, so that the molten powder does not sag.

[0066] Figure 8C The illustration shows the melting of powder in a portion of a powder layer 807 outside of a hanging region 809. In this case, the modified print command 817 can instruct the deflector 815 to reduce the beam scan rate to achieve a slower scanning energy beam 821, which can be a beam scan rate used to melt powder not in the hanging region.

[0067] Figure 9 This is a flowchart illustrating an exemplary method of feedforward compensation for a PBF system. The PBF system can obtain (901) information related to the melting of the powder material. For example, this information may include a physics-based model predicting the shape of the component (e.g., deformation, sagging, etc.), the completeness of melting, etc. The PBF system can determine (902) discrepancies with the object model based on the information. For example, if the information predicts sagging in a specific area, the system can determine the amount of sagging. The PBF system can modify (903) the energy applied to the powder material based on this information. For example, the system can increase the scan rate of the energy beam based on information about the predicted amount of sagging to prevent sagging. In various embodiments, modifying the energy application may include modifying print instructions. The PBF system can generate (904) an energy beam and can apply (905) the energy beam to melt the powder material to create a three-dimensional (3-D) object with the object model.

[0068] Figure 10A -E illustrates an exemplary PBF device 1000 with post-processing closed-loop control. Figure 10A The PBF apparatus 1000 after a printing run is shown. The PBF apparatus 1000 includes a build plate 1001. A powder bed 1003 and a first completed build piece 1005 are on the build plate 1001. The PBF apparatus also includes an energy application system 1007, which includes an energy beam applicator 1009 having an energy beam source 1011 and a deflector 1013, a memory 1015 including an object model 1017, a print command 1019, a comparator 1021, and a compensator 1023. The PBF apparatus 1000 also includes an object scanner 1025.

[0069] In this example, the first completed construct 1005 is the first construct printed based on object model 1017. For example... Figure 10A As shown, print instruction 1019 retrieves object model 1017 from memory 1015, and the print instruction is based on the object model. However, compared to object model 1017, the first completed construct 1005 has parts with incorrect shapes. Therefore, PBF device 1000 performs a compensation procedure, such as... Figure 10B -E is shown.

[0070] Figure 10B The diagram illustrates the object scanning procedure of the PBF device 1000. Specifically, the object scanner 1025 scans the first completed component 1005 to obtain dimensional information about the shape of the first completed component. This dimensional information is sent as scan information 1027 to the comparator 1021. Furthermore, the comparator 1021 receives the object model 1017 from the memory 1015. The comparator 1021 then executes... Figure 10C The comparison operation shown is used to determine the differences between object model 1017 and scan information 1027.

[0071] Figure 10C The operation of comparator 1021 is illustrated. Comparator 1021 receives object model 1017 from memory 1015 and scan information from object scanner 1025. Comparator 1021 performs a comparison operation 1029 to determine the difference 1031 between object model 1017 and scan information 1027 and sends the difference to compensator 1023.

[0072] Figure 10D The operation of compensator 1023 is illustrated. Compensator 1023 receives object model 1017 from memory 1015 and receives difference 1031 from comparator 1021. Compensator 1023 performs compensation operation 1033 to determine compensated object model 1035. The print instruction generated from compensated object model 1035 will cause the printing of the construct that matches object model 1017. In other words, compensated object model 1035 compensates for the error that occurred when the first completed construct 1005 was printed. Compensator 1023 sends compensated object model 1035 to be stored in memory 1015.

[0073] Figure 10EThe illustration shows a second completed component 1037 obtained by printing using the compensated object model 1035. When printing the second completed component 1037, the print instruction 1019 is based on the compensated object model 1035. In this way, for example, the shape of the second completed component 1037 can match the shape of the object model 1017. In fact, once the compensated object model 1035 has been determined, each subsequent component can be matched to the shape of the object model 1017.

[0074] Figure 11 This is a flowchart illustrating another exemplary method for compensation in a PBF system. The PBF system can provide (1101) printing instructions for printing 3-D objects, and can print (1102) 3-D objects based on these printing instructions. For example, the system can print a first construct, such as... Figure 10A The first completed construct 1005 is generated. The PBF system can sense (1103) the shape of at least a portion of the printed 3-D object. For example, the first construct can be scanned by an object scanner (such as object scanner 1025). The PBF system can compare (1104) the shape of the printed 3-D object with a reference shape (such as object model 1017) to determine difference parameters, such as dimensional differences in the shape. The PBF system can update (1105) the printing instructions based on these difference parameters. For example, the printing instructions can be updated based on a compensated object model (such as compensated object model 1035) that can be determined by the difference parameters.

[0075] The preceding description is provided to enable any person skilled in the art to practice the various aspects described herein. Various modifications to these exemplary embodiments set forth throughout this disclosure will readily be apparent to those skilled in the art. Therefore, the claims are not intended to limit themselves to the exemplary embodiments set forth throughout this disclosure, but are to be accorded the full scope consistent with the language of the claims. All structural and functional equivalents of the elements of the exemplary embodiments described throughout this disclosure are intended to be included by the claims, which are known to or subsequently known to a person skilled in the art. Furthermore, nothing disclosed herein is intended to be exclusive to the public, whether or not such disclosure is expressly stated in the claims. Claim elements shall not be construed under 35 U.S.SC §112(f) or similar laws in applicable jurisdiction, unless the element is expressly stated using the phrase “method for…” or, in the case of a method claim, using the phrase “step for…”.

Claims

1. An apparatus for additive manufacturing, comprising: A depositor configured to deposit powder materials; An energy applicator configured to apply energy, the energy applicator including an energy beam source configured to generate an energy beam, wherein the energy applicator applies the energy beam to melt powder material to create molten powder of a building component; as well as A compensator is configured to modify the energy application to the powder material according to a physics-based model, wherein the physics-based model is configured to predict the properties of the printed component based on physical modeling, and the compensator modifies the energy application based on the predicted properties to achieve the desired properties of the printed component.

2. The apparatus of claim 1, wherein the desired characteristics include at least the desired shape of the component, desired density, desired internal stress, or desired completeness of melting.

3. The apparatus according to claim 1, further comprising: A comparator is configured to compare a physics-based model with an object model of the building block to obtain a comparison result, wherein the compensator modifies the energy application based on the comparison result to achieve the desired properties of the printed building block.

4. The apparatus of claim 3, wherein the desired characteristic includes the desired shape of the component, the comparison result includes the shape difference with the object model, and the compensator modifies the energy application based on the shape difference.

5. The apparatus of claim 4, wherein the shape difference includes shrinkage.

6. The apparatus of claim 1, wherein the compensator modifies the energy application such that the energy application changes during the printing of the component.

7. The apparatus of claim 1, wherein the compensator modifies energy application by modifying printing instructions, wherein the printing instructions are instructions for printing a 3D printer to construct a part.

8. The apparatus of claim 1, wherein the physics-based model characterizes the sagging of the molten powder material, and the compensator is configured to compensate for the sagging.

9. The apparatus of claim 1, wherein the physics-based model characterizes the loss of the molten powder material, and the compensator is configured to compensate for the loss of the molten material.

10. The apparatus of claim 9, wherein the loss of the molten powder material is caused by evaporation.

11. The apparatus of claim 1, wherein the physics-based model characterizes the melt pool viscosity of the molten powder material, and the compensator is configured to compensate for the melt pool viscosity.

12. A method of additive manufacturing, comprising: Deposited powder materials; Applying energy, wherein applying energy includes generating an energy beam and applying the energy beam to melt the powder material to create a building block; as well as The energy application to the powder material is modified according to a physics-based model, wherein the physics-based model is configured to predict the properties of the printed component based on physical modeling, and the energy application is modified based on the predicted properties to achieve the desired properties of the printed component.

13. The method of claim 12, wherein the desired characteristics include at least the desired shape of the component, desired density, desired internal stress, or desired completeness of melting.

14. The method of claim 12, further comprising: A physics-based model is compared with an object model of the building block to obtain a comparison result, wherein the energy application is modified based on the comparison result to achieve the desired properties of the printed building block.

15. The method of claim 14, wherein the desired characteristic includes the desired shape of the component, the comparison result includes the shape difference with the object model, and the energy application is modified based on the shape difference.

16. The method of claim 15, wherein the shape difference includes shrinkage.

17. The method of claim 12, wherein the energy application is modified such that the energy application changes during the printing of the component.

18. The method of claim 12, wherein modifying the energy application includes modifying the printing instructions, wherein the printing instructions are instructions for printing the building blocks by the 3D printer.

19. The method of claim 12, wherein the physics-based model characterizes the sagging of the molten powder material, and the energy application is modified to compensate for the sagging.

20. The method of claim 12, wherein the physics-based model characterizes the loss of the molten powder material, and the energy application is modified to compensate for the loss of the molten material.

21. The method of claim 20, wherein the loss of the molten powder material is caused by evaporation.

22. The method of claim 12, wherein the physics-based model characterizes the melt pool viscosity of the molten powder material, and the energy application is modified to compensate for the melt pool viscosity.

Images