material solutions for additive manufacturing
MATERIAL SOLUTIONS FOR ADDITIVE MANUFACTURING Capra J’neva & Bill Yackabonis May 2017
INTRODUCTION
Image courtesy of Briggs Automotive Company, Ltd.
The additive manufacturing space promises unprecedented freedom in what can be manufactured—offering greater complexity, more design freedom, and fewer constraints around economies of scale and supply chain. To fully capitalize on these opportunities, engineers need to be able to better predict the product’s manufacture and optimize end part performance.
Current gaps include: • Effects of part orientation • Part performance including mechanical, physical or flammability • Part manufacturability and design constraints • Part costs, especially controlling for the entire manufacturing process and cost of machines • End-part aesthetics and rendering • Comparison methods, especially across manufacturing processes • Clear description of the process and post-processing requirements
Material Solutions for Additive Manufacturing
2
Introduction Lacking this information, designers and engineers can only guess how to design for these processes and face uncertainty whether they will be able to reliably manufacture their end products. With more fundamental data as to how additive processes perform, the amount of iteration required to reach a positive outcome can potentially be reduced, thus lessening the time and cost incurred for each job. Other advanced digital manufacturing processes, such as composites, face similar hurdles. These manufacturing processes involve complex interactions between hundreds of machine and process variables, a material’s inherent properties, effects of thermal processing on bulk materials, and the transformation of models into toolpaths by software preparing the machine instructions. Because of the number of variables, there is no simple way to predict the outcomes of changing any single parameter or combination of parameters. Additive manufacturing, in particular, has a long history of being used only for prototyping, and thus has been inadequately optimized and studied for end-part performance.
Material Solutions for Additive Manufacturing
To help capture and control for the intricacies of advanced digital manufacturing processes like additive, we characterize end-part properties of materials with their full process description and call these material solutions. Material solutions can give designers the ability to compare the cost, aesthetics, performance and manufacturability, and to run simulations to improve their designs before investing time and money creating products. Material solutions should be viewed as a proxy to help advance industrial applications of additive manufacturing while a better physics process model is being understood and created. Once a material solution is created and tested for end part properties, it can easily be exported from one factory to another anywhere in the world, providing the “recipe” for simulating and creating additive parts.
3
SUPPORTED CHARACTERIZATION
A team of materials scientists and process experts at Autodesk spent a year studying which factors in additive manufacturing contributed to end part property variability. These included inconsistent toolpathing and support building in software, changes to process parameters, machine calibration, hardware/firmware variations, and inadequate documentation of proper postprocess procedures. Software that uses a deterministic model to create supports and toolpaths can improve repeatability in additive manufacturing. Changes to either the support structure or toolpath are the equivalent of changing a gate position or cooling channel in traditional techniques such as injection molding and casting. Therefore, software such as Autodesk® Netfabb®, which controls for these important factors, can improve reliability and repeatability in additive.
Material Solutions for Additive Manufacturing
To achieve a truly accurate characterization of as-manufactured additive parts, the process description for a material solution must specify: • The machine along with its hardware, firmware, and tooling • A material • Deterministic software that will yield the same support structures and toolpath each time the same model is processed • And a locked set of reliable process parameters tuned to achieve a particular result Any heat treatment, finishing, subtractive, or post-processing steps should also be well defined in the solution and can be captured separately to understand their effects on tolerance, appearance, and surface quality of the part.
4
Supported Characterization Once defined, material solutions are characterized for mechanical performance and quality to enable their comparison and specification by designers and engineers hoping to use additive manufacturing for production. Basic solutions contain the information needed to make cost comparisons, understand how the part will look through sophisticated shaders, and consider manufacturability of small features and overhangs. Solution providers may also include data on the performance of the solution to enable a variety of mechanical and other physical simulations, generative design according to manufacturability rules, and process simulation to predict the effects of the process on part build and warping. It is important to note that while material solutions can approximate end part properties, they may not account for all the effects of an individual additive build, such as heat cycling from beds packed with multiple parts or variations in toolpathing affected by part geometry or infill. Full process analysis run on a completed build tray with a tool such as Autodesk® Simulation Utility for Netfabb can help to predict all the effects of an actual additive build.
Comparison data
To enable comparison, we run each material solution through a cost model that allows increasing refinement of cost estimation as more is known about the specific manufacture of the part. For a
Material Solutions for Additive Manufacturing
first-glance cost estimation, we base the costs shown in search results on a cubic centimeter part built with the solution. The cost model accounts for material, supports, machine and machine maintenance, and waste factors of each solution. Labor and overhead can be added in integrated factory environments with ERP integration.
Figure 1: Shader thumbnail.
The Autodesk rendering team has created specialized shaders to allow more realistic rendering of solutions for many types of printing. These shaders show the approximate effects of layering, resolution, and material on parts. Shaders can be used to help communicate to clients the look and feel of end parts as well as the effects of resolution, but won’t accurately reflect the effect of layers on end parts unless the parts are oriented in the direction they will be printed when rendered. The renderers do not capture exact toolpath effects at this time, but rather approximate the effects of toolpaths.
5
Supported Characterization As a final tool to help in comparison and determining if a material solution is appropriate to a particular project, we conduct three manufacturability tests: overhangs, and both small positive and negative feature tests. Each test is printed in 3 locations on the print bed to capture the dynamics of the printer’s bed and cooling strategy. Overhangs face in all four ordinal directions of the print bed to capture the effects of the cooling strategy on overhang success, and the worst results are reported. Test results for overhangs include a design angle—the angle at which the overhang quality looks the same as the top of the part—and a critical support angle—the angle at which the supported part has fewer flaws than the unsupported part.
Figure 2: Manufacturability test sample part.
Material Solutions for Additive Manufacturing
Small feature test results include the smallest feature size that reliably prints and measurements of the accuracy of printed small features, measured by micrometer. Small features include columns and holes printed in three different heights for columns and three wall thicknesses for holes. These test results are self-reported by the solution provider. The data from these tests can allow designers to determine if a material solution is appropriate for their project and to design for overhangs, shrinkage, and overprinting. They can help during design to dimension features and overhangs appropriately for the manufacturing technique and to reduce post-processing requirements. These factors may also be used to place and orient prints for best success or to improve the types of features possible on a print. The manufacturability tests are not meant to be exhaustive, but to lower the barrier to entry for providing information in the dataservice, while creating a minimum standard. We encourage contributors to provide data for any additional testing of physical properties or manufacturability characteristics.
6
Supported Characterization Simulation data
We chose to characterize each material solution with the most commonly required mechanical tests to enable basic structural simulation and generative design, as well as to provide information likely needed by design engineers that are not used in simulations. Physical and mechanical properties can also improve comparison of the suitability of a material solution to a particular application. These properties include tensile strength, impact strength, compressive strength, and flexural strength testing to enable linear static, linear buckling, non-linear static, non-linear transient response, and modal response analyses using the Autodesk® Nastran® solver. The tests are conducted in the yx, yz, and zy directions to capture anisotropic z axis layer bonding effects on mechanical properties. For each test and test direction, the solution developer prints a bed of 12 samples which are laid out across the entire bed surface and labeled for their position and direction. Eight of these samples are tested and their print bed locations and orientations captured.
Material Solutions for Additive Manufacturing
The tests are conducted to ASTM standards and all tests are conducted to failure. For each, except impact testing, tabular data is collected and stored. This data can be entered into FEA solvers to provide approximate performance for parts printed using the material solution specification with the Netfabb profile from the material solutions database, but should not be substituted for actual testing of final parts as part geometry and build tray set up will impact results. Some of this data is bundled into the Autodesk® Optimization Utility for Netfabb and can drive generative processes to enable topological optimization and latticing.
7
Supported Characterization Contributors to the database can also provide other tested properties appropriate for their material solution, and we encourage them to note the direction the tested samples were printed based on this diagram:
Figure 3: Conventions for denoting test sample orientation.
Material Solutions for Additive Manufacturing
8
Supported Characterization Manufacturing data
Each material solutions’ characterization is accompanied by a profile and process instructions to enable factories to create the solution the same way each time on the specified software and machines. Changing any process parameters, materials, machines, or software will invalidate the mechanical and physical characterizations of the material solution, so care should be taken to ensure that machines are properly calibrated and that each part of the material solution definition is followed.
Material Solutions for Additive Manufacturing
9
CONCLUSION Our hope is that material solutions will help to drive adoption of additive manufacturing by providing the engineering data that allows designers and engineers to understand how the additive manufacturing process will affect their designs. We also hope to spur innovation within additive manufacturing by providing access to parameters development, testing, and information that encourages major chemical suppliers and machine manufacturers to contribute. Driving competition to provide innovative and useful end-part properties will make additive manufacturing a more viable solution, allowing the industry to realize its potential and connect the designers who use our products with the information they need to succeed in the new design possibilities additive opens for the future of the manufacturing industry.
Material Solutions for Additive Manufacturing
10
Autodesk, the Autodesk logo, and Autodesk Netfabb, are registered trademarks or trademarks of Autodesk, Inc., and/or its subsidiaries and/or affiliates in the USA and/or other countries. Nastran is a registered trademark of the National Aeronautics and Space Administration. All other brand names, product names, or trademarks belong to their respective holders. Autodesk reserves the right to alter product offerings and specifications at any time without notice, and is not responsible for typographical or graphical errors that may appear in this document. © 2017 Autodesk, Inc. All rights reserved.
Material Solutions for Additive Manufacturing
INTRODUCTION
Image courtesy of Briggs Automotive Company, Ltd.
The additive manufacturing space promises unprecedented freedom in what can be manufactured—offering greater complexity, more design freedom, and fewer constraints around economies of scale and supply chain. To fully capitalize on these opportunities, engineers need to be able to better predict the product’s manufacture and optimize end part performance.
Current gaps include: • Effects of part orientation • Part performance including mechanical, physical or flammability • Part manufacturability and design constraints • Part costs, especially controlling for the entire manufacturing process and cost of machines • End-part aesthetics and rendering • Comparison methods, especially across manufacturing processes • Clear description of the process and post-processing requirements
Material Solutions for Additive Manufacturing
2
Introduction Lacking this information, designers and engineers can only guess how to design for these processes and face uncertainty whether they will be able to reliably manufacture their end products. With more fundamental data as to how additive processes perform, the amount of iteration required to reach a positive outcome can potentially be reduced, thus lessening the time and cost incurred for each job. Other advanced digital manufacturing processes, such as composites, face similar hurdles. These manufacturing processes involve complex interactions between hundreds of machine and process variables, a material’s inherent properties, effects of thermal processing on bulk materials, and the transformation of models into toolpaths by software preparing the machine instructions. Because of the number of variables, there is no simple way to predict the outcomes of changing any single parameter or combination of parameters. Additive manufacturing, in particular, has a long history of being used only for prototyping, and thus has been inadequately optimized and studied for end-part performance.
Material Solutions for Additive Manufacturing
To help capture and control for the intricacies of advanced digital manufacturing processes like additive, we characterize end-part properties of materials with their full process description and call these material solutions. Material solutions can give designers the ability to compare the cost, aesthetics, performance and manufacturability, and to run simulations to improve their designs before investing time and money creating products. Material solutions should be viewed as a proxy to help advance industrial applications of additive manufacturing while a better physics process model is being understood and created. Once a material solution is created and tested for end part properties, it can easily be exported from one factory to another anywhere in the world, providing the “recipe” for simulating and creating additive parts.
3
SUPPORTED CHARACTERIZATION
A team of materials scientists and process experts at Autodesk spent a year studying which factors in additive manufacturing contributed to end part property variability. These included inconsistent toolpathing and support building in software, changes to process parameters, machine calibration, hardware/firmware variations, and inadequate documentation of proper postprocess procedures. Software that uses a deterministic model to create supports and toolpaths can improve repeatability in additive manufacturing. Changes to either the support structure or toolpath are the equivalent of changing a gate position or cooling channel in traditional techniques such as injection molding and casting. Therefore, software such as Autodesk® Netfabb®, which controls for these important factors, can improve reliability and repeatability in additive.
Material Solutions for Additive Manufacturing
To achieve a truly accurate characterization of as-manufactured additive parts, the process description for a material solution must specify: • The machine along with its hardware, firmware, and tooling • A material • Deterministic software that will yield the same support structures and toolpath each time the same model is processed • And a locked set of reliable process parameters tuned to achieve a particular result Any heat treatment, finishing, subtractive, or post-processing steps should also be well defined in the solution and can be captured separately to understand their effects on tolerance, appearance, and surface quality of the part.
4
Supported Characterization Once defined, material solutions are characterized for mechanical performance and quality to enable their comparison and specification by designers and engineers hoping to use additive manufacturing for production. Basic solutions contain the information needed to make cost comparisons, understand how the part will look through sophisticated shaders, and consider manufacturability of small features and overhangs. Solution providers may also include data on the performance of the solution to enable a variety of mechanical and other physical simulations, generative design according to manufacturability rules, and process simulation to predict the effects of the process on part build and warping. It is important to note that while material solutions can approximate end part properties, they may not account for all the effects of an individual additive build, such as heat cycling from beds packed with multiple parts or variations in toolpathing affected by part geometry or infill. Full process analysis run on a completed build tray with a tool such as Autodesk® Simulation Utility for Netfabb can help to predict all the effects of an actual additive build.
Comparison data
To enable comparison, we run each material solution through a cost model that allows increasing refinement of cost estimation as more is known about the specific manufacture of the part. For a
Material Solutions for Additive Manufacturing
first-glance cost estimation, we base the costs shown in search results on a cubic centimeter part built with the solution. The cost model accounts for material, supports, machine and machine maintenance, and waste factors of each solution. Labor and overhead can be added in integrated factory environments with ERP integration.
Figure 1: Shader thumbnail.
The Autodesk rendering team has created specialized shaders to allow more realistic rendering of solutions for many types of printing. These shaders show the approximate effects of layering, resolution, and material on parts. Shaders can be used to help communicate to clients the look and feel of end parts as well as the effects of resolution, but won’t accurately reflect the effect of layers on end parts unless the parts are oriented in the direction they will be printed when rendered. The renderers do not capture exact toolpath effects at this time, but rather approximate the effects of toolpaths.
5
Supported Characterization As a final tool to help in comparison and determining if a material solution is appropriate to a particular project, we conduct three manufacturability tests: overhangs, and both small positive and negative feature tests. Each test is printed in 3 locations on the print bed to capture the dynamics of the printer’s bed and cooling strategy. Overhangs face in all four ordinal directions of the print bed to capture the effects of the cooling strategy on overhang success, and the worst results are reported. Test results for overhangs include a design angle—the angle at which the overhang quality looks the same as the top of the part—and a critical support angle—the angle at which the supported part has fewer flaws than the unsupported part.
Figure 2: Manufacturability test sample part.
Material Solutions for Additive Manufacturing
Small feature test results include the smallest feature size that reliably prints and measurements of the accuracy of printed small features, measured by micrometer. Small features include columns and holes printed in three different heights for columns and three wall thicknesses for holes. These test results are self-reported by the solution provider. The data from these tests can allow designers to determine if a material solution is appropriate for their project and to design for overhangs, shrinkage, and overprinting. They can help during design to dimension features and overhangs appropriately for the manufacturing technique and to reduce post-processing requirements. These factors may also be used to place and orient prints for best success or to improve the types of features possible on a print. The manufacturability tests are not meant to be exhaustive, but to lower the barrier to entry for providing information in the dataservice, while creating a minimum standard. We encourage contributors to provide data for any additional testing of physical properties or manufacturability characteristics.
6
Supported Characterization Simulation data
We chose to characterize each material solution with the most commonly required mechanical tests to enable basic structural simulation and generative design, as well as to provide information likely needed by design engineers that are not used in simulations. Physical and mechanical properties can also improve comparison of the suitability of a material solution to a particular application. These properties include tensile strength, impact strength, compressive strength, and flexural strength testing to enable linear static, linear buckling, non-linear static, non-linear transient response, and modal response analyses using the Autodesk® Nastran® solver. The tests are conducted in the yx, yz, and zy directions to capture anisotropic z axis layer bonding effects on mechanical properties. For each test and test direction, the solution developer prints a bed of 12 samples which are laid out across the entire bed surface and labeled for their position and direction. Eight of these samples are tested and their print bed locations and orientations captured.
Material Solutions for Additive Manufacturing
The tests are conducted to ASTM standards and all tests are conducted to failure. For each, except impact testing, tabular data is collected and stored. This data can be entered into FEA solvers to provide approximate performance for parts printed using the material solution specification with the Netfabb profile from the material solutions database, but should not be substituted for actual testing of final parts as part geometry and build tray set up will impact results. Some of this data is bundled into the Autodesk® Optimization Utility for Netfabb and can drive generative processes to enable topological optimization and latticing.
7
Supported Characterization Contributors to the database can also provide other tested properties appropriate for their material solution, and we encourage them to note the direction the tested samples were printed based on this diagram:
Figure 3: Conventions for denoting test sample orientation.
Material Solutions for Additive Manufacturing
8
Supported Characterization Manufacturing data
Each material solutions’ characterization is accompanied by a profile and process instructions to enable factories to create the solution the same way each time on the specified software and machines. Changing any process parameters, materials, machines, or software will invalidate the mechanical and physical characterizations of the material solution, so care should be taken to ensure that machines are properly calibrated and that each part of the material solution definition is followed.
Material Solutions for Additive Manufacturing
9
CONCLUSION Our hope is that material solutions will help to drive adoption of additive manufacturing by providing the engineering data that allows designers and engineers to understand how the additive manufacturing process will affect their designs. We also hope to spur innovation within additive manufacturing by providing access to parameters development, testing, and information that encourages major chemical suppliers and machine manufacturers to contribute. Driving competition to provide innovative and useful end-part properties will make additive manufacturing a more viable solution, allowing the industry to realize its potential and connect the designers who use our products with the information they need to succeed in the new design possibilities additive opens for the future of the manufacturing industry.
Material Solutions for Additive Manufacturing
10
Autodesk, the Autodesk logo, and Autodesk Netfabb, are registered trademarks or trademarks of Autodesk, Inc., and/or its subsidiaries and/or affiliates in the USA and/or other countries. Nastran is a registered trademark of the National Aeronautics and Space Administration. All other brand names, product names, or trademarks belong to their respective holders. Autodesk reserves the right to alter product offerings and specifications at any time without notice, and is not responsible for typographical or graphical errors that may appear in this document. © 2017 Autodesk, Inc. All rights reserved.
Material Solutions for Additive Manufacturing
