Browsing by Subject "design for additive manufacturing"
Now showing 1 - 15 of 15
- Results Per Page
- Sort Options
Item Cellular and Topology Optimization of Beams under Bending: An Experimental Study(University of Texas at Austin, 2019) Gopal, Arjun; Parihar, Gaurav; Holt, McKay; Stinson, Tanner; Sharma, Manasvi; Bhate, DhruvDesign for Additive Manufacturing (AM) includes concepts such as cellular materials and topology optimization that combine the capabilities of advanced computational design with those of AM technologies that can realize them. There is however, limited experimental study of the relative benefits of these different approaches to design. This paper examines these two different approaches, specifically in the context of maximizing the flexural rigidity of a beam under bending, while minimizing its mass. A total of 23 beams were designed using commercially available cellular design, and topology optimization software. The Selective Laser Sintering (SLS) process was used to manufacture these beams with Nylon 12, which were then tested per ASTM D790 three-point bend test standards. The effect of varying the size and shape of cells on the flexural rigidity was studied using 15 different cellular designs. These results were then compared to six different topology optimized beam designs, as well as three solid and hollow baseline beams. These preliminary findings suggest that topology optimized shapes underperform their cellular counterparts with regard to specific stiffness, and that stochastic cellular shapes deserve deeper study.Item Computer-Aided Process Planning for Wire Arc Directed Energy Deposition(University of Texas at Austin, 2019) Xiong, Yi; Dharmawan, Audelia Gumarus; Tang, YunlongWire arc directed energy deposition provides a rapid and cost-effective method for fabricating low-to-medium complexity and medium-to-large size metal parts. However, the complex nonequilibrium phase transformations, inherent to this process, make it a challenging task to produce consistent and high-quality parts, especially for parts with materials or geometries that have not been manufactured before. This study outlines a holistic and data-centric computer-aided process planning framework utilizing a knowledge base to assist engineers in selecting optimal process parameters that reduce dimensional deviations, and therefore to obtain near-net-shape parts using directed energy deposition only. The knowledge base has a data-knowledge-service architecture and is proposed to synthesize information from various sources, e.g., characterization tests. Based on these collected data, several knowledge representations, including database, metamodels, and planning rules, are constructed to support decision-making in the process planning. The proposed framework is demonstrated in the fabrication of components from industrial applications.Item Conceptual Design for Additive Manufacturing: Lessons Learned from an Undergraduate Course(University of Texas at Austin, 2019) Thompson, Scott M.Design for additive manufacturing (DfAM) guidelines continue to emerge and evolve as various additive manufacturing (AM) technologies, and the knowledge of their associated end-users/designers, matures. This work summarize important pedagogical and technical lessons learned from the conceptual re-design of several, diverse parts/assemblies submitted by ~50 undergraduate students cognizant of recent DfAM strategies and guidelines. All students were enrolled in a traditional, semester-long metals AM course designed by the author herein. Students were instructed to select an existing, metallic product and provide a conceptual redesign of that product for subsequent, effective laser-powder bed fusion (L-PBF). Students were instructed that the redesigned concept should have enhanced functionality/specifications and consist of features (e.g. thin walls, bore diameters, etc.) that can be fabricated with minimal risk via current L-PBF systems. The presented results include the types of parts that attract an ‘AM redesign’ effort and the most popular AM-enabled detailed design decisions made. To encourage a more detail-inspired design, DfAM topics were presented ‘backwards’; from post-manufacturing considerations to conceptual design while considering emerging design rules and heuristics. Results indicate that students can become preoccupied with DfAM rules to a point where the design failure modes are not properly accounted for.Item Design and Manufacture of a Continuous Fiber-Reinforced 3D Printed Unmanned Aerial Vehicle Win(University of Texas at Austin, 2021) Jayashankar, Dhileep Kumar; Devarajan, Aarthi; Dong, Guoying; Rosen, DavidThe Markforged Mark Two 3D printer is capable of printing various orientations of continuous fiber reinforcement. An initial study of how the orientation of the fiber influences the strength characteristics (tensile and flexural properties) was conducted. Four combinations of carbon fiber reinforcement orientations were tested, specifically unidirectional, isotropic, concentric and a combination of isotropic and concentric, with the Markforged Onyx matrix material. The results will aid in designing a wing with the optimum fiber configuration that will give the desired mechanical properties based on the forces acting on the wing. Design for Additive Manufacturing (DfAM) concepts and tools will be used to design and manufacture a large UAV wing. Topology optimization, based on a CFD computed pressure distribution, was used to determine geometric regions where carbon fiber reinforcement could be best utilized. From there, a honeycomb structure was designed to ensure stiffness and light weight based on desired densities. A wing section was fabricated using the Mark Two printer to identify the capabilities and limitations of the system in realizing the design objectives.Item Design for Additive Manufacturing: Simplification of Product Architecture by Part Consolidation for the Lifecycle(University of Texas at Austin, 2019) Kim, Samyeon; Tang, Yunlong; Rosen, David W.Additive manufacturing (AM) can support the fabrication of the complex design and generate new design opportunities for improving products. To identify and leverage these opportunities, design studies in early product design stages are required. Since part consolidation is one of AM design potentials in conceptual and embodiment design stages, this study proposes a design method to reconceptualize existing product design in the context of part consolidation. Function requirements and physical relations between existing parts are used to investigate AM design potential and identify candidates for consolidation. After identification of consolidation candidates, function sharing between parts and modules is checked because they have high possibilities to be consolidated if they share the same functions. Furthermore, AM design potential is identified to help designers add value in part design. In order to support designers, it is required to link AM design potential to the part candidates in order to explore AM design benefits. A case study with motorcycles is performed to demonstrate the proposed method. The AM design potential for the case study contains the lifecycle considerations related to fuel savings due to lightweight, and simplified and less expensive assembly operations due to simplified product architecture by part consolidation.Item Design for the Additive Manufacturing Process Chain(University of Texas at Austin, 2021) Rosen, David W.Post-processing operations are required for most additive manufacturing (AM) processes. For production parts, consideration of these post-processing operations during design is critical to achieve design requirements. For both metal and polymer parts, the sequence of steps in the process chain can be extensive. A design framework called the Process Chain Map (PCM) is introduced in this paper that explicitly relates design requirements for the part to each step in the AM process chain. This PCM visually shows the role of each step in the process chain and facilitates communication among design and manufacturing personnel. Software implementation of the PCM enables generation of system-level problem formulations of multidisciplinary design optimization problems. An example of a metal AM part demonstrates the PCM and the formulation of such a design problem.Item DESIGN FOR(E!) ADDITIVE MANUFACTURING: IN SEARCH OF A COMPREHENSIVE DESIGN CHALLENGE SUITABLE ACROSS AM EDUCATION(University of Texas at Austin, 2023) Meisel, NicholasModern engineering design education relies heavily on the concept of problem-based learning (PBL). Driven by the constructivist theory of education, PBL enables students to build knowledge organically, rather than through rote memorization. As such, design for additive manufacturing (DfAM) education also tends to emphasize the use of PBL to encourage student learning. Unfortunately, dedicated DfAM education is still nascent. The result is a wide range of educators leveraging an equally wide, and often unproven, range of design challenges to support DfAM PBL. Because of this, there is the possibility that a chosen design challenge will not represent AM as a true end-use manufacturing process nor promote a design space that can benefit from the full consideration of all opportunistic and restrictive DfAM concepts. In this paper, the author discusses the creation and implementation of a comprehensive design challenge that is suitable across the range of AM education. Specifically, the author proposes the use of a golf putter DfAM design challenge. This concept draws from lessons learned over years of DfAM instruction at undergraduate and graduate levels and is based in the need for three key aspects for a successful DfAM challenge in education: (1) clarity, (2) applicability, and (3) demonstrability.Item A Designer's Guide for Dimensioning and Tolerancing SLS Parts(University of Texas at Austin, 2012) Seepersad, Carolyn Connor; Govett, Tyler; Kim, Kevin; Lundin, Michael; Pinero, DanielBecause additive manufacturing (AM) is a relatively novel industry, with the first commercial machines introduced in the late 1980s, many designers are unaware of the capabilities of AM technologies. Many engineers also find it difficult to utilize AM because of a lack of “Design for AM” knowledge in the public domain. Reliable information on material properties, dimensions and tolerances, and other process-related specifications is often scattered throughout the literature, if it is publicly available at all. The objective of the research reported in this paper is to begin to create a designer's guide for dimensioning and tolerancing parts that are additively manufacturing using selective laser sintering (SLS) technology. The guide is based on a series of experiments designed to determine the limiting feature sizes for various types of features fabricated in commercially available SLS machines. The features include slits, holes, letters, mating gears, and shafts built in a preassembled state. The impact of part thickness, orientation, clearance, and dimensions on the resolvability of features is examined. Results are reported in a series of matrices that relate realizable feature sizes to other important variables such as part thickness.Item An Experimental Study of Design Strategies for Stiffening Thin Plates under Compression(University of Texas at Austin, 2019) Ramirez Chavez, Irving E.; Noe, Cameron; Sekar, Vigneshwaran; Jogani, Shainil; Israni, Siddharth; Bhate, DhruvIncreasing stiffness and failure loads while minimizing mass is useful in many engineering applications, including the design of thin plates and shells. In this paper, the performance of thin plates using a range of stiffening approaches were studied for the specific instance of compressive loading. Periodic, graded, stepped, “Voronoi” stochastic, and topologically optimized patterns were explored. These stiffening designs were realized using different software tools and manufactured with the Selective Laser Sintering (SLS) process. These 3D printed specimens were tested under compression to assess their mechanical response. Videos of these tests were recorded to study the shape of the failure modes. This data was analyzed to determine the performance of the different stiffener designs, in comparison to the performance of baseline plates without any stiffening. The study concludes with a discussion of the results and their implications for stiffening thin plates, showing that triangular and stochastic stiffening strategies show particular promise in increasing specific compressive stiffness and specific buckling load.Item GUIDED MANUAL DESIGN FOR ADDITIVE MANUFACTURING OF TOPOLOGICALLY OPTIMIZED LEGACY TOOLING PARTS(University of Texas at Austin, 2023) Luben, Hannah; Meisel, NicholasDesign for Additive Manufacturing (DfAM) is a unique conceptual way to adapt a part for Additive Manufacturing (AM). While some of the choices made in DfAM become second nature to seasoned AM designers, inexperienced designers may not know the nuances involved in what is still a developing manufacturing technology. Topology Optimization (TO) in particular tends to create organic shapes that may not be immediately conducive to printing through AM. This paper proposes a comprehensive workflow tool to guide a designer, regardless of their experience, through the decision-making process inherent to DfAM. The guide helps the designer manually edit a legacy tooling design into a topologically optimized part that is readily manufacturable through AM. Discussion of a relevant case study follows the outline of the design tool to exemplify its use.Item Lattice Design Optimization: Crowdsourcing Ideas in the Classroom(University of Texas at Austin, 2018) Bhate, DhruvCrowdsourcing is a powerful method of generating ideas, particularly when there are many possible solutions to a particular problem with no obvious process towards arriving at the optimum one. In this paper, results of a crowdsourcing exercise conducted in a 30-student classroom are reported. Students were tasked with using lattice design concepts to minimize the weight of a beam under bending, tension and torsion. Using the nTopology software, they approached the problem in three steps: (1) Selection/design of a unit cell, (2) Distribution of cell size, and (3) Optimization of the thickness of individual members. The first two steps were design decisions made by the students, the last step used nTopology's native solver. This work shares insights gained both in lattice design itself, as well as on the use of crowdsourcing in the classroom, particularly in the context of the rapidly evolving field of Design for Additive Manufacturing.Item Optimization of a Worm Gear Assembly Design for Additive Manufacturing(University of Texas at Austin, 2021) Borstell, D.; Georg, M.-C.Worm gears are widely used to transmit power at predominantly low speeds and highspeed ratios. Their self-locking characteristic makes them unique to many drive applications. Industrial power requirements are causing forces and tensions mostly prohibiting thermoplastic materials in worm gear drive trains. Double basses are tuned using a worm gear assembly made from machined steel and brass or cast bronze. Neglectable power requirements, few hours of operation, esthetic expectations and the classic luthier’s approach to making such an instrument by hand have excluded the double basses’ tuning assembly from all engineering approaches regarding optimal design, efficiency and costs. Manufacturing the traditionally designed double bass worm gear assembly using Additive manufacturing processes requires the application of general design rules and the rules of Design for Additive Manufacturing (DFAM) resulting in an optimized gear assembly regarding weight, costs and design properties.Item Powder Bed Fusion Metrology for Additive Manufacturing Design Guidance(University of Texas at Austin, 2017) Allison, Jared; Sharpe, Conner; Seepersad, Carolyn Conner; Kubiak, StevenDesign for additive manufacturing (DFAM) guidelines are important for helping designers avoid iterations and fully leverage the design freedoms afforded by additive manufacturing (AM). Guidelines can be generated via metrology studies that use test parts to characterize statistically the geometric capabilities of specific AM processes. Towards that end, a test part is designed for polymer selective laser sintering (SLS) that incorporates an array of geometric features in an extremely compact volume, such that it can be easily inserted into existing builds. The part is then built in multiple materials, build orientations, and locations within the build chamber in a factorial-style study to assess the variation attributed to each processing parameter. Both part resolution and accuracy are investigated. Upon measurement of the test parts, tolerances and design allowables are established and compiled into a set of design guidelines for SLS. The guidelines are then made publicly accessible through an online web tool to be used by designers creating parts for polymer SLS.Item A Triz-Based Analysis of the Fundamental Limits of Fused Filament Fabrication(University of Texas at Austin, 2021) Weaver, J.M.; Patternson, C.Each category of additive manufacturing (AM) has specific fundamental limitations bounded by the physics and material properties involved. For example, the speed of fused filament fabrication (FFF) processes is bounded by how quickly thermoplastics can be melted, deposited, and resolidified while retaining material properties and dimensional accuracy. Incremental improvements approaching these theoretical limits will continue to occur, but more radical changes are necessary to completely overcome the current constraints. This paper considers some of the fundamental limits bounding FFF processes and investigates possible avenues for future research to overcome these limits. The framework for this analysis is the “Theory of Inventive Problem Solving” (TRIZ), a formalized problem solving and ideation tool that generalizes design-specific problems into contradicting engineering parameters, then suggests universal design principles based on analogy to solutions in other systems and patents. TRIZ has been used in many fields successfully, including the design of parts to be more manufacturable through AM, but literature on its application to additive manufacturing processes themselves is limited. Two case studies are shared demonstrating how TRIZ-based analysis can lead to radical improvements in FFF and other AM technologies.Item Using Autoencoded Voxel Patterns to Predict Part Mass, Required Support Material, and Build Time(University of Texas at Austin, 2018) Murphy, C.; Meisel, N.; Simpson, T.W.; McComb, C.Additive Manufacturing (AM) allows designers to create intricate geometries that were once too complex or expensive to achieve through traditional manufacturing processes. Currently, designing parts using features specific to AM, commonly referred to as Design for Additive Manufacturing (DfAM), is restricted to experts in the field. As a result novices in industry may overlook potentially transformational design potential enabled by AM. This project aims to automate DfAM through deep learning making it accessible to a broader audience, and enabling designers of all skill levels to leverage unique AM geometries when creating new designs. To execute such an approach, a database of files was acquired from industry-sponsored AM challenges focused on lightweight design. These files were converted to a voxelized format, which provides more robust information for machine learning applications. Next, an autoencoder was constructed to a low-dimensional representation of the part designs. Finally, that autoencoder was used to construct a deep neural network capable of predicting various DfAM attributes. This work demonstrates a novel foray towards a more extensive DfAM support system that supports designers at all experience levels.