2023 Annual International Solid Freeform Fabrication Symposium (SFF Symp 2023): Plenary Session
Program Organizers: Joseph Beaman, University of Texas at Austin
Monday 8:00 AM
August 14, 2023
Room: Salon HJK
Location: Hilton Austin
Session Chair: Joseph Beaman, University of Texas at Austin
8:00 AM Introductory Comments
8:15 AM Plenary
Design, Analysis, Optimization, and Fabrication of
Porous/Heterogeneous Microstructures via Spline-based Volumetric Representations: Gershon Elber1; 1Technion
Boundary representations (B-reps) can no longer satisfy the needs of modern (additive) manufacturing (AM) technologies. AM requires the representation and manipulation of interior fields and materials. Further, while the need for a tight coupling between design and analysis has been recognized as crucial almost since geometric modelling (GM) has been conceived, contemporary GM systems only offer loose links between the two. For more than half a century, the (trimmed NURBs) surface representation has been the B-rep of choice for the GM industry. Fundamentally, B-rep GM has evolved little during this period. Herein, we present a, B-spline based, volumetric representation (V-rep) that successfully confronts existing and anticipated design, analysis, and manufacturing foreseen challenges. Among others, full support of AM of porous heterogeneous artifacts.Examples and applications of V-rep GM, that span design, analysis and optimization, and AM, of (heterogeneous) lattice- and micro-structures, including compliant microstructure mechanisms, will be demonstrated.
8:40 AM Plenary
Implementing In-situ Process Monitoring in Additive Manufacturing to Accelerate Qualification: Michael Heiden1; Jesse Adamczyk1; Dan Bolintineanu1; Anthony Garland1; Ana Love1; Hyein Choi1; David Moore1; Catherine Appleby1; David Saiz1; 1Sandia National Laboratories
Studies across industry have demonstrated in-situ monitoring’s ability to track build events during metal additive manufacturing (AM) processes. However, process monitoring hasn’t yet been implemented in production environments due to the slow, resource-intensive nature associated with acquiring/analyzing large datasets with multiple complex outputs. The goal is to equip AM production machines with deployable sensor hardware, along with a simple user interface and software that provides a manageable amount of data for designer decision-making. This presentation describes a multi-year effort to develop a resource-effective in-situ instrumentation toolset for AM production environments that incorporates machine learning to ultimately form a common data processing framework. The study aims to assist AM process development, ensure process consistency, and contribute to part acceptance for metal AM production. Furthermore, this work investigates how in-situ process monitoring contributes to an AM part’s digital thread.SNL is managed and operated by NTESS under DOE NNSA contract DE-NA0003525.
9:05 AM Plenary
Machine Learning Applied to
Process Monitoring for Laser Hot Wire Additive
Directed Energy Deposition
: Brandon Abranovic1; Elizabeth Chang-Davidson1; Jack Beuth1; 1Carnegie Mellon University
In order to reliably print quality parts using additive manufacturing, process monitoring is a crucial step in the build process. During deposition, large quantities of data are collected and must be analyzed in order to detect and eliminate process anomalies. Unsupervised deep learning techniques are valuable in executing this analysis due to their ability to recognize flaws without the need for vast quantities of labeled data. A convolutional long-short term memory autoencoder model was trained on process data from a laser hot wire additive manufacturing process. This model used, as input, data from both a visible-light camera and an infrared camera, to encompass melt pool disturbances as well as near-melt pool part temperatures. This model is shown to be feasible as a real-time monitoring technique capable of detecting known characteristic process flaws, as well as a post-deposition data analysis tool for directing part testing towards suspected flaw areas.
9:30 AM Plenary
Experimental and Computational Study of Area Printing™ Additive Manufacturing: Inconel 718 and M300 Maraging Steel Density Improvement: Ben Fotovvati1; Subin Shrestha1; Nicholas Ferreri1; Ning Duanmu1; 1Seurat Technologies
The low manufacturing speed of laser-powder bed fusion (LPBF) additive manufacturing has hindered its adoption in conventional manufacturing methods. Large-area pulsed laser powder bed fusion (LAPBF), also known as “Area Printing”, has addressed this limitation by replacing the point laser with a large-area pulsed laser. Each pulse melts a region of the powder bed on the order of square millimeters, which enables shorter manufacturing time and lower final production costs in addition to higher quality parts due to the lack of spatter compared to conventional LPBF methods. In this study, process parameters are optimized to achieve near full-density parts, and a computational model is developed to understand the multi-physics governing the process. It is observed that the shallow depth and large aspect ratio of the melt pool lead to a unidirectional solidification front extending along the build direction where grains grow epitaxially, and highly directional microstructures are created.
9:55 AM Plenary
Gas Flow and Delivered Laser Power Effects on Mechanical Properties: Joy Gockel1; Edwin Glaubitz1; Clay Perbix1; Sage Frontella1; Allan Huntington2; Ryan Fishel2; Jeff Shaffer2; 1Colorado School of Mines; 23D Systems
Reducing process variability to achieve predictable and repeatable part performance is necessary to accomplish successful machine qualification of AM technology. Machine processing variables (e.g.- gas flow) and laser window contamination by process byproducts lead to within-build and build-to-build anomalies that are currently unexplained and hinder the translation of processing knowledge from one build, or one machine, to another. This work investigates a measured grid of gas flow and laser power and relates to material structure and mechanical properties. Investigated structure and properties include microstructural characteristics, surface features (protrusions vs notches), forms of porosity (interlayer, spatter, sub contour, etc.), tensile strength and fatigue life. The outcome of this work will identify additional process metrology needs, monitoring requirements, and processing pedigree documentation that must be included in AM qualification standards.
10:20 AM Plenary
NSF Advanced Manufacturing Program and Funding Opportunities: Khershed Cooper1; 1National Science Foundation
NSF’s Advanced Manufacturing (AM) program supports transformative advances in materials engineering, processing, and manufacturing and fosters multidisciplinary research that applies innovative manufacturing approaches to accelerate new product development, customize products, increase production efficiency and reduce production cost. The AM program encourages industry partnerships through collaborations with the Manufacturing USA Institutes and international partnerships through collaborations with DFG and funding agencies of other countries. The AM portfolio includes the entire manufacturing enterprise including additive manufacturing, all scales (nano to infrastructure), all materials systems and most applications. The AM program participates in several cross-cutting activities such as Future Manufacturing (FM), Critical Aspects of Sustainability (CAS), Engineering Research Centers (ERCs) and others, and supports education and outreach activities through solicitations such as BRITE and supplements such as INTERN. These programs address national priorities such as AI/ML, semiconductors, quantum, sustainability, climate change, clean energy, biomanufacturing, and workforce development. This presentation describes program goals, basic research activities and accomplishments of the AM and related programs.
10:45 AM Break
11:15 AM Plenary
You Must Unlearn What You Have Learned: Establishing a DfAM Mindset in the Face of Centuries of Traditional Manufacturing: Nicholas Meisel1; 1Pennsylvania State University
With the rise of additive manufacturing, we as designers need to dig deeper into our designs, with the aim of getting straight to the heart of functionality, without limiting ourselves to previous notions of what something is “supposed” to look like. In this talk, the speaker will discuss his lab’s ongoing research into establishing an evidence-based design for additive manufacturing mindset in human designers as derived from years of research. This will include discussion of the types of designs that engineers are likely to generate early-on in their design process, as well as how these designs align with the principles of traditional manufacturing or additive manufacturing. Practical strategies for changing designers’ thinking about manufacturing will also be demonstrated, with findings reported from the use of virtual reality, problem-based learning activities, and online design evaluation tools. At the end of this presentation, attendees will be able to better understand not only the challenges facing engineers when it comes to unlearning their previous notions of manufacturability, but also how to address these challenges head-on to encourage the generation of additive manufacturing-appropriate designs earlier in the design process.
11:40 AM Plenary
Co-design of 3D Printing, Parts and Microstructure in High-temperature and High-pressure Heat Exchangers: Anthony Rollett1; 1Carnegie Mellon University
We illustrate the importance of co-design with the example of high temperature heat exchangers that employ a pin-and-plate structure and integral headers. 3D printing complex geometries entails mostly empirical adjustments to, e.g., pin geometry and avoidance of sharp changes in cross-section. Concurrently, optimal print conditions must be identified for each printer that minimize porosity. Qualification involves predicting the process window from the physics of laser powder bed fusion and then printing test parts to measure porosity as a function of P, V, H etc. Characterization of single tracks calibrates absorptivity, e.g. The resulting mechanical properties in both Haynes 230 and 282 are comparable to standard product despite the fine grain structure. Microstructure simulation points to strategies for optimization but surrogate models offer potential for substantial speedup. High-speed visualization with synchrotron x-rays provides calibration data of all sorts and we end with select recent results on oscillation and pore formation detection.