2024 Annual International Solid Freeform Fabrication Symposium (SFF Symp 2024): Process Modeling
Program Organizers: Joseph Beaman, University of Texas at Austin
Tuesday 8:00 AM
August 13, 2024
Room: 615 AB
Location: Hilton Austin
Session Chair: Emmanuel Ekoi, University of Texas at Austin
8:00 AM
Understanding Melt Pool Variation Due to Geometric Features and Scan Pathing: David Anderson1; Shawn Hinnebusch1; Haolin Zhang1; Praveen Vulimiri1; Alaaeldin Olleak1; Albert To1; Xiayun Zhao1; 1University of Pittsburgh
Melt pool variation in laser powder bed fusion printing processes is a potential source of porous defects that are inconsistent and difficult to model. To better understand the effects of laser scan pathing and local geometric features on residual heat and melt pool variation, simulations were conducted using PAMSIM (Pittsburgh Additive Manufacturing SIMulation), a GPU-based finite element method simulation tool and validated against in-situ monitoring data. From these simulations, a predeposition temperature is extracted based on the occurrence of a local minima in the thermal history of a given point prior to laser scanning. This metric for residual heat is correlated with changes in melt pool depth, which indicates transitions to and from a potential “keyhole” or overheating melt pool regime. From these results, a correlation between residual heat accumulation (due to local geometric features and scan vector length) and the corresponding melt pool geometry is presented.
8:20 AM
Numerical Simulations of Melt Pool Physics in Laser Powder Bed Fusion Processes: Craig Weeks1; Jonathan Malen1; Satbir Singh1; 1Carnegie Mellon University
In metal additive manufacturing, information about the physics and properties inside of the melt pool can prove invaluable for predicting the microstructure, defects, and strength of a printed part. Experimental approaches alone cannot explain the underlying physical processes and require complementary models. Analytical models can only capture rudimentary heat conduction effects in the build medium. Numerical models can yield a much better understanding of the fluid advection and vaporization physics present inside the high temperature regions of the melt pool. Effective implementation of underlying physics into open-source numerical codes can enable simulations with greater physical accuracy. In this talk we will share progress on open-source multiphysics modeling of conduction, advection, and vaporization of laser powder bed fusion (LPBF) meltpools using the OpenFOAM platform. Model development and methodology will be explored along with ongoing validation by experiments.
8:40 AM
Physics-based Modeling of Polymer-Metal Composites in Fused Filament Fabrication: Tyler Paupst1; Paromita Nath1; 1Rowan University
Manufacturing polymer-metal composite products of desired quality using fused filament fabrication (FFF) technologies requires a comprehensive understanding of the material and the process. There has been a substantial effort in recent years on modeling polymer FFF, however, polymer-metal composites exhibit unique physical behaviors. These unique conditions, such as particle-fluid interactions and visco-elastic effects on melt flow, complicate simulation dynamics rendering existing modeling methods ineffective in predicting the melt behavior. This work leverages a computational fluid dynamics modeling approach to simulate the deposition of a homogeneous polymer-metal composite in the FFF process. This approach allows for an efficient investigation of the effects of the process parameters, such as nozzle temperature, nozzle speed, and material composition, on the application-relevant quality of the deposited material. Quantities of interest such as the profile of the deposition and the thermal history are studied. Future considerations for a heterogeneous simulation are also discussed.
9:00 AM
Additive Manufacturing Beyond the Gaussian Beam: Insights from Microstructure-based Modeling Studies: Daniel Moore1; Giovanni Orlandi2; Theron Rodgers3; Daniel Moser3; Heather Murdoch4; Fadi Abdeljawad1; 1Lehigh University; 2Clemson University; 3Sandia National Laboratories; 4DEVCOM Army Research Laboratory
The use of Gaussian laser beams in state-of-the-art metal additive manufacturing (AM) has been shown to introduce large thermal gradients, melt pool defects, and undesirable microstructural features. Recent experimental studies employed spatially extended laser beam profiles to provide greater control over the thermal fields, allowing for tailored grain morphologies and opening the door to enhanced AM process control. With the aid of a recently developed coupled thermal transport-microstructure Monte Carlo model, we investigate the impact of using spatially extended laser beam profiles, such as ring and Bessel beams, on the evolution of thermal fields and grain microstructures. Simulation results show that the use of spatially extended laser beams expands AM processing windows and allows for increased laser scan speeds. Furthermore, our results reveal notable variations in the resultant grain microstructures across a wide range of processing conditions, underscoring the enhanced adaptability and process control using laser beam shaping.
9:20 AM
Multi-Layer Temperature History Prediction During DED Process with Physics-Informed Neural Network: Bohan Peng1; Janos Plocher2; Ajit Panesar1; 1Imperial College London; 2Ansys
In this work, we present a framework that applies physics-informed neural networks (PINN) in obtaining the temperature history in a multi-layer DED process. The presented work is one of the first attempts in applying PINN to the multi-layer problem. It overcomes the conflict between the intrinsic discontinuous nature of the DED process and the typical shortfall of NN in working with discontinuity with some simple but efficient techniques. The result from PINN is compared against ANSYS benchmarks to demonstrate the accuracy while achieving potential time savings for large scale parts with transfer learning. The proposed framework also promises immediate availability of thermal gradient information and capability of super-resolution which opens up pathways to alternative means for thermomechanical prediction of DED process.
9:40 AM Break
10:00 AM
High-Fidelity Modeling of Multi-Component/Material Additive Manufacturing with Chemical Reactions: Wentao Yan1; 1National University of Singapore
I will present our high-fidelity models of multi-component/material additive manufacturing with chemical reactions. The first scenario is oxidation, which is inevitable. Our high-fidelity model reproduces oxidation (also other gas-liquid metal reactions), oxygen mass transport, and the effect on molten pool flow. The second scenario is AM of mixed powders for particle-reinforced composites, and we developed high-fidelity models for both micro- and nano- particles considering size effects. The third scenario is AM of mixed powders for in-situ alloying, where our model enables synergistic design of alloy compositions and AM parameters. We developed phase field and cellular automaton models to simulate grain and dendrite evolutions, incorporating the effects of nano-particles, melt flow and locally varied solute concentration, providing guidance for the design of new AM alloys with intrinsic hot cracking resistance. Finally, we leveraged crystal plasticity simulation and in-situ tensile experiments to reveal how the heterogeneous structures impact mechanical properties.
10:20 AM
Numerical Simulations and Experimental Validation of Laser-Foil-Printing Additive Manufacturing: Tunay Turk1; Tao Liu1; Emmanuel Olugbade1; Chia-Hung Hung2; Jonghyun Park3; Ming Leu1; 1Missouri University of Science and Technology; 2National Cheng Kung University; 3Missouri Univ of Science and Technology
Laser-foil-printing (LFP) is a foil-fed metal additive manufacturing (AM) that offers many unique features over powder-fed systems, such as faster cooling, lower cost feedstock, and uniform heat distribution. This study investigates LFP using thermal finite element analysis (FEA). The analysis incorporates an absorptivity model to account for Gaussian beam laser energy input, surface roughness-dependent thermal contact conductance, and the temperature-dependent thermophysical properties of the feedstock. The use of adaptive mesh refinement allows the FEA model to simulate a single-pass of a line-raster scan at variable speeds by focusing on critical regions like the melt pool. The model predicts melt pool size and cooling rate. Validation against experimental data was done for LFP of 304L stainless steel using scanning electron microscopy. This study demonstrated that, by incorporating material, foil thickness, and geometry in the simulation, the FEA model could be a cost-effective tool for defect prevention in the LFP process development.
10:40 AM
Modeling the Effect of Thermal Contact Resistance on Mechanical Properties in Fused Filament Fabrication: Ahmed Adisa1; David Kazmer1; Amy Peterson1; 1University of Massachusetts Lowell
This study explores the impact of thermal contact resistance on the mechanical properties of Fused Filament Fabrication (FFF) parts. A one-dimensional heat transfer model is developed to simulate a single road-width polycarbonate system. This model not only accounts for the influence of process variables on heat transfer within the structure but also incorporates experimentally measured environmental temperatures and thermal contact resistance values. The predicted tensile strength of the FFF part is then compared against models that neglect the effect of thermal contact resistance. The results demonstrate that, as layer deposition time increases, thermal contact resistance plays a more crucial role in determining the tensile strength. This highlights the importance of considering thermal contact resistance in modeling the mechanical properties of FFF parts.
11:00 AM
Part-Level Heat Buildup Across Directed Energy Deposition Additive Manufacturing Processes: Elizabeth Chang-Davidson1; Jose Loli2; Brandon Abranovic3; Jack Beuth4; 1Northeastern University; 2Northrup Grumman; 3Relativity Space; 4Carnegie Mellon University
In large-scale additive manufacturing directed energy deposition (DED) processes, heat buildup on a part level has a strong impact on residual stresses, deformation, and microstructure in the final printed part. In this work, geometric and thermal issues in thin walls are studied and then mitigated through the use of specific build strategies and interpass delay times in laser hot wire DED. The rapid semi-analytical models used for delay time selection are then generalized across process space, geometries, and DED technologies. However, in more complex geometries, part-scale thermal modeling is often prohibitively computationally intensive. To address this, a surrogate model for predicting temperatures over the height of the part across geometries and process parameters is trained on a select set of FEM results. The results from this modeling provide insight into build planning for DED systems in a process-agnostic way, enabling pre-build optimization and cutting down on waste from failed builds.
11:20 AM
Finite Element Analysis for Sintering of Binder-Jetted SS316L Parts: Emrecan Soylemez1; Mehmet Fatih Yalcin1; Omer Faruk Anlayis2; Recep Onler2; 1Istanbul Technical University; 2Gebze Technical University
Binder jetting is a multi-step additive manufacturing process in which metal powders are formed into a 3D shape through the deposition of binder, followed by sintering. The surface quality of the final parts surpasses that of prevalent powder bed fusion processes at lower costs. However, achieving the final part geometry while considering the sintering shrinkage in all three axes is a critical task that affects the reliability of the process. Employing trial and error for each design is not efficient. Thus, predicting the sintering behavior with numerical calculations is crucial. In this study, we conducted sintering experiments for binder-jetted SS316L parts using an in-situ camera monitoring set up. Then, we developed empirical fitting parameters for a continuum finite element model to calibrate the simulation predictions. This approach helps to achieve the desired shape and design guideline for binder jetting additively manufactured SS316L parts.
11:40 AM
Probing the Effect of Dislocation Cell on the Plasticity and Anisotropy of Additively Manufactured Copper via Crystal Plasticity Finite Element Modeling: Md Mahabubur Rohoman1; Caizhi Zhou1; 1University of South Carolina
Due to the high cooling rate during additive manufacturing (AM) of metals, especially laser-powder bed fusion process, the microstructure within AM metals has unique features, such as interlayer interfaces, molten pool boundaries and dislocation cell structures. It is critical to understand the relationship between the spatial heterogeneity of the microstructure and the mechanical properties of AM metals. In this work, the crystal plasticity finite element (CPFE) model will be used to explore the effect of dislocation cells on the strength and anisotropy of AM copper for the first time. We will analyze how dislocation cell size and dislocation density affect the evolution of geometrically necessary dislocation density and statistically stored dislocation density during plastic deformation in AM copper. We will also explore the influence of the dislocation cell structure on the anisotropy of plasticity in AM copper. These results will deepen our understanding of the plastic deformation of AM metals.