2023 Annual International Solid Freeform Fabrication Symposium (SFF Symp 2023): Thermal Modelling
Program Organizers: Joseph Beaman, University of Texas at Austin
Tuesday 8:15 AM
August 15, 2023
Room: Salon G
Location: Hilton Austin
Session Chair: Bradley Jared, University Of Tennessee, Knoxville
8:15 AM
Thermal Simulation of the Material Extrusion Process with Different Print Bed Boundary Conditions: Orkhan Huseynov1; Mohammad Al-Shaikh Ali1; Ismail Fidan1; 1Tennessee Technological University
The temperature evolution in the material extrusion (MEX) process significantly affects the stability and bonding of 3D printed parts. Numerous studies have focused on developing models to capture the temperature history of the MEX process. However, there remains a need to explore the influence of different boundary conditions applied to the print bed. Additionally, the size of the bed relative to the 3D printed object has not been extensively investigated. This study aims to analyze the thermal behavior of the first layer in MEX by considering various boundary conditions and bed sizes. The obtained results will contribute to the development of faster yet reliable models for simulating the temperature variation in the MEX process.
8:35 AM
Analytical Modeling of Cooling Rates in PBF-LB/M of Bulk Metallic Glasses: Hanna Schonrath1; Jan Wegner1; Maximilian Frey2; Erika Barreto3; Arno Elspaß1; Norman Schnell1; Benjamin H. Erdmann1; Julian Neises1; Nils Ellendt3; Ralf Busch2; Stefan Kleszczynski1; 1Universität Duisburg-Essen; 2Saarland University; 3Leibniz Institute for Materials Engineering
Additive manufacturing through laser powder bed fusion (PBF-LB/M) inheres great potential for the processing of bulk metallic glasses (BMGs). The size-independent high cooling rates during processing issue a breakthrough for amorphous fabrication of large and complex structural components. Albeit partial crystallization is frequently observed in additively manufactured BMGs, which limits the resulting mechanical properties. In this matter the complex thermal history during processing states a remaining uncertainty. Besides temperature measurements and numerical estimation, analytical models might be useful for a deeper understanding of the transient temperature evolution. In this work, an iterative solution to the analytical Rosenthal equation was developed and applied to ZrCuAlNb- and CuTiZrNi-BMGs to predict melt pool dimensions and cooling rates during PBF-LB/M. Therefore, temperature dependent thermal properties were determined via laser flash measurements. The effective absorptivity of the two materials was additionally measured and single line experiments were performed as a validation for the approach.
8:55 AM
A Robust Local Preheat Temperature Dependent Stochastic Finite Element Heat Source Model for Inconel 718 Laser Powder Bed Fusion: Seth Strayer1; Albert To1; 1University of Pittsburgh
Thermal field prediction of laser powder bed fusion (L-PBF) via the finite element method can help optimize the process while avoiding the cost of experiments. However, these models abstract critical physics into an effective heat source model that does not readily capture the experimentally-measured melt pool size magnitude and variance, especially for multi-track cases. This work presents a novel local preheat temperature dependent stochastic heat source model to help address these issues. First, the heat source parameters are calibrated to the mean melt pool sizes for Inconel 718 L-PBF multi-track experiments. These parameters are predicted during the simulation to establish the role of a local preheat temperature metric. Second, random sampling techniques are employed to match the experimentally-measured variance within each track. Accordingly, the simulated melt pool sizes are within 10% error regarding experimental measurements up to five consecutive tracks while more closely matching the measured melt pool size variance.
9:15 AM
Enabling Part-scale Melt Pool Prediction in Laser Powder Bed Fusion via a Global-local Thermal Process Simulation Model: Shawn Hinnebusch1; William Templeton2; Alaa Olleak1; Praveen Vulimiri1; Florian Dugast1; Sneha Narra2; Albert To1; 1University of Pittsburgh; 2Carnegie Mellon University
Predicting accurate thermal history in laser powder bed fusion (LPBF) is a challenging problem. Layerwise simulations are geometry dependent for calibration and cannot capture the local heat accumulation due to the laser scanning process. Scanwise simulations are far more accurate but are restricted in size to just a few millimeters. An infrared (IR) camera is mounted on an LPBF system to calibrate and validate the interpass temperatures. Using a GPU-accelerated finite element based solver, the geometry-agnostic layerwise calibration was completed with less than 6% mean absolute percentage error. The layerwise simulation provides an accurate thermal boundary condition for the local scanwise simulations at a reduced computational cost. Melt pool width and depth can be predicted in any location before printing. Integrating high-speed layerwise simulations with scanwise simulations results in a low-cost yet accurate thermal history that identifies problematic regions before costly builds.
9:35 AM Cancelled
Component Geometry Feature-based Heat Source Model for Temperature History Fast Prediction in the Directed Energy Deposition Process: Lei Yan1; Wei Gao1; Frank Liou2; 1Nanjing University of Aeronautics and Astronautics; 2MIssouri University of Science and Technology
Temperature prediction is critical in metal directed energy deposition (DED) for process parameters optimization, especially comes to components with large sizes and complex geometry features. To improve prediction efficiency with accuracy guaranteed, a component geometry feature-based heat source model is proposed and developed with ANSYS APDL. The heat source model is applied layer-wise and has energy intensity redistributed according to the geometry features of each slice. Predicted temperature history is validated with thermocouple data and shows a maximum 50K temperature difference. This model provides a promising tool for high-efficiency process window optimization.
9:55 AM Break
10:25 AM
A Preliminary Understanding of Process-property Effects on the Thermal Response via High-throughput Finite Element Models of Wire Arc Direct Energy Deposition: Jeffery Betts1; Matthew Register1; Matthew Priddy1; 1Mississippi State University
Historically, process parameters for wire arc DED have been found in a heuristic manner for both experimental and computational efforts since process parameters are material- and geometry-dependent. However, finite element (FE) analysis can provide a low-cost, high-throughput method to simulate many process parameters. This worked utilized a sequentially coupled thermo-mechanical framework in Abaqus, using progressive element activation and the Goldak double-ellipsoidal heat source. A full-factorial design of experiments was constructed for the thermal analysis, varying the boundary conditions, material properties, time increment, mesh density and heat input. The results were post-processed to analyze the effects of each parameter on the maximum temperature and weld pool evolution at the start, middle, and end of the single-pass weld. An ANOVA test was conducted to examine relationships between process parameters (or combinations of parameters) and the nodal temperatures (at the start, middle, and end of the weld) as well as the melt-pool dimensions.
10:45 AM
CIFEM: Elucidating the Role of Local Preheat Temperature on Multi-track Melt Pool Morphology Variation for Inconel 718 Laser Powder Bed Fusion: Seth Strayer1; William Frieden Templeton2; Alaaeldin Olleak1; Florian Dugast1; Sneha Narra2; Albert To1; 1University Of Pittsburgh; 2Carnegie Mellon University
Despite advancements in finite element (FE) thermal simulation techniques for laser powder bed fusion (L-PBF), these models employ an effective heat source model, which invokes a tedious calibration process and provides inaccurate thermal fields compared to high-fidelity computational fluid dynamics (CFD) simulations. Accordingly, the driving force behind melt pool size variation, especially in the multi-track case, has remained enigmatic up to this point. In this work, the authors extend CIFEM to multi-track scenarios for Inconel 718 L-PBF to help address these issues. CIFEM's data-driven heat source model is trained to predict the thermal fields from multi-track CFD simulations with different scan lengths to establish the role of a local preheat temperature metric. By imposing these fields on the desired FE solution domain, the simulated melt pool sizes are within 10% error regarding experimental measurements up to five consecutive tracks while providing substantially more accurate thermal fields to traditional FE models.
11:05 AM
Thermomechanical Modeling and Fabrication of Tungsten Carbide-nickel Geometries Through Laser Powder Bed Fusion: Alexander Gourley1; Edgar Mendoza Jimenez1; Reeja Jayan1; Jack Beuth1; 1Carnegie Mellon University
The combination of hardness, abrasion resistance, and fracture toughness make cemented carbides desirable for machining and tooling, but these properties limit achievable geometries when fabricating parts through traditional techniques. Laser powder bed fusion (LPBF) provides a fabrication pathway through which cemented carbide parts can be achieved with greater geometric freedom. Our lab explored various process parameter combinations with a tungsten carbide - 17wt% Ni agglomerated powder, but recoater blade collisions limited achievable geometries to 15 mm tall cylinders. Leveraging a commercially available thermomechanical modeling software, Netfabb, the effects of interlayer timing, parameter combinations, and geometric features on recoater clearance guided designs for two prints evaluating printing larger and taller parts with more geometric features. All parts were printed to completion except for a long geometry that delaminated from the build plate. the measured density and hardness values in the taller parts were similar to the previous 15 mm tall cylinders.