Open Collections

Pourabdollah, Pegah

University of British Columbia

This thesis investigates the key factors required to efficiently predict distortion using numerical methods in two metal Additive Manufacturing (AM) processes: Laser-based, Powder-fed Directed Energy Deposition (LP-DED) and Electron Beam Powder Bed Fusion (EB-PBF). For LP-DED, a part-scale thermomechanical model was developed to predict distortion, incorporating layer agglomeration and time-averaged heat input for computational efficiency. A new approach was developed in which the deposited material was activated at an initial temperature above the solidus to generate plastic strain. The energy input from the beam was adjusted for the enthalpy of deposited material, ensuring a correct system heat input. Additionally, a material-centric approach accounted for the thermal strain mismatch between the deposited material and the start plate, contributing to part distortion. The model was validated against experimental data. Three models were developed for EB-PBF. First, a novel system-scale thermal cavity radiation model was formulated, which included component fabrication to predict heat transport within the built chamber and component. The model incorporated layer agglomeration and time-averaged heat inputs for each computational layer, calculated based on EB-PBF system data considering beam efficiency and enthalpy associated with activating layers at preheat temperature. Experimental data was used to validate the model. The second model conducted a thermomechanical analysis of a component with overhang features fabricated using EB-PBF. It incorporated layer agglomeration, time-averaged heat input, and an Inherent Strain (IS) strategy. Using a new approach, the IS values were applied as initial anisotropic thermal strains at layer activation, establishing static equilibrium layer by layer. By adjusting IS values, the model was tuned to predict the experimentally measured distortion. Lastly, a sub-domain thermomechanical model was developed to explore the role of substrate temperature in plastic strain generation. It described a small region, including the substrate and four powder layers. The thermomechanical properties were defined as a function of temperature and material form. Additionally, the yield stress was defined as a function of strain rate, which was found to be important. A linear regression model was proposed to link substrate temperature to plastic strain, which may be used to estimate in-part variation in the IS.

Text

2024-04-04

Attribution-NonCommercial-NoDerivatives 4.0 International

http://hdl.handle.net/2429/87694

Materials Engineering

University of British Columbia

UBCV

http://creativecommons.org/licenses/by-nc-nd/4.0/