Thermo-mechanical Model Development and Experimental Validation for Directed Energy Deposition Additive Manufacturing Processes
Open Access
- Author:
- Heigel, Jarred Christopher
- Graduate Program:
- Mechanical Engineering
- Degree:
- Doctor of Philosophy
- Document Type:
- Dissertation
- Date of Defense:
- February 24, 2015
- Committee Members:
- Panagiotis Michaleris, Dissertation Advisor/Co-Advisor
Ashok D Belegundu, Committee Member
Eric Russell Marsh, Committee Member
Allison Michelle Beese, Committee Member
Todd Palmer, Committee Member
Edward William Reutzel, Special Member - Keywords:
- Additive manufacturing
laser cladding
distortion
convection
finite element analysis
residual stress
in situ measurement - Abstract:
- Additive manufacturing (AM) enables parts to be built through the layer-by-layer addition of molten metal. In directed energy deposition (DED) AM, metal powder or wire is added into a melt pool that follows a pattern to fill in the cross section of the part. When compared to traditional manufacturing processes, AM has many advantages such as the ability to make internal features and to repair high-value parts. However, the large thermal gradients generated by AM result in plastic deformation. Thermo-mechanical models must be developed to predict the temperature and distortion produced by this process. Thermo-mechanical models have been developed for AM by several investigators. These models are often validated by measuring the temperatures during the deposition of a small part and the final distortion of the part. Unfortunately this is not a sufficient validation method for the non-linear thermo-mechanical model. Although good agreement between the thermal model and the temperatures measured during a small depositions can be achieved, it does not necessarily mean that the model will be accurate for an industrially relevant part that requires 10^2 - 10^4 tracks and hours of processing time. The relatively small deviations between the model and the validation will propagate when modeling large depositions and could produce inaccurate results. The errors in a large part will be increased further if the assumptions made of the thermal boundary conditions are not appropriate for the system. The objective of this work is to develop and experimentally validate thermo-mechanical models for DED. Experiments are performed to characterize the distortion induced by laser cladding. The depositions require many tracks and nearly an hour of processing time, during which the temperature and the deflection are measured in situ so that the response of the plate to each deposition track is understood. Measurements are then made of the convection caused by two different laser deposition heads. Thermo-mechanical models are developed by implementing the measured rate of convective heat transfer and the temperature dependent material properties. The models are validated using in situ measurements of the temperature and the deflection generated during the process, as well as post-process measurements of the residual stress and the distorted shape. Finally, experiments and models are used to investigate the impact of feedstock selection, either powder or wire, on the DED process.