Date of Award

5-2017

Document Type

Dissertation

Degree Name

Doctor of Philosophy (PhD)

Legacy Department

Mechanical Engineering

Committee Member

Dr. Georges Fadel, Committee Chair

Committee Member

Dr. Nicole Coutris

Committee Member

Dr. Mohammed Daqaq

Committee Member

Dr. James Gilbert

Committee Member

Dr. Henry Rack

Abstract

Ultrasonic additive manufacturing (UAM) is an additive manufacturing technolo-gy that combines an additive process of joining thin metal foils layer by layer using ultra-sound and a subtractive process of CNC contour milling. UAM can join similar or dis-similar materials and allows for embedded objects such as fibers and electronics. Despite these advantages, the UAM process exhibits a critical bonding failure issue as the height of the built feature approaches its width. Based on previous studies, we believe that the loss of bonding is due to complex dynamic interactions between the high frequency excitations of the sonotrode and the built feature. While the previous investigations have qualitatively explained the cause of the height to width ratio problem by showing the change of dynamic states as new layers of foils are deposited, they do not explain how the change of dynamics affects bond formation. Specifically, a UAM model is needed to be able to predict the bond quality, i.e. bond or debond, as the dynamics of the substrate state change. In order to establish the model, a comprehensive understanding of the welding process and bonding mechanisms is required. Due to the complexity of the bonding process, the model is first decomposed into several sub-models based on the different factors that affect the process. The key factors that govern the bonding process: material plasticity, heat transfer, friction, and dynamics need to be characterized. An experiment setup is designed to investigate and characterize the effects of ultrasound on aluminum 6061-O, 6061-T6, 1100-O, and Copper 11000-O. A plasticity model is proposed by modifying the Johnson-Cook plasticity model to introduce strain-rate hardening and acoustic softening effects. A lumped parameter model consisting of mass-spring network is proposed to re-place the finite element dynamic model for reducing computational cost. An asperity lay-er model based on sinusoidal shape solid asperities is proposed to associate the plastic deformation of the material to the linear weld density of the bonding at the interface. Other sub-models (thermal and friction models) are defined based on studies in the literature. The sub-models are implemented in the commercial software ABAQUS by using user subroutines and are integrated into one UAM model. The model is validated by comparing its prediction with experimental results in the literature. The proposed model can thus be used to understand the effects of dynamics on the stress state close to the bond inter-face, understand the energy flow within the UAM system, and evaluate the effects of different process parameters on the bond quality for process optimization.

Share

COinS