Date of Award:

5-2011

Document Type:

Thesis

Degree Name:

Master of Science (MS)

Department:

Mechanical and Aerospace Engineering

Advisor/Chair:

Dr. Thomas H. Fronk

Abstract

The stress and strength behavior of cylindrical tubular adhesive joints composed of dissimilar materials was explored. This was accomplished with the finite element method (FEM) and stress-based failure theories. Also, it was shown how a design of experiments (DOE) based method can be used to objectively organize the process of optimizing joint strength by using stress-based failure criteria.

The finite element program used in this work was written in-house from scratch to implement the FEM for the purpose of solving both axisymmetric and three-dimensional linear elastic governing equations of static equilibrium. The formulation of the three-dimensional model is presented, and the required operations to arrive to the axisymmetric model are also presented. The axisymmetric model is two dimensional, capable of using four and eight node quadrilateral elements. However, only four node elements are used because a mesh of eight node elements requires more memory and increased mesh refinement. The three-dimensional model is capable of using eight and twenty node brick elements, but only eight node brick elements are used for the same reason.

Both of the axisymmetric and three-dimensional models calculate the nodal displacements, strains, stress values for each material, and strength values for each material. The external static loads can be individually applied, or coupled together. The outputs seem to be most useful for interpretation when plotted through-the-thickness (TTT) and along-the-length (ATL) of the joint or tube. Outputs are valid only for materials that behave linearly elastic up to(or near) failure, and the stress-based failure criteria are used to define that limit.

A small laboratory-sized joint was modeled to look at the theoretical stress and strength distributions plotted along-the-length of the joint at different radial locations. These stress and strength distributions can be correlated to the type of load being applied because of unique or prominent features seen in the stress and strength distributions. The load can be a uniform temperature change, axial load, torque load, internal and external pressure, and/or bending load. A variance in the stress or strength for different joint sizes and materials is not examined closely due to the many possible combinations of these parameters.

Comments

This work was made publicly available electronically on September 30, 2011.

Share

COinS