Date of Award:


Document Type:


Degree Name:

Master of Science (MS)


Mechanical and Aerospace Engineering


Thomas Fronk


In order to incorporate fiber-reinforced composite materials in space structures, adhesive joining techniques are required. Because analytical models have a hard time capturing the complex stress state inherent to adhesively joining dissimilar materials, a different modeling technique was deemed necessary. A two-dimensional axisymmetric finite element model capable of capturing the three-dimensional stress state of cylindrical adhesive joints was developed. In order to rigorously validate the model, testing was undergone to ensure the model accurately predicted joint displacements.

Displacement data was acquired via an Epsilon axial extensometer. Load data was taken simultaneously via the load cell incorporated in the Tinius Olsen tensile test machine used. The measured force vs. displacement data was found to agree with the model’s predicted displacement for a given load. Displacement data was also taken, again with the extensometer, as the joints were rapidly cooled to liquid nitrogen temperature. It was found that the joints behave much like laminated plates in that after the first several cycles they “settle down”. The term “settle down”, in this context, means that after the first several cycles the displacements of the joints when placed from a room temperature environment to a cryogenic environment become consistent and smooth. This result allows for the joints to be modeled. The finite element model was shown to accurately predict the settled down displacement given the prescribed temperature change.

The joints were also shown to maintain structural integrity post thermal cycling. Transient temperature tensile tests were performed until sample failure. One result with major design implication coming from this test was that the material properties do not change significantly enough over the temperature range tested to affect the joint’s behavior. The same properties used in the room temperature model were used to model the measured data of the transient temperature data and were found to match satisfactorily.

Having validated that the developed axisymmetric finite element model accurately predicts cylindrical joint displacement fields, the model becomes an invaluable tool in design. The model can now be used in confidence, in conjunction with design requirements for a specific joint, to reduce the maximum displacements below any specified operating requirements.


Publication made available electronically December 21, 2011.