Biaxial testing of adhesively bonded composite joints
Debonding has been noted as the main cause of failure in bonded composite assemblies and there have been many studies to characterize the failure models surrounding the complex nature of adhesive bond failure. In this study, the author has fabricated and recycled Glass Fiber Reinforced Polymer (GF...
Saved in:
| Main Author: | |
|---|---|
| Other Authors: | |
| Format: | Final Year Project |
| Language: | English |
| Published: |
2018
|
| Subjects: | |
| Online Access: | http://hdl.handle.net/10356/74876 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Institution: | Nanyang Technological University |
| Language: | English |
| Summary: | Debonding has been noted as the main cause of failure in bonded composite assemblies and there have been many studies to characterize the failure models surrounding the complex nature of adhesive bond failure.
In this study, the author has fabricated and recycled Glass Fiber Reinforced Polymer (GFRP) specimens to be joined using epoxy adhesive to perform mixed-mode biaxial loading experiments using a modified Arcan Setup. Three types of adhesive joint profiles were used to test and compare the fracture characteristics of the specimens. ABAQUS® finite element analysis software was then used to simulate the experimental measurements of load-displacement to determine a suitable failure model that could fit the fracture mechanics of the debonding of the specimens through the use of cohesive elements and traction-separation law.
It was determined that the quadratic nominal stress failure criterion fitted with power law damage evolution was suitable for the straight profile samples. For the rectangle profile, the failure criterion was only suitable for very low loading angles, however the Benzeggagh and Kenane (BK) mixed mode damage evolution criterion was able to predict crack patterns which the power law damage evolution was not able to. The failure criterion did not fit well for the triangle profile as most stresses were either over or under predicted by both damage evolution criterions. However, for loading angles at 60° and 75°, the power law damage evolution was able to predict peak loads. For the straight and rectangle designs, it was noted that the fracture energies predicted by the failure criterion was suitable for loading angles below 45°. For the triangle profile, the fracture energy was suitable for loading angles at 0°, 15° and 60°. |
|---|