Optimization of analysis using VCCT’s for prediction of failure initiation of adhesively bonded joints
Authors
Advisors
Issue Date
Type
Keywords
Citation
Abstract
Adhesively bonded joints are used on numerous applications involving light-weight composite materials. Consequently, the successful modeling and subsequent analysis of bonded joints involving more than one adherend material is crucial for optimization of structural components. An increasing effort to analytically predict the failure strength of adhesively bonded joints has driven the industry’s attention to energy methods such as the virtual crack closure technique. This research attempts to optimize the modeling of such joints in order to acquire accurate analytical predictions with a reasonable computational cost. A detail study of the application of virtual crack closure technique in Abaqus® showed the existence of considerable dependence of results accuracy in mesh densities, element type, and mixed mode criteria used for post processing of the results. The investigation reported herein involves the use of Hysol® EA9394 adhesive to bond components made of two different carbon/epoxy materials, as well as aluminum into single lap shear, double lap shear, T pull-off, and thick lap shear specimens. The results of this research are not intended to provide the analyst with exact parameters for constructing a FEM for a bondline and post processing the results. Rather, this paper provides a procedure for optimizing the finite element model in order to maximize the accuracy-tocomputational time ratio. Additionally, a procedure for correlating the analytical results to test data is provided. It is concluded that adhesively bonded joints can be accurately analyzed using VCCT’s in Abaqus®, provided the finite element models are developed using the technique described herein for obtaining converged analytical solutions. Finally, the correlation of analytical results to test data is not a trivial task that is achieved utilizing the procedure developed during this research.