In the aerospace industry, fiber reinforced composites are gaining increasing use and attention in structures such as the Ares V. As a result, adhesively bonded joints are increasingly popular, although often hard to analyze because of an eccentric load path and adhesives which often exhibit nonlinear behavior. Additionally, joints can fail in multiple locations: in the adhesive, in the adhesive/adherend interface, or in the adherend.

Because of a lack of confidence in joint prediction methods, heavy mechanical fasteners have often been used along with adhesive as an insurance against catastrophic failure. Therefore, accurate analysis of adhesively bonded joint behavior and failure is becoming more critical than ever.

Finite element Analysis (FEA) has been widely utilized in industry to assess joints. However, a dense mesh is required to capture the complex behavior of joints. Such models are costly, and are impractical to use in vehicle-scale models to perform initial sizing and assessment early in the design cycle.

Therefore, a mesh-independent element is being developed for joint evaluation in global finite element models. This macroscopic joint element is constructed by imbedding an analytical model into the element, thereby giving the single element the behavior of the whole joint. In addition, failure models are being added to the element to predict the failure mode, location, and post-failure behavior of the joint. In conjunction with model development, experiments will be used to validate the accuracy of the model.

This project is supported through sponsorship from the NASA CUIP project.