Adhesively bonded joints under cyclic loading spectra

Current designs which involve the use of composite materials in primary aircraft structures are often conservative. This, in turn, significantly lowers the weight advantage that composites have over established metallic airframe materials. Strain restrictions are often applied because the failure mechanism(s) in (fibre) composite joints and stiffener runouts where the stress state is often complex, are not fully understood. Nevertheless, from the air-worthiness perspective it is essential that both the static strength and the fatigue behaviour of the components subjected to complex multiaxial stress conditions are both understood and predicted. This topic is extremely complex, and numerous criteria ranging from the purely empirical to the theoretical have been proposed. In both cases, it is necessary to know the localised stress-strain history. One common design methodology is to keep the stresses so low that fatigue will not be an issue. However, this can lead to an overly conservative design. On the other hand, while a detailed (nonlinear) finite element analysis can be performed it is often both resource-intensive and time-consuming. The present paper shows that Glinka's hypothesis can be used in order to calculate the localised stresses and strains for a bonded joint subjected to cyclic loading. This is a new result and has not previously been noted. It has the potential to extend the Hart-Smith design methodology to the adhesively bonded joints in order to encompass durability considerations. This formulation also raises the possibility of enabling the degree of conservatism inherent in traditional joint design to be relaxed provided that failure occurs in the adhesive. This paper also addresses the problem of variable adhesive thickness. We show that while variable adhesive thickness can change the stress and the energy fields, the peak in the strain energy density is relatively insensitive to the stress-strain relationship for the adhesive and that Glinka's hypothesis still appears to be true. This means that, for the present class of problems, even if there is variability in the thickness of the adhesive bond the energy field and, hence, the strength of the joint can be estimated from a purely linear elastic analysis of the joint, provided that failure occurs in the adhesive.