Evaluation of strength of steel bonded joints undergoing axial-torsional non-proportional fatigue loading

Real-life applications are generally subjected to multiple, out-of-phase sources of fatigue loading, i.e. Multiaxial Non-Proportional Fatigue (MNPF). On the other hand, most of the literature research is focused on Multiaxial Proportional Fatigue (MPF) of adhesive joints, with just a limited number of recent papers dealing with MNPF. The objective of this work is therefore to identify a failure criterion that can represent MNPF better than others, helping to fill the gap due to the limited literature (novelty 1). For this purpose, some criteria were selected from the literature, belonging to three different groups: i) pressure-independent; ii) pressure-dependent and iii) critical plane. A bonded butt joint was designed to have adhesive thickness self control, to be joined without a template and tested under simple axial, torsional and axial-torsional MPF and MNPF (novelty 2). A two-component epoxy adhesive was selected, since no data were found in the literature for room temperature-curing adhesives, while they are a very common solution for bonding metallic and/or composite adherends (novelty 3). The adherends were manufactured out of S235 construction steel. Tests were conducted at room temperature. The adhesive joint was found to be sensitive to MNPF loading, with a significant reduction in strength (between 13% and 4% for a number of cycles to failure ranging from 10 4 to 10 6 , respectively) compared to the MPF case (main result 1). The Matake and Findley critical plane criteria provided a good correlation of all the fatigue tests results, highlighting the appropriateness of a critical plane approach for adhesive joints (main result 2). However, a complete physical interpretation of the differences among criteria could not be fully established based solely on the present experimental findings, thus indicating the need for further investigation.