PSI - Issue 57

Cristian Bagni et al. / Procedia Structural Integrity 57 (2024) 859–871 Author name / Structural Integrity Procedia 00 (2019) 000 – 000

864

6

wide and with a 30 mm overlap, 2 mm thick adherends and 0.5 mm thick adhesive. The CP specimen was modelled as 130.5 mm long, 60 mm wide and with a 25 mm overlap, 2 mm thick adherends and 0.5 mm thick adhesive. In both geometries the mechanical joints were modelled with a diameter of 10 mm. The convergence study was carried out, starting from initial meshes where the adhesive was modelled as one single layer of solid elements and the membrane shell elements had the same size as the solid elements. The mesh of the solid elements was then refined four times, by doubling the number of layers of solid elements at each refinement, while the mesh of the membrane shell elements wrapping the solid elements was kept constant. For comparison purposes the peel stresses from the centroid of both membrane shell elements and solid elements were extracted. The results of the FE analyses are graphically summarised in Fig. 3 and Fig. 4. As expected, the predicted highest peel stresses were in the adhesive, for both lap shear and coach peel geometries. Furthermore, the maximum peel stress estimated at the centroid of the solid elements and of the membrane shell elements was approximately in the same location, independently of the mesh refinement. Fig. 5 shows that while the maximum peel stress calculated at the centroid of the solid elements increased at each mesh refinement, the maximum peel stress calculated at the centroid of the membrane shell elements remained reasonably constant. This means that wrapping the exposed faces of the solid elements, that model the adhesive, with membrane shell elements is an effective way to recover the peel stresses from the solid elements, with the advantage of making the stresses of interest reasonably mesh insensitive. 3. Fatigue parameters derivation through physical testing The performance of hybrid joints is strongly influenced by several factors, such as adherend material type and thickness, type of adhesive, surface preparation of the adherends, presence of adhesive spew and its amount and shape, as well as the type, size, distribution and quality of the mechanical joints. Therefore, for more accurate fatigue analyses, it is important to derive bespoke fatigue parameters of the hybrid joints by testing specimens that are representative of the production parts and processes. In this Section, a test-based process to derive bespoke fatigue parameters of the hybrid joints is proposed. Lap shear and coach peel are the most common specimen geometries used when testing hybrid joints. More complex specimen geometries can be tested if required, however, these are usually more difficult and expensive to test. When deriving fatigue parameters for hybrid joints it is essential to choose the most appropriate failure criterion, or in other words to appropriately identify the number of cycles to failure. Three possible failure criteria are:  Specimen separation . When the test stops either because the specimen is physically separated or because a maximum allowable displacement, set to avoid damaging the test equipment, is reached. This last condition can leave the specimen physically intact, but with significant cracks. In both cases, the specimen is considered ‘separated’ for fatigue purposes. This failure criterion is the simplest choi ce, as no further data analysis effort is needed. However, it is inherently non-conservative.  Crack initiation . This can be detected in different ways (for example from visual inspection of a video recording of the test). The detection of crack initiation is usually a time-consuming process and is not a trivial exercise due to the very small adhesive thickness. Furthermore, the interpretation of crack initiation could depend on the capability and opinion of the person carrying out the analysis and therefore, it is not a consistent and repeatable process. Finally, using crack initiation as the failure criterion might be over-conservative.  Stiffness drop . This is based on the variation of the specimen stiffness throughout the test duration, as described below in more detail, and is calculated based on displacement data captured during the test.