In situ tomographic study of a 3D-woven SiC/SiC composite part subjected to severe thermo-mechanical loads.

A high-temperature multi-axial test is carried out to characterize the thermo-mechanical behaviour of a 3D-woven SiC/SiC composite aeronautical part under loads representative of operating conditions. The sample is L-shaped and cut out from the part. It is subjected to severe thermal gradients and a superimposed mechanical load that progressively increases up to the first damage. The sample shape and its associated microstructure, the heterogeneity of the stress field and the limited accessibility to regions susceptible to damage require non-contact imaging modalities. An in situ experiment, conducted with a dedicated testing machine at the SOLEIL synchrotron facility, provides the sample microstructure from computed micro-tomographic imaging and thermal loads from infrared thermography. Experimental constraints lead to non-ideal acquisition conditions for both measurement modalities. This article details the procedure of correcting artefacts to use the volumes for quantitative exploitation (i.e. full-field measurement, model validation and identification). After proper processing, despite its complexity, the in situ experiment provides high-quality data about a part under realistic operating conditions. The influence of the mesostructure on fracture phenomena can be inferred from the tomography in the damaged state. Experiments show that the localization of damage initiation is driven by the geometry, while the woven structure moderates the crack propagation. This study widens the scope of in situ thermo-mechanical experiments to more complex loading states, closer to in-service conditions.

Data Merging of AE Sensors with Different Frequency Resolution for the Detection and Identification of Damage in Oxide-Based Ceramic Matrix Composites.

In this paper, acoustic emission data fusion based on multiple measurements is presented for damage detection and identification in oxide-based ceramic matrix composites. Multi-AE (acoustic emission) sensor fusion is considered with the aim of a better identification of damage mechanisms. In this context, tensile tests were conducted on ceramic matrix composites, fabricated with 3M™ Nextel™ 610 fibers and aluminosilicate matrix, with two kinds of AE sensors. Redundant and complementary sensor data were merged to enhance AE system capability and reliability. Data fusion led to consistent signal clustering with an unsupervised procedure. A correlation between these clusters and the damage mechanisms was established thanks to in situ observations. The complementarity of the information from both sensors greatly improves the characterization of sources for their classification. Moreover, this complementarity allows features to be perceived more precisely than using only the information from one kind of sensor.