Mechanical properties of as-printed 316L stainless steel thin-walled structures obtained by directed energy deposition are investigated. In-situ tensile and fracture tests are performed on small samples obtained from a additively manufactured square section tube and extracted with three different orientations with respect to the part build direction. Despite a strongly oriented microstructure resulting from the process, as-printed specimens exhibit a reduced anisotropy in comparison with thick or polished samples commonly reported in the literature. Moreover, it is shown using a simple model that the reduced dentified anisotropy can be explained by considering the material thickness variation pattern only, resulting from the layer stacking process. Fracture tests are analyzed using an adapted digital image correlation procedure that evaluates the specimen fracture toughness from experimentally computed J-integrals. Using time reversal, strain fields in regions close to the crack path are identified. Stress fields are then computed from the constitutive behavior identified in tensile tests. A regularization procedure is proposed to enforce the stress equilibrium. Finally, the J-integral is computed using various integration contours in order to validate its path-independance. On this basis, a nearly isotropic fracture toughness is identified. Additional scanning electron microscope observations show that fracture surface features are independent from specimen orientation. This apparent isotropy is explained by the isotropic distribution of lack-of-fusion defects driving crack initiation and propagation.
The High Resolution Wavevector Analysis (HRWA) is presented and its application illustrated. Extending the High Resolution Wavenumber Analysis method [1] to 2D signals, it allows the wide-band and local characterization of the linear elastic behavior of anisotropic plates. The method belongs to the family of experimental wavenumber-based characterization methods and uses the high resolution signal processing algorithm ESPRIT (Estimation of Signal Parameters via Rotational Invariance Techniques) and the ESTER criterion (ESTimation of ERror) to overcome some of the limitation of Fourier-based methods. Three experimental applications on composite plate specimens are presented. First (i), from the out-of-plane velocity field of a sandwich plate with a foam core, different wave types (bending, shear and compression) are extracted. The results are compared with numerical predictions. Second (ii), individual layer contributions are separated on a honeycomb sandwich plate by means of the observation of the dependence of the extracted complex wavevectors as a function of wave propagation direction and frequency. Third (iii), a local wavenumber extraction is performed on a 4-layer carbon-epoxy plate made of fiber patches with spatially varying orientations. The local specific bending stiffness of the plate is identified from the extracted wavevectors and compared with theoretical results.
The High-Resolution Wavenumber Analysis (HRWA) is presented. It identifies complex wavenumbers and amplitudes of waves composing the harmonic response of a beam. Based on the frequency dependence of these wavenumbers, experimental dispersion branches corresponding to various beam motions (e.g bending, torsion) can be retrieved. The HRWA method is compared to the Mc Daniel and the Inverse Wave Correlation (IWC) methods. It overcomes some drawbacks of these methods: the wavenumber resolution is enhanced. Also, the wavenumber search problem is expressed as a linear problem, making the method computationally efficient. A number of wavenumbers can be identified automatically, thanks to a statistical criterion. First, the noise sensitivity of each method is investigated in the basis of synthesised measurements. For this criterion, the HRWA and Mc Daniel method performances are close and much better than IWC. Moreover, the HRWA is five to twenty times faster to compute than other methods, depending on the mesh size. Second, an experimental case is presented where bending and torsion waves are identified, yielding an apparent viscoelastic Young and shear moduli on a wide-frequency range.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.