Issue 38

S. Fu et al, Frattura ed Integrità Strutturale, 38 (2016) 141-147; DOI: 10.3221/IGF-ESIS.38.19 142 one concern in the design of stents. The low level ratcheting under biaxial loading is observed and simulated by means of macro-scale based kinematic hardening rules of the Ohno-Wang (O-W) and Chen-Jiao-Kim (C-J-K). E XPERIMENTAL PROCEDURE micro tension-torsional fatigue apparatus for thin wires, as shown in Fig. 1, is applied in this study. The coupled tensile and torsional loading at the scale of 10 0 N-10 2 μNm is achieved by a tensile and a torsional load frame that actuated and measured independently without interfering with each other. A wire specimen is clamped between the frames that have been aligned by an x-y translation stage. A linear motor connected with a load cell is applied in the tensile frame, which allows for the precise control of axial force and the free axial movement of specimen. The micro-torque of the specimen in test is loaded by a DC micro motor and measured by a high precision torque transducer through the transference of a thrust air bearing. The axial deformation of the wire specimen in tension-torsion tests is measured by a grating sensor built in the linear motor and corrected based on data acquired by a non-contact displacement detection system (NDDS) in uniaxial tension tests. More details about this apparatus can be found elsewhere [3]. Figure 1 : Schematic illustration of the tension-torsional fatigue apparatus for micro-scale components. The material used in this study is 316L stainless steel soft tempered wire, in nominal diameter of 100 μm, annealed with pure hydrogen. The chemical composition of the material is (wt%): C 0.03%, Cr 16.93%, Ni 12%, Mn 0.95%, P 0.63%, S 0.21%, Mo 2.39%, Si 1%. The actual diameters of specimens were measured before testing by using a scanning electron microscope (SEM). The gauge lengths of the specimens were denoted as the distance between the edges of clamps, which was set to be around 6 mm for all tests. Uniaxial tension tests and torsion tests were conducted to derive the basic mechanical properties. The cyclic tension-torsion tests were conducted at room temperature under load control for axial loading and angle control for torsional loading. The loading paths in the axial stress-shear strain plane (σ-γ plane) are illustrated in Fig. 2, and test conditions are given in Tab. 1. In all cases the shear strain amplitude is 0.55%. As the strains in this study are small and consist of low level of plasticity, the shear stress is derived with the hypothesis of linear distribution in the cross section. The shear strain and stress are calculated as follows: D l    (1) T D 3 16    (2) where  is the shear strain,  is the shear stress,  is the twisting angle, T is the corrected torque, and D and l are the diameter and gauge length of wire specimen, respectively. In addition, a completely reserved torsional test with shear strain amplitude of 2% was performed to characterize the isotropic softening behavior of the material. A minimum axial stress of A