13th International Conference on Fracture June 16–21, 2013,Beijing, China -3- machined with the rolling direction parallel (T-L) and perpendicular (L-T) to the notch. The specimens were fatigued in air and at room temperature. The FCGR test were conducted in an Instron‘s servo-hydraulic machine with closed loop to computers for automatic test control and data acquisition. The crack extension was measured with a krak-gages® technique, which is principally an indirect DC potential measurement procedure. The krag-gages® theory is explained in reference [16]. For instance, the crack gauge KG-B20, whose full scale is 20 mm, has a sensitivity of measurement to crack extension that is better than 0.02mm [17]. The crack extension was also measured using the compliance technique by means of a clip gage in the crack mouth and a strain gauge fixed in the back surface. The crack closure measurementwere made by using a sampling rate of 400 data pairs (load and displacement) per cycle according to some of the recommendations made by Song et al. [18]. The signal noise was reduced by using low-pass digital filters [19]. The determination of the opening force was done by comparing slope segments of 10 percent of the load- displacement data with the linear region of the load-displacement curve. The fatigue tests were performed at a frequency of 20 Hz. Three different values of load ratio R (=σmin/σmax) 0.1, 0.3, 0.5, 2 different stress level at the same load ratio, and 3 decreasing ∆K at different load ratio were employed. For ∆K decreasing test the rate of force shedding with increasing crack size was according to the ASTM standard (20). The details of the experiments are listed in table 3. The martensite transformation around the crack tip was observed by optical microscopy. The material was ground in the surface where the fatigue crack was going to appear with SiC emery paper up to a roughness of 1200 grit and then polished. Because of the mechanical grinding can induce martensitic transformation,the material was electro-polished with a solution consisting of 5vol% perchloric acid and 95% ethanol at 45V for 15 s. The martensitic phase was revealed by chemical etching in a solution of 100 ml ethanol, 20 ml HCl, 1.5 g K2S2O5 and 2 g NH4F·HF. The electro-polished was used until no peak (100) (200) (211) (α`martensite peak according to M. Karimi et al [21]) was found by X-ray diffraction. Table 3.Details of test configuration Specimen R (σmin/σmax) Load level [N] (Pmax- Pmin) Notch [mm] Test Type w [mm] C [mm-1] 1 NL2 0.5 2625 8.729 Constant amp. load 35 - 2 NL3 0.1 2625 8.600 Constant amp. load 35 - 3 NL4 0.1 1815 13.817 Constant amp. load 35 4 NL6 0.3 1890 14.020 Constant amp. load 35 - 5 NL7 0.7 2625 8.927 Constant amp. load 35 - 6 NL9 0.1 variable 9.777 ΔK decreasing 40 -0.08 7 L3G 0.5 variable 11.55 ΔK decreasing 40 -0.065 8 NNL4 0.7 variable 9.088 ΔK decreasing 35 -0.055 9 NNL7 0.7 variable 8.897 ΔK decreasing 35 -0.09 10 NNL1 0.5 variable 8.912 ΔK decreasing 35 -0.08
RkJQdWJsaXNoZXIy MjM0NDE=