PSI - Issue 2_A

3

Abhishek Tiwari et al. / Procedia Structural Integrity 2 (2016) 690–696 Author name / Structural Integrity Procedia 00 (2016) 000–000

692

The fracture specimens were pre-cracked to obtain width normalized crack length in the range of 0.2 to 0.6. After carrying out the fracture tests in three point bending, the broken specimens were observed for pre-crack length measurement using 9 point average method under low magnification stereo microscope. The specimens after measurement of pre-crack length, were categorized according to their a/W values in ranges of 0.2-0.3±0.04, 0.4±0.04, 0.5±0.04, 0.6±0.09, as shown below.

Group 1 (Gr-1) = 0.19-0.34 Group 2 (Gr-2) = 0.35-0.44 Group 3 (Gr-3) = 0.45-0.54

   

Group 4 (Gr-4) = 0.55-0.69 The fracture tests were performed in an environmental chamber at -110 o C, -120 o C, -130 o C, -140 o C and -150 o C. The temperature was maintained with ±1 o C accuracy by recirculating liquid N 2 in the environmental chamber fixed with the universal testing machine. The test data in these categories were analyzed using MC approach. 3. Numerical Analysis To understand the effect of crack depth on the volume sampled ahead of the crack tip under the maximum principal stress dominance, numerical analyses were performed on a/W of 0.3, 0.4, 0.5, 0.6 and 0.7. The geometry used for modelling was 5×10×55mm 3 sub-size charpy specimen with 2×2×2 reduced integration and incremental loading algorithm. The tensile properties were used for -110 o C obtained from 4mm diameter and 20mm gauge length round bar tensile specimen. The quarter symmetric geometry modelled is shown in Fig. 1.

Fig. 1 Quarter symmetric charpy geometry with active volume at crack front

The numerical analysis was post processed for the σ*- V * behavior investigation. The cleavage failure is described in this approach by the volume ahead of the crack tip encompassed under maximum principal stress σ 1 higher than a threshold. The threshold value is generally taken to be twice of yield strength σ YS . The details of calculation involved in σ*- V * approach can be found in earlier work by Tiwari et al. (2015). Another post processing was carried out to define a new constraint parameter derived from Weibull stress approach of micro-mechanical cleavage model. The triaxiality or constraint has been described in numerous forms such as, elastic T stress , Q -stress and q . The parameter ‘ q ’ is described as the ratio of equivalent stress σ e and hydrostatic stress component σ h . This ratio is known to be affecting the damage in the material resulting in void