PSI - Issue 18

Mehdi Mokhtarishirazabad et al. / Procedia Structural Integrity 18 (2019) 457–471 M. Mokhtarishirazabad / Structural Integrity Procedia 00 (2019) 000–000

469 13

3.5. Failure Assessment Diagram The fracture data of the plane-sided samples were analysed using Option 1 in a Failure Assessment Diagram (FAD) (R6: Assessment of the Integrity of Structures Containing Defects). Although using the Option 2, can reduce conservatisms in Option 1, both options result in similar curves for materials with high strain hardening capacity like austenitic stainless steel. The failure assessment line, K r and L r are defined as follows: � � � � � �� � ��� � � � ���� ���3 � ��� �������� �� �� (5) � � �� ��� � � �� � �� � ���� � � �� � � ��� ��� � � � ��� ���� ���� (6) � � �� � �� ������ �� � ���� � ���� � � � ������� � ��� � ������� �� � � �� � ��� � ��� ��� � (7) where � ����� ��� �� � � � � 2 , � � ��22��� � � � �, � � , � � √ 2 3 (8) K mat is defined as follows (Anderson 2017): ��� � � ��� �� � � � � � � (9) and K I is calculated based on Murakami (Murakami 1987). Here we used J Q (measured after applying linear scale correction for crack extension) as the J 0.2 . The FAD for samples with a/W = 0.5 and two different thicknesses of 10 mm and 5 mm are shown in Fig. 17. The sample with 20 mm thickness was not plotted due to considerable error in crack extension measurement by UC. Each point in this figure corresponds to the load just before the unloading in the fracture test. The highlighted marks in this figure show the corresponding load at J 0.2 . The main point of this figure is that the samples failed by plastic collapse much earlier than any crack growth occurred. That is uncracked ligament failed during crack tip blunting. 4. Conclusion The fracture toughness of Stainless Steel 316L has been assessed by ASTM standards, with crack extensions estimated by the compliance method. It has been found that:  The compliance method significantly under-predicted crack extension, particularly for plane-sided thick samples. Two correction methods were used to improve the UC estimation, but both of these resulted in inconsistent estimations by variation of the thickness and initial crack length of samples.  By decreasing the sample thickness (loss of out-of-plane constraint) and the initial crack length (loss of in-plane constraint), the estimated value of J Q also decreases. This was contrary to expectations. The main reason for this observation was the significant underestimation of the crack extension when using compliance method. Therefore, when considerable deformation is expected for fracture test, the equations used in compliance method for samples would need to be modified and validated by a FE analysis.