PSI - Issue 47

3

Teresa Morgado et al. / Procedia Structural Integrity 47 (2023) 882–887 Author name / Structural Integrity Procedia 00 (2019) 000 – 000

884

Fig. 1. SENB specimen.

For the first study, established in table 1, five models were created, as shown in table 3. The elements type used was the first order, CPS4, with linear interpolations and second order, CPS8, with quadratic interpolations (Abaqus, 2014). The CPS4 elements are quadrilateral elements of four bi-linear nodes, and CPS8 are quadrilateral elements of eight bi-linear nodes. As shown in table 3, the number of elements and nodes decreases as the crack length increases because when the crack is smaller, a large number of elements are needed to transition between the refined zone (crack edge) and the coarse zone (off-edge). Fig. 2 can be observed four meshes for different crack locations.

Table 3. Models used in 2D study 1.

Length [mm]

Element Type

Number of elements

Number of Nodes

1.5

CPS4

1570

1656

3

CPS4

498

547

5

CPS4

926

979

7

CPS4

576

623

10

CPS4

344

382

Fig. 2. Mesh for crack location at (a) 1.5 mm; (b) 3mm; (c) 5mm; (d) 10mm.

All simulations were performed using the implicit method (ABAQUS/Standart) because the conventional method (Contour Integral) cannot be performed using an explicit method. This method also assumes that the type of analysis is static, and therefore a simulation time of 0.1 seconds was imposed. Since this type of analysis is not time-dependent, the results and the number of increments tend to be the same even if one increases the simulation time. For each increment, the program has required the program to provide Integral J and FIT values for Six contours. In Fig. 3, only contours One, Two and Six are exemplified.