PSI - Issue 5

Jan Raška et al. / Procedia Structural Integrity 5 (2017) 225–232 Author name / Structural Integrity Procedia 00 (2017) 000 – 000

229

5

To input the information about the crack path, the user has to edit a text file. At the first line, the FE model file name, the crack path and the other data must be defined, as is shown on the figure 6.

Beginning& End boundaries of the crack (i.e. the crack will not necessarily open and end AT these nodes, but BETWEEN them)

Type of input file

Name of input file

CRAC2Dmaterial

CRAC2D ID number

Fig. 6. Input text data – crack path definition [3]

Moreover, the user has to specify the crack growing states (defined by the crack tip nodes), where he request the stress intensity factor computing. In this pre-defined crack tip nodes, the procedure described in the chapter 2 is applied: The crack is modeled by “element separation”, the 4 elements, neighboring the crack tip, are removed and the CRAC2D element is created. For some next applications, there is useful to save the FE model with the crack, but without applied CRAC2D element. For this reason, the user has to define the model output according to the figure 7:

Fig. 7. Input text data – output definition [3]

In the example given above (figure 7), the software will create a total of 4 output FE models (4 text files) [3]:  1 with the crack tip at node 1003, with only basic shell elements  1 with the crack tip at node 1005, with a 0.5-thick CRAC2D element at the tip  2 with the crack tip at node 1007 : 1 with shell elements only; 1 with a 0.5-thick CRAC2D. After the execution of the main software procedure, the group of the models, including the requested crack opening, with and without CRAC2 element, is generated. The file organization structure and the model name convection is shown on the figure 8. In additions to the models, the batch file for the successive execution of models in NASTRAN is generated. Also, the crack lengths are computed. After the execution of all the models, the last procedure can start. The software reads in all the NASTRAN output files (.f06 files) and links the result stress intensity factors – K I and K II – with the crack opening. The final result is the stress intensity factor as a function of crack length. In fact, two discrete functions are defined by the software output data table (example figure 9): K I vs. crack length and K II vs. crack length.