Page 308 - Handbook of Materials Failure Analysis
P. 308

304    CHAPTER 12 A nonlocal damage-mechanics-based approach




                                    400
                                            1TCT (P =0.05)
                                                 f
                                   Fracture toughness K JC  (MPa m)  200  2TCT (P =0.50)
                                            1TCT (P =0.50)
                                                 f
                                            1TCT (P =0.95)
                                                 f
                                    300
                                            2TCT (P =0.05)
                                                 f
                                                 f
                                            2TCT (P =0.95)
                                                 f
                                            4TCT (P =0.05)
                                                 f
                                            4TCT (P =0.50)
                                                 f
                                            4TCT (P =0.95)
                                                 f
                                    100
                                      0
                                           –100     –80      –60      –40      –20
                                                        Temperature (°C)
                         FIGURE 12.19
                         Effect of specimen size (1T, 2T, and 4T) on fracture toughness variation in DBTT regime of CT
                         specimen.
                         6.9 EFFECT OF SPECIMEN SIZE ON PROBABILITY OF CLEAVAGE
                         FRACTURE IN THE DBTT REGIME
                         The effect of specimen size on the fracture toughness variation in the DBTT region
                         was also studied. Figure 12.19 presents the predicted results of fracture toughness
                         variation (with temperature and probability of fracture) for the 1T, 2T, and 4T CT
                         specimens, respectively. The 2T and 4T CT specimens are scaled versions of 1T
                         CT specimens with scale factors of 2 and 4, respectively. It can be observed from
                         Figure 12.19 that the scatter in the fracture toughness decreases for the specimens
                         with larger dimensions, signifying the increase in the fracture toughness transition
                         temperature. Hence, specimens with larger dimensions are more susceptible to the
                         DBT for a given value of loading. This is in line with the experimental observations.



                         6.10 ANALYSIS OF SEB SPECIMEN WITH CRACK AT
                         THE FERRITE-BUTTERING INTERFACE
                         The standard dimensions of SEB specimen are same as discussed in Section 5 and these
                         are shown in Figure 12.2a. The schematic representation of crack at the ferrite-buttering
                         interface of the dissimilar metal welded joint is shown in Figure 12.20a. The FE mesh of
                         the SEB specimen with different material properties (Table 12.7) at different regions
                         and initial crack at the ferrite-buttering interface is shown in Figure 12.20b.The full
                         specimen is meshed with FEs as we have different materials at both sides of the crack
                         and hence the symmetric conditions cannot be used.
                            The load versus CMOD and J-resistance behavior of the SEB specimen is obtained
                         from the analysis and compared with those of experiment in Figures 12.21 and 12.22,
   303   304   305   306   307   308   309   310   311   312   313