PSI - Issue 28

B.W. Williams et al. / Procedia Structural Integrity 28 (2020) 1024–1038 Author name / Structural Integrity Procedia 00 (2019) 000–000

1033

10

The MMC-PW damage model coefficients given in Table 3 yielded poor agreement with the measured quasi-static response. As mentioned above, this was not unexpected as Charpy tests were not used in the model calibration. An alternate set of damage model coefficients that results in better agreement with the quasi-static Charpy response measured at room temperature is given in Table 4. The initiation and evolution failure surfaces for this new set of parameters is shown in Figure 9a. These model parameters would need to be verified under different loading conditions (e.g. notched tension, shear, biaxial, SEB, SET and DWTT). Figure 9b compares the damage evolution ( D = D c ) curves between the two room temperature parameter sets given in Table 3 and Table 4. The curves are only for a normalized Lode angle of unity. The plot indicates that the trend with stress triaxiality for the new parameter set is maintained compared to the original MMC-PW model. These model coefficients at 24 °C and 100 s -1 were used as a starting point for the low temperature parameters.

Table 4: Quasi-static Damage Model Parameters for the Current TC128B at Room Temperature

� � ����� � � �

Strain-Rate (s -1 )

A (MPa)

n

m

0.001

112.87

0.180

0.145

62.94

0.969

1.0

2.6

0.05

a)

b)

Figure 9: Quasi-static failure response for TC128B at 24 °C; a) damage initiation and evolution surfaces and b) comparison of damage evolution curves for two different sets of MMC model coefficients; plotted for normalized Lode angle of unity

5.2. Dynamic Response at Multiple Temperatures without Brittle Fracture The predicted and measured dynamic response at each temperature is shown in Figure 10. For each temperature, the peak load predicted is in reasonable agreement with the measured response and the gradual drop in force is somewhat captured by the model for the ductile responses. However, brittle fracture (indicated by the sudden drop in force) was not well captured by the MMC model. In all cases, the MMC predicted a gradual drop in force with increasing displacement and could not capture the sudden force drop. The MMC damage model coefficients are given in Table 5. The parameters A and n were based upon tensile data given in Table 2. C 3 defines the influence of shear on the fracture and for the final parameter was not modified from the value of 0.969 reported by Paredes et al. (2018). The damage evolution curves at each temperature at a normalized Lode angle of unity are compared in Figure 11. All the curves have the expected trend of decreasing failure strain with increasing stress triaxiality. Also, there is a good trend with decreasing temperature. At -80 °C a ductile fracture curve was calculated which is significantly less than the other temperatures, but only brittle fracture was active at this temperature, so the ductile response is not of major consequence.

Made with FlippingBook Ebook Creator