Issue 35

A. Tzamtzis et alii, Frattura ed Integrità Strutturale, 35 (2016) 396-404; DOI: 10.3221/IGF-ESIS.35.45 401 Cyclic hardening The cyclic tests performed in [14] and reproduced in Fig. 5 show that cyclic strain hardening occurs in the overaged materials A250 and A300. Cyclic strain hardening after overaging is more pronounced compared to the base metal (T3) as obtained from the cyclic strain hardening exponent n’ values in Tab. 1. Parameter n’ increases with increasing overaging temperature (A300 material). This is associated with the HAZ location close to the weld center. The dependency of increasing cyclic strain hardening with increasing crack closure was investigated in [14]. Here, this dependency is used to analyze the crack growth behavior of A250 and A300 materials using the different cyclic hardening characteristics (value n’) in Eq. (7) to obtain an estimation of the respective behavior of the related positions in the HAZ in Fig. 3. Yield strength σ 0.2 (MPa) Cyclic yield strength σ c0.2 (MPa) Strain hardening exponent n Strength coefficient H (MPa) Cyclic strain hardening exponent n΄ Cyclic strength coefficient H΄ (MPa) T3 375 445 0.120 694 0.042 576 A250 245 245 0.161 604 0.148 568 A300 135 185 0.242 594 0.211 674 Table 1 : Cyclic and monotonic properties of T3, A250 and A300 materials. C R ACK GROWTH ANALYSIS rack closure under cyclic loads is influenced by the cyclic hardening mechanism at the crack tip. In [14-17] it was analytically and experimentally demonstrated that increased cyclic hardening contributes to increased crack closure and retards fatigue crack growth. In Eq. (7), the cyclic hardening exponent n’ was used to analyze the local fatigue crack growth behavior of the various positions in the HAZ (A250 and A300). In the analysis, the material properties included in parameter B (Eq. 10) for all materials have been taken from the reference, base metal, 2024 T3 alloy and are presented in Tab. 2. The material volume length r c was fitted from the 2024 T3 data. In Fig. 6 the analytical results obtained using a numerical integration of Eq. (7) for T3 (base metal), A250 (mid-point HAZ gradient) and A300 (HAZ/TMAZ) materials are compared against experimental crack growth rates. Analytical and experimental crack growth rates agree well for the stress intensity factor range examined (12-25 MPa m 1/2 ), specifically for the position associated with A250, while for the case of A300 (HAZ/TMAZ interface) a small overestimation in predicted crack growth rates exists. Cyclic hardening exponent n΄ Parameter r c (m) Coffin- Manson exponent c Coffin- Manson parameter ε f ΄ critical stress intensity factor K cr (MPam 1/2 ) [23] Young’s modulus E (GPa) Poisson ratio ν T3: 0.042 A250: 0.148 A300: 0.211 1.291x10 -6 -0.6 0.13 60 73.1 0.33 Table 2 : Material parameters used in Eq. (7) for FCG analysis. In Tab. 3 the average value of parameter r c for the three materials T3, A250 and A300 is given. In Fig. 7 the analytical results are compared against the experimental crack growth rates using in the analysis the mean value of r c in order to evaluate the sensitivity of the analysis on the fitting parameter. The differences in this case are negligible, which shows that analytical predictions are not influenced by the value of r c . C