ANALYSIS OF LOADING RATE EFFECTS ON CLEAVAGE FRACTURE TOUGHNESS OF FERRITIC STEELS

摘要

The effects of loading rate on the Weibull stress model for prediction of cleavage fracture are examined in this paper for a low-strength pressure vessel steel (A515-70). We focus on low-to-moderate loading rates (K_I ≤ 2,500 MPa {the square}m/s). Tregoning and Joyce tested a large number of 1T SE(B) specimens for this material with different a/W ratios (0.15, 0.55) at several loading rates over this range. They also conducted comparative, quasi-static tests using 1T C(T) specimens and shallow-cracked SE(B) specimens (a/W＝0.2). We describe very detailed, 3-D finite element analyses of these specimens employing rate-sensitive material flow properties characterized by a viscoplastic constitutive model with uniaxial, tension stress-plastic strain curves specified at varying plastic strain rates. To quantify the probability of cleavage fracture, we adopt the three-parameter Weibull stress model as modified earlier by the authors to bring the microscopic model into better agreement with the macroscopic fracture toughness distribution adopted in ASTM E1921. The analyses described here examine dependencies of the Weibull stress parameters on K_I. The study shows that the Weibull modulus (m) remains reasonably rate independent at constant temperature over the range of loading rates considered for this material. Rate dependencies of the scaling parameter (σ_u) and the threshold parameter (σ_(w-min)) can be computed using the calibrated m, and the results indicate these dependencies are not overly strong. However, the predicted cumulative probability for cleavage exhibits a strong sensitivity to the calibrated value of σ_u, analogous to the strong sensitivity of K_0 to loading rate. Consequently, use of a calibrated σ_u value based on static tests leads to significant errors in the predicted cumulative failure probabilities for dynamic loading.