PSI - Issue 28

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

1026

3

– ductile (Gurson-Tvergaard-Needleman or Roussilier) model to predict fracture. Reasonable prediction of K 1c values were obtained but the accuracy of the model needed to be improved. Alternative to the Beremin model is the Richie Knott-Rice (RKR) criterion in which cleavage fracture occurs when the stress (over a specified length) exceeds a critical stress. Nam el al. (2018) utilized a coupled Gurson-RKR model to calculate a Strain Energy Density (SED) criterion for predicting fracture in offshore structures. These models considered the focused stress at a notch/crack. Though the current work only considers the stress/strain state at the crack front in the Charpy test, any brittle-ductile criterion would also be expected to be applied to a wider range of loading as typical in impact events. 2. Material and Experimental Results Tensile and Charpy specimens were machined from a section of TC128B extracted from a tank car built and qualified in 2015. All tensile specimens had a gauge length of 25.4 mm and diameter of 6.35 mm with the gauge length oriented in the longitudinal (rolling) direction of the steel. Charpy specimens were machined in both the longitudinal and transverse orientations. An instrumented pendulum Charpy machine with maximum available energy of 750 J was used in the testing. Quasi-static tensile tests were performed between 24 °C and -80 °C. The engineering stress-strain curves are shown in Figure 1. As consistent with steel testing, the strength of the steel increased with decreasing temperature. Elongation and reduction of area remained very similar in the temperature range. Lüders strains were in the range of 2-2.9% and increased slightly with decreasing temperature.

Figure 1: Quasi-Static engineering stress vs. strain response of TC128B at various temperatures

Charpy tests were performed at temperatures between 24 °C and -80 °C with the measured CVN energy as a function of temperature as shown in Figure 2. The initial velocity corresponds to a dynamic strain-rate of about 100 s -1 . Longitudinal CVN values were higher than transverse values at temperatures above -60 °C. This anisotropic fracture behaviour was consistent with observation of elongated inclusions along the rolling direction. It was not attributed to plastic anisotropy which was measured to be small based on longitudinal and transverse tensile specimens. Modelling of anisotropic fracture is not addressed in the current work but should be studied further in future work. The load versus displacement response at each temperature is shown in Figure 3. Ductile fracture is dominant in the high CVN range, whereas cleavage fracture is dominant in the low CVN range.