PSI - Issue 28

V.V. Sudin et al. / Procedia Structural Integrity 28 (2020) 1637–1643 Sudin V.V./ Structural Integrity Procedia 00 (2019) 000–000

1642

6

The presence of a ductile crack under the notch when the pendulum moves about 0.5 mm indicates the start of its growth immediately after the specimen reaches the stress corresponding to the yield point on the load-displacement curve Fabry et al. (1993). The reason for the differences in the processes of metal fracture with different microstructures may be a significant difference in the systems of cleavage triggers Chen and Cao (2017). In particular, large non – metallic inclusions brittle particles of silicates, aluminosilicates and oxides-can act as the main triggers in welded joints, while for modern steel containing a low amount of harmful impurities, the main triggers can be large ferrite grains surrounded by small grains Cao et al. (2014). Early cleavage can significantly reduce the reliability of welded joints, not only by reducing the impact strength, but also by incorrectly assessing their real ability to resist brittle fracture. In the case of laboratory experiment, as reason of stopping of brittle crack can act exhaustion of pendulum energy and consequent slowing of its movement or deformation at the edge of the specimen opposite the notch. At the same time, in real structures, the energy reserve is several orders of magnitude higher than that in impact bending tests and the origin of a brittle crack during loading is likely to lead to the destruction of the structure due to the different stress strain state. To register the early formation of a brittle crack, it is not necessary to analyze dynamic fracture curves, which may be complicated by the need to use an instrumented tester. The shown correlation between the displacement of the pendulum to the formation of a brittle fracture and the distance from the notch root to the cleavage facets can be used to estimate the energy spent on the initial growth of a ductile crack and act as an additional standard value. At the same time, replace the measure of the ductile cracks depth by value of shear fracture area impossible due to the low correlation of the values of the share of ductile fracture and the depth of the ductile cracks for specimens with a large energy final fracture. For example, for specimens of welded joints № 4 and № 5, and steel 17G1S, showing a Pearson correlation coefficient less than 0.85 in the figure 5.

Fig. 5. Correlation of the values of the shear fracture area of the specimen and the depth of the initial ductile crack.