Measurement of Crack Arrest Fracture Toughness of a Ship Steel Plate

Abstract Two types of full thickness compact crack arrest (CCA) specimens have been used to characterize the crack arrest toughness of a 15.5-mm-thick ship plate steel in both the L-T and T-L orientations. One was a modified version of that used in ASTM Standard E 1221, and the other was based on the proposed procedure of Crosley and Ripling [1] for non-plane strain testing. The latter specimens employed a strain age embrittled chevron notch as the crack starter. The testing was performed over the temperature range from −40° (the nil-ductility transition temperature (NDTT) of the steel) up to −5°C. None of the results from the ASTM E 1221 type specimens met the validity requirements of the ASTM procedure. Even though they had the largest dimensions allowed in the test specification, the limited thickness of these specimens meant that they were too small for valid plane strain testing. In the non-plane strain test procedure, the in-plane dimensions are scaled to the expected crack arrest toughness at the test temperature, and the main qualifying requirement is for the crack to be arrested at a point in the specimen lying within certain defined limits. Although difficulties were encountered with variability in run/arrest behavior, this specimen gave some useful crack arrest toughness data for this steel, particularly with regard to the relatively high values obtained when testing near the top of the test temperature range. The effectiveness of the crack starter technique used in both type of specimens is also discussed, as well as possible modifications for future tests.