Abstract Because of the larger specimen dimensions and more significant ligament for crack growth, the drop-weight tear test (DWTT) has become more widely used than other impact tests. On the other hand, very little information is available from DWTT experiments on weld metal (WM). In the present study, in addition to the standard chevron notch (CN) DWTT specimen with a notch depth (ND) of 5.1 mm, API X65 specimens with 10- and 15-mm ND were tested on the WM and also the base metal (BM). Fracture energies versus the ND were plotted, based on which tests can be performed successfully with lower-capacity DWTT machines to predict the absorbed energy for each ND, by specimens with increased ND and reduced absorbed energy. The standard requires thickness reduction for specimens thicker than 19 mm. With the method proposed in the present study, high-grade specimens can be tested with their original thickness. In the DWTT, regardless of specimen materials (BM or WM), about 44 % of the absorbed energy is spent on crack initiation, and 56 % is spent on crack growth. In all specimens, the BM absorbed energy was more than that of the WM. Therefore, this study proposes that the API standard, which does not include DWTT on the WM, should require performing DWTT on WM specimens. Hence, the results of the current study can form the basis for further studies that can initiate the development of new standards on the WM.