Redesign of an industry test for hot tearing of high performance aluminium casting alloys using casting simulation software

Hot tearing propensity in aluminium alloys is commonly measured using dog-bone (or "I" beam) and ring tests. Hot tearing occurs as a result of a number of factors including; level of stress and strain, hot spots and nucleation sites. This paper presents the results of a study to redesign a dog-bone type hot tear test using casting simulation software to ensure that the location of the tearing was always in the same location to improve the reproducibility of the test. In the simulation of the original five fingered die both the stress and strain were sufficiently high for hot tearing but there was no defined hot spot implying that the random hot tear locations would result depending upon suitable nucleation sites. A number of design iterations were carried out to produce more focussed hot spots and to ensure that the die was easy to manufacture and use, and was economically viable.