The influence of structural dimensions on crack arrest

An energy balance has been used for the prediction of toughness values required to ensure arrest in double tension crack arrest tests. The analysis required calculation of the total energy that depended on the force applied to the main plate and the effective length of the structure including the test frame. Theoretical calculations indicated that the energy release rate at crack arrest was governed by both the width and the length of the structureGs=σ12/E(1/πa+a/WL)where σ1 is the initial stress; a is the crack length; L is the effective length that, in cylinders, is the circumference; W is the width; E is the elastic modulus. ent was obtained between the measured and calculated values of the gross section stress at arrest derived from the above analysis. The data were explained in terms of a static analysis so that the kinetic energy did not require consideration in the calculations. The mean dynamic fracture energy per unit area, Rs, derived from small impact tests was sufficiently close to Gs to enable prediction of arrest crack lengths for all materials. The agreement was improved by including an estimate of the effect of the force from the secondary crack initiation jack on the elastic energy available. he expression Gs values are higher in structures with large W and/or L dimensions. A conservative approach to crack arrest would be to ensure that, for all structural dimensions, the dynamic toughness exceeds the conventional value of G for a structure of infinite size, having no finite size factorGs=σ12πa/E.