An assessment of some methods of obtaining fracture toughness values from invalid test data

Abstract

Several methods used in estimating the fracture toughness from data obtained in a single invalid test are discussed. The methods were, Witt's equivalent energy, Rice's formula, Chell and Milne's load/displacement curve fitting procedure and a new empirical technique called the pseudo-elastic load/displacement curve fitting method. Using published computed results and an analytical model, a quantitative investigation of the accuracy of each method is made. For deeply-cracked specimens ( c/ w = 0.7) which failed in the post-yield regime, all the methods gave results within 10% of the known solutions. For shorter cracks ( c/ w < 0.5), Rice's method underestimated and Witt's method overestimated toughness values by 10%, whilst the other two methods were still accurate to within 10%.