A single specimen method was developed to predict the fracture toughness behavior for steels in the transition. The prediction is based on a single specimen estimate of lower bound toughness in the transition and order statistics. A fracture toughness from a single test is subjected to a lower bound estimate to generate a single lower bound value. This value is transformed into a fictitious distribution of toughness values using order statistics. From this a median value of toughness, KJc(med), is determined and hence To where To is the temperature at which the master curve reaches a median toughness of 100 MPa√m. With the determination of To the entire distribution of toughness in the transition is determined.
The single specimen method of To determination was evaluated for transition toughness tests where a large number of fracture toughness values were available at a single temperature. Single specimen values of To were compared with the To from the entire population of fracture toughness values and from random selections of six toughness values taken one at a time. The result of the evaluation shows that the single specimen could estimate To to within a reasonable value if the test temperature was close to To. Compared with the result of a random selection of six values to determine To the single specimen method gave similar results.