This paper describes the results of an experimental and analytical study of the effect of specimen size on the fracture toughness behavior of A36 steel, primarily in the lower-transition region. This region is beyond the limits of plane strain as defined by ASTM Method E 399, but in the region where failure occurs prior to any macroscopic fibrous tearing. Experimental results were obtained using crack-tip opening displacement (CTOD) three-point bend specimens, and analytical results were obtained using a 3-D elastic-plastic finite-element analysis. Two specimen sizes, full thickness [31.8 mm (1.25 in.)] and reduced thickness [12.7 mm (0.50 in.)], were investigated using the preferred t × 2t × 8t specimen size.

Testing was conducted according to the procedure outlined in BS 5762:1979, Methods for Crack Opening Displacement Testing. CTOD-temperature transition curves were developed for both specimen sizes. A sufficient number of specimens were tested at two temperatures in the lower-transition region to allow statistical characterization of the CTOD results. Both normal and Weibull distributions were used for that characterization.

Three-dimensional elastic-plastic finite-element analyses were conducted to determine the effect of specimen size on the near-crack-tip opening stresses through the specimen thickness. The near-crack-tip stress fields were compared at various CTOD levels well into the elastic-plastic regime, but prior to any stable crack growth. At each of the applied CTOD levels, the maximum near-crack-tip opening stress occurred at the specimen centerplane and was found to be essentially independent of specimen size.

The results of this study showed that the lower bound of the elastic-plastic CTOD toughness values of A36 steel for small specimens is essentially equal to those of larger specimens in the lower-transition region prior to any stable crack growth. Thus it may be possible to characterize the notch toughness of low-strength structural steels, such as A36 steel in the lower-transition region, using small specimens provided that a sufficient number of specimens are tested. However, additional study is required before any recommendations can be made.