0 , a partir da metodologia da curva mestra também fornece um suporte adicional para o uso de corpos de prova SE(B) não padronizados como uma alternativa geométrica nos procedimentos rotineiros de avaliação de tenacidade à fratura, incluindo o uso de corpos de prova de pequenas dimensões nas medições de tenacidade quando a disponibilidade limitada de material e a capacidade de carregamento da máquina são as principais preocupações.Palavras-chave: fratura por clivagem, corpo de prova SE(B) não padronizado, corpo de prova PCVN não padronizado, efeitos de restrição, temperatura de referência, curva mestra ABSTRACT This work addresses a numerical-experimental investigation on the cleavage fracture behavior of an ASTM A572 Grade 50 high strength, low alloy structural steel using standard and nonstandard SE(B) specimens, including a non-standard PCVN configuration. The main purpose of this study is to contribute to the development of a fracture toughness test procedure applicable to bend geometries with varying specimen span over width ratio ( / ) and loaded under 3-point and 4-point flexural configuration. The secondary purpose, but none the less important, is to investigate the effects of geometry and loading mode on experimentally measured fracture toughness data distributions and implications for the characterization of the temperature dependence of toughness based on the Master Curve methodology. Based on these purposes, a large new set of plastic -factors applicable to these non-standard bend geometries which serve to estimate the experimentally measured toughness values in terms of load-displacement records, including the -integral and the crack tip opening displacement ( ), is provided. In order to facilitate contact with other test protocols, a new set of rotational factors, , to determine the based on the plastic hinge model is also described. Fracture toughness testing conducted on various bend geometries with varying specimen span ( / ) extracted in the T-L orientation from an A572 Grade 50 steel plate provides the cleavage fracture resistance data in terms of the -integral at cleavage instability, . The experimental results show a potential effect of specimen geometry and loading mode on -values which can help mitigating the effects of constraint loss often observed in smaller fracture specimens. An exploratory application to determine the reference temperature, 0 , derived from the Master Curve methodology also provides additional support for using nonstandard bend specimens as an alternative specimen geometry in routine fracture assessments, including the use of small-scale fracture specimens to measure fracture toughness properties when limited material availability and test machine capacity are of major concern are of major concern.