Abstract
This paper describes the adaptability of Hillerborg fictitious crack model to timber. Using the regression line between the cumulative loss of the potential energy and the fictitious splitting length from the test data, the fracture process zone length was defined the length between the origin and x-intercept of the regression line, and the fracture energy was defined the slope of the regression line. The strength calculated by the fracture mechanical model on the specimen assuming initial splitting length equal to the fracture process zone length were agreement with the experimental results. However, the strength calculated by the fracture mechanical model using the conventional assuming initial splitting length equal to zero, was dangerously overestimated with the experimental results.
