The selection of appropriate constitutive models is critical for accurately predicting the mechanical behaviour of timber in structural applications. This study evaluates the influence of material model selection on the tensile response of timber using a combined single-element simulation and experimental approach. A controlled single-element framework was employed in LS-DYNA to isolate the constitutive behaviour of MAT-22, MAT-54, MAT-143, and MAT-261 under uniaxial tension parallel to the grain. All models were calibrated using a consistent elastic modulus and mean experimental strength. The numerical results showed identical elastic stiffness and peak stress across all models, while significant differences were observed in the post-peak response. Experimental stress–strain curves exhibited a predominantly brittle to quasi-brittle behaviour with minimal post-peak softening. Among the evaluated models, MAT-54 demonstrated the closest qualitative agreement with the experimental response, whereas continuum damage models overpredicted ductility. The findings highlight that material model selection has a limited influence on elastic response but significantly affects failure prediction, with important implications for the structural modelling and design of timber elements.
Partha et al. (Tue,) studied this question.