Evaluation of the shear size effect in glued laminated timber using a stochastic FE model calibrated on 17000 glue-line tests

Abstract

Glue-line tests provide the shear resistance and failure mode of samples extracted from laminated timber products. Several papers questioned the practical use of such tests affirming the lack of correlation between the capacity of the beam and the outcomes of glue-line tests. However, no studies provide a probabilistic framework for interpreting glue-line tests. This paper presents and discusses the largest database of glue-line tests ever published, consisting of approximately 17000 tests on glued laminated timber (GLT) beams, carried out at the Treteknisk, the Norwegian Institute of Wood Technology (Norway). This database has been used to develop in Abaqus a stochastic FE model to predict the shear capacity of glued laminated timber beams (GLT) using advanced modelling concepts, including discrete cracking and plasticity. The authors examine the dependence of the shear capacity on the beam size to assess the shear size effect in GLT components. The methodology is based on Montecarlo simulations. The mechanical properties of the elementary volumes composing the beam are assigned using a random field model calibrated on the glue-line tests.