Nonlinear shear properties of spruce softwood: Numerical analyses of experimental results

Determination of material parameters from experimental tests often rely on simplifying assumptions like the existence of uniform stress and strain fields within the considered part of the test specimen. However, more detailed analyses usually show that the stress and strain fields differ from the assumed (nominal) uniform distributions. In order to utilize the potential of numerical analyses of wooden structures by the FEM method, the nominal material parameters measured directly from tests need to be re-evaluated in order to make them more useful for FEM models and to make FEM models more reliable. Experimental data from shear testing of clear wood from Norway spruce was analysed numerically with a bilinear material law in shear. The inherent material parameters were fitted to the experimental behaviour by means of optimization methods in conjunction with FEM analyses. The study included six Arcan test configurations comprising the three orthotropic material planes of wood, and covered the whole loading range until failure. Compared to numerical results, it was found that stiffness values measured were too high, and that downward adjustments in the range of 5-30% were required. Linear limit stresses between 40% and 60% of the nominal shear strengths were found, whilst the tangent moduli ranged between 30% and 70% of the linear elastic shear moduli. The rolling shear plane RT showed most nonlinearity and the LT plane least. Analyses with modified bilinear parameters showed good correspondence with experimental findings. The parameters were found to be relatively well adapted by Weibull distributions. (C) 2009 Elsevier Ltd. All rights reserved.