Avaliação de concentradores de tensão em madeiras sólidas furadas

Abstract

Holes in wood are necessary for structural applications, wherein they become stress concentrators, reducing the strength and stiffness of the wooden part, altering the flow of stresses and creating a probable region to the occurrence of failure. The present study aimed to mechanically characterize solid wood samples drilled with three different hole diameters. The study material consisted of young trees from Pinus elliottii, which were selected at random in Piratini/RS. Prismatic samples were made with dimensions of 50 × 50 × 200 mm3 (largest dimension oriented in the longitudinal direction) for compression parallel to the fibers, 50 × 50 × 760 mm3 (largest dimension oriented in the longitudinal direction) for static bending and 50 × 50 × 63 mm3 (largest dimension oriented in the tangential direction) for the tensile perpendicular to the fibers, all following the ASTM D143 standard. 4 mm, 8 mm and 12 mm center holes were drilled using a bench drill. These diameters were chosen because they are common in structural wood applications. Numerical simulations were performed using AbaqusTM finite element software, using orthotropic elastic properties evaluated in experimental tests of compression parallel to the fibers and elastic ratios indicated in the literature in order to calibrate the properties in the numerical model. The statistically analyzed compression results showed that the influence of the holes was insignificant. However, the onset of longitudinal crack failures was apparently influenced by holes with diameters above 4 mm. In addition, it was possible to prove through numerical modeling that the maximum stresses occurred in the hole region, which can be considered a region with a high incidence of damage. In the place chosen between indenter and support for the holes in the samples tested in static bending, through experimental and numerical tests, it was possible to prove that the holes did not change the damage mechanisms, regardless of the hole size. It was possible to compare the stiffness of the samples from stress vs. deformation graphs for compression and force vs. displacement graphs for the static bending, which were numerically and experimentally obtained. The compressive graphs were calibrated and remained similar within the elastic limit. In the bending test, compared to the experimental ones, it was possible to notice a greater stiffness in the curves obtained by simulation, which was attributed to the geometry of the samples proposed in ASTM D143. In this sense, a correction of the stiffness levels based on a relationship proposed in the literature was able to better match the numerical and experimental results.