Title: Studies of the fibre direction and local bending stiffness of Norway spruce timber − for application on machine strength grading
Subject: Building Technology
Faculty: Faculty of Technology
Date: Thursday 1 February 2018 at 10.00 am
Location: Room N1017, building N, Växjö
External reviewer: Professor Mats Ekevad, Lulea University of Technology, Sweden
Examining committee: Professor Geir Vestøl, Norwegian University of Life Sciences, Norway
Professor Karl-Gunnar Olsson, Chalmers University of Technology, Sweden
Associate Professor Eva Frühwald Hansson, Faculty of Engineering LTH, Sweden
Chairman: Professor Sigurdur Ormarsson, Department of Building Technology, Linnaeus University
Supervisor: Professor Marie Johansson (main supervisor), Professor Anders Olsson and Professor Charlotte Bengtsson, Department of Building Technology, Linnaeus University
Examiner: Professor Marie Johansson, Department of Building Technology, Linnaeus University
Spikning: Thursday 11 January 2018 at 10.00 am at the university library in Växjö
Machine strength grading is a production process in the sawmill industry used to grade sawn timber boards into different strength classes with specific characteristic values of the bending strength, modulus of elasticity (MOE) and density. These properties are called grade determining properties. Each of these is predicted on the basis of a statistical relationship between the property and a so-called indicating property (IP), which is based on non-destructively assessed board properties. In most cases, the prediction of strength is crucial for the grading. The majority of commercial grading machines rely on a statistical relationship of strength to an IP, which is either a global dynamic MOE or an averaged flatwise bending MOE measured over a board length of about one meter. The problem of today's machine strength grading is that the accuracy of the strength prediction is rather poor with a coefficient of determination of about R2 ≈ 0.5 − 0.6. One consequence of this is that much of the strength potential of timber is unused.
The intention of this research is to contribute to a long-term goal, which is development of a method for prediction of bending strength that is more accurate than the methods available today. The research relies on three hypotheses. First, accurate prediction of bending strength can be achieved using an IP that is a localized MOE value (determined over a short length) that represents the lowest local bending stiffness of a board. Second, knowledge of the local bending stiffness with high resolution along a board's longitudinal direction can be established on the basis of fibre direction within the board in combination with dynamic MOE. Third, fibre directions in the interior of a board can be determined by application of fibre angle models utilizing data of fibre directions on the board's surfaces obtained from tracheid effect scanning. Following these hypotheses, this work has included laboratory investigations of local material directions, and development of models for fibre directions of the interior of boards. The work also included application of one-dimensional (1D) analytical models and three-dimensional (3D) finite element models of individual boards for the mechanical behaviour, analysis of mechanical response of boards based on experiments and based on the suggested models. Lastly, the suggested models were evaluated by comparisons of calculated and experimentally determined local bending stiffness along boards, and of predicted and experimentally determined bending strength.
The research contributes with in-depth knowledge on local fibre directions close to knots, and detailed information on variation of the local bending stiffness in boards. Moreover, fibre angle models for fibre directions in the interior of boards are presented. By application of the fibre angle models in the 3D model of the whole board, the local bending stiffness along timber boards can be determined over a very short length (l < 50 mm). A comparison with results determined on an experimental basis show a very close similarity implying that the applied models are sufficient to capture the variation of local bending stiffness, caused by knots and fibre distortions, with very high accuracy. Furthermore, it is found that by means of IPs derived using the suggested models, bending strength can be predicted with high accuracy. For a timber sample comprising 402 boards, such IPs results in coefficient of determination as high as R2 = 0.73. However, using IPs based on the 3D finite element model did not improve the R2 value achieved when using the IPs based on the 1D model.