Quality of Structural Timber - End-User Requirements and Performance Control

The performance and quality of structural timber products cannot solely be defined by mechanical properties. It is the totality of features of a product that determines its ability to satisfy the customer. This thesis covers three aspects within the context of structural timber quality: Quality requirements for structural timber products. Performance specifications have been established for a number of structural timber products based on analysis of end-user requirements. The work has been focused on the demands related to constructability (ease of construction). Performance control of short-term stiffness using dynamic modal tests. Dynamic modal tests have been carried out and evaluated for prediction of short- term (static) modulus of elasticity of deep and short timber beams. Influence of raw material parameters on geometrical performance (warp). The spatial variation within trees with respect to propensity to warp has been studied for fast-grown Norway spruce. The aim has been to form the basis for optimization of log conversion with respect to geometrical performance. Requirements in terms of geometrical and mechanical properties differ substantially between end-uses of structural timber. Therefore, a product-oriented classification system is needed both for the optimal utilization of timber and to assure future supply of suitable timber products for the building industry. The current grading rules for timber do not comply with the end user's requirements regarding warp. Furthermore, the two general timber grading systems of today do not utilize the great potential that lies in the variability of the raw material in relation to the various end-uses. The geometrical performance of sawn timber products, especially with respect to twist, must be improved if timber is to keep its traditional position as a building material. It has been shown that it is possible to accurately predict the static bending stiffness of structural timber beams with a length to depth ratio as low as 5 using dynamic modal tests and Timoshenko beam theory. The applicability of Euler beam theory, however, should be questioned for length to depth ratios lower than 18 for free-free bending vibrations, since the static bending stiffness will be underestimated by more than 5%. Twist and crook of Norway spruce timber were found to be larger for pith-associated wood. Twist decreased much more rapidly with the distance from pith compared with crook. Crook was larger in the butt log than in the top log while twist and bow appeared not to vary longitudinally. Heartwood formation in the juvenile core of the butt log did not diminish warp propensity. Twist was highly correlated to the distance between pith and centroid of the stud cross section. The ratio of grain angle to pith distance was the best parameter for twist prediction. Presence of compression wood increased bow and crook significantly.