In Europe, strength classes for structural timber and glulam lamellae are defined by minimum requirements of characteristic values of the grade determining properties (GDPs). To fulfill these minimum requirements of characteristic values in the daily production at sawmills, indicating properties (IPs) to GDPs are calculated for each board and based on predetermined limits of the IPs (settings) boards are assigned to the graded class, or rejected. The aims of this paper is to address and discuss

more »

... two different grading procedures/models that can be applied when settings for IPs that reflects a local board property are derived and to show how the yield in different T-classes depend on the model applied. It is not always that a board's weakest cross-section is evaluated in a destructive test. An IP representing a local board property can therefore be determined either as the lowest property of the tested part of the board or as the lowest property along the whole board when applied to derive settings. Results presented in this paper show that too low settings and too large yields are obtained when the latter IP is employed. Similarly, IPs reflecting a global board property, like axial dynamic MOE, also give too low settings and too high yield in strength classes. This paper is the second and closing part of a series of two paper on prediction of GDPs and procedures for grading sawn timber into T-classes. Keywords Grading of timber Á Laser scanning Á Fibre direction Á Dynamic modulus of elasticity Á Norway spruce List of symbols E a,90,nom Lowest local axial modulus of elasticity (MOE) of the destructivly tested part of the board, calculated on the basis of observed fibre directions E a,90,nom,test Lowest local axial MOE of the destructivly tested part of the board, calculated on the basis of observed fibre directions, i.e. E a,90,nom,test = E a,90,nom E a,90,nom,whole Lowest local axial MOE of the whole board, calculated on the basis of observed fiber directions E b,90,nom Lowest local bending MOE of the destructivly tested part of the board, calculated on the basis of observed fibre directions ().,-volV) (0123456789().,-volV) E b,90,nom,test Lowest local bending MOE of the destructivly tested part of the board, calculated on the basis of observed fibre directions, i.e. E b,90,nom,test = E b,90,nom E b,90,nom,whole Lowest local bending MOE of the whole board, calculated on the basis of observed fibre directions E dyn,12% Axial dynamic MOE at a moisture content of 12% E t,0,12% Grade determining tensile MOE parallel to grain E t,0,12%,test Grade determining tensile MOE parallel to grain i.e. E t,0,12%,test = E t,0,12% E t,0,12%,whole Tensile MOE parallel to grain of the weakest cross-section of the whole board E t,0,mean Mean characteristic MOE parallel to grain f t,0,h Grade determining tensile strenght parallel to grain f t,0,h,test Grade determining tensile strenght parallel to grain, i.e. f t,0,h,test = f t,0,h f t,0,h,whole Tensile strenght of the weakest cross-section of the whole board f t,0,k 5-percentile characteristic tensile strength parallel to grain IP E,a Indicating property (IP) for prediction of the grade determining properties (GDPs), derived by means of multiple linear regression using E a,90,nom,test and E dyn,12% as predictor variables and the investigated GDP as dependent variable IP E,b IP for prediction of the GDPs, derived by means of multiple linear regression using E b,90,nom,test and E dyn,12% as predictor variables and the investigated GDP as dependent variable IP E,b,f,1 (u,v) IP for prediction of strength, derived by means multiple linear regression using E b,90,nom,test and E dyn,12% as predictor variables and f t,0,h,test as dependent variable IP E,b,f,2 (u,v) IP for prediction strength, derived by means multiple linear regression using E b,90,nom,whole and E dyn,12% as predictor variables and f t,0,h,test as dependent variable q 12% Grade determining density q k 5-percentile characteristic density q s,12% Board density at a moisture content of 12% r 2 Coefficient of determination