Abstract:

Objective: This study aims to explore the influence patterns of knots on the longitudinal compressive strength of Pinus koraiensis and Larix olgensis, and establish a predictive model, so as to provide a theoretical basis for the safety design of timber structures. Method: The size of the longitudinal compressive strength specimens was 40 mm (radial) × 40 mm (tangential) × 100 mm (longitudinal), with the largest knot positioned in the middle of the specimen as much as possible. Knot size and volume ratios were measured in accordance with GB/T 29897—2013. Longitudinal compression tests were performed following GB/T 28993—2012, and specimen failure modes were recorded according to ASTM D143—25. Based on the actual specimen size, a half symmetrical model was used for finite element simulation. The specimens with dead knots were subjected to surface-to-surface hard contact to simulate interface separation and slippage, while the specimen with live knots was subjected to shared-node connections to reflect fiber continuity. The stream line approach was used to simulate the complex deflection behavior of wood fibers around knots and predict the longitudinal compressive strength. Result: 1) The average longitudinal compressive strength of Pinus koraiensis specimens with knots (all single live knots) was 20.49±4.69 MPa, which was 23.66% lower than that of the specimens without knots (26.84±4.82 MPa). The average longitudinal compressive strength of Larix olgensis specimens with knots (all single live knots) was 30.68±5.82 MPa, which was 26.21% lower than that of the specimens without knots (41.58±8.29 MPa). In terms of failure types, knot-free wood predominantly exhibited type Ⅲ and type Ⅵ failures, while knotted wood mainly showed type Ⅲ failures, along with mixed failure types combining type Ⅲ and type Ⅵ. 2) In Pinus koraiensis specimens with knots, the measured knot sizes ranged from 0.10 to 33.00 mm, divided into three grades: T1 (0.10?4.00 mm), T2 (4.00?10.00 mm), T3 (10.00?33.00 mm). The average longitudinal compressive strengths were 22.39, 20.86, and 17.87 MPa, respectively. There were significant differences in the longitudinal compressive strengths between T3 and both T1 and T2 (P_1,3<0.001, P_2,3=0.021<0.05). In knotty Larix olgensis specimens, the measured knot sizes ranged from 0.18 to 18.75 mm, divided into three grades: T4 (0.18?4.00 mm), T5 (4.00?10.00 mm), T6 (10.00?18.75 mm), with average strengths of 35.35, 28.59, and 26.22 MPa, respectively. There were significant differences in the longitudinal compressive strengths between T4 and both T5 and T6 (P_4,5=0.001<0.05, P_4,6<0.001). The influence of knot volume ratios on longitudinal compressive strength followed the same trend as knot size. 3) The coefficient of determination (R²) between the grain deflection angle near knots and the longitudinal compressive strength was 0.82399 for P. koraiensis and 0.83056 for L. olgensis. The slopes of the linear regression equations were ?0.26334 and ?0.34501, respectively. 4) In the predictive model, the failure type of specimens containing dead knots was manifested as cracking along the knot-matrix interface, and as the grain deflection angle near dead knots increased, the crack gradually deviated from the straight-grain direction. the failure type of specimens containing live knots depended on the modulus of elasticity (MOE) difference between knot and the wood matrix. Conclusion: The presence of knots significantly reduces the longitudinal compressive strength of both P. koraiensis and L. olgensis and leads to mixed failure modes. The increase in the knot size and volume ratio reduces longitudinal compressive strength, with L. olgensis exhibiting a more pronounced decline. Fiber deflection angle near knots shows a negative correlation with longitudinal compressive strength for both species. The longitudinal compressive strength of L. olgensis decreases more significantly as the grain deflection angle near knots increases. The failure mechanisms induced by dead knots and live knots are distinct. In specimens with dead knots, failure primarily occurs at the interface between the dead knot and the wood matrix, and the crack direction is influenced by grain deflection. In specimens with live knots, the failure mode depends on the relative magnitude of the MOE between the live knot and the wood matrix.

Key words: knot, longitudinal compressive strength, type of failure, experimental method, finite element method