POTENTIAL OF SHORT KNOT-FREE BOARD SECTIONS FOR GLUED LAMINATED TIMBER PRODUCTION

摘要

Considering the strength of small clear wood specimens enormous strength potential can expected for full-size and defect-free board sections. Based on this, the research project presented here was initiated in cooperation with Schaffer Holz Tirol GmbH in order to evaluate the suitability of short knot-free board sections for the production of glued laminated timber (GLT) of the highest strength classes according to EN 1194. The raw material, a side product of scantlings for the window frame production, is sawn from the mature part of large diameter timber of European Larch (Larix decidua Mill.) and European Silver Fir (Abies alba Mill.). The knot-free, optically graded quarter and flat-sawn board sections range from 250 mm to 1,100 mm in length and 24 mm in thickness. The board sections are finger jointed in length to form glulam lamellas, which are then assembled via an adhesive to form the product PURE-LAM® (trade name).The main differences between PURE-LAM® and standard glued laminated timber (GLT) are the short lengths and the small thicknesses of the board sections when compared to standard GLT. Due to the rejection of all visually detectable defects, especially knots, it is expected that variation within and between tension strengths of the board sections is low compared to standard GLT.The aim of this project was the evaluation of the suitability of short knot-free board sections of the two wood species for the production of GLT of the highest strength classes according to EN 1194, namely GL32h and GL36h. Furthermore, the physical and mechanical potential of the raw material were investigated by means of tension tests according to EN 408 and EN 1194.PURE-LAM® beams were tested in four-point bending according to EN 408 to verify and confirm the assumptions. Moreover models should be verified by the test data.Tension tests were performed on four lamella series (two with finger joints in the free testing length and two without) for each wood species. Dimensions for tension specimens were approximately 1 x w x t = 1,600 mm x 94 mm x 24 mm and for bending samples 1 x w x d = 5,000 mm x 90 mm x 310 mm. Due to the short length of board sections free testing length was set to 200 mm. Deformation of bending specimens was measured in the load range of 10 kN to 50 kN and 6 kN to 40 kN corresponding to 10% and 50% of the estimated bending strengths for Larch and Fir, respectively.To quantify the influence of annual ring orientation on finger joint and timber strength, specimens were produced from flat-sawn and quarter-sawn board sections. Bending specimens were only produced from quarter-sawn specimens. Additional five beams made of Larch were produced and tested after the analysis of the first test series for the evaluation of the suggested production changes.Moisture samples were taken as close to the area of failure as possible to determine moisture content according to EN 13183. Density was determined for each specimen on the basis of dimensions and weight of the whole specimen for bending samples and on moisture samples for tension specimens. Density was corrected to the reference moisture content of 12% based on EN 384Length was determined for all board sections in the top layers and in core lamellas of three Fir and four Larch beams in order to gain information about length distribution of board sections in PURE-LAM® beams and to verify the production parameters. Despite the fact that at least the first 250 mm of timber specimens in the left and right clamping area remained without finger joints many specimens failed in a finger joint or in the wood within the clamping area due to the high strengths of the timber in the free testing length. Only those timber specimens and those finger joint specimens that showed a failure in the free testing length and in the finger joint in the free testing length were taken into account for calculation of timber and finger