A study was conducted with the primary objective of examining the efficacy of delamination test using cylindrical core specimens to assess the bond quality of cross laminated timber (CLT) products. A prototype coring drill bit was fabricated to prepare a cylindrical-shaped specimen, the height of which corresponds to the full thickness of the CLT panel. A secondary objective was to examine the effect of pressure, adhesive type, number of plies, and specimen shape on the delamination resistance of CLT panels. The wood material used for the CLT samples was Select grade nominal 1 x 6-inch Hem-Fir boards. Examples of three adhesive types were evaluated, which were designated as A, B, and C. The delamination tests used were as described in CAN / CSA O122-06 and EN 302-2.
Cylindrical specimen extracted as core was found satisfactory as a test specimen type for use in delamination testing of CLT product. Its efficacy was comparable to that of a square cross-section specimen. The former is recommended as it can be extracted from thicker panels and from any location in the panel. It would also be more convenient to plug the round hole.
Adhesive type had a strong effect on delamination resistance based on the two delamination tests used. Adhesive A exhibited the greatest delamination resistance, followed in decreasing order, by adhesives C and B. It should be noted that no effort was made to find the optimum CLT manufacturing parameters for each type of adhesive. Therefore the relative rankings of the adhesives tested may not be representative. However, for the purposes of this study, the different performance levels from the three adhesives are useful in providing insight into how the proposed delamination test responds to significant changes in CLT manufacturing parameters.
Pressure used in fabricating the CLT panel showed a strong effect on delamination resistance as demonstrated for one of the adhesives. Delamination resistance decreased with decreasing pressure. The effect of the number of plies in the CLT panel was dependent upon the type of adhesive, and this was probably related to the adhesive’s assembly time characteristic. These results provide support as to the effectiveness of delamination test in assessing the moisture durability of CLT panels. It was able to differentiate the performance in delamination resistance among different types of adhesives, and able to detect the effect of manufacturing parameters such as pressure and increased number of plies in CLT construction.
The test procedure described in CAN / CSA O122-06 appears to be reasonable in the delamination resistance assessment of CLT panels for qualification and quality control testing. Based on the results of the study along with some background information and guidelines, delamination requirements for CLT panels are proposed. The permitted delamination values are greater than those currently specified for laminated and fingerjoined lumber products. This is in recognition of the higher bond line stresses when bonded perpendicular laminations (i.e. CLT) are exposed to the delamination wetting and drying cycles, as opposed to parallel laminations (i.e. glulam or fingerjoints).
Forintek has completed a two-year investigation of the NLGA SPS 6 Standard, Special Products Standard for Structural Face-Glued Lumber. The NLGA SPS 6 Standard prescribes product specifications and qualification and quality control requirements for structural products created by edge-gluing and/or fingerjoining lumber segments. Under the NLGA SPS 6 Standard, the design values assigned are based on the visual grade and the stress level achieved in qualification tests on the glue joints.
The project assessed the effect of the following three factors on strength of the NLGA SPS 6 product:
1. Tension proof-loading;
2. Relative location of fingerjoints in adjacent members when fingerjoined material is edge-glued;
3. Strength of the material used to make the NLGA SPS 6 product.
Results showed a positive effect of proof-loading, a minor effect of staggering of fingerjoints, and a highly significant effect of density of raw material on tensile stress of edge-glued specimens. It was confirmed that SPS6 products of greater commercial value can be obtained from lower grade lumber. However, visual grading of SPS 6 products proved to be more difficult than visual grading of lumber, because grade-determining wood characteristics were sometimes hidden in the bond line, and could not be properly identified.
The findings of this project can be used to fine tune the NLGA SPS 6 standard and the other NLGA fingerjoint and face-glued lumber product standards. The project will help the wood industry maximize the utilization of their raw material resource, resulting in increased profitability.