3.1.1 Bending properties of GLT, CLT, and ply-lam CLTs by static test
The MOE and MOR values obtained for the GLT, CLT, and ply-lam CLTs are summarized in Table 4. The changes in the out-of- and in-plane MOE values with various loading directions were dependent on the type of material. The CLT and ply-lam CLTs exhibited significant out-of-plane MOE values, whereas the GLT showed a high in-plane MOE value, which is similar to the value reported by Li et al. (Li et al. 2022). The plywood used in CLTs and ply-lam CLTs showed a low in-plane MOE value because of the load along the cross section of the lumber. Compared with the value obtained for the GLT, the out-of-plane MOE values (F*** = 9.019) for three ply-lam CLTs (i.e., P-A, P-B, and P-C) were improved by 11.5–29.5%. However, the MOE value of the P-D ply-lam CLT was 8.2% lower than that of the GLT. Additionally, the MOE values obtained for the three ply-lam CLTs were 11.6–23.4% higher than those for CLT. When larch lumber was used as the outer layer of ply-lam CLTs, higher MOE values were obtained, particularly for P-A and P-C.
Wang et al. (Wang et al. 2015) studied the bending properties of CLT using a LSL and pine lumber. When the second layer was changed to LSL, MOE and MOR increased by 1,248 MPa and 8.58 MPa, respectively. Comparing P-A and CLT, the MOE and MOR increased by 2,257 MPa and 9.63 MPa, respectively. When LSL and plywood were applied as component materials of CLT, the increase in MOE and MOR in the fiber direction was consistent.
The in-plane MOE value obtained (F*** = 7.803) of GLT was the highest, followed by that of ply-lam CLTs and CLT. The average MOE values obtained for the GLT, ply-lam CLTs, and CLT were 13.5, 10.9–12.1, and 8.9 GPa, respectively. There were no significant differences in the MOE values found among the ply-lam CLTs in the significance test at 5% level (F = 1.566, p = 0.229). Thus, we concluded that the location of plywood unaffected the measured in-plane MOE values.
Table 4
Bending properties of the GLT, CLT, and ply-lam CLTs.
Panel type | Out-of-plane | In-plane |
MOE (MPa) (F*** = 9.019, p = 0.000) | MOR (MPa) (F*** = 12.848, p = 0.000) | MOE (MPa) (F*** = 7.803, p = 0.000) | MOR (MPa) (F*** = 10.025, p = 0.000) |
GLT | 12,157ab (12.4%) | 56.56d (21.7%) | 13,502c (16.3%) | 53.98c (9.3%) |
CLT | 11,839ab (13.3%) | 39.66a (17.6%) | 8,860a (3.5%) | 34.13a (17.0%) |
Ply-lam CLTs | P-A | 14,096cd (9.7%) | 49.29c (14.7%) | 11,307b (5.0%) | 42.47b (7.3%) |
P-B | 13,210bc (11.7%) | 40.91ab (10.0%) | 12,142bc (8.2%) | 42.79b (7.1%) |
P-C | 14,612d (13.6%) | 46.37bc (14.5%) | 12,037bc (9.3%) | 42.9b (11.2%) |
P-D | 11,212a (12.6%) | 37.81a (11.2%) | 10,888b (13.4%) | 43.51b (11.6%) |
Wang et al. (Wang et al. 2015) (laminated strand lumber (T) and pine lumber (L)) | LLL (∥-⊥-∥) | 9,727 (7.37%) | 35.37 (9.85%) | | |
LTL (∥-∥-∥) | 10,975 (6.19%) | 43.95 (9.94%) | | |
The GLT and CLT showed higher out-of-plane MOR values than the ply-lam CLTs. The MOR values among the ply-lam CLTs were different, and the out-of-plane MOR value was high when lumber was used as the outer layer. Meanwhile, the in-plane MOR value was high when plywood was used as the outer layer. The out-of-plane MOR value (F*** = 11.957, p = 0.000) of GLT was the highest, followed by those of ply-lam CLTs and CLT. The trend for the out-of-plane MOR at 5% lower limit was GLT (34.4 MPa) > P-B (32.4 MPa) > P-A (32.2 MPa) > P-C (31.6 MPa) > P-D (29.2 MPa) > CLT (26.1 MPa), which was similar to the average of MOR values.
The in-plane MOR value (F*** = 9.99, p = 0.000) of GLT was the highest, followed by those of ply-lam CLTs and CLT. The significance test across the studied ply-lam CLTs exhibited no differences in their in-plane MOR values at the same level (F = 0.056, p = 0.982). The in-plane MOR values at 5% lower limit for P-A, P-B, P-D, P-C, GLT, and CLT were 34.9, 33.9, 32.4, 31.8, 43.2, and 21.7 MPa, respectively. The difference in the in-plane MOR values for the studied ply-lam CLTs was 3.1 MPa. When plywood was utilized as a layer, the homogeneity of the material was guaranteed because the coefficients of variation for the ply-lam CLTs were lower than those for GLT and CLT.
On the basis of the MOE and MOR values, we believe that the MOE and MOR values obtained for P-A and P-C can be improved. Additionally, if plywood is used as the outer layer (e.g., P-B and P-D), we predict uniform bending characteristics. Specifically, the MOE and MOR values of P-D were similar or greater than those of regular CLT. Jang and Lee (Jang, and Lee 2019) maximized the use of small larch logs by designing a CLT panel that consisted of bonded plywood and glued laminated board made from small-diameter lumber as the outer and inner layers, respectively. They concluded that the P-D method is acceptable for the fabrication of structural panels from low-mechanical-grade lumber and small-diameter logs (Jang, and Lee 2019).
Niederwestberg et al. (Niederwestberg et al. 2018) compared the bending properties of LSL and lumber (L) as CLT (5-C1), 5-B2 (LSL∥-L∥-LSL∥-L∥-LSL∥), and 5-A1 (L∥-LSL∥-L∥-LSL∥-L∥). The bending moment and stiffness of 5-B2 were the highest, followed by those of 5-A1 and 5-C1. In this study, the bending properties of P-C were the highest, followed by those of P-B and CLT. This is judged to be the difference between plywood and LSL applied as a layer. LSL is a composite structural lumber consisting of oriented wood stands (Wang et al. 2015), and plywood is glued together with adjacent layer having their wood grain rotated up to 90º to one another (Shreeranga et al. 2017); thus, plywood has lower bending properties than LSL. Accordingly, it is judged that when plywood is used as the outer layer, it acts as a factor impairing the bending properties.
The failure modes were analyzed after the out-of- and in-plane bending tests, and the results are shown in Fig. 5. The out-of-plane failure modes were classified into tensile, tensile + shear, and shear to highlight the differences in the modes depending on the compositions of the specimens. All the GLT specimens failed during the bending tests, and the CLT, P-A, and P-C specimens exhibited tensile + shear failure. The P-B specimen exhibited tensile + shear and shear failure values of 13 ea and 2 ea, respectively. When plywood was used as the outer layer, the RF occurred between the veneers, which are arranged at the intersection that inhibited strength. The P-D specimen showed tensile, tensile + shear, and shear failure values of 9 ea, 4 ea, and 2 ea, respectively. When lumber was used continuously across two to four layers (P-D type), the tensile failure mode of the P-D specimen was identical to that of the GLT specimen.
The changes observed in a failure mode was closely related to the MOE and MOR values. We found that the ratio of shear failures existing in the material had an influence on the degradation rate of the bending qualities. Therefore, we concluded that the use of a layered material as the outer layer can ensure that the specimen would have better bending properties than plywood. When plywood was used as the outer layer to enable the use of low-mechanical-grade larch lumber, the specimen showed uniform bending properties. However, to avoid failure under shear of the outermost layer, it is necessary to consider both the thickness and shear properties of the layer.