Verifikasi Empiris Persamaan Lentur Statis Glued Laminated Timber (Glulam)
Abstract
INTRODUCTION. Wood in construction is often loaded by bending moment. Since the development of wood engineering increasing, the solid wood tends to be substituted by wood’s engineering such as glulam. Glulam is usually made from some laminas. A Transformed Cross Section method has widely used to calculate the mechanical properties of glulam if each lamina’s properties are known. The method transforms each lamina based on its MOE. The longitudinal width is decreased if the MOE is lower, and increased if the MOE is higher. However MOE is a material property which is independent from the geometric property, calculated MOE based on geometric properties is not accurate. Bahtiar (2008) was developed a new methods to calculate the bending properties of glulam. The method is mathematically proven. However it would be better if empirical experiment is conducted to prove it. METHOD. Materials used in this research are 2x10x170 cm3 boards from 4 wood species, namely : Nangka (Artocarpus heterophyllus), Kapuk (Ceiba pentandra), Afrika (Maesopsis eminii), and Sengon (Paraserianthes falcataria), noted by N, R, A, and S respectively, and isosyanate adhesive to bond three layers glulam. Face and back of glulam are N and core is R, A, or S. Test of wood small clear specimen and glulam bending properties is conducted with center point loading configuration based on ASTM D143. Then the equation of load-deformation curve is built on two ways. The first way is only linier regression for data below proportional (elastic) limit; the second way is dummy variable regression which allows both linier regression for data bellow proportional limit and continued by quadratic regression for data upper proportional limit until maximum load. The empirical MOE and MOR are calculated by and equation respectively. The theoretical MOE and MOR are calculated by Bahtiar’s methods (2008). RESULTS AND DISCUSSION. The results of this research show that the average value of MOE calculated by first method for Nangka is 6,3x104 Kg/cm2, Afrika 5,9x104 Kg/cm2, Sengon 5,3x104 Kg/cm2, and Kapuk 2,7x104 Kg/cm2. Average value of MOE calculated by second method for Nangka 6,0x104 Kg/cm2, Afrika 5,6x104 Kg/cm2, Sengon 5,0x104 Kg/cm2, Kapuk 2,6x104 Kg/cm2. The MOE calculated by second method is similar with the first methods for Kapuk but lower for three other species. Average value of MOR of Nangka is 680 Kg/cm2, Afrika 490 Kg/cm2, Sengon 460 Kg/cm2, Kapuk 273 Kg/cm2. The average value of theoretical MOE for glulam N-A-N formation is 6,2x104 Kg/cm2, N-S-N is 5,8x104 Kg/cm2, and N-R-N is 5,9x104 Kg/cm2. The average value of theoretical MOR for glulam N-A-N is 617 Kg/cm2, N-S-N is 584 Kg/cm2, and N-R-N is 675 Kg /cm2. The average value of empirical MOE for glulam N-A-N is 4,6x104 Kg/cm2, N-S-N is 4,9x104 Kg/cm2, and N-R-N is 4,5x104 Kg/cm2. Meanwhile the average value of empiric MOR for glulam N-A-N is 256 Kg/cm2, N-S-N 261 Kg/cm2, N-R-N is 223 Kg/cm2. Theoretical calculation of MOE and MOR of glulam are significantly different with empiric testing result of glulam. Empirical testing shows that the average value of MOE and MOR lower than theoretic calculation. The adhesion of glulam is weaker than the shear strength of lamina, so it doesn’t match with Bahtiar’s methods which assume the first destruction of glulam should happened on lamina (not on glue layer). Since the assumption is not fulfilled, the horizontal shear on glulam should be considered as important variable. CONCLUSION. The MOE value calculated by dummy variable which allows both linier regression continued by quadratic regression are lower than conventional methods which only allows linier regression bellow proportional limit. Theoretical value of MOE and MOR of glulam are higher than empirical value because glue lines have lower strength than the lamina. Keyword: MOE, MOR, Glulam
Collections
- UT - Forestry Products [2376]