Study on the estimation of the strength properties of structural glued laminated timber I: determination of optimum MOE as input variable

Abstract

There have been many attempts to predict the performance of glulam beams. Several approaches have been taken, from early empirical techniques to more sophisticated stochastic methods. In recent years, more emphasis has been placed on the modeling of material properties. Generally, the modulus of elasticity (MOE) has been used as a criterion of laminar strength for the prediction of glulam performance in the traditional models. Most of the current models are based on MOE that was measured using the long span test; that is, they account only for variability between pieces of lumber. Therefore, these models do not account for the variation of material properties within a given piece of lumber. Five methods were considered to choose the appropriate one that could effectively predict the performance of glulam in this study. Prediction of glulam performance was done by the transformed section method. MOEs measured with the five methods were applied to a strength prediction program to compare the actual test results and the predicted results. MOEs used as input variables are as follows: long span MOE of the static bending test, localized MOE of the static bending test, long span MOE of the stress wave test, localized MOE of the stress wave test, and MOE of the machine stress rating (MSR) test. Results of the localized test showed excellent signification compared to those of the long span test. The MSR method, when used as input variable, obtained the most approximate result, so it is considered adequate for predicting the strength of glulam.