Stressed-Skin Panels of Mixed Construction - Using Wood-Based Materials, Especially Chipboard

Doctoral thesis, 1993

Stressed-skin panels are structural elements widely used as floors, roofs and so on. In such panels, flanges of sheet material interact with webs of beams (joists). The aim of the present study is to summarise various aspects of the short- and long-term behaviour of stressed-skin panels, mainly those made of timber and/or wood-based materials together with a thin steel-sheet flange. In order successfully to predict the long-term behaviour of a stressed-skin panel, knowledge of the long-term behaviour of possible component parts and joints in this panel was needed. As a result, the first part of this study focuses on the long-term behaviour of component parts and joints with particular emphasis on chipboard. One aim of the experimental part of the chipboard study was to examine and quantify creep behaviour under different loading conditions (bending, compression and shear). Another aim was to develop methods for the prediction of long-term deformation from short-term creep data. The use of weighting factors at the end of the observed creep data period can significantly improve the predictive capacity of a model. The effects of interlayer slip on composite action for three-layer symmetrical and asymmetrical stressed-skin panels were studied under short- and long-term loading. Theoretical simulations using a closed-form solution of a three-layered system with linear properties showed very good agreement with the measurements obtained from short-term tests on some stressed-skin panels with various flange and web combinations. The stiffness of both flanges has a greater effect on the total stiffness of stressed-skin panels as the degree of interaction increases. Flange curling in a very thin tension flange (0.6 mm steel sheet) should always be taken into account even at relatively low stress levels. All the long-term laboratory tests of stressed-skin panels were conducted in a controlled climate (20°C, 65%RH) under a constant load. The elastic solution for a panel with partial composite action was adapted to creep conditions by using an effective modulus of elasticity for all the component parts and joints. The elastic modulus was reduced for each specific time to obtain the effective modulus of elasticity by using a simple creep model, namely a "power function". The predictions of the long-term behaviour using this simple method appeared to be extremely good. By obtaining the elastic and creep properties for material specimens matched with the component parts of stressed-skin panels, the prediction of the behaviour appears to be more reliable than that obtained in some other ways. The simulation of long-term deflection of a few stressed-skin panels tested by others showed good agreement even when predicting creep at stepwise-increasing stress levels. Using a thin steel sheet as a tension flange in order to reduce deflection and creep in a stressed-skin panel should be extremely advantageous, provided that the durability aspects of its joints are carefully considered.