Determination and prediction of the mechanical behaviour of architectural fabrics

Abstract

This thesis concerns the material behaviour of architectural fabrics for use in the construction of tensile fabric structures, particularly the determination and prediction of biaxial and shear behaviour. Original contributions to knowledge include a novel shear test frame design, an understanding of the influence of biaxial stress on shear behaviour and an improved predictive unit cell model. While tensile fabric structures are subject to a combination of biaxial tensile stress and shear stress, there is no accepted test methodology for accurately determining shear behaviour of architectural fabrics. Shear behaviour is absent from some analysis methodologies used by industry and broad assumptions must be made by design engineers. A novel picture frame shear test design and associated test protocol is presented that aims to provide a practicable solution for the accurate determination of the shear stiffness of architectural fabrics. Strains are shown to be homogeneous across the test specimen during shear testing. The influence of biaxial stress on fabric shear behaviour is explored through tests conducted on polyvinyl chloride (PVC) coated polyester fabrics, PVC coated glass fabrics and polytetrafluoroethylene (PTFE) coated glass fabrics. Results of the tests, conducted at increasing levels of biaxial prestress, and the implications for analysis are presented. Existing predictive fabric models based on constituent material properties are unable to predict fabric behaviour with a level of accuracy which is sufficient for their use in design. An improved predictive fabric model is proposed using a sinusoidal description of yarn geometry. A system of compatibility and equilibrium equations is derived which aims to realistically simulate principal deformation mechanisms within real fabrics. The improved model predicts non-linear yarn behaviour and hysteresis using input parameters obtained using non-specialist test equipment, i.e. test equipment which is available in typical material testing laboratories. The model is validated by comparing predicted data with experimentally obtained data for a range of PVC coated polyester fabrics, PTFE coated glass fabrics and silicone coated glass fabrics. Safer and more efficient structural solutions will be possible if accurate material tests are available to characterise material behaviour. Reliable predictive models will make accurate design parameters easily accessible to designers.