Graduation date: 1994

No description available.

Moisture ingress and thermal effects on carbon fibre reinforced plastic is a well understood phenomenon. For aircraft structures where safety is paramount this results in the use of worst case material properties, known as HOTAA/ET properties. In reality most structures are not fully saturated and are therefore penalised by using these worst case properties. This project attempts to fully understand the environmental effect on mechanical performance and accurately model a structures exposure to the environment, while still maintaining conservatism, to realise structural weight savings for aircraft. From the literature study it appears that this is the first attempt to link the mechanical property degradation brought about by environment, to classical laminate theory. By modelling individual ply property performance, based on each ply's level of saturation and linking it to a bespoke set of materials properties generated within the project, it is possible to accurately model the mechanical performance of a component. The model and modelling process derived within this project have been successfully validated by structural testing.

Moisture diffusion studies were performed using unidirectional (UD) tape and quasiisotropic (QI) woven 5-harness satin fabric, carbon fibre reinforced (CFR) epoxy composite materials. Firstly the moisture constants, (i.e. diffusion coefficient, D[x], and equilibrium moisture content, M[max]) were experimentally derived at 70°C and 85% relative humidity (%RH), for the two CFR materials. To investigate moisture absorption as a function of %RH test coupons were conditioned to differing equilibrium moisture levels viz., 70°C/60%RH, 70°C/75%RH, 70°C/85%RH, and 70°C/95%RH. Also oven dry (OD) and as-received (AR) tests were performed for baseline comparison. The effect of moisture absorption on the mechanical behaviour was investigated; lamina properties were studied by measuring tension, compression, shear (inter/intralaminar) strength and stiffness of the UD material. This comprehensive set of testing provided quantitative relationships between moisture content and mechanical properties. The quasi-isotropic lay-up was then utilised to investigate multi-directional laminate lay-ups using open hole tension and compression testing. The experimental data showed that the uptake of moisture in both the materials studied was described well by Fick's Second Law and the properties most affected by moisture ingress were matrix-dominated properties. More specifically, the transverse tensile strength, F[t][2] was most affected by the ingress of moisture, with a near 50% reduction in strength when conditioned to equilibrium moisture content at 70°C/95%RH. Such information is a necessary prerequisite if improved design procedures are going to be developed in the future. The initial phase of testing produced mechanical property/moisture relationships that were employed to predict the strength and stiffness of the material containing specific moisture gradients through-the-thickness (TTT). To be able to predict mechanical properties with different moisture distribution, firstly moisture distribution TTT of the material was modelled using an analytical solution to Fick's Second Law. Then moisture content was considered on a ply-by-ply basis TTT of the laminate; reductions were applied to each individual ply property dependent on the moisture content using the experimentally derived relationships, essentially applying environmental knock-down factors (KEKDF'S) to each individual ply. Classical Laminate Analysis (CLA) was then performed using the Max Stress failure criteria in order to predict the overall laminate failure. A second phase of mechanical testing was then performed to validate these predictions. The mechanical property predictions compared well to the experimental data showing similar reductions in strength for a given profile of moisture in the laminate. The predicted strengths also fell within the measured standard deviation of the experimental data in a significant proportion of the results.

Abstract not available