No abstract

No abstract

Static and dynamic burst tests were carried out on filament wound glass reinforced plastic (GRP) tubes. The tubes were loaded under internal pressure with minimum end constraints. In the static tests, the pressure was supplied to the rig by a Losenhausen servo hydraulic intensifier system, whereas in the dynamic tests a drop hammer apparatus was used to generate the pressure, giving medium strain rates of the order of 10 sec -1 . Higher strain rate burst tests were carried out using suitable explosives. A system for fast data acquisition was developed so that during each test, the internal pressure and strains in the central section of the specimen could be recorded, the data processed and the results automatically plotted in the form of pressure-time, strain-time and stress-strain curves. Some high speed photographs were also taken during the dynamic bursting of the tubes to record the failure mode for each winding angle.The results showed no significant change in the strength, strain to failure or the secant modulus of the tubes at strain rates up to about 10 sec -1 , typical to the drop hammer apparatus. There was, however, a large increase in burst strength as the strain rate was further increased to 150 sec -1 during the explosive tests. The values of strain to failure were also increased, thus leaving the secant modulus of the tubes virtually unaltered.

A comprehensive experimental study was undertaken to evaluate strain rate effects on the hoop burst strength, hoop modulus, strain to failure and Poisson's ratio of Kevlar reinforced plastic (KRP) angle ply filament wound tubes. Both static and dynamic tests were performed on tubes of winding angle 25 ° , 45 ° , 55 °, 65° and 75 ° , which were burst under internal radial loading with minimum end constraints. For each test, the inter nal pressure and the strains in both circumferential and longitudinal directions were re corded on suitable digital processing equipment. For a particular batch of tubes tested at different strain rates, the results showed a signifi cant rise in the hoop burst strength with increasing strain rate, for all tubes. The high winding angle tubes (65° and 75°) failed by catastrophic fibre fracture; the strain to failure values increased with increasing strain rate and the hoop modulus remained fairly constant over the tested strain rate range. The lower winding angle tubes were characterised by an "initial failure" associated by resin cracking and leading to a subsequent degradation in tube modulus. The use of a non-structural liner during quasi-static tests leads to a signifi cant increase in ultimate strength and strain of all tubes tested with the exception of the 25°.

Several theoretical procedures are proposed to extract unidirectional ply properties from the results of dynamic tests on angle ply laminates. The tests were carried out on filament wound tubes under hoop loading. It has been shown that the response of certain laminates is dominated by particular ply properties which can be isolated and extracted with accuracy. The extracted properties include longitudinal, transverse and inplane shear moduli and strengths. The procedures are based on conventional 2D laminate analysis with extensions to cater for lamina non-linear and progressive failure behaviour. These procedures are equally valid for static or dynamic test data. Full descriptions of methodology and computational flow charts are presented.