Non-monotonic variation of stress intensity with flaw size in metal-lined fibre-reinforced pressure vessels

Abstract

Metal-lined continuous fibre reinforced plastic (FRP) over-wrapped pressure vessels are used in aerospace applications, for storage of breathing air in fire fighting and scuba diving and for the storage of compressed gaseous fuels on natural gas and hydrogen vehicles. Continuous fibres, generally about 15 micro m in diameter and made of glass, carbon or kevlar, are embedded in a polymer matrix. The metallic liner is made of a ductile material - generally steel or aluminum alloy.

After the metallic liner has been wrapped, the composite vessel is subjected to a process termed Autofrettage. In that process the vessel is internally pressurized to the point that the ductile metal liner undergoes a small amount of plastic deformation (unlike elastic deformation which disappears upon removal of stress, plastic deformation remains after the stress has been removed). Upon de-pressurization of the vessel, the metallic liner remains under compression and the FRP under tension.

Acoustic emissions associated with fiber breakage are being developed currently as a non-destructive means of assessing the structural integrity of metal-lined continuous FRP over-wrapped vessels. Laboratory experiments have been carried out with flaws such as cracks and saw cuts of varying dimensions oriented in an axial-radial plane and located in the metallic liner, in the FRP or in both. Pressurization of the flawed vessel leads to fiber breakage, the extent of which is being examined with the intent that it will be a measure of structural integrity of the vessel. The results, however, suggest that the acoustic emissions attain a maximum for an intermediate flaw size. Low emissions are recorded when on the one hand the vessel has insignificant flaws, or on the other if the vessel has serious flaws. This non-monotonic variation of acoustic emission occurs whether the flaws are located in the metallic liner, in the FRP or in both.

The experiments suggest that the stress intensity at the discontinuity (crack tip) attains a maximum at an intermediate flaw size. A mathematical corroboration is desired.