The response of partially confined right circular stainless steel cylinders to internal air-blast loading

Abstract

This article presents the results of an experimental and numerical investigation into the response of partially confined right-circular stainless steel cylinders to air-blast loading. The blast loading was generated by detonating spheres of plastic explosive at two axial positions along the centre line of the cylinders. Partial confinement was created by closing one end of the cylinder and leaving the other end free to vent to air. Numerical simulations were performed to gain insight into the blast wave propagation and the transient response of the cylinders. As expected, the diametric deflection of the cylinders increased with increasing charge mass, and was a maximum in the same axial location as the charge. For the centrally located charges (L = 150 mm), the diametric deflections increased linearly with increasing charge mass. The numerical simulations showed that the reflected pressure from the closed end of the cylinder (that is, the axial component) interacted with the radially developed pressure reflected from the cylinder walls. This caused the pressure to be driven out of the open end of the cylinder when L = 150 mm, meaning that the expected quasi-static pressure accumulation had little effect on the deformation of the cylinder.When the charges were placed closer to the open end, at L = 225 mm, the experimental diametric deflections increased exponentially with increasing charge mass, and were significantly higher than the deflections measured when L = 150 mm. The simulations predicted linear increases similar to those for L = 150 mm, but at slightly higher magnitudes. This eliminates the lower mechanical support at the open end from being the main cause of the higher experimentally observed deflections, as this would also have been observed in the numerical simulations. Since the numerical simulation results were unable to fully predict the response and pressure accumulation when L = 225 mm, there must be some physical phenomena must be present in the experiments that did not affect the response when L = 150 mm (that was also not captured by the numerical simulations). One possible explanation is that afterburning of the explosive products was a significant factor when L = 225 mm, but this requires further investigation to confirm.