Deformation,fragmentation and acceleration of a controlled fragmentation charge casing

Two different finite element software, LS-DYNA and Impetus, have been evaluated to test their ability to predict the deformation, fragmentation and acceleration of a controlled fragmentation charge casing. The general-purpose program LS-DYNA was used with a multi-material ALE formulation and a mass-preserving erosion criterion coupled to a Johnson-Cook fracture criterion. In the Impetus simulations, a third order Lagrangian element formulation was used for the casing and a node-splitting element erosion treatment coupled to a Cockcroft-Latham failure criterion was used to describe casing fracture. The high-explosive gases were described by a discrete particle formalism.In order to acquire data to validate our computational tools and constitutive models, a series of experiments have been performed using a laboratory charge with an internal grooved casing. In the test series, the charge geometry was fixed except that the groove depth were varied from very shallow to very deep resulting in different deformation patterns, fracture modes and terminal velocities. Various diagnostic tools captured the different stages of the expansion and fragmentation of the casing. A high-speed framing camera depicted the deformation pattern before fragmentation and was used to determine the moment when the casing failed. Three different complementary techniques were used to follow the acceleration of the fragments; a Photon Doppler velocimetry to determine the initial acceleration of the casing, double exposed radiographs to estimate the fragment velocity after break-up and a high-speed video to determine the terminal velocity of the fragment after leaving the fireball. In addition, the fragments were soft recovered in a set of sawdust pit tests and their final shape and weight were measured. A SEM was used to characterise the fracture surfaces and to determine the modus of fracture (tensile or shear failure).Comparisons to experiments show that both software can predict the change in deformation behaviour when the groove depth increases, from tangential necking for shallow grooves to radial punching for deep groves. Both software could also reasonable well predict the acceleration of the fragments, though both overestimates the terminal velocity for the charge with the deepest grooves.