An integrated investigation of rod-on-rod (symmetric Taylor) impact of annealed copper was conducted using the single-stage gas-gun facility at the Cavendish Laboratory as a validation study of the Armstrong-Zerilli constitutive model, as modified by Goldthorpe. Two main techniques were used for obtaining data from the experiments: high-speed photography (up to 20 million frames s K1 framing rate) and a velocity interferometer system for any reflector (VISAR). The symmetric configuration was used to minimize friction effects and eliminate target indentation seen in classic Taylor tests, where a rod is fired against a massive target block. However, the need for coaxial alignment of the two rods made the experiments considerably more challenging to perform than the classic case. The propagation of plasticity along the rods was monitored using high-speed photography and VISAR. It was found to propagate with a logarithmically decelerating velocity. The rod profiles and VISAR traces can be understood in terms of material properties such as strain hardening. No asymmetry between the responses of the two rods involved (moving and stationary) was observed within the resolution of the techniques employed. A modified Armstrong-Zerilli material model for copper predicted intermediate profiles well, but slightly overestimated the material strength.

The ballistic impact properties of a borosilicate ('pyrex') glass was studied using mild steel rods accelerated using a light gas gun. High-speed photography at sub-microsecond framing rates was used along with schlieren optics to investigate the propagation of elastic shock waves and fracture fronts. Flash X-radiography was used to visualize the deformation of rods as they penetrated the comminuted glass normally. The rod was seen initially to dwell on the surface for at least 3 s creating a Hertzian cone-crack. Later on, between 40 and 60 s, self-sharpening of the projectile was observed as the 'wings' of the heavily deformed front end sheared off. After this event, the front of the rod speeded up.X-rays also showed that the pattern of fissures within the comminuted glass was observed to be very similar shot-to-shot. X-radiography was also used to examine the mechanisms occurring during oblique impact of rods at 45˚. In oblique impact, bending of the rod rather than plastic deformation ('mushrooming') takes on the role of distributing the load over an area larger than that of the original rod diameter. High-speed photography of the rear surface of a glass block on which a fine grid had been placed confirmed that the comminuted glass moved as larger interlocked blocks. The experiments were modelled using the QinetiQ Eulerian hydrocode GRIM making use of the Goldthorpe fracture model. The model was found to predict well the transition from dwell to penetration.

The impact and penetration of brittle solids has been the subject of study for many years. Recently, the development of gauge techniques to measure the deviatoric response of various materials has highlighted the role of the shear strength in determining the ballistic performance. Recent experiments on various glasses using plate impact and ballistic geometries have characterised materials and provided a detailed analysis of penetration behaviour using simultaneous X-ray, high-speed photography and gauges. We present a study of impact on a borosilicate glass using a combination of these techniques.RLsumL : La penetration des solides fragiles fait l'objet de nombreuses etudes depuis plusieurs annees. Rkemment, I'utilisation de jauges de contrainte pour mesurer la reponse deviatoire des matkriaux a mis la lumiere sur le r61e de la resistance au cisaillement pour interpreter les performances balistiques. De recentes experiences sur divers vemes ont ee conduites tant en impact de plaque qu'en impact balistique. L'utilisation sirnultanee de la radiographic, de la cinematographie rapide et des jauges a permis d'apporter des elements sur le componement durant la penetration. Nous presentons une etude sur fimpact d'un verre borosilicate utilisant cette approche.

No abstract

No abstract