Experimental Measurements of Residual Stress in ARMOX 500T and Evaluation of the Resultant Ballistic Performance

Abstract

To better elucidate the processing route used in the manufacture of High Hardness Armour steel, the crystallographic texture and the residual stress in the through thickness of an ARMOX 500T plate was evaluated. Crystallographic texture can play a part in the armour's ballistic response, and authors note that the hot rolling process used in the manufacture of these steels often leads to an inhomogeneous through thickness texture variation. Neutron diffraction experiments highlight two texture components, a shear texture near the surface, and rolling texture, which extends from 1.5 mm below the surface to the plate’s centreline. Moreover a pronounced twin peaked residual stress field was observed using neutron diffraction and verified through the contour method, with stresses ranging from − 150 to 150 MPa for the rolling and transverse directions. Using the experimentally measured stress profiles, and recently derived Johnson–Cook flow stress parameters, the authors mapped the stresses to a finite element model and subsequently carried out numerical analysis on the ballistic response of the plate against a 0.30 cal APM2 round. Analysis of the two starting conditions, with and without residual stress, allowed the authors to highlight a small improvement in the ballistic response by way of: (a) a reduction in the positive tri-axiality experienced by the plate’s back face and (b) an increase in the volume of material ahead of the bullet with negative tri-axiality. The presence of compressive stresses on the strike face coincides with an increase in the adiabatic heating of the plate specimen which counters some of these benefits. The newly generated ballistic limit diagram is compared with the available literature and a number of variations are apparent which are discussed in light of the flow stress regimes. Comparisons of the Finite Element results to a newly modified version of the analytical Forrestal–Warren model show excellent correlation, thus providing another rapid verification method.