| An experimental study was conducted to determine local strain, local strain rate, and temperature distribution within an adiabatic shear band during its formation. Thin-walled tubular specimens were deformed at dynamic rates in a torsional Kolsky bar (torsional split-Hopkinson bar) and direct observation of the strain localization process was made through ultra high-speed photography of a grid pattern deposited on the specimen's surface. Temperature measurements were effected using an array of sixteen infrared radiation detectors. The optical system employed allows each detector to focus on a band 17 {dollar}mu{dollar}m wide. Three different steels were tested: (1) an AISI 1018 cold rolled steel, (2) a low alloy structural steel (HY-100), and (3) an AISI 4340 VAR steel tempered to either of two hardnesses, HRC 44 or 55. The principal goal of the investigation was to relate initial geometric defects in specimen geometry to the timing of stress collapse during shear band development. A comparison of experimental results with predictions based on the analysis of Molinari and Clifton showed good agreement. In addition, quasi-static tests on the same materials revealed two significant differences between dynamic and quasistatic plastic deformation. First, localization of strain into a shear band occurs in dynamic deformation, but never quasi-statically, and second, the magnitude of the strain at fracture is always considerably greater in quasi-static deformation than dynamically. The infra-red detector system indicates that the average temperature rise in the narrow shear bands (3 to 10 {dollar}mu{dollar}m) attains a maximum of about 600{dollar}spcirc{dollar}C. This does not eliminate the possibility of a phase transformation since the temperature distribution is unknown. Transmission electron microscopy would be an effective technique for future investigation of the microstructure of the adiabatic shear bands. |
