The influence of impact location and angle on the dynamic response of a large omnidirectional child dummy head and neck complex

Head injuries are a common cause of fatality and long-term impairment in child occupants in motor vehicle crashes. The National Highway Traffic Safety Administration (NHTSA) has developed the Large Omnidirectional Child (LODC) Anthropomorphic Test Device (ATD) where the head was designed to match pediatric biomechanical impact response targets from previous literature. The purpose of this study was to compare experimental and computational results for eight impact directions at 45-degree increments around the LODC head under two levels of impact severity: low and high, corresponding to nominal velocities of 3.08 m/s and 5.42 m/s, respectively. The experimental setup consists of the LODC head and neck assembly rigidly attached to a circular fixture plate and a hemispherical-shaped impactor 76.2 mm in diameter. The acceleration and angular velocity responses were measured and computed from the LODC finite element (FE) head CG and compared against the experimental data. Experimental peak acceleration for the low severity impacts ranged from 100 g to 145 g, which resulted in Head Injury Criteria (HIC) scores between 150 and 300 for all directions. Experimental peak acceleration for the high severity impacts ranged from 258 g to 428 g, with HIC scores between 904 and 2868. However, LODC FE peak acceleration responses for low severity tests ranged from 74 g to 206 g, with corresponding HICs ranging between 88 and 542; high-severity-impact responses ranged from 155 g to 427 g, with corresponding HIC values ranging between 672 and 2885. Experimental peak angular velocity ranged from 600 to 1800 deg/ms for low severity impacts and from 958 to 2355 deg/ms for high severity impacts. LODC FE responses for the peak angular velocity of low severity impacts ranged between 800 and 1684 deg/ms, whereas they ranged from 908 to 2230 deg/ms for high severity impacts. Correlation and analysis (CORA) scores for the head resultant acceleration resulted in mostly poor correlation and for the primary axes of angular velocities were graded in the spectrum of fair to good correlation. The FE head acceleration response is found to be strongly influenced by the assumed coefficients of friction between two interfaces, e.g., the head flesh to skull and head flesh to impactor interfaces. Recommendations for modeling refinements are discussed.