Restraint System Performance and Injury Potential to Belted Occupants in Automobile Rollover Crashes

It has long been recognized within the automotive safety community and by numerous biomechanical research studies that providing effective occupant protection in automotive rollover crashes requires effective occupant restraint. Effective occupant restraint includes, at the most basic level, preventing occupant ejection and providing sufficient control of occupant kinematics through the rollover event to prevent potentially injurious contacts with interior vehicle components. This paper examines both laboratory and real-world accident analysis of restraint performance in rollover-type environments. This includes studies involving static and dynamic testing with human surrogates and anthropometric test devices (ATDs). Additionally, the effects of rollover roof deformation on the restraint systems ability to control or affect occupant kinematics, when those restraint systems are anchored to the dynamically deforming structural components of the vehicle, are examined. Finally, various production and alternative restraint system designs are considered and discussed relative to their ability to control occupant kinematics and their influence on belted occupants' injury potential in the rollover crash mode. This paper will focus on the effect of seat belt looseness, or slack, and its relationship to occupant excursion during a rollover. Literature is referenced establishing that increased occupant excursion produces increased injury potential in rollovers, both by increasing the likelihood of injurious contacts with interior vehicle components as well as an increased risk of full and partial ejection. Four complete vehicle inversion studies (spit tests) are conducted with live surrogate occupants in production vehicle restraint systems. These studies document occupant excursions under a 1G inverted environment with various amounts of seat belt slack in production restraint systems as well as comparison tests using various alternative restraint configurations. Additionally three complete vehicle inverted drop tests are conducted in which the vehicles' roof structures and the upper torso belt anchors (D-ring) are instrumented to document their displacement while producing typical real-world type roof structure damage. The effect of this restraint anchor deformation is then examined relative to the occupant excursions evaluated in the spit tests. Lastly, a complete dolly rollover test conducted on a contemporary production mini van with production restrained anthropometric test devices (ATDs) is examined with a focus on the restraint system's geometry alterations and effectiveness through the multiple roll/roof contact events.