Abstract
<jats:p>Designers of vehicle barriers face a gap between two regimes. At the low end, AS/NZS 1170.1 prescribes fixed car-park barrier forces whose assumed crumple zones (0.10–0.15 m) have no stated empirical derivation and do not scale with vehicle mass or speed. At the high end, hostile-vehicle-mitigation standards certify specific barrier products at 48–80 km/h but publish no calculation method. In between, for impacts at 30–60 km/h or by vehicles heavier than a passenger car, the designer has no empirically grounded estimate of vehicle crush or average impact force. This paper offers that basis: an empirical research envelope. From 10,876 crash-test records of the United States National Highway Traffic Safety Administration (NHTSA), a documented quality-control chain yields a modern cohort, model year 2004 on, age-matched to the Australian passenger fleet. The 614 tests with directly derived dynamic crush supply the primary fit, a non-crossing quantile envelope in speed and mass, with average force following from the energy identity F_avg = mv²/(2d), checked independently against barrier load cells. The result is a rigid-barrier crush and force envelope supported over 1,007–3,448 kg and 33–57 km/h, read in both directions: the stiff-vehicle 5th percentile bounds force demand, the soft-vehicle 95th percentile bounds penetration. The load cells locate the height at which the force acts. Beyond the tested speeds, external evidence bounds a crush ceiling the data alone cannot locate, keeping the high-speed force treatment conservative. It is offered as a candidate basis for future calibration, not as characteristic design actions.</jats:p>