HORK Enterprises
3221 Quick Road
Holly, MI 48442, USA
 +1 (248) 328-0231
 consult@hork.com

Roll Over

Sport Utility Vehicles (SUV) have been around since the 1930s, but only gained broad popularity in the second half of the 1990s. The early models were often based on light trucks, which gave them a high stance and stiff suspension. The high center of gravity made them especially prone to roll-over when swerving at highway speeds.
Roll-overs frequently led to occupants being ejected from the vehicle. That may be expected for unbelted occupants, but first responders also found occupants outside the vehicle, while the seat belts were securely buckled. Research showed that seat belts that were designed primarily for protection in frontal collisions had shortcomings for roll-over. Slack in the shoulder portion of a 3-point belt would allow webbing to slip to the lap belt portion. The violent nature of the roll-over then allowed the occupant to slip from under the lap belt and be prone to being ejected.
A first innovation that improved the effectiveness of the seat belt for roll-over was the retractor pretensioner that takes slack out of the belt. Subsequently, a device that cinches the webbing at the buckle latch plate, prevented the slippage and thus made the lap belt more effective.
In 2011 the U.S. Government issued a separate "Ejection Mitigation" standard in FMVSS-226. Unrestrained occupants are likely to be ejected from a rolling vehicle, because "without a force acting on a body, it will continue with constant velocity and in a straight line" and of course the car doesn't. Therefore, separation is all but inevitable. Exactly how and when that happens is affected by the smallest of factors. Even the size of your shoes can determine whether you slip freely out of the side window, or get hung up on the steering wheel for just a fraction of a second. And that can make the difference between the car rolling over you, or you being flung high into the sky.
If ever "no two events are the same" applied to anything, it applies to roll-over accidents. Most roll-overs start with a rotation about the longitudinal axis, which is the axis with smallest moment of inertia and highest kinetic energy. As kinetic energy is absorbed with each impact with the ground it will transition to rotating about the lateral axis, which has the largest moment of inertia and lowest kinetic energy. That means rolling end-over-end. Exactly how that transition occurs depends on how much energy is absorbed with each impact. The vehicle occupants will be thrown in every possible direction in the process. Compared to frontal impact and side impact, a roll-over is a long duration event. Whereas a frontal impact may be over in just 0.15s, a roll-over may take several seconds. Side curtain airbags, therefore, are designed to remain inflated for such a longer duration. They deploy out of the headliner and look somewhat like an air matrass, when inflated. They are sealed and generally use a cold, inert gas such as argon or helium, for inflation. To keep an occupant inside the vehicle they must be strong and span the gaps that are the side windows. Several methods have been developed to achieve that, one such method is described in that we have contributed to. The government and industry have long searched for a test that could be used to evaluate the effectiveness of restraint systems in case of a roll-over. Tests with the names "curb trip", "cork-screw", and "dolly test" were explored for suitability and repeatability. For the "curb trip" test the vehicle is placed laterally onto a cart, with the leading wheels, up against a curb. The sled deceleration then determines weather the vehicle trips into a roll or falls back onto its wheels. This allows for precise sensor calibration.
For the "cork screw" test the vehicle is brought up to speed and then driven up on a ramp on one side. This induces a twist that quickly transitions into a lateral roll. This most closely mimicks a swerve induced roll-over.
The NHTSA eventually settled on the J2114 dolly test.   As the name suggests, for this test the car is loaded sideways and slanted onto a dolly, which is brought up to speed (typically 30mph or 50 km/h). When the dolly is abruptly stopped, the car is launched into a roll.
To assure the restraint systems are deployed under a variety of accident scenarios, the industry generally evaluates the roll-over sensor system for all three test modes.