Evaluation of Shoulder-Complex Motion in Frontal and Oblique Impact Loadings - A Comparison between Frontal Impact Dummies and the Human Body

Offset and angled frontal impacts represent a crash type that is more common than full-frontal collisions. Several studies have shown that over 70% of all frontal impacts are offset or oblique. These studies thus indicate the importance of having crash tests representing offset and oblique frontal collisions so that protective systems may be evaluated that offer offset or oblique occupant protection. The evaluation of such protective systems requires access to biofidelic crash test dummies that accurately replicate the human body motion in frontal impacts with oblique dummy loadings. This thesis investigates how the performance of today’s 50th percentile male frontal impact dummies may be improved in oblique collisions with oblique frontal occupant motions. In the first part of this thesis the kinematics of the 50th percentile male frontal impact dummies, Hybrid III, THOR 99 and THOR Alpha, were compared with corresponding kinematics of PMHS in oblique crash tests. The PMHS data included results from oblique frontal, near side (the occupant moves into the shoulder belt anchor) and far side (the occupant moves away from the shoulder belt anchor), sled tests at 15°, 30° and 45° angles. The test subjects were belted with three-point belts and the impact speed was 30 km/h. The results indicated that the frontal impact dummies slipped out of the shoulder belts more easily than the PMHS. This was mainly an effect of stiffer shoulder complexes than those of the PMHS. The results from the first part of this thesis was the foundation of the second part were the shoulder range-of-motion and resistance of motion i.e. stiffness was compared between volunteers and that of the 50th percentile male frontal impact dummies, Hybrid III and THOR Alpha. The test conditions were in static loading conditions simulating frontal collisions. Five volunteers, a Hybrid III and a THOR Alpha were positioned in a test rig where both arms were loaded statically in the forward-upward direction at 90, 135 and 170 angles while sternum was supported. The results showed that the volunteers’ motions are at least three times larger for the maximum load (200 N/arm) than those of the Hybrid III and the THOR Alpha. The results in this study indicate that the biofidelity of the dummies used today in full-frontal, oblique and offset frontal collisions may be improved by redesigned shoulder complexes. The dummies should then better predict human head and T1 kinematics, and load distribution of the chest