Coefficient of Restitution Testing: Explicit Finite Element Modeling and Experiments

Abstract

Particle based modeling approaches, such as the discrete element method (DEM) approach, require the definition of accurate contact (collision) models. An essential parameter within these models is the coefficient of restitution (e), which defines the ratio of post-collision to pre-collision relative velocity during the collision of two materials. In this study, e of various steel-material combinations is predicted through both physical experiments and explicit finite element modeling of a falling sphere colliding with a stationary plate, and examined against a theoretical formulation of e. Experiments are performed on various sphere materials including steel, brass, chrome steel, tungsten carbide, aluminum, polybutadiene, and nitinol 60, Experimental results for metals colliding with steel, match what is predicted by theory, as they show a decreasing trend in e with increasing impact velocities. Additional experimental results for polybutadiene rubber show no velocity dependence, remaining constant across the impact velocities examined; this matches with theory. Modeling results, obtained via the explicit finite element method (FEM) approach, are compared against experimental results for verification. Current explicit FEM results show very good quantitative agreement with experimental results for various materials (brass, steel, tungsten carbide, and chrome steel) impacting steel, but do not display the proper decreasing trend in e for increasing impact velocity, over the range of impact velocities examined (∼2 m/s–3.2 m/s). However, for simulations at impact velocities above this range, a decrease in e is witnessed. Current and future work focuses on refining the models to correctly display this qualitative trend within the current range of experimental impact velocities, as well as performing experiments on a wider range of impact velocities.