Investigation of Dynamic Characteristics of Rolling-Ball Dampers

  • Nagashima, Kirino (Graduate School Engineering and Science, Shib)
  • Saeki, Masato (Shibaura Institute of Technology)

Please login to view abstract download link

Tuned mass dampers consisting of an auxiliary mass, spring and damper system are widely used to mitigate the response of lightly damped structures. However, frequent maintenance is required due to age-related deterioration of the spring and damper. Komatsuzaki et al.[1] have investigated the damping performance of a cycloidal pendulum dynamic vibration absorbers. As these dampers are simple devices characterized by a steel ball rolling on a cycloidal shaped surface, frequent maintenance isn’t required. In these dampers, additional damping is adjusted by changing the coefficient of friction between the steel ball and the rolling surface. Tsuda et al.[2] have investigated the effectiveness of tuned rolling-cylinder dampers. The dampers consist of some cylinders rolling on a cylindrical surface and the optimum additional damping is obtained by the friction between the cylinders. Ozawa et al.[3] presented a parameter optimization method of the tuned rolling-cylinder dampers and confirmed that a high vibration damping could be obtained through experiments. Chen et al.[4] showed experimentally that rolling ball dampers, characterized by multiple steel balls rolling in a spherical bawl, could reduce the dynamic response of a wind turbine tower. Although the study described the damping effect in detail, the results focused on the motion of the multiple steel balls with the same size. In this study, we investigated the characteristics of a rolling ball damper in a horizontally vibrating system. This damper consists of balls with different radii on an arc-shaped container with a lid. The effects of the size of the balls and the number of particles used in combination on the damping performance were investigated experimentally. The simulation was performed by combining EDEM software and MotionSolve software. The comparison of the experimental and numerical results show that the numerical computation is effective for prediction of the amplitude response of the primary system.