This paper presented possibility of the design of a low frequency shaker using MR(magneto rheological)dampers. In order to understand the possibility of the development of a low frequency shaker based on the theory that interpreted by the simulation, the components used in the experiment were designed and manufactured firsthand. The experimental system used in this paper consists of two steel plates without any deformation, MR dampers and 4 springs.
Prior to the experiment of the system, the performance of the damper was tested multi-directionally. In order to calculate the damping coefficient, the damping ratio was calculated previously by the harmonic response curve showing half power points and bandwidth. Meanwhile, the critical damping coefficient was calculated by the mass of the plates and the natural frequency. And based on the calculated critical damping coefficient, the damping coefficient can be obtained.
The syntony and responses were measured while the current was set to 0 A, 0.25 A, 0.5 A, 0.75 A, 1.0 , 1.25 A, 1.5 A, and the frequency ranging from 2 Hz to 11.25 Hz with the intervals of 0.125 Hz. And as the results, it can be identified that the higher the current was, the higher the damping coefficient presented. Also in another experiment where the frequency ranged from 3 Hz to 10 Hz at the intervals of 1 Hz, the higher the current was, the lower amplitude of the responses presented. The optimum value of the damper switch frequency f2 was 1 Hz, and the upper limit of the value was 2 Hz.
To do experiment using the manufactured system, signals were amplified by the power amplifier with selected frequencies that function generator needed. The amplified signals were connected to the shaker and it made the steel plate excite by using the frequencies selected in the function generator. The frequency component was analysed by signal analyzer connected to force sensor and an accelerometer on the vibrating plate. A force sensor was installed in middle of the plate to measure the exact vibratory force, and accelerometers were installed in middle and edges of the plate to measure the frequency components of vertical displacement and angular displacement.
The feature of the vertical displacement frequency components in the middle of the plate and the angular displacement frequency components on the edges of the plate which was verified in the simulation were analyzed. The feature of the angular displacement frequency content at the edge of the plate contains the frequency component that occured in the middle of the plate. Thus while the frequency components in the middle and the edges of the plate being measured simultaneously, the vertical frequency component was eliminated and the pure frequency component of the angular displacement was shown by a diagram.
The result showed that when there is no switch frequency f2 of the MR damper, either in the middle or at the edges of the plate the frequency component of f1 was highest, followed by 3f1 and 5f1. When f2 and the current became stronger, the measured value in the middle of the plate presented almost no differences, but the value of f1 at the edge decreased considerably, rather the frequency components of f1+f2 and f1-f2 increased.
As the result, the low frequency resonance force that we needed can be obtained by controlling the two frequencies rather than oscillate the shaker in vertical direction of the vibratory force by MR damper.
The components of the natural frequency fnd on vertical displacement and the natural frequency fnr on angular displacement that verified by simulations can''t be found in the damping coefficient and non-linearity of the system. As far as solving the problem which contains eliminating the residual magnetism effect of MR damper, the problem of the damping coefficient and non-linearity, development of a low-frequency shaker using MR damper will be more easier.