The Design Theory And Application On The Resonant Intake Silencer

Because engine induction noise is mostly in the low frequency region and the controlling measures for reducing the induction noise are different from that of exhaust noise. It is difficult to attenuate the induction noise entirely with a common expansion chamber silencer or an air filter. The dynamic performance and economic performance of an engine will be affected if the intake system is modified a little. Due to its inherent characteristics in reducing low frequency, small pressure loss and simple geometry, the resonant silencer is being extensively used in engine induction noise reduction. But the resonant silencer produces narrow bands of high transmission loss near the resonance frequency, it is very important that the resonant frequency is predicted as precisely as possible in the design of a silencer in order to save time, reduce experiment number and raise the success rate. The formula derived on the classical lumped-parameter theory is commonly used to predict the resonance frequency of the silencer and but limited by some assumptions: the dimensions of cavity and neck greatly smaller than the wavelength; the connector volume greatly smaller than cavity's, etc, due to the simplification involved in reducing distributed, multi-dimensional phenomena to lumped parameters and wave motion neglected by the classical approach. In realistic application, the volume and dimension of the silencer are usually out of foregoing assumptions due to the limitation of objective conditions. This can lead to situations in which the result estimated by the classical formula deviates from the real data; the cavity dimensions, which are neglected by classical theory, can have a significant effect on the resonance frequency. The effect of changing the cavity geometry dimension on the resonance frequency of concentric simple resonant silencer with rectangular cross-section, which is used usually in engine noise control, is investigated on the base of acoustic theory by three-dimensional finite element. The results have shown that the depth to width ratio of the cavity has a significant effect on the resonance frequency. The calculated resonance frequency reached a maximum when the cavity was approximately cubical, and then decreased as the cavity broadened. The cross-sectional aspect ratio (length to width) of the cavity has a little effect on the resonance frequency too. The resonance frequency decreases with increasing the aspect ratio and the more the aspect increases, the more the resonance frequency decreases. Following this introduction, the suitable region and error of the formula obtained from the lumped-parameter theory and the analytical expression obtained from the one-dimensional acoustic theory for calculation of resonance frequency is discussed. The results calculated by the foregoing two formulas differ noticeably from that by three-dimensional finite element even though the connector ends are corrected by the connector end correction formula used usually and the differences vary with the depth to width ratio of the cavity. For the lumped-parameter formula, the error is the least when the depth to width ratio varies from 0.34 to 1.0 and then become great when the depth to width ratio is smaller than 0.34 or greater than 1.0, the relative error reaches 88% especially when the depth to width ratio is greater then 3.05. Therefore, when the depth to width ratio is smaller then 1.0, the estimating error when used to predict the resonance frequency of a silencer is smaller. For the one-dimensional formula, the results approach that obtained by the lumped-parameter formula when the depth to width ratio is smaller, but approach gradually that obtained by three-dimensional finite element and so the estimating error when used to predict the resonance frequency of a silencer is smaller with increasing depth to width ratio. By considering various effective factors, the suitable connector end correction for the lumped-parameter formula used to predict the resonance frequency of the silencer of automobile engines is presented, the general principles and methods for the design of the induction silencer are summarized. Next, the experimental setup for silencers is designed and constructed in order to prove the analytical conclusions and investigate the effect of the flow in the main pipe on the resonance frequency of a silencer. The results have shown that the experimental data are agreement with the results computed using three-dimensional finite element approach and no matter how the flow direction is the same as or opposite to that of sound propagation, the effect of the flow velocity on the resonance frequency is all very small. Even though the flow velocity in the main pipe reaches 50m/s, the changing amount of the resonance frequency is only approximately 3Hz. Therefore, the effect trend of the cavity geometry on the resonance frequency is not changed and the connector end correction formula is suitable when there is flow in the main pipe. Finally, with application of the obtained conclusions, a three-side-branch resonant silencer is designed for SC7080A automobile engine in terms of its characteristics and is tested on the experimental setup and the automobiles. The obtained results have shown that the designed silencer can attenuate the engine intake noise effectively and only has a little effect on the engine output power, which demonstrate the methods and principlesused in the course of silencer design are correct and workable.