Characteristics of side-branch array aeroacoustics and its application in low mach number flow duct noise control

Abstract

The objective of this study is to investigate the noise attenuation devices consisted of multiple quarter-wavelength resonator. This study is threefold. The first objective is to understand the acoustic behaviour under no-flow condition. The second part is to comprehend the behaviour of the resonator if the sidebranch only occupy a portion of the spanwise duct width. This provides a new direction on improving the attenuation performance by installing other kinds of sound attenuation device. The third part of the study is about the array performance under slow flow. It aims to find out a solution for reducing aerodynamic deterioration to the array performance. The study is carried out by using theoretical model, FEM simulation and experiment. There are two sidebranch length variations adopted throughout this study. They are Linear Length Variation (LLV) and Linear Frequency Variation (LFV). An array of multiple quarter-wavelength resonator is flush mounted on an infinitely long rectangular duct for the examination. The resonance frequencies of sidebranches forming the array is compared to the resonance frequencies obtained from the 11-sidebranch array. They cannot perfectly align due to the fluid loading interactions between sidebranches. A more complex design is done by flush mounting the second array on the opposite side wall of the existing array, which can be hanged either in the same or reversed orientation. Their performances are different. The normal arrangement is able to improve the sound attenuation power and the effective range, when compared to the single array. The reversed arrangement can give severe coupling effect at certain frequencies, resulting in zero transmission loss. The combinations of half spanwidth arrays are tested by using the finite-element method. The base case is the basic full spanwidth array. There are dipole-like and quadrupole-like radiations emitted into the main duct from the sidebranches, because of the interactions between the coupled arrays. For the case involving two sidebranch arrays of different tube length hanged side-by-side on one duct wall, the spectral uniformity of sound transmission loss is enhanced. The whole system can be further enhanced by installing one more set of arrays on the opposite duct wall. The separation wall between the sub-arrays of one of the coupled sidebranch arrays can be removed for further improvement. However, the low frequency performance of the array is worsened if one of the coupled array is reversely installed. There are also trials on replacing one half spanwidth array with an expansion chamber. The result shows that such arrangement can shift the effective span to lower frequency range, while the attenuation performance in higher frequency range is weakened. This attempt introduces the potential of integrating other sound attenuating devices for further development. The 11-sidebranch array is tested with flow applied. The first examination is for the relationship between the incident pressure level and the TL performance under flow condition. The attenuation power of the array is less influenced by the flow when the incident sound pressure level is high. This result is related to the strength of acoustic pressure in the sidebranches. Therefore, a relatively simple 2-tube model is solved theoretically for a better understanding of the acoustic interaction between sidebranches. The orientation of the array is found to be important to the acoustic pressure distribution in the array system. This finding is validated by the experiment. The reversely installed array has stronger resilience to the aerodynamic disturbance.