Index of content:
Volume 118, Issue 5, November 2005
- NOISE: ITS EFFECTS AND CONTROL 
118(2005); http://dx.doi.org/10.1121/1.2047207View Description Hide Description
This study examines the effects of a mean flow and turbulent flow excitation on the performance of the recently conceived device which was tested under the no-flow condition [J. Acoust. Soc. Am.112, 2014–2035 (2002)]. The silencer consists of two cavity-backed membranes lining part of the duct walls. When a certain optimal tension is applied, the silencer gives a broad stopband in the low-frequency regime. Similar performance is predicted for the condition with a mean flow, and tests conducted for flow speeds from validated these predictions. The spectrum of transmission loss without flow features three resonance peaks, and the mean flow is found to smooth out all peaks and shift two of them through cross-modal coupling. The silencer was tested in a wind tunnel, and no flow induced flexural instability was found on the membrane in the range of flow speeds tested. Insertion loss measurement was also conducted in a natural ventilation condition where a turbulence intensity of 3% was recorded, and the results were close to those without flow. It is concluded that no noticeable extra sound is produced by the turbulent excitation of the membrane under the optimal tension required by the silencer.
118(2005); http://dx.doi.org/10.1121/1.2047127View Description Hide Description
The performance of local active noise controlsystems is generally limited by the small sizes of the zones of quiet created at the error sensors. This is often exacerbated by the fact that the error sensors cannot always be located close to an observer’s ears. Virtual sensing is a method that can move the zone of quiet away from the physical location of the transducers to a desired location, such as an observer’s ear. In this article, analytical expressions are derived for optimal virtual sensing in a rigid-walled acoustic duct with arbitrary termination conditions. The expressions are derived for tonal excitations, and are obtained by employing a traveling wavemodel of a rigid-walled acoustic duct. It is shown that the optimal solution for the virtual sensing microphone weights is independent of the source location and microphone locations. It is also shown that, theoretically, it is possible to obtain infinite reductions at the virtual location. The analytical expressions are compared with forward difference prediction techniques. The results demonstrate that the maximum attenuation, that theoretically can be obtained at the virtual location using forward difference prediction techniques, is expected to decrease for higher excitation frequencies and larger virtual distances.