Abstract
Starting from numerical models which are used in engineering, it is shown that it is possible to bring to light the main features which are responsible for the generation of self-sustained tones in turbulent flows.
The numerical study is based on our experimental results. The set-up consists of two diaphragms in tandem in a duct, leading to self-sustained tones up to 150 dB.
The modelled flow is assumed to be incompressible. It does not allow estimation of noise levels. However the calculation of the flow field exhibits the necessary conditions for the conversion of the kinetic energy towards the acoustical field.
An initial stationary model shows the location of the acoustical source. A second non-stationary L.E.S. model describes the dynamics of the detachment and impingement of the coherent structures.
An experimental noise reduction method is performed. The numerical approach contributes an explanation of the noise reduction mechanism.
