Abstract
Nelson and Morfey have derived an expression for the sound power radiated from an in-duct spoiler subjected to the action of fluctuating drag forces arising from the flow turbulence in the vicinity of the spoiler. Although the derivation of an analytical expression for the fluctuating drag forces is impractical, an alternative approach could be the use of Computational Fluid Dynamics (CFD). The Navier-Stokes equations, which are the basis of CFD, provide a valid description of nearly all practically relevant flows including turbulent ones. However, in order to solve the relevant equations with sufficient accuracy to describe the fine detail of turbulent flow would mean working with a three dimensional mesh large enough to cover the region of interest but with spacing smaller than the smallest turbulent eddy present in the flow. Therefore the computational time required is unrealistically large even on modern super computers and using very efficient algorithms.
In this paper an engineering approach is followed similar to that of Nelson and Morfey and other researchers to explore the possibility of an alternative method in which information provided by standard turbulence models can be employed as the basis of an airflow noise prediction method. In view of the success of early investigators in collapsing experimental data based upon the assumption of a proportional relationship between steady and fluctuating forces, an attempt was first made to confirm the validity of the model developed against this assumption using computational fluid dynamics. Experimental results presented by previous investigators have been processed and compared with CFD predictions. As well as providing a degree of validation for the approach developed in this paper, this investigation sheds further light on the assumption implicit in much of the early work of a single constant of proportionality relating the fluctuating to steady state drag forces.
Get full access to this article
View all access options for this article.