Experimental Investigation of Installation Effects in a Low-Speed Fan System
2.7 Installation Effects
The reduction of the noise generated by industrial fans that is imposed by ever increasing regulations requires an improved understanding of aeroacoustic mechanisms, in particular of the interactions taking place between the components of HVAC systems or household appliances. Indeed, while the noise produced by an isolated blower is usually characterized under ideal inflow conditions with uniform inlet profile and low turbulence intensity, such conditions are in practice rarely met in the full system, due to product compactness and integration constraints. This leads to extraneous noise generation that can compromise the acoustic performance of the equipment, and thereby increase its development time and cost.
The paper describes an experimental investigation of installation effects for a low-speed fan, based on an advanced multi-ports modal eduction technique. This approach, implementing arrays of microphones and loudspeakers, permits discriminating between aerodynamic and acoustic effects and hence to isolate the influence of flow distortions located upstream of the fan. The outcome of the procedure is a quantitative measure of the noise that is produced by the fan, irrespectively of artifacts of the test rig that would inevitably contaminate the results otherwise.
The cornerstone of the procedure is the projection of the acoustic field measured by the microphone arrays on a modal basis corresponding to the Green’s function of the circular duct, parameterized by its axial wavenumber, radial and azimuthal mode orders. In this study, the planar, first two azimuthal and first radial modes are included in the analysis. The modal acoustic reflection of the test rig terminations, as well as the modal scattering properties of the fan, are obtained using the loudspeaker arrays. The influence of the positioning of the microphones is discussed with regard to the accuracy of the matrix inversions involved in the post-processing of the acquired data.
The results provide a quantitative assessment of the types of distortions that cause the most significant noise increase, depending on their geometry, distance with respect to the fan inlet pane, and ingestion through a contraction. A further analysis of the results will be conducted in the final paper, in order to provide recommendations regarding the allowable inflow quality.