Prediction and reduction of internal blade-passing frequency noise of the centrifugal fan in a refrigerator

Abstract

The internal blade-passing frequency (BPF) noise of a centrifugal fan in a household refrigerator is computed using a hybrid method. The unsteady flow field of the centrifugal fan in a duct is predicted by solving the incompressible Reynolds-averaged Navier–Stokes equations with conventional computational fluid dynamics techniques. The principal sources of noise are then extracted from the predicted flow field through the acoustic analogy. Finally, the internal BPF noise is predicted using the modeled sources in combination with the boundary element method. A parametric study using this hybrid technique shows that the BPF noise levels predicted for various non-dimensional cutoff distances closely follow the experimental data. Based on this result, a low-noise design for the centrifugal fan is proposed. A BPF noise reduction of approximately 3 dB is achieved in comparison with the original model, which is in good agreement with the prediction using the hybrid techniques.