Experimental Validation of an Analytical Prediction Model for Fan Buzz-Saw Noise

Abstract This paper is focused on the application of an analytical model, which was previously developed and published for the prediction of buzz-saw noise emitted by supersonic fan rotors. The application test cases are six different fan stages equipped with hardwall inlet ducts and clean inflow conditions. The model is validated through a comparison with experimental sound power levels measured along the working line of each fan in high subsonic, transonic and supersonic regimes. In terms of overall evolution and maximum levels, the model provides a reasonable prediction; the sensitivity to parameters such as the intake axial Mach number or the intake duct length is well reproduced. However the significant discrepancies observed locally show that the prediction of buzz-saw noise is still a difficult task. The model is then extended to the case of distorted mean flow impinging on the rotor. Following this extension, the maximum levels of buzz-saw noise are expected to occur over a wider range of Mach number compared to the case with clean inflow.