Mode detection in turbofan inlets

An inverse technique for determining the mode amplitudes generated by turbofan inlets both for tonal and broadband noise is proposed using pressure measurements made in the near-field. The motivation of this research is to make use of the Turbulence Control Screen (TCS). This TCS offers a useful platform for locating microphones to implement a non-intrusive inverse technique since it is often fitted to aero-engines during ground testing to remove the integrated flow. The knowledge of such model content is very useful for characterizing source mechanisms of broadband noise and for determining the most appropriate mode distribution model for duct liner predictions and for sound power measurements of the radiated sound field. The near-field sound pressure radiated from a duct is modelled by directivity patterns of cut-on modes. The resulting system of equations is ill-posed and it is shown that the conditioning of the inverse problem, which depends greatly on the positions of the microphones, is important in assessing the sensitivity of the modal solution to measurement noise and thus the modal reconstruction accuracy. An optimal array geometry for robust inversion is investigated. It is then shown that the presence of modes with eigenvalues close to a cut-off frequency results in a poorly conditioned directivity matrix. A physical interpretation of the Singular Value Decomposition (SVD) of the directivity matrix throws light on the understanding of the issues of ill conditioning as well as the detection performance of the radiated sound field by a given sensor array. The detection of broadband modes generated by a laboratory-scaled fan inlet is performed using the optimal array geometry. This experiment provides a milestone for detecting modal content of broadband noise produced by real fan inlet engines.