Improving the performance of aeroacoustic measurements beneath a turbulent boundary layer in a wake flow Stefan Haxter ∗ and Carsten Spehr † German Aerospace Center DLR, D-37073 G¨ottingen, Germany Hartmann, M. ‡ Ocker, J. § Tokuno, H. ¶ Wickern,G. ‖ I. Introduction Experimental measurement and subsequent numerical prediction of the excitation of flat plates or car windows beneath a turbulent boundary layer have become important for the development of novel cars and airplanes. A wavenumber spectrum can be used to define the load on a plat caused by the pressure fluctua- tions on the surface. Wavenumber spectra from measurements are used to validate the numerical predictions of the acoustic and hydrodynamic portions of the pressure fluctuations. When measuring wavenumber spec- tra, the design of the experiment can have a large influence on the outcome. In this paper, the effects of both array design and the application of deconvolution algorithms on the experimental determination of the wavenumber-frequency spectrum are evaluated. The pressure load on a surface beneath a flow can consist of contributions from several characteristic source mechanisms, as summarized by Bremner: 1 first, pressure fluctuations caused by the eddies in a turbulent boundary layer in the flow above the surface; secondly the distinct acoustic pressure field, caused, for instance, by a side mirror in the flow field; and third and fourth, a diffuse acoustic sound field and a broadband rise in the amplitude in the wavenumber domain due to spatially and temporally random events, referred to generally as ”rain-on-the-roof” loading. The importance of each source mechanism depends on the scenario under consideration. In a flight test scenario, the hydrodynamic pressure fluctuations are of importance, since the convective speed of the turbulent eddies in the boundary layer is in the vicinity of the natural oscillation of the fuselage panel. The frequency of this aerodynamic coincidence is nearly the same as the acoustic coincidence on the inner side of the panel facing the cabin. Therefore, the fluctuation energy can be transported efficiently from outside to inside of the cabin. In an automotive scenario, the convective velocity of the eddies is somewhat lower, so that the excitation of a side window by this mechanism is less efficient. The response of a plate to the fluctuating acoustical pressure field comes to the fore. For the numerical prediction of aeroacoustic excitation of a surface, the proportion of energy introduced by each of the mechanisms is of importance. The determination of the contribution of these various energy sources is the aim of many experimental studies, which can themselves then be used to validate the numerical simulations. Several authors have successfully measured the time-averaged pressure wavenumber-frequency spectrum below a turbulent boundary layer. Arguillat et. al. 2, 3 used a rotational array to increase the number of ∗ PhD student, Institute of Aerodynamics and Flow Technology - Experimental Methods, stefan.haxter@dlr.de † Research Engineer, Institute of Aerodynamics and Flow Technology - Experimental Methods, carsten.spehr@dlr.de ‡ Volkswagen AG, Letterbox 1777, D-38436 Wolfsburg, Germany § Dr.-Ing. h.c. F. Porsche AG, D-71287 Weissach, Germany ¶ Daimler AG, D-71059 Sindelfingen, Germany ‖ Audi AG, D-85045, Ingolstadt, Germany 1 of 17 American Institute of Aeronautics and Astronautics