Investigation on the impact of hydrodynamic pressure perturbations, induced by a turbulent boundary layer, on sound pressure measurements with wall mounted microphones

Abstract

The the tonal and broadband insertion loss of liners in turbo engines is often determined in the build-in situation, i.e. in a scaled test rig or real engine. The insertion loss is defined as the logarithmic ratio of the sound pressure without the acoustic measure, i.e. hardwall and the sound pressure measured with the liner installed. In case of broadband measurements non-acoustic noise can compromise the measurement results if it exceeds the acoustic signals. For wall mounted microphones such noise is mainly due to hydrodynamic pressure fluctuations induced by the turbulent boundary layer. In order to determine the insertion loss correctly the acoustic fraction of the measured signal needs to be separated from hydrodynamic pressure perturbations. In order to make a first qualitative and quantitative statement on the hydrodynamic signal components, the spectrum of the hydrodynamic pressure fluctuations will estimated. For measurements on flat walls and circular pipes, there are empirical models that describe the spectrum of pressure fluctuations. In addition to the flow velocity and the different boundary layer thicknesses, the shear stress, friction velocity are required as input parameter for these models. In a first step, methods for determination of the theoretical signal-to-noise ratio of the acoustic signal components against the hydrodynamic pressure fluctuations induced by the turbulent boundary layer are investigated. In the next step was investigated experimentally the reduction of the hydrodynamic perturbations by means of acoustic treatment and changes in the probe design. Furthermore, the use of empirical models for the prediction of the turbulent spectrum was analyzed and existing analytical techniques used for the separation of hydrodynamic and acoustic components. This report summarizes the results of the work.