Comparison of techniques to measure the low wavenumber wall pressure spectrum of a turbulent boundary layer

The agreement between low wavenumber measurements of the turbulent wall pressure spectrum by various investigators has not been too good. Different techniques have been used in different facilities and have yielded data in different nondimensional frequency and wavenumber ranges. The current measurement program has utilized the wavenumber filtering techniques used by three primary investigators, the Martin plate, the Jameson plate and the Farabee and Geib 6-element microphone array in the same facility. Also a 12-element collinear and stagger array and a lateral array were used. The agreement between the different techniques in the M.I.T. facility is generally good. In addition, the difference of results between facilities has been reduced by displaying the data in a non-dimensional form that does not include the effect of boundary layer thickness. However, there was still a significant scatter in the data for each technique. Scatter is believed to be due to variation of each filter's wavenumber response away from the main acceptance lobe where the filter is responding to the acoustic and convective contamination and also the variation of the contamination with flow speed.