Measurement, prediction, and reduction of the transmission of separated flow noise through panels

The transmission of separated flow noise through a panel was studied. Separated flow noise is important in the understanding of automobile wind noise. A broad investigation of several aspects of the problem was made. One phase of the research consisted of measurements of the wall pressure fluctuations beneath turbulent and separated flows. Since separated flows represent the source of most automobile wind noise, this data is crucial, and this work represents only the third known study of this subject. The vibration of the plate excited by a separated flow was measured using a miniature accelerometer in another phase of the research. The total damping of a plate bonded to a frame with urethane adhesive was also measured. A major portion of the project entailed developing a bench scale flow noise facility using the Upson Hall wind tunnel as a flow source. Several hundred windows and other panels were tested in the facility. This data was shown to correlate well with full-scale automobile data taken in a much larger wind tunnel. A basic model for the damping at the edge of a plate supported by a bead of adhesive was given. It was shown that edge damping can be significant. A model for predicting the damping level of laminated glass was also developed, and a way to deal with a dual interlayer was also shown. Practical considerations in the development of advanced laminated glass windows were discussed. An empirical model to predict the acoustic transmission loss of a single layer panel was also developed. A model to predict the overall transmission of flow noise was presented, which agreed well with experimental results. The research provides a broad foundation for further research in automobile wind noise.