Abstract:
Scale-model roadside barriers were studied, using an air-jet to recreate a line source of traffic. Scale-model materials were selected by measuring their excess attenuation, at scale factors of 1:20, 25, 31.5, 40, and 50. For each material, the residuals between a measurement and a theoretical best-fit curves were calculated for each cell of a 2-D array whose axes were scale factor versus effective flow resistivity. An optimal scale factor of 1:31.5 was selected, in conjunction with the selection of three model materials to simulate earth berms and soft ground (150--300 cgs rayl), and vertical walls or roadways (20 000 cgs rayl). Roadways were simulated by varnished particle board, walls by dense polystyrene, and both earth berms and soft ground were simulated using extruded polystyrene. The insertion losses, with and without an A-weighted traffic-noise spectrum, were determined for 38 different noise-barrier configurations. Walls outperformed earth berms by 2 dBA; for earth berms, shallower slopes degraded insertion losses, but when a wall was present the opposite trend was observed. One of the most relevant references is [Hutchins et al., J. Acoust. Soc. Am. 75, 1817--1826 (1984)]. [Work supported by the B. C. Ministry of Transportation and Highways.]