Scale-model investigation of highway traffic noise barriers

Barriers, such as walls, earth berms, and earth-berm/wall combinations, are used to control highway noise. Factors that affect noise attenuation include profile, height, location, and surface composition. A literature review was conducted on the topics of outdoor sound propagation, ground impedan...

Full description

Bibliographic Details
Main Author: Busch, Todd
Format: Others
Language:English
Published: 2009
Online Access:http://hdl.handle.net/2429/5844
Description
Summary:Barriers, such as walls, earth berms, and earth-berm/wall combinations, are used to control highway noise. Factors that affect noise attenuation include profile, height, location, and surface composition. A literature review was conducted on the topics of outdoor sound propagation, ground impedance, noise barriers, and acoustical scale modelling. The noise attenuation provided by vertical walls, earth berms, and earth berms crested by a vertical wall — for a line source of vehicular traffic — has received little attention. Using an ultrasonic air-jet source, scale-model materials were selected by measuring their Excess Attenuations, at scales of 1:20, 25, 31.5, 40, and 50. A new method was employed to select materials: the residuals between measurements and 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 of 1:31.5 was selected, in conjunction with specifying three model materials to simulate berms and soft ground (expanded polystyrene), vertical walls (dense polystyrene), and roadways (varnished particle board). Berms, typically 4 m in height, were tested with slopes of 1.5, 2 and 3:1; for each of these slopes, five crest profiles were tested: wedge, flat-top (widths of 1 m and 2 m), and round-top (radii of 1 m and 2 m). For berms crested by a vertical wall, the relative heights of berm and wall were either lm/3m, 2m/2m, or 3m/lm; for each height combination, berm slopes of 1.5, 2, and 3:1 were tested. The surface composition of berms were altered using dense polystyrene, and felt. Weighted Insertion Losses (ILAs) for a line source were obtained by applying A-weighted, traffic-noise spectra, before integrating the 80-2500 Hz full-scale, thirdoctave bands. Contrary to current highway practice — where a "soft-top correction" of +3 dBA is often applied to earth berms — walls out-performed berms by 1-2 dBA. For berms, shallower slopes degraded third-octave ILs in the 250 Hz band, but had little effect on a berm's ILA. For berms crested by a vertical wall, shallower slopes increased ILAs. === Applied Science, Faculty of === Mechanical Engineering, Department of === Graduate