James A. TenCate
Appl. Res. Labs. and Mech. Eng. Dept., Univ. of Texas at Austin, P.O. Box 8029, Austin, TX 78713-8029
The question of whether one sound beam can interact with another at nonzero angle and scatter nonlinearly generated sound outside the mutual interaction region has been debated since the 1950s. Experimental work on this problem has left the question unresolved. This presentation describes experimental difficulties associated with measuring scattered sound produced by real diffracting primary beams. Optimal conditions for observing scattered sound, as outlined by Berntsen et al. [J. Acoust. Soc. Am. 86, 1968 (1989)] and by Darvennes and Hamilton [J. Acoust. Soc. Am. 87, 1955 (1990)], are reviewed in relation to the design of our own experiments. Our experiments were performed with either two uniform circular sources in water (megahertz frequencies), or with one circular source and the other a shaded source with lower sidelobes. A variety of primary frequency ratios, interaction angles, and other parameters were considered. Comparison of the primary beam patterns with measured sum and difference frequency field patterns reveals the difficulty in identifying which components of the latter correspond to ``scattered'' sound. We conclude that two Gaussian-type sources with exceedingly good sidelobe suppression are needed to perform a reasonable experiment. [Work supported by the Packard Foundation and ONR.] [sup a)]Present address: Earth and Environmental Sciences Division, Mail Stop D443, Los Alamos National Laboratory, Los Alamos, NM 87545.