Index of content:
Volume 119, Issue 3, March 2006
- ULTRASONICS, QUANTUM ACOUSTICS, AND PHYSICAL EFFECTS OF SOUND 
119(2006); http://dx.doi.org/10.1121/1.2168427View Description Hide Description
The sound attenuation at ultrasonic frequencies caused by small spheroidal particles in a fluid is examined with regard to the size parameters that determine the shape of the attenuation spectrum. Our investigations are based on a coupled phase model for spheroids with arbitrary orientation, thus facilitating the calculation of average attenuation for a given orientation distribution. Since the model just considers the visco-inertial coupling, its applicability is restricted to small solid particles with high density contrast. The calculated attenuation spectra of mono-sized, randomly aligned spheroidal particles are compared with the attenuation spectra of mono-sized spheres. When the latter approximate the former to a reasonable degree the size of the spheres is called attenuation equivalent diameter. It is shown that the concept of attenuation equivalent diameter can be applied only to slightly elongated prolates. Oblates and very stretched prolates yield considerably broader attenuation spectra than mono-sized spheres. While for oblates and for slightly elongated prolates the characteristic frequency of the attenuation spectrum is determined by the volume specific surface, no such attenuation determining size parameter could be identified for very stretched prolates.
119(2006); http://dx.doi.org/10.1121/1.2161450View Description Hide Description
The longitudinal and shear wave phase velocities and attenuation are measured as function of frequency for random particulate composites, consisting of near mono-disperse spherical glass particles imbedded in a homogeneous epoxy matrix. Experimental results are compared to two theoreticalscattering models: the Waterman and Truell method (WT) and the dynamic generalized self-consistent model (DGSCM). It is observed that both models work well in the case of a low inclusion concentration. At higher concentrations both the WT and DGSCM matched the phase velocity, in most cases, within the range of experimental error of about 3%. The WT agrees with the attenuation data only at low concentrations, and the disparity between theory and measured values increase with increasing concentration. The DGSCM agrees well with the attenuation data.
119(2006); http://dx.doi.org/10.1121/1.2165000View Description Hide Description
The characterization of the thermal properties of regenerator materials has traditionally been studied by measurement of the heat transport of a gas passing through a regenerator flanked by two heat exchangers. The results for heat transfer between the gas and regenerator are typically expressed as a nondimensional Nusselt number. The newer thermoacoustic perspective, developed over the past two decades by Swift and his collaborators, has characterized the thermal properties of regenerators through the thermoviscous function used by Rott in the 1960s. This paper derives the relation between those two perspectives. The complex Nusselt number for oscillatory flow is calculated in terms of the thermoviscous function. The asymptomatic results of the real part of the complex Nusselt number at zero frequency converge to the steady (direct current) result and at high frequency it matched qualitatively with the limit derived by other researchers interested in the behavior of heat exchangers used in oscillating flows.
119(2006); http://dx.doi.org/10.1121/1.2161429View Description Hide Description
One implementation of the vibro-modulation technique involves monitoring the amplitude modulation of an ultrasonic vibration field transmitted through a cracked specimen undergoing an additional low frequency structural vibration. If the specimen is undamaged and appropriately supported, the two vibration fields do not interact. This phenomenon could be used as the basis for a nondestructive testing technique. In this paper, the sensitivity of the technique is investigated systematically on a set of mild steel beams with cracks of different sizes and shapes. A damage index was measured for each crack. The correlation obtained between the crack size and the strength of the modulation is fairly poor. The technique proved extremely sensitive to the initial state of opening and closing of the crack and to the setup due to the modulating effects of contacts between the specimens and the supports. A simple model is proposed which explains the main features observed and approximately predicts the level of sideband obtained experimentally.