A gas in a tube spontaneously oscillates when the temperature gradient applied along the wall of the tube is higher than the critical value. This spontaneous gas oscillation is caused by the thermal interaction between the gas and the tube wall. The stability limit of the thermally induced gas oscillation is numerically investigated by using the linear stability theory and a transfer matrix method. It is well known that an acoustic wave excited by the spontaneous gas oscillation occurring in a looped tube is different from that in a straight tube with two ends; a traveling acoustic wave is induced in a looped tube, whereas a standing acoustic wave is caused in a straight tube. The conditions for the stability limits in both tube types were calculated. The calculated and measured conditions were compared and were found to be in good agreement. Calculations performed by varying the value of the Prandtl number of the gas were used to determine the reasons for the existence of the stability limits of the looped and straight tubes.

An acoustic field spontaneously induced in a thermoacoustic prime mover consisting of a looped tube and resonator is determined through simultaneously measurements of pressure P and velocity U. A thermal efficiency of the thermoacoustic prime mover of this type has been reported to reach 30%. The measurements of the acoustic field in the present system revealed that a phase lead of U relative to P takes a negative value of about −20° in the regenerator where the output power of the prime mover is generated. It was concluded that the possession of a negative phase lead at this position is taken as a clue in a significant increase in the output power. Moreover, the analysis in the thermoacoustic mechanism shows that a precise position for the location of a second regenerator acting as a heat pump exists in the looped tube. Indeed, by locating the second regenerator at the position, a thermoacoustic cooler was constructed. The thermoacoustic cooler could generate a low temperature of −25 °C without involving any moving parts.

Two-sensor method proposed by Fusco et al. [“Two-sensor power measurements in lossy ducts,” J. Acoust. Soc. Am. 91, 2229–2235 (1992)] is a novel technique that determines acoustic intensity of a gas column in a wide duct from measurements of pressure based on the boundary layer approximation. For further development of this method, its validity is experimentally tested through comparison with the direct method measuring the pressure and the velocity simultaneously, and its formulation is modified to include the narrow duct range where the duct radius is smaller than the viscous boundary layer thickness of the gas. It is shown that the modified two-sensor method enables quick and accurate evaluation of the acoustic intensity seamlessly from narrow to wide duct ranges.

This paper reports on the acoustic streaming in a looped-tube thermoacoustic prime mover equipped with an asymmetric constriction called a jet pump. The time-averaged mass flow velocity was determined using visualization methods and using acoustic field measurements. It was demonstrated that the magnitude and the direction of the velocity were dependent on the orientation of the jet pump. From the observed velocities, we estimated the heat carried away from the hot heat exchanger by the mass flow. It was shown that the heat loss was decreased from 30 W to 6.5 W by reversing the orientation of the jet pump, when the input heat power supplied to the prime mover was 100 W. The influence of the acoustic streaming was also studied from the cooling temperature of the looped-tube cooler.

The experimental evaluation of the wave number and characteristic impedance of stacked-screen regenerators is described. First, a two-by-two transfer matrix of a stacked-screen regenerator was estimated from pressure measurements performed at four different positions; then, the wave number and characteristic impedance of the regenerator were evaluated using a “capillary-tube-based” theory that models a stacked-screen regenerator as an array of pores having a uniform cross section. The evaluation was applied to seven types of stacked-screen regenerators. The experimental results show that these stacked-screen regenerators can be modeled as arrays of circular-cross-section tubes. Moreover, an empirical equation used to estimate the radius of the circular cross section of the tubes comprising the modeled stacked-screen regenerators was addressed.