No abstract

The simple model of pressure pulse propagation in slug flow proposed by Henry, Grolmes, and Fauske has been extended by considering wave reflection and wave transmission at gas-liquid interfaces. A frequency-response model applied to a series of idealized gas and liquid slugs yields a pulse propagation speed that approaches the homogeneous model value as the number of slugs is increased for a given void fraction. All characteristic roots from the solution to a three-equation drift-flux model are related to the velocity of the center of mass of the mixture. The pulse propagation speed relative to this velocity is exactly equal to the homogeneous model value, however. Measured pulse propagation speeds in vertically downward slug flow are, as anticipated, much less than those predicted by the simple model of Henry, Grolmes, and Fauske, but slightly greater than the homogeneous model value. Measured pressure surges produced by the rapid closure of a downstream valve in a pipeline are reasonably well predicted by the drift-flux model. For the range of void fractions, pressures, and velocities encountered in this study, it is concluded that pressure pulse speeds and the magnitude of pressure surges in slug flow can be adequately predicted by a homogeneous model.

Note: On Tuesday, October 31, the afternoon session of the Sheet Metal Forming symposium at the TMS/ ASM Materials Week in Cleveland, Ohio, (October 2~No vember 2, 1995) will focus on high·velocity deformation and forming by electromagnetic force.High-velocity metal forming was the focus of much research approximately 30 years ago. It appears that this technology responds well to many of the challenges facing sheet metal forming today: formability is enhanced, springback is minimized, and die tryout time should be minimized.

The rolling motion of a sphere on a smooth plane boundary in a simple-harmonic water motion has been analytically and experimentally investigated. For spheres having specific gravities ranging from 0·09 to 15·18 the sphere motion was found to be sinusoidal for both low and high values of the period parameter defined by Keulegan & Carpenter. The knowledge of the sphere motion, and hence the resultant force, allowed the determination of inertia and drag coefficients from Fourier-averaging techniques. Experiments in the inertial range yielded an added-mass coefficient of 1·2, compared with 0·67 from inviscid theory for translating spheres. For values of the period parameter greater than 30 the drag coefficient is reported to be approximately 0·74.

No abstract