Analysis of centrifugal convection in rotating pipes

Citation
V. Shtern et al., Analysis of centrifugal convection in rotating pipes, PHYS FLUIDS, 13(8), 2001, pp. 2296-2308
Citations number
8
Categorie Soggetti
Physics
Journal title
PHYSICS OF FLUIDS
ISSN journal
10706631 → ACNP
Volume
13
Issue
8
Year of publication
2001
Pages
2296 - 2308
Database
ISI
SICI code
1070-6631(200108)13:8<2296:AOCCIR>2.0.ZU;2-N
Abstract
New exact solutions, obtained for centrifugal convection of a compressible fluid in pipes and annular pipes, explain axially elongated counterflow and energy separation-poorly understood phenomena occurring in vortex devices, e.g., hydrocyclones and Ranque tubes. Centrifugal acceleration (which can be up to 10(6) times gravity in practical vortex tubes), combined with an a xial gradient of temperature (even small), induces an intense flow from the cold end to the hot end along the pipe wall and a backflow near the axis. To account for large density variations in vortex devices, we use the axial temperature gradient as a small parameter instead of the Boussinesq approx imation. For weak pipe rotation, the swirl is of solid-body type and soluti ons are compact: v(z)/v(za)=1-4y(2)+3y(4) and (T-T-w)/(T-a-T-w)=(1-y(2))(3) ; where y=r/r(w), the subscripts w and a denote values of axial velocity v( z), temperature T, and radial distance r, at the wall and on the axis. The axial gradient of pressure, being proportional to 3y(2)-1, has opposite dir ections near the wall, y=1, and near the axis, y=0; this explains the count erflow. With increasing pipe rotation, the flow starts to converge to the a xis. This causes important new effects: (i) the density and swirl velocity maxima occur away from the wall (vortex core formation), (ii) the temperatu re near the axis becomes lower than near the wall (the Ranque effect), (iii ) the axial gradient of temperature drops from the wall to the axis, and (i v) the total axial heat flux (Nu) reaches its maximum Nu(max)approximate to 4000 and then decreases as swirl increases. These features can be exploite d for the development of a micro-heat-exchanger, e.g., for cooling computer chips. (C) 2001 American Institute of Physics.