Top view:
The flow considered is fully developed flow through a square-sectioned straight duct, the side walls of which are roughened with square-sectioned ribs in a staggered arrangement as shown in the figure. Data are available for both the stationary and rotating cases
The experiment was conducted at a Reynolds number, based on bulk streamwise velocity Um
and duct height D, of 105. In the rotating case, the rotation number Ro=
D/Um is
0.2. Geometric parameters of the duct are:
| Duct height and width: | D=5cm |
| Pitch to height ratio: | P/h=10 |
| Rib height: | h/D=0.1 |
The heat transfer measurements were made in a separate experimental apparatus, at the same Reynolds number, with air as the working fluid.
The measurements were obtained as part of a study on the flow and heat transfer in rotating square-sectioned U-bends with rib-roughened walls. Since the flow through the ribbed passage develops rapidly, fully developed periodic conditions, where the flow is unaffected by the presence of the bend, are obtained both upstream of the bend and by about 6 diameters downstream of the bend exit.
The velocity measurements were obtained using LDA in a U-bend mounted on a motor-driven turntable in a water tank. Measurements were taken for the stationary case and for rotation numbers of ±0.2, although for the fully-developed conditions considered here, the positive and negative rotations are equivalent.
Heat transfer measurements were made in a separate apparatus, at the same Reynolds number but in an air flow, by heating the side walls of the duct and using liquid crystal techniques to obtain the surface temperatures. Measurements are currently available only for the stationary case, but it is hoped that data for the rotating case in water (Pr=7) may also be available before the workshop.
Since the flow is fully developed, one need only consider a domain extending, for example, from the leading edge of one rib to the leading edge of the next. Periodic boundary conditions can then be applied between the upstream and downstream boundaries (in the case of temperature, account has to be taken for the heat input into the flow through the wall). The flow is symmetric about its vertical mid-plane, so only the upper (or lower) half of the duct need be computed.
For the thermal boundary conditions, note that only the duct side walls were heated, whilst the top and bottom walls were insulated. A constant heat flux should therefore be applied through the duct side walls, and adiabatic conditions applied on the top and bottom walls and the rib surfaces.
We plan to compare profiles of mean velocity and Reynolds stresses at a number of stations in the symmetry plane, in addition to the distribution of the local heat transfer coefficient on the walls.
Velocity field data:
case7_3a.tar.gz (for gunzip)
case7_3a.tar.Z (for uncompress)
Heat transfer data:
case7_3b.tar.gz (for gunzip)
case7_3b.tar.Z (for uncompress)
Iacovides, H., Jackson, D.C., Ji, H., Kelemenis, G., Launder, B.E., Nikas, K. 1996 "LDA study of the flow development through an orthogonally rotating U-bend of strong curvature and rib roughened walls" Paper No. ASME-96-GT-476 International Gas Turbine and Aerospace Congress, Birmingham, UK.
For some recent computations, see
Iacovides, H., 1997 "Computation of flow and heat transfer through rotating ribbed passages" Proc. 11th Turbulent Shear Flows Symposium, Grenoble.