Boundary Layer in a Concave Bend

Developing boundary layer submitted to destabilizing concave curvature at \(Re_{\theta} = 1450\) without longitudinal pressure gradient (\(S = \partial p/\partial x \approx 0\)).

The test facility is a free surface water channel, shown schematically in figure 1, consisting of an asymmetric 4:1 contraction, a 4.88 m long straight development section with dimensions 25 cm (width) by 1.25m (span), a 90 degree constant radius curved wall (\(R = -136\) cm), and a straight recovery section. The channel walls are contoured to minimize pressure gradient in the straight development section and along the concave test surface. The boundary layer is tripped 1 m downstream of the nozzle exit using a 0.476 cm square rod. Spanwise surveys show good spanwise uniformity; specifically, the roll cells which appear in the curved section are not fixed in space, but are randomly distributed in space and time.

Experimental arrangement Fig. 1: Experimental arrangement

LDV measurements of all three velocity-component were made down to \(y^+ \approx 5\); \(u'\) and \(w'\) are at even smaller \(y^+\) at a number of streamwise locations, as indicated in figure 2.

One set of measurements were taken with no grid-generated turbulence present (without the grid shown in figure 2). For the other sets, a square, biplanar, square bar grid (mesh spacing \(M = 6.35\) cm; ratio of mesh spacing to bar width \(M/w = 4\)) was placed at various locations upstream of the curved wall and used to generate nearly axisymmetric turbulence. In most cases the free-stream turbulence levels were nominally \(u'/U_{pw} = 0.05\).

Flow geometry Fig. 2: Flow geometry and measurement locations

Velocities measured using LDV.

For most cases profiles are available at a number of streamwise locations around the bend for

  • Mean velocity
  • Reynolds stresses
  • Triple moments, skewness, flatness

Sample plots of selected quantities are available.

The data can be downloaded as compressed archives from the links below, or as individual files.

The file readme.txt contains a description of the files and data formats.

xdata.dat contains a summary of the channel width as a function of streamwise distance, and values of wall skin friction at measurement locations.

Cases Without Turbulence-Generating Grid
Location Mean velocity & Reynolds stresses Triple moments
\(x=-56\) cm fn1.dat fn2.dat
\(x=35.5\) cm (\(15^o\) station) cn151.dat cn152.dat
\(x=71\) cm (\(30^o\) station) cn301.dat cn302.dat
\(x=106.5\) cm (\(45^o\) station) cn451.dat cn452.dat
\(x=142\) cm (\(60^o\) station) cn601.dat cn602.dat
Cases With Turbulence-Generating Grids
Location Mean velocity & Reynolds stresses Triple moments Notes
\(x=-56\) cm fg191.dat Grid placed at \(x=-177\) cm
\(x=-56\) cm fg101.dat Grid placed at \(x=-120\) cm
\(x=35.5\) cm (\(15^o\) station) cg151.dat cg152.dat Grid placed at \(x=-127\) cm
\(x=71\) cm (\(30^o\) station) cg301.dat cg302.dat Grid placed at \(x=-92\) cm
\(x=106.5\) cm (\(45^o\) station) cg451.dat cg452.dat Grid placed at \(x=-56\) cm
\(x=142\) cm (\(60^o\) station) cg601.dat cg602.dat Grid placed at \(x=-32\) cm
  1. Barlow, R.S., Johnston, J.P. (1988). Structure of a turbulent boundary layer on a concave surface. J. Fluid Mech., Vol. 191, pp. 137-176.
  2. Johnson, P.L. (1990). PhD thesis, Stanford University.
  3. Johnson, P.L., Johnston, J.P. (1989). The effects of grid-generated turbulence on a flat and concave turbulent boundary layer. Report MD-53, Thermosciences Division, Dept. of Mech. Eng., Stanford University.

Indexed data:

case023 (dbcase, confined_flow)
titleBoundary Layer in a Concave Bend
flow_tag2d, curvature