Flow Visualization and Turbulence Measurements in a Three-Dimensional Turbulence Wall Jet
Paper i proceeding, 1991

The modelling of turbulence has recently been directed towards the handling of anisotropic flow fiends, i.e. the interest has been focused on the use of models base on the transport equations for the Reynolds stresses (RST models). To improve these turbulence models well-defined, simple and fundamental experiments are needed in which gradients of the different turbulence parameters are determined. Together with direct simulations of the Navier-Strokes equations these experiments yield a good base for the improvement of different terms in the Reynolds equations. A fundamental, well-defined and simple flow case is the three-dimensional wall jet, where an interaction between a wall boundary layer and a free shear layer forms the anisotropy as well as the inhomogeneous character of the flow field. The purpose of the present work was to study the turbulence field of a three-dimensional wall jet, without an outer disturbing flow field. This was accomplisher by using smoke visualization and hot-wire technique for mean velocity and turbulence measurements. In the present measurement a Reynolds number104 was used, and the extension of the measurements in the flow direction was in the range x/h = 25 through x/h = 62 (x – coordinate in the flow direction and h – slot height.) Preston tubes were employed for the determination of friction of the friction velocity. The spreading rates in the normal as well as lateral direction were visualized for a rectangular and a circular orifice. Different sizes of the vertical wall perpendicular to the outlet were also tested. All spreading rates were determined for the outer boundary of the smoke plumes. It was found that the normal spreading rate was influenced by the wall size as well as the orifice geometry. High wall and circular orifice implied a reduction of the normal spreading. In the lateral direction, no significant difference could be noticed due to either wall size or orifice geometry. Profiles of mean velocities and turbulent quantities were determined at five positions on the centre-line. At three of these positions, equally spaced, the mean velocity and turbulence intensity were also determined at positions off the centre-line to enable the determination of the lateral spreading. From the measurements the normal spreading rate, based as usually on the half-width, was determined to 0.055 and the lateral spreading was estimated in the same way to 0.27. Considering the turbulence measurements, the typical two-point maxima can be found in the normal stress component. A comparison of the profiles for different orifice geometry yields that the orifice not seems to have any significant influence on the turbulence intensity. From the off centre-line measurements it can be noted that the profile shape is approximately the same as on the centre-line. Cross-wire technique was employed for the determination of shear stress profiles along the centre-line.


Martin Eriksson

Chalmers, Institutionen för termo- och fluiddynamik

Bert Johansson

Chalmers, Institutionen för termo- och fluiddynamik

Lennart Löfdahl

Chalmers, Institutionen för termo- och fluiddynamik

First European Fluid Mechanics Conference in Cambridge September 16-20


Strömningsmekanik och akustik