UFR 3-09 Description
Underlying Flow Regime 3-09 © copyright ERCOFTAC 2004
The flow under consideration consists of a turbulent jet issuing from a circular pipe and impinging on a flat plate. A schematic view of the flow is shown above.
The turbulent jet impinging orthogonally onto a plane surface produces, in the vicinity of the stagnation point, among the highest levels of Nusselt number encountered in single-phase convection. This is thus a flow configuration which is extensively used in the process industries to achieve intense heating, cooling (present case) or drying rates. From the point of view of turbulence modellers, the turbulent impinging jet is an excellent test case for validation, since it differs in several important respects from flows parallel to walls which are primarily used to calibrate the model. Indeed, in the vicinity of impingement point:
- turbulence is generated by normal straining (shear in parallel flows);
- the velocity fluctuations normal to the wall are larger than those parallel to the wall (in a parallel flow, fluctuations normal to the wall are much smaller than other components);
- the local turbulent length scales are strongly affected by the length scales of the jet turbulence (in a parallel flow, length scales are usually determined by the distance from the wall alone);
- convective transport of turbulence energy towards the stagnation point is important (in a parallel flow, convective effects are usually negligible);
- just beyond the impingement region, the flow structure will be significantly affected by the strong curvature;
- at greater radii, the flow turns into a radial jet;
- the Nusselt number exhibits a hard-to-predict distribution along the plate, with a global maximum at the stagnation point, and a secondary local maximum or at least a plateau around r/D=2 which appears when the nozzle-to-plate distance is small enough.
Review of UFR studies and choice of test case
The overall heat- or mass-transfer performance of jet-impingement configurations has been examined in numerous experiments. The experiments anterior to 1992 are listed in:
Jambunathan, K., Lai, E., Moss, A., Button, B., L. 1992, A review of heat transfer data for single circular jet impingement, Int. J Heat Fluid Flow 13, 106-115.
Impinging jets are very challenging for modellers: whereas standard models are developed by reference to flow parallel to the wall, the flow is orthogonal to the wall in the impingement region, which leads to all the hard-to-reproduce features described in section 1.
In the particular case selected here, the nozzle-to-plate distance is H=2D, where D is the pipe diameter ; the Reynolds number is Re=23,000. The jet is at ambient temperature and the plate is heated from below (constant heat flux). This particular case, for which flow field and temperature measurements are available, has been computed by many different research teams, from academics as well as from industry, and is widely recognized as a comparison exercise: the prediction of the Nusselt number distribution is particularly difficult, due to the presence of the secondary peak. The experiments were devised for the purpose of evaluating turbulence models: in particular, the length:diameter ratios 80:1 is sufficient to consider that the flow, at some diameters before the pipe exit, is a fully developed pipe flow, which is very convenient for computations, since all the necessary inlet quantities (depending on the turbulence model) can be evaluated from a separate pipe flow computation.
© copyright ERCOFTAC 2004
Contributors: Remi Manceau - Université de Poitiers