Momentum boundary-layer characterisation from a pulsed impinging jet

Abstract

The development of the unsteady boundary layer produced from an impinging jet with a temporal puslation forms part of the complex flow physics occurring within an internal combustion engine and its characterisation is crucial for improved modelling with lower-fidelity tools. The momentum boundary layer prediction of these flows currently relies on wall-modelling based on the logarithmic law of the wall for equilibrium zero pressure gradient boundary layers. However, it is not fully understood how these non-equilibrium boundary layers evolve and whether the reliance on the law of the wall is justified for modelling such flows. Thus, in this paper, we attempt to characterise the momentum boundary layer developing from a pulsed impinging jet through the use of highly-resolved large eddy simulations of a canonical setup. The simulations are conducted for three peak jet Reynolds numbers of 10,000, 20,000 and 50,000, achieved at the peak of the temporal pulse. The resulting boundary layer produced by the jet impingement is examined for the mean flow, the scaling of the outer-layer and the boundary layer parameters, such as the boundary layer thickness, the friction Reynolds number and the wall-scaled profiles. The effect of the jet Reynolds number on the different parameters is also evaluated statistically as well as temporally. The results show that the boundary layer’s development is highly unsteady, even statistically, with a large variation in the friction Reynolds number over the statistical sampling period. However, most of the values were found to be lower than the values seen for turbulent boundary layers. This leads to an at least 40% overprediction in the wall-scaled velocity when using the equilibrium law of the wall for modelling these boundary layers. The predictions of the wall-friction using the equilibrium law of the wall and select turbulence models was also tested and it was found that the latter could be useful as a recourse to the near-wall modelling of these flows.