The previous research realized that Coanda effect can be used for the thrust vectoring for the vertical-take-off and landing of an aircraft for civil aviation. But, the fundamental flow behavior on the curved surface is little known till date. Therefore, an attempt has been made towards enhancement of the knowledge on the flow behavior on the Coanda effect for the use in VTOL (Vertical Take-Off and Landing) as well as in the various engineering applications where nozzle flow is embedded. In the present paper the computational study of the Coanda effect has been performed at different jet Reynolds number from 12000-24000 using k-zeta-f turbulence model.
It has been realized that due to damping of the normal velocity on the curved surface, the turbulence becomes anisotropic and the k-epsilon model does not take into account the anisotropic effect. Therefore, with k-epsilon model the prediction of the adhesion angle is more than k-zeta-f model. The adhesion angle is very important parameter for the determination of the thrust in along the flow and normal to the flow. It has been also found after grid independence check that, one need to resolve viscous sub-layer properly for the grid independent solution of angle of adhesion. It has been also found that the convective flow (non-linear part in left hand side of Navier-Stokes equation) and viscous stress both are important. Therefore, it has been concluded that there exist the wide range of the turbulent length scale. The effect of entrainment of ambient fluid on the flow has been also investigated and found that normal boundary layer theory (developed for the flow over horizontal wall) does not hold longer. Therefore, the requirement of the reinvestigation of the boundary layer phenomena for the Coanda effect has been realized and the current research makes some contribution into this area.