The myth of the re-entrant jet

The re-entrant jet flow model is one of the options of wake closure in the Helmholtz/Kirchhoff free-streamline theory. It is a mathematically consistent steady flow model, without a physical equivalent because mass is not conserved. However, since numerous experimentalists have reported a liquid stream penetrating under a sheet cavity which they have identified as a re-entrant jet, it has played an important role in the discussion of dynamic cavitation behaviour. It is said to be responsible for the dynamic behaviour by impinging on the forward end of the cavity, upon which the cavity breaks and a vapour cloud is shed. In this paper we use numerical flow simulations to show that the re-entrant jet model is not an adequate description of the phenomena occurring, in spite of the apparent experimental evidence. These simulations have been carried out for the NACA0015 foil at 6 degrees angle-of-attack, i.e. a flow free of boundary layer separation in non-cavitating conditions. We demonstrate that partial cavitation soon after the inception phase invokes flow separation. Partial cavitation is therefore as a rule accompanied with a flow recirculation zone. This flow recirculation zone is present while the cavity is more or less stationary as well as during the major part of a shedding cycle in dynamic conditions. For developed shedding, the flow separation zone is swept from the suction side of the foil during each cycle. The flow phenomena found under dynamic conditions have therefore significant similarity with dynamic stall phenomena. The paper concludes that the story of the reentrant jet as it has repeatedly been described in the literature on dynamic cavitation is very likely based on a visual illusion. Experimentalists are advised to reconsider and re-interpret their observations and results with a more realistic model of the underlying flow.