This study is part of an ongoing research project, which aims to increase the physical understanding of the plume formation on Saturn’s icy moon Enceladus. The experimental setup presented in this study is the third iteration of the physical models aiming to recreate the main plume characteristics, where in this study the effects of the channel length, width, type (straight, converging/diverging or diverging), temperature and material are explored and linked to Enceladus’ crevasses. This is done by conduct- ing experiments with 7 different models, where the temperature and pressure are measured along the channels. The models are placed on top of a water-filled reservoir, where the reservoir conditions can be controlled to a limited extent by varying the heating power supplied to the reservoir water. The experimental setup is placed in the test section of the Hypersonic Test Facility Delft (HTFD), which functions as a vacuum chamber.

It is shown that the varying geometry of the channel imposes constraints on the maximum expansion angle before flow separation occurs, the sonic point location and the length/width combination in order to achieve a certain vent Mach number and mass flow rate, although this is also decided by the reservoir and ambient conditions. However, besides the physical properties of the model, there is evidence that the flow properties are dominated by the thermal processes occurring inside the setup. Condensation occurs only in the reservoir, releasing latent heat and making the isentropic flow assumption invalid by definition. It is demonstrated that it is likely that the thermal radiation from the test section of the HTFD onto the model is sufficient to thermally choke the flow. It is unlikely that the flows become choked due to the effects of friction alone. Cooling the models by 10-15°C did not result in significant changes in flow properties, and to have noticeable effects on the flow, the models would have to be cooled to much lower temperatures so that condensation occurs in the channel instead of the reservoir. Although the vapor remains unsaturated in the channels, there are signs of local temperature spikes in the ex- panding sections of the channels, near the vent, when either no or low heating power is supplied to the reservoir water and the pressure in the channel is reduced. This implies that either the temperature at the center of the channel is lower than what could be expected from the temperature measurements and heat is thus released by the deposition of the vapor, or the particles that condensed in the reser- voir partially evaporate after the throat of the channel, after which the evaporative cooling freezes the remainder of the particles, with the accompanying latent heat release. It is also not expected that a pressurized reservoir is necessary to create the plumes on Enceladus, nor is the presence of a geomet- rical throat, due to the combined effects of friction and condensing vapor. The results of one physical model are compared to a computational fluid dynamics model using the same geometry, and the main difference between the physical and numerical model is that the vent pressure of the numerical model is approximately half the vent pressure of the physical model, and the temperature of the numerical model dropped to about -50°C at the expanding section of the channel where the temperature only increased throughout the channel for the physical model. This, and the small heat spike near the vent under low-power conditions, has questioned the accuracy of the temperature measurement method and further research would be required to improve this accuracy.