Issues with Mixing
 

Hi all, I currently am doing my thesis at the University of Sydney based on the simulation of a saline cavity at the bottom of a river being purged by an overflow of fresh water. My initial testing began using a laminar viscous model which did in fact demonstrate turbulence within the boundaries of the channel and cavity. I was able to simulate this process using the bousinessq approximation with both the saline inside the cavity and the water inside the channel having the same density but different temperatures to initiate the process of diffusion (i.e. purging). The driving forces are highly buoyant in this situation!! :confused:

However I have moved on to testing the k-epsilon model which do not seem to give the same results as the laminar solution. I know this is true due to the fact that laminar solution's instability matches those of previous results obtained through numerical solution testing. The k-epsilon models (both standard and RNG) seem to merely create a singular vortex trapped within the cavity which is self perpetuating by the shearing effect of the overflowing fresh water. Therefore there is no ejection of 'saline fluid' (lower temp) above into the channel.

If anyone has any methods on how to properly initiate a turbulent diffusion process between the two fluids in the following problem and or a reason why the k-epsilon models are incapable of achieveing proper purging, please share. It would be most appreciated.