compressibleInterFoam versus sonicLiquidFoam
I am interested in simulation of pressure waves propagation in the two-phase flow, e.g. pressure wave propagetes in a liquid first and then hits a liquid-gas interface. compressibleInterFoam looks like a suitable candidate for this problem. To start with I set up a simulation of a pressure wave propagation though a liguid phase only, and run it with both compressibleInterFoam and sonicLiquidFoam (that is designed to propagate pressure waves in liquids).
Physical problem is a sphere with two pipes (top and bottom) opened to the atmosphere filled with an ambient water. Gravity was set to zero and alpha1 for the compressibleInterFoam simulation was set up through the boundary conditions (air can enter the pipes, but only water cal leave the pipes). At a particular time a pressure pulse was applied to some part of the wall of a sphere and pressure waves propagation from the walls of the sphere towards its center was followed. Attached below are pressure contours at two different times for both sonicLiquidFoam and compressibleInterFoam. It can be seen that while at early stages fields look almost identical, once wave start to propagate farther, its shape differes between two solvers, and the one obtained with sonicLiquidFoam looks much more physically realistic. Values of velocity and pressure are not miles away from each other (apart of negative absolute pressure seen with sonicLiquidFoam, in the compressibleInterFoam a minimum pressure is constrained by Pmin). Does anybody have an idea what is the reason for such differences? Is there difference in the goverment equations between sonicLiquidFoam and compressibleInterFoam for each phase.
Attached figures are:
1. Pressure Contour plots at t=t_1 with sonicLiquidFoam
2. Pressure Contour plots at t=t_2 (t_2>t_1) with sonicLiquidFoam
3. Pressure Contour plots at t=t_1 with compresssibleInterFoam
4. Pressure Contour plots wt t=t_2 with compresssibleInterFoam
|All times are GMT -4. The time now is 14:55.|