[rupak504]

Hello,

I am using Implicit VOF multi-phase model, to simulate a phase change problem. Also, i have a square 2D domain with Non-uniform structured quadrilateral mesh, finer at center and coarser towards ends. I am also using variable Time step size method. PISO for Pressure velocity coupling. and all other at at their default values.

I have given a convergence criteria of 1e-04 for continuity equation and max 200 iterations/timestep, but just after phase change phenomenon, the continuity residual goes up at 1e+02 and gets stable over there.



I don't know what is causing this problem, and what could be the solution ?


any kind of help would be appreciated

regards

[vinerm]

Few more details would be more helpful. Is the domain closed or open; what are the boundary conditions? Which mass-transfer mechanism is being used. And is it liquid->gas or gas->liquid mass transfer? If yes, then you have to ensure either the domain is open or the gas is being modeled as ideal gas.

[rupak504]

Originally Posted by vinerm

Few more details would be more helpful. Is the domain closed or open; what are the boundary conditions? Which mass-transfer mechanism is being used. And is it liquid->gas or gas->liquid mass transfer? If yes, then you have to ensure either the domain is open or the gas is being modeled as ideal gas.

Domain is open from top with Pressure outlet boundary condition. All other sides have No-slip BC. Lee model for pool boiling. Liquid to gas mass transfer.

[vinerm]

One possibility is time-step. If the simulation does not converge within 30-40 iterations in a time-step it implies that the time-step is larger than it ought to be. 200 iterations is not suitable. Even if the simulation converges in, say, 80 or 100 iterations, results will appear jaggered over time. Second point is numerics, i.e., space discretization and URFs. If physical setup is good, you will be able to control the run using numerics and time-step.