I am trying to simulate evaporation in unsaturated porous wicks. I have run two set of simulations with different geometries and have incorporated symmetry B.C inside porous region. But behavior of vapor at symmetry is non-realistic. Can anybody tell me about suitability of symmetry B.C in porous media ?
Many Thanks.
Depends on what you want to simulate. In case that your porous wick is in contact with an infinite reservoir a steady-state will develop in which the evaporative flux is balanced by a constant influx at the other boundary. Your moisture profile will not depend on time and at every position in space the sum of the liquid flow and vapor flux should be constant. At the boundary you will have to use a constant flux BC.
In case that your are dealing with a drying porous material not in contact with an infinite reservoir, your moisture distribution will change in time. At the boundary not in contact with air, you should use a no-flux conditions for both the liquid flow and the vapor flux. Effectively this means that gradients in the vapor density and water content should be zero at this boundary.
Depends on what you want to simulate. In case that your porous wick is in contact with an infinite reservoir a steady-state will develop in which the evaporative flux is balanced by a constant influx at the other boundary. Your moisture profile will not depend on time and at every position in space the sum of the liquid flow and vapor flux should be constant. At the boundary you will have to use a constant flux BC.
In case that your are dealing with a drying porous material not in contact with an infinite reservoir, your moisture distribution will change in time. At the boundary not in contact with air, you should use a no-flux conditions for both the liquid flow and the vapor flux. Effectively this means that gradients in the vapor density and water content should be zero at this boundary.
I agree with the previous answer. I want to underline that the more accurate way to enforce the zero gradients at boundaries not in contact with air is the following. Insert two additional points (cedll centers) located outside in a simmetric geometrical location. At these locations you should insert values of vapour density and water content equal to the values at the corresponding points located inside. In such a way the profile of vapour density and water content will have a maximum or a minimum at the boundary thus ensuring a zero gradient at the boundary. Then you perform your computations neglecting the boundaries and using data at the two auxiliary points. Obviously two points are required to obtatin a second order approximation, while only one point is needed if a first order approximation is required.
- Badges
- Science topic