[mayank199325]

Hi
How can i apply boundary conditions at the interface(moving) between fluids in VOF model.


for ex. the interface should always have certain value of a scalar

[Madeinspace]

Originally Posted by mayank199325

Hi
How can i apply boundary conditions at the interface(moving) between fluids in VOF model.


for ex. the interface should always have certain value of a scalar

Hi Mayank,


It has been quite sometime but if you recall this, were you able to implement the BC on the interface? Could you please suggest?


Thanks.

[vinerm]

Usually, no boundary condition is required to be applied at the free surface or interface between two fluids in VOF. All the fluxes are conserved at the interface and boundary specification could lead to conservation issues. However, if this needs to be done, then it has to be applied either as a source term or as flux since the interface is not a real boundary.

[Madeinspace]

Hi Venerm,


Thanks for your reply. And I agree with you. I have applied BC on the interface as a source term in the past. e.g. heat source on the metal surface as a surface heat flux. But in this particular case, I can't implement it as a flux on the surface. And in order to solve this equation, I know the condition that should be satidfied on the interface. Well, in literature people have computed by applying this BC on top of the domain but it gives very small magnitude of the forces on the interface. I have detailed my issue with graphics, equation and BC on this post (at the end). Would you please have a look at it?


Interface boundary conditions in chtMultiRegionFoam


Thanks!

[vinerm]

It appears that the meaning of the interface is different for both of us. In my post, interface implies fluid-fluid interface, such as, free surface of water in a partially filled vessel. In your post, you seem to be talking about Fluid-Solid Interface, which is a real wall, a real physical boundary for the flow. It is easier to apply condition for a scalar on physical boundary than on fluid-fluid interface. I do not know how OpenFOAM handles it but in Fluent you can apply value for a scalar at the fluid-solid interface.

[Madeinspace]

Hi Vinerm again,


Well, It is indeed free surface interface. But at the initial condition, the metal in my case is treated as a fluid with very high viscosity. As temperature increases upto melting temperature and phase change occurs, then we will get real fluif-fluid free surface interface. Ideally, the condition should satisfy on the free surface interface at each time step. But meanwhile my assumption is that the free surface deformation is not large enough to cause significant change in the term I am interested in. That's why I only need to initialize the force and use it as a constant body force in my momentum equation.


Thanks!

[vinerm]

I am afraid I could not understand what do you mean by body force at the interface. Which body force are you talking about?

[Madeinspace]

Sorry for I was not very clear in my earlier post.

In my other post, when I solve the equation of electric potential with the BC on the interface. It eventually generates electromagnetic body force. The concern for now is to solve the equation of electric potential only as the other term to generate electromagnetic force depends on the electric potential.

[vinerm]

Let me rephrase to be clear

You want to apply an electric potential at a fluid-fluid interface. Is that correct?

[Madeinspace]

Yes, that's correct. The gradient of electric potential at the fluid-fluid interface is a known value that I want to apply.


regards,

[vinerm]

Well, gradient implies flux. Now, there are two possible scenarios

1. In one of the phases, say, solid phase, the electric potential field is more or less constant. Then, the life becomes easier. You can simply apply a constant value in whole of that phase. The gradient will be calculated automatically and field calculated.

2. If that is not the case, then you will have to determine value using the gradient since directly apply the value of gradient could cause issues. So, determine the field values from the gradient and then apply the values in each phase around the interface. Or, using divergence theorem, similar to what is done for surface tension force model, convert this into volumetric force and apply source value.

[Madeinspace]

Hi,


Electric potential is not constant in thw whole domain. Electrical conductivity is. Electrical conductivity is needed to calculate electric potential. The latter option using the divergence theorem seems to be doing the trick. I am exploring into it. Thanks.