Model Transitions in Multiphase CFD: From DPM to VOF and Eulerian Wall Film in Ansys Fluent

Hello, today I would like to show how to implement a model simulation where we basically have a VOF converting to Eulerian wall film and VOF.

So basically what we have here for our geometry is a U-shaped bent pipe, pretty much, a short section of relatively small mesh where we are injecting air from one side and we get air from the other side as well as some DPM particles, in this case our DPM droplets of water, small droplets of water that can circulate through this pipe and can interact with the walls of the pipe and or you can use the same method by using a water liquid.

I have created a material for that, basically copying the same property with water liquid, that I am using in my DPM model. So the diameter is 50 microns, and basically injects for the whole course of the simulation, the transient simulation, with a small velocity magnitude as well.

This is the setup, fairly simple in terms of the DPM model. We also have to enable a multiphase volume of fluids model. This is what I've added here. We are using volume of fluids with implicit formulation and sharp dispersion interface modeling.

The phases are basically air, which is my primary phase, and my secondary phase is water-liquid, which is the same property as the DPM particles. And then phase interaction is very important when we set up basically the transition from one model to another, from in this case DPM to VOF.

In this case the transition criteria for particle in VOF liquid So basically you can set up the particle without secondary break up and particle with secondary break up as well. And you can activate adaptation during transition as well.

So, these are some of the properties basically for setting up this DPM to VOF model. And I click Apply here. And another important model is the Eulerian Wall Film, the EWF.

The Eulerian wall film can be converted to VOF We solve for momentum here, so first we have to enable Eulerian wall film So the Eulerian film model, when we were solving the momentum, the DPM coupling and the phase coupling.

And basically once we activate the phase coupling and DPM coupling, we have these two tabs here, the DPM interaction, where we define the critical Weber number and the critical angle, and the separation model.

With edge separation enabled, and the phase interaction, basically we have VOF coupling, Transition based is on volume fraction. The solution method here is what basically we specify to resolve the Eulerian wall film on the walls.

So basically we're using first order for time, continuity, and momentum. and we also basically apply some time marching time step controls basically to ensure convergence.

Another important feature is the maximum thickness, so this is the thickness that basically is the maximum thickness that the Eulerian wall film will model. Once you get into that thickness, you get a conversion to VOF.

So basically these are some of the setups for the EWF model and once we have that setup we can basically define our boundary conditions. So basically it's important also to notice the materials before you actually go to the boundary conditions.

So the fluid that we have is air and water liquid and basically these are my primary and secondary phases My inert particle has to have the same properties as my water liquid shown here, including surface tension, viscosity, density, and here we have the properties density and viscosity shown here.

and basically for boundary conditions, now going to boundary conditions, we have basically one inlet where we are injecting air.

We don't have water at all in the inlet, so water has a volume fraction of zero, but air is my primary phase and I'm defining a velocity inlet, a very small velocity inlet of 0.1 m/s. I also have to define an outlet, which is just a pressure outlet at this side here, shown in red.

And we also have a wall where we define where basically the Eulerian wall film can form. So we have DPM interactions in the wall film here enabled. The Eulerian wall film is defined, so that basically allows us to have the Eulerian wall film forming on that wall.

Once all your model setup is defined, you can basically run. Here I'm running a transient simulation at 1 x 10-4 seconds. Basically this allows me to track the particles efficiently. I've also created a number of contours as well as surfaces to capture the volume fraction at different levels.

I also have defined some scenes to define for example the EWF. Here we can see after running the simulation what the EWF looks like. We can see that there is a lot of, basically the wall film thickness is larger near the bottom.

This is a VOF of 0 so we can also look at the DPM This is the DPM only and we can see that the particles basically the trajectory that the particles follow they kind of go closer to the wall and interact with the VOF and the wall film in these zones.

If you want to start and integrate Ansys modeling, we will need a built-in schema as well as optiSLang, and other tools.

As they interact with the wall, you can start to see that the wall film, which is the colored walls, are starting to get yellow, so that indicates that the wall film is growing in these zones. And basically to set up this wall film is basically a contour on the walls of the system, of the domain.

So this will take quite a long time, but I've uploaded some of the results. Basically this will allow me to capture the growth of the film on the wall as well as of the water accumulation near the bottom of the domain.

This largely depends on the velocity of the particles, the number of droplets that you are injecting at each time step, and the mass flow as well. So basically this is how to set up a model transition within Ansys Fluent whereby we transition from DPM to EWF and VOF.

Please contact us at https://ozeninc.com/contact for more information.