Fluids

Fluids

two separated liquid “phases”

    • BillT
      Subscriber

      Hi,


      I am trying to model a flow of 2 liquids (both water) through a channel filled with air. I want them to spread in different directions (see below), which I tried by defining different momentum source terms. In my simulation the two water phases always mix up and flow downwards only. I set up the eulerian multiphase model with 3 eulerian phases (air, water1, water2). At my inlet I set the volume fractions of both water phases to 0.5 and the same velocity. Do I have to set any phase interactions I forgot or use the discrete phase model to get a result as shown below?


      Thank you in advance!


    • Rob
      Ansys Employee

      If the two water phases are supposed to mix you need two phases with the liquid phase comprising two species. What would cause the two water materials to flow in different directions in reality? 

    • BillT
      Subscriber

      I want to model the flow over two inclined plates in structured packings with porous media. Therefore I need to model two separate liquid phases. My primary phase is air and I have defined two secondary phases with water. The liquid phases are not supposed to mix. What exactly do you mean that I need two phases with the liquid phase comprising two species?


       

    • DrAmine
      Ansys Employee

      The picture is from which reference? I will go for just two-phase system wit only one water phase.


      Even if it does not make sense for me:


      If both phases of water are having the same density or do not have a pressyre jump due surface tension they will not be sparated (perhaps via sorce terms)

    • BillT
      Subscriber

      The picture is from phd thesis from Soulaine. He modeled this behaviour in openfoam. Basically, I try to model this:


      Title: Gas-liquid flow modeling in columns equipped with structured packing


      Link: https://onlinelibrary.wiley.com/doi/full/10.1002/aic.14550


      Any suggestions how to model this behaviour in fluent?

    • Rob
      Ansys Employee

      I don't think you need species, they're modelling how liquid behaves as it pours over the packing. This is going to need a huge amount of mesh, and can't work in 2d unless you leave gaps for the liquid to drop down the system & gas to rise.  From quickly reading the public access document I think the source terms are referring to the bulk model, not the plates? 


      Multiphase modelling requires a very detailed understanding of the flow physics before you go anywhere near CFD software: so to aid we need to understand the purpose of the model. We will also struggle given the limitations we're working to within the Community. Please can you outline what you're wanting to find out, and we'll take it from there? 

Viewing 5 reply threads
  • You must be logged in to reply to this topic.