Using multiphase VOF model for stagnant liquid-gas interface

    • dodonn4

      Hi everyone,

      My simulation is of a blood bag warming up in ambient air. The bag is initially at 5C, and I need to find the time it takes for the center of the blood to reach 10C. The bag is about 90% blood and 10% air inside. From my research, and some help from people on this forum, I figured out I need to use the VOF multiphase model to properly represent the interface between the blood and air inside the bag.

      Every example I have found online involves fluids with velocity and fluid mixing (i.e. air and water flowing through a pipe, or water falling into a tank of air), but in my example there is little to no velocity, and virtually zero mixing of the fluids. The only motion is caused by buoyancy, and with the temperature changes I am dealing with, this will create very small magnitudes of motion in the fluids.

      So my question is, how would you recommend I model this with the multiphase model? I made the blood phase 1 and the air phase 2, and I tried different attempts using explicit and implicit methods, but I feel like I'm running around blindfolded here.

      In my many attempts it causes the blood bag to warm up far too rapidly, or the simulation just fails.

      Thanks in advance!

    • DrAmine
      Ansys Employee
      What about first starting just with the bag without any surrounding and just using a convective boundary condition at the outer walls of the bag. If the bag contains air and blood then both are separated and not mixed you might use the VoF approach.
      You need first to check the theory guide before you start. With 21R1 the setup is made straightforward so that in a VOF case by clicking on Settings Optimization , Fluent will help setting up suitable parameters. Implicit is more stable if you want to use larger time step sizes. Explicit is more high end if surface tension effect to be considered but requires smaller time step size.

      If phases are not separated: VoF is the bad choice. You will need then to rely on more generalized approach.
Viewing 1 reply thread
  • You must be logged in to reply to this topic.