Fluids

Fluids

momentum balance, physical origin of ‘Domain Src (Pos)’

    • Geoff
      Subscriber

      I am performing a buoyancy driven two-phase flow simulation in CFX. Air (ideal gas) is injected into the bottom of a vertical pipe filled with water. Pressure boundary conditions are used for the pipe and mass flow rate of air is specified. 


      As part of my post processing, i am trying to perform a momentum balance on a segment of the pipe:


      pressure differential force = wall shear + weight of column + momentum differential 


      I have been unable to get the equation to balance so i looked at the output file of my run (included below). I noticed a relatively large momentum contribution from Domain Src (Pos). I looked it up in solver modeling guide 16.10.3.2  and found the explanation unsatisfactory. I would like to understand where it came from physically and how i can properly account for it in my momentum balance. 


        +
      
      +
      | V-Mom-Bulk |
      +
      +
      Boundary : Periodic -1.0541E-06
      Boundary : air inlet -3.5050E-05
      Boundary : outlet -7.4481E-03
      Boundary : symmetry 5.2063E-19
      Boundary : wall -8.6698E-02
      Boundary Sources : wall -5.1424E-05
      Boundary : water inlet -1.1001E+00
      Domain Src (Neg) : fluid -6.4758E-02
      Domain Src (Pos) : fluid 1.3152E+00
      Neg Accumulation : fluid -5.5979E-02
      Domain Imbalance : 1.8272E-04
    • DrAmine
      Ansys Employee

      Which multiphase model are you using? What about Residual behavior and other domain imbalances? Actually Pos and Neg should sum up to near zero at convergence.

    • Geoff
      Subscriber

      I am using Homogeneous, with Continuum surface force model. It is a transient run where the flowrates in and out are fluctuating a lot . I have a max residual criteria of 1E-3 and a conservation target of 1% set. The RMS residual is typically around 5E-6. The domain imbalance is the highest for water mass, but the residual is highest in V momentum 

    • DrAmine
      Ansys Employee

      Can you please add screenshots of all you described. 

    • Geoff
      Subscriber

       Hope this helps, if i am missing anything specifically let me know. 


    • DrAmine
      Ansys Employee

      Imablances are looking good: but what about the water phase it sill have a huge accumulation?

    • Geoff
      Subscriber

      I do not understand what you are asking by 'what about the water phase it sill have a huge accumulation?' 


      It is a transient simulation, it is not a problem to instantaneously have a large accumulation of water. in fact i expect it based on the physics of them problem. 


      If mass is being accumulated in the domain id expect it to be represented by the 'Neg Accumulation' term.

Viewing 6 reply threads
  • You must be logged in to reply to this topic.