multiphase steam jet pump (ejector) (mixture model)

    • Jabaly

      Dear All 

      I'm trying to apply a multiphase model (mixture model) on an ejector ( as a steam jet pump). so I am trying to validate a paper about this topic. where, 
      1.The primary fluid (water) with flow rate Q0 and high-pressure P0 is accelerated through nozzle and obtained subsonic speed results in a low-pressure PS (vacuum) at the outlet section of the nozzle.
      The secondary fluid (air, with flow rate QS) is entrained and accelerated into the mixing chamber.


      Paper title: Numerical study on mixing flow behavior in gas-liquid ejector

      I applied all Bc's from that paper, Which are: 

      1.Mixture model –noslip velocity –implicit-without phase interactions
      2.Steady state
      4.Air-water (mixture)
      5.K-e model (turbulence)
      6.V water =3m/s …………..nozzle
      7.V air=3m/s  …………….domain
      8.Outlet= atm pressure
      9.Methods= coupled
      10.Default properties for materials (water and air)    

      The problem is: 
      There is a problem with the behavior of water when I get the volume fraction results. as you see in fig (A1 to A5) the water flow like a transient behavior ( 200 iterations is different from 400 iterations and so on). 

      Note#2: I applied steady-state, not transient 

      I think I need to apply phase interactions between water and air. but I don't know what is good interaction between them that I should apply to get the volume fraction as the paper (figure B).

      may you help me to solve this problem?


      Thanks a lot

      My result (figure A)

      paper result (figure <img src=">

    • DrAmine
      Ansys Employee

      As we cannot read any paper can you please double check your boundary conditions? Are they accounting for any other thing like mass transfer from water to water-vapor in air? Moreover from your picture there is clear stratification from the other pictures a sort of diffusion which tells me that diffusion force is missing in your case.

    • Rob
      Ansys Employee

      Please also post the velocity field at 1200 iterations. 

    • Jabaly

      Thanks for your reply.

      plz see attached link (pdf file), this is a presentation about that paper. It will give us a good and quick view of the ejector.

      The link:

      and yes you are right, the flow looks like stratified flow in my result, and I need to apply something like ( evaporating -condensation interaction) or ( cavitation) , I should reach to bubble flow as the paper. 
      but you know, evaporating -condensation interaction can be used when you have one fluid like water, and then convert liquid to vapor. 

      but in our system we have air-water mixture, so what is the interaction that I can apply in fluent ?! 

      Note#1: the paper does not mention anything about the interaction between water and air. 
      Note#2: you can see the velocity and pressure  counters, also velocity profiles 

    • Jabaly

      Thanks for your reply.

      yes sure, plz see attached photo for the velocity contours and profiles, and this link (pdf file), this is a presentation about that paper. It will give us a good and quick view of the ejector. also, you will find the velocity profile in slide #17. 

      The link:


Viewing 4 reply threads
  • You must be logged in to reply to this topic.