Fluids

Fluids

Topics relate to Fluent, CFX, Turbogrid and more

Is this the correct approach for a thermal analysis on a chamber? (Multiphase)

    • Vanessa Patino
      Subscriber

      Hello, I am new to Ansys and have been learning from tutorials and previous forums posts, please forgive my lack of technical knowledge and bad wording.  

      I am trying to do a simple thermal analysis on a enclosed chamber where half is filled with (static) water, the rest air, and a Silicon wafer heat source at the bottom. I'm assuming Fluent is the best option and the option I was told to use. I've also thought about steady state thermal but at some point I will have to simulate a moving body of water so I was hoping to get adapted to Fluent.

      The model was made in SolidWorks and imported. In the meshing stage I created named selections for the air, water, Si wafer, and the container itself (steel 0.2mm thickness). I changed the mesh sizing for water and air to .006 m and the container .009 m in hopes of getting a more accurate result.

      The Silicon wafer will be set to 65C. I would like to have convection and conduction included in the results. I am trying to see the temperature contours in general and see if they align with the physical experiments. 

      Initially I was trying to do this wthout the use of Multiphase but I was getting:

      ''Warning: zone of type interior found between different fluids!..." 
       
      After some googling someone said it's not as simple as setting cell zone materials and I would need to turn on multiphase but now I'm more lost as to how to complete the analysis. I'm also unsure if this is the right approach. 
       
      I've gotten contours but they're always very off and so I'll change something else or try another method and still they don't seem very accurate. I'm sure I'm missing something or trying to oversimplify/overcomplicate the process. 
       
      Please let me know if this is possible and if I need to provide any calrification. 
       
      Thank You!
    • Rob
      Forum Moderator

      No worries, good to see you started with the tutorials. Too many don't. 

      OK, it looks like you've not quite got it right. Fluent needs telling it's a multiphase model, so have a look at the options (Learning & Help) to help choose the right one. Also look up temperature dependent properties and buoyancy. When you are looking at the models you also want to consider what you need to know from the results, and what else you may want to know later on: we use models for specific functions so need to know ahead of choosing the starting point. Set up is fairly easy, working out what to set up is the hard bit. 

      I also suggest looking at "thin walls" in Fluent as you generally don't mesh the container unless there's a very good reason. 

      • Vanessa Patino
        Subscriber

         

        Oh yes I forgot to mention I chose VOF for my multiphase model. From looking at the ansys guide it says VOF “is designed for two or more immiscible fluids where the position of the interface between the fluids is of interest”, also “the steady or transient tracking of any liquid-gas interface”. So because the fluids in my model are Silicon, water, and air I went with it. However since the Si Wafer is meant to be solid in the materials I selected Si for its phase. 

        Looked up temperature dependent properties and buoyancy and how tempertaure affects buoyancy. It all seems to make sense (from what I understand an appropriate understanding of heat transfer is needed for Ansys but my knowledge is lacking).

         As far results go we’re just trying to look at the temperature contours: static temperature, total temperature, and wall temperature.

        I have some questions about the VOF Multiphase model. I would have 3 Eulerian phases correct? Would the phases go Phase 1: Si, Phase 2: water, Phase 3: air? Or would the opposite order make more sense? 

        Thin walls makes sense, never realized that meshing the container was uneccessary. 

         

        • Vanessa Patino
          Subscriber

          Also how would I apply the Si material to the Si wafer cell zone, same for air and water. Before I switched to multiphase I was able to assign those materials to those zones but now I don't have the option. 

    • Rob
      Forum Moderator

      If the wafer isn't flowing (or going to melt) it's a solid zone. That means we just worry about conduction. 

      That leaves two phases for the fluid, water & air. Those are the two phases. The cell zones are "mixture", and you patch the phase in. Check the VOF tutorials, you may not find anything specific example but the basic steps are there. 

Viewing 2 reply threads
  • The topic ‘Is this the correct approach for a thermal analysis on a chamber? (Multiphase)’ is closed to new replies.