7 Replies Latest reply on Feb 20, 2014 5:07 PM by Jared Conway

    Why does my transient simulations give the same results?

    Jonas Burman

      Hello!

       

      I've been trying to simulate a system for this past week and have been pulling my hair out for just as long.

       

      What i'm trying to do is to simulate a burner, with gas exiting a diffuser into a burning chamber. What i want to see is how the flame will be affected by draging out some of the flue gases from the top of the chamber (a flame lying against the wall is no good). Since you can't simulate combustion in sw (or atleast it's hard), i'm faking it with hot air instead.

       

      So to the problem. Every simulation i do practically gives the "flame" the same appearence every time: The flame points in a steep angle upwards, even if the flue gas exits straight out. Since the project is to see how the flame is affected by taking out some of the flue gases at the top of the burner no conclusions can be made if the flame is touching the wall even without the new exit. I've been trying to change temperatures flowrates etc with the same result. If the flame angel would differ randomly between every run it would be ok but even when i do a time dependent study the result is the same (FOR THE LOVE OF GOD WHY!?). I'm unsure how the solver works but i find it strange that my simulations of transient flow is so consistent.

       

      You can see the result in the attached picture, unfortunately i can't show you the whole system but you can se the flow exiting the nozzle and exits straight out to the right.

       

      Would be wery happy if anyone could explain why this is happening

       

      /a Swede

        • Re: Why does my transient simulations give the same results?
          Jared Conway

          Why transient?

           

          What are the inlet conditions?

           

          What are the outlet conditions?

           

          What are the initial conditions?

           

          Have you setup a simple version to see if it is also affected? This version could be posted.

          • Re: Why does my transient simulations give the same results?
            Mark Keown

            Remember the old videos for rocket development where the rocket flips around seconds after launch.  The flow out of the nozzle will 'snap' to one side if the nozzle diverges too quickly.  Sorry I cannot tell you how to design a proper nozzle but an investigation in this area is needed or other may be able to help.  Cheers.

              • Re: Why does my transient simulations give the same results?
                Jonas Burman

                Remember the old videos for rocket development where the rocket flips around seconds after launch.  The flow out of the nozzle will 'snap' to one side if the nozzle diverges too quickly.  Sorry I cannot tell you how to design a proper nozzle but an investigation in this area is needed or other may be able to help.  Cheers.

                 

                Good example. Well the thing is that while it´s absolutely crucial for a rocket to have a straight flow out of the nozzle it isn't for a burner of this type, actually it's good to create some turbulence and circulation in the burner chamber to increase the mixture of gases. So the phenomenon itself isn't the problem, it is the fact that the solver (in my opinion) seems to converge to pretty much exactly the same result every time even with time dependent studies up to 20 sek. It takes like 1 s and then the wake region is created at the bottom of the diffuser and the flow goes like in the pictures. Observe that the nozzle is off in this picture, it doesn't matter if it's on or off the wake region is created in exacly the same place anyway. You may think that the outlet in the upper right corner may affect it but it doesn't.

                Burnersim.png

                Actually i´ve tried to build the burner completely symmetrical without inlets and instead used the walls as inlets. That actually worked! then i could see a "flame" that were unstable and pointed in every directions during a 10 sek time dependent simulation. Athough as soon as I start to add the inlets for the combustable gases and the secondary air, that is added around the venturi, it becomes incredibly steady.

                 

                Im starting to think that the mesh is created in a asymetrical way because of this inlets and that´s the reason for the weird results. Anyway, I´m out of ideas...