CFD Online Discussion Forums

CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   FLUENT (http://www.cfd-online.com/Forums/fluent/)
-   -   Solving the turbulence equation (http://www.cfd-online.com/Forums/fluent/45367-solving-turbulence-equation.html)

summer July 12, 2007 09:24

Solving the turbulence equation
 
Hello, everyone. Now I am trying to model two-phase flow using Eulerian multiphase model with dispersed k-e model. However, I have a convergent problem. When I tried to ingore the turbulence equation in the solution panel, I found it converges well. Does anyone know that whether my solution is reasonable without the turbulence equation? Or is it possible that I will get some reasonable solution when I add this turbulence equation later? Thank you very much.

Joe July 12, 2007 10:13

Re: Solving the turbulence equation
 
You need to give much more explicit on what you are modelling for anyone to be able to answer your question.

summer July 12, 2007 11:20

Re: Solving the turbulence equation
 
O.K. I am sorry that I stated my problem simply. Actually, I am trying to model air(gas)-water(liquid) two-phase flow problem using Eulerian multiphase model. It is unsteady solver with standard k-e turbulence model. I use vel-in for both phases and pressure-out boundary condiition. After the simulation runs, I noticed that the residual of the continuity is quite big (~ 0.1) and it stays flat. I tried to reduce the under-relaxation factor of momentum. The residual went down to ~ 0.001 but also stayed flat (did not converges!) Finally, I noticed that if I did not select the equations of turbulence under the panel of "solve->controls->solution", the residuals go down perfectly. I want to know the effect of the turbulence in the solving processing. Is it possible to get the reasonable solution in this way? BTW, I also tried to refine my mesh or reduce the time step during the calculation but the residuals did not go down. Thank you very much.

Joe July 12, 2007 16:30

Re: Solving the turbulence equation
 
Turbulence is multiphase flow is a thorny issue ... Reynolds number dependancy etc.

Summer July 13, 2007 09:57

Re: Solving the turbulence equation
 
I use air-water two-phase flow. Since the superfacial velocity for air is about 0.23 m/s and for liquid is 2 m/s and the void fraction is 10%. The Reynolds number for gas is only 736 but for liquid is 1e5, which is quite large.


All times are GMT -4. The time now is 07:12.