
[Sponsors] 
October 14, 2009, 02:14 
I hope you can point out problems in my FLUENT solution

#1 
New Member
TangPing
Join Date: Sep 2009
Posts: 3
Rep Power: 9 
Hello,friends, thank you for your help.
I want to simulate methane explosion in pipe(diameter:0.2,length:5m). I will show you my simulation process in the following. I hope you can point out problems in my work and give me some advice. Thank you sincerely. My FLUENT version is Gambit2.2.30, Fluent6.2.16. Simulation process is as follows: 1 Model Simulate methane explosion in pipe. The pipe is blocked at one side and opened to atmosphere at the other side. Methane is ignited at the blocked side. The boundary type of pipe wall is WALL and the open side's is pressureoutlet. The mesh shape is QUAD. I do not show you the picture of my model because I am sure you have understood my model. It is quite simple. 2 Solving by FLUENT solver The solution process in FLUENT solver is as follows: (1) Grid check and grid scale are not mentioned for simplicity. (2) Define solver. Definemodelsolver. I choose coupled in slover, implicit in formulation, 2D in Space and Unsteady in Time, default for others; (3) Energy equation was chosen. kepcilkon for viscous model because many papers use it and defaul for model constants (4) Define Species model I choose species transport because some papers use it. But I am not sure whether I should use premixed combustion because explosion is of this kind. Moreover if premixed combustion is chosen, I can not select mixture properties. So at last I use species transport. Then I choose volumetric in reaction, eddydissipate in turbulencechemistry interaction (because some papers use it, but Finiterate/eddydissipation was also used a couple of papers). (5) Define materials Default. (6) Define boundary conditions The open end of the pipe: In turbulence specification method, I choose intensity and hydraulic diameter. back hydraulic diameter=0.2. In specied mass fraction, O2=0.21. Default for others. (7) solution control solvecontrolssolution, default is chosen here. (8) Initialization Compute from the blocked end of the pipe. guage pressure=101325. Input mass fraction of ch4 and o2. Then default for others. (9) Monitors Residual monitor is mot mentioned here. Surface monitor is also used. In this monitior, time step is chosen. (10) Define burnt zone I am not sure how to define it. I think I need to Adapt a region and then Patch. So I do it. I adapt a circle region. And then Patch in sloverinitializationpatch. Pressure=101325, temperature=2000K (In some papers, high temperature zone was defined, but I am not sure whether it was done in this way). CH4=0 O2=0. Input mass fraction of CO2 and H2O. (11) Iterate I just show you the main steps in my solution. Some steps such as save, scale and etc. were quite common and I did not mention it. I hope you can point out the problems in my solution. Thank you all of you. Thank you. I wish all of you be happy everyday!!! 

August 2, 2010, 03:18 

#2 
New Member
Dong Liang
Join Date: Jan 2010
Location: Shanghai, China
Posts: 12
Rep Power: 8 
thx for your work.
i'm interested in your boundary set. the wall is fixed temperature or heat transport? 

March 15, 2012, 12:40 

#3 
New Member
Join Date: Mar 2012
Posts: 4
Rep Power: 6 
Hi TangPing,
I am currently working on exactly similar thing. Same setup and I am getting results that are physically impossible. Can you please help me? 

Thread Tools  
Display Modes  


Similar Threads  
Thread  Thread Starter  Forum  Replies  Last Post 
[GAMBIT] 3.5 years overtake by Ansys: Where are the enhancements in grid generation  Volker P.  ANSYS Meshing & Geometry  22  January 17, 2012 18:27 
Solution technique/ Mesh type/ Cell limit/ Multizone regions in FLUENT  zandi  FLUENT  0  April 6, 2009 04:04 
Gmsh and samplesurface  touf  Open Source Meshers: Gmsh, Netgen, CGNS, ...  2  December 10, 2007 03:27 
Solution file to be used in Fluent  asif  FLUENT  5  July 9, 2003 09:41 
Wall functions  Abhijit Tilak  Main CFD Forum  6  February 5, 1999 02:16 