
[Sponsors] 
May 14, 2003, 11:35 
Continuity

#1 
Guest
Posts: n/a

Hi
I am modelling the flow of an incompressible fluid through a structure. Due to the complexity of this structure, the use of hexahedral elements is not practical, thus tetrahedral are being used. My question to anyone that can help is, do you know why I am getting a net gain of mass of 1.6kg/s when the massflowrate is 6kg/s. The solver is steady state. The inlet is a velocity inlet, and the outlets are velocity inlets with negative velocities. I have tried pressure outlets, but for my problem, setting the correct pressure on the boundary is not easy to get continuity right, thus the use of pressure outlets is not practical either. Each velocity inlet has a minimum of 6 cells across the diameter, thus providing one either side of the inlet for a crude boundary layer, and 4 for the bulk of the flow. Thanks 

May 14, 2003, 15:23 
Re: Continuity

#2 
Guest
Posts: n/a

Why don't you try using Outflow as boundary conditions for your outlets?
Outflow BC imposes a zero diffusion flux of all variables and applies a mass balance correction. All others variables are calculated supposing a fully developed flow. This boundary can't be used if flow density changes, if you're using Pressure Inlets, or if you're modeling compressible flows (not your case). Hi ap 

May 14, 2003, 16:33 
Re: Continuity

#3 
Guest
Posts: n/a

I think the solution is not convergent.


May 15, 2003, 00:17 
Re: Continuity

#4 
Guest
Posts: n/a

Hi
The solver is given to you a solution for the condition that you impose. In other word, you must to 1) increase the number of elements 2) impose a velocity profile at the inlet 3) inpose a turbulent kinetic energy at the inlet 4) impose a dissitation rate of the TKE at the inlet. 5) Use second order upwind 6) Use Presure outlet 7) Use pressure velocity SIMPLEC Finally, Tetrahedral mesh are characterizad for numerical diffusion. Therefore, You must to increase the number of elements per volume. Keep in mind that not too many works has been than with it, and also the ones reported are 2D. It is my guess that yo have a large domain in 3D. By the way, Mr Liu coment about lack of convergence take it as a lack of convergence to the true solution not to a mathematical solution. Regards Alex Munoz 

May 15, 2003, 06:41 
Re: Continuity

#5 
Guest
Posts: n/a

I don't think your comments are right at all: the mass is always conserved in finitevolume based method regardless the mesh density, mesh topology, turbulent or laminar, SIMPLE or SIMPLEC, upwind or HOD.


May 15, 2003, 17:13 
Re: Continuity

#6 
Guest
Posts: n/a

If you are specifying velocity at the both the inlet and outlets of your model, then you are fixing the mass coming in and leaving, assuming constant density. Go back and double check that your inlet velocity multiplied by your inlet area is IDENTICALLY equal to your outlet velocity multiplied by your outlet area.


May 20, 2003, 12:33 
Re: Continuity

#7 
Guest
Posts: n/a

Hi DAE,
Are you getting reverse flow at the outlet ? As you don't know the exact pressure outlet bc, you try outflow bc. Run for about 100 to 200 iterations. take the total pressure value at the outlet and put it as pressure outlet bc. (i have tried this for my problem this works very well) bye, Ashish 

October 26, 2011, 01:14 

#8 
New Member
satyendra
Join Date: Jun 2010
Posts: 15
Rep Power: 9 
hi ashish,
i am trying to solve an incompressible ideal gas problem having two mass flow inlet conditions and two pressure outlet conditions ( i cannot use outflow BC since according to chapter 7, fluent user guide this is not allowed ). hence I have set target mass flow at outlet. But the continuity is not converging. any idea where i am going wrong. rana 

October 26, 2011, 08:57 

#9 
New Member
amir mofakham
Join Date: Feb 2011
Posts: 8
Rep Power: 8 
I think you cannot define the velocity outlet and inlet.
For example, if the velocity inlet is specified the outlet velocity is less than inlet velocity because of the drop pressure (viscosity). So one of them can be defined and the other one will be calculated. 

Thread Tools  
Display Modes  


Similar Threads  
Thread  Thread Starter  Forum  Replies  Last Post 
How to write k and epsilon before the abnormal end  xiuying  OpenFOAM Running, Solving & CFD  8  August 27, 2013 15:33 
Convergence moving mesh  lr103476  OpenFOAM Running, Solving & CFD  30  November 19, 2007 15:09 
Transient simulation not converging  skabilan  OpenFOAM Running, Solving & CFD  12  September 17, 2007 17:48 
IcoFoam parallel woes  msrinath80  OpenFOAM Running, Solving & CFD  9  July 22, 2007 02:58 
Could anybody help me see this error and give help  liugx212  OpenFOAM Running, Solving & CFD  3  January 4, 2006 19:07 