Convergence problem

Alimohamadi_nasr · July 26, 2013, 03:10

Hello,
in my problem, there are an inlet and 3 outlets, I know mass flow inlet and the mass fraction of each outlet. The flow is incompressible and laminar and I use Ansys Fluent 14.0 to solve it. I have decided to use velocity inlet and pressure outlet as boundary condition. However, I do not know the amount of pressure outlet. Therefore, I have done over a number of trials to achieve suitable mass fraction in each of my outlets. I have used transient mode for this problem. The amount of residual is good (10-4 for continuity and 10-6 for velocity X & Y), also mass balance is very good; it is less than 10-5%. My problem is that after reaching to suitable mass fraction in each of outlets it stayed constant beyond 700-1000 iterations but then it again started to change and the changes is not periodically. Can I consider it as converged solution during that 500 or 700 iterations? What is your suggestion? I should mention that, mesh study has been done and it did not affect solution. It is kind of instability problems in its physics. How can I get completely steady mass flow outlet.

flotus1 · July 26, 2013, 08:36

Since you only have 1 inlet and know the mass flow rate at each of the 3 outlets, you could use velocity outlets and a pressure inlet.
Because the fluid is incompressible, the pressure level at the inlet wont change the solution.
There are other possible combinations of boundary conditions for such a setup, but i am not not familiar with all of them.

As long as the solution changes significantly with additional iterations, you can not consider it as converged somewhere in between.

Alimohamadi_nasr · July 26, 2013, 08:50

Thank you for your reply. I don't think its good choice to use velocity out let or mass flow out let. because i only can guess the mean velocity in out let and also the direction in outlet is not clear for me. moreover, the physics of my problem is unstable especially near the out let. therefore, velocity outlet cannot be a good choice.

flotus1 · July 26, 2013, 08:57

For these cases I think that fluent provides an option to specify the mass flow fraction at several outlets, just the way you need it. I never used this option, so you will have to consult the manual.

Alimohamadi_nasr · July 26, 2013, 09:05

yes, its name is outflow. in this boundary condition, users should only determined the fraction of mass in each outlet. the only problem about this boundary condition is that it is only suitable for fully developed flow or flow without any gradient. in my case that the flow is laminar in the pipe, if i want to have fully developed flow I should increased outlet pipe more than 60 times of its origin.

July 26, 2013, 03:10	Convergence problem	#1
Alimohamadi_nasr Member Ali Mohamadi Join Date: Aug 2012 Posts: 32 Rep Power: 13	Hello, in my problem, there are an inlet and 3 outlets, I know mass flow inlet and the mass fraction of each outlet. The flow is incompressible and laminar and I use Ansys Fluent 14.0 to solve it. I have decided to use velocity inlet and pressure outlet as boundary condition. However, I do not know the amount of pressure outlet. Therefore, I have done over a number of trials to achieve suitable mass fraction in each of my outlets. I have used transient mode for this problem. The amount of residual is good (10-4 for continuity and 10-6 for velocity X & Y), also mass balance is very good; it is less than 10-5%. My problem is that after reaching to suitable mass fraction in each of outlets it stayed constant beyond 700-1000 iterations but then it again started to change and the changes is not periodically. Can I consider it as converged solution during that 500 or 700 iterations? What is your suggestion? I should mention that, mesh study has been done and it did not affect solution. It is kind of instability problems in its physics. How can I get completely steady mass flow outlet.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
convergence problem when use pisoFoam, LES for wind tunnel case	Forrest_Lei	OpenFOAM	3	July 19, 2011 06:00
convergence problem	commonyue	Main CFD Forum	1	December 1, 2009 03:54
Convergence of CFX field in FSI analysis	nasdak	CFX	2	June 29, 2009 01:17
3D Fluid Flow Convergence problem	Emily	FLUENT	2	March 21, 2007 22:18
Non Convergence of 3D Heat transfer cfd problem	Balraj	Main CFD Forum	3	December 9, 2004 00:24

July 26, 2013, 08:36		#2
flotus1 Super Moderator Alex Join Date: Jun 2012 Location: Germany Posts: 3,399 Rep Power: 46	Since you only have 1 inlet and know the mass flow rate at each of the 3 outlets, you could use velocity outlets and a pressure inlet. Because the fluid is incompressible, the pressure level at the inlet wont change the solution. There are other possible combinations of boundary conditions for such a setup, but i am not not familiar with all of them. As long as the solution changes significantly with additional iterations, you can not consider it as converged somewhere in between.

July 26, 2013, 08:50		#3
Alimohamadi_nasr Member Ali Mohamadi Join Date: Aug 2012 Posts: 32 Rep Power: 13	Thank you for your reply. I don't think its good choice to use velocity out let or mass flow out let. because i only can guess the mean velocity in out let and also the direction in outlet is not clear for me. moreover, the physics of my problem is unstable especially near the out let. therefore, velocity outlet cannot be a good choice.

July 26, 2013, 08:57		#4
flotus1 Super Moderator Alex Join Date: Jun 2012 Location: Germany Posts: 3,399 Rep Power: 46	For these cases I think that fluent provides an option to specify the mass flow fraction at several outlets, just the way you need it. I never used this option, so you will have to consult the manual.

July 26, 2013, 09:05		#5
Alimohamadi_nasr Member Ali Mohamadi Join Date: Aug 2012 Posts: 32 Rep Power: 13	yes, its name is outflow. in this boundary condition, users should only determined the fraction of mass in each outlet. the only problem about this boundary condition is that it is only suitable for fully developed flow or flow without any gradient. in my case that the flow is laminar in the pipe, if i want to have fully developed flow I should increased outlet pipe more than 60 times of its origin.