Divergence detected in Valve simulation

mehn · July 22, 2009, 14:13

Dear friends,
I am doing valve simulation, wherein the flow Bcs are stagnation pressure at inlet and pressure at outlet. I want to calculate mass flow rate by this simulation. The problem is ,as soon as I start the simultaion in starccm, the flow diverges within 10 iterations.

The mass flow rate value at outlet shoots very high (in terms of thousands, which is not possible). What may be the reason?. I also tried to change Inlet BC as velocity. But the results are same. I am using compressible state. (Ideal gas). My intention is to find the mass flow at inlet and outlet by specifying pressure drop. The state of the flow is steady.

bernarde · July 23, 2009, 03:25

Hi

The problem might be related to your mesh. One thing to try is to create threshold plots for different variables (pressure, temperature) to see if there are certain cells where the values are exploding. You might have a couple of bad quality cells in your mesh.

Regards

vishyaroon · July 23, 2009, 09:13

Have you tried using the Mass Flow Split outlet BC instead of the pressure BC at the outlet?

I'm not certain, but my guess is that by specifying the Stagnation BC at Inlet and Pressure at Outlet you are specifying the pressures at both the inlet and outlet and you may be over-specifying the system.

mehn · July 23, 2009, 09:28

Dear friends,

I got the solution.

Thank you.

bernarde · July 23, 2009, 09:32

What was the problem?

mehn · July 23, 2009, 09:38

Well, I was not getting the converged solution. The solver diverged within 10 iterations. I initialized with 2 bar (as suggested by support engineers). May be, its easy to relieve gradient, but it is very difficult for solver to create pressure gradient.
I took 2 bar as initial pressure and pressure inlet 2 bar and pressure outlet 1 bar.

July 22, 2009, 14:13	Divergence detected in Valve simulation	#1
mehn New Member mehul Join Date: May 2009 Location: Bangalore Posts: 7 Rep Power: 17	Dear friends, I am doing valve simulation, wherein the flow Bcs are stagnation pressure at inlet and pressure at outlet. I want to calculate mass flow rate by this simulation. The problem is ,as soon as I start the simultaion in starccm, the flow diverges within 10 iterations. The mass flow rate value at outlet shoots very high (in terms of thousands, which is not possible). What may be the reason?. I also tried to change Inlet BC as velocity. But the results are same. I am using compressible state. (Ideal gas). My intention is to find the mass flow at inlet and outlet by specifying pressure drop. The state of the flow is steady.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Divergence detected when using parallel solver	Phil	FLUENT	3	May 9, 2008 09:17
divergence detected in AMG solver !!!	yansheng	FLUENT	0	September 27, 2007 12:22
user subroutine error	CFDUSER	CFX	2	December 9, 2006 07:31
user defined function	cfduser	CFX	0	April 29, 2006 11:58
problems simulation ideal gas, divergence in AMG S	Ralf Schmidt	FLUENT	11	October 1, 2005 14:21

July 23, 2009, 03:25		#2
bernarde Member Bernard Esterhuyse Join Date: Mar 2009 Location: Pretoria, South Africa Posts: 50 Rep Power: 17	Hi The problem might be related to your mesh. One thing to try is to create threshold plots for different variables (pressure, temperature) to see if there are certain cells where the values are exploding. You might have a couple of bad quality cells in your mesh. Regards

July 23, 2009, 09:13		#3
vishyaroon Senior Member Aroon Join Date: Apr 2009 Location: Racine WI Posts: 148 Rep Power: 17	Have you tried using the Mass Flow Split outlet BC instead of the pressure BC at the outlet? I'm not certain, but my guess is that by specifying the Stagnation BC at Inlet and Pressure at Outlet you are specifying the pressures at both the inlet and outlet and you may be over-specifying the system.

July 23, 2009, 09:28		#4
mehn New Member mehul Join Date: May 2009 Location: Bangalore Posts: 7 Rep Power: 17	Dear friends, I got the solution. Thank you.

July 23, 2009, 09:32		#5
bernarde Member Bernard Esterhuyse Join Date: Mar 2009 Location: Pretoria, South Africa Posts: 50 Rep Power: 17	What was the problem?

July 23, 2009, 09:38		#6
mehn New Member mehul Join Date: May 2009 Location: Bangalore Posts: 7 Rep Power: 17	Well, I was not getting the converged solution. The solver diverged within 10 iterations. I initialized with 2 bar (as suggested by support engineers). May be, its easy to relieve gradient, but it is very difficult for solver to create pressure gradient. I took 2 bar as initial pressure and pressure inlet 2 bar and pressure outlet 1 bar.