Convergence issue on a very simple case

Kalabagh · January 23, 2024, 14:18

Hey foamers,

I'm not pretty new at OpenFOAM. I have worked on some cases of medium dificulty (non newtonian, turbulence, complex geometry, etc.) in the past.

Now I want to work on a simple venturi on a pipe. But first of all, I have started just simulating water flow through a straight pipe.

It is a straight pipe diam 50 mm and lenght 300 mm. Boundaries are:

- U
inlet (zeroGradient)
outlet (fixedValue uniform 0 0 1)
walls (noSlip)
- p
inlet (fixedValue uniform 0)
outlet and walls (zeroGradient)

Mesh is done by blockMesh macro to adapt well to cylinders (extended checkMesh is fully OK).

Always in laminar condition, starting with 10 cPs, the simulation (simpleFoam) converges so well. Using final data as 0 and reducing the viscosity to 5 cPs, everything continues to be OK. Again, now with good data as 0, when I reduce viscosity to 1 cPs (water) the p residual starts to climb smoothly performing a pretty constant wave and it ends diverging...

Does someone have any idea?

Thanks.

Alczem · January 24, 2024, 04:31

Hey!

Can you share the case?

tas38 · January 24, 2024, 07:50

Based on the flow conditions and geometric dimensions provided, I suspect you may be in the laminar-turbulent transitional region for the lower fluid viscosity values $\left( \mathrm{Re}\approx 10000 \right)$ . A simple check is to turn-on turbulence and see if the solution converges.

Kalabagh · January 26, 2024, 10:54

Quote:

Originally Posted by tas38

Based on the flow conditions and geometric dimensions provided, I suspect you may be in the laminar-turbulent transitional region for the lower fluid viscosity values $\left( \mathrm{Re}\approx 10000 \right)$ . A simple check is to turn-on turbulence and see if the solution converges.

Quote:

Originally Posted by Alczem

Hey!

Can you share the case?

Here you can see the .zip of the case: https://drive.google.com/file/d/1EVv...ew?usp=sharing

As I said, pretty simple. I have done everything with a p residual set of 0.001. If I do it with normal 0.01 it becomes even worse.

Kalabagh · March 23, 2024, 15:03

Hi Troy,

As you said, turbulence model was required. With the straight pipe I have been able to solve it, but when I add just a seondary inlet orthogonal to that pipe, convergence problems appear again.

In attached the full case already run: https://drive.google.com/file/d/1IYn...ew?usp=sharing

January 23, 2024, 14:18	Convergence issue on a very simple case	#1
Kalabagh New Member Santi Parera Join Date: Jul 2019 Location: BARCELONA Posts: 25 Rep Power: 6	Hey foamers, I'm not pretty new at OpenFOAM. I have worked on some cases of medium dificulty (non newtonian, turbulence, complex geometry, etc.) in the past. Now I want to work on a simple venturi on a pipe. But first of all, I have started just simulating water flow through a straight pipe. It is a straight pipe diam 50 mm and lenght 300 mm. Boundaries are: - U inlet (zeroGradient) outlet (fixedValue uniform 0 0 1) walls (noSlip) - p inlet (fixedValue uniform 0) outlet and walls (zeroGradient) Mesh is done by blockMesh macro to adapt well to cylinders (extended checkMesh is fully OK). Always in laminar condition, starting with 10 cPs, the simulation (simpleFoam) converges so well. Using final data as 0 and reducing the viscosity to 5 cPs, everything continues to be OK. Again, now with good data as 0, when I reduce viscosity to 1 cPs (water) the p residual starts to climb smoothly performing a pretty constant wave and it ends diverging... Does someone have any idea? Thanks.

January 24, 2024, 07:50		#3
tas38 Senior Member Troy Snyder Join Date: Jul 2009 Location: Akron, OH Posts: 219 Rep Power: 18	Based on the flow conditions and geometric dimensions provided, I suspect you may be in the laminar-turbulent transitional region for the lower fluid viscosity values $\left( \mathrm{Re}\approx 10000 \right)$ . A simple check is to turn-on turbulence and see if the solution converges. Alczem likes this.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Issue setting up a case with periodic boundaries	sampy	CONVERGE	12	February 10, 2023 14:29
axial compressor mass flow convergence issue	jyotir	SU2	4	November 24, 2021 06:03
Continuity convergence issue on a 3d wing	Nick R	FLUENT	19	April 6, 2020 14:52
[DesignModeler] DesignModeler Scripting: How to get Full Command Access	ANT	ANSYS Meshing & Geometry	53	February 16, 2020 15:13
Disabling SIMPLE convergence check in simpleControl.H	Mojtaba.a	OpenFOAM Running, Solving & CFD	0	August 8, 2012 17:31

January 24, 2024, 04:31		#2
Alczem Senior Member Join Date: Dec 2021 Posts: 207 Rep Power: 5	Hey! Can you share the case?

March 23, 2024, 15:03		#5
Kalabagh New Member Santi Parera Join Date: Jul 2019 Location: BARCELONA Posts: 25 Rep Power: 6	Hi Troy, As you said, turbulence model was required. With the straight pipe I have been able to solve it, but when I add just a seondary inlet orthogonal to that pipe, convergence problems appear again. In attached the full case already run: https://drive.google.com/file/d/1IYn...ew?usp=sharing