Nonsymmetric result from symmetric flow and geometry?

jh1998 · August 27, 2022, 15:24

Hello all,

I have a fairly good experience in commercial CFD solvers (Fluent and the likes) but wanted to try OPENFOAM. I started with a simple geometry and flow to get used to the openfoam file structure but received an unexpected result.

The flow seems to stick to one side of the nozzle after a certain number iterations. I presumed it was due to a numerical error that accumulated over time, but running it several times produced the same result.

This does not seem to happen when the restricted section in the middle is wider. I run the simulation using icoFoam (following blockMesh)

P.S. only one cell is in the z-direction. fvSchemes and fvSolution are directly from icoFoam tutorials.

Files and velocity field screenshot are attached below.

Any ideas would be appreciated. Thank you.

Alczem · September 8, 2022, 04:10

Hey

I am not 100% sure about this, but I simulated venturi nozzles a while back and had similar behaviors. From what I can remember, which is not a lot, at certain reynolds number (or some other adimensional number), the flow is expected to deviate like your picture. I guess that if you tried in a transient simulation, the flow might oscillate and not stick to only one side.

(I might be totally wrong! Check some scientific papers, I know there are some talking about this)

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
An error occurs.'Increase the file catalog size. '	dhehdxhdaus	CFX	3	April 10, 2022 21:13
Issues on the simulation of high-speed compressible flow within turbomachinery	dowlee	OpenFOAM Running, Solving & CFD	11	August 6, 2021 07:40
Initialize flow from different geometry	B.Axel	FLUENT	2	January 24, 2018 14:15
Laminar pipe flow using 2D axis symmetric geometry	nandiganavishal	OpenFOAM Running, Solving & CFD	4	April 1, 2013 01:48
Simulation of Flow through Complex 3D Geometry	EmersonKB	CFX	5	July 2, 2009 09:17

September 8, 2022, 04:10		#2
Alczem Senior Member Join Date: Dec 2021 Posts: 251 Rep Power: 6	Hey I am not 100% sure about this, but I simulated venturi nozzles a while back and had similar behaviors. From what I can remember, which is not a lot, at certain reynolds number (or some other adimensional number), the flow is expected to deviate like your picture. I guess that if you tried in a transient simulation, the flow might oscillate and not stick to only one side. (I might be totally wrong! Check some scientific papers, I know there are some talking about this)