Solution diverge with small element size.

stone123 · October 22, 2013, 16:36

Hi All

I am using FLUENT to study the flow in an aorta. Specifically, I am performing a mesh quality analysis to assess the best element size and the optimal value of wall shear stress.

The problem is that the solution diverge with small element size. Specifically, a mesh with 700000 element (skewness is about 0.75) shows convergence while meshes with 1400000 (skewness is about 0.84) elements or 2100000 elements (skewness is about 0.86) show divergence.

Meshes with high number of elements diverge using either Simple or Piso algorithm at 3e-4 of residuals. The model has an inlet (velocity inlet) and four outlet with flow split boundary conditions (outflow), and the flow is assumed laminar.

Does anyone has suggestions on my problem?

Many thanks

vigii · October 23, 2013, 08:19

Try adjusting your relaxation factors

stone123 · October 23, 2013, 17:13

Thank you very much for your suggestion!. I have decreased the under relaxation factor for moment to 0.1... should I decrease more?.... by the way, I will work on it and let you now... thanks

flotus1 · October 25, 2013, 13:09

Maybe the assumption of laminar flow does not hold.

stone123 · October 26, 2013, 06:48

I fixed it!!... apparently the problem was that I have split the mesh in several zones.... Once I have merged all fluids in the analysis, the solution converged rapidaly

October 22, 2013, 16:36	Solution diverge with small element size.	#1
stone123 New Member salvatore Join Date: Oct 2010 Posts: 7 Rep Power: 16	Hi All I am using FLUENT to study the flow in an aorta. Specifically, I am performing a mesh quality analysis to assess the best element size and the optimal value of wall shear stress. The problem is that the solution diverge with small element size. Specifically, a mesh with 700000 element (skewness is about 0.75) shows convergence while meshes with 1400000 (skewness is about 0.84) elements or 2100000 elements (skewness is about 0.86) show divergence. Meshes with high number of elements diverge using either Simple or Piso algorithm at 3e-4 of residuals. The model has an inlet (velocity inlet) and four outlet with flow split boundary conditions (outflow), and the flow is assumed laminar. Does anyone has suggestions on my problem? Many thanks

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
CFL Condition	Matt Umbel	Main CFD Forum	19	June 30, 2020 09:20
[ANSYS Meshing] Problem with specifying first element size in 2D Mesh	sigmundab	ANSYS Meshing & Geometry	9	April 14, 2016 10:53
Pros and cons of the method of Space-Time Conservation Element and Solution Element	Dwayne	Main CFD Forum	0	October 11, 2013 08:43
Mesh element size	sunilpatil	Main CFD Forum	1	April 16, 2013 21:41
[Commercial meshers] ST_Malloc: out of memory.malloc_storage: unable to malloc Velocity SA,	cfdproject	OpenFOAM Meshing & Mesh Conversion	0	April 14, 2009 16:45

October 23, 2013, 08:19		#2
vigii Member Vignesh KPV Join Date: Oct 2011 Location: Bangalore Posts: 83 Rep Power: 15	Try adjusting your relaxation factors

October 23, 2013, 17:13		#3
stone123 New Member salvatore Join Date: Oct 2010 Posts: 7 Rep Power: 16	Thank you very much for your suggestion!. I have decreased the under relaxation factor for moment to 0.1... should I decrease more?.... by the way, I will work on it and let you now... thanks

October 25, 2013, 13:09		#4
flotus1 Super Moderator Alex Join Date: Jun 2012 Location: Germany Posts: 3,427 Rep Power: 49	Maybe the assumption of laminar flow does not hold.

October 26, 2013, 06:48		#5
stone123 New Member salvatore Join Date: Oct 2010 Posts: 7 Rep Power: 16	I fixed it!!... apparently the problem was that I have split the mesh in several zones.... Once I have merged all fluids in the analysis, the solution converged rapidaly