Weird artefacts ... mesh-problem?

Eike · December 12, 2012, 02:47

Hi,

I get some weird artefacts like dents and rows ... is ther e anyone who has seen this before?

There is a constant mesh (Trimmer 20x20x20mm) with prism layers, do I have to refine it? The velocity of the fluid is around 0.1 to 0.3 m/s and I use a timestep of 0.025s at 8 iterations. I'm using k-omega SST ...

Best wishes
Eike

sail · December 12, 2012, 10:27

let me guess: big triangular mesh without boundary layer in tat area and your mesh does something like this

_ _ _ _ _ _ _
\/\/\/\/\/\/\/
- - - - - - --

i'd say that this is beacause one edge of the triangule touch the wall and the next cell have one whole face on the wall.

maybe interpolation issues? you might try to reduce this behaviour using higher order schemes.

Eike · December 13, 2012, 12:00

The mesh is cubic 20x20x20mm with prism layers at the walls ...

Wall Y+ varies at these spots:

This are some plane sections cutting the spots, scalar is velocity:

Martin Hegedus · December 13, 2012, 15:57

I've seen this sort of problem for low Mach number flows when a compressible solver without Mach number preconditioning is used. But, if you are using an incompressible solver than this is not an issue. I've also seen this sort of issue when not enough artificial damping is used with a central difference method.

December 12, 2012, 10:27		#2
sail Senior Member Vieri Abolaffio Join Date: Jul 2010 Location: Always on the move. Posts: 308 Rep Power: 17	let me guess: big triangular mesh without boundary layer in tat area and your mesh does something like this _ _ _ _ _ _ _ \/\/\/\/\/\/\/ - - - - - - -- i'd say that this is beacause one edge of the triangule touch the wall and the next cell have one whole face on the wall. maybe interpolation issues? you might try to reduce this behaviour using higher order schemes. __________________ http://www.leadingedge.it/ Naval architecture and CFD consultancy

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Dynamic Mesh Problem.	Tom Clark	FLUENT	10	June 21, 2021 05:27
Gambit problems	Althea	FLUENT	22	January 4, 2017 04:19
[snappyHexMesh] snappyHexMesh won't work - zeros everywhere!	sc298	OpenFOAM Meshing & Mesh Conversion	2	March 27, 2011 22:11
Icemcfd 11: Loss of mesh from surface mesh option?	Joe	CFX	2	March 26, 2007 19:10
Convergence problem when ustructured mesh is used.	SangJin Ryu	Siemens	3	October 5, 2000 21:26

December 13, 2012, 15:57		#4
Martin Hegedus Senior Member Martin Hegedus Join Date: Feb 2011 Posts: 500 Rep Power: 19	I've seen this sort of problem for low Mach number flows when a compressible solver without Mach number preconditioning is used. But, if you are using an incompressible solver than this is not an issue. I've also seen this sort of issue when not enough artificial damping is used with a central difference method.