Low orthogonal quality after gradient adaptation

herrolsen · May 11, 2015, 05:59

Hi everyone!
I'm running a VOF simulation with water and air. The problem is 2D axisymmetric and the water is going to be falling freely. In order to get a sharp and clean interface I am currently using gradient adaptation (normalized) based on air-fraction (Fluent can't dynamically refine based on water fraction), the mesh is coarsened when the normalized gradient is below 0.1 and refines if it's over 0.1. My problem is that the refined cells automagically turns shitty when refined (orthogonal quality <0.1) which then leads to instability and eventually diverges.
Ansys mesher reports a quality of >0.95 in this area.

I've tried blocking the mesh by making several smaller connected domains in which I manually set the face size, but I can't make it axisymmetric for some reason. Fluent reports that the mesh is below the x-axis, ansys mesher says it's fine though.

Does anybody have an idea on how I can get a clean interface and a stable mesh?

And is it possible to do manual blocking of the mesh in Ansys Mesher?

Any help would be greatly appreciated!

Cheers

herrolsen · May 11, 2015, 14:40

Realized I've been using the wrong tool for the job. If you want to be able to block the mesh (set face sizes manually), you'll need to draw some sketches of your refinement area and the use "face split" to split the face but still keep it as one surface body.

May 11, 2015, 05:59	Low orthogonal quality after gradient adaptation	#1
herrolsen New Member Join Date: Jun 2014 Posts: 29 Rep Power: 12	Hi everyone! I'm running a VOF simulation with water and air. The problem is 2D axisymmetric and the water is going to be falling freely. In order to get a sharp and clean interface I am currently using gradient adaptation (normalized) based on air-fraction (Fluent can't dynamically refine based on water fraction), the mesh is coarsened when the normalized gradient is below 0.1 and refines if it's over 0.1. My problem is that the refined cells automagically turns shitty when refined (orthogonal quality <0.1) which then leads to instability and eventually diverges. Ansys mesher reports a quality of >0.95 in this area. I've tried blocking the mesh by making several smaller connected domains in which I manually set the face size, but I can't make it axisymmetric for some reason. Fluent reports that the mesh is below the x-axis, ansys mesher says it's fine though. Does anybody have an idea on how I can get a clean interface and a stable mesh? And is it possible to do manual blocking of the mesh in Ansys Mesher? Any help would be greatly appreciated! Cheers

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
[ICEM] Low quality mesh at curve-surface intersections	AndBra	ANSYS Meshing & Geometry	4	July 3, 2013 10:00
[ICEM] Hex refinement negative quality afterwards	Anorky	ANSYS Meshing & Geometry	8	March 22, 2010 04:25
Circular Cylinder at low Reynold number mesh and non orthogonal corrections	kassiotis	OpenFOAM Running, Solving & CFD	6	May 15, 2008 07:07
Multicomponent fluid	Andrea	CFX	2	October 11, 2004 06:12
pressure gradient term in low speed flow	Atit Koonsrisuk	Main CFD Forum	2	January 10, 2002 11:52

May 11, 2015, 14:40		#2
herrolsen New Member Join Date: Jun 2014 Posts: 29 Rep Power: 12	Realized I've been using the wrong tool for the job. If you want to be able to block the mesh (set face sizes manually), you'll need to draw some sketches of your refinement area and the use "face split" to split the face but still keep it as one surface body.