CFD Online Discussion Forums

CFD Online Discussion Forums (
-   ANSYS Meshing & Geometry (
-   -   [GAMBIT] Mesh causes divergence (

villager June 15, 2012 11:03

Mesh causes divergence
Good day to everybody.
I've made a tetrahedral mesh in GAMBIT, non-structured, with sizing function - initial element size 0.1, step 1.1, max 0.2 form walls, geometry is several connected spheres with diam. 1 (united geometry). No highly (>0.97) skewed elements were found (but max skewness 0.77, mesh quality in FLUENT = 0.257). Then I've imported mesh to FLUENT to solve my problem with species transport. After twenty iterations some species (O2, hydrocarbons) residuals begin increasing, then all residuals start to oscillate at values about e-03.
I tried to repair mesh or improve it in FLUENT, and suddenly after restarting FLUENT, the problem solved.
I've created a new mesh with same size with periodic b.c., but the problem repeated, and here no action was helpful. Contours of species show that species concentrations are not adequate at one place of mesh (in fluid zone, where many very fine elements are connected together).
Can you explain, why it's happening? What meshing guidelines I've violated? Or how can I check & heal my mesh in GAMBIT for example?
Thank you very much, need your help.

hadikhayyamian June 16, 2012 14:08

This might be due to your inappropriate solver settings in FLUENT. you can lower amount of under relaxation for example.

All times are GMT -4. The time now is 21:27.