CFD Online Discussion Forums

CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   FLUENT (http://www.cfd-online.com/Forums/fluent/)
-   -   Divergence error in AMG solver in axisymmetric problem. (http://www.cfd-online.com/Forums/fluent/119653-divergence-error-amg-solver-axisymmetric-problem.html)

gan1990 June 21, 2013 02:53

Divergence error in AMG solver in axisymmetric problem.
 
Hello eneyone,
I am modelling a axisymmetric natural convection problem in fluent.

I have set the X axis as the axisymmetric axis and also given X acceleration as -9.81 m/s2.

The problem is that in the first iteration itself I am getting an error of Temperature divergence in AMG solver.

When I change the boundary condition of axis to symmetry and the problem type to planar-2D,the problem goes off and the solution converges.

The meshing seems to be fine since it is giving feasible solution to planar problem.

I have come across suggestions like there should not be any node below X axis while solving axisymmetry.I have ensured that this does not happen while modelling itself.But is there any way of explicitly checking this?



I request everybody out here to give some solution to this problem.

Thanking in advance.

gan1990 June 21, 2013 12:51

Can anybody please help me.I really do not understand what the difficulty is out here.

nitman118 June 21, 2013 13:28

try decreasing the value of energy from 1 to 0.9 or less in--- solution controls--->under relaxation factor---> energy

gan1990 June 23, 2013 02:24

Quote:

Originally Posted by nitman118 (Post 435327)
try decreasing the value of energy from 1 to 0.9 or less in--- solution controls--->under relaxation factor---> energy

Thanks for the suggestion Mr nitish, I have tried with under relaxation of upto 0.2 but of no avail. It is not working out.

The problem seems to be in the axisymmetric set up.

Is there any other method that can be followed?


All times are GMT -4. The time now is 23:09.