CFD Online Discussion Forums

CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   FLUENT (http://www.cfd-online.com/Forums/fluent/)
-   -   Divergence detected in AMG solver:Temperature (http://www.cfd-online.com/Forums/fluent/101414-divergence-detected-amg-solver-temperature.html)

foamcfd May 3, 2012 08:43

Divergence detected in AMG solver:Temperature
 
Hi All,
I am getting an error:divergence detected in AMG solver:temperature. Residual for energy is fine (below 1E-06). However monitored temperature is oscillating and diverges. What can be the reason for this and how can I solve this?
I changed energy cycle type to F with BCGSTAB and double precision but no convergence.
Using coupled solver would not result in divergence. However monitored temperature would oscillate in 3 deg band. I would prefer avoid this oscillation so tried with low energy URF(0.85) but no success. Also I would avoind using coupled solver because of the hardware constraints.

Any suggestions to get solution with segregatted solver?

Many Thanks

LuckyTran May 3, 2012 10:48

Quote:

Originally Posted by foamcfd (Post 359068)
Also I would avoind using coupled solver because of the hardware constraints.

Any suggestions to get solution with segregatted solver?

Go ahead and use it. It's as simple as a click of a button. Any reason why you're not already using it? There are no visionaries here, we cannot know beforehand why you are doing anything a certain way. Please give some details about your simulation if you want any fruitful discussion.

foamcfd May 4, 2012 03:25

Hi,
Here is the problem description. This is basically natural convection problem of pipe flow and water jacket around it. Temperature of fluid inside pipe is around 120 degC and water is at 25degC. Water jacket is insulated from outside and so is the open area of pipe. I need to analyse this system first in steady state to get the temperature and flow field. Then carry out unsteady state analysis to see how much time does it take to reach a specified value of temperature at a specified location inside the pipe.
While running for the steady state analysis I am getting the dievergence error as explained in the last post.

Thanks,

LuckyTran May 4, 2012 12:23

Quote:

Originally Posted by foamcfd (Post 359235)
This is basically natural convection problem of pipe flow and water jacket around it.

Segregated solver will work. It will just take more iterations than the coupled solver (but it's faster too so that should make up for it).


All times are GMT -4. The time now is 02:24.