CFD Online Discussion Forums

CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   CFX (http://www.cfd-online.com/Forums/cfx/)
-   -   Good convergence but without a good heat balance (http://www.cfd-online.com/Forums/cfx/24186-good-convergence-but-without-good-heat-balance.html)

Juan Martin Catelén June 27, 2007 16:51

Good convergence but without a good heat balance
 
Hello. I'm modelling a pool that is loosing heat trough a walls (one of them has velocity). Also, it has inlets of fluid with fixed temperatures and outlets where the temperatures are calculated. The problem I'm having is that although the residuals of the heat transfer are below 10-4, when I compare the heat entering to the pool through the temperature difference of inlets and outlets with the heat that is flowing through the walls, I see a difference of 20%. Is this possible with such a good convergence? What could be happening? The pool is of 22m long, 0.9m and 2.5m width. It has 750000 elements. I'll be waiting for help. Thanks in advance.

opaque June 27, 2007 16:56

Re: Good convergence but without a good heat balan
 
Dear Juan Martin,

Are you doing the comparison manually, or looking at the imbalance in the ANSYS CFX Solver Manager?

Do you have any Fluid Solid domain interfaces? The heat transfer reports are a bit different when they are present.

Opaque


Juan Martin Catelén June 27, 2007 17:16

Re: Good convergence but without a good heat balan
 
I'm doing the comparison with the CFX-Post functions in the table viewer. I'm taking the enthalpies at the inlets and at the outlets and comparing the diffrence with the areaint applied to the walls that are not adiabatic. I dont have solid fluid interfaces in the domain. What do you think? Should I continue the calculation?

Glenn Horrocks June 27, 2007 18:38

Re: Good convergence but without a good heat balan
 
Hi,

If you are looking for good heat balances then you should add this to the convergence criteria. This is usually caused by not running the simulation for enough physical time so the inlet flow fluid has reached the outlet and established steady state. Sometimes the timescale for the temperature to settle out is far long than for the flow so the default convergence options stop the run before the heat profile has reached steady state.

Glenn Horrocks

Juan Martin Catelén June 27, 2007 18:41

Re: Good convergence but without a good heat balan
 
Thanks.


All times are GMT -4. The time now is 13:57.