CFD Online Discussion Forums

CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   Main CFD Forum (http://www.cfd-online.com/Forums/main/)
-   -   Convergence in Free convection (http://www.cfd-online.com/Forums/main/117219-convergence-free-convection.html)

myty May 4, 2013 19:02

Convergence in Free convection
 
Hi all,
I am new user of ANSYS/FLUENT 14.0, sometime i am running simulation of natural convection with boussinesq approximations for density, the simulations is going on with no error, residuals decreases rapidly at start and then constant. so i am confused that either i need to run for more iterations or it will not converge at all.
please guide

ztdep May 5, 2013 06:08

Quote:

Originally Posted by myty (Post 425249)
Hi all,
I am new user of ANSYS/FLUENT 14.0, sometime i am running simulation of natural convection with boussinesq approximations for density, the simulations is going on with no error, residuals decreases rapidly at start and then constant. so i am confused that either i need to run for more iterations or it will not converge at all.
please guide

which level of does the residual decreases to.
could you pease upload your mesh system, and what is the Ra number.

myty May 5, 2013 12:00

All residuals decreases quickly to value close to convergence but then constant, means no further decrease. The mesh is simple rectangular box, with three walls adiabatic q''=0, and one side heating wall. Is there a role of Ra number to simulate natural convection. Also gravity value at -0.0098 gave convergence but when i increased it to -0.098 or -0.98, -9.8, situation is like i told.

ztdep May 5, 2013 20:04

Quote:

Originally Posted by myty (Post 425373)
All residuals decreases quickly to value close to convergence but then constant, means no further decrease. The mesh is simple rectangular box, with three walls adiabatic q''=0, and one side heating wall. Is there a role of Ra number to simulate natural convection. Also gravity value at -0.0098 gave convergence but when i increased it to -0.098 or -0.98, -9.8, situation is like i told.

I high Ra number means a convection intensity, i strongly suggest you to increase the mesh density as the increases of Ra number(increases of gravity).

myty May 6, 2013 17:22

Thanks, i decreased the mesh size and now its working, one more thing is that when i am using velocity inlet BC for natural convection flow with very small value 10^-m/s, its working but when i am using pressure inlet BC which is preferred with natural convection flow, its not converging with constant wall T & q''. Do u suggest me something to run it.


All times are GMT -4. The time now is 02:56.