Fluent, scale domain help!
 

Hello everyone!

For a 3D computational domain like 1000m x 1000m x 50m, to simulate its velocity and temperature field, after setting boundary conditions and calculation, the scaled residuals are not convergent, just after several iteration steps, the lines are flat at the level of 1e-01. However, when scaling like 100m x 100m x 5m, and the same boundary conditions, the results are convergent very well.

Somebody can tell me why, and how to set for the larger domain like 1000m x 1000m x 50m to make the calculation convergent well. Really frustrating for the problem.

Really appreciating for your help.

Bill

you can decrease the under relaxations ....

Thanks guy.
I have found the problem. Just because I set one side of the square as air velocity inlet, and the others as pressure outlets. If there is only one pressure outlet opposite to inlet, with other two sides as walls, then the calculation is perfect. However, in fact, all of the other three sides should be pressure outlets, I will try further to solve the problem properly.
Thanks for your attention.