[jhssyb]

I use the two nozzles to inject water droplets of 50 um into hot air of 313K, this process is used to model to cooling effect in 20m-height tower.The support for the tower is 5m, where hot air is sucked in and the nozzle is placed at 3m height with an upwind injection. In view of the symmetrical property of a tower, I just cut 30 section wedge from the tower to carry out this simulation. But the residuals are totally irregular. They will remain stable at first and then experience a sharp increase or decrease, accompanying great change of velocity and temperature and some other things. Various attempts have been made to solve this problem but turned out to be fruitless. The residuals and monitored parameter iteration process have been attached as pictures.

1. I have tried to decrease the under relax factor but found it useless.

2. I also tried change the default method for SIMPLE to SIMPLEC, PISO, COUPLED, the problem still occurs.

3. Different turbulent models are also investigated and have little influence on this disgusting results.

4.The mesh cells have been varied from 130,000 to 1,790,000 with uniform and structural cells. But the problem has not been avoided.

So I have no idea about what is wrong with my simulation. Could you give me some suggestions?