[giack]

Hi to all,
i run the following case:
A horizontal closed pipe is filled with water. At initial time the right wall is removed, allowing the liquid to exit the domain and at the same time, allowing the gas to enter. My target is to measure the velocity of the front of the bubble.
Initially I make a grid sensibility but I obtained some strange result. I tested the following case:
1)20x20x500
2)30x30x600
3)40x40x800
4)50x50x1000
5)60x60x1200
where the number are respectively number of cell in x direction x number of cell in y direction x number of cell in z direction. the section of the pipe is the xy plane and the dimensions of the pipe are d=50cm and l=3m
I obtained a very high fluctuating solution in case 1), less fluctuation in case 2), more less fluctuation in case 3). but the surprise is that in the 4) case I obtained more fluctuations than case 3)
Note that the solution should be constant and no fluctuating.
Someone have an idea for this strange behavior?
thanks for your help