I have this problem. I'm trying to simulate a tank with water and air at 6 bar. Suddenly a valve at the bottom is opened and I want to see how the air moves the water through a pipe that is connected to the bottom of the tank. To set the initial conditions I created three domains and I set the volume fractions for each one. When I star the solver I get the following message: floating point exception-overflow. I have made some free surface problems before and I didn't have any problem. What could be wrong? What should I do? Is it possible to fix the initial conditions as I did?

Hi,

If the valve and piping are at the bottom of the tank can't you just model the tank as water only with no free surface and just impose a pressure on the top surface? You don't need a multiphase model then and it will be much easier.

Glenn Horrocks

But the one who asked me to do the work wants to know if a "water bullet" can be created with the system. We will have, in the pipe, air, water and air pushing the water. I want to see how this "package of water" flows through the pipe and deforms. And also how could I fix a pressure on top if it would vary with time?

Hi,

If you want to know the air/water interface then you need multiphase. Have you read the best practises guide in the documentation on multiphase?

Glenn Horrocks

Yes. I have made a simulation without any problem. But in that simulation I had an inlet and an outlet. The only difference here is that I have only an outlet and the initials conditios are a little more complicated(I have also tried with an opening). Now I'm getting the "floating point exception". Thanks in advance.