[LA_RA]

Hello to everyone!

This is my first post so if have to change or modify something I apologise. I'm new into the CFD world and I'm learning how to use OpenFoam for my university projects.

I'm trying to simulate the following:

Closed cavity (I have 1 inlet and the the other faces are walls). (difficult task according to what I saw in other posts)

Through that inlet I'm injecting a gas at a very low speed. (Reynolds very low so laminar flow)

I've read all the posts I could about which type of solver should I use but I have no clue.

I have a Mach number very close to 0, so Ma<0.3 and that means that the flow is incompressible.

If the flow is incompressible it means that the density doesn't change with time but here as we are injecting gas into the closed cavity, if the volume is the same it means that the pressure must rise (I think that density should increase also).

So I don't know which approach would be the right one. For more information, the cavity is a big hexagonal prism (3 meters high and with 2 meters width) and the nozzles are very small (about 1 cm diameter).

The injected flow is also low, about 10l/min.

The temperature is about 1200K and the pressure inside the cavity is close to vacuum, 500Pa.

I want to do a transient simulation, see the velocity distribution and the pressure increase in the cavity and then add a transport equation with a passive scalar to simulate the diffusion of another gas.

I've tried buoyantBoussinesqPimpleFoam for example, but after I initalise the pressure fields with the 500Pa, in very next timestep that total pressure dissapears and I get positive and negative values.

I attach 2 pictures of a test I did with simple cube.

In the first moment I have the 500Pa but after that I get that vertical distribution with low negative and positive values that I don't know the meaning.

I also tried other solvers like rhoCentralFoam but I read that it doesn't work well with such low Mach numbers.

Any hints to know what type of information to search would be great.

Thanks for reading this!

Leandro.