[Cluster] Negative pressure

[mfournier01]

Hey,

I'm currently starting my PhD project and am setting up a simulation starting from the cluster.in environment. I've slightly modified the default set up so that I start with a perturbed density field (basically a clumpy atmosphere). When I let the simulation run, everything goes fine (the gas is cooling down, and start collapsing due to gravity).

However at some point the pressure reach negative values in some cells for a reason I could not identify. I added some couts in the code to check for the values of the variables involved in the computation of the pressure, and it look like the energy density is balancing the kinetic term so that they cancel out (w_p being defined in the code as gm1*(u_e - e_k)) and the result is slightly negative.

Any idea where this could come from ? I work with periodic boundary conditions by the way.

Thanks a lot !