Discussion Forum

Hi everybody

I'm currently trying to simulate the evaporation von water out of meat during a vacuum-packaging process.

I'm using Comsol 5.0. The fluid flow in the chamber is simulated with a turbulent k-e physics. The vapour is modelled with a transport of diluted species. I also have heat transfer physics in the air domain and in the "meat".

My problem is, that I cannot couple the calculated evaporation into the fluid flow. I used an "Inlet" BC and inserted the evaporation mass flux, but this model does not find initial conditions. So I'm missing the additional flow due to the evaporation.
Is there an easy possibility to get this to work? (Attached File: "evaporation_BC")


Then I tried to use the "Previous Solution Node" as described here: www.comsol.com/blogs/using-the-previous-solution-operator-in-transient-modeling/
I want to use the evaporation massflux from the previous time step, to insert it into the "Inlet" BC in the fluid flow. This should in my opinion benefit the solver because the coupling gets eased.
I could implement it and it is converging.

The problem is, that the BC with the value of the previous time step is not doing anything in the fluid flow. I see no difference in the solution whether I activate or deactivate this BC. Furthermore, the value of the previous time step (in my file “u2”) is changing whether I use the value furthermore or if I just calculate it and do not use it. (The value “u2” is correct if do not use it but horribly wrong if I use it. Have a look at the plot “Prev Time step vs actual value” in the model: “previous_solution”)
I have no problem using the value in other physics (for example the “heat transfer”), the problem only occurs if I use it in the fluid flow.

I would appreciate it if someone could help me with my first or my second problem.
Thank you, Seraphin