Default Solver and two-way flow model

[Gilles Marie Josep Flamant]

 

 

Dear All,

 

I am restarting CONTAM simulations after some while and I have two questions:

1. ‘File Vivienda1_5h_d.prj’ : When I run the simulation with the ‘Default Solver (Implicit Euler)’, I receive the following error message : “too many density iterations”. Simulation runs only if I change the integration method to ‘Short Time Step (Explicit Euler)’ while modifying at the same time some parameters (using shorter calculation simulation time and increasing supply and return system volumes of the simple AHS system I modeled to non-realistic values). Do you know the reason for such error message and what I should change to be able to run the ‘Default Solver’ integration method?

 

2. When modeling an open internal door using the ‘two-way flow’ model (with one opening), the result file gives two flow values : ‘Flow0’ and ‘Flow1’. I found some information about that model in the user manual (p58, p65 and p273) but I am not sure about the exact meaning of these two flows. Does ‘Flow0’ correspond to the flowrate in the lower part of the opening and ‘Flow1’ in the upper part? In the example below, the airflow rate through the opening (internal door) is quite constant (50 m³/h) (design value) but Flow0 and Flow 1 oscillate between negative and positive values without any obvious explanation (to me). What could be the reason for this?  The first figure below shows the (almost) constant air flowrate of 50 m³/h and the second one the same information but this time clicking on ‘Show two-way flows’.

 

 

Thanks a lot in advance for your help!

I also attach the .PRJ file.

 

Kind regards,

 

Gilles Flamant

Ingeniero civil - investigador

Centro de Desarrollo Urbano Sustentable (CEDEUS)

El Comendador 1916, Providencia

 

No sienta la obligación de contestar este mail fuera de horario laboral.

[Anders Bjork (Thunderhead Engineering)]

Gilles,

Did you learn any more about the first part of your question?

I was just experimenting with some Simulation Parameters and non-trace Species and found a similar set of behaviors. The short- time step method seems to help, and changes to the value of "Absolute Convergence Factor" and "Vary density between time steps" values have a significant impact on computations and solver stability/ability to solve.

Anders

Thunderhead Engineering

ME Masters' Student studying CFD at Kansas State University