[ziemowitzima]

Dear All,

I encountered, a very strange behavior, which I hope could be explained with help of other Foamers.

I am running calculations using reactingFoam, but without reactions and combustions. I am interested in passive gas mixing.

The problem is:
I run the same calculations (exacly the same set-up, mesh, etc.) with two different max Courant Number:
1. maxCo = 2
and
2. maxCo = 0.25

I would expect that calculations with maxCo = 0.25 should be more numerically stable and give more presice results, because dt is smaller.
BUT
-results for maxCo = 0.25 are numerically unstable, they blew up after some time,
- results for maxCo = 2 are stable and gives nice smooth results...

Attached are results, which show O2 content in few points of the domain. They are compared with some experimental results.
Why for maxCo = 0.25 results are so different and so "bad", and why does it blew up ?

here results with maxCo = 2


Here results with MaxCo = 0.25

Thanks,
ZMM

[markusrehm]

Hi,

it has been a long time since I have bee using reactingFoam. I guess that there is a problem with the chemical time scales and or the ODE-Solver.
Might be that the reactions have not really started and after each time step they are averaged. Which ODE-Solver are you using?

Maybe you can check your mechanism in chemfoam?

Regards, Markus.

[ziemowitzima]

Hi Markus,
Thank you for your interest !
In my case I do NOT use chemical reactions, just passive mixing.
But in the chemistryProperties I have:

chemistrySolver ode;

but later:

chemistry off;


Regards,
ZMM

[markusrehm]

Hi,

I remember that there was an issue with the diffusion terms even leading to wrong temperature. But as mentioned above I haven' t used the solver for a long time.

Maybe that helps:

http://www.cfd-online.com/Forums/ope...-update-2.html

Good luck.

Markus