CFD Online Discussion Forums - buoyantPimpleFoam

Hi all,

maybe I found a bug in the buoyantPimpleFoam solver.

Information:

small case (2D - wedged)
closed volume
T initialized with setFields
dt = 1e-8
save date all 1e-8 (so every iteration)

1. Step

No relaxation factors in fvSolution
I get 24 iterations
after that simulation blow up

2. Step

added relaxation to fvSolution
After 24 iterations the simulation blow up

3. Step

relax factors to 0.001
After 24 iterations the simulation blow up
Solution in paraview like Step 1 and 2

4. Step

relax factors changed to 1e5
After 24 iterations the simulation blow up
Solution in paraview like 1, 2 and 3

So in my opinion the relaxation factors are not taken into accout because if I set the factors to 1e5 I should never reach the second/third iteration, arent I ?

Additionally I had a look into the code - relaxation is included:

Code:

    EEqn.relax();

    pEqn.relax();

    UEqn.relax();

So now the question is, why I get always same solutions if I change the relaxation factors - thats not correct.

If I change the factors in simpleFoam to 100 the simulation blow up at the beginning.

Thanks in advance
Tobi