[Yuri Almeida]

Dear FOAMers,

I've been simulating a burner using a wedge-type mesh, like the one in Fluent tutorial, and I'm having difficulties when I turn on the FVDOM radiation model.

Apparently, the rays in the FVDOM model which point to the axis (center line of the wedge) do not converge. The initial residual for these rays keeps rising after each radiation iteration until the solver blows up.

I've started to think that an incompatibility between the FVDOM and the wedge boundary condition exists.

I'm using OF 2.2.x and the fireFoam solver.

Has anyone here successfully applied the FVDOM in a wedge-type simulation?

Best,

[shininiu]

Hi Yuri,

Did you have any luck getting fvDOM model to work in a wedge geometry?
I ran into a similar problem- reactingFoam+radiation solver, wedge geometry, and fvDOM model. The problems I noticed are;

1. even though it's a 2-D solution, when nTheta=0 gives a floating point exception error. nTheta=1 will give no error.

2. solution won't converge, always hitting the 1000 iteration limit.

I'm very new to CFD in general, so please let me know if I missed something obvious.
Thank you,

[Yuri Almeida]

Hi shininiu,

Unfortunately, I couldn't overcome this issue with FVDOM in a wedge geometry. I tried to use P1 instead of FVDOM but the results were poor compared with no radiation.

I should highlight that using a wedge geometry is different than simulate a 2D case. Actually, a wedge geometry is a 3D case, as you can see from the 3 velocity components calculated during the simulation. So, it's expected that FVDOM gives an error if you try to put nTheta=0, since it's a 3D case.

However, FVDOM seems to have a problem with wedge BC. The axis should work as a reflective surface, returning back to the domain all radiation carried by the rays (as supposed to occur with a symmetric case), but I noted that the radiation starts to accumulate in the axis, always leading to blow up the solver. And I have no idea how to fix it...

A simple solution could be to set a fixed source term in the energy equation, for example -0.1*Q_c, considering that the system is losing 10% of the generated heat by the radiation. But it is only a suggestion.

Good luck!

[shininiu]

Ah... wedge being 3D makes a lot of sense... I was just wondering about non-zero z-components!

I think I'll move on without validating fvDOM model. (The wedge case was my validation case.) At least fortunately, my research case is not a wedge...

Thank you for the suggestion and the insight.

[jaydeep]

Hi Guys,

please take a look at this thread: fvDOM OpenFOAM wedge type patch