[fireFoam] Pyrolysis of a wood crib case, janaf temperature divergence
 

Hello
I'm trying to model an experimental model-scale case that consists in burning a wood crib with heptane pools. Case description is shown in this pdf link:

https://www.iafss.org/publications/f...ss_10-1305.pdf

I have made the model of the wood crib and exported it as an stl file, then I took the tutorial case 'oppositeBurningPanels' from the LES fireFoam solver in OpenFoam-v2012 as a base, tweaked the parameters and boundary conditions to match my case, used snappyHexMesh for the wood crib and the compartment meshing.

this is the first model of the wood crib:
https://imgur.com/cvM37RR.png


I was trying to extrude the wood crib stl as a pyrolysis region with the extrudeToRegionMesh utility. I don't know how to use its other options but I tried to extrude the region around the stl face zone of the wood crib that's inside the internal mesh, and I get extruded cells inside the stl (which is outside the internal mesh).
The extruded cell layers eventually intersect, but I ignored it at that time and tried to run the simulation. it was set for 20 seconds at an adjustable time step.
Eventually the simulation stopped halfway with an error from the janaf table:

I got outrageous values of the temperature.
I refined the mesh, tweaked the solution schemes to gain stability over precision, and still got the same thing.

I tried to simplify the wood crib model and I made this:

https://imgur.com/HELXCAP.png

I tried to extrude the panel region the same way using the stl as a face zone.
The image below shows how the intersected cells of the panel region:

https://imgur.com/S4FEwWA.png

The two images below show the compartment and the wood crib (second version) together:

https://imgur.com/hM9B9Mh.png

https://imgur.com/H7WUPDb.png


I can't find the root of the divergence (the exact area) but I'm assuming it's the panel region that's causing the problem.

I really appreciate any advice, but I also want to know is there any way to properly extrude a panel/pyrolysis region out of such geometry inside the internal mesh without causing any problems or cell intersections.

There's a kind of a similar case published, but I can't find enough resources on that, so I can at least make a proper panel region.
here's the link:
https://www.sciencedirect.com/scienc...79711217301893

Thanks in advance. :)

Hi there
is a bit old, but I think the mesh in the depth direction is not fine enough.
What you can do is make a simpler model, where you model one or two rows and columns of the crib, this will allow you to refine the mesh in the depth direction and see if runs well.