Numerical Instability - FDS Stopped

[Nicola Tunzi]

Good evening everyone.

I'm trying to perform a FDS analysis onto a simply-supported beam with a constant burner placed under the bottom face of the beam. Furthermore, I have to set a Simulation Time to
 4h (14400 s), no devices but both the AST and the heat transfer coefficient boundary functions are enabled and the mesh cell size is 0.10 m (cubic mesh). I disabled the Plot3D data (in order to save time) and I launched the analysis. After reaching approximately 5000 s of simulation time, I receive an error due to numerical instability. What could be the issue? I'm reporting the FDS input file I'm using:

&HEAD CHID='beam'/
&TIME T_END=1.44E4.0/
&DUMP NFRAMES=14400, SMOKE3D=.FALSE./

&MESH ID='Mesh01', IJK=60,20,20, XB=0.0,6.0,0.0,2.0,0.0,2.0/


&REAC ID='POLYURETHANE_GM27',
      FYI='SFPE Handbook, GM27',
      FUEL='REAC_FUEL',
      C=1.0,
      H=1.7,
      O=0.3,
      N=0.08,
      CO_YIELD=0.042,
      SOOT_YIELD=0.198,
      RADIATIVE_FRACTION=0.35/

&SURF ID='CONCRETE',
      COLOR='GRAY 40'/
&SURF ID='BURNER',
      COLOR='RED',
      HRRPUA=1000.0/

&OBST ID='Obstruction #1', XB=1.3,4.9,0.8,1.1,1.1,1.1, SURF_ID='CONCRETE'/
&OBST ID='Obstruction #2', XB=1.3,4.9,0.8,1.1,0.7,0.7, SURF_ID='CONCRETE'/
&OBST ID='Obstruction #3', XB=1.3,1.3,0.8,1.1,0.7,1.1, SURF_ID='CONCRETE'/
&OBST ID='Obstruction #4', XB=1.3,4.9,1.1,1.1,0.7,1.1, SURF_ID='CONCRETE'/
&OBST ID='Obstruction #5', XB=4.9,4.9,0.8,1.1,0.7,1.1, SURF_ID='CONCRETE'/
&OBST ID='Obstruction #6', XB=1.3,4.9,0.8,0.8,0.7,1.1, SURF_ID='CONCRETE'/
&OBST ID='Obstruction', XB=1.9,4.2,0.8,1.1,0.0,0.5, SURF_IDS='BURNER','INERT','INERT'/
&OBST ID='Obstruction', XB=1.8,1.9,0.8,1.1,0.0,0.7, SURF_ID='INERT'/
&OBST ID='Obstruction', XB=4.2,4.3,0.8,1.1,0.0,0.7, SURF_ID='INERT'/

&VENT ID='Mesh Vent: Mesh01 [XMAX]', SURF_ID='OPEN', XB=6.0,6.0,0.0,2.0,0.0,2.0/
&VENT ID='Mesh Vent: Mesh01 [XMIN]', SURF_ID='OPEN', XB=0.0,0.0,0.0,2.0,0.0,2.0/
&VENT ID='Mesh Vent: Mesh01 [YMAX]', SURF_ID='OPEN', XB=0.0,6.0,2.0,2.0,0.0,2.0/
&VENT ID='Mesh Vent: Mesh01 [YMIN]', SURF_ID='OPEN', XB=0.0,6.0,0.0,0.0,0.0,2.0/
&VENT ID='Mesh Vent: Mesh01 [ZMAX]', SURF_ID='OPEN', XB=0.0,6.0,0.0,2.0,2.0,2.0/

&BNDF QUANTITY='ADIABATIC SURFACE TEMPERATURE'/
&BNDF QUANTITY='HEAT TRANSFER COEFFICIENT'/


&TAIL /

Thanks all are gonna answer me.

Best regards

Nicola Tunzi

[Kevin McGrattan]

Unless you are using the latest FDS release (6.7.5), you need to declare the interior of the beam as a pressure ZONE

&ZONE XYZ=1.5,1.0,1.0 /

[Nicola Tunzi]

This is what I'm getting when running helpfds command on CMDFDS:

FDS revision       : FDS6.7.5-0-g71f025606-release
MPI library version: Intel(R) MPI Library 2019 Update 7 for Windows* OS

Current number of OpenMP threads per MPI process:
OMP_NUM_THREADS=4

[Kevin McGrattan]

That appears to be the latest version. Add the line anyway and run again. I am running the case too.

[Nicola Tunzi]

Splitting the mesh may help to run the analysis faster? Or may it cause instabilities?

[Nicola Tunzi]

Meanwhile I'm running as a single mesh obviously

[Kevin McGrattan]

In cases like this, splitting the case on multiple meshes might make it more difficult to diagnose the problem. One thing to try is add 

&SLCF PBX=1, QUANTITY='PRESSURE ZONE', CELL_CENTERED=T /

and make sure that the interior of the beam has a value of 1.

[Nicola Tunzi]

They're running. When I'll have some news, I'll notice you. Thanks a lot for the suggestions!

[Kevin McGrattan]

Your beam has an INERT boundary, but it is hollow. There is something causing a problem inside this hollow beam. If you do not care about the thermal properties of the beam or its interior, create one single OBST for a solid beam, not 6 OBSTs for a hollow beam.

[Nicola Tunzi]

On FDS I have not set any thermal parameters for the material. It is called CONCRETE and I use it for thermal Mechanical analysis on Ansys. This is the reason of the boundary functions

On Mon, 26 Apr 2021, 22:33 Kevin McGrattan, <mcgr...@gmail.com> wrote:

Your beam has an INERT boundary, but it is hollow. There is something causing a problem inside this hollow beam. If you do not care about the thermal properties of the beam or its interior, create one single OBST for a solid beam, not 6 OBSTs for a hollow beam.

--
You received this message because you are subscribed to the Google Groups "FDS and Smokeview Discussions" group.
To unsubscribe from this group and stop receiving emails from it, send an email to fds-smv+u...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/fds-smv/CAAJimDEz85cpg05uiw-9ScoGxzzz1Nvqkr5L%2BMRN2p0rPeOqbQ%40mail.gmail.com.

[Kevin McGrattan]

OK, but there is no reason to have a sealed, hollow obstruction when the thermal parameters are not set. It is more stable to make the obstruction solid.