Issue With Fire Extinguishing On Its Own

[drjfloyd]

Discussed in #16124

^{Originally posted by Asher19000349 April 7, 2026}
I am developing a hydrogen fire scenario using the PyroSim version 2022.3.1208 running on FDS 6.5.3. To develop the model, I am following the exact source information, and mesh configuration from this paper: https://doi.org/10.1016/j.ijhydene.2021.01.130

The model replicated at my end matches the HRR, and flame characteristics appropriately. However, apparently, the combustion extinguishes after around 15-17 secs of simulation time. I tried the following workarounds:
(1) Using HRRPUA instead of MLRPUA. This does not change the physics, but I hoped if it could lead to better numerical stability.
(2) Using all OPEN boundaries, as well as, INERT/closed for all boundaries, and in one case all closed with ZMAX open, to reduce the excessive air entrainment and prevent flame blow-off.
(3) I also ran the software using latest FDS and PyroSim, but the results became disturbed, and the version I used was given correct results jut that the fire extinguishes.

I am pasting below the FDS code. and the smokeview video showing the fire extinguishing.

FDS CODE

&HEAD CHID='validation6'/
&TIME T_END=70.0/
&DUMP DT_RESTART=300.0, DT_SL3D=0.25/
&MISC SIMULATION_MODE='LES'/

&MESH ID='Mesh01', IJK=50,50,70, XB=4.0,9.0,2.55,7.55,6.0,13.0/

&REAC ID='Reaction1',
FUEL='HYDROGEN',
HEAT_OF_COMBUSTION=1.42E+5/

&DEVC ID='T1', QUANTITY='TEMPERATURE', XYZ=7.0,4.45,6.85/
&DEVC ID='T2', QUANTITY='TEMPERATURE', XYZ=6.8,4.45,6.85/
&DEVC ID='T3', QUANTITY='TEMPERATURE', XYZ=6.6,4.45,6.85/
&DEVC ID='T4', QUANTITY='TEMPERATURE', XYZ=6.4,4.45,6.85/
&DEVC ID='T5', QUANTITY='TEMPERATURE', XYZ=6.2,4.45,6.85/
&DEVC ID='T6', QUANTITY='TEMPERATURE', XYZ=6.0,4.45,6.85/

&SURF ID='Surface01',
FYI='Jet fire',
COLOR='RED',
MLRPUA=0.5,
ADIABATIC=.TRUE./

&OBST ID='jet release', XB=7.2,9.0,4.4,4.5,6.2,6.9/

&VENT ID='Mesh Vent: Mesh01 [XMAX]', SURF_ID='OPEN', XB=9.0,9.0,2.55,7.55,6.0,13.0/
&VENT ID='Mesh Vent: Mesh01 [XMIN]', SURF_ID='OPEN', XB=4.0,4.0,2.55,7.55,6.0,13.0/
&VENT ID='Mesh Vent: Mesh01 [YMAX]', SURF_ID='OPEN', XB=4.0,9.0,7.55,7.55,6.0,13.0/
&VENT ID='Mesh Vent: Mesh01 [YMIN]', SURF_ID='OPEN', XB=4.0,9.0,2.55,2.55,6.0,13.0/
&VENT ID='Mesh Vent: Mesh01 [ZMAX]', SURF_ID='OPEN', XB=4.0,9.0,2.55,7.55,13.0,13.0/
&VENT ID='Mesh Vent: Mesh01 [ZMIN]', SURF_ID='OPEN', XB=4.0,9.0,2.55,7.55,6.0,6.0/
&VENT ID='Vent', SURF_ID='Surface01', XB=7.2,7.2,4.4,4.5,6.8,6.9/

&SLCF QUANTITY='TEMPERATURE', VECTOR=.TRUE., CELL_CENTERED=.TRUE., ID='temp', PBY=4.4/
&SLCF QUANTITY='TEMPERATURE', VECTOR=.TRUE., CELL_CENTERED=.TRUE., ID='temp 2', PBY=4.45/
&SLCF QUANTITY='VELOCITY', VECTOR=.TRUE., CELL_CENTERED=.TRUE., ID='Vel', PBY=4.4/
&SLCF QUANTITY='VELOCITY', VECTOR=.TRUE., CELL_CENTERED=.TRUE., ID='Vel2', PBY=4.45/

&TAIL /

VIDEO

fire.extinguishing.mp4

[drjfloyd]

The original case had a couple of input choices that seemed to be the cause of the issue but turned out not to be.

I cut down the mesh some so this runs more quickly. In the images you see at 1.4 s things look like you might expect for a case where the inlet vent is just a single grid cell. Then quickly there is a step change in the flow behavior. Now in the velocity we see jets up, down, and out where the velocity is much higher than the vent velocity (~20 m/s given the MLRPUA and TMP_FRONT). Looking at the hrr at ~1.4 there is a sudden spike in the hrr which then goes back down to the expected value. Mass file and the various slice outputs look OK (no suddenly missing or new mass or bad looking values of specific heat or enthalpy when compared to the gas fractions and hrr).

&HEAD CHID='h2'/
&TIME T_END=1.6/
&DUMP MASS_FILE=T,NFRAMES=10000/
&MISC SIMULATION_MODE='LES'/

&MESH ID='Mesh01', IJK=50,40,40, XB=4.0,9.0,2.5,6.5,6.0,10.0/

&REAC ID='Reaction1',
FUEL='HYDROGEN'/

&DEVC ID='T1', QUANTITY='TEMPERATURE', XYZ=7.0,4.45,6.85, ORIENTATION=1.0,0.0,0.0/
&DEVC ID='T2', QUANTITY='TEMPERATURE', XYZ=6.8,4.45,6.85, ORIENTATION=1.0,0.0,0.0/
&DEVC ID='T3', QUANTITY='TEMPERATURE', XYZ=6.6,4.45,6.85, ORIENTATION=1.0,0.0,0.0/
&DEVC ID='T4', QUANTITY='TEMPERATURE', XYZ=6.4,4.45,6.85, ORIENTATION=1.0,0.0,0.0/
&DEVC ID='T5', QUANTITY='TEMPERATURE', XYZ=6.2,4.45,6.85, ORIENTATION=1.0,0.0,0.0/
&DEVC ID='T6', QUANTITY='TEMPERATURE', XYZ=6.0,4.45,6.85, ORIENTATION=1.0,0.0,0.0/

&SURF ID='Surface01',
FYI='Jet fire',
COLOR='RED',
MLRPUA= 0.5,
TMP_FRONT=710./

&OBST ID='jet release', XB=7.2,9.0,4.4,4.5,6.2,6.9/

&VENT MB='XMIN',SURF_ID='OPEN'/
&VENT MB='XMAX',SURF_ID='OPEN'/
&VENT MB='YMIN',SURF_ID='OPEN'/
&VENT MB='YMAX',SURF_ID='OPEN'/
&VENT MB='ZMIN',SURF_ID='OPEN'/
&VENT MB='ZMAX',SURF_ID='OPEN'/

&VENT ID='Vent', SURF_ID='Surface01', XB=7.2,7.2,4.4,4.5,6.8,6.9/

&SLCF QUANTITY='TEMPERATURE',CELL_CENTERED=T,PBY=4.45/
&SLCF QUANTITY='VELOCITY',CELL_CENTERED=T,PBY=4.45/
&SLCF QUANTITY='SPECIFIC HEAT',CELL_CENTERED=T,PBY=4.45/
&SLCF QUANTITY='SPECIFIC ENTHALPY',CELL_CENTERED=T,PBY=4.45/
&SLCF QUANTITY='HRRPUV',CELL_CENTERED=T,PBY=4.45/
&SLCF QUANTITY='INTEGRATED INTENSITY',CELL_CENTERED=T,PBY=4.45/
&SLCF QUANTITY='VOLUME FRACTION',SPEC_ID='HYDROGEN',CELL_CENTERED=T,PBY=4.45/
&SLCF QUANTITY='VOLUME FRACTION',SPEC_ID='AIR',CELL_CENTERED=T,PBY=4.45/
&SLCF QUANTITY='VOLUME FRACTION',SPEC_ID='PRODUCTS',CELL_CENTERED=T,PBY=4.45/

&TAIL /

[drjfloyd]

If I double the resolution the issue seems to go away. At 37 s and still OK. I added CONDUCTIVITY and VISCOSITY those numbers get quite large.

[rmcdermo]

OK, I will work on it.

[rmcdermo]

My theory is that the explosion is related to the extinction model interacting with the surface temperature ramp. I added TAU_T=0 and things seem to behave. Can you send me the full case and I'll test?

[rmcdermo]

Or if it is just the one from the original post, I'll try that. But what is the TMP bc, ADIABATIC or TMP_FRONT=710?

[drjfloyd]

Original post but should be with 710 and delete the HoC from REAC (that is the HHV value). ADIABATIC and use of the HHV values turned out to be red herrings but the case moving forward will not have them.

[drjfloyd]

I ran the short case with SUPPRESSION=F and it is still happening.

[rmcdermo]

Yeah, I'm eventually running into this with the big case at about 5.5 s. Back to the drawing board.

[rmcdermo]

Seems to me to be a time step issue. For the h2.fds case I changed to:

&REAC ID='Reaction1', FUEL='HYDROGEN'/
&COMB SUPPRESSION=F/
&PRES SOLVER='UGLMAT', MAX_PRESSURE_ITERATIONS=1/
&MISC CHECK_HT=T/

and get this. Note that both ADIABATIC and high TMP_FRONT often lead to this violation of the energy equation CFL.

[drjfloyd]

I should have thought of that. I'll start up and let it run for a while to see if that indeed fixes it. Other iterations of this input have had failures out ~10-12 s.

[rmcdermo]

Well that may be part of it, but the bigger case puked at about 2.4 sec anyway.

[drjfloyd]

I made a "heavy" hydrogen using the thermo props and HoC from the .out file. So far out to 5 s and still going.

&SPEC ID='H2',FORMULA='H8',RAMP_CP='CP',RAMP_K='K',RAMP_MU='MU',RAMP_D='D'/

&RAMP ID='MU',T=20, F= 1.531E-05/
&RAMP ID='MU',T=223, F= 2.287E-05/
&RAMP ID='MU',T=723, F= 3.838E-05/
&RAMP ID='MU',T=1223, F= 5.158E-05/
&RAMP ID='MU',T=1723, F= 6.319E-05/

&RAMP ID='K',T=20, F= 2.369E-02/
&RAMP ID='K',T=223, F= 3.652E-02/
&RAMP ID='K',T=723, F= 7.012E-02/
&RAMP ID='K',T=1223, F= 1.041E-01/
&RAMP ID='K',T=1723, F= 1.356E-01/

&RAMP ID='CP',T=20, F= 1.252/
&RAMP ID='CP',T=223, F= 1.288/
&RAMP ID='CP',T=723, F= 1.458/
&RAMP ID='CP',T=1223, F= 1.601/
&RAMP ID='CP',T=1723, F= 1.695/

&RAMP ID='D',T=20, F= 2.024E-05/
&RAMP ID='D',T=223, F= 5.230E-05/
&RAMP ID='D',T=723, F= 1.711E-04/
&RAMP ID='D',T=1223, F= 3.378E-04/
&RAMP ID='D',T=1723, F= 5.457E-04/

&REAC ID='Reaction1', FUEL='H2',HEAT_OF_COMBUSTION=119960.5135/