Char oxidation

[AmirrezaHBSH]

Hello, based on the User Guide, when setting SURFACE_OXIDATION_MODEL=T, FDS uses Eq. 17.8, in which the unit of 'A', as given in Table 17.1, is kg/m²/s. Therefore, based on the same equation, the unit of the char oxidation rate will be kg/m³/s. Consequently, if A's unit is m/s, we need to multiply it by the density of the char to achieve the aforementioned unit and then use it as an input parameter. Is this correct?

However, when SURFACE_OXIDATION_MODEL=T is not set, what will be the unit of A? Additionally, in this case, which equation will FDS use to calculate the rate of char oxidation?

[ericvmueller]

A in the surface oxidation model is an empirical coefficient specific to fitting this model and I’m not sure it makes sense to adapt a value obtained from fitting a different model (that has different units).

For oxidation without SURFACE_OXIDATION_MODEL=T see equations 9.1 to 9.3 in the most recent version of the user guide:

https://github.com/firemodels/test_bundles/releases/tag/FDS_TEST

The A value is assumed to have whatever units are required to obtain a rate of kg/m3/s, so it depends on the reaction order (N_S) and temperature dependence (N_T).

[drjfloyd]

Without knowing your inputs there is no way to suggest anything. Also the fact that your answer changes with the grid size does not mean that there is anything that needs to be corrected. Radiation scales with temperature to the fourth power. If making your grid cells have 1/8 the volume means you now resovle something closer to flame temperatures, then near field radiation can increase a lot.

Predicting fire growth and spread puts you in the realm of research, and you should do some legwork to understand what is happening. You have access to essentially every computed variable via the various outputs. You should attempt to determine why the heat flux is changing. It may be helpful to create simple test problems where you examing the impact of your inputs and the grid size.

[AmirrezaHBSH]

I understand this will get you to the correct units, but as I mentioned at the top of the thread I cant confirm this would give you a reasonable behavior. The A for the surface oxidation model assumes the reaction is not dependent on char density but on available surface area. If you are starting with an A with different units then it will have come from fitting an equation of a different form (I'm guessing from Porterie) and it is not obvious to me that you can just re-scale it and use it here. You could see if this gets you reasonable results in an idealized scenario (for example just solving the reaction rate at different temperatures) ...but at a certain point you may as well just tune A to get the result you want. It is a fitting coefficient not a fundamental physical parameter.

Thank you for your assistance and consideration. I have plotted the total heat flux at a specific point in my pine needle bed, as shown in the attached file. The highest heat flux value should correspond to the combustion reaction, while the second peak is expected to relate to char oxidation, which is exothermic. However, my plot shows a reversed trend.

To analyze this, I defined a solid particle at a specific location and placed the total heat flux device on that solid. Below are my input file details and the resulting plot:

&OBST XB=0.4,0.41,0.04,0.05,0.03,0.04, SURF_ID='solid_particle' /
&SURF ID='solid_particle', COLOR='GRAY',SURFACE_VOLUME_RATIO = 4550, LENGTH= 0.07, MATL_ID='solid_material' /
&MATL ID='solid_material', DENSITY=500.0, CONDUCTIVITY=0.5, SPECIFIC_HEAT=1.5 /
&DEVC ID='hfTOTAL1', XYZ=0.405,0.045,0.035, QUANTITY='TOTAL HEAT FLUX', IOR=1 /

The fire spreads from left to right across my fuel bed, so I set IOR=1. However, I believe the plot might need to be flipped to accurately represent the heat flux values over time. The current plot appears to show an inverted trend, which might not align with the expected progression of heat flux during the fire spread.

Do you have any suggestions for how I can adjust or verify this to ensure the plot correctly reflects the heat flux over time?

[drjfloyd]

Without knowing your inputs how are we possibly supposed to make suggestions?

[AmirrezaHBSH]

Without knowing your inputs how are we possibly supposed to make suggestions?

&HEAD CHID='Test_1_0', TITLE='Flame Spread Over Bed of Pine Needles' /

&MISC GVEC=0,0,-9.81 /
&RADI RADIATION=.TRUE. /

&TIME T_END=450., DT= 0.01/
&MISC TMPA= 27/

&CLIP MINIMUM_DENSITY=0.1 /

&MESH ID='mesh1', IJK=10,17,50, XB=0.0,0.1,0.0,0.17,0.0,0.30, COLOR='BLUE', CHECK_MESH_ALIGNMENT=T /
&MESH ID='mesh2', IJK=10,17,50, XB=0.1,0.16,0.0,0.17,0.0,0.30, COLOR='RED', CHECK_MESH_ALIGNMENT=T /
&MESH ID='mesh3', IJK=10,17,50, XB=0.16,0.22,0.0,0.17,0.0,0.30, COLOR='BLUE VIOLET', CHECK_MESH_ALIGNMENT=T /
&MESH ID='mesh4', IJK=10,17,50, XB=0.22,0.28,0.0,0.17,0.0,0.30, COLOR='BROWN', CHECK_MESH_ALIGNMENT=T /
&MESH ID='mesh5', IJK=10,17,50, XB=0.28,0.34,0.0,0.17,0.0,0.30, COLOR='COBALT', CHECK_MESH_ALIGNMENT=T /
&MESH ID='mesh6', IJK=10,17,50, XB=0.34,0.4,0.0,0.17,0.0,0.30, COLOR='DIM GRAY', CHECK_MESH_ALIGNMENT=T /
&MESH ID='mesh7', IJK=10,17,50, XB=0.4,0.46,0.0,0.17,0.0,0.30, COLOR='FLESH', CHECK_MESH_ALIGNMENT=T /
&MESH ID='mesh8', IJK=10,17,50, XB=0.46,0.52,0.0,0.17,0.0,0.30, COLOR='GOLD', CHECK_MESH_ALIGNMENT=T /
&MESH ID='mesh9', IJK=10,17,50, XB=0.52,0.58,0.0,0.17,0.0,0.30, COLOR='GREEN', CHECK_MESH_ALIGNMENT=T /
&MESH ID='mesh10', IJK=10,17,50, XB=0.58,0.64,0.0,0.17,0.0,0.30, COLOR='HOT PINK', CHECK_MESH_ALIGNMENT=T /
&MESH ID='mesh11', IJK=10,17,50, XB=0.64,0.7,0.0,0.17,0.0,0.30, COLOR='PURPLE', CHECK_MESH_ALIGNMENT=T /

&SURF ID ='WALL1',
TMP_FRONT = 21.0 /

&VENT XB = 0.0,0.0,0.0,0.17,0.0,0.3,
SURF_ID = 'OPEN' / ! BC left side

&VENT XB = 0.70,0.70,0.0,0.17,0.0,0.30,
SURF_ID = 'OPEN'/ ! BC right side (pressure outlet)

&VENT XB = 0.0,0.70,0.0,0.0,0.0,0.30,
SURF_ID = 'OPEN'/

&VENT XB = 0.0,0.70,0.17,0.17,0.0,0.30,
SURF_ID = 'OPEN'/

&VENT XB = 0.0,0.70,0.0,0.17,0.30,0.30,
SURF_ID = 'OPEN',/

&OBST XB = 0.0,0.7,0.0,0.17,0.0,0.0,
SURF_ID = 'WALL1',
COLOR = 'WHITE'
TRANSPARENCY = 0.5/

&OBST XB = 0.0, 0.7, 0.0, 0.08, 0.0, 0.012,
COLOR = 'PINK'
SURF_ID = 'ASBES'/

&SURF ID = 'ASBES'
MATL_ID = 'ASBESTOS'
THICKNESS = 0.012
TMP_FRONT = 21.0 /

&MATL ID = 'ASBESTOS'
FYI = 'Autoclaved aerate concrete'
CONDUCTIVITY = 0.8
SPECIFIC_HEAT = 1.0
DENSITY = 2500.
EMISSIVITY = 0.80 /

&OBST XB = 0.1, 0.6, 0.08, 0.081, 0.0, 0.032,
COLOR = 'GRAY'
SURF_ID = 'FIRECLAY'/

&SURF ID = 'FIRECLAY'
MATL_ID = 'FC'
THICKNESS = 0.032
TMP_FRONT = 21.0 /

&MATL ID = 'FC'
FYI = 'Autoclaved aerate concrete'
CONDUCTIVITY = 0.5
SPECIFIC_HEAT = 1.0
DENSITY = 2500.
EMISSIVITY = 0.80 /

&SURF ID = 'PINE_NEEDLE_BED'
MOISTURE_FRACTION = 0.009
SURFACE_VOLUME_RATIO = 4550
LENGTH = 0.07
GEOMETRY = 'CYLINDRICAL'
COLOR = 'TEAL'
MATL_ID(1,1) = 'LIPID'
MATL_ID(1,2) = 'DIGEST'
MATL_ID(1,3) = 'HEMI_CELLULOSE'
MATL_ID(1,4) = 'GLUCAN'
MATL_ID(1,5) = 'PHENOLIC'
MATL_ID(1,6) = 'ASH'
MATL_MASS_FRACTION(1,1) = 0.2273
MATL_MASS_FRACTION(1,2) = 0.0990
MATL_MASS_FRACTION(1,3) = 0.1744
MATL_MASS_FRACTION(1,4) = 0.201
MATL_MASS_FRACTION(1,5) = 0.272
MATL_MASS_FRACTION(1,6) = 0.0262 /

&PART ID = 'Needles'
DRAG_COEFFICIENT = 2.81
COLOR = 'GREEN'
SURF_ID = 'PINE_NEEDLE_BED'
STATIC = .TRUE./

&INIT PART_ID = 'Needles'
XB = 0.1,0.6,0.0,0.08,0.012,0.032
N_PARTICLES_PER_CELL = 1
CELL_CENTERED = .TRUE.
MASS_PER_VOLUME = 31.429
DRY = T /

&CATF OTHER_FILES='vegetation_model.txt' /

&SURF ID = 'IGNITOR',
TMP_FRONT = 1300.,
COLOR = 'BLUE',
RAMP_T = 'RAMP_IGN'/

&RAMP ID = 'RAMP_IGN', T = 0.0, F = 1.0 /
&RAMP ID = 'RAMP_IGN', T = 15, F = 1.0 /
&RAMP ID = 'RAMP_IGN', T = 16, F = 0.0 /

&VENT XB = 0.094,0.098,0.0,0.08,0.012,0.012, COLOR = 'BLUE', SURF_ID = 'IGNITOR'/

&OBST XB=0.4,0.41,0.04,0.05,0.03,0.04, SURF_ID='solid_particle' /
&SURF ID='solid_particle', COLOR='GRAY',SURFACE_VOLUME_RATIO = 4550, LENGTH= 0.07, MATL_ID='solid_material' /
&MATL ID='solid_material', DENSITY=500.0, CONDUCTIVITY=0.5, SPECIFIC_HEAT=1.5 /
&DEVC ID='hfTOTAL1', XYZ=0.405,0.045,0.035, QUANTITY='TOTAL HEAT FLUX', IOR=1 /

[drjfloyd]

This model cannot be run without your vegetation_model.txt file assuming that has the definition for LIPID, DIGEST, etc.

You have no SLCF outputs for temperature, heat release, etc.
You have no PART outputs for burning rate, heat flux, etc.
Your have only one single DEVC for heat flux to a surface

If you are modeling the growth of fires you are doing research. You need to take the time to try and understand what is happening in your simulation. That cannot be done without having outputs that let you understand what is happening.