Optically Thin & Radiative Fraction

[Francois]

Hi Guys

Any help will be greatly appreciated.

I am simulating a large open wood chop / pile fire and I am interested in the thermal radiation emitted to a nearby structure.

1. My D*/dx ratio is 19.4 which points to a fine mesh given the HRR.
2. My model has Number_Radiation_Angles=1000.
3. My model uses the default Gray Gas model as the large wood pile fire will be a sooty fire.
4. My model uses a simple chemistry fuel with composition CH1.7O0.74N0.002 (Douglas Fir, SFPE).
5. My model uses the standard LES turbulence model.
6. My simulation time is only 30seconds as I am only interested in the thermal flux not temperatures. I can calculate the steady state temperatures if I know what the radiative heat flux values are (&BNDF  Radiative_Heat_Flux). Note: I did verify that the HRR stabilized before I used the results (stabilized at approx. 10s after ignition). 
   

I have studied the FDS user guide and technical reference guide but I am still a bit unsure of a few things.

 

1. I am well aware that the radiative fraction reduces as the fire size increases due to inefficient combustion / large production of smoke and the shielding of the luminous flame zone. Therefore I am aware that I cannot simply use typical values for small luminous fires given in the literature (0.3 considering Douglas Fir SFPE). Problem is I cannot find any correlation for the radiative fraction relating to large wood fires. I could find lots of information / correlations for large hydrocarbon fires (SFPE and McGrattan). The FDS user guide (Section 16.1, page 189) briefly discussed the reduced radiative fraction with increasing fire size but the reference given (Beyler [35]) refers to large open hydrocarbon fires. I know that the same principle will apply for large wood pile fires but surely it is not correct to use the hydrocarbon correlations (large hydrocarbon fires produce thick black smoke whilst large wood/timber produce a more whitish smoke as can be seen from large forest fires)? Note, my wood chop fire is a large 440m² fire (not the test attached).  
2. I would like to confirm if I understand Section 16.1.3 correctly. When a radiative fraction is specified is it applied to all of the “hot” byproducts / lumped byproducts of the fire (soot, CO2, CO, H2O, etc.) where xrq”’>10kW/m³ hence the hot plume? Elsewhere (xrq”’<10kW/m³ ) “cold” gasses (soot, CO2, CO, H2O, etc.) absorb and emit the thermal radiation emitted by the hot plume (considering the default Optically Thick scenario).
3. I am not sure which “cold” gas constituents are all involved in the reabsorption is it soot, H20, CO and CO2?
4. The following is what really bothers me. I did a quick test scenario (attached). When I compare two thermal radiation scenarios I get drastically different results (order of magnitude difference). The only difference between the models are Optically Thick (default) vs Optically Thin. In this case I am using a radiative heat fraction of 0.3.
   1. Optically Thick results approx 2.4kW/m².
   2. Optically Thin results approx 20kW/m².

            It does not make sense that the cold gasses absorb so much thermal radiation,  

            hand calcs point to a reduction of approx 15% not 833%. What am I doing wrong? 

[Kevin McGrattan]

10 m by 10 m is a big fire. What is the difference in the steady-state value of Q_RADI in the _hrr.csv file?

[Francois]

Hi Kevin

The 10m x 10m fire is just a test I did. My actual fire is 440m² (27m effective diameter). I am simulating a fire of the entire wood stockpile and the thermal radiation received by a structure nearby. We are evaluating the requirement for passive fire protection (structural fire protection).

When I average the values out from 10s to 30s I get the following results:

1. Optically Thick
   1. Q_RADI = 27441kW
   2. HRR = 303187kW
   3. Q/RADI / HRR = 0.09

1. Optically Thin
   1. Q_RADI = 93286kW
   2. HRR = 314735kW
   3. Q/RADI / HRR = 0.3

[dr_jfloyd]

1. I don't think I have seen radiative fraction data for very large wood fires.

2. The radiative fraction just applies to cells where there is a heat release rate.

3. Those gasses plus fuel and any other gas species where a RADCAL species is specified on SPEC for that gas.

4. It may be that the default path length used to generate the absorptivities for the radiation transport is too small. OPTICALLY_THIN will be a conservative answer. You could try increasing the default path length. You could also set HUMIDTY=0 and Y_CO2_INFTY=0 on MISC which would still let smoke absorb radiation but prevent the ambient air around the fire from absorbing radiation..  

[Francois]

Thank you for the inputs, I really appreciate it.

I will run a few tests with your suggestions.

[Francois]

Hi

I ran some tests on the tests case discussed above where I specified a Path Length of 10m and the results I got is in line with expectations (seems to do the trick).