FDS6 - CO_PRODUCTION - Finite Rate Combustion

[winklerm...@gmail.com]

Hi guys,

I am new to this group and have a couple of questions regarding the option CO_PRODUCTION=.TRUE. from FDS5. Does anyone know why it was removed in the current version?

I am trying to do a toxicity assessment in a tunnel environment (train is on fire) based on output data from fds. My design fire is based on a full-scale test and it grows to about 45MW within 20 minutes – it goes into ventilation controlled.

As my FED calculation is sensitive to CO inputs, I am relying on more-or-less accurate CO values. In FDS 5 I would have done this with CO_PRODUCTION. The FDS 6 manual refers to the Finite Rate Combustion chapter, which is recommended for DNS only. Furthermore I want to use a surrogate fuel with an average CO yield. The average CO_yield of the top materials of my train in well ventilated fires about 0.05. In using the finite rate combustion approach, I guess I cannot specify any representative CO yield for a mix of materials anymore.

Any ideas how I could achieve reliable CO yields in FDS6 (under ventilated) without using the finite rate combustion approach? And if this is the only way, how could I achieve CO yields for a mix of materials and not just for the one material I have specified in the reaction?

Any help would be appreciated!

Michael

[dr_jfloyd]

Fundamental changes were made to the combustion model in going from FDS 5 to FDS 6.  The prior approach to do CO_PRODUCTION (which was somewhat limited) was no longer possible.   The new combustion model allows you to mix fast chemistry and finite rate chemistry.  If you look at the example in 12.3 of  the user's guide, you will see the fuel to CO step is fast chemistry and the CO to CO2 step is finite rate chemistry.  This is the approach that was used for cases involving CO production in the validation guide (Smyth slot burner [DNS], Beyler hood [LES], and NIST RSE/FSE [LES]).

[winklerm...@gmail.com]

Thank you for your helpful advice, I will have a look into that. 

Lets say I would pick PU foam with a certain chemical composition as a dominant fuel of my train and set up the equations like in example 12.3 ...then I would achieve CO for this material. The problem is that the train is not just PU foam. How could I incorporate the varying CO yields of other materials? Do you have any thoughts on that? In FDS5 I would have specified a CO_yield mix in order to represent different materials.

Thank you so much!

[dr_jfloyd]

You can only use CO_YIELD if you are using simple chemistry which limits you to one fuel.  If you wish to have multiple fuel species, you will need to define a detailed chemistry reaction for each fuel following the examples in the guide.

[spring-tiger]

I have two question about CO PRODUCTION.

1.

When using Finite Rate Reaction model(e.i. multiple &REAC), one cannot specify a desired CO yield.

CO yield depens only on stoichiometry.

-> RIGHT?

2.

When using Finite Rate Reaction model(e.i. multiple &REAC), one cannot use HRRPUA on any SURF line.

-> How can I set a given HRR ?

2014년 3월 5일 수요일 오후 7시 58분 27초 UTC+9, winklerm...@gmail.com 님의 말:

[Craig Weinschenk]

1. If you use finite chemistry, the CO/CO2 concentrations are coming from the finite rate not from a specified yield.

2. Specify the mass flux based on the heat of combustion from the first reaction.

--
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 post to this group, send email to fds...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/fds-smv/baf60e3b-8ac1-4549-bf50-98d4c82ac969%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

--

Craig Weinschenk

Fire Research Division

National Institute of Standards and Technology
www.cweinschenk.com

[Randy McDermott]

Why do you say "first reaction"?

I would think hrr for a boundary condition assumes complete combustion. No?

To view this discussion on the web visit https://groups.google.com/d/msgid/fds-smv/CAP8fiQwsUQ6PoAM6qLqeFtUumtasJ6FHc6D11dYmaEsX2xTfag%40mail.gmail.com.

[dr_jfloyd]

1)

You could still do a fixed yield.  You could define two different CO species.  In the reaction for the fuel oxidation step you could split the CO between the two species with the fixed yield assigned to one and the rest to the other

2) specify the fuel mass flux

[Craig Weinschenk]

Jason - I didn't think about that -- good point about specifying two different CO species.

If you specify HRRPU that assumes a single reaction. If you are going to use multi-step chemistry or multiple fuels, the user needs to explicitly specify the mass flux. In the case of multi-step chemistry this would be the fuel(s) that breaks down to CO.

To view this discussion on the web visit https://groups.google.com/d/msgid/fds-smv/CAC6wax11r2GyM5pNc%3Ds32mkOD3DX6G-n%2Bi539hViJJ99iN8%2BhQ%40mail.gmail.com.

For more options, visit https://groups.google.com/d/optout.

[Randy McDermott]

Craig,

I don't think I stated my question clearly.  Typically when I think of a "1 MW fire" translating this into a fuel mass flux would be done using the heat of combustion of the single reaction for complete combustion (regardless of whether we are considering multiple reactions in FDS or not).  The user should simply realize that if the fire is under-ventilated then they should not expect to recover the 1 MW in the HRR file.

R

To view this discussion on the web visit https://groups.google.com/d/msgid/fds-smv/CAP8fiQx2_Kg%3Dn83ieGBVkZkMOQHXeLOSF0erjHiOOyQELcw0Kw%40mail.gmail.com.

[Craig Weinschenk]

Ahh, yes you are correct. Sorry I definitely misunderstood what you were saying.

To view this discussion on the web visit https://groups.google.com/d/msgid/fds-smv/CAC6wax2snLTt3bZw0UBmQP7FT672pTd%2BCxDEuD5ojKquw9xkBQ%40mail.gmail.com.

For more options, visit https://groups.google.com/d/optout.

[spring-tiger]

Jason Floyd,

1)

You mean this?

('CARBON MONOXIDE_fixed' is additional species for fixed stoichiometry coefficient=0.1. corresponding to desired CO_YIELD)

&SPEC ID='CARBON MONOXIDE_fixed' C=1, O=1/

&REAC ID = 'R1'
             FUEL = 'PROPANE'
             SPEC_ID_NU = 'PROPANE','OXYGEN','CARBON MONOXIDE_fixed', 'CARBON MONOXIDE','WATER VAPOR'
             NU = -1,-3.5,0.1, 2.9,4
             CRITICAL_FLAME_TEMPERATURE = 730./

&REAC ID = 'R2'
             FUEL = 'CARBON MONOXIDE'
             A = 1.5e9
             E = 41840
             SPEC_ID_NU = 'CARBON MONOXIDE','OXYGEN','CARBON DIOXIDE'
             NU = -1,-0.5,1
             SPEC_ID_N_S = 'OXYGEN','CARBON MONOXIDE','WATER VAPOR'
             N_S = 0.25,1,0.5/

&REAC ID = 'R3'
             FUEL = 'CARBON DIOXIDE'
             A = 6.16e13
             E = 328026
             SPEC_ID_NU = 'CARBON DIOXIDE','OXYGEN','CARBON MONOXIDE'
             NU = -1,0.5,1
             SPEC_ID_N_S = 'OXYGEN','CARBON DIOXIDE','WATER VAPOR'
             N_S = -0.25,1,0.5
             N_T = -0.97/

2)

If stoichiometry co. is exactly specified, HoC does not have to be specified?

2014년 4월 2일 수요일 오전 10시 46분 10초 UTC+9, dr_jfloyd 님의 말:

[dr_jfloyd]

No. Look in the users guide. There is an example. Define CARBON MONOXIDE as a species but as a LUMPED_COMPONENT_ONLY. Then define to species mixtures whose only species is CARBON MONOXIDE. This way you can refer to the two "mixtures" in reactions bit still use CARBON MONOXIDE to get mass fraction outputs that aggregate the two.