FDS Liquid Fuels

[Afehksion]

Hello,

I'm new in FDS.

I wish to realize a fire simulation from a liquid. In the User Section 8.5.7, I saw that it was necessary to characterize the liquid with the MATL function.

Like this : 

&MATL ID = 'ETHANOL LIQUID'

      EMISSIVITY = 1.0

      NU_SPEC = 0.97

      SPEC_ID = 'ETHANOL'

      HEAT_OF_REACTION=836.98

      CONDUCTIVITY=0.17

      SPECIFIC_HEAT= 2.4398

      DENSITY= 789.

      ABSORPTION_COEFFICIENT = 1534.3

      BOILING_TEMPERATURE=78.5 /

But I fail to understand how the fire will be start. Because we can't use HRRPUA in the SURF function as MATL forced to use its liquid pyrolysis model.

I've tried to put an ignitor particle in the middle of my 'ETHANOL LIQUID' surface. As it's presented in the example : 'couch.fds'

But this involves to define the fuel with REAC function, and REAC does not accept MATL input. The only way, is to put FUEL = 'ETHANOL' in the REAC line.

I'm worried that FDS does not consider the ETHANOL LIQUID. FDS use ethanol just to start the fire ?

What were some other possible solutions to start the fire?

I'm sorry if it is not clear but I'm not an expert on the fire simulation.

Thank you in advance.

Jeremy

[Kevin]

https://github.com/firemodels/fds/blob/master/Validation/Pool_Fires/FDS_Input_Files/ethanol_1_m.fds

[Afehksion]

Thanks you very much, I will look into it.

[Afehksion]

Hello,

I have a question about REAC function, the specified fuel is only a gas ?

For my work, I have has to create a database for a lot of liquids and materials. And I'm asking which parameter is very important ?

For specified a liquid with SPEC function, I saw that FDS need the following properties :

- molecular weight MW

- ENTHALPY_OF_FORMATION : because the liquid will participate in a reaction

- conductivity or Lennard-Jones potentiel parameters

- diffusivity or Lennard-Jones potentiel parameters

- enthalpy or specific_heat and reference_temperature

- viscosity or Lennard-Jones potentiel parameters.

I tried to specify ethanol (liquid) in SPEC function with this properties. But my fire is different from the fire created with ethanol included in Table 12.1

Thank you in advance.

Jeremy

[Kevin]

Properties of gases and liquids used in FDS are stored in this file

https://github.com/firemodels/fds/blob/master/Source/data.f90

I'm not surprised that your results are different because these properties can be complicated. Can you tell me more about what you are trying to do?

[Afehksion]

Thanks for the file.

In fact, I'm an intern. My company works on the secondary circuit of a nuclear power plant. On part of my work is to create a database listing all the materials, liquids or gas present which may be present.

So I need to know which thermal proprieties are essential to realize the best simulation.

If i understand, the fuel in the REAC function is always a gas? And this gas results from a material or liquid burned ? In this case, REAC needs just FORMULA, CO_YIELD and SOOT_YIELD.

I saw in the user guide that materials are defined like that :

- CONDUCTIVITY

- DENSITY

- EMISSIVITY

- SPECIFIC_HEAT

This definition can be more complexe with N_REACTIONS, but it's not a problem for me I understood this part.

And for liquids : 

The first part of the definition with SPEC function :

- molecular weight MW

- ENTHALPY_OF_FORMATION : because the liquid will participate in a reaction

- conductivity or Lennard-Jones potentiel parameters

- diffusivity or Lennard-Jones potentiel parameters

- enthalpy or specific_heat and reference_temperature

- viscosity or Lennard-Jones potentiel parameters.

And the second part with MATL function : 

- EMISSIVITY
- NU_SPEC

- SPEC_ID = 'ETHANOL'
- HEAT_OF_REACTION
- (CONDUCTIVITY)
- (SPECIFIC_HEAT)
- DENSITY
- ABSORPTION_COEFFICIENT
- BOILING_TEMPERATURE

It's correct or I don't understand something ?

Thank you in advance

[Kevin]

Many of these parameters that you list are not necessary for design applications. If your objective is to SPECIFY (as opposed to PREDICT) the heat release rate, then you need only specify on the REAC line the name of the fuel (if it is listed in the FDS database) and the soot and CO yields. If the fuel is not listed, you need to specify the chemical formula. You do not need solid or liquid properties unless you are trying to predict the burning rate of a liquid pool or droplets. Or the pyrolysis of a solid slab, like wood or plastic. The properties required are given in the examples in the User's Guide.

[Afehksion]

Thank you for you reply, It allow me to see better.

My compagny needs to see the fire spread resulting from a heat release rate specified.

If I understand to start the fire. FDS need juste a REAC line with formula, soot and CO yields. But the other liquids or materials need to be more specified, because the fire can spread and burn them.

For the liquids and materials, who aren't present in the reac line. I need to use all the parameters quoted previously ?

Other questions :

I have been looking the file and I don't understand some parameters used to specify the liquids or gases

1/ H_L is HEAT_OF_VAPORISATION on the SPEC function ?

2/ H_V is REFERENCE_ENTHALPY ?

3/ Why sometimes T_REF is equal to T_BOIL and sometimes is equal to T_MELT ?

4/ G_F (ENTHALPY_OF_FORMATION ?) and CP can be defined by a function? (I know that it's possible to vary CP with RAMP function)

5/ To which temperature, the DENSITY and CONDUCTIVITY  must be supplied ? (because in sight of tabuled values, I found no temperature corresponding at the same time to the density and to the conductivity)

6/ What is BETA ?

I'm so sorry about all this question, but I need to understand everything in detail.

Thank you in advance.

Jeremy

[Kevin]

15.3.1 Thermal Properties

The following parameters should be specified to control the evaporation. The INITIAL_TEMPERATURE of
liquid droplet; assumed ambient, TMPA ( C) is specified on PART input. If the fluid given by the SPEC_ID is
included in Table 12.1, then no further inputs are required. Otherwise, you must provide all of the following
properties of the liquid on the SPEC input:

DENSITY_LIQUID The density of the liquid or solid droplet/particle (kg/m3).

SPECIFIC_HEAT_LIQUID Specific heat of liquid or solid droplet/particle (kJ/(kg K)).

RAMP_CP_L Ramp of temperature vs. specific heat for the solid droplet/particle.

VAPORIZATION_TEMPERATURE Boiling temperature of liquid droplet ( C).

MELTING_TEMPERATURE Melting (solidification) temperature of liquid droplet ( C).

HEAT_OF_VAPORIZATION Latent heat of vaporization of liquid droplet (kJ/kg).

ENTHALPY_OF_FORMATION The heat of formation of the gas (kJ/mol).

H_V_REFERENCE_TEMPERATURE The temperature corresponding to the provided HEAT_OF_VAPORIZATION
( C).
15.3.2 Radiative P

[nguyen dac]

Hi, 

I am also new to FDS. As I know, only liquid described on REAC line can burn. For liquids which are not presented on the REAC line, their thermal parameters only afftect the heat transfer process.

[Afehksion]

Thanks

As I understand, if I have 10 liquids in my simulation I need to use 10 REAC lines ?

And I define liquids with SPEC function :

DENSITY_LIQUID The density of the liquid or solid droplet/particle (kg/m3).

SPECIFIC_HEAT_LIQUID Specific heat of liquid or solid droplet/particle (kJ/(kg K)).

RAMP_CP_L Ramp of temperature vs. specific heat for the solid droplet/particle.

VAPORIZATION_TEMPERATURE Boiling temperature of liquid droplet ( C).

MELTING_TEMPERATURE Melting (solidification) temperature of liquid droplet ( C).

HEAT_OF_VAPORIZATION Latent heat of vaporization of liquid droplet (kJ/kg).

ENTHALPY_OF_FORMATION The heat of formation of the gas (kJ/mol).

H_V_REFERENCE_TEMPERATURE The temperature corresponding to the provided HEAT_OF_VAPORIZATION
( C).

I can apply directly this liquid to the surface or I have to define liquid as a liquid fuel with MATL function ?

[Randy McDermott]

You need a REAC line for each gas phase volatile reaction you want to occur independently.  You could define a LUMPED SPECIES fuel (with many species) and still only need a single REAC line.  In that case, you would add LUMPED_COMPONENT_ONLY=T on SPEC.  See, for example, sec 12.2.1 of user guide, "Complex Fuel Molecules" (which should probably be renamed "Complex Fuel Mixtures").  There we define a "natural gas" as a mixture of several species, and this requires only one REAC line.

On the other hand, if you wish to mimic a distillation curve of, say, gasoline, where lighter compounds volatilize and burn at faster rates, then you will need independent SPEC and REAC lines.

--
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+unsubscribe@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/3b98a40f-8086-4b09-8e9a-783011110d82%40googlegroups.com.

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

[Afehksion]

Thank's a lot, I'm going to look at this.

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.

[Afehksion]

I'm looking https://github.com/firemodels/fds/blob/master/Source/data.f90, to compare my values of the specific heat for ethanol gas and the value in this script. And I was surprised to that the values of specific heat in the script is always negative.

I don't understand because this script don't use the coefficient of NASA/TP-2002-211556 as for other liquids.

[dr_jfloyd]

There are two specific heat values used in FDS

H(T) = cpbar(T) T = Integral( cp(T') dT', T'=T0,T) + H(T0)

Since enthalpy, H(T), can be negative, the value of cpbar(T) can also be negative.

cp(T), however, is always positive.

The cp calcluation in data.f90 is for cp(T).  cpbar(T) is computed in read.f90

[Afehksion]

Ok I see, thank you

[Afehksion]

Excuse me but I don't understand why T_BOIL = 358.8 K. Because in the literature, the boiling point of ethanol is 78.4°C so 351.55 K

CASE('ETHANOL') !NASA/TP-2002-211556
	T = MIN(390._EB,MAX(REAL(I_TMP,EB),159._EB))
	CP = 450111.594_EB*T**(-2)-10208.2899_EB/T+101.426678_EB-0.387467261_EB*T+0.000712139261_EB*T**2 &
	-0.000000185707145_EB*T**3+0.000000000203762257_EB*T**4
	CP = CP * 8314.472_EB / 46.06844_EB !J/kg/K
	H_L = -6656155.11985071_EB !J/kg
	H_V = 1075310.47205154_EB !J/kg
	T_REF = 158.8_EB !K
	T_MELT = 158.8_EB !K
	T_BOIL = 358.8_EB !K
	DENSITY = 790._EB !kg/m^3
	FUEL = .TRUE.
	MU = 0.001095_EB !N.s/m^2
	K = 0.203_EB !W/m/K
	BETA = 0.00118_EB !1/K

[Afehksion]

I'm so lost with the specification of liquids and gases who participate in a combustion.

I understand that for gases or liquids not listed in the FDS database, we have to specified only FORMULA, SOOT_YIELD and CO_YIELD.

But gases or liquids need also to be specified by SPEC ? And if I specified a liquid with spec, I have to use this liquids as fuel  otherwise FDS makes no link between the fuel in REAC and my liquid.

I did a lot of fire simulation to find the same simulation between my liquid specification and the liquid listed on the FDS database. But there were great varations between this different variation.

So my real question is : If I have to use liquids not listed in FDS data base (as oil). Which proprieties must I fill in FDS ?

I have to specify a gas phase and a liquid phase with two different spec ?

I tried also to work on the fire spread. I've made a simulation with 2 liquids pool (listed in FDS data base). The two liquids are specified in two differents REAC lines. But the fire started directly in the two pools. Is there any way to start the fire only in one pool, and spreads in the other pools when the thermal conditions are respected ?

Thank you in advance.

[Randy McDermott]

The thermal properties in the FDS database come from the NASA polynomials.  It's possible there was an error in copying these over (Jason or Topi, please double check).

"But gases or liquids need also to be specified by SPEC ?"

When you specify a gas with SPEC then FDS tracks this gas (allocates memory for an array) for that gas unless you add LUMPED_COMPONENT_ONLY=T.

"And if I specified a liquid with spec, I have to use this liquids as fuel  otherwise FDS makes no link between the fuel in REAC and my liquid."

The link between REAC and SPEC is given by the parameters on the REAC line.  You need a FUEL and you need the stoichiometry is the reaction is not "simple chemistry", as discussed in the user guide.

"So my real question is : If I have to use liquids not listed in FDS data base (as oil). Which proprieties must I fill in FDS ?

I have to specify a gas phase and a liquid phase with two different spec ?"

I think it is best here to simply follow the Pool_Fires example.  But you may want to post a minimal input file and it might help us understand where you are stuck.

"Is there any way to start the fire only in one pool, and spreads in the other pools when the thermal conditions are respected?" 

There are at least a couple ways to do this.  First, you could set an IGNITION_TEMPERATURE on the second pool, and fuel would only evaporate after the pool reaches that temperature.  However, since vaporization and ignition are really separate phenomena, another way would be the following.

If you want to model ignition then you can specify a pilot ignition zone around the base of one pool, but not the other.  To do this set an INIT with XB around the base of the first pool with AUTO_IGNITION_TEMPERATURE=-273.  Then on your REAC line set AUTO_IGNITION_TEMPERATURE=?? (540 C for methane, but look it up in SFPE handbook for your fuel gas).

To unsubscribe from this group and stop receiving emails from it, send an email to fds-smv+unsubscribe@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/41fdfa38-e375-4c07-ae7d-0521bbe3053f%40googlegroups.com.

[dr_jfloyd]

Not sure why we have the incorrect boiling point for ethanol.  I'll give a pass through the species using the NASA polynomials and check to see if I notice any errors.

Specifying liquid properties depends on how you are tracking the liquid.  If you are using liquid droplets, then you need to specify properties on SPEC,  If you have a pool of liquid defined using SURF and MATL, then you need to define the properties using MATL. Gas properties are always specified on SPEC and in the case of droplets you would use the same SPEC input for both the gas and the liquid.

[Afehksion]

Thank you Randy and Jason.

I will try you advice for the fire spread.

 

With regard liquid specification :

Thus the specification of a liquid depends on its use in the simulation.

I made 4 simulations of liquid pool :

- the first with 'ETHANOL' (listed in the FDS data base)

- the second by specifying gas properties of ethanol

- the third by specifying ethanol as a liquid droplets

- the fourth by specifying gas and liquid properties with the maximum of data found on the link : https://github.com/firemodels/fds/blob/8778c4310a642c93ae281e734c1418816d3cd99d/Source/data.f90

 

I asked FDS to measure the temperature every second. But for each simulation the temperature is differente.

I need to find approximativly the same temperature that I find with 'ETHANOL'.

I give you the four simulations for you to see what I try to make.

 

Thank in advance

[Kevin]

The confusing part of all this is that the droplet evaporation model and the liquid pool evaporation model in FDS were developed independently, and the specification of parameters is not the same. For example, there is BOILING_TEMPERATURE on the MATL line and VAPORIZATION_TEMPERATURE on the SPEC line.

We need to rework the guides to make things clearer. But let me just say there are two situations where we work with liquids -- evaporating liquid droplets and evaporating liquid pool. Consider each

Evaporating Droplets:  Specify a PART line with a SPEC_ID. On the SPEC line, add the properties in Section 15.3.1 "Thermal Properties".

Evaporating Pool: Specify a SURF line with a MATL_ID. On the MATL line, specify SPEC_ID of the gas species. You do not need to list any liquid phase properties on the SPEC line. In an LES calc, don't list things like VISCOSITY and CONDUCTIVITY and SPECIFIC_HEAT. These are all modeled as part of the LES model. You really just need the molecular weight if the species is not a known gas. On the MATL line, you need the BOILING_TEMPERATURE, ABSORPTION_COEFFICIENT, CONDUCTIVITY, DENSITY, SPECIFIC HEAT of the liquid. HEAT_OF_REACTION is essentially the heat of vaporization.

[Afehksion]

Thank you Kevin, you have given a lucid explanation.

I will try this.

[Afehksion]

I tried with your explanation, I have just specified gaz properties.

But the result was inconclusive.

I give you the 2 simulations and a curve representing the temperature of the two simulations.

I think some properties are missing but I don't know which ones.

Thank in advance.

[Afehksion]

"Is there any way to start the fire only in one pool, and spreads in the other pools when the thermal conditions are respected?" 

There are at least a couple ways to do this.  First, you could set an IGNITION_TEMPERATURE on the second pool, and fuel would only evaporate after the pool reaches that temperature.  However, since vaporization and ignition are really separate phenomena, another way would be the following.

If you want to model ignition then you can specify a pilot ignition zone around the base of one pool, but not the other.  To do this set an INIT with XB around the base of the first pool with AUTO_IGNITION_TEMPERATURE=-273.  Then on your REAC line set AUTO_IGNITION_TEMPERATURE=?? (540 C for methane, but look it up in SFPE handbook for your fuel gas).

I tried the two solutions but :

- The first : FDS error because a SURF cannot have a REACting MATL and IGNITION_TEMPERATURE

- The second : I tried the file in attachment. But the fire start in the 2 pools at the same time.

May be, I forgot something but I don't know what.

Thank you in advance.

[Randy McDermott]

You need AIT on the REAC line:

&REAC FUEL = 'ETHANOL'

SOOT_YIELD = 0.001

CO_YIELD = 0.008

        AUTO_IGNITION_TEMPERATURE=365 / from SFPE handbook

--
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+unsubscribe@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/709fba09-4b7c-4628-9309-bcfd734298bc%40googlegroups.com.

[Randy McDermott]

Regarding the first approach, I forgot that this only works with specified MLR.

[Kevin]

Let's return to the problem of defining ETHANOL. First, compare the HRR of each case, not the temperature.

Ethanol has a radiative fraction of 0.2 --> RADIATIVE_FRACTION=0.2 on the REAC line

Ethanol has radiative properties similar to METHANOL  --> RADCAL_ID='METHANOL' on the SPEC line.

There are probably other reserved properties of ethanol, but this will get you closer.

[Afehksion]

Thank Kevin and Randy,

I have tried your advice Kevin, and I obtained an approximate error of 10% between HRR of ethanol listed and HRR of ethanol built. 

But for other liquids/gases, how I can find radiative_fraction or radcal_id ?

I suppose that each liquid/gas don't have a radcal_id or can have a radcal_id of an liquid not listed in a data base. It's right ?

I have tested also the temperature, because it's the parameter the most important for my company.

If you have an other idea of properties missing, i'm taker.

Concerning fire spread,

I forgot AIT on the REAC, I will test with it. I come back to you if it doesn't perform.

[Afehksion]

Concerning fire spread,

For 1 fuel, Randy advice perform. But when I tried the fire spread with 2 pools (each pool has a liquid different), the fire start in same time in each pool.

I didn't forget AIT for the second liquid.

I give you my file, in case or I did something bad.

Thank you

[Afehksion]

[Randy McDermott]

You have SUPPRESSION=F.  In that case, none of the checks to stop the reaction take place.

On Mon, Apr 24, 2017 at 8:26 AM, Afehksion <afe...@gmail.com> wrote:

--
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+unsubscribe@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/3f71a66b-88e5-4f58-9fe4-1d1805c9a7f8%40googlegroups.com.

[Afehksion]

But when I put SUPPRESSION=T, FDS displayed an error message : The default EXTINCTION MODEL is designed for 1 reaction.

Le lundi 24 avril 2017 14:59:03 UTC+2, Randy McDermott a écrit :

You have SUPPRESSION=F.  In that case, none of the checks to stop the reaction take place.

On Mon, Apr 24, 2017 at 8:26 AM, Afehksion <afe...@gmail.com> wrote:

--
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.

[Randy McDermott]

Try EXTINCTION_MODEL='EXTINCTION 1'.  This invokes the O2-based model (see tech guide appendix).

Unfortunately, this is under development---the current extinction model was not designed for multiple reactions.

To unsubscribe from this group and stop receiving emails from it, send an email to fds-smv+unsubscribe@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/54bce734-ab1c-4da1-82a2-94537c8e069a%40googlegroups.com.

[Jeremy Q]

Thank Randy, the fire spread between 2 fuels performs with your advice. 

Le mardi 25 avril 2017 12:46:05 UTC+2, Randy McDermott a écrit :

Try EXTINCTION_MODEL='EXTINCTION 1'.  This invokes the O2-based model (see tech guide appendix).

Unfortunately, this is under development---the current extinction model was not designed for multiple reactions.

On Tue, Apr 25, 2017 at 3:29 AM, Afehksion <afe...@gmail.com> wrote:

But when I put SUPPRESSION=T, FDS displayed an error message : The default EXTINCTION MODEL is designed for 1 reaction.

Le lundi 24 avril 2017 14:59:03 UTC+2, Randy McDermott a écrit :

You have SUPPRESSION=F.  In that case, none of the checks to stop the reaction take place.

On Mon, Apr 24, 2017 at 8:26 AM, Afehksion <afe...@gmail.com> wrote:

--
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/3f71a66b-88e5-4f58-9fe4-1d1805c9a7f8%40googlegroups.com.

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

--
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.

[Jeremy Q]

I have tested fire spread between liquids and liquids, it's performs.

But now, I tried between liquids and solids. I have tested with PVC, but the PVC does not seem to catch fire. However, in the PVC specification I put REFERENCE_TEMPERATURE = 200 and the temperature is above 200°C.

Somebody can explain me why PVC does not catch fire ?

I give you my file

Thank you in advance

[Kevin]

I don't know if it's possible to have two fuels.

[Jeremy Q]

I don't think so, because it performs with 2 liquids differents.

I tried with only one fuel but the PVC does not catch fire even if I have put REFERENCE_TEMPERATURE at 100°C.
I think the problem comes from the specification of PVC but I don't know where.

If someone have an idea, I am taker.

[Kevin]

I added

&BNDF QUANTITY='WALL TEMPERATURE' /
&BNDF QUANTITY='MASS FLUX', SPEC_ID='PVC_GAS' /

and I see that your PVC slab is generating this PVC_GAS. But I don't know if the input file is set up appropriately to have two gas phase reactions.

Randy or Jason -- could you look at the file and see if two gas phase reactions are possible in this case.

[Randy McDermott]

Well, there are 2 REAC lines, but both contain CHECK_ATOM_BALANCE=F, which means the stoichiometry of probably wrong.

--

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+unsubscribe@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/ff93d8cc-a7cf-4e21-a8ca-cf563a3ad091%40googlegroups.com.

[Jeremy Q]

Ok thank you, 

I will work on it tomorrow to remain CHECK_ATOM_BALANCE=F and test your suggestion Kevin.

Le jeudi 27 avril 2017 15:37:42 UTC+2, Randy McDermott a écrit :

Well, there are 2 REAC lines, but both contain CHECK_ATOM_BALANCE=F, which means the stoichiometry of probably wrong.

On Thu, Apr 27, 2017 at 9:26 AM, Kevin <mcgr...@gmail.com> wrote:

I added

&BNDF QUANTITY='WALL TEMPERATURE' /
&BNDF QUANTITY='MASS FLUX', SPEC_ID='PVC_GAS' /

and I see that your PVC slab is generating this PVC_GAS. But I don't know if the input file is set up appropriately to have two gas phase reactions.

Randy or Jason -- could you look at the file and see if two gas phase reactions are possible in this case.

On Thursday, April 27, 2017 at 3:54:30 AM UTC-4, Jeremy Q wrote:

I don't think so, because it performs with 2 liquids differents.

I tried with only one fuel but the PVC does not catch fire even if I have put REFERENCE_TEMPERATURE at 100°C.
I think the problem comes from the specification of PVC but I don't know where.

If someone have an idea, I am taker.

Le mercredi 26 avril 2017 21:53:57 UTC+2, Kevin a écrit :

I don't know if it's possible to have two fuels.

On Wednesday, April 26, 2017 at 11:06:46 AM UTC-4, Jeremy Q wrote:

I have tested fire spread between liquids and liquids, it's performs.

But now, I tried between liquids and solids. I have tested with PVC, but the PVC does not seem to catch fire. However, in the PVC specification I put REFERENCE_TEMPERATURE = 200 and the temperature is above 200°C.

Somebody can explain me why PVC does not catch fire ?

I give you my file

Thank you in advance

--
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.

[Jeremy Q]

The stoichiometry was wrong for PVC, I modified it and remained CHECK_ATOME_BALANCE=F.

Now the stoichiometry is right. But the PVC does not catch fire.

When Kevin said that : 'But I don't know if the input file is set up appropriately to have two gas phase reactions.'

Are there anything to add to have two gas phase reactions ? FDS does not manage this alone ?

It performs with 2 liquids pool. I don't understand where is the difference because at 200°C 'PVC_GAS' is generated and normally he should burn.

Thank you

[dr_jfloyd]

The PVC is burning in the gas phase; just not a lot of heat release rate (see the attached image).  Your gave your first surface reaction a heat of combustion of 2025 kJ/kg which is a small number.  Are you missing a digit ?

Add IDs to your REAC lines:

&REAC FUEL = 'ETHANOL',ID='ETHANOL'

      SPEC_ID_NU = 'ETHANOL','AIR','PRODUCTS_1'

      NU = -1, -2.967931, 1 /

&REAC FUEL='PVC_GAS',ID='PVC'

HEAT_OF_COMBUSTION=16400

SPEC_ID_NU='PVC_GAS','AIR','PRODUCTS_2',

NU=-1,-1.5975,1 /

and add these SLCF lines:

&SLCF PBY = 0.6, QUANTITY='HRRPUV REAC', REAC_ID='ETHANOL'/

&SLCF PBY = 0.6, QUANTITY='HRRPUV REAC', REAC_ID='PVC'/

[Jeremy Q]

Thanks a lot Jason.

I took the PVC defined here : https://github.com/firemodels/fds/blob/e089e4ed0c2439d17beefa70d8b40d0baee186d5/Validation/FAA_Polymers/FDS_Input_Files/FAA_Polymers_PVC_3_75.fds

I have just modified A and E by REFERENCE_TEMPERATURE.