Extinction model

[Stephen Olenick]

The FDS5 extinction model was a simply oxygen volume % and temperature correlation to determine if burning was able to occur.  In FDS6, though, works by resolving the cell temperature and comparing it to a critical flame temperature.  The tech ref guide indicates that one can return to the simpler oxygen species concentration model mentioned in Appendix D, which appears to be the same sub-model from FDS5.  But looking through the User's Guide, it wasn't readily apparent what command/option is necessary to use this simple model instead.  Can you advise what the syntax for that is?  Thanks.

Stephen Olenick

Principal Engineer

Combustion Science & Engineering, Inc.

[Randy McDermott]

Thanks, Stephen, for pointing this out.  We will update the guides.  The syntax should be:

&MISC EXTINCTION_MODEL='EXTINCTION 1'/

--
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/dc7bcb24-4842-4501-9146-690458eb9c4a%40googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

[Stephen Olenick]

Thanks Randy.  I was going to suggest you update the Guides too.  Am I correct, therefore, in saying that if nothing is put in, it uses the default critical flame temperature model, but if the syntax below is added, it goes to the limiting oxygen concentration model?

Also, I assume if one puts in the &MISC lines below, the CRITICAL_FLAME_TEMPERATURE and X_O2_LL can be set on the &REAC line similar to the old FDS5 functionality?

Thanks,

Stephen.

[Craig Weinschenk]

The Critical Flame Temperature is a REAC parameter and it applies to both Extinction Models. EXTINCTION 1 uses the Critical Flame temperature to solve for the O2 limit and than compares that to the available O2 in the cell. X_O2_LL is not currently an input variable.

To view this discussion on the web visit https://groups.google.com/d/msgid/fds-smv/82a76478-4877-41ad-9f7c-9f42a28fa836%40googlegroups.com.

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

--

Craig Weinschenk

Fire Research Division

National Institute of Standards and Technology
www.cweinschenk.com

[Stephen Olenick]

Thanks Craig.  Makes sense.  Any reason the development team removed X_O2_LL as an input if someone uses EXTINCTION 1?

[Craig Weinschenk]

Simplification and consistency. Both extinction models require the same input information. One can calculate a critical flame temperature using the lower limit of combustion for the fuel in their simulation and get the impact there. If you think this is something that should be added back in, I would start an issue with feature request and we can take a look at it more closely.

[Stephen Olenick]

Good point.  No, I think that works.  One suggestion I might have, though, would be to alter Appendix D of the Tech Ref Guide.  As noted in the User's Guide, the default critical flame temperature is 1327C (1600K).  Which I think is based on Beyler, SFPE.  But Appendix D of the Tech Ref Guide on the oxygen extinction model of Mowrer is written as if the default is 1700K, which it was in FDS5.  I think 1600K is the right number, but that would necessitate making a few tweeks to Appendix D.  Thoughts?

[Craig Weinschenk]

Good catch. I thought I caught that change. I will make the corrections.

To view this discussion on the web visit https://groups.google.com/d/msgid/fds-smv/6573e8fb-261f-47be-a063-01f564212c67%40googlegroups.com.

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

[Craig Weinschenk]

The appendix is updated. It will appear in the the next nightly build.

[Stephen Olenick]

Randy, I tried this syntax but got the old ERROR:   Problem with MISC line   Since it is the only command I have on the MISC line, can you check the syntax again?  Many thanks,

Stephen.

On Wednesday, January 15, 2014 8:34:04 AM UTC-5, Randy McDermott wrote:

[Craig Weinschenk]

I just tried using

&MISC EXTINCTION_MODEL='EXTINCTION 1'  /

and didn't get an error.  Using SVN 18123.

[Stephen Olenick]

yup yup.  got it working.  was using a depreciated version of FDS.  ~svn 15000.  6.0.1 official release works great.  Thanks,

Stephen.

[Chris]

For a Simulation of PVC I Need to have a LOC of 16.9%. How can I handle this in FDS 6?

Thanks

Chris

[Craig Weinschenk]

I would calculate the critical flame temperature based on a reaction where the oxygen is at 16.9% compared to the stoichiometric value. There is an example in Chapter 1 in the SFPE handbook.

To view this discussion on the web visit https://groups.google.com/d/msgid/fds-smv/cc61b127-c7c2-4d6a-8dc9-ed7e5effaf5a%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[Chris]

You suggest using Chapter 5 1-97 Calculation of Adiabatic Flame Temperatures? But the reaction of PVC seems highly complicated...

[Craig Weinschenk]

Chris,

In FDS, unless you specify otherwise PVC is going to react and form CO2 and H2O (and soot and CO if you specific yields). I would calculate the critical flame temperature based on that reaction (yes, I would use that equation). 

An alternative would be use the EXTINCTION_1 extinction model. Note that EXTINCTION_2 is the default model and is outlined in the FDS Tech Guide. EXTINCTION_1 is slightly simpler focused on just oxygen concentration. See Appendix D in the FDS Tech Guide. I would recommend following Equation D.4 and solving for T_f assuming T_m (bulk temperature) is your ambient temperature and the Y_{O_{2},lim} is the extinction concentration you mentioned for PVC - 0.169. This should provide the appropriate cutoff of combustion you are looking for. Once you find T_f - it can be set using CRITICAL_FLAME_TEMPERATURE on the REAC line. Also make sure you adjust the extinction model to EXTINCTION_1 on the MISC line

-Craig

To view this discussion on the web visit https://groups.google.com/d/msgid/fds-smv/6e2b9016-fa64-420c-a222-617058ced2eb%40googlegroups.com.

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

[Randy McDermott]

So that we are all on the same page, Craig is suggesting solving the following equation for T_c:

0.169 = 1.2*(T_c-293)/13100

T_c = 1551 - 273 = 1279 deg C

To view this discussion on the web visit https://groups.google.com/d/msgid/fds-smv/CAP8fiQxHw%3DV9j7cXYwijmT9DAi5p3i1dBeY0BurcrDiPkYuKQg%40mail.gmail.com.

[Chris]

Ok, thanks a lot!

[Stephen Olenick]

Just to add in some discussion Craig and I had off the board and my understanding of it.  This approach should work, but depending on the conditions in the submitter's actual model, he may not see suppression at 16.9%. At the time the model hits 16.9% and suppression is expected, Cp (1.2) may not be exactly correct, but Tm, the control volume bulk temperature, is not likely 293K.  Its probably higher, which means the oxygen level at suppression is going to be lower.  So should X_LL_O2 be added back in as an input or not?  Craig makes the very fair point the if the data for the suppression O2 is known, knowledge from that experiment could/should be used to do this calculation a little more properly.  But the result is that it may not suppress at the same O2 level.  Now, making this adjustment based on temperature is probably a more correct answer.  But if the user, for whatever reason, can guarantee suppression at a particular O2 level, should it be an input or not?  Or maybe just a little bit of text in the appendix should be added to note that, when using this approach, one is calculating the critical flame temperature for suppression in ambient temperature conditions, and that suppression will likely occur at some lower oxygen concentration.

Stephen.

[Randy McDermott]

Stephen,

In my mind we are just trying to avoid redundant inputs for EXTINCTION_1 (for those of you not yet with us, see the simple model described in Appendix A of the Tech Guide).  If X_LL_O2 is an input, it has to be pegged to a temperature, presumably ambient temperature (TMPA).  If so, it is identical to what I showed before.

The other option is two have two input parameters for the "simple" extinction model.  This draws a different slope for the line, effectively just adjusting the mean heat capacity assumed in the model.

As I am new to this discussion, please let me know if I'm off base.

R

To view this discussion on the web visit https://groups.google.com/d/msgid/fds-smv/86c49155-f3be-4e8d-b890-57edd710e9cd%40googlegroups.com.

[Stephen Olenick]

No, I think you are pretty spot on.  I think the issue relates to a general, yet incorrect, assumption that if one wants suppression at 16.9% and follows this guidance, the answer may be more physically correct, but it is highly unlikely that one will actually get suppression at 16.9% because the Tm will be elevated.  So I don't think adding in the extra input is the answer.  You could do one of two things I think.  Less desired, but possible, is just to add X_LL_O2 as an input without need for the CRITICAL_FLAME_TEMPERATURE input.  Don't peg it to a temperature, just state that is the volume fraction of oxygen in a cell drops below X_LL_O2 %, do not allow combustion.  I think a better approach, though, is doing what you are doing with the critical flame temperature as EXTINCTION=0, and if someone chooses EXTINCTION=1, do it just the way you are doing it, but perhaps add a note that the user should expect suppression at a lower oxygen concentration than calculated due to heightened temperature.

Stephen.

[dr_jfloyd]

I don't think we want to add any empirical models that are likely to be non-conservative.  If we just used the room temperature X_LL_O2, we would be.  Under this model we would extinguish a fire at 16.9 % even though we all know that if the room temperature is significantly higher than ambient, the fire would have burned longer.  Any hazard we might be evaluating (temperature, heat flux, toxicity, etc.) would be under predicted.  

[Stephen Olenick]

Right, so I think you agree with me that it is good now.  My only suggestion, if you think it warrants it, is to add a sentence to that effect in the Mowrer model appendix.  That way, the next time the next user does that calculation, they do not expect suppression at the oxygen concentration they are solving to, but at some lower concentration adjusted by temperature.  Or if you think it is clear, leave it be.

Stephen.

[Randy McDermott]

I think there used to be a plot in the appendix that made the model clear.  I've asked Craig to add that back.

To view this discussion on the web visit https://groups.google.com/d/msgid/fds-smv/c727fdd1-794c-46aa-9505-349f19181142%40googlegroups.com.

[Chris]

Randy,

I calculated another value:

0.169 = 1.2*(T_c-293)/13100

T_c = 2144 Kelvin

[Chris]

Actually I found slightly higher X_LL_O2-values in the Simulation then calculated with the CFT-Approach: I wanted 16.9% but it extincts already around 18%. The same for the standard-values: Should be around 12.2 % but extincts already at about 15%. 

[Stephen Olenick]

Chris,

I also calculated it to ~2140K.  

Not really sure why you would get higher values.  Perhaps Randy or Craig have some thoughts.

[dr_jfloyd]

Which extinction model are you using?  There are two.  The more complicated, default one (EXTINCTION 2) uses the actual enthalpies of the species present.  If you haven't given a meaningful enthalpy for PVC, then this approach may not work well.  The simpler one (EXTINCTION 1) assumes cp=1.2 (which works OK unless you have a lot of water vapor or other high cp species around).

[Randy McDermott]

Apparently, I can't use a calculator correctly (I subtracted 293 instead of added).  Sorry.

To view this discussion on the web visit https://groups.google.com/d/msgid/fds-smv/45cb70d9-167f-454e-8e2a-301a96e7b981%40googlegroups.com.

[Chris]

I just put

CRITICAL_FLAME_TEMP = 1865 (°Celsius)

AUTO_IGNITION_TEMP =  0.0

I could understand if these two approaches lead to slightly other results as in the X_O2_LL-approach you had to check the cell for the O2-concentration only and with the CFT-approach you check whether your flame temp (which is certainly also dependant somehow on X_O2) is lower than a specified value. 

[dr_jfloyd]

The default extinction model is EXTINCTION 2 which uses eq 5.19 in the Tech Reference and not the method in App D.  

On Friday, August 15, 2014 8:29:30 AM UTC-4, Chris wrote:

I just put

CRITICAL_FLAME_TEMP = 1865 (°Celsius)

AUTO_IGNITION_TEMP =  0.0

I could understand if these two approaches lead to slightly other results as in the X_O2_LL-approach you had to check the cell for the O2-concentration only and with the CFT-approach you check whether your flame temp (which is certainly also dependent somehow on X_O2) is lower than a specified value. 

[Chris]

One other question:

Is this right to set the Auto-Ignition-Temp for PVC to 0 °C? Actually something like a Auto-Ignition-Temperature doesn't exist for solids...

[dr_jfloyd]

We set the temperature to 0 K so users don't have to engineer some kind of ignition source.  In typical engineering uses of FDS, the user just wants to put in a prescribed fire and have it burn.  Having a low AIT does that.