CFAST doesn't save the fire settings!

[MartinR]

Dear Ladies and Gentlemen,

 

a week ago I downloaded CFAST to compare the results I got with FDS with zone model results. Althought I read everything in the cfast manual I had huge problems so far.Here is the biggest problem:

I want to add my own fire scenario under "Fire". So I add a fire, edit the fire object and add a new fire object (not t², I use a linear correlation). Here is the problem: When I write in my Heat of Combustion (45000 KJ/kg) it doesn't show an error. When I save the settings everything is fine. Then I close the program. When I start CFAST again and open the file I saved before, the data I wrote in before is gone. Instead it shows the old value (this problem refers to the folder "fire", in the other folders that delete of data doesn't exist). And not enough, that old value I mentioned is 0,004 kJ/kg. That number is to small and causes an error! The same problem with the molar mass, it just doesn't save my settings.

I hope you can help me with that problem.

Maybe it's a bug? I use CFAST 6.

 

Best Regards

 

Martin Rudert

[CFAST Development]

We'd need to know a bit more to see what's wrong.  What's the exact version of CFAST you are using?  You should be able to select Help, then About from the menu to see the complete version number. It should be something like 6.2.0.290 if you are using the latest officially released version of the model.  If it's this version or older, we'll need a copy of any input file (.in) and fire object file (.o) that you are having problems with.  If it's the latest test version (currently 6.3.0.559), we'll only need the input file.

Before any of that, please check to make sure your computer is set to use the U.S. English numeric units (at least for the account that uses CFAST). See http://www.nist.gov/el/fire_research/cfast-faqs.cfm#Q6 for details. This can cause a range of problems with numbers in CFAST.

Richard Peacock

[MartinR]

Hello,

I installed CFAST version 6.2.0, if its 290 I don't know. On the cfast homepage it is just named cfast 6.2.0 I downloaded last week, so I guess it is the newest official version, not the test version.

In the language settings changed the decimal separating item from komma to dot. Now it semms to save the changes I make, nice!

But there is still the confusing error message about the fire area that is less than 0 or bigger than 10000 m². But none of the fires in fire folder has an area less 0 or bigger 10000 m². Tis is why I still would like you to take a look on the attached files.

 

Thanks for your help!

 

Best Regards

Martin Rudert

[Richard Peacock]

What you are seeing is indeed an bug in the version of the user interface you are using so that the comparison made is actually <= zero rather than < zero.  We corrected it earlier this year for the next release version.  At this point, you should just ignore the error message.

On Tuesday, November 27, 2012 6:10:57 AM UTC-5, MartinR wrote:

[MartinR]

OK, and it still will calculate correctly´without complaining?

 

[CFAST Development]

Yes.  Messages you see in the GUI interface are just warnings intended to highlight model inputs that you may want to check.  The model will run just fine regardless of most messages that are displayed in the GUI interface.

[MartinR]

In order to the size of my compartment (36mx54m), would you recommend me to split that compartment into 2 separate compartments?

 

[CFAST Development]

My guess is that it won't make much difference in the results.  While it is a large compartment, your scenario also has a large fire so a single compartment is likely appropriate.  I'd pay particular attention to the fire inputs in you scenario to make sure the species inputs are appropriate for your fuel.  These typically have a greater impact on the results than dividing the compartment into smaller pseudo-compartments.

[MartinR]

Good to know that it is not so important to create more then one compartment. The next question that pops into my mind is about fire props. As you already mentioned that is very important. My fire develops from 0 sec to 30000 kW after 328 sec and remains constant for 15 min on that HRR. But my complete mass of fuel (polypropylene) is 150000 kg. The highest mass I can implement in CFAST is 10000 kg! Does that huge difference influence the result a lot?

 

 

[CFAST Development]

This input is actually not even used in the model's calculations.  It's been removed from the latest version that's under development where we've totally redone the combustion chemistry to be easier to use, understand, and document.

[MartinR]

Another issue: I read that when using vents in zone models I need to consider a factor of 0.65 or 0.7 as a flow factor. In my settings for the vents I alredy considered a flow factor of 0.7, because the German Standard indicates that. Do I now need to imply another factor of 0.7 because of the zone model (maybe the factor balances out a calculation issue?) or not?

 

Best Regards

Martin

[CFAST Development]

There is no way to change the vent coefficient in CFAST.  It is set in the code at 0.7 based on emperical studies conducted by NIST and others in the 1970s and 1980s.

 

Richard Peacock

[MartinR]

OK, thanks. Just another question to make sure I do the right thing:

The vertikal flow vents in the programm are the vents on the sides of the compartment (like inlets for the air), aren't they? And the vertical flow vents are the vents that connect the top of the building to the outside (like smoke flues)? Or is it the other way round?

I am realy not sure if I implemented the location of the vents correctly, because in CFST there are just a few parameters that describe the location. For ex. for the vertical flow vents I can't specifiy the detailed location. So I guess just the sum of the vent surface is important, not where exactly they are located. For the horizontal flow vents the parameters Sill, Soffit and Width can be used to describe the location of the vent. But I am not sure what the reference point is. For example, if my Vent is located at the front and 2 m away from the left corner, then I use a vent offset of 2 m, right? And when I give a width of 40 m in addition that means that my went goes from 2 m to 42 m?

 

I attached my CFAST-file and a screenshot of the compartment. Can you check if the parameters I took make sense or if they are completely nonsense? Or is that too much effort?

 

Also I don't need to give a smoke yield. Does CFAST take an average value for that or is there a way to implement that.

Refering to the fire description: As the normal vector of the fire I used x=0. y=0 and z=1, because I thought that this makes sense, if the fire "looks up". In the User's Manual I don't find any clear statements refering to these questions.

So. sorry for that "bombardement of questions". I hope you can answer some, because I am really unsue about the correctness of my values...

 

Best Regards

 

Martin Rudert

[MartinR]

Attached you can find some documents that you may need to get a better understanding about the questions above.

[CFAST Development]

On Tuesday, December 4, 2012 6:28:32 AM UTC-5, MartinR wrote:

OK, thanks. Just another question to make sure I do the right thing:

The vertikal flow vents in the programm are the vents on the sides of the compartment (like inlets for the air), aren't they? And the vertical flow vents are the vents that connect the top of the building to the outside (like smoke flues)? Or is it the other way round?

In CFAST, Horizontal Flow Vents are openings in the walls such as doors or windows.  Vertical Flow Vents are openings in floors and ceilings.  In the yet to be released new version, we have renamed these Wall Vents and Ceiling/Floor Vents to make it easier to understand (we hope).

 

I am realy not sure if I implemented the location of the vents correctly, because in CFST there are just a few parameters that describe the location. For ex. for the vertical flow vents I can't specifiy the detailed location. So I guess just the sum of the vent surface is important, not where exactly they are located. For the horizontal flow vents the parameters Sill, Soffit and Width can be used to describe the location of the vent. But I am not sure what the reference point is. For example, if my Vent is located at the front and 2 m away from the left corner, then I use a vent offset of 2 m, right? And when I give a width of 40 m in addition that means that my went goes from 2 m to 42 m?

In a zone model, the exact position is not critical, only the vent's location relative to the two gas layers/zones.  The offset for Horizontal Flow Vents is largely just for visualization although it is used in the calculation for corridor flow if included in the simulation.  The reference point is relative to the zero value for the wall of interest.  For the front and back walls, this is the "left" side of the structure viewed from the front.  For the side walls, this is the front of the structure.

 

 

I attached my CFAST-file and a screenshot of the compartment. Can you check if the parameters I took make sense or if they are completely nonsense? Or is that too much effort?

 

Also I don't need to give a smoke yield. Does CFAST take an average value for that or is there a way to implement that.

Refering to the fire description: As the normal vector of the fire I used x=0. y=0 and z=1, because I thought that this makes sense, if the fire "looks up". In the User's Manual I don't find any clear statements refering to these questions.

The smoke yield can have a significant impact on the overall temperatures in a simulation since it is key to getting the radiation calculation correct.  In the simplest combustion reaction, methane reacts with oxygen to form only carbon dioxide and water.  Real-life fires are seldom that simple so it is important to get the definition of the fire as correct as possible.  There are a number of sources for these data, most notably the SFPE Handbook of Fire Protection Engineering. The Chapter on "Generation of Heat and Chemical Compounds in Fires" includes numerous tables of measured yields for important inputs to fire models, including CFAST.

The normal vector is only important for ignition of secondary fires if ignition is specified as a critical value of temperature or heat flux. For a single fire or one that is specified to ignite as a particular time, the input is not used.

[MartinR]

OK, so an example to make it clear:

I want to position my horizontal flow vent on the front side 3 m away of the left corner of the front side. It is 2 m high, starts from the very bottom of the compartment (my building). The width of the vent (from the left side to the right side) is 4 m.

 

Then I use the follwing settings:

First compartment: Compartment X (my building), Vent offset: 3 m

Second Compartment: Outside (so I guess this is the environment of the building, so basically air), Vent offset: 3 m

Sill: 0 m

Soffit: 3 m

Width: 4 m

 

If I want the vents to open 10 s in delay: Initial opening fraction: 0, change fraction at: 10s, final opening fraction: 1

 

Wind Angle: 0 ° (no WInd)

Face: Front (or left?)

 

Are these assumptions correct?

 

Best Regards

Martin

[CFAST Development]

You are correct for placement of the vent.  It is on the front face. You can always check the placement by viewing the simulation in Smokeview.  The vent should be placed where you requested in the visualization.