Fire objects with heat flux and temperature ignition conditions do not ignite when conditions are met.

[andersonad]

As per the title, fire objects in a CFAST 7 simulation do not appear to respect the heat flux and temperature ignition conditions.

To reproduce, run the following input file, and note that the bunsen fire never ignites, despite the coincident target showing it has surpassed the heat flux condition to ignite. A similar exercise can be repeated for temperature ignition conditions.

This issue was encountered using the gnu_linux compilation of CFAST.

VERSN,7,CFAST Simulation !! !!Scenario Configuration Keywords !! TIMES,900,-50,10,10 EAMB,293.15,101300,0 TAMB,293.15,101300,0,50 !! !!Material Properties !! MATL,GYPSUM,0.16,900,790,0.016,0.9,Gypsum Board (5/8 in) MATL,METHANE,0.07,1090,930,0.0127,0.04,"Methane, a transparent gas (CH4)" !! !!Compartment keywords !! COMPA,Compartment 1,3.6,2.4,2.4,0,0,0,GYPSUM,OFF,GYPSUM !! !!Vent keywords !! HVENT,1,2,1,0.91,2.1,0,1,1,0,1,1 !! !!Fire keywords !! GLOBA,10,393.15 !!sofatriangle FIRE,1,1.8,1.2,1,1,1,0,0,0,1,sofatriangle CHEMI,1,4,0,0,0,0.3,5E+07,METHANE TIME,0,200,400 HRR,0,3000000,0 SOOT,0,0,0 CO,0,0,0 TRACE,0,0,0 AREA,0.1,1,0.1 HEIGH,0,0,0 !!bunsen FIRE,1,0.5,1,1,1,3,15000,0,0,1,bunsen CHEMI,1,4,0,0,0,0.33,5E+07,METHANE TIME,0,60,120,180,240,300,360,420,480,540,1800 HRR,0,100000,150000,200000,150000,125000,100000,90000,80000,75000,75000 SOOT,0,0,0,0,0,0,0,0,0,0,0 CO,0.001047221,0.001047221,0.001047221,0.001047221,0.001047221,0.001047221,0.001047221,0.001047221,0.001047221,0.001047221,0.001047221 TRACE,0,0,0,0,0,0,0,0,0,0,0 AREA,0.1,0.1,0.1,0.1,0.1,0.1,0.1,0.1,0.1,0.1,0.1 HEIGH,0,0,0,0,0,0,0,0,0,0,0 !! !!Target and detector keywords !! TARGET,1,0.5,1,1,0,0,1,GYPSUM,IMPLICIT,PDE,0.5

[mcgratta]

Thanks, we'll take a look at it.

[mcgratta]

When I run the case on my Windows 7 machine, I see the bunsen fire ignite. Or let's say that I see two fires in the Smokeview rendering.

[andersonad]

Having now built CFAST 7 on my Windows 7 machine using the gnu_windows script and run the simulation, it is still not showing secondary ignition of the bunsen fire.

Both the windows and linux instances of CFAST I have tried were built using the gfortran compiler. Was your CFAST instance built using the intel compilers?

[mcgratta]

Yes. I am cloning the CFAST repository on my linux machine now. I'll see if I can replicate the problem with the gnu compiler. We normally compile with Intel Fortran. We are at version 15 or 16; that is, fairly recent.

[rpeacoc]

I've run things on both OS X and Linux with GNU and at least the sample test case shows an ignited fire. We can check more examples.

Sent from my iPhone

On Nov 5, 2015, at 4:51 PM, Kevin McGrattan notifications@github.com wrote:

Yes. I am cloning the CFAST repository on my linux machine now. I'll see if I can replicate the problem with the gnu compiler. We normally compile with Intel Fortran. We are at version 15 or 16; that is, fairly recent.

— Reply to this email directly or view it on GitHub.

[andersonad]

I am using the default gnu fortran compiler for Ubuntu 14.04 in this instance, which is gfortran 4.8.4. What version of gfortran are you using when compiling? I might just be too far out of date.

[rpeacoc]

I'm using the same default Ubuntu installation. I'll take a more careful look tomorrow morning

Sent from my iPhone

On Nov 5, 2015, at 6:10 PM, andersonad notifications@github.com wrote:

I am using the default gnu fortran compiler for Ubuntu 14.04 in this instance, which is gfortran 4.8.4. What version of gfortran are you using when compiling? I might just be too far out of date.

— Reply to this email directly or view it on GitHub.

[andersonad]

Running this test case on Mac OS X 10.11.1 (El Capitan) compiled using gfortran 5.2.0 also failed to produce a secondary ignition from the bunsen fire.

At this juncture I have reproduced this issue on Windows 7, Mac OS X, and Ubuntu 14.04 using gfortran 4.6.3 (windows cygwin default), 5.2.0 (Mac OS X most recent available), and 4.8.4 (ubuntu 14.04 default), respectively.

To restate what you are looking for, the !!bunsen fire object should ignite secondary to the main !!sofatriangle fire at 15 kw/m^2 incident heat flux. Presently I am confirming this condition by opening the _n.csv and examining whether the bunsen fire ever has a non-zero HRR. You can also examine the _w.csv target 1's total heat flux to confirm that the condition is indeed met. Target 1 is coincident with the bunsen fire, so if it ignites, you will likely start seeing NaN for its heat flux.

[rpeacoc]

I"ve reproduced the problem with a test case that fails on windows and should also fail with gnu. Once I resolve the error on windows, I'll verify it works on gnu.

I should note that you do need to update the input file with the new version of CFAST since it no longer creates automatic targets for fires. Fires are now just linked to a user specified target.

[rpeacoc]

With the latest source code, 56e8f45, objects do ignite in the windows intel compiler version but does die with the gnu compiler. The submitted data file is in an older format and needs to be updated for the release version of CFAST. I've attached an updated file (that I just opened in the GUI and saved the file to update it). We'll look and see why it's failing with the gnu compiler.

VERSN,7,CFAST Simulation !! !!Scenario Configuration Keywords !! TIMES,900,50,10,10 EAMB,293.15,101300,0 TAMB,293.15,101300,0,50 !! !!Material Properties !! MATL,GYPSUM,0.16,900,790,0.016,0.9,Gypsum Board (5/8 in) MATL,METHANE,0.07,1090,930,0.0127,0.04,"Methane, a transparent gas (CH4)" !! !!Compartment keywords !! COMPA,Compartment 1,3.6,2.4,2.4,0,0,0,GYPSUM,OFF,GYPSUM,50,50,50 !! !!Vent keywords !! HVENT,1,2,1,0.91,2.1,0,1,1,1 !! !!Fire keywords !! !!sofatriangle FIRE,1,1.8,1.2,1,1,TIME,0,0,0,0,sofatriangle CHEMI,1,4,0,0,0,0.3,5E+07 TIME,0,200,400 HRR,0,3000000,0 SOOT,0,0,0 CO,0,0,0 TRACE,0,0,0 AREA,0.1,1,0.1 HEIGH,0,0,0 !!bunsen FIRE,1,0.5,1,1,1,FLUX,15000,Ign_bunsen,0,0,bunsen CHEMI,1,4,0,0,0,0.33,5E+07 TIME,0,60,120,180,240,300,360,420,480,540,1800 HRR,0,100000,150000,200000,150000,125000,100000,90000,80000,75000,75000 SOOT,0,0,0,0,0,0,0,0,0,0,0 CO,0.001047221,0.001047221,0.001047221,0.001047221,0.001047221,0.001047221,0.001047221,0.001047221,0.001047221,0.001047221,0.001047221 TRACE,0,0,0,0,0,0,0,0,0,0,0 AREA,0.1,0.1,0.1,0.1,0.1,0.1,0.1,0.1,0.1,0.1,0.1 HEIGH,0,0,0,0,0,0,0,0,0,0,0 !! !!Target and detector keywords !! TARGET,1,0.5,1,1,0,0,1,GYPSUM,EXPLICIT,PDE,0.5,Targ 1 TARGET,1,0.5,1,1,0,0,1,METHANE,EXPLICIT,PDE,0.5,Ign_bunsen

[rpeacoc]

I've updated the compiler options for gnu fortran and the object ignition seems to work correctly for a correct version 7 input file in 5756bd6. Here's a file that has three fires, two ignited by heat flux or temperature by the initial fire. See if it works and let us know.

VERSN,7,CFAST Simulation !! !!Scenario Configuration Keywords !! TIMES,900,50,10,10 EAMB,293.15,101325,0 TAMB,293.15,101325,0,50 !! !!Material Properties !! MATL,SOFTWOOD,0.12,1380,510,0.019,0.9,"Wood, Softwoods (fir, pine) (3/4 in)" !! !!Compartment keywords !! COMPA,Comp 1,5,5,5,0,0,0,OFF,OFF,OFF,50,50,50 !! !!Vent keywords !! HVENT,1,2,1,1,2,0,2.5,1,1 !! !!Fire keywords !! !!Main FIre FIRE,1,2.5,2.5,0,1,TIME,0,0,0,0,Main FIre CHEMI,1,4,0,0,0,0.35,5E+07 TIME,0,1000,3600,3610 HRR,0,1000000,1000000,0 SOOT,0.05,0.05,0.05,0.05 CO,0.0195,0.0195,0.0195,0.0195 TRACE,0,0,0,0 AREA,0.3,0.3,0.3,0.3 HEIGH,0,0,0,0 !!Fire by Temp FIRE,1,2.5,3.5,0.5,1,TEMP,523.15,Fire 2 Targ,0,0,Fire by Temp CHEMI,1,4,0,0,0,0.35,5E+07 TIME,0,1,999,1000 HRR,0,100000,100000,0 SOOT,0.05,0.05,0.05,0.05 CO,0.01954722,0.01954722,0.01954722,0.01954722 TRACE,0,0,0,0 AREA,0.3,0.3,0.3,0.3 HEIGH,0,0,0,0 !!Fire by Flux FIRE,1,2.5,2.5,0.5,1,FLUX,6000,Fire 2 Targ,0,0,Fire by Flux CHEMI,1,4,0,0,0,0.35,5E+07 TIME,0,1,999,1000 HRR,0,100000,100000,0 SOOT,0.05,0.05,0.05,0.05 CO,0.01954722,0.01954722,0.01954722,0.01954722 TRACE,0,0,0,0 AREA,0.3,0.3,0.3,0.3 HEIGH,0,0,0,0 !! !!Target and detector keywords !! TARGET,1,2.5,3.5,0.5,0,0,1,SOFTWOOD,EXPLICIT,PDE,0.5,Fire 2 Targ TARGET,1,2.5,1.5,0.5,0,0,1,SOFTWOOD,EXPLICIT,PDE,0.5,Fire 3 Targ

[rpeacoc]

Here's the log file from the test case in the previous comment, run with gnu fortran on OS X

CFAST

Version : CFAST 7.0.0 Revision : Gitv7-59-g5756bd6-dirty Revision Date : Fri Nov 6 11:03:52 2015 -0500 Compilation Date : Nov 06, 2015 11:04:36

Object #3 (Fire by Flux) ignited at 345.157 seconds Object #2 (Fire by Temp) ignited at 399.426 seconds Total execution time = 2.82 seconds Total time steps = 40028 Normal exit from CFAST

[andersonad]

Confirmed on all three OSes, the updated code works with the newly specified input file and produces the expected secondary ignitions.

I am closing this issue now.

[mcgratta]

Great, thanks for your help.

[rpeacoc]

We'll add a note to the release notes on the need to check igniting fires to make sure target information is correctly included.