Ceiling jet heat transfer removes more heat than physically possible

[mcgratta]

For the attached case, the HGL temperature drops below ambient in the first few minutes. The mechanical extraction rate is high, but I do not know why that would drop the temperature below ambient.

VERSN,7,CFAST Simulation
!!
!!Scenario Configuration
!!
TIMES,3600,50,10,10
EAMB,293.15,101325,0
TAMB,293.15,101325,0,0
!!
!!Material Properties
!!
MATL,FB XL,1.486015,1015.1,616.3487,0.453,0.9,Fußboden XL
MATL,D XL,1.651613,579.8,324.1935,0.031,0.9,Dach XL
MATL,AW XL,0.5841,510,95.604,0.091,0.9,Außenwand XL
!!
!!Compartments
!!
COMPA,Room,32,25,3,0,0,0,D XL,FB XL,AW XL,50,50,50
!!
!!Vents
!!
HVENT,1,2,1,7,1,0,6.5,,1,TIME,,,0,0,0,1,0,0
HVENT,1,2,2,7,1,0,6.5,,2,TIME,,,0,0,0,1,0,0
HVENT,1,2,3,7,1,0,6.5,,3,TIME,,,0,0,0,1,0,0
HVENT,1,2,4,7,1,0,6.5,,4,TIME,,,0,0,0,1,0,0
MVENT,1,2,1,H,3,1,H,3,1,8.333,200,300,TEMP,373.15,,,1,,1,16,12
!!
!!Fires
!!
!!New Fire
FIRE,1,16,12,0,1,TIME,0,0,0,0,New Fire
CHEMI,1,0,0,0,0,0.2,2.6E+07
TIME,0,30,60,90,120,150,180,210,240,270,300,330,360,390,420,450,480,510,540,570,600,630,660,690,720,750,780,810,840,870,900,930,960,990,1020,1050,1080,1110,1140,1170,1200,1230,1260,1290,1320,1350,1380,1410,1440,1470,1500,1530,1560,1590,1620,1650,1680,1710,1740,1770,1800,1830,1860,1890,1920,1950,1980,2010,2040,2070,2100,2130,2160,2190,2220,2250,2280,2310,2340,2370,2400,2430,2460,2490,2520,2550,2580,2610,2640,2670,2700,2730,2760,2790,2820,2850,2880,2910,2940
HRR,0,169200,676800,1522800,2707200,4230000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000
SOOT,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16
CO,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0
TRACE,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0
AREA,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20
HEIGH,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0

[preneke]

I have run this a reproduced the bug. I will let you know when I figure it out.

[rpeacoc]

This is related to the enhanced ceiling heat transfer due to the presence of a ceiling jet. If the ceiling jet heat transfer calculation is turned off (defaulting to the regular heat transfer), the problem goes away.

[rmcdermo]

Rick,
In FDS we have a check on the DT (optional) that keeps the heat transfer from overshooting. Here is a brief write up. In short, you can form a velocity out of the heat transfer coefficient, density, and specific heat. Then the DT needs to satisfy a CFL-type condition. Not sure exactly how to implement this in CFAST... it would depend on what you use for the length scale. But maybe it would help.

4.5.5 Heat Transfer Constraint
Note that the heat transfer coefficient, h, has units of W/(m2 K). Thus, a velocity scale may be formed from h/(ρ cp). Anytime we have a velocity scale to resolve, we have a CFL-type stability restriction. Therefore, the heat transfer stability check loops over all wall cells to ensure δt ≤ δxρ cp/h. This check is an option. It is not done by default.

[JaegerMJ]

I have revised the file, there is one compartment, the same as the fire compartment, above. In the results the temperature profile is the same, the ULT is falling.
VERSN,7,CFAST Simulation !! !!Scenario Configuration !! TIMES,3600,50,10,10 EAMB,293.15,101325,0 TAMB,293.15,101325,0,0 !! !!Material Properties !! MATL,FB XL,1.486015,1015.1,616.3487,0.453,0.9,Fußboden XL MATL,D XL,1.651613,579.8,324.1935,0.031,0.9,Dach XL MATL,AW XL,0.5841,510,95.604,0.091,0.9,Außenwand XL !! !!Compartments !! COMPA,Room,28.28,28.28,3,0,0,0,D XL,FB XL,AW XL,50,50,50 COMPA,Room 2 ,28.28,28.28,3,0,0,3,D XL,FB XL,AW XL,50,50,50 !! !!Vents !! HVENT,1,3,1,7,1,0,6.5,,1,TEMP,373.15,Temp,,0,,1,0,0 HVENT,1,3,2,7,1,0,10.64,,2,TEMP,373.15,Temp,,0,,1,0,0 HVENT,1,3,3,7,1,0,10.64,,3,TEMP,373.15,Temp,,0,,1,0,0 HVENT,1,3,4,7,1,0,10.64,,4,TEMP,373.15,Temp,,0,,1,0,0 MVENT,1,3,1,H,3,1,H,3,1,8.333,200,300,TEMP,373.15,Temp,,0,,1,14.14,14.14 !! !!Fires !! !!Fire FIRE,1,14.14,14.14,0,1,TIME,0,0,0,0,Fire CHEMI,1,0,0,0,0,0.2,2.6E+07 TIME,0,30,60,90,120,150,180,210,240,270,300,330,360,390,420,450,480,510,540,570,600,630,660,690,720,750,780,810,840,870,900,930,960,990,1020,1050,1080,1110,1140,1170,1200,1230,1260,1290,1320,1350,1380,1410,1440,1470,1500,1530,1560,1590,1620,1650,1680,1710,1740,1770,1800,1830,1860,1890,1920,1950,1980,2010,2040,2070,2100,2130,2160,2190,2220,2250,2280,2310,2340,2370,2400,2430,2460,2490,2520,2550,2580,2610,2640,2670,2700,2730,2760,2790,2820,2850,2880,2910,2940 HRR,0,169200,676800,1522800,2707200,4230000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000 SOOT,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16 CO,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875 TRACE,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0 AREA,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20 HEIGH,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0 !! !!Targets and detectors !! TARGET,1,10,10,3,0.6293043,0.6293043,-0.4560176,D XL,EXPLICIT,PDE,0.5,Temp
I'm sorry for the formatting.
@rpeacoc , how and where could i enter this attitude?

This is related to the enhanced ceiling heat transfer due to the presence of a ceiling jet. If the ceiling jet heat transfer calculation is turned off (defaulting to the regular heat transfer), the problem goes away.<

[JaegerMJ]

I've found a parameter in the code witch is called zin (position to calculate plume centerline temperature (m)) in the 'fire.f90' file in the part 'get_ceilingjet_tempandvelocity', but i didn't find where this parameter get a size. It can be possible that i had overlook it.

[rpeacoc]

Unfortunately, it is not a user-selectable setting. I had to change the code (commenting out line 68 of convection.f90) to eliminate the calculation. Initial testing seems like this part of the ceiling heat transfer calculation does not change the results of all of our V&V results. There's a bit more testing before we put out a patch release.

…

On Tue, Apr 18, 2017 at 3:30 AM, JaegerMJ ***@***.***> wrote: @rpeacoc <https://github.com/rpeacoc> , how and where could i enter this attitude? This is related to the enhanced ceiling heat transfer due to the presence of a ceiling jet. If the ceiling jet heat transfer calculation is turned off (defaulting to the regular heat transfer), the problem goes away.

[rpeacoc]

The parameter zin is an argument to the routine get-ceilingjet_temperatureandvelocity

[JaegerMJ]

I would want to ask how far you are, entire task is for my bachlor thesis. At the moment i write the document in latex and expand my VBA code, which edit the CFAST results. The thesis is not announced yet, up to june there is time avaible. I have to very big thanks at you. I work very much with pleasure at your program, so I would support you with this program.

[rpeacoc]

Are you running the stock Windows version or a compiled version on another computer platform?

[JaegerMJ]

We have a compiled version, because of the fire equation. The only code we have circumscribed is a the heskestd part in the fire.f90 module. I show you for this a german criteria in a guideline.
Also I gave you our compiled version, is it possible that anything is damaged as a result of the compile?
And here is the link of the compiler, which we have used.
https://software.intel.com/en-us/intel-parallel-studio-xe
`! --------------------------- heskestad -------------------------------------------

subroutine heskestad_plume (q_t, q_c, z, t_inf, emp, ems, eme, area, xfx, xfy)

!     purpose: calculates plume entrainment for a fire from heskestad's variant of zukoski's correlation
!     inputs:    q_t   fire size (w)
!                z     plume height (m)
!                t_inf ambient temperature at base of the fire
!                emp   mass loss rate of the fire (kg/s)
!                area  is the cross sectional area at the base of the fire
!                xfx   distance from fire to wall in x direction (m)
!                xfy   distance from fire to wall in y direction (m)
!     outputs:   ems   total mass transfer rate up to height z (kg/s)
!                eme   net entrainment rate up to height z (kg/s)

real(eb), intent(in) :: q_t, q_c, z, t_inf, emp, area, xfx, xfy
real(eb), intent(out) :: ems, eme

real(eb), parameter :: cpg = cp/1000._eb ! correlation uses different units
real(eb) :: d, qj, z0, z_l, deltaz, xf, factor, qstar, rho_inf
real(eb) :: c1, c2

! determine which entrainment factor to use by fire position.  if we're on the wall or in the corner, entrainment is modified.
! by reflection, entrainment on a wall is 1/2 the entrainment of a fire 2 times larger;
!                            in a corner, 1/4 the entrainment of a fire 4 times larger
xf = 1.0_eb
if (xfx<=mx_hsep.or.xfy<=mx_hsep) xf = 2.0_eb
if (xfx<=mx_hsep.and.xfy<=mx_hsep) xf = 4.0_eb

! qstar and virtual origin correlation are based on total HRR
qj = 0.001_eb*q_t*xf
if (z>0.0_eb.and.qj>0.0_eb) then
    d = sqrt(area*xf/pio4)
    rho_inf = 352.981915_eb/t_inf
    qstar = qj/(rho_inf*cpg*t_inf*gsqrt*d**(2.5_eb))
    z0 = d*(-1.02_eb + 1.4_eb*qstar**0.4_eb)

    ! entrainment is based on convective HRR and the mean flame height
    qj = 0.001_eb*q_c*xf
    z_l = max(0.0001_eb,d*(-1.02_eb + 3.7*qstar**0.4_eb))
    if (z>z_l) then
        factor = 1.0_eb
        deltaz = max(0.0001_eb, z-z0)
    else
        factor = z/z_l
        deltaz = max(0.0001_eb, z_l-z0)
    end if
    c1 = 0.196*(grav_con*rho_inf**2/(cpg*t_inf))**onethird  ! under normal conditions, 0.071_eb
    c2 = 2.9_eb/((gsqrt*cpg*rho_inf*t_inf)**twothirds)      ! under normal conditions, 0.026_eb
    eme = (c1*qj**onethird*deltaz**(5.0_eb/3.0_eb)*(1.0_eb+c2*qj**twothirds*deltaz**(-5.0_eb/3.0_eb)) * factor)/xf
    ems = emp + eme
else
    ems = emp
    eme = 0.0_eb
end if

end subroutine heskestad_plume`

The code is only a part of the whole fire.f90 module, i thougt ist was nearly row 511.

CFAST exe.zip

At last, the equation whith tat we work and the study about that.

[rpeacoc]

If you're compiling your own executable, the latest repo version works with the test case you submitted above.

[JaegerMJ]

The results of both versions is a falling ULT at the beginning. In my last post i gave you our compiled version, in an .zip file.
Here are results of on variante, in both versions
3600_XL_3_800_BMG3_30000_n_NIST.xlsx
3600_XL_3_800_BMG3_30000_n_VDI.xlsx
The difference between your version and ours is that the plume is differently calculates.
The other plume is the result of the code from the last post.

[rpeacoc]

If you have test cases which fail with the unmodified version of CFAST, post them and we'll be happy to look at them. I had to correct errors in you original post file just to allow it to run. With the errors corrected, the latest repo source code, unmodified, does not exhibit a falling layer temperature. We cannot diagnose user-modified versions of the model.

[JaegerMJ]

I think i can limit the bug.
If we change the air flow all variants (3600_XL_3_800_BMG3_var) have this falling ULT.
If i only change the high, as example the high is 4 m this error is away (3600_XL_4_800_BMG3_40000).
If i change the area on 1000 m², the simulation with high as 3 m the ULT is also falling (3600_XL_3_1000_BMG3_40000).
In the results of area at 1000 m² and high at 4 m, the bug is once again away (3600_XL_4_1000_BMG3_40000).
All cases where it fails:
3600_XL_3_800_BMG3_var.zip
The cases where only the high is changed at an area by 800 m²:
3600_XL_var_800_BMG3_40000.zip
The cases where only the high is changed at an area by 1000 m²:
3600_XL_var_1000_BMG3_40000.zip

[rpeacoc]

I'm a bit confused. I've run all the cases in your zip file 3600_XL_3_800_BMG3_var.zip which you say are the failing cases and none of them show a dropping layer temperature. Here's a graph of the results with the most recent source code:

[rpeacoc]

I did have to make corrections to most of the files to allow them to run. Only the 30000 file would run as supplied. All others exited with an error caused by an incomplete mechanical ventilation specification. If you want to open a vent at a selected temperature, you have to create a target at the appropriate location and associate that target with the vent. Your 30000 test cases had the vent open at a specified time so it worked as supplied.

[rpeacoc]

We've added a check to the input routine to ensure that the associated target is specified. I'd make sure you delete you existing output files before you run all the cases just in case you are seeing old output files when a cases fails before creating new output files.

[JaegerMJ]

VERSN,7,CFAST Simulation !! !!Scenario Configuration !! TIMES,3600,50,10,10 EAMB,293.15,101325,0 TAMB,293.15,101325,0,0 !! !!Material Properties !! MATL,FB XL,1.486015,1015.1,616.3487,0.453,0.9,Fußboden XL MATL,D XL,1.651613,579.8,324.1935,0.031,0.9,Dach XL MATL,AW XL,0.5841,510,95.604,0.091,0.9,Außenwand XL !! !!Compartments !! COMPA,Room,28.28,28.28,3,0,0,0,D XL,FB XL,AW XL,50,50,50 !! !!Vents !! HVENT,1,2,1,7,1,0,6.5,,1,TIME,,,0,0,0,1,0,0 HVENT,1,2,2,7,1,0,10.64,,2,TIME,,,0,0,0,1,0,0 HVENT,1,2,3,7,1,0,10.64,,3,TIME,,,0,0,0,1,0,0 HVENT,1,2,4,7,1,0,10.64,,4,TIME,,,0,0,0,1,0,0 MVENT,1,2,1,H,3,1,H,3,1,8.333,200,300,TEMP,373.15,Temp,,0,,1,14.14,14.14 !! !!Fires !! !!Fire FIRE,1,14.14,14.14,0,1,TIME,0,0,0,0,Fire CHEMI,1,0,0,0,0,0.2,2.6E+07 TIME,0,30,60,90,120,150,180,210,240,270,300,330,360,390,420,450,480,510,540,570,600,630,660,690,720,750,780,810,840,870,900,930,960,990,1020,1050,1080,1110,1140,1170,1200,1230,1260,1290,1320,1350,1380,1410,1440,1470,1500,1530,1560,1590,1620,1650,1680,1710,1740,1770,1800,1830,1860,1890,1920,1950,1980,2010,2040,2070,2100,2130,2160,2190,2220,2250,2280,2310,2340,2370,2400,2430,2460,2490,2520,2550,2580,2610,2640,2670,2700,2730,2760,2790,2820,2850,2880,2910,2940 HRR,0,169200,676800,1522800,2707200,4230000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000,6000000 SOOT,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16,0.16 CO,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875,0.06059875 TRACE,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0 AREA,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20 HEIGH,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0 !! !!Targets and detectors !! TARGET,1,10,10,3,0,0,1,D XL,EXPLICIT,PDE,0.5,Temp

This are my results of runs with the orginal and compiled version.
Can you please look over the code for mistakes from my part.

[rpeacoc]

I sorry, but I don't see a falling temperature with the latest test case you supplied. Here's the output from the latest repo version of CFAST with the normal Heskestad plume, repo version 0589e0a:

and here's the input file I created from your file, above since it copies without any line formatting (You'll need to change the extension from .txt to .in):

Test.txt

[rpeacoc]

Make sure you are using the latest source code. The significant change is to convection.f90. Lines 69-71 should look like

! limit the heat transfer area to the valid limit of the correlation (r/H<4)
lw_eff = min(pi*(4.0_eb*roomptr%cheight)**2,roomptr%cwidth*roomptr%cdepth)
qconv_avg = 0.27_eb*qconv/(lw_eff**0.68_eb*roomptr%cheight**0.64_eb)

There are other changes to define variables, add error checking, and make use of this new ceiling jet calculation, but this should indicate whether you've updated your repo to the latest revision.

[JaegerMJ]

No, the lines in the version i have are not the same, i download the new source code and the new version, i let you know about the next results.
The CEdit says version 7.2.0.
In release notes you write version 7.2.2, do you have a download?

[JaegerMJ]

I have to thank you very much. I downloded the new version and the new source code and compiled the version for my calculations.

[rpeacoc]

Problem still exists for very large rooms, particularly with very high ceilings. New test case attached.

testcase.txt

[rpeacoc]

Underlying correlation (eq. 5.24 in Tech Ref. assume ceiling is at ambient temperature, thus overestimating the convective flux when the ceiling is actually warmer (and thus closer to the HGL). Need to reformulate to properly calculate enhanced heat transfer coefficient. Information is still in SFPE Handbook.