1 HEAT RELEASE RATE HEAT RELEASE RATE NATURAL VENTILATION NATURAL VENTILATION - - CONTROLLED CONTROLLED ROOM FIRES ROOM FIRES Dr Penh LAMUTH (1) and Dr Jean Pierre VANTE Dr Penh LAMUTH (1) and Dr Jean Pierre VANTE LON (2) LON (2) (1) Commissariat (1) Commissariat à à l l ’ ’ Energie Atomique Energie Atomique - - CEA/DPSN, CEA/DPSN, 16 16 - - 18, Route du Panorama, BP 06, 92265 18, Route du Panorama, BP 06, 92265 Fontenay Fontenay - - Aux Aux - - Roses Roses Cedex, Cedex, France France [email protected][email protected]33 (1) 46 54 94 40 33 (1) 46 54 94 40 (2) Laboratoire de Combustion et de D (2) Laboratoire de Combustion et de D é é tonique tonique – – CNRS/LCD CNRS/LCD 1, Avenue Cl 1, Avenue Cl é é ment Ader, BP 40109 ment Ader, BP 40109 – – 86961 Futuroscope 86961 Futuroscope Chasseneuil Chasseneuil Cedex, Cedex, France France GDR GDR Incendie Incendie – – CORIA UME 6614, July 2 CORIA UME 6614, July 2 - - 3, 2009 3, 2009
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
(2) Laboratoire de Combustion et de D(2) Laboratoire de Combustion et de Déétonique tonique –– CNRS/LCDCNRS/LCD
1, Avenue Cl1, Avenue Cléément Ader, BP 40109 ment Ader, BP 40109 –– 86961 Futuroscope 86961 Futuroscope ChasseneuilChasseneuil Cedex,Cedex,
FranceFrance
GDR GDR IncendieIncendie –– CORIA UME 6614, July 2CORIA UME 6614, July 2 --3, 20093, 2009
2
GDR GDR IncendieIncendie –– CORIA UME 6614, July 2CORIA UME 6614, July 2 --3, 20093, 2009
The study of thermodynamic aggressions of the equipment and structures of the closed rooms needs data concerning HRR
The fire evolution in a room can be affected by:
– Quantity and arrangement of fuel in the fire room
– Oxygen supply
If the ventilation is great enough (sufficiency of oxygen), the fire is said to be fuel-controlled
However, if the ventilation is small, relative to the size of the fire, there is not enough oxygen to combust all the pyrolysis fuel, the fire is said to be ventilation-controlled. The associated HRR depends mainly on the amount of available oxygen (ventilation condition).
3
GDR GDR IncendieIncendie –– CORIA UME 6614, July 2CORIA UME 6614, July 2 --3, 20093, 2009
The four different fire phases, characterized by a HRR, can be distinguished in this figure
IgnitionAnd
Growth phase
Steady-state phase
(full-developed fire)
Decay phase
T
t
Fig 1: Temperature evolution in a fire room – Fire phases
4
GDR GDR IncendieIncendie –– CORIA UME 6614, July 2CORIA UME 6614, July 2 --3, 20093, 2009
1. Ignition phase: fire starts with the ignition of burning material (usually in a single location of the room)
2. Growth phase: fire starts to propagate within the room. It is characterized by an exponentially increasing HRR that depends on:
• the type and geometry of fuel,
• interaction with the surrounding,
• access to oxygen.
The development can evolve towards the maximum of HRR
5
GDR GDR IncendieIncendie –– CORIA UME 6614, July 2CORIA UME 6614, July 2 --3, 20093, 2009
3. Fully developed phase: the HRR is relatively unchanging and leading to small variations in temperature.
The situation can evolve toward one of the following situation:
• Fuel-controlled situation:
� Spread of the fire to the whole room (flashover phenomenon): the gas temperature, became so elevated, can cause the sudden ignition of “every object” and unburnt gas in the room
� No spread of the fire to the whole room: if the propagation is slow, the gas temperature rise is not sufficient to cause flashover, fire can find no combustible material in its closed vicinity. The fire remains localized and, with time, dies out.
The HRR is related to the pyrolysis rate by equation:
GDR GDR IncendieIncendie –– CORIA UME 6614, July 2CORIA UME 6614, July 2 --3, 20093, 2009
2O,RH∆airdryair MM =If we take = 13,100 kJ/kg, = 29 kg/kmol,
2OM =ρ0= 32 kg/kmol and
the time to reach the HRR peak becomes
( )3/1
peak 3,0LOI.2,3lnV.k
.37t
+α
−=Eq. (6)
This time remains “independent” of LOI under 4%. Therefore, if we take LOI = 0%, Eq. (6) becomes
3/1
peak
V.k4.39t
α= Eq. (7)
1.18 kg/ m3,
18
B. PostB. Post--flashover model (steady phase)flashover model (steady phase)
After the peak, the HRR decreases. It is assumed to tend towards an asymptotic direction corresponding to a fire which is controlled by natural ventilation.
By taking Eq. (2) becomes0dt
dX2O =
( )22 O,R
0OLOIimax H.XZ.mQ ∆= && Eq. (8)
where
21LOI21
23
LOI.2932
23
X
XXZ
0O
0,O0O
LOI
2
22−≅
−=
−= Eq. (9)
GDR GDR IncendieIncendie –– CORIA UME 6614, July 2CORIA UME 6614, July 2 --3, 20093, 2009
19
The expression of the pressure differences, ∆Pe in vents at lower level and ∆Pi upper level, using the Bernoulli equation, can be expressed as
gHTT
1PP 00ie
−ρ+∆−=∆ Eq. (10)
where H is the difference in height between the air inlet and outlet (m)
We neglected the combustion mass inside the cabinet to obtain a simple mass balance relating the incoming and exhausting mass flows in a steady phase.
11
gH2CAmm 0eeei +µτ−τρ== &&
Eq. (11)
where293T
TT
0
==τ
2
ee
ii
CACkA
=µ
: area of the room air inlet (m2),: area of the room air outlet (m2): vent inlet coefficient determined by specific test (flow resistant) (no unit): vent exhaust coefficient determined by specific test (no unit)
iA
eA
iC
eC
GDR GDR IncendieIncendie –– CORIA UME 6614, July 2CORIA UME 6614, July 2 --3, 20093, 2009
20
The dimensionless parameter τ increases from T / 293 to reach the maximum value
2
ii
eemax CkA
CA11
++=τ Eq. (12)
The asymptotic HRR becomes
( )
µ++
ρ∆= ∞1
11
1.C.A.H.g.2..H.Y.ZQ ee0O,R,OLOImax 22
&
Eq. (13)
If we take g = 9.81 m/s2 ; kg/kJ000,313100x23.0H.X22 O,R
0O ≅=∆ and
18.10 =ρ kg/m3, Eq. (13) becomes simply
Eq. (14)
GDR GDR IncendieIncendie –– CORIA UME 6614, July 2CORIA UME 6614, July 2 --3, 20093, 2009
( )2
ii
ee
eemax
CkACA
11
HCA.LOI21.747Q
++
−=&
21
GDR GDR IncendieIncendie –– CORIA UME 6614, July 2CORIA UME 6614, July 2 --3, 20093, 2009
The following expression can be used for most of fuels (polymers, etc.)
15CkACA
11.315T01.0LOI2
ii
ee +
++−=+×−=
Figure 3: Laboratoire National d’Essais (LNE) test results: Evolution of LOI as a function of temperature (used by permission)
22
RemarksRemarks
The flow coefficients and can be determined by means of specific tests.As most materials found in nuclear facilities are “polymers”, the LOI can be expressed as a function of temperature
eCiC
Eq. (15)
Finally, we obtain the asymptotic HRR to represent the post-flashover phase
2
ii
ee
ee
2
ii
eemax
CkACA
11
HCA.
CkACA
13.2240Q
++
++=& Eq. (16)
GDR GDR IncendieIncendie –– CORIA UME 6614, July 2CORIA UME 6614, July 2 --3, 20093, 2009
15T01.0LOI +−=
15CkACA
11.3LOI2
ii
eeimummin +
++−=
23
GDR GDR IncendieIncendie –– CORIA UME 6614, July 2CORIA UME 6614, July 2 --3, 20093, 2009
VERIFICATION & VALIDATION OF THE PROPOSED MODELVERIFICATION & VALIDATION OF THE PROPOSED MODEL
Comparison between tests and calculationsComparison between tests and calculations
ObjectiveObjective: : blind calculation of time to reach the peak and asymptotic HRRblind calculation of time to reach the peak and asymptotic HRR
and comparison with test resultsand comparison with test results
Tests selectedTests selected: : 10 IRSN tests in a closed steel box 1m(wide) X 0.6m (deep) X 2m 10 IRSN tests in a closed steel box 1m(wide) X 0.6m (deep) X 2m (height), (height), naturally vented, are usednaturally vented, are used
Initial objective of testsInitial objective of tests: : evolution of mass moss rate considering influence ofevolution of mass moss rate considering influence of
Table 2: Standard growth parameter, entrainment coefficient and difference in height between
the air inlet and outlet
No experimental data is used .The zero of the calculation time axis is set at the end of the incubation phase (ignition reference time ignored).
26
GDR GDR IncendieIncendie –– CORIA UME 6614, July 2CORIA UME 6614, July 2 --3, 20093, 2009
Formula and data used for the comparison LOI = 8% and 1CC ie == (default values) are used to calculate
• 3/1
peak
V.kx4.39t
α=
• ( )2peakpeak tQ α=&
• 2
i
e
emax
kAA
11
HA.9700Q
++
=&
The results of the comparison can be presented as follows
27
GDR GDR IncendieIncendie –– CORIA UME 6614, July 2CORIA UME 6614, July 2 --3, 20093, 2009
0,00
200,00
400,00
600,00
800,00
1000,00
1200,00
1400,00
1600,00
1800,00
0 200 400 600 800 1000
test1
calculation 1
calculation 2
Test 1
Influence of fire growth parameter
Data: α = 0.1876 (ultra-fast) and α = 0.0469 (fast) : k = 2 and H = 1.90 m: (top outlet vent)
Remarks
• The early extinction has been observed after reaching the maximum HRR
• The influence of fire growth coefficient appears:
28
GDR GDR IncendieIncendie –– CORIA UME 6614, July 2CORIA UME 6614, July 2 --3, 20093, 2009
0
200
400
600
800
1000
1200
0 200 400 600 800 1000
calculation
test2
Test 2Remarks
The peak of HRR is correctly estimated
The theoretical result is not conservative during the fire extinction phase for two main reasons:
• influence of fire growth factor
• “2nd steady state” (?) and/or decay phase ignored in the model (junction between peak and steady state)
Peak and steady state
Data: α = 0.0469 (fast) : k = 2 and H = 1.90 m: (top outlet vent )
29
GDR GDR IncendieIncendie –– CORIA UME 6614, July 2CORIA UME 6614, July 2 --3, 20093, 2009
-100
0
100
200
300
400
500
600
700
0 200 400 600 800 1000 1200 1400
calculation
test 3
Test 3
Maximum HRR
Data: α = 0.01172 (medium) : k = 2 and H = 1.90 m: (top outlet vent )
Remarks
• Peak of HRR is correctly estimated• Steady state HRR is overestimated: after reaching the peak, the fire
doesn’t become necessary controlled by the ventilation
30
GDR GDR IncendieIncendie –– CORIA UME 6614, July 2CORIA UME 6614, July 2 --3, 20093, 2009
0
50
100
150
200
250
300
350
0 200 400 600 800 1000 1200 1400
calculation
test4
Test 4
Test 4
Steady state phaseData: α = 0.01172 (medium); k = 2 and H = 1.90 m (top outlet vent)
Remarks
• Fire controlled by fuel (peak is not reached)• Steady state HRR is overestimated (fire controlled by fuel and not by ventilation)
31
GDR GDR IncendieIncendie –– CORIA UME 6614, July 2CORIA UME 6614, July 2 --3, 20093, 2009
-100
0
100
200
300
400
500
600
700
0 200 400 600 800 1000 1200 1400
calculation 1
test8
calculation 2
test5
Tests 5 and 8Remarks
• Experiment 5 (top vent area) : early extinction has been observed (plotted as an indication).
• Experiment 8 (wall vent area) : no peak has been observed
• Maximum of HRR : correctly estimated
Influence of outlet vent location
Data test 8: α = 0.01172 (medium) and 0.00293 (slow); k = 1 and H = 1.80 m (wall outlet vent)
32
GDR GDR IncendieIncendie –– CORIA UME 6614, July 2CORIA UME 6614, July 2 --3, 20093, 2009
Influence of nature of fuels and configuration
Data: α = 0.01172 (medium); k = 2 and H = 1.80 m (wall outlet vent)
0,00
50,00
100,00
150,00
200,00
250,00
300,00
350,00
400,00
450,00
0 200 400 600 800 1000 1200 1400
test7
calculation
test9
test10
Tests 7, 9 and 10
Remarks• Great influence of nature of fuels and configurations on combustion
phenomena • HRR during the stationary phase is better predicted than that of the growth
phase
33
GDR GDR IncendieIncendie –– CORIA UME 6614, July 2CORIA UME 6614, July 2 --3, 20093, 2009
CONCLUSIONCONCLUSION
A comparison of the IRSN test with the calculated data gives encouraging results.
It should be noted that It should be noted that calculations results are strongly conditioned calculations results are strongly conditioned byby the fire fire growth parametergrowth parameter (function of all parameters such as vent flow areas, nature and(function of all parameters such as vent flow areas, nature andconfiguration of fuels)configuration of fuels)
It will be necessary to complete the comparison in order to determine the weaknesses or strengths of the formula, especially in rooms of large volume.
The following formula can be recommended to be used for engineering computation
.
2
i
e
emax
3/1
peak
A.2A
11
HA.20000Q;
V.50t
++
=
α= &
34
Thank you for your attention
GDR GDR IncendieIncendie –– CORIA UME 6614, July 2CORIA UME 6614, July 2 --3, 20093, 2009