0
5
10
15
20
25
30
35
40
0 30 60 90 120
150
180
210
240
270
300
330
360
390
Num
ber o
f Tes
ts
Peak HRR (kW)
139 tests < 10 kW
0123456789
10
400
550
700
850
1000
1150
1300
1450
1600
1750
1900
2050
2200
2350
2500
2650
Num
ber o
f Tes
ts
Peak HRR (kW)
0
5
10
15
20
25
30
35
40
0 10 20 30 40 50 60 70 80 90 100
Num
ber o
f Tes
ts
Total Energy Release (MJ)
166 tests < 5 MJ
0123456789
10
100
140
180
220
260
300
340
380
420
460
500
540
580
620
660
700
Num
ber o
f Tes
ts
Total Energy Release (MJ)
0
10
20
30
40
50
60
10 100 1000 10000
Heat
of C
ombu
stio
n (M
J/kg
)
Peak HRR (kW)
CellulosicMixedOtherThermoplastic 0
0.51
1.52
2.53
3.54
10 100 1000 10000
Q*
Peak HRR (kW)
CellulosicMixedOtherThermoplastic
0
0.05
0.1
0.15
0.2
10 100 1000 10000
CO Y
ield
(kg
CO/k
g fu
el)
Peak HRR (kW)
CellulosicMixedOtherThermoplastic
0
0.02
0.04
0.06
0.08
0.1
0.12
0.14
10 100 1000 10000
Soot
Yie
ld (k
g so
ot/k
g fu
el)
Peak HRR (kW)
CellulosicMixedOtherThermoplastic
0
5
10
15
0 500 1000 1500
HRR
(kW
)
Time (s)
Single Rag5 RagsBag of 10 Rags
0
10
20
30
40
50
0 200 400 600
HRR
(kW
)
Time (s)
Single Rag
5 Rags
0
5
10
15
20
0 1000 2000 3000 4000 5000
HRR
(kW
)
Time (s)
Repeat 1Repeat 2Repeat 3
010203040506070
0 500 1000 1500 2000 2500 3000 3500
HRR
(kW
)
Time (s)
Repeat 1Repeat 2Repeat 3
0
20
40
60
80
100
0 700 1400 2100 2800 3500
HRR
(kW
)
Time (s)
Draped 1
Draped 2
Folded
0
50
100
150
200
0 800 1600 2400 3200 4000
HRR
(kW
)
Time (s)
Draped 1Draped 2Folded
0
20
40
60
80
0 900 1800 2700 3600 4500
HRR
(kW
)
Time (s)
Repeat 1Repeat 2Repeat 3
0
5
10
15
20
25
30
0 900 1800 2700 3600 4500
HRR
(kW
)
Time (s)
Repeat 1Repeat 2Repeat 3
0
20
40
60
80
100
0 500 1000 1500
HRR
(kW
)
Time (s)
Quarter FullHalf FullFull
050
100150200250300350
0 500 1000 1500 2000 2500
HRR
(kW
)
Time (s)
Quarter FullHalf FullFull
0100200300400500600700
0 100 200 300
HRR
(kW
)
Time (s)
Emptyw/Peanutsw/Paper
0
50
100
150
0 100 200 300
HRR
(kW
)
Time (s)
Emptyw/Peanutsw/Paper
010203040506070
0 100 200 300
HRR
(kW
)
Time (s)
Emptyw/Peanutsw/Paper
THE MEASURED HEAT RELEASE RATES OF TRANSIENT FUEL ITEMS FOUND IN NUCLEAR POWER PLANTS
Jason Floyd and Matthew DiDomizio, Jensen Hughes, USAKevin McGrattan and Matthew Bundy, National Institute of Standards and Technology, USA
Marko Randelovic and Ashley Lindeman, Electric Power Research Institute, USAInterflam 2019 – July 2019
Hot Gas Layer (HGL)Model (typically CFAST) to determine damage
Fire Modeling for PRA*
in a Nutshell1
Cable Trays (Targets)
Zone of Influence (ZOI)Maximum distance where plume temperature or radiative flux can cause damage
In ZOI = damaged
Not in ZOI or HGL = undamaged
In HGL = ?
Transient Fires in PRA2 PRA Realism3A transient fire is a fire that results from combustible materials not fixed in place like a motor, transformer, or electrical cabinet. A transient fire could occur anywhere that there is a surface where transient materials could be placed.
The current recommended approach on treating transient fires is contained in NUREG/CR-6850 and NUREG/CR-6850 Supplement 1. The recommendation is that transient fires are represented by a gamma distribution with a 75th percentile HRR of 142 kW and a 98th percentile HRR of 317 kW with a growth rate of 2 minutes (unconfined) or 8 minutes (confined – e.g. trash can).
In a PRA, transient fires are typically the number two or three source of fire risk in a nuclear power plant.
At first fire PRAs were used for NFPA 805, to demonstrate that a plant met an overall limit on core damage frequency. It was recognized that conservatisms in the PRAs existed, but they were tolerated since plants were still able to meet risk targets.
*PRA = probabilistic risk assessment
Now PRAs are being used for risk informed maintenance activities. All plants have technical specifications which define minimum amounts of safety equipment that must be operational. If a plant drops below this it must go offline. For some equipment a brief time period for repair is allowed. Being offline can mean $1,000,000 a day or more in lost revenue. Technical specification times originated long before PRAs and were largely based on expert judgement. A plant can get a longer time for repair if the PRA shows the risk is small. PRA conservatism makes this more difficult. This has resulted in a number of research efforts to improve realism in fire PRAs.
The Realism Issue4 Project Outline5 Ignition and Fuels6Phase 1: Burn fuel packages representative of transient fire events in the Fire Events Database (FEDB).
• 290 tests of 99 fuel package were done using a 100 kW calorimeter at Jensen Hughes in Baltimore, MD USA (EPRI funded tests) and the 1 and 3 MW calorimeters at the National Institute of Standards and Technology in Gaithersburg, MD USA (US Nuclear Regulatory Commission [NRC] funded tests). Fires burned without intervention until there was no visible flame.
• Ignition sources were selected to be representative of the strength and duration of those in FEDB events.
Phase 2 (Future Work): Using the FEDB transient fire events, weight the test data and generate new gamma distributions for HRR and ZOI plus guidance on how to model the time-dependence of a transient fire.
• Butane lighter – fast igniting fuels• Nylon wick – 6 cm x 1 cm nylon rope with 5 mL of heptane• Continuous flame – ~100 W propane flame from 0.5 cm tubing• Radiant panel – 15 cm x 25 cm panel at 30+ kW/m2 representing a
halogen work light.
Cotton rags, paper, cardboard boxes, duct tape, tarps, debris piles, plastic buckets, vacuums, trash bags, mop+bucket, rope, plastic chain, power cords, safety cones and chain stanchions, plastic tubing, canvas tool bag with tools, wood scaffolding and pallets, laptop computer, tablet computer. Flammable liquids in plastic bottles, oil absorbing pads, tablet computer, temperature ventilation ducts
The 317 kW 98th percentile heat release rate is based on a set of 27 experiments from the literature at the time NUREG/CR-6850 was being written. Some of the experiments contain fuel items not typically found in risk significant areas of a plant. As a whole, the set of experiments appear to be more severe than operational experience (all plants report fire events to a centralized database maintained by EPRI).
A typical cable tray arrangement is for trays to run 0.3 to 0.6 m above electrical cabinets that are 1.8 to 2.1 m tall. This is within the ZOI of a 317 kW fire.
This suggests that not only does the guidance lack realism (not matching operational experience) but it is also likely resulting in significant conservatisms in fire PRAs. For each fuel type there were multiple configurations and repeat tests.
Peak HRR and Total MJ7 Fire Characteristics8 Cardboard Boxes9
Trash Cans10 Plastic Cart w/ Laptop11 Rope/Hose12
Safety Cone/Stanchion13 Plastic Tarps14 Cotton Rags15
Medium Empty Box
15.2 m of 2.5 cm Nylon Rope
15.2 m of 1.9 cm Vinyl Water Hose
Fire Resistant Tarp
Non-Fire Resistant Tarp
Dry Cotton Rags
Cotton Rags w/5 mL Heptane Per Rag
Stack of Four 23 cm Tall Plastic Safety Cone
102 cm Tall Plastic Chain Stanchion
0.11 m3 Metal Trash Can
0.13 m3 PE Trash Can
Cans have PE bag, plastic water bottles (empty), paper
Laptop, 3-ring Binder w/paper, and printer on two-shelf plastic work cart
Small (0.006 m3) Medium (0.03 m3)
Large (0.2 m3)
0
500
1000
1500
2000
2500
3000
0 2000 4000 6000
HRR
(kW
)
Time (s)
Repeat 1Repeat 2Repeat 3
This test represents the worst transient fire event in the FEDB
Test data will be published as a joint EPRI/NRC report (NUREG)
This project was sponsored by the Electric Power Research Institute
© 2019 Jensen Hughes, Inc. All rights reserved.
Tests ≤ 400 kW Tests > 400 kW
Tests ≤ 100 MJ Tests > 100 MJ