ML19209B743
| ML19209B743 | |
| Person / Time | |
|---|---|
| Site: | Yankee Rowe |
| Issue date: | 10/03/1979 |
| From: | Fung F, Martin T NUTECH ENGINEERS, INC. |
| To: | |
| Shared Package | |
| ML19209B738 | List: |
| References | |
| SSP-01-001, SSP-1-1, NUDOCS 7910100382 | |
| Download: ML19209B743 (87) | |
Text
SSP-01-001 I
I I
SMOKE SIMULATION PROTOTYPE TEST CONDUCTED AT YANKEE ROWE NUCLEAR PLANT Prepared for:
BOSTON EDISON COMPANY COMMONWEALTH EDISON COMPANY FLORIDA POWER AND LIGHT COMPANY NORTHEAST UTILITIES COMPANY YANKEE ATOMIC ELECTRIC COMPANY I
Prepared by and j
g NUTECH Approval AAAA N
i Tim D.
- Martin, P.E.
LL tt(1 Reviewed by Francis C.W.
Fung, PhlD.
/
/
October 3, 1979 1138 154 7910 03
SSP-01-001 I
EXECUTIVE
SUMMARY
Seven Smoke Simulation tests were conducted at Yankee Rowe Nuclear Plant on August 23 and 24, 1979.
Each test simulated an incipient fire in the switchgear room and was a demonstra-tion of NUTECH's method to monitor simulated smoke movement to aid in locating smoke detectors.
Results of the testing demonstrate that the testing is a valid method to evaluate smoke detection systems and to select eptimum fire detector locaticn s.
The test was sponsored by and conducted for five sponsoring utilities.
The intent of the prototype testing was to demonstrate to the utilities and the Nuclear Regulatory Commission (NRC) that an acceptable method was available to satisfy requirements for "...a method for conducting insitu tests with a suitable smoke generation device to verify that a fire would be promptly detected by installed smoke detectors."1 NUTECH's Smoke Simulation Testing uses portable space heaters to simulate an incipient fire.
The heaters create a plume of 1.
Fire Protection Action Plan, NUREG 0298, Vol.
1, No.
4, page I-38, December 4, 1978.
1138 155 i
nutech
SSP-01-001 I
heated air, or simulated smoke, that is similar to the heated air created by a fire as it is just starting.
Both the simulated fire (hot air) and smoke from an incipient fire will be distributed within the room primarily by the ventilation system because neither have enough energy to disturb the ventilation flow pattern.
To monitor movement of the simulated smoke, a very small amount of tracer gas (SF ) is continuously released into the plume of hot air from 6
the heaters and serves as a tracer of the smoke movement.
The I
SF is detectable in extremely low concentrations such as one 6
part per billion (ppb) and provides a means of mapping the concentration profile of the simulated smoke in each area of the test room as the simulated fire progresses.
Thus, with the knowledge of how the simulated smoke is distributed, the optimum location for smoke detectors can be selected.
This report presents the results of the testing that demonstrates the validity of the technique and provides a basis for NRC concurrence that this testing satisfies NRC requirements.
The NUTECH prototype Smoke Simulation Testing identified optimum locations for smoke detectors in the switchgear room at Yankee Rowe.
These locations were selected based on the generally counter-clockwise movement of simulated smoke in I
156 ii nutech I
\\
SSP-01-001 room.
Additionally, the testing was shown to be a workable method for evaluating smoke detection systems in other plant locations and in other plants.
I I
I a
P I
I U" #
I nutech I
SSP-01.001 I
TABLE OF CONTENTS Page EXECUTIVE
SUMMARY
i LIST OF FIGURES v
LIST OF APPENDICES vi
1.0 INTRODUCTION
1.1 1.1 History 1.1 1.2 NUTECH Testing 1.2 1.3 Prototype Test 1.2 1.4 Report 1.3 2.0 TEST PROCEDURES 2.1 2.1 Pre-Test Activities 2.1 2.2 Testing 2.4 3.0 RESULTS AND CONCLUSIONS 3.1 3.1 Test Results 3.1 3.2 Conclusions 3.6 4.O RECOMMENDATIONS 4.1 4.1 Smoke Detector Locations 4.1 4.2 Continued Testing 4.1 4.3 Test Improvements 4.2 5.0 ANALYSIS 5.1 5.1 Data Analysis During Testing 5.1 5.2 Post-Test Data Analysis 5.3 5.3 Individual Test Analysis 5.4 I
1138 158 I
iv nutech g
SSP-01.001 LIST OF FIGURES Figure Title Pagt 1.1 Switchgear Room, Yankee Rowe 1.5 2.1 Grid Areas in Switchgear Room 2.9 2.2 Fire Simulation Locations 2.10 2.3 Sampling Locations 2.11 4.1 Optimum Fire Detector Locations -
4.4 Descriptions 4.2 Optimum Fire Detector Locations -
4.5 Grid 5.1 Smoke Simulation Testing Summary 5.5 5.2 Simulated Smoke Profile - Test 1 5.8 5.3 Test 1 SF Plots 5.9 6
5.4 Simulated Smoke Profile - Test 2 5.12 5.5 Test 2 SF Plots 5.13 6
5.6 Simulated Smoke Profile - Test 3 5.17 5.7 Test 3 SF Plots 5.l'8 6
5.8 Simulated Smoke Profile - Test 4 5.21 5.9 Test 4 SF Plots 5.22 5.10 Simulated Smoke Profile - Test 5 5.25 5.11 Test 5 SF os 5.26 6
5.12 Simulated Smoke Profile - Test 6 5.29 5.13 Test 6 SF 6
I 1138 159 I
V nutech I
SSP-01.001 I
LIST OF APPENDICES I
Appendix I
A Pre-Test Meeting Attendees B
Switchgear Room - Description C
Smoke Simulation Principles D
SF Concentrations - Normalized 6
I I
I I
I I
I 38 160 vi nutech g
SSP-01-001 l.0
];'1TRODUCTION l.1 History Following the fire at the Browns Ferry Nuclear Station on March 27, 1975, the NRC initiated a program to investigate the causes of the fire.
This investigation resulted in activities by each licensee to upgrade the fire protection program at their nuclear plants.
One phase of this review concentrated on smoke detectors and the smoke detection system at each plant.
Specif-ically, the NRC questioned how the smoke detectors were positioned to provide early warning.
To answer this question the NRC required that licensees conduct insitu tests to verify performance of their smoke detection system.
This test would demonstrate that a fire would be promptly detected by the installed smoke detection system in each critical space of each nuclear plant.
In 1978, when the Safety Evaluation Reports were issued to plants, there was no acceptable testing methodology to conduct an insitu test to satisfy this requirement.
A test using a real fire within the critical spaces of a plant was not allowed, and other suggestions, such as smoke bombs or ventilation mapping, were inadequate or had disadvantages that prohibited their use.
1.1 1
1138 161 nutech
SSP-Cl-001 The utilities approached the NRC in July of 1978 and explained that they were willing to comply with this requirement, but an acceptable method was not currently available.
1.2 NUTECH Testing In January 1979, NUTECH proposed a testing methodology to The satisfy the smoke detection system testing requirement.
proposed testing would verify the adequacy of the smoke detec-tion system in two phases: Phase 1 would be a smoke simulation test to determine the optimum location of smoke detectors; Phase 2 would be in-place testing of each smoke detector head with a portable smoke generating device to verify the alarm set point.
This proposed testing was presented to the utilities and the NRC.
Five utilities contracted NUTECH to conduct a prototype or demonstration test to answer NRC detailed questions and to provide a basis for evaluation of the testing methodology proposed for Phase 1.
Evaluation of the portable smoke detector testing device for Phase 2 was not included in the prototype testing.
1.3 Prototype Test The smoke simulation prototype test was conducted on August 23 and 24, 1979 at Yankee Rowe.
A series of seven fire simulation tests were conducted by NUTECH in the switchgear room, shown in Figure 1.1 - Switchgear Room.
The purpose of the. testing was:
1.2 1138 162 nutech
SSP-01-001 (1) to demonstrate the validity of the testing methodology to satisfy the NRC requirements, and (2) to identify the optimum smoke detector locations for the switchgear room at Yankee Rowe.
l.4 Report The following sections of this report describe the Smoke Simulation-Prototype Test conducted at Yankee Rowe in August 1979:
2.0 Test Procedures This section describes test preparations, rest execution, and the data collection.
3.0 Test Results and Conclusions Overall results of the testing are presented as the basis of the conclusions and recommendations.
Conclusions about the testing are presented, both for individual tests and the overall methodology.
4.0 Recommendations This section contains recommendations for optimum smoke detector locations in the switchgear room at Yankee Rowe Nuclear Plant, and recommendations for continued testing in g
other locations and plants.
In addition, recommendations 5
and improvements for subsequent Smoke Simulation Tests are presented.
I 1.3 1138 103 I
nutech g
SSP-01-001 5.0 Test Data Analysis This section describes analysis of the data during the test for first-look confirmation and data analysis upon completion of testing.
Detailed results and analysis of each of the seven tests are provided.
I I
I 1138 164 I
1.4 nutech g
mm?
,o$
wN ud u@
l R.
R.A A
E P g Z
G U
r.
H u
C Tl T
I v
W 8
5 4 S
Tt.
E V
V 0
0 G
0 0
N 4
4 2
2 I v W
Tm WC E T
OmE i
lU L
[
[
O us J
N w
T S
C M
N L
R I
N E
A NT O
L
\\
1 P
O r
A R
R
/
A
]
E
[
L T
L I
O R
R V
R C
1 A
TR V L p
0 U
"O NE r
OT 0G 6
A N
1 CN 8H 4
E e
C
'5 RE 4T 0
E r
OC W
0 G
W u
I I
T 4
I O
S g
H O
2 M
7 i
s
~
R r
C F
[
Tt.
R T
E E
R g
I E
N E
K g
W
-T T L y
N R E g
N E
E N w
YR S
YA V
M.
N%
S NA l E I F PL
- h. Tau Rz s I
R uN E
T QA O
dP G
Ew o N
N D
R Sc o Y
A R
I O
R E
T YG g
N RH E
H E
v T
G EC z
Tw TT 1
TI A
e I
' L
%W v
t
- G tS Tx GH e
H 2
Ti.
lG l
A.
TR HA o
L 6H N
A#
LC C E EE NR I
G RRI yARA M._
GZ
[N O
D V.
EL p
r E
U a
wD
._ o u s'
i C
O 1,$
EQGQy
~'m i
ei
SSP-01-001 2.0 TEST PROCEDURES 2.1 Pre-Test Activities I
2.1.1 Pre-Test Conference The day before the testing began, a conference was held with NUTECH engineers, engineers from Yankee Atomic Electric Company, members of the Yankee Rowe staff, representatives from partici-pating utilities, NRC representatives, and NRC consultants.
The attendance list is found ir, Appendix A - Pre-Test Meeting Attendees.
The purpose of the nteeting was to review the testing methodology, explain the mechanics of the test, and jointly inspect the switchgear room where the prototype test was to be conducted.
These same people witnessed tests 1, 2, and 3 conducted on August 23, 1979.
I 2.1.2 Simulat.ed Fire Locations To facilitate identification of portions of the switchgear room, a grid designation was made as shown in Figure 2.1 -
Grid Areas in Switchgear Room.
This grid was used in the selection of fire simulation locations and sample locations.
Grid areas were selected to conform to the major components in the room and the general arrangement.
Some grid lines were selected to eliminate areas such as the stairwells in C-1 and A-7.
The vertical grid line betwecn areas 4 and 5 I
2.1
}]'O
'f nutech I
m SSP-01-001 was positioned to evenly divide the 480 volt switchgear and the catwalk across the top of the switchgear.
This allowed coverage of 2 grid areas from the catwalk.
A total of 18 grid areas were designated, each covering an area of approximately 200 squa're feet.
The switchgear room is approximately 115 feet long, 38 feet wide, and 15 feet high.
It has a very complex ventilation flow pattern and an air flow rate of approximately 30,000 SFM passing from the fan room into the switchgear room and exhausting into the turbine room.
A more detailed description of the switchgear room is found in Appendix B - Switchgear Room Description.
The room contains several major components, many cable trays and ducts.
This combination of complicated air flow and many components provided an excellent proving ground for the prototype test becaus.: proving the test methodo)agy in this complex room would create greater confidence that the testing would work in othet spaces.
After inspection of the switchgear room seven fire simulation locations were selected.
These are shown in Figure 2.2 - Fire Simulation Locations.
These locations were selected so that the smoke distribution pattern could be identified for a fire in any location of the switchgear room.
2.2 lbL
}b7 nutech
SSP-01-001 A fire was possible from almost any location in the switchgear room because major components were located uniformly throughout the room.
The fire simulation locations were selected to allow identification of the smoke movement from a fire in each corner (tests 3, 4,
5, and 7) and from a fire in the center of the room (tests 1 and 6).
All the smoke simulation locations were near potential fire components and at the same time provided adequate information so that tha smoke patterns from a fire in any location could be determined.
Two fire simulation locations were selected in areas of low flow or unique geometry: at the workbench in area B-1, test 4, and in the battery room, test 7.
Smoke from a fire in these areas was not expected to enter the main part of the switchgear room and would not be distributed in a similar manne as smoke from other fire simulation locations.
2.1.3 Sampling Location Selection Proper identification of the' smoke distribution pattern requires adequate sampling within the test space.
A total of 27 samplc points were selected in the switchgear room following a detailed examination of the arrangement of beams, cable trays, ducts and major components.
These are shown in Figure 2.3 and described in Figure 2.4.
I 2.3 1138 16:
nutech g
SSP-01-001 Of the 27 sample points, all but 4 were selected high in the room, or about 6" from the ceiling.
Points were selected inside areas bounded by beams, areas where pockets were formed by cable trays or duct work, and areas such that at least one sample point was located in each grid area.
Of the four sample points that were not taken next to the ceiling, two were taken at a height of 5 feet from the floor.
These sample points (B-21 and B-32) were taken to determine the amount of vertical distribution that would occur.
The other two sample points away from the ceiling were taken in pockets formed by cable trays and ducting.
These points were A-62 and C-62, both taken at a height of about 8 feet from the floor.
2.2 Testing 2.2.1 Test Equipment Equipment used during the testing was arranged in the switchgear room and consisted of:
A.
Sampling syringes - approximately 200 - 20cc syringes were each labeled with the proper sample location and placed next to each sample point.
f.4 1138 i6?
nutech
_ SSP-01-001 d.
SF
- a container of SF with a regulator and high 6
6 accur:.cy flow meter were assembled; this included tubing to relems.e the SF in the vicinity of the heaters.
6 SF was selected as a tracer gas because it is stable 6
and non-reactive to 800 C and is detectable in extremely low concentrations such as 0.01 ppb.
A trace amount of SF6 released into the heated air stream has no effect on the p'lysical characteristics of the hot air plume but allows determination of how the air /SF mixture is distributed.
6 Because of the case of detection a very low flow of 1-2cc/
minute was used during each test resulting in a total of approximately 60cc of SF at STP being released during the 6
test.
C.
Heaters - 2 portable space heaters were used to generate the hot air plume of the simulated incipient fire.
Each heater was rated at 12.5 amps, for a total of 25 amps, which generated 3,000 watts or 10,250 BTU /hr.
The heaters were selected to emit energy at a rate equal to that given off by an incipient fire, or fire just as it is starting.
For comparison of fire size, a 1-pound block of wood that burns at, slow rate so that it is consumed after an hour will release about 9,000 BTU's in that hour.
Burning one quart of flammable liquid over a one-hour period will release about 6,000 BTU's in that hour.
The heaters were selected large enough to generate a plume of hot air with 2.5 1138 i70 nutech g
SSP-01-001 sufficient bouyancy to raise tne simulated smoke (hot air with SF6) off the floor so that it could be distributed by the ventilation system.
Another criteria for heater size selection was that the energy release not disturb the ventilation flow pattern in the test room.
In the switchgear room the dominant air flow was from the main air outlet in grid area B-3.
The plume from the heaters did not appear to affect the normal ventilation flow patterns.
A more detailed discussion of smoke simulation is found in Appendix C - Smoke Simulation Principles.
D.
SF Counting Equipment - the SF detector and 6
6 associated equipment was located in an adjacent room to facilitate counting during the testing and to avoid contamination of the test room by the exhaust from the SF detector as samples were 6
counted.
The SF detector used in the prototype test has a 6
range of 0.01 ppb to 50 ppb SF6*
2.2.2 Test Procedures The execution of the test was very simple after all the equipment and sampling personnel were in place.
The space heaters were energized and then the flow of SF w s started.
6 The flow meter used to control the release of SF was precise 6
enough to allow flow as low as 1-2cc per minute of SF 6
released into the heated area between the heaters.
2.6 1138 17i nutech
SSP-01-001 At pre-determined times after commencement of the test, a signal was given.and individuals took 20cc samples at each designated location with the 20cc syringes.
These grab samples were all taken within one minute of the signal and were then set aside and measured for SF content upon completion of each 6
test.
Upon completion of the testing and after all samples had been taken, the SF fl w was stopped and the heaters were de-energized.
6 Thus, the flow of SF was continued for about 3 minutes after 6
the signal was given for the last sample of each test.
2.2.3 Data Collection During the testing some of the collected samples were measured to determine if the release rate of SF was adequate within the 6
room.
The majority of the 20cc syringe samples were simply left at the sampling location until completion of the test.
They were then brought to the SF detector to measure the SF 6
6 content within each sample..
For the prototype test there were 27 sample points and as many as 7 samples taken at each sample point.
This resulted in 189 samples that required counting with each sample requiring 45 seconds to be analyzed in the SF detector.
6 2.7 1138 172 nutech
SSP-01-001 During the approximately 2 hours2.314815e-5 days <br />5.555556e-4 hours <br />3.306878e-6 weeks <br />7.61e-7 months <br /> required to count the samples, some initial data evaluation was conducted.
I 1138 173 2.8 g
nutech
M M
M M
M M
M M
M t i s' - o" 1
2 3
4 5
6 7
LK,HTING TRANSFCRMER EUMAT s2 UP-gg A
ELEV.
STORME MOTOR GJNTRot.
F% net.
N I
64 5000 EQUIPMENT CEM ER r
PANEL
[j
,)
q
_- w
~%
Ll6HTING # 15ATTERY INVERTER ausmx w a.
s4L E
'~ v '~-
B
~
B
-e~
G ANALI R BATTERY S WITCHGEAR
- guyggygg, e
k E C
.k_
y m
m u
XEWTTERY j
- e up ow N
7
/
[
2400 y swnCHGEAR.
7 -s 7
C C
\\
em -
No.2.
NO. I 24CO 480V TRANSFORMERS 1
2 3
4 5
.6 s 7
E!
SWITCHGEAR ROOM R'l 3
YANKEE ROWE NUCLEAR PLANT N
g Figure 2.1 Grid Areas in Switchgear Room O7
ama oms sum num ums 11 5' - O "
1 2
3 4
5 6
7 I
I I
;5#payDE Jg7g"r ac,Hris e m ~ -sen ug.
A N
A sto=
ce'*'a MH tH F-noc,4 mo minuEur
--]
r PANEL g
,___m LIGHT 1HG # E5ATTERY INVERTE1 QMRGING w PANEL sh 1
E
' " '""=
B B
-~
=
HGEARj a
-, =e e um a i
S WITC INVERTU'.
1 v%A L
.A UP DN
{
b8ETTEW j
N N
7 p
/
[
24o0 y swnCHGEAR.
C l
eme- -s C
3 t
- 80. 2.
so.i zum ev TRAssomens F" 1 i
I 1
1 1
2 3
4 5
.6 7
hh 1) i 1
t.2 i
cra SWITCHGEAR ROOM g
- 3 ~~^
YANKEE ROWE NUCLEAR PLANT hN O
g Figure 2.2 - Fire Simulation Locations E
o
?
- r
M M
i s s' - o" 1
2 3
4 5
6 7
~
c a tA Au UP-
"f-
+-
[g LK,HTIN 1E
-5 ON mm.
[,-
~,=]
b
=_
ROO EQUIPMENT M ANEL__
O O'
$c, 33 cggmKes N 2m--
B
~
s
_ ~ $ Q BATTE HGEAR]
d RY ANALI 2'eR
$[
S WITC
- inyggygg, e
h m
A T
s
,s UP DN l
BATTtRY A
&- U V"
[
- ~ ~ " ~ ~
C C
r 1
/
g@
g4 *gy "oj "o '
l 1
2 3
4 5
6 7{fghfbk SWITCHGEAR P. O O M s u 5 cc E E YANKEE ROWE NUCLEAR PLANT-h _N H Figure 2.3 - Sampling Locations g 7
SSP-01-001 I 3.0 RESULTS AND CONCLUSIONS 3.1 Test Results This section discussec over-all results of the NUTECH Smoke Simulation Testing. Conclusions from these results are found in Section 3.2. A more detailed review of each individual test is found in Section 5.3. 3.1.1 Smoke Distribution Pattern The NUTECH testing clearly identified the movement of smoke in the switchgear room to be in a counter-clockwise direction. This is best demonstrated by looking at the simulated smoke profiles for tests 3 and 5, Figures 5.6 and 5.10. In these " cloud" diagrams, the simulated smoke can be clearly seen starting from the fire locations and propagating in a counter-clockwise direction prior to accumulating in the lefthand part of the room. Both tests were started in the righthand side of the switchgear room. Tests 2 and 4 were initiated in the left hand side of the room and also resulted in accumulation of smoke in the left hand portion of the room in areas A-2, A-3, and B-2, B-3. This is best seen in cloud pictures for tests 2, 4, and 6. In these tests, the simulated smoke did not move to the right portion 1138 177 nutech
SSP-01-001 of the room nearly as much as the smoke had moved to the left in previous tests. This demonstrated the accumulation in the lefthand portion of the room. Testing also showed that a fire in the right side of the room would result in an accumulation of smoke along the A-row of the grid away from the ventilation exhausts. This is best demonstrated in tests 3, 5, and 6. In all three tests the steady state concentration of smoke along the A-row was higher than in other portions of the room. 3.1.2 Time I Time for each simulated fire was adequate to reach a steady state dispersion of simulated smoke. This can best be seen on the SF pl ts for tests 3, 5, and 6. These plots are in 6 Section 5.3. The plots show that there was an initial increase in the SF concentration, a peak and then a decrease 6 or steady value for the remainder of the tests. The plots demonstrate that the steady state or decreasing conditions occurred in all cases by Time 15. 3.1.3 Sample Points The 27 sample points spaced about 10 feet apart and utilized for the first six tests were adequate to determine the simulated smoke movement throughout the room. The sample points showed .2 l );3g )73 nutech l
SSP-01-001 that the simulated fire resulted in a higher concentration of SP in the immediate vicinity of the fire and then the simulated 6 smoke spread into other areas of the room. I The sample points were adequate to show the distribution of the simulated smoke during tests where the smoke was spread through-out the whole room and also during those tests where the smoke remained localized. The smoke was dispersed throughout all the room in tests 1, 3, 5, and 6, and was kept localized to a small area in tests 2 and 4. The sample point selection would have provided more information about vertical distribution if more samples at chest height had been used. Of the 27 sample points selected, only four were below ceiling height. These four points did indicate that the simulated smoke concentration was higher near the ceiling, but additional chest high points would have provided mc re confirma-tion. For rooms with a higher ceiling, additional samples at various heights will be required. Data on vertical distribution came primarily from grid point B-3, where two sample points were located near the ceiling and I one was at chest height. Some vertical distribution did occur and was most obvious in tests 2, 3, and 5 where the lower sample, B-32, was consistently lower than the samples near the ceiling. ff30 )[3 3.3 nutech
SSP-01-001 ~ The stratification did not appear very clearly in tests 4 and 6 because all three samples taken within grid B-3 were approximately the same. 3.1.4 Fire Simulation Locations The seven fire simulation locations did provide information on smoke movement from a fire in any location in the room. These seven positions included all four corners, two in the middle and one in the battery room. The smoke distribution patterns were duplicated for tests in each end of the room as shown in tests 3 and 5. Both were conducted in the righthand side and resulted in similar simulated smoke distribution patterns. Tests 2 and 4 were conducted in the lefthand side of the room and resulted in similar distribution patterns. Tests 1 and 6 were conducted in the middle and also showed similar distribu-tion characteristics. 3.1.5 Smoke Simulation The simulated smoke movement in the switchgear room was consistent with the expected movement of real smoke. At the initiation of fire, a very small amount of simulated smoke was dispersed almost throughout the entire room. After this initial dispersion the concentrations continued to increase as the ventilation system distributed the simulated smoke throughout the room. The best examples of this are tests 2, 3, and 5. In all three cases, there was a detectable amount of simulated 3,9
- 1138, a
4 nutech
SSP-01-001 smoke in almost all areas of the room followed by accumulation in pockets and distribution by the ventilation system. An I initial thin dispersion followed by accumulation in the pockets was observed during the testing and is consistent with expected movement of real smoke. Further discussions of smoke simulation principles is found in Appendix C. 3.1.6 Heaters The 10,000 BTU / hour heaters used for the prototype test were adequate because they created a sufficient plume to raise the SF and heated air off the floor. At the same time the heaters 6 did not appear to disrupt the normal air flow fron. the ventila-
- ion system.
The SF was dispersed throughout the switchgear room for all 6 seven tests. In considering heater size, it was not necessary that the heaters generate a plume large enough to drive the sintulated smoke to the ceiling, but rather simply get the simulated smoke moving so that the ventilation system could r disperse it throughout the room. It would be very difficult to simulate the heat source of an incipient fire because of the many variables of any fire. This was not the purpose of the heaters but rather they were used simply to begin the air movement and generate an adequate bouyant force to raise the hot air and SF mixture off the floor so that the ventilation 6 system could then disperse it. 3.s 1138 181 g nutech g
SSP-01-001 3.2 Conclusions The testing was successful because it accomplished two primary purposes: (1) demonstrated that the methodology was valid and could generate usable results, and (2) the results did provide enough information to select those optimum smoke detector locations or those locations where a smoke detector would be most likely to detect a fire as it was just beginning. I Based on the results of the tests, it can be concluded that the NUTECH Smoke Simulation Testing can be used for future tests in other plants. The purpose of this testing is to determine smoke distribution patterns and provide an engineering basis for selection of optimum smoke detector locations within spaces of nuclear power plants. The recommended locations for the smoke detectors are presented in Section 4. 0 - Recommendations. Based on the results of the I testing it was possible to predict smoke movement and accumulation in the switchgear room. Selection of the optimum smoke detector locations resulted from examination of the results of each test, identifying the smoke detector location that would first see the smoke from a fire in the test location. By collating the optimum fire detector locations from all the tests, it was possible to identify the optimum smoke detector locations for a fire in any location of the switchgear room. I 3.6 l]}] } g '? nutech g
SSP-01-001 4.O RECOde1ENDATIONS 4.1 Smoke Detector Locations Based on the results of the Smoke Simulation Prototype Test conducted at Yankee Rowe, a system of 12 smoke detectors is recommended in the switchgear room. A description of the optimum smoke detector locations is shown in Figure 4.1, Optimum Smoke Detector Locations - Description, and Figure 4.2, Optimum Smoke Detector Locations - Grid. These detector locations will provide early warning of a smoke producing fire in the switchgear room at Yankee Rowe. These 12 locations are based on the results and conclusions of testing and account for ventilation flow patterns and the geometry of the room. Other configurations of smoke detectors will provide protection for the room with the 12 detector arrangement providing the most protection and highest confidence of early warning. 4.2 Continued Testing It is recommended that, based on the results of the Prototype Test, this Smoke Simulation Testing be used to fulfill, in part, the NRC requirements for insitu smoke testing. 4.1 1138 183 nutech
SSP-01-001 4.3 Test Improvements The tests conducted at Yankee Rowe were prototype tests or demonstration tests. The success of these tests indicates that the testing can be used to locate smoke detectors or to confirm-proper placement of installed smoke detector systems. The following improvements could be made to the NUTECH Smoke Simulation testing technique to enhance its value in future tests: Use additional SF detectors for counting samples. 6 Thc testing required a 2 hour wait between each test to count the SF SamP es. An additional SF dete t r would cut the l 6 6 sample measuring time in half, and allow a much more efficient use of manpower for the testing. Other metbeds to reduce the sample measurement time are also being investigated. 2. Additional heaters should be used to heat the simulated smoke as it rises in the plume. The two heaters used for the simulation test did succeed in causing the simulated smoke to rise as indicated by higher SF concentrations 6 near the ceiling compared to those taken at chest height. Additionally, the heaters need to be modified so they do not trip off and on. Further discussion of smoke simulation is found in Appendix C - Smoke Simulation Principles. 3. More vertical distribution data should be obtained in the testing. Only four of the 27 sample points were located other than near the ceiling or the top of the room. 4" 1138 184 nutech
SSP-01-001 Some vertical distribution was observed in all except test 6. Ir. formation on vertical distribution will be important in tests of rooms with high ceilings. I I I i138 18=. I I I I 4.3 nutech I
SSP-01-001 Figure 4.1-OPTIMUM FIRE DETECTOR LOCATIONS
- DESCRIPTIONS PRIORITY GRID DESCRIPTION 1
B-3 High, on ceiling, on non-turbine side of light fixture 2 B-2 High, on ceiling, by end of lighting and battery charging panels 3 A-3 High, on ceiling, near wall 4 A-2 High, on ceiling, near penetrations to control room 5 A-6 High, on ceiling, in middle of conduit runs 6 A-5 On bottom of 6" beam 7 A-4 On ceiling, near top of cable tray 8 B-6 In middle of pocket formed by I cable trays and ducts 9 C-7 High, on ceiling near edge of I 2400-volt switchgear 10 C-2 High, on ceiling, near ventilatio.' duct passing across battery room 11 C-5 7.figh, on ceiling, over transformers 12 B-5 High, on ceiling, near end of catwalk I I l 1133 185 I nutech 4.4
SSP-01-001 1 2 3 4 5 6 7 I A 4 3 7 6 5 A 12 DETECTOR a 2 1 12 ~8 B gg c 10 l1 9 c 1 2 4 5 6 7 ^ 4 3 6 5 A B 2 1 7 B 10 DETECTOR SYSTEM c. 9 10 8 c I 1 2 3 4 5 6 7 ^ ^ 7 3 2 B 8 5 1 7 DETECTOR ' :52M i c c o 4 1 2 3 4 s s 7 ^ 3 2 I B 5EMM B 5 1 SYSTEM c c 4 1 2 3 4 5 6 7 ^ ^ 3 2 B B 3 DETECTOR y SYSTEM c c Figure 4.2 - Optimum Smoke Detector Locations - Grid nutech 4.s ll38 187
j SSP-01-001 s _J 5.0 ANALYSIS
- a 7
j 5.1 Data Analysis During Testing Preliminary data analysis was conducted Detween each of the tests during the 2 hours required to measure the SF ontent 6 ^ of the samples from each test. The SF concentration profile, 6 a indicative of how smoke would be distributed, was plotted on a sketch of the switchgear room. = = Evaluation of these initial plots showed that the SF concen-6 ~trations were within the range of the SF detector. Additionally, 6 this preliminary examination confirmed that the simulated smoke movement was generally as expected and that results of the tests were reasonable. Similarly, the trends were roughly identified to insure no errors had been committed and the testin T method did not require drastic modification. _j As an example, during preliminary evaluation of the second fire simulation test, Test 2, the data showed all SF concen-6 , trations increased slightly between minutes 15 and 20. Review ,f of the Test History showed that the flow of SF ncreased 6 slightly at t=17 minutes. This was confirmation that the SF6 _7 _Q readings were responding to variations in SF f 6 s 1138 !8o e nutech =
SSP-01-001 This preliminary data analysis was also used to determine if the samples were being taken at the correct times. It was important that the first sample be taken soon enough to show concentrations still increasing and show where the simulated smoke was traveling. Additionally, the last sample had to be taken long enough af ter the start of the testing so that the steady conditions were reached as indicated by no change in SF concentrations. 6 Examination of sample times between tests resulted in a reduction of the number of samples taken during each test. Seven samples were taken durirg the first two tests and only five samples during the remaining tests. This change was made because the first two tests showed concentrations changed slowly enough that five samples would be adequate to show how the simulated smoke moved. This reduction in samples also reduced the wait between tests while the samples were measured for SF content. 6 I Examination of the first three tests also showed steady state was reached within the first ten minutes. Thus, the final sample for each test was changed to Time =15 minutes versus Time =30 minutes used for the first three tests. 5.2 1138 18? nutech
SSP-01-001 5.2 Post-Test Data Analysis Post-testing data analysis consisted of several steps: 1. Data Tabulation: Data was re-tabulated from the readings taken at the site. This data was reviewed and checked to insure that it was consistent and the necessary calculations were done correctly. 2. Normalizing the Data: For each test the higm 'st SF 6 concentration was used as the denominator to normalize each data point to the maximum concentration. Thus, the numbers used in the data analysis are normalized values for each test. These normalized values are listed in Appendix D - SF6 Concentrations - Normalized. This normalization was done to provide a clearer picture of smoke movement. Sample concentrations varied from 0.05 ppb to 50 ppb and normalizing to the peak concentration makes the distribution easier to visualize because the data presenta-tion is only 2 digits. The normalized values are the percent of peak value at each sample location. 3. Grid Concentrations: The normalized values were then plotted on a simplified grid of the sw!.tchgear room for the five sample times for each test. Thus, a single sheet of paper contains the history of the concentrations for each simulated 5.3 l}32 1, c 1 6 7 I ^f[ N "NK $N '7" ^ e'5 k \\u 51 55 Nat Ng 2 41 25 sh 2 / B T = 19 54N M.QpC7 30 35 28 I Figure 5.2 - Simulated Smoke Profile - Test 1 !l38 i95 nutech
SSP-01-001 TEST 1 to D ~ UMW e0 N eik~i4{ W i - 4FA M. _P. M 2 T F W W W O m = ,p_ KfWi hii-4ths: r Mphrit?- _~ rwg- .p== = s + +43-
==m m =a =p 54-Q= LM=;id=E~,34.$1.= : +M.~ = = --
==l-id:=#L2MM W EsT.=W@==Mi- = - - i= mi ;=
=e :
=====u + =- = n g _==ga==
% =v==-@s e +=w vM=
i=PR== wt'Em
== ,a.;=/,. - = W. m 3=b Ms =i:xw 2+. sw-- s== +gy - m im +2+==- t=-is== =-mtg r=5# =w wrs=a -w --+ 2 -d =-f g4 ,p
m!
a=3 g g G _ % = lass F M M B; W##i 0 44~=EfE =4 TbNL=66+1 = =-EI== =rE - =
==E mt3E Z 3 p=~=q~.ks. f f -s 'w%- ___8., ag _ m-
- -s
,1 cyz 39 / ~ ~ i f' 'w g/ y \\,' a E l i e-x tQ NORMALIZED SF 1 6 ~ I I to l 9 -~ - ~~ ~ - ~ - = - - 9.~mt -- AMLer ;* = ~ ~ ~ - = -= T: .:: ^ P- ' t.- = = ' ~2 Y v= =-. = '22.-:? ^ l '='m==~ -~'F" ~~"! --M.. =- ^ = - - = " ~ - - - 4 -" -'-
==r 8 =T '~ n2h 131! ifthiit== T?5 TEME nift T:tf55???-:YP 12 1.11. O : i = - - - - - =
-
==-
- - =: --- ---
- -
= - - fs==#7=~.MEI=isi-I='_ .. _ :14E --5 =r._' 45#=s =~- V=E_Mi-- --l--Mi =h-- E
_LiiEf4 =@f= -;
--===z 6 F N N ~5= E -E -5E~ I =--N = -- -E - - ~
==- I .. _. - -: =% _L_ = ; ::2 +===:_ ti==== :==="_. _ XL= t=== ;-; : ~
= =
..=
- s. _ - z-- :--
5=
-= =- 4 l I 2 I s 2. a. 4_ 6._ t'o ti g (g is TIME (Minutes) Figure 5.3 - Test 1 SF Plots 6 Q 3b i96 nutech l
SSP-01-001 5.3.2 Test 2 Location: Test location was in grid area A-2, slightly to the right of the lighting transformers. The heaters and SF6 apparatus were placed on the floor. B sampling: Sampling locations were the same 27 samples used for the first test Seven samples were taken at each of the 27 locations; sample times were at minutes 2, 4, 8, 12, 16, 20, and 30. Results: This test was valuable in showing that a fire in the lefthand side of the room will keep the smoke primarily con-centrated in that portion of the room. It also confirms the ventilation flow pattern as counter-clockwise, and vividly demonstrates that the airflow over the top of the battery room (contained in the switchgear room) into the turbine building is so high that concentrations became too dilute for detection. Thus, a smoke detector next to the exhaust duct over the battery room would be ineffective in detecting a fire from any location because of the high air flow. During the conduct of the test, the problem with the portable space heaters tripping on and off became more pronounced with one heater tripping off three times during the 30 minutes of testing. In all cases, the heater cycled back on approximately three miautes after it turned off. 5 1138 197 nutech
SSP-01-001 Temperatures between the heaters where the SF as released 6 again measured 110 F. The flow meter was set at 2cc/ min. and was adjusted at time 17 minutes because the flow had decreased slightly. This adjustment to the flow was detected by an increase in SF concentration at almost all sample locations 6 between samples 5 and 6. I The results of the tests are consistent with expectations and .the results found in other areas. The smoke was localized to the vicinity of the simulated fire and remained in the left-hand portion of the room throughout the test. Physically, the ventilation flow in a counter-clockwise direction was simply pushing the simulated smoke directly into the turbine room through the exhaust duct directly over the battery room. At the time of highest concentration over the battery room, however, the level was still signficantly less than found in adjacent areas of the room. This would indicate the area above the battery room is of minimal value for location of a smoke detector. I I I 5.11 1138 90 nutech
SSP-01-001 TEST 2 1 3 4 5 6 7 \\ A 42 ( -16 1 1 1 0 u A v u B T=2 15 3 6 0 0 C C 1 h,l _ 4 5 6 7 / 51 46 4 0 1 0 1 ~ g \\ f A 57 h 57h 18 [ 1 3 2 1 1 1 0 1 7\\ \\ ioO ')3 B T = El 8 Y 12 0 1 0 0 0 C 1 2_, A 4 5 6 7 g \\ ^ ^ 1 B( B T=8 17 1 1 1 1 1 C C l 1 ,M 4 5 6 7 jfN '5 Nj' x x T=E B B( Mdg 1 1 1 1 1 C C I 1 M 4 G j 1 1 0 1 33 ^ _M \\ \\ l 0\\ 2 5 '2 2 3 1 2 T=$ B[ 1 2 1 1 1 C I Figure 5.4 - Simulated Smoke Profile - Test 2 nutech I, 1138 199 5.12
SSP-01-001 TEST 2 h + *g - i -I4 p-. h g---y =._ -+' 4-- 7 -.t - A- - N<-. -;w-- f-'", ';"' ';---T M--' + ib .,[ -^-n='- -M;;n %+. L.a.., a-- M-htet ;;=twm ..= +t +- F c-- ,g __.a W - I :.1. e. _ _Egznd; {-- =1 ..==-}n=-=.a?gi - - =? = t mnag 4 ,a - n .= = _ q g.,- r = +-.h. W. g._ A l2 M_.. h 5 =. m_htN~5.A =~5$. $_ N*^ L-. ?.S. i.H1-". =;?, '$....2$.,i 51 -m = ,e m ' T-~j/- i s4-- i i= hE-t i-PM ;M = %N
- si_14 Uni R %=r?
_=- M = - rr*= =-. = : f-. - =3 5 z.g.5_ - -355_ 53.: 4
- -- _=_--.
. u: _~ 'O y%:-p: :- :7- -^ : _-..: e :::& p-e T: d =. z - - - _, - - _.... A_ m __ w t- =. -. -== g _._ <a-._-.- r M,43 30 fx Gli %N xy s' NN s' i s x t._._. NORMALIZED
- 's'
'&hw._.. 2n s' y-r-- yx W ~- SF. B3 /b -- _~x Q 6 ~, I I x s' l ' ~ CW I ~ ~ l I I N 10 x 9" -w Ngid=:-F q4[ff {Q P% -- " !.~ 6 ~T g 5 r+.- -3__ -i = - r= --b -r t:;
- m ;M; _
.:-- :- 9 m 3 L:2-- =- = - - - - - =- ---~f~--
- - z
--i'.~ -: - 1.~-
- =. :_-
y . = - - - r . v += = . = = + -b,t <= r1
t
- w=--
-= r.:-4 = =m ._ =c =- _ 6 (I~ -- D # ~~~~~#" -[s== 2 r i-'~
- E 2 ~- -~ ~ * ' - T-
!~ ~ ~ --Z - N- - f= =y =mi -- _g==2
- :: = =-4
--:
:-
u== == =.. f:1". w=~- tl =a~--L:--- Z ~= ' =_ -- /f = "- @l '=- - =^ &si=-t =+W Z=lt J ~ = =-___ I': 3 l l 7 / 4 1 I e/ r a i i
- 2-
,1
- A 1
1' i' s / SV c' .. 1 0 t i 34' 5' 6 7 8 S to u t2. is (8 2.1 ' z4 z7 So TIME (Minutes) Figure 5.5 - Test 2 SF Plots 6 J ;33 7g nutech s.n
SSP-01-001 5.3.3 Test 3 Location: The location of the test was in grid area C-6, outboard the 2400-volt switchgear next to the wall to the turbine building. The heaters and SF apparatus were placed 6 on the floor. Sampling: Sampling locations were the same 27 sampling I locations used in the first test. However, only five samples were taken at times 3, 6, 10, 20, and 30. Results: This test was one of the most conclusive demonstra - tions of the dominant counter-clockwise smoke movement. Similarly, the spread of SF --simul' ted smoke-- approached 6 the ideal as demonstrated by the simulated smoke profile. The test also confirmed that a test in the far corner of the room would eventually migrate into the area where higher concentrations were seen during tests 1 and 2. The results of the test identified a very strong counter-clockwise flow of simulated smoke which is best seen on the " cloud" sketch showing the simulated smoke movement, page 5-16. The previous tests had been in the lefthand portion of the room, and the smoke had tended to accumulate in that area. In test 3, the simulated fire location was in the lower righthand corner of the room, and a similar phenomena occurred in that I 5.14 1138 201 nutech
SSP-01-001 the smoke accumulated in the lefthand side once the ventilation system moved it over there. Thus at time =20, the A-row of grid points is at the highest concentration, and the B-row area or grid also shows some accumulation of smoke. As with the other tests, the B-4 and 5 areas showed lower concentrations because of the high air flow in that area out of the main supply outlet in grid area B-3. This test also showed that the steady state condition is reached between t=10 and t=20 minutes. There was very little change in the concentration profiles of the simulated smoke between t=20 and t=30. This observation became the basis for the sampling times used in subsequent tests. This test also denonstrated quite vividly that once the test begins, there is a general dispersion of a small amc.unt of smoke throughout the space. SF concentrations at t=3 were 3% 6 to 4% of peak SF concentration almost uniformly throughout 6 the room. By t=6 minutes, the concentrations had increased to about 10% of peak throughout the room, except where the ventilation system was distributing the smoke. By t=10 minutes, the entire room remained at approximately 10% of the peak concentration except in those areas where the air flow was concentrating the smoke, where concentrations reached up into the upper 20% of the peak. These general room concentrations s s g 1138 202 nutech
SSP-01-001 remained constant throughout the remainder of the testing. This demorstrated that although there would be some initial uniform dispersion of SF ug u e g pace, de 6 testing did result in a definable SF gra ent hom one grid 6 space to the other, which was being affected by the ventilation system. I The conduct of the test was much smoother than the first test with the exception that one sensitive heater continued to trip off for a minute or two at a time. The heater tripped off at t=5 minutes, t=10, t=16, and t=26. In all cases, it came back on within two minutes after it tripped off. The temperature in the simualted fire was 110 F, dropping down to 106 F during the time the one heater was off. Ambient temperature in the room was 85 F. The SF flow meter setting was at 10, slightly 6 higher than other tests. 1138 203 5 16 nutech
sse-o1-oot g l rest 3 I 1 2 3 4 5 6 7 3 1 9 9 4 16 1 3 2 5 -3 4 4 6 8 17 B T-3 5-5 ' ((2g 1 g c c I ~ l 2 3 4 5 m 7 A A j]' B l 7 u u or 8N 9 (N C %) c x 1 2 3 7 y Kx x W. l g ),W y e n 1. B T = 10 17 p 16 16 16 18 g 1 ('(D x'N'N ^ ^ I BQN'2[ # ]yT)a T - 20 n 1. n I 1 2 3 L -_ % 7 1 2h' 2j A f A sx x. uu X 3 l ( 2 N W T7 9 8 12 9 B T = 30 (.,g3xy 1. 1. 2, 1. .c I Figure 5.6 Simulated Smoke Profile - Test 3 !l38 20; nutech g
SSP-01.001 TEST 3 10 0 .-'+4* Ti-y !~ '-Wr ~ ".w ' - _--~ -:-*-~ r & V[ t
- A-k h r, '^ r i- -!j
,p 3 ~-y= r j -l-Ul } =3 W- , p_ 'C
- O M lf
= C+ .? J : - t--i Mt4 W i~ 7 =++MI %;2 = MdMy 4 ?-%Wwi iq#b~E.'tL_!. ?!. ;=;;,. si..=+. Lii? L : :M. LL:, _ '. .. =_ t L. = ,p .i=.- i rm -- i =.== =-===1 = . :a -- +L-da at =r.42 = 'M- = ,a 4=+T -EiiD:- =42 M =-L&+
- E t = ~=- #T=i
= = + = = 4= iM :-==E= :r = mi 7
- _.. __ _2 - = y-_;
- -;. g ___._; ta=E+2m ;.::
- '-- --Lgg^ =E5 y = =- += q m
aq -_---r--- --t-0 3 I ^ 6 a W ,i 633 ' j'rK 'x l NORMALIZED 6fjd. ih I 6 Ml ! h- \\ f t0 11 v
- a: 2
+ l
- = _-,
,, a 1-. :- w-._ J=. L }.- t,t.
- -l
- , _.
-jwe m, c _c
- LJ Ok/=
~ 5- : _ Zl3pid:A e-;R. Den jn M y T 2f;-$ 7 = + i a=-., = u-q ===-g.m-w =_ w -.g + n m + p3[o s+; w.g=_=g w t + = -. - -- p+
- = H Miri +=-4=
5iam = E %E* + + W= ht-a4=E-F"m== E F.+5K+ 4===4y s/ =-# 4d1+1:t =EI= + g-h_i=+ zarrht-ME 2=;=a gg=.--- a =- =- 4i2=4 MNfr E#i- -M i E-i= M?EE i'_=== 3E+=.
2 =iiYu=-; ; tE:::. - i=--;
i , gy_ygj_==_-.:
==_ = p- - -
== = g -- - - - - = 7 -- -
=
.= 5silEf=1 ~ ~~ Y 3 I d+Y I 2 1 s 3 to is z.o 24 27 3o TIME (Minutes) Figure 5.7 - Test 3 SF U 8 6 nutech 5-1138 205
SSP-01-001 5.3.4 Test 4 Location: The location of the fire was on the workbench in grid area B-1. This was selected because it was in an extreme corner of the room, and it was suspected that smoke from a fire would have a unique distribution profile. This was also a likely source of fire because of the maintenance work that is done on the workbench. The location of the heaters and the SF released 6 was on the workbench approximately 2 feet above the floor. Sampling: The sample points were the 27 sample points used in the first test. Five samples were taken at each samples were taken at minutes 1, 3, 6, 10, and 15. Results: The results of the tests were interesting and valuable be.ause they showed that smoke from a fire at the workbench in B-1 would immediately be exhausted into the turbine room with virtually no mixing in the switchgear room. Additionally, the high concentrations in the immediate vicinity of the simulated fire were signficantly diluted as they passed across the top of the battery room roof being exhausted from the switchgcar room. Prior to this test it was thought that a smoke detector placed over the battery room roof would be a logical location because the smoke would pass over that area as it was exhausted from the switchgear room. 5.19 1138 20 nutech
SSP-01-001 Test 4 demonstrated that even with an extremely high concen-tration in adjacent grid aree, the concentration abo..e the battery room remained quite lcw. In fact, it reached only 5% of the peak shown for this test. Thus, this test was valuable because it demonstrated that a fire at the workbench in grid area B-1 would not be detected exceo. by a smoke detector in areas B-1 or B-2. I I I 113s 207 l s.2 nutech I
SSP-01-001 I TEST 4 I 1 2 3 4 5 6 7 O O O O O O A I 0 pc \\ 0 0 0 0 0 0 0 0 0 1 B / O O T"1 ) I -N O O O O 0 0 u C C 1 2 3 4 5 6 7 2 0 0 0 0 0 0 A A h B 0 B T=3 y ' "1 0 0 0 o u u I C C 1 2 3 4 5 6 7 4 2 0 0 0 0 3 Mh B T=6 B \\ '2 1 1 1 1 0 1 U' 1 2 3 4 5 6 7 I 4 3 1 1 1 1 A 1 F%,s B T = 10 B \\ j '1 1 1 1 1 1 I 1 L, 3 4 5 7 1 b3 ( 6 1 1 1 1 A A 1
- M B
T = 15 y N"5 1 1 1 1 1 C C I Figure 5.8 - Simulated Smoke Profile - Test 4 l 1138 ?03 nutech
SSP-01.001 T' . 4 I 10 0 -m +a o a..- e . : =, v 4. :- - -R. } w 9M .. +.
- m
_n1 r r-4 ?fA-_ :4_g+ i s-g e tti ;; 3 b g = q L =,- g - s-y g ~ - ~ 33 (~ f _ -_)j hj' =5 -i- -h -C q _ _ _3-[h%i$1- -:-it=, </ -i. 4 N -c; _=.2 sq- - =-; -& t-i-c+ 3 - =++= if,-- . 7..=3;;y _ f nz_ .; g_g =-.=i =. - - -5 .= -_ i;= c-i=_= g- - I
- M5 us cd M ~d =
/stt -t ri = =- T:- a : +si'S 4LEMirF=ak . = - - ,a =+ =tz-i Gi^= -MW EDM JLE^E Mi+?== M +#EM
- EE== KEt=_=
t =E=l + E =3 Y :~== z= =,J#f:- --_Tr~f "" ~ ~~~ ~ ~ " ~ ~ ~ ~ ~ ~ ~ ~ ' ~ ~ ~ ~ ~ ' .a ~ t-n! i 3a ,i / l "l' I I I I ~ I ~ 2O I i r NORMALIZED SY 6 I i CL my f,E QL2f M r.-SL LL =- y ---- -{ L- =-
- 4
;,h w "-'
_-~t' - _= __; - _- g+; y-gg 9_Q n ~ um . = _~ f - x- - -- b m + O r=-h;se ,e - - --- - -- s C - m -u I .+ =+n f p = wi=xy- = mv 7 . = + - - -;g qwy y +,== =. --- ---:=== .=+1:2 i e =+w+=- &wa-w =1a ===k =- i= z#v M e-- c- - = = =
====+ t ===- t = - - - +=cz= n-z: a= =:=g m- _=_i ; m+ =u -_ -;q ;-.1._--- =a =a: -==== = g =;=_ _5$TT ?!? $$?: ??? WYtW5W5 ???? ?OYWW if E I F2 =E'~-E=~~= -- _-~-===-~1=------ = s. = ~ L^~ &= --T ~=~-32WW 4 [- L I 5-: = l 3 ,/ I l / I 2 i BIL- / n Il I IV2 l 1 i 1 ( I t 3 6 . LO 15' TIME (Minutes) Figure 5.9 - Test 4 SF' Plots P00R OR M 1 nutech e.22 1138 207
= SSP-01-001 5.3.5 Test 5 Location: The location of the test was in grid area A-6, to the right of the pneumatic control panel close to where the major conduits to the control room are bracketed. The heaters and SF apparatus were placed on the floor. 6 Sampling: Sample points were those same 27 identified in the first test, and were taken at time = minutes 1, 3, 6, 10, and 15. The heaters both operated continuously throughout both tests and maintained a constant temperature of 110 F between the heaters. There was some flow directly into the vicinity of the heaters from the door leading to the stairwell in grid area A-7. Pressure differential across the door was +0.06" of water and a measurable draft was observed in the corner near the door during the testing. Results: Test 5 again confirmed the flow cattern for smoke from a fire in the switchgear room. The smoke progressed counter-clockwise with concentrations in the same areas as those found from other tests. There was high concentration in the vicinity of the test, higher concentrations along the top of the room, or the A-row and peak concentrations seen in the B-2, B-3 area of the room. a 1l38 210 5x23 3 nutech h
SSP-01-001 Upon initiation of the test there was a very rapid dispersion in the immediate vicinity of the test, possibly due to the draft from the door. By three minutes into the test, the distribution of the smoke was very similar to that observed in previous tests. The smoke began to migrate counter-clockwise along the top of the A-column of the room, headed toward increased concentration in the B-2, B-3 area by the time =6 minutes. By time =10 minutes, the pattern was well established and maintained itself relatively constant throughout the remainder of the test. }}38 211 5.24 nutech
SSP-01-001 TEST 5 1 2 3 4 7 (826 \\ 31 s90 2 2 s A N% A f 1 5 18 3 2 4 1s 16 87 2 B 13 11 B T=1 ) 2 1 1 l 2 C C 1 2 3 4 _ e M_ 7 A A s ' 'W> ~ p )'" l .oxics j 12 2 1o C C I 1 2 3 ~~~ [Y YNYhN ^ ^ I hu q 3,' u sa 3,x e 22 3 )12 22 s ,,g N .w ~ I ~ ~ 26 is 12 to 74 N5 / 6 C C l 1 2 /?xNVxX 'h"b^ I ['N^,dj\\jgN"q-e 1 - 10 w q se 1, 1. c l 1 A s A l a ( x ;' N 'Agg(Nsj 22 o 1 - is ->s 2o is 4, s, j c l w c ~ Figure 5.10 - Simulated Smoke Profile - Test 5 1138 217 nutech 5.25
SSP-01.001 TEST 5 , too JM ++ g-5... M-gn%.' ,e>t 7 7 q 4g %.q['gm,-g , %N fpg M-ew +M+LE+%ud;l 4 h, ..i wlI g ye &-. $ ;' ? L. Ex.yi_1. - w .= m - _V WW,. a ' _R[m.??'e ^ ..N-- - BP i tr f +- -=. .. =. = = +,
==- m, = :, - _.-_ nt .=. +s =%=. - --+-.==y +:43 g .n,s 3 p -{l r$ ~~,, , f 7j-O. p_-f:9 ::. p- - ph
- .~.--
. F9_ F5.2. g-pgi - -i - '^ r_{ Y _ _: II[= ' --37,6 -F-5 -5 zi E~5-NUsii --UN1:.._=_- 55-it. U E-M -QU_b5'd ;E.i-- j- -E -.=. Euh.=.. -s_i L i_i_ :i_L_E-i.31 5. ^i.'r. 3 %...^ d, _ - - - _..
- f..
_ 3%3 ELL ;.p;q = =. = _ = _ _ = _
-
...,- 'gg---. - = r =d =%._=_.L i . -p- - -...,=,. m p _. y_ -=r -g_ _- 4g gx=. _% $ MT=h E 30_ l
- f r
l .-/--,' w I
- -1 r,'
I y I A I If I sf 3 ....l, _.,{ '. _.... ~ T NORMALIZED / l SF 1 6 __/- l' 10 !&+F VWt' yML Wi+h n+ k+ + %- ~ _--- -* M a 7 I "!!* UlO EM TE 2= r^ n- - w. -.= y, = - '. = -
- 1-x_*.:
+ i 4 =_: +-- p 4 = ^i. x_.--.i;_---
- =-=
_:_ } ~ _:.== ~~'--i =. _ = -
- A i = :
--g:- -.5 5 i_
- ..-- E G-~.
- -+ =.: i =._=- } 4 -[Q
, : _ q. . :-- 5 s - :: =l _i_= _. =_=1=,
- iti=
gv=_ = = a g=t==: 1+=_+=g wy== =-== +:: =r i'. _ _... _ u r:.: __.--_=_p-
- __=_'-'~ ~:~irt-FE- =EIES _r -_E -~ 'E_2.g7 7;_; =gg _r+ =:gg_ :g - -- -
- . _.- g ;=;;;_j-y
~ m = s - +- ~= = - - - = = = = = = - - - - _y ___1 n: :: - hM-4. -t l 3. ~ i i1 2 I ,I 1 1 1 o i t 3 4 s 6 , _g 3 in 4 , a. .., to,y TIME (Minutes) Figure 5.11 - Test 5 SF Plots P00R BREM. nutech 6 e.2e i i 3s 213 I
SSP-01-001 5.3.6 Test 6 Location: The test location was in grid area B-3 to the inverter panel. This location was near to the main ventilation s.pply outlet that caused a very high air flow down the middle of the switchgear room. The heaters and SF apparatus were placed on 6 the floor. Sampling: Sampling locations were the same 27 identified in the first test. Sample times were at minutes 1, 3, 6, 10, and 15. The test was uneventful except that the same 27 sample points were covered by a total of eight personnel compared to the 18 5 'tsed during the first three tests. Thus the prototype was I successful in improving the ability to take samples with fewer people. The simulated smoke was distributed uniformly throughout the switchgear room in this test because the simulated fire loca-tion was very close to the primary ventilation supply outlet. Although there was some increasing concentration along the A-wall of the room as the test progressed, the increasing concentration was consistent with the increased SF readings throughout the 6 room as the test progressed. This test did confirm that the ventilation system is the dominant motive force for distributing the simulated smoke 5.27 1138 ?)A nutech ~
SSP-01-001 in the switchgear room. The rapid air flow from the switchgear room into the turbine room was so high that the simulated smoke was relatively evenly distributed throughout the room. This result was somewhat predicted from other tests and engineering judgment. However, the simulation test did serve to confirm this judgment by actually showing how the smoke will be distributed. I I 1138 215 I I I 5.28 s l nutech
SSP-01-001 I TEST 6 I 1 2 3 4 5 6 7 31 39 36 41 30 43 m s39 33 43 5 5Q ) 46 51 45 38 N I = s a ,.1 = 48 48 49 56 44 48 50 C C 1 2 3 4 5 6 7 I 45 42 36 51 44 45 A ,ik
- 52' \\ 61, (3 9(67 51 3 48 40 so 35
\\\\ j 5\\ \\ \\ B T=3 g N k '5{ \\ \\58\\;>46 55 45 C k C p 1 2C 4 5 6 7 [ 57 50 48 53 48 54 I . s\\ 'N N \\ ^ ^ m 49 ( 5 1 72 100's 60 56 3 46 55 \\ \\ \\ B T"6 2 I 6k b' \\\\53\\ 53 50 C \\ C AM I 1 M 4-5~ 6 7 4 6s 35 50 r $y 3
- , g 5
( A A 31 6 O 18' 78 59 63 34 2A \\ja2 '95 \\ \\ I 65 \\61 su ("66 v' - C mAh s C I 1 mm e 7 / 57 60 5 37 1 54 l / \\ N5o s\\ 3 s ~ B T-15 k C C I Figure 5.12 - Simulated Smoke Profile - Test 6 nutech I b LiU 3,, s.29
SSP-01.001 TEST 6 ,'q w l )e de-e d &+ yd P% M +== .c my-- m+-x. "_m E]-:=N_ :. "- " " - " -"-i*b f r+mxmt7 .. m .st em- = Yi_ E C li p Z r3 N NW +Osiy=d ~ "57 ~+ 5 '- - - 4 ,q h$ Y [W f +l %-2-W N_$. & ~i$$ Ti -Mi - ,o
===zHs===h==3ikg3 y.- g _==2WQ M =.=E=-
- +2 =
yt= a =1:-+ 1 m HbbW+p +i -- ?===?DN5 =Q- ,p p.m_.=3_-gy y--_- = r-q-_ - --- g c =
==
I 8(;me ~ ' - ---/-- _._t_ 4o A4b@ 3D 20 I NORMALIZED. SF 6 tQ wu o_.- q L a ; - _g; t + pin. - - j_m - ~ S ^ -- ' * ^ ~ I ~ ' 8 mw w -= = +- -:.=ing -ck-==- + =:= 1== -r::=z = 1 ;=Jci =--+--@ -. k =._a=-w = = -== .5-i_ 2 -g E- =.2__
- r._--:
E. ; ~~ iEip = g = -:_--J :- - - ~=1 'E:' _E d'izi-7_ c.:--- 4~.p.r- .. - - = - ..=.g. g= g = . = g= y g = = - a-_+ u= a =. g = _._.g== ===-1 -:.+e au.== - = =-c=g _== m g = g=y _5 =-p-
===-1 _--;; ;-g. yz p = - = - +====r-t=t?=.22t =t'i + = Mi=L- =;=i;==- = ==== h =n#12 =i -h=== e L+:== t ' = = ZEEEET.: =Ca ~~ -H"=E i====' r --a ~~ z- .- _Z
=
3- -= ___=_. - _ 4 3 I 2 I 1 a a i '6 4 .lo_. IT TIME (Minutes) Figure 5.13 - Test 6 SF Plots 6 5.30 ?00R DMNAL nutech 11~38 217
SSP-01-001 5.3.7 Test 7 Location: Location of this test was in Battery Room 2. The heaters and the SF apparatus were placed on the floor close 6 to the door of Battery Room 2. Sampling: Samples were taken in Battery Room 1, Battery Room 2, and just outside the battery room in the vicinity of the door. Samples were taken one minute prior to commencement of the test, and then at minutes 1, 3, 6, 10, and 15. The purpose of this test was to determine the amount of leakage of smoke from the battery room into the switchgear room or into the adjacent battery room. No leakage was expected since the battery rooms have a separate exhaust that discharges outside the building. This test was initiated one hour af ter completion of test 6. The background SF concentration at the start of test 7 was 6 too high to allow any realistic conclusions to be drawn from this test. Samples were taken in both battery rooms and in the switchgear room prior to commencement of the test that indicated the SF concentration had significantly decayed to a low level 6 from test 6. However, the samples taken at t=10 in all locations indicated a significant mmount of SF
- ^ ""
6 5.31 1138 213 nutech
SSP-01-001 time the test was initiated. This high ambient level in the background invalidated any conclusions that might have been drawn from this test. I I 1138 219 5.32 nutech
SSP-01-001 APPENDIX A NUTECH SMOKE SIMULATION PROTOTYPE TEST PRE-TEST MEETING ATTENDEES NRC AND CONSULTANTS Calvin Heit NRC Leo Derderian NRC Mario Antonetti - Gage & Babcock Ingemar Asp Gage & Babcock Jack Clavan Brookhaven John Boccoio Brookhaven NORTHEAST UTILITIES Mario Blancaflor John Roncioli Gary Dylinski Joe Summa COMMONWEALTH EDISON Terry Pickens NUTECH Tim D. Martin Francis Fung BOSTON EDISON Rick Venkateraman YANKEE NSD Bob Atkisson Jay Thayer Ed Sawyer YANKEE ROWE Ed May Dave Tong 8 '20 nutech A-1
SSP-01-001 APPENDIX B SWITCHGEAR ROOM DESCRIPTION NUTECH's Prototype Smoke Simulation Testing was conducted in the switchgear room at the Yankee Rowe Nuclear Plant. The switchgear room is approximately 115 f t x 38 ft x 15 ft. It contains the major electrical switchgear and control panels for the plant, such as: 480-volt switchgear 2400-volt switchgear Rod eqaipment panel Motor control center Pneumatic relay rack Inverter and inverter panel 2400-volt test cabinet Battery switchgear Cable penetrations to the control room. The arrangement of components and the dominant ventilation supplies and exhausts are shown on page B-3 (arrows) of this Appendix. The fan room is located to the left on the drawing, and generates the large air flow indicated exiting grid area 3-3. This major outlet generates a flow down the center aisle. 1138 221 nutech
SSP-01-001 There are two other s;pply outlets along the wall that discharge across the switchgear room. The room contains three exhausts to the turbine building located at the bottom of the drawing. All three of these exhausts are approximately 3 ft x 4 ft and contain dampers that provide no resistance to air flow. Ventilation flow rate is approximately 30,000 CFM and generates a very complex flow within the switchgear room. The switchgear room air flow is very complex because the c space contains a high ventilation flow, a high ceiling, multiple N supplies and exhausts, several large pieces of equipment to distribute and split the flow, and many cable trays and bus ducts that can interrupt flow in the room. Additionally, the room contains two doors that are used quite frequently, a door to the fan room, an elevator shaft, and two battery rooms that contain their own ventilation system. The switchgear room did provide a good test location for the NUTECH Smoke Simulation Testing. The high ceilings, high air flow, unique ceometry and equipment-filled space prompted discussion about and solutions to many problems. Thus, the state of the NUTECH Smoke Simulation Testing was much improved because the switchgear room was used for the prototype test. 1138 222 B-2 nutech
ums mum sua s i s'- o" 1 2 3 4 5 6 7 I I I 1 JZ \\ LZ";" A rise musn: eden gg{f c ug A smaee / -o~ j o~ j-Oc,4 mo mu,~ew ceme= [] PANEL [, t m r 1 7 - i, u ea 'tav i,ue= Tea c-mwi m a. s#L E ' ~ '""""^' B ~ B ~~ ~ ,$.^." & N. m $,m ~2 & E m ~ UP DN BATTERY j N 7 p / { 2400 y sw CHGEAR C fr C eme" -s NO. 2. NO. I 24CO 480V TRANSmWERS s [ g V. .N 9 1 2 3 4 5 .6 s 7 i 100R BRBIL L ~ = _S_WITCHGEAR ROOM 8 D i'J YANKEE ROWE NUCLEAR PLANT c) hN 7 G Appendix B - SWITCHGEAR ROOM DESCRIPTICN - VENTILATION O lllr
SSP-01-001 APPENDIX C - SMOKE SIMULATION PRINCIPLES
1.0 INTRODUCTION
Smoke movement is the movementof heavy gas molecules and sooty particulates suspended in warm air. The heavy gaseous molecules and particulates are generated by the pyrolysis of combusion products. The warm air that mixes with these pyrolysis products is ambient air entrained by the upward convection of the fire plume. This up-ward convection is caused by the heat input from the fire. Thus to simulate smoke we must examine the basic principles of smoke movement. 2.0 CONCEPTS OF SMOKE MOVEMENT DERIVED FROM BASIC PRINCIPLES Following is a summary of useful concepts applicable to smoke move-ment study and control. The inclusion of certain concepts is guided by their usefulness in practical smoke control rather than their elegance, and to facilitate understanding of smoke simulation experiments. An attempt will be made to illustrate these concepts by derivations from elementary principles. The purpose of these derivations is to present the concepts in simple expressions to make them availabic to a wider spectrum of interests in the fire community. 2.1 Movement of Air Containing Smoke Fluid motion in general, including air or smoke, is caused by the action of natural and mechanical forces. In a fire the two motivation forces that cause smoke movements are buoyancy force and pressure forces. Buoyancy forces in a fire are caused by density changes due to heating. The relative differences in density then sets up a convective flow in a ll#3 n i 1 1/ t C-1 nutech
SSP-01-001 gravity field. Pressure forces can originate in a number of ways, e.g. by restriction of volume expansion, by stack effect or by the intro-duction of ventilation controls. Before discussing these forces, consider smoke motion caused by a generalized ptessure force, p. The moment 2m equation for the flow field in this case can be simply written as, pYVi = -Vp (1) where p: the gas density, V: the gas velocity vector i p: the pressure distribution of the flow field V: the. gradient operator. Integrating (1) along the direction of smoke motion one obtains 1/2pV =lapl (2) I where i op: the pressure difference Assume that gas density change is governed by the ideal gas law, p=h (3) where I T: the absolute temperature. R: the gas constant Combining (2) and (3) and evaluating for air under one atmosphere condition, with R equaling 53.3 ft.lb/lb
- R one obtains V = 174/ apt (4)
I where I V: velocity in feet per minute j l } } } } r*, nutech c-2 g
SSP-01-001 a-T: Temperature in degrees Rankine Ap: Pressuredifferencein'inchesofHf0 If Ap is expressed in lb/ft we have V = 104/ apt (5) Equations (4) and (5) clearly shows that it takes very little pressure difference to move air at a considerable velocity, e.g., 0.01 inch water pressure er 0.05 lb/f t can generate velocity of 400 ft/ min or 4.5 mph at room temperature. 2.2 Buoyancy Force Consider the bulk ambient air temperature to be T and the corresponding density p The buoyancy force F Per unit volume due to B locally heated air at temperature T and density p in the gravity field is simply, FE " (P - p)g (6) o Substituting (3) in (6) at constant pressure, one obtains, =f( - ) F 3 o Evaluating for air at one atmosphere pressure we have, FB" ~ where F s in l W t B Transforming to pressure difference for. given flame height H, we have ( ~ )" (0) APB" o where ApB: Pressure difference due to buoyancy force in inches H O 2 H: flame or hot gas column height in feet C-3 Il38 226 nutech
SSP-01-001 I The above formula shows that for a fully developed fire with gas temper-ature reaching 1600*F and confined to one floor with a ceiling height of 10 ft the pressure difference due to buoyancy can reach 0.1 inch H 0' 2 2.3 Estimate of Volume Outflow From a Fire Consider the following simplified cellulose burning reaction in air as representing one possible form of wood burning in a fire, (9} 2 + 5 H O + 24 N2 6"10 5 + 6 02+ "2 0 2 The above reaction indicates that for every lb-mole of cellulose 30 lb-moles of air are required to sustain burning, resulting in 35 moles of total combustion products. Knowing that one lb-mole of cellulose weighs 162 lbs 3 and one 1b-mole of any gas occupies 359 ft at STP, one can calculate the volume of combustion products per pound of cellulose consumed as follows, 3 V= x 35 ft = 77.5 ft at STP (10) Thus for a given burning rate R lb/ min of cellulose and a burn room average air temperature T we have
- T V = 77.5 R p (10a) o o
We see that in a fire where temperature can reach as high as 1600 F the effect of temperature increase can represent a five fold expansion of gas at room temperature. The stoichiometric reaction stated in equation 9 also indicates an air to cellulose ratio of 5.3 to 1 by weight. This checks with Thomas' calculation [1 J that in a ventilation controlled. fire the air to fuel ratio can be estimated to be in the range of 5.45 to 1 by weight. It 1 may be pointed out that the roughly 5 to 1 to fuel ratio fire is a lower s Number in [ ] brackets refer to References. 1138 227 c-4 nutech
SSP-01-001 limit estimate. More typical ratios of air to fuel appears to be approximately 10 to 1 according to actual. measurements of hot gaseous flow from a fire in Ref. [2]. An upper limit of hot gaseous flow from a fire appears to be 20 to 1 according to experimental measure-ments in Ref. [ 3 ]. Thus the estimate of hot gaseous flow per pound of burned cellulose given in equation (10) is a conservative extimate. In a more typical fire the amount of outfloa can be doubled. In general we can write, .T
- T 77.5 R p <_ V < 310R 7 (10b) o o
2.4 Pressure Force Due to Volume Outflow From a Fire The hot gases generated by a room fire will expand according to ideal gas law and be pushed out of the room through openings. In so doing the gas will exchange heat for kinetic energy. In a fire situation where the absolute pressure remains near one atmosphere the velocity of the hot gases can be estimated from the burning rate and the size of the room openings to the outside. In section 2.3 we have estimated that for ventilation controlled fire the volume of gaseous product per pound of burning cellulose is 3 77.5 ft at STP. Let R be the burning rate in pound per minute, then T V = R x 77.5 x ( } TA o where V: the average flow velocity in fpm through openings T: the burn -room temperature I T: the ambient room temperature A: area of cpenings in f t 1138 220 C-5 nutec I
SSP-01-001 The pressure due to this flow rate can then be obtained from equation (4) or (5). Substituting (11) in equation (4) and solving for Ap, we obtain the pressure force due to this volume outflow given by, Ap = 2.0 x 10 (T A) T for stochiometric burning o where Ap: pressure difference in inch H O 2 T, T: temperature in degree Rankine g R: burning rate in Ib/ min in general for varying degree of air supply to a fire as discussed in section 2.3, we have T (12) f 2.0 x 10-1 ( R ) T 1 Ap i 8.0 x 10~1 (T A) A o o The above formula shows that the pressure difference across the outlet of j a burn room is directly proportional to the square of burning rate, and inversely proportional: to the square of the burn room outflow area. For example in a ful],y developed room fire with a burning rate of 10 lb/ min and 1600*F gas temperature, a 20 ft doorway opening can induce a pressure difference of.0004 to.002 in H 0, whereas a 2 ft small window opening 2 can induce a pressure difference of.04 to 0.2 inch H O across the window 2 opening. 2.5 Pressure Force Due to Stack Effect The well known stack effect.in a high rise building is caused by the l difference in hydrostatic pressure dhe to two air columns at different temperatures. Thus the pressure difference is given by }}38 22} C-6 nutech
SSP-01-001 Ap -(p - p) gh where I p,: the outside density p: the inside density g: the gravity constant h: distance from neutral plane Assuming the ideal gas low are given in equation (3) we have Ap=f( -h)h o Considering one atmosphere condition, we can express the above as Ap = 7.7 ( - )h (13) I o where h: is in ft T: in degrees Rankine Ap: in inches of water The above formula indicates that for a 100 ft tall building with neutral plane at mid height and a 70*F temperature differential, a maximum of 0.1 inch H O pressure difference can be induced by stack effect. 2 2.6 Determination of Location of Neutral Plane To calculate the pressure force due to stack effect in formula (13) one needs to determine the location of neutral plane, The stack effect causes a circulatory motion with a transition neutral plane. The stack effect causes a circulatory equilibrium pressure. This neutral plane location can be easily determined by invoking the law of conservation of mass. C-7 1138 230 I nutech g
SSP-01-001 Consider the simple case.of a building with only top and bottom I openings. I
- f rea, A2 A
Velocity,Vz'- - ~P - Gas Density,p Gas Dens.ty,pg ht i Neutral Pressure Plane ~ Absolute Temp., T hg Absolute Temp., To I Velocity,Vi% AArea, Ai With variables indicated as in above sketch, by conservation of mass, we pAVy y y = p2 AV C14) 22 From equa~ ion 4 and equation 13 we obtain t v'h T and -V ~ y9 V ~~ (15) 2 gT Combining (14), (15) and solve for th e ratio h /h, and assuming p2/Pl~ l 2 y from small temperature difference, during each fire we have, I _ 2_, A h T 1 (16) 1 A T 2 This is a formula for determining the location of neutral plane when given size of top and bottom openings, and inside and outside temperatures. 3.0 Discussion or Smoke Simulation and Smoke Control In terms of smoke simulation and smoke control results of the pre.ious section can be summarized as follows: 1. To simulate a room fire in terms of gaseous outflow two additional factors need to be considered besides heat. They are the volume outflow and pressure difference created by a room fire. We have shown that the rate of volume outiflow can be simply.related to the burning rate as e !!38 231 nutech ce
SSP-01-001 indicated in equation (10) and the pressure dif ference due to this overflow is proportional to the square of the burning rate. Furthermore in a room fire with the burn room door open the pressure difference created by the outflow from the burn room is on the order of.01 in H O or less. Thus the 2 simulation of smoke movement during early stages of a fire by warm rather than hot smoke.is relatively simple and can now be related to a real fire in terms of volume flow and pressure difference. 2. The pressure created by a buoyancy force due to the presence of a layer of hot gases is proportional to the thickness of the layer. Thus the effect of buoyancy force only makes itself felt when the vertical heignt of the fire becomes extensive. From equation (8), a hot gaseous layer 1 foot thick with an average gas temperature of 1600*F only sustains i a pressure of.01 in H O across it. In other words in terms of smoke 2 control in high rise buildings the buoyancy force created by the fire itself is not the critical factor until the building is extensively involved in fire. However, if two stories are fully involved in fire with a temperature of 1600*F the pressure difference due to buoyancy can be as high as.2 inch H O according to equation (8). 2 3. The pressure created by stack effect on the other hand is propor-tional to the height of the building. Thus in a high rise building when there is a significant temperaturc difference inside and outside due to weather, the stack effect is the major factor in smoke control even at the early stages of a fire, e.g. from equation (13), a 10 story building in the winter with a 100*F. temperature difference, can produce pressure difference of.16 in H 0. 2 }}38 2}} C-9 nutech g
SSP-01-001 REFERENCES
- 1) Thomas, P.
H., Heselden, A. J. M., " Fully Developed Fires in Single Compartments", Fire Research Note No. 923, August 1972, Fire Research Station, Joint Fire Research Organization.
- 2) Fung, Francis C. W., Suchomel, M.
R., Oglesby, P. L., " Corridor Fires, Energy and Radiation Models, NBS Technical Note, Washington, D. C. 1973. 1
- 3) Quintiere, J., " Growth of Fire in Building Compartments", ASTM, STP 614, 1977.
Also "The Spread of Fires From a Compartment, A Review" ASTM, STP 685, 1979. I 1138 233 l I ~ I I I nutech l
APPENDIX D - NORMALIZED SF CONCENTRATIONS TEST _1_ ~ 6 SAMPLE 1 2 3 4 5 6 7 51.4 ss.4 sl.4 60.8 se.1 A-2 81.1 l A-3 100-81 1 64.9 56.8 52.7 ss.4 ss.4 A-41 43.2 74.3 62.2 60.8 47.3 45.9 45.9 l A-42 37.8 60.8 35.1 36.s 36.5 43.2 40.s A-5 54.1 40.s 31.1 35.1 39.2 41.9 41.9 A-61 37.8 37.8 27.0 29.7 27.0 28.4 31.1 l A-62 45.9 27.7 25.7 28.4 37.8 36.5 35.1 B-1 82.4 67.6 s4.1 56.8 ss.8 55.4 l B-21 74.3 63.s 50.0 50.0 44.6 59.s 54.1 B-22 43.2 64.9 45.9 50.0 47.3 58.1 s4.1 B-31 56.8 s2.7 48.s 51.4 48.s 48.6 s1.4 s0.0 s2.7 51.4 52.7 B-32 62.2 50.0 ;48.6 50.0 ss.4 54.1 54.1 ss.4 B-33 ss.4 62.2 B-41 64 9 43.2 35.1 40.s 32.4 39.2 B-42 37.8 32.4 27.0 27.0 28.4 28.4 B-51 24.3 21.s 24.3 21.s 21.s 31.1 l B-52 24.3 24.3 52.7 23.0 73.0 93.2 B-61 27.0 28.4 27 9 21.6 24.3 23.0 24.3 l B-62 40.5 32.4 27.0 25.7 28.4 2s.7 B-7 33.8 29.7 27.0 21.6 31.1 27.0 C-2 55.4 50.0 4s.9 45.9 48.6 43.2 l C-3 s4.1 52.7 45.9 48.6 51.4 47.3 C-4 29.7 27.0 21.6 23.0 21.6 18.9 21.6 l C-5 27.0 27.0 18.9 18.9 16.2 14.9 16.2 C-61 43.2 31.1 28.4 29.7 28.4 29.7 C-62 3C.8 s9.s 24.3 32.4 32.4 27.0 3s.1 C-7 33.8 24.3 17.6 18.9 18.9 18.9 28.4 $$S 1 3 5 7 9 14 , 19 nutech 1138 234 D-1
SSP-01.001 APPENDIX D NORMALIZED SF ~~ 6 SAMPLE 1 2 3 4 5 6 7 A-2 52.3 si.1 84.1 70.5 8s.2 32.7 44.3 A-3 48.9 4s.s 51.1 50.0 7s.0 s9.1 36.4 g A-41 0 0 11 1.2 1.4 1.3 1.s A-42 0 .s .9 1.4 1.4 1.3 2.6 A-5 0 .s .8 1.3 1.4 1.4 1.3 A-61 0 0 1.1 1.4 1.4 .3 1.3 A-62 0 .7 10 18 1.6 13 1.s g B-1 42.6 56.8 61.4 61.4 67.0 60.2 54.s B-21 45.0
- 100, 90.9 83.0 4s.s s0.0 41.8 B-22 6s.9 75.0 88.6 81.8 90.9 64.8 68.2 l
B-31 15.9 18.2 27.3 31.4 20.0 21.8 30.9 B-32 1.0 2.7 2.4 1.9 1.6 1,6 1.9 I B-33 .9 1.3 2.0 4.1 21.1 19.1 9.2 l B-41 .e 2.7 2.e 4.7 2.> 2.0 4.e B-42 .5 2.4 2.8 4.s 3.5 4.8 3.0 B-51 .2 .9 1.7 2.6 2.1 1.9 3.4 y B-52 0 .9 1.7 2.6 1.7 1.s 1.7 d B-61 .7 11 1.7 2.3 2.1 2.s 1.6 l B-62 .1 .3 .7 2.0 1.2 .9 1.2 B-7 .3 .7 1.0 1.5 1.1 1.9 1.1 f-? 11.4 13.4 21.1 28.6 16.4 15.9 1s.9 C-3 s.7 11.6 16.8 23.6 14.1 19.5 12.7 g C-4 0 .9 1.5 1.1 1.3 1.3 l C-5 .3 .6 .9 1.s 1.2 1.6 1.2 C-61 .3 .5 .e 1.0 1.0 .l. 3 -C-62 .2 .4 .8 1.3 1.3 1.3 1.2 C-7 0 .4 .6 .9 .8 .9 6.8 g "EEs* 2 4 8 12 16 2P gj . g, g r 1133 235
APPENDIX D - NORMALIZED SF N N ONS TEST 3 UM 6 SAMPLE 1 2 3 4 5 A-2 3.2 1o.8 18.o 21.6 21.6 A-3 1.4 7.7 17.1 21.0 16.4 A-41 8.5 18.4 23.1 21.6 18.7 A-42 9.2 '16.4 21.0 19.0 19.0 A-5 9.2 20.3 23.6 20.0 19.3 A-61 4.3 21.0 24.3 23.3 20.7 A-62 16.4 21.0 20.o 20.0 24.6 B-1 1.1 6.9 14.9 22.3 19.7 B-21 4.7 11.3 17.4 19.7 11.8 B-22 2.7 11.2 17.7 20.0 14.4 B-31 2.2 10.5 15.1 17.7 13.8 B-32 4.8 13.8 19.7 18.4 8.2 B-33 s.3 11.3 17.4 13.1 17.4 B-41 6.3 8.7 8.5 9.2 B-42 2.8 7.9 8.7 8.2 8.0 B-51 3.8 6.9 10.2 9.8 11.8 B-52 4.3 6.9 8.9 12.3 9.2 B-61 s.9 8.2 10.0 8.5 10.3 B-62 7s 7.s 7.5 4.6 5.7 B-7 17.4 14.8 23.9 18.7 15.4 C-2 2.2 6.6 15.9 17.1 15.7 C-3 2.8 11.3 15.7 18.o 21.0 C-4 13.1 16.4 28.2 23.6 32.8 C-5 9.0 13.1 17.7 15.1 18.4 1138 .?36 C-61 82.0 80.3 88.5 85.3 67.2 C-62 100. 83.6 88.5 36.1 93.4 C-7 .5 21.6 26.9 23.3 19.7 NDM Sangte l TIMES 3 6 10 20 30
TEST 4 SSP-ol. col APPENDIX D - NORMALIZED SF CONCENTRATIONS 6 SAMPLE 1 2 3 4 5 A-2 .2 2.1 4.3 4.3 4.5 A-3 .1 .1 2.0 2.s 6.4 g A-41 0 .05 .2 .8 .9 l A-42 .o2 .1 .3 .6 .e A-5 .o7 .2 .4 .7 .s l A-61 .1 .1 .2 .6 .5 A-62 o .2 .4 .8 .7 g B-1 100. 98.9 95.5 95.5 95.5 l B-21 .1 .8 .4 .8 .9 B-22 1.6 67.1 27.3 83.0 59.1 l B-31 1 .6 2.6 5.6 s.5 B-32 o .3 ..e .8 .7 g B-33 .3 .3 1.0 3.0 2.8 l B-41 .1 .3 .5 .8 1.o B-42 .4 .6 .9 .8 .9 B-51 .3 .4 .8 .8 .7 B-52 .5 .3 .7 .7 .8 g B-61 .1. .3 .6 .7 .7 l B-62 .2 .2 .4 1.2 .6 B-7 .1 .4 .7 .5 .8 C-2 .2 .5 2.1 4.1 4.8 C-3 o .2 .s 1.0 1.1 g C-4 .3 .4 .7 .9 .9 l C-5 .3 .4 .8 .8 .9 C-61 1 1 .2 .6 .s C-62 .1 .3 .6 .s .7 C-7 .1 .4 .6 .6 .7 I SAMPLE TIMES 1 3 6 10 15 nutech g jj}} ?}7 D-4
APPENDIX D - NORMALIZED SF CONCENTRATIONS -- TEST 5 ssp-oi.ool 6 SAMPLE 1 2 3 4 5 A-2 1.8 18.4 44.2 46.3 A-3 1.7 16.3 35.3 52.6 s2.6 A-41 26.3 57.9 63.2 76.3 71.1 A-42 s7.9 6s.8 69.7 84 ? 80.3 A-5 31.6 39.8 s2.6 44.7 2.1 A-61 89.s 76.3 92.1 97.4 100. A-62 ss.3 78.9 44.7 47.4 80.3 B-1 1.1 7.1 21.6 34.2 42.1 B-21 13.7 30.0 36.8 43.7 41.6 B-22 s.o 22.6 37.4 44.2 4s.3 l B-31 3.3 21.6 38.9 51.0 47.4 B-32 11.1 22.9 29.5 32.6 36.8 B-33 18.1 42.1 49.0 52.6 50.0 B-41 3.6 45.3 so.o 50.0 44.7 B-42 1.8 11.8 12.2 is.2 12.9 B-51 16.8 30.0 50.0 31.6 31.6 B-52 1s.3 29.s 36.8 32.1 35.8 B-61 8.2 ,7.1 21.6 23.1 23.1 B-62 86.8 68.4 75.0 85.8 96.0 B-7 2.2 lo.o 11.7 20.8 C-2 1.1 12.0 26.3 35.8 34.7 C-3 .9 7.2 17.9 14.2
- 2. 0. 3 g
C-4 1.1 7.0 11.8 16.6 14.7 C-5 2.0 8.4 lo.s 14.o 13.4 C-61 48.8 60.s 54.0 73.7 89.s l C-62 s2.6 lo.s 15.0 17.1 19.2 I.2 C-7 2.0 lo.o c.2 7.s SAMPI E 1 3 6 10 15 nutech TIMES I 1138 23')
APPENDIX D - NORMALIZED SF CONCENTRATIONS TEST 6 SSP-01.00i 6 SAMPLE 1 2 3 4 5 A-2 30.7 44.8 57.2 55.9 57.2 A-3 38.6 42.0 49.7 55.2 59.3 A-41 35.9 36.5 48.3 50.3 51.0 A-42 39.3 47.6 49.0 53.8 49.7 A-5 41.4 51.o 53.1 57.2 56.6 A-61 30.0 44.1 48.3 51.o 51.0 A-62 43.5 44.8 53.8 58.o 53.8 B-1 33.1 34.5 49.0 31.0 58.0 l B-21 39.3 5'. 7 54.5 60.o B-22 43.5 35.1 ' 51.7 55.2 55.9 B-31 52.4 63.5 61.4 65.5 60.3 B-32 48.3 6 c,., 69.o e,.2 63.8 B-33 53.1 62.8 72.4 60.0 53.5 B-41 77.6 91.4 100. 98.3 81.0 B-42 69.0 67.2 62.o 77.6 60.3 B-51 46.2 51.7 60.0 59.3 55.9 B-52 51.7 53.1 55.9 62.8 52.4 g B-61 45.5 48.3 53.1 53.8 l B-62 38.o 40.0 45.5 47.6 46.2 B-7 50.3 54.5 51.0 3a.6 E= C-2 48.3 60.0 69.0 65.5 69.0 C-3 48.3 55.2 60.0 65 5 61.4 g C-4 49.o 55.9 59.3 60.7 60.7 l C-5 55.9 58.0 53.1 58.0 61.4 C-61 44.1 45.5 55.2 58.0 55.9 C-62 48.3 54.5 53.1 52.4 53.1 C-7 50.3 44.8 50.3 sl.7 g SAMPLE 1 3 6 10 15 ggg TIMES D-6 1133 23')
] APPENDIX D - NORMALIZED SF CONCENTRATIONS -- TEST 7 SSP-01.001 6 a f SAMPLE O 1 2 3 4 5 A-1 84.0 100. 94.0 99.0 100 100. ~ ~ A-2 32.8 83.0 87.0 96.0 100. 100. 4 B-1 20.8 '9.6 19.6 20.0 20.4 19.6 B-2 19.2 20.8 21.2 20.4 21.2 21.2 i C-1 19.6 18.4 19.2 16.8 . ] S"2" o 1 3 5 1 10 15 J 1 4 2 1138 240 i 1 b i l 1 ] flutech o_, = _. _ _.........}}