ML20066H778
| ML20066H778 | |
| Person / Time | |
|---|---|
| Site: | Midland |
| Issue date: | 11/16/1982 |
| From: | Gitnick B, Toth K NUS CORP. |
| To: | |
| Shared Package | |
| ML20066H714 | List: |
| References | |
| NUS-4242, NUDOCS 8211230349 | |
| Download: ML20066H778 (21) | |
Text
--
.j APPENDIX D Natural Gas Build Up In Selected Non-Safety-Related Structures At The Midland Site 2
, 8211230349 821116
' ~
~
=
- PDR ADOCK 05000329 A
I I
NUS-4242 NATURAL GAS BUILDUP IN SELECTED MIDLAND PLANT STRUCTURES Prepared for CONSUMERS POWER COMPANY by B.J.
Gitnick K. J. Toth November 16, 1982 Mu Approved:
S.
J.
Nathan, Manager Radiological Analysis Department Consulting Division NUS CORPORATION 910 Clopper Road Gaithersburg, Maryland 20878 I
I TABLE OF CONTENTS I
Section and Title Page No.
1.0 INTRODUCTION
1 2.0 TECHNICAL APPROACH 2
2.1 Break Mass Flow Rate Analysis 2
2.1.1 Guillotine Rupture 3
2.1.2 Smaller Breaks 4
2.1.3 Breaks Inside Structures 5
I 2.2 Gas Concentration at Intakes 6
2.3 Gas Buildup Inside Structures 7
3.0 RESULTS 9
4.0 REFERENCES
10 I
I I
I ii NUS COAPOAATION
I LIST OF TABLES Table No.
Title Page No.
BUILDING FREE VOLUMES AND 2-1 12 AIR INTAKE CHARACTERISTICS (4,5)
I 3-1 MAXIMUM CONCENTRATION OF NATURAL GAS INSIDE STRUCTURES FROM A POSTULATED PIPELINE BREAK 13 I
I I
I I
I I
l I
I I
I iii I
NUS CORPORATION t
LIST OF FIGURES Figure No.
Title Page No.
1-1 PIPELINE AND BUILDINGS CONSIDERED 14 2-1 TIME VERSUS BREAK FLOW RATE FROM PIPELINE NEAR THE MIDLAND PLANT 15 I
2-2 TIME VERSUS BREAK FLOW RATE FROM PIPELINE BREAK NEAR THE MIDLAND PLANT 16 I
I I
I I
I
~ e c- - ~
I I
1.0 INTRODUCTION
I This analysis was performed to determine the maximum buildup of natural gas in five buildings at the Midland site following a postulated rupture in the natural gas pipeline.
These five buildings are the permanent structures closest to the natural I
gas pipeline.
The pipeline runs adjacent to the plant's northern boundary with a spur extending onto the site, (See Figure 1-1).
The five buildings are:
o Evaporator building o
Combination Shop building o
Warehouse No. 1 o
Mechanic Shop building o
Condensate Return Pumphouse Consumers Power Company (CPCo) will install a computerized leak detection system and two isolation valves, one approxi-mately one half mile from the pressure reducing station, and the other in the flood plain, approximately 165 feet from the pressure reducing station (See Figure 1-1).
This system will shutoff the natural gas flow within 5 seconds following a guillotine rupture.
In addition, a flow limiter will be installed to restrict the steady flow to 20 lbs/sec.
A computer analysis was carried out to determine the mass flow rate resulting from a rupture of the natural gas line adjacent to the Midland site.
A basic Gaussian plume dispersion model was then used to determine cloud centerline concentration at I
the air intakes of the five buildings using 0.5 percentile directionally dependent meteorological conditions III.
I I
1 NUS CORPORATION
I HVAC characteristics were reviewed to develop a ventilation flow model for each of the structures identified above.
Based on these models differential equations for gas buildup were developed.
These equations were numerically integrated to I
determine the peak natural gas concentration in each building.
2.0 TECHNICAL APPROACH 2.1 Break Mass Flow Rate Analysis In order to evaluate the break flow rate as a function of I
time, a model for the natural gas line was developed for the COMPARE-MOD 1 code.I I COMPARE uses ideal gas parameters R and C
to descri.be the fluid.
Calculation of parameters for p
CPCo's natural gas constitutive mixture was performed using the procedures of the American Petroleum Institute Data Book.I I I
Separate models were used to evaluate the transient prior to isolation valve closure and after isolation valve closure.
Prior to isolation valve closure, a 25 node model (each node 528 feet long) was used.
Only the last six nodes were analyzed after isolation.
The geometric description of the pipeline is as follows:
OLATION MW No.1 ISOLATION VALVE No.2 AND FLOW LIMITER (river flood plain) w 2475 feet w n 2
LJ H
to Stewart road pressure 165 feet I
station - C,[
reducing evaporator ;
[ccinbination I
building l shop
......2 t.....
I 2
NUS CORPORATION
I I
At the beginning of the event, the gas pressure in the break node was taken as 390 psia (based on the known reservoir pressures and the pipeline lengths) and the temperature is 45 F based on CPCo operating experience.
For these conditions and assuming adiabatic, isentropic flow, I
the choked maximum flow rate can be analytically predicted as:
k+1 2
G
-k-1 2
=k I
k+1 9
- o p
c o
where I
k = the dimensionless ratio of the specific heats p
= density g
G
= Specific critical mass flow rate c
g
= gravity term c
p
= initial pipeline pressure g
I 2.1.1 Guillotine Rupture For a 6 inch pipe at these conditions the theoretical maximum flow rate is 220 lbs/sec and, in fact, COMPARE does predict approximately 210 lbs/sec immediately following the initiation of the transient.
I However, as the fluid is accelerated and the frictional losses become significant, the pressure loss upstream of the break increases.
These losses decrease the pressure and the gas density, causing the break flow rate to decrease.
I l l i
3 NUS CORPORATION
I I
After the initial break, a decompression wave travels uostream at sonic velocity.
Approximately two seconds after the break this decompression wave will have reached the flow limiter and shutoff valve and will continue to travel back to the reservoir.
Shortly after two seconds the fluid at the flow limiter has been accelerated to its maximum of 20 lbs/sec and the rate of the pressure change will decrease.
At five seconds, the isolation valve closes.
The remainder of the transient is simply the evacuation of the remaining gas in the pipeline after isolation.
By thirty one seconds, the break flow terminates.
Natural gas flow versus time is presented graphically in Figure 2-1.
I 2.1.2 Smaller Breaks The maximum concentration inside each structure is a function of the intake ventilation flow rate and the natural gas release rate.
The guillotine rupture, for example, may result in too rapid a release for structures whose ventilation flow is toe small to ingest the entire passing cloud.
Very small natural gas releases will be exhausted from the buildings as fast as they are ingested, resulting in only a modest buildup of natural gas.
Parametric studies of natural gas concen-tration vs. break flow size were performed to determine the worst break for each structure.
The break flow rate was conservatively held constant for the duration of the release.
From the COMPARE analysis performed above, it was determined I
that the total release of natural gas resulting from any size break is simply the inventory of gas in the pipeline plus the amount of gas passing the flow limiter up to isolation, minus I
I 4
NUS COAPORATION
I I
the small amount of gas which remains in the pipeline after I
the pressure has equalized.
The isolation time is a function II of break size.
CPCo supplied data for isolation time vs.
break flow rate is presented graphically in Figure 2-2.
The duration of the release may, therefore, be determined from the following relationship:
M + W*TISO (0)
-R T
=
rel Q
I where duration of the release (sec)
T
=
rel release rate (lb/sec)
Q
=
initial inventory of natural gas in M
=
I segment of pipeline beyond isolation valve (1bs) refill flow iato segment beyond isolation W
=
valve (1bs/sec)
TIS 0 (Q) isolation time (sec)
=
natural gas remaining in pipeline after R
=
blowdown (lbs)
The refill W, is constrained by the flow limiter to a maximum of 20 lbs/sec.
2.1.3 Breaks Inside Structures An eight inch natural gas line extends into the auxiliary boiler room and a ten inch pipeline branches into the high pressure boiler rooms.
These lines are downstream of both the I
pressure reducing station and the second isolation valve which I
5 j
NUS CORPORATION
I I
is located in the flood plain (See Figure 1-1).
The maximum I
pressure in these lines is 50 psia.
Using the critical flow formula in Section 2.1 above, the maximum flow rate from a 10 inch low pressure line is 71 lbs/sec.
Conservatively assuming that this maximum flow rate persists for the period up to isolation of the valve in the flood plain (5 seconds), the mass available for release is 400 lbs.
The release of this mass of gas will not result in an explosive concentration in I
either boiler room.
2.2 Gas Concentration at Intakes This analysis was performed using 0.5 percentile directionally dependent X /O 's developed for the Midland site.III The I
intake concentration is determined using a Gaussian dispersion model.
For intakes less than 165 feet (50 meters) from the release point, the concentration at the intake is conserv-atively assumed to be the density of the gas at ambient condi-tions.
The natural gas ingested with the intake air was limited such that no more than the amount of gas contained in the passing cloud was admitted into the building.
I The distance from the break to the air intakes was conserv-i atively chosen to be the shortest distance to the pipe for each intake.
For buildings with natural inleakage only, the distance used was the shortest distance from the pipe to the building of interest.
The assumed natural inleakage was conservatively taken as two air exchanges per hour.(6)
I Free volumes and air intake characteristics for the buildings considered are shown in Table 2-1.
This table is based on I
information provided in References 4 and 5.
l l I 1
6 NUS COAPORATION
I I
2.3 Gas Builduo Inside Structures The HVAC systems of the nearby structures were studied to determine the ventilation flui paths.
For those structures where the air can freely mix, a single volume model was used.
All other structures were modeled as two interconnected compartments, with one volume representing the potentially limiting (e.g.,
highest air exchange rate) subcompartment.
So~te of the structures have several independent HVAC systems.
For these structures, each independent HVAC zone was modeled separately.
The buildup of natural gas in the two compartments can be determined from the following equations:
I dC I9 I
I9L+901) C V
II 1
1 dt l
I dC I9 I + I9 I
- (902)C2 V
I2 L
1 2 dt 2
I X = cloud concentration at distance D (lbs/ cubic feet) y intake rate (cubic feet /sec) q7
=
flow rate from volume 1 to volume 2 (cubic feet /sec) g
=
g
= exhaust rate (cubic feet /sec) q0 V = volume (cubic feet)
I C = concentration in compartment (lbs/ cubic feet)
D = distance in feet (from postulated break site) l 7
I NUS CORPORATION
I I
These equations were solved numerically using a forth order Runga-Kutta-Gill procedure. The volumes of the buildings were reduced 10%
to 15%
account for equipment and interior structure.
The Evaporator building has several zones served by indepen-dent HVAC systems.
For modeling these zones have been divided into the Northern Zone (Control Room, Laboratory, and equip-ment areas), the Auxiliary Boiler area, and the Evaporator area which occupies the south half of the building.
The Evaporator building north has the subvolume with the highest air exchange rate.
The remaining spaces have been grouped together.
The Evaporator building south is taken as one, freely mixed volume with the two air intake locations.
The Auxiliary Boiler area is represented as a single volume I
structure.
The Combination Shop building and Warehouse No. 1 both consist of two noncommunicating compartments. (5)
The Combination Shop building and its High Pressure Boiler Room are both modeled as simple one volume structures.
The east side of the Warehouse No.
1 has as its second compartment the Auxiliary Diesel-I Generator Room.
This room has the highest air exchange rates of the compartments which draw air from the Warehouse No. 1 east. The Warehouse No. 1 west side is modeled as a simple one volume structure.
I The Mechanic Shop and the Condensate Return Pumphouse are also modeled as simple one volume structures.
I I
I e
g
~ m __ ~
t
3.0 RESULTS The maximum concentration of natural gas inside each of the five structures analyzed is presented in Table 3-1.
Based on these results four structures have potentially hazardous natural gas concentrations:
the Condensate Return Pumphouse, the Mechanic Shop, the High Pressure Boiler room and the Auxiliary Boiler room.
None of these structures are safety-related.
The Mechanic Shop and the Condensate Return Pumphouse have been analyzed elsewhere (8) and were shown to present no hazard to any safety-related structure.
The Auxiliary Boiler room has been constructed so as to result in minimal effects to the plant in the event of an internal explosion.
The south and west walls are block construction while the north and east walls are of butler construction which will act as blow-out panel.s and relieve to the north and east.
Therefore, in the event of an explosion in the Auxiliary Boiler room, the potential air shock and missiles generated will be directed to the north away from safety-related structures. The HVAC for the Auxiliary Boiler room operates at a reduced capacity ( < 40,000 cfm) when the boiler is not producing, therefore, the peak percent concentration during this time will be less than 8.5 lbs/sec shown in Table 3-1.
When the HVAC is operating at full capacity the boiler is also fired up and natural gas drawn in from a postulated pipeline break would ignite and result in a deflagration which would not be expected to cause damage to any safety-related structures.
j The high pressure boiler room is located on the southwest corner of the Combination Shop building.
The predicted frequency at which a natural gas cloud of 5% concentration reaches this building is 3.7x10-7.(9)
Gince the High Pressure Boiler and the HVAC for the Combination Shop building only operate one third cf the
- time, the expected frequency of ingesting a 5%
concentration is reduced to 1.2x10-i 9
l NLJS CORPORATION
I (3.7x10
/3=1.2x10-7)
When the Combination Shop building HVAC
-7 is not operating the expected in-leakage results in a peak concentration of approximately 0.7 volume percent (See Table 3-1) which is well below the flammable limit.
Again, when the HVAC is operating so is the boiler, and as with the Auxiliary Boiler room, natural gas drawn into the building from a postu-lated pipeline break would ignite and result in a deflagra-I tion.
Such a deflagration, combined with the low probability (1x10-7) that a flammable natural gas cloud concentration from a postulated pipeline break reaches the Combination Shop building air intake while the HVAC and boiler are operating, is not expected to pose a hazard to any safety-related structures.
I It should be noted that the criterion for flammability is a 5%
gas concentration which corresponds to the lower flammability limit for methane.
The criterion for a highly explosive mixture is 9.5% gas concentration, which is approximately the stoichiometric mixture of methane in air.
4.0 REFERENCES
1.
Report NUS-3963, "Off-Site Hazardous Chemical Definition Program for the Midland Plant",
Appendix B,
dated January 29, 1982, 2.
LA-7199-MS, " COMPARE-MODl",
R.
G.
- Gido, et.al.,
- LASL, MARCH 1978.
3.
" Technical Data
- Book, Petroleum Refining",
American I
Petroleum Institute, Third Edition 1976.
4.
Bechtel Drawing - Trailer Layout Annotated by Haldon Smith, August 6, 1982.
I I
I
-o-o~
I I
5.
Personal communication M.
- Ferens, Consumers Power I
Company, to K. Toth, NUS Corporation, August 20, 1982.
6.
1981 ASHRAE Fundamentals Handbook, ASHRAE, 1981.
7.
Personal communication M.
- Ferens, Consumers Power Company, to K. Toth, NUS Corporation, October 28, 1982.
I 8.
Report NUS-4235, " Midland Nuclear Power Station Methane Explosion Evaluation for Overpressure and Missile Effects at Safety-Related Structures," October 1982.
9.
Report NUS-4146, Rev.
1,
" Analysis of Flammable Concen-trations at the Midland Nuclear Plant from Natural Gas Pipeline Breaks," October 1982.
I I
I I
I I
I I
l I 11 NUS CORPOAATION
M M
M M
M M
M M
M m.
m m
m y
a e
m TABLE 2-1 BUILDING FREE VOLUMES AND AIR INTAKE CHARACTERISTICS (4,5)
AIR INTAKE CHARACTERISTICS Nearest Distance From Floy Rate Free V }ume 3
Building Compartment 10 FT Source Pipeline (FT)
(10 CFM)
Evaporator a) north storage tank cubicle 25 outside 50 3.2 b) remainder of 144 outside 10 14.6 north area c) auxiliary boiler 143 outside 10 40 area d) south area 1672 a) Outside 225 101 (No. Wall) b) outside 40 587
~"
(So.&W. Wall)
Combination Shop a) shop area 622 outside 50 23 b) high pressure 151 outside 10 75 boiler room Warehouse No. 1 a) east area 1995 outside 375 73 b) diesel room 8.7 from east area 35.5 l
c) west area 194 outside 250 8
Mechanic Shop 60 outside 25 2.2 l
l l
Condensate Return Pumphouse 27 outside 15 22
W M
M M
M M
M M
M M
M M
M M
M M
M TABLE 3-1 MAXIMUM CONCENTRATION OF NATURAL GAS INSIDE STRUCTURES FROM A POSTULATED PIPELINE BREAK Peak Building Compartment
% Concentration Comments Evaporator a) north storage tank 4.5 4.6 lbs/sec break flow cubicle b) remainder of 3.7 4.6 lbs/sec break flow north area c) auxiliary boiler 8.5 8.4 lbs/sec break flow area d) south area 1.0 guillotine break Combination a) shop area 2.8 11.0 lbs/sec break flow Shop b) high pressure 8.9/0.7*
guillotine 'reak a
boiler room H
Warehouse No. 1 a) east area 0.5 guillotine break b) diesel room 0.5 guillotine break c) west area 3.5 guillotine break Mechanic Shop 22.0 1.0 lbs/sec break flow Condensate 47.0 16.3 lbs/sec break flow Return Puuphouse When HVAC is not operating
m M
M M
M M
M M
M M
M M
M M
M M
ISOLATION VAVLE No. 2 IVER FLOOD PLAIN d'
s#
MECilAN SIIOP PRESSURE G
REDUCING STATION Q
cot-ENSATE l
l l
lS/
+
D
' URN PUMP b
IlOUSE NORTil COMBINATION Nbk" SilOP 1
BUILDING ISOLATION VALVE No.1 EVAPORATOR BUILDING one half mile from pressure reducing station towards Stewart road station SOUTil pres sure
~
boil ar room u
WAREllOUSE No. 1 WEST EAST FIGURE l-1 N
PIPELINE AND BUILDINGS CONSIDERED
r' i
i i
- -i i
i i
i f
i i
i i
i-i l
l l, i -
l, i i-i- i l
r 186 1se g
W cn 140 s
a; 120 5
Wg 100 g
d 80
{
ISOLATION VALVE CLOSURE w
i g
80 i
40 20 l
lI lI l
lI I
I I
I 3
I I
I 0
O 5
10 15 20 25 30 TIME (SEC)
FIGURE 2-1 TIME VERSUS BR"AK FLOW RATE FROM PIPELINE NEAR'THE MIDLAND PLANT l
g I
- I l
60 I
I 50 l
I 0
w (n
40 v
l t
w I
H i--
30 z
o I
H F-
<t l
_J 20 o
m E
H l
10
~
I I
O
'I'I'I'I'=
0 5
10 15 20 25 LEAKAGE FLOW RATE CLB/SEC) l r1 cure 2-2 TIME VERSUS BREAK FLOW RATE FROM PIPELINE BREAK NEAR THE I
MIDLAND PLANT l
16 I
l
~
l t
s I.
{
- ~*
,,[
h'
- I I-I I:
r,
,-,%,..~,
.,-y.,,.
ft I
I I
I I-
^
I
' 3o I.
hA Halliburton Company L
j