ML20056E262
| ML20056E262 | |
| Person / Time | |
|---|---|
| Site: | La Crosse File:Dairyland Power Cooperative icon.png |
| Issue date: | 07/31/1993 |
| From: | WESTINGHOUSE ELECTRIC COMPANY, DIV OF CBS CORP. |
| To: | |
| Shared Package | |
| ML19310D623 | List: |
| References | |
| NUDOCS 9308230096 | |
| Download: ML20056E262 (25) | |
Text
., - - _ _
Dairyland Natural Gas Hazards Westinghouse Class 3 l
i i
ANALYSIS OF TIIE POTENTIAL BLAST EFFECTS ON TIIE LACBWR REACTOR j
i BUILDING DUE TO A 16 INCil NATURAL GAS PIPELINE FAILURE i
i i
SUPPLEMENTAL ANALYSIS FOR FAST CLOSING VALVE i
July 1993
^
I i
i l
1 1
5
.)
i i
1 i
Energy Systems Business Unit Nuclear and Advanced Technology Division P.O. Box 355 Pittsburgh, Pennsylvania 15230 01993 Westinghouse Electric Corporadon All Rights Reserved l
i l
i I:0025.wpf:Ib-081093 9308230096 930812
~:i PDR ADOCK 05000409 m
P PDR
$2
]
I Dairyland Natural Gas llazards Analysis of the Potential lilast Effects on the LACBWR Reactor Building Due to a 16 Inch Natural Gas Pipeline Failure Summary This report summarizes the methodologies used by Westinghouse to evaluate the potential effects on i
the Lacrosse Boiling Water Reactor (LACBWR) Reactor Bailding which would be associated with a j
release of natural gas from a 16 inch diameter pipeline, located at its closest approach 240 feet away.
j Rese effects are primarily blast wave pressure due to a postulated unconfined vapor cloud explosion (UVCE) of the natural gas.
Re evaluation of the postulated unconfined vapor explosion provided an estimated maximum " worst case" over-pressure of 1.5 psi at the LACBWR Reactor Building, assuming a rapid depressurization of the gas piping,4,827 pounds in 4 seconds due to a double-ended pipe rupture. His analysis assumed stable meteorological conditions (Pasquill Gifford Class F) and a low wind speed,3.28 feet /sec.
(1 meter /sec.). His analysis assumes the release bchaves as a venical jet originating from a buried pipeline. He analysis does not account for a horizontal jet release since it was not considered credible from the buried pipe. The analysis also does not account for localized atmospheric turbulence; local turbulence could result in higher, near-ground level concentrations. However, such behavior is not included in most dispersion models. For such events, the ground level concentration would not be expected to exceed the elevated plume concentration predicted at the same downwind distance.
The results of this analysis, performed with the DEGADIS computer dispersion modeling code, indicated an over-pressure of approximately 1.5 psi at the Reactor Building. This is assuming a detonation at the 4.36 volume percent Lower Flammability Limit (LFL) cloud edge some 312 feet above the Reactor Building at its nearest approach.
Description of Release LACBWR is located in the Mississippi River Valley near Genoa, Wisconsin. The proposed 16 inch natural gas pipeline runs along the Mississippi River and is located to the north of the existing f.eactor Building and the proposed Combustion Turbine Building / Figure 1 shows the general site layout with the proposed natural gas line routing. The general terrain between the piping and the Reactor Building is flat, with few obstructions,i.e., trees, buildings, etc. The release of natural gas from the nearby pipeline is postulated to occur due to a pipe break associated with accidentally digging into the pipe.
De release conditions analyzed assumed that a pipe break of 204 square inches occurred in the buried piping and that the area of the excavation was equivalent to a ten foot diameter hole. This resulted in an initial vertical jet of natural gas. A ten foot diameter is a reasonable assumption; assuming a smaller diameter hole would result in higher velocities and increased vertical dispersion, assuming a larger hole would result in additional horizontal dispersion, due to increased initial release area.
l 1:0025.wpf:Ib-081093 1
l i
. -,.. ~.
Dairyland Natural Gas Hazards l
he gas release was estimated as follows. De natural gas pipeline rupture is assumed to occur at its closest approach to LACBWR (240 feet). The complete guillotine break results in two legs of 16 inch piping venting to the atmosphere at the same time. De section of piping to the Combustion Turbine l
Building is approximately 210 feet. We assume that check valves prevent the backflow of any gas from the combustion turbine. Bis dead leg depressurites from 400 psig to atmosphere. - We have j
calculated that 330 pounds of natural gas are released over a.6 second period. [
}"
Re other end of the pipeline runs approximately 1,080 feet to an automatic isolation valve at the LACBWR propeny boundary. Upon sensing a sudden loss in pressure in the pipeline, the isolation -
valve will close'. Bis valve closing takes approximately 4 seconds. [
j I
i i
i l
l
}"
Based on these assumptions, the total gas released is thus equal to this steady state release for four seconds (2,801 lbs.) plus the vented volume of the 1,080 foot pipe section (1,696 lbs.) and the j
210 foot pipe section (330 lbs.), or 4.827 lbs. De resultant average release rate over four seconds is j
therefore 1,207 lbs/sec.
' Personal Conununications. Nonhern States Power, March 1993.
t:0025.wpf:lb-081093 2
l a
l Dairyland Natural Gas Hazards This analysis assumed very stable meteorological conditions (Pasquill Gifford Class F) and a low wind speed,3.28 feet /sec. (1 meter /sec.) in order to maximize the concentration at the Reactor Building, the wind direction is assumed to be directly from the release point toward the Reactor Building.
Methods Description I
The a ialysis was performed using a two step approach; first the downwind dispersion of the natural gas was estimated using a Gaussian plume dispersion model, then the blast effects were estimated by calculating the TNT equivalent of the energy in the vapor cloud and estimating the blast effects based on measured TNT blasts.
J l
i ne DEGADIS, Version 2.1, PC computer, dispersion code was used to analyze the atmospheric i
dispersion of the natural gas. Development of this code has been sponsored by die United States Environmental Protection Agency (EPA), and the code is commonly used for dispersion modeling and emergency planning work.
Re Automated Resource for Chemical Hazard Incident Evaluation PC computer code, ARCHIE version 1.0, developed by the EPA, and the Federal Emergency Management Agency (FEMA), was used to perform the blast effects calculations. Rese calculations are performed consistent with die TNT equivalent method described by The Society of Fire Protection Engineers Handbook of Fire Protection Engineering.
Calculation Basis The primary consequence consideration was for the possible formation of a flammable cloud mixture between the 16 inch pipeline and the Reactor Building and the over-pressurization resulting from a postulated UVCE.
The following assumptions were used:
f The natural gas is assumed to be buoyant, with a molecular weight of 16.G4, significantly -
lower than air,(Molecular Weight = 28.8). The natural gas was assumed to be 100%
methane. The DEGADIS code accounts for the gas molecular weight.
The atmospheric stability is a Pasquill "F" Class very stable.
The terrain in the immediate area is level, rural and the wind is blowing toward the Reactor Building; a wind velocity of 3.28 feet /sec. (1 meter /sec.) was used.
The gas yield fraction is 0.03. (i.e., this is the fraction of gas energy which will contribute to the blast effects, reference The Handbook of Chemical Hazard Analysis Procedures.)
1:0025.wpf:Ib-081043 3
~~
j Dairyland Natural Gas Hazards 4
Elevated detonation of the vapor cloud at point closest to Reactor Building.
j i
Re vapor cloud was assumed to detonate rather th;n deflagrate; however such detonations j
are extremely improbable and require special conditions, such as panial confinement and an ignition source of sufficient energy to trigger a high combustion velocity through the l
f natural gas vapor cloud. Appendix A discusses the extreme improbability of an unconfined natural gas cloud detoaation and the conservadsm in the TNT equivalence i
method.
Calculation Procedure Dispersion Analysis
)
l ne DEGADIS code utilizes a Gaussian plume dispersion model, similar to many other consequence codes. De DEGADIS code, including the applicable equations,is described in DEGADIS (DENSE l
GAS DISPERSION) - VERSION 2.1, JUNE 1989 - Users Guide. [
]" Due to the uncertainties associated with actual releases and meteorological variations, the accuracy of the predicted results are typically estimated to be plus or minus a factor of two.
l De dispersion analysis provides the plume centerline pollutant concentration and elevation versus distance from the release location. De ccde also estimates the concentration profiles at selected elevations. [
]" He code calculates the peak, average concentration at a given location. [
l
]" For this analysis, the concentration level of concern was the LFL. [
1" LC ne trajectory of the LFL cloud was estimated by performing several dispersion analyses and determining the LFL concentration profiles at various elevations. For the release, the maximum downwind distance to the LFL is approximately 436 feet at a 640 foot elevation. [
1 i
1:0025.wpf:lb-081093 4
l Dairyland Natural Gas Hazards l'
l
]" We have very conservatively assigned the total weight of gas released (4,827 lbs.) to the LFL cloud directly over the Reactor Building. This assumes none of the gas is dispersed.
Whereas our analysis predicts no LFL concentration beyond 436 feet from the release point.
It should be stressed that this release would form a transient cloud, and these concentrations would
.I exist for only a short period, i.e., approximately equal to the release duration (4 seconds). Beyond that, dispersion effects will cause the puff / plume to diminish rapidly.
Due to the conservative assumptions presented above, it is likely that the mass of gas in the actual flammable cloud at any given instant will be much less than this value which is utilized to estimate the possible over-pressure resulting from the postulated UVCE.
l 111ast Effects Re blast effects were estimated by convening the energy in the vapor cloud into an equivalent mass of TNT then estimating the over-pressure versus distance, and associated damage based on known TNT measurements. He Handbook of Chemical Hazard Analysis Procedures, Appendix B, pages l-B-39 to B-45, discusses the application of this model to the ARCHIE code. His method is identical l
to that described in he SFPE Handbook of Fire Protection Engineering, Society of Fire Prctection l
Engineers. He energy available for ignition is calculated by:
E = a x H, x ny where:
l E
Blast Energy, kJ
=
Yield Factor 0.03 for methane a
=
Gas net heat of combustion, kJ/kg H,
=
i I
Mass of flammable vapor in cloud, kg m
=
r l
l 1:0025.wpf:lb-OS1093 5
l
Dairyland Natural Gas Hazards j
h t
i The TNT equivalent (kg) is then calculated by:
l Wm = E / 4200.
The over-pressure is then obtained from Figure 2-5.14 of The SFPE Handbook of Fire Protection Engineering by determining the scaled distance,i.e., the actual distance divided by the cube root of the TNT equivalent. Note, the ARCHIE code utilizes the following equation for determining the over-pressure, which was developed based on regression analysis of the data used to develop the above l
figure.
~
i X = (Wm) exp (3.5031 - 0.7241 In(0,) + 0.0398 x (In (0,))2)
'l where:
X
= distance to given over-pressure, ft.
i Peak over-pressure, psi O,
=
A postulated detonation at the edge of the LFL cloud could result in an overpressure of 1.5 psi. 'Ihis is based on the edge of the LFL being 312 feet from the top of the Reactor Building.
F Conclusion l
i The results of this analysis indicate a worst case UVCE at LACBWR to be 1.5 psi overpressure. The -
containment can withstand this overpressure. Moreover the probability of a complete rupture of the 16 inch gas line and ignition of the released gas is diminishingly small. We conclude that the risk
'l from a UVCE is diminishingly small and tolerable, i
- 0025.wpf:lb-081093 6
Dairyland Natural Gas Hazards
References:
1.
NRC Information Notice 91-63: Natural Gas Hazards at Fort St. Vrain Nuclear Generating j
Station, NRC Office of Nuclear Reactor Regulation, October 3,1991.
9 2.
John lacovino et. al., Development of Accident Scenarios for the Decommissioning of the Fon i
St. Vrain Nuclear Generating Station, Waste Management '92, Tucson, Arizona, P farch 1992.
3.
J. Havens and T. Spicer DESADIS, A Dispersion Model for Elevated Dense Gas Jet Chemical Releases, EPA-4501, April 1988.
j 4.
ARCHIE 1.0, Autorrred Resource for Chemical Hazard Incident Evaluation, U.S.
[
Environmental Protection Agency,1988.
t 5.
Tne SFPE Handbook of Fire Protection Engineering, Society of Fire Protection Engineers, April 1990, pp. 2-101.
6.
Handbook of Chemical Hazard Analysis Procedures, Federal Emergency Management i
Agency,1988.
l t
7.
J.E. Shepard, G.A. Melhem and P. Athens Unconfined Vapor Cloud Explosions: A New
'i Perspective, the International Conference and Workshop of Modeling and Mitigation the j
Consequences of Accidental Releases of Hazardous Material, AlCHE 1991.
l 8.
Regulatory Guide 1.91, Evaluations of Explosions Postulated to Occur on Transportation Routes l
Near Nuclear Power Plants, U.S. Nuclear Regulatory Commission February 1978.
{
9.
D.C. Bull and J.A. Manin, Explosions of Unconfined Clouds of Natural Gas American Gas Association Transmission Conference, St. Louis, Missouri, May 1977.
l 4
1:0025.wpf:1b.081093 7
r Dairyland Natural Gas Hazards
,/
Scale: 1 inch equals 150 feet Isoiction Volve j/
/
~
y '., -
l; ll
'. '. < l:
'/
D l:
!!A 4
..Il J
I
- 's
.t'-.
..v l'
/,
~
~~
N.'.,,J
.l t
i
- i i
si i
M
/
Q f
I
.. _.j 3
N d
l..
1 0o
/
N
/
2Yproech I
ucsa Noe H-N
/
e's
(
Contcinment
-O'g Building
'l N
Ac,.
l Q.,O j
s, N
{
_g
- g-O El
-l l
q N
h LUT I L
\\
w b- {'
L b(
.(
"~
L..
R0' ut r
l f
1 Combustion l
Turbine 7N O 4I Buriding l
l i
Figure 1. Lacrosse BWR Site and Proposed Natural Pipeline Routing WPF1286A:!Bo?;803 g
Dairyland Nr. rural Gas Hazards 1
l l
l 1
1 i
l i
i I
I J
).e 1:0025.wpf:lb4)S1093 9
Dairylands Power 2n Blast Domoge Enfimof e b5 5.5 J
- - --- - r -
~
7 l
2 j
5
+
l l
l 4.5 b-m
'\\
l m
E
\\_._
___-4-_
n 4
f O
d a
3.5 s
-3
\\
a 5
J
\\
z e
u
\\
i i
E E
3 3
\\.
__e_._....._
l E
5 1
E N
l E
8 o
N
}
j o
8
\\-
[
j M
G c
2.5 v
N g
n N
m o
g y
a m
2 4
N 3-N
N 5
1.5 l~dq- -
-+
1 L ---
}-
0.5 120 160 200 240 280 320 360 400 140 180 220 260 300 340 380 Distance f rom Explosion, feet N'1 -
,'e Dairyland Natural Gas llazards APPENDIX A NATURAL GAS EXPLOSION CONCERN AT THE LACROSSE HOILING WATER REACTOR S
[This appendix is based on general work performed by Mr. Stan Manin, a nationally known expert on unconfirmed vapor explosions. This work was done in support of our landmark analysis on Public Service of Colorado's Fon St. Vrain plant. It is equally applicable to LACBWR.
Introductory Remarks By far, the majority of cases of accidental or experimental, unconfined vapor-cloud explosions have failed to produce significant airblast effects. Notable exceptions - the Flixborough and Port Hudson events are about the only examples - are often held up to those who would dismiss the prospect lightly, as a reminder that experience shows that the threat of structural damage and/or personal injury / death cannot be ruled out. Dese events were special, however, because the first involved a i
massive release of cyclohexane, while the second was propane, and may have involved detonation initiated within a strong walled building. Rese two examples are therefore not directly appropdate to i
the circumstances of current consideration, that is, a natural gas release in open country, in the absence f
of enclosures where strong shocks might be generated. Both experience and theory-based analyses unconfined natural-gas clouds in air (lacking some peculiar, extreme, and very special conditions) say:
are not a blast damage / injury threat.
i The Concept of TNT Equivalence r
For years, attempts have been made to express the hazard of such uncondensed explosives as fuel-gas / air and fuel-gas / oxygen mixtures in terms of the energy release (or yield) equivalent to some standard condensed-phase explosive such as TNT. Re concept has some merit, as long as the state of a uniform mixture of the reacting gases is well known, and the explosion-to-target distance is large compared to the size of the cloud (i.e., the target is sufficiently remote that the extended source approximates s point source.) His is usually not the case, nor of practical interest, in accidental gaseous releases into the atmosphere. De amount of gas that may participate in the combustion reaction, soon enough to support a pressure wave,is commonly much less than any calculation of the time-averatied lower-flammability-limit envelope would s'uggest. Any estimate of yield is, therefore, very sensitive to variations in cloud geometry and eddy fine-structure within the cloud, which in turn depends on wind-speed profiles, gustiness, terrain roughness, etc., that are hardly knowable. Forecasts are quite beyond doing, with any confidence, at the present stage of development of the technology.
i Another very important source of variability in gas-phase explosions is the kind of explosion:
- 1. deflagration; 2. detonation; 3. thermal explosion. In the order given, the explosion energy yields increase from a small fraction of that potentially available (order of 0.01) to a large fraction (approaching 100%). For natural-gas / air, we can eliminate any prospect of a thermal explosion, l
1:0025.wpf:Ib-081093 11
)
r l
Dairyland Natural Gas Hazards because alkane hydrocarbons are stable in air at ambient temperatures; and detonations can occur only I
in very special circumstances, such as in channels with high Reynolds' Numbers or in enclosures with strong walls. In pure methane / air, detonations can be mied out. Higher hydrocarbons are more subject to detonation, unsaturated hydrocarbons, even more, and certain chemical additions to the l
hydrocarbon stmeture increase this tendency dramatically. This explains, in part, the differences in experience-based energy yields of explosive vapors and gases. Alkane hydrocrebons are in the group i
having yields of 0.03. His would be appropriate for deflagration of natural-gas / air. He higher yields are for ethers, nitrated paraffins, olefins, and acetylenics. This is entirely consistent with the findings of Braisie and Simpson (1968). In the incidents they surveyed, including a wide range of hydrocarbon / air explosians, yield values were commonly less than 0.M. Deflagrations having yields as high as 0.1 require mixtures containing the more reactive components, listed above, or special circumstances. Analysts often use this more conservative value to cover any and all eventualities, to 1
be very conservative. It would seem, however, inappropriate to do so here, unless it can be shown that detonations are both possible and probable. This issue is addressed in the following section.
The Possibility of Detonation 3
here is considerable, quality evidence that detonation cannot be directly.uiated in a natural-gas / air mixture with anything less than heroic efforts (e.g., kilograms of high explosive or a strong shock s
emanating from a pipe; see MIT-GRI LNG Workshop,1982). Further, there is considerable doubt that, once initiated, a detonation could sustain itself in practical situations of nonuniform mixing. nat may have happened at Port Hudson because of special circumstances, but altemative explanations for j
the damage have been offered. It must be remembered that the Port Hudson event was fueled by propane not natural gas.
Results of all the tests conducted to the 1982 date of the MIT-GRI Workshop, to study the detonability of unconfined vapor clouds, indicated the following:
No detonations are possible in an unconfined stoichiometric mixture of pure methane vapor and air, even under the initiating influence of 2 kg. explosive Sustained detonations are possible in unconfined vapor clouds containing stoichiometric fuel air mixtures when the fuel vapor contains methane and propane in the molar ratios of 60/40,70/30, and 85/15. Above 85% (molar) methane concentrations, no sustained detonations have been observed.
Propagating detonations (from a pipe) have been sustained in unconfined mixtures of methane / propane / air (stoichiometric) for methane volume fractions of less than 85% in the fuel vapor.
1:0025.wpf:Ib-081093 12 1
a
7 Dairyland Natural Gas Hazards it is evident from these results that relatively small fractions of heavier hydrocarbons, such as ethane and propane, in the fuel vapor increase the propensity for the detonation of unconfined vapor / air mixtures. This phenomenon is of great importance in evaluating the detonability consequences of natural-gas releases in the form of late-timt. boil-off from LNG, but usually does not apply to releases from pipelines, because gaseous releases are usually not rich in higher hydrocarbons.
As a practical matter, as we understand the circumstances of the possible release of natural gas at the Lacrosse BWR site, we conclude that the likelihood of a detonation is an incredible event (less than once in a million years).
Flame Generated Pressure In view of the large uncertainties in any attempt to establish an explosive-energy yield, an arguably better approach to setting an upper bound on airblast pressures from unconfined gas-phase explosions is one based on the theoretical development of Kuhl, Kamel, and Oppenheim (1974). A flame propagating through a premixed gas of uniform composition and properties pushes unburned gas ahead of its flame front, causing the pressure to rise. In the free field, the flame propagates in three dimensions as a sphere. The compression is less than it would be if the flame were constrained by a nonyielding surface (e.g., the ground or a wall) to propagate with only two degrees of freedom, and still less than situations in which expansion is constrained to just one dimension. In any case, if the flame has a fixed speed - neither accelerating nor decelerating-it will maintain a fixed pressure rise in the unburned gas it pushes ahead. Unless the speed of the flame front is up around 40 misec., the overpressure just ahead of the flame cannot exceed one atmosphere (14.7 psig). To shock up and (possibly) detonate requires much higher speeds. 'Ihe required "run-up speed" for natural gas in air is not known, but certainly approaches - and may exceed - Mach I (Lee,1977), that is, alx)ut 330 misec., which is patently unattainable in the open. Typical burning speeds for methane are 7 m/sec.,
in quiescent conditions. Turbulence may increase that by a factor of four (to, say,30 misec., see Kanury,1975, pp. 303, 304).
It is important to note that this air blast will not " shock up" - that is, the pressure wave will have a finite rise time, and it is therefore, inappropriate to use a doubling rule for reflected pressure loading of walls. Burning speeds would need to exceed 230 misec. to develop any significant shocking of the pressure wave (Lee,1977). The pressure rise time will be comparable to the time taken by the flame to transit the near-stoichiometric (and richer) portion of the cloud. If this were on the order of 100 ft.,
the time could be a second or two.
- 0025.wpf:Ib-081093 13
Dairyland Natural Gas Hazards Structural Loading and Response Side-on overpressures have little to do with structural respense in this case. These buildings are more likely to respond to drag loading than to diffraction (or crush) loading'. If the rise time is long compared to diffraction times snd cleving times (and for leaky structures, long compared to inside-to-outside pressure equalization times), then there would be little net force acting on a wall except for dynanuc (i.e., drag) forces. An air-tight box would be subjected to crushing forces, but such structures j
are inherently sinng - even within the cloud, no pressures capable of crushing damage would be expected. De pressure-pcise duration (and the accompanying drag-phase flow) is undoubtedly long compared to the natural period (roughly 1/2 sec.) of the kcactor Building. Pressure durations from uncondensed explosives are i to 2 orders of magnitude longer than ideal explosions (see Bodurtha.
1980, p.108); therefore, the dynamic pressure pulse can be compared to a wied gust. Accordingly, it is appropriate to consider the wind force of the dynamic pressure wave to be the principal mode for mechanical damage, and to compare its magnitude to the design value.
In ideal blast waves with a true shock front, peak dynamic pressures can be appreciable in magaitude compared to peak (side-on) overpressures. For example, a 4 psi peak (side-on) overpressure is accompanied by 0.4 psi peak dynamic pressure. That represents a particle velocity of 200 ft/sec. or 136 mph. A 6-psi overpressure is accompanied by 0.8 psi dynamic pressure, corresponding to a 260 ft/sec. particle velocity, representing a 177 mph wind gust. A useful rule-of-thumb: The particle velocity associated with a 1-psi overpressure is about 50 ft/sec., and increase (or decreases) in linear proportion to the overpressure. Rus, a 300-mph air blast (440-ft/sec. particle velocity) accompanies a peak overpressure of 8.8 psi.
Alkane / air deflagrative explosions, by contrast, do not generate shocks. Clearly, therefore, the peak particle velocity driven by the explosion cannot exceed the flame speed driving it As a result, flame speeds in quiescent, premixed methane / air mixtures would be incapable of driving wind gusts above 16 to 18 mph at the edee of the explodine cloud! Allowing a factor of four for turbulence, the limit would still be a factor of six to seven below the designed wind force for the Reactor Building, and would rapidly dissipate with distance from the explosion.
' For a fuller development of the mechanisms of airtlast loading and responses of structures (and definitions of such terms as dran. diffraction, and clearine times), refer to Glasstone and Dolan.
1977.)
1:0025.wpf:Ib-081093 14
V Dairyland Natural Gas Hazards Conclusions These are our basic conclusions:
1.
The likelihood of a detonadon is negligibly small.
2.
The air blast from a deflagrative explosion would exhibit a weak pressure pulse having a long rise time compared to any characteristic building-response time. 'Ihus, the response would be entirely like a response to natural wind forces.
3.
Responses to side-on overpressures, if any, would be limited to window / door / light-panel breakage or deformation, but (except for glass window breakage) this is expected only from overly conservative estimates of yield.
4.
Dynamic pressures would be inconsequential relative to the 360-mph design wind loading for LACBWR.
References Allan, D.S., and P. Athens,1968:
- Influence of Explosions on Design," CEP Tecimical Manual, Loss Prevention Vol. 2, AIChE, New York.
Bodurtha, F.T.,1980: " Industrial Explosion Prevention and Protection," McGraw-Hill, New York.
Brasie, W.C., and D.W. Simpson,1968: AIChE, Loss Prevention, Vol. 2, pp.91-102.
Brode H.L.1955: " Numerical Solutions of Spherical Blast Waves," J. Applied Physics,26,766.
6 Glasstone, S., and P.J. Dolan,1977: "'Ihe Effects of Nuclear Weapons," Third Edition, US DoD and ERDA, Gov't Printing Office.
Kanury, A.M.,1975: " Introduction to Combustion Phenomena," Gordon and Breach, New York.
Kuhl, A.L., M.M. Kamel and A.K. Oppenheim,1974: " Pressure Waves Generated by Steady Flames "
14th (Int'l) Combustion Symposium, The Combustion Institute, Pittsburgh.
Lee, J.H.,1977: " Initiation of Gaseous Detonation," Vol. 28, An. Review of Physical Chemistry.
Lee, J.H., et al.,1977: " Blast Effects from Vapor Cloud Explosions," paper presented at the 1977 AIChE Loss Prevention Symposium, March 1977. Houston, TX.
- 0025.wpf:Ib-081093 15 m
l
r Dairyland Natural Gas Hazards 4
i Lind, C.D.,1975 : "What Causes Unconfined Vapor Cloud Explosions?" CEP Technical Manual.
Loss Prevention Volume 9_,.American Institute of Chemical Engineers, New York.
j Raj, P.K. (ed),1982: "MIT-GRI LNG Safety and Research Workshop, Volume Ill: LNO Fires -
l Combustion and Radiation," GRI 82/0019.3, Gas Research Institute, Chicago.
i Strehlow, R.A., and W.E. Baker,1976: "The Characterization and Evaluation of Accidental i
Explosions," Prog. Energy Combustion Science 2,27.
j t
Stull, D.R.,1977: " Fundamentals of Fire and Explosion," AIChE Monograph Serics 23, No.10 American Institute of Chemical Engineers, New York.]"
i E
i I
. l l
l t:0025.wpf:lb.081093 16 l
l
- .-.,---n-----,,,,,.
.n,..,~
,,r.-.-
Dairyland Natural Gas 11azards Westinghouse Class 3 ANALYSIS OF TIIE POTENTIAL llLAST EFFECTS ON TIIE LACIlWR REACTOR IlUILDING DUE TO A 16 INC11 NATURAL GAS PIPELINE FAILURE SUPPLEMENTAL ANALYSIS FOR FAST CLOSING VALVE July 1993 l
l-I Energy Systems Business Unit Nuclear and Advanced Technology Division P.O. Box 355 Pittsburgh, Pennsylvania 15230 01993 Westinghouse Electric Corporation All Rights Reserved td)025.wpf:lb-081093 l
~
?
Dairyland Natural Gas Hazards i
Analysis of the Potential lilast Efrects on the LACllMI Reactor Iluilding c
Due to a 16 Inch Natural Gas Pipeline Failure Summary This report summarizes the methodologies used by Westinghouse to evaluate the potential effects on the Lacrosse Boiling Water Reactor (LACBWR) Reactor Building which would be associated with a release of natural gas from a 16 inch diameter pipeline, located at its closest approach 240 feet away.
Rese effects are primarily blast wave pressure due to a postulated unconfined vapor cloud explosion (UVCE) of the natural gas.
f The evaluation of the postulated unconfined vapor explosion provided an estimated maximum " worst case" over-pressure of 1.5 psi at the LACBWR Reactor Building, assuming a rapid depressurization of the gas piping 4,827 pounds in 4 seconds due to a double-ended pipe rupture. Tius analysis assumed stable meteorological conditions (Pasquill Gifford Class F) and a low wind speed,3.28 feet /sec.
l (1 meter /sec.). Bis analysis assumes the release behaves as a verticaljet originating from a buried pipeline. De analysis does not account for a horizontal jet release since it was not considered credible from the buded pipe. De analysis also does not account for localized simospheric turbulence; local turbulence could result in higher, near-ground level concentrations. However, such behavior is not
)
included in most dispersion models. For such events, the ground level concentration would not be expected to exceed the elevated plume concentration predicted at the same downwind distance.
The results of this analysis, performed with the DEGADIS computer dispersion modeling code, indicated an over-pressure of approximately 1.5 psi at the Reactor Building. His is assuming a detonation at the 4.36 volume percent Lower Flammability Lanit (LFL) cloud edge some 312 feer above the Reactor Building at its nearest approach.
I Description of Release LACBWR is located in the Mississippi River Valley near Genoa, Wisconsin. The proposed 16 inch natural gas pipeline runs along the Mississippi River and is located to the north of the existing Reactor Building and the proposed Combustion Turbine Building? Figure I shows the general site layout with the proposed natural gas line routing. The general terrain between the piping and the Reactor Building l
J is flat, with few obstructions,i.e., tree.s buildings, etc. The release of natural gas from the nearby pipeline is postulated to occur due to a pipe break associated with accidentally digging into the pipe.
The release conditions analyzed assumed that a pipe break of 2(M square inches occurred in the buried piping and that the area of the excavation was equivalent to a ten foot diameter hole. His resulted in
]
an initial vertical jet of natural gas. A ten foot diameter is a reasonable assumption; assuming a smaller diameter hole would result in higher velocities and increased vertical dispersion, assuming a f
larger hole would result in additional horizontal dispersion, due to increased initial release area.
1:0025.wpf:Ib-081093 1
i i
i
}
Dairyland Natural Gas Hazards i
i ne gas release was estimated as follows. De natural gas pipeline rupture is assumed to occur at its l
closest approach to LACBWR (240 feet). The complete guillotine break results in two legs of 16 inch piping venting to the atmosphere at the same time. The section of piping to the Combustion Turbine
{
Building is approximately 210 feet. We assume that cheek valves prevent the backflow of any gas from the combustion turbine. This dead leg depressurizes from 400 psig to atmosphere. We have l
calculated that 330 pounds of natural gas are released over a.6 second period. [
i
}"
1 The other end of the pipeline runs approximately 1,080 feet to an automatic isolation valve at the LACBWR property boundary. Uptm sensing a sudden loss in pressure in the pipeline, the isolation valve will close. This valve closing takes approximately 4 seconds. [
l
(
i I
I f
,1 1"
Based on these assumptions, the total gas released is thus equal to this steady state release for four seconds (2,801 lbs.) plus the vented volume of the 1,080 foot pipe section (1,696 lbs.) and the 210 foot pipe section (330 lbs.), or 4,827 lbs. The resultant average release rate over four seconds is i
therefore 1,207 lbs/sec.
l i
j i
' Personal Cosmnunications, Northern States Power, March 1993, a
1:0025.wpf:Ib-081093 2
i e
~=r-.i
--++r'
?-P
~
d
-M
i Dairyland Natural Gas Hazards i
Ris analysis assumed very stable meteorological conditions (Pasquill Gifford Class F) and a low wind speed. 3.28 feet /sec. (1 meter /sec.) In order to maximize the concentration at the Reactor Building, the wind direction is assumed to be directly from the release point toward the Reactor Building.
Methods Description De analysis was pe formed using a two step approach; first the downwind dispersion of the natural gas was estimated using a Gaussian plume dispersion model, then the blast effects were estimated by calculating the TNT equivalent of the energy in the vapor cloud and estimating the blast effects based on measured TNT blasts.
The DEGADIS, Version 2.1, PC computer, dispersion code was used to analyze the atmospheric dispersion of the natural gas. Development of this code has been sponsored by the United States Environmental Protection Agency (EPA), and the code is commonly used for dispersion modeling and j
emergency planning work.
The Automated Resource for Chemical Hazard Incident Evaluation PC computer code, ARCHIE version 1.0 developed by the EPA, and the Federal Emergency Management Agency (FEMA), was used to perform the blast effects calculations. Dese calculations are performed consistent with the i
TNT equivalent method described by The Society of Fire Protection Engineers Handbook of Fire t
Protection Engineering.
Calculation Basis he primary consequence consideration was for the possible formation of a flammable cloud mixture between the 16 inch pipeline and the Reactor Building and the over-pressurization resulting from a pos'ulated UVCE.
The following assumptions were used:
The natural gas is assumed to be buoyant, with a molecular weight of 16.04, significantly lower than air. (Molecular Weight = 28.8). The natural gas was assumed to be 100%
methane. The DEGADIS code accounts for the gas molecular weight.
The atmospheric stability is a Pasquill F' Class very stable.
The terrain in the immediate area is level, rural and the wind is blowing toward the Reactor Building; a wind velocity of 3.28 feet /sec. (1 meter /sec.) was used.
he gas yield fraction is 0.03. (i.e., this is the fraction of gas energy which will contribute j
to the blast effects reference De Handbook of Chemical Hazard Analysis Procedures.)
1:0025.wpf:lb-081093 3
L l
Dairyland Natural Gas Hazards l
Elevated detonation of the vapor cloud at point closest to Reactor Building.
]
l The vapor cloud was assumed to detonate rather than deflagrate; however such detonations are extremely improbable and require special conditions, such as partial confinement and l
an ignition source of sufficient energy to trigger a liigh combustion velocity through the natural gas vapor cloud. Appendix A discusses the extreme improbability of an l
unconsned natural gas cloud detonation and the conservatism in the TNT equivalence j
method.
Calculation Prec4re j
-1 I
Dispersion Analysis I
)
ne DEGADIS code utilizes a Gaussian plume dispersion model, similar to many other consequence i
codes. The DEGADIS code, including the applicable equations,is described in DEGADIS (DENSE
~
GAS DISPERSION) - VERSION 2.1, JUNE 1989 - Users Guide. [
]
]" Due to the uncertainties associated with actual releases
~
and meteorological variations, the accuracy of the predicted results are typically estimated to be plus j
or minus a factor of two.
ne dispersion analysis provides the plume centerline pollutant concentration and elevation versus distance from the release location. He code also estimates the concentration profiles at selected elevations. [
]" The code calculates the peak, average concentration at a given location. [
f i
}" For *is analysis, the concentration level of concern was the LFL [
1" I
l 1"
The trajectory of the LFL cloud was estimated by performing several dispersion analyses and
]
determining the LFL concentration profiles at various elevations. For the release, the maximum downwind distance to the LFL is approximately 436 feet at a 640 foot elevation. [
l t:0025.wpf:Ib-081093 4
i l'
l
i 1
l Dairyland Natural Gas Hazards i
l l
}" We have very conservatively assigned the total weight of gas released (4,827 lbs.) to the LFL cloud directly over the Reactor Building. This assumes none of the gas is dispersed.
Whereas our analysis predicts no LFL concentration beyond 436 feet from the release point.
It should be stressed that this release would form a transient cloud, and these concentrations would exist for only a short period, i.e., approximately equal to the release duration (4 seconds). Beyond that, dispersion effects will cause the puff / plume to diminish rapidly.
Due to the conservative assumptions presented above, it is likely that the mass of gas in the actual fb mmable cloud at any given instant will be much less than this value which is utilized to estimate the possible over-pressure resulting from the postulated UVCE.
Blast Effects i
The blast effects were estimated by converting the energy in the vapor cloud into an equivalent mass l-of TNT then estimating the over-pressure versus distance, and associated damage bas'ed on known TNT measurements. The Handbook of Chemical Hazard Analysis Procedures, Appendix B, pages B-39 to B-45, discusses the application of this model to the ARCHIE code. This method is identical to that described in The SFPE Handbook of Fire Protection Engineering, Society of Fire Protection Engineers. The energy available for ignition is calculated by:
E = a x H, x m, where:
I E
= Blast Energy, kJ j
a
= Yield Factor. 0.03 for methane H,
= Gas net heat of combustion, kJ/kg m,
= Mass of flammable vepor in cloud, kg 1:0025.wpf:lb-081093 5
i i
l l
I P
l l
+
l Dairyland Natural Gas Hazards h
De TNT equivalent (kg) is then calculated by:
Wm = E / 4200.
De over-pressure is then obtained from Figure 2-5.14 of De SFPE Handbook of Fire Protection Engineering by deternuning the scaled distance, i.e., the actual distance divided by the cube root of the TNT equivalent. Note, the ARCHIE code utilizes the following equation for determining the over-pressure, which was developed based on regression analysis of the data used to develop the above i
figure.
X = (Wm)"5 exp (3.5031 - 0.7241 In(0,) + 0.0398 x (In (0,))')
r where:
X
= distance to given over-pressure, ft.
l 0,
- Peak over-pressure, psi i
A postulated detonation at the edge of the LFL cloud could result in an overpressure of 1.5 psi. This is based on the edge of the LFL being 312 feet from the top of the Reactor Building.
t Conclusion The results of this analysis indicate a worst case UVCE at LACBM R to be 1.5 psi overpressure. The
{
containment can withstand this overpressure. Moreover the probsility of a complete rupture of the
{
16 inch gas line and ignition of the released gas is diminishingly small. We conclude that the risk l
from a UVCE is diminishingly small and tolerable.
)
i l
i 1
i
- 0025.wpf:lb-081093 6
t l
Dairyland Natural Gas Hazards l
References:
1.
NRC Information Notice 91-63: Natural Gas Hazards at Fort St. Vrain Nuclear Generating l
Station, NRC Office of Nuclear Reactor Regulation October 3,1991.
I t
2.
John lacorino et. al., Development of Accident Scenarios for the Decommissioning of the Fort
[
St. Vrain Nuclear Generating Station, Waste Management '92 Tucson, Arizona, March 1992.
j 3.
J. llavens and T. Spicer, DEGADIS, A Dispersion Model for Elevated Dense Gas Jet Chemical l
Releases, EPA-4501, April 1988.
4.
ARCHIE 1.0, Automated Resource for Chemical Hazard Incident Evaluation, U.S.
Environmental Protection Agency,1988.
5.
Ti,? SFPE Handbook of Fire Protection Engineering, Society of Fire Protection Engineers. April 1990, pp. 2-101.
6.
Handbook of Chemical Harard Analysis Procedures, Federal Emergency Management f
Agency,1%.
f i
7.
J.E. Shepard, G.A. Melhem and P. Athens, Unconfined Vapor Cloud Explosions: A New l
Perspective, the International Conference and Workshop of Modeling and Mitigation the j
Consequences of Accidental Releases of Hazardous Material, AICHE 1991.
i 8.
Regulatory Guide 1.91, Evaluations of Explosions Postulated to Occur on Transportation Routes Nel Nuclear Power Plants, U.S. Nuclear Regulatory Commission Febniary 1978.
9.
I LC. P -
d J. A. Martin, Explosions of Unconfined Clouds of Natural Gas, American Gas Asst
.e.r a P -mission Conference, St. Louis, Missouri, May 1977.
1
.)
i I
t:0025.wpf:lb-081093 7
t Dairvland Natural Gas Hazards
./
Scale: 1 inch equals 150 fee:
l-1 isolction Volve
[
/
f, l',
?k
/ e'
/,,.j
/:,l:/
iik l!
ii ti
! L.. 5:
,/, ~ ~~j!
/
I
~
' '..) l l
\\
i
- s
~.
u
.' 8
/ ~Z' l
_;l.:
l
~
U@
D i
l i
i i
/
us.
.m 2 M it tJCEWR N.*** /
/
k Containment i
'Q'-b 4
sutiding I
{
'h z.
N r
U7 j
l m
ILP i o"
.l I J j'
t h{
y-M2-1
(
J. ~
f0
\\\\\\
\\
1 I
J u
Combustion
< M S
'.....=
nnlL Turbine
'"" ~
i J
Figure 1. Lacrosse BWR Site and Proposed Natural Pipeline Routing WPF1266A1B07 801 8
r r
Dairyland Natural Gas Hazards l
6 f
t f
i
. i
-i, i
t b
4 h
i
- i u
i r
l t
- i
\\
i I
j j
i i
I j.,
l t:0025.wpf:Ib-081093 9
l l
i
=
-+n
+
r o-
e Dairylands Power 8y Blast Domoge Estimate 5.5 2
b 4
~
g 5
g
~
\\
4.5 21 m
Ea 4
n h
o
.G m
J o_
3.5
{
t 5
si Z
m u
a 3
3
[
g T
U t
O a
u a
c 2.5 i
55
\\
k C...
N a-2 s
5 N
F.T N
5-N
.o
.5 m x N
~
1
- - l '~
4 0.5 120 160 200 240 280 320 360 400 140 180 220 260 300 340 380 Distance from Explosion, f eet -
-__m._._._-_.-t.m_m_m._______m__._____m.______m_.___.-_____mmv.m_,m__m%
,er,%.,m,i.,-.,.c.,4,_,,..%.,
..mm.,%%,.,we,_
,,.y7
.,.. -.,,,_.,.y
Dairyland Natural Gas llazards APPENDIX A NATURAL GAS EXPLOSION CONCERN AT TIIE LACROSSE IlOILING WATER REACTOR
[This appendix is based on general work performed by Mr. Stan Martin, a nationally known expert on unconfirmed vapor explosions. This work was done in support of our landmark analysis on Public Service of Colorado's Fort St. Vrain plant. It is equally applicable to LACBWR.
Introd ctory Remarks By far, the majority of cases of accidental or experimental, unconfined vapor-cloud explosions have failed to produce significant airblast effects. Notable exceptions - the Flixborough and Port Hudson events are about the only examples - are often held up to those who would dismiss the prospect lightly, as a reminder that experience shows that the threat of structural damage and/or personal injury / death cannot be ruled out. These events were special, however, because the first involved a massive release of cyclohexane, while th: second was propane, and may have involved detonation initiated within a strong walled building. These two examples are therefore not directly appropriate to the circumstances of current consideration, that is, a natural gas release in open country, in the absence of enclosures where strong shocks might be generated. Both experience and theory-based analyses say: unconfined natural-gas clouds in air (lacking some peculiar, extreme, and very special conditions) are not a blast damage / injury threat.
The Concept of TNT Equivalence For years, attempts have been made to express the hazard of such uncondensed explosives as fuel-gas / air and fuel-gas / oxygen mixtures in terms of the energy release (or yield) equivalent to some standard condensed-phase explosive such as TNT. The concept has some merit, as long as the state of a uniform mixture of the reacting gases is well known, and the explosion-to-target distance is large.
compared to the size of the cloud (i.e., the target is sufficiently remote that the extended source approximates a point source.) This is usually not the case, nor of practical interest, in accidental gaseous releases into the atmosphere. The amount of gas that may participate in the combustion reaction, soon enough to support a pressure wave, is commonly much less than any calculation of the j
I time-averat'ed lower-flammability-limit envelope would s'uggest. Any estimate of yield is, therefore, very sensitive to variations in cloud geometry and eddy fine-structure within the cloud, which in turn depends on wind-speed profiles, gustiness, terrain roughness, etc., that are hardly knowable. Forecasts are quite beyond doing, with any confidence, at the present stage of development of the technology.
Another very important source of variability in gas-phase explosions is the kind of explosion:
- 1. deflagration; 2. detonation; 3. thermal explosion. In the order given, the explosion energy yields increase from a small fraction of that potentially available (order of 0.01) to a large fraction (approaching 100%). For natural-gas / air, we can eliminate any prospect of a thermal explosion, t:0025.wpf:lb481093 11
Dairyland Natural Gas llazards because alkane hydrocarbons are stab!c in air at ambient temperatures; and detonadons can occur only in very special circumstances, such as in channels with lugh Reynolds' Numbers or in enclosures with strong walls. In pure methane / air, detonations can be ruled out. liigher hydrocarbons are more subject to detonation, unsaturated hydrocarbons, even more, and certain chemical additions to the hydrocarbon structure increase this tendency dramatically. This explains, in part, the differences in experience-based energy yields of explosive vapors and gases. Alkane hydrocarbons are in the group having yields of 0.03. This would be appropriate for deflagration of natural-gas / air. The higher yields are for ethers, nitrated paraffins, olefins, and acetylenics. "Ihis is entirely consistent with the findings of Braisie and Simpson (1968). In the incidents they surveyed, including a wide range of hydrocarbon / air explosions, yield values were commonly less than 0.(M. Deflagrations having yields as high as 0.1 require mixtures containing the more reactive components, listed above, or special circumstances. Ans!ysts often use this more conservative value to cover any and all eventualities, to be very conservative. It would seem, however, inappropriate to do so here, unless it can be shown that detonations are both possible a,Lnd probable. Ttus issue is addressed in the following section.
The Possibility of Detonation There is considerable, quality evidence that detonation cannot be directly initiated in a natural-gas / air mixture with anything less than heroic efforts (e.g., kilograms of high explosive or a strong shock emanating from a pipe; see MIT-GRI LNG Workshop,1982). Further, there is considerable doubt that, once initiated, a detonation could sustain itselfin practical situations of nonuniform mixing. That may have happened at Port Hudson because of special circumstances, but alternative explanations for the damage have been offered. It must be remembered that the Port Hudson event was fueled by propane not natural gas.
Results of all the tests conducted to the 1982 date of the MIT-GRI Workshop, to study the detonability 1
of unconfined vapor clouds, indicated the following:
No detonations are possible in an unconfined stoichiometric mixture of pure methane vapor and air, even under the initiating influence of 2 kg. explosive Sustained detonations are possible in unconfined vapor clouds containing stoichiometric fuel air mixtures when the fuel vapor contains methane and xopane in the molar ratios of 60/40,70/30, and 85/15. Above 85% (molar) methane concentrations, no sustained detonations have been observed.
Propagating detonations (from a pipe) have been sustained in taconfined mixtures of methane / propane / air (stoichiometric) for methane volume fractions of less than 85% in the fuel vapor.
t:OO25.wpf:Ib-081093 12 Y
Dairyland Natural Gas liaiards it is evident from these results that relatively small fracdons of heavier hydrocarbons, such as ethane and propane, in the fuel vapor increase the propensity for the detonadon of unconfined vapor / air mixtures. His phenomenon is of great importance in evaluating the detonability consequences of i
natural-gas releases in the form of late-time boil-off from LNG, but usually does not apply to releases from pipelines, because gaseous releases are usually not rich in higher hydrocarbons.
As a practical matter, as we understand the circumstances of the possible release of natural gas at the Lacrosse BWR site, we conclude that the likelihood of a detonation is an incredible event (less th i once in a million years;.
Flame Generated Pressure In view of the large uncertainties in any attempt to establish an explosive-energy yield, an arguably better approach to setting an upper bound on airblast pressures from unconfined gas-phase explosions is one based on the theoretical development of Kuhl, Kamel, and Oppenheim (1974). A flame propagating through a prendxed gas of uniform composition and properties pushes unburned gas ahead of its flame front, causing the pressure to rise. In the free field, the flame propagates in three dimensions as a sphere. %c compression is less than it would be if the flame were constrained by a j
nonyielding surface (e.g., the ground or a wall) to propagate with only two degren of freedom, and still less than situations in which expansion is constrained to just one dimension. In any case, if the flame has a fixed speed - neither accelerating nor decelerating-it will maintain a fixed pressure rise in the unbumed gas it pushes ahead. Unless the speed of the flame front is up around 40 misec., the overpressure just ahead of the flame cannot exceed one atmosphere (14.7 psig). To shock up and (passibly) detonate requires much higher speeds. he required "run-up speed" for natural gas in air is not known, but cettainly approaches - and may exceed - Mach I (Lee,1977), that is, about 330 misec., which is patently unanainable in the open. Typical burning speeds for methane are 7 misec.,
in quiescent conditions. Turbulence may increase that by a factor of four (to, say,30 misec.; sec 1
Kanury,1975, pp. 303, 3N).
It is important to note that this air blast will not " shock up" - that is, the pressure wave will have a finite rise time, and it is therefore, inappropriate to use a doubling rule for reflected pressure loading of walls. Burning speeds would need to exceed 230 misec. to develop any significant shocking of the pressure wave (12e,1977). De pressure rise time will be comparable to the time taken by the flame to transit the near-stoichiometric (and richer) portion of the cloud. If this were on the order of 100 ft.,
<g the time could be a second or two.
4 1
4 t:0025.wpf: I b-081093 53
Dairyland Natural Gas Hazards i
l Structural Loading and Response l
i Side-on overpressures have little to do with structural response in this case. These buildings are more likely to respond to drag loading than to diffraction (or crush) loading'. If the rise time is long compared to diffraction times and clearing times (and for leaky structures, long compared to inside-to-outside pressure equalization times), then there would be little net force acting on a wall except for dynamic (i.e., drag) forces. An air-tight box would be subjected to crushing forces, but such structures are inherently strong - even within the cloud, no pressures capable of crushing damage would be expected. De pressure-pulse duration (and the accompanying drag-phase flow) is undoubtedly long compared to the natural period (roughly 1/2 sec.) of the Reactor Building. Pressure durations from l
uncondensed explosives are I to 2 orders of magnitude longer than ideal explosions (see Bodurtha, 1980, p.108); therefore, the dynamic pressure pulse can be compared to a wind gust Accordingly, it is appropriate to consider the wind force of the dynamic pressure wave to be the principal mode for j
mechanical damage, and to compare its magnitude to the design value.
In ideal blast waves with a true shock front, peak dynamic pressures can be appreciable in magnitude i
compared to peak (side-on) overpressures. For example, a 4 psi peak (side-on) overpresst re is j
accompanied by 0.4 psi peak dynamic pressure. 'Ihat represents a particle velocity of 200 ft/sec. or 136 mph.' A 6-psi overpressure is accompanied by 0.8 psi dynamic pressure, corresponding to a 260 ft/sec. particle velocity, representing a 177 mph wind gust. A useful rule-of-thumb: The particle velocity associated with a 1-psi overpressure is about 50 ftisec., and increase (or decreases) in linear proportion to the overpressure. Thus, a 300-mph air blast (440-ft/sec. particle velocity) accompanies a
}
peak overpressure of 8.8 psi.
l Alkane / air deflagrative explosions, by contrast, do not generate shocks. Clearly, therefore, the peak i
i particle velocity driven by the explosion cannot exceed the flame speed driving it. As a result, flame speeds in quiescent, premixed methane / air mixtures would be incapable of driving wind gusts above j
16 to 18 mph at the edce of the ext 31 odin 2 cloud! Allowing a factor of four for turbulence, the limit would still be a factor of six to seven below the designed wind force for the Reactor Building, and i
would rapidly dissipate with distance from the explosion.
l i
l t
}
i
' For a fuller development of the mechanisms of airblast loading and responses of structures (and definitions of such terms as dran, diffraction, and clearine times), refer to Glasstone and Dolan, 1977.)
i
.I t:0025.wpf:lb481093 14 i
t Dairyland Natural Gas Hazards i
i l
l Conclusions l
These are our basic conclusions:
1.
The likelihood of a detonation is negligibly small.
t 2.
The air blast from a deflagrative explosion would exhibit a weak pressure pulse having a long rise time compared to any characteristic building-response time. 'Ihus, the response would be entirely like a response to natural wind forces.
l l
3.
Responses to side-on overpressures, if any, would be limited to window / door / light-panel
(
breakage or deformation, but (except for glass window breakage) this is expected only from overly conservative estimates of yield.
4.
Dynamic pressures would be inconsequential relative to the 360-mph design wind loading for l
LACBWPw l
L l
References l
Allan D.S., and P. Athens,1968: " Influence of Explosions on Design," CEP Technical Manual, Loss Prevention Vol. 2, AIChE, New York.
Bodunha, F.T.,1980: " Industrial Explosion Prevention and Protection," McGraw-Hill, New York.
Brasic, W.C., and D.W. Simpson,1%8: AIChE, Loss Prevention, Vol. 2, pp. 91 102.
l Brode, H.L.1955: " Numerical Solutions of Spherical Blast Waves," J. Applied Physics,26.,766.
j 6
i I
l i
Glasstone, S., and P.J. Dolan,1977: "The Effects of Nuclear Weapons," Third Edition, US DoD and ERDA, Gov't Printing Office.
l Kanury, A.M.,1975: " Introduction to Combustion Phenomena," Gordon and Breach, New York.
l Kuhl, A.L. M.M. Kamel and A.K. Oppenheim,1974: " Pressure Waves Generated by Steady Flames,"
14th (Int'l) Combustion Symposium, The Combustion Institute, Pittsburgh.
Lee, J.H.,1977: " Initiation of Gaseous Detonation," Vol._24 An. Review of Physical Chemistry.
Lee, J.H., et al.,1977: " Blast Effects from Vapor Cloud Explosions," paper presented at tie 1977 AIChE Loss Prevention Symposium. March 1977, Houston, TX.
i 1:0025.wpf:lb-081093 15 l
.m
. __ ~-
Dairyland Natural Gas 11azards l
Lind, C.D.,1975 ; "What Causes Unconfined vapor Cloud Explosions?" CEP Technical Manual, l
Loss Prevention Volume 9, American Institute of Chemical Engineers, New York.
t Raj, P.K. (ed),1982: "MIT-GRI LNG Safety and Research Workshop, Volume III: LNG Fires -
Combustion and Radiation," GRI 82/0019.3, Gas Research Institute, Chicago.
Strehlow, R.A., and W.E. Baker,1976: "The Characterization and Evaluation of Accidental l
Explosions," Prog. Energy Combustion Science 2,27.
Stull, D.R.,1977: " Fundamentals of Fire and Explosion " AIChE Monograph Series 3 No.10, American Institute of Chemical Engineers, New York.]"
i i
~!
l l
l 1
i l
l l
l i
)
i t:0025.wpf:1b-OS1093 ~
16
O O
ATTACHMENT 3
O
- _ _ _ _. _ _ _ _ _ _ _ _ _. - _ _ _ _ _ _.. _. _ _