ML19338C751
| ML19338C751 | |
| Person / Time | |
|---|---|
| Site: | Sequoyah  |
| Issue date: | 09/02/1980 |
| From: | TENNESSEE VALLEY AUTHORITY |
| To: | |
| Shared Package | |
| ML19338C749 | List: |
| References | |
| NUDOCS 8009030666 | |
| Download: ML19338C751 (450) | |
Text
{{#Wiki_filter:. O 4 TENNESSEE VALLEY AUTHORITY l SEQUOYAH NUCLEAR PLANT CORE DEGRADATION PROGRAM VOLUME 2 REPORT O ON THE SAFETY EVALUATION OF THE INTERIM DISTRIBUTED IGNITION SYSTEM l l SEPTEMBER 2,1980 l goo 903066fo __ _ _ . - - - __. _ _ _ _ _ - . - - - - _ _ - - - - - - - _ . 1- - - - -
TABLE OF CONTENTS I. BASIC CONCEPT OF CONTROLLED IGNITION () A. Amount of hydrogen from events beyond the current regulatory requirements for design basis B. Potential impacts on containment C. Controlled ignition as a potential means of limiting impact II. INTERIM DISTRIBUTED IGNITION SYSTEM A. Overview B. Igniter C. Location D. Power Supply E. General Requirements F. Igniter Tests G. Operation H. Surveillance Testing ( I, Qualification III. EVENTS A. Overview B. Characterization of Events _ C. Mass, Energy, and Gas Releases D. Basic Assumptions E. Calculational Methods _ i IV. DEFLAGRATION AND DETONATION A. Overview B. Concentration Effects C. Impact of Water Droplets D. Flame Temperatures - 1 p E. ' Flammability Limits V - F. Ignition Requirements i
G. Flama sps ds
-H. Burn Efficiency
( V. CONTAINMENT PROCESSES AND DESIGN A. Overview B. Combustible Gas Control System C. Containment Structures D. Ice Condenser System I E. Containment and Residual Heat Removal Spray Systems F. Air Return Fans VI. HYDROGEN DISTRIBUTION A. Overview B. Factors Promoting Nonuniform Concentrations C. Factors Promoting Uniform Concentrations D. Program Plans VII. ANALYSIS OF CONTAINMENT CONDITIONS A. Overview B. Methods, Model, Assumptions, Inputs, Environmental Conditions, Results, and Verifications 4 C. Containment Environmental Conditions VIII. ENVIRONMENTAL IMPACT A. Overview B. Containment Boundary C. Containment Interior Structures D. Critical Containment Systems E. Containment Shell Temperature
- IX. CONCLUSIONS A. Overview B. Ignition Source Effectiveness
() C. Containment Heat Removal D. Environmental Effects 11
E. Scope of Events F. Margins Available G. Uncertainties APPENDICES A. Computer Codes B. Event Definition C. Phenomenal Description of Hydrogen Distribution D. Nonsymetric Containment Loads E. Critical Components JI
, j F. Thermal Effects on Components !
G. Pressure, Shock, and Transient Loads on Components H. Containment Response to Detonations I. Containment Systems J. Containment Structural Capability Analysis O E. Conteinment Monitor 1ns Svetems L. Igniter System Design M. Igniter Tests - Endurance and Acceptance Tests N. Igniter Tests - Environmental Conditions
- 0. Hydrogen Combustion P. Sensitivity Analysis Q. Analysis of Effects of TMI-2 Events R. Operating Procedures S. TVA Degraded Core Instigation Program T. CLASIX Program Description U. Summary of Analysis of Ice Ccndenser Containment Responses to Hydrogen Burn Transients V.
Progrv..; Report on Verification of CLASIX O ')E03;INDEX Q iii
, . - . - - . .~.
SECTION I ( ). BASIC CONCEPT OF CONTROLLED IGNITION A. Amounts of hydrogen from events beyond the design basis accident The limiting design basis accident for hydrogen production at_the Sequoyah Nuclear Plant has been based on a large loss-of-coolant accident (LOCA). This is in agreement with current Nuclear Regulatory Comission regulations regarding hydrogen control (10 CFR Section 50.44 and GDC 50 in appendix A to 10 CFR Part 50). Using conservative licensing calculations, relatively small amounts of hydrogen are a , predicted to be produced due to oxidation of z realoy cladding O in the reactor core in the short term during a LOCA and corrosion of various materials in the containment in the long
~
. term following a LOCA. Thase design basis amounts of hydrogen are currently capable of being accommodated by hydrogen recombiners with a low-volume purge system as backup. The accident at Three Mile _ Island 2 occurred when a stuck-open relief valve vented the primary system to the containment; effectively a small LOCA. Even though the automatic safety
. systems responded as designed, manual override by the operators led to the core being uncovered for longer periods than-assumed in the design ~ basis calculations. As a result, significant core damage occurred and more hydrogen was 1-1 D
, - , ,n.
, , . . ,,w -- u a r.- - -~ -, - r--,
generated at a faster rate from the oxidized circaloy than had been considered previously. (GD The TMI-2 accident has caused TVA and the NRC to reassess the effect of hydrogen on plant safety at Sequoyah. Even though it is believed that the present emergency core cooling systems are completely adequate and that significant improvements in instrumentation, other system hardware, and operator treining have been made since TMI, evaluations are being made of that class of events in which significant hydrogen production does result from inadequate core cooling. The NRC is planning to issue an advance notice of rulemaking on degraded core accidents shortly which will result in examination of these events and possible reco=mendations for their mitigation. Presently, for Sequoyah, TVA's att5ntion n (_) is focused on events with core damage limited to approximately 75 percent since more severe degradation would almost certainly lead to complete core melt which has a different associated set of conditions to mitigate. Even with this restriction in scope, the rate of hydrogen production and release is extremely accident scenario-dependent. A representative sequence selected for a number of TVA's evaluations has been S 2D from WASH-1400, a small loca followed by loss of core injection. B. Potential impacts on containment Most of the hydrogen produced in the TMI accident was released U 1-2 .-v+.
to the containment in the first few hours. About ten hours into.the event, a sudden surge of pressure of about 28 psig , was recorded inside the containment by the steam generator , /'N U pressure sensors. Almost immediately, several thermocouples distributed throughout the containment showed about a 50 F temperature increase. The pressure spike decayed very quickly, while the temperature fell more slowly. Although it is generally agreed that the transient was due to rapid hydrogen combustion, estimates of the amounts of hydrogen responsible for the transient vary widely. i ! 4 The Sequoyah containment is designed for a-differential pressure of 12 psi and a peak temperature of 250 F. Due to the margins in design, TVA believes that the containment would withstand the pressure and temperature loadings actually {} recorded during the TMI-2 accident. However, because of the smaller volume of the Sequoyah ice condenser containment when compared to the TMI dry containment, the addition of heat from adiabatic burning of some of the estimated amounts of hydrogen produced at IMI would result in press".es exceeding the ultimate Sequoyah containment capability. However, a i determination of the containment capability is very scenario- ' ~ dependent, since the rate of energy release is at least as important as the total magnitude. In addition, the degree of conservatism versus realism in the calculational technique and assumptions used in the evaluation is extremely important i to the overall conclusion. () 4 1-3
~
- + Mr +.. - e. 4.: 3 , - 4W/4@Jd;'l;[dli.gs 4.KdMS*i Tf"U Le ,
. . e w e% %i , m 's, ,
C. Controlled ignition as hydrogen mitigation TVA is convinced that the current capability of the Sequoyah containment for hydrogen control is acceptable. However, l to increase this capability in the short term, several steps 3 are being taken. Foremost among these is the installation of a controlled ignition system consisting of thermal glow i plugs distributed throughout the containment available after a degraded core accident to initiate combustion of any significant amount of hydrogen released. Since the [ containment capability is more dependent on the rate of hydrogen release and combustion than on the total magnitude, the rationale behind the ignition system is to ensure burning in a controlled manner as the hydrogen is released instead 4 of allowing it to collect and then be ignited by a random () source. This more gradual addition of the heat of combustion will. allow the active and passive containment heat sinks to reduce the overall impact. Although the rate of hydrogen release predicted for various accidents is very sequence-dependent, the ignition system should be effective for a wide range of accidents and associated release rates. Even though the release locations also depend on the accident scenario, mixing due to containment turbulence and proper location of efficient igniters should allow combustion of hydrogen at~even lean concentrations as it is released. The containment spray systems, ice condenser, and internal struct'ures all serve as effective heat sinks to moderate the l I temperature increases due to controlled combustion. Another
'k-) 1 1-4 e
g r A <-'ec*e - w> womeo omano ww .e *-n
- .- r e - --- , . ,w,----owg-- --e,- .-+---.,-,w w --,m--gv-.m e --
advantage of controlled ignition is in reducing the potential for local detonations due to pocketing over a period of time. ( x, The cnly potential drawback of initiating combustion in an ( ) undesirable manner with the ignition system is already present due to the numerous random sources of sparks in the containment. Therefore, TVA has concluded that the controlled ignition system should result in a net benefit by providing added margin to the containment capability for hydrogen control during degraded core accidents. (^N t/ m C1 1-5
INTERIM' DISTRIBUTED IGNITION SYSTEM t A. Overview Q TVA has designed'an igniter system that we believe will burn l volumetric quantities of hydrogen at low concentrations. The igniter assemblies are seismically-designed not to damage other safety-related equipment inside containment. They are powered by a reliable ac scurce and in case of loss of offsite power will be powered by the diesels. TVA performed preliminary tests in
; selecting the igniter to verify that it would burn hydrogen and operate for reasonable periods of time. TVA has also scheduled additional environmental and endurance tests on the entire igniter assembly to assure that it is a reliable and effective design.
TVA has installed the igniter assemblies throughout primary containment at Sequoyah unit 1 and we have made arrangements to periodically perform surveillance tests on the installed system to verify operability. B. Igniter The igniter that TVA has selected is a G'eneral Motors AC Division Model 7G. This glow plug is commonly used in diesel engines and is readily commercially available. The igniter is powered directly from a 120/14V ao transformer contained inside the igniter assembly box. The box has overall dimensions.of 8"w x
- 8"h x 6"d'and is constructed of 1/8-inch steel plate. It is covered by a spray shield and has a 1/16-inch copper heat sink I I l
4 e O z.1 1 f
,, . 5 , -: + .h . . , ~ - . - mc7p.-o,y9;3.-< ,~.mem mm-Nr.7%F.C .
-- . ,v-y.3-v* mm.m.4:s1 Y<
CA T :,;ck .. '" * -
. G :Ui' ' + t ~ 22'
- N Nf*MO u ;
i
mounted to ths glow plug on the face of the igniter assembly. A more detailed description of the igniter assembly is contained in appendix L to this report. C. Location The interim distributed ignition system has 31 igniter assemblies distributed throughout the lower, upper, ar.d ice condenser compartments. FSAR figures 6.2-142 through -146 contained in appendix L to this report show the ignitor locations by-elevations. There are a total of nineteen ignitors in the lower compartment, seven inside the crane wall at elevation 731.0' and 12 outside of the crane wall in various rooms, compartments, and pipe chases. The upper compartment has three igniters which are suspended 35 feet from the top of containment. The ice condenser has nine igniter assemblies, five uniformly distributed below the ice condenser baskets and four also evenly distributed above the ice condenser but below the top deck blanket. A more detailed description is contained in appendix L. The igniter location and quantity were chosen based on the following considerations.
- 1. The interim ignition system is intended to provide ignition if the bulk concentration in the lower compartment, upper compartment, or ice condenser exceeds the lower combustible limits. This is the basis for the preliminary analyses by TVA and the more complete CLASIX calculations. No credit has been taken for local combustion in a smaller segment
~.,
l 2-2
)
t-lof these compartments. Therefore, igniters have been . distributed in each major region to provide assurance that ignition occurs if the bulk concentration reaches or l At levels above approaches the lower combustible limit. l the combustible linit, one igniter is all that is needed to ignite a major co.'partment. In phase 2 of TVA's program, we intend to study the detailed NB concentration profiles and the behavior of hydrogen ne1r the combustible limit in order to take advantage of localized and partial combustion in optimum locations. This will provide additional margins of safety but is not l required in an interim system. Igniters were located at or near the tops of each major () 2. subcompartment. -When the height to width ratio was small, igniters were located at several places in order to assure adequate ignition as early as practicable.
~
1, 3 Our preliminary judgment was that on the order of 20 to 30 igniters would be needed to provide adequate assurance that the objectives in one and two above would be met. l Detailed placement resulted in 31 igniters which was in good agreement with the preliminary estimate. j
- 4. Within a given area, the maximum size of a " cloud" of gas I r
not exposed to igniters is on the order of the distance between igniters. Since the analysis is based on combustion I 2-3 . i e-W* 74 a i 41, i* hk @
- d-
of upper nor lower compart, ment, the characteristic dimension of the analysis is on the order of 75 to 100 feet. The igniters were.'.3cated to provide a separation distance of significantly less than this.
~S. Because of the tendency of hydrogen-rich mixtures to rise, i the igniters were located near the tops of compartments in order to provide ignition as early as practicable.
4
! 6. The lighting circuits were found to meet all our requirements (power supply, number, compartments covered, l spacing,-height, and location within compartments. This is not surprising since these circuits are designed to 4
- provide coverage of compartments from above. Fixtures were found in all compartments where igniters were felt to be needed, although all fixtures were not used. If additional or modified locations are found necessary based ~on our continuing studies, new circuits or wiring will be installed as needed; possible-igniter locati5n is not limited by lighting circuit availability.
D. Power Supply The igniter assemblies are supplied with 120v ac from the standby lighting system. The standby lighting system which receives its
' power through' 480/208V transformers has alternate ac supplies from the 480v MOV shutdown boards. In case of loss of offsite
~
power the shutdown boards are automatically energized by the diesel _ generators and therefore the IDIS is capable of operating i 2-4 j' .
in such an event. A description of the standby lighting system and the 120/14V ac transformer inside the Egoiter assembly box is contained in appendix L. E. General Requirements The igniter assemblies are seismically designed not to fall down in the event of an carthquake and damage safety-related equipment inside containment. This is accomplished by attaching a steel cable to the igniter assembly and then anchoring the other end to a bolt embedded in a concrete wall or ceiling. This is required of all components inside containment that were'2ot individually seismically qualified by test or analysis. s. F. Igniter Tests s
%-)
TVA conducted a testing program at Singleton Laboratories to obtain preliminary information about the performance of commercially available igniters. From that testing program it was determined that the GMAC model 70 g16w plug could produce sufficiently high surface temperature (1720 F), could remain energized for prolonged periods of time (148 hours), and could burn hydrogen in small sealed containers containing hydrogen concentrations of 12 percent and less. From these results TVA gained considerable assurance that commercially available igniters would be suitable for use in the IDIS. see appendix M for further details, r (hj Hrvever, TVA feels that additional testing is necessary to r 2-5 , I
demcastrate that the igniter assembly will initiate a volumetric burn of hydrogen for various environmental conditions of pressure, temperature, and steam. TVA, therefore, has centracted with Fenwal Laboratories of Ashland, Massachusetts, to perform hydrogen burn tests with the igniter assembly in an enclosed vessel. For a description of the igniter assembly testing see appendix N. G. Oyeration The Interim Distributed Ignition System (IDIS) is designed to be energized immediately following the start of an accident and remain energized until the unit reaches cold shutdown. In order for the operator to initiate the IDIS he must dispatch an Assistant Unit Operator (AU0) to the standby lighting panel and switch on lighting circuits 10, 11, and 12. The appropriate modifications to the Emergency Operating Instructions (EDI's) have been proposed. These modifications will not be incorporated until NRC approval for the system is obtained. Further details on the proposed modifications to the E0I's is contained in appendix R. H. Surveillance Testing Periodically the igniters will be subjected to surveillance testing. Surveillance testing will consist of energizing the IDIS at the standby lighting cabinet and taking current readings of circuits 10, 11, and 12. These three circuits will have only O isniter essemh11ee connected to their out1ete, a11 other outiete 2-6 ,
on these circuits will have the light bulbs removed. During the post modification testing, once all the igniters are inctalled () and operating, initial current reading of the three circuits will be taken. These current readings will become the base data that will be compared to the readings taken during the surveillance tests. The comparison of the two readings will indicate whether or not all the 16 niter assemblies are operational. If the readings do not compare favorably then all the igniters will be checked visually. I. Qualification The initial testing TVA ce9 ducted at Singleton Laboratories (see appendix M) identified a commercially available igniter that TVA
- chose to use in the IDIS. However, that initial testing was not U conducted in sufficient detail to provide a high level of confidence in the GMAC model 70 glow plug. TVA has therefore procured 300 of these glow plugs and will perform a statistical qualification program on them to obtain a high level of confidence in their reliability. The qualification program will consist of selecting a statistical random sample according to Military Standard, MIL-STD-105D, " Sampling Procedures and Tables for Inspection by Attributes," then testing the ability of this sample to withstand cycling, prolonged operation, and reponse time from anbient temperaure to the minimum required surface temperature.
TVA then plans to select from this same captured lot of 300, 31 glow plugs to be installed in the igniter assemblies. O 2-7
- em - e e ..e .mr,,. a
SECTION III EVENTS O The ignitor syLa nas been designed to prevent the accumulation of explosive concentrations of hydrogen inside the containment following accidents exceeding the design basis. The system is not required for any transients with severity less than or comparable to the design basis large break loss-of-coolant accident. However, it is currently planned to initiate operation of the system for all accident sequences since use of the glow plug igniter system for events where they are not required will not result in unacceptable consequences. The design basis large break loss-of-coolant accident is one such event that does not result in the production of excessive quantities O of hydrogen. The proper operation of the emergency core cooling systems act to prevent fuel rod heatup beyond 2200 F with subsequent large cladding oxidation rates. Tha redundant containment hydrogen recombiner system installed at Sequoyah has been designed with sufficient capacity to adequately remove all hydrogen forration predicted by ECCS computer performance models in addition to that resulting from the assumption of conservative water radiolysis and zinc / aluminum coating reaction rates. The igniters are expected to provide some local recombination for events of this severity but which is not required. Therefore, events of this and lesser magnitude have not been selected for the igniter design basis. Events where functional operability of the igniters cny be required to mitigate n hydrogen accumulation are those that include significant periods of U 3-1
inadequate core cooling. Production of such an event sequence requires the assumption of the failure of redundant core cooling systems in combination with primary system coolant loss or, as in the case of the Three Mile Island accident, operator errors which terminate coolant injection improperly during a coolant loss event. The Reactor Safety Study (WASH-1400) quantitified the probabilty of degraded core events to determine which events were more likely to occur. In addition, cathematical models were created and applied to approximate the core thermal-hydraulics, containment response, and radiation release consequences of these events. From this study several events have been selected, based on probability of occurrence and hydrogen release potential, for the design of the igniter system (see volume I, section 5). - A. Overview I The event sequences selected have a low probability of occurreisce but relatively severe consequences as predicted by conservative physical models. Five accident sequences were chosen to represent the spectrum of hydrogen producing events from metal-water reactions. Each event results in inadequate core cooling and, if cooling is not restored, core melt. Fuel-cladding oxidation is excessive prior to clad and fuel melt thereby resulting in the production of large quantities of hydrogen which cannot be accomodated by the installed design basis recombiner system. Each event represents a significant challenge to any 3-2 .
l t hydrogen removal scheme. t 1 () For the purpose of this study, the alphameric designation of events presented in the Reactor Safety Study (WASH-1400) has ! been retained in this report. A description of these l I designations can be four.1 in appendix B. The sequences selected j are designated as follows: l t i : 1 ; l 1. AD - Large break LOCA with concur ent failure of i i j Emergency Core Cooling Injection. j i l l
- 2. S2 D - Small break LOCA (1/2 inch A D 4 2 inch) with l j concurrent failure of Daergency Core Cooling Injection.
3 Sn-S a 11 break LOCA (1/2 inch 4 D di 2 inch) with
, failure of Daergency Core Cooling Recirculation.
4
- 4. TMLB' - Failure of the Feedwater Delivery System (Power 3 Conversion and Auxiliary Feedwater Systems) given Loss l
4 of Offsite AC Power with a failure to Recover either onsite l or offsite AC power within 3 hours as the initiating event. 5. TB) - Transient Event of Loss of Main Feedwater followed by failure of the DC Power Supply System. t Each selected event sequence is characterized in the following 3 section. A further condensation of scenarios was performed for i the purpose of hydrogen production. The gas release rates for these scenarios is described in section C.
~
!O 3-3 e ., r--w,-w-, , - . . . w eg --W,4 -3 yw
-y-w m +- -mw- - -y 4 --
--w r . -y.3w y y -- _
gp vy. -y -m,ww.,-.yy-7-m r,- 9i.w&.m
B. Characterization of Events AD - Large LOCA with concurrent failure of emergency core cooling J injection. I l This accident sequence uses the following event assumptions: i j A - Large LOCA (break area greater than that of a 6 inch i diameter pipe) Z - Ice condenser functions properly B - Electric power available X - Air return fans operable . C - Containment spray injection available D - Emergency coolant injection fails E - Emergency coolant function lost due to D H - Emergency core recirculation lost due to D
- F - Containment spray recirculation operable j G - Containment heat removal systems operable I
l l This sequence assumes the emergency cooling injection system would fail to adequately cool the core following the occurrence of a large LOCA. All containment engineered safety features would operate as' designed. However, tince failure of emergency core cooling would cause a core melt in approximately 0.5 hour , after the occurrence of the LOCA, operation of the emergency core cooling system in the recirculation mode would not prevent continuation of the core melt. Note that the reactor protection I (} system _is not required to function early in this event since 3-4 e t
, ,--c. --.-- e---,-- ,, - - . ----c.- -,wyme ,,,-.-- <,-,g--- - , , -- -, - < , , ,,y-,..--y ---y-- , - , , , , . -w,,,y= -- e,- e s -eg--r,-ew,- e
the rapid loss of r 'derator will force the core to become suberitical. The auxiliary feedwater system is also not required since energy removal by critical flow discharge of fluid at the (' } break exceeds the energy input from decay heat. Un fortunately, withcut coolant injection, mass is depleted from the reactor coolant system until the core uncovers, heats, and begins to molt. Early restoration of emergency core cooling injection may terminate this event short of complete core melt but may result in significant hydrogen generation as the fuel rods are quenched. S D - Small LOCA with concurrent failure of emergency core 2 cooling injection. This accident has the following events postulated to occur: O S - S=all LOCA initiates the event 2 Z - Ice condenser functional B - Electric power availble
~
. \
K - Ecactor protection system functional X - Air return fans operable - L - Auxiliary feedwater system available C - Containment spray injection available D - Emergency core injection fails E - Emergency coolant function lost due to D F - Containment spray recirculation available G - Containment heat removal systems available 3-5 9
In this sequence of events, the emergency core cooling system would fail to adequately inject water to the core following a x_; small LOCA having a break equivalent diameter between about 1/2 inch and two inches. All containment engineered safety features would cperate as designed to control containment pressure and leakage and would serve to remove radioactive iodine airborne in the containment. Since the emergency core cooling system aas considered to fail in the injection mode, its operation in the recirculation mode would not prevent core meltdown. The accident is characterized by a slower loss-of-coolant when compared to the large break. This permits more time for restoration of coolant injection. Complete core melt and hydrogen generation occurs over a longer period of time resulting in slower release rates due to the extended presence of water in the core. The igniter system is anticipated to be more effective for the small break scenarios where the hydrogen release rates are slower as exemplified by the Three Mile Island accident. _ S 11 - Failure of emergancy core recirculation system given 2 a small LOCA. This event proceeds through the following sequence: S - Small LOCA initiates the event 2 Z - Ice condenser functional B - Electric power available 3-6
K - Reactor protection system functional X - Air return fans operable L - Auxiliary feedwater system available C - Containment spray injection available D - Emergency core injection functions E - Emergency cooling functional H - Emergency core recirculation fails F - Containment spray recirculation available G - Containment heat removal syste=s available In this sequence of events, the emergency coolant injection system functions properly but in the transition to the recirculation mode, failure occura. This event follows the same sequence as does the S D case except for the above 2 failure difference. Hence, core meltdown proceeds unobstructed during the recirculation mode. The meltdown of the core is delayed even further in , time than the S 2D case since inadequate core cooling does not occur until switchover to recirculation. In addition the decay heat at this time has decreased somewhat resulting in a slower core heatup. Time is available to attempt to correct the recirculation problem if external to the containment or to provide nonsafety grade water supplies to the reactor water storage tank to reinitiate the injection mode. , Containment design will limit the amount of additional water that can be injected. O 3-7 e
- l TMLB' - Failure of the feedwater delivery systems (power conversion and auxiliary feedwater syste:m) given the
{} initiating transient event of loss of offsite ac power. l-l This event proceeds through the following sequences: k T - L ss of all ac power A K - Reactor proection system functional B2 - L ss f ac p wer cours L - Auxiliary feedwater system fails 4 M - Secondary side relief \ tves functional P - Primary side relief valves open Q - Primary side relief valves close i 4 Loss of all offsite ac power occurs such that neither the power I conversion system nor the auxiliary feedwater system is available to remove heat from the RCS. In determining the probability - of the sequence, it was considered that neither the offsite nor onsite emergency sources of ac power were recovered within about 3 hours. Since all ac power was lost and not recovered in i i sufficient time to prevent an excessive coolant loss through I the RCS safety and relief valves, a core melt occurs. Also, since all ac power sources were not recovered, containment engineered safety features could not operate to mitigate the i radioactivity released from the melting core. Failure of the ac power system also results in consequential failure of the coolant injection systems. Although no break 3-8
~
t7 y -- c -'
--.a er --
fr M - w - - gywmyav -m -yy- T p- wrnT--- ,,
+-99w9- y -et*- ++eme-
occurs, the inability to remove decay heat from the primary system due tc auxiliary feedwater failure results in primary g.s Q system pressurization and pressurizer relief valve cperation. The system slowly boils dry as the relief valves cycle open and closed. Core molting results when sufficient water no longer rcrains in the vessel te adequately cool the core. T - Failure of the de power supply given the initiating event B of loss of main feedwater system. This accident proceeds through the following events: T - Transient loss of main feedwater system initiates the B accident K - Reactor protection system functional B - AC power available B) - DC power fails L - Auxiliary feedwater system fails due to B) M - Secondary side relief valves functional P - Pricary side relief valves open Q - Primary side relief valves reclose D - Emergency coolant injection fails due to,E O - Containment emergency safety features fail due to B) Loss of the main feedwater system initiates a primary system transient (failure to remove primary side heat). This is followed by a loss of de power. Loss of do power results in p failures to the engineered safety features due to the dependence Q . 3-9 - 6
. j
of the control systems of these features on de power. Hence, de power failure causes loss of ECI and the containment emergency
- ( }
safety features resulting in core melt and containment failure. The Three Mile Island accident sequence has also been examined in relation to WASH-1400 by the Kemeny Commission technical staff (III-1). It was found that the accident was roughly equivalent to the TP'QU' sequence. This is similar to other small breaks discussed above but with termination of high pressure safety inje,ction unavailability before complete core damage resulted. TVA plans to study this accident in relation to the specifics l of the Sequoyah design as discussed in appendix Q. The TMI accident would encompass the following events (III-1): T ~ Transient event K - Scram availability P' - PORV of safety valve operation Q - PORV or safety valve stuck open L - Secondary side cooling restored U - HPSI available U' - HPSI interrupted and PORY open for sufficient period to cause core damage U - HPSI restored in time to avoid core damage. C. Mass, Energy, and Gas Releases Each of the core ccit sequences (AD, S2D, S2H, TMLB', TB, etc.) _
#J described above generates a significant quantity of hydrogen 3-10
.,e p=1E. s- .1.,,.. + a _ a e a as.,. m.m -
a .m a .- . .-aw - *+ which is subsequently released to the containment. Several of i the cases are expectad to yield similar hydrogen production rates 4 () perhaps merely offset in time. For example, the S D and the S2 H 2
- sequences are similar accidents, except the loss-of-core cooling does.not occur in the S 2H case until the sump water recirculation is required. This delays the onset of hydrogen production l significantly in the accident. Similarily, the consequences i
of the TMLB' and TB accident may be about the same. Therefore, due to the restricted amount of time available for this study and because of limited availability of existing data, the
~
i sequences considered have been limited to the AD, S D, and TMLB' 2 accidents as discussed in volume I, section 6.1 for which data J was available. Other nequences may be analyzed inhouse, depending on the availability of the appropriate analytical tools as discussed in appendix Q sometime in the future. i The'three cases mentioned provide a wide range of mechanistic hydrogen release rates ranging from very rapid, in the case of the AD sequence, to long term as in the_ case of the TMLB' sequence. The cases also encompass several different release i mechanisms. Both AD and S D accident are initiated by pipe 2 ruptures in the reactor coolant pressure boundary. The hydrogen generated for-these events should be lost through the break 4 1 ! opening. The TMLB' accident does not postulate a break in the
- primary system. For this accident, mass will be lost through i
the pressurizer relief. valves into the pressurizer relief tank. O 3-11 J' % . A u > .yy; 'g..~m.
...u e a xp, " ," N, . '. t
> l$ h~ -a
. - . ._. - - , - . . .~ ~ . _ - . . . . _ - . ... ___ -
.,_.m, , _ . , . - _ . _ , _ , , _ . _ . - . - - , , , . ., ,~, -
~
Hydrogen will enter the containment either through the rupture disk on the relief tank or from the head vent line if it is used. [- 't V The break sequences therefore represent unknown release points (all should be in the lower compartment); whereas, the release point for the TMIB' sequence is known. The release rates have been describet in volume I, section 6.2 3 1. They are repeated here for convenience in figures III-l thr ough -9 The values werG generated by Rattelle Columbus in an earlier study obtained by Westinghouse. They are specific to Sequoyah. Each event may terminate prior to complete core celtdown, provided adequate core cooling is restored. This is discussed in reference III-2. Several nonsafety systems may also be employed to cool the core after core melt to prevent () rupture of the containment. Unfortunately, cladaing oxidation may be as high as 75 percent by this time. The igniters may be effective for metal-water reactions as high as 60 percent. Release points, as mentioned above, wil1 not be known for the break events. For large breaks, the release may occur in either a hot leg or a cold leg. The hot leg break has the potential for releasing hydrogen near a steam generator dog house where it may collect. However, the temperature at which the hydrogen is released will probably be in the spontaneous ignition regime for this break. Small breaks, as typified by the S D sequence, 2 may occur in a wide range of locations, including the letdown line and pressurizer attached piping. Hydrogen from the sources (j (_ i may accumulate in the pressurizer dog house, steam generator 3-12 e w
I N t o
%) -O,
-4
- ...................p....................
l
................... . . . . . . . . . . . . . . . . . . . 6. . . . . . . . . . . . . . . . . . . . ................... .
4....................g...................(...................t....................{...................
. . . . . . . . . . . . . . . . . . . .................... 6................... . . . . . . . . . . . . . . . . . . . ., . . . . . . . . . . . . . . . . . . . . . ....... ....... ...
..................;...................g...................(...................t................... 4...................
..................g........ ~~~.---- . ]
.........t.~ ~ ~. - - ~ ~ ~ ~ 4.- ~ ~. - - . - - - .i.. - ~ ~ ... ~ .- ~..i..
.........g.......................................p....................
........q...................i....................{...................i.................
.o 4 .
..g
..................i.......... .......
................... 1.......................................f...................
.................(........... .......g...................t....................g.................. 4....................
..p........... ......... . . . . . . . . . . . . . . . . . . . .p...................;...................p...................
m
........... ......(.......................................;.......................................
. . H H
..................g.............
.......................t.......................................g....... .
Os OH
~
OH
..................g......................................t....
...................q .................. (fj
. N/o
. to M,
g
..........g...................)....
........ ............... ...................g................... . ymQ
- . :. - M O
n
?
_o
. ,.4 g4
..................... . . . ..............{ .
........ g .........
....... ..........;...................p... ................3...................)...................
pu : : : :
.. O <s ........
- y. : : : -
p e . . . q O O M. : : W. . v_ m .
)4 - .e A h :. :
A !
!. !. ee Q< .
g) i i i i i ,4 0 dT 0 0T 09 0'9 OT 0d O'O pr. (mq0 ssm 3-13 .
o hi -- __O
\s# .................... ..................'....................,..................t..................;,...................l-H g
...................;...................q..................4....................t................>...;...................,.
...................:...................p...................:..................6....................:...................L.
..................;..................g..................4...................4...................4....................r 7
....... ...........;...................(...................;...................<..................
. . . 4.................... .
.....;..................4....................;...................[...................;...................
.........4....................{..................4....................{...................
a . g . . s
.e
. a . .
; .' -g c, w- . .................g............................ .........j................... ..................t...................
.t.................... .......... . . . . . . . . . . . . . . . . .....................
..t....................
t-l - . . 4....... ..........(..............-..4....................
.................4....................t..............
............... ....t......U...............................p................... ..
se.,,,
...................t..................... .
~
- m no
....................{................. ........ W ..........{.......................................
gs
- N
...... . .........g...................................... g......m............;....................g...................
QH p* l l QH
- : : . l
! ! ! U
..........q...................g................... N...... vMo
*..........~t.-
q
- - - -.- ~ ~ ~t.. - ~ ~ ~.- ~ ~ ~
u
. .. E4 . w
. p . .s
- : :r -
*=-4 p.
............;.)4.,................t.................... pq O ......... .........g...................t......
g O i. i.
- i. Mo i EH o .
O ! !-q '
- i a
>- : : : w .
O H ..... ..
-d 4- M .......... ;.......................................
A o, . . 8 4 . ...... y m y
......... y...................)...................t...................).................
.~ ., .
,3, r . . . .
............................................................ ....................f
........ 3 .........
1 . c . . . bs
;, 8i
? .
-4
=
v . . . A ,, . 1 1
' i .! 'm ,
- i. i. . i. g 1 s s s s s 4 0'0003 0'0938 0'0003 0'093T O'00M 0"009 0'0 (TH91) SSVR 3-14 .
l
O se _, O
- : : : . . v-4
...................g.........................................
.t................... ....................g...................
...................n........................................a............................................................
. ......................................C.23.................... ....................
I:1....................
............... 4....................q...................{................
. . 3 ~.
(.............. ...t.................... o: H. :. :
- 9> :* :* -
A. . 3 . .
................g.. .
~ ................... 3..............................J......~.,.
I w .................... : ................... . 0: .
. 3 .
. t g O .
g . 4.................... ...................(............. ,....q...................{...................
.............. 9 .
. . s . .
. s. . .
. . s
.g . .
sg .
. s .
. s .
s .
.....;...................t.......... .........;........s g.........p.........
. s .
. e. s .
. . . g .
. s. .
%! ^
>- -4 :. % M h
.g
% m
. . . . % i s : : : : :. s g
OH gH
! I ! ! ! \ fo g H g
vw mo l :
..................t.......................................!.....................................
............sg
...... _. M.
w
.U m
g .-.. O l
....................................................................:.............s,... 4 9
O 8
............................j..................................................... '.
i ! ! ! , g ........ _ O ........ 04 :
.........,...................;...................g...................;............
- \
. . . .t T.
N . .
........ O.4 e .........{...................:.................. 4...................:................. .
f4 A h I i I i M
..........;...................p....................g...................t...................
4 . . .
......... s, o . .
a 9 : : L4 p i i . M, le f< , . k8
. ....... .........g...................(...................g..............-...{................... .
i n Z . . n' .
..e
- e~
A
- A ,w-4 s :. :. : :. :.
. . e .
i i a i i O g 0 JI 0'OT O'O O'9 OT 0"d 0V
,0T, (urq1) ssyn 3-15
na O m . ; . g 1. n
...............,....... ..........).......................................
.... . ....... ....l....................p.. .
. . o o.g oo o o.o.o.o o . .
.. . . . . . . . . . . . . . . . . ( . . . . . . . . . . . . . . .....
oo t.o o . o o
................................... o o m . o . ( o....................
. o o . om oo.~ .
. t.o oo o.
. ...q...................;...................
.... ........... 4....................{...................;................
. . . . . . . . g . . . . . . . . . . . . . . . . . . . ; . . . . . . . . . . . . . . . . . . . t.o o o o . o o o o o 4. o o o o . .
. . ..............(.. .
. g
..... ......... ..:......... ........g..................4..................3...................3.....................
(
.e
. . -o g 4
- N. . : .
.............a,........ ......... ................... 1.................... .. .
..................i...........................
...................(................... .
....g........................... .........j .................. .
m oo ooo. ogoooooo.omoyomoo.omooo .,
..................;-~~~...oooogooo.oooooo; 7 .
>- (
... .............. 4................... : : Ht d
..... ........... ...................t..............
OH
.............;...... ............. . . . . . . . . . . . . . . . . . . .g...................
H
............................................... : i : :
od O y) i
- : :* L U2 O v L s p. I..................*..................j.....................................j................... : .
. . B.
e M ..................: . Me p
. . . . . . . . . .. . . . . . . . . . . . . . . . . . . . .p...................:
9.................... 6-4
- E-4
}- : : ~
w !. !. O i. i.
!. !. M O
i.
. a i
,4 qq ....................t........................ .............
rn M .........t............................:...................t...................t....................t.. . UJ ........ A . . 8 ......... i
. . . . . . . . . ....................)........................................).........
i
- ........... ...... . . )
H 3 ..............................
- 1 8
X< . vd r
.s-y ......... .
a Ae D !. !. . _"o T ! V Oh Oh Ob 0'O 0?T O'OT Oh (mq0 ssm
- p. 3-16
' *QM g'h gag . ! Ab e,4 -
(m%.)' ----. O es
...................i...................j...................i...................i...................i...................t--
T-4
..................{..................g..................4....................q..................4....................p
...................................... 8.................... ................... .................... ....................
.................. ...................{..................
4....................q.................. 4.................... -
. .......g..................4....................g.................4....................
t..... ............l....................p...................$....................
..................;...................(..................
. 4.............. g& ...t................... 4.................... .
CO i. !
. !. ES
~ .
i i
~
4 . . . p . . .
- :* : N
- m : . P A :
- m : d;
. ~ .
,e T"'T*
4 I
........... ......q.................
. ...................j......... ... 4..... ...
.............g................... ,. N
.O .
'm .
r H
..... .............g..
. ................;.....................t...
........;...@...............t................... ,
..................(...................g...................{............ .....j...p............{...................r H
t....................;.....................t................. t..................7 m
... ................{.......................................{.............
F-- l : : : : : 'A p ..................)......................... \ v :
.............t-
- = *- - ----l=*--
=
.. g O?
Os QH
................t................... ................... .................... ................... ...................
- 03 o N
. P,
-4 .
m
- :. : w O
...........t....................;...................t....................
_ pet g 1
>p : :
l b ! ! ! ! O i en F7 *
?
F-4
............................. ........... ...................O
.- v-4
,1 e- ,-4 .
Ao e
..............................;o.~.--.-~~.;.o.-~~~~~~..; . ~ ~ ~ ~. o o- o ..
J ........ 8 > V' + y .-,
..........g....... ...........l....................g...................t...................
J p '.
..... .. a .........
~ O .
M <'" *v. l Ls l .....................................................................!................... r-. 0 . s .
? g . .
;-+ : : : :
. .... . 4 '
s-w An : :. :. O i i e i i g 00000 0'0938 0'0003 0 096T O'00ST O'099 0*0 (m0 3-17 ssvn
O : : : : W i . v-4 b m ....................t........................................
............... .t........................................t....................
~
,4
.t................
Q .
...............w. ................... .................. q................... .... .............;................... .
Ai . i
. i. ~ :.
. . i . .
..............d....................g..................4....................g...9.............t....................
s
.s s s
. s s
sg
.................q...................t.o-o-o.~~~..t.oo-~~~.~. .
~.-~...-~~~4.o-..-~~K.-- s 4 .
g
. . . g
. . . s .
..................(...................{...................j...................j...................i.............
A H sa
. . . s
. . . H m
-4 . . . . .
, w.
*Osg C)
.....................................i,...................i....................i,..................i....................
- . "4 O gb:3 6
ec
. , e 7-4
. ........ .......g...................;...................g................... ...................g................... . g O .. .... .........i....................!....................!...................!...................
O p .i i
- i. i E-4
_t O 04 . p_ Q g
.........p...................;...................p...................;...................
w M o : : : : .
...... . M .........:................... ................... ...................:................... .
mA A e2 O U I :. : (f) ......... m : : :
.........;...................g...................g...................p..................
.A.4 .
a HM : : : : y ;) o ut u . . . dr .
........ ke e.
8 . .
.........t................... 4....................g...................q...................
Z . . 1 C) . 4 . I a . .
- a.f < .
A *w O : :. :
. ii
( . g 6 e i i i -
,.4 02T OtT O'O 01 OT 02 0'O
,oh 3-(mq0 18 ssvn
a s
- g
-O
.... .................................g...................g.................. 6.. ... ..... ..
................ 4..
..................g...................{...................g..................4....................;.
......................................;.......................................g.......................................,.
...............j...................j...................i...................(...................i....................I.: . .
...........(....................g...................t.................... g...................(...................
.................. ) ............
.......i....................g...................;...................
. . . s h
..................;............. ... g...................i....................(..................4....................
C !. 4
- g
_Q
..H
.....................l.................... .................. . ................. -
1.......................g...................g..................................
.... ............g. .............. .
.....g.....-
....{...................g...................(...................g...................(..............
. L m
...i................. ...: .
H4 ................:.................;......................................(....................................... l OHH
. . . . . . . . . . . . . . . . . . . .g....................: . . . . . . . . . . . . . . . . . . . .t.....................
.......... ........g..................... .
O H
% F :
G
-1 .
. CQ c
..g...................: ...................t.................... ....................t....................
s .
%./
g.
"3
. to
- . N 4
.............l.............
.....:................... j; ...................
e.=4
; =
h4 O ! ! I O !.
~
. A
? ! . O,-(
..........................4.. . . . . . . . . . . . .... . . . . . . .
y . . ... M
..... ....}...................j..................j..................j...................
r.C .. ... . p .
..g...................8...................g .........e..e..........................
........ 8 .......
...................)...................
g ........ .. ...... ..)...................
-9 . g g;A .
H M
- yi
.... ....g...................g..............
J
.( ,
g.:J . .
#y . . .
...... .. g8 ......
w< M : :. i. M :. . E. e U<
. V g v& w m .
C4 ew ,13 . g .
. g .
- g . .
e R A 4
.e
- e. e. .
_Q g i w/ . i e i 0't i Old O'O 0'01 00 0'9 0 31 pn (m0 3-19 ssvn r..u- sh s a-a,
*F
ki o
\ _
p...................)...................p.............,.....:,................... 4
,O
...................t.....................
..................{.................. 4...................{...................q..................
. . 4.................... .
. ..................................... ........................................g.................... ...................
4...... .............{.................. 4................... 4....................;...................
.........4....................g...................4....................(....................;................... . .
- . . .P .
- 4 : .
.:...................I...................'...................p.(.3...............:...................
- : n 8:
. A .
- :. : : N ^!
. . . tn .
.......q..................4...................{..w...........g.................... .
!. !. !. $ o!
O. t .
. .s . . .
.c .
H' .
.O
.q 4....................;...................q...................;...................q...................
. .. ...... .......................... ...................t.. ................. . . . . . . . . . . . . . . . . . ....................
..................(................... : ...................(...
)-l . ................t....................{................... .
. . . . . . . . . . . . . . . . . .;................................. ...... g.........................................t................... m F- ..................
....................{.......................................(.......................................
ei s O g
.t-*-***-**-.t**--*t-**....
O O y
................ .g................... ...................q................... ..................q................... . M .
-4 . :. .
N./ 4 O
. . ,. .ta-
. . . . ~
.. .... ..........;...................g........................................p...................
. pel H : : : . >-4 w i i i !. 8 O
--4 :. :. : : e O
C;3
)
...... . ga ................... .
1 O . - . . i br- (
~
........ o . . . .
d ...... ..
.............................p.......................................p...................
E. i : : : 3 . . i. . O
- e. .
V. r, 8 . o . . c y 4 g . 4 4' ......... .................. . . . . . . . . . . . . . . . . . . . , . ................... ................... . q' < : :. :. : Q se . . n . q
. , ,. a
\ . i O H 6 6 4 6 0'008C Ot930 0'0093 0006T O'00ST O'029 0'0 inq1) sgyg
-20
s
.o C3
. : : : v4
................M}...................*...................[......................................{...................,.
^ .
m
- t. . .
.............,,,.1 .,
O- ........ ............................... H. . . .
.............A... . . . . . . . . . . . . . . . . . . . , . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . , . . . . .
+4
~. .
Q o.
. . . 4 .
..............m....................g.......................................t........I.............................
O . (G. . t O .
. t .
.................g.................... ......... .......... ........l....g.....
g
, . . g .
's s
! ! 's !
.................4...................g.... .
..........(...................t..................A1...................
. . s.%
. . .g
. s
. . g y . . .
. . s s
. . . . . . . . . . . . . . . . . . . > . . . . . . . . . . . . . . . . . . . .y...................p....................g...... ....>.........s.s........
. s
. . s
. . . . s 5
H *
- :. :, :, \ .
- :. : A
- . M V" ^J . . .
Tj
- o w :
- . :. :. . . A' 1
O HH
- . : :. : + O H b
gC
- gW
.v '
F, . . .
- : : : . d
---t ......................................,...............................................................................
Lr O . . .,4
...... . ...... ..{................... ...................a...................
W b. C p .....
.............................;.......................................y...................
4 y
- ! i i i P g ........
.........j..................j...................j..................j...................
W . . M
-4 ...... Q u .........t...................:...................)...................)...................
ds .
~
O .
........ w . . . . . . . . . , . ................... ...................,.. .................. ................... .
M A : : : . i ! ! g H~ : . - 4 m : . . s A .........g...................)...................t...................)................... a h *w A : : : : X 3 O j .! l l S Y< . M o 8MY
- y. .
>- .........t....................t....................t....................t....................
v . y . a
, 4 w A- o .
O ! : i : : n i U :. i
. i. i. e i i i i i y
- OTT O'OT O'O 0*9 OT 02 0'0
,0T. (mq1) ssyg l
3 -- -, ,
dog houses, or nay be discharged directly into the lower compartment volume. Each of the dog house areas does not contain /~ k_) cn igniter. Ignition in such a confined space may not be advisable. Instead, a hydrogen mixing system inlet is provided to circulate hydrogen accumulated in these areas into the air return fans where it eventually becomes mixed in lower compartment air. Release points, as mentioned above, will not be know for the break events. For large breaks, the release may occur in either a hot leg or a cold leg. The hot leg break has the potential for releasing hydrogen near a steam generator dog house where it may collect. However, the temperature at which the hydrogen is released will probably be in the spontaneous ignition regime for this break and it will burn before it collects. Small breaks, as typified in the S D2 sequence, may occur in a wide range of locations, including the letdown line and pressurizer attached piping. Hydrogen from these sources may accumulate in the pressurizer dog house or steam generator dog houses, or may be discharged directly into the lower compartment volume. Each of the dog house areas does not contain an igniter. Ignition in such a confined space may not be advisable. Instead, a hydrogen mixing system inlet is provided to circulate hydrogen accumulated in these areas into the air return fans where it eventually becomes mixed in lower compartment air. D. Basic Assumptions (v) 3-22 O
In each of the events used, the sequences have been described. These accidents result from the gross failure of emergency core ( cooling systems either directly or by consequential failure as in the case of loss of ac power. The inputs used for thase simulations are plant specific and are factored into the MARCH computer code. Probabilities of the sequences are extremely low as discussed in Volume I, section 5. The hydrogen release accidents up to the point of core melt are the ones considered for the purpose of the 16 aiter system design. At core melt, the AD sequence generates hydrogen in a rate that would probably exceed the useful applicability of the ignition system. Similarly, the TMLB sequence has a large hydrogen spike late in the accident. Work thus far has, therefore, concentrated on the response of the containment to use of the igniters in the S D sequence. This accident is similar to that of TMI. 2 E. Calculational Methods Calculational techniques used to generate the gas and energy releases mainly consist of the MARCH computer code. This tool is discussed in Volume I, section 6.2.2, and Volume I, Appendix C. The studies which defined the probabilities of each sequence and resulted in the selection of the sequence for further study are detailed in Volume I, section 5, and Volume I, Appendix B. (3 N.! 3-23 , . kitsa . 2:. . ~a .. . -. ~: ~ L::: a
References:
@ LII-1 " Technical Staff Analysis Report on WAS%1400-Reactor Safety
} Study to President's Commission on the Accident at Threo
!!ile Island," Kemeny Commission, October 31, 1979 III-2 "Miti Sation of Small-Ereak LOCA's in Pressurized Water i
Reacter System," NSAC-2, March 1980. 3-24
~~ - . , _ ,
SECTION IV DEFLAGRATION AND DETONATION l A. ' Overview i r The behavior of the hydrogen combination process with oxygen i-4 is a complex phenomena affected by many variables. At hydrogen concentrations in air as low as 4-volume percent, burning can i } be sustained in an upward direction although burning efficiency j will be low. Flame propagaticn will not be preferential in i direction above about 9 percent by volume and burning will become essentially complete. Above 18-percent detonation can occur. l Several variables influence the behavior of each of these burn ' f } sones. In the course of a degraded core accident within a small volume containment, each of these burn regimes is possible; and many of the influences affecting them are found. If the burning j_ can be controlled within the lower concentration regimes, the resulting containment pressures and temperatures will be relatively mild and manageable. i I
- The success of a controlled ignition system in preventing i 1
~
excessive containment environm' ental conditions is greatly l j dependent on the many variables which affect the burn
- characteristics. These include concentration effects, heat transfer effects caused by mixture composition, temperature and pressure of the unburned mixture, presence of steam, and flame propagation. Each of these variables will be examined for the l () influences exerted on the combustion behavior of hydrogen and n
1 4-1
.- , , -- . - - - . - - . , , - , . - , - , , - . , . . . . .- , _ , , - , . , . - _ , - - - - - .y
1 therefore on the design of the igniter system. [ B. Concentration Effects Figure IV.1 (reference IV-1) shows the flammability concentration limits for mixtures of steam, air, and hydrogen. At a concentration of approximately 4-percent hydrogen by volume with approximately 95% air, combustion can be sustained but inefficiently. Apparently, below 4-volume percent the effects of heat loss away from the reacting gases is sufficient to quench the reaction. This lower concentration limit is not significantly affected by replacement of air with steam until the 0 e ncentration limit of 5 percent is 2 is approached. The limit is, however, sensitive g to density of the gases in the presence of steam. As indicated V , by the dashed line curve in~ Figure IV.1, an increase in the volume percentage of hydrogen is required in order to achieve flammability. For example, at 40 percent steam the volume percentage of hydrogen required has shifted from about five percent to about 12 percent for mixtures of constant pressure but different temperatures (75 F and 300 F, respectively, in this example) or, i.e., different gas densities. The presence of steam is very important to the operation of an igniter from the concentration standpoint. It is interesting to note that at concentrations of steam above about 35 percent, no detonation can occur and above 58 percent, hydrogen in the l l mixture cannot burn. Hence, very different behaviors may be ()N im l 4-2
O Flammobility Limits igg o 75 F-O psig 10% reaction container
---C---- 300 F - O psig g ambient temperature 216 F
--&--- 300 F -100 psig 80 \ -- 20 IO% reaction container 100% reaction -
c om n 216 F c
/ }
ombient 280 F 100 % reaction container ent E F 60 ' , , '. 40
'. ap f -
f
/rAssumed g p' ! ! ! / t detonation limits ,g
!I 60~C 40 's * -
O
/ '
Flammobility limits o,f \ [/ \ / - \, 20.7 / =4 7 +- V- . x 80 00 80 60 40' 20 0 Hydrogen, v/o Figure IV Flammability and Detonation Limits of Hydrogen - Air-Steam Mixtures O l 4-3 1
observed for different typss of igniters in the various compartments of containment and for different breaks. O An example of the problems which the above may suggest is a mixture with a steam content around 55 percent and a hydrogen content around 10 percent. This would be just outside the fla== ability limit and a spark type igniter would not ignite such a mixture; however, a glow plug may sufficiently heat up N-the surrounding mixture such that the local steam concentration is lowered and ignition will occur. This could propagate by virtue of preheating the surrounding mixture and changing the concentration just ahead of the flame front. Another possibility, however, is that the hydrogen concentration in the vicinity of a glow plug may diffuse away from the plug sufficiently fast to render a marginally flammable mixture i inflammable. Fortunately, studies of small breaks performed for ice condenser J l plants suggest that the concentration of steam during the long
~ \
term when hydrogen could be expected to be present is sufficiently low to have little effect on the limits and, therefore, the behavior of the igniters should not be I significantly affected. Concentration effects caused by steam are a concern only for the upper region of the ice condenser where the removal of steam
- can cause en enrichment of hydrogen. A mixture consisting of
~
10-percent hydrogen, 45-percent air, and 45-percenc steam (^\ w) 4-4 l i
-,--t .-..%. s u, , . u gggggeg:ggggggp ,- .
entering tho'ico condenser would discharge into the upper plenum at the detonable limit. Air is always present in this regior. which will dilute this flow; however, this region is likely to have a high concentration. Burning will be likely in this region ! if an ignition source is present. 'I 4 C. Impact of Water Droplets A number of mechanisms are known by which water droplets dispersed within a combustible mixture of hydrogen and air might
; act to limit combustion and/or detonation. These mechanisms include:
(1) Providing a large mass of dispersed water droplets per unit volume of gas mixture is an effective method of dramatically raising the specific heat of the mixture, decrease heat
}
diffusion distances and times, and increase surface heat transfer area, thereby limiting temperatures and pressures should combustion occur; (2) The presence of supersaturated water vapor inhibits chain bonding in the hydrogen-oxygen reaction which may preclude detonation by slowing the kinetics of the reaction; and, (3) Many small droplets of water dispersed within a gas provide i an effective particulate damping of steep-fronted pressure waves by viscous shear action, whereby the extra surface energy of the smaller droplets comes from the wave. These O 4-5 1
.+wt*-_- _m,-w
-___e,:. -
~
- a. ;W. ' e < - u+ ;; $ V;% o *% . if
smaller _ droplets are then more effective as a heat' sink in the ensuing flow behind the pressure combustion wave. Since, at the present time, it is not possible to predict the combined effect of these suppression mechanisms acting together, various experimental programs have been, or are in the process, performed. These experimental findings are the subject of this section. Different modes of combustion have been observed over various ranges of hydrogen concentrations. An understanding of these modes is important from the standpoint of mitigative concepts designed to preclude or control hydrogen deflagrations. For instance, it has been established that flame propagation is erratic with incomplete burning over the range from 4 to 9 volume percent H2 in air. In this concentration range, less than 50 percent of the hydrogen burned. At higher - concentrations, 20 to 24 volume percent 2H air mixtures, detonation waves were produced but not sustained. Detonation waves are well established in 28~voluma percent mixtures. In dry air, hydrogen exhibits the following characteristics in addition to those stated above: (1) A well-established rhme would not propogate in 5 volume percent H
- 2 (2) ,An ignitor which sparks 60 times per second across a 0.050 inch gap would not ignite air mixtures up to 9 3 volume 4-6
' percent H ; it would ignite 12 percent H 2*I* DUP ** 2 i ! (3) A flame will not propagate downward in less than 9 volume O percent H mixture. 2 i (11) Ignition of mixtures containing 12 percent H2, laminar i i i O
=
1 O 4-7
,w ,n-- - - - - - - -
, , -- , , - . . , - . - , , - - , , - . - - , , . - - - - - , . , - , . , ,..,-.,a., . , ,.g,., - , .
,,,-w___., w-e.v, ,,,-, _,_
___y-ww-,,,,s
burning occurs with a flame. speed of approximately 5 fps,
. pressures increase by a factor of 2 to 2.5.
q b (5) Turbulent flaming occurs in mixtures containing 16 percent
'H2 , with or without water spray.
These dry air characteristics contrast with the following water spray results: (1) A well-eatablished flame would not propagate at less than 7 percent H
- 2 (2) Ignition of mixtures containing 12 percent H , water spray 2
agitates the mix and may increase the flame speed up to approximately 50 fps. The spray also absorbs much of the i heat energy and the pressure rise is limited to a factor of 1.8. (3) Water spray was effective in quenching detonation waves. l. In summary, water spray has been shown to be an effective mixing j mechanism which suppresses the temperature and pressure rises l caused by combustion or detonation. It is also effective in cooling and suppressing ignition. The containment sprays should therefore be useful to prevent detonations in the upper I compartment during their operation. In addition, the sprays will serve to suppress a large temperature spike by absorbing considerable amounts of burn energy. It should be noted that 12 percent H mixtures were able to burn in the spray 2 atmosphere; therefore, it is not expected that detonable
) concentrations of hydrogen can accumulate in the upper 4-8
The igniter compartment prior to ignition by the ignitor system. design uses a spray shield to prevent the direct spray of water O onto the glow plug unit to ensure proper operating temperature thus no problems are anticipated from the concurrent use of the j containment spray system and igniter system. l D. Flame Temperature 4 Flame temperature is one of the most important factors that characterize and influence combustion behavior. Flame temperature refers to flames burning at constant pressure with The flame no appreciable external heat losses or gains. temperature is a function of the mixture composition, initial mixture temperature, and pressure as shown in figure IV-2 (reference IV-7). The maximum flame temperature is obtained () with a slightly rich mixture (approximately 31 percent hydrogen in air). These values drop off regularly on both sidr.s of this Initial maximum as the flammobility limits are approached. mixing temperature has the effect, that except for mixtures near stoichiometric, the flame temperature increases almost linearly with initial temperature. In very rich or lean mixtures, where flame temperatures are low and there is little dissociation, flame temperature increases degree for degree with mixture temperature. -As the composition approacnes stoichiometric, however, dissociation becomes more important and flame temperature becomes less dependent on initial mixture temperature. The effect of pressure is such that dissociation
/")T
(_ of the burned gas is favored by reduced pressures, so that flame 4-9
.aC ,: ":
, _ _ . . , _ . - . _ _ ,_ _ _ ~ . . . , , . -
4 o
~
d 2
,.', ,i, *, , , .
[.d . .' .".*i?.. E "..d..i.'i. N..*..bb1Ikfi
- . . . . - ..3_,. .g. .:: .n ::ag ...m _o_ =_)....
~..,'
-. ,.;;~,; ,;; ,:77- .
~3 r - f: :ii31iiidYN N!! ;*k .b3.3i . * *9i.Y b*-' i**T--' I
~~ '
.'"' [[
=:=.[r b WW. '.
' " : iib.i=5: h j5Mh@$:lM $ iif SilINEIU- Y5 N5 : %I 2i- E S_ M i =hE MMn/ iViW::iNMid Pressure, Et" nii -% 5% i] Q "'y [F'.
("N 200c :...;, ;i14 3=1.5= 4;gigep eimp;+i::W P, o t m 4:q jgg-- hip =: 1iali:pG=jE M=; =i ==@ (,) M. I:[M5551thdi'= Mi.k@r iE H"dlMid kWif XEp$5MME'M i. .f @Y @]"$ N7 f.;.'.
;-j'1 ~ ' 5i~.55. U= 'I"r N "j. . ; E $i ! h'
+
- '- r s
NEiN NYvl"< =F "MM'. Tib,idik aE'i bliiPMrO iniMENM5 E!*s #. 3 2G00 Q.i..s . ;.s 3;m/:ppy e ;iiw . g;;Ml RiddeVi44%
@. K". .[fi 91 $"" j gi tr . I b 'f- t N 'II-Disiw n= M / MM : . "Ethir .M*!2 4M.T*M PM Y!CM '.313f
..-.,_ yetM/tre 'WMNa meNN# Nw= M. W=9 N 5 %T ~= ~
7400 . %s N,'.._i:.py. W.iWf,C $
! '. T.. i t i$ i E . %".
- u Y{" 2; '="h ii
'"i N N M @
;i '. ,p.-
F g M *_'. iiN.. \ OM..
* - a .n. J;qi,c 4,: :Mt4p
} *H E
~
= Y -q3 "* '
jI.
. a 5:y :'t:ff[:dN[. /P. hit % @i 'V ;M.I
. . *. .*" t .
3 .iNIL'h3h $ % U.Missifai r.rf.af///ojp.s 7/N#ei WN @ .4SE eiM =$=- 5#= M- E-'-:.-M w.Nai= win:iNuiEGE Wat==M E=ENER =='g .d
'S' .
.G;r.b.GG Y$ f ff5 /digg//=.5tE5U Si N dMuMN
/Aufs E-Si.NYE E $F*b#5iN WI hN i*M "$$W9E i- i r Qa ~ ~- %
-S TSEilA\ M SVIN. E MS 000::.i:s.Q_mNs/u,Ms=iEW5
& km.mm mam m m Q U.TbisM"MW=4***
E w w wm em= a=l M *MP."?& r. . < !
=
E IMeNEre,'! f Eva
- 1800 i;'.;;.f;fefigsF/ii3' t ~=f-tal F @ #' W Elis M 9 N M T D =i-g is p ,.y,. g g i"IME4MEM g g ap g gjg gsrgs:is iN:=e 5t%904 MAN
- E =WF=i =9 ii%E W%"=Ed~~i d=
O Es_i=.iMUJ/? E!Ori=EE l ia' S44#5MN ME V
- k% MiWl?[ M 25 b i400E2 J .u p fg;ze u= sci-sq a am 5si@ E.ii M 515i MN Mr:VE 6iE Fi4M = =:"" EWEE E N.ilj=..aGy'!:hi@/5 f dei h EU EEiEE!EIE iY-EnW ^lM.$ EYNsN bN 1600 di.. i.-Jj F=hh.- -..- :iiinE iEL-Esyi:-3-M is.## iMN%k. -YE. #-SitThi Ei5 N= Nr-*S55
.E .y.' Mp ?d &f.j&q.ii.iis!55 ) X= E5 5\{gi 55]Ene Wii3L=Y-M ;!t M ==*EE1E M**%*% M 5M.
$ :h si U54 El31; ele'il"ih 5M 'W E M E N N :
. 5 =[rj/Nf gapids 5Nnit.2 z nia is a:tcits Edsjn4 E" -Wii @iNE M iid 6"EMEf3Mii - N@.=. EWio00ET; N Y b - @"
g49g [f. .f,;1:i7;[,; } g :n,qgg. -g.. .g.i . MNpf-HiGFJM
= =7r. 7 ..,]yapning , n. -. :..-. t.g 3.j.gi:
E : a = f-
~= =le u
, ; .g3._-
M,-k.igDEjgy. 3#pM M $h.}.isp=s
'a,. = sis ze"# M e=r@E *=0ais GM*
*=*
~ni
. eMSD.- . . .$b...S ?.: "
- f. *:
df =i-ifibtF". 5 ab h -
~~ % 2 b -i ' . -
1200 e_. i :h. m ;;u. - m...=...i:a.i.. e...Ls.w..- i.mi dii -ii: ;iE F=.-dis idl;i.k &W VW iWh 9 e2 ' o E 1
?..
u .a f-.:- /m. .=/g.=u;a r: c
- 6. .; -
.= $ ii3Q 2?.}i!YTh*M hif.f.b iN5hi=U~
. - - - - - ~ -
k, k- I'* !
- b M..\. 3E Eim h."- =
c.;
-~
d lij t $ E
#f h. O
[ 1000 hz: ~enranwara
; ; u- J 15:fJe/:= ?.:[..-=-r.
m@ =winalm,wm r e-- - wi c.F.!5 Esta.:. 8m = =a=* ii=Ei :--:asi :di.E;niti-E dEMJ@E iM: EM5 D F E4N=E iiE mm&=J% i E= i =r:- - 9 m ta ~=b - . - - - - - -
-; ;fr_1 :r. : ::- -v 9.g[ Y
. ;rs--.:
..e q . . . .-
;_ .s ;
==H/[2 pEl' pie ldidiU*fNi.II@i=JEb T'5!5N$$7EEG 'iF "E9 M - '""" [-
N NM .= x._ :. . M-
" 5N yi uM._ .;_h. , .-ih_.iH. = iii.iii@@ flitis iE.lisL5% @ES El5
!.5 $ 5 . ;-_
a49 = , , = ' 5 'p..m g..
.c. - :E.. R. i..iE.ME. -
mig ;sts Wiiit Mei43El'iMs.. .. _.. ..e!s = .. = r= *
-~
_. fi= cafa# 800Pif5r+:iay;E!E Elis EEM AA" =E # *298 i NN=i"9 ist 7 =~ g- .ig:is sidii Ajea Wiglia iMMi shaia;isii=Pik (=!N
;;E i M =5ms
-j ~*
iisis@ni:.=liig Uh5E kbi;fi fh
#ta#WE Mid= #Wai Wi@= Rs# #is EIA is iE2 I*
W N$ C7{ n
.f. ?
oca:g w g a;a m we .aa gm m wi hw'it!Wh, tB_ m;;=bt;.. k....b.M..a- o. .-
.W. .
.=: .= c soc e p:/l?J=, 3:,si gi;g g;jggg;;g g;g pgjiig:iimg gglE ggr := =. =4. . .. _agm
. - - r 1g%..; p. 7.-
i m q7D,, ~E "N, M,' tg i='ii si fia 4i45 Fine -ifi% "9i! EMETE stEEM N 5; M'EW m e""
=%ghg .M
. ww.
O!SjM5s i m i::i=e. a s#illi!!! m It is-iWiE a=m ME!Eii# EEtEiidiMMEW l I
-* =
e?!Ni '.?:NN 9N 400 f ri.] U E E=s".i=- =w4tiw igin MEiEmii ds#! MEs Essi3E!!# # E= ~ C =:= Ei M.EN f.,...... . . - . . . . . l . -- . - k*
- E5 . '
.. .- ;,n- ; .,_ .d : 4J iii35i*i3kd3UNiiiN"k!ib:N!5 inh"*t =
~
'#' ~'
;.g :
.'J- . - U. N. 't .U. Id. idf-- - r--
- - ..-- ETb5 h i *
~ ..
.M.#M W.,'.'..
. --..&- - . - ~ .
200 a.pdifis3 sjis: u...... iip iinaw giajingisjv gg is,5 il+r % uui:#iMM p 7-5 %..~=. . MN.n .- k(t :iis= 'i=hiii Hii# EM=#'EM5:E E" iWWMiWW= M SR 7 ! .. 7 M Ei5i l. E 4U S5M[i-i5 M Ti !i 45 HUN @.*3= ? # h Q m - 5Ei 29 p= -;r4 =ihF="ii# 5: 1a ip:il.n: 5p Ei!:i : siHE I "~.Mi?E k Ai@ d4 LEA 3=I M i5 8
-=
! o -
4Ni E ENN4 NF
? &
fibi{NiMih.=.r 70 80 90 80 20 30 40 SO 60 10 0 O Hydrogen in oir, percen' by vclume Figure 2
\'
4-10 6 I M* -
<w. 4M.,Q 8:%$ % k h '. -
' w a** he- I E .- * "A v e=M ? 'k . f.\ # 4-% a
temperature decreases as pressure is decreased. However, the
/~ - size of the effect depends strongly on the general level of flame k_;)
temperatures produced by a given mixture. Near-stoichiometric mixtures show a strong dependence of flame temperature on pressure, while lean and rich mixtures have little or no dependence. Mixtures that are quite lean or rich have flame temperatures too low to cause much dissociation, thus, pressure has little effect. E. Flammability Limits i The flammability limits vary, depending on whether they are measured for upward- or downward-propagating flames, because convection assists flames traveling upward. The rich limit of hydrogen is the same for both directions of flame travel, 74
' percent by volume in air. The lean limit is affected, but not in the usual way. It is 9.0 percent for downward propagation; however, for upward propagation, there are two lean limits.
One is called the limit of coherent flames; it is 9 percent and is the leanest mixture that burns competely. Leaner mixtures down to the noncoherent limit of 4-percent hydrogen are still flammable, but the flame is made up of separated globules that slowly ascend. The noncoherent flames occur because of the high diffusivity of hydrogen. The flammable range is widened by heating the unburned mixturen. The lean limit occurs at lower concentrations, and the rich limit at higher concentrations as the mixture temperature is increased. O~ This is shown in Figure IV-3 There is a linear change in the 4-11
9equaga17 Egbrugen 9tubg s
.) .
C.
c n
j.. .. . . . . . . . . .
-o 2: . . . .. . .
- i.
I i
. . . . . . : j O
M i :
. . ,o ,
-a . s. .a o, .
5> m
- i . .~ . . _ .
ct; : : : g
-szO r
x . . o . : .. .. - -
. - )
~
e .- g t; o 4
- c. M w
co q~ a. p
. . .uvZ j
4 ~ 2. x x . . . . . : o > A
' - m . .
-u . .
..a i
a<4 7-4 .
.ea u s< ~m e- . . -
e z un< : . m . . . . . . . .
- m m
a o<- o i. o a a z,a o
. : .n
>- a .
- m .
. . m 4 . .
,a i
~
!. . '-h I : ;
<d ;
Q .. .. ... . . .... . i : . N i.. : . :. * : .
*t
. .. . - )
-o o
O b 0$9 00b ONO 09N Ob b GOI D' DT Ov0h 0thb , (a.) annivaaanal Figure IV-3 {-} 4-12
.,,g
' -' --,../__ - ,
, ' MMD dP~- . _ # *
'~
..ss .
I
d limits with mixture temperature, and the rich limit is somewhat more strongly affected than the lean. (' The addition of inert gas to a flammable hydrogen-air mixture may dilute the mixture to nonflammability. The rich limit is sharply decreased as inert gas is added, whereas the lean limit is scarcely changed. This effect varies between diluents. For instance, it takes more nitrogen than carbon dioxide to prevent flame propagation, presumably because of the greater heat capacity of the latter. F. Ignition Requirements Hydrogen-air mixtures are extremely easy to ignite as evidenced () by the minimum spark energy required to initiate combustion i of only 0.000019 joules. This spark is not visible to the human eye (reference IV-8). Figure IV-4 illustrates the variation of this energy with hydrogen concentration in air. Since the igniter system only uses glow plugs and not spark igniters this only is important to demonstrate the minimal energy required for ignition. It is anticipated that sparks may be abundantly available; however, particularly for the large LOCA where flooding of nonqualified electrical equipment occurs. Small static electric charges may also be present. Strong sources of ignition may initiate deflagration or recombination in nearby z mixtures even if they are outside the flammability limits. , L . ss Reference IV-1 suggests one such source may be combustion 4-1,3
- e -- - .- .e.- ~- , - -
y g ~n- e ,-w ,,,-n-~. ..,.,g, y< , - - ' + ,-r-,eme-nv, r , g--
O . 6, i 3 i i . 4-1 g 2 - w J o PRESSURE, P, - o, 1 - l 3 _ atm _
.a l
- .2 -
6 -
> 4 _
c .3 c:: w . z , w ~
~
z O 2
.s s .1 -
a_. - e 08 - t 3 --
- 2. .06 ~
Z - 2 04
.02 -
t , i , 0 , i , i . i , 1 . 20 30 40 50 60 70 0 10 HYDROGEN IN AIR, PERCENT BY VOLUME Figure IV l+ 4-14 l l
--,,--..-..--.--.r - - , - - - - - - - - -
- - - - - - _ . _ _v. . _ _ , , - r-
4 3 of a flammable mixture in a partially confined area such as 1 a duct or dead ended volume. The resultant strong deflagraticn
) - would propagate energy into the lesser concentrated mixtures l i
j exterior to the space. I Spontaneous ignition of hydrogan has also been examined , (IV-l ). Figures IV-5 and IV-b illustrate the variation of the spontaneous ignition temperature with water vapor concentration and with exit velocity. The spontaneous i ignition temperature is measured by heating the entire mixture to each measured temperature, discharging the mixture into l unheated air and observing whether ignition occars. The spontaneous ignition temperature is expected to be lower than i i the temperature required of a heated igniter since the -igniter () must heat the surrounding mixture to the ignition temperature and therefore must be at a higher temperature for heat transfer to exist. The spontaneous ignition temperature will influence l whether the hydrogen immediately burns on exit from the primary system. (For cold leg breaks where the hydrogen must travel l through the steam generator before leaving the break, it may , l be well below the spontaneous ignition temperature. Also for l ! hydrogen release through the pressurizer relief tank, the i I bubbling of the mixture through the relier line sparger into i the tank fluid and out of the rupture disk will cool the gas. However, for a-hot leg break concurrent with inadequate core 1 ~ cooling,-the hydrogen-steam mixture may exit the break at l-i tem,peratures well into the spontaneous ignition region. It O would then mix with air and deflagrate near the break opening. (_/ _4-15 l ~. a .- . ., w - , . . . - I
Bequagall Eghragm !biuby O
}. . . -d
. o
% J.s O
i l
! : O
- : : ..... .. ..... .. . . . . . . . . . .. .. ., 4
; ?
o 4 : : . . .
- . A e :
. a Q
W Q g :. .
. O N
- : t :* '
p
- : w P . . . . . ..
i ; : o 7. _, d T C Z : 1
- a-g m :. : : -
F O N N M H :
- p . . . . . . y J. tw H ** .
c4
- - O C
4 i *
. O Q
>=4 4 : *
. . . . . . . .. . . . . . . . . .. . . . . . . . . . . . . ............ 6.......
rn p. :. :. CO g . 9 i* i i i i : O
- : : : :* :. A
% r.-4 >4 D
- Ea3 :
d O
.......................................i............ ........!..........:!...........:..
9 9
. c4
- :. :. -? :. '
4 i i. i :. i
- 4 O .
1 C i : 1 I ! I i * *
................}... . . . . .. . . . . . . . . . .. . . . . . .... ... . . . . . . ........ o
... .. . . . . . . _g l :. * .
. g
- . :. : L CR
, i i i i i -
. . . . 6 o
OTKi71 0*MCT 0V091 07.d 1 OMI O*00ti O'MC1 O*E3CT OM O W 210'0331 (a.) anna.vusawaa, FIGURE IV-5 p s 4-16 t.
~ , . ,,m,. , \[ '
.%.,. *o%, .. a V-=+o.g y r' s ,. w - w "*
h.*P.*.."m3*
.n*es .%s d* '
Dequngal7 Egbragert retuby
~
O
) C5 s : :
- :. : i : i : u l,t. C.
v : 4 .
. ~
.e
.gs W : : : *
- gvm y 4 : : : : .
" !. : . :* : : M O
b : ! !
- g CD
....3..........:...........................................f _g t'3 \=
ai : : :
. ! i J
O I 1 .i i. .i i. i !
- O +
H . :* % i i i v p. f, 4 . ....
. . . . . . . . . . . . . . . . ........* ... . .... .. . . . . . . . . :. . . i... ....i......,, [Ct ~
H w :. : :. o
. . : i.
C3 M
~ h w
O
* : i
$ i i i i I i O
- . . . 4 ....i...........i.........:.........i.........:......,,.;......,:,,,,,,,,,
i {o CO
< N>o. i i
i i i I [
+
. e. - :. : :. . : : *
. E-*
4 D w.<e
* . e-.e O .. . . . . . .
-. Oy e
. . . . . . . . . . . , . . , ,,, y
., . w
- . 3
- e o
O
.! .. . ......... . ...I..........
l .
,5
.i : I *
. i ;
~
C
. . . i.
g oosi6d: 1 64r ud:t orkt oais hdW osi ovkr oidmoun (a.) aan,Lvasawat s FIGUE IV-6 4 4-17 i I
~-
'j d burn by the above mechanism, If the hydrogen does not ignite an to function to prevent it will be important for the igniters The proper m
s., U the accumulation of largeanhydrogen concentrations. n estimation of the functioning of the igniters requires e to initiate combustion. necessary igniter surface temperatur the flow velocity past the i This temperature depends in turn onselected is expected t3 mainta ,. The glow plug system arface. igniter. F at the element temperatures between 1600 and 1700 I N allowable flow velocity A preliminary estimate of the maximumhas been made using past the hydrogen igniter glow plugnm pentane-air .~ experimental dath 9; (reference IV-9) produced for a A ratio was determined from s in data which heated solid 5.: - i both pentane and hydrogen atmospherewere passed throug I spheres of varying diameters i ately 13 ft/sec and air mixtures at velocities of approx mignition temperatu approxi:ntely 4 f t/see and the ignition The ratio of the hydrogen to pentane of 5 cm passing measured. I was determined at a sphere diameter N temperatures t approximately 13 ft/sec, through the fuel gas-air mixture a l gas to air mixtures for and at close to stoichiometric fue both gases. e This ratio was then used to adjust a curv .h rod with a velocity. given for a fully heated 1/4-inct d agairmt stream at pentane ignition threshold plot e toiciometric mixture of
)
This curve had been evaluated for a s s,_- 4-18 my
~. %.__ - -..
- . ~ . . D +- [ *
- v %- %
t
pontane and air at an ambient temperature of 160 1 10 F. The adjusted curve is shown in figure IV-7 The figure shows that if the ignitor is maintained at 1700 F, a flame will be sustained at flow velocities up to 158 ft/sec, which is well above any expected containment velocities. While this curve is for a 20 percent mixture of H in 2 air the data a used in determining the ratio indicated that the ignition temperature was much more sensitive to the stream velocity than the percentage concentration of H 2
. This was noted in part because the 10 percent 2H mixture at approximately 4 ft/sec Save lower ignition temperatures than a 20 percent mixture at 1
approximately 13 ft/sec. () It should be further noted from figure IV-7 that at rod surface temperature much below 1500 F a flame would be hard to sustain even at very low stream velocities. t G. Flame Speed l Laminar flame speeds are measured using several techniques, although each has sources of inaccuracy. Recent theoretical calculations of laminar flame speed using laminar-flame-structure models are in good agreement with experimental work, as described ) in reference IV.10 . Hydrogen-oxygen reactions have been studied to the point where the detailed chemical reactions of importance j , in hydrogen-oxygen combustion are believed known, and the reaction rates are fairly well known. For hydrogen-air mixtures 4-19 s
I I I I I I Figure IV *3 Rod surface temperature at ignition threshhold versus strea n velocity. 20% mix-ture of g in air. 1 2000 1900 Q 1800 L M a 3 i / h1600 x 1500 l 20 40 60 80 100 120 114 0 160 180 200 220 240 260 STREAM VELOCITY (FT/SEC) 4-20 ,
e at atmospheric pressure, the thickness of the laminar flame front has been determined to be about imm O Figure IV.8 shows a comparison of laminar flame speed for hydrogen-air mixtures. One investigator has computed the laminar flame speed and compared his results with those of other experimenters. After corrections were applied to early work believed to be in error, the revised data are in good agreement I illustrating the theoretical understanding of laminar burning now exists. Figure IV.9 shows the effect of different ratio: of oxygen to nitrogen concentration on laminar flame speed. The maximum laminar flame speed of hydrogen-air mixtures is about 10 feet /second near a concentration of about 42-percent hydrogen. As the flammability limits are approached, the flame speed O becomes much smaller. Diluents such as nitrogen reduce flame speed by reducing flame temperature, apparently by removing energy at the reaction front. Steam also reduces flame speed, but by less than the amount expected from equilibrium flame temperature considerations. Moderate changes in ambient temperature and pressure do not For instance, a significantly alter the laminar flame speed. 90 F temperature rise above room temperature, the increase in laminar flame speed is less than 0.6 ft/sec for pressure change. In the range of interest for reactor containments, the expected I variation of hydrogen-air flame speed with pressure will be w small. 21
- ~.n->.a - . -- .
- . u, g -- - ,w .--=,ew<- - - . ,7, yy. .e, . -,. . ,,.m-e9-re c --ee--^
4b0 . . Q O O ! 300 - 9 [
/* +8x
+
g 0 5 g . tb . x [ \ Ej 200 -
> =x O x Z o I Z -
E 100 - D ,/ STOICHOMETRIC
?5tXTURE f f
% 9 F f 9
'O4 0.6 0.8 0.2 .
~
HYDROGEN, PERCENT O LAMINAR FLAME VELOCITY OF HYDROGEN-AIR MIXTURES T = 298 K, P = 100 kPa 37 Line - CalculategValues, Warnitz e Jahn ( x l.2) 4 Scholte and Vaags ( xif)24 , O Edmondson and Heap D Bartholoms a Burwasser and Pease c Gunther and Janisch CG Unther and Janisch Andrews and Bradley e Miller, Evers and Skinner x Gibbs and Calcote
& Smith and Pickering C*
Figure IV-8 (reference IV-10) 4-22
Figure IV-9 (referenceIV-9) 1200 , , , , , , , , , i O 1100 - 02
=.985 :
02 +N 2 , 1000 -
.90 .
~
.80 .
900
.70 -
800 -
.60 .
m .50 E 700 - - o - { F- - l ' I U ,40
/ t O
i 2 800 -
> .35 l a -
l . Z
.0 h 500 -
l [ _ , C3 -
.25 _
400 -
.21 l
1
- .175 _.
I 300 - 4
^.
.15 -
)1 -
- i 2,00 ;
. 1 . .
.125
' ~ -
100
/
O -
; y -
- ,; / , , , , u a 0 *10 20 30 40 50 60 70 80 90 100
' HYDR,0 GEN, PERCENT, ,
. . - . - = , ._ - . - _ .
h
' The expanding laminar flame of a plane or spherical front has
' O
\/ been shown to be unstable. The importance of the instability of the flat laminar. flame front is that the mean speed of a laminar flame will be somewhat higher than.the laminar firme speed. The laminar flame speed is the normal component of velocity of the unburned gas moving into the flame front. If
- the normal to the flame front is at an angle to the direction of flame propagation, then the propagation speed of the front
- is equal to the laminar flame speed divided by the cosine of the angle. The Mach number of the front will be extremely small, in any event, if the front stays laminar. The pressure is then expected to be uniform in containment, and to rise monotonically.
The containment structure loads will be quasi-static. However, t (} the instability of the laminar flame front can lead to " tert lence, flame acceleration and possibly transition to i detonation, if the mixture is within the detonation limits, i I A laminar deflagration is likely to be~'ome c turbulent in containment. The "self-turbulization" of laminar flames has been discussed by Shivashinsky (IV-jp. Many turbulent flames have mean flame speeds in the range two to five times the laminar flame speed. However, even at these speeds, the Mach number
-of the flame front will be very low. The pressure in containment will be nearly spatially uniform, and the pres sure will rise monotonically.
(/ The action of obstacles in the path of flames is under study' i 4-24'
by various researchers. The flame front is stretched and turbulence is promoted. Increases in flame speed have been (~N observed when a flame front passes through a field of obstacles \-) such as would be expected in the lower sections of PWR containments that contain many pipes, tanks, beams, etc. However, much of the upper portion of such containments is fairly open, particularly as in the case of the upper compartment at Sequoyah. Unfortunately, when the flame front leaves an obstacle field and enters an open region, there have been no experiments to study flame speed behavior. The cancensus of several researchers suggests that the flame speed might decrease after leaving the obstacle field. Detonation waves travel at a speed very close to that corresponding to the Chapman-Jouget point, except that "darginal gs (_) detonations," (detonations in mixtures near the detonation limits) travel at a speed lower than the Chapman-Jouget speed. The pressure ratio across the detonation wave for hydrogen-air detonations will be approximately 16, the detonation velocity approximately 6500 ft/sec, and the temperature behind the detonation wave will be about 4600 P. H. Burn Efficiency From several small- and medium-scale laboratory experiments, it has been found that when hydrogen-air mixtures with hydrogen concentrations in the range 4-8% were ignited with a spark, much of the hydrogen was not burned. The resultant pressure rise - was far below that predicted for complete combustion. /- LT) 4-25
Experimental results with a spark ignition source, shown in Figure IV.10 indicate the completeness of combustion increases () -with increasing hydrogen concentration, and is nearly complete at about 10% hydrogen. The range of incomplete combustion corresponds to the range in which the mixture is above the upward propagation flammability limit, but below the downward propagation flammability limit. " Separated globules" of flame have been observed.in upward propagation of lean hydrogen-air flames. Even when ignition occurs at the bottom of a chamber, the upward propagating flame fails to burn some of the hydrogen. References IV-2 Carlson, L. W., R. M. Knight, and J. O. Henrie, " Flame and ,_ Detonation Initiation and Propagation in Various Hydrog'en-Air V Mixtures, With and Without Water Spray," Rockwell International, AI-73-29, May 1973 The above document also draws from several analytical and experimental investigations listed below: IV-1 McLain, Howard A., " Potential _ Metal-Water Reactions in Light-Water-Cooled Power Reactors," ORNL-NSIC-23, August 1968 The above document also draws from several analytical and experimental investigations listed below:
~
IV-2 Carlson, . W., R. M. Knight, and J. O. Henrie, " Flame and 4 .
g . _ . _ _ . . _ _ _ . _ _ _ _ _. _ _ _ _ _ . . . . _ - O T C) F la m) INITIAL 2 ea e . u 's 12
/*'
y as 1., 14 5 7 w - g 71 1.2 17-21
/
O FURNOT DATA
*/
TseORETICAL d P FOR COMPLETE COMBUSTION
/ ~
/
/
1 M .
/ -
U /
/ .
f
/ \
20 ,
/
/ 7
/ .
/ '
. /
/ -
se - ! '
/
/
/
/ .
- A
/ , i oc-ec ^ct, g4 ,
0 2 4 6 3 10 12 e HYOROGEN CONCENTR ATION tv0L %)
. Figure IV-10 PRESSURE RISE VERSUS HYDROGEN CONCENTRATION, SPARK IGNITOR 4-27
Air Detonation Initiation and Propagation in Various Hydrogen-tional, Mixtures, with and Without Water Spray," Rockwell Interna O 31-73-29. Mez 1973
" Hydrogen-Oxygen Explosions in Exhaust Ducting," m.
IV-3 Ordin, Paul M., NACA-TU-3935, Lewis Flight Propulsion Laboratory, Cleveland, T Ohio (April 1957). b tion IV 4 Drell, I. L, and F. E. Belles, " Survey of Hydrogentory, Com us ( Properties," NACA-RM-E-57024, Lewis Flight Propulsion Labora Cleveland, Ohio (July 1957). ity of Gases IV-5 Coward, H. F. , and G. W. Jones, " Limits of Flammabil ;e t of Interior. and Vapors," Bureau of Mines Bulletin 503, Departmen tion "The Effect of Temperature on the Detona IV-6 Moyle, M. P., f Characteristics of Hydrogen Oxygen Mixtures," University o [ l Michigan, Doctoral Dissertation IP-195 (1956). ; IV-7 Drell, Isadore L. and F. E. Belles, " Survey of Hydrogen tion l O- Combustion Properties," Report 1383, Lewis Flight Propulsa f Laboratory. h IV-8 " Hydrogen Safety Manual," NASA TMX-52454, Lewis Researc ' ; Center, 1968. l i s IV-9 Lewis, Bernard and G. Von Elbe, " Combustion Flames and Exp os 5 of cases," 1961 During et.al., "The Behavior of Hydrogen I -10 Bieniarz, Peter P., Accidents in Light Water Reactors," Sandia National Laboratory and Energy Incorporated, DRAFT 1980. J d n Note the above reference contains an excellent discussion of hy roge # t combustion which has been used as background in this repor . f O 4-28 ___ _ _ _ - - = -
, e.mMsMkT 'I fo4" C'a.Q y M M W G , ' . .' . *
, e u -, , N 1.a. .' . .. ~ .E * ,'i w wh # e t ,o *).
. +
I.- 11 Shivashinsky, "On Self-Turbulization of a Laminar Flame," Acta Astron 5: pg. 569-591, 1973. O 1 1 I 1 l i 'O 0 l 2 !. O 4-29
+e 5ee - -e - * * * - e .g* - .. . ,,, ggg y ,*A 4why4g 9 5 e,,, +
- - . .-----v., - -, , , , . - - - - ,,w,,,,,,,--,,,y,,-,yy.ww,,,on,,,..ww--w--%e. . ere-=<
SECTION V CONTAINMENT PROCESSES AND DESIGN 4 A. Overview There are several systems at Sequoyah which, by performing their normal design functions, will assist the igniters in the controlled combustion of hydrogen. One system, the Combustible Gas Control System, directly aids the igniters by recombining hydrogen in the containment. The other systems will act as heat sinks, serve to mix the containment atmosphere, or do both. Heat sinks, by absorbing energy generated in a hydrogen burn, reduce the containment atmosphere temperature and pressure. Mixing the containment atmosphere aids in the prevention of hydrogen pockets. Pockets of hydrogen with concentrations in () the detonation range may result in very high temperatures and pressures if combustion were to occur. Thus heat sinks and the mixing of the containment atmosphere will reduce the temperatures and pressures inside containment and reduce the probability of containment rupture. Details of various systems are given in Appendix I. B. Combustible Gas Control System The Combustible Gas Control System is a safety system designed into the Sequoyah containment to mitigate the effects of hydrogen production.- This system is composed of four subsystems which will adequately process all of the hydrogen postulated in the original design basis accident scenario with 100-percent marging. They include hydrogen analyzers for postaccident j 5-1 i . g [+ B 2 i.wm A4.+ed 5.d e. r Af $
~
Tl l gyd , , 6 4 4h .. * . #.%s M ,'E -\, ,I # e \ D e -
'a .- ' '[ d i j V [ e-
monitoring, electric hydrogen recombiners for removal of hydrogen, a containment mixing system to prevent pockets of () hydrogen at higher concentrations than the general containment g airspace, and a small hydrogen purge system as a backup to the recombiners. The Combustible Gas Control System will assist the igniters in recombining hydrogen in the containment atmosphere. C. Containment Structures 1 The containment is a large structural volume designed to contain the radiation, mass, and energy released from a breach of the reactor coolant system and the nuclear fuel cladding. The Sequoyah containment is a free-standing steel structure with a free volume of 1.2 million cubic feet. The design pressure of this structure $s 24.6 psia which exceeds the calculated pressures resulting from any of the present design basis events. The containment is divided into three compartments: The lower compartment, the upper compartment, and the ice condenser compartment (see figures below). The lower compartment completely encloses the reactor coolant system equipment. The upper compartment contains the refueling canal, refueling equipment, and the polar crane used during refueling and maintenance operations. The ice condenser connects the lower conpartment to the upper. O 5-2 l J
In the event of a hydrogen burn, the steel containment shell, the internal concrete floors, walls, and ceilings, the internal /~') k/ structural steel, and the piping and equipment located inside containment will act as heat sin.c that will remove energy from the containment atmosphere. D. Ice Condenser System The ice condenser is a compartment inside containment enclosing large quantities of ice which is used to condense high-energy steam escaping from a LOCA or MSLB. By condensing the steam, containment temperatures and pressures are kept relatively low. During a LOCA or MSLB, the lower compartment pressurizes, forcing the lower compartment air-steam atmosphere mixture into th ice (~) condenser. Inside the ice condenser, the steam is condensed V and drained back into the lower compartment while the air is cooled and forced into the upper compartment. The ice condenser will assist ccatrolle'd hydrogen combustion not only by providing a heat sink, but also by helping to mix hydrogen with air as it passes through the ice condenser. E. Containment and Residual Heat Removal (RHR) Spray Systems The containment spray and RHR spray systems are designed to act as containment atmosphere heat sinks by the addition of spray water into the upper compartment. Each system consists of redundant spray headers located in the top of the upper n ( i compartment. x/ 5-3
The functioning of these systems as heat sinks contributes to controlled hydrogen combustion. Furthermore, t'he spray systems do a very effective job of mixing the upper compartment atmosphere. F. Air Return Fan System The air return fan system aids controlled hydrogen combustion by assisting in the cooling of the containment atmosphere and by mixing the air inside containment. The air return fan system enhances ice condenser and spray heat removal by circulating air from the upper compartment into the lower compartment. The fans thus establish a recirculation path ( which allows more of the lower compartment air-steam atmosphere to be transported through the ice condenser and into the upper compartment spray area. Ducts which lead to a header on the suction side of the air return fans draw air from the containment dome, accumulator rooms, steam generator and pressurizer enclosures, and other dead ended spaces where hydrogen may accumulate'. (See figure below.) When both fans are in operation, there is an air circulation rate of one lower compartment volume in less than five minutes. This kind of highly turbulent flow will also help prevent O ce==o1ettee oc averose# 1 voeuete me no 181e eeto" tio# - 5-4
e l
'm *
. .'.
- i / . } .
[ jR k G h *
=.
. I I T I ,1
=
,u, .
109 CM L (iteeg ' vto 9:Cel 'I * *
,J . . ...._...--_.g_.,-.._,k/\3 k b #
y--- --J
--i. '" ~~C say g l Kt C0wenefWmf ET
[ .,. v(f.103Can W
- fag
? ,
'~-
'" C ot)d a t mm , sum -
lI
$3Uedatt
} *
(Mit (
,, -if t (>J0t$lt Costa ta Mgwf- .
, _ m.
' n W0lbad
( tai tuu113 g I -~ 0di f aga n
- 3f8 8'Id
. '-- LAlt h'I i
f y
.I i
-7 -
w -LAt0Ge t eLacice rue Spact gm
' ~L~
f 4A*11
, ggg ga.c4 - -- -
E -
} s ee.4 gie,t e..
s , g . ! --, .,n ,, u. . , p 'd hw' *
' El * .g ; ./ -.... * . (04!a% e00 088t( ' , '.m.-
\
, g. 1 .. . \ ,
aulltd Sm(LS -=7 , ,
-- = J- I
[ ' 7m
'...e... , *(I bf 3 +
__ _ , __L' '
'8 g--
f**I}f i ..u , y , DXal
-- i
- i
,.y.
9 1 h*,; g ,
.,-- . t&.t a
._w-. .-%.ois ILACfCO k\ s l
MM" - 1 i l .. Qd n ,<
$ Pa((
i m
- s. s
'l I.
uco.mu- ~ ,
- s. L, ,
; , i p . -
g
; q l .'.
.. .y 1:- e. . .
s g .. 3 6 u, . m , muu w. s a, xv . i _ . _ . . ri 4
.!....+. . .
^
' s u'"* "
. -[ 6'.
'l g3 , ..
g f '. ' : {N. <y.'
.: h N '- j r .
w, O ':...
.. A? O w, . .
. : < r.
M . .- e .
. P <* * *
, a*- "*:- ....*.s
- * . ,t '.*: , , .
.-[1:=
.. : .y 1. , ,,w; . . *. * .:;*1D.*;.'e' U
Centas Cws
- atta estet avtg CLM l4 wkM ALACfCt I t' .
.~- ,= . ;.
y ,o . i.y... r; . n t:. . * . (._)
'.*i.: . : .i
. . . h'.
FIGUF2 . CONTAIm/ENT CROSS SECTION c_c
w.ren.av.or i
\
O. g N V . Y.. - I, s s .
- r ap 1 I l l I It IO ;
I t ince . . ,, i. h - a- A. -
- 1 n e . .
ciw .na -. (. .p .
) .
ma __ . s te-3.- -ice co otsu. I
" _. L l ,
l . , i i
- 1. ,r.,
6 :
- j. ..aw wi.
l Ii l f l' e , [ (
/7 N
-~
b-~ - , ici nas --
' ' , . 1 I A s- t (o>
. . _ . . _ ... a. j, .,. _
m ,d )
; , a.
k . w __' . -
.- u
. %e.L, m.i
! f.~lq; -
i.ui eu, _. . .m.twum Xs r-
.m :
/n-t -
m . ., , _ . ._ f 3 l f l 3 ,i l.-_.- I.G_"u fM#2)~ i 4i~ . y l u. cjj
' l 'i
~~s f
- s' i, m ._.Ja mul , ,p.i~a
.S.J e r ,i ,g.. .._..
, pc, _
g-E, ,q , . fo-(,>;. . x.
' r,', ,i; i. it i' s . h.
- ,/, ,
m e,.-
+ i
. 4 .y .', . ' . .
a r
., . . . m, ,.s. ..
.. w. n .
h,'.di . ,.. \. .h" ... ;. {;k.O
.. k. ,.
/ .woa -
.._..i.
p i.*l '.:.:.. .
- R .. '. .. ;f.
y . FIGURE . CQJITAENENT CROSS SECTION 5-6 .
L-G --_ . . . . I 9 ho C > ?- ~
-O *:
mn sun wmN - b: " & % .} g i
.c, c.v O
- IL eq.
e.
... C.3 *
,.,j j O
' . 4 II
{-. 9 _. _ _ _ . . . b .
- b. u(T O
.1 tJ
- u. se-- -
y H.>;_ -J d) Cf I IT h 4 *r-- bt C 9 L T f '.4 $'
!*, ~I J$.'D iOhIID.U' .
G.!:]CC$ 6 ed
>l o
co l<~i s-
.'4 x s;'" ~.s L.>.
- g. -
2 % 4-hr
- vim Sl.3 L . N) k G Y.ijf*7)N39N'VS15- d~
~~
'Y*f- U UC ! O h
(
#'t {.
d Q ,' e-NO no v., F- ,,g. 2 0jM q t.1
~4 .
.# o t:1 ta }--
c1 'g [ m 5 tu Ett
- wo si.c 3 .
.d 9 'l, 1'8 ts
..c.r-4.1.7AYD # ts,. ,
4D '-1 2:.o21:fr.w.rrJ3 4 1 o - ,- 0tak 6,?es[9 'O Q= O po
.. /
c.uosst 5o . -1 a to o
& N ~C' V p' *2 t'l O E .(-@ ,
g -- d cf n Cl g 4o 1-N, ef. bd ., w y -
- c. 0 -A c y&o lo oy >
c/ 12 u-- j c2r5 1 W ,,.
. 4e
~4.'
13/WNG *- m w.T1'E; vl2 hr- m'-:.c- O CQ<q. 3 h Cad CYO I 3 g- t,lN e t.- tn c-N- Wf.'& C1 l
~5lIA ND4 $l$,e.-nLg ,,, y l < -' i
( g.LJ COG (G ;
.. 4-a%" D; @t. - .- S l
j O C', t-vri t' O
)
L.x er m . ...
~
c o in ,_ ,3 2 s N i 0 g?Jro9 U.%!!n vm2!:]
.y72'--Q-.-[E-.,..h'C'Q t y*1.,
T q 3
'--------=- -
_______L.__,._,_,,_.,_,, l/ 5 . . . - . . . . . . _ _ _ . . . _ _ _ _ _ . . . _ . _ . _ . _ _ . _ . . _ _ _ _ . _ . ..-
SECTION VI q HYDROGEN DISTRIBUTION A. Overview Appendix C provides a discussion of factors that can cause or reduce nonuniform hydrogen concentrations in the containment, postaccident. Since the mass, energy, and gas release occurs in the lower compartment, the hydrogen concentration tends to be higher there. Air is swept to the upper compartment, depleting the concentration of oxygen in the lower compartment. The air return fans return air to the lower compartment. This gross circulation pattern is modeled in the CLASIX computer code which provides the bulk hydrogen, oxygen, nitrogen, and water concentration in each major compartment. (} While certain processes can establish nonuniform concentrations within major compartments, the design of the ice condenser containment provides a substantially better situation than large dry containments in regard to mixing. The ice condenser and air return fan flow provide a relatively high flow of air through the major compartments. The hydrogen mixing subsystem provides flow through all significant compartments. In addition, the pressure changes and temperat'ures associated with hydrogen burning provide both flow between compartments and thermal gradients which enhance natural convection.
\-) 6-1
# *b** * ~
af. ..L 4 ed a a s * ~ e e-
- As a result of these factors, there appears to be no reasonable -
mechanism that could cause a substantial volume of combustible / ' i gas at hydrogen concentrations significantly above the bulk L) Sufficiently detailed hydrogen concentration of the compartment. analys:s have not been completed to determine if small localized cu.e+M. rations could develop, but it appears that even if such c= ventrations develop, they would be too small to serious 1.y jeopardize containment. N B. Factors Promoting tionuniform Concentrations Two mechanisms have been por,tulated that could lead to localized hydrogen concentrations: the " plume" or " jet" emanating from the RCS release point and removal of water from the air by condensation. \# The jet" or plume is composed of steam and noncondensible gases. In general, one cannot get hydrogen genecation and release without significant steam being present. Therefore, For this the hydrogen comes into containment diluted in steam. mixture to be combustible, the effluent must be mixed in substantial amounts of air (to provide oxygen). This process furt her dilutes the hydrogen, resulting in a concentration not significantly different than the bulk lower compartment concentration. In any event, the " plume" is limited in size and is rapidly entrained in the flow through the lower compartment. Although the steam generator and pressurizer dog houses may provide some trapping of this plume, the low volume, s high-design pressure capability, v 6-2 .
" ** " W et pJir5X*e p We= , . - - ,
i annular geometry, and high natural circulation flows minimize i the impact if such concentrations were unexpectedly formed. i !O I Condensation increases the hydrogen concentration by reducing i the partial pressure of water. This can occur in the ice h I condenser, in lower compartment coolers, and due to upper i compartment sprays. The extent of such an increased l t I concentration is small for the ice condenser and cooler effluent due to rapid mixing downstream.
-Ihis limits the total energy IS f l /
! released from combustion of the effluent; the effects are bounded J by the current analysis assumptions of burn throughout the major l compartment. The containment sprays remove substantial water . i i For small LOCA's analyzed,
- by condensation only after ice melt.
j this occurs after the hydrogen has been released and burned. I In addition, the spray water suppresses combustion and provides j e heet einx thet minim 1 zee the pres =ure end temperetere dee to l O combustion in the upper compartment. .i i
- C. Factors Promoting Uniform Concentrations _
1
' The ice condenser containment design includes many features that i enhance nearly complete mixing. These include the following:
i, 1 I e } l l ' O 6-3 i ' , 'n*> -:g8~L .; 2} , *}
' }* R~~~~ 3 - -_ -<- , - -
* ^ ' ~ ~ , ;; . . .. Q
~~.. . - -
i
- 1. Recirculation flow through the air return fans, accumulator rooms, ran rooms, main lower compartment, and ice condenser i
l O promotes rapid mixing by induced turbulence within each compartment and by moving air between compartments. Mixing within each of these compartments will be essentially j complete, considering the mean residence time (1-5 minutes)
- compared to the time period for hydrogen releases (15 minutes to a few hours).
1
- 2. The hydrogen mixing system provides suction from all j significant compartments.
, 3 The recirculation flow paths tend to isolate "deadend"
- compartments from the release point, preventing preferential buildup in these areas.
- 4. The upper compartment sprays provide very effective mixing of the upper compartment due to spray-induced turbu'lence.
I. In addition to these mechanisms, mixing is promoted in each compartment by the pressure changes due to burning by temperature differenceinduced nattiral circulation and diffusion. The first two are so strong that'they overshadow the effects of diffusion.for events of interest in an ice
~ ~
condenser containment.
~
D. Program Plans a TVA intends, as part of its Phase 2 studies of controlled ignition, to quantify the effects of mixing. These efforts include: O - #ere ee e der er =eae 1# eer =eet 1 e=t ee 11 ee eeee-6-4
to identify more sabvolumes and flow paths,
- Analytical studies and literature research to place bounds t g on the extent and magnitudes of local concentrations,
- Studies of the behavior of release point plumes and jets, and
- Sensitivity studies to evaluate the effects of different event and failure scenarios.
1
=
l l l l O 6-5 _.. . , - - - .r-,
SECTION VII () ANALYSIS OF CONTAINMENT CONDITIONS A. Overview An analysis was performed to determine the environmental conditions inside the Sequoyah containment which wi?.1 result from igniter operation in an accident environment. The Westinghouse /0PS computer code CLASIX was used to perform calculations of containment conditions. CLASIX is a multi-compartment containment code which calculates pressure and temperature response for the individual compartments while monitoring the distribution of oxygen, nitrogen, steam, and hydrogen. CLASIX can model many of the features of an ice () condenser plant such as the ice bed, ice condenser doors, containment sprays, and recirculation fans. Except for the ice bed, it does not, however, have the capacity to model passive i heat sinks or ECCS recirculation. CLASIX does not model the priaary system, thus energy and mass data for blowdown and hydrogen must be provided as input to the code. Hydrogen burning can be modeled with variable criteria for the hydrogen burn and burn propagation between departments. Further discussion of the code is presented in appendix T. l () 7-1
A small break LOCA followed by failure of ECCS injection (designated S D in WASH-1400) was used as the ba n case 2 for analysis to determine the conditions inside containment for an accident plus hydrogen burn scenario. This particular accident results in large amounts of hydrogen being produced due to a lack of core cooling. The Battel-Columbus Laboratory code MARCH was used to calculate the primary system blowdown mass and energy releases and hydrogen mass and temperature releases as a function of time for an S D 2 accident. Hydrogen production versus time is illustrated in figure 1. The hydrogen mass upper limit of 1550 pounds corresponds to a 75 pc Mt zirconium-water reaci, ion in the core. Containment initial conditions'for about the first 3500 seconds of this accident (the time prior to hydrogen production) were calculated using the Westinghouse LOTIC code. This code calculates the long term ice condenser containment pressure and temperature responses. The MARCH and LOTIC calculated data was input into CLASIX for the base case analysis. Other initial conditions and information which were input into the CLASIX calculation are shown in appendices T and U. For the base case, the hydrogen ignition in a given compartment was assumed to take place at an atmospheric concentration of 10 voluta percent. Propagation of hydrogen burning to other comractavnts could only occur when the volumetric O V hydrogan concentration in the atmosphere of that compartment was 7-2
HYOR0GEfl PRODUCTION ' DURING S 0 2 CORE MELT SEQUEtlCE Q 1600 - (MARCH CODE RESULTS) 1500 - 1400 - , 1300 - 1200 - 1100 - 1000 - 900 - O 800-
'P 700-600 500 -
400 -
?00 -
200 - 100 - 5000 6000 7000 8000 3000 4000 TIME (SEC) - 7-3 ,
~
- FIGUf!EI
i 10 percent or above. O Results of the CLASIX base case analysis are shown in appendix U. The results of the base case analysis indicate that the hydrogen will be ignited in a series of nine burns in the irar compartment. The time interval for the series of burns is approximately 3300 seconds. As a result of the action of engineered safety features, such as the ice condenser, air return fans and upper compartment spray, the pressure and temperature spikes were rapidly attenuated between burns. The pressure was decreased to its pre-burn value roughly 2 minutes after the burn occurred. After the last ignition of hydrogen, at 6800 seconds after accident initiation, approximately 300,000 pounds of ice is left in the ice condenser section. Tha results of the base case analysis show an increase in containment pressure of 4-6 psi, with r,he containment remaining well below the estimated failure pressures. All burning in the analysis occurs in the lower compartment since this is the area of hydrogen release, thereby gaining the advantage of heat removal l l by the ice bed. By burning at a givea concentration in the lower compartment there is also the advantage of burning less total l hydrogen at a time since the lower compartment volume is only around 1/4 of the total containment volume. It should be noted that passive heat sinks were not considered l 7-4 l 1
.o - _ _ _ . . . _ . m. ,_ - - - - . , . , - - _ _ _ - - _ - -- .-- -
in the CLASIX analyses, thus neglecting the heat capacities of the containment ressel, steam generators, pressurizer, structural concrete and structural steel, and other concrete and steel materials inside containment. These masses will absorb large quantities of heat during a LOCA and hydecgen burn. If considered, these passive heat sinks would shift the calculated compartment temperature and pressure curves downward, resulting in containment atmosphere temperatures and pressures which are lower than those calculated. B. Methods, Model, Assumptions, Inputs, Environmental Conditions, Results, and Verifications These subjects are described in appendices T, U, and V. O C. Analyses hue been performed to scope the environmental conditions inside the containment following a hydrogsn deflagration. The major factors influencing these conditions are flame speed, flame temperature, heat sinks, specific heats of the gases, relative humidity and the composition of the gaseous mixture. These variables affect the environmental conditions in various ways. The flame speed and flame temperature concrol the temperature peak of the atmosphere during , the burn. The specific heats also help to determine this peak though to a lesser extent. The composition of the mixture and the relative humidity are important in that these, two factors control most of the major characteristic of the mixture (specific p heat, density, etc.). The heat sinks tend to moderate the i 7-5 l
temperatures and greatly reduce the temperature peaks in the atmosphere. The flame speed information available for hydrogen mixtures at presssures near atmospheric pressure indicates a range of speed varying from 6 to 10 rps, the exact speed being on gas concentrations. Structural heat sinks inside the lower compartment consist of over 56,000 ft2 of concrete surface area with each concrete sink having a thickness over 1 foot, and over 24,000 ft of steel surface with thicknesses ranging from .53 to 1.7 inches. These heat sinks act to lower the atmospheric temperatures both during and following deflagration, as they absorb the heat given off by the hydrogen burn. Both convection and radiation heat transfer mechanisms are active during this process. Since the trend of the specific heats of the gaseous components is to increase with increasing temperature this phenomena assists in keeping the environmental temperature lower. The humidity of the atmosphere is important in determining the peak temperatures of the systems. Water vapor has the highest specific heat of an component gas in the atmosphere (excluding hydrogen present). It is postulated that the relative humidity of the atmosphere is at or near 100 percent due to the presence of steam escaping from the break along with the hydro 6en. This water vapor lowers the adiabatic flame temperature to about 670 F and removes a large amount of heat once it condenses on the cold containment surfaces. O V Preliminary calculations using 100 percent relative humidity 7-6
and approximately 25 pcreent of the available concrete heat sinks (} (no steel) shows that temperatures are considerably below those of the S 2D scenario as calculated by the CLASIX code. The method used involves computing the minimum natural convection, based on that of a horizontal steel plate. The value of the natural convection coefficient selected is conservatively low considering the fact that considerable turbulence should exist, generated by both the air return fans and the combustion process. Although ignored, the natural convection will be augmented by the forced convection provided by air return fan operation. The radiation heat transfer coefficients used are based on radiation calculations that are from furnace formulas developed by the International Flame Foundation. All heat produced by the hydrogen burn is assumed transferred directly into the gaseous ( mixture and subsequently it is transferred into the heat sinks. These calculations assume that the heat is uniformly generated in the lower compartment, that all heat is contained within that volume and there is no heat loss to the upper compartment or to the dead-ended compartments. There is 65.8 pounds hydrogen burned in each of nine burns spaced in time according to the burns in the S2 D scenario. This is a total of 592.2 pounds of hydrogen burned (this corresponds to approximately a 9 percent mixture of hydrogen by volume for each burn). Using the above situation with only about 20 percent of the lower compartment heat sinks and none of tha upper and dead-ended compartment heat sinks, temperatures were found to peak in the ()_ range-of 1250 to 1720 F depending on the timing of the burn. 7-7 m - - - - ~ g
Temperatures dropped rapidly between burns to below 600 F. Further studies may be appropriate. It is anticipated that calculations incorporating all the available heat sinks, the ice condenser (which has potential for removing additional large quantities of thermal energy), and the heat absorbing capacities of the upper compartment and the dead-ended compartments, will demonstrate that the effective temperatures encountered following a deflagration, are expected to be on the order of those which occur for a main steam line break. This is in contrast to current studies which only include the ice condenser as a heat sink and indicate high lower compartment post-burn temperatures.
- C
- ) .
e r i i l O 7-8
7- SECTION VIII V ENVIRONMENTAL IMPACT i A. Overview To fully evaluate the Sequoyah capability for hydrogen control following a LOCA, the containment environmental conditions discussed previously in Section VII must be applied to the 4 plant's structures and systems. The conditions to be evaluated are elevated pressure (including shockwave) and temperature transients. The structures include the containment shell and J interior while the system componenth include numerous types of electrical and mechanical equipment. O B. Containment Boundary The Sequoyah containment boundary has been reviewed 'for its capability to withstand the environmental conditions associated ' ~ with the combustion of hydrogen. The containment boundary is divided into two distinct regions which have sigetficantly different capabilities. These are the upper compartment free-
, standing shell and the lower compartment shell. A discussion of the capability of these containment regions to withstand pressures and temperatures resulting from the hydrogen burn and 1
the potential shock waves of a detonation is discussed below, section VIII.C, and appendices H and J. [) The lower region of the containment houses the primary system 8-1
,_ _ , . , , . - , - --w-w- w--v v +
and associated emergency core cooling systems. A thick concrete wall heavily reinforced with steel which supports the large () building crane surrounds the primary system separating it from l the containment barrier. In this region of containment the steel shell varies from a thickness of approximately one and three-eights inch to one and one-quarter inch. The space between the crane wall and the steel shell largely contains ECCS accumulators 1 and containment cooling units. See figure 9 7, volume 1. s The upper compartment is communicated with the lower compartment i through the ice condenser. This space is a large relatively open volume. The containment boundary, however, is largely hidden by the ice condenser which occupies the space between i the crane wall and the steel shell. The steel shell varies in i thickness also, having a minimum section of one-half inch at the containment spring line near the ice condenser t'op vent. Both upper and lower compartment shell have horizontal and vertical stiffeners. The horizontal stiffeners are installed on approximately 20-foot stiffeners and_ the vertical stiffeners are spaced at approximately four degrees. In addition, penetrations of the containment have been locally stiffened. All aspects of the containment shell were evaluated for the load bearing capability in order to identify the weakest elem ent
! of the shell. This includes penetrations, valves, welds, and membrane materials, and effects of stiffeners. Each of these items were found capable of withstanding substantial pressures ranging between 60 and 100 lb/in2g. The stiffeners were found 8-2
to be spaced to the point that the hoop stress capability of the membrane was largely unaffected in the mid area between stiffeners. Hence, the one-half inch shell plate between the [ stiffeners at elevation 778'-6" and 788'-0" in the upper i compartment was identified as the controlling critical area. The capability of the lower compartment region where the membrane is more than twice the thickness has correspondingly much more capacity than the critical upper containment section. The actual 4 property yield capacity of the containment at this critical 2 section of the upper compartment is 33 1 lbs/in g. See appendix J for structural capability. The environments which may exist in the upper and lower compartment will be very different. In all events, the release of mass and energy occurs in the lower compartment. The O temperature will be high, the atmosphere will be turbulent, and the pressure will be somewhat higher than the upper compartment. The upper compartment will be relatively cool, and the atmosphere will contain mostly air under pressure- The upper compartment sprays will most likely be on. In the event that significant hydrogen is generated, the hydrogen will be introduced in the lower compartment. . In the presence of igniters, the possibility of burning the hydrogen in the lower compartment is the most probable. Because of the geometry of the lower compartment and the potential for rich hydrogen release points, local regions exceeding the detonable threshold may be possible but is not probable because 8-3
-w.- - , , _.
y - , . , _ . , . , - . . _ _ _ . . e- - - - ,r.,
of the many violent mixing mechanisms occurring in this region. In the upper compartment the effluent from below enter via the ice condenser where the diluting effect of the steam is removed. Here again it is possible for a pocket of highly concentrated
-H 2
mixture to exist momentarily in the upper plenum of the ice condenser if high local concentrations exist in the lower compartment. However, mixing in the upper plenum is relatively strong and pockets of detonable gas will be quickly diluted. Effluent from the ice condenser will be further diluted rapidly as it mixes with the upper containment atmosphere under the influence of the containment sprays and the hydrogen mixing system. In summary, it is believed that the containment shell in the lower compartment may be exposed to quasi-static pressure loadings and corresponding heat loads from hydrogen burns in addition to the steam environment from the LOCA. Also, the shell of the lower compartment could be exposed to impulse loadings ~ occurring from localized detonations. _The upper compartment, on the other hand, will be exposed to the quasi-static pulse of a possible hydrogen deflagration and the normal LOCA environment. _ The analysis performed by Offshore Power Systems has shown the effects of burns within compartments. The investigations have shown a base case of pressure no greater than about 14 lbs/jn g. This is far below even the yield strength of the weakest containment section. Preliminary analysis discussed in section O 8-4 l
VIII.F of the heat loads indicate that the temperature rise in the containment walls will not be significant. While the (j containment boundary may not withstand the pressure following a full containment volume detonation or large burn, preliminary scoping analysis indicate that sizable local detonations are within the capability of the containment. Local detonations and the structural capability analytical methods are briefly discussed in appendices H and J. C. Containment Interior Structures 1 The principal interior structures of the containment are primarily the crane wall and the divider barrier separating the upper and lower compartment. These structures (see figures 9.7 and 9.8 of volume 1) are heavily reinforced concrete structures. They were designed to take the large differential pressure loads associated with either the controlling LOCA or main steam line i breaks in combination with other loads. Although a detailed reevaluation of these structures has nqt been performed, an inspection of the original design suggests that these structures will exhibit satisfactory behavior during the event postulated to occur in the containment. _ The Westinghouse / Offshore Power Systems studies of burning hydrogen in the Sequoyah containment have produced a maximum differential pressure of about 14 psig. Much of these structures were designed for around 20 psig and larger if other loads in the' design combination were converted to differential pressure. 8-5 e - - -
- n. n ,+- ~~ . . . ~ + -- - n. - -
TVA scoping studies as described in appendix D have shown that differential pressures across structures will in general not prasent a problem. It is also our best engineering judgment that these structures would survive a substantial local hydrogen detonation with only minor concrete cracking based on the massive amount of , reinforcement bar contained in the structures and on the results of high impulse loadings on similarly reinforced test specimens
; from tornado missiles. TVA plans to perform additional investigation into the effects of local detonation on interior structures. Further discussion on local detonations and structures may be found in appendices H and J.
t D. Critical Containment Systems
- O In addition to the environmental effects of hydrogen. combustion
~
on containment structures, there are several classes of electrical and mechanical components whose capability must be examined. These include various types of instrumentation, valves, cables, and protective coatings. When applying the environmental conditions discussed in section VII, it must be remembered that the CLASIX code and other state-of-the-art calculational techniques capable of analyzing ice condenser containment response to hydrogen combustion are conservative in their temperature predictions since structural heat sinks are neglected. Therefore, the peak temperatures presented previously should be applied with caution if a reasonable
+
(g 8-6 . i l l r l
evaluation is to be made. TVA has begun analytical efforts to predict the temperature effects of a series of hydrogen burns O
\,,/ on small components. In addition, work is progressing on the development of more reasonable methods to calculate the containment environmental response.
In the meantime, as a part of the phase II test program on hydrogen ignition, it is TVA's intention to conduct a limited equipment qualification test program on items in the containment which may be subjected to the hydrogen burn. The majority of these items will be instrumentation and control components such as an analog transmitter, a solenoid valve, a limit switch, and lengths of cables. The number and type of components to be tested will be dictated somewhat by the size of the test chamber (1000 gal sphere), by the size of the opening into the chamber l O. (18 inches), and equipment availability. This will not be a dynamic test; that is, none of the components placed in the chamber will be energized or operated before, during, or after the test. However, after the test, the- components will be inspected for gross failures, obvious deformations, and distortions that would indicate that the components could not withstand the extreme environmental conditions. produced by a hydrogen burn. It does not appear that the pressure transients due to hydrogen combustion will exceed the environmental conditions of pressure for which the containment instrumentation is qualified. Shock (~ wave effects have not been examined for small components. TVA v 8-7 l
\
7 i is continuing its analysis and testing of environmental effects on critical containment components. t E. Containment Shell Temperature An estimate was made of the temperature rise in the containment i shell due to hydrogen burning in the lower compartment. It was assumed that the gas will lose heat to the containment shell by radiation and convection and to the ice condenser. Due to the relatively low temperature of the gas in the dead-ended compartment, it was assumed that only the water vapor emitted and absorbed radiation. The containment shell will reradiate a portion of the energy it receives from the gas, some of which will be absorbed by the gas. Simple finite difference equations O were =eed to represent the heat be1ances for the centainment shell and gas for a time increment t. The gas and shell temperatures wer1 updated at the end of each time step and the calculation repeated until thermal equilibrium was reached.
~
For a single burn of 100 pounds of hydrogen in the lower compartment, the average temperature of a 1" thick steel containment shell increased by approximately 8 F. Assuming a similar tempoerature rise for each of the nine burns for the S2D accident scenario, the mean temperature of the shell should increase by roughly 72 F. This corresponds to a total energy deposition in the wall of about 4.5 x 10 Btu. The one inch thick containment shell was modeled as a one-dimensional slab. The total heat input of 4.5 x 10 Btu was
'8-8
- p.
- added to the shell over a 200 second time interval which corresponds to a surface heat flux of about 5370 Btu /hr-ft .
The TAP-A computer program was utilized to compute the transient temperature distribution in the shell. The maximum temperature difference in the wall from inner surface to outer surface was approximately 21.4 F. The corresponding temperature difference from the inner surface of the shell to the center
- of the shell was roughly 15.7 F.
An estimate was also made for an assumed accident scenario similar to that for the S D base 2 case except that no ice remains in the ice condenser after the first two burns. The calculational assumptions are similar to those used in making the previous estimate of the shell temperature rise for the S D 2 base case. i v Since ice is available for the first two burns, it was assumed that the average temperature of 1" thick steel shell increased by 8 F for each burn as previously calculated. When no ice is available in the ice condenser the average shell temperature increases by approximately 17 F per burn. There are a total of seven burns, two with ice and five with no ice. The total temperature rise in the shell is estimatad to be 101 F (2 x 8 x 5 x 17). This corresponds to e tot 11 Caposition in the wall of about 6.3 x 106 Btu. l
)
It was assumed that this amount of heat is added to the Q l kJ containment shell over a 200 second time interval which ) 8-9
i corresponds.to a heat flux of approximately 7520 Btu /hr-ft . l- The Tl?oA computer program was utilized to. compute the transient l temperature distribution in the shell. The maximum temperature difference in the shell from inner surface to o outer surface was approximately 32 F. The corresponding i temperature difference from the inner surface of the shell 4
- to the center of the shell was roughly 22 F?
4 i s i O 7 l ~ l [ i 'l 4 9
- O 8-10 l
a p p r'-- y... , -y- __.,e ,_y.,y---. . ,- y g ,_ g,.,,, 9p, y gg, ,, y.9 9 9 _g,..g- . _paa.p,99y9 - g m . m .e--
SECTION IX
' CONCLUSIONS i D V
4 A. ' Overview j TVA believes that, given a degraf.ed core accident with hydrogen
- generation, spontaneous ignition of that hydrogen will occur.
. In order to mitigate the gross impact of such an event, TVA has
- designed a controlled ignition system which we believe will burn the hydrogen at combustible levels and therefore mitigate the l adverse impact of an uncontrolled combustion at high hydrogen concentration inside Sequoyah's containment.
B. Ignition Source Effectiveness i (} TVA conducted preliminary tests to obtain reasonable assurance that the igniters we were considering would indeed ignite hydrogen at low combustible concentrations. The details of this testing is discussed in Section II and Appendix M of Volume 2.
~
TVA found that the GMAC model 7G glow plug could ignite hydrogen concentrations'at 12 volume percent and less. These results l allowed TVA to proceed with the design of the interim distributed 1 ignition system (IDIS) and the' igniter assembly enclosure. Once l the design was cc=plete, TVA, through Westinghouse, contracted with Fenwal Laboratories to conduct larger scale tests on the entire igniter assembly under various environmental conditions of pressure, temperature, steam, and flow past the igniter. These tests, which will be conducted around the second week in () September 1980, are discussed in detail in Appendix N of Volume 9-1
-_ .-- - - ~ . - - . - . - . - , - . . - . . - . - . .
2. -p v Also discussed in Section II and Appendix L of this report are the 31 igniter assemblies installed in containment at Sequoyah unit 1. These igniters have been located so as to be able to
;rn hydrogen in all three containment compartments and the majority of enclosed spaces. TVA believes that the location and number of igniters is sufficient to guarantee that hydrogen will not be able to randomly collect and form high concentration pockets. To ensure that the igniters are available whenever hydrogen begins to be released into containment, TVA is proposing to modify the emergency operating procedures (E0I's) as discussed in Volume 2, Section II, and Append!x R. The proposed procedures instruct the operator to manually initiate the IDIS immediately after all automatic equipment. The igniters then continue to O,
operate until the unit reaches cold shutdown. TVA has not been able to identify any accident scenerio that becomes worse due to the operation of the igniters. C. Containment Heat Removal As discussed in Section V of Volume 1, the Sequoyah containment is equipped with systems that-actively remove heat generated inside containment and also inherently aid controlled burning l by the IDIS. Although these systems were not originally designed to aid the 1 DIS, TVA feels that their operation adds significantly to the ability of the containment to withstand 1 this postulated event. 1 i O 9-2 l
A .*J+ J -- Two of the three containment heat removal systems actively remove heat from the containment atmosphere, the ice condenser and the () containment residual heat removal (RHR) spray system. The RHR j spray system also promotes mixing of the hydrogen in the containment volume. TVA expects good mixing of the containment air volume mainly due to the third containment heat removal system, the containment air return fans, which take suction from the upper compartment and dead ended volumes and forces the air l into the lower compartment at a rate of 40,000 cfm. The mixing 4 effect by the RHR spray system and the containment air return fans promote an even distribution of the hydrogen through large volumes in containment. 4 In order to monitor the hydrogen concentration inside containment, TVA has installed redundant hydrogen analyzers. These analyzers are discussed in Appendix K, Volume 2. They - have two sampling points one in the upper.and lower compartments. TVA is not certain that the present monitoring system is adequate for the life of the plant. However, due to the mixing effect obtained from the spray system and the
- containment air return fans and the planned continuous operation j of. the igniters, we feel that-the gross hydrogen concentration
- reading obtained from the present system is satisfactory for l
the next. year and a half until an improved system can be designed, purchased, and installed. 1 i D. Environmental Effects () TVA has analyzed, at least in a preliminary fashion, the i 9-3
= -
<w-+ ~e, m , , ps-, > m-m , - - , - - - ,ev---sa -,,- - ., ,w- ,
environmental effects of hydrogen burning on the containment shell, containment interior structures, critical containment O systems and critical components. A detailed discussion of these analysis are contained in Volume 2, Section VIII, and Appendices D through J. Preliminary results indicate that the temperature rise in the containment walls will not be significant and the containment structure as a whole can withstand sizeable local detonations. TVA analyzed the effects of local detonations on interior structures (Volume 2, Section VIII, Subsection C) and found that the differential pressures generated by these local detonations would not be more severe than the major accident loadings for which thes'e structures are presntly designed. Based on this analysis, it is our best engineering judgment that these structures would survive a substantial local hydrogen detonation with only minor concrete cracking. Analysis of the effects of hydrogen burning on essential equipment is much more difficult to analyze. Preliminary results indicate significant temperature rises across vital equipment. 1 This is contradicted by the physical results experienced at TMI-4
- 2. Thermocouples, which by design are one of the most sensitive pieces of instrumentation, at TMI-2 experienced only a 50 F rise concurrent with the ignition of the-hydrogen. To resolve the discrepancy between analytical and actual results, TVA plans 4
to include representative samples of vital equipment inside the test vessel at Fenwal Laboratories during phase 2 of the hydrogen burning tests. O
,9--4 d
- - - - . . . _, .m,_,_ - - - _ - _-
E. Scope of Events TVA s<lected five low probability, but severe consequence,
.AAC'it scenarios to represent the spectrum of hydrogen producing events fron metal-water reactions (see Volume 1, Section 5). From these five originals, three were input into the MARCH computer code. From the computer analysis (see Volume 1, Section 6), TVA selected hydrogen generation rates from the S2 D case, small break LOCA, as the representative basis for determining containment response characteristics.
The Westinghouse CLASIX code uses the output of the MARCH cobe for the S D case in their analysis of the ice condenser 2 containment. The CLASIX code is discussed in Volume 2, Appendix T. The CLASIX's results for the S D2 case assuming hydrogen (} burning at 10 volume percent indicate that containment pressure is reduced below the containment design pressure by the use of igniters. The results of the CLASIX analysis is contained in Volume 2, Appendix U. F. Margin Available Inherent in any type of simplified analysis are factors which
~
could not or were not included in the final results. Several factors which TVA feels provide conservative margin for the CLASIX code are:
- 1. The CLASIX code does not take into account the heat sinks available in the containment walls and interior structures.
TVA feels that this causes CLASIX to produce much higher () peak containment temperatures than what would be produced
- 9-5
in an actual hydrogen burn.
- 2. The CLASIX code does not take into account the water spray present in the containment atmosphere prior to a hydrogen burn. This water is an additional heat sink and a flame suppressant.
3 The CLASIX code does not model the flow from the upper compartment to the lower compartment created by the cooling and contraction of the air in the lower compartment. The contraction of the air in the lower compartment could cause a differential pressure much larger than that created by the air return fans. This type of pressure would create significant additional inflow of air from the upper compartment which would further aid containment heat removal. O G. Uncertainties Inherent in any simplified analysis are certain factors which cannot.be quantified. Throughout this entire analysis, TVA has
";d to make very basic assumptions as to the accident scenario and critical parameters. Several factors, however, are still fairly uncertain. The most prominent factor is that until NRC rulemaking is resolved, the accident scenario that will become the design basis is undetermined. Until the accident scenario J
is specified, results obtained from the MARCH code, which assumes particular modes of core failure, may or may not give , conservative hydrogen generation rates. The hydrogen generation rate is the single most critical factor in determining the
' structural capability of the containment and the usefulness of
,. 9 t e
, - - - , . -, - , ~. , - . - - -
the' hydrogen igniters. Another less central but important uncertainty is the effect i the heat sinks have on temperature and pressure peaks during hydrogen _ burns. TVA feels that t.he massive heat sinks help to i reduce the peaks, but this has yet to be quantified. Also, the effect of steam on the upper and lower hydrogen combustible f limits and to a lesser degree the effect it has on flame 4 propagation has not been quantified. TVA hopes to be able to reduce these uncertainties in all these areas through testing in the short term with the Phase 2 Fenwal tests and in the next 18 months with the more definitive studies that EPRI is proposing. k
- ([)
L
~
J 4 a 4 Y j 9-7 e- --- y + w ---
- -c- rr,,.-+e--- --g, *,.-.--..4 ,%-- -y w v s, e-- t w - +- * - --++eyM
l l r O APPENDIX A COMPUTER CODES i l i l 4 SEQUOYAH UNIT 1 REPORT OF THE SAFETY EVALUATION OF THE INTERIM DISTPTBUTION OF IGNITION SYSTEM I. 1 O { ! TENNESSEE VALLEY AUTHORITY SEPTEMBER 2, 1980 - I ( i ] 1 I
APPENDIX A () COMPUTER CODES a , The computer models used in the TVA hydrogen study include both i primary system and containment thermal hydraulic codes. These tools are described in Volume I, sections 4.2.2 and 6.2. In addition, TVA has contracted with Westinghouse to study the Sequoyah containment with the CLASIX computer code. The CLASIX code is described in Appendix T of Volume II; Appendix V describes code verification j efforts. Due to the importance of hydrogen distribution and 3- ! deflagration, an inhouse TVA effort has also been initiated to modify I' TVA existing containment code to include all physical processes believed to be important to the hydrogen problem. This effort is
, long term and is not expected to yield results before the end of the 1980.
i 1 M d i 4 4 i 3
\.)
s
'I
4 APPENDIX B EVENT DEFINITION 4 SEQUGUH UNIT 1 REPORT OF THE SAFETY EVALUATION OF THE INTERIM DISTRIBUTION OF IGNITION SYSTEM O TENNESSEE VALLEY AUTHORITY SEPTEMBER 2, 1980 - 1 l J ! O l l
i 4 APPENDIX B EVENT DEFINITION i The potential exists for significant core damage from several hypothetical accident sequences. Each sequence involves many of the 4 same event. For convenience, the Reactor Safety Study used alphameric designations for the major events that would alter the behavior of each accident sequence. These abbreviations have been retained and are listed below with the characteristic of each event. Event A: Large Break in the Reactor Coolant System Pressure Boundary This initiating event is a random rupture in the Reactor Coolant O ' System (RCS) that creates a break area equivalent to that resulting from a 6-inch diameter pipe rupture and ranging up to the full area
, of a double-ended rupture of the largest Reactor Coolant System pipe and causes rapid depressurization of the RCS. Also included are - reuptures of the reactor pressure vessel itself that place no more stringent demands on the engineered safeguards than the double-ended cold leg break.
4 Event B: Electric Power (EP) Operability of the engineered safety features (ESF) depends on the availability of electric power. Hence, this event is generally listed soon after all initiating events because of its importance to the accident ~ sequence. This event represents the availability of ac power O to the buses that furnish power to the ESF's. The principal t
. . . _ . . _ _ _ , , , - , __. . - . ._ .~ _ _ . ... _ _ -- - - - - _ . - _ . - -
i l i 4 components of the. electric power system are comprised of the of" site () ac. network and the onsite ac diesel generators together with the de control systems. The definition of EP failure is failure to provide sufficient ac and i de power for the operation of the ESF's required to mitigate the initiating event. EP failure affects the structure of event trees through the dependence of other systems on elect:ic power. Hence, when EP fails, other choicese are eliminated so as to imply consequential failure of other systems. } The electric power system event is subdivided into two separate . events - availability of direct current and alternating current. { The direct current (de) system is designated system / event B1. .The I () buses provide control power for ESF equipment, emergency lighting, 1 J vital inverters, and other safety-related de powered equipment for the entire plant. Similarly, the alternating current (ac) system is designated B2. Ac power is also required by all ESF's. N Event C: Containment Spray Injection System The containment spray injection system (CSIS) delivers spray to the containment to scavenge airborne radioactivity (iodines in particular) I frca the containment atmosphere during the time immediately following i- the RCS break. In addition, the spray aids in reducing containment pressure, thereby reducing containment leakage to the environment and lowering the probability of containment failure. The CSIS consists of redundant spray headers and pumps that deliver water from the refuelin5 water storage tank (RWST) which contains approximately i
375,000 gallons of water. Failure of the CSIS system is considered f- to be failure to delivery borated water from the RWST to the \_/ containment atmosphere at a rate at least equivalent to the full delivery from one of two containment spray pumps. A schematic of the containment spray system for Sequoyah is shown in figure B-1. Event D: Emergency Coolant Injection The emergency coolant injection (ECI) system is composed of four subsystems that operate in various combinations to provide emergency coolant for a range of RCS break sizes to precl'ude core damage. These four subsystems are: (1) the accumulators (ACC), (2) the high presure injection system (HPIS), (3) the low pressure injection system (LPIS), and (4) the upper head injection (UHI) system. The accumulators discharge stored borated water into the RCS cold legs when the RCS (~'s pressure drops below the pressure setpoint of the accumulator tank. %.) The accumulator system is a passive sytem. The HPIS injects borated water from the RWST into the RCS cold legs by using redundant, electrically driven centrifugal pumps. The LPIS injects borated water from the RWST into the RCS cold legs by using a separate set of redundant, electrically driven pumps. The UHI system is passive and discharges borated water from the UHI accumulator directly into the upper reactor vessel head when the RCS pressure drops below the UHI setpoint. ECI failure is (1j delivery of less borated water than would result from the discharge of two accumulators into the RCS cold legs immediatley following a large pipe break; (2) delivery of borated (~' L,) water at a flow rate less than the design output of either one
, g4_ , , ,
"-Y.'
- 4
- N-
*.#'-
- j, I ' . ,
g ,, h , [.-k 4', , , 4,* ,g
\ t .
e }
- 4 [/
a w$oO m R f
~> O c oQ '
- - 5 T ~i N 5 AL M .
I O)\
)L E I
/
N 4 U O M O W2 O - 0 S ~ R L S A U F 4 F C E 9 V N C E R A T E T U A R N T C U E E M R DI G I N NF 5 R I AI 7 I N T 3 A ' S O N 4 R R O O E D C Bi, E
. 5 T 2,AL M ;f E
S 4 U OO T U A 5 7 R ; ONB 3 O M N1. 2. h - T L .
,U R C E C A S C E S A R E 0
R Y V EI T 8 R PV O
,8 PA T 0 UC C A
; \u.
1 S EO MT R R w .E R m
)
OC E - O DE Z m - T O L I c_ % '. 5 N d, T L R I 7 E i ( NO U S wS y' ; MM O _ EC S O)\ b M) E l
+ U R O 1 0 IN N 10, A G E
R P .wI
; y-t T R S B E
( TO ,;
. ' IN R NR ,
U OD G CY I H F 5~ 7 3
/ -
R O
~ - 5 T
, S k' 4.< 2 A L M R 7 O
3 S
, + U O O T R S M R C O R U E T O C L A T C L R C A O E E C N L E N L N A O E G F C G M L' O AR N R E D EU G R Y TT O- O H S E R 0 T R O 4 A g!? Y L M IR H .
T U O A O I M R 3)u._ ( U T l p4_ ' o e C C lA
charging or one safety injection pump to the RCS cold legs (starting () at about 30 seconds following a large pipe . break andr lasting 1/2 hour); or (3) failure to deliver UHI accumulater water into the upper head. Failure of the passive injection systems is less likely than failure of the active systems. Figure B-2 lists parameters of the Sequoyah ECCS system. Event E: Emergency Cooling Functionability . Emergency cooling functionability (ECF) relates to the probability of failing to cool the core due to accident phenomenology even though the ECI functions properly. Several possiblilities exist for failure of the emergency core cooling system to reflood and cool the core even though emergency coolant is delivered to the primary system cold legs. Included are steam binding,. failure of core supporting structures resulting in an undefined coolant flow path within the vessel, and an uncoolable core geometry as a result of blowdown loads and subsequent thermal-mechanical distortion of the core. Event F: Containment Spray Recirculation Containment spray recirculation system (CSRS) provides for the recirculation of containment sump water through heat exchangers to spray headers inside the contain=ent for the purposes of pressure control and the removal of radioactivity and heat from the containment. This system is composed of four trains, each of which contain a separato pump, spray header, and heat exchanger. Failure of'the CSRS is considered to be a flow rate less than the [) equivalent. normal output of two recirculation spray pumps for the 5
EMERGENCY CORE COOLING SYSTEM COMPONENT PARAMETERS Component Parameters Cold Leg Injection Accumulators Number 4 Design Pressure, psig 700 Design Temperature, "F 300 Operatin( Temperature, *F 60-150
'ormal
,- Operating Pressure, psig- 660 Minimum Orurating Pressure, psig 600 Total Volume ft 1350 each Minimum Water Volume, ft 925 each Volur:e N2 gas, ft 425 Boric Acid Concentration, nominal, ppm 1900 Inleakage Alarm Sounds, ft 14 Relief. Valve Setpoint, psig 700 Upper Head Injection '
Accumulators Number- 2 Design Pressure, psig 1800 O oes1 8 re=Per t re. r 300 Operating Temperature, 'F 70-100 Normal Operating Pressure, psig 1350 MinibiumOperatingPressure,psig
- 1300 Total Volume, ft 1800 esch Water Voll:me, ft -
1800 Volume N gas, ft 1800 2 Boron Concentration, ppm min. 1900 Relief Valve Setpointi psig 1800 Centrifugal Charging Pumos Number 2 Design Pressure, psig 2800 I Design-Temperature, *F 300
- Design Flow Rate, gpm 150
. Design Head, ft. 5800
- Includes miniflow TO FIGURE B-2.
d^
EMERGENCY CORE COOLING SYSTEM COMPONENT PARAMETERS Component Parameters Centrifugal
- Charging Pumps (Cont'd) Max. Flow Rate, gpm 550 Head At Max. Flow Rate, ft. 1400 Discharge Head at Shutoff, psig 6000 tMotor Rating, b b 600 Minimum Starting Time,-sec. 6 Safety Injection Pumps *
, Number ,' 2 Design Pressure,'psig -
1700 Design Temperature, *F 300 D,esig'n Flow Rate, gpm 425
~
Design Head, ft. 2500 Max. Flow Rate, gpm .650 Head At Max.. Flow Rate, ft. 1500 Discharge Head, psig 1520 tMotor Rating' bhp 400
~
Minimum Starting Time, sec. 6 NORMAL OPERATING STATUS OF EMERGENCY CORE COOLING SYSTEM COMPONENTS FOR CORE COOLING , s . Number of Safet'y Injection Pumps Operable 2 Number of Charging Pumps Operable ' 2 Number of Residual Heat Removal Pumps Operable 2 Number of Residual Heat Exchangers Operable 2 Refueling Water Storage Tank Volume, Cal. ,370,000 Boron Concentration in Upper Head. Injection Surge Tank," min, ppm l',900 Boron Concentration in Refueling Water Storage Tanks, minimum ppm 1,950 I Boron Concentration in Cold. Leg Accumulator, minimum ppm 1,900 i Boron Concentration in Upper Head Injection Accumulator, minimum ppm 1,900 Number of Accumulators 6 Minimum Cold Leg Accumulator Pres'sure, psig 600 Minimum Upper. Head Accumulator Pressure, psig 1300 Minimum Cold Leg'Acetmulator Water Volume, ft 925
' Minimum Upper Head Ac:umulator Water, Volume, ft '
1800 System Valves, Interlocks, and Piping Required for the Above Components which are Operable All 7 TTnffDD D.9 [ nan 4-4 n ) .
s first 24 hours or the normal output of one recirculation spray pump thereafter. Figure B-1, although illustrating spray injection, also shows diagrammatically the recirculation system. Event G: Containment Heat Removal System i The containment heat removal system (CHRS) provides for the removal of containment heat by passing service water through heat exchangers I in the CSRS. There are four heat exchangers in the plant, one for each of the CSRS trains. To successfully remove heat, service water must flow through a heat exchanger located in an operating train of i j the CSRS. Failure of the CHRS is the operation of less than two of the four containment spray heat exchangers during the first 24 hours and operosion of less than one of the four heat exchangers thereafter. l Event H: Emergency Core Recirculation System (ECRS) When ECI has succeeded, the continuity of emergency core cooling is provided by the recirculation mode of the safety injection system, which is realigned to take coolant from the containment sump. Alignment into this recirculation mode of operation occurs j approximately 1/2 hour after the de_ sign basis large_LOCA when the a I RWST nears depletion. Operation action is required to switch the i
- valve arrangements to the appropriate configuration for recirculation.
3 The initial alignment for recirculation is for the safety injection
- pump to deliver into the'RCS cold legs only. However, unlike the
~
injection phase, the delivery of fluid into the RCS hot legs is not considered as failure. The ECRS should be aligned to deliver into
the RCS hot leg piping to help avoid potential accumulations of
,o
( ,) residuc er debris in their reactor vessel, which may result from continuous boiling, within about one day after a large LOCA initiated in the cold leg piping. Failure of the ECRS is defined as failure to inject into the RCS from at least one safety injection pump. Failure to realign to hot leg delivery is also considered failure. Event K: Reactor Protection System Reacter protection system failure is conservatively defined as the failure of more than two full-length control rou assemblies to insert into the core within approximately 30 seconds after the initiating g3 event. The failure of the required control rod assemblies to insert () may be caused by electrical faults in the signals or equipment required to gravity insert the rods into the core, or by mechanical faults that result in a hangup of more than two full-length control rod assemblies. - Event L: Auxiliary Feedwatar System (AFWS) The auxiliary feedwater system supplies, in the event of loss of the l main feedwater supply, sufficient feedwater to the steam generators I to remove primary system stored and residual core energy from decay l heat. It may also be required to provide cooling during a small break LOCA by maintaining a water head in the steam generators. Each system has two electric-motor-driven pumps and one turbine-driven pump.
~
Each electric pump serves two steam generators, the turbine all four. (^} 1_ Failure of auxiliary feedwater delivery is considered to be less than i
the full delivery from one or two half-size electric-deiven feedwater () pumps or the equivalent flow from the full-size steam-driven auxiliary feedwater pump. The period of demand and operation of the AFWS is about 1/2 day for a small break LOCA. Event M: Secondary Side Relief Valves The secondary side relief valves vent steam generatSd from the water provided to the steam generators by the auxiliary feedwater system. This steam is vented to the outside atmosphere via power-operated relier valves or safety valves to augment RCS heat removal. Failure of this system could result in steam generator overpressurization and failure or inadequate primary system heat removal for accidents where the steam line isolation valves and atmospheric dump valves cannot be opened. Figure B-3 details the valve parameters for (n s_/ I Sequoyah. Event 0: Containment Emergency Safety Features The containment emergency safety features are comprised of the containment spray injection system, the containment spray recirculation system, and the containment heat removal systems. This event is simply a general heading for coupling the three components either as a general fialure or success. Event P: Primary Side Relief Valves Open This event represents the opening of the RCS pressurizer safety or safety and relief valves to limit the rise in the reactor coolant pressure immediately following the initiating transient event. Not ('3 all anticipated transients require operability of the safety valves 10
PRIMARY AND SECONDARY SAFETY RELIEF VALVES
, DESIGN DATA Accumulation Blowdown
- 4 Valve MK # Set Pressure Percent PSIG Percent PSIG Max. Set Pressure 47W400-101 1064 10.3 1174 96 1021 1075~
~
3 47W400-102 1077 9.0 1174 96 1034 1088 I . ! 47W400-103 1090 7.7 1174 96 1046 1102 i
'47W400-104 1103 6.4 1174 96 1058 1115 47W400-105 1117 5.1 1174 96 1072 1132 .
Valve Design Considerations - Five safety valves per steam generators. Each set of safety valves should have a minimum combined capacity of 3,917,000 PPH. Maxi =um actual capacity of any one safety ' valve shall not exceed 890,000 PPH at 1,100 psia. , Valve settings consider a 21 psi pressure drop from the steam generator outlet j. () to the most remote valve inlet.' The balanced safety valve design and sonic flow limiting orifices assure set points independent of back pressures. Valve
. set tolerances are u maximum of +1 percent. Maximum accu =ulated pressure equals -
i 1,100 psia times 110 percent minus 21. psia which equals 1174 psig. . i Valve 47W400-101 has a maximum set pressure of 1075 psig, 'which'is 10 psi below system design pressure. ' ~ Valve 47W400-105 has a maximum set pressure of'1132 psig, which provides a minimum of 3.7 percent accumulation.- 1 l i i I - i i FIGURE B-3 o 4 sy- -rs; P -+p-r w.T t p. ,+g.q q e w -- y-e pi t-+1-' ---W--- ==e4,< *+y P
I due to the surge capacity of the pressurizer to mitigate small () pressure surges within the reactor coolant system.For more severe transients, such as those involving failure of the RPS to terminate core power, when required during a transient, the operability of the presst'izer safety valves would be required to prevent a rupture of i the RCS from the resultant energy and pressure pulse. Typical parameters for these valves are shown in figure B-4. i Event Q: Primary Side Relier Valves Reclose The RCS pressurizer safety / relief valves that open must reclose to prevent a discharge of an excessive quantity of coolant from the RCS. , T Otherwise, a valve sticking open following the transient event of
- interest would result in a loss of coolant event similar to the event that occurred at the Three Mile Island facility.
, Event R: Reactor Pressure Vessel Rupture ; This event considers the gross failure of the reactor vessel in which l I coolant loss exceeds that of a double-ended 'old leg break. l I 1 l Event S,: Small Break LOCA (2" D 6") This even considers the range of small breaks in the RCS with a break area equivalent to that of a 2-inch diameter up to and including a 6-inch diameter. The separation of small breaks into two categories is due to the difference in mitigation procedures. For a break in ' j this range, accumulator discharge and high pressure injection is j required to arrest the fuel-clad temperature rise and to prevent inadequate core cooling. f' s
, r2.
L y- -e--- , ,, , , - - . . _ _ _
.w,,,- , . ,,_,.,.,.~.,y. ,----w - , , >,*
/'
(_T / PRESSURIZER VALVES DESIGN PAPJLMETERS I Pressuriser Spray Control Valves Number 2 Design pressure, psig 2485 Design temperature, "F 650 i Design flow for valves full open, spm 800 Pressurizer Safety Valves Number
. 3 Maximum relieving capacity, ASME rated flow, lb/hr (per valve) 420,000 Set pressure, psig . 2485 (+1.0%)
Fluid Saturated steam Backpressure:
- Normal, psig 3 Maximum during discharge, psig 500 Full lift pressure .
. <3% of set pressure Blowdown <5% below set pressure RCS pressure at the reactor coolant <110% of set pressure pump discharge when valve is at full
() lif t (including pressure drop between safety valve and reactor coolant pump) 1 Pressureizer Power Relief Valves 1 Number 2 ' Design pressure, psig 2485 Design temperature, "F 680
~
Relieving capacity at 2350 psig, lb/hr (per valve) 179,000 Fluid Saturated steam
=
9 O FIGURE B-4 1,3 E' s v v .-.,,c-.----,
- - -_ , ,- n- - -,,r ,, ---
Event S : Small Break LOCA (1/2" D 2") (^- Small ruptures of the RCS equivalent to the area of a 1/2-inch diameter pipe up to and including 2-inch diameter are included under this event. This event requires ECC injection via the high pressure injection system. Event T.: Transient Event Initiated by Loss of AC Power This event can be characterized as a failure in the plant's electrical network which results in a transient being imposed on the RCS and core that (1) leads to a demand for the operation of the RPS to cause a trip of the reactor control rods to terminate power generation, and (2) requires operation of the plant normal or alternate heat removal systems to ensure adequate cooling of the reactor core. / Event T3 : Transient Event Initiated by Loss of Main Feedwater In this event, a failure of the main feedwater system imposes a loss of heat removal transient on the RCS. Overheating and pressurization of the RCS may result in posing a demand on the relief valve system. The same conditions apply on the RPS and alternate heat removal systems as stated in Event T
- A Event X: Air Return Fans The primary purpose of the air return fan system is to enhance the ice condenser and containment spray heat removal by circulating air from the upper compartment to the lower compartment, through the ice condenser, and then back to the upper compartment. The air return fans also serve to mix the hydrogen generated during the design basis
( ) accident precluding concentration pocketing in the containment. The lY
~ -,
r . I f system is illustrated in figure B-5. Failure is assumed to be l capacity less than the equivalent of one fan. f
' O V
Event 2: Ice' Condenser The ice condenser containment relies on the melting of a minimum amounct of ice to limit the containment pressure in the event of a large LOCA or other event that releases energy into the containment structure. The ice condenser is assumed to meet the technical specification surveillance requirements for minimum poundage of ice. O 15 wemx.nd#www=9
. =- _.
- w. ,.-
~
l.(.
~O .
O' o !l l' f XN X - AAAAAA N ' ~ AAAAAA d X "
'd \
g ><
=
. AAAAAA AAAAAA d \ YA
- SPRAY HEADERS I UPPER COMPARTMENT m m m n
< ICE CONDENSER >
R.W.S.T. LOWER COMPARTMENT CORE i g COOLING -- -- -- -- CONTAINMENT i RESIDUAL SPRAY
. S MP HEAT HEAT REACTOR-+ EXCHANGERS EXCHANGERS e e __ __ __ __
; V ) () ( RESIDUAL SPRAY PUMPS
' HEAT REMOVAL s PUMPS r 1 r p
9 J L ; L 4 s T I S ( / / 4
. (
i
- b~
g FIGURE B-5
O APPENDIX C A DESCRIPTION OF PHEll0MENA WHICH MAY AFFECT THE POST-CORE-DAMAGE-EVENT DISTRIBUTION OF HYDROGEN IN THE SEQUOYAH CONTAINME!!T i SEQUOYAH UNIT 1 REPORT OF THE SAFETY EVALUATION OF THE INTERIM DISTRIBUTION OF IGNITION SYSTEM O TENNESSEE VALLEY AUTHORITY SEPTEMBER 2, 1980 l I O
i ~ APPENDIX C A
'V A DESCRIPTION OF PHEN 0!ENA WHICH MAY AFFECT 4 THE POST-CORE-DAMAGE-EVENT DISTRIBUTION OF HYDROGEN IN THE SEQUOYAH CONTAINMENT J
Purpose and Scope I. i This report provides a qualitative description of the subject j phenomena. While quantitative information is not provided, the a report does provide an evaluation of the relative impact of each phenomenum. 4 The effect of hydrogen combustion inside a reactor containment ( ). is dependent on various factors, the most important of which are: (1) the total amount of hydrogen burned; (2) the rapidity l of combustion (including detonation); and (3) the concentration j 4 prior to the combustion. These factors affect the :
~
transient and longer term pressures and temperatures that result 1 i from combustion. It is, of ccurse, easiest to analyze the results if the pre-combustion concentration in the whole j containment (or in a major region) is uniform over the whole 1 l volume. l l 1 I If concentrations are not uniform, the results of combustion may exceed or be less than those estimated for uniform concentrations. A nonuniform concentration -(such as " hydrogen ( pocketing") might lead to a condition where detonation, large . ~,
. ~ . - - _ . . _ . _ _ - - . __ _. , _ _ , _ , , _ . . . . . - . , . . _ . . . _ _ . _ _ . . . _ . . _ . . _ . ~ . . _ . . - _ , .
.. .-. . - . - . . . . . --. - - - _ . . - - .. =-
If concentrations are not uniform, the results of combustion may O execed or be less than those estimated for uniform 1 concentrations. a nonuniform concentration (such as " hydrogen 4 I pocketing") might lead to a condition where detonation, large local pressure and temperatures, or earlier combustion might occur. On the other hand, a nonuniform concentration may involve
, distributions where a significant amount of hydrogen is in regions with a concentration less than the combustible limit leading to less total hydrogen burned than predicted by uniform distribution estimates, and correspondingly reduced bulk pressures and temperatures.
i i II. General Overview (} A. General Hydrogen Circulation For essentially all events of interest, hydrogen is released j ,at high temperature (approximately 300 - 700 F) from the reactor coolant system. The RCS is located in the central 4 are' of the lower compartment. This release may be continuous over a period of time (with the major fraction l! j of the release occurring over a time period as short as 15-30 minutes or over several hours), if the initiating event is a pipe break or stuck open valve. On the other 4 1.
. . . _ . _ - - , . _ m_..-_- ,, . . , _ . - . _ _ ._
hand, the release may be in rapid, sporadic bursts if the RCS is essentially intact (releases would be by way of p/ s, m relief and safety valves or RCS vents). In either case, the hydrogen is accompanied by steam, the amount of which depends on the particular event. The release of these gases to the local compartment increases compartment pressure, forcing flow from the lower to upper compartment. Condensation and cooling in the ice condenser (when ice is present) and upper co=aprtment (due to sprays) also forces flow from the lower to upper c ompartment. Air return fans force flow from the upper to lower
~
es c ompartments. In addition, the fans take suction from the top of such lower compartment subcompartments as the steam generator and pressurizer doghouses and returns this flow to the general spaces of the lower compartment. Circulation within each subcompartment is quite vigorous. Phenomena causing this circulation include turbulence due to containment spray (upper compartment) a nd flow through coepartments; natural circulation due to different l l
/h l L) 3 l
temperatures of gas,_ walls, ceilings, floors, and equipment;
~ turbulence induced by jet effects from RCS breaks and vents; and, if operable, flow from fans in HVAC systems. Because ; of the significant pressure and temperature transients that 1 exist for events of interest, natural convection alone is sufficient to assure that no major stagnant regions exist for at least many days after the event.
i B. Mechanisms Affecting Distribution to Hydrogen in Containment Several phenomena have been postulated as establishing or
- minimizing differennes in hydrogen concentration in different regions of containment. These are summarized 4
below and discussed in more detail in the body of the j (} report, ,
- l. " Jet" or " plume" effects - Hydrogen is released to containment mixed in steam. The concentration in the release jet or plume may be higher than the bulk concentration (at least during the time periods of i
most importance), leading to the jet being a region of high hydrogen concentration. I i 4
~ ~' ' * - - -+ ' " * = - ' -- ~- -
-+-- -* --e ea.w - . L.- ,m_. - . -- --____
- 2. Steam condensation effects - Steam is an important
'constituant of the postaccident containment CE) i e r.. ironment. k'here massive condensation takes place (in the ice condenser, due to upper compartment sprays, t
or regions with significant heat sinks), the removal of steam increases the relative concentration of I l hydrogen. 1 1 3 Fan-forced convection - The air return fans force flow from the upper compartment and from lower compartment , i high point to the lower compartment. This flow tends to induce a counter flow through the ice condenser and bypass flow paths back to the upper compartment. t This tends to even out the hydrogen concentration i in the major compartments (and prcvides oxygen to the
.( )
lower compartment). i l
! 4. Spray-induced turbulence - The upper compartment sprays i
remove steam, thereby inducing flow through the ice condenser from the lower compartment. This effect is much more pronounced, of course, after the ice has j melted out. In addition, the spray flow, by momentum ) transfer, thermal gradients, and condensation, induces I i 1 O i , sr i
.. + w <e,- --.44- -.w w r-,,, ,.7-,,.,g -- ,,vr, , ,_,,ny v ,,. -- - ~.,--e.,-v-- . - -,-- e.
strong turbulence in the upper compartment, assuring nearly perfect mixing of the upper cccpartment (this i O effect is supplemented by the hydrogen mixing ducts in the domo). 1
! 5. Natural circulation - Natural convection occurs on l
both large and small scales. Hydrogen / steam releases. i (being both light and hot) rise rapidly. Heat sinks
! such as structures and equipment provide cooling in the early stages of the event, creating strong natural ,
- convection flows in each subcompartment, promoting l
mixing in the compartment. Therefore, while natural convection causes gas with high concentrations of hydrogen to rise to the top of the compartment, the
^
turbulence induced by this flow and by cooling from g heat sinks tends to break down these concentrations. i 6. Diffusion - Gas diffusion tends to even out the i j hydrogen concentration within each subcompartment. Diffusion is pedbably not of significance however, because of the presence of turbulence which, postevent, ' is a much stronger mixing force. 4 o y
C. Initial Containment Environmental Conditions l~ For loss-of-coolant accidents, significant mass and energy
' releases to containment occur prior to formation and release of significant amounts of hydrogen. Therefore, such I
important containment systems as the ice condenser, containment spray, and the air return fans have come into 1 play prior to hydrogen release, assuring that the release 4 is into a dynamic,. turtulent environment. The lower I cocpartment would be expected to be at a pressure slightly above that of the upper compartment, and to have an j atcosphere composed of more water vapor / steam and less air , (in the early stages, the lower compartment is pressurized principally by steam and temperature, while the upper () co=partment is pressurized by air from the lower co=partment and temperature). While the lower compartments' bulk oxygen concentration is somewhat depleted, sufficient oxygen exists i in most regions to support combustion. IV. Release Point Effects i For the events of interest, the release point for hydrogen is in the lower compartment. The most probable region is the main f i co
~
r GP L
r annular volume between the reactor vessel cavity wall and the crane wall. Release points might be a small hole in a large pipe, a broken small pipe, the pressurizer relief tank (relief, safety, and vent valve discharges), or the reactor vessel vent l to ccntainment. 4 i While there may be mechanisms which break up jet flow from the j release point (such as release under water and ' impingement on
- structures and equipment), in general, the plume of released j gas can rise relatively unimpeded to the top of the lower i
compartment. Significant mixing with the atmosphere due to f turbulence can be expected, particularly in the margins of the plume. At the top of the compartment, the flow spreads laterally I with a significant portion entering the ice condenser ' inlet (} doors. Another portion enters the dog houses (steam generator or pressurizer compartments). Since the dog houses are deadended volumes (roughly annular in shape), the flow into the space drops i after the space is filled with gas since volume removal from l the space is relatively slow (turbulence at inlet, condensation, hydrogen mixing flow, natural circulation). While a region of relatively high hydrogen concentration can be ! formed by the mechanism described above, its size and i O 9
ccncentration are severely limited by several conditions. First, I the hydrogen is released highly diluted by steam, setting an upper limit on _ the hydrogen concentration. If mechanisms exist to strip this steam, then the temperature is also correspondingly decreased, reducing the jet and bouyancy forcing, and preventing significant stratification. Second, before a highly combustible mixture can exist, air must be added by turbulent mixing and 4 ! hydrogen diffusion; both effects significantly reduce the i j hydrogen concentration. Third, there is strong circulation of air through the compartment due to air incoming from the air i return fans and existing through the ice condenser. Considering these three effects, TVA does not believe that pockets of ccmbustible mixtures with hydrogen concentrations significantly
- above the compartment mean concentration will occur.
~
(:) I If the release point is below the sump water level, the steam may be at least partially removed and the hydrogen cooled. The hydrogen would then be released at the pool surface with greatly reduced buoyancy. The air flow from the air return fans, being colder than bulk compartment air, tends to sweep this pool surface, providing a strong dilution mechanism as well as i breaking up vertical flows due to buoyancy. l l 2 + i i i
.O
-__ ~ . .- ..__ _ . - _ . _ _ _ . _ . _ . _ _ _ _ _
V. Steam Condensation (, ~) Steam (or water vapor) not only dilutes the hydrogen, but also tends to suppress combustion. Therefore, if the steam is removed by condensation, the combustibility is increased. There are three strong mechanisms for such condensation in the ice condenser; cooling in the ice beds, by the containment sprays, or by the lower compartment coolers. The flow into the ice beds comes from the lower compartment. In essence, this flow is composed of air from the return fans mixed with the RCS release flow. The exit flow is composed, then, of the noncondensibles (including hydrogen) from the RCS release and water vapor, diluted in the air return fan flow. The ice bed exhaust is rapidly mixed in the very turbulent air rm ( / of the upper compartment. Therefore, the physical size of a high concentration region is limited to the ice condenser upper plenum and a few feet outside the plenum exit. Since such a region would compromise only a small fraction of the upper compartment since propagation of combustion into the j well-sprayed upper compartment is suppressed by the falling spray droplets, and since there is little or no sensitive critical l 1 l l l l l l l
/
10 a..- -
. .uwwuxana;z =.:
equipment in the area, strong combustions or even minor-detonations :4111 not endanger the plant. The upper compartment sprays remove any excess moisture fecm the upper containment. However, if the flow through the ice bed is well cooled (the case when flow is moderate to low and substantial ice remains), the sprays can actually heat the air and add water vapor. After the ice is gone, steac from the lower compartment is condensed in the upper compartment. While this can Icad to increased concentration of hydrogen, the spray droplets plus the increased water vapor concentration at the higher temperatures involved (post-ice melt) should effectively suppress any substantial combustion.
~
The lower compartment coolers are not engineered safeguards; however, since they are believed to be capable of operating in a postaccident environment and their use would be desirable in a degraded core event both for mixing and heat removal, their effects are considered. The coolers will remove substantial amounts of steam, thereby increasing the hydrogen concentration. However, the turbulence of the exhaust promotes very rapid mixing and, within a short distance, nearly complete mixing. This has been observed when the fans were run in a steam environment. u
- .= ..
. . , . -. ...~ .. . .- c , _ , _
This results in limited volume of higher concentration, limiting-the possible effects of rapid ccmbustion. If a sufficiently strong combustion or detonation occurred, the coolers could suffer damage, but other equipment should not be affected. 1 VI. Fan-Forced Convection The air return fans exhaust to accumulator rooms from which the flow passes through the fan rooms and out through holes in the j crane wall into the lower compartment min compartment. Since i this air is colder than the lower compartment, it tends to sweep the sump water surface. The mjor components in this region j (EC pumps, RCS piping, steam generators, and RCS supports I interferes with this flow, causing turbulence and mixing, ensuring a well-mixed flow to the ice beds. The flow out, the j ice beds into the upper compartment is colder than the air in j the upper compartment; therefore, it tends to move easily into i the flow pattern established by the upper compartment sprays.
~
l The air flow to the accumulater and fan rooms can also go i (through holes in walls and floors) to the other compartments f 1 outside the crane walls (two accumulator rooms, instrument room, I and raceway). These are deadended volumes whose only exhaust
~
O 11
is through the mixing system ductwork back to the air return fans. These rooms therefore tend to centain air the same
/ temperature and concentration as the upper compartment, but lag upper compartment trends because of.the relatively low flow.
i Because the fan flow effectively blocks releases from the RCS from entering the deadended volumes, there is no major hydrogen i source in these rooms. The only source of hydrogen is evolution ' of hydrogen from sump water in the lower receway. However, wat'er
- only gets to the raceway after completion of ECCS injection and substantial ice melt. For small LOCA's with ECCS failure, the
. majority of hydrogen has been released prior to water entering
! the raceway. The water that does get to the raceway has had I
l a significant residence time in the main compartment, allowing dissolved hydrogen to be released. The hydrogen mixing system 4 flow will keep the deadended compartment concentrations from I i () building to high levels. The air return fans also pull flow from the upper containment dome to prevent hydrogen buildup there. However, while the containment sprays operate, the induced turbulence is sufficient to provide rather complete mixing. The mixing system flow to the air returns is needed therefore only in the long term. 4 4 4 4 r 0 13
,. - n...,...-. _ - - . .- ...... . ... ~.~. ...
er %- .y w ---. -p y -4m. p- - - g i>=y-wy
% pv. mw y-w-w---~v gi gum wy,r-- y-g 9my- y.-m--.wgyv--*-y--esew'.- -
-ym*+ -
VII. Spray-induced Turbulence The upper centaintcent spray system consists of four subsystems,
~
each with its own spray header; two redundant centaincent spray systems and two redurdant RHR containment spray cubsystems. The centainment spray subsystems provide 4750 gal / min each and the RHR subsystems 200 gal / min each. The RHR subsystems are intended for use in the long term, but could be used in the short term for certain ECCS failures. The four headers are located between the 838 and 842' elevation, and spray the upper comparttent inside the crane wall. The spray causes strong turbulence and mixing in the upper compartment due largely to momentum transfer between the spray () \_) droplets and the air. This turbulence has been observed during spray flow tests. Shear forces between the sprayed region and the relatively small unsprayed region promote mixing in the latter areas. VIII.!!atural Circulation The postaccident environment in the containment is a highly dynamic situation with pressure and temperature changes, high p i ,) 1+
flows through most compartments, and heat transfer at walls, ficors, ceilings, equipment, and structural elements. This leads ( to to:perature differences between. gases and structures / equipment
, that produce natural circulation flows. Such flows are present i
in the cost quiescent compartments due to the large heat capacity of many of the structural elements. This natural convection mixing is present for design basis events. For events beyond the design basis, it is augmented by the piston effect of pressure rise due to burning and by the relativol'y higher temperatures after burn. Since it requires realtively low temperature differences (a few degrees) to 1 J establish turbulent c~onvection, strong local turbulence and mixing can be expected postaccident. For the major compartments, 4 forced flow and spray-induced turbulences should predominate.
' (:) For deadended volumes, natural circulation should provide sufficient mixing to prevent stratification.
IX. Diffusion - Diffusion would tend to reduce nonuniform concentrations. Although hydrogen has a relatively high diffusion rate, diffusion is not expected to have a significant effect on concentration 4 e e O f5
-- . = _ - . - . - , - . . - - - . . . - . . - , . . . . . - - - - . . - . - . _ . , _ . . - . , - _ ,
differences because the cuch stronger turbulent mixing will predominate. Diffusion will enhance turbulent mixing by , increasing mixing within eddies. Diffusion can play a substantial part on a microscopic scale, however, since for hydrogen lean mixtures, it plays an important role in the combustion process. O O IL _ , . - . , _ y.--- u% t --'---w-pw-y- mwr--w' P +--'WW " ' - ' ' - * - ' " ~ 7*"'#
--+m * +~ ---a,- a m.4 +---+<=- - m <m , -- ---- - w _- , - -- - -- .,,- --
1 APPENDIX D NONSYMMETRIC CONTAINMENT LOADS SEQUOYAH UNIT 1 PEPORT OF THE SAFETY EVALUATION OF THE INTERIM DISTRIBUTION OF IGNITIOil SYSTEM i O i TENNESSEE VALLEY AUTHORITY SEPTEMBER 2, 1980 4 4 0
APPENDIX D (3 V NONSYMMETRIC CONTAINMENT LOADS An evaluation of potential nonsymmetric loads on the containment shell and internal structures has been made. There are two basic phenomena that could result in such nonuniform loads. The first is due to burning in a local area. This would occur as a result of pockets of hydrogen collecting in one or more restricted areas'or a general burn that starts in a restricted area of the containment. The second cause of local pressure perturbations would be due to a pocket of hydrogen reaching a detonable concentration and exploding. The potential for local pocketing and the effects of detonation are discussed in Appendices C and H, respectively. Discussions in this appendix will be limited to the cases involving hydrogen burning. To evaluate the potential for asymmetric loads at Sequoyah due to hydrogen burns, the containment was modeled using nine nodes. Six nodes were used to model the lower compartment inside the crane wall, and the upper compartment, reactor cavity, and an accumulator room comprised the remainder of the model. The model did not include other dead end volumes to reduce the overall volume in the lower portion of the containment which maximizes pressure differentials. Flow areas and loss coefficients were taken from the calculations of assymmetric
. pressures due to large pipe breaks for the original plant design bases.
Typical data on hydrogen burning in dry air shows the flame speed is generally accepted to range from 6 to 10 ft/sec. Velocities of l
this magnitude mean the energy release is relatively slow, ( particularly when compared to the energy releases in a large LOCA or main steam line break. This means that nonuniform loadings due to burns should not be as severe as loads already considered in the design of the plant. Our evaluations show maximum differential pressures for localized lower compartment burns to be less than 11 psi. The 11 psi is based on a flame speed of 30 ft/see and assumes that the flame radiates in a spherical fashion from the center of the area. It has also been assumed that there vns no hydrogen burning in other areas of the containment. These are sxtremely conservative assumptions that maximize the pressure differential. If the flame speed is reduced to 10 ft/sec, the maximum differential pressure on internal structures is reduced to approximately 3 psi. In comparison, the maximum differential pressure for a large LOCA is 14.3 psi. The design values used for internal structures is 1.4 times the calculated differential pressure plus any margins required by the construction and design codes (American Concrete Institute, etc.). Figures D/1, 2, and 3 show the transient results for various burn velocities. It is concluded that nonuniform pressure effects for local hydrogen burning are bounded by present design values. 0
DIVIDER DECK DIFFERENTI AL PRESSURE c! r2 m M
- m
- e. . .
e,. _- . . . ... .... .. .. ... . . . .. . ..... . . . . . . . . . . M
- : 9 NODE MODEL e
- BURN VE10 CITY OF 7.5ft/sec :-
6y ng . .......3.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 100% DURN EFFICIENCY ; -- g3 i -c-CL s_el . W . D [/) gc_. M M - W . . . g j g gy .
+4
- a. . . . . . . . . . .
i a : * -
.g - . : .
m . . .
- 1 E-* ot _ ... . ... .. :. ...... ....... . . . . . . . . . . . . :. . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . .. . . . . . . . . . . . . .
W 3 . Q; m o! : : : & . C .
............i................
[14
- h .
r4 . . Qc. .. ...... ...*. . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . ........ ... .... . . . . . . ........
- f. .
U- 6I*********************** . O- . . . . . . . . -
. . O . .
.j FIGURE D-1 1 Q
&cquagah IGgbragen &tuby
- :* i . : . :
- 5
.o 9
* . O
- l
* ! *O
. . . . . . . . . . . . . . . . . . . . .: . . . . . . . .. .. .:.. . . . . .,O. '
f:Q i : 1 . c g y .................,..... . . . . . . . . . - 5
- 4 b : : . . . :
h -4 $ o Z M
. . . . . y -
w
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . - .. .-.OA .-
A :-
- M N > :
t o : . OZ
- d zdD :
. O O
OJ
= . . . . g o.v0 A Cp *
= ;
m o Fd .
- : o M cy b
........e. . . .
o i : C.) .. !. !.
' C:
. . . . n
? .
2; .
. i. ! .
i.
- * ; : * .o
-g H .
- i : ! -
l : : H . : : : . Q : g N- 6 O'9 7t 0*t 7C O'c 73 0*3 9'I 0'T 70 CV (1sd) 3HOSS3Hd 'IVI1N383ddIG t I l
Dequagah Egbrugen &fuby a 6 m
. . g g
f p:: i ; : i : . : i : : : -e y) : ; . g . . . . . . . . . . . . . . 3................;......................... *
-g
, 8 .
e . :. : - N,
- +
w *
- h>N g .
I C,b O w 7 C . M . *
.C W ; : .
<C w w~b a
N O Z
- - -v -
~ . . . - -
p
... 32 h: C: . . . . . . . . . . . . . . . QA (f)
Z ca >D a c : .
. t M
- c:
O Z Zmg ; g, : . O me cm 3 C4 f w
. - La
. . . . . . . . d. U Q p
- O lf) v
!
- O w -
e s W
- qPce *
.......g................... . . . . . . . .. . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . .
O : *
- i - ; -
- w
. . . . . . .. . . . . . . . . . . . . . O{
. . . . . . ... g. . . . . . . .. . . . . . . . . .
*O 3 .
I w
-f H :. :. :. :.
- j H
...........g.................
g s 8 j . j
. M. .
x g 6 i ' e
- 6 6
6 6 . .
e U1! 8*6 0'8 E4 99 99 Vfr C'C 2'E TI 0'O (Isd) 38flSS3Hd 'IVllN383 MIG I \ . . 5 l l
i G APPENDIX E i 1 CRITICAL COMPONENTS I l
. 1 l
i l SEQUOYAH UNIT 1 l l REPORT OF THE SAFETY EVALUATION OF THE INTERIM DISTRIBUTION OF IGNITION SYSTEM ! l l i i ! l ) TENNESSEE VALLEY AUTHORITY i 1 1 SEPTEMBER 2, 1980 - j i I I ! I i l - i I 1 i O
(_,, APPENDIX E CRITICAL C0!'20NENTS The following table lists components which may be required to function after a hydrogen burn within containment. It was assumed that hydrogen could be generated during any size LOCA. Note that all of the equipment listed is not required for all of the different LOCA events considered. To generate Table E-1, all systems within containment were considered. Those systems which could be required to perform a function following a hydrogen burn were selected for detailed investigation. The I\/ ; individual components of those systems within containment were determined and then evaluated to see if their function could be required. Based on current best estimates of minimum time required for core melt and significant hydrogen production and release, all equipment whose function was completed in the first 20 minutes following a LOCA was deleted. Table E-2 includes all components: which must function after a degraded core event; whose function might be desirable; or which should not be allowed to malfunction.
-~
% j
/
hs
TABLE E-1 Components Required After Hydrogen Burns (O~') Location Component Inside Containment Function Limit cwitches Upper and lower FC767-87, -95, -103, Lower compartment Lower containment cooler
-111 discharge FCV67-295, -296, -297, Upper compartment Upper containment sent
-298 cooler FCV70-87 Lower compartment RCP thermal barrier FCV7-89 Lower compartment RCP oil cooler LT3-148, -156, -164, Lower compartment Steam generator level
-171, -172, -173, transmitters (narrow
-174, -175 range) 30-1AA, 30-1BB Lower compartment Containment air return fans
(~] Penetrations Medium voltage power, low voltage power, (_/ and control and instrumentation Splices Juncticn boxes Lower compartment RTD connections to measure TCS temperature FCV62-61 Lower compartment Valve motor operator FCV63-67, -80, -98, Lower compartment Valve motor operator
-118, -172 TE68-2A&B, -14, -25, Lower compartment Narrow range RTD's
-37, -44, -56, -67, .
-79 (-410, 411, 420, 421, 430, 431, 440, 441)
TE68- 1, - 18, -24, -41, Lower compartment Wide range RTD's
-43, -60, -65, -83 (413, 423, 433, 443)
NC41, 42, 43, 44 Lower compartment Excore neutron detectors H2 recombiners Upper compartment ("') x_-
PT68-322, -323, -334, Lower compartment Pressurizer pressure
-340 (455, 456, 457, transmitters
() 458) PT68-320, 335, 339 Lower compartment Pressurizer level (459, 460, 461) transmitters i PT68-66 (403) Lower compartment RCS wide range pressure transmitter LT3-56, -111 (502, Lower compartment steam generator level 504) transmitters (wide ; range) FT1-3A&B, -10A&B, Lower compartment Steam flow transmitters
-21A&B, -28A&B (512, 513, 522, 532, 533, 542, 543)
LT3-43, 98 (501, 503) Lower compartment Steam generator level transmitters (wide range) LT3-38, -39, -42, -51, Lower compartment steam generator level
-52, -55. -93, -94, transmitters (narrow
-97, -106, -107, -110, range)
-111 (517, 518, 519, gs 527, 528, 529, 537, O
v 3
PRELIMINARY EQUIP!E!iT LIST ,ews TABLE E-2 V System /Cemponent Time Function Comments RCS TE-68-1 0 to end PAM (TE-68-1 and 24 See PAM list
-24 only) & Margin to sat.
-43 Hot leg, wide-range
-65 temp.
TE-68-60 See PAM list Wide-range, cold leg See PAM list 83 temp. PT-68-68 2 hours Loop 4 RHR suction line Need for RHR cool-
/~'i -68A,D pressure. Pressure down V
interlock for RHR suction valve. Possible use with
~
" margin to sat." meter.
Cold overpressure pro-tection input. PS-68-66 2 hours " " PS-68-0>A,D p.
]
s_-
(
%/
') XE&XT-68-363 0 to end Safety valve acoustic Need for valve coni-
-364 monitor toring
-365 XE&XT-68-340A 0 to end PORV acoustic monitor Need for valve moni-
-334 toring PCV-68-334 0 to end PORV & PORV air supply Needed for pressure PSV-68-334A solenoids control and venting PSV-68-334B PCV-68-340A 0 to end PORV & PORV air supply Needed for pressure PSV-68-340AA solenoid control and venting n
U PSV-68-340AB FCV-68-332 0 to end PORV block valve Needed for possible
-333 PORV isolation 1
L,) 1 4 Sj 1 1
PT-68-322 0 to end Used for control of See also PAM list (3 (_/ -323 pressurizer heaters,
-334 spray, & PORV's. Also
-340 provides PAM. ( Also for RPS & SI initiating signals.)
LT-68-320 0 to end Pressurizer level info. See PAM list
-335 (PAM). (Also for RPS
-339 & SI initiation.)
Valve-68-563 0 to end Presurizer safety Under certain circust
-564 valves ces, thee valves coul
-565 be required to operat 7-s
\ or not spuriously ope RHR FCV-74-1-2 2 hr. RCS Cooldown 1. Failure would cak RHR cooldown impossible.
- 2. Actually located in RHR valve roo:
,m, Y
75-505 2 hr. RCS Cooldown Failure or reduced se-point from elevated (~~) temperatures would cause a small LOCA if the RHR were used for RCS cooldown. Safety Injection FCV 63-67 2 hr. Isolate the C.L.
-80 accumulator during RCS
-98 cooldown.
-118 FCV 63-172 2-15 hrs. To initiate RHR cooldown b
s/ or to shift SIS dis-charge to the RCS hot legs FCV 63-72 1/4-6 hrs. Cont'ainment sump isola- Located in the BHR
-73 tion valves to SIS. valve room Na l
7 i
FC7 63-72/73 P.I. O to cnd Logic inputs to RHR (Cear switch & valves atem switch) LT-63-176 0 to end Logic inputs to safety Cont sung level
-177 injection automatic
-178 suction switchover
-179 sequence CVCS FC7 62-69 0 to end RCS makeup-letdown These are valves in t'.
-70 norcal makeup and let-
-72 down path. The seal
-73 injection path does n 0 -74 have any ecmponents
-84 insido containment.
-85
-86 1
l l O 8 1 1 l
62-662 0 to end Relief valve on Failure or reduced s: x, letdown line point frca elevated - eratures would cause small LOCA if the let cause a small LOCA it line were used. 62-636 0 to end Relief valve on RCP Failure or reduced seal discharge line setpoint from elevated temperature: would cause a small ! RCS and Containment Sampling ()' s-FCV 43-1 0 to end Sample RCS and PZR The new postaccident
-2 sampling station wil:
-10 use the existing sam:
-11 lines inside contair.:
-20
-21
-22
/~T 4 t x
FCV 43-201/202/ 0 to end sanple containment " 207/208 atmosphere including ( Also note that addi-the existing hydrogen tional hyd. monitors monitors and penetrations are expected.) Contair.nent Ventilation Air return fans 10 min.-end Containment pressure and hyd. control O O 10
Hyd Recembiners 0 to end Centainment hyd. con-O tro Hyd Igniters 0 to end Hyd. control PAM (as presently installed) RCS Temp TE-68-1 0 to end PLM
-24 TE-58-60
-83 O RCS Press. O to end PAM PT-68-66 PT-68-68 PZR Level 0 to end PAM LT-68-320 LT-68-335 i
O . 1 il
SG Level 0 to end PAM (~'3 LT 3-39 LT 3-52 LT 3-94 LT 3-107 LT 3 h3 LT 3-56 LT 3-98 LT 3-111 RG 1.97, Rev. 2, Draft 2 (6-4-80) The specific instruce Neutron Flux have not been identif Source Range because the RG, Rev. Intermed. Range has not been issued a () RCS Hot Leg Temp specific instrument RCS Cold Leg Temp numbers are not avail RCS Loop Flow able. This list is RCS Pressure only for information
~
PZR Level and possible future
?,
G IA.
PZR Safety and consideration. Relief Valve f- s, j Position Indi-cation Contair. ment Sump Level Containment Iso-lation Valve Position Indi-cation Core Exit Temp. Containment High Radiation SG Pressure SG Leve] gg
\
Cold Leg Accumu-lator Level or Pressure and Isolation Valve Position Indi-cation
, ~ .
k )
- I3
, .-~,n-,.,..~.
l _ . - _ . . -..;, ...:. .- _. _ ._= x = .-- , s:: . . :;. .
- . . . . . - . , . . ~ .
l
Cent-Isol-Valves ( (_,) 1-507 5 min to end Valves must remain open isolation valves 1-506 closed & maintain a cust, of course, clos. 1-501 tight seal to maintain to establish containm< 1-564 containment isolation. isolation, but this 1-505 Additional requirements active function is fu'. I-504 may be specified else- filled prior to any I-563 where, hydrogen generation. I-511 Also note that this 1. 1-510 a listing of valves 1-503 inside containment us 1-566 for the integrated le 1-509 test, and therefore 1-508 includes more valves t 's l 1-502 than those that 1-562 actually perform a 3-860 containment isolatien 3-351B function. 3-351A 3-859 3-924 1-184 1-810 O \J lY
- --m - _- _ * - - ,
d F 1-811 h 1-803 1-183 1-808 1-809 1-181 1-804 1-805 1-800 1-182 1-806 1-807 1-801 26-1261 26-1260 26-1295 26-1296 4 30-8 30-10 30-15 30-17 30-50 30-20 0 t IS
i l l I l l , 30-56 O so-52 l l 30-40 1 , 30-58 l i 30-134 , 31c-223 31c-750 31c-752 31c-225 j l 31C-732 l 31c-749 i 31c-734 l 31c-230 l 31c-713
~
0 31c-715 l 31c-232 31c-695 31c-712
~
31c-697 ) 32-335 32-287 1 32-281 32-292 O Ib
32-297 32-336 32-377 32-373 33-704 43-75 62-656A 43-34 63-614A 43-2 68-576A 43-11 68-575A 43-22 . O 68-548^ 59-633 61-1161 61-192 61-534 61-532 61-1162 61-679 l i 61-681 i O I)
i 61-194 61-691 61-97 61-693 61-692 61-746 61-122 l 62-571 i 62-575 , 62-567 62-569 I t t
)
62-573 1 62-565 ., 62-570 ' I Q 62-574 62-566 62-568 62-572 ' 62-564 62-61 62-639 62-73 62-72 i
! l l
I f i ! I 18 l l
62-74 62-707 62-545 62-544 63-667 63-661 63-112 63-111 63-660 63-659 63-174 ! l 63-24 63-666 1 63-71 ' O 63-653 1 63-121 63-325A 63-326A 63-319A 63-320A 1 63-655 l t 63-321A 1 63-322A l 1 l i l O l l '1
63-323A Q 63-324A 63-654 63-656 63-648 i 63-167 63-560 63-313A 63-314A 63-317A 63-318A 63-649 63-21 63-657 h 63-658 63-315A 63-316A 63-311A 63-312A 63-629 63-172 63-636 63-158 O ao
63-642 67-691D 67-579D 67-581D 67-298 67-584D 67-692D 67-691B 67-579B 67-581B 67-6923 67-297 67-584B 67-579C O 67-581C 67-296 67-584C 67-6920 67-579A 67-581A 67-691A 67-295 67-584A s O 21
67-692A 67-572D 67-111 67-574D 67-695D 67-700D 67-572D 67-5670 67-561D 67-564D 67-697D 67-103 67-695B 67-574B O 67-5728 67-564B 67-561B 67-562B 67-567B 67-295C 67-95 , 67-574C 67-572C O u
- 67-567C 67-561C 67-562C 1
67-565C I 67-572A l 67-87 ! 67-574A 67-695A 67-572A 67-567A
; 67-561A 67-562A 67-564A 67-797A -
O 68-308 i 68-574A 68-574B 68-562 68-560 68-571 70-702E 70-702B i 70-691B l lO 23
. _ _ _ _ _ _ _ _ _ _ _ _ , , ,,-e-w- ,%.., , , - - - .y, . - , .. r..,_.,,- ,- _,,m,,n ,,,--.-r y-,-, gym.,e,,,,,,y..p
70-692 O 70-693 70-679c 70-680 70-702F 70-89 70-698 70-734 70-737
- 70-87
- 70-697 70-736 63-172 63-636 74-503
- 74-2 74-504 74-502 i 77-518 77-9 77-519 4
77-517A 77-517B i l l 1 O l a+
77-127 g 77-906 81-528 81-502 78-558 78-560 78-559 77-16 77-508 77-18 77-591 > 33-214 43-207 43-208 O 43-202 43-201 90-108 90-109 90-110 90-114 90-115 90-116 87-523 l O , s I 25' \ i _ _ . . . . . . _ - - . . . . . - _ _ - _ . _ . . - . _ , _ _ _ _ _ . ~ . . _ . . -
87-7 O 87-8 77-867 77-886 77-848 77-880 61-744 61-745 62-563 62-561 6:!-562 62-560 62-54 62-13 O 63-633 67-580D 67-585D i i 67-580B 67-585B 67-580C 67-585C 67-580A 67-589 ; 1 l O u-
67-575D j 67-575C j 67-575A 5-100 30-571 30-572 30-573 61-680 61-533 X-003 0 to end Blind flange provides Blind flanges depend X-054 containment boundary on 0-ring seal to fur. X-079A tion X-079B X-lli X-ll2 X~113 Electrical 0 to end Provides Containment Electrical penetratic - Penetrations boundary may use caterials su, as epoxy to provide ' seal. O V 17 -
RV Level Sys. O to end Indicates coolant The system design has t (a) (All instruments level in RV not progressed sur-
& valves inside ficiently to date to containment) identify individual v2
& instrument numbers.
RV Head Vent Sys. O to end Provides ability to The system design has (All instruments vent noncondensibles not progresses suf-
& valves inside from RV head to ficiently to date to containment) containment identify individual valve & instrument numbers.
Airlock seals 0 to end Maintain a containment (~\ s)m equipment hatch boundary
& personnel air-locks Divider barrier 0 to ice Maintains pressure seal melt differential between upper and lower containment compart-ments to assure o
2B
l proper flow through r~' ice condenser V..)t Ice Condenser Upper & Lower 0 to ice Open to direct airflow Doors nust not close doors melt through ice condenser & jam upon local Hyd. burns Auxill.ary Feedwater T-3-171 0 to end Provides input to AFW T-3-175 LCV's to control SG T-3-164 level to provide T-3-174 containment isolation for large LOCA.
,3 N_ ) T-3-156 Provides input to AFW Time requirements may T-3-175 LCV's to control SG vary considerably, T-3-148 level for decay heat depending on ability T-3-172 removal for small & to achieve cold shut-TMI LOCA's, down.
Cor.conent Cooling Water 1-FCV-70-87 0 to end Containment isolation Pump seal protection, valve en RCP thermal if needed, can also te barrier. Must be provided by seal opened after cont. injection. phase A isolation to protect pump seal integrety if desired. Valves 683A 0 to end Spring safety valves on 683B CCW RCP thermal barrier 683C loops. 683C 1 DE01;Atti 30 . t n
- ~
w ,-- - - - , - , , , . ,
i s i 4 i APPENDIX F 4 I t TilERf'AL EFFECTS ON COMPONENTS i { f i f ,I SEQUOYAH UNIT 1 ! PEPORT OF THE SAFETY EVALUATION OF THE INTERIM I DISTRIBUTION OF IGNITION SYSTEM 1 i 1 !O 4 4 1 i TEN!!ESSEE VALLEY AUT!!0RITY '
~
! SEPTEMBER 2, 1980 i i l ) 3 l - 1
1 I j' - l APPENDIX F THERMAL EFFECTS ON COMPONENTS i TMI experience showed the temperature rise in instrumentation as a result of the hydrogen burn to be on the order of 50 F. This represents the best information presently available on the temperature response of instrumentation. Present analytical tocls for evaluating 4 the transient response of an ice condenser containment to a hydrogen i burn are not sufficiently sophisticated to be used in the evaluation of small components. The co;tainment models use extremely conservative assumptions and simplifications to provide conservative upper bound calculations of the containment response. For. example, the CLASIX code does not model the containment structural heat sinks O and the computed atmospheric temperatures are above the adiabatic flame temperature for hydrogen in a steam-air mixture. Studies performed Ly Westinghouse of the Sequoyah containment response that resulted in the maximum temperature for an S D 2sequence assumed that 900 pounds of hydrogen burned. Estimates of the amount of hydrogen burned at TMI range from 450 to 1160 pounds.
.Several conclusions can therefore be drawn:
- 1. The actual temperature rise from a hydrogen burn can be expected to be close to those observed at TMI.
O
. I
- 2. The TMI temperature profile is representative of the response of the most sensitive instrumentation to a hydrogen burn.
Thermocouples are at 1 cast as sensitive to temperature changes as other equipment due to their purpose and size. 3 Advancements to the CLASIX code (which have been planned) are required before analytical results (such as radiative heat transfer) that have reasonable levels of conservatism can be produced. O O
.1
-.. . . - ~ . - - . _ _ . . _ _ . _ _ . . - _. . . . . . . - - - - _ - . _ _ _ _ _ . __ -
l !h i APPE!! DIX G i
- PRESSURE, SHOCK, AND TRA?JSIE!!T LOADS ON COMPONE!!TS i
i i } l l 1 1 SEQUOYAH UNIT 1 i l REPORT OF THE SAFETY EVALUATIO?I 0F THE If;TERIM i DISTRIBUTIO'I 0F IC'JITION SYSTEM , i l 9 ) i-i TEt:NESSEE VALLEY AUTHORITY SEPTEMBER 2, 1980 - i 1 i e l l l l - S
APPENDIX G PRESSURE, SHOCK, AND TRANSIENT LOADS ON COV20NENTS 7 N.] The shock wave generated by a detonation is very rapid. The response time of components within containment are such that they will not respond to this very rapid shock wave. Static pressures or pressure changes due to a burn will not prevent the components from continuing to operate because the p.w- "tres to which they are qualified are greater than the calculated containment yield pressures. During the pressure transient, instruments which have a vented reference will indicate an incorrect (low reading due to the increase in their reference. Once the transient is complete, the instruments () will return to the correct reading. This scenario is confirmed by a review of steam generator pressure plots from TMI which show a low steam generator pressure during the time of the pressure transient which was initiated by the hydrogen burn. Thermal effects on components are described in Appendix F. k l N/ I j eaw h:a ~. ,. L. .
- C ',
_ _ . . _ _ _ . = - _ . _ . _ _ - - _ _ _ .- __-. _-. _.__. . . . . . _-- -. Y I 1
- O I
l APPE!! DIX H C0!! TAI!!ME!IT RESPO!!SE TO DETO'IATIONS 1 l 1 SEQUOYAH Uf!IT 1 f REPORT OF THE SAFETY EVALUATION OF THE I!!TERIM i DISTRIBUTION OF IGflITIOff SYSTEM O TE?IIIESSEE VALLEY AUTHORITY SEPTEMBER 2, 1980 ~ i a
=
b i . 9
APPENDIX H CONTAINMENT RESPONSE TO DETONATIONS \_/ The introduction of igniters into containment has greatly reduced the possibility of hydrogen buildup to detonable limits within the containment during a degraded core event. With successful operation of the igniters, only small pockets of detonable gas would be possible. The containment response to a detonation is very dependent on the duration of the shock wave in comparison with the response time of the shell. Assuming a single degree of freedom system, loads which are applied sicwly are static in nature, and the containment response f) or deformation can be determined from static analyses (i.e., th'e
'%./
dynamic amplification will be unity. For a suddenly applied constant load, the structural response will vibrate with a deformation twice what one would expect from a static determination, i.e., a dynamic amplification of two. Thus, any load such asla blast wave which was suddenly applied to the containment and was maintained for a long time relative to the period of the structure (on the order of 40 periods) would result in the maximum deformation of about twice that expceted from static analysis. If the load decays more rapidly than this, the deformation is reduced. If the suddenly applied load decays very rapidly with respect to the period of the structure, the load will be impulsive. Deformation of the structure under impulsive load is not dependent on the peak load but rather on the time integral (s) of the lead. Thus, the containment can withstand a pressure peak
/
if the loading on the containment is of several thousand lbs/in sufficiently short in duration. The dynamic amplification would be ,
/h
(' / much less than one. In reference H.1, the pressure pulse forces are calculated for several detonable mixtures and are compared to the Surry Nuclear Plant containment using the relationship: g I ot p 0.32 Pd T I where I is the impulse pressure. A t is the time duration of the detonation wave in seconds. P is the loading that produces the maximum elastic d I . deflection for the structure. T is the natural period of the structure.
/--
-5 see was taken as an upper bound for the duration In WASH-1400 10 of the detonating wave.
Okrent and Chan (reference H.2) applied this study to an ice condenser plant. Based on the McGuire containment, they have estimated the i It is taken as about maximum elastic deflection for the structure. twice design pressure at 60 lbs/in a. Also, the natural period of the i structure was assumed to be about the same as the Surry Plant used in WASH-1400. Using this information, the ice condenser containment Using is expected to withstand impulses up to about 0.4 lbs/in -sec. 2 a detonation pressure of 1725 lbs/in as given in Wash-1400, an r~s b :t
, , ~., 4 ,g
.:Ofz:.'
2 -see was obtained, far below impulse loading of only 0.017 lbs/in O the estimated containment capability (see also Appendix J). V l Staff A Los Alamos evaluation for TMI-2, as discussed in the Technica - ANALYSIS REPORT ON CHEMISTRY to the ld PRESIDENT'S g ACCIDENT AT THREE MILE ISLAND showed that an explosion at TMI-2 wou [ ltimate V produce a load which was Jess but close to the building's u fi It was the conclusion of the staff's report that the pulse strength. hN i: duration of 10 microseconds in WASH-1400 is based on the thickn They reason that the shock front would strike of the shock front. re the wall and be reflected, thereby producing a sudden rise in pressu - This increased pressure level would be sustained until on the wall. rarefaction waves followed the shock and lowered the pressure. For - l , this reason, they believe that the WASH-1400 results are probab y .- 4 in error, although the conclusion arrived at concerning the WASH-1 00 The approach to the detonation problem has results may be correct. The assumption has been a large open volume 5 been oversimplified. - In actuality, the mixture in the with a uniform detonable mixture. iig containment will not be uniform; spaces are generally small conta n n N-large equipment which tend to break up and dissipate any shock waves, 1 the and in the case of the lower compartment of an ice condenser, steam content may be too high to support detonation. It is difficult to identify the impulse loading capacity of a local region of a containment structure; however, the general structure i d Chan of the Sequoyah plant is on the order calculated by Okrent an (reference H.2) for the McGuire Nuclear Plant. J _g=n-c.,
' ' ' -y- ,,% .
As indicated in the Kemeny Commission technical staff's report on -s U chemistry, the key to the question of containment failure due to The hydrogen detonation is the duration of.other pulse loading. w conclusion drawn by the staff's report is that the containments have The inclusion only marginal capability to withstand a detonation. i of an igniter system greatly improves that margin. k In Sequoyah, the igniter system will prevent the buildup of large 5EE
- j concentrations; henca only small pockets of high regions of high H2 s
b Detonations of these pockets would result concentration could exist. in a much reuuced challenge to the containment than previously considered. Because the amount of gas detonated would be small, the duration of the pressure pulse, between the time of the arrival of _ the detonation wave and the ensuing rarifaction, would be considerably O shorter than the few millisecond impulse duration estimated by the Kemeny Commission staff on chemistry. Another factor which will [ reduce the impulse is the change that takes place at the edge of the ~ At this point, the detonation wave is no longer detonable mixture. sustained, and the shock wave will begin to subs'lde. If sufficient distance were available, the wave would eventually degrade into an Thus, both peak and duration are reduced acoustical disturbance. in local detonations. In the Sequoyah containment, the lower compartment shell is protected from detonations by the crane wall. Detonations outside the crane wall are extremely remote possibilities, but in the event of an occurrence, the duration would be extremely short because of the very V- :
)
- - ~ , &cr.h- .;_.
...-~,<n
. "===aw===mesensus i
, s v: . ..
cannot occur because of the operation of the igniters in the lower (m () compartment and the upper. A remote possibility of a local detonable mixture exists in the upper plenum of the ice condenser. Again, no damage would be expected because of the very short pulse duration. In addition, the many pipes exiting the ice bed and the open doors exiting the upper plenum would significantly contribute to a dispersal of the shock wave. We thus conclude that the Sequoyah containment with an operating igniter system has ample margin to sustain detonations. References
- 1. WASH-1400 ,
h' 2. C. K. Chan, "On the Failure Modes of Alternate Containment Designs Following Postulated Core Meltdown," UCLA-ENG-7661, June 1976, Principal Investigator D. Okrent, pp. 58, 59,'60, and 93~ f (/
+ 9 ., APPETDIX I i l ! C0'ITAI!IMENT A!!D ASSOCIATED SYSTEMS ! I i i i ! 1 i ) SEQUOYAH UtiIT 1 l REPORT OF THE SAFETY EVALUATI0t10F THE I!!TERIM i , DISTRIBUTI0tl 0F IG!!ITIO!! SYSTEM s !e l i l TEllMESSEE VALLEY AUTHORITY l I SEPTEMBER 2, 1980 ~ l i l t O O _ _ _ _ ~ _ _ . . _ _ _ _ _ _ _ _ . . _ _ _ . _ _ _ . . . . _ - _ _ _ _ _ _ _ _ . _ _ _ _ _ _ _ .-
APPENDIX I /' CONTAINMENT AND ASSOCIATED SYSTEMS The containment system is designed to assure that an acceptable upper limit of leakage of radioactive material is not exceeded. Aiding the containment are several associated systems whose functions support the overall containment system function. In general, the containment system acts as a physical barrier to radioactive material release. The associated systems function to provide containment heat removal capability and/or prevent hydrogen accumulation. (Excessive pressures inside containment from excessive heat or hydrogen combustion could compromise containment system integrity.) The associated systems include: (1) Ice condenser (2) Containment spray systems (3) Air return fan system - (4) Combustible gas control system Adequate heat removal capability for the containment is provided by the ice condenser, the air return fan system, and two sepa' rate containment heat removal spray systems. One of these heat removal spray systems is the containment spray system, and the second , is the residual heat removal spray system which is a portion of the j l residual heat removal system. The combustible gas control system works to prevent excessive hydrogen concentration inside containment. I
1 The containment and the above associated systems are discussed in more detail in the following sections.
- 1. Containment System The Sequoyah containment system is designed to assure an acceptable upper limit of radioactive material leakage by containing the radiation, mass, and energy released from a breach of the reactor coolant system and the nuclear fuel cladding.
The containment system is a dual containment structure composed of a primary containment and a secondary containment which encloses the forrLer. a O' It is the primary containment which acts to physically contain the radiation, mass, and energy resulting from an accident. For this reason in this appendix and in the other sections of this document primary contair. ment is referred _ to as containment. The containment is a freestanding, welded steel vessel with a vertical cylinder, hemispherical dome, and a flat circular base. Free volume of the containment vessel is 1.2 million cubic feet. The design internal pressure for the containment is 12 psig, ano the design temperature is 250 F. The design basis leakage rate is 0.25 percent /24 hr. The containment is subdivided into three compartments: the upper , compartment, the lower compartment, and the ice condenser. The
-lower compartment completely encloses the reactor coolant system 1
equipment. The upper compartment contains the refueling canal, refueling equipment, and the polar crane. The ice condenser, which connects the lower compartment to the upper compartment, is discussed in a later section. The secondary containment encloses the containment (primary) and , provides an effective barrier for airborne fission products that , may leak from the containment during a LOCA. Gases which leak from containment are diluted with the air enclosed within the secondary containment, held up, and filtered before being bled to the environment. The containment type used at Sequoyah was selected for the following reasons:
- 1. The ice condenser containment can accept large amounts of energy and mass inputs and maintain low internal pressures -
and leakage rates. A particular advantage of the ice condenser is its passive actuation not requiring an actuation system signal.
- 2. The containment combines the required integrity, compact size, and a carefully considered advanced design desirable for a nuclear plant.
3 The double-enclosure concept affords minimal interaction between the containment vessel (leakage barrier) and ("% (_) the reactor building (protected structure), a margin of
-]
l
~ _ _ . - _ . - . ,. .. _ _ , - _ - _ , _ - _ , _ - - - -_ .
i 1C8 l nouar 1 condenser i O .
.i u _-
-r m
- l j -'
,M__ a w_2=
- _]
cm
< a -- -
r - . , - s , AIR NAnDLlu3
%j -
CRAhE 4,
. imTERut0IAT '
DECK
%4 i ,
AIR DISTRIBUTION
', [M. ..j
'" 'I ICE sAsxET h 3g g
*> CRAuE WALL q ,y.!
. a: -
LATTICE FRAwE Dd % , f(NEATOR ENCL 0suRE
D R; Y't./ /3 .. .
.l OO; 0: %;.* . ..;,'
l., t.*.'.
' ' : .*..::,$E' 1 l SHELL
?t *,
3
/ ~
e , WALL PANELS , ll , l
, ='.
STE'AW GENERATOR f, : *, VAuts
.s r
f
OR Y ,
% ;9 .- . , -
^'
MS ,....? h ', f. .*#'p QF ..
,,.f g
p
< w -
j - DIVIDER DECK s, Q LDWER i c:..'. i . - INLET 000R3
.h..
' a. .').
' b . 'J'.*
4 : /, /^
LOWER SUPPORT STRUCTURE O v . ( \
conservatism in leakage rate from the use of two structures A
!j and the EGTS, and a reduction of gaseous and particulate radioactive releases due to annulus mixing and holdup prior to filtering and release.
- 2. Ice Condenser The ice condenser is one of the three compartments inside primary containment. It is designed to limit the containment pressure below the design pressure for all reactor coolant pipe break sizes up to and including a double-ended pipe severance of the largest main reactor coolant pipe.
The ice condenser concept utilizes a large mass of ice to condense escaping high-energy steam from postulated loss-of-coolant accidents (LOCA) of steam line break accidents. The rapid condensation of steam in the ice bed keeps the maximum containment pressure relatively low while maintaining the capacity to absorb a continuing high energy input from the reactor core and reactor coolant systems. The ice condenser is made up of 24 individual bays which form a 300 arc inside containment. Each bay consists of three major sections: A lower plenum; an ice bed; and an upper plenum (figure 1). The lower plenum is isolated from the lower compartment by doors in each bay that open at a differential pressure of 0.007 psi. The ice bed contains a minimum of 2.45 million pounds of ice. The ice is stacked in columns one foot in diameter and 48 feet high. The upper plenum contains cooling bV units used to maintain the low ice bed temperature during normal E t i
plant operations. The upper plenum is separated from the ice bed and the upper compartment by two sets of doors that will open with a differential pressure of 0.028 psi. In the event of a LOCA, steam pressurizes the lower compartment which opens the lower inlet doors. An air-steam mixture enters the ice bed where all the steam is condensed. The rising air then causes the top two sets of doors to open and then flows into the upper compartment. To provide maximum use of the ice bed, air return fans are provided which circulate containment atmosphere through the ice bed condensing steam released after the initial pressurization. When all the ice has melted, the spray system located in the upper compartment removes the remaining energy released to the containment. O 3 Containment spray systems The containment spray systems consist of two separate trains of equal capacity with each train independently capable of meeting system requirements. Each train includes a pump, heat exchanger, ring header with nozzles, isolation valves and associated piping, and instrumentation and controls. During normal operation, all ! of the equipment is idle and the associated isolation valves are
- closed. Upon system activation during a LOCA, adequate containment cooling is provided by the containment spray systems i whose components operate in sequential modes. These modes are
l (1) spraying a portion of the contents of the refueling water storage tank into the containment atmosphere using the containment 1 1
/
,- , ,- y.-, - ,, ,-. ,o -
, , y , - .---,-- ---n4-
spray pumps; (2) after the refueling water storage tank has been drained but while there is still ice remaining in the ice condenser, recirculation of water from the containment spray pumps through the containment spray heat exchangers and back to the containment. This spray is useful in reducing sump water temperatures and increases the effective life of the ice; (3) diversion of a portion of the recirculation flow from the residual i heat removal system to additional spray headers. The latter operation occurs in the event the containment pressure starts to rise after the ice condenser has been depleted. The diversion will be by manual operation of system components. The spray water froc the containment and RHR spray systems will , be returned from the upper compartment to the lower compartment t through two 14 inch drains in the bottom of the refueling canal. A small portion of this spray water is diverted to the floor and equipment drain sump through four 3 inch drains located on the~ operating deck. The primary design basis for the heat removal spray systems is to spray cool water into the containment atmosphere when , appropriate in the event of a loss-of-coolant accident and thereby i ensure that the containment pressure cannot exceed the containment shell design pressure of 12.0 psig at 250 F. This protection is afforded for all pipe break sizes up to and including the hypothetical instantaneous circumferential rupture of the reactor coolant loop resulting in unobstructed flow from both pipe ends. , The containment spray system supplements the ice condenser until 7 n-- - - - , , *- w -
- all the ice is melted (approximately 7,000 seconds after the LoCA) at which time it arid the residual heat removal spray system become the sole systems for removing energy directly from the containment. The containmerit heat removal systems are designed to provide a means of removing containment heat without loss of functional performance in the postaccident containment environment and operate without benefit and maintenance for the duration of time to restore and maintain containment conditions at atmospheric pressure. Although the water in the core after a loss-of-coolant accident is quickly subcooled by the emergency core cooling system the design of heat removal capability of each containment heat removal system is based on the conservative assumption that the core residual heat is released to the containment as steam which eventually melts all ice in the ice condenser.
- O The secondary design basis for the containment heat removal spray systems is the suppression of steam partial pressure in the upper volume due to operating deck leakage from a small break before a full loss-of-coolant accident. The requirement is that the containment spray systems be able to absorb the steam leakage through the operating deck at the maximum possible long-term deck
- differential pressure of one pound per square foot equivalent to the ice condenser door opening and differential pressure in I the upper compartment due to dec leakage in the " double accident" situation. The " double accident" is a small break followed by a large break up to the double ended severance of the largest pipe.in the reactor coolant system.
i e
The system is designed such that both trains are automatically started by high-high containment pressure signal. The signal actuates, as required, all controls for positioning all valves to their operating position and starts the pumps. The operator can also manually actuate the entire system from the control room. Either of the two trains containing a pump, heat exchanger, and associated valving and spray headers is independently capable of delivering the design flow of 4750 gpm.
- 4. Combustible Gas Control System The combustible gas control system is designed to monitor hydrogen levels inside containment and, should concentrations exceed preset levels, act to control the hydrogen gas in the containment O atmosphere. The combustible gas control system is composed of the following subsystems:
(1) Hydrogen monitor _ (2) Hydrogen recombiners (3) Hydrogen purge The hydrogen monitor sutsystem is discussed in detail in appendix K and will not be examined herein The hydrogen recombiner subsystem consists of two electric hydrogen recombiner units, located in the upper containment j compartment, and separate control panels and power supplies for O each recombiner unit, located outside the containment in an area 9
that is accessible following a loss-of-coolant accident. The
/'N U recombiners are completely redundant.
The recombiner unit consists of a preheater section, a heater-recombination section, and an exhaust section. Containment air is drawn into the unit by natural convection, passing first through the preheater section. This section consists of the annular space between the heater-recombination section duct and the external housing. The temperature of the incoming air is increased by heat transferred from the heater-recombination section. This results in a reduction of heat losses from the unit. The preheated air passes through an orifice plate and enters the heater recombination section. This section consists of a thermally insulated vertical metal duct
' enclosing five assemblies of metal-sheathed electrical heaters.
Each heater assembly contains individual heating elements, and the operation of the unit is virtually unaffected by the failure of a few individual heating elements. The incoming air is heated to a temperature in the range of 1150 to 1400 F, where recombination of hydrogen and oxygen occurs. Finally, the air from the heater-recombination section enters i the exhaust section where it is mixed with cooler containment air and discharged from the unit. l l Tests have verified that the recombination of hydrogen and oxygen l in the unit is not the result of a catalytic surface effect but l > s occurs as a result of the increased temperature of the process 10 , l i
. - --=
gases. The performance of the unit is, therefore, unaffected b'v by fission products or other impurities which might poison a catalyst. The hydrogen purge exhaust subsystem consists of a single penetration in the primary containment wall equipped with two normally closed, remote manually operated isolation valves, one on either side of the containment wall, and one pneumatically operated annulus purge exhaust valve located within the annulus. 2 With these valves open, a flow path is established between the primary continment and annulus which will permit purging of the containment for hydrogen control subsequent to a LOCA. Tne impetus for flow will be provided by a differential pressure f 3 at least 0.5 inches of water gauge which will be maintained by the' annulus air cleanup (emergency gas treatment) system. The containment effluent purged for hydrogen will flow directly to the annulus where it will mix with the annulus atm'osphere and
~
be filtered by the air cleanup system prior to discharge. In the event that the hydrogen concentration exceeds 3 percent this subsystem can be used for containment purging. This f subsystem is designed to provide a backup to the redundant inplace i , hydrogen recombiners. It's use is not anticipated in l a design basis accident even if one recombiner system fails. The actuation of this system is covered in TVA operating procedures. The system is described in section 6.2.5 of the SQN FSAR and the environmental consequences are analyzed in section 15.5.3 of the SQN FSAR. Il
=_. - ~ _ _ . - - - _ - _ . , . . _ - . _ - _ . ,
d #N Use of this system is not consistent with the use of the Interim Distributed Ignition System. The purge subsystem was analyzed assuming the design basis LOCA. For degraded core events the : hydrogen release is faster and occurs earlier. If the purge sub-system were used for these events, large offsite doses would occur without appreciably reducing the containment hydrogen { concentration significantly, at least in the short term. The system was required by Regulatory Guide 1 7 and was intended for operation several days after the event (see figure 6.2-96a of , the SQN FSAR). TVA's studies of the containment response to degraded core events has shown that the plant can safely withstand hydrogen
- (} concentrations in excess o' 10 percent. Therefore, when use of the interim distributed ignition system is approved by NRC for use, TVA will' modify the appropriate operating procedures to prohibit use of the purge subsystem any earlier than four days after a severe loss of coolant accident. This time period allows sufficient time to evaluate the nature of the event and obtain samples of the containment atmosphere, while allowing sufficient time before the subsystem is needed for the design basis event (greater than 8 days).
- 5. Air Return System The primary purpose of the air return fan system is to enhance
( the ice condenser and containment spray heat removal operation I 1.
. __ _ _ - ~ . . _ . - - . _ - . __ . _ . - . . _ ._,_ _
by circulating air from the upper compartment to the lower compartment, through the ice condenser, and then back to the upper compartment. The operation will take place at the appropriate time following the design basis accident including loss-of-coolant 4 accident. The secondary purpose of the system is to limit hydrogen concentration in potentially stagnant regions by ensuring a flow of air from these regions. There are two 100 percent capacity air return fans. Each will remove air at the rate of 40,000 cfm from the upper containment through a main duct to an accumulator room of the lower compartment. (See figure 2.) The discharged air will flow from each accumulator room through the annular equipment areas into the lower compartment. Any steam produced by residual heat will O mix with the air and flow through the lower inlet doors of the . ice condenser. The steam portion of the mixture will condense as long as ice remains in the ice condenser and the air will continue to flow into the upper compartment through doors at the top of the ice condenser. Each main duct contains a non-return damper which prevents flow from the lower compartments to the upper compartment during the initial stages of a loss-of-coolant accident. a Both fans will start automatically after receipt of an isolation signal. In addition, either fan may be controlled manually from the control room. Each fan can develop sufficient head to keep the non-return dampers and ice condenser inlet doors open after !
./"T/ ;
. blowdown is complete. A flow indicator upstream of each fan and )
i l- , I3 \ l
a pressure differential indicator across each fan are provided. (l In addition,.the discharge flow rate is indicated in the control room. , Simultaneously with the return of air from the upper compartment to the lower compartment, post-LOCA hydrogen mixing capability is provided by the air return fan system in the following regions of the containment: containment dome, each of the four steam generator enclosures, pressurizer enclosure, upper reactor cavity, each of the four accumulator rooms, and the instrument room. 4 These regions are served by seven hydrogen collection headers which terminate on the suction side of either of the two air return fans. A schematic of this system is shown in figure 2. !O ,1 l i HYD.12 1 1 h i. k 4
)
I+
, - _ . _ _ __ _ .. _ .~ _ _ _ _ _ . -_ _ _ . _ _ . _.-
l, ;) I ll
\ .
}
[ f e 0h m
- R f I.
og S O c 5 TA N 5 M I
)L E
- 1
/ _}
L U O M RO W2 O - 0 S N L S A U F 4 E F C 9 V N C E R A T E T U A R N T C U E E M R DI G I N NF I I 5 IR A 7 N T 3 A S O N - R R O O E D C B E 5 T A : MS 2 M L E T U A 5 7 R U OO O NB m 3 O M R N1 .2 . e k - T C E L U t s C y A S C S E S A R E n V a t R Y F EI T 8 R PV O n PA T r
, 8 UC C u 8 t A e
\
S R R w1 EO MT R E
.E Ts i
r
" (
)
O DEOC i.
- - A O
Z .
. T d, T L L I
R
' 5 N f o
i ( NO EC U S I f 7 E MM O c
)
S I U O i 5 M E l R T R t a 0 NI N R t I E P - S m 9A G ,
" - - e 1
( TO - ' IN h c NR i. . S OD CY H 2 5 e 7 / r 3
/ R O
i F u g _ 5 T S k7 t f
- 2 A L M R
O 3 S U OO T R S M R C O R U E T O C L A T C L R C A O E E C N L E N L N E GA F C O G MN O AR N R E D EU G I l R Y T T O O H S E R 0 T R D 4 A Y L M R H I U O ( O I ) A 1 M U R C
. I 8h C A
~*
m* :E { y .
- o. E*3* $#
= _= . _
I 1 i i j 2
- e f APPE!! DIX J CONTAINffE!!T STRUCTURAL CAPADILITY A!!ALYSIS 1
i b l 1 4 1 I 1 SEQUOYAH UNIT 1 } PEPORT OF THE SAFETY EVALUATION OF THE INTERIM i j DISTRIBUTION OF IGilITION SYSTEM i l l l TENNESSEE VALLEY AUTHORITY j SEPTEMBER 2, 1980 - 4 J I i i 4 l l O \ l I
APPE!! DIX J [] C0!1 TAI!!!2!iT STRUCTURAL CAPABILITY A!!ALYSIS v An evaluation of the containment capabil.'.ty to withstand pressure has been performed. The burning of hydrogen in the containment produces a pressure transient which is essentially static when compared to the response time of the containment vessel. Hence, the evaluation has focused on static pressure application. On the other hand, detonations may be possible within some areas of the containment. Detonations generally have very short periods when compared to the response time of the containment and are, therefore, impulsive loadings on the containment should they occur. Some scoping analysis of the impulsive loading capacity has been made. It is (e m
's- recognized that the actual capability is dependent on many variables and may be quite different at various containment locations.
Influences such as pulse shape, flexibility of the containment, obstructions, etc., can have a large effect on the loading capacity. TVA plans to review these influences in more depth as appropriate to achieve a better understanding of the containment capability. Detonation Load Capability TVA has considered the case of 100-percent metal-water reaction (about 2,000 pounds of hydrogen and about 25 percent of hydrogen by volume). We considered that the containment was a simple cylinder (which is structurally weaker than a sphere) consisting of only a 1/2-inch-thick steel shell (the minimum shell thickness at Sequoyah) . We used the n k_) " impulse" loading information provided by C. K. Chan. We have l
concluded that failure of the containment wall due to detonation shock wave is not expected to occur; however, even though the containment t {} can withstand the detonation loading, due to it's short duration, the resulting relatively long term pressure due to the oxidation of a large amount of hydrogen would exceed the ultimate capability of 1 the containment. This would be true for 600 Kg of hydrogen uniformly distributed and completely burned in one short period. (A 68-percent metal-water reaction would generate 600 Kg of hydrogen. If all the hydrogen is released to the containment and uniformly distributed, l it would represent about 18 percent by volume of hydrogen. This could be a detonable mixture.) l l l Static Pressure Capacity The calculated strength of the Sequoyah containment vessel is.33 1 lbs/in g based on tha actual minimum yield strength of the SA 516 grade 60 material and 43 5 lbs/in g2 based on the minimum specified (} j ultimate strength. These two pressure capacitiee are conservative because: i (1) Based on the Von Mise criterion, an increase of 15 percent is allowable for the yield strength over the value based on the maximum shear stress. (2) An increase of 10 percent in the ultimate strength of the material results from the use of the certified test report (CTR). (3) A margin of 26 percent over minimum ultimate is available when the CTR properties and Von Mises criterion are used. O - x w.wwww w
~^^ (4) Short duration loads (quasi-static or impulse) may give the a
containment an apparent strength response which would exceed its static properties. The above is based on an evaluation of all components of the containment boundary, i.e., shell plate, stiffenera, penetration, valves, and welds. n The evaluation of membrane stress, effect of stiffeners, discontinuities, etc., were based on standard stress-strain relationships which can be found in most textbooks on plate and shell theory such as " Theory of Plates and Shells" by Timoshenko et. al., McGraw Hill Book Company, New York 1959. The results were also
,f s
(_, ' confirmed by a computer analysis using the ANSYS computer program. Evaluation of penetration valves and other components were performed either in a similar manner or have been tested under high pressure conditions. A detailed report will be prepared documenting the various evaluations and will be available later.
-w l
3
'N** f #
#_9 =
i t G APPE!! DIX K l I l C0i!TAINMENT M0!!ITORING SYSTEM l SEQUOYAH UNIT 1 i REPORT OF THE SAFETY EVALUATION OF THE INTERIM DISTRIBUTION OF IGNITIOff SYSTEM O TENNESSEE VALLEY AUTHORITY SEPTEMBER 2, 1980 ! G I l
. . - _ , - , , _ . . _ - . . - - - , . - _ - - . _ - - - - _ - - _ . . - . . . . _ _ . _ _ , - . _ . _ . _ _ ~ _ . - . - , - _ _ - _ . . . . _ _ _ _ . _ _ . _ _ . .
. APPENDIX K CONTAINMENT MONITORING SYSTEM The containment monitoring system consists of hydrogen analyzers to monitor hydrogen concentrations inside containment, pressure control loops to monitor differential pressure between the annulus and the containment, and temperature sensors distributed throughout containment to monitor containment air temperatures.
Hydrogen Monitors Detailed information on the h'ydrogen monitoring system may be found in Volume I, section A.S.1., of this report. In addition, each analyzer takes a well-mixed sample from several points in both the upper and lower compartment for an indication of an overall average containment hydrogen concentration. f Pressure Monitors Five differential pressure control loops monitor the differential pressure between the annulus and the lower comp a~ rtment of the containment. Four loops have indication and annunciation in the main control room with a range of -1 to 15 lbs/in g as well as actuating ) l the safeguard systems. The fifth loop indicates and alarms in the l auxiliary control room as e backup. There are also two wide-range , pressure control loops with a range of 0-60 lbs/in g and 0-100 2 lbs/in g to monitor large pressure transients. Each wide-range l s l l
,% . - .,~. ww.m. _ - . _
_ _ _ . - _ -wfg _.nw.
,p ;
24w:s:5 M: 4 + . K (
pressure control loop has trained power and indication in the main control room. O Temperature Monitors Thirty-four temperature sensors are strategically located throughout containment with indication in the main control rool. Temperature sensors are located on the intake and exhaust of various containment coolers in both upper and lower compartments to ensure that the air
- temperature requirements for proper operation of equipment are maintained. Other sensors are used to establish an average temperature which is monitored in the main control room to assure personnel comfort during occupancy.
o
/
n 9 e o
i APPEf! DIX L IC'IITER SYSTE" DESIG'I SECUOYAH UNIT 1 FEPCFT OF THE SAFETY EVALUATIO'I 0F THE INTERIM DISTRIBUTION OF IGNITIO?! SYSTEfi l I l l TEN!!ESSEE VALLEY AUTHORITY l SEPTEMBER 2, 1990 l l l
i 1 1 APPENDIX L IGNITER SYSTEM DESIGN The interim distributive ignition system (IDIS) is designed to burn hydrogen inside the containment in the event of an accident in which excessive hydrogen is generated inside the reactor vessel and released into containment. .It is designed to ignite the hydrogen prior to it reaching a dangerously high level. This system is intended to back up the safety-grade hydrogen recombiner system, but it is not a safety-grade system itself. The system consists of 31 thermal resistance-type igniter assemblies distributed throughout containment. The locations of these assemblies are shown in the attached figures, These assemblies (typical assembly (', shown in SQN FSAR figure 6.2-147, attached) are located in three different compartments inside containment (see FSAR figure 6.2-141). Lower Compartment Nineteen igniters are located in the lower compartment; four at elevation 689.0', eight at elevation 700 3', and seven at elevation 731.0'. At elevation 689 0'(see FSAR figure 6.2-142), three igniters are located on the outside of the shield wall. This area is primarily a pipe raceway. The area inside the crane wall is the containment sump and all lighting fixtures in this area have been capped and sealed against the-flood. The area cutside the crane wall is sealed to prevent flooding and also provides access to the containment fan rooms and accumulator rooms above at elevation 693 0'. At elevation (~) k- 700 3' (see FSAR figure 6.2-143), all the igniters are located outside I
c of the crane wall. There are four igniters located in three () accumulator rooms (one in accumulator room number 2, one in 4 accumulator room number 3, and two in accumulator room number 4), one igniter each located in the two lower containment cooling fan rooms (180 degrees apart ), and two located in the area adjacent to the personnel hatch. Seven igniters are mounted inside the crane wall at elevation 731.0' (see FSAR figure 6.2-144). There is one igniter located above each of the four reactor coolant pumps, and s one above the pressurizer relief tank, one between steam generators i 1 and 4, and one between steam generators 2 and 3 Ice Condenser Compartment There are a total of nine igniters in the ice condenser area. At elevation 731.0'(see FSAR figure 6.2-144), there are five igniters () located at.the bottom of the ice condenser. At elevation 792.0' (see FSAR figure 6.2-145), there are four igniters mounted on the crane wall above the ice condenser and below the top deck blanket. . Both sets of igniters above and below the ice condenser are generally evenly distributed around containment. Upper Compartment
~
In the upper compartment there are four dome lights distributed roughly 90 degrees apart. Three of the four dome lights have been replaced by igniters (see FSAR figure 6.2-146). i 4 Igniter Assembly l l
> The individual ign' car assembly (shown in FSAR figure 6.2-147) is
]
( ) a very simple design. /
= man -
M
-- ~ . . - . . - - . , - - , . - . . - - - - - -
y
\,
Power Supply Each assembly is supplied power from one of three 120V ac standby lighting circuits. The three 120V ac circuits are supplied through transformers from the 480-volt shutdown boards which have normal and alternate ac power supplies and in the event of their failure are
~
j fed from the diesel generators. i
; Igniter Assembly Enclosure Each igniter is partially enclosed in a 1/8-inch steel plate box which also houses the igniter 120/14V ac transformer and all electrical connections. (In describing the box, the dimensions are oriented j with respect to figure 6.2-147.) Looking at FSAR figure 6.2-147, the box is eight inches long by eight inches high and six inches deep.
A typical assembly weighs approximately 21 pounds. Each box is j covered by a 1/16-inch thick spray shield which is 14 inches long and eight inches deep. Not shown on the FSAR figure is a copper heat a sink 'on the face of the igniter assembly. This 1/16-inch thick copper plate is six incheo deep and six inches high. The plate is bent outward at a 45 angle away from the box at approximately one inch to each side of the glow plug. It is secured to the glow plug and box by a washer then_a nut which is screwed onto the outer threads of the glow plug. Access to the box interior is made through one side of the box. This side is coated on the interior of the box with a rubber seal material and is attached to the box with four bolts, one in each' corner, d
. i
_ -- , . _ ~, _ _ _ _ _ _ . . _ _ _ _ _ _ _ _ .
4-Transformer () The Dongan transformer lo'cated inside the I/8-inch steel plate box receives the 120V ac from the standby lighting circuit and steps the voltage down to 14V ac.
~
The transformer is wound with class H insulation and is rated for operation in greater than 400 F l environments. Glow Plug The transformer supplies 14V ac to the General Motors AC Division i Model 7G glow plug. In laboratory tests conducted by TVA, at 14V i ac the exposed portion of the igniter glows a red-orange and reaches 1720 F. The glow plug is basically a coiled nicrome wire surrounded l by magnesium oxide inside an Iconel Alloy 601 sheath. 1 CE)
~
l-I' i 4 m l.
l = i
!'i f
O I 7 y ~~
~
.(3) .
818.0ft. 792.0 ft. .
.(4) i s t l
I i ( O l .
; F
{-: i O~
~
i : '
' ' 7 3'- i
*(5) -
(7) m Aia RcruRn 7,
/ u fat 4
=_=
., i 1 -
7 0 0.0 ++. i
*(8) i....
* , ! j
. - L -
T. 6 89. 0f+. a> ~O . _u- l
. 6 69.0 f+. '
l l l 4 5
,,._,-,-n-.--,-_ _,,,,,,----,_..--n,. , , , _ , - - - - - - -
e i l . KEY TO FIGURES i Denotes nor:r.a1 lighting fixtures (LC) - Q Denotes standby lighting fixtures (LS) - @ 1 Fixture Description For=at: "X Y"-
, where ,
s . X is one of the following; L- fixture containing a lightbulb l G- fixture containing glowplug type igniter ! Y is one of the following; A- phase A' electricity B- phase B electricity d C- phase C electricity 1 0 - a
~
l i ' i 1 l i i i 4 I 4 o !O - 6
%-- 5Y '"
wy^$-.
- - 4 7 -m - * - - - = y y ,yw- p-.3 ,,yway, y 4 w ,,,y-.,,y9 gym y,3 -se. gg -y-we,,=e- ,yrwy.--w ey w --y .-.-y,yw..,g-wy g,-.uy--+-- -
------wyw--- -g.nwwy
, r-
] .7 : *- - ,~.'%
's.
v [
\
\
, ,e ,
4 6
, j --
818.0 f t. w\- d/ ; ._ ._._ m , L
.: % \ ,! _ y-d , ;e -~
-- , . , ---s ' ,
u-
,r_ ---3 i,1 .; y / . s y ; l--. 792.0 f t.
- p. -
c-- t + r ==* ur J.m._p{;~ ._ .e - e '
.t
'*n,
~ ~ " " - ~
i i i, ! !i. t r a,
.\. q\.,,/
, \. 1
'17 - * <*-.s
- =:.~e q -)
.A t, X.y :
s . -
~,-
-kr 4 )6c. f..
~ ~ ~% :X Xt -
t, -
. . t. ,
g E. ,, i . _M.7 pf :( . pr - f" ~ }tl'
- *# ' f.fIm l ar', ~saf.-~l
- R : ! f.-~ ~ ---
- =:;._ _: --pt,m w..): g ,u,,e- l' . 2. 54,
- -__. i (m
a 4 - 4i. 1 1 l .
,.r ~ . . ' ;.,r ',F . . .
8 , .
.-,,n ,.
y p. , 731.0 f t.
.d ,1 '. s - 4 cg u; c- _- W - ~, .
* .I.
. .o, i aX +. ..2 a
,t,~.~ r y- . e_)(1
[ ;..- ~9 J_-. x 3;'. t ; 1
, , ~ . , .- -.
Jfi1 ner j j c: u ll :l -
- , wu+e . j
.. ~
-p s
. ~~c -- a;.j u. . ,: ,, l W-l i y- .1- -
=.= f h 1 s(-- ql r,,..
,,r i
,. r ,_.._t._. <.,~9,,,.. * ;/', k700 0-p ,s /s r .' .. ,
. * ,i .I
*~ , ,, , .
g
.s* ..
..p_.- l.t V, <,.,IL. yt4-E. g \
- g .. . . ;
..+_r_,..p._ c.
j.. , . ( 1 _.7.,..':.:':.l
, ~. ,' '
-t l
, ., = - --x, i .bs: _. e : .
f 4 ; ;;;,,. l<-- , = .,
- 1. J ~-
e_. , , _ _
-y ._.:s v- -'. -(6 89.01 Eig,t =_tln
-- - 7 5.e'..- mT , 51 - 3 t.
b _ v< . > , - @L , ,sp1 i.t. . 1
.- \. -
. . t.
- l-5;
.c .w.;;
. _.c
= - .. . . .
. 93 . ,..., .
- 6 6 9.0 ft.
- . . . i L,. _r, .
..g= ,.g .
. . :c , .
. T-r , . . . , .
F IGU R E 1
.:.. c. -
. . .. j . . .
.I . .
(7
%./ . -
CON TAI N M ENT l L IG HTING, FJX T UR E > EL E VATIO N S . ,
/~ w; O
..- l- B
.~.
a F.
-C '/ [~ %.' N G-A O
g$ ,
. L- C
/ L-
/
/ ,*
/ * ~
L- A
'.Tg'
/// - .'
t '. .o.s
' _ _ ./
, '.:i,. r.
l,' ' L- A ' L- IL-A :; L- A ; goo 6 p
-}- --
23 **' l ? ']
,_j ' R. , ; . ;. . . :- l.
- e. q . . .
e- , 4 - s y L- B i
/
/
'/
/
L- ' L- C
'sf,C
.R'B 9 ~
6 G-A N _ __ ._ _ _ t i b L-B SEQUOYAH NUCLEAR PLANT FINAL SAFETY ' ANALYSIS REPORT CONTAINMENT LIGHTING FIXTURES h EL. 689.0'
, FIGURE 6.2-142 s
,si: c
. .c. . .
-Q v . 0
.. ::s- ...~
,@L-C 2
- k o,
/h G-B .
- 's 03 %
/
I
.'\,y .
\
\
5 ,
;! "(. '
'. I gg O E
-'. . . .. . '.kb ht.c 5
bl-B
~
h G .- A Q ,
,i
\N s . ,, M -C Pw-g w
o . -s . . , ' . . . SEQUOYAH NUCLEAR PLANT FINAL SAFETY ANALYSIS REPORT CONTAINMENT LIGHTING FIXTURES pA-EL. 700.3
, FIGURE 6.2s143
... , :;.s . . :: . -
q v O A?a .~. g-- ,_ _, _G -C
, L-C c
g@ d' s,
, ,' G -CD \
. ,- N\ \
./ s L-C
,'G-A g, \ h
'd L-B '"i ' ' #' '{* '
i .
's 90 p v i .
O
- 4 G-Ah -
a, t
\ **) '.' 4 i ' 3?.'s ..,%
l ls' . N' '
,I O -A G - C s' ,/ ,
\ /
\
\ 's /
\ 's -
' L-C N 's , G-B '
& .3. _ _ . - '
'h .,
l
-fC w
i SEQUDYAH NUCLEAR PLANT FINAL SAFETY i j ANALYSIS REPORT
- i n
( ,1 i CONTAINMENT LIGHTING FIXTURES l EL 731.0'
; FIGURE 6.2-144 in
e 8 8
-- LS CONTROL PANEL (LS 2 or LS4) o O
~ '
~
G -A f- 'N -B L- B '
-c
~
I
' i
' u go o o ,
-t- -
k.J - L- A
./
l / G-As N y N
/
/ L-B
\
s s -
/
8 - L-B G-A-
; ,y .
CONTAINMENT LIGHTING FlXTURES EL 79 2. O' FIGURE 5
% r' c
1 0 ' O v .' > B j
/
/- -
1 G- / . 's \
\
\
s
\
s 90 O ; -
-t- }
I I e I l
/
/
/
/
L-B @fG -C S. ~ ~
/
l
'_ _- )
I
~-
' SE.
1 CONTAINM EN T. LIGHTING FIXTURES l EL. 818. 0' I i O v FIGURE 6 l i r y - . ,- w,
4 270 AIR R AIR FI N T
,.. . ~ l' /
..) ~f'
/ j oE)
E)6' 6 g p$4 WQ sy n y <g g a
*'e
. efd8 -I f
- : = = _-
P = REACTOR COM7 SG =~ STEAM GDIERATOR I3
O APPENDIX l' ! l PRELIMINARY TESTING TO IDENTIFY SUITABLE l C0!?!ERCIALLY AVAILABLE IGilITERS - TESTIt!G CONDUCTED AT TVA'S SI!!GLETON LABORATORIES SEQUOYAH UNIT 1 REPOPT OF THE SAFETY EVALUATION OF THE INTERIM DISTRIBUTIO!! 0F IGNITION SYSTEM l O TEMNESSEE VALLEY AUTHORITY SEPTEMBER 2, 1980 -
. - . - _ . - , - - - . - - _ . _ .,_,.__-,._____.-..--_,.___......-..-,.,_._,_._...._____._.._,.....-.--.m.. .
1.0 Int roduction j() TVA has a testing program which is being conducted at TVA's Singleton' Laboratory to obtain preliminary information about The purpose the performance of commercially available igniters. of these tests was to screen alternative igniters and to gain a degree of confidence that the igniters could ignite hydrogen. The tests were not run under normal laboratory test conditions since the objective was to identify which igniters, if any, were 55 ' most promising as subjects for more detailed testing and evaluation. Nontheless, TVA gained considerble information and assurance that commercially available igniters could ignite hydrogen. 2.0 Preliminary Screening A number of igniter types were evaluated, ranging from high O energy spark igniters to large diameter (1-1/2" I.D.) heater coils. Although the spark plug type igniter was considered an excellent candidate for this application, it was rejected prior to preliminary testing due to potential problems with instrumentation. electromagnetic interference (EMI) with-critica) : TVA's Electrical Engineering Branch is researching the problems l associated with EMI generators, and spark type igniters may be i considered at a later date for use in Sequoyah unit 2 or Watts Bar. Two other potential candidates, both coil heaters, were rejected l after the first one, a large diameter (1-1/2" I.D.) coil, couJd l not reach sufficient surface temperature, and the second one l There fore ,
'- failed at the connector in less than five minutes.
l
.. - - ~ . .
4- g i,ye ,'s -Y9h yO % AMR & 'Pg& *M, ? @ =^ 'A- $ m,WA 4\. a ) _j
, testing was restricted to diesel engine glow plugs, since they were known to be capable of achieving the 1500 F minimum surface i temperature' desired by TVA and because of their rugged I design, i i TVA determined that at 12 volts ac, acceptable surface temperatures could be achieved but that considering line losses, variances in system voltages, possible plug cooling due to high humidity, and other effects, TVA would need to operate the plugs at 13 volts ao 1 1 volt.
- Since the possibility existed that TVA could overstress the plugs by overvoltage, TVA consulted glow plug manufacturers and
- identified two types of failure modes which could be expected.
The first type of failure caused by overstressing would be the () failure of the heater wire within the glow plug sheath. This . type of failure due to the breaking of the circuit would l outwardly cause the plug to discontinue glowing. The second 1 l type of failure caused by overst' ssing would involve offgassing
~
of the glow plug tip. Unlike the first type of failure after offgassing, the' glow plug may continue to glow; however, the surface temperature would drop significantly. 3.0 Description of Glow Plugs Glow plugs manufactured by three different companies have been tested to'date. They include: General. Motors AC Division, Model 7G, 12 Volt l ? BOSCH 10.5 Volt 4
, k.
y e-e--- -
--m- * - , , . . ,,e- , . x ..----...,,g-- ,.--~m , - - - -
- - -- .. - - - .- ~ .
ISUSI 10.5 Volt
~
4.0 Endurance and. Temperature Tests
-(
Diesel glow plugs are not usually intended for continuous service in an air environment. Therefore, TVA undertook tests to determine that the plugs:
- could reach and maintain the desired temperature;
- could withstand the effects of overvoltage and temperature; and i
- that the plugs could operate for extended periods of time at high temperatures.
1 h.1 -Testing Equipment [) For these tests, the plugs were bench mounted and operated for various periods of time. The power source was 120 V ac wall socket which was reduced by a variable transformer (Variac, model no. Staco, Inc., ty.pe 3 P/N 1010L) to the I appropriate voltage levels for each test. The voltage 'i levels both on the primary and secondary side and at the 4 plug were measured by a digital voltmeter.(Fluke model' number 8024A), and the current leveles were measured by i an amp meter (Triplett model number 10 type 2).
~
The surface [ . temperature of each of the glow plugs was measured by either a thermocouple (type S) in contact with the surface of the plug and connected to a potentiometer (Leeds and Northrop (~j ' model number 8690-2) <n by an optical pyrometer (Pyro model .
%)
2
-%4 +.W - W .idth- *M*WAEppg y ...,,,4.--_ ',. e g , y ,,pgy p --- -
' ^
4.Z,_ 4_ ' #, >N N *
~
% 7 4[,
- -- - , , , . . - _ . _ . . . -, , , . , . , . , , , , _ . _ . . . _ , - . , . . , _ . - _ _ _ _ . . . ~ . _ , , .-. - - , , _
number 85). A total of 12 plugs have been tested to date. 4.2 Surface Temperature A GMAC model 7G plug was operated at 12, 14, and 16 volts ac. Surface temperatures as measured by the thermocouple were 1480, 1550, and 1650 F, respectively. Since the thermocouple would be expected to increase local heat loss and hence reduce the measured local surface temperature of the thin-walled plug sheath, these valves were probably somewhat lower than actual surface temperatures. This conclusion was supported by later readings with the pyrometer while testing another GMAC model 70 at 14 volts ac and getting 1720 1 15 F. A Bosch plug has been tested at 13 volts ac. It produced (} a surface temperature of 1700 F as measured by an optical pyrometer. Based on these results, TVA concluded that the diesel glow plugs could produce the desired surface temperatures.
- 4.3 Voltage Tests -
Voltage tests have been completed on only the GMAC model 7G plugs. Based on tests on 5 GMAC 7G plugs, reliable operation.at 14 volts was confirmed but two other 7G plugs failed at 16 volts ao after a few minutes. Inconclusive testing on Bosch plugs resulted in two failures when: operated at 14 volts ac; however, one Bosch plug
-operated satisfactorily at 13 volts ac. In addition, one
.(} Isusi plug was tested at 14 volts ao but lasted for only y h, -hm44 A '" [ . i6Y b i-_ ' e M4 FVA a4, g L
30 minutes. 4.4 Extended Operation Endurance tests have been performed on only two plugs for extended periods of time. A GMAC model 7G plug was operated continuously for 148 hours without failure and was later used in the hydrogen burning tests. A Bosch 10.5 volt plug was operated at 13 volts for 90 hours, then cooled down for two hours and turned back on. It has been running continuously after being reenergized since August 20, 1980, at 10 a.m. 5.0 Hydrogen Testing-One igniter (AC 7G) was installed in a "PARR" (229HC6-T316-031579-5142) pressure vessel in order to determine feasibility of igniting hydrogen in a sealed container. The vessel lid has a silicone rubber sealed gas injection sampliag port. Hydrogen O V concentrations in the-vapor phase were determined before and after ignition intervals. An ignition interval is the time current flows through the igniter circuit. The hydrogen was measured by a Perkin-Elmer gas chromatograph equipped with 3920 thermal conductivity dector and an M-2 integrator. The chromato-graph was standardized with hydrogen and air mixtures prepared
-from research grade hydrogen and laboratory air.
Temperature measurements were made with a mercury r.nd glass (484635, ASTM 9C) thermometer. Temperatures reported are ambient for tests.1 through 3 Prior to tests 4 through 10, 100 grams of water were added to the vessel. The vessel was heated by a temperature adjustable hot plate to saturation temperature i e 7,:3. m ;; w.e .y - 34.~.,
, .5 - x - - ; my-~ -~--- - = - - = ~ - ~ ~ -
. aa..
. .. .. m -
. ... ~ .. . . . . . . w u ,, ,,-
of the water and maintained throughout the test. The reported temperature is the water temperture after conpletion of the test. Results of the 10 ignition tests are given in table 1. 6.0 Future Tests at Fenwal Laboratories TVA and Westinghouse have contracted with Fenwal Laboratories to perform hydrogen burn testing on the AC igniter and its mounting enclosure in an enclosed vessel. Appendix N contains the proposed Test Plan for the testing. The final test plan is being prepared by Fenwal and should be available in the near future. These tests are designed to prove the effectiveness of this igniter assembly to burn a volumetric quantity of hydrogen in environmental conditions which approximate postulated accident conditions inside containment. 7.0 conclusions and Summary . O The purpose of these tests at Singleton was to select a commercially available igniter that was capable of igniting hydrogen. From the results obtained, the GMAC model 7G glow plug produces more than adequate temperatures at a range of voltages that can be provided inside the Sequoyah containment. In addition, the plug seems capable of extended operation at high temperatures and has been shown in small tests to be able to ignite 12 percent and lower volumetric quantities of hydrogen. Although it has not been tested as thoroughly, the Bosch plug appears like it may also be an optional igniter. 1
~1 O
TABLE 1 fs HYROGEN IGNITION TESTS b Initial Final Ignition Test Vessel Temp. Hyd. Conc. Hyd. Conc. Intervals No. Contents ( F) (% Hyd.) (% Hyd.) (Min.) 1 Hyd., Air 90 12.5 0.1 5 2 80 7.0 0.1 5 3 80 3.5 0.1 5 4 Hyd. Air, Water 120 12.0 0.1 3 5 180 14.0 0.5 3 6 180 4.0 2.5 1 7 180 2.5 1.5 1 8 180 1.5 1.3 1 9 180 11.0 5.0 1
~
O to 180 so 2o 13 Vessel Volume 1.1 dm3 (0.039 ft3 )
~
Operating Voltage 12V do
~
w j
} 4 l APPEilDIX N TESTI!!G OF THE HYDROGE!! IGNITER ASSE!iBLY UMDER VARIOUS E!1VIR0!!MEllTAL C0'IDITIO?IS SEQUOYAH UNIT 1 REPORT OF THE SAFETY EVALUATION OF THE II;TERI!1 DISTRIBUTION OF IGt!ITION SYSTEM , . O TEfNESSEE VALLEY AUTHORITY SEPTEMBER 2,1980 - 1 I i [ l .l ,
() APPENDIX N TESTING OF THE HYDROGEN IGNITER ASSEMBLY UNDER VARIOUS ENVIRONMENTAL CONDITIONS TVA and Westinghouse have contracted with Fenwal Laboratories of Ashland, Massachusetts, to perform hydrogen burn tests with the igniter assembly in an enclosed vessel. The purpose of the tests is to demonstrate that the igniter will initiate a volumetric burn of hydrogen for various environmental conditions of pressure, temperature, and steam. The approved package of 13 tests is expected to demonstrate the ability of the igniter to burn hydrogen at 8 and 12 volume percent, concurrent with pressures of 6 and 12 lb/in gauge and 100 percent humiditv. The package also includes tests that add N- the effect of fan ca flow speeds across the igniter. The draf t Hydrogen Igniter Test Plan is included as attachment 1 to this appendix. A final test plan is being prepared by Fenwal and will be available prior to the start of testing. rx ) ( 1 e
?
() ATTACHMENT 1 SUMMJLRY SEQUOYAH PLANT HYDROGEN IGNITER TEST PLAN
- 1. Introduction The following describes tests to be conducted on a type of hydrogen igniter to be installed in the Sequoyah Nuclear Plant.
The igniter consists of a " glow plug" as used in diesel engines, the surface of which exceeds 1500 F and serves as a hot surface to initiate hydrogen burning, and a power transformer and an enclosure for the unit. The function of the igniters in the (~'3 (_) nuclear power plant containment is to burn hydrogen, in accidents where it could be released, when it reaches a burnable , concentration thereby precluding its buildup to high concentration levels. The tests will be conducted by Fenval,
~
Incorporated, at their facilities in Ashland, Massachusetts. The unit, consisting of glow plug and enclosed transformer, will be placed in a test vessel and subjected to a range of environmental conditions (including hydrogen concentration, temperature, pressure, and steam), and its hydrogen ignition performance monitored. 1.1 Purpose of Tests The primary purpose of the tests is to demonstrate that
/~N
(_) the igniter will initiate a volumetric burn of the hydrogen Y
for the specified environmental conditions (pressure, temperature, water vapor). A secondary objective of the tests is to narrow down the hydrogen concentration range for rhich a volumetric burn of hydrogen will be initiated. 1.2 Acceptance Criteria For the initial set of tests, the following acceptance criteria will be used:
- 1. Data generated are internally consistent (i.e.,
@l ignition at 8% consistently produces low pressure rise). _
- 2. Data gathered confirm theoretical predictions.
3 Igniters reliably ignite mixtures at high (12%) concentration and provide relatively complete combustion.
- 2. Description of Igniter _
The igniter is a General Motors AC Division Model TG glow plug (thermal resistive heating element) requiring 14V ac supply at a maximum of 8-1/2 amps. The surface temperature of the plug . as measured by an optical pyrometer should be a minimum of 1500 F. TVA has measured 1720 F surface temperature on one of the The igniter is powered by 120V glow plugs at their facilities. ac stepped down to 14V ac. The power transformer is a Dongan Electric, Incorporated, Model 52-20-187 specially wound transformer having the following characteristics: 120V RMS AC on primary side 14V RMS AC on secondary side 200VA Min. Ib hh m v.p's,,.
Class H (High temperature insulation) Open style with 18" flexible leads Certified capability that transformer will operate at 220 C. The igniter and transformer are mounted as a unit with the glow plug extending from the side. The unit is encased in a 1/8-inch steel plate box type casing and sealed with a rubber seal for water tightness. 3 Description of Test Facility The tests will be conducted by Fenwal, Incorporated, at their facilities in Ashland, Massachusetts. 31 Test Vessel The igniter un.it will be tested in a spherical vessel in excess of six feet in diameter. The internal volume of the test vessel is 1000 gallons (134 ft 3). The vessel is constructed of carbon steel (exterior) and is lined with stainless steel. The vessel is designed for a working 2 pressure of 500 lb/in . The vessel is equipped with five 18-inch diameter access ports (fou'r on circumference, 90 apart, and fifth at the top), one of which is drilled to attach to a manifold with valves and connecting lines to air, steam, and hydrogen makeup sources. The vessel is heated externally via electrical heaters. The vessel will be equipped internally with a fan to promote mixing and also to create a draft at the igniter heating surface during testing when desired. n U. 3.2 Instrumentation and Measurements il
- The vessel is instrumented with two pressure transducers
,_s to ronitor the pressure including the pressure transient
) during the hydrogen burn. The output is carrier amplified and feeds to an oscillograph device. Thermocouples are provided which will monitor vessel atmosphere temperature prior to and after a burn. In addition, a thermocouple will be used to measure the temperature of its heated surface. Gas mixtures will be formed using pressure instrumentation and a partial pressure method in which a given gas is added until the appropriate partial pressure is indicated. Sampling capability exists via a 1-inch by 1-foot lecture bottle. Samples will be taken before and after each burn and then analyzed by a gas chromatograph.
- 4. Test Plan 4.1 Identification of Tests The unit consisting of the glow plug and encased transformer will be positioned in the test vessel (via 18-inch port) with the glow plug heating surface located near the center of the test vessel. Various mixtures of H2 ' 8D* "' and air will be adjusted with pressure and temperature as specified and then the igniter turned on. The pressure transient will be recorded and the mixture analyzed for H and 0 2 2 content prior to and after the burn. The test matrix for the first series of 12 tests is shown in Table 1. Initial total pressures of 6, and 12 lb/in g will be covered at hydrogen concentrations of 8 and 12-volute percent.
Initial temperature wil vary from 180 F (dry case) to 350 F fl '(superheated steam) with most of the tests being mj 1 L' )
, . x., ,. , m' _ um.w.e u.
i conducted at saturation temperature corresponding to the pressure to'be tested. In addition, a fan will be located O in the test vessel to provide drafts of 5 and 10 FPS in , the vicinity of the glow plug to simulate turbulence which may be developed in the vicinity of the igniters. 1 Further testing will be developed based on the outcome of test i series #1, and may include addition of an instrumented transmitter, steel, or concrete surfaces with thermocouples In attached to measure temperature response on hydrogen burn.
! addition, means to simulate spray droplet entrainment in the atmosphere are under investigation.
4.2 Test Procedure The basic procedure is to adjust mixture concentrat. ion temperature and pressure, take a sample, then energize the glow j i plug and record the pressure and temperature transient. After. [ the pressure and temperature have been recorded, the mixture j will be sampled again, then, to measure the effectiveness of l a spark igniter versus the glow plug, a 105 joule spark will l i be set off and the mixture sampled once again. 2. In one of the tests with a steam environment, the glow plug will be energized after the steam, pressure, and temperature environment conditions are reached, but before hydrogen is added, and allowed to stand for two hours. Then the glow plug will be deenergized, i i hydrogen adjusted, and then the glow plug energized. The purpose 1 of this is to allow for preburn exposure to the environment. 4.3- Test Schedule l The test schedule is tentatively planned as follows: IJ'
...._.n_ __.-.7 - , , ;q q - . . ,
- :y n
! Facility Preparation 8/18 through 8/29 t O rest sectee ne- , 9'8 terecsw 9',2 Subsequent Tests 9/15 through 10/6 Test Evaluation 10/6 through 10/10 , l l l t 1 O 1 l \ O M 1
--_-,.-_._..._-_.,,,,,y-7 ,y,-,,,,+,,,,mgm,pw_4.,.-% -wg,,,g.,me,y,,cm,m,p..., _,,,w, e--,y-mewmwew-ep
I i l l l I O l l
- APPE!! DIX 0 l
HYDROGEN COMBUSTIO!! (TO BE SUPPLIED LATER) I i l I i i 4 l SEQUOYAH U'!IT 1 1 REPORT OF THE SAFETY EVALUATION OF THE INTERIM DISTRIBUTION OF IGNITION SYSTEM l 1 ! O i 1 TENNESSEE VALLEY AUTHORITY i l ! SEPTE!GER 2, 1980 i 1 1 l 5 I 1 I i i ?O i i
l l O I APPE!! DIX P SE!JSITIVITY ANALYSIS SEQUOYAH UNIT 1 REPORT OF THE SAFETY EVALUATION OF THE INTERIM DISTRIBUTIO!! 0F IGNITIO!! SYSTEM i O TEM::ESSEE VALLEY AUTHORITY SEPTEMBER 2, 1980 1
APPENDIX P SENSITIVITY ANALYSIS O - A sensitivity analysis has been performed by Westinghouse /0PS to evaluate the effects of ignition criteria and containment safety system performance on containment response. Calculations were performed using the Westinghouse /0PS CLASIX code. The sensitivity analysis is discussed in Appendix U and CLASIX is discussed in Appendix T.
/
O 9 I l l 1 1 o
^
l 1 - i 1
-7. , ,- , ,_! l-#: - - - - - . - - - - - - : <mme??- g-; - h .-' .7 ; T > >~.,--r
-l
' - - ' ~ ~ ^' ~ : \
__ , _,, , -- - . . _ - . - , - -. -- -. l
O APPE!! DIX 0 A!!ALYSIS OF EFFECTS OF TMI-II EVEf;TS l l l l l SEQUOYAH U!!IT 1 REPORT OF THE SAFETY EVALUATIO!! 0F THE INTERIM DISTRIBUTIO!-I 0F IGNITIO'1 SYSTEM l l l N l TE!!ESSEE VALLEY AUTHORITY SEPTEMBER 2, 1980 - l EO
APPENDIX Q ANALYSIS OF EFFECTS OF TMI-2 EVENTS in The accident at the Three Mile Island nuclear facility resulted i d . core damage and the production of significant quantities of hy rogen as described ~in Volume I, section 4.3. Estimates of the hydrogen burned in the containment have ranged from 452 pounds to as much as A reasonable estimation of the burned hydrogen appears 1,160 pounds. , This would represent to be somewhere between 500 and 600 pounds. between a 30- and 40-percent metal water reaction at Sequoyah from figure A 4 of Volume I and would result in a concentration of 9-12% The by volume (neglecting the presence of steam for conservatism). containment pressure response to the hydrogen burn at TMI was a 2 Wessure spike of approximately 28 lbs/in g, as described Volume I. However, Figure Q-1 repeats the pressure response for convenience. it has been noted (reference Q-1) by investigators that the temperature rise TMI was only on the order of 50 F as measured at Figure Q-2 shows the recorded several containment locations. I Conservative calculations of the temperature and temperatures. pressure response of the Sequoyah containment design to a similar hydrogen deflagration have been compared by-others in recent weeks TVA does not believe at this to the actual data obtained at TMI. point that such a comparison is valid since the analytical models are not yet sophisticated enough to perform a truly best estimate It is expected burn calculation (see Section VIII.E and Appendix F). that application ,of the same conservative calculation methodology to the TMI system would yield results far in excess of those observed 1 i~ _ _ .
._g,+, -
9equagall Egbragen 9tuhg C d
- : : ' p
...... ... . $,M O
C l
.o 9
g
............... o.
.O
. . m W
- : : : - :: : s epa : :. : :.
O c-H .
^ :. :. .
Q W O. t .
............ .............;. ........ gy Q .
g b-1 o H
- . W p.
- : :. : :. g,
. w
.o P 4 . . . . . .
. . . . . ............... . . . . . . . . . . . .. . . . . .......... . . . . . . .............. ........................ d H . .
~ .
I
- . :. 1 l.
&e .
. i g .
- . l l
p
. g
.i .
F- ( I. : :. !. 4 k- ( aM w
- . :. : l r2'i m .
. o l
* : : ............. . . . . . . . . . :..... ... .... .Lg 3 . . . . . .. . . . . . . . . . . . ..: . . . . . . . ............ :. . . . .. . . .: . . . . . . . :. .
+ n :. :. . .
3 : i. i. i. M fm A 4 a
- i. .
i. i. M . t . H. m . ..
>M ........ . , .....: ........... : ........... :........... :........... :........... :......... :
b ep m : e
- : - : 6
- : i n . i , 6 i 0'T- j l
V 01 09 0'8 O'63 OT3 0 C3 0'03 0'a OTT OIT I (SI sd) EHnSSSHd a. 1
-_--=.m
. . -. ; y ; .r<~-.-.. e ,.... n ,w. . , _ _ ' -
D bN2
* * * +
id[[.*
- f.9 A ,,( -w i 1
- - - - . - _ , _ _ _ _ w - . - . .
Se O . o . o . s- _ 100 < 5 Supply air 180 - G Supply air
, 7 Ambient steam and Icedwater 170 -
penetration 8 Ambient sump area 1G0 - w 9 Ambient let down cooler area 150 -
- 13 i C . . -
4 L 140 _
' 12 l $ .,
7j E 120 - i y q ' h N
^
110 - 0 f l 100 - 6 10 Ambient drain tank area i -
' 5 11 Ambient clovation 353-1 90 -
-12 Ambient elevation 353-2 13 Ambient c!cvation 330-1 80- 14 Ambient elevation 330 2 5 70- ! , , ,
1400 1500 ! 1300 *
'. , Time ,
Reactor building temperaturos durin0 hydrogen 10nition. l 1 FIGURE Q-2 g . {'. l [
at the Three Mile Island facility, particularly in the temperature
)omparison. Such a comparison should provide insight into the importance of each physical phenomena neglected in present analytical models and will provide some indication of the response that could be expected from the ice condenser pressure suppression containment to a hydrogen deflagration. Therefore, TVA is working toward this type of comparison.
! Plant differences also exist between TMI and Sequoyah. In particular, j the nuclear steam supply system at TMI was manufactured by the Babcock l & Wilcox Company and consists of a two-loop design with once-through i L steam generators. Sequoyah employs a four-loop U-tube steam generator of Westinghouse design. 4 (:)Differences also exist in the containment design beyond the obvious 4 pressure suppression versus dry containment concept design. A study i is in progress to examine the TMI events in detail for determining l the approximate response of systems that could reasonably be expected at Sequoyah if similar events occurred. Low power tests have also _been run at Sequoyah to demonstrate the ability of the system to achieve and sustain natural circulation hea_t removal at power levels j typical of decay heat following a small break LOCA. Positive results 1 were obtained in these in-plant experiments. Plant operations ] j personnel has also been trained to recognize events such as TMI and to take appropriate action to maintain adequate core cooling at all times. O. To perform the above analytical tasks, TVA is also pursuing the t
procurement of outside training en the use of appropriate computer
,_. codes for analysis of accidents beyond the design basis. The MARCH
(/
)
computer code written by Battelle Columbus Laboratories appears to be a good candidate for this type of training and accident simulation. In anticipation of obtaining MARCH or a similar tool, work has been initiated on the generation of appropriate plant parameters and accident sequences for study. Results from these studies will be used to improve the hydrogen mitigation system as appropriate. References Q-1 Analysis of Three Mile Island - Unit 2 Accident, NSAC-1 Appendix HYD, July 1979 (-~ v e V s
4 ,
! t i
e 4 i I. j APPENDIX R 1 I OPERATING PROCEDURES i i i ' 1 } l l SEQUOYAH UNIT 1 i REPORT OF THE SAFETY EVALUATION OF THE I!!TERIM , DISTRIBUTI0tl 0F IGtJITION SYSTEM ! i I 1 ! U I 4 I t TE!WESSEE VALLEY AUTIIORITY 1 ! ! SEPTEMBER 2, 1980 - '1 4 4 I 4 , 1 l> ) 0 9
1 APPENDIX R OPERATING P30CEDURES 4 O To reficct the installation of the Interim Distributed Ignition System 1 (IDIS) at Sequoyah unit 1, the emergency operating procedures had to be revised. The existing emergency operating instructions (E0I's) were checked for procedures that would; (1) Conflict with the operation of the IDIS. 4 (2) Prevent the IDIS from accomplishing its function, and (3) Lead to a more severe accident scenerio than those previously l evaluated because the procedures did not include instructions I l for energizing and deenergizing the IDIS. After reviewing all of the Sequoyah emergency operating instructions, () TVA did not. find any procedures that conflicted with the expected ! operation of the IDIS or lead to a more severe accident. J Emergency Operating Instruction 1A (E0I-1A), " Loss of Reactor
~
Coolant," however, would prevent the IDIS from accomplishing its function. E0I-1A instructs the operator to " place the (postaccident) j H purge system in service . . . if the containment atmosphere reaches 2
~
i 35 by volume . .. The postaccident purge system is designed as i i a backup to the redundant hydrogen recombiner system. The system f consists of a single penetration in the primary containment wall equipped with two normally closed, remote manually operated isolation l ' l valves, one on either side ~of the containment wall, and one pneumatically operated annulus purge exhaust valve located within _() the annulus. With these valves open, a flow path is established I
- 6
between the primary containment and annulus which will permit purging (N, of the containment for hydrogen control subsequent to a LOCA. V The operation of this system, however, is based on an assumption of a maximum 5-percent metal-water reaction which will not produce the much larger volumetric quantities of hydrogen that a degraded core accident would. Therefore, allowing this system to purge would lead to unnecessary release of radioactivity to outside the containment instead of making full use of the IDIS to burn off the hydrogen and keep as much radioactivity inside the containment as possible. The E0I-1A instructions, therefore, had to be modified to eliminate operation of the H2 purge system while the interim ignition system is installed. () Procedure'odification M In addition to revising E0I-1A to eliminate operation of the H 2 purge system. TVA also had to modify E0I-0, "Immediate Actions and Diagnostics," to ensure initiation of the IDIS before it would be needed. Table 1, attached, lists the procedures that required modification in order to reflect the addition of the interim distributed ignition system. Special Note : None of the modifications to procedures listed in Table 1 are to be incorporated into the procedures or implemented prior to NRC approval of the system and these proposed E0I changes. l l i ex ' i 1
) l
APPENDIX R OPERATING PROCEDURES /~') TABLE 1 \d PROPOSED EMERGENCY OPERATING INSTRUCTION CHANGES E0I No. Change Text E0I-0 Add to Section II.B 11. Energize power supply to U-1 controlled hydrogen ignition system by closing breakers 1Q , 11 , and jlj[ in Standby Lighting Cabinet LS-4 (near CCS surge tank). E0I-1A Add to Section II.G 2. Ensure controlled hydrogen ignition system is it. aervice per E0I-0, section II.B.11. E0I-1A Delete from Section Place H2 purge system in service as II.00.5 follows:
- a. If the containment atmosphere reaches 3% by volume, place the H purge system in service per S8I-831.
f) h) f
% j' 8
O APPE!JDIX S TVA DEGRADED CORE I!!VESTIGATIO!! PROGRAM d SEQUOYAH U: LIT 1 r REPORT OF THE SAFETY EVALUATIO!! 0F THE INTERIM DISTRIBUTION OF IG!!ITION SYSTEM l l 1 O
)
l TEt!NESoEE VALLEY AUTHORITY SEPTEMBER 2, 1980 ~ I' l l O i
--.--eer,. ...-.yy.,,e, m a . e,. 4m. , w .ag-e-y,,m w_ ,_. _ _ . , -y yg wW- Wmgteg , VTN7*
APPENDIX S TVA DEGRADED CORE INVESTIGATION PROGRAM O In November 1979, in the course of the TVA Nuclear Program Review I l of the Sequoyah Nuclear Plant and the Report of the President's l Commission on the Accident at Three Mile Island, the TVA Task Force on Nuclear Safety recommended that "TVA . . . (4) Study ways to contain larger amounts of hydrogen and to backfit feasible features I into the Sequoyah design." . 1 I To fulfill that commitment, the Nuclear Engineering Branch of the Division of Engineering Design was charged with coordinating a study l and making a report on the effects of postulated degraded core ' accidents on the Sequoyah Nuclear Plant and on design features to O mitigate those effects. The study and draft report were completed q on April 15, 1980, and included input from Westinghouse, Burns and Roe, Sargent & Lundy, Stone & Webster, and TVA's Office of Power.
~
The study began with a summary of the original design bases of the i Sequoyah Nuclear Plant, including accident sequences, analytical tools used to calculate phenomena associated with the accidents, and W features and equipment required to mitigate the accidents. Then, the Three Mile Island accident significance was assessed by investigating the accident phenomena and their ramifications on the current design bases for Sequoyah. An analysis was perforced to compare the current risk posed by Sequoyah with Surry, a WASH-1400 reference plant. Accident sequences beyond the design bases were 4
, n- - ,.- , ,
-- _ - _,, ,- , , . . _ _ ,_. __.r,-_- ,_ _,,m_#.,_.-_y ._,_.,,x,-,, w-,.-.,--, ,,
i
- l identified and investigated. Since the calculational tools to hnalyce l
these accidents were of uncertain accuracy and limited availability, hand calculations were performed to scope the magnitudes of accident 4 parameters. Information from the Batelle Columbus computer code MAPCH j was then obtained to estimate the magnitudes and rates of the j parameters. The present margins in the containment safety systems for degraded core accidents were evaluated. Later, a number of f concepts for study were proposed to mitigate accident phenomena by preventing or minimizing hydrogen combustion (e.g., controlled ignition, Halon inerting) or by increasing the containment capacity , for overpressure events (e.g., filtered venting). Another analysis was performed to evaluate the risk poced by the Sequoyah Nuclear Plant, assuming the mitigating features were installed. Conceptual j designs by TVA and others of the mitigations were also evaluated on , the bases of physical effectiveness, technical feasibility, additional risks, reliability, cost, and schedule. The report of the study recommended that further investigation of both degraded core accidents and their mitigations be undertaken before commitments were made to
; specific mitigation devices. However, controlled burning of hydrogen and postaccident inerting with Halon were suggested as the most i promising mitigations for further study. Nitrogen inerting was rejected after a thorough study, mainly because of the extreme hazards it would pose to personnel during the almost daily containment entries i required for the inspection and maintenance of ice condenser and other vital equipment. Filtered venting was rejected because of the large radiation dose to the public that would result from venting the containment during a severe accident.
t O 10 L
To act on the recommendations of this report and to continue to fulfill TVA's Nuclear Program Review commitment, a Degraded Core Task (j~T s Force was established in June 1980. Organized in the Nuclear Engineering Branch, it is composed of seven nuclear, mechanical, chemical, and electrical engineers whose two-year mission is to further investigate degraded core accident sequences, their analyses, and their mitigations. The primary efforts in mitigation are to support the installation of igniters at Sequoyah to allow controlled burning of hydrogen and to evaluate the use of Halon to suppress hydrogen combustion. Other mitigation concepts such as catalytic recombiners are also being investigated. The primary efforts in accident analysis are to develop MARCH-type code packages capable of realistic calculations of accident parameters. Since overconservative or arbitrary computer models may be () counterproductive to determining the need for mitigation or to the selection of optimum mitigations, code improvements will be made by the incorporation of realistic assumptions or physical models bench-marked to present and planned experiments. Special efforts are also
~
being made to analyze the containment structural integrity for uniform and localized pressure effects and to analyze and qualify necessary equipment incida containment for accident environmental effects. A comprehensive effort related to all of the above work is the technical support of utility input during future NRC rulemaking on degraded core licensing issues. I l 1 1 HYD.2 I
<8 l t> .
I
//
l i l t l l ! APPE!! DIX T i ! CLASIX PROGRAM DESCRIPTIO'l i \ I i 1 i i
- I 1
+ l i } SEQUOYAH UNIT 1 1 l REPORT OF THE SAFETY EVALUATION OF THE I!!TERIM l DISTRIBUTION OF IGNITION SYSTE!! l l9 I l 1 l TENNESSEE VALLEY AUTHORITY t 1 4 - SEPTEMBER 2, 1980 i l I _ 1 i t j
INTRODUCTI0ft As a result of the incident at Three Mile Island, Offshore Power Systems began the development of a computer program to investigate the response of an ice-condenser containment to a degraded core condition and the subsequent ignition of the hydrogen released to the containment. Because of the compartmentalized nature of the ice condenser containment, an analytical model representing a number of volumes would be required. None of the programs available at the time represented anything other than steam and air. To adequately track the migration of hydrogen and eventually convert hydrogen and oxygen to steam and heat, each constituent must be individually considered. It was concluded that none of the existing programs could be readily modified and a new program was required. The new program was designated CLASIX to C) reflect the analysis of Class IX accidents. ANALYTICAL MODEL The analytical model selected for the CLASIX program is shown schematically in Figure 1. The containment can be represented by up to six volumes interconnected with appropriate flow paths. A seventh volume is included for repri. :ation of a blowdown volume for venting the containment. Volume 1 in the figure characterizes the lower compartment in the volume of the containment below the divider deck and outside the ice condenser doors but exclusive of any further subcompartmentalization such as Volumes 5, 6 and 7. Volume 5 characterizes the equipment and instrumentation volume off the lower compartment. Volume 6 can represent the pressurizer shed and Volume 7 the steam generator dog house. Volume 2 is the ice condenser volume in that an ice heat sink may be specified. q w/
. = _ - - _ - - - . . _- .
I () Volume 3 represents the upper compartment and generally has the largest net free volume. ! The flow paths from Volume 1 to 2 and Volume 2 to 3 have representation of the lower inlet doors and intermediate doors respectively. The flow , path from Volume 3 to 4 permits specification of a rupture disc, submergence of the discharge under water and specification of a length of water which must be expelled before gas discharge may begin. Also shown in Figure 1 is a recirculation fan. The fan may be turned on and off at optional times during the transient, The fan takes suction from the upper compartment and discharges into Volumes 1, 5, 6 and 7 in optional proportions. The most significant capabilities of the CLASIX program are listed in [} Figure 2. The vent, ice condenser, recirculation fan and the ice condenser doors were discussed above. Each of the non-condensible gasses j and the steam inventories in each compartment are calculated individually. The steam properties are from the computerized ASME steam tables and the steam may be saturated or superheated. Sprays are optional in all volumes 4 except the ice condenser. The sprays are individually controlled by tables specifying time, flow. rate, temperature and heat transfer coefficients. The spray drop size and fall time may be varied. . Hydrogen, nitrogen and heat addition tables are available in Volumes 1, 6 and 7. There are break flowtables available in the same volumes and the break flowmay be subcooled,in the saturated region or superheated. Burning control of the hydrogen is available in all compartments, n
U The burn control parameters are shown in Figure 3. When thc. volume percent of hydrogen exceeds the value for ignition, burning is assumed to occur provided that the volume percent of oxygen exceeds the value for ignition. Once the hydrogen is ignited, the flame will proceed at some velocity thus establishing a burn time for the hydrogen in the volume and a propagation delay time or elapsed time for the flame to reach an adjoining volume. Since the ignition source may require a relatively high hydrogen concentration, the flame may propagate to adjoining compartments at much lower concentrations. Depending upon the hydrogen concentration at ignition, only a fraction of the hydrogen may burn. Finally,because of depletion of oxygen below a minimum concentration, (3 v the flame may be extinguished, PROGRAM DESCRIPTION A flow diagram of the CLASIX program is shown in Figure 4. The program has a restart capability. During any analysis with CLASIX, the instanta-neous conditions cui be written, periodically, to a restart file. By reading this file in a subsequent run, the transient can be picked up at the time of the restart file and the transient continued with altered input. The first decision in the program is whet;er this is a restart run or the start of a new transient. Whether or not it is a restart run, a complete input edit is written. Once this edit is complete, the finite difference integration cycle is initiated. q s_/ O The output options include a short output which prints the total
- pressures and the temperatures of all compartments on a single line.
A long form output is a detailed printout of containment conditions and consists of a full page of information. A restart file may also be written and all three forms may be intermingled. After each
, printout a test for completion of the transient is made.
Based on the flow path parameters and the differential pressures, a volumetric flow rate is calculated. Then, based on the source volume conditions, the individual constituent mass flow rates and energy flow rates are determined. The spray flow rate, temperature and film coefficient of heat transfer are linearly interpolated from input tables. If the temperature of O the steam is above the saturation temperature corresponding to total pressure of the volume, the spray is assumed to flash immediately to steam and water based on the saturation enthalpies. -The droplet size, 1 mass rate of flashing and the energy addition rate to the gas phase are thus determined. Knowing the droplet size, temperature, heat transfer coefficient and the residence or fall time of the spray, heat transfer rates to the gas can then be calculated. If the initial spray temperature is below the saturation temperature, only heat transfer rates need to be calculated. During those periods of time when the gas temperature is above the saturation temperature, heat transfer will result in vaporization at the surface of the droplet so that there is a corresponding mass and energy addition rate to the 9 V
+
h gas phase of the volume. The heat addition rate to the volume is linearly interpolated from i- the input tables. Ilitrogen and hydrogen addition rates and their corresponding temperatures are also interpolated from the input tables. Using the specific heat at constant pressure, the energy . I addition rates are calculated. The break flow rate and energy flow rate. is linearly interpolated from the input tables. If the
- j. breakflow enthalpy is above the saturation enthalpy corresponding j to the total pressure of the break volume, the rates are added directly to the gas phase. If the break flow enthalpy is below the
{ saturation enthalpy, the break' flow is assumed to flash instantly ) l. to water and vapor at the saturation eathalpies giving corresponding flow and energy rates. i The flow entering the ice condenser flows over the ice, transferring heat to the ice, condenses steam and melts some of the ice. The CLASIX model for heat transfer is the same as that in LOTIC, but the correlations are based on vapor content in air. Tiierefore an air equivalent of the nitrogen, hydrogen, oxygen mixture of non-condensiblesmust be approximated, t i Based on the constituents in the volume and the input burn control parameters j
- discussed above, .it is determined whether or not ignition occurs.
t If. ignition occurs, clocks are intialized to calculate propagation delay time to adjacent compartments, burn rates and amounts of hydrogen to be burned are also calculated. If the propagation time from some adjacent compartment has expired, conditions are checked to see if the
/
1 ./
(]) critiera for propagation are satisfied. If so, the same calculations as for ignition are completed. A maximum time step from the input is determined for comparison with an internally calculated stable time step. With all the rates of change calculated, equations for the conservation of total energy in the compartment and the conservation of the masses of the individual constituents are written in finite difference integration form. Knowing the total internal energy of the compartment and specific volume of the vapor, a temperature and partial pressure of the vapor can be assumed and steam properties determined from the steam tables. Based on these values the total energy can be calculated and compared to the actual values. Iteration leads to a convergence within n (/ ' acceptable limits. With the converged results, parameters are updated for the next time step. The finite difference integration is continued until the transient has been completed. ~
SUMMARY
The long form printouts provide sufficient information so that hand calculations of summations were used to verify conservation of mass and energy. Comparison of printed output with the ASME steam tables confirm that the steam properties are properly evaluated by the iterative scheme discussed above. - O p
.)
. . . _ .- . .. ~ _ - . - - . . .
4 O VENT VOLbME
/
3 r 3 UPPER COMPARTMENT U t 2 ICE CONDENSER O f
~
I LOWER COMPARTMENT -
/ t A ~
t l 7 6 5 I ! STEAM PRESSURIZER DEAD I GEHERATOR VOLUME I ENDED I l VOLUME VOLUME
+ + ' i t
L________L_______1 ' l h CLASIX MODEL l j FIGURE 1
O CLASIX CAPABILITIES
- 1. VENT FROM UPPER COMPARTMENT 2.
ICE CONDENSER
- 3. RECIRCULATION FAN Lt . DOORS - LOWER INLET AilD INTERMEDIATE
- 5. INDIVIDUAL REPRESENTATION OF 0 , H , N AND H O 2 2 2 2 O 6. SATURATED AIID SUPER-HEATED STEAM
- 7. SPRAYS
- 8. H2, N AND HEAT ADDITIONS 2
- 9. BREAK FLOW
- 10. BURN CONTROL FIGUE
O BURN CONTROL
- 1. v/o H2 IGNITION
- 2. v/o H 2 PROPAGATION
- 3. o/o H2 CONSUMED 14 . v/o 02 IGNITION O .
- s. v/o 02 supe 0RT C0nsusTiON
- 6. PROPAGATION DELAY TIME
~
- 7. BURN TIME O
FIGURE 3
.~
START YES NO
" o READ RESTART FILE READ INPUT d
READ INPUT CHANGES I T WRITE INPUT EDIT 1 :. v WRITE OUTPUT e STOP? e CALCULATE MASS AND ENERGY RATES BETWEEN h-COMPARTMENTS A CALCULATE SPRAY FLOW RATE AND , HEAT TRANSFER RATES h HEAT ADDITION RATES, HYDROCEN ADDITION RATES ( NITROGEN ADDITION RATES FROM TABLES E BREAK FLOW AND ENERGY ' FROM TABLES i l CALCULATE FLASHING RATES 4 HEAT TRANSFER RATES TO ICE. MELTING RATES
) 2 .s FIGURE 4 1
2 l o ,, l
-j DETERMINE IGNITION, BURN RATE, PROPAGATION h
DETERMINE MAXIMUM TIME STEP FROM INPUT h
' SIMULTANEOUSLY CALCULATE STABLE TIME STEP & ITERATE TO CONVERGED STEAM PROPERTIES AT END OF TIME STEP BASED ON RATES o
UPDATE MASSES, INTERNAL ENERGY & PRESSURES OF EACil COMPARTMENT CHECK FOR DEPLETION OF HYDROGEN OR OXYGEN f; IN COMPARTMENTS WHERE BURNING EXISTS o 1 FIGURE 4 (Con't.)
O APPENDIX U
SUMMARY
OF ANALYSES OF ICE CONDENSER CONTAINMENT RESPONSE TO HYROCEN RURN TRANSIENTS SEQUOYAH UNIT 1 PEPORT OF THE SAFETY EVALUATIO?! 0F THE INTERIM DISTRIBUTION OF IGNITION SYSTEM
- O i
l TENNESSEE VALLEY AUTHORITY SEPTEMBER 2, 1980 O l
Introduction
' (dD 'A series of analyses have been performed to study ice condenser containment response to hydrogen burn transients for an accident sequence similar to the TMI-2 accident. The particular sequence studied is that designated as S2D in WASH-1400. This is a small break loss of coolant accident (LOCA) accompanied by the failure of emergency core cooling injection.
l The S2D transient may be divided into three phases. The first phase is the i period from accident initiation to the beginning of hydrogen generation. This period is similar to the small break LOCA transient. Existing calculational techniques such as the Westinghouse Long Term Ice Condenser Containment Code (LOTIC) are currently used for analysis of the containment response during this period. i The second phase is the period from hydrogen initiation through the end of i 4 core melt. This phase, which progresses thrcugh various stages of degraded core conditions, is representative of the TMI-2 accident. Analyses of the containment response during this phase of the transient are performed using the CLASIX computer program which was developed. by Offshore Power Systems. These CLASIX analyses are the subject of this report and are discussed in the following pages. The third phase of the transient is th'e period following vessel melt. This period is beyond the scope of these analyses. l O 1
CLASIX Base Case Analysis The parameters for the base case analysis were selected by consideration of
- 1) available experimental data on hydrogen burn characteristics, 2) the potential effect of the containment engineered safeguards systems parameters, and 3) the containment geometry. The mass and energy releases from the break (steam, hydrogen, and fission products) were based on calculations by Battelle Memorial Institute at Columbus using the MARCH code. These parameters are summarized in Tables 1 through 3. The conditions inside the containment prior to the onset of hydrogen generation, including subcompartment volumes, temperatures, air and steam partial pressures, and ice mass,were determined from LOTIC analyses and the MARCH generated blowdown. These parameters are summarized in Table 4.
O For the base case, it was assumed that ignition would occur at a hydrogen concentration of 10 percent by volume (V/o). Consistent with ignition at V 10 /o, the flame was assumed to propagate at 6 feet per second. Propagation was assumed to occur to any connected volume with a hydrogen concentration of at V least 10 /o. Ignition was assumed to occur at~the center of the lower compart-ment, along the circumference of the dead ended region, or at the top of the upper compartment. Ignition in and propagation to any compartment with less V V than 5 /o oxygen were suppressed. Consistent with ignition at 10 /o,it was assuiaed that complete combustion occurs in any compartment in which a burn ' is initiated either by ignition or by propagation. These parameters are summarized in Table 5. The spray system was modeled to provide a constant flow of 6000 gallons of 125 F L.) 2 4
water per minute to the upper compartment. The spray drops were assumed to have a diameter of 680 microns, a fall time of 10 seconds, and a heat transfer coefficient to the upper compartment atmosphere of 20 BTU /hr ft F. The i spray system is automatically initiated 30 seuonds after the containment reaches 3 psig. In this transient, spray initiation occurred prior to the beginning of the CLASIX analysis. The spray parameters are summarized in Table 6. 1 The fan system was modeled to provide a constant flow of 80,000 cubic feet per minute from the upper compartment to the lower compartment. About 0.55 percent of the fan flow was directed to the dead-ended region of the lower compartment to represent the hydrogen skimmer system. The fan system is initiated 10 minutes after the containment reaches 3 psig. In this transient, fan initiation occurred prior to the beginning of the CLASIX analysis. The fan parameters are summarized in Table 6. The ice condenser lower inlet and intermediate deck doors were modeled based on i the door representation in the Westinghouse Transient Mass Distribution (TMD) analyses for ice condenser plants. Both types of doors were modeled to act as check valves, preventing reverse flow. The lower inlet doors were restricted to a maximum opening of 55 degrees for this small break transient. The door parameters are summarized in Table 6. Flow parameters for the flow paths between the various subcompartments are based on typical ice condenser containment geometry and are consistent with similar parameters used by Westinghouse on TMD calculations. Flow path parameters are I summarized in Table 7. E The results of the base case CLASIX analysis, identified as JV900 in Tables 8 l l l. L
,, -. - -,- -. ~_. _ _ , _ - , . ,
O and 9, indicate that hydrogen will be ignited in a series of nine burns in the lower compartment over a period of about 3300 seconds beginning about 5000 seconds after accident initiation. One of the burns propagates into the ice condenser. Each burn in the lower compartment consumes about 100 pounds of hydrogen and the burn in the ice condenser consumes about 37 pounds of hydrogen, giving a total burn of about 900 pounds of hydrogen. For the first burn, calculated peak pressures were 26.5 psia in the lower compartment and 28.5 psia in the ice condenser and upper compar' rent,with a preburn pressure of 22.5 psia. Subsequent burns resulted in successively lower pressure peaks. Peak temperatures of 2200 F, 1220 F, and 150U F were calculated in the lower 4 compartment, ice condenser, and upper compartment, respectively. Only small i differential pressures occur across containment structures during the transient. As a result of the action of engineered safety features, such as the ice condenser, air roturn fans and upper compartment spray, the pressure and temperature peaks were rapidly attenuated between burns with pressure returning to the pre-burn value approximately two minutes after the burn. At the end of the transient, 7080 seconds after accident initiation, 650 pounds of hydrogen remained distributed
- in the containment at a concentration insufficient for ignition and 300,000
! pounds of ice remained in the ice coadenser.
CLASIX Sensitivity Studies To determine the effects of ignition criteria and safeguards performance on containment response to hydrogen. transients, a number of sensitivity studies t were performed on the parameters that have been judged to have the greatest potential impact. O
O In the first sensitivity cese, identified es Jv901 1n Tebies 8, 10 end 11, , the hydrogen concentrations for ignition and propagation were reduced from 10V/o to 8 /o. Based on experimental data, the burn fraction was also reduced from 1.0 to 0.5. All other parameters were the same as those used in'the base case. In this case there was a series of seventeen burns in the lower compartment, j eight burns in the ice condenser, and one burn in the upper compartment. Altnough the number, magnitude, and distribution of the burns varies considerab'y from i the base case, the total amount of hydrogen burned and the peak containment pressure during the transient do not vary appreciably from the base case. Peak . temperatures in the lower compartment and ice condenser are considerably lower than in the base case due to the smaller magnitude of the individual burns in these compartments. In the second sensitivity study, identified as JV913 in Table 8, the hydrogen concentration for propagation was reduced from 10 V
/o to 8V /o while the concentration for ignition was kept fixed at 10V/o. For consistency with V
experimental data, the 8 /o burns were restricted to a 50 percent burn fraction V while the 10 /o burns had complete combustion.- All other parameters were identical to those used in the base case. In this case there was a series of three burns initiating in the lower compartment. All three burns propagated to the upper compartment and two of the burns propagated to the ice condenser. Again, although the number, magnitude, and distribution of the burns varies - ccnsiderably from the previous cases, the peak pressures and temperatures are similar to those calculated above. In the third sensitivity case, identified as JVTC4 in Table 8, the hydrogen J Ci
V concentration for propagation was again reduced from 10 /o to 8V /o while the concentration for ignition was kept fixed at 10 V/o. For conservatism, all burns were assumed to have complete combustion. In addition, the fan flow rate was reduced from 80,000 cubic feet per minute to 40,000 cubic feet per minute. The results for this case are similar to the JV913 case, with.an increased peak pressure and temperature in the upper compartment due to the greater magnitude of the upper compartment burn. In the fourth sensitivity case, identified as JV914 in Table 8, the flame speed was increased from 6 feet per second to 12 feet per second. All other
- parameters were idantical to those used in the base case. The results of this case were very similar to those for the base case.
In the fifth sensitivity case, identified as JV915 in Table 8, the flame speed O was increased from 6 feet per second to 12 feet per second, the hydrogen concen-trations for ignition and propagation were reduced from 10 V
/o to 8V/o, and the burn fraction was redu:ed from 1.0 to 0.5. All other parameters were identical to those used in the base case. The results of_this case are similar to the first sensitivity case.
- In the sixth sensitivity case, the fan flow rate was reduced from 80,000 cubic j feet per minute to 40,000 cubic feet per minute with all other parameters identical to the base case. The results of this transient are almost identical to those for the base case.
j In the seventh seasitivity case, the fan flow rate was reduced to zero in the
- CLASIX part of the analysis (i.e., from the beginning of H2 generation). All i
l 1
h other parameters were identical to the base case. In this case, steam and hydrogen from the break push air out of the lower compartment reducing V the oxygen supply below the minimum 5 /o required for burn initiation. As the transient continues, hydrogen accumulates in the upper compartment and eventually ignites there. This burn propagates to the ice condenser. The combined upper compartment and ice condenser burns cause a redistribution of the containment atmosphere, adding oxygen to the lower compartment and hydrogen to the upper compartment. With the addition of oxygen to the lower compartment there is an independent lower compartment ignition which also propagates to the ice condenser and forces more hydrogen into the upper compartment in which hydrogen is still burning. 4 The net result is a burn of approximately 1200 pounds of hydrogen in total, , of which 860 pounds burn in the upper compartment. The peak calculated pressures are 92 psia in the upper compartment, 86 psia in the ice condenser, and 46 psia in the lower compartment. J In the eighth sensitivity case, the ice condenser drain temperature was increased from 32 F to 132 F with all other parameters identical to those used in the base case. The results of this case are almost identical to the base case results except for the ice remaining at the end of the transient. This indicates the drain temperature is important to the ice condenser efficiency but not to its i effectiveness. -l In the ninth sensitivity case, the initial ice mass was reduced by 1.17 x 10 0 1 pounds. Thus the ice mass input to the CLASIX part of the analysis was 1 6 5 i j reduced from 1.67 x 10 pounds to 5 x 10 pounds with all other parameters ! 4 b .
'O V identical to those used in the base case. This is a non-mechanistic study to determine the total effect of ice on a hydrogen transient. In this transient, ice melt out occurred during the second of a series of seven burns in the lower compartment. The peak containment pressure was about 10 psi higher than in the base case.
The remaining sensitivity cases were performed varying spray parameters. Since spray operation is much more important for upper compartment burns than for lower compartment or ice condenser burns, these variations were performed for the upper compartment burn in the first sensitivity case, i.e., the case with ignition and V propagation at 8 /o hydrogen and 50 percent burn fraction, identified as JV901.
'In the first spray sensitivity case, the spray heat transfer coefficient was 2
reduced from 20 BTU /hr ft F to 2 BTU /hr ft F. In the second spray sensitivity case, the spray temperature was increased from 125 F to 180 F. In the third spray sensitivity case, the spray flow rate was reduced from 6000 gpm to 4700 gpm. In the fourth spray sensitivity case, the spray flow rate was increased from 1 6000 gpm to 9400 gpm. In the fifth spray sensi.tivity case, the spray flow was reduced to zero in the CLASIX part of the analysis. In the sixth spray sensitivity case, the spray drop diameter was reduced from 680 microns to 400 microns. In the last spray sensitivity case, the spray drop diameter was increased from 680 microns to 1000 microns. .The results of the spray system sensitivities are summarized in Table 12. These results indicate that while spray temperature and some minimum flow rate are important, the remaining parameters are relatively unimportant for containment response to hydrogen burn transients. In the spray temperature sensitivity case, the increased spray temperature resulted in an increased ambient pressure. The increased pressure required additional hydrogen to achieve l b s-I
)
l
._ - - . - = . , - - . -. , , - - - - -l
O the hydrogen concentration required for ignition so that a greater quantity of hydrogen was consumed in each lower compartment burn,and an upper compartment burn did not occur. i O
==
l t ? l O
-9
Summary and Conclusions (v'] The calculations described above represent the first attempt to perform a realistic assessment of hydrogen transients in an ice condenser containment. The geometry and engineered containment safeguards parameters used in the study were based on the Sequoyah containment design. Therefore the results are directly applicable to the Sequoyah plant. The results of this study indicate that over a wide range of cases, the Sequoyah containment pressure response to hydrogen transients would not cause containment failure. Future analyses will be performed to expand the range of parameters studied, such as the burn criteria in the S2D transient,and to extend the analyses to other transients. (~s. L.) \_/ TABLE 1 7_, MARCH REACTOR COOLANT f1 ASS AND ENERGY RELEASE RATES '~' S2D SEQUENCE TIME MASS RELEASE RATE ENERGY RELEASE RATE (sec) (1bm/sec) (BTU /sec) 0.0 197.167 116722.67 2172 190.500 109728.00 2478 44.850 52295.10 3180 53.533 65471.27 3804 34.817 42615.60 4428 21.400 28419.20 4752 48.417 55582.33 5700 19.417 21824.33 6012 14.067 15825.00 O
\> 6960 5.253 5988.80 7062 4.718 5388.34 7206 4.060 4693.36 w
9 I ()
~
TABLE 2 f1 ARCH HYDR 0 Gell GENERATI0f1 RATES At1D TEf4PERATURES S2D SEQUENCE TIllE f1 ASS RELEASE RATE TEf4PERATURE (sec) (lbm/sec) (9F) 0.0 0.0000 61.24 3480 0.0000 61.24 3804 0.0413 66.56 4116 0.2600 1582.29 4428 0.7400 795.45 4752 1.0700 771.47 5700 0.4300 611.53 6330 0.2233 555.39 6648 0.1600 535.22 ({]) 6960 0.1167 519.43 8070 0.0367 519.43 9 h
%)
i
, -n~,.--~ ,
TABLE 3 (MRCH FISSI0fl PRODUCT EllERGY RELEASE RATES S2D SEquEllCE TIf1E EllERGY RELEASE RATE [sec) (BTU /sec) 0.0 , 0. 0 3810 O.0 4116 1803 4428 4800 4752 6708 5376 7000 7080 7135 0 . ee 4 0
__ _ . . . _ . _ . . ._. . . .... _- _ __. ._ _ _ . . _.= __.. . . _ . . .. . . O O O. TABLE 4 , SUBCOMPARTMENT PARAMETERS
- CLASIX BASE CASE ANALYSES S2D'SEQUErlCE LOWER ICE UPPER DEAD ENDED COMPARMENT** CONDENSER COMPARTMENT + REGI0fl VOLUME (ft )
3 3.03 X 10 5 7.85 X 10 4 6.98 X 10 5 7.87 X 104 02 PRESSURE (psia) 2.53 3.70 3.62 3.62 N2 PRESSURE (psia) 9.56 13.98 13.67 13.67 i H2 0 PRESSURE (psia) 6.10 0.40 0.90 0.90 TEMPERATURE (OF) 171 75 98 98 ICE MASS (1bm)' 1.67 X 10 6 2
- . ICE HEAT TRANSFER AREA (ft ) 2.02 X 10 5 l
- BASED ON LOTIC RESULTS i ** INCLUDES MELTED OUT PORTION OF ICE CONDENSER 1
+ INCLUDES ICE CONDENSER UPPER PLENUM i
1 1 1
TABLE 5 ([) HYDR 0 GEN BURN PARAMETERS CLASIX BASE CASE ANALYSES H2 V/o FOR IGNITION *10 H2 V/o FOR PROPAGATION 10 H2BURN FRACTION 1 02 V/o FOR IGNITION 5 MINIMUM 02v/o TO SUPPORT COMBUSTION 0
** BURN TIME LC 12 sec IC 5.5 sec UC 22 sec DE 2 sec
** PROPAGATION DELAY TIME LC-IC 12 sec IC-UC 5.5 sec UC-LC 60 sec f
q.;
-) LC-DE 12 sec
- EXCEPT IN THE ICE COND$NSER; ASSUMED NO IGNITION SOURCES AVAILABLE BASED ON A FLAME SPEED OF 6ft/sec I
/\ \
TABLE 6 SYSTEM PARAMETERS CLASIX BASE CASE ANALYSES Spray System Flow Rate 6000 gpm Tempera ture 125 F Drop Diameter
- 680 Fall Time 10 sec Heat Transfer Coefficient 20 BTU /hr ft2 og Initiation Time S2D during LOTIC Air Return Fans Flow Rate 80000 cfm Fraction of Flow to DE Compartment 0.0055
.O Initiation Time 52D during t0nc Ice Condenser Lower Inlet Doors Maximum Opening Angle (degrees) 55 Differential Pressure for Maximum Opening - 0.0206 psi Maximum Flow Area 840 ft 2 Ice Condenser Intermediate Deck Doors MaximumOpeningAngle(degrees) 85 Differential Pressure for Maximum Opening. 0.493 psi Maximum Flow Area 982.47 ft 2 O
O O O . TABLE 7 FLOW RATE PARAMETERS
- CLASIX BASE CASE ANALYSES LC-IC IC-UC UC-LC DE-LC 2 ** **
Flow Area (ft ) 2.2 108.6 Flow Loss Coefficient 1.12 2.26 1.5 3.0 i
- Based on TMD Models for ice condenser containments
** Function of door opening
4 4 A W<>%+V. IMAGE EVALUATION
$<s.i>
TEST TARGET (MT-3) l.0 'EM BLA El0 EM l.l [m IllE l.8 (25 IA 1.6
- . e., =
i MICROCOPY RESOLUTION TEST CHART p% + /4 Ihf'N 4,,,,,/ 4*)phS!E p 7 (, ;e 4
. . ._ . - 2. = . u. = :. . . . -
k h'4 xxf'%+$, s ////4+M V IMAGE EVALUATION TEST TARGET (MT-3) l.0 l# flu DM yly lie I.I ['" OLE 18 I l.25 1.4 i.6 I . , l
< 6" MICROCOPY RESOLUTION TEST CHART #lll4 4%
?$fi/hr ///// ( ;
$+h4$
9
. _ _ . . , s
. . . . . . = .
pj (' , , c - TABLE 8
SUMMARY
OF RESULTS S2D BURN SENSITIVITY STUDIES JVTC4 JV900 JV901 JV913 ~S2D JV914 JV915 Base Case (10/8%) (10/8%) (10%) (8%) (100/50% Burn) (100% Burn) (12 fps LC) (12 fps UC)
# Burns LC 9 17 3 3 9 IC 1 8 2 2 1 UC 0 1 3 1 0 Magnitude of LC s100 s45 sll5 90 s100 Burns (1bm) IC 37 16-45 25-49 80 48 UC -
200 s230 430 - Total H2Burned (lbm) $900 s1050 1100 950 900 H2 Remaining (lbm) s650 m 500 450 600 650 Peak Temp. (OF) LC m2200 s120Q $1900 2100 2100 IC S1200 m 700 m 630 1500 1370 UC m 150 m 260 m 275 480 160 Peak Press.(psia) LC s26.5 s28.5 29 34 27 29 IC m28.5 s28.5 29 44 30 29 UC 28.5 s30.5 33 53 29 36 5 5 5 5 5 Ice Remaining (lbm) 3x10 3.2x10 4.5x10 5x10 3.2x10 Figures 1-8 9-16 17-24 25-32 33-40 41-48
. -.. . . . ___. .- - - _ - - - . . - . . . _ - . _ - - - - . - . . _ _ . . . . .- - . .. .. . . . - .~ - .,
o O O TABLE 9
SUMMARY
OF RESULTS S2D FAN AND ICE CONDENSE'R SENSITIVITY STUDIES JV900 JV402 JV903 JV904 JV905
'(Base Case) (1 Fan) (no Fan) (Less Ice) (Drain Temp)
# Burns LC 9 9 1 7 9 IC 1 1 2 0 1-UC 0 0 1 0 0
] Magnitude of LC' N100 $100 s130 $100 Burns (1bm) IC 37 s 60 - S 39 .j UC - - - - Total H2Burned (lbm) s900 s900 S1200 s850 s950 H2 Remaining (1bm) s650 $650 350 s700 s600 . 1 Per' Temp. (OF) LC 2200 m2200 2370 m2400 s2000 IC 1200 s1350 2583 m2000 s1270 2 UC 150 s 160 1088 s 270 m 150 , I Peak Press. (psia) LC 26.5 s26.5 46.4 s41 s26.5 i IC ' 28.5 m26.5 86.4 s41 s28.5 UC 28.5 m29.5 92.4 S41 s26.5 5 5 5 5 Ice Remaining (1bm) 3x10 3.7x10 6.3x10 0.0 8.3x10 Figures 1-8 , 49-56 57-64 65-72 73-80 1 I a w f 1
. O O O .-
TABLE 10
SUMMARY
OF CASES CLASIX SPRAY PARAMETER' SENSITIVITY STUDIES i Case Number JV901 JV?n6 JV907 JV908 JV909 JV910 JV911 JV912 Spray Flow Rate -(gpm) 6000 6000 6000 4700 9400 0 6000 6000 Droplet Size /f 680 680 680 680 680 - 400 1000 i Heat Transfer to Drop 20 2 20 20 20 - 20 20 (BTU /hr ft2 p) Spray Temperature (F) 125 125 180 125 125 - 125 125
- Figures 9-16 81-88 89-96 97-104 105-112 113-120 121-128 129-136
) I i i-4 1 t i
.mw
rs TABLE 11 f
SUMMARY
OF RESULTS CLASIX SPRAY PARAMETER SENSITIVITY STUDIES Peak Pressure Peck Temperature Case (psia) (oF) JV901 30.5 260 JV906 34.0 340 JV907 This case did not have an UC burn. JV908 33.8 270 JV909 30.9 255 JV910 52.3 930 JV911 31.5 255 JV912 32.2 280 0 .
~
l
\_/
1 I
O
- ______.. L . __
1 l l l l i u mea.e 1 E ,, t O h i
-I i !
lao.e l
- i i i l 1 t
\
i . l! l
!\ '
) \
\ \ . 'i \ \\ t \ t. t
\. Y
' ' \ \ \ \ \
\ \ \
w
-/
8.0 8.0 1800.0 aTO.0 3XD.e - 4XO 0 , TIr1E (SECONDS) S2D CASE 12 F#11 SPRAY BLRN 100 PCT AT 16 410 GFPS T+3480 BAW1 l l [J
'u_
4 t l Jiq.I
.14 M
E . -- 't./ +- 159.0 5 Y W __
\ , n
%Jun_%J% wa n--a ~-
1 0.0 - 0.0 1a9.0 HKO.0 3XD.0 m.0 Tire (SECONDS) S2D CEE12 FM 1 SPRAY BWN 100 PCT AT 10 V O 6 FPS T+34S0 BASCI f c,g ' . a_
R U
-ZZZ 19B.8
__u Zd' 1e.0 3 g m.0 O I M U0 2 1
! l \
m#'vN4a~ %' J
/
s'1t n , L
\ j I
""s '
\'Al 7
n L J\a/
]\
11e.0 / I I I I I I I IW.0 SELO 0.0 HEO.0 iE0iLO 3m0 ( 10 0. 0 TIFE (SECONDS) 52D CASE 1 2 F#1 1 SPRAY BUMt 100 PCT AT 10 0 0 GFPS T+34SO BASE 1 l l 4 J O ehy 3
V 1142.0 f I . I ? i :n i i :t i . 1930.0 it i A in r
,n, f b- -7 ,I f; JB04 i ). ij
~b y 11 f!
SBA ! !
! i i l i i ! i 8 u r ; i i i
' w .e I" 1 I i i k i 1 i i i i i l' L- 1 )
I g F 3 3 ; g g GE.0 i i i 4 V s-I
\
i E \. \ \ 55"2 B i
\
1
\
W
\
y , 40.0 ;I 39.0 ^ .
/
./
3D.0 ./ J I i' 153.0 O.0 ' IW HKO.0 mo CV.0 TIFE (SEC0roS) S2D CASE 12 FM11 SP9ny BLRN 100 PCT AT 10 V 0 6 FPS T+3430 BASE 1 f" N)) i c, . Y
'm
m ( )1 27.e T I Y r i 25.0 E l [ u 34.0 -
?. '
d 1 I [ I -
'1 1 :-
1 I I E
- ELO
-i - ,
i +t 1 1 E Se I i r , 1,, 1, 1 I I 1 1, W n.
, 1 1 ' I
'. _: 1
_E , ' o 8.0 ,
,' ~
3
'I l l
20.0 _--er-19.0 I f i 18.0 . 8.0 1000.0 iEco.e any,o 4w,0 l T!rE (SECONDS) l
$2D CASE 12 F#t 1 sppgy gtprq ggg PCT AT 10 V O 6 FPS T+3480 BASE 1 l
. E
A Q I 27.0 I I I I l I E.0 ' I 1 1 Y I f 5.0 1
~;~~ ~
I I H I I g d*
---t- , 4-.
m a. 0 1 1 1 1 I 1 I I i Z L I . .
+w--
i Ihh,- : cr i
- 1 i 1 1
- gLo '
E i I
!? _t
-t-+
w - I /- 1 l 1 eg 7 I E 1 1 'N) 'I ' h , 1
- , , 1, x
2.0 v I i 1 CL. w ,
.d_ r o 2 m
. =
a.0 J I - - L 1
"'I y I l .
I f 2.0 -r I J' 1 I I. T ' G.0 ; I I I I I I I I I 1 I IfLe 0.0 ICEO.0 350.0 3W.o Aw.e T!fE (SECONDS 1 S2D CASE 12 FM 1 SPRAY DLRN 100 PCT AT 10 V 0 GTPS T+34SO BASE 1
~
,A, s ._/ G 9 -
(~') v 2.e (
-. $ (
s \lU\n \ .\ \ , J \
' \ \ \ . k {
s 7- % [ w a.e / ( ,1 y , -- - O E l d b 20 4
# 350.0 yy,g 41V.0 .
TITE (SECONDS) D MK12 FAN g $s94y M 100 PCT AT 10 0 0 6 FPS T+NSO BAm 1 1 1 I eN h , ,/
'7 l
y. \ ] F.0 1 I I I I - '* I = E 50 -g--
._._w. ; ,
- r y
.h,/.
I I I I I I 1 25.v ;- E
' I ,,
_. qq .:: r
.+ E I
g 24.0 1
'i .
I 1
.E I 1
I 1 I I l I I I I II I
.. i 1 1 I I a
I', T l 1 I e r 1 1 I i Cg ' ' m
- ELO o i 1
.i = - - '1 1I I
i 1 f g i I 1 ;3 1 I I > a, I i \ 1 1 1 Nw .,/ - f 1 E I 1 I I I I E I _* J Y cc -I I 1 i 1 : 1 I . r i A -Q,
. I d
~
I z J '. O I I 'F I p ' ' a.0 - l 1 I m- x -__- 1 I J pl--- 1 1 I 2D.e fr 11 - I I I 4 I 1 I
"'A-
- I I
J I f 19.0 .. n- .. f, r_ 1 I I I 1 ' 18.0 O.0 lad.0 aW.0 310.0 60 T!riE (SECONDS)
$2D CASE 12 FM 1 SHMY BtRt 100 PCT AT 10 V 0 6 FPS T+3430 BASE 1 0
\vb
. T<
n v 4
+
iMilbe i I ' I $ 1 l ' V - 1000.0 ..-. g ' w
- s
- ll k
f ! l lf
!l ! !) h ;
} } } l l il \ i , t ( ( {
t !
\ ) \ t \' \ \
\ \ .
\ 4 \ s
\
s
\
l
, /
0.0 -- N he 2m,9 g ,p TIPE (SECONDS) 2 rm 1 my PJ<M 50 PCT AT 8 V 0 G'PS T+3480 BASE 2 l f - I 9
.O i ! . %J g E
I I I I i i I I
*Ete I
I I am.e u E0.0 i 6 I
'I O. % 40.0
( )
%/ E w
h O j -- 30.0 I I 1k aE.0 I"k
'I
?
Is . t+
.II 1 m , t II..li 1 I L l' 1 B i1. l 1. L.
' ' v ' - '
200 4 '. ' 1 . 1 L 'l I' I\'u I
,f &R I 3.
. 's vA I l
m' -- L _.s'.'s[L O.e 0.0 tare.e 3 9.0 3W.0 4W.0 , TIFE (SECONDS) S2D CASE 2 2 Frd 1 SPRAY B'Ji'N 50 PCT AT 8 U 0 6 FPS T*3480 BA%2 l I N..) [I - t L
w w 38.8 l
-b._
M G E
\
(v3 g w N T I d,. ;- \ ,
)
-~ ~~-~-
1 12.0 1 1 3y,p 3mp 0.0 0.0 '1000,0 me ' l TIME (SEC0tCS) f BASE 2 50 PCT AT 8 U 0 EFPS T+34So S2D CASE 2 2 FAN 1 SPRAY BURN ! o o eduy.it
I i (O,/ .,i i ! b. lalle.e L n
; fi l' !
f H in
. 1.1 L' 1 I i r1 0!1 i n!i l L1 ' I s i i 1
l, I 0 a 1 n 1 1 h ,
\ \' l1 l' T yg \ t \ t i
. i 1 i i \
\
b i di \ L I \ l 1 1 1 I i i 1
' ', l '
\
O g~ > 1
\
\
f
\ t 1
. r
\
( \ Ti g 49.0 \ i 3D.0 1
/
/
/ ..
l 59.0 ,
/ .'y.
I i 12.0 I i ! I e.e 3h1 Ante 0.0 asse a800.0 TIFT 4SECCNDS3 S2D CASE 2 2 FAN 1 SPRAY BURN Se PCT AT 8 U 0 GFPS T+3480 O) iw ,'
/
,/'%
\s 5 m
a
\
l .
)
e i \\rdh \
\\
8 I y e 8 k \ ( h -
/ x sx x s
,Q a0 v , f f s' e 10.0 mgo g ,p . I U *O 900.0 TIN (SECONDSI r, 2 2 N 3 N BLRN 50 PCT AT 8 0 0 W T+3410 1 Or o i i I 4 . /O
(' %, b F 38.0 1 I
\
1 i
- i. \\n k\( (
e ._
\(\ >
i l
\
) i
/
t
' \T \ \ \
& s Ns \s
/
m.e - (v3 , - w
$ f d
e
<ny,0 10.0 c"IXB.O 3ee.o .
00 1 @ .0 71f1E (SECONDSI 50 PCT AT S V O 6 FPS T+34 M BASE 2 S2D CASE 2 2 FAN 1 $ PRAY BLRN i o
[ ( e.e g 2.0 e G k b ('N V g lLJ
#1
\
I m t lYWj!% \ { \ \ ) i d ' \
\, b \ \ \ g 5 j \ \ 's /
y,, ---- g f sae '- g as s e .e pp.'8 hw T TIME (SECONDSy
$2D CASE 2 2 Fret i sppgy W 50 PCT At 8 U 0 & PS T+34SJ BASE 2 O ~
./C
i a
?
3).O I' l
\
\
\ .._-
, i.
1 ! \ IS[ ,' k k 8
\\ ! I 5
f k\ ' \ ( ( - r e x xs yJ ps a,.. i W c { d i 5 1 l l 10.0 300.0 4w.e E0 1000.0 ake.0 TIFT (SECONDS 1 S2D C/G.2 2 FAN 1 SPRAV R.RN 50 PCT AT 8 U e GFPS T+34Se a f 1
%J css /' . &
L
d ama.e i I i S _ l E ! w p IM.O --" 'V E ) l 1 5
- 1
\ \ \
1
\ \
__. \s \ \s s s % ---_< s _
, r"
/
, _ m e.e m **
T!rE (SEM) 52D CASE 14 2 F#t 1 SPRAY BURN 100 PCT +cO W 10 Wo + 3 N 6 yps T+ 3WO v4 edil9 17
f f
\-v l i
I se.e l
; r ;
E sne e
; 49.e ,o
(.) h E g aae.e ane 1 fl e 100.e : ( , .
- ~
., l -~
t --- (Le e.e laae ano.e sm.o 4xne TIPE (SECONDS) S2D CASE 14 2 F#1 1 SPRAY BtRN 1ee PCT .6-sDO ~PC.,T 10 U/O + 3 U/O l 61:96 T + 3 A/ 70 O v 4tp /5' I
O G 28.0 I il i.
' l! 11 w
ae.e I il I il C \ th w t \ ! (- C' 5 \ \ j \' ks) .E w --- 1 ( ; (
\ \
i . I \
# li \J \-
\---_x, -
1m.0 ! l
" j ev.o ame m .e e.e um.e I
Tire (SECONDS) 1 S2D CASE 14 2 F#41 SPRAY BURN 10e PCT +4070.7 IO ufg g v/o 6I-~PS T + a yya l l O I
Ms.4 8 ii; Ou .E
- 1980.0 J i !)
K
- I i i
\
l L ( (
\ \ \
\ >
\
g
\s
_j \ ' --._,_j ' J u - 44 3 Ass.e 3188.8 3e0.0 4so.e TIFE (SECONDS) S2D CASE 14 2 FNs 1 SPmW EUEN 10 V 4 + B U 0 GFPS T+3480 lo oPCT + @o T-U.,'T c"9"
,,) 'w e 6
v 3.e I
\ \
t < A
\ \
! ) . \ (
3 1 \ i t' i .
/\ \ \ \
g / \ \
' / \
O gm 7 x E / d e 18.0 3NP.0 4 00.0 , 0.0 120.0 3E0.0 TIFT (SECONDS) S2D CASC14 2 FM 1 Smv BLA 100 PCT t@QPC,7, IOV/ot $ U/O toFPS TtO4%O O \'
= h
r 3. f k A
\
l \ \ l \ 1 ' I, I m
~ \ ! V i
/ \J \ i i
_ / s / \ p ---__
' w__;
s 1 () g m.e , - x r r d e
** "* me.e m., SP.0 ,
Titt (SECONDS) 1 N U w 190 PCT % ggy gy 6 F70s T.f.3g (3/ \, l
A V 4 e.a M ** 3
\
e \ ! tl
' \\
0 w/ w
-\ \ \
\
)
y
' i
\
L
\
( Yl
! (
\
o -- Ns \ 1 8 / \. x ---;
\
S g,, ----
, / ( ,
f I , 10.8
&O 1s2.0 ige.g . BED.0 4X0.0 t
TIRE (SECONDS) san cat 4 a m t sev susm tee ect +6o FC,r tov /o +g v/o laFPO T+GLl%O v
.d
/S %s]
3A
! k
- l. , .
\ \
I \ , \ \ s \ \ t' i i sJ \
/ -
p
- ----.- i r e d
e Sea N <M184 320.0 eise.o TIE (SECONDS) S2D CASE 14 2 F#t 1 my RRt 19 U e + 8 U 0 SFPS 7+3480 lCCTTCT*-t- QRT l G d
(,9"
)
n
- , 'i .
n 3 18. 4 0 3 l 6 w
^ E (d
j' l I 1000.0
\
1
\' i i
\ s s \s i s ,
- k. " /~ l l
6.8 l M 1818.0 3900.0 'E00.0 e20.0 TIPE (SECONDS 1 TVA1 S2D BASE 1 F#41 SPRAY BURN 100 PCT AT 10 U 6+ 8 U/9 6 FPS T+3480 l l O ed ' . cQE l 1
<~
-0 te:,
( g la0.0 sa h
- e d , \.
; a _ -
#8 38E04 320.0 3100.0 4319.0 .
TIfE (SECues) M1 S2D BASC 1 F#t 1 SPRAY BLFm 100 PCT AT 10 V 6+ 8 Ve9 6 FPS T+3480 l i V t
b-
=c* ,,
ea.e 8 3 10 .0 E D- $ V g Y W < ae.e 8 ( h t
\
\
l \ t l' Aw W_. 0.0 A8 188E8 SES.0 3ee.o egg,9 TIT ISEC0rfDS) M3 55 8ASE 1 Fart 1 SPRAY BLMf 100 PCT AT 19 U 9+ 8 U/9 GFPS T+3484 O O l i l
' Q9 '
f
O I t 1 8 k 1 s J 180.0 fh
! s r
- r ;
l k s -----
/
s
-:=-.--
J u m iB00.0 he TIFE (SECONDS) 1 1 N BWN 180 PCT A7 Je U b 8 U/0 (; FPS T+348e Q 'u) edig.aS'
( < v' e.e l l
\
y 3.0
\
\
1 e Ga (
\ \
s i d(m $ e m g M / \
' / ) \
d ', / ' 5
" ' ~ _- -
3.0 7 ,
}
f 1 l \ u , i M 3e80.8 ins.e 3x10,0 4co,o , TIFE (SECONDS) TUR152D BASE 1 F#81 SPRAY SLRN 190 PCT AT 19 U 9+ 8 U4 GFPS T+3480 l l l l l 1 l l p,5 t
\j i
1 1
rs ) (
%)
9Le e.e hf 5 0 i . . . 2.0 f i i -f p) ( m y 1 I
\
y
- a. I L d \ \
B
,s J
i
\,
s"/ k
-~ ~
y e ( 18.8 Me afe.e g,, 410.0 72 T ISEC0 resp M I FM 1 SPRAY RJM tee PCT AT 18 U 0+ 3 g/g GPS T+3m i V ed . &D
v B.4 M ._ . S , l e.0 6 M s s {J N W 1 l n. g 3.0 \
\
5
*- 1
! k.
..___.,i i>k \
m J % 3
"' \
---~
_,- -" s% l9.0 y
~__._
38.8 1818.0 # 3100.0 60 TIFE (SECONDS) I V W 100 PCT AT le y e+ 8 U/9 SFPS T434so O I v l
C\ / lw M l l 8 24 I i
' \ -.
1 E r (^ p) ~-- \ \
\ ._
x 3 Y i
. '\,
= /
w O
> , #~
^".'<" % w ,
j ( ses 8.0 1829.0 h4 NA MO TIT (SECCNDS) I M I % 1 SPRAY m 100 PCT AT 1e V 0+ g ufe g,rPS T+N m r"~% %)
C' s f 3ame w_ .._ _ n E l kJ E w a w I i 3 , n r tom.e I u
. t I l ! i
( \ \
\- \ \ \
i ( \ ( < \ ( i \ i i ; i \ ( \
\ ,
\ \s \
'~
en . p l u asse amo amo . IIE (SECONDS) S2D CASE 15 2 F m 1 SPR M Bl7 tit 100 PCT AT 18 V 012 FPS T* 3436 ,~ v
l O U G
/"N .- 150.0 . 's) b f
I e \ t : Y
't I
L. _ . , 2 6. \
,_sfw xawa s_.; -..-- ----
0.0 MeELS ales 8 M.0 4200 TIPE tSECONDSI S2D CASE 15 2 F#t 1 SPRAY R.Rt 100 PCT AT 10 V O 12 FPS T+ 3480 1 n I (-) l
e
/m \
O
!= I I I I I I , I I I I I I I I I I I 3 y I
I I I late , x I I I' I 1 SLO I I 3 se.e - I I em I b i W 1 f"N. b 12.0 (s._) E m . I. d 13.8
! \is:. ,
y i '
.g_
i a
!{-
I I l_ Jl I L-
.l x-
. Ev"1 I
1 . 1
,1 1 1 12.e ' >
1
! 1 i -a \
l l M8.0 1 1 98.4 . 1 1 i M 3edr.e igue.e 3w.0 <ac.e i 1 1 TIPE tSECONDS1 SED CASE 15 2 Fat 1 SPRev msm see PCT AT f e U e 12 FPS T+ 3480 l l v 35 : l
. /~N G
u.4
. i t I
i 1e2.0 i f. go.e i , i l I' !j i 82.0 l. 8 6 o i m .e
" r , t ! ,
b l 1 1 i GB.e i I i 1 W i b 1 i i <! i 5 ! 1 1 i I '
/ E . I ,
Y n \ \
$ ge.e t1 L 1 1 n
1 i, 'i i i 49.e <
\,
\
t _. 3D.9 J
/
/
/
3B.0 .
/
1e0.0 O.8 - E0 1810.8 ins.e 3go.e 43.e . TIRE (SECONDS) SED CASE 15 2 F#t 1 SPRAY pt.RN 100 PCT AT le U e 12 FPS T+ 34SO i [
sm. i%.)\ 27.0 e ,. - . , r l -+- I I I I 1 I I I I I I I I I w
- . *N.
I I I u 39.0 I
' ! ;+ ' '
ti; I I I l
&m -.'I+v +i . .
E. ys m r i i .. 1.- II
. I I
4. I I I g- i . I 2 I
' ' ', ' p !_ , g, ,
,f %i I I i 5+
- i i 1 V "'
E.0 I 1 I I
-m i '
1 , I I _ u r a I i F I I 1 1 I i I I 1 5 I I De I LI
^
'-w-
,O,_ I 1 i I 21.0 ' I i 1 i I 11 y 9 I l~ -
1 I I
- 1 I I I I iB 0 J
4.r; I 1 I I I
.. I I
ggg , 1 I II I I I I f I I r I I I I I I I 1EELO 2E00.0 300.0 420.0 , TIPE iSECONDS3 S2D CEE1E 2 FM 1 SPRM BURN 100 PCT AT 10 V O 12 FPS T+ 3480 r ; s / w
- 77
[ I hf I E E A I l [\ i k L. k i. l
@ ts \ \ \ \ \ s ! (
\
/ \ \ \ (
$ / \ 's ~ \
? - / m a'.0 , ___,.
/
f 18A l l1 e.e 2 2.0 am.o anw c o.o TIPE (SECorcS) S2D Cd E !S 2 F#4 1 SPRAY BURN lee PCT AT 10 V e 12 D 5 T+ 34SO sj dp
[D v o 3.0 4 l N \ l N r \ (
\
\
~
\ \ {
/ \ --
[
\ T
( h /
\
\
E N
/
(g3 f me p
? (
d 2.0 he ggg,p , TIPE iSEconos, 2 % 3 m W 100 PCT AT 10 V 612 FPS T+ 3480 4 [o ll
\
[G 27A I I I I 15.0 F <_
..b. ., t I
E g as.e I g;;j __: I 1-- I 3 I1 e' f. , N I 1 1 I I
;g e
- 1 I ..
.1. . ,
41 c i 1 h 1
'~ ' I
- p. .
-', I I I II l I I '% 2
{%
- m.0 , I1 l I' '
l 1 I W w -- -- I I I I 1 I , 1I I
, I, Y 1
c> 3.0 / i t i , j - I I I iD.O l l __~- < ,
-r I I
I I I 19.3 E 1 l 28.8 ---- M 3808.8 3B20.0 T00.0 490.0 TIPE NCONDS)
- $2D CASE 15 2 F#41 SPRAY SURN 100 PCT AT 10 U 012 FPS T+ 3480
'd Q(jf. YO
de U
- l W
- a f
s f. g C ,
>- a
' H X- 0%
Q () i
\
ts
\
\
l l L E w N I i
, a 3 $
8 8 v w wTaun 8 i . 5 O - Wig V / l
O N I: k
- - ! k
[[$- 8"
, gg O '
i E
!=
i 1 1
~
i m 3 i i i i i i i i 8 31 (A 3dnitsfMG1 l E . O . e ax. ._
I
/
O E
==
W
/ 8
/ 6 I
( s d,
- f k O !a
_t_ -
, 1 1 I
-a ., g e
i 3 K m R R 3 9 g an a wuwwu r ry (_./ .
& L/b
O E
/-
-mm*Y
&pr-
,y- N h E'
.-- e.
Nur e p C[.. 1-g#
!y
^ *a N
T
\ r ;
i g I 5 u N WI
= a ---, ,
w) *E MM
g l
-e I:
/
_~- '
- l, n
C
+ I.
i 7-~ r C
~
O 'x s i
\
\ l k
\
b
~
\ !
2 1
' b 2 I I 31 (UISdl MISSM T101 E k
O . - 4 6
l O
. A V
/ g
, s "
g k s v> . _5
~
h
- g. =".
] y*
4.- h O x s
\
\
l l
\
k I ~
\ in
\
I I E I S m mu msw wi E 4 46
J V 9 O O I
/ lz
/
s - 4
-eC
=
s. g-I 7 g>* 2_ y O '
<-, g d \ W E
\
\ 3 E
_ g E
\
1 E w N ee
\
s s g I 3 1
, .u msw mm !!
l l l e 1 a. 1 G - dj . 4"/ ;
O i o
~
5 5 E
, /
- /
,# # , l
+
t 7 ^
}.- y l
# 7 E i
\ E I s
\ i I l
i w N l 2
\
}
2 2 I 0 E LON) JunSS M T101 E r l t O . V
O ' S aine I I E
-I I <
s ' 1 t i n g I w V & 0 y I ---- t i Ll i l l ! l i \ \ \ \ i me \
. \ \ \ , \ '
\ \. \ '
\ i \ >
( \
', li \, 1 s
\ \
! \ N s
,/
0.0
' IU 0 SD3.0 MO g TIT f SECCt4DS)
W31 FM 1 SPRAY ptm 100 PCT AT 10 U 0 GrPS T+34s0 casc3 (~~\ > r
\
I I f l
f%. O a l M G lap.0 w h u lk _ m
-J. s c'~
JsP i r. s_ys_,J ~
&O Igne amo 3g,o 4g,p ,
T!PE (SECONDS 1 S2D CASE 31 Fret 1 SPRAV BtFN le PCT AT 10 U e GFPS T*3430 CASE 3 v 50
A 6 d b# p
- b i
M i 1 I late i I I I
!!DA 8
10.0 E I . 1: y' - O s e r E w e- p .7--.- _ P A i G.0 iA :
--( S -'
qygf4 ^_ _ .- =u- - -
-m
[i i b I 110.e ; j . 6:::,
~
m.e I
**1 I I 1
- 1 g I aIO.0 310.0 4W.0 0.0 1000.0 ,
T!tt (SECONDS) CASE 3 52D CASE 31 F#1 1 SPRAY BURtt 100 PCT AT 10 V 0.6 FPS T+3480 l V
f%, L) ase.e ' I s i i <
* \
U f\
-_-\-_I\
~ . } }
" t
/D b lad.0 '
V E i 0 o \ l )
* \
\ \ V \ \ \ (\ \
\ , \ \ \ \\ \
N '4 lI \
\
l s
/
= a=-
O.0
- 09 10EU.0 31D.0 3m.0 cv.0 TIrt (SECONDS)
S2D CASE 31 FM 1 SPRAY BURN 100 PCT AT 10 U 0 EPS T+34S0 CASE 3
- 'v
\h
D.
\
N,/
$ 'W I
en a 27.4 I EE.0 '
'., i l L 5.0 5 '
i v. 1 i 1 I, i I 1 u iM.0 [1 - - i ; e I I L 1 1 i , g " , g 23.0 ,
,' 1 1
- n. i
~ i
{s O
'y 9
/
> 1.
w/ ' ' C 2.0 ' cc Q. I
.d_ I I o
C.0 , M I y-Y I I I
~I s.O '
I I
! I 1
19.0 I E 18.0 M .0 4W0 0.0 10W.0 BD3.0 . TIf1E (SECONDS) CASE 3 S2D CASE 31 FAN 1 SPRAY BURN 100 PCT AT 10 V e FPS T+34SO f . Iwj
r ai t
%/
.' S i
D.0 I ; I I I 1 YI I I l E h'4 I I E I 1 l 1 g I 3 g I I r E I 7 4 I I i I r i 1 a I 1 1 33
,_ , i =
y 1 I I I Y I u 24.0 ,
; l' 1 1 I I I i 1 E I I I I 1 i I I I I l 7
y 1 I E e ' G D0 E
~
y r ] g I I I 1 s -H l A I , ' '
' f O E.0 I
l E --
/i o I g i / 1 i
1 . e' - c;h o
/ --M' y I i 1 i i Y I 3
y I 3,0 I ' ' _I 7 I I I I J I 19.0 ft 1
-+
+ -
44 I t
. I e.0 1020.0 M0 M*0 TIriE (SECONDS)
CASE 3
$2D CASE 31 FfW 1 SP9AY BLRN 100 PCT AT 10 0 0 (IPS T+34SO
()
FN, d W 3Le
'pWimmW k t s1 '
\s \ \ {t ( \ \ h t
, L \ \ \
.N
\s \
x B / .,
/ N N
/
E /
/
, s b/ w m.0 '
h ) O '
/
E d 5 m-I I te.e y,g e ,o l e,e t G0 e T]rE (SECONDSI CAK3 ( S2D CASE 31 FAN 1 SPRoy BURN 100 PCT AT 10 V e W T+3450 i [ l l') t 1
%./
36 ) 4 l l
O (v) ! a 27.4 I I I 1 i T [
- M,
. 1 I I I
y L.
?, ,
I II I I E I
, . .. 1 1 I I
. t 1 y 1 1 I N I I E E
m 23.0 1 1 1 ' 1 '
- o. 1 1 v) ;&
1 C L. _'
\
, ; 1 T0 J O 1 1 "hf 1 -
- n. 1
__s.
- I O .J
.r :
ag,e
- i 1 m> ,
_J k E I I 3.0 --- I E 29.0 f 1 I T I E ggg,e aca.o 393.0 60 g,g TIK (SECONDSI S2D CASCI 1 FM 1 SPRW BLRN 100 PCT AT to V 9 GFPS T+3M i I I a/ I
i v
)
a Z
*B20.0 I
I 3 ame c l (O _,/ $ Ei Jl 5 n l %. / 130.0 I 's; /
/
, s #
, /
/,
J p - '
/
0.0 Imc 4ni,o 00 1000.0 &W.0 7IPE iSECONDS) CASE 4 S2D CASE 4 e FAN 1 SPRAY BLRN 100 PCT AT 10 V 0 G PS T+34SO Om.
()
't ) :
I: 3K9.0 o 310.0 C n s-
;a E
t 6_ 1a2.0 , m
~~ . ,
I II I ! ! I 0.0 e.0 lae.e axe.e arc.o ue.0 TIriE (SECONDS) S2D CASE 4 0 FAN 1 SPRAY BURN 100 PCT AT 10 U 0 6 FPS T+3450 CASE 4 l 1 1 O o l l Yi 5$
f V = m 9 i i 11C0.0 ' I. 10M.0 1 92.0 ,
..i i
4 KO.0 3 re.0 C i ] GV.0 bs s a e sc.e , f I 40.0 ; i
~00.0 t
x s 3U.0 x N 13).O O.0 1W.0 4W 0 1 630.0 aTD.0 0.0 TIf1E tSEcor4DS1 CASE 4 S2D CASE 4 0 FAN 1 SPRAY BlFN 100 PCT AT 10 V O 6 FPS T+3480 ()~s L. Ek 1
,n I'u) anas.e 8 \
\
c t h / V Y J 1a'0.0 \ j
\ s_ /
Y e
/
, /
, / -
/
/
/
/
l e.0 tesa.0 a200.0 310.0 4KD 0 T!rE (SEcorcS) S2D CASC4 0 F#1 1 SF94Y BLRN 100 PCT AT 10 U 0 GFPS T+34S0 I I l l bb l l
(D
%)
num j
%'* l i
,1) 1 II I i
_ _ 4 } l \
!I i \
4.0 1 Ji it t P \ it \
'd \
li p i r ( I E ! \ O $ l' _ V T.0 y . _-- - s T d _- ' y / 1 s
- -s d ./l \
E s '
" t ,
t 7 y
/
m.0 j A\v.0 10.e iDU.0 m v.o e.0 !aV.0 TIME (SECONDS) CASE 4 S2D CASE 4 0 FAri 1 SPRAY BLRft 100 PCT AT to V 0 UPS T+34SO f%
,)
wQ 6/
/%
U so.e I I I I I I I I I I 1
! 1 1
I I I I I I I I I I I I I I f I I 1 I I I I i I I 1 1 I 1 I I I
! I ] I I I I 1 I I 1 1 T I 1
i I .9 I I I I I f I I T I 1 I I I I 1 I I . 1
" I .
1 I 1 I t I i g
]
1 I I 1 .I i I 1 . 1
] I y I I I I I
1 I
@.0 l
.I I I i I I I I I I. I. I ,,,
C I 1 . i I i i p I 1 "U 1 I 1 J' *+ I I a. \s) 1
. L #.i 1
- 1 i i
' ' I .' ;
(D.0 1 1 . A _ f 1 k 1 I . I I I 1 I I T i i - I . r,. 1 g 1 I I 1 _ f I , , y ! I 1 I I 1 . g 1 I 1 Ii I - 1 I g . I . I g, 1 I L h I ' 7 1 I i N @.0 2 ' ' ' i E
. lt I . i
.- 7I!
I i 1 i 1 I _ I _I I _.Il 1 I I
. I I .L Y '
i i ! i i i ',C i I I I i . I . . li I
-T' T.0
- LF l
. __4 -
i I
.j_
I .L d I 1 _I_- -
, P
]f ! I I
_C? " L _, /" _~ 1 1 I 1 g [I I 1 w I I 1 I I I c 1
-+
4 1
~
1 I ii i I I I 1 i I I
~[
l[
-*~
se.e <m0 e.o iae.o ame aw.o TInc (SEC0t4DS) CASE-t 52D CASE 4 0 FFd 1 SPRAY BtFN 100 PCT AT 10 0 0 ffP5 T+3450 i l l l l i sw s> 1 1
\ ,
i v , I 1 12.0 I I I I I I I I I I I I I I I I e 1 I I 1 , k I I W.0 I I I Y I I I I I g m.0 , , I I I I I I 2 I I I F I I Y
! I 1 I I I I I I I I I 1
- I 1 I I I Og
%g 1
I I I I 1 7
.f 1
1 I I
' l O
C t.D.0 -i . - , 1 I I I I i L f I I I I I I I I , O ' CO 1 l l I E 1 I 1 I 1 I I I 1
& I 1 1 I
I I I I I I I
- j ,
1 I I I i I
- j,-
L.-- P, - I I I I I. ,-is -- 3
'dF' I I I 'l I
I
- I I .'. 1 I I E
I I . CmM "i.a7' I
' I I I I
e ' ' ' ' l I 3.0 +
. , I I I
^
^
I I j b g g I I I 1 10.0 cTN.0 3W.0 CEO 0.0 13D.0 TIr1E (SECONDS) CA$E4 52D CASE 4 0 Ffti 1 SPRAY BtRN 100 PCT AT 10 0 0 GTPS T*3430
- A
(~% L' p mm M i i i
\
1 L
\
\
* \
1 i I \ g ! \ 4 \ i 1 E (
\
i $ *
\
%) ~ w 31.0 s " Q ' y ~' \\
..s y ' \
' e
\
d / N l .. E '
/ l N
I f
/
35.0 [
.-e
-m, 18.0 'DD.0 G0 e.0 1000.0 HKr.0 7IFE (SECONDS 1 S2D CASE 4 0 Fftt 1 SPRAY BLRN 100 POT AT 10 V 0 6 FPS T+34W h
(o
Th. Cl a m e 3 eke ffl ._ l I 3 3 am.e ! . i i I l l k E ' I l i l O v g 1 1 i i I 1 5 l I l i (
- hk
\ '
\
'\
! 11 i (
) 4 \ \
I\ \ \ \ l \ \ 1GRO J !! \ ) \ i \
\ li ( \ 19 x
T _ { l \ i i
\i m.
--_f" e.e Im.0 uno
- N REO.0 TIPE (SECONDS)
-!CE S2D CASES 2 FAN 1 SPRAY BLJN 100 PCT AT 10 V 0 6 FPS T+3480 O
L) Oc7 ' .Y
- - - - ' - ~ -
a , h i ; I f i
" ~~~ .
i \ \ \
= \ \ \ \ \
n
" i~
g e
\
i \
\ ;
\ 1
\
\
\
s {\ \ \ \ s > G
~
&O --
3W.0 g,g TIME (SECONDS) 5 2 FM i speny gtpH 100 PCT AT 10 U e m pm q a i L
f% k v
)
I j I y ane Lx s \ x f s 6 lr \ \ s xs y .'
- a E w y Y U __-
IfD.0
*A t
'# 1812.0 gj TIT (SEC0PfDS) 2 F M 1 SP % y m ISO PCT AT 19 U 9 6 FPS T+3480 .gg Oi
(_) q1cQ 47 l
O i l l
/ t 8 u r 1
\
\
6 h I\ \
' \ l \ \ \ \
O 120.0 3 k
\ \ \ \
I
* \
\
\ \
\ \
n ( u
,/
M g N BEELO 330,8 TIPE (SECONDS) 52D CASES 2 F#v 3 SpRM SL5tN 100 PCT AT 19 U 9 diFPS T+3480 -ICE' O
O U SLe 1 6 4Le i
! I I
\ \ sl
\ 1 g
\ \ i i
\ h \
E N i-> ! /~% M \ \ (~) S
' \ \ (
V i ( y 34 j \ T R i \ \ 0 1
\ J E
- d 5
~
)
/
/
~- /
- 5,Q _
1 ( e 8.9 HEO.0 h.0 3EP.0 (f0.0 TIPE (SECONDS) 5 2 Fm 3 SppAY BURN lee pc7 AT 10 0 0 gps y,pg , (^) v
SILO I 4.0 i
\ \ k 1:f __.._
I \ 1)
\ l
\
\ \ i T \ \ i (3
E E N s x s
\ <
3
\
i
\j 2.0
# \' ' ' h v
k i %
$ t i N 'N w
a. s w 5 /
/
../
E.0 _ ( 10.0 A0 3eleA 319.0 3se.e 4xe 0 TIPE (SECONDS) SED CASES 2 FM 1 SPRAY DLRN 100 PCT AT 10 V e 6 FPS T+3480 -ICE O_
gx s )
._.J 50.0 .
e i i e.0 1' l ll '
! 6 i i i 3 \ i t fl ti \J V \ 11 1
\ \ i \. \ 1 i N \ip i ij
'i i \ it i k e i
.T.0 i N ti ii li l'
\ i N !
L 0 cc \ ll i i
' v l d
- e i
t
/
B / j
/
./l l cT.0 ---"' ' I I i e p ee i
/ l i
10.0 00 1ED.0 aYao.O m,o ~ ~ h,g T!rE (SEcorfDS) S2D CASES 2 FM 1 SPRAY BLF11 100 PCT AT 10 V O 6 FPS T+3480 -ICE rm L,
5BA 4.e l l I l i d \ \ \
\ ( \
\ \ 't !
\ T \ \
e N \ _ \ (
~B.O V \ s
' N T i N N e t d
a.
\
d
/
y _-- - ' -/ ( l i 1 t 28.e u agite gog o 4 33,, l seesA TIPE (SCCONDS) S2D M 2 FM 1 SPRAY BURN 100 PCT AT 19 U 0 GFPS T+3480 -ICE sq u
Oh (_)
=,
game G l r s '
- f. .
i y I $ I; W \ li I (
/
lae.e 1 i l I fl 1 \ \ ! L < \ \ i\ ---- i ( \ U \ \ \ \ '
\ \ t i \ \ \ \ \
y y f
'J
,/
he 3m0 g ,p , f I TIPE (SEcCros s 2 M6 2 Fm 1 sppey BURN 100 PCT AT 10 U e ErPS T+34So sgm f; i v( 19 i l l l
rh ki
'a
;uuan a M
G
- 16 9.0 E
sa i
' k%I + kk._fw_J -.
is_ e.e MA i55 0.0 3XV.0 60 TIPE (SECONDS)
- 2 N 2 FM 1 SPRAY BWN 100 PCT AT 10 U 0 GrPS T+3480 DRAINT I
1 , r's v)
?
) W
,<~ ,
4 I
\ /
m,. 298.0 148 3 E T U - me 5 l_ y -
-~--_
~_.
k-
~
e' h, , i _%"' j \/ j r/ ft-.,- 1, A M ? I I I f I 1m.0 98.0 N N 3XD.0 er.o TIRE (SEC0rcS) M N 2 Fm 1 SPRW BLRN 100 PCT AT 10 U 0 GFPS T+3490 DRAINT bi ) v 74 i
8 w' jag.g I I ji I i i i , l i .i
$EB.O P s .! =
' )
i i i I .11 1 f i 'i l, MA \ \ I_I I I. t Ii i i i
.i l I , 1 d I '
I!! l !' H "AB.O l .: i i ! k .' 1i 1. ii 1 1 I 1. 1 i 1, ; y 5 i' u I 1 i
' I i 1 1
g I L EB.O i, 1 l (
>= . I 1 \
4D.0 \ 3 m& s
/
/
J
/
HB.O j
/
I I 20 i 6.8 3xe.e ega,e M 38E9.0 ERe.e TIPE (SECONDS) 52D CASE 6 2 FAN 1 m BLRN 100 PCT AT 16 U 9 GFPS T+3480 DfMINT s-*
1 I l 1 l 1 I 1 l l l 1 l I l 27.4 l I l l I I I I I l l l I l iLe i E u EH.0 '
.a I I I
, 1, I -1 _ ~ y e Y I
.C. m g , ,i Ib ,
=
1 I 1 I I I h I 1 A E I I M , e ' 2E.0 ' M
- o. 1 .
I
,_ 1 1
-: I ,
' ' ~
g l . 1 N . 1 I O E 2.0 - I I 1, 1 I J' I
- m. -
k I I '. ' I .- 1 I f : I 2.0 . , Z I I I y _- - I A I I I I I I I I I I I I I I I I I r I I 1 I 1 i . I I 1 I I I l s8.0 1 e.e Make iges.e axe.0 4me j i TITE (SECONDSI
$2D CASE 6 2 F#v i SPRAY DIEN 100 PCT AT 10 V e GFPS T+34Se DRAINT l
1 77
o N, 3Le i
! \ \ I, l u t \ \ ("ll i { 4
\\ ;
OO j ae -
/ \
s s
\ ,
s l\ y t-p M t 5 Idle n SeBB.O BDO.e ~ 400.0
)
0 E YM 1 SPpsy y e . ~ 1.,. , , , )* wu9w
+ ) 3.8 k i ( , 8 ( \\ i ( TA
\ i
( i
)
# \ 3
( (
/
cr
\ \ \ ,
h f sM \ D) l
'./ ,_---
/
f m.e m l 5 f d e
~
MLO
" =0 m., ,,
TIfE (SECONDS) 2 FM 1 my W 180 PCT Af 16 U 0 6 FPS T+3e pgg O \
\ 'w) 79
n' %s 27A '
!++
I I
-E Ee .
E
> I I
5.0 - I I I I I g B4.8 I
=
I L i 7 1 I K I I I I E I 1 I ae
- I m , i I I O
i
- a. 1 ,
= I i - ii t u i I g I I I I [ [
m . e y i _
~ '
M n. 1 i 1 I I I 1 . . I A I E I E,
.- 1 1
_ I I l 21.0 , l 1 I 1 I 1
# I I I I J g
I I . . Y T I I I I 1 I kJ
' 1 j ,
T^^p I I ISLO I I I I I l l i 28.8 s.e Isme mes.e mes.o mes.e i TIPE (SECONDS) S2D CASEG 2 F#t 1 SPRAY BLRN 100 PCT AT 19 U 9 GFPS T+34Se DRAINT O ( ig TO
7s V e
~
I
" U 8
A g
~
$p V \
\ .
*3
\ 1 I i l N
, i i m i
a unwwu a g 4 r (_) MsyTI
4 O v i 9 R
-_ y f.
u ia e, k h
- - o
~~~' '"" - ,-m
.., q ;g,-
Nj nC7 2, M
*L I ;: 3 U .
1 1 1 E u fU s E I i
- 8 m
v a g i a g m aumuu 5 i r , b,)
~.
l l
T v y - - m a 6
/ l [
i m b I t 3
=
4 $* 1 oa f, W* g f~ h3 1 I I u N
\
I i i i i : 8 x w m-u 6 1 W l O 1 V ! l
. i
o
- A
- ~
I: 6 s 2 A
-- k k
=== ,
~
;> _a fN -
m uu N 6-g A N E3 V s .r 4 I
~
! Nm i i i i i a i a i i a E
x a man qW
O V e W 7 i $ 2 A
, C
-- l k
-ac:5 _
g$. 1+ 2 m- gg4 e
-s V s s
$3 X s i I 8
1 m ( ' h N s
* %m 9 Q n en msw wi I 9
r ,/~N G dkW
13
- V 2
I b .- A i.
--C .
-- g_ a
-g.. "
_-4 8% x '
$e3 s
O* . 1 l I i i g I h I I l h s h
~ ~_
q D ,
- 9
- 9 e e 9 + Q ,
N D d 5 d Kl d b t l l 31 (BISdl 3fC5i d T101 E 1 I E 2 W l\ v
O O 1 W 7
$ Y i
i
, l, 2
_ _:5-
-=C
=
L M
~
i
'N $g
/'l (J \
'3
\ i I
_ i 1
~
c.
\
I $ $ ] 0 x mso msmuu i r
.. ._ 48'7
P 4 R
= # 7 h y a
A
. A
_j- k h __ c e
--= g :>
WM
.;/-
t 1 - -- W
-- -E O x r s
.s
\ "
\
- l t % 0 2 3 E 2 2 i E @ E 2 8 8 0 30 ftf! Sal 3dnSSM %01 t s
2 W O edy18
og#6 46 O _a ! ! g a f
- t 8
A A hG g
~ ;;; l
[g O x
!ga x s s x R =
T
<=
id n ,, --21 3 r edig 79
V
==
=
, a Y
8 A E
, k
_= p+ 3
"~""10:1n'. ann:;--- .
g s, :E
,E (J i i
Ma l
,E N
i i i i i i i i 8 8 31 13 3Wnitsw11 s 5 2 In)
's_/
')
A N
~
?
- I 3
~
A
----- I. . E 1s 5- , ,oe p
/
' Eg !
p '
.3 ,
N i
,s -
')<
5 fu ' x I i i i i 3 3 I I g I 12 3611W34G1 2 W l j r's l l C) l 4 9/
fx W 5 I 8 A.
,,[, 50f L-W "
-, +.: ,
f r, p<,. <d" I, D
; f, a
~
gg g t,.g ()3 -
..s a
O 6 1 u N
% ~-_
i i i i i i i i g s' a x ,, man a 4 2 e n U . k --
e e O f% \ l l
- A W
i I 6
,W" 1m
-2: o
/ 80
--~,pe >
-~~
,s-
' s y@3 s *
\ S s
k g ( s a s
* %m 9 U n mso msw w1 3 r
93 y - - - -
O
- i
= ,
- , I_
8 A
. A
'I, k h x 4 ,
s
=
2 > 7 d -, k O N !a i I
~
\
s, I l m
\
\ Q A g I
I I I @ 3: mu msm wm . 8 r
$ 9" -
O _i I: I 3 A l,' h _2 T
/ >
4 y O - N . s
\ a I
\
(
- N s
X 2 I O 3R 101541 Jun553Wd 1101 E i i o l L) l i
y ** O i a I k
'[ k k z '
a 09 O _---- W-
~
C
^x !gs s
5 s I i ' t
- q co m ,
30 tu! Sal 3dn553WJ 1H101 8 e l i t O. 1 4%
I l OV
~
5 I*
/ ,
g g T-TE ^$ i ku o 9
\ !! S
\ .-
\ g
\,
N i v i w uTwwu i a 5 r . G .. edug97
g b 1 5 k e
' 9
.u .--....-.. ---. - -- ~
l ,
,,,r" o.
.,f-_]_
\
- - 4 g
N i i i i i i i i 8 8 31 tal 3Wn1W h 11 s i W 3 (G - d
v l i I: k
- , e i
k k i g-s
> Bt f W-O ~
ii i] . E i
\ "
\
I l l l 3 0 x w mwwu i 1 r O ..
. W'
O 1 1 2 A
. 5
__;m;E ._.._ 1 i w_ c _=_ -1
<=
is w o __ -
\ x'~
NX i
\
a. I N m ,, i i i i i i i i i i i 8 8 x ,,mawn k n NJ .. odQ 100
O I I: Ia aM
,3 n $ee 1 1- ge
-,c - '~ gf n i -- - -v V r '-
s g3 s . m
\
t ( ' 8 n T
,y
% %- w I E E 2 E E @ 2 2 8 3 R
- n wu msw mu i a
4 O .. edig tot
. . _ _ _ _ _ _ . _ - - ~ _ _ . - . _ _ _ _ _ _ _ . _ . _ _ _ _ _ - _ _ . _._ __ _. __ . - _ - _ - . ._ _ _ _ _ . _ - - . . _ _ _ __ ._.. . _ - . _ _.
I i. ! i i l Figure 102 is not included herein, but will i be supplied at a later time. i 6
~
1 l I l l I I l A l
- -.-----.-,.,,,---,--_v, -
,wn,-,.., ---e.-n-,n. .,n,-,-me.,.-.~-,,,
O l I I l S I k
. c
{ S o ( h*
--y 8 'i L. 6 9 -
i ., s . i
\ 1 l
s. N i \ i x , h x wu msm wi E r ? i l0
O b Y
- 6
~~.
. a Ig
- k ST ww _.
o s g
,,, _- - .g q, x '
\
c
~%~
N
\ ,
\ in b
I
~
) "
\ -
fu
.k __
- 3 2 g a
= ma, wom,,, &
i 2 W ,q 's/ .. 6:fdg/0hl
F n v G k
- a
/
--_ a 3- se c
g$
\ $
s .3
\ t I i I
m N l i i s' a ! m i.e awnwan f , l i}}