ML20039D700

Station Blackout at Browns Ferry Unit One - Accident Sequence Analysis

ML20039D700
Person / Time
Site:	Browns Ferry
Issue date:	11/30/1981
From:	Cook D, Greene S, Harrington R, Hodge S, Yue D OAK RIDGE NATIONAL LABORATORY
To:	NRC OFFICE OF NUCLEAR REGULATORY RESEARCH (RES)
References
CON-FIN-B-0452, CON-FIN-B-452 NUREG-CR-2182, NUREG-CR-2182-V01, NUREG-CR-2182-V1, ORNL-NUREG-TM-4, NUDOCS 8201060024
Download: ML20039D700 (225)

Text

,

NU R EG/CR-2182 -

ORNL/NUREG/TM-455/V1 e

e J

Station Blackout at Browns UNION Ferry Unit One-Accident CARBP.DE Sequence Analysis D.H. Cook S. R. Greer.n R. M. Harrington S.A.Hodge D.D.Yue t

Prepared for the U.S. Nuclear Regulatory Commission Office of Nuclear Regulatory Research Under Interagency Agreements DOE 40-551-75 and 40-552-75

' 4 Lf) b O Yf L DEc2 C

[ "'%/,yw1981u [,

~

s 8j 8201060024 811130" DR ADOCK 05000 gQ f ,

'/

-. . . . )

r ,

e t

Printed in the United States of Amenca Available from National. Technical Information Service U.S. Department of Commerce 5285 Port Royal Road, Spnngfield, Virginia 22161 Avaiiable from GPO Sales Program Division of Technical information and Document Control U.S. Nuclear Regulatory Commission Washington, D.C. 20555 This report was prepared as an account of work sponsored by an agency of the g

Unn3d States Government. Neither the U nited S tates Government nor any agency thereof, not any of their employees, makes any warranty express or emphed, or assumes any legal liability or responsibihty for the accuracy. completeness, or usefulness of any information. apparatus, product, or process disclosed, or represents that its use would not en fnnge prrvately owned nghis. Reference herein to any specific commercial product, process, or service by trade name, trademark, manufacturer, or otherwise, does not necessanly constitute or imply its endorsement. recommendation, or favonng by the United States Government or any agency thereof The views and opinions of authors expressed herein do not necessarily state or reflect those of th s United States Govemment or any agency thereof.

d

NUREG/CR-2182, Vol. I ORNL/NUREG/TM-455/VI Dist. Category RX, 15 o Contract No. W-7405-eng-26

/

Engineering Technology Division STATION BLACK 0UT AT BROWNS FERRY UNIT ONE -

ACCIDENT SEQUENCE ANALYSIS D. H. Cook S. R. Greene R. M. Harrington S. A. Hodge D. D. Yue Manuscript Completed -- October 6,1981 Date Published - November 1981 Notice: This document contain., information of a preliminary nature. It is subject co revici<:n or correction and therefore does not represent a final report.

Prepared for the U. S. Nuclear Regulatory Commission Office of Nuclear Regulatory Research Under Interagency Agreements DOE 40-551-75 and 40-552-7 5 NRC FIN No. B0452 o Prepared by the OAK RIDGE NATIONAL LABORATORY Cak Ridge, Tennessee 37830

, operated by UNION CARBIDE CORPORATION for the DEPARTMENT OF ENERGY l

+

111 CONTENTS Page

SUMMARY

........................................................ y ABSTRACT ....................................................... 1-

1. INTRODUCTION ............................................... 1

2. DESCRIPTION OF STATION BLACKOUT. ............................ 5 j 3. NORMAL RECOVERY ............................................ 7

4. COMPUTER MODEL FOR SYSTEM BEHAVIOR PRIOR TO LOSS OF INJECTION CAPABILITY ....................................... 17

5. INSTRUMENTATION AVAILABLE DURING STATION BLACKOUT AND-NORMAL RECOVERY ............................................ 30

6. OPERATOR ACTIONS DURING STATION BLACK 0UT AND NORMAL RECOVERY ................................................... 33

7. COMPUTER PREDICTION OF THERMAL-HYDRAULIC PARAMETERS FOR NORMAL RECOVERY ............................................ 36 7.1 Introduction .......................................... 36 7.2 Conclusions ...............'............................ 36 7.2.1 Normal recovery -- conclusions .................. 36 7.2.2 Loss of 250 vde -- conclusions .................. 37 7.3 Normal Recovery ....................................... 37 7.3.1 Normal recovery - assumptions .................. 37 7.3.2 Normal recovery - results ...................... 38

. 7.4 Loss of 250 vde Batteries -............................. 50 7.4.1 Loss of 250 vde batteries -- assumptions ........ 50 7.4.2 Loss of 250 vde batteries - results ............ 50 e 8. FAILURES LEADING TO A SEVERE ACCIDENT ...................... 59 8.1 Induced Failure of the HPCI and RCIC Systems .......... 62 8.2 Stuck-Open Relief Valve ............................... 65 8.3 Loss of 250-Volt DC Power ............................. 68

9. ACCIDENT SEQUENCES RESULTING IN CORE MELTDOWN .............. 71 9.1 Introduction .......................................... 71 9.2 Accident Phenomenology ................................ 71 9.2.1 Accident progression resulting in core melt ............................................ 71 9.2.2 containment failure modes ....................... 72

9. 4 Event Trees ........................................... 74 9.3.1 Event trees for accident sequences resultin'g in core melt .................................... 74 9.3.2 Core damage event tree .......................... 74 9.3.3 Containment event tree .......................... 74 9.4 Accident Sequences .................................... 78 9.4.1 MARCH computer code ............................. 78 9.4.2 Accident progression signatures ................. 78 8 9.5 Containment Responses ......................... ........ 123 9.5.1 Drywell responses ............................... l?).

• j 9.5.2 Wetwell responses ............................... 149

10. PLANT STATE RECOGNITION AND OPERATOR MITIGATING ACTION ...................................................... 161 10.1 Introduction .......................................... 161 10.2 Plant State Recognition ............................... 161

- -.__-- - - _ _ _ _ _ _ _ _ - _ _ _ ~

iv Page 10.3 Operator Key Ac tion Event Tree . . . . . . . . . . . . . . . . . . . . . . . . 163 10.4 Operator Mitigating Actions ........................... 163

11. INSTRUMENTATION AVAILABLE FOLLOWING LOSS OF 250 VOLT DC POWER . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

165

12. IMPLICATIONS OF RESULTS ..................................... 167 ,

12.1 Instrumentation ....................................... 167 12.2 Operator Preparedness ................................. 170 12.3 System Design ......................................... 172- -%

13. REFERENCES .................................................

173 ACKNOWLEDGEHENT ................................................. 177 ,

APPENDIX A. Computer Code used for Normal Recovery ............. 179 APPENDIX B. Modifications to the March Code .................... 197 APPENDIX C. March Code Input (TB') .............................. 199 APPENDIX D. Pressure Suppression Pool Model .................... 205 D.1 Introduction .................................. 205 D.2 Purpose and Scope ............................. 205 D.3 Description of~the System ..................... 206 D.4 Identification of the Phenomena ............... 207 D.5 Pool Modeling Considerations .................. 211 APPENDIX E. A Compendium of Information Concerning the Browns Ferry Unit 1 High-Pressure Cooling Injection System .......s . . . . . . . . . . . . . . . . . . . . . . . . . . 217 E.1. Purpose ........... .......................... 217 E.2 System Description ............................ 217

• E.3 HPCI Pump Suction ............................. 2 *.9 E.4 System Initiation ............................. 210 E.5 Turbine Trips ................................. -221

• E.6 System Isolation .............................. 221 E.7 Technical Specifications ...................... 222 APPENDIX F. A Compendium of Information Concerning the Unit 1 Reactor C6re' Isolation Cooling System ....... 223 F.1 Purpose ....................................... 223 F.2 System Description ............................ 223 F.3 RCIC Pump Suction ............................. 225 F.4 System Initiation ............................. 225 F.5 Turbine Trips ................................. 226 F.6 System Isolation .............................. 227 F.7 Technical Specifications ...................... 228 APPENDIX G. Ef fect of TVA-Estimated Seven Hour Ba ttery Life on Normal Recovery Calculations ...............

229 a

e

v

SUMMARY

This study describes the predicted response of Unit I at the Browns Ferry Nuclear Plant to a hypothetical Station Blackout. This accident would be initiated by a loss of of fsite power concurrent with a failure of all eight of the oasite diesel generators to start and load; the only re-maining electrical power at this three-unit plant would be that derived from the station batteries. It is assumed that the Star - Blackout occurs at a time when each of the Browns Ferry units is oper .ing at 100% power so that there is no opportunity for use of the batteries for units 2 or 3 in support of unit 1.

The design basis for the 250 volt DC battery system at Browns Ferry provides that any two of the three unit batteries can supply the electri-cal power necessary for shutdown and cooldown of all three units for a period of 30 minutes with a design basis accident at any one of the units.

It is further provided that the system voltage at the end of the 30 minute period will not be less than 210 volts, and that all DC equipment supplied by this system must be operable at potentials as 1:a as 200 volts.

It is clear that the 250 volt system was not designed for the case of a prolonged Station Blackout, and the period of time during which the DC equipment powered by this system could remain operational under these con-ditions can only be estimated. It would certainly be significantly longer than 30 minutes since all three batteries would be available, the equip-ment is certified to be operable at 200 volts, and there would be no de-sign basis accident. In response to AEC inquiry in 1971, during the pe-riod of plant construction, TVA estimated that the steam-driven High Pres-sure Coolant Injection (HPCI) and Reactor Core Isolation Cooling (RCIC) systems, which use DC power for turbine control and valve operation, could remain operational for a period of four to six hours. A period of four hours has been assumed for this study.*

Within 30 seconds following the inception of a Station Blackout, the reactor would have scrammed and the reactor vessel would be isolated be-hind the closed main steam isolation valves (MSIV's). The initial phase of the Station Blackout extends from the time of reactor vessel isolation until the time at which the 250 volt DC system fails due to battery ex-haustion. During this period, the operator would maintain reactor vessel water level in the normal operating range by intermittent operation of the RCIC system, with the HPCI system available as a backup. Each of these water-injection systems is normally aligned to pump water from the conden-sate storage tank into the reactor vessel via a feedwater line.

• It should be noted that the events subeequent to failure of the 250 volt DC system are relatively insensitive to the time at which this fail-ute occurs. This is because the only parameter affecting the subsequent events which is dependent upon the time of failure is the slowly-varying decay heat.

As part of a requested review of the results of this study, TVA per-formed a battery capacity calculation which shows that the unit batteries can be expected to last as long as seven hours under blackout conditions.

In Appendix G, the effect of a battery lifetime of seven vice four hours upon study results is shown to be limited to timing.

vi The operator would also take action during the initial phase to con-trol reactor vessel pressure by means of remote-manual operation of the primary relief valves; suf ficient stored control air would remain avail-able to permit the desired remote-manual valve operations for well over four hours. The Control Room instrumentation necessary to monitor reactor vessel level and pressure and for operation of the RCIC and HPCI systems would also remain available during this period. .

There is no Emergency Operating Instruction for a Station Blackout at Browns Ferry. However, the existing written procedure for operator action following reactor isolation behind closed MSIV's provides for the reactor .

vessel level control and pressure control described above. The primary relief valves would actuate automatically to prevent vessel over pressuri-zation if the operator did not act; the purpose of pressure control by recote manual operation is to reduce the total number of valve actuations by means of an increased pressure reduction per valve operation and to parmit the steam entering the pressure suppression pool to be passed by different relief valves in succession. This provides a more even spacial distribution of the transferred energy around the circumf erence of the pressure suppression pool.

The initial phase of a Station Blackout has been analyted in this study by use of a relatively simple computer code developed specifically for this purpose. This coding uses the Continuous Systems Modeling Pro-gram (CSMP) language of the IBM computer system to simulate the response of Browns Ferry Unit 1 to postulated operator actions. The analysis shows, because of the loss of the drywell coolers, t hat it ir necessary for the operator to begin to reduce the reactor vessel pressure to about

• 0.791 MPa (100 psig) within one hour of the inception of the Station Blackout. This depressurization reduces the temperature of the saturated fluid within the reactor vessel and thereby decreases the driving poten-tial for heat transfer into the drywell, yet keeps the vessel pressure high enough for continued operation of the RCIC system steam turbine.

With this action, the drywell average ambient temperature can be kept be-low 149*C (300*F) throughout the initial phase of a Station Blackout;.

tests have shown that both the drywell structure and the equipment located therein can be expected to survive temperatures of this magnitude.

The analysis also reveals an important second reason for operator ac-tion to depressurize the reactor vesel early in the initial phase of a Station Blackout. This depressurization removes a great deal of steam and the associated stored energy from the reactor vessel at a time when the RCIC system is available to inject replacement water from the condensate storage tank and thereby maintain the reactar vessel level. Subsequently, when water injection capability is lost for any reason, remote-manual re-lief valve operation would be terminated and there would be no further water loss from the reactor vessel until the pressure has been restored to the setpoint [7.72 MPa (1105 psig)] for automatic relief valve actuation.

Because of the large amount of water to be reheated and the reduced level ,

of decay heat, this rapressurization would require a significant period of time. In addition, the subsequent boiloff* would begin from a very high

• The term "boilof f" is used to signify a monotonic decrease in re-actor vessel water level due to intermittent loss of fluid through the primary relief valves without replacement.

vii vessel level because of the increase in the specific volume of the water as it is heated and repressurized. Thus, an early depressurization will provide a significant period of valuable additional time for preparative and possible corrective action before core uncovery af ter injection capae bility is lost.

One design feature of the HPCI system logic was brought into question

, by the analysis. The existing logic provides for the suction of the HPCI system pump to be automatically shifted from the condensate storage tank to the pressure suppression pool upon high sensed suppression pool level.

4 During a Station Blackout, this would ocpur af ter about three hours when the average suppression pool temperature has reached about 71*C (160'F).*

Since the lubricating oil for the HPC1 turbine is cooled by the water be-ing pumped, this would threaten the viability of the HPCI system.

The rationale for the automatic shif t in HPCI pump suction on high sensed pool level is not explained in the literature available to this study. There is no corresponding provision for a shif t of the RCIC pump suction on high pool level, and a separate logic is provided for an auto-matic shif t of the HPC1 pump suction should the supply of water from the condensate storage tank become exhausted. For these reasons, it is recom-mended that the desirabilty of the automatic shift of HPCI pump suction on high sensed suppression pool level be reexamined.

The plant response during the initial phase of a Station Blackout can be summarized as an open cycle. Water would be pumped from the condensate storage tank into the reactor vessel by the ACIC system as necessary to maintain level in the normal operating range. The injected water would be

• heated by the reactor decay heat and subsequently passed to the pressure suppression pool as steam when the operator remote-manually opens the re-lief valves as necessary to maintain the desired reactor vessel pressure.

• Stable reactor vessel level and pressure control is maintained during this period, but the condensate storage tank is being depleted and both the level and temperature of the pressure suppression pool are increasing.

However, without question, the limiting factor for continued removal of decay heat and the prevention of core uncovery is the available of DC power.

The sequence of events used for the fission product release analysis for a prolonged Station Blackout was established by this study under the assumption that no independent secondary equipment failures would occur.

Dependent secondary failures, i.e. , those caused by the conditions of a Station Blackout, were included in the development of the sequence, but with the assumption that the operator does take action to depressurize the reactor vessel and thereby prevent drywell temperatures of the magnitude that could severely damage the equipment therein. The point is important because the operator would probably be reluctant to depressurize; current training stresses concern for high suppression pool temperatures (based on LOCA considerationst) and the operator would recognize that suppression

, pool cooling is not available during a Station Blackout.

• It is expected that any accident sequence resulting in high sup-pression pool level would also produce an associated high pool tempera-ture.

tThere is currently no written procedure for the case of Station Blackout.

i

viii This study has established that the possible dependent secondary failure of a stuck-open relief valve would not reduce the reactor vessel pressure below that necessary for operation of the RCIC steam turbine dur-ing the period in which DC power would remain available. However, during this period all of the energy transferred to the pressure suppression pool would pass through the tailpipe of this one relief valve and would be con-centrated near its terminus. This would reduce the steam quenching capac- .

ity of the pool to that provided by the ef fective volume of water sur-rounding this one tailpipe. local boiling in this area would pass most of the relief valve discharge directly to the pool surface and could result .

in early containment failure by overpressurization.

The question of suppression pool effectiveness in quenching relief valve steam discharge in cases which involve highly localized energy transfer from the reactor vessel is important to a complete analysis of any severe accident in which the pressure suppression pool is used as a heat sink. This question has not been resolved, at least in the non-pro-prietary literature, and has been recently adopted as a dissertation topic for a University of Tennessee doctorial candidate working with the Severe Accident Sequence Analysis (SASA) project. Pending the results of this work, pool-averaged temperatures are used in this study.

The MARCH code has been used in support of the analysis of the second phase of a Station Blackout, i.e., the period after the 250 volt DC system fails because of battery exhaustion. The existing versions of MARCH are too crude to permit modeling plant response to a series of postulated op-erator actions such as those previously discussed for the initial phase of a Station Blackout. Therefore, in the event sequence modeled by the MARCH +

code, the reactor vessel remains pressurized with pressure control by automatic relief valve actuation and level control by automatic operation of the HPCI system. As before, averaged suppression pool temperatures are

• used, and it is assumed that injection capability is lost af ter four hours, when the unit battery is exhausted.

In the MARCH event sequence, the reactor vessel water level is in the normal operating range and the vessel is fressurized at the four-hour point when boiloff begins due to loss of injection capability.* The MARCH i results predict core uncovery 62 minutes af ter the beginning of bollof f, followed by the inception of core melting 53 minutes later. The model provides that the melted core slumps down to the bottom of the reactor vessel and this results in a predicted failure of the reactor vessel bot-too head at approximately three hours af ter injection capability is lost.

l The subsequent breaching of the primary containment because of failure of the electrical penetration modules by overtemperature is predicted at about four and one-half hours after the inception of ooiloff. The detailed

• Tnese conditions at the beginning of boiloff are similar to those predicted by the previously discussed sequence which models the plant re-sponse to operatcr actions during the initial phase because in that sequ-

• ence, the reactor vessel would have to repressurize before boilof f could begin. The difference is in the timing. If injection capability were lost at the four-hour point, the boilof f would begin immediately if the -

vessel is pressurized, but would be delayed if the vessel were depressur-ized.

ix thermo-hydraulic parameters needed for the analysis of fission product re-lease from the f uel rods and the subsequent transport of fission products to the environment were taken from the MARCH results for this sequence.

An estimate of the magnitude and timing of the release of the noble gas, cesium, and iodine-based fission products to the environment is pro-vided in Volume 2 of this study. Under the conditions of a Station Black-

, out, fuel rod cladding failure would occur while the reactor vessel was pressurized. The distribution of both solid and gaseous fission products within the Browns Ferry Unit I core at the present time has been obtained a through use of the ORIGEN2 code; the results show that internal gas pres-sure under Station Blackout conditions could not increase to the magni-tude necessary to cause cladding failure by rod burst. Accordingly, the ciadding has been assumed to fail at a temperature of 1300*C.

S i

O s

STATION BLACKOUT AT BROWNS FERRY UNIT ONE --

ACCIDENT SEQUENCE ANALYSIS D. H. Cook S. R. Greene R. M. Harrington S. A. Hodge

, D. D. Yue 4 ABSTRACT This study describes the predicted response of Unit 1 at the Browns Ferry Nuclear Plant to Station Blackout, defined as a loss of offsite power combined with failure of all onsite emergency diesel generators to start and load. Every effort has been made to employ the most realistic assumptions during the process of defining the sequeace of events for this hypo-thetical accident. DC power is assumed to remain available from the unit batteries during the initial phase and the oper-ator actions and corresponding events during this period are described using results provided by an analysis code developed specifically for this purpose. The Station Blackout is as-sumed to persist beyond the point of battery exhaustion and the events during this second phase of the accident in which DC power would be unavailable were determined through use of the MARCH code. Without DC power, cooling water could no longer be injected into the reactor vessel and the events of

, the second phase include core meltdown and subsequent contain-ment failure. An estimate of the magnitude and timing of the concomitant release of the noble gas, cesium, and iodine-based fission products to the environment is provided in Volume 2 of this report.

1. INTRODUCTION The Tennessee Valley Authority (TVA) operates three nearly identical reactor units at the Browns Ferry Nuclear Plant located on the Tennessee River approximately midway between Athens and Decatur, Alabama. The Cen-eral Electric Company and the Tennessee Valley Authority jointly partici-pated in the design; TVA performed the construction of each unit. Unit 1 began commercial operation in August 1974, Unit 2 in March 1975, and Unit 3 in March 1977.

. Each unit comprises a boiling water reactor steam supply system fur-f nished by the General Electric Company. Each reactor is designed for a 7 power output of 3440 MW t , f r a corresponding generated 1152 MWe ; the l

• maximum power authorized by the operating license is 3293 MW ,t or 1067 l ne t MW, . The primary containments are of the Mark I pressure suppres-sion pool type; the drywells are light bulb shape and both torus and dry-l well are of steel vessel construction. The three units share a reactor r

2 building-secondary containment of the controlled leakage, elevated release type.

Safety systems for each unit include a Reactor Protection System, a Standby Liquid Control System for Poison injection, and the Emergency Core Cooling Systems: High-Pressure Coolant Injection (HPCl), Automatic De-pressurization (ADS), Residual Heat Removal (RHR), and Core Spray (CS).

The Reactor Core Isolation Cooling (RCIC) system is also provided for the .

removal of post-shutdown reactor decay heat as a consequence limiting sys-tem.

Several components and systems are shared by the three Browns Ferry -

units. A complete description of these shared features is given in the Final Safety Analysis Report;l the shared Safeguards systems and their supporting auxiliary equipment are listed in Table 1.1, With the assumption that the interfaces with the other two units do not interfere with the op-eration of any shared system as applied to the needs of the unit under study, the existence of the shared systems does not significantly compli-cate the analysis of an accident sequence at any one unit.

The results of a study of the consequences at Unit 1 of a Station Blackout (loss of all AC power) at the Browns Ferry Nuclear Plant are pre-sented in this report. Section 2 provides a description of the event and discussion of the motivation for consideration of this event. The normal recovery from a Station Blackout is described in Sect. 3, the computer model used for the normal recovery analysis is discussed in Sect. 4, and the instrun.entation available to the operator is described in Sect. 5.

l The actiono which the operator should take to prolong the period of decay l heat removal are discussed in Sect. 6, and computer predictions of the be-

! havior of the thermal-hydraulics parameters during the period when a nor-mal recovery is possible are displayed in Sect. 7.

A Severe Accident by definition proceeds through core uncovery, core l meltdown, and the release of fission products to the surrounding atmo-sphere. The equipment failures which have the potential to extend a Sta-tion Blackout into a Severe Accident are discussed in Sect. 8, and the Severe Accident sequences which would follow during a prolonged Station Blackout are presented in Sect. 9. The consequences of each of these se-grances after the core is uncovered differ only in the timing of events; the ac: tons which might be taken by the operator to mitigate the conse-quences of tne Severe Accident are discussed in Sect. 10. The instrumenta-tion available following the loss of injection capability during the pe-riod in which severe core damage occurs is described in Sect. 11.

The conclusions of this Station Blackout analysis and the implica-tions of the results are discussed in S,ct. 12. This includes considera-tion of the available instrumentation, the level of operator training, the existing emergency procedures, and the overall system design.

Appendix A contains a listing of the computer program developed to model operator actions and the associated system response during the per-iod when normal recovery is possible. The MARCH code was used f or analy- ,

ses of the severe accident sequences; the modifications made to this code are described in Appendix B and an input listing is provided in Appendix C. ,

The pressure suppression pool is the key to the safe removal of decay heat f rom an isolated Boiling Water Reactor, but no satisfactory method

3

.able 1.1 Shared rafeguards systems and their austitary support systems Systes Quantity Func tion Standby AC Power Supply Systen Four Jtesel generators each cou- Supply eurgency power during loss of pied as an alternate source of of f ette power conditions power to f our independent shutdown board, f or Unit s 1 and 2. Fout e additional diesel generators serve as altes nate power sources to four Unit 3 shutdown boarJs.

j 250V DC Power Supply System Three batteries, one per unit. Supply DC power when battery hargers The 250V DC power supplies f or the are not functioning DC powered edundant services of each unit are normally derived f rom separate batteries. A fourth battery is provided f or common ststion service.

Cont'al Rod Drive Hydraulic 4 spare control rod drive hydrau- Provide driving f orce f or norrsal rM System lic pump is shared between Units I movement and 2.

Reactor Butiding Closed Cooling Two pumps and two heat eschangers Provide cooling water to reactor Water System per unit with one common spare auxiltary equt;wnent pump and heat exchanger f or oli three units Gaseoas Radwaste System Systes is unitized except far a Ohtatn elevated discharge f rom the common discharge plenum, duc ting, Standby Gas Treitsent System and the 600-f t stack Station Drainage Systes One common drain header into wutch Pass reactor building dratnage to re-each of the three unit reactor ceiver tanks butiding floor dratnage peps dis-

, c ha rge Standby Coolant Supply System Completely shared systes Provide means f or core cooling f ol-lowing a cosplete failure of the Re-s tdual Heat Removal (RHR) cooltig

, to sple x RHR Se rvice Water Systei Completely sharei systes Provide an assured heat sink f or long-teri removal of decay heat when the main condensers are not sva11sble f or any reason Emergency Equi;nsent Cooling Completely shared systes Distribute raw cooting water to equip-Water Systes ment and aualitary systems which are required for shutdown of all three unite Standby Gas Treatment System Completely shared systee used only Filter and exhaust the air f rue a when an abnormal activity release unit zone and the ref ueling gone, the occurs refueling aone only, or the entire seconJary containment for the analysis of the response of the torus and pool under severe acci-dent conditions currently exists, at least in the non proprietary litera-ture. The nature of the problem and the measures being taken at ORNL to improve the analysis capability are discussed in Appendix D.

. In the event of a Station Blackout, the reactor vessel will be iso-lated behind the closed Main Steam Isolation Valves (MSIVs) with pressure maintained by periodic blowdowns through the relief valves to the pressure

, suppression pool. Makeup water to maintain the reactor vessel level must be injected either by the High Pressure Coolant Injection (HPCI) or the Reactor Core Isolation Cooling (RCIC) systems. The operation of these im-portant systems is discussed in Appendices E and F.

4 As a portion of their review of the draft of this report, the Elec-trical Engineering Branch at TVA performed a battery capacity calculation which shows that the unit batteries at the Browns Ferry plant can be ex-pected to provide power f or as long as seven hours under station blackout conditions. Since a battery life of four hours was assumed for the calcu-lations of this work, a new Appendix G has been added in which all of the possible failure moCes other than battery exhaustion that were considered in Sections 3, 7, and 8 are re-examined f or applicability to the period between four and seven hours af ter the inception of a Station Blackout.

The magnitude and timing of the release of the fission product noble gases and various forms of iodine to the atmosphere are discussed in Volume 2. Thia includes the development .of release rate coef ficients f or the escape of these fission products from fuel as a function of tempera-ture, specification of the various chemical forms, determination of the optimum strategy for the modeling of precursor / daughter exchange, and the study of the particular auxiliary systems involved to determine the appli-cable fission product release pathways. These results _re incorporated into a vehicle for the calculation of the transport of th,e individual fis-sion , products from the reactor core, through the primary and secondary containments to the atmosphere, control volume by control volume.

The primary sources of information used in the preparation of this report were the Browns Ferry Nuc2aar Plant (BFNP) Final Safety Analysis Report (FSAR), the Nuclear Regulatory CotLission BWR Systems Manual, the BFNP Hot License Training Program Operator Training Manuals, the BFNP Unit 1 Technical Specifications, the BENP Emergency Operating Instructions, end

• various other specific drawings, documents, and manuals obtained f rom the Tennessee Valley Authority *. Additional information was gathered by means of one visit to the BFNP and several visits to the TVA Power Operatious ,

Training Center at Soddy-Daisy, Tennessee. The excellent caoperation and assistance of Tennessee Valley Authority personne) in the gathering of in-formation necessary to this study are gratefully acknowledged.

• The setpoints for automatic action used la this study are the safety limits as given in the FSAR. In many cases these differ slightly from the actual setpoints used for instrument adjustment at the BFNP be-cause the instrument adjustment setpoints are established so as to provide margin for known instrument error.

I L

5

2. DESCRIFr10N OF STATION BLACK 0UT A Station Blackout is defined as the complete loss of AC electrical power to the essential and nonessential switchgear buses in a nuclear power plant.2 At the Browns Ferry Nuclear Plant, a Station Blackout would be caused by a loss of offsite power concurrent with the tripping of the turbine generators and a subsequent failure of all onsite diesel gen-erators to start and load. After these events, the only remaining sources

, of electrical power would be the battery-supplied 250 volt, 48 volt, and 24 volt DC electrical distribution systems; AC power would be limited to the instrumentation and control circuits derived from the feedwater in-verter or the unit preferred and plant preferred motor generator sets which are driven by the DC systems.

The reliability of of fsite power at Browns Fctry has been excellent.

There are two independent and separated sources oi 161 KV offsite power to the plant, each from a different nearby hydroelectric station. In the near future, of fsite power will become even more reliable. System modifi-cations will permit any of the three reactor units at Browns Ferry, in the event of a generator trip, to receive reverse pouer from the TVA 500 KVA grid which these units normally feed.

However, should all of fsite power we lost together with the tripping of the turbine generators, there remain eight diesel generators at Browns Ferry which are designed to automatically start and load whenever normal AC power is lost. By design, all equipment required for the safe shutdown

, and cooldown of the three Browns Ferry units can be powered by six* of these diesel units, even with the assumption of a design basis accident on any one unit.3

. Therefore, a Station Blackout is sn extremely unlikely event. Never-theless, the consequences of a Station Blackout should be studied. It is important to recall that one Browns Ferry unit suf fered a loss of electri-cal power to all of its emergency core-cooling systems during the March 22, 1975 electrical cable tray fire." To determine if Browns Ferry and similar nuclear plants can recover from the loss of other combinations of AC powered machinery, it will prove most efficient to first consider the plant response to a loss of all AC powered equipmer.t, as in a Station Blackout.

This study considers the effect on Unit 1 at Brom.s Ferry of a Sta-tion Blackout which begins with a loss of offsite powoc, associated tur-bine generator trip, and failure of all diesels to start. Barring further equipment failures, the battery powered systems have the capability to supply the power necessary to operate the high pressure water injection systems to maintain the reactor in a stable cooled state for several hours. The first seven sections of this report pertain to the methods which can be used to prolong this period of water injection capability and decay heat removal for as long as possible.

• With operator action, any six diesel generators would be satisfac-

. tory. However, for adequate short term shutdown and cooldown response without cravat er action, the six operable diesel generators would have to comprise three of the four provided for the Unit 1/ Unit 2 complex and three of the four provided for Unit 3.

6 However, injection capability would ultimately be lost during an ex-tended Station Blackout, either through exhaustion of the installed bat-tery capacity or earlier, through secondary failures of vital equipecat.

Once injection capability is lost, the Station Blackout would become a Severe Accident with inevitable core uncovery. The remainder of-this re-port proceeds from an assumption that the Blackout does develop 'into a Severe Accident, involving core meltdown and subsequent fission product .

release to the atmosphere.

e

\

i

.- . - . . - -,.-,,.,..r---,- . - , , - - - - . , . . - , - - , , . - . . . , - - . .-

7

3. NORMAL RECOVERY The ability to inject cooling water into an isolated reactor sessel for a significant period of time under Station Blackout conditions at Browns Ferry is provided by the Reactor Core Isolation Cooling (RCIC) and the High Pressure Coolant Injection (HPCI) systems. A normal recovery

' f rom a Station Blackout is defined as the restoration of AC power prior to the loss of these cooling water injection systems; no core damage is in-volved, because as long as the RCIC or HPCI system is operable, the reac-tor vessel water level can be cont, rolled within normal operating limits.

The control of water level and other aspects of the response of Browns Ferry Unit I during the period of a Station Blackout in which a normal re-covery is possible are discussed in this section. Equipment failures which might ultimately lead to loss of injection capability are discussed in Section 8 and the sequence of events following this loss is discussed in Sections 9 through 11.

A Station Blackout at Browns Ferry would be initiated by a total loss of offsite power and the concomitant reactor scrams, turbine-generator trips, and Main Steam Isolation Valve (MSIV) closures. At each of the three units, the reactor would be shut down and isolated behind the closed MSIV's within 30 seconds. If the on-site diesel generators fail to suc-cessfully start and load, the conditions for a complete Station Blackout are established. The remainder of this discussion will concern the sub-sequent events at Unit I under Station Blackout conditions.

Immediately following MSIV closure, decay heat generation will cause the reactor vessel pressure to increase to the setpoints of as many relief valves as are required to terminate the pressure increase. The affected

, relief valves would open to pass steam and the associated energy directly from the reactor vessel to a terminus near the bottom of the pressure sup-pression pool. The emerging steam would be condensed by the mechanism of heat transfer to the suppression pool water; the temperature of the pool water would begin a monotonic increase since there would be no means for suppression pool cooling under Station Blackout conditions.

The events up to this point would have all occurred automatically, leaving the reactor vessel isolated behind the closed MSIV's while decay heat generation adds energy which must be removed. The two major consid-erations requiring immediate attention are the means of reactor vessel level control and pressure control.

Level Control. Two independent systems, HPCI and RCIC, are provided for water injection into an isolated reactor vessel at high pressure.

Both require only DC electrical power for operation and comprise steam-turbine driven pumps; the steam is taken from the main steam piping up-stream of the MSlV's and the turbine exhaust steam is discharged into the pressure sur pressioa pool. Both systems are normally aligned for the pumping of inter from the condensate storage tank into the reactor vessel

. via a connection into a feedwater line, and are automatically initiated if l the vessel water level :ecreases to 12.1 m (476 in.) above the bottom of the vessel. This automatic initiation point is 2.lf m (85 in.) below the

• normal operating level of 14.25 m (561 in.) and 2.95 m (116 in.) above the l top of the active fuel in the core. The turbines in ooth systems will trip if the vessel water level rises to 14.78 m (582 in.).

l i

l i -- ___- - - _______

r 8

The HPCI system has an injection capacity of 0.315 m 3/s (5000 CPM),

and is capable of cycling the reactor vessel water level over the 2.69 m (106 in.) range between the HPCI automatic initiation point and the HPCl turbine trip point without operator action. The design and operation ot' this important system are described in detail in Appendix E.

~

The RCIC system has an injection capacity of 0.038 m /s3 (600 GPM).

Operator action is required for long-term level control usins this system .

since once the turbine has been tripped, it will not automatically re-s ta rt. This important system is discussed in detail in Appendix F.

Both of these high pressure injection systems will isolate, i.e. , .

their steam supply valves will automatically shut if the reactor vessel pressure becomes too low to permit turbine operation. The purpose is to prevent excessive leakage from the seals of an immobile turbine; the HPCI setpoint is 0.793 MPa (100 psig), while the RCIC system isolates at 0.448 MPa (50 psig). These systems will also isolate if the ambient temperature in the vicinity of their turbines reaches 93.3*C (200*F); this is to pro-tect against a steam leak in the system piping.

Immediately following the inception of the Station Blackcut, the core water void collapse caused by both the scram and the pressure increase following MSIV closure would result in o rapid drop in reactor vessel water level. The level would decrease to a point beneath the range of level indication available in the Control Room under Station Blackout Con-ditions [13.41 to 14.94 m (528 to 588 in.) above the bottom of the ves-sel]. Standard Browns Ferry Emergency Operating Instructions for immedi-ate action following reactor scram and MSIV closure call for the operator to manually initiate both the HPCI and the RCIC systems. When the level -

has been restored into the indicating range, the operator would turn off the HPCI system. Subsequently the operator would be able to maintain the reactor vessel water level by intermittent remote-manual operation of the '

RCIC system alone, as illustrated in Section 7.

For this analysis of the sequence of events during a Station Black-out, it is intended to employ the most realistic assumptions concerning equipment operation and operator actions. Accordingly, it will be assumed that the operator does take the actions described above to control level over the long term using the RCIC system. However, it should be recalled that the HPCI system is capable of automatic level control should the op-erator fail to act.

The condensate storage tank contains enough stored water to replace that lost from the reactor vessel in the form of steam passed to the pres-sure suppression pool for well over eight hours. The tank capacity is 1419.4 m3 (375,000 gallons) of which 511.0 m3 (135,000 gallons) is a guaranteed reserve for the low pressure emergency cooling systems (which are inoperable during a Station Blackout) and for the HPCI and RCIC sys-tems. This reserve is guaranteed because the pumps of these emergency in-jection syitens take suction on the bottom of the condensate storage tank whereas all other demand is taken from a standpipe within the tank; there is a 511.0 m3 (135,000 gallon) capacity below the standpipe entrance.

• As discussed in Appendices E and F, the pump suctions for each of the two high pressure injection systems can be remote manually shif ted to the ,

pressure suppression poal by the Control Room operator. This would be useful if the condensate storage tank source failed for any reason during l

l l

9 about the first three hours of a Station Blackout. Af ter this , the pres-sure suppression pool water would be so hot (as illustrated in Section 7) that the RCIC or HPCI system turbine lubricating oil, which is cooled by the water being pumped, would overheat; this would significantly threaten the continued operation of these systems.

Pressure Control. Overpressure protection for the isolated reactor

. vessel is provided by thirteen two-stage Target Rock primary relief valves, with a roughly even circumferential distribution of the tailpipe discharges into the pressure suppression pool. These valves are located

• on the main steam lines upstream of the MSIV's and are capable of auto-matic actuation requiring no external source of power other than reactor vessel steam pressure. Four of these valves are set for automatic actua-tion at 7.722 MPa (1105 psig), four are set for 7.791 MPa (1115 psig), and the remaining five are set for 7.860 MPa (1125 psig). Af ter automatic actuation, each valve will reseat following a 0.345 MPa (50 psi) drop in vessel pressure.

Each of the primary relief valves can also be remote manually oper-ated from the Control Room. This requires the availability of pressurized control air for physical operation of the relief valve as well as the availability of DC power for actuation of the solenoid valve which opens to admit the control air to the relief valve operator.

The pressurized air for remote manual relief valve operation is pro-vided by the Drywell Control Air system. The two air compressors in this system normally operate intermittently as necessary to maintain the pres-sure in two 1.61 m3 (57 ft3 ) receiver tanks between the limits of

• 0.689 and 0.793 MPa (85 and 100 psig). Under the conditions of a Station Blackout, the air compressors would be inoperable but there would be an initial supply of stored pressurized air available in the receiver tanks.

Six of the primary relief valves are associated with the Automatic Depressurization System (ADS). (This system, which is designed to auto-matically reduce the reactor vessel pressure during a LOCA so that the high-capacity, low pressure injection systems can operate would not be functional during a Station Blackout.) For improved ADS system reliabili-ty, each of the six ADS relief valves is fitted with an individual air ac-cumulator sized to permit five valve operations without replenishment.

Therefore, under Station Blackout conditions, an initial supply of stored control air is provided for all thirteen primary relief valves by the drywell control air system receivers. An additional special stored supply is provided for the six valves associated with the ADS system; this special air supply alone will permit 30 valve operations, i.e., five remote-manual operations per ADS valve.

For pressure control of an isolated reactor vessel, standard proce-dure is for the operator to remote manually operate the primary relief valves in succession as necessary to cycle the vessel pressure between about 7.688 and 6.309 MPa (1100 and 900 psig). This avoids the automatic actuation ci the relief valves [ lowest setpoint: 7.722 MPa (1105 psig)]

l and greatly reduces the total number of relief valve actuations since the j pressure reduction associated with an automatic valve actuation is only l , 0.345 MPa (50 psi). Lowering the number of actuations decreases the op-portunity for a relief valve to stick open, and permits the operator to evenly distribute the decay heat energy transferred to the pressure sup-pression pool.

l l

l

10 As in the case of reactor vessel. level control, it is intended that-the most -realiscic assumptions concerning pressure control be used in this study. Accordingly, it will be assumed that the operator does take the actions in regard to pressure control as described above. . Howeve r,11t should be noted that without operator action, the pressure would cycle be-tween the limits of _7.377 and 7.722 MPa (1055 and 1105 psig); unfortunate-ly, it is probable that this would occur through the repeated operation of ,

the same relief valve causing severe localized heating in the vicinity of the tailpipe discharge in the suppression pool.

Short-term Plant Status. Within minutes after the inception of a .-

Station Blackout, both reactor vessel level control iand pressure control will be achieved with or without operator action. An open cycle will exist, with water pumped from the condensate storage tank into the reactor vessel, converted to steam there by reactor decay heat, and subsequently' passed to the pressure suppression pool and condensed. Satisfactory re-actor cooling is provided..but both the pressere suppression pool level and temperature are increasing, as is the drywell ambient tempetature due to loss of the drywell coolers. The reserve DC power stored in the bat-teries is being dissipated.

There is -a 250V DC battery for each unit at Browns Ferry, plus a sta-tion battery for common loads. Each unit battery is designed to provide a guaranteed supply of direct current for the first 30 minutes of a design basis accident. Under the less severe conditions of a Station Blackout, and assuming prudent actions by the operator to conserve battery poten-tial,DCgowershouldbeavailableforthefirst Blackout.

4-6 hours of a Station

• Long-Term Considerations. Once DC power is lost , the HPCI and RCIC systems will be inoperable and cooling water can no longer be injected into the reactor vessel. With prudent and realistic operator actions,

• this should not occur unti3 4-6 hours

• af ter the inception of a Station Blackout. However, there are other considerations which must be properly addressed during this 4-6 hour period during which all efforts would be made to accomplish the restoration of AC power.

First, since all plant ventilation is stopped during a Station Black-out, the ambient temperature near the RCIC turbine may reach 93.3*C (200'F), causing an automatic isolation of the RCIC system. This is un-likely since the RCIC system will be operated only intermittently, and the turbine is not located in a closely confined space. If isolation does oc-cur, the isolation signal can be overridden by the placing of rubber insu-lation between the appropriate relay contacts in the auxiliary instrumen-tation room, or the HFCI system can be used as a backup.

Second, a relief valve might stick open although it should be noted that the currently ir. stalled two-stage Target Rock valves are much less prone to this malfunction than their three-stage predecessors. A stuck-open relief valve would produce difficulties with localized suppression pool heating, but reactor vessel pressure would not decrease to the point ,

that the HPCI or RCIC system turbines could not be used. This eventuality is discussed in more detail in Section 8.2.

• Recent battery capacity calculations have indicated that DC power may remain available for as Jong as seven hours.

11 A third consideration would be the continued availability of conden-sate storage tank water for injection into the reactor vessel. As previ-ously discussed, there is a guaranteed minimum stored supply of 511.0 ud (135,000 gallons) available to the HPCI or RCIC systems. As illustrated in Section 7, only about 359.6 m3 (91,000 gallons) would be used during the first five hours of a Station Blackout. It can be concluded that an

• adequate supply of condensate storage tank water is provided for injection during the period in which DC power will remain available.

A fourth consideration is the availability of sufficient stored dry-well control air to permit the desired number of remote-manual relief valve actuations during the period of Station Blackout. The accumulators provided for the six relief valves associated with the ADS system are sized to permit five operations per valve, or a total of 30 actuations.

As illustrated in Section 7, less than 25 relief valve actuations will be required during the first five hours of a Station Blackout. It can be concluded that there is enough stored air in the ADS relief valve accumu-lators alone to permit the desired number of relief valve operations dur-ing the period in which DC power is available. In addition, the stored air in the Drywell Coatrol Air system receivers provides an ample backup to the accumulator supply.

A fifth consideration concerns the steady heatup of the pressure sup-pression pool water with no means of pool cooling available during Station Blackout. The pressure suppression pool of approximately 3785 m3 (one million gallons) of water is contained within a torus of 33.99 m (111.5 ft) major diameter and 9.45 m (31.0 ft) minor diameter. The torus sur-rounds the drywell and reactor vessel as shown in Fig. 3.1.

The large heat sink afforded by the pressure suppression pool water can be effectively utilized during a Station Blackout only as long as the steam being discharged into the pool via the relief valves is condensed.

If the local temperature of the water surrounding the relief valve tailpipe terminus in the pool is excessive, condensation oscillations may occur causing gross unstable vibrations of the torus assembly and pressu-rization of the torus airspace due to tha escape of saturated steam from the water surface. This effect is discussed in detail in Appendix D. For the "T quencher" type of discharge header which is installed at the ter-minus of each relief valve tailpipe at Browns Ferry Unit 1, the condensa-tion oscillations are not expected to occur if the local pool temperature is limited to 93.3*C (200*F) or equivalent to 87.8'C (190*F) average pool temperature.6 The T-quencher dischargers on the tailpipes of the thirteen primary relief valves are distributed approxir tely evenly around the torus, as shown in Fig. 3.2. This diagram is posted on the Browns Ferry Unit I con-trol room panel which contains the relief valve operating switches. Dur-ing a Station Blackort, it would be incumbent on the operator in his ef-forts to manually cot trol reactor vessel pressure to ensure that oppo-

. sitely located primary relief valves are operated in turn so that the energy input to the pressure suppression pool is evenly distributed.

As illustrated in Section 7, the average pressure suppression pool

• temperature will have reached approximately 82.2*C (180*F) five hours af ter the inception of a Station Blackout. It can be concluded that with operator action to alternate the relief valve discharges, the pressure suppression envl temperature will not reach the point when condensation

12 ORN L-DWG 81-8602 ETD 1 <

1. REACTOR VESSEL DRYWELL  ; # F

- 6 s F \

< s f #6 3

~

f . .

f ya p )

P .f,g_. -

B' 4 e s .

b. 7. . g ~ -b ..}i:p'!I'.5.'F -

, w PRESSURE SUPPRESSION POOL =

Fig. 3.1 Arrangement of drywell and torus.

oscillations might occur during the first five hours of a Station Black-out.

As a related matter, the HPCI system is equipped with logic which will automatically shift the suction of the HPCI pump to the pressure sup-pression pool if the indicated pressure suppression pool level increases to a certain point, as discussed in Appendix E. Depending on the initial water level in the pool, this will occur following an increase in pool volume of between 257.4 and 370.9 m3 (68,000 and 98,000 gallons), or

• sometime between two and four hours into the Station Blackout. As previ-ously discussed, the HPCI system is not expected to be utilized but merely to serve as a backup to the RCIC system. However, to preserve the HPCI

13 ORNL-DWG S1-8 03 ETD 4.- t t-

\ I 33r* g3 l .cv ,..

• PC,1-41

...- .ev ,... / _ v ,... .,

ecv i-i.e Pev 1-13 ser~

0 eci i-ir, sra ,

v .a-3..- eev i-i. \ ecv i-si ~,,,.

/

• eev ,-4 acv -a s -

j mm o D cv -,. l ,cv ,.

j (\; ' -

m ~. ~ ~ . or . .r

~ , " ' ' ",** -

\ - m ,..

.. - m. ../- I i ..- ,cv i-. ,e , ,-i.

eev w

. . -.cv ,- i m ,*s ,

m* p

_c ]

vs H,cv ,-s.t atm-i

! .- e ecv i-i.F ecvs.tr.]

Fig. 3.2 Unit i relief valve discharges.

l system as a viable backup, the operator should take action to maintain the pump suction on the condensate storage tank. This can be done by racking out the motor-operated breakers for the suction valves, and will prevent i

the excessive HPCI system lubricating oil temperatures which would result

. from the pumping of the hot pressure suppression pool water.

The sixth and most important consideration for long term stability following a Station Blackout is the heatup of the drywell atmosphere. The ten drywell coolers immediately fail on loss of AC power, while the rate i

I

14 of htat transfer to the drywell at that instant from the hot [287.8'C (550*F)] reactor vessel and associated piping is about 0.98 MW (3.35 x 106 Bt.i/ h) . Under this impetus, the drywell temperature rapidly in-creases. Af ter a few minutes, the drywell temperature will have signifi-cantly increased so that the rate of heat transfer f rom the reactor vessel is reduced while a substantial rate of heat transfer from the drywell at-mosphere to the relatively cool drywell liner has been established, as -

discussed in Section 4.

The design temperature for the drywell structure- and the equipment located therein is 138.3*C (281*F). The response of this structure and of

• certain safe-ty components within the drywell to higher temperatures has been previouely investigated 7 and the results are summmarized below:

It was determined that the drywell liner will not buckle under liner temperatures as high as 171.1*C (340*F) nor would this temperature produce higher than allowable stress intensity.

The drywell electrical penetrations were purchased with a specified short term (15 din) temperature rating of 162.8'c (325*F) and a long term rating of 138.3*C (281*F).

It was determined that the stresses in the drywall piping penetra-tions would not exceed the allowable stress intensities at a temperatute of 171.I'C (340*F).

A DC solenoid control valve used for the remote manual operation of the primary relief valves was tested at 148.9'C (300*F) for ten hours with 118 actuations during the test period. An AC-DC solenoid control valve for the Main Steam Isolation valves was tested at 148.9'C (300*F) for 7 hours8.101852e-5 days <br />0.00194 hours <br />1.157407e-5 weeks <br />2.6635e-6 months <br /> with 83 actuations each on AC, DC, and the AC-DC combination. The ,

maximum temperature achieved during each of these solenoid tests was 153.3*C (308'F). .

The electrical cable which feeds the safety equipment inside the dry-well has a 600-V rating with cross-linked polyethelene insulation. It has seven conductors of number 12 wire with no sheath. It is estimated that the ten-hour temperature rating is in excess of 160.0*C (320*F)./

The results of these tests and investigations indicate that it can be confidently predicted that the primary relief valves will remain operable and the drywell penetrations will not f ail if the drywell ambient tempera-ture is prevented f rom exceeding 148.9'c (300*F) during the normal recov-ery phase of a Station Blackout.

With the operator acting to maintain reactor vessel level by use of the RCIC system and to control vessel pressure between 7.688 and 6.309 MPa (1100 and 900 psig) by remote-manual relief valve actuation as previously discussed, the drywell ambient temperature would reach 148.9'c (300*F) in one hour (curve A of Fig. 3.3). Before this occurs, the operator should act to reduce the reactor vessel pressure by blowdown to the pressure sup-pression pool. This will reduce the temperature of the saturated liquid within the reactor vessel and thereby reduce the driving potential for heat transfer into the drywell. .

The vessel pressure reduction would be by remote-manual actuation of the primary relief valves, and should proceed at the Browns Ferry Techni-cal Specifications limit of a rate equivalent to a 55.6*C/h (100*F/h) de- .

crease in saturated liquid temperature. The response of the drywell ambi-ent temperature to such a reactor vessel depressurization begun at one hour af ter the inception of a Station Blackout is shown by curve B of Fig.

3.3.

15 oRNL-owG 81-7841 ETD 400 I I I e 3s0 - -

a:

3 A

, 300 - g ~

DRYWELL DESIGN g ___ ______ ________._____ _TEMPE RAJUR{ {28pD_ ,

=

52s0 -

k D w

A = NO DEPRESSURIZ ATION o B = 1000F/h DEPRESSURIZATION STARTING AFTER 1 h C = 1000F/h DEPRESSURIZATION STARTING AFTER 20 min D = DEPRESSURIZATION IN 10 min STARTING AFTER 20 min 150 - . -

l l l 0 1 2 3 TIME (h)

Fig. 3.3 Average drywell temperature during a station blackout.

Once th( reactor vessel is depressurized, it is recommended that the operator manually control the vessel pressure between the limits of 0.965 and 0.621 MPa (125 and 75 psig). This pressure range is high enough to permit continued operation of the RCIC system, yet the corresponding aver-age saturated liquid temperature [170*C (338'F)} is low enough to keep the drywell ambient temperature well below 148.9'C (300*F).

As discussed above, a reactor vessel depressurization begun one hour into the Station Blackout would limit the maximum average.drywell ambient temperature to 148.9'c (300*F). Alternatively, a depressurization at the Technical Specifications limit of 55.6*C/h (100*F/h) begun 20 minutes after the inception of the Station Blackout would limit the maximum aver-age drywell ambient temperature to about 141.7'C (287*F); the subsequent drywell temperature response is shown by curve C of Fig. 3.3.

Curve D of Fig. 3.3 represents the average drywell ambient tempera-ture response to a rapid depressuri7ation (within ten minutes) of the re-actor vessel begun at 20 minutes af ter the inception of the Station Blackou t. This would violate the Technical Specifications limit for re-actor cooldown. It can be seen that the drywell temperature only briefly l exceeds the design value of 138.3*C (281*F), and is relatively quickly l brought down to below 115.5'C (240*F).

16 The curves of Fig. 3.3 show that the maximum drywell average temper-ature can be kept below 148.9'C (300*F) provided that the operator takes action to depressurize within one hour following the loss of drywell cool-ers concomitant with a Station Blackout.* A depressurization at the Technical Specifications limit of 55.6*C/h (100*F/h) will produce a grad-ual reduction of the drywell ambient temperature; a rapid depressuriza-tion will produce a faster drywell temperature reduction, but is probably *

not necessary. Certainly, if the operator perceives developing difficul-ties with the in-drywell equipment such as the relief valve solenoids, he ,

should convert an ongoing 55.6*C/h (100*F/h) depressurization into a more rapid one.

Summary. Assuming no independent secondary failures of equipment, reactor vessel level control and pressure control can be maintained during a Station Blackout for as long as the DC power supplied by the unit bat-tery lasts. This is expected to be a period of from four to six hours, and this portion of the Station Blackout severe accident sequence is graphically displayed with commentary in Section 7. If AC power is re-gained at any time during this period, a completely normal recovery would follow.

If AC power is not regained, and DC power is lost, remote-manual op-eration of the primary relief valves would no longer be possible and the reactor vessel pressure would increase to 7.722 MPa (1105 psig), the low-est setpoint for automatic actuation of the relief valves. Therea f t er ,

the reactor vessel pressure would cycle between 7.722 and 7.371 MPa (1105 and 1055 psig) while the vessel level steadily decreased because injection ,

capability would be lost as well. The sequence of events following the loss of injection capability is described in Sections 9 through 11.

• If the backup HPCI system were used for injection, the average re-actvr vessel pressure af ter depressurization would have to be maintained at about 1.14 MPa (150 psig), which corresponds to an average saturated liquid temperature of 185'C (365'F). This is only 15*C (25*F) higher than the average temperature associated with RCIC operation and would not significantly increase the drywell temperatures shown in Fig. 3.3.

- . , - . . - - ,, - -~r -- - - . - , - ,

17

4. COMPUTER MODEL FOR SYSTEM BEHAVIOR PRIOR TO LOSS OF INJECTION CAPABILITY The purpose of this chapter is to describe the BJR-LACP (HJR-Loss of AC Power) computer code that was developed to calculate the effects of operator actions prior to loss of injection capability af ter Station

• Blackout. Topics covered include: general features, simplifying assump-tions, solutions to more important modeling problems, and model validation efforts.

BWR-LACP is a digital computer code written specifically to predict general Browns Ferry thermal-hydraulic behavior following a Station Black-out event (defined in Sect. 2). The code consists of the differential and algebraic equations of mass and energy conservation and equations of state for the reactor vessel and containment. The code was written in the IBM CSMP (Continuous System Modeling Program)B language to minimize program-ming and debugging time. The listing in Appendix A specifies all required input pa rameters.

Some of the important variables calculated by the B4R-LACP code in-clude:

1. Reactor vessel levels (above the fuel as well as in the downcomer annulus),

2. Reactor vessel pressure,

3. Total injected water volume,

4. Safety-relief valve (SRV) flow rates,

5. Containment pressures and temperatures, and

6. Suppression pool water level and temperature.

These variables may be calculated for arbitrary time periods follow-ing the loss of AC power - typically several hours. Run specific in-put includes:

1. Injection flow vs time or a law to represent operator control of re-actor vessel level with injection flow.

2. SRV opening (s) vs time or a law to represent operator control of re-actor vessel pressure with SRV opening.

3. Initialization parameters: initial time elapsed since reactor trip, initial reactor vessel pressure and level, and initial containment pressures, temperatures, and suppression pool level.

Specific characteristics of the Station Blackout event allowed sim-plifying assumptionn to be made, or required specific assumptions for ade-quate description of mystem response. These assumptions are discussed be-low:

. 1. Reactor trip at time zero: heat input from the reactor is as-sumed to consist of decay heat and is calculated from the ANS standard . 9

2. No fuel temperature calculation: the fuel remains covered for

. transients considered here and 100% of the decay heat is assumed to be transferred directly to water.

18

3. Density of steam-water mixture calculated by drif t-flux correla-tion: the low steaming rate of decay heat operation requires a model, such as the drif t-flux modello that can allow relative slip between steam and water.

4. Phase separation is likely in the standpipes: the low steaming rates of decay heat operation allow phase separation to occurli if the 2-phase level is below the steam separators (in standpipes or core outlet -

plenum).

5. Recirculation pumps t rip at time zero: these pumps require AC power for operation. During the loss of AC power transients, the water in the recirculation piping is treated as an inactive volume.

6. Steam flow through the SRV's is calculated by ratioing the flow at rated conditions:

l W=W r (pP)/(pr P)r l l

i

where, W r= rated SRV flow at pressure Pr and density pr W = SRV flow at pressure, P, and density p.

This relationship is rigorously true for choked flow of an ideal gas with 12 and should give reasonable results for choked constant C p /C y ratio flow of dry steam. ,

7. A realistic estimate of SRV relieving capacity is calculated by taking into accouat the ASME code rule which requires nominal valve flow to be only 90% of the flow measured at 103% of the nominal actuation pres- .

sure. For the two-stage Target Rock SRV's at Browns Ferry, the resulting estimate of actual valve capacity is:

Wr = 121.2 kg/s @ 7.79 MPa

(= 960,000 lb/h @ '.115.0 psig).

8. Steam flowing to RCIC and HPCI turbines is calculated by ratio-ing the steam flowing at rated conditions:

W = Wr *(P/Pr )*(AHr/AH)

where, Wr = turbine flow required at conditions P rand All r (available enthalpy dif ference for isentropic expansion ,

between P r and Pexhaust) to pump rated injection flow.

W = turbine flow required at conditions P and AH (isen-tropic expansion between P and Pexhaust) to pump the same ,

(rated) injection flow.

19 This equation predicts a linear relationship between steam pressure and turbine flow providing steam pressure is high enough to establish choked flow across the turbine nozzles. Turbine steam flow is zeroed when the turbine is tripped.

Several modeling problems had to be solved in order to adequately -

simulate behavior of the system following loss of AC power:

1. Reactor vessel steaming rate,

2. Reactor vessel natural circulation flow,

3. Suppression pool water temperature (s), and

, 4. Drywell air temperature.

These particular problems are highlighted because of their importance in determining the overall results and to point out approximations that were made in formulating the solutions.

Reactor vessel steaming rate is defined here as the rate at which steam flows from the liquid regions in the reactor vessel into the steam volume in the upper part of the vessel. When pressure is constant the steaming rate equals steam production rate in the reactor core which is a function of core inlet enthalpy, flow, and pressure. However, when pres-sure is changing, the steaming rate is influenced by the significant amount of water that normally exists in the steam-water mixture above the core in the core outlet plenum, standpipes and steam separators. This water adds to the steaming rate when pressure is decreasing by flashing.

It is assumed to subtract from the steaming rate when pressure is increas-ing by condensing part of the core steam production. In modeling rhis ef-fect, it was assumed that this water remains at saturation at all times.

. The net effect on the steaming rate was calculated using the following re-lationship:

dP My dhg

• Wn"Mc E

• dP (hg - h g )'

where W n = net steam production rate We = core steam production rate P = reactor vessel pressure hf = saturated fluid enthalpy h = saturated vapor enthalpy M = mass of water in core boiling region and above the core in outlet plenum, standpipes, and steam separators.

The inherent natural circulation capability of BWRs is an important feature that should be considered even in calculation of thermal-hydraulic transients at decay heat level. The natural circulation flow path is from the area outside the standpipes and steam separators (referred to here as

. the downcomer annulus), down through the jet pump diffusers into the lower plenum and then into the reactor. As it flows up through the core the

~

fluid becomes saturated and begins to boil. The steam / water mixture flows up through the outlet plenum and standpipes to the steam separators, where the water is returned to the downcomer annulus and the steam flows through 4

20 the steam dryers to the steam dome. The natural circulation rate is de-pendent on a number of variables including: water level in the downcomer annulus, water-steam level above the core, core steaming rate, and density of the water-steam mixture in and above the core. One phenomenon that makes the calculation more dif ficult is that when water level gets low enough the recirculation of water from the steam separators ceases and water flows from the downcomer annulus only as required to counter-balance ,

water lost by boil-off from the core.

The equation used to calculate core inlet flow is:

KgWei2 . Pdc *Ldc - I'tp

• Ltp where Kg = empirically determined friction coef ficient Wei = core inlet flow pdc = downcocer water density Ldc = downcomer water level (ref. to zero at bottom of active fuel)

Ptp = elevation-averaged density of water-steam mixture in and above core Ltp = level of steam-water mixture above bottom of active fuel.

The value.of the friction coefficient, Kr, was calculated from the

• natural circulation curve on Fig. 3.7-1 of the BFNP FSAR by forcing agree-ment with the predictions of this equation at the 30% thermal power point.

• This procedure resulted in a reasonably good approximation ci the flow at other points, as shown on Fig. 4.1.

The rate of recirculation of water back to the downcomer annulus is calculated as a function of the water-steam mixture level in the steam separators:

Wrecir " (Wei --W n }

• f(Ltp) where Wrecir = water recirculated back to downcomer annulus f(Ltp) = empirical function of 2 phase level Wn = net steam production rate Wei = core inlet flow.

The function f(L t P) is 1.0 at normal level in the steam separators, decreasing to 0.0 when level is below the lower edge of the steam separa- ,

tors.

Calculation of local temperature of suppression pool water is neces-sary for simulation of those events that require the assumption of ex- ,

tended SRV discharge into one local area of the pool. For the normal re-covery accident sequences discussed in this report it was assumed that the operators would follow the Browns Ferry procedures for reactor vessel isolation which require manual cyclic sequential operation of the SRVs

21 ORNL-DWG 81-8804 ETD 100 I I I I I 90 -

-DIRECTLY FROM FSAR -

FIGURE 3.7-1

$ 80 -

O POINT USED FOR BWR-LACP 3 INPUT PREPARATION b 70 - -""

O RESULTING BWR-LACP FLOW

, j CALCULATION

!w - -

E o

w - -

n ge - -

d 30 - -

e N 20 - -

10 - -

i I I I I o

O 10 20 30 40 50 60 70 PERCENT RATED CORE FLOW Fig. 4.1 Natural circulation flow.

such that the SRV discharge is distributed throughout the pool. There-fore, it wasn't necessary to calculate local temperatures, and the model was programmed to calculate only the whole pool volume-averaged tempera-ture. It was assumed that the pool would quench all of the SRV discharge as long as the average pool temperature is below the boiling point. The SRV T qt ncher dischargers are located at a depth below the midpoint be-tween pool top and bottom and thus tend to "see" a temperature close to or below the average temperature. The possible occurrence and consequences of non-homogeneous suppression pool temperatures are discussed in Appendix D.

During normal full power operation, plant measurements have shuwn that the Browns Ferry Unit I drywell coolers remove about 1%0 kW

[6.7(10)6 Btu /h) of thermal energy from the drywell atmosphere. This heat comes from various sources, including heat transfer from the hot re-actor vessel and steam lines and from the operation of AC powered equip-ment inside the drywell. During a Station Blackout event, the AC powered equipment is lost so only about half of the heat source remains; accord-ingly, the initial drywell heat load following loss of AC power was taken

to be 980 kW [3.35(10)6 Btu /h). This heat load Qdw is assumed to be

22 proportional to the dif ference between reactor coolant temperature (Trc) and drywell atmosphere temperature (Tdw). Therefore, Qdw = 980*(Trc -- Tdw)/(T rc -- Tdw)o -

To test the validity of the 980 kW heat load, an independent calcu-

• lation was performed under the following tssumptions:

1. Total heated surface area is 1115 m**2 (12,000 sq ft). This includes 539 m*?2 (5800 sq f t) for the reactor vessel, 372 m**2 (4000 sq f t) -

for the steam lines, and 204 m**2 (2200 sq ft) for other heated pip-ing inside the drywell.

2. 90% of the heated surface is insulated. This is a rough estimate to account both for uninsulated piping and for imperfectly installed insulation.

3. Total heat loss through insulated surfaces is 0.25 kW/m**2 (80 Btu /h sq ft). This is a good nominal estimate for MIRROR insulation when total hot surface to ambient temperature difference is about 220*C (400*F).

4. Radiant heat loss from uninsulated surface is figured using emmis-sivity = 0.8 (i.e., unpolished surface)

5. Convective heat loss from uninsulated surface figured using the rela-tion Nu ' O.13 (Gr*Pr)**0.3333 (as discussed in a following para-graph).

The results were as follows:

• Qins, total = 281 kW Qunins, convective = 170 kW Qunins, radiant = 375 kW ,

Qtotal = 826 kW This shows that the heat losses assumed here for Unit I are realistic and perhaps slightly conservative. The Station Blackout analysis presented in the Browns Ferry FSAR (response to AEC Question 14.2) assumes a much more conservative heat loss of 2000 kW.

In addition to heat transfer from hot surfaces, the rate of steam es-caping directly to the drywell must be known in order to cciculate dry-well pressure and temperature. Figure 9.2-2 of the Browns Ferry FSAR specifies a total drywell drain sump ir.put during normal power operation of 22 m**3 (5800 gallons) per day or 4 gpm as measured at the normal sump temperature of about 38'C (100*F). Recirculation pump seals and control rod drive flange seals are identified as the major source of this leakage.

This 4 gpm leak rate is believed to be realistic for Browns Ferry Unit I during a Station Blackeut. The f raction of the leak that flashes to steam was calculated in terms of saturation properties:

X flash = (h @P.

g -- h @Pdw)/(h g @Pdw g -- h @Pdw) g where hg and hg are saturated liquid and vapor enthalpies and P r and Pdw are reactor vessel and drywell pressures. .

Drywell temperature was calculated using a very simple noding scheme

-- the model assumes that the drywell atmosphere is at a uniform tem-perature. This is equivalent to assuming that effective natural circula-tion paths will develop for heat transfer from hot surfaces, to drywell

23 atmosphere, to cold surfaces (such as the drywell liner). Such natural circulation probably exists in the lower part of the drywell where the main steam isolation valves and drywell coolers are located. However it is likely that temperatures in certain areas near the top of the drywell could be much hotter than the single atmosphere temperature this model calculates.

Heat transfer between drywell atmosphere and liner was calculated

- using the relation

, Nu = 0.13*(Cr*Pr)**0.333 where Nu = Nusselt Number (ht/k)

Gr = Grashof Number (p2 ggt3 AT/p2 )

Pr = Prandtl Number (p C p/K) .

This forunla is also used for the same purpose in the MARCH and CONTEMPT-LT codes.13:14 Internal heat transfer resistance of the steel drywell liner is low relative to the resistance between liner and the atmosphere. This allows the use of one temperature to represent the whole thickness and it also ,

means that the entire ~364,000 kg (800,000 lbs) of drywell metal is ef-fectively available as a heat sink during drywell heat-up. The heat sink effect of the drywell liner is important because without this heat sink the drywell atmosphere would very quickly approach primary coolant tem-perature after loss of the drywell coolers.

. Various code comparison activities were pursued in order t o test the validity of the BWR-LACP simulation:

1. Comparisons of BWR-LACP results to calculations reported in the FSAR.

2. Comparison to results computed by the RELAP-IV computer code.

The first FSAR comparison was the " Loss of Auxiliary Power -- All Grid Connections" transient reported on Figs. 14. 5-12 and 14. 5-13 of the BFNP FSAR. Plant conditions assumed by the FSAR analysis include:

1. Loss of all external grid connections at time = 0.

2. All pumps tripped at time = 0.

3. Reactor trip and main steam isolation at about time = 2 s.

The BWR-LACP calculation was started at time equals 20 s with reactor vessel level and pressure equal to those shown for t = 20 s on FSAR Figs.

14. 5-12 and 14. 5-13. Figure 4.2 compares the code and FSAR predictions of reactor vessel pressure and core inlet flow between 20 and 50 s. During this period the core inlet flow is decreasing because steam production is falling off rapidly; reactor vessel pressure is being con-trolled between the assumed 7.59 MPa (1100 psia) and 7.41 (1075 psia) SRV set and reset points. Results are reasonably close. Figure 4.3 shows j . BWR-LACP and FSAR calculations of reactor vessel level from 20 to 2200 s.

( During this period, the vessel level decreases slowly until the mass addi-l tion rate of the 0.0378 m**3/s (600 gpm) injection flow exceeds the mass l - loss rate due to steam produced by decay heat. The results show similar behavior. The FSAR minimum level is 11.76 m (463 in.) at time equals 1600 s and the BWR-LACP predicted minimum level is 11.56 m (455 in.) at time equals 1200 s.

24 ORNL-DWG 81-8806 ETD (MPa) (psi)

I m I I o

$ 0.6 -

z 5 -

s 0.5 -

\ p\

s -

N s

%'% % \ f N

m 0.4 -

% / \ /

e  %,/ / y/

a. 50 - -

{ij 0.3 -

N

$ 0.2 - ---- CURVE A - TAKEN DIRECTLY FROM FSAR e 25 - FIGURE 14,5-12 -

O CURVE B - BWR-LACP RESULTS USING FSAR g0.1 INPUT ASSUMPTIONS w

" I I 0- 0 -

$ i I ,

_J

"- TAKEN DIRECTLY FROM FSAR FIGURE 14.5-12 H

uj O BWR- LACP RESULTS USING FSAR INPUT .

.z ASSUMPTIONS m

e o

o g 25(7 -

Q l =

u.

I O l s O z

U m

E O I I 20 30 40 50 TIME (s) l Fig. 4.2 Comparison of FSAR and BWR-LACP results for loss of

• l auxiliary power transients.

i l

l

~

l 1

t l

d 4

i ORNL-DWG 81-7858 ETD (m) (in.)

57 4 II llll ll Illi II Illi j l l Illl i 14 -

LEVEL REFERENCED TO BOTTOM OF VESSEL J 13 -

U$

in z 12 470 DIRECTLY FROM FSAR w

$ 11 FIGURE 14.5-13 i u. 420 O O CALCULATED BY BWR-LACP d 10 _ USING FSAR INPUT B

370 TOP OF ACTIVE FUEL -

9-320 II IIII II IIII I I IIII I I I III O 2 5 10 2 5 100 2 5 1000' 2 5 10,000 TIME (s)

Fig. 4.3 Comparison of FSAR and BWR-LACP results for reactor vessel level.

.26 Comparison of FSAR and BWR-1ACP calculations of drywell temperature was performed for the " Loss of All AC Power" results shown on Fig.

Q.14.2-1 of the BFNP FSAR. Plant conditions assumed by the FSAR calcula-tions include:

1. Initial pc,.er = 100%.

-2. Loss of- all AC power at time = 0.

3. Drywell perfectly insulated on outside.

4. Heat capacity of interior equipment and structures is negligible. *

5. Energy input from primary surfaces is initially 2000 kW.

6. Heat transfer coefficient between atmosphere and wall is 11.36 w/m**2*C (7 istu/h ft2 .p).without steam condensation. *

7. Drywell coolers lost at time = 0 and not restarted.

8. ' Steam leak = 0 The conditions assumed for this analysis are of the same order as those assumed for the " Normal Recovery" part of - the Station Blackout as dis-cussed in this report. Comparisons of various predictions of drywell atmosphere temperature to the FSAR result are shown on Fig. 4.4. The dry-well temperature climbs rapidly during the first 5 to 10 min, before heat oRNL-DWG81-8606 ETD

(*Cl ('F)

O I I i i i I I 180

- 3W - -

160 -

,#~

/

300 -

/,# '-

,140 -

p# ,'

,s#"

ggg _ 250 l s' -

$ /

CURVE 1- DATA TAKEN DIRECTLY FROM 100 -

FSAR FIGURE 014 2-1 (T-ATM 200 WITHOUT LE AK) -

I -- - CURVE 2: CALCULATED BY BWR-LACP I

USING FSAR INPUT 80 -

f ------ CU RVE 3 SAME AS CURVE 2 EXCEPT OW LINEAR AREA LARGER BY FACTOR OF 1.4 ,

60 -

45 -

I l I t f t t 100 0 15 30 45 60 75 90 105 120 TIME (men)

Fig. 4.4 .Drywell atmosphere temperature: ccmparison of FSAR and ORNL BWR-LACP results.

L.

27 transfer to the drywell liner is established. After the first 10 min, the temperature climbs much more slowly, with the metal liner acting as the heat sink. Curve 2 on Fig. 4.4 was calculated using as input to the con-tainment model the FSAR assumptions as listed above. The resulting curve is similarly shaped, but about 17'C (30*F) above the FSAR prediction. In order to estimate the possible magnitude of input parameter difference,

. the calculation was repeated with identical input except a 40% increase in drywell liner area. Results are shown on Curve 3 and are very close to the FSAR prediction.

A boil-off transient was selected for comparison to RELAP-IV and WR-LACP predictions. The RELAP-IV calculation was prepared and run by Idaho National Engineering Laboratory. It starts at nominal full powet plant conditions. During the first several seconds the reactor trips and the main steam isolation valves close. Initially, several SRVs are re-quired to control vessel pressure, but af ter about 40 seconds one SRV is sufficient. Injection flow is zero throughout. The WR-LACP calculation was intiali.:ed at 30 seconds because it is programmed only for decay heat operation. Figure 4.5 compares the results of the calculations of reactor oRNL-DWG 81-7859 ETD (m) (in.)

25 -

l I I I

()

pff77 e 14 - 550 13 -

500 12 -

• a Sli AM.' WATER ABOVE CORE g 450 -

$ 11 -

d 3 400 -

/ -

> 10 -

DOWNCCMER ANNULUS

_ _ _ . . _ _ _ _ _ TOP.OF, MTig fRE_L,,, , , , _.... ._ y..........

9 350 -

8 -

300 -- O RELAP-4V RESULTS _

. 7 -

I I i I

,so O 500 1000 1500 2000 2500 TIME (s)

Fig. 4.5 Comparison of RELAP-IV and BWR-LACP rcralts for reactor vessel level.

28 vessel level. The steam / water evel above the core is shown for both codes. Results are very similar, s.ith RELAP predicting 33 minutes and BWR-LACP predicting 30 minutes to begin uncovering active fuel. Figure 4.6 shows RELAP and BWR-7 results for reactor steam pressure control.

Results are similar alti gn RELAP predicts a longer SRV cycle (about 60 seconds) than BWR-LACP (about 38 seconds).

Results presented in this Section show that the BWR-LACP code is cap-

• able of providing reasonable predictions of overall thermal-hydraulic ORNL-DWG 81-7860 ETD (MPa) (psia) 7.60 1100 , , ,

i $ i 7.25 1050 - -

RELAP-IV .

I I I I 1000

' 7.60 1100 7.25 tg -_ _

BWR- LACP (b) , , i  ; .

1000 0 500 1000 1500 2000 2500 TIME (s) .

Fig. 4.6 Comparison of RELAP-IV and BWR-LACP results for reactor vessel pressure.

29 variables such as reactor vessel levels and pressure and containment tem-perature for extended periods af ter reactor trip prior to system datage due to loss of injection capability. Features of the code that have con-tributed to its utility in the analysis of the Normal Recovery portion of Station Blackout include:

1. The code is specifically designed for BWRs; therefore, parameter

~

changes are straightforward and easily made.

2. The code is designed for flexibility to model the plant response to different operator actions.

3. It is locally available and has quick turnaround time.

4 6

l l

6

--v-- * - - + - - - ,, -. , , _ _ _ _ _ _ , _

30

5. INSTRUMEhTATION AVAILABLE DURING STATION BLACK 0lTr AND NORMAL RECOVERY For some time following a Station Blackout, the reactor vessel water level can be maintained in the normal range by operation of the HPCI and/

or RCIC systets, and the pressure can be maintained in any desired range ,

by remote-manual relief valve operation. Restoration of AC power at any time during this period of stable level and pressure control would permit complete and normal recovery without core damage. The Control Room in- .

strumentation available and necessary to monitor the plant status during the period of stable control following the inception of a Station Blackout is discussed in the following paragraphs.

1. Electrical system status. The Station Blackout would be clearly discernible by the loss of much of the control room indication and the loss of normal Control Room lighting. Emergency Control Room lighting is available. Numerous instruments would indicate the loss of voltage, am-perage, and frequency on the electrical supply and distribution boards.

2. Reactor vessel water level. Two channels of Control Room instru-mentation would provide indication of the reactor vessel water level over a range between 13.41 and 14.94 m (528 and 588 in.) above the bottom of the vessel. This range includes the normal operating level of 14.15 m (561 in.) and extends over the upper portion of the steam separators, a distance of 4.27 to 5.79 m (14 to 19 ft) above the top of the active fuel.

Mechanical Yarway indication available outside of the Control Room would provide an additional range f rom the low point of the Control Room indica- .

tion,13.41 m (528 in.) above vessel zero down to a point 9.47 m (373 in.)

above vessel zero, which is 0.33 m (13 in.) above the top of the active fuel in the core.

The level instrumentation derives the reactor water level by compar-ing the head of water within the downcomer region of the reactor vessel to the head of water within a reference leg installed in the drywell. With the loss of the drywell coolers, the drywell ambient temperature will in-crease significantly, heating the reference leg water to above-normal tem-peratures. This will reduce the density of the water in the reference leg, causing an error in the indicated water level, which will be too high. For example, if the drywell ambient temperature increases from its normal range of 57.2 to 63.6*C (135 to 150*F) to 171.l*C (340*F), the in-dicated level can be as much as 0.76 m (30 in.) too high.15 As dis-cussed in Sect. 3, the drywell ambient temperature is expected to reach about 148.9'C (300*F). However, the lower boundary of reactor vessel level indication is 4.27 m (168 in.) above the top of the active fuel.

There for e, if the operator maintains the level in the indicating range, a 0.76 m (30 in.) error is not significant as far as the goal of keeping the core covered is concerned.

3. Reactor vessel pressure. Two channels of pressure instrumenta-

~

tion would provide indication of reactor vessel pressure over a range of 8.274 MPa (0 to 1200 psig).

4. Main steamline flow. Two channels of steam flow instrumentation

• with a range of 2016 kg/s (0 to 16 x 10E lbs/h) would provide verifica-tion that steam flow had ceased following Main Steam Isolation Valve (MSIV) closure.

31 5.- Feedwater flow. One channel of flow instrumentation with a range of 1008 kg/s (0 to 8 x 106 lbs/h) for each of the three feedwater pumps would indicate the decay of feedwater flow following loss of steam to the feedwater turbines.

6. Neutron flux. The power range meters . fail on loss of ' AC power.~

The source range and intermediate range monitors remain operational, but

.' the detectors for these systers are withdrawn from. the reactor core during power operation and could not be reinserted under Station Blackout condi-tions. Nevertheless, the operator could verify the reactor scram which

• - occurs when the Station Blackout is initiated by observing the decay of the indicated levels on these monitors.

7. Control rod position indication. The control rod position indi-cation system remains operational so that the operator could verify that the control rods had inserted with the scram. ,

8. Relief valve position. There is no provision'for direct indica-tion of the actual position of any primary relief _ valve, even under normal-operating conditions. Thus the operator has no indication of automatic actuation of a specific relief valve other than the recorded tailpipe tem-peratures available on charts behind the control room panels, or the re-lief valve acoustic monitors, all of which would be inoperable under Sta-tion Blackout conditions. However, remote-manual actuation of a relief valve is accomplished by energizing its DC solenoid operator; lights on the control panel for each valve indicate whether or not these solenoids are energized, and this capability is maintained during Station Blackout..

9. RCIC instrumentation and controls. The RCIC system can be oper-ated and monitored under Station Blackout conditions.

10. HPCI instrumentation and controls. The HPCI system can be oper-g ated and monitored under Station Blackout conditions.

11. Condensate Storage Tank Level. The condensate storage tank level indication [ range: 9.75 m. (0 to 32 f t)] remains operational so that the operator can determine the remaining amount of water available for re-

~

actor vessel injection via the RCIC (or HPCI) system.

12. Drywell Pressure. Drywell pressure instrumentation [ range:

0.55 MPa (0 to 80 psia)] remains operational'so that the operator can mon-itor the increase in drywell pressure due to the ambient temperature in-crease following loss of the drywell coolers.

13. Pressure Suppression Pool Water Level. Indication of torus level [ range: -0.63 to 0.63 m (-25 to +25 in.)] remains available allow-ing the operat to monitor the increasing pool level caused by both the relief valve blowdowns and the increasing pool temperature.

The above instrumentation is powered during Station Blackout either by DC power directly from the installed batteries or by AC power-indi-rectly obtained from the battery systems. The sources of AC power during Station Blackout are the feedwater inverter and the unit preferred and plant preferred systems where single phase 120V AC power is provided by.

. generators driven by emergency battery powered DC motors.

It is important to note that Control Room temperature instrumentation -

for the drywell and the pressure suppression pool does not remain opera-

= tional under Station Blackout conditions. However, the ambient tempera-tures at various points in the drywell would be available at an indicator-meter and a recorder mounted on panel 9-47, which is located on the back of the Control Room panels and is accessibir from outside of the Control

32 Room. Also, using a portable self powered potentiometer, station person-nel can monitor local pressure suppression chamber.and drywell tempera-tures which are sensed by installed thermocouple elements.

Following the restoration of normal AC power, either by recovery of the offsite power sources or by the delayed but finally successful loading of the onsite diesel generators, the Instrumentation and Control buses would be re-energized, restoring all normal Control Room instrumentation for 31se in monitoring tha normal recovery.

I e

6 1

4 0

-- - - - . n , ~ . , . . - - - - , , , , - - . , c..- -, .. v . - -- - . - , - , - - - - .-e- - - - -v . . ,- . - , ,.,-----w - ,- w - , -- - --

33

6. OPERATOR ACTIONS DURING STATION BLACK 0UT AND NORMAL RECOVERY In the event of a Station Blackout, the immediate result would be a turbine trip, reactor scram, and Main Steam Isolation Valve (MSIV) clo-sure, which would all occur automatically with no operator action; the re-actor would be shut down and isolated behind the closed MSIVs within 30 seconds following the initiating loss of offsite power. Immediately fol-

. Iowing the isolation, the reactor vessel pressure would increase to the setpoints of as many relief valves as are required to terminate the pres-sure increase. The affected relief valves would open to pass steam di-rectly from the reactor vessel to a terminus beneath the water level in the pressure suppression pool so tnat the released steam is condensed.

The core water void collapse caused by both the scram and the pres-sure increase following MSIV closure would result in a rapid drop in re-actor vessel water level to some point below the lower limit of level in-dication available in the Control Room under Station Blackout conditions

[13.41 m (528 in.) above vessel zero]. Standard Browns Ferry Emergency Operating Instructions for immediate action following reactor scram and MSIV closure call for the operator to manually initiate both the HPCI and 3

the RCIC systems. The combined injection flow of 0.353 m /s (5600 CPM) would rapidly restore the level into the indicating range and the operator would secure the HPCI system when the indicated level reached a point equivalent to about 13.72 m (540 in.) above vessel zero. The level would

• then continue to increase due to the remaining 0.038 m 3/s (600 GPM) of RCIC system flow and the heatup and expansion of the injected water, but at a slower rate.

For pressure con *.rol, standard procedure is for the operator to remote manually operate the primary relief valves as necessary to cycle the reactor vessel pressure between 7.69 and 6.31 MPa (1100 and 900 psig).

This prevents the automatic actuation of the relief valves [ lowest set-point: 7.72 MPa (1105 psig)], and since the pressure reduction following automatic actuation is only 0.34 MPa (50 psi), greatly reduces the total number of relief valve actuations. The purpose is to decrease the oppor-tunity for a relief valve to stick open, and to allow the operator to al-ternate the relief valves which are in effect passing the decay heat energy to the pressure suppression pool; this more evenly distributes the pool heatup.

Throughout the period of Station Blackout, the operator would operate the RCIC system intermittently as necessary to maintain the reactor vessel level in the indicating range, i.e., between 13.41 and 14.94 m (528 and 588 in.) above vessel zero. Thus, it is expected that the operator would turn on the RCIC system at an indicated level of about 13.61 m (538 in.)

and turn the system off at about 14.68 m (578 in.). The periods of RCIC system operation would become less frequent as the decay heat intensity subsides during a prolonged Station Blackout; af ter the first hour, the l operator wotid find it necessary to run the RCIC system for only about

. one-t,hird of the time. This is illustrated in Sect. 7.

As previously discussed, the operator would control reactor vessel l pressure by remote manual operation of the relief valves. It is important

( that the operator take action to begin to lower the reactor vessel pres-sure to about 0.79 MPa (100 psig) within 60 min following the inception of i

34 a Station Blackout. This is necessary because of the loss of drywell cooling concomittant with a Station Blackout, and would be accomplished by remote manual actuation of the primary relief valves. Analysis indicates that the average ambient drywell temperature will not exceed 148.9'C (300*F) if this action is taken within one hour. It should not be neces-sary for the cooldown rate to exceed the Technical Specifications limit of 55.6*C/h (100*F/h) for reactor vessel cooldown, but the operator should increase the cooldown rate if temperature-induced problems with the opera-tion of the relief valves are perceived. This reactor vessel depressuri- ,

zation is necessary to preclude an excessive ambient temperature in the drywell, designed for 138.3*C (281*F). Excessive drywell ambient tempera-ture could cause serious and expensive damage to drywell equipment and f rom a safety standpoint, would threaten the primary relief valve manual actuation solenoids and the integrity of the primary containment.

When the operator acts to reduce the reactor vessel pressure from 7.00 to 0.7 9 MPa , (1000 psig to 100 psig) the corresponding reduction in saturation temperature within the vessel is from about 285 to 170*C (545 to 338"F); this reduces the driving potential for neat transfer into the drywell and limits the maximum drywell average ambient-temperature follow-ing the Station Blackout to about 148.9'c (300*F). The RCIC system steam turbine is capable of operation with an inlet pressure of 0.79 MPa (100 psig) and is of ten run under these conditions. It is reasonable to expect the operator to act to manually control vessel pressure between the limits of 0.97 and 0.62 MPa (125 and 75 psig) during a prolonged period of Sta-tion Blackout while DC power remains available. Analysis shows that this .

will not require an excessive number of relief valve actuations.

The operator must alternate the relief valves used for pressure coa-trol so that the suppression pool water is evenly heated by the condensing -

steam. Otherwise, the local water temperature surrounding the tailpipe of an overused relief valve may become too high for ef fective steam condensa-tion.

It should be recalled that there are temperature sensors located near the HPCI and RCIC turbines which are designed to detect steam leaks and consequently shut down these systems by closure of their primary contain-ment inboard and outboard steam supply isolation valves. The setpoint is 93.3*C (200*F) and it is not unreasonable to suggest t hat this space tem-perature might be reached during a prolonged Station Blackout when these systems are operated without benefit of space coolers, although neither of these systems is closely confined. If this occurs, the operator would have to take action to override these system isolation signals. This can be easily done in the auxiliary instrument room by the placement of rubber insulation between the temperature-activated relay contacts.

It is important to note that no emergency procedure for Station Blackout at Brown's Ferry currently exists. The operator actions discus-sed in this section are those indicated by the ORNL analysis of the casu-alty. In summary, these show that during a prolonged Station Blackout the operator should maintain reactor vessel level within the indicating range by intermittant operation of the RCIC system and should control pressure between 0.62 and 0.97 MPa (75 and 125 psig). This range of control in-sures sufficient pressure to run the RCIC turbine, but keeps the reactor vessel temperature as low as possible to minimize drywell heatup.

35 There are ample supplies of Condensate Storage Tank water available for injection, and the period of stable level and pressure control follow-ing a Station Blackout can be maintained for a minimum of four hours pro-vided that independent secondary equipment failures do not occur. Beyond four hours, the availability of battery-supplied DC power is in question, and the pressure suppression pool water temperature will have increased to the point where the successful' condensation of the relief valve steam dis-charges begins to be in doubt.*

The required frequency of RCIC system operation.and the manual actua-tions of relief valves by the operator during the period of Station Black-out are shown on the plots of thermal-hydraulic parameters as functions of time included ir. Section 7.

During the period of Station Blackout, the 0.038 m 3/s (600 GPM) RCIC capacity is sufficient for vessel level control, and the HPCI system need not be used. However, if the RCIC system should malfunction, the HPCI sys-tem can be used as a backup. It is important to recall that the suction-of the HPCI pump will automatically shif t from the condensate storage tank to the pressure suppression pool when the pressure suppression pool level increases to an indicated level of 0.18 m (+7 in.) as discussed in Appen-dix E. This level will be reached after about three hours of Station .

Blackout, when the pressure suppression pool temperature will have in .

creased to about 71.l*C (160*F). Since the HPCI turbine lubricating oil is cooled by the water pumped, and because the high water temperature may provide an inadequate net positive suction head, the operator should' take action to prevent this shift in pump suction under Station Blackout condi-

, tions. This can be done by manually racking out the circuit breakers for the DC-motor-operated suction valves. ,

When AC power is restored, the operator must take action to implement

. long-term cooling for the removal of decay heat. It is important that he recognize that the increased drywell pressure caused by the heatup of the drywell atmosphere following the loss of the containment coolers combined with the reduction of vessel pressure form the classic indication of a LOCA, i.e., high containment pressure [setpoint 0.012 MPa (? psig)]^ and low vessel pressure [setpoint 3.21 MPa (450 psig)]. Unless the operator takes action to prevent it, these signals will cause both the Low Pressure Coolant Injection (LPCI) mode of the Residual Heat Removal (RHR) system and the Core Spray system to automatically actuate as soon as AC power is restored. These systems would then inject a combined flow of approxi-mately 2.271 m 3 /s (36,000 GPM) into the reactor vessel entirely .unneces-sarily. This flow would quickly fill the vessel and raise the pressure to the shutoff head of the dore Spray and RHR pumps. Several actions might be taken by the operator to prevent this occurrence; for example the high drywell pressure signal can be overridden, or the core spray and RRR sys-tem pumps can be turned off before AC power is restored. The important thing is that the operator anticipate this occurrence. ,

The normal recovery from a Station Blackout would be established when

+ the operator has manipulated the RER system into the shutdown cooling mode and the suppression pool cooling mode following the restoration of AC

! power.

i t

• See Appendix G for analysis of the case where DC power f rom the l

l unit batteries is assumed to remain available for seven hours.

tN

36

7. COMPUTER PREDICTION OF THERMAL-HYDRAULIC PARAMETERS FOR NORMAL RECOVERY 7.1 Introduction The purpose of this chapter is to present the results of BJR-LACP .

calculations (see Sect. 4) of system behavior for two different portions of the Station Blackout:

1. Normal Recovery: DC power from the unit battery remains available

• during this period; therefore RCIC and HPCI injection flows are assumed available throughout, and

2. Loss of 250 vde Batteries: The batteries are assumed to fail after four hours, causing failure of RCIC and HPCI injection, ultimately leading to uncovering of the core about eight hours af ter the blackout.

The normal recovery portion of the Station Blackout is discussed in detail in Sects. 3 and 6. Operator actions assumed here are consistent with those sections. Results were calculated to five hours

• for normal recovery in order to provide some overlap with the MARCH code severe ac-cident calculations (see Sects. 9 and 10), which assume that the boil-off begins af ter four hours from a fully pressurized condition. The Loss of 250 vde Batteries results are given in this section in order to provide an estimate of system behavior possible af ter loss of de power if the plant is depressurized during the normal recovery period.

7.2 Conclusions Major conclusious drawn from the calculated results are given below, .

succeeded by a detailed discussion of code input assumptions and transient res ult s.

7.2.1 Normal Recovery -- Conclusions If power were recovered within five hours of the inception of the Station Blackout, a normal recovery would be possible. System parameters are within acceptable ranges after five hours:

1. The 250 vde batteries, by assumption, last the full five hours.

2. Reactor vessel level is wi thin the normal control range, about 5.08 m (200 in.) above the top of active fuel.

3. Reactor vessel pressure is being controlled at about 0.69 MPa (100 psia).

4. About 354 m3 (93,500 gal.) of water have been pumped from the Con-densate Storage Tanks, which had an assured capacity of 511 ud (135,000 gal.) before the blackout.

5. Suppression pool temperature is about 82*C (180*F), but this should not be a problem for the T quencher type of SRV discharge header pip- -

ing.

6. Containment pressures are elevated to about 0.17 MPa (25 psia), well

• below the 0.53 MPa (76.5 psia) design pressure.

• These results are extended to seven hours in Appendix G.

I

37

7. Drywell atmosphere temperature is below the 138'c (281*F) design tem-perature.

7.2.2 Loss of 250 vde Batteries -- Conclusions The results of this transient show very clearly how fuel damage is postponed because the reactor vessel was depressurized early'in the Sta-tion Blackout:

1. The repressurization time af ter loss of the 250 vde batteries is

, greater than one _ hour, during which time there is no significant coolant loss from the reactor vessel.

2. When the boil-off does begin, it takes a much longer time to uncover fuel because of the higher starting inventory of water.

Although fuel damage is significantly delayed, the ability to avoid ultimate fuel damage is compromised because of the elevated drywell team perature experienced af ter. loss of the 250 vde batteries. As discussed in Sec t. 3, a containment temperature of 149'c (300*F) would not prevent nor-mal recovery. This temperature is reached about 40 min. af ter loss of the batteries. At about four hours af ter. the battery loss, the fuel is begin-ning to be uncovered, and the drywell temperature is above 191*C (375'F).

This elevated temperature may cause failure of the drywell electrical pen-etrations and may fail the solenoid operators necessary for operation of the SRVs, inner isolation valves, and containment cooler dampers (which fail closed on loss of AC power). Even if electrical power were fully re-stored at this point, considerable operator ingenuity would be required to

. effect a normal recovery if the MSlV and SRV solenoid operators have failed.

In addition, the chances for a normal recovery are compromised by the

+

elevated suppression pool temperatures experienced at the end of the tran-sient. Condensation oscillation during SRV discharge (see Appendix D),

which can damage the suppression pool pressure boundary, becomes more likely at higher pool temperatures. Steam discharge without oscillation from the Browns Ferry T quencher type discharge piping is assured up to 88'C (190*F), but average pool temperature at the end of the Loss of 250 vde Batteries calculation is above 93*C (200*F).

7.3 Normal Recovery 7.3.1 Normal Recovery -- Assumptions

Assumptions specific to programming and input preparation for the

! normal recovery calculation include:

i 1. The calculation begins 30 s following the Station Blackout initiation with the reactor tripped and the main steam isolation valves closed.

Values for system parameters at the 30-s point are taken from the re-

. sults reported by Sect. 14.5.4.4 of the Browns Ferry FSAR. The cal-

! culation ends at the 5 hour5.787037e-5 days <br />0.00139 hours <br />8.267196e-6 weeks <br />1.9025e-6 months <br /> point.

l 2. The RCIC system is used in an on-off mode to control vessel level be-tween 13.7 m (538 in.) and 14.7 m (578 in.) above vessel zero. The

38 HPCI system ic actuated only when level is more than 0.25 m (10 in.)

below the 13.4-m (528-in.) lower limit of control room indication under Station Blackout conditions. Suction for both systems is from the condensate storage tank (CST). The operators actuate HPCI and RCIC 1 min. after the blackout.

3. Following 1 min. of automatic SRV actuation, reactor pressure is con-trolled by remote-manual actuation of the SkVs. Control range dur-ing the first' hour is 7.52 MPa (1090 psia) to 6.31 MPa (915 psia).

• Af ter 1 hour1.157407e-5 days <br />2.777778e-4 hours <br />1.653439e-6 weeks <br />3.805e-7 months <br /> the reactor is depressurized to about 0.79 hPa (115 psia) at a rate consistent with the 55.6*C/h (100'F/h) cooldown rate

• limit. Af ter depressurization, pressure is controlled between 0.62 MPa (90 psia) and 0.86 MPa (125 psia).

Manual sequential SRV action is used to spread SRV discharge around the suppression pool and thereby avoid localized heating of the pool.

4. The assumed reactor system leakage into the drywell is 0.0089 m3 /s (4 gpm).

7.3.2 Normal Recovery - Results Results for normal recovery are shown on Figs. 7.1 through 7.9.

Each system variable is discussed below.

7.3.2.1 Reactor vessel pressure. Figure 7.1 shows reactor vessel pressure. During the first minute, reactor pressure cycles between 7.72 MPa (1120 psia) and 7.38 MPa (1070 psia) on automatic SRV actuation.

After 1 min. the operator opens one SRV to lower pressure to 6.31 MPa (915 .

psia). Since the HPCI is running, pressure continues to decrease to below 6.21 MPa (900 psia) until the HPCI is shut down. During the first hour a single SRV is cycled open and shut seven times to keep pressure in the de- .

sired range.

Af ter one hour, the depressurization begins. Initially, intermittent operation of one SRV is suf ficient to meet the target depressurization rate [i.e. , 55.6 *C (100*F) /h] . As pressure decreases, the rate of de-crease begins to slow until finally an additional SRV is opened. For choked flow of dry steam, the mass flow rate is linearly proportional to reactor pressure . Theref ore an extra valve is required at the lower pre-ssure. When depressurization is complete, the operator controls pressure between 0.62 and 0.86 MPa (90 and 125 psia).

Af ter depressurization, the steam flow through one SkV is nearly equal to the core steaming rate, resulting in much slower pressure change.

Under these conditions, the most rapid pressure change is caused by the RCIC system. When RCIC comes on the 600 gpm injection of cold water tends to lower the core steaming rate, causing pressure to decrease. This ef-fect can be seen in Fig. 7.1. At 256 min., the RCIC comes on, reducing pressure until the open SRV is shut. When the RCIC is shut off 23 min.

later, pressure increases rapidly until the SRV is reopened.

Steam flow f rom the reactor vessel is shown in Fig. 7.2. The large .

spikes are due to SRV actuation. Also included in the total steam flow are the smaller amounts of steam flowing to the RCIC and HPCI turbines when they are running. ,

e D 0 0

T 3 E E

. R 9 -

U l 8

4 F

, S e 8 F Y S s C 0 E s 1 R 3 e v

I 8 $

. C - P 2 G R S W N N L r D O O o

- R t L

N T c R N 0 a O Oa 2 e Ci sp r R -

O 52

- T1 A

RO 0 t

k u

o

- E T P0 2

c O9 a l

b 0 n s -

8 1 o N i t

O )

a I

T in t A (

m s Z 0

. 3 I R D

$ E r U E t

M e

. N ST I

T t

O SE f I

T EL a A RP Z

I

,lV PM EO 0 2

y r

. R DC 1 e U v S . o SS A c E N Y e RI r PG c 0 EE 9 l DB a a n

l, n

> l o

N

~ 0 6

( 1

)f .

l f

F 7 l F O . .

I 0 ge f C 3 ir l P F u

)l l s s

)l l H_ e n\

r p

0 8 ~ g g 8 S R g

- _ ~ -

nE Eham2bM

- ~ - -

e s ~ - * " ~ -

1Eehum2dm

tO o

O O 8 w

O *

• 8 d < z r > 9 g I

8 j

> 50 5 8 -!! -

o 02 w= yt j g a

2 o $>

m- mm

$ NM r

E e yz om

$ $o mm ZC 82 "Q rz mO ~$

M o m

,Hy m$ <O u H

m ON om 2 m

m H3 D: m e ZC om HO l <

• @ i -o Y" fO mp "

I E3 q ENI8 RUT CICR HGUORHT WOLF h $

ER E$ _

h og -8 7

HC mO -

5 D

C< I o

1. '3 s g l m m Qm n ,1 -$

- m n a

d mm 1 n

$> ' -E 4

>m rO i

Q

~

8 l o

-8 8 l "

W p

?

.g 8 o

z l

ENIBRUT CICR HGUORHT WOLF e l

l

-8 N SENIBRUT CICR DNA ICPH HGUORHT WOLF N "

• c l

052 52'2 00'2 57'1 05'1 52'1 00'1 5'7 0'5 Q

5'2 0 H

)s/bI( WOLF MAETS LESSEV m i 1 9 a f i t R R E 011 001 09 08 07 06 05 ot 1 0 03 02 01 O '

)s/gk( WOLF MAETS LESSEV

41 7.3.2.2 Reactor vessel level -- normal recovery. Figure 7.3 shows vessel water level (distance sbove vessel zero) in the region outside the core outlet plenum, standpipes, and steam separators. This is the level measured by vessel level instrumentation and available in the control room. As discussed in Sect. 4, there is a corresponding level of two-phase mixture in and above the core. For an undamaged core in a mild

, transient such as Station Blackout, the level of steam / water mixture in and above the core will be higher than the downcomer level whenever down-comer level is below the top of the steam separators.

. Downcomer level is initialized at 12.7 m (500 in.) and, at first, de-creases rapidly until the operator initiates injection via the RCIC and HPCI systems. When level recovers to 13.7 m (540 in.), the operator shuts of f the HPCI system, and when level reaches 14.7 m (578 in.), the RCIC injection is shut off. Level continues to increase due to the continuing heatup of the large quantity of cold water injected by the RCIC and HPCI systems and due to formation of additional voids in and above the core as the steaming rate increases. The level peaks at 15.2 m (600 in.). This is about 0.30 m (12 in.) above the top of the range of level indication available in the control room during Station Blackout.

Control Room level indication during a Station Blackout is limited to a range of 3.66 m (0 to 60 in.) with instrument zero corresponding to 13.41 m (528 in.) above vessel zero, which is the bottom of the reactor vessel. The basic cycle of RCIC actuation by the operator when level reaches an indicated 0.25 m (10 in.) and shut-off when level reaches an indicated 1.27 m (50 in.) is repeated eight times on Fig. 7.3. The fine structure is caused by action of the SRVs. When the SRV opens, vessel level first swells due to increased steaming, but then begins to decrease due to inventory loss. When the SRV shuts, level at first continues to decrease due to decrease in steaming rate, but then begins to recover due to heatup of subcooled liquid still within the downcomer.

Injection flow from the CST to the reactor vess21 is shown in Fig.

7.4. The operator controls the RCIC and HPCI systems in an off-on mode in order to maintain vessel level within the desired control band as speci-fled in Para. 7.3.1.2. The HPCI system is only run during the initial few minutes. The periods of RCIC system operation become less frequent as the decay heat subsides with time. Total injected flow (i.e. , total amount of water taken from the CST) is shown in Fig. 7.5. The injected total in-creases rapidly at first during the brief period of combined HPCI and RCIC operation; the increase is much slower during the subsequent intermittent periods of RCIC operation.

7.3.2.3 Suppression pool level -- normal recovery. When steam flows from the reactor vessel to the suppression pool, it is condensed by the colder pool water. Suppression pool level increases both because of the extra mass of the condensed steam'and because of the temperature increase of the original pool water. Suppre:sion pool level is shown in Fig. 7.6, referenced to instrument zero of the torus. The torus has a 9.45 m (31 f t) inside diameter and instrument zero is 0.1 m (4 in.) below the mid-

. plane. (A zero level indicates the torus is about half filled with water.) The pool level is normally maintained between -0.051 and -0.152 m

(-2 and -6 in.) indicated; an initial pool level of -0.102 m (--4 in.)

was assumed for this analysis. The pool condenses steam both f rom the i

OflNL-DWG 81-8491 ETD 8

N

t. -

LEVEL SPlKES DUE TO g

..a

\ SRV ACTUATION m.

~

TOP OF CONTROL ROOM INDICATION

• - .. F ...... ..........

/ I..... ... ....

....k . ...... . . . . . . . . . . . . . . . .

e.

/

h)  %

g n

/

a y y I-w a

h v v "

v vv v

>n. W>

w- j s

$ /

w -8 g c.~ ~

• BOTTOM OF CONTROL ROOM INDICATION e w $

W RCIC SYSTEM OPERATION TO MAINTAIN N

> LEVEL IN INDICATING RANGE

> 8,.

o. h

m. g

......... .......... ......... .................. . . . . . . T.O.P .O.F. .A.C.T.I V.E. .F.U.E.L. . .

n e.

0 30 60 90 120 150 180 210 240 270 300 TIME (min)

Fig. 7.3 Nonnal recovery af ter station blackout - reactor vessel level.

OR NL-OWG 81-8492 ETD E

~

3 4 5000 gpm HPCI PLUS 600 gpm RCIC

g. TOTAL INJECTED FLOW t

i M- g.

1 R-2-

i -3 N ~

2 @_

~ M-E _E fn 8 g-a

• 3 R- 3

' 3 e- u.O g.

m y t-W 8 E- S B_

a b

- g- 8 3"

I- HPCI OFF e- RCIC

• ~

g_ ON OFF ON OFF

~

RCIC OFF

,_ . /,- - ( \ [ h re -

o- o 1 O f 9'O 1$0 1$o 10 210 250 do M TIME (min)

Fig. 7.4 Normal recovery after station blackout - injection flow.

a ORNL-DWG 81-8493 ETD g.

e-b.

g_ I.CIC

~ OFF ON

/

)

~

n RCIC IS OFF DURING PERIODS WHEN

5. g. INDICATED VESSEL LEVEL IS ABOVE ~ - - -- - - - -

g g DESIRED LOWER LIMIT S S OO O

> xi- >

8 88 ,

o w G8 w

~

E E_ a_

/ e

.a a o S- h 12.

RCIC /

/

8 OFF ON /

9 HPCI OFF 8-h 1 INJECTION BY COMBINED HPCI f- AND RCIC SYSTEMS O~ O

,  ! l l O 30 6'O IO lb 150 180 21'O 240 2$0 300 TIME (min)

Fig. 7.5 Normal recovery af ter station blackout - total injected Ilow.

D 0 0

T 3 E

4 9

4 8

/

1 0 8 7 2

G

. W n D

- o L i N 0 s

R 4 s O s 2 e r

f p

p u

s 0

1 -

2 t

u E o

[ SV AR ES 0

k c

a

/

8 l E R F 1 b

S C O A N )

in n E IG o RI C DN m i CC II PH

(

t NR AC 0 5 E a

/ M t IF RN 1

s LO A F EW I

T UG OUO r f_ e DNTS 3 QL F t AIHU , DO 0 f RCA OTM A 2 a GNH I RE E 1

y

. EX E UT r UE PDS e TQ U E v H ON o STIB 0 c

' 9 e VER ~

r RUU ~

SDT J l a

}v

\ 0 6 N m

r o

6 7 .

0 3

gv

.le i e Fl l

o 0 o p

S  % o~ ' " n  ?

au ah 8 hm&B

- ~ - -

. n ~ o 7

_ { i

" a ! z92Ye %v 1

I s

s

46 SRVs and the RCIC and HPCI turbine exhaust. The effect of RCIC turbine exhaust shows up in Fig. 7.6 as a very slow background rate of increase against the dominant ef fect of the SRVs.

7.3.2.4 Suppression chamber temperatures - normal recovery. Su p-pression chamber temperatures are shown in Fig. 7.7. For all practical purposes, the average pool temperature responds solely to condensation of reactor vessel steam. Therefore, this curve has almost the same shape as

• the pool level curve. The pool atmosphere can be influenced by mass and heat transfer from the pool water and by mass transfer from the drywell ,

a tmosphere.

During the first 30 min. the suppression chamber atmospheric tempera-ture increases rapidly because of mixing with the inflow of hotter gases from the expanding drywell atmosphere. Drywell temperature (Fig. 7.8) in-creases rapidly at first due to loss of the drywell coolers. As a result, the drywell atmosphere expands into the suppression chamber atmosphere through the pool water via the eight vent pipes, the single ring header, and the 96 downcomer pipes.

In the long term, mass transfer fr om the drywell ceases and the sup-pression chamber temperature continues to increase due to heat transfer and evaporation from the suppression pool surface.

7.3.2.5 Containment pressures normal recovery. The suppression chamber and drywell pressures are shown in Fig. 7.9. The drywell and sup-pression chamber atmospheres are coupled in the following manner:

1. When suppression chamber pressure exceeds drywell pressure by 0.00345 .

MPa (0.5 psi), the vacuum breaker valves will open and allow flow of suppression chamber etmosphere into the drywell.

2. When drywell pressure exceeds suppression chamber pressure by more 2 than about 0.0121 MPa (1.75 psi), this will clear the 1.22 m (4 f t) of water from the downcomer pipes and allow drywell atmosphere to bubble up through the pool water into the suppression chamber.

Immediately after the Station Blackout, drywell pressure increases rapidly due to the concurrent rapid heatup of the drywell atmosphere.

Even after the drywell atmosphere reaches its peak temperature and starts down, the drywell pressure continues to increase under the influence of the assumed 0.0089 m3/s (4 gpm) leak of hot reactor coolant, some of which flashes to steam. Throughout the first 2 hours2.314815e-5 days <br />5.555556e-4 hours <br />3.306878e-6 weeks <br />7.61e-7 months <br />, conditions in the drywell bring the suppression chamber pressure up at about the same rate by expansion (via the downcomers) into the suppression chamber. After the first two hours, suppression chamber pressure is increasing f aster than dictated by the drywell because of increasing rates of heat transfer and evaporation from the overheated suppression pool. Af ter about 190 min.

the pressure is high enough to open the vacuum breakers and caose expan-sion of suppression chamber atmosphere back into the drywell.

7.3.2.6 Drywell temperature normal recovery. Drywell atmosphere ~

temperature response is shown in Fig. 7.8. The initial rate of increase is very rarid because the drywell coolers are lost at time zero, and about half of the 100% power drywell heat load is assumed to remain. The rate of increase becomes much slower when drywell temperature is high enough

- - - - ~ . . .

l i

SUPPRESSION CHAMBER TEMPERATURES (OC) 9 9 9 9 7p 9 9 190 110 SUPPRESSION CHAMBER TEMPERATURES ( F) .-

n 90 110 130 150 170 190 210 230 250 I h l l l -

f.

e WINITIAL RAPID INCREASE DUE TO INFLUX FROM Qy DRYWELL HEATUP on E* 8 8? -

0" n

k *-- SLOW INCREASE WHEN QUENCHING

• k[ 8 ONLY RCIC TURBlfE EXHA3r

• l g

-[

RAPID INCREASES DURING QUENCHING OF SRV DISCHARGE \

k b -.

S a -,

4 / 8 r

. g

~

y a s a \ .

s .

& E," 9 I s

g..

g a m ,- .

~ f '

$. 8 i g g . .

@ dd8  % K 8

a 8z5

,pm QS e

M,>o aS 85e

% P'O Q

$ $$I

=>5 I z 7

589m O

8 O

gg mgy ____ ,

>El 7 80 ) t

• I E

$E

' \ O 4V

s ORNL-DWG 81-8496 ETO 8

g.

s ri

-DEPRESSURIZATION G E BEGINS 9 9 y y DRYWELL DESIGN TEMPERATURE 3 u

> S. >3 f

~ ~

/

g w

g w

1. N /

, , .... ............ ......... .......... N......... .y....... ......... . . . . . . . . . , .........

! ! 1 7 ^#

w m

y w

ch /

E DEPRESSURIZATION g $ COMPLETED 3

00 O O 2 2 r8 s

.J a

<k

.J

_a w w m

m O O 8 *----LOSS OF DRYWELL COOLERS

g. ,

8

.O 30 60 90 120 150 '180 210 240 ,270 300 TIME (min)

Fig. 7.8 Normal recovery after station blackout - drywell temperatures.

ORN L-DWG 81-8497 ETO o

h- l l

l DRYWELL ATMOSPHERE EXPANDING iI SUPPRESSION CHAMBER ATMO ~

INTO SUPPRESSION CHAMBER VIA SPHERE EXPANDING INTO DOWNCOMER PIPING

_ l DRYWELL VIA VACUUM BREAKERS f El- _

.2 I

{

8 ER ~ l m

X w l

$ b

= s l 3 g I cc D- >-

a. z l f y g

$ $ -#! /

w z DRYWELL _

/"

! E \ /

$ \ / /

o8

{O \ f

^

/

7 U

C!- d

- f f \ ,

SUPPRESSION CHAMBER A= O

~

0 30 60 90 120 150 180 210 240 270 300 TIME (min)

Fig. 7.9 Normal recovery after station blackout - containment pressures.

50 to establish natural convection heat transfer to the drywell liner (see discussion in Sect. 4). Drywell temperature begins decreasing af ter about 60 min. because a reactor coolant system (RCS) depressurization is begun at this point. The depressurization reduces saturation temperature within the RCS, hence surface temperatures of the reactor vessel and piping.

7.4 Loss of 250 vde Batteries ,

7.4.1 Loss of 250 vde Batteries -- Assumptions Assumptions specific to programming and input preparation for the Loss of 250 vde Batteries calculation include:

1. When the batteries are lost, injection capability is lost; in addi-t ion, the SPUs can be opened only by automatic actuation.

2. Under automatic actuation, the lowest set SRV will open at'about 7.72 MPa (1120 psia) and close at about 7.38 MPa (1070 psia).

3. The calculation begins 4 h after the Station Blackout and proceeds to about 8 h.

4. Initial values of system parameters, except reactor vessel level, are equal to those calculated at the 4 h point in the normal recovery re-sults (see Figs. 7.1 through 7.9). Reactor vessel level is assumed to begin at 14.2 m (558 in.), the midpoint of the control range as-sumed for the normal recovery calculation.

7.4.2 Loss of 250 vde Batteries -- Results ,

Results for the Loss of 250 vde Batteries case are shown in Figs.

7.10 through 7.16. Each system variable is discussed below. .

7.4.2.1 Reactor vessel pressure. Reactor vessel pressure is shown in Fig. 7.10. During the first 90 min. , pressure climbs slowly f rom the 0.86 MPa (125 psia) initial value to the 7.72 MPa (1120 psia) actuation point of the lowest set SRV. The rate of increase is slow because the de-cay heat af ter 4 hours4.62963e-5 days <br />0.00111 hours <br />6.613757e-6 weeks <br />1.522e-6 months <br /> is small compared to the thermal inertia of liquid water within the vessel.

Af ter repressurization, the pressure is controlled between 7.72 MPa (1120 psia) and 7.38 MPa (1070 psia) by one SRV, actuating automatically.

As shown in Fig. 7.11, no steam flows from the vessel until pressure re-aches the SRV automatic actuation setpoint.

7.4.2.2 Reactor vessel level. Reactor vessel level is shown in Fig.

7.12. No injection is provided at any time in this transient, since DC power is assumed lost.

The initial response of level is a brief but rapid decrease caused by closing of the previously open SRV at time zero. Within several minutes level begins slowly rising aad peaks at about 15.88 m (625 in.). Level .

swells during repressurization because the large mass of water in the ves-sel is heated from about 165*C (330*F) to about 288*C (550*F) with about a 25% increase in specific volume, and there is no coolant loss (other than the assumed leakage) during repressurization.

. . . . , e OFINL-DWG 81-8498 ETD m-g . _ _ _ _ _ . m _ _

..7,..,, ,

pxx,,,r, - - -

STEAM PRESSURE CONTROL VIA h s_ g AUTOMATIC ACTUATION OF e~ LOWEST SET SRV

-CALCULATIONS STOPPED e-0 2~ z u.

b m m /

m 3

eo 38 a

\

$ *~ M 'NO SRh OPERATION AS E E REACTOR VESSEL w

a REPRESSURIZES n-@ w 2 2 b r- b u by

/

c. - O

/

wLOSS OF E - de POWER

- REMOTE-MANUAL SRV OPERATIONAL CAPABILITY

- RCIC OPERATIONAL CAPABILITY o- o l l l l l l 240 270 300 330 360 390 120 450 480 510 540 TIME (min)

Fig. 7.10 Loss of 250 VDC batteries - reactor vessel pressure.

e t

+

52 J

r 9 U .

8 m 3 z i d, p 2 o o .

E s

$r o

ur a

O

_a z

ao

<F M

W c Um 4*

O Q

o u

m O  ;

a- -  :>

z 0 --

a e 2 m a>-

u cc 5

ah C

g r1 y E 9w~

-m hws &

e

a I omzm I tc 8 < v> * 'Baozw <s " 3rES 5 mmg<E sa 2 = 6auw' swo$m e a C "Owu $ ". w <sf z > O cc o o I OSZ StZ 002 541 ost szt c0! sz es st o (5/91) Mold WV31S 13SS3A oil oct 06 0s 04 09 CS 06 ct ct Lt 0 (s,61 ) Mold WY31S 13SS3A . i f i r i ORNL-DWG 81-8500 ETD 8 l

g. TOP OF CONTROL ROOM INDICATION 8

BOILOFF INITIATED BY DISCHARGE 2- [ THROUGH LOWEST SET SRV m. b ..-.y-... ..-...-- / /- ....... ..... ..- ......... .................. ......... . . . . . . . . . ...........

• - ,8 _

j LEVEL SWELL DUE TO HEATUP OF g E .5

• VESSEL WATER a  ; .......y.........L........[......... ..... .. .................. ......... .......... .........

w w A w SRV CLOSES RCIC INJECTION CAPABILITY LOST \

3. " ~@R , i a .a l l M n. w M BOTTOM OF CONTROL ROOM INDICATION U w

> >g \ 8 9- \ TOP OF ACTIVE FUEL

m. g n

ma k 210 270 300 330 360 390 420 150 480 510 540 TIME (min) , Fig. 7.12 Loss of 250 VDC batteries - vessel level. 54 Af ter repressurization, the SRVs begin to discharge steam so level decreases throughout the remainder of the transient. Fuel is beginning to be uncovered at the end of the transient. 7.4.2.3 Suppression pool level. The suppression pool level is shown in Fig. 7.13. During the first 90 min. , there is no SRV discharge, so pool level is constant. Af ter SRV discharge resumes, the pool level be- , gins increasing steadily. 7.4.2.4 Suppression chamber temperatures. Figure 7.14 shows sup- - pression pool and suppression chamber atmosphere temperature. Pool tem-perature, like pool level, is constant for the first 90 min. and then be-gins a steady increase. Atmosphere temperature continues to rise during the first hour due to heat transfer and evaporation from the pool sur-face. Atmosphere temperature rise accelerates slightly between the first and second hours due to an influx from the expanding drywell atmosphere. Af ter about 2 hours2.314815e-5 days <br />5.555556e-4 hours <br />3.306878e-6 weeks <br />7.61e-7 months <br />, the mass interchange ceases, and the suppression chamber atmosphere temperature is responding solely to the increasing pool temperature. 7.4.2.5 Containment pressures. Figure 7.15 shows containment pre-ssures. During the first 90 min., drywell pressure increases rapidly due to heatup of the drywell atmosphere during repressurization. This brings drywell pressure from slightly below suppression chamber pressure to about 0.0125 MPa (1.8 psi) above, allowing some drywell atmosphere to expand into the suppression chamber. Af ter about 2 hours2.314815e-5 days <br />5.555556e-4 hours <br />3.306878e-6 weeks <br />7.61e-7 months <br />, the drywell and sup- ' pression chamber presvires are increasing independently at about the same rate, with little mass interchange. , 7.4.2.6 Drywell atmosphere temperature. Drywell atmosphere tempera-ture is s.own in Fig. 7.16. During the repressurization, the coolant tem-perature within the reactor vessel increases by over 111 C' (200 F'), thereby increasing the drywell heat source. Atmosphere temperature rises to new quasi-equilibrium values, with increased natural circulation heat transfer from the atmosphere to the drywell liner. Af ter repressuriza-tion, the coolant temperature becomes approximately constant, and the tem-perature of the drywell atmosphere rises more slowly at a rate limited by the thermal inertia of the liner. . e e e o

• ORNL-DWG 81-8501 ETD S

/ 7 - NO DISCHARGE INTO POOL 7 - m p 3 ES x d d 3' INTERMITTENT SRV DISCHARGE @ "" INTO POOL BEGINS WHEN VESSEL , s j PRESSURE REACHES SETPOINT g og OF LOWEST SET SRV $6 9- M 0 M E E t, & 8 8 O- ? T- '? . 240 270 300 330 360 390 420 450 480 510 540 TIME (min) Fig. 7.13 Loss of 250 VDC batteries - suppression pool level. ORNL-DWG 81-8502 ETD k e_ R N G C o @- 09 S x x / h 0 POOL -POOL TEMPERATURE STABLE / < 8- < DURING PERIOD OF NO SRV / 5 5g DISCHARGE j /g/ 2  % ~ o Y y - = =2 - - 2 2 gp  % POOL TEMPERATURE INCREASES WHEN $ z O R- I O ATMOSPHERE . ' "^". * * -HEATUP ACCELERATES DUE TO INFLUX $ $b OF HOTTER DRYWELL ATMOSPHERE 5 g_ 5 -CONTINUED HEATUP BY HEAT ^ y y AND MASS TRANSFER FROM g gg HOTTER POOL 8 s-8~ S ?- 8 R- 240 270 300 330 360 390 120 150 180 510 540 TIME (min) Fig. 7.14 Loss of 250 VDC batteries - suppression chamber temperatures. a o e

• 9 o ORNL-DWG 81-8503 ETD Q

4-W- / ? M- / z 6 1

, /

// X $ SUPPRESSION CHAMBER '/ s M B E u, // / g c:- r a ~' c - cL H [ 1 , 3 E DRYWELL DRYWELL ATMOSPHERE BEGINS EXPANDING 2 < INTO SUPPRESSION CHAMBER VIA { g DOWNCOMERS 8_2 e ~- g 240 270 300 330 360' 390 420 450 480 510 540 TIME (min) Fig. 7.15 Loss of 250 VDC batteries - containment pressures. ORNL-DWG 81-8504 ETO 8 Q. SLOWER INCREASE BECAUSE REACTOR VESSEL TEMPERATURE APPROXIMATELY . CONSTANT - SRV DISCHARGE BEGINS, # SL /. h l l/_ '- G E

• O E y y y ORYWELL DESIGN TEMPERATURE R s. R

<~ < k~ / e r m 'X ,,/ ..... ............ . .. .-.. .......... ......... ......... F...... -........ .......... ......... w w t [ w DRYWELL TEMPERATURE INCREASES x m g .. AS REACTOR VESSEL COOLANT y y TEMPERATURE INCREASES w 2 2 k b~ k d w d R-w h m h m o o

s. _

i .

s-8 210 270 -300 330 360 390 420 450 480 510 540' TIME (min)

Fig. 7.16 Loss of 250 VDC batteries - drywell atmosphere-temperature. 1 , e e e 6

• 59

8. FAILURES LEADING TO A SEVERE ACCIDENT As discussed in Sect. 3, control can be maintained over reactor.ves-sel pressure and water level at' Browns Ferry Unit I for a significant period.of time during a Station Blackout provided the installed equipment

., operates as designed and the operator takes the required actions. The first objective of this Section is to briefly diecuss the other sequences which might occur given significant secondary independent equipment fail- - ures or improper operator actions. Subsequently, the equipment failures which have the potential to directly convert the relatively stable period of level and pressure control of the isolated reactor vessel during the early stages of a Station Blackcut into a Severe Accident will be dis-cussed in detail. The several sequences which are considered most probable in a Station Blackout during the period in which DC power remains available are dis-played in the " event tree" of Fig. 8.1. Each of these sequences is dis-cussed bel'w.o Sequence 1: Following the complete loss of AC power, the operator takes remote-manual control of the RCIC system and maintains the reactor vessel water level within.the limits of the available Control Room indica-tion. Since the RCIC System is being used for level control, the status of the HPCI system is not a factor in this sequence. The operator also controls the reactor vessel pressure by remote-manual operation of the primary relief valves and takes action to begin depressurization of the

• reactor vessel within one hour; this lowers the temperature of the satu-rated liquid within the vessel and thereby decreases the driving potential for heat transfer into the drywell atmosphere. As a result of this de-pressurization, the maximum average ambient temperature in the drywell is limited to 148.9'c (300*F), as discussed in Sect. 3. The operator will be able to maintain reactor vessel level and pressure control for as long as DC power is available from the Unit Battery. This is considered to be the most realistic sequence should a Station Blackout occur, and served as the basis for the plots of thermohydraulic parameters as functions of time which were presented in Sect. 7. This sequence will also serve as the starting point for the degraded accident behavior discussed in Section 9 in which the Station Blackout develops into a Severe Accident when DC power is lost so that cooling water can no longer be injected into the re-actor vessel.

Sequence 2: This sequence differs from Sequence 1 only in that the operator does not act to depressurize the reactor vessel. The average ambient drywell temperature continues to increase beyond the one-hour point, reaching 176.7'c (350*F) about four hours af ter the inception of the Station Blackout (cf. Curve A of Fig. 3.3). This excessive tempera-ture existing over a period of several hours can cause serious damage to , the equipment located within the drywell and a breaching of the primary containment due to failure of the drywell electrical penetration assembly seals. If the DC solenoid operators for the primary relief valves fail because of the excessive temperature, the operator will lose remote-manual relief valve control. If the solenoid operators for the inboard Main Steam Isolation Valves also f ail,. the operator will be unable to depres-surize the reactor vessel through the primary relief valves even after AC OR NL-DWG 81-1544A ETD OPERATOR ACTION OPERATOR ACTION COMPLETE LOSS HPCI SUCTION SHIFTS TO PSP AT TO CONTROL LEVEL HPCI OPE R ABLE TO DEPRESSURIZE OF AC POWER AND RCIC OPERABLE PSI LE VE L = + 7 WITHIN 1 h \ e / N g 8 7 \ @ O Fig. 8.1 Event tree for initial phase of a station blackout in which DC power remains available. h $ $ Y Y 61 power is restored. There is no reason to accept these risks to the dry-well integrity and installed equipment; the operator should take action to depressurize, as in Sequence 1. Sequence 3: Either the RCIC system is inoperable or the operator does not choose to employ it for long-term reactor vessel level control. The HPCI System is operable, and either automatically functions as neces-

• sary to cycle the reactor vessel level between 12.1 and 14.8 m (476 and 582 in.) above vessel zero, or is remote-manually controlled by the oper-ator to maintain the water level within the indicating range, 13.4 to 14.9 m (528 to 588 in.). The operator controls the reactor vessel pressure by remote-manual operation of the relief valves and takes action to begin de-pressuration within one bour. Up to this point, this sequence is similar to Sequence 1, except that the injection of condensate storage tank water into the reactor vessel is by the HPCI System instead of the RCIC System.

When the indicated water level in the pressure suppression pool in-creases to 0.18 m (+7 in.), the HPCI pump suction is automatically shif ted from the condensate storage tank to the pressure suppression pool. De-pending on the initial suppression pool level, the necessary amount of water to cause this shif t will have been transferred from the reactor ves-sel via the primary relief valves at some time between two and four hours after the inception of a Station Blackout. Since the lubricating oil for the HPCI turbine is cooled by the water being pumped, and the average sup-pression pool cemperature will be about 71.1*C (160*F) at this point , an .early failure of the HPCI system is threatened. Sequence 4: This sequence is the same as Sequence 3, except that the operator has recognized the dif ficulties with the elevated suppression pool temperature and has taken action to prevent the shif t in HPCI pump . suction. The operator will be able to maintain reactor vessel level and pressure control for as long as DC power remains available from the Unit Ba :te ry. The continuing increase in suppression pool level will not cause ar i significant problems during the remaining period in which DC power would remain available. It should be noted that pressure suppression pool temperature instru-mentation would be inoperable during a Station Blackout. Operator train-ing and the Station Blackout Emergency Operating Instruction (which does not now exist) must provide for operator understanding of the need to pre-vent a shif t of the HPCI pump suction to the overheated pressure suppres- ! sion pool. l Sequence 5: This sequence represents the case of Station Blackout I with no operator action and no secondary independent equipment failures. The HPCI System automatically cycles the reactor vessel water level be-tween the HPCI system initiation point of 12.1 m (476 in.) and the HPCI j turbine trip point of 14.8 m (582 in.) above vessel zero. Over the long l term, one primary relief valve would repeatedly operate as necessary to l maintain the reactor vessel pressure between about 7.722 and 7.377 MPa i . (1105 and 1055 psig). l There are three problem areas with this case of no operator action:

1. the dif ficultie with high drywell ambient temperature as discussed in l

• connection with Sequence 2,

2. the increased probability of loss of injection capability after the undesirable shift of the HPCI pump suction to the overheated pressure suppression pool as discussed in connection with Sequence 3, and

62

3. the decrease in suppression pool effectiveness caused by tt s highly localized heating of the pool water surrounding the discha ge of the one repeatedly operating relief valve. The possible consequences are discussed in Appendix D.

Sequence 6: This sequence differs from Sequence 5 only in that the HPCI pump suction is maintained on the condensate storage tank, either by operator action, or by fortuitous failure of the petiinent HPCI system . logic. This reduces the probability of HPCI system failure due to inade-quate lube oil cooling during the period of Station Blackout in which DC power is still available. The difficulties with high drywell ambient tem-perature and localized suppression pool heating as discussed in connec-tion with Sequence 5 remain. Sequence 7: The RCIC system is not available because of equipment failure, or lack of operator action. The HPCI system has failed because of a system malfunction. The represents the worst case of loss of injec-tion capability while DC power remains available. A boiloff of the ini- , tial reactor vessel water inventory begins immediately after the complete loss of AC power, leading to a relatively quick core uncovery and subse-quent meltdown. In this sequence, the core uncovery is hastened by a de-pressurization of the reactor vessel either by means of a stuck-open re-lief valve or because of inappropriate operator action; the depressuriza-tion increases the rate of loss of the irreplaceable reactor vessel water inventory. Sequence 8: This sequence is the same as Sequence 7, except that there is no reactor vessel depressurization. The uncovery of the core would begin about one-half hour af ter the loss of AC power. The plant re-sponse to Sequences 7 and 8 will be discussed further in Section 9. The preceding discussion of alternate sequences again shows that a , Station Blackout at the Browns Ferry Nuclear Plant will not evolve into a Severe Accident as long as reactor vessel water injection capability is maintained. The remainder of this section contains a discussion of the methods by which this injection capability might be lost. There are three main areas of concern for the viability of the injection systems: Station Blackout induced direct failure of the HPCI and RCIC systems, a stuck-open relief valve which might reduce reactor vessel pressure below that neces-sary for HPCI and RCIC turbine operation, and the ultimate loss of DC power. Each of these will be discussed in turn. 8.1 Induced Failure of the HPCI and RCIC Systems The design and the principles of operation of the HPCI and RCIC sys-tems are discussed in Appendices E and F, respectively. Each of these systems has the capacity to provide the necessary reactor vessel water level control during a Station Blackout; both must fail to cause a total loss of injection capability. The failure modes considered in this Section will be limited to those induced by the conditions of a Station Blackout. Independent secondary ' failures, whose occurrence during a Station Blackout would be purely coin-cidental, have been considered. in other studies 16 and will not be dis-cussed here. 63 The first threat to the continued operation of the HPCI and RCIC sys-tems under Station Blackout conditions is due to the loss of the Reactor Building cooling and ventilation systems. The major components of the HPCI and RCIC systems are located in the basement of the Reactor Building, where the average temperatures during a Station Blackout would be signif- 1 icantly affected by the temperature of the pressure suppression pool. The . HPCI and RCIC systems are designed for continuous operation at an ambient temperature of 64.4*C (148'F). As was shown in Section 7, the average pressure suppression pool temperature would reach 71.l*C (160*F) af ter

• about three hours under Station Blackout conditions. However, it la ex-pected that the ambient temperature in the vicinity of the HPCI and RCIC systems would significantly lag the increasing suppression pool tempera-ture; it is estimated that the design t,mperature of 64.4*C (148'F) would not be exceeded for at least four hours. In any event, exceeding the de-sign ambient temperature woulu not assure failure of the HPCI and RCIC systems.

A more probable cause of HPCI or RCIC system failure is the automatic isolation of these systems due to the tripping of the temperature sensing circuits designed to dete'e t steam leaks in the system piping. As des-cribed in Appendices E and F, each of these systems has a set of four trip logics with four temperature sensors per logic. The 16 sensors are physi-cally arranged in four groups placed near the HPCI or RCIC equipment, with the foer sensors in each group arranged in a one-out-of-two taken twice trip logic. If the trip setpoint of 93.3*C (200*F) is reached, an auto-matic closure signal is sent to the RCIC or HPCI primary containment in-board and outboard steam supply valves, and the system turbine is tripped. Since the inboard steam supply valve for each of these systems is AC-motor-operated, it would not close, but the system would be effectively isolated due to closure of the outboard (DC-motor-operated) steam supply valve. As shown in Section 7, it is expected that the RCIC turbine would be operated only intermittently during a Station Blackout, while the HPCI system would serve only as a backup in the event of RCIC system failure. Also, the RCIC equipment is not located in a closely confined space, and some natural convection within the reactor building will certainly occur. Nevertheless, it is conceivable that the local temperature in the vicinity of the steam leak detection sensors could reach 93.3*C (200*F) during RCIC system operation af ter the average space temperature has increased to over 60*C (140*F). If this occurs, the resulting RCIC system isolation signal can be overriden in the auxilia ry instrumentation room and the steam sup-ply valves reopened and the turbine trip reset. Thereafter, the HPCI sys-tem and the RCIC system, which are located on opposite sides of the sup-pression pool torus, might be alternately used for reactor vesssel level control so as to reduce the heatup in the vicinity of the RCIC system.* Thus, the tripping of the temperature sensing circuits is not expected to , lead to a total loss of injection capability.

• The HPCI room has not been designed to permit natural circulation and the HPCI system has much more surface area for heat transfer into its surrounding space. Therefore, it is expected that the heatup associated with HPCI operation would be more severe than that associated with RCIC operation.

i 64 Another challenge to the viability of the HPCI and RCIC systems would be posed by the loss of control air pressure during a prolonged Station Blackout. The plant control air system is supplied from three air receiv-ers with a combined capacity of 22.60 m3 (798.0 ft3 ). Under Station Blackout conditions, the air compressors which normally run intermittently as necessary to maintain the receiver pressure in the tange of 0.66 to 0.86 MPa (80 to 105 psig) would be inoperable; the stored inventory of . pressurized air would gradually be lost due to valve operations and leak-age. Control air is used in the HPCI and RCIC systems to operate the steam supply line drain isolation valves. In each of these systems, the two primary containment isolation' valves in the steam line to the turbine are normally open so that the steam supply piping is kept at elevated tempera-tures; this permits rapid turbine startup on demand. The water formed by the stean which condenses in this line when the turbine is not operating is removed to the main condenser via a thermostatic steam trap. The two air-operated drain isolation valves close to prevent the flow of water from the steam trap to the main condenser when the primary containment is to be isolated; the closing signal is automatically generated on low re-actor water level [12.10 m (476.5 in.) above vessel zero]. The steam supply line drain isolation valves fail closed on loss of control air, which would eventually occur during a prolonged Station Blackout. This should not cause serious difficulties for the RCIC system which would be run inte rmittantly so that the accumulation of water in the steam supply piping between runs would not be excessive. However, a sub-

• stantial amount of water would collect in the HPCI steam supply piping if this system is not operated over a long period following the loss of con-trol air pressure. Subsequent initiation of the HPCI system might cause ,

turbine damage. However, it should be noted that the HPCI and RCIC tur-bines are two-stage, non-condensing Terry turbines, which are of very rug-ged construction and are designed for use under emergency conditions, and the operator could take action to run the HPCI turbine for short periods to clear the lines of water as necessary. Thus the loss of control air during a prolonged Station Blackout should not lead to a total loss of in-jection capability. A third challenge to the viability of the water injection systems should affect the HPCI system only. This challenge would occur because of the HPCI system logic which provides for an automatic shif ting of the HPCI pump suction from the condensate storage tank to the pressure suppression pool when the indicated pressure suppression pool level reaches 0.18 m (+7 in.). As previously discussed, this would occur between two and four hours af ter the inception of a Station Blackout, when the pressure sup-pression pool temperature has increased to about 71.1*C (160*F). Since the turbine lubricating oil is cooled by the water being pumped, and the oil cooler is designed for a maximum inlet water temperature of 60.0*C (140*F), the oil would become overheated, possibly leading to failure of . the turbine bearings. This can be avoided if the operator takes action to maintain suction on the condensate storage tank for any HPCI system oper-ation. There is an ample supply of condensate storage tank water avail- . able for injection, and the continuing increase in pressure suppression pool level will in no way threaten the continued operation of the 65 injection systems during the remaining period in which DC power remains available. The effects of increased temperature in the vicinity of the HPCI and RCIC turbines, of the loss of control air, and of an automatic shift of the HPCI pump suction to an overheated pressure suppression pool under Station Blackout conditions have been examined in this subsection. None

• of these Station-Blackout-induced events are expected to lead to a total loss of injection capability by means of a direct failure of the HPCI and RCIC systems, but the operator should be aware of the potential for fail-ure and be prepared to take the appropriate corrective actions when re-quired.

8.2 Stuck-Open Relief Valve Since both the RCIC and the HPCI steam turbines are driven by steam from the reactor vessel, question arises as to whether enough steam pres-sure would be maintained in the event of a st-ck-open relief valve to per-mit the operation of these injection systems. Each of these systems will automatically isolate on low reactor vessel pressure to prevent the escape of large quantities of steam to the atmosphere through the gland seals of an immobile turbine; the trip setpoint is 0.793 MPa (115 psia) for the HPCI turbine and 0.448 MPa (65 psia) for the RCIC turbine. The isolation is similar to that which occurs if the sensed equipment space temperature e reaches 93.3*C (200*F) as previously described. The teactor vessel pressure as a function of time during a Station Blackout in which one relief valve sticks open as a result of the initial relief valve liftings is shown in Fig. 8.2. The points for this figure were calculated using the computer program described in Sect. 4; this pro-gram was also used in the development of the figures included in Sect. 7. The reactor vessel steam pressure shown in Fig. 8.2 decreases very rapidly during about the first six minutes of the Station Blackout. This rapid decrease is due both to the stuck-open valve and to the injection of relatively cold water from the condensate storage tank by the combined operation of the HPCI and RCIC systems at a rate of 0.353 m3 /s (5600 GPM). The operator uses both of these systems as shown in Fig. 8.3 as he strives to bring the reactor vessel level back into the Control Room indi-cating range. Once the level has been restored into the operating range, the operator turns off the 'HPCI system; the subsequent pressure decrease is at a much slower rate. (The mass flow through the relief valves is approximately proportional to reactor vessel pressure, as explained in Section 4). When the reactor vessel indicated level is near the top of the indi-cating range, the operator turns off the 0.038 m /s3 (600 GPM) RCIC sys-tem; this occurs at about 29 minutes into the Station Blackout. The re-sult as shown on Fig. 8.2 is that the reactor vessel pressure begins to increase; more steam is being generated by decay heat within the vessel than the stuck-open relief valve can remove. At about time 48 minutes, the indicated reactor vessel level has decreased to the point where the operator again turns on the RCIC system. This pattern is continued over the five-hour period shown in Fig. 8.2; the steam pressure increases dur-ing the periods when the RCIC system is off and decreases when the RCIC 99 s,, s I 1 - .? i ,, a - 3  ::G e i 8 8 O  % i l 0i W ) E a \ - O -g Y O ~m m E

• Z 5  %

O o ." 9 m / 8 2 e.n n- GO o mW n . g n x -8 9- <[ re z O

• r 42 nO a na k --8 o o

9 m 2

• 37 l

N i 8 .d 00 G* o . 4, o 0011 0001 009 008 007 006 005 004 003 002 001 0 j )aisp( ERUSSERP MAETS ! $ 0  ! i  !  !  !  ! * )aPM( ERUSSERP MAETS t ORNL-OWG 81-8506 ETD o a N- -~ 5600 gpm COMBINED HPCI AND RCIC INJECTION g_ ! M-S" _ g_ g_ N~ 8 ? "E s- y$ 8 @n-x  % o N, o g -.8 o O u e 2- a' l-w O @ S- h H , 8 *- 3 2 ~- ~ o_ HPCI OFF m-600 gpm RCIC INJECTION FLOW o RCIC RCIC

• ~ O' OFF ON OFF ON

.- 4 a / \ ,, n-o- 0 0 $ $ N 120 153 iso 2io 24o 29o 300 TIME (min) Fig. 8.3 Reactor vessel water injection with a stuck-open relief valve during a station blackout. 68 system is operating. The average steam pressure decreases with time as the power generation due to decay heat slowly decreases. As shown in Fig. 8.2, the average reactor vessel steam pressure will have decreased to the point where continued itPCI system operation is ques-tionable [0.793 MPa (115 psia)] after about two hours of the Station Blackout. Howcser, the RCIC system should remain operational for over five hours. The minimum pressure reached during this period would be about 0.58e ppa (85 psia) whereas the RCIC system is operational at steam

• pressures as low as 0.448 MPa (65 psia).

It can be concluded that reactor vessel level control can be nain-

• tained for at least five hours (or until DC power is lost) during a Sta-tion Blackout with one stuck-open relief valve.

8.3 Loss of 250-Volt DC power There are eight 250-volt DC battery systems at the Browns Ferry Nuclear Plant. Each system consists of a 120 cell lead acid battery, a battery charger, and the associated distribution equipment. Four of these battery systems provide control power to the four 4160-volt AC shutdown boards and would not be used during a Station Blackout. The fifth 250-volt DC system provides power for common plant and transmission line con-trol f unctions and would supply drive power for a 120-volt AC plant pre-ferred motor generator set during a Station Blackout. The 120-volt AC plant preferred system loads include common plant equipment such as the communications room, the CO2 fire protection system, the sequential events recorder, the computer clock, and the stack gas monitors. Each of the Browns Ferry units is provided one of the three remaining- ~ 250-volt DC battery systems as a source of power for unit control func-tions and for certain unit DC motor loads. During a Station Blackout, each unit battery would also provide drive power for a 120-volt AC unit preferred motor-generator set. The 120-volt AC unit preferred system loads are listed in Table 8.1. The major 250-volt DC loads which would be supplied by each of the unit batteries during a Station Blackout are listed in Table 8.2. It is possible to feed the 250-volt distribution system of one unit from another units' battery, but there is no provision for supplying DC power into any of the unit distribution systems from the shutdown board control power or plant battery systems discussed in the preceding paragraph. A total loss of injection capability at each unit will certainly oc-cur during a prolonged Station Blackout at the time when the unit battery, which supplies 250 volt DC logic and valve-control power to the RCIC and HPCI systems becomes exhaus;ed. The design basis for the Browns Ferry 250V-DC Power Supply and Dis-tribution System provides that:17 " Battery capacity shall be adequate so that any , two unit batteries can supply for 30 minutes, without chargers available, the DC power required to operate the engineered saf eguards systems on any one reactor , unit in the event of a design basis accident as well as the DC power required for the safe shutdown and cooldown of the other two units with a final terminal voltage of 210 volts." L. 69 Table 8.1 Urit preferred system loads

1. Containment Isolation panel

2. Feedwater Control panels

3. Reactor Manual Control panel

4. Drywell Ventilation and Reactor Building Closed Cooling Water System Control panel

. 5. Feedpump Turbine Control panels

6. Rod Position Information System

7. EHC Control Unit

8. Unit Computer and Rodworth Minimizer panels

9. Reactor Water Cleanup panel Table 8.2 Major unit 250-volt DC loada

, 1. Turbine Building Distribution Board

2. 480-volt Shutdown Board Control Power

, 3. Emergency DC Lighting

4. Unit Preferred AC Motor-Generator

5. Circuit Breaker Board 9-9 (Feedpump Turbine Controls)

6. Reactor Motor-Operated Valve Boards a) Primary Relief Valve DC Solenoids b) Main Steam Isolation Valve DC Solenoids c) Recirculation Motor-Generator Set Emergency 011 Pumps d) Backup Scram Valves e) RHR Shutdown Isolation Valves f) Engineered Safeguards Logic Power Supplies g) HPCI System Controls, Valves, and Auxiliaries h) RCIC System Controls, Valves, and Auxiliaries

70 "The engineered safeguards syste=s that are sup-plied from the 250 volt DC syste= shall be designed to operate at a mini =us of 200 volts." It nas previously been esti=ated5 that under the less severe con-ditions of a Station Blackout, eith all unit batteries avaiable, and as-su=ing prudent actions by a well-trained operator to conserve battery po- . tential by mini =izing DC loads, the necessary 250 volt DC power for HPCI or RCIC system operation would re=ain available during the first four to six hours of a Station Elackout. - In su==ary, it is reasonable to ' expect that control can be =aintained over reactor vessel pressure and water level at each of the Browns Ferry Units during a Station Blackout for as long as the 250 volt DC power re-mains available. This period will depend on the operator's ability to conserve the unit battery potential by minimizing the DC loads, but is ex-pected to last from four to six hours.* During this period the operator =ay have to take other actions to maintain the operability of the HPCI and RCIC systens as discussad in subsection 8.1.

• A battery life of seven houra is considered in Appendix G.

9 O 1 l 71

9. 0 ACCIDENT SEQUENCES RESULTING IN CORE MELTDOWN 9.1 Introduction This section deals with the various accident sequences which might

- occur during a complete Station Blackout (CSB) at the Browns Ferry Nuclear Plant which can lead to core meltdown.18 Event trees are presented for what are considered to be the six most probable sequences. The character-istics and event timing for each of these sequences were determined by use of the MARCH . code, and include consideration of the operator's role in re-actor vessel pressure and level control. The progression of core damage and containment failure following core uncovery will be discussed in this section; an event tree for operator key actions in mitigating the accident progression will be discussed in Sect. 10. Of the se sequence TB'* quences is deemed to modelled by MARCH of be most representative and thepresented in this section, the events which would occur af ter core uncovery following a four-hour period during which the availtbility of DC power permitted reactor vessel level and pressure control. For this analysis, it is assumed that no independent secondary equipment failures occur, and this sequence was chosen to serve as the basis for the fission product transport analysis discussed in Volume 2 of this report. 9.2 Accident Phenomenology 9.2.1 Accident Progression Resulting in Core Melt Upon a loss of offsite and onsite AC power, a number of reactor safety systems respond immediately and automatically. First of all, a full load rejection (i.e., fast closure of the turbine control valves) oc-curs, followed by the tripping of the recirculation pumps and the main condenser cooling water pumps. With the load rejection, the scram pilot valve solenoids are deenergized and the control rods start to move toward the f ully inserted position. The main steam isolation valves (MSIVs) be-gin to close, resulting in a rapid reactor vessel pressure increase. The-se events are closely followed by a main turbine trip (i.e. , closure of the turbine stop valves) and the tripping of the feedwater turbines. Af ter the automatic actions described above, core flow is provided by natural circulation and excess reactor vessel pressure is relieved by steam blowdown through the safety / relief valves (SRVs) into the pressure suppression pool. If any SRV fails to reclose after actuation, the reac-tor vessel will continue to depressurize with an accompanying loss of

• The nomenclature TB' follows that used in the Reactor Safety Study,19 where T denotes a transient event and B' denotes failure to

• recover either offsite or onsite AC power within about I to 4 hours4.62963e-5 days <br />0.00111 hours <br />6.613757e-6 weeks <br />1.522e-6 months <br />. In effect, TB' is a loss of decay heat removal (TW) sequence with the loss of both offsite and onsite AC power as an initiating event.

I 72 water inventory similar to that occurring in a small break LOCA. Accord-ing to the Reactor Safety Study,13 the probability of a stuck-open re-lief valve (SORV) event is estimated to be 0.10 with an error spread of 3. However, this estimate was based upon consideration of three-stage SRVs such as those originally installed at the Browns Ferry site. with the two-stage relief valves of improved design which have a recently been in-stalled at the Browns Ferry Plant, the probability of a SORV event is sub-stantially reduced and is estimated to be less than 10-2 The occur- - rence of a sequence involving a SORV, designated sequence TPB', is in-cluded in the event trees of this section. The reactor vessel water level decreases rapidly during the initial moments of a Station Blackout due both to the void collapse caused by the increased pressure and to the water inventory lost through the SRVs. When the water level sensed by the wide range detector reaches the low water level setpoint (Level 2), the DC powered HPCI and RCIC systems are auto-matically initiated, with their turbine-driven pumps supplied by steam generated by the decay heat. The HPCI and RCIC systems are assumed to re-main operational until the unit battery is exhausted at four hours into the Station Blackout. Following the loss of these injection systems, the reactor vessel level decreases until the core is uncovered about one hour later. Subsequently, the core begins to melt. The failure probability for HPCI has been estimated 19 to be 9.8 x 10-2 with an error spread of 3 and for RCIC has been estimated to be 8 x 10-2 with an error spread of 3. The overall failure probability of both HPCI and RCIC to provide makeup water during the sequence TB', is designated TUB' and has been estimated 19 to be 2 x 10-3 with an error spread of about ,

4. The failure of both HPCI and RCIC together with a SORV occurrence is designated TUPB'.* Sequences TUB' and TUPB' have been included in the event trees of this section. .

9.2.2 Containment Failure Modes Any sequence resulting in core melt will eventually lead to contain-ment failure if electrical power is not restored before the reactor vessel fails. According to the Reactor Safety Study,19 BWR containment failure could occur in the following categories: o - Containment failure due to steam explosion in vessel, g -- Containment f ailure due to steam explosion in containment, y - Containment failure due to overpressure - release through reactor building, y' -- Containment failure due to overpressure-release direct to atmos-phere, 6 -- Containment isolation failure in drywell, c - Containment isolation f ailure in wetwell, c -- Containment leakage greater than 2400 volume percent per day. Containment failure by overpressurization was considered to be the - dominant failure mode for a BWR inerted containment in the Reactor Safety

• TUB' and TUPB' are in effect TQUV and TPQUV sequences with the loss of both of fsite and onsite AC power as an initiating event.

, ( . ns 73 s #1 Study.19 It was postulated that failure would occur when the steel ' - liner and the inner layer of reinforcement were stressed to a level be-tween the yield and the ultimate tensile strength. For this reason, the _, study provided an upper and lower bound for the containment failure pre-ssure [i.e., failure was assumed to occur between 1.34 and 1.21 MPa (195 and 175 psia)]. Other failure modes resulting in high initial leakage from the con-

• tainment have historically been considered to be of less significance from the standpoint of radiological release because they would not lead to gross containment failure by overpressurization and it was believed that the leaked radioactivity would be significantly reduced by the filtration and absorption capacity of the standby gas treatment system.

However, failure modes involving high initial containment leakage have been found to be important for the Station Blackout sequences con-sidered in this study, in which the standby gas treatment system would be inoperable as a consequence of the loss of AC power. Gross containment failure due to overpressure is discounted for the following reacons:

1. Containment failure due to a steam explosion following core melt has unlikely at the Zion plant,20 and recent been found to experiments at be highg{ have shown that corium does not undergo SANDIA violent explosions upon interaction with water. Therefore, contain-ment failure due to steam explosions is not considered in this study.

2. The containment at Browns Ferry has an inert atmosphere. This means it is highly unlikely that the containment would fail as a result of

. overpressurization caused by hydrogen burning, and this failure mode is not considered in this study. Containment failure by deterioration of the drywell electric penetra- . tion assembly (EPA) seals due to high ambient temperature resulting from core meltdown would occur before the containment pressure increase caused by suppression pool heatup reached failure proportions. Accordingly, EPA seal deterioration has been identified as the dominant failure mode in the BWR.18,22 However, for sequences in which excessive amounts of superheated steam and noncondensibles are discharged into the suppression pool within a short period of time, the wetwell might fail before the dry-well due to forces of steam jet impingement and condensation oscilla-tions.18 These forces are discussed further in Appendix D. In the case of failure of the drywell EPA, the initial high leakage would not prevent an ultimate gross containment failure. The drywell tem-perature would continue to increase, causing the EPA seals to further de-compose and to finally be blown out of the containment wall. If, on the other hand, failure occured in the wetwert, radiological consequences from the releases wculd be less than those from a drywell failure because a large fraction of the tission products would have been dissolved or deposited in the suppression pool and therefore not released' from the containment.

• As will be shown, the postulated mode of containment failure by de-composition of the electrical ~ penetration seals due to overheating would . -

occur at a pressure approximate'ly 30% less than that considered,necessary # , for containment failure in' the Reactor Safety Study.19 , , A h A ._t f) 74 9.3 Event Trees '9.5.1 Event Trees for Accident Sequences Resulting in Core Melt . It has been concluded in this study that the sequence of events dur- . 'ing a prolcaged complete Station Blackout -(CSB) at the Browns Ferry Nu-clear Plant (BFNP) would most probably be described by one of the follow- . ing titles:

1. CSB + RPC1/RCIC (TB')

2. CSB + HPCI/RCIC + SORV (TPB') *-

3. CSB + Manual RCIC & SRV (Ty B')

4. CSB + Manual-RCIC & SRV + SORV (Ty PB')

5. CSB + No HPCI/RCIC (TUB')

6. CSB + No HPCI/RCIC + SORV (TUPB')

As indicated in this listing, a stuck-open relief valve (SORV) corresponds to a small break LOCA. The terminology " Manual RCIC & SRV" refers to the assumption that the operator controls level and pressure by remote-manual operation of the RCIC system and the relief valves, respectively. The phrase "No HPCI/RCIC" means that both of these water-injection systems are assumed to be unavailable due to mechanical failure. The event tree identifying these six accident sequences is shown in Fig. 9.1. Sequence No. 1, traced by the dotted line in this figure, is the sequence TB' which involves no independent ' secondary equipment fail-ures and was selected as the basis for the fission product transport anal-

n. ysis, discussed in Volume 2. Other possible but much less likely sequences M such as failure of reactor scram are shown on Fig. 9.1 but were not con- ,

r sidered in this study. 9.3.2 Core Damage Event Tree Without re-toration of offsite or onsite AC power and with the con-sequent inevitable loss of DC power, any'of the numbered sequences-shown on Fig. 9.1 will lead to core-uncovery and subsequent core' melting, reac-tor vessel failure, aid corium-concrete interaction. A tree for the -c events subsequent to core uncovery is shown in Fig. 9.2, with the path used for the fission product transport analysis of Vol. 2 identified by a dotted line. 9.3.3 Containment Event Tree A containment event tree for the occurrences following core melt is given in Fig. 9.3. The particular path which has been chosen for the fis-sion product transport studies is again indicated by the' dotted line path. It should be noted that if AC power is restored af ter core melt, but before a failure of the reactor .*essel occurs, the low pressure Emergency Core Cocling Systems (ECCS) wo 1d function and a breach of the containment . ' in the drywell or wetwell areas r.ay be averted. ..e s L , i , 1 l o' l b f~' - ,ff. '. . ,r ; h, l. ~ { _l w _+ n-,. ..-- - - = ~ , , - w - '* " ' " " ~ ~ ~ ~ ' ' ' ' ' ' ,l l'll .l 'I , [ yU / >,I' , 6 / Q 1, b D D D E E E r i y . Y E Y E Y E E ES A E C AG A L G A L G A A L G A RW / , 4 6 N F) A E A E E OT f 1 8 E DM DM DM A DM A CA U / ,(T A T A T T 1 8 Q . L A D L L D A T L L LD A T L LD A L L - , E E I E E I E E E I E E I E E AI E G W S N M T R M TR M M T R M T R M TG D N O NO E NO NO NA O E E E E E E E E E E - C } R T C R T C R R T C R T C R T M <N L O O O O O O O O O O OA PJ <R C P C P . C C C P C . , t O E ") l C # B 1 'B) 1 1 1 B e o N P, B P 8, , E / P U B U t T T T U Q ( ( T( T( t T( e r ? E S h h @ @ h - @ o c i e - , <. - n _ i / r . - g e / - n i 1 , - - t l - u s - e , r l - s ' l l l r e N c O - n VTn. I e ram e - u S2 q LR1 l5 - e , s A UN USI NSH - t AET n M EPW RI - e d D _ i c it O E - c 'N CB L - a AI A r ICR - o CR P E P f H O -, 3 = e NV E e EL - t r PA OV - t KF n CE - e t UIL T SER - v E ,l slli CS - I E MTS EAN - 1 TM SOPO - 9 S Y TS . UE AR - g -r! lsg i F TET - NI ES E - VFTR EFSE I ONW S - s 6 G NF OO P E Y - VON TIOD 5Na c AISA - TO e I L N - I _ , ,l !ll ll l1'l1Il l! l Onset-gnnG ei-sane ETO de pow &R RESTORED OFFSITE ao POWER RESTORED W 8587 88 R I AFTER $4GNIFICANT AFTER $#GNIFICANT dr POWER RESTORED ' INtTI ATING EVENT. RE ACTOR VESSEL SEFORE EXTENSIVE CORE MELT. BUT CORI MELT. RUT OFF3:Yi as POntR RESTORED AFTER VE$5EL CORE UNCOvf RY IDAMA OUTCOest DEPRESSURl2ED CORE DAMAGE . LOW PRE $$URE Eh 8 FOREVE55EL OEFORE VE5SEL FAILURE LP ECC$ (ftRASLE MPCURCIC OPE R ABLE WERABLE F AILURE - F AILuRE-WURCIC OPE RA8LI LP ECCS DPE R ASLE SLIGHT CORE DAMAGE. LITTLE On NO CORE wtLT CORE MELT. DEGRs5 CONT Asht D Im V1SSE L YES O CORE MELT. RPV F A# LURE, NO CONCRf TE ATTACK N 8 Cont MELT.RPV @ ' FAJLURE. AND CONCRfTE , ATTACK e '. SLsGMT CORE DAWAGE. V e e LITTLE OR NO CORE vf LT NO e a '..........= CORE uf LT. DESRis a COe:TAsNED im yEssEL .................................'s CORE WILT.RPV e l F Astunt. NO CONCRE 78 '..............................J ^"ac" . e.................................... CORE .5tT.R,, F AILORC. Aho CONCRETE A' TACK Fig. 9.2 Core damage event tree. . e . . . . 77 m - ,. LA#GE nesPLITUDE " 'L " " ' - umTtarwo3 tyget Co=0E auBYM3st T-cue most a vis5E L 0#fetLL De mtLL OUTConst m e:Ett OEC Lav e omgwel g p estwa t E*a vanteesG epa ranuet , 4 se<m Ritta&E tef fMLL pagLyegg v5 AT *e6 Gas PELEaEE EDeYwtLL Fattuou e toast SELEau 408veELL wttYe8uG) vf RT ewGee #f lea 5E tLov'atLL pa: Lua n esO COpetan%esibf f aaLu#t to vt35EL f 4 Luet eso CONTA' host %f A.Lu i

• • • GW #E lea 54 W Tat LL Fa Luet e W E #' se' Gee AS LE a.% 40e ret L L p aLuett goest mELEa$4 #0AvetLL vt4 Tie G e VE3 vt#T weGes #Etta58 iDerwtLL pastumo O

.e Co.81a .,.a.Lu.i - - - - - - - - .so vesu L .a Lu .so co=, .sw.r d aiku*f l . i I vtav Gn actsau soevenu sasuats I v e sout esttasa toevesu vtsrwers ,,o , I , vsev e.Ga esteasa scaveau eanuati 1 a seO CC9 eta #4etts e ? $astunt g i 1 no visu L eamas. iso centaewaar a I asLU81 g I .. u , a,t v , L,_____,, g . . . . .. . .... .... .. ... v .v ,esGee ., a au ,o ' saw .itsau ,o.<=u v1 1,.eG, l I ,....... E . vt#v *** Gas REggast soaywtLL partuats g g B g E- - - - - . . - - - - _ - seO CCestat%est=T f &#LumE esQ vl35(L Fattuet. as0 CDesTarisestti Fattuet Fig. 9.3 Containment event tree. . _ . .-- . ..~ - - .. . . . _. 78 9.4 Accident Sequences 9.4.1 MARCH computer code Detailed calculations fpr the accident sequences have been performed - with the MARCH computer code ', version 1.4B. This version is based-on version 1.4 at Brookhaven National laboratory, but contains a different subroutine ANSQ for the decay heat power calculations and modifications to i' the subroutine HEAD which were necessary to improve the prediction of ves - ., i sel failure time.18 In the MARCH code, the fission product decay heat source term is based on ANS Standard ANS-5.1 (1973)23 which does not account for the 7 decay of U-239 or Np-239. The new standard ANS-5.1 (1979)*" provides ., decay parameters for U-239 and Np-239, but does not account for the decay power generated from these or the numerous other heavy nuclides (acti- ~ i nides) which exist in power reactors. The new subroutine ANSQ used in this study (Cf. Appendix B) is based on the actinide decay heat source in 1 a BWR following a depletion of 34,000 mwd /t. The actinide heating calcu--

l. .lations25 were performed using the EPRI-CINDER code and include all sig-4-

nificant actinides from T1-208 through Cm-246. Furthermore, the initial fission energy release from the fuel is assumed to last 3 seconds rather than 5 seconds as used originally in the MARCH code. This_has been found to yield better agreement with test data in implementing the REDY code at-GE.zb A comparison of MARCH calculations for the sequence TB' as per- . f ormed by versions 1.4 and 1.4B are given ir. Figs. 9.4 through _9.11 for the time- dependence of (1) the maximum core temperature, (2) the water- " steam mixture level, (3) the containment wall temperature, and (4) the . containment volumetric leak rate. It should be noted that the times pre-dicted by version 1.4B for core uncovery and core melt are less by approx.- imately 18 and 26%, respectively, and that the maximum containment wall temperature is increased by more than 120%. Other comparisons have shown that the 1.4B version produces results in better agreement with the core uncovery time calculated by the RELAP4/ Mod 7 Code for the sequence TUB ' at INEL. 27 j 9.4.2 Accident prograssion signatures This section contains the accident progression signatures for the l, six most probable sequences following complete station blackout (CSB). . The early part of the plant transient characteristics follow that con-tained in the Browns Ferry Nuclear Plant Final Safety Analysis Report. All other accident signatures have been based on the results calculated by the MARCH computer code.14 9.4.2.1 CSB + HPCI (TB'). The accident progression signature for

• the sequence TB' is given in Table 9.1. This sequence has been chosen

^ for the fission product transport studies presented in Volume 2. 9.4.2.2 CSB + HPCI'+ SORV (TPB'). The accident progression signa-ture for the sequence TPB' is given in Table 9.2. The SORV event has the same ef fect as a small break LOCA, ensuring that the reactor vessel 4 4. 4 ORNL-DWG 81-8515 ETD 3000.0 CORE MELTING BEGINS 2500.0-2 E o 2000.0 - 5 a. 2

1. 1500.0 -

E 8 g s 3 1000.0 - E DECREASE FOLLOWING SCRAM X 4 CORE UNCOVERY BEGlhS 2 500.0 - de POWER LOST O.0 0.0 50.0 1$0.0 1$0.0 2bO.0 2$0.0 3b0.0 3$0.0 400.0 TIME (min) Fig. 9.4 Maximum core teEPerature (MARCH / MOD 1.4B). ORNL-DWG 31-8516 ETD 3000.0 CORE MELTING BEGINS 2500.0 - t _2 m $ 2000.0-e a: E E

1. 1500.0 -

E o 8 O s $ 1000.0- DECREASE FOLLOWING SCRAM x < CORE UNCOVERY BEGINS-2 i 500.0 - de POWER LOST-0.0 - , , , , , , , , , 0.0 50.0 100.0 150.0 200.0 250.0 300.0 350.0 400.0 450.0 500.0 TIME (min) Fig. 9.5 Maximum core temperature (MARCII/ MOD 1.4) . ORNL-DWG 81-8517 ETD

12.0 l

10.0 - LOSS OF dc POWER [ BOILOFF BEGINS x 4 E,, 8.0 - _J m m m 6.0-C o . H X 4.0 - TOP OF CORE 3 2 m 4 e m M 2.0-e m Q ROTTOM OF CORE , 0.0 - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -- CORE SLUMPS -2.0 - -4.0 , , , , , , , 0.0 50.0 100.0 150.0 200.0 250.0 300.0 350.0 400.0 TIME (min) Fig. 9.6 Water-ateam mixture level (MARCH / MOD 1.4B). 4 9 c" o D 0 T 0 E 0 - 5 8 1 5 8 1 8 0 G - - S 0 W P i5 C L - - - M U 4 . N L R - - S O E . R 0 - - O 0 C i0 4 - - O - - ) 4 0 R - ,5 0 1 E S 3 D WN OI - - O M P EG / e d B - - 0 H C R F FF - 0 A i0 M OO SL - 3 ( SI - l e OO LB - ) n ve - 0 im l [ - ,5 E 0( 2M I T t e r u x i - - 0 m . m ,0 0 2 t a e s r e - - 0 t a 0 W ,5 1 7 - - 9 0 . - - ,0 i g 0 F - E- 1 7 E- C O-R F O 0 C- O- i0 M 5 F-0 O-T T 9-O O- . T- 0 I B- 0 0 0 0 0 0 0 0 0 0 2 0 8 6 4 2 0 2 4 1 1 E_Jw>$ wEDH5E 3<wt aw&<3 .l

• i

ORNL-DWG 81-8519 ETO , 1400.0 t ' 2

1200.0 -

c 0 O ', Z 1 1000.0 - P oc E 3 800.0- HEATUP DURING CORIUM-F CONCRETE REACTION d

• 3 z 600.0 -

m lE GROSS EPA FAILURE E $ VESSEL BOTTOM HEAD FAILS z o 400.0- EPA VENTING 200.0 , , , , , , , , , , 0.0 100.0 200.0 300.0 400.0 500.0 600.0 700.0 800.0 900.0 1000.0 1100.0 TIME (min) Fig. 9.8 Containment wall temperature (MARCH / MOD 1.4B). l OllNL-DWG 81-8520 ETD 600.0 S 550.0 -

=

5 0 Z E 500.0 - HEATUF DURING S CORlUM-CONCRETE g HEACTION r b 450.0 - H ~. ! $ '

• VESSEL BOTTOM HEAD FAILS /

l it m 400.0 - 2 8 350.0 - __ ___. . / s' 1' 300.0 . . . . . . i i . . i O.0 100.0 200.0 300.0 400.0 500.0 600.0 700.0 800.0 900.0 1000.0 1100.0 1200.0 TIME (min) Fig. 9.9 Containment vall temperature (MARCH / MOD 1.4). i O D 0 T 0 E 0 1 1 1 2 5 8 1 8 0 0 G ,0 W 0 D 1 L N R 0 O 0 . ,0 9 0 0 ,0 . 8 ) B 4 1 0 0 D 0 O ,7 M A 0 / H C R A M 0 ) n ( ,0 L 6 im ( t e a f 0 E M I 0 T k r a ,0 e 5 l t . n e 0 m - 0 0 i n S a L 4 t A n E o S C A 0 P 0 0 E ,0 1 F 3 9 O E g R 0 i U ,0 0 F L I 2 A F S S 0 O 0 R ,0 G 1 0 0 - - - - ~ 0 0 0 4 0 0 5 3 0 0 0 3 0 0 5 2 0 0 0 2 0 0 5 1 0 0 0 1 0 0 s 2E$- = N < m x b J oE'":3O>5o D 0 T E 0 0 2 2 1 2 5 8 - 0 1 8 0 G 0 1 W 1 D L 0 N R 0 O 0 0 1 , 0 0 0 9 b ^ ) 4 0 1 0 i0 D 8 O M / H 0 C S 0 R L ,0 A A 7 M ( E S ) n e t A 0. i a P E ,0 0 (m r E k F 6 O M a I T e , E l R 0 t U 0 n L ,0 e . 5 m IA n F i S a S 0 t O ,0 n R 0 o 4 C G 1 1 0 ,0 0 9 3 g i F 0 ,0 0 2 0 0 ,0 1 0 . 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 5 0 5 0 5 0 5 . 4 3 3 2 2 1 1 225 W<crMbaocbs3O zo I ' i W 4 -a i.-. e L+A,+4+Y' -. 87 Table 9.1 Browns Ferry Nuclear Plant: Complete Station Blackout Sequence of Events CSB + HPCI/RCIC .. (TB') , . Time (sec) Event O.0 Loss of all AC power and diesel generators. The plant is initially operating at .100% power. Initial drywell temperature - 66*C (150*F)

Initial wetwell temperature = 35'C . (95'F)

<, 0.2 Full load rejection (i.e., fast' closure of turbine control valves) occurs. -0.2 Recirculation pumps and condenser circulatory water pumps trip off. Loss of condenser vacuum occurs.

Core flow is provided by natural circulation.

. 0.2 Reactor pressure increases suddenly due to load rejec-

. tion. a ~ 0.3 Scram pilot valve solenoids are deenergized due to i -load rejection. Control rod motion begins. 0.5 Turbine bypass valves start to open due to load rejec-l tion. -1.0 Neutron flux starts to decrease after an initial in-crease to over 100% rated power level. 1.0 Reactor power starts to decrease slowly af ter an ini-tial rise. 1

2.0 Control rods are 40% inserted from fully withdrawn position. 4 2.0 Main steamline isolation valves (MSlVs) start to close

, (relay-type reactor trip system), result _cg in a rapid i steam-line pressure rise. 1 . , 2.0 Turbine. bypass valves are tripped to close. i i . 3.0 Control rods are 75% inserted from fully withdrawn

position.

I i i i i . _ , . , , . , - , , - - ,_.- ,,,,...,...._.._...-._......_-....-m__--- . . . . , - - - , - - . . . - - _ . . , - ,, . - ~ . 88 Time (sec) Event 3.0 Turbine trips off (turbine stop valves fully closed). 3.5 Power generation due to delayed neutrons and fission product decay drops to 10% of initial rated power gen-eration. . 4.0 Feedwater turbines trip of f. 5.0 MSlvs are fully closed, result'ng in a momentary 0.69 MPa (100 psi) pressure ' acrease and 1.02 m (40-in.) drop of water-steam mixture level due to collapsing of voids. 5.0 All control rods are fully inserted. 5.0 Reactor vessel pressure exceeds the lowest setpoint at 7.52 MPa (1090 psi) of safety / relief valves (S/RVs). 5.0 Sev eta (1) out of thirteen (13) S/RVs start to open in response to pressure rise above the setpoint. 5.2 Water-ste m mixture level recovers 0.51 m (20 in.) from the previous momentary 1.02 m (40-in.) drop. 5.5 S/RV steam blowdowns into the pressure suppression pool through the T quenchers begin. , 7.5 Feedwater flow drops below 20%. 9.0 Feedwater flow decreases to zero. 10.0 Power generation due to fission product decay drops to approximately 7.2% of rated power generation. 15.0 All 7 S/RVs are completely closed. 15.7 Four out of 13 S/RVs start to open. 17.0 Neutron flux drops below 1% of initial full power level. 21.0 Narrow range (NR) sensed water level reaches low alarm (Level 4), i.e. , 5.98 m (235.50 in.) above Level 0, or 5.00 m (196.44 in.) above TAF. , 22.0 Suppression pool water average temperatut

• rises to 35.13*C (95.24*F) in response to the first S/RV .

pops. 29.0 All 4 S/RVs are completely closed. 89 Time. * (sec) Event 1 29.7 Two out of ' 13 S/RVs start to open. 47.0 All 2'1/RVs are completely closed.

• 47.7 One out of 13 S/RVs starts to' open.

56.0 Suppression pool water average temperature is approxi-mately 35.3*C (95.54*F). 56.0 NR sensed water level reaches low level alarm' (Level 3), i.e., 5.50 m (216.00 in.) above Level 0, or 4.50 m (176.94 in.)'above TAF. 90.0 Suppression pool water average temperature is approxi-mately 35.4*C (95.72*F). 101.0- The S/RV is completely closed. The'same.S/RV contin-ues to cycle on and of f - on setpoints throughout 'the subsequent cyclings of HPCI and RCIC injections. 625 Wide range sensed water level reaches low water level setpoint (Level 2), i.e., 4.18 m (164.50 in.) above Level 0 at 2/3 core height, or 2.96 m (116.50 in.) above TAF. .- 625 HPCI and RCIC systems are automatically turned on. The HPCI and RCIC turbine pumps are driven by steam generated by decay heat. System auxiliaries ... powcred by the 250 V de system. , 655 HPCI and RCIC flows enter the reactor pressure vessel at 315 t/s (5000 gpm) and 381/s (600 gpm), re-r spectively, drawing water from the condensate storage tank. 12.5 min. Narrow range sensed water level reaches Level 8 set-point, i.e., 6.86 m (270.00 in.) above Level 0, or 5.64 m (222.00 in.) above TAF. e 4 + - - , - - - , +--- , . , - - - . , , , , - --.-----.--,-n, -- -.e .. - y ,- - y - 90 Time (sec) Event 12.5 min. HPCI and RCIC trip off. 20 min. Drywell and wetwell temperatures exceed 70*C (158'F) and 42*C (108'F), respectively. , 26.5 min. Wide range sensed water level reaches Level 2 setpoint and HPCI automatically restarts. (RCIC does not automatically . restart.) 27.0 min. HPCI flow enters the RPV. 29.0 min. Narrow range sensed water level reaches Level 8 setpoint and HPCI trips off again. The HPCI system, driven by steam gen-erated by decay heat, turns on and off automatically between sensed Levels 2 and 8 until the batteries run out. 80 min. Auto-isolation signal initiates as increase of drywell pressure exceeds 13.8 KFa (2.0 psi). The HPCI/ RCIC systems are not isolated. 240 min. The HPCI pump stops when the batteries run out. 260 min. Wide range sensed water level reaches Level 2 setpoint.

• Drywell and wetwell temperatures are 85'C (185'F) and 87'C (188'F), respectively. Mass and energy addition rates into the wetwell are:

• Mass Rate Energy Rate (kg/s) (lb/ min) (w) (Btu / min)

Steam 27.82 3.68 x 10 3 7.68 x 10 7 4.37 x 10 6 Hydrogen 0 0 0 0 302 min. Core uncovery time. Steam-water mixture level is at 3.58 m (11.73 ft) above bottom of the core. 320 min. Average gas temperature at top of core is 485'c (904*F). Drywell and wetwell temperatures and pressures are 101*C (213*F) and 0.21 MPa (31 psia), respectively. Mass and energy addition rates into the wetwell are: Mass Rate Energy Rate (kg/s) (lb/ min) (w) (Btu / min) 5.40 x 10 7 Steam 18.75 2.48 x 10 3 3.07 x 10 6 Hydrogen 6.65 x 10 ~7 8.80 x 10 5 3.34 1.90 x 10-1 4 0 4 . _ . . - -_. . . . . , . - _ . , = _ ,_ _ , _ , , . . . . _ , ,. . . 91 Time (sec) Event 340 min. Average gas temperature at top of core is 821*C (1509'F). Drywell and wetwell temperatures and pressures are 103*C (218'F) and 0.23 MPa (33 psia), respective'y. Mass and energy addition rates into the wetwell are: 1iass Rate Energy Rate . (kg/s) (1b/ min) (w) (Btu / min) Steam 11.75 1.56 x 10 3 3.76 x 10 47 2.14 x 10 6 Hydrogen 1.52 x 10-3 2.00 x 10 ~1 1.10 x 10 6.25 x 10 2 355 min. Core melting starts. 389 min. Water level in vessel drops below bottom grid elevation. 390 min. Bottom grid fails and temperature of structures in bottom head is above water temperature. 392 min. The corium slumps down to vessel bottom. 394 min. Debris starts to melt through the bottom head. . 426 min. Vessel bottom head fails, resulting in a pressure increase of 0.34 MPa (49 psia). - 426.04 min. Debris starts to melt the concrete floor of the containment building. Temperature of debris is 1433*C (2611*F) initial-ly. Internalheatgenerationinmetalsandoxidesare1.05 x 107 and 1.95 x 10 watts, respectively. 503.27 min. Drywell electric penetration assembly seals have failed as the containment temperature exceeds 204*C (400*F) and start to vent through the primary containment at a leak rate of 118 t/s (250 ft /3 min). 513.59 min. Containment f ailed as the containment temperature exceeds 260*C (500*F) and all electric penetration modules are blown out of the containment. Mass and energy addition rates into the drywell are: Mass Rate Energy Rate (kg/s (lb/ min) _ (w) (Btu / min) Steam 4.61 610 1.59 x 10 5 9052 Hydrogen 0.11 15 0 0 CO2 1.01 133 . CO 2.35 311 The leak rate through the drywell penetration seals is ~3.04 x 104 g/s (6.44 x 10 4ft /3 min). I 92 Time (sec) Event 613 min. Drywell and wetwell pressures are at 0.10 MPa (~14.7 psia) and temperatures are 661*C (~1222*F) and 98'C _(~209'F), respectively. The leak rate through the containment failed area is ~2.96 x 104 1/s (~6.27 x 109 ft 3/ min).

• 695 min. Drywell and wetwell temperatures are 623*C (1154*F) and 97'C . .

(207'F), respectively. The leak' rate through the contain-ment 3 failed area is ~6.47 x 104 1/s (~1.37 x 105 ft / min). 1028 min. Dryvell and wetwell temperatures are 614*C (~1138'F) and 97'C (~207'F), respectively. The leak rate through the containment 3 failed area is ~1.34 x 103 1/s (~2.83 x 103 ft / min). e 93 Table 9.2 Browns Ferry Nuclear Plant: Complete Station Blackout Sequence of Events CSB + HPCI/RCIC + SORV (Small Break LOCA) (TPS') Time (sec) Event 0.0 Loss of all AC power and diesel generators. The plant is initially operating at 100% power. Initial drywell temperature - 66*C (150*F) Initial wetwell temperature = 35'c (95'F) 0.2 Full load rejection (i.e., fast closure of turbine control valves) occurs. 0.2 Recirculation pumps and condenser circulatory water pumps trip off. Loss of condenser vacuum occurs. Core flow is provided by natural circulation. 0.2 Reactor pressure increases suddenly due to load rejec-tion. 0.3 Scram pilot valve solenoids are deenergized due to load rejection. Control rod motion begins. 0.5 Turbine bypass valves start to open due to load rejec-tion. 1.0 Neutron flux starts to decrease after an initial in-crease to over 100% rated power level. 1.0 Reactor power starts to decrease slowly after an ini-tial rise. 2.0 Control rods are 40% inserted from fully withdrawn position. 2.0 Main steamline isolation valves (MSlVs) start to close (relay-type reactor trip system), resulting in a rapid steaa-Jine pressure rise. 2.0 Turbine bypass valves are tripped to close. 3.0 Control rods are 75% inserted from fully withdrawn position. i [ 94 Time (sec) Event 3.0 - Turbine trips off (turbine stop valves fully closed). 3.5 Power generation due to delayed neutrons _and fission product decay drops to 10% of initial rated power gen - , eration.- 4.0 ~ Feedwater-turbines trip off. - 5.0 MSlVs are fully closed, resulting in a momentary 0.69 MPa (100 psi) pressure increase and 1.02 m (40-in.) drop of water-steam mixture level due to collapsing of voids. 5.0 All control rods are fully inserted. 5.0 Reactor vessel pressure exceeds the . lowest setpoint at 7.52 MPa (1090 psi) of safety / relief valves (S/RVs). 5.0 Seven (7) out of thirteen (13) S/RVs start to open in response to pressure rise above the setpoint. 5.2 Water-steam mixture level recovers 0.51 m (20 in.) from the previous momentary 1.02 m (40-in.) drop. ' 5.5 'S/RV steam blowdowns into the pressure suppression , pool through the T quenchers begin. 7.5 Feedwater flow drops below 20%. 9.0 Feedwater flow decreases to zero. 10.0 Power generation due to fission product decay drops to i approximately 7.2% of rated power generation, j 15.0 All 6 S/RVs are completely closed. One S/RV is stuck open (SORV); this has the same effect as a small break LOCA of equivalent break area of 0.015 m2 (0.1583 s ft 2), 15.7 Four out of 13 S/RVs start to open. 17.0 Neutron flux drops below 1% of taitial full power level.- 21.0 Narrow range (NR) sensed water level reaches .iow alarm (Level 4), i.e. , 5.98 m (235.50 in.) above Level 0, or . 5.00 m (196.44 in.) above TAF. 95 Time (sec) Event 22.0- Suppression pool water average temperature rises to 35.13*C (95.24*F) in response to the first S/RV pops. s 29.0 All 4 S/RVs are complately closed. ) . 29.7 Two out of 13 S/RVs start to open. 47.0 All 2 S/RVs are completely closed. 47.7 One out of 13 S/RVs starts to open. 56.0 Suppression pool water average temperature is approxi-mately 35.3*C (95.54*F). 56.0 NR sensed water level reaches low level alarm (Level 3), i.e., 5.50 m (216.00 in.) above Level 0, or 4.50 m (176.94 in.) above TAF. 90.0 Suppression pool water average temperature is approxi-mately 35.4*C (95.72*F). 101.0 The S/RV is completely closed. The same S/RV contin-ues to cycle on and off on setpoints throughout the subsequent cyclings of HPCI and RCIC injections. 625 Wide range sensed water level reaches low water level setpoint (Level 2), i.e., 4.18 m (164.50 in.) above Level 0 at 2/3 core height, or 2.96 m (116.50 in.) above TAF.

625 HPCI and RCIC systems are automatically turned on.

l The HPCI and RCIC turbine pumps are driven by steam generated by decay heat. System auxiliaries are powered by the 250 V de system. '655 HPCI and RCIC flows enter the reactor pressure vessel at 315 t/s (5000 gpm) and 38 1/s (600 gpm), re-spectively, drawing water from the condensate storage tank. 12.5 min. Narrow range sensed water level reaches Level 8 set-point, i.e. , 6.86 m (270.00 in.) above Level 0, or 5.64 m (222.00 in.) above TAF.~ . 12.5 min. HPCI and RCIC trip off. 4 .--- . _ _ , ~ . , _ . _ ., --< ,r-4-- , y -. p 96 Time (sec) Event 20 min. Drywell and wetwell temperatures exceed 70*C (158'F) and 50*C (122*F), respectively. Mass and energy addition rates into the wetwell are:

• Mass Rate Energy Rate *

(kg/s) (lb/ min) (w) (Btu / min) Steam 35.04 4.63 x 10 3 9.87 x 10 7 5.61 x 106 Hydrogen 0 0 0 0 25.5 min. Auto-isolation signal initiates as increase of drywell pres-sure exceeds 13.8 KPa (2.0 psi). The HPCI/ RCIC systems are not isolated. 26.5 min. Wide range sensed water level reaches Level 2 setpoint and HPCI automatically restarts. (RCIC does not automatically restart.) 27.0 min. HPCI flow enters the RPV. 29.C min. Narrow range sensed water level reaches Level 8 setpoint and HPCl trips off again. The HPCI system, driven by steam gen-

• erated by decay heat, turns on and off automatically between sensed Levels 2 and 8 until the batteries run out. ,

240 min. The HPCI pump stops when the batteries run out. 261 min. Wide range sensed water level reaches Level 2 setpoint. Drywell and wetwell temperatures are 100*C (212*F) and 96*C (205'F), respectively. Mass and energy addition rates into the wetwell are: Mass Rate Energy Rate (kg/s) (lb/ min) (w) (Btu / min) Steam 10.45 1.38 x 10 3 2.92 x 10 7 1.66 x 10 6 Hydrogen 0 0 0 0 315.07 min. Core uncovery time. Steam-water mixture level is at 3.53 m (11.58 ft) above bottom of the core. 321.5 min. Average gas temperature at top of core is 211*C (412*F). Drywell and wetwell temperatures at i pressures are 106*C , (223*F) and 0.24 MPa (35 psia), respectively. Mass and energy addition rates into the wetwell are: Mass Rate Energy Rate (kg/s) (lb/ min) (w) (Btu / min) Steam 2.66 3.00 x 10 3 6.33 x 10 7 3.60 x 10 6 Hydrogen 0 0 0 0 97 Time (sec) Event 384 min. Average gas temperature at top of core is 1046*C (1915'F). Drywell and wetwell temperatures and pressures are 114'C (237'F) and 0.28 MPa (41 psia), respectively. Mass and . energy addition rates into the wetwell are: Mass Rate Energy Rate = (kg/s) (lb/ min) (w) (Btu / min) Steam 2.46 325.0 8.30 x 10 6 4.72 x 10 5 5 Hydrogen 0.021 2.78 1.14 x 10 8.19 x 10 3 387.77 min. Core melting starts. 418.77 min. Water level in vessel drops below bottom grid elevation. 420.77 min. Bottom grid fails and temperature of structures in bottom head is above water temperature. 421.67 min. Tae corium slumps down to vessel bottom. 4 22.12 min. Debris starts to melt through the bottom head. . 515.18 min. Vessel bottom head fails, resulting in a pressure increase of 0.34 MPa (49 psia). . 515.20 min. Debris starts to boil water fron containment floor. 515.20 min. Drywell electric penetration assembly seals have failed as the containment temperature exceeds 204*C (400*F) and start to vent through the primary containment at a leak rate of 1051/s (222 f t 3/ min). 515.21 min. Debris starts to melt the concrete floor of the containment building. Temperature of debris is 1771*C (3219'F) initial-ly. x 107Internalheatgenerationinmetalsandoxidesare1.01 and 1.86 x 10 watts, respectively. 579.24 min. Containment failed as the containment temperature exceeds 260*C (500'F) and all electric penetration modules are blown out of the containment. Mass and energy addition rates into the drywell are: Mass Rate Energy Rate - (kg/s (1b/ min) (w) (Btu / min) 6 Steam 0.63 83.33 1.59 x 10 9052 , Hydrogen 0.22 29.10 0 0 CO2 2.08 275.13 CO 4.63 612.44 The leak rate through the drywell penetration seals is 5 ~2.53 x 105 t/s (5.35 x 10 ft /3 min). 98 Time (sec) Event 681.88 min. Drywell and wetwell pressures are at 0.10 MPa (~14.7 psia) and temperatures are 674'C (~1245'F) and 98'c (~209'F), respectively. The leak rate through the containment failed area is ~3.61 x 1@ 1/s (~7.65 x 104 ft 3/ min). . 810 min. Drywell and wetwell temperatures are 1006'C (1843*F) and 97'c (207'F), respectively. The leak rate through the

• containent 4

failed area is ~3.10 x 104 t/s (~6.57 x 10 ft / min). 3 1128 min. Drywell and wetwell temperatures are 591*C (~1095'F) and 97'C (~207'F), respectively. The leak rate through the containment 3 failed area is ~1.27 x 103 1/s (~2.70 x 103 ft / min). e , -- .- , - . , , - , - --,w. - , , - . - ,- , - , - , , , - , - . , - -- - - - - , - - - -- 99 is depressurized during core uncovery and melt. However, the pressure re-mains high enough so that the HPCI turbine can be operated for as long as the battery lasts. Compared with the sequence TB', the SORV results in a ~30 min. delay in the core melt, ~90 min. delay in vessel failure, and ~65 min. delay in the containment failure. 9.4.2.3 CSB + Manual RCIC & SRV (TyB'). The accident progression signature for the sequence T yB' is given in Table 9.3. In this sequ-ence, it is assumed that the HPCI system is inoperable. The opera *9r re- , mote-manually opens one SRV af ter 15 minutes of Station Blackout to de-pressurize the vessal in order to lower the vessel temperature and to pre-pare for use of the low pressure ECCS injection systems upon restoration of AC power. At the'same time, the operator attempts to maintain a con-stant vessel water level by manually centro 11ing the RCIC injection. It is noted that the fluid loss through the SRV causes the core to become momentarily uncovered, but it refloods immediately with the continued RCIC water injection. This short period of core uncovery is believed not to cause any core damage at that stage. 9.4.2.4 CSB + Manual RCIC & SRV + SORV (Ty PB'). The accident pro- ! gression signature for the sequence T yPB' is given in Table 9.4.. In this sequence, it is assumed that an SORV event occurs at the beginning of the transient, and the operator remote-aanually opens a SRV which causes , the reactor vessel to depressurize at a faster rate. Compared.with the sequence Ty B', the two sequences have produced very similar results . with respect to the times of core melt, vessel failure, and containment failure. ~ 9.4.2.5 - CSB + No HPCI/RCIC (TUB ') . The accident progression signa-ture for the sequence TUB' is given in Table 9.5. The events of this sequence as calculated by MARCH have been compared with results from the RELAP4/ MOD 7 code 27 up to the time of core uncovery. With the modifica-tions incorporated in MARCH version 1.4B, the core uncovery times pre-dicted by the two codes are in agreement. 9.4.2.6 CSB + No HPCI/RCIC & SORV (TUPB'). The accident progression signature for the sequence TUPB ' is given in Table 9.6. This is the severest of the six Station Blackout signatures studied, and includes a prediction of core uncovery at 17 minutes af ter the initiating loss of AC -power. The mitigating action which might be taken by the operator in re-

sponse to this and the other accident progression signatures will be dis-cussed in Section 10.

9.4.2.7 Summary of key events. A comparison of the predicted times to key events following core uncovery for each of the six sequences as calculated by MARCH version 1.4B is given in Table 9.7. 9.4.2.8 Key results for sequence TB'. Key results from the MARCH E calculations for the sequence TB' which is used as the base case for the fission product transport calculations in Volume 2 are presented in Figs. 9.12 through 9.21. l l -.- -- ._ --,- _ - . . . . - - . , _ . , _ . . , _ - . . . _ _ _ _ , .m, - 100 Table 9.3 Browns Ferry Nuclear Plant: Complete Station Blackout Sequence of Events CSB + Manual RCIC & SRV (TyB" ) . Time * (sec) Event 0.0 Loss of all AC power and diesel generators. The plant is initially operating at 100% power. Initial drywell temperature = 66*C (150*F) Initial wetwell temperature = 35'C (95'F) 0.2 Full load rejection (i.e., fast closure of turbine control valves) occurs. 0.2 Recirculation pumps and condenser circulatory water pumps trip off. Loss of condenser vacuum occurs. Core flow is provided by natural circulation. 0.2 Reactor pressure increase ( suddenly due to load rejec- , tion. 0.3 Scram pilot valve solenoids are deenergized due to . load rejection. Control rod motion begins. 0.5 Turbine bypass valves start to open due to load rejec-tion. 1.0 Neutron flux starts to decrease after an initial in-crease to over 100% rated power level. 1.0 Reactor power starts to decrease slowly af ter an ini-tial rise. 2.0 Control rods are 40% inserted from fully withdrawn po sition. 2.0 Main steamline isolation valves (MSlVs) start to close (relay-type reactor trip system), resulting in a rapid steam-line pressure rise. 2.0 Turbine bypass valves are tripped to close. 3.0 Control rods are 75% inserted from fully withdrawn . position. .n, 7 o / j ',/- a, 101 '3' ' , ~' '"( [ ' Time h (sec) Event - c['f ' /g .c ~ 3.0 Turbine trips off ,(tu,rbine stop valves fully closed). ,; ./ /, ' ~' 3.5 Power generation due to delayed, neutrons and fiss'ioo' , . product decay drops s to 10% ofg initial rated pouer gen-

• eration.

- ' ;.,, e 4.0 Feedwater turbines trip off. . +' e p , 7 5.0 tiSlvs a'fe fully closed, resulting in,a momentary 0.69 11Pa (100 psi) pressure increase and 1.02 m-(40-in.) '[' ~ drop of water-steam m'1xture level dee to collapsing of . , voids. N , % ~ .s .n 5.0 All control rode are fulif inserted. f 5.0 Reactor vessel pressure exceeds D.d lowest :s tpoh at' ' 7.52tiPa (1090 pui-) of safety / relief valves (S/RVs). - ~' r, , , -l 5.0 Seven (7) out of'linirteerM13) S/RVs start to dpan in response to pressure rise above the setpoint./;, uA ': ? - 5.2 Wafer-steam mixture'lefel' recovers 0.51 m (20 in.)- > . ir6m' the previous momenta'ry 1.02 m (40-19,.. .); drop. . , s 5.5 S/RV st_eam blowdoGnq $6tothepressuresuhression , pool tiirough the T-<iuyachers begin. , s ,- , . 7.5 Feed'aater flow d,rops' below 20%. '^

• 9.0 Feedwater flow decreases to zero. ,

2 10.0 Pover generation due' t'o fission product decay drspi[to . approxir. ate 1y 7.2%+of rated power generation.

/ , ,

~ 15.0 All 6 S/RVa are completely clased. One S/RV is stick open' (SORV)'; this'has the same ef fect.as a small break - LOCAL of equivalent' break area of 0.015 m2- (O.158.i f t2 ), / -- , 15.7 Four out of 13 S/RVs start to open.- ' , 17.0 Neutron flux drops below 1% of initial full power level. ,, e - , 21.0 Narrow range (NR)' sensed water level reaches low alarm (Level 4), i.e., 5.98 m (235.50 in.) above Level'O, or l *- 5.00 m (196.44 in.) above TAF. [ 22.0 Suppression pool water average temperature rises to 35.13*C (95.24*F) in response to the first S/RV , Pops. .-i-i i c,, _ I p ,,.g;e.l:r v

f. !, . ,; ,.

l +* nr, ,, s' 102 y , > e, )  : ,. lM <# , " , %, Time ,. e ' -A (sec) _'__ Event V /- , g ye [ , '29 0 , ,, All'4 S/RVs are completely closed. ,j! 7 'Two out of 13 S/RVs start to open. / [ $ [F 47.0 , All 2 S/RVs are completely. closed. (' ( - M/ f ., , t-' '47.7<, One out of.13 S/RVs starts to open. * , ,<' / . ',s' , ^ / -5(.M Suppression pool water average temperature is approximately 'I 7 / '.,' / 35.3*C (95.54*F). x ~ i e, '5620 . , NR sensed water level reaches low level alarm (Level 3), ' '~ ~'f'* % ,p - -1.e. , 5.50 m (216.00 in.) above Level 0, or 4.50 m (176.94 ~ " I ,' U - , ' ,in.) above TAF. / 3,,; < . ,/ j e ' Suppression pool water average temperature is approximately " 30.0". s.'J , e/, L - 35.4*C (95.72'F). ,./ ,' ;-l, 'l r .-  : ,101.0 ' The S/RV is completely closed. The same S/RV continues to ^~ f ,- cycle on and off on setpoints throughout the subsequent RCIC / 4 , injections. .<625 Wids range sensed water level reaches low water level set- ,( point (Level 2), i.e., 4.18 m (164.50 in.) above Level 0 at 6-2/3. core height, or 2.96 m (116.50 in.) above TAF. , ' g '4f- . gprator manually controls RCIC injection to maintain con-t 625 ,, ,, .,stant vessel water level. The RCIC turbine pump.is driven j f , by steam generated by decay heat. System auxiliaries are l power,ed by-the 250 V de system. ,' , ,, 1 - T[~" 655 / TiCIC flows enter the reactor pressure vessel at 38 1/s Q- ., li(600 gpm) drawing water from the condensate storage tank. w, / , , 15 min. Operator manually opens one SRV to depressurize the vessel. ,y 20 reYn.' Drywell and wetwell temperatures exceed 76*C (169'F) and ' ' 50*C (122*F), respectively. Mass and energy addition rates '/ m / - ,- r1rito i.he wetwell are: '2 Mass Rate Energy Rate [ , , t. .# (kg/s) (lb/ min) (w) (Bta/ min) / # - Steam 829.75 1.10 x 105 2.32 x 108 1.32 x 107 -

1. 'I > ,

', '.iyd rogen 0 0 0 0 [/[6/' ,; f i / , ,.-

• i mg ,  !, >

/ , , , ~,y _s , 1 w. t

• t

• r* Y

• ,n ,/n '_ s' ,

~/..~ . t _ ,c " ,

• N, , , . - ._

s F. . ~ x; , x / (. '

t. <' '

b 103 & Time (se'c} Event 21.14 min. Core uncovery time. 22.0 min. Core reflaods. 30 min. Auto-isolation signal initiates as increase of drywell pressure exceeds 13.8 KPa (2.0 psi). The RCIC system is not isolated. o 240 min. The RCIC pump stops when the batteries run out. 266.3 min. Wide range sensed water level reaches Level 2 octpoint. Dryw~211 and wetwell temperatures are 99'C (210*F) and 100*C (212*F), respectively. Mass and energy addition rates into , the wetwell are: Mass Rate Energy Rate (kg/s) (lb/ min) (w) (Btu / min) Steau s 19.16 2.53 x 103 5.20 x 10 7 2.96 x 10 6 Hydrogen 0 0 0 0 347 min. Core uncovers again. e 366 min. Average gas temperature at top of core is 491*C (916*F). Drywell and wetwell temperatures and pressures are 113*C o (236*F) and 0.28 MFa (40 psia), respectively. Mass and energy addition rates into the wetwell are: Mass Rate Energy Rate (kg/s) (lb/ min) (w) (Btu / min) Steam 9.26 1.22 x 103 2.97 x 10 7 1.69 x 10 6 5 3 Hydrogen 4.09 x 10 5.41 x 10 222.28 12.64 386 min. Average gas temperature at top of core is 855'C (1571*F). Dr/well and wetwell temperatures and pressures are 115'C (239'F) and 0.29 MPa (41 psia), respectively. Mass and energy addition rates into the wetwell are: Mass Rate Energy Rate (kg/s) (lb/ min) (w) (Btu / min) Steam 5.05 6.68 x 10 2 1.81 x 10 7 1.03 x 10 6 Hydrogen 1.68 x 10-2 2.23 1.35 x 10 5 7.70 x 10 3 9 104 Time (sec) Event 395.3 min. Core melting starts. 449.3 min. Water level in vessel drops below bottom grid elevation. A 451.2 min. Bottom grid fails and temperature of structures in bottom head is above water temperature. o 452 min. The corium slumps down to vessel bottom. 452.9 min. Debris starts to melt through the bottom head. 539.3 min. Vessel bottom head fails, resulting in a pressure increase of 0.0047 MPa (0.68 psia).. 539.3 min. Debris starts to boil water from containment floor. .539.3 min. Drywell electric penetration assembly seals have failed as the containment temperature exceeds 204'C (400'F)' and start to vent through the primary containment at a leak rate of 118.1/s (250 ft 3/ min). 539.3 min. Debris starts to melt the concrete fJoor of the containment building. Temperature of debris is 1750*C (3182*F) initial-

• ly.

Internal heat generation in metals and oxides are 9.99 x 106 and 1.84 x-10 watts, respectively. , 601.05 min. Containment failed as the containment temperature exceeds 260*C (500'F) and all electric penetration modules are blown out of the containment. Mass and energy addition rates into the drywell are: !! ass Rate Energy Rate (kg/s (lb/ min) (w) (Btu / min) Steam 4.70 '621.51 1.59 x 10 5 9052 Hydrogen 0.14 18.27 0 0 CO2 1.29 170.23 CO 2.88 331.21-The leak rate through the drywell penetration seals is ~5.33 x 104 1/s (1.13 x 105 f t /3 min). ~ 718.8 min. Drywell and wetwell pressures are at 0.10 MPa (~14.7 psia) and temperatures are 700'C ( 1293*F) and 98'C (~209'F), ' respectively. The leak rate through the containment failed area is ~5.18 x 104 1/s (~1.10 x 10 5 gg3/ min). . ,. ,. . . . . , , ,. ~ , . . 105 Time (sec) Event 821.5 min. Drywell and wetwell temperatures are 737'C (1359'F) and 93*C (199'F), respectively. The leak rate through the contain - ment failed area is ~4.23 x 10 4 1/s (~8.96 x 10" , ft /3 min). 1127.5 min. Drywell and wetwell temperatures are 468'c (~875'F) and s 86*C (~188'F), respectively. The leak rate through the containment 4 failed area is ~4.79 x 104 1/s (~1.02 x 10 ft / min). 3 4 4 i O 4 1 't 106 Table 9.4 Browns Ferry Nuclear Plant: Complete Station Blackout Sequence of Events CSB + Manual RCIC & SRV + SORV (Small Break LOCA) (TyPB')

• Time (sec) Event 0.0 Loss of all AC power and diesel generators. The plant is initially operating at 100% power.

Initial drywell temperature = 66*C (150*F) Initial wetwell temperature = 35*C (95*F) 0.2 Full load rejection (i.e., fast closure of turbine control valves) occurs. 0.2 Recirculation pumps and condenser circulatory water pumps trip off. Loss of condenser vacuum occurs. Core flow is provided by natural circulation. 0.2 Reactor pressure increases suddenly due to load rejec-

• tion.

0.3 Scram pilot valve solenoids are deenergized due to - load rejection. Control rod motion begins. 0.5 Turbine bypass valves start to open due to load rejec-tion. 1.0 Neutron flux starts to decrease after an initial in-crease to over 100% rated power level. 1.0 Reactor power starts to decrease slowly af ter an ini-tial rise. 2.0 Control rods are 40% inserted from fully withdrawn position. 2.0 Main steamline isolation valves (MSIVs) start to close (relay-type reactor trip system), resulting in a rapid steam-line pressure rise. 2.0 Turbine bypass valves are tripped to close. 3.0 Control rods are 75% inserted from fully withdrawn

• position.

~ 107 Time (sec) Event 3.0 Turbine trips off (turbine stop valves fully. closed). 3.5 Power generation due to delayed neutrons and fission product decay drops to 10% of initial rated power gen-eration. , 4.0 Feedwater turbines trip off. 5.0 MSlVs are fully closed, resulting in a momentary 0.69 3 MPa (100 psi) pressure increase and 1.02 m (40-in.) drop of water-steam mixture level due to collapsing of voids. 5.0 All control rods are fully inserted. 5.0 Reactor vessel pressure exceeds the lowest setpoint at 7.52 MPa (1090 psi) of safety / relief valves (S/RVs). . 5.0 Operator manually opens one S/RV to depressurize the vesse]. 5.5 S/RV steam blowdowns into the pressure suppression

• pool through the T-quenchers begin.

7.5 Feedwater flow drops below 20%. 9.0 Feedwater flow decreases to zero. 10.0 Power generation due to fission product decay drops to approximately 7.2% of rated power generation. 15.0 All 6 S/RVs are completely closed. One S/RV is stuck open (SORV). This has the same effect as a small break LOCA of equivalent break area of 0.0147 m2 (0.1583 ft2 ), l 17.0 Neutron flux drops below 1% of initial full power level. 30 Operator manually controls RCIC injection t< maintain constant vessel water level. The RCIC turbine pump is driven by steam generated by decay heat. System l auxiliaries are powered by the 250 V de system. 60 RCIC ? lows enter the reactor pressure vessel at 38 1/ r, (o00 gpm) drawing water from the condensate storage tank. 108 Time (sec) Event 10.52 min. Core uncovery time. Steam-water mixture level is at 3.56 m (11.68 ft) above bottom of the core. 11.52 min. Core refloods. . 20 min. Auto-isolation signal initiates ~as increase of d rywell pressure exceeds 13.8 KPa (2.0 psi). The RCIC system is e not isolated. 240 min. The RCIC pump stops when the batteries run out. 270 min. Wide range sensed water level reaches Level 2 setpoint. Drywell and wetwell temperatures are 100*C (212*F) and 102*C (216*F), respectively. Mass and energy addition rates into the wetwell are: Mass Rate Energy Rate (kg/s) (lb/ min) (w) (Btu / min) Steam 19.50 2.58 x 103 5.29 x 107 3.01 x 10 6 Hydrogen 0 0 0 0 337 min. Core uncovers again. . 356 min. Average gas temperature at top of core is 313*C (595'F). Drywell and wetwell temperatures and pressures are 113*C . (236*F) and 0.28 MPa (40 psia), respectively. Mass and energy addition rates into the wetwell are: Mass Rate Energy Rate (kg/s) (lb/ min) (w) (Btu / min) 7 Steam 12.25 1.62 x 10 3 3.64 x 10 2.07 x 106 Hydrogen 3.05 x 10~8 4.03 x 10-6 0.115 6.53 x 10~3 376.4 min. Average gas temperature at top of core is 650*C (1202*F). Drywell and wetwell temperatures and pressures are 115'C (239'F) and 0.29 MPa (41 psia), respectively. Mass and energy addition rates into the wetwell are: Mass Rate Energy Rate (kg/s) (Ib/ min) (w) (Btu / min) Steam 7.20 9.52 x 10 2 2.43 x 107 1.38 x 10 6 , Hydrogen 1.60 x 10 ~3 0.21 1.07 x 10 4 6.09 x 10 2 109 Time (sec) Event 396.36 min. Core melting starts. 450.4 min. Water level in vessel drops below bottom grid elevation.- 452.3 min. Bottom grid fails and temperature of structures in bottom head is above water temperature. 'd 453.1 min. The corium slumps down to vessel bottom. 454. min. Debris starts to melt through the bottom head. 542.5 min. Vessel bottom head fails, resulting in a pressure increase of 0.005 MPa (0.7 psia). 542.5 min. Debris starts to boil water from containment floor. 542.5 min. Drywell electric penetration assembly seals have failed as the containment temperature ' exceeds 240*C (400*F) and start-to vent through the primary containment at a leak rate of 1041/s (221 f t 3/ min). 542.5 min. Debris starts to melt the concrete floor of the containment

• building. Temperature of debris is 1766'c (3210'F) initial-ly. Internalheatpenerationinmetalsandoxidesare1.00 x 107 and 1.83 x 10 watts, respectively.

596.4 min.- Containment failed as the containment temperature exceeds 260*C (500*F) and all electric penetration modules are blown out of the containment. Mass and energy addition rates into the drywell are: Mass Rate Energy Rate (kg/s (1b/ min) (w) (Btu / min) 5 Steam 7.69 1017 1.59 x 10 9052 Hydrogen 0.038 5.08 0 0 CO2 2.58 342 CO 0.80 160

The leak rate through the drywell penetration seals is 4 5

! ~9. 20 x 10 1/8 (1.95 x 10 ft 3/ min). i l l i L 110. Table!9.5

Browns Ferry ' Nuclear Plant: ' Complete Station Blackout Sequence of Events i

CSB + No HPCI/RCIC (TUB") . Time d- -(sec) Event .' 0.0 Loss of all- AC power and diesel generators. The plant 4 is initially operating at 100% power. Initial drywell temperature - 66*C (150*F) Initial wetwell temperature = 35'C-(95'F) h 0.2 ~ Full load rejection (i.e. , fast closure of turbine - control valves) occurs. 0.2 Recirculation ' pumps and condenser circulatory water pumps trip off. Loss of condenser vacuum occurs.. Core flow is provided by natural circulation. 0.2 Reactor pressure increases suddenly due to load rejec- , tion. 0.3 Scram pilot valve solenoids are deenergized due to .. load rejection. Control rod motion begins.- , 0.5 Turbine bypass valves start to open due to load rejec-tion. 1.0 Neutron' flux starts to decrease after an initial in-crease- to over 100% rated power level. 1.0 Reactor power starts to decrease slowly af ter an ini-tial rise. 2.0 Control rods are 40% inserted from fully withdrawn position. 2.0 Main steamline isolation valves (MSIVs) start to close (relay-type reactor trip system), resulting in a rapid. i steam-line pressure rise. 2.0 Turbine bypass valves are tripped to close. 3.0 Control- rods are 75% inserted from fully withdrawn . position. . 3.0 Turbine trips off (turbine stop valves fully closed). a , i i 111 Time (sec) Event 3.5 Power generation due to delayed neutrons and fission product decay drops to 10% of initial rated power gen-eration. 4.0 Feedwater turbines trip off. g 5.0 MSlVs are fully closed, resulting in a momentary 0.69 MPa (100 psi) pressure increase and 1.02 m (40-in.) drop of water-steam mixture level due to collapsing of voids. 5.0 All control rods are fully inserted. 5.0 Reactor vessel pressure exceeds the lowest setpoint at 7.52 MPa (1090 psi) of safety / relief valves (S/RVs). 5.0 Seven (7) out of thirteen (13) S/RVs start to open in response to pressure rise above' the setpoint. 5.2 Water-steam mixture level recovers 0.51 m (20 in.) from the previous momentary 1.02 m (40-in.) drop. & 5.5 S/RV steam blowdowns into the pressure suppression pool through the T-quenchers begin. 7.5 Feedwater flow drops below 20%. 9.0 Feedwater flow decreases to zero. 10.0 Power generation due to fission product decay drops to approximately 7.2% of rated power generation. 15.0 All 7 S/RVs are completely closed. 15.7 Four out of 13 S/RVs start to open. 17.0 Neutron flux drops below 1% of initial full power level. 21.0 Narrow range (NR) sensed water level reaches low alarm (Level 4), i.e. , 5.98 m (235.50 in.) above Level 0, or 5.00 m (196.44 in.) above TAF.

• 22.0 Suppression pool water average temperature rises to.

35.13*C (95.24'F) in response to the first S/RV pops. 29.0 All 4 S/RVs are completely closed. 29.7 Two out of 13 S/RVs start to open. 112 Time (see) Event 47.0 All 2 S/RVs are completely closed. 47.7 One out of 13 S/RVs starts to open. 56.0 Suppression pool water average temperature is approximately s 35.3*C (95.54*F). s 56.0 NR sensed water level reaches low level alarm (Level 3), i.e. , 5.50 m (216.00 in.) above Level 0, or 4.50 m (176.94 in.) above TAF. 90.0 Suppression pool water average temperature is approximately 35.4*C (95.72*F). 101.0 The S/RV is completely closed. The same S/RV continues to cycle on and off on setpoints throughout the sequence. 625 Wide range sensed water level reaches low water level set point (Level 2), i.e., 4.18 m (164.50 in.) above Level 0 at 2/3 core height, or 2.96 m (116.50 in.) above TAF. 625 RPCI and RCIC systems are not turned on because they are assumed to be unavailable. ' 20 min. Suppression pool water average temperature reaches 46*C (114*F).

• 33 min. Core uncovery time. Steam-water mixture level is at 3.54 m (11.61 ft) above bottom of the core.

40 min. Auto-isolation signal initiates as increase of drywell pressure exceeds 13.8 KPa (2.0 psi). The HPCI/RCIC systems are not affected. Drywell and wetycl1 temperature are 72*C (162*F) and 55*C (130*F), respectively. Mass and energy addition rates into the wetwell are: Mass Rate Energy Rate (kg/s) (Ib/mir) (w) (Btu / min) Steam 33 9 4.36 x 10 3 9.25 x 10 7 5.26 x 10 6 Hydrogen 6x 10 8.62 x 10-7 2.92 x 10 -2 1.66 x 10 ~3 60 min. Mass and energy addition rates into the wetwell are: Mass Rate Energy Rate (kg/s) (lb/ min) (w) (Btu / min) 5.06 x 10 7 Steam 15.6 2.07 x 10 3 2.88 x 10 6 Hydrogen 2.8 x 10 ~3 3.76 x 10 -1 2.15 x 10" 1.22 x 10 3 J- ~ b._,s -- -,3- _,e,: i 113 Time (sec) Event 70 min. Core melting starts. 80 min. Drywell and wetwell temperatures are 75'C (167'F) and 63*C I' (145*F), respectively. ' Mass c.nl energy addition rates into - J the wetwell are: , Mass Rate Energy Rate (kg/s) (Ib/ min) (w) __{ Btu / min) 5.68 7 Steam 7.51 x 102 2.22 x 10 1.26 x id6 Hydrogen 0.19 2.53 x 101 2.29 x id6 1.30 x 10 5_ f 96 min. Water level in vessel drops below botton grid elevation. 97 min. Bottom grid falls and temperature of structures in bottom head is above water tenperature. 99 min. The corium slumps down to vessel bottom. 101 min. The debris is starting to melt through the bottom head. Drywell and wetwell temperatures are 97'C (207'F) and 71*C (159'F), respectively. Meanwhile, local pool water tempera-ture at the discharging bay exceeds 149'C (300*F). Steam condensation oscillations could accelerate due to the con-tinuous discharge of superheated noncondensable gases into , the suppression pool. Mass and energy addition rates into i the wetwell are: Mass Rate Energy Rate (kg/s) (lb/ min) (w) '(Btu / min) Steam 18.6 5.46 x 10 3 5.42 x 10 7 3.08 x ids l, Hydrogen 6.8 x 10-2 8.93 3.59 x 10 5 2.04 x 10" 129 min. Vessel bottom head fails, resulting in a pressure increase i of 0.34 MPa (49 psia). 129.03 min. Debris starts to melt the concrete floor of the containment building. Temperature of debris is 1546*C (2815*F) ini-tially. Internal heat generation in metals and oxides are 1.36 x 107 and 2.50 x 10 7 watts, respectively. e r I 114 Time (sec) Event 165 min. Drywell and wetwell temperatures are 141*C (286*F) and 74*C (166*F), respectively. Mass and energy addition rates into the drywell are: 4 Mass Rate Energy Rate ? (kg/s) (Ib/ min) (w) (Btu / min) 5 ' $ Steam 5.46 722.83 1.59 x 10 9052 Hydrogen 3.3 x 10-2 4.38 'O O CO2 2.58 341.88 CO 0.69 91.35 190 min. Drywell electric penetration assembly seals have failed as the containment temperature exceeds 204'c (400*F) and start to vent through the primary containment. 193 min. Containment failed as the containment temperature exceeds 260'C (500'F) and all electric penetration modules are blown out of the containment. 219 min. Drywell and wetwell pressures are at 0.10 MPa (14.7 psia). Drywell and wetwell temperatures are 598'C (1109'F) and 78'C (173*F), respectively. Mass and energy addition rates into . the drywell are: Mass Rate Energy Rate * (kg/s) (Ib/ min) (w) (Btu / min) Steam 0.70 92 1.59 x 10 5 9052 Hydrogen 0.24 32 0 0 002 2.32 307 CO 5.03 666 The leak rate through the containment failed areas is 5 ~2.90 x 10 1/s (~6.15 x 105 ft /3 min). 250 min. Drywell and wetwell temperatures are 675'C (1247'F) and 78'C (173*F), respectively. Mass and energy addition rates into the drywell are: Mass Rate Energy Rate (kg/s) flb/ min) (w) (Btu / min) 5 Steam 6.84 905 1.59 x 10 9052 Hydrogen 0.25 33 0 0

• CO2 1.53 203 CO 5.25 695

115 Time (sec) Event The leak rate through the containment failed area is 5 ~4.91 x id4 1/s (~1.04 x 10 ft /3 min). j 309 min. Rate of concrete decomposition is ~4.65 x id4 _gm/s. Kate of heat added to atmosphere is ~1.20 x IT kW. . 367 min. Drywell end wetwell pressures are at 0.10 MPa (~14.7 psia) and temperat:ures are 854*C (1570*F) and 77'C (171*F), re-spectively. The leak rate through the containment failed area is ~3.94 x id4 t/s (~8.35 x 10N f t3 / min). 733 min. Drywell and wetwell temperatures are 546'c (~1014*F) and 77'C (170*F), respectively. The leak rate through the con-tainment 3 3 failed area is ~2.12 x 10@ 1/s (~4.50 x t 10 ft / min). 4 O 9' l l w g ++ m, v * - ~ w a , - , y ma- - - - -r - 116 i Table 9.6-Browns Ferry Nuclear Plant:; Complete Station Blackout Sequence of Events CSB + No HPCI/RCIC & SORY (Seall Break LOCA)- (TUPB') Time-(sec) Event 0.0 Loss of all AC power and diesel generators. . The plant is initially operating at 100% power. Initial drywell tesperature = 66*C (150*F) Initial wetwell temperature = 35'C (95'F) 0.2 Full load rejection (i.e., fast closure of turbine control valves) occurs. 4 0.2 Recirculation pumps and condenser circulatory water pumps trip off. Loss of condenser vacuum occurs. Core flow'is provided by natural circulation. 0.2 Reactor pressure increases suddenly due to ' load rejec- . tion. 0.3 Scram pilot valve solenoids are deenergized due to - load rejection. . Control rod motion begins. l . 3 0.5 Turbine bypass valves start to open due to load rejec-tion. 1.0 Neutron flux starts to decrease after an initial in-crease to over 100% rated ' power level. 1.0 Reactor power starts to decrease slowly af ter an ini- . tial rise. 2.0 Control rods are 40% inserted from fully withdrawn position. 4 2.0 Main steamline isolation valves (MSIVs) start to close (relay-type reactor trip system), resulting in a rapid steam-line' pressure rise. 2.0 Turbine bypass valves are tripped to close. 3.0 Control rods are 75% inserted from fully withdrawn a position. I 3.0 Turbine trips off.(turbine stop valves fully closed). l' l ew-- ~ rmyn- - ~ . , - - w. - , , ,.-n.,,,c , , - y-my -,,-r-w->e-,,e--e,-- - - - - , .m ,mg-- neg. vev-- + - - , , , - - - am m- r . 117 Time , 4 (sec) Event ll 3.5 Power generation due to. delayed neutrons and fission product decay drops to 10% of initial rated power' gen-eration. , a 3 4.0 Feedwater turbines trip off. ' l* 5.0 MSIVs are fully closed, resulting in a momentary 0.69 MPa (100 psi) pressure increase and 1.02 m'(40-in.) drop of water-steam mixture level due to collapsing of , voids. 5.0 All control rods are fully inserted. 5.0 Reactor veseel pressure exceeds the lowest setpoint at i 7.52 MPa (1090 psi) of safety / relief valv'es (S/RVs). 5.0 Seven (7) out of thirteen (13)'S/RVs start to open in response to pressure rise above the setpoint. , 5.2 Water-steam mixture level recovers 0.51 m (20 in.) < from the previous momentary 1.02 m (40-in.) drop. 5.5 S/RV steam blowdowns into the pressure suppression pool through the T quenchers begin. l 7.5 Feedwater flow drops below' 20%. - i l 9.0 Feedwater flow decreases to zero. t

10.0 Power generation due to fission product decay drops to l approximately 7.2% of rated power generation.

[ 15.0 All 6 S/RVs are completely closed. One S/RV is stuck' open (SORV); this'has the same effect as a small break

LOCA of equivalent break area of 0.015'm2 (0.1583

, f t2 ), ! 15.7 Four out of ~13 S/RVs start to open. 17.0 - Neutron flux drops below 1% of initial full power level. f, 21.0 Narrow range (NR) sensed water level _ reaches low alarm (Level 4), i.e., 5.98 m (235.50 in.) above Level 0, or f 5.00 m (196.44 in.) above TAF. i

• j- 22.0 Suppression pool water average temperature rises 'to ,

35.13*C (95.24*F) in response to the first S/RV pops.

29.0 All 4 S/RVs are completely closed. 29.7 Two out of 13 S/RVs start to opea. I 118 Time (sec) Event 47.0 All 2 S/RVs are completely closed. 47.7 One out of 13 S/RVs starts to open. s 56.0 Suppression pool water average temperature is approximately 35.3*C (95.54'F). 4 56.0 NR sensed water level reaches low level alarm (Level 3), i.e. , 5.50 m (216.00 in.) above Level 0, or 4.50 m (176.94 in.) above TAF. 90.0 Suppression pool water average temperature is approximately 35.4'C (95.72*F). 101.0 The S/RV is completely closed. The same S/RV continues to cycle on and off on setpoints throughout the sequence. 625 Wide range sensed water level reaches low water level set point (Level 2), i.e., 4.18 m (164.50 in.) above Level 0 at 2/3 core height, or 2.96 m (116.50 in.) above TAF. 625 HPCI and RCIC systems are not turned on because they are , assumed to be unavailable. 17.2 min. Core uncovery time. Steam-water mixture level is at 3.62 m . (11.88 f t) above bottom of the core. 20 min. Auto-isolation signal initiates as increase of drywell pressure exceeds 13.8 KPa (2.0 psi). The HPCI/RCIC systems are not affected. Drywell and wetwell temperature are 73*C (163*F) and 55'C (130'F), respectively. Mass and energy addition rates into the wetwell are: Mass rate Energy Rate (kg/s) (lb/ min) (w) (Btu / min) Steam 60 7.94 x 103 1.69 x 10 8 9.61 x 106 -7 Hydrogen 8.53 x 10-13 1.13 x 10-10 3.19 x 10-6 1.81 x 10 40 min. Mass and energy addition rates into the wetwell are: Mass Rate Energy Rate (kg/s) (lb/ min) (w) (Btu / min) 7 Steam 10.30 1.36 A 10 3 3.41 x 10 1.94 x 10 6 Hydrogen 2.38 x 10-4 3.15 x 10-2 1.61 x 10 3 91.56 119 Time (sec) Event 56.6 min. Core melting starts. 60 min. Drywell and wetwell temperatures are 75'C (167'F) and 63*C _; (145'F), respectively. Mass and energy addition rates. into the wetwell are: 6 Mass Rate Energy Rate (kg/s) (lb/ min) (w) (Btu / min) Steam 2.73 3.61 x 102 1.23 x 10 7 6.99 x 105 Hydrogen 0.49 6.48 x 10 1 7.05 x id6 4.01 x 10 5 78 min. Water level in vessel drops below bottom grid elevation. 79 min. Bottom grid fails and temperature of structures in bottom head is above water temperature. 81 min. The corium slumps down to vessel bottom. 81.5 min. The debris is starting to melt through the bottom head. Drywell and wetwell temperatures are 82*C (180*F) and 71*C (159'F), respectively. Meanwhile, local pool water tempera- . ture at the discharging bay exceeds 149'c (300*F). Steam condensation oscillations could accelerate due to the con-tinuous discharge of superheated noncondensable gases into = the suppression pool. 101 min. Mass and energy addition rates into the wetwell are: Mass Rate Energy Rate (kg/s) (lb/ min) (w) (Btu / min) Steam 1.25 165.35 3.47 x 10 6 1.97 x 10 5 -4 3 Hydrogen 4.45 x 10 0.06 1.03 x 10 58.58 142.5 min. Vessel bottom head fails, resulting in a pressure increase of 0.34 MPa (49 psia). 152.5 min. Debris starts to boil water from containment floor. 162.5 min. Debris starrs to melt the concrete floor of the containment building. Temperature of debris is 2013*C (3655'F) ini-tially. Internal heat generation in metals and oxideu are g 2.43 x 107 and 1.26 x 10 7 watts, respectively. 120 Time (sec) Event 162.5 min. Drywell and wetwell temperatures are 128'C (262*F) and 74*C (166*F), respectively. Mass and energy addition rates into the drywell are: Mass Rate Energy Rate (kg/s) (1b/ min) (w) (Btu / min) Steam 0.057 7.54 1.59 x 10 5 9052 Hydrogen 0 0 0 0 167.8 min. Drywell electric penetration assembly seals have failed as the containment temperature exceeds 204*C (400*F) and start to vent through the primary containment. 175.2 min. Containment failed as the containment temperature exceeds 260*C (500'F) and all electric penetration modules are blown out of the containment. 185.3 min. Drywell and wetwell pressures are at 0.10 MPa (14.7 psia). Drywell and wetwell temperatures are 314*C (598'F) and 78'C (173'F), respectively. Mass and energy addition rates into the drywell are: Mass Rate Energy Rate (kg/s) (Ib/ min) (w) (Btu / min) Steam 1.65 5 218.26 1.59 x 10 9052 Hydrogen 0.025 3.31 0 0 CO2 5.79 765.88 CO 0.526 69.58 The leak rate through the centainment failed areas is 4 ~3.00 x 10 1/s (~6.36 x 104 ft /3 min). 206 min. Drywell and wetwell temperatures are 610*C (1130*F) and 78'C (173*F), respectively. Mass and energy addition rates into the drywell are: Mass Rate Energy Rate . (kg/s) (lb/ min) (w) (Btu / min) CO2 1.83 242 Hydrogen 0.20 26 0 0 Steam 1.36 180 1.59 x 10 5 9052 CO 4.15 549

• 121 1

Time (sec) Event The leak rate through the containment failed area is ~2.94 x 104 1/s (~6.24 x 104 f t /3 min). 4 ., 222.5 min. Rate of concrete decomposition is ~4.46 x 10ja/s. Rate of heat added to atmosphere is ~3.71 x 10' kW. .- 254.5 min. Drywell and wetwell pressures are at 0.10 MPa (~14.7 psia) and temperatures are 746*C (1375'F) and 77'C (171*F), re-s pectively. The leak rate through the containment failed area is ~5.54 x 104 1/s (~1.17 x 105 f t 3/ min). 501 min. Drywell and wetwell temperatures are 815'c (~1500*F) and 77'C (170*F), respectively. The leak rate through the con-i ^ tainment 4 3 failed area is ~2.34 x 104 1/s (~4.96 x 10 ft / min).. s e 1 e . - . . ,, _ - -. , ~ - - _ . , _ _ _ _ . _ _ - . - _ . . - _ - . . .; Table 9. 7. Predicted times to . key events Accident progression time (min)' Sequence Event core Core Core Start Core Vessel Wetwell" Drywell - Drywell uncover reflood q melt slump _ _ f ailure failure EPA vent' . f ailure ! CSB + HPCI/RCIC 302 355- 392 426 .503 514 2 CSB + HPCI/RCIC + SORV 315 388 419 515, 515 : 580 3 CSB + Manual RCIC/SRV 21 22 347 395 449 539 539- 601' 4 CSB + Manual RCIC/SRV + SORV 11 12 337 396 453 543 54 3 5% 5 CSB + No HPCI/RCIC 33 69 95 128 # 190 -.193 6 CSB + No HPCI/RCIC + SORV 17 57 78 143 130f 145 168 175

1. ~

H Wetwell failure is due to forces of steam jet imp 1ngement and condensation oscillations resulting from excessive thermal; stratification in the suppression pool. b Drywell electric penetration assembly seals start to vent when the ambient temperature exceeds 204*C (400*F). %rywell electric penetration assembly seals become decomposed and are blown out of the containment when the ambient ' tem- ~ perature exceeds 260*C (500*F). bstestimatevalue.

1. ~

Containment would fail at 288 min. using WASH-1400 f ailure criterion of static pressurization at 190 psia. fBest estimate value. k 9 , e e , gp 123 Figure 9.12 shows the rec tor vessel pressure distribution; it re-mains approximately constant during the first 240 minutes when the steam mass flow through the SRV discharge into the suppression pool (Fig. 9.13) is balanced by HPCI injection (Fig. 9.14). The pressure _ increases after the HPCI injection has stopped upon the assumed loss of de power at 4 h into the transient. The mass of water and the mass of steam within the reactor vessel are e shown in Figs. 9.15 and 9.16 respectively. The relatively rapid decrease in steam inventory beginning at about 305 minutes as shown in Fig. 9.16 is due to depletion in the production of hydrogen (Fig. 9.17) by the Zirconium-water reaction (Fig. 9.18) before the onset of core melting (Fig. 9.19). The large amount of energy released by the Zr-H 2O reaction (Fig. 9.20) before the core slumps into the lower head causes an increase in the steaming rate and the reactor vessel pretsure. The large amount of energy release in the predicted corium-water in-teraction after reactor vessel failure would cause a significant contain-ment pressure spike. The vertical concrete penetration resulting from corium attack is shown in Fig. 9.21. 9.5 Containment Responses As previously discussed, containment failure could occur either in the drywell or in the wetwell if a core meltdown accident were to occur at the Browns Ferry Nuclear Plant. The radiological con =equences of a wet-

• well failure would be less severe than those of a drywell failure, because

> a significant fraction of the released fission products would be dissolved - or deposited in the suppression pool. 9.5.1 Drywell Response Drywell failure would occur in the EPA when the elastomeric sealing materials undergo degradation and lose sealing integrity at ambient ten-peratures greater than 204*C (400*F) 18,22 This does not preclude the possibility of an esrlier failure of the wetwell by overpressurization. An EPA typical of those installed in the Browns Ferry unit i drywell liner is shown in Fig. 9.22. The masses of steam and hydrogen accumulated in the drywell and wet-well following core melt in the base case TB' are shown in Figs. 9.23 through 9.26. For this case, wetwell failure by overpressurization is not preducted to occur. The drywell temperature responses for the sequences TB' and TUB ' are shown in Figs. 9.27 and 9.28 respectively; the general trends are very similar in the two sequences, except that high temperatures are reach 3d much sooner in sequence TUB ', as would be expected.* The corresponding drywell pressure responses for these two saquences are shown in Figs. 9.29 through 9.36.

• *In sequence TUB ', it is assumed that the HPCI and RCIC systems are unavailable from the inception of the Station Blackout. Thus boiloff, core uncovery, and the subsequent events would occur sooner.

ORNL-DWG 81-8523 ETD (x 105) 77.0 t 76.0 - 2 75.0 - E S w.0 - 10 E 3 73.0 - t; E h >- 72.0 - @ PRESSURE INCREASE FOLLOWING MSIV CLOSURE a cc ' 71.0 - 0.0-dc POWER LOST 69.0 , , , , 0.0 50.0 100.0 150.0 200.0 250.0 300 0 350.0 400.0 TIME (min) Fig. 9.12 Reactor vessel pressure. , l l(jll U D 0 T 0 E 0 4 4 2 5 8 Y - R 1 E 8 V G W O C ' 0 0 D M N i5 L U U 3 N N E ~ R E R O L , P O C R E - 0 W T i0 0 O 3 . L m O e t T s N y I s S y P 0 M 0 r U e5 a L f, 2 i m S - S r p E P R O m O T ) n o C S r 0 im f T N 0( e SO OI i0E t L C T 2 M a I r . RE N T w E J O o WIN T I l f OI PC E C e P J 0 s dH N 0 s I i5 a I 1 m CD m a P E e HC t DN S N LA AA 0 3 0 1 WB e 0 1 9 O L E_ F M . O g VC i RE F SB 0 i0 5 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0

• 0 0 0 0 0 0 1

6 5 4 3 2 7)_ uci . 2O@ $$ 45 3bM r ' I jl l ORNL-DWG 81-8525 ETD 120.0 1 i 100 0 - 5 4 d 1 a Lu > 80.0 - E a a E 60.0 - R - E I" $ 40.0- SRV FLOW AND HPCI INJECTION 3: BECOME BALANCED s 1 $ dc POWER LOST S 20.0- j __ HPCI INJECTION STOPS 0.0 i , , i , , , 0.0 50.0 100.0 150.0 200.0 250.0 300.0 350.0 400.0 TIME (min) Fig. 9.14 HPCI flow rate into pressure vessel. ORNL-DWG 81-8526 ETD 4 (x 10 )

30.0 i

-LOSS OF dc POWER i 25.0- BOILOFF BEGINS g -_ _ 5 a: 20.0 - E E E a- 15.0 - H N t: 3 u. y3 10.0 - n 2 5.0-CORE SLUMPS 0.0 0.0 50.0 1d0.0 1d0.0 2d0.0 2$0.0 3$0.0 3dO.0 400.0 TIME (min) i Fig. 9.15 Mass of water in primary system, i i ORN L-DWG 81-8527 ETD 3 (x 10 ) 20.0 18.0 CORE UNCOVERY l 16.0 - 2 m @ 14.0-u >. de POWER LOST $ 12.0- BOILOFF BEGINS a E H z 10.0 - $ h BEGIN CORE MELT 2 8.0-2 b $ 6.0 - 4.0 - CORE SLUMPS 2.0 , , , , , , , 0.0 50.0 100.0 150.0 200.0 250.0 300.0 350.0 400.0 TIME (min) Fig. 9.16 Steam mass in primary system. s . . ~ , _ ORN L-DWG 81-8528 ETD 300.0 a 1 250.0 - Tn 200.0 - di vi i S 7 4 H2BUILDUP DUE TO Zr-H 2O REACTION , i, z m 150.0 - < 1' a o x ~r 1 g 0 e a l f I 100.0 - r  : ~ l , i g . s.

a 50.0- ,

ICORE-SLUMPS ' CORE UNCOVERY -- l , 0.0 i i i i i a i 00 50 0 ~ 100.0 150.0 200.0 250.0 - 300.0 350.0 4000 .-s TIME (min) .s 3

s, ' r 1 ,

s , Fig. 9.1 Hydrogen teass in prir.ary systeini. z\ e , .S - .( i s

t.  %

\ c . t A ,g / ,g .g 4 s .s., - ~ I .i ) 'gX f. 'y + j_ I _,'(a 4 3 '\ ( - , '? _) ]', .t .,, . .,. . . . _ . . . - . . ,, x ';b ..; -s 'A s ~ s 1 g i s' s , c e .

• a . .g 1., e. y

, t . s '. , s s -ef' ', s ,l , . 1 1 ~. , 3 !' 's y ,. 's - s 4, , ,- , s. 4 s s , s v - s -' . CHNt.40WG 88-8529 ETD , <> s ,

3. >

,~ 0.20 .3 q'  ? '-x'3 1 .,s, . . 't x 1. .s ,, j. , r N l \ x 0.15 - 3' o s X w , o w O _J o 0.10 -

u. ~

O u o 7 o_ n o e u. 0.05-I CORE UNCOVERY-- n 0.00 , , , , , , , 0.0 50.0 100.0 150.0 200.0 250.0 300.0 350.0 400.0 TIME (min) Fig. 9.18 Fraction of clad reacted. l t l ~ - . . , _ l ORNL-DWG 81-8"30 ETD 1.0 - 0.8- , O .J w 2 0.6-O .J U u. O 2 s 9 0.4-- # w O cr i 0.2-CORE UNCOVERY 0.0 v j 00 b0 1000 l$0.0 2$0.0 2$0.0 3$0.0 3$0.0 400.0 TIME (min) l Fig. 9.19 Fraction of core melted. I t l-ORNL-DWG 81- 8531 ETO (x 107 ) 10.0 8 z o 8.0 - U ' < L y HIGH INITIAL REACTION RATE O . m I 6.0 -

n 2

O cc M LI-Q M

• 4.0 - l cc w

z w 0 g 2.0 - m z CO R E--* CORE UNCOVERY- SLUMP 0.0 , , , , , ." , 0.0 50.0 100.0 150.0 200.0 250.0 300.0 350.0 400.0 TIME (min) Fig. 9.20 Energy generated from Zr-H 2O reaction. V ORNL-DWG 81-8532 ETD 800.0 CONCRETE PENETRATION COMPLETED 700.0-E f 600.0-9 E fm500.0-5 a. p 400.0 - 5 8 300.0- _a 8 p 200.0 - VESSEL BOTTOM @ HEAD FAILS 100.0 - CORE UNCOVERY 0.0 , , h , , , , , , , 0.0 100.0 200.0 300.0 400.0 500.0 600.0 700.0 800.0 900.0 1000.0 1100.0 TIME (min) Fig. 9.21 Vertical concrete penetration. ORNL-DWG 81-8533 ETD FIELD WELD CONNECTOR END SHIELD PRIOR TO MOUNTING FOR RADIATION INSTALLATION HOLES CONTAINMENT PROTECTION SHIELD ___ FlELD WELD SUPPORT i rp,Ygstg$$ki u >L"om777, _m _

e-

'_E??b5 " , =. / in n 'JJTfillTI' " iiiiiii13  % 9 sf / x; f INSULATION \ ' \ PENETRATION n CD / "ozz'e D:0. x4  :  ; k Q d h DD % #h d. k n - DR y ( kk fr, _ kMI[ - l l I hfh L CABLE b _J g gg COLLAR SUPPORTS PRESSURE VALVE AND SHIELD GAUGE PRESSURE TAP . HEADER PLATES Fig. 9.22 Typical electrical penetration assembly canister. ORNL-DWG 81-8534 ETD 3 (x 10 ) 18.0 16.0 - 14.0 - i 3 12.0 - 1 E VESSEL BOTTOM HEAD FAILS DRYWELL EPA SEALS Fall g 10.0-m 8.0- h 2 lE 6.0 - 4.0-2.0- DRYWELL VENTING THROUGH -CORE UNCOVERY FAILED EPA SEALS 0.0 i i i r i i i tn k i k[ i i i 0.0 100.0 200.0 300.0 400.0 500.0 600.0 700.0 800.0 900.0 1000.0 1100.0 TIME (min) Fig. 9.23 Mass of steam in drywell. ORN L-DWG 81-8535 ETD 250.0 SPlKE DUE TO H2RETURN FROM WETWELL 200.0 - Tn di ~ FAILURE OF DRYWELL EPA SEALS 8 $ DRYWELL H2SWEPT INTO WETWELL g AS VESSEL BOTTOM HEAD FAILS 100.0 - h M 2 50.0 - Zr-H 2O REACTION BEGINS [ 0.0 } 0.0 100.0 200.0 300.0 400.0 500.0 600.0 700.0 800.0 900.0 1000.0 1100.0 TIME (min) Fig. 9.24 Mass of hydrogen in drywell. . . . . . o ORNL-DWG 81-8536 ETO 9000.0 8000.0 - VESSEL BOTTOM HEAD FAILS l 7000.0 - gFAILURE OF DRYWELL EPA SEALS _ 6000.0 - CORE SLUMPS / .I 2 g 5000.0-i !7;

u. " s o 4000.0 -

8 U 2 CONTINUED RELIEF M.0 - VALVE ACTUATION 2000.0 - 1000.0 - -CORE UNCOVERY O0 , , ." , , , , , , , 0.0 100.0 200.0 300.0 400.0 500.0 600.0 700.0 800.0 900.0 1000.0 1100.0 TIME (min) Fig. 9.25 Mass of steam in wetwell. ORNL-DWG 81-8507 ETD 800.0 700.0 - f .0-DRYWELL H SWEPT INTO WETWELL Ad VESSEL BOTTOM HEAD FAILS 0-RETURNED TO DRYWELL AS EPA @ q bdALS Fall AND CONTAINMENT VENTS $ 400.0 - u. o 0-8 * < 300.0 - lE 200.0 - 100.0 - Zr-H 2O REACTION BEGINS 0.0 f , , , , , , , , , , 0.0 100.0 200.0 300.0 400.0 500.0 600.0 700.0 800.0 900.0 1000.0 1100.0 TIME (min) Fig. 9.26 Mass of hydrogen in wetwell. ORNL-DWG 81-8538 ETD 1600.0 4 1400.0 - k 52 ~ 1200.0 - 5 ( / V t - 5 1000.0 - 9 f INCREASE DURING F- CORIUM-CONCRETE $ 800.0- REACTION U s f E 600.0 - VESSEL 30TTOM HEAD FAILS 400.0 - / FAILURE OF DRYWELL EPA SEALS 200.0 , , , , , , , , , , 0.0 100.0 200.0 300.0 400.0 500.0 600.0 700.0 800.0 900.0 1000.0 1100.0 TIME (min) Fig. 9.27 Drywell temperature distribution (TB'). l l ORNL-DWG 81-8539 ETD 1800.0 1400.0 - Y [ 1200.0-W < INCREASE DURING CORIUM-L ' \ Y CONCRETE REACTION h 1000.0-F w $ 800.0- $ 2 2 g M.0 - FAILURE OF DRYWELL EPA SEALS U VESSEL BOTTOM HEAD FAILS 400.0 - 200.0 , , , , , , . 0.0 100.0 200.0 300.0 400.0 500.0 600.0 700.0 800.0 TIME (min) Fig. 9.28 Drywell temperature distribution (TUB') . ORNL-DWG 81-8540 ETD 5 (x 10 ) 10.0 9.0-g 8.0- !b E o 7.0-v'1 E o. s 6.0 -- b / / FAILURE OF DRYWELL EPA SEALS k VESSEL BOTTOM HEAD FAILSm g g 5.0- N - 9 o j 4.0 -

5 o

• 3.0-2.0-

-CORE UNCOVERY 1.0 , , , v k 0.0 100.0 200.0 300.0 400.0 500.0 600.0 700.0 800.0 900.0 1000.0 1100.0 TIME (min) Fig. 9.29 Drywell total pressure (TB'). ORNL-DWG 81-8541 ETD (x 105) 8.0 7.0 - y 80-E 3 5.0- FAILURE OF DRYWELL EPA SEALS VESSEL BOTTOM HEAD FAILS 2 5 4.0 - r

$ 0 f

2 3.0 - d t; 2.0-1.0-CORE UNCOVERY ~ 0.0 - 0.0 1d0.0 2d0.0 3d0.0 4$0.0 5d0.0 6$0.0 7d0.0 8d0.0 9d0.0 1000.0 1100.0 TIME (min) Fig. 9.30 Drywell steam partial pressure (TB') . ORNL-DWG 81-8542 ETD 4 (x 10 ) 10.0 8.0 - E @ FAILURE OF DRYWELL EPA SEALS d 6.0 - E A h H2SWEPT INTO WETWELL AS $ g VESSEL BOTTOM HEAD FAILS z 4.0- $ / e Q lf b \ ' 2.0-V [JL l Zr-H 2 O REACTION BEGINSq / \/ ti / w 0.0 , , , , , , , , , i 0.0 100,0 200.0 300.0 400.0 500.0 600.0 700.0 800.0 900.0 1000.0 1100.0 TIME (min) Fig. 9.31 Drywell hydrogen partial pressure (TB'). l l l l ORNL-DWG 81-8543 ETD (x 104 ) 30.0 i 3 25.0- NON-CONDENSABLES SWEPT INTO WETWELL [ oc AS VESSEL BOTTOM HEAD FAILS 3 l h 20.0- f [ I < FAILURE OF DRYWELL EPA SEALS H Zr-H O REACTION 15.0 - BEGibS y !0 E d HEATUP OF DRYWELL Z ATMOSPHERE 10.0 - e ff ATMOSPHERIC PRESSURE ( z ( w @ 5.0- ] 0.0 , , , , , , , , , , 0.0 100.0 200.0 300.0 400.0 500.0 600.0 700.0 800.0 900.0 1000.0 1100.0 TIME (min) Fig. 9.32 Drywell non-condensables partial pressure (TB'). + . W O 1 ? ORN L-DWG 81-8544 ETD 5 (x 10 ) 7.0 1 6.0-b E g 5.0 - m W FAILURE OF DRYWELL EPA SEALS [ VESSEL g BOTTOM z HEAD g 4.0- FAILS 7  : o. 2 8 3.0 - _s i-o H 2.0-1.0 , , , , , , , 0.0 100.0 200.0 300.0 400.0 500.0 600.0 700.0 800.0 TIME (min) Fig. 9.33 Drywell total pressure (TUB'). ORNL-DWG 81-8535 ETO 5 (x 10 ) 6.0 l 5.0-4 - S g 4.0-3 VESSEL $ BOTTOM FAILURE OF DRYWELL EPA SEALS E HEAD E FAILS g 3.0 - P x \ 1 $ I k 2.0 - Y m 1.0-

0.0 , , ,

k, , , , 0.0 100.0 200.0 300.0 400.0 500.0 600.0 700.0 800.0 TIME (min) Fig. 9.34 Drywell stearn partial pressure (TUB'). ORNL-DWG 81-8546 ETD (x 104 ) 8.0 7.0 - 2 6.0 - [ FAILURE OF DRYWELL EPA SEALS 5 y 5.0 - e H2SWEPT INTO ' WETV/ ELL AS Y VESSEL BOTTOM p 4.0 - HEAD FAILS

• a

$ 3.0 - 8 F \ i 8 g 2.0 - ( Zr-H O 1.0- REAdTION BEGINS 0.0 , , , , , , , 0.0 100.0 200.0 300.0 400.0 500.0 600.0 700.0 800.0 TIME (min) Fig. 9.35 Drywell hydrogen partial pressure (TUB'). O RNL-DWG 81-8547 ETD 4 (x 10 ) -25.0 NON-CONDENSABLES SWEPT INTO WETWELL AS VESSEL BOTTOM HEAD FAILS g 20.0 - 8 10 E 15 0 - FAILURE OF DRYWELL EPA SEALS $ Zr-H2O A REACTION $ BEGINS Z s 5 10.0 - 5 ( ( $ ATMOSPHERIC I ( 8 PRESSURE I A \ @ 5.0 - i bW 0.0 , , , , i i i O.0 100.0 200.0 300.0 400.0 500.0 600.0 700.0 800.0 TIME (min) Fig. 9.36 Drywell non-condensables partial pressure (TUB"). \ e

• . . O O 9

1 - 149. i 4 As the drywell ambient temperature exceeds 204*C (400*F), the EPA elastomeric sealing materials start to deteriorate and venting of the i drywell begins. When the ambient' temperature has exceeded 260*C (500*F), the EPA elastomeric sealing materials have decomposed to such'an extent that the EPA seals are blown out of the containment wall by the elevated ' pressure within the drywell. This greatly increases the containment leak-age (Fig. 9.10). 9.5.2 Wetwell response + The condensation of the SRV steam discharge within the pressure sup- - pression pool is generally accomplished without undue pressure stresses on the containment wall. If, however, conditions are such that complete. con- - densation of the steam discharge does not occur, pressure loads signifi-

cantly in excess of design limits can result, leading eventually to wet-well rupture.

i Relief valve steam discharge into the pressure. suppression pool is accompanied by -pressure oscillations of varying characteristics which are

functions of the steam mass flux and local pool water temperatures as well j as the type of sparger device installed at- the discharge line terminus.

It has been observed that condensation instability can occur when a- - l submerged pipe vent discharges steam at flow rates higher than critical l~ (sonic discharge) with a sufficiently high ambient pool temperature- the . so-called "Wurgassen effect". The threshold of instability is character-i ized by an increase in the amplitude of pressure oscillations which nor- + mally accompany condensation at supercritical flow rates. At present, ramshead-type sparger devices are installed on the relief i valve tail pipe discharge lines at .the Browns Ferry Plant. The ramshead !* device consists of two 90* elbows welded-back-to-back to form a modified ! T-junction, and provides an improvement in condensation performance over !~ that of a straight vertical pipe. The horizontal discharge from the ram- ! shead allows the rising convection currents and induced secondary flows to circulate cooler water around the steam plumes.rather.than stagnate l against a downward flow of steam. However, small scale tests have shown. I that it also has the potential to-produce unstable condensation and con-comitant large pressure stresses when sufficiently high pool water temper-ature is reached during supercritical (sonic) discharge. For this reason, certain limitations on plant operation are established in the Browns Ferry. Technical Specifications to preclude any possible steam discharge through the ramshead devices at pool temperatures greater than 77'C (170*F). In- the near future, the ramshead devices at the Browns Ferry Plant will be replaced by T quencher sparger devices similar to that'shown in . Fig. D-2 of Appendix D. Tests have shown that the T quencher spargers produce lower loads during the initial air-clearing upon relief valve ac-tuation and permit smooth condensation of the steam discharge at pool tem-peratures up to 93*C -(200*F). For this study, it was assumed that the SRV discharge lines terminate in T quencher spargers. Although the use of T quenchers considerably improves the steam con-densation characteristics related to SRV' steam discharges, the constricted flows introduced by the T quencher have caused insufficient thermal mixing in the suppression pool. The resultant thermal stratification in the pool m-w +e -vg-g y-e ,.A g.. < ~-._4m ,m,,- ,,. ,,,,.,w ,,,,9 .,y,y%,&y ~7-,4-pi..,.y.-y, . - - -, , .%,9 ,- . - - - . . i_.,g ,-,,-,9.e.,-7 150 can produce much higher local water temperature in the vicinity of the T-quenchers although the average pool temperature remains quite low. The problem of thermal stratification has been f ound to be signific-ant for the sequences TUB' and TUPB', in which the HPCI and RCIC sys-tems are assumed inoperative. Without injection capability, core uncovery and subsequent degradation occur early in the Station Blackout when the level of decay heat is relatively high so that a large amount of super- . heated steam and noncondensibles is discharged from the T quenchers into the suppression pool in a short period of time. The average suppression pool temperature is plotted as a function of time for sequences TB' and

• TUB' in Figs. 9.37 and 9.38 respectively.

In sequence TUB', it is realistically assumed that without operator action, reactor vessel pressure control over the long term would be by re-peated cycling of the same relief valve.* Because of the high steam mass flux into the suppression pool bay in which the discharging T quencher is located, significant thermal stratification would be expected. MARCH com-putations show that the difference between the local and average suppres-sion pool temperatures can be estimated to increase f rom about 5'C at the beginning of the transient to about 40*C 100 minutes later. This means that the suppression pool would lose its condensation effectiveness; the resulting pressure loads f rom tha SRV discharge of steam and noncondensi-bles would rapidly increase, leading to a possible rupture of the wetwell which could occur before the overtemperature-induced failure of the dry-well. The drywell pressure signatures for the sequences TB' and TUB' are , shown in Figs. 9.29 through 9.36. The general trends are again very sim-ilar, except for the timing of events. The pressure peaks for the sequence TB' are generally higher than those for the sequence TUB'; this is attri- . buted to the presence of a larger water inventory as a result of the four-hour period of HPCl injection during sequence TB". The wetwell temperature signatures are given in Figs. 9.37 and 9.38, and the average pressure distributions are given in Figs. 9.39 through 9.46 f or s,.quences TB' and TUB'. The plotted pressures do not include the effect of pressure waves resulting from condensation instabilities or the vapor jet plumes emanating from the T quencher spargers.

• Although four of the thirteen SRVs share the lowest nominal set-point of 7.72 MPa (1105 psig), practical considerations dictate that there would be one loweet-set valve.

i ll 1 l l llI J li E D 0 T 0 E 0 1 8 1 4 5 8 1 8 0 0 G 0 W ,0 D 1 L N R 0 O 0 . ,0 9 F O E R S L 0 U ,0 L A 0 ) I E 8 ' AS B T FAP [ ( Y B E 0 n L ,0 o GL 0 i t NE I 7 u TW b NY i r ER 0 t VD 0 s - i 0 )n d 6i ( m e r E u t . Y 0 M a R I / E ,00 T r e V 5 p O m e . C t N U 0 l 0 l E ,0 e R 4 w A O t e E C W H M 0 7 O 0 3 S , 0 T N 3 9 T I O G . B E g L B 0 i E F S N ,0 0 S O 2 E I T V C F A O E 0 R 0 E ,0 R O 1 U 2 L H I - A Z r . F 0 0 0 0 0 0 0 0 0 0 0 0 0 0 5 0 5 0 5 0 5 5 4 4 3 3 .e~

IE2N 52$I28 I' ll ' I lllll Il iI l i

ORNL-DWG 81-8549 ETD 550.0 FAILURE OF VESSEL BOT OM HEAD p 2 VENTING BY Fall'JRE OF DRYWELL EPA SEALS o i 450.0 - Y b U s 400.0-oc I s 8 350.0 - Zr-H 2O REACTION BEGINS 300.0 , , , , , , , 0.0 100.0 200.0 300.0 400.0 500.0 600.0 700.0 800.0 TIME (min) Fig. 9.38 Wetwell temperature distribution (TUB'). ' s . . , , ORNL-DWG 81-8550 ETD (x 105) 10.0 - 9.0-g 8.0-E g 7.0 - m $ VENTING BY FAILURE OF DRYWELL EPA SEALS 6.0-5 e 2 FAILURE OF VESSEL BOTTOM HEAD U E 50-f s o j 4.0 - sO Zr-H 2O REACTION BEGINS

• 3.0-

/ 2.0- -~ 1.0 , , , . . . . . . . 0.0 100.0 200.0 300.0 400.0 500.0 600.0 700.0 800.0 900.0 1000.0 1100.0 TIME (min) Fig. 9.39 Wetwell total pressure (TB'). ORNL-DWG 81-8551 ETD (x 104 ) 25.0 l FAILURE OF VESSEL BOTTOM HEAD 20.0-VENTING BY FAILURE OF g S / [ DRYWELL EPA SEALS E 3 $ 15 0 - E o. k ATMOSPHERIC PRESSURE g 10.0 - 2 h t; 5.0- -CORE UNCOVERY u 0.0 . . , , , , , i , i 0.0 100.0 200.0 300.0 400.0 500.0 600.0 700.0 800.0 900.0 1000.0 1100.0 TIME (min) Fig. 9.40 Wetwell steam partial pressure (TB'). ORNL-DWG 81-8552 ETD 4 (x 10 ) 25.0 i 20.0 - E o g FAILURE OF VESSEL BOTTOM HEAD g 15.0 - VENTING BY FAILURE OF DRYWELL EPA SEALS 5 / U "z 10.0 - e 8 ? I 5.U-Zr-H 2O REACTION BEGINS I 0.0 - , , , , , , , , , , 0.0 100.0 200.0 300.0 400.0 500.0 600.0 700.0 800.0 900.0 1000.0 1100.0 TIME (min) Fig. 9.41 Wetwell hydrogen partial pressure (TB'). 1 ORNL-OWG 81-8553 ETD (x 105) 7.0 6.0-E 3 5.0 - VENTING BY FAILURE OF $ DRYWELL EPA SEALS E a h4.0-FAILURE OF VESSEL BOTTOM HEAD $ 0 e g 3.0 - 0 2.0 - o V Zr-H 2O REACTION BEGINS Z O Z 1.0 - ALL NON-CONDENSABLES ATMOSPHERIC PRESSURE REPLACED BY STEAM 0.0 , , , , , , , , , , 0.0 100.0 200.0 300.0 400.0 500.0 600.0 700.0 800.0 900.0 1000.0 1100.0 TIME (min)

Fig. 9.42 Wetvell non-condensables partial pressure (TB').

ORNL-DWG 81-8554 ETD 1' (x 10 53 7.0 6.0 - 7 !b 3 5.0 - 8 $ FAILURE OF VESSEL VENTING BY FAILURE OF EPA SEALS [ BOTTOM . z y 4.0- HEAD y $ U 2 2 0 3.0-Zr-H O '0- REAdTION i. BEGINS 1.0 . . . . . . . 0.0 100.0 200.0 300.0 400.0 500.0 600.0 700.0 800.0 TIME (min) f Fig. 9.43 Wetvell total pressure (TUB'). 's s s ORNL-DWG 81-8555 ETD 4 (x 10 ) . 60- , VENTING BY FAILURE OF DRYWELL EPA SEALS , 5.0- WETWELL ATMOSPHERE IS NOT PURE STEAM g FAILURE OF !b VESSEL w 4 .0 - BOTTOM $ HEAD a E T o. p 3.0- - E 2 ! 2.0 -

• u) 1.0-0.0 , , , , , , ,

0.0 100.0 200.0 300.0 400.0 500.0 600.0 700.0 800.0 TIME (min) Fig. 9.44 Wetwell steam partial pressure (TUB') . ORNL-DWG 81-8556 ETD 4 (x 10 ) 20.0 2 fh 15.0 - w -VENTING BY FAILURE OF DRYWELL EPA SEALS W 3 v) FAILURE OF ' VESSEL d ....q' BOTTOM , j- h.* HEAD r m t ? $ t z  : m o o x n > 5.0-SOME HYDROGEN REMAINS IN WETWELL ATMOSPHERE 0.0 , , , , , , i 0.0 100.0 200.0 300.0 400.0 500.0 600.0 700.0 800.0 TIME (min) Fig. 9.45 Wetvell hydrogen partial pressure (TUB'). ORNL-DWG 81-8557 ETD (x 10 5g 6.0 + _ 5.0 - E E 3 h4.0- mVENTING BY FAILURE OF DRYWELL EPA SEALS E FAILURE OF _a VESSEL 5 BOTTOM A

• HEAD

$ 3.0-a. v> \ g m o "i E 2.0 - 8 z 8, 5 1.0 - - SOME NON-CONDENSABLES REMAIN IN WETWELL ATMOSPHERE 0.0 , , , i i . . 0.0 100.0 200.0 300.0 400.0 500.0 600." 700.0 800.0 l TIME (min) Fig. 9.46 Wetwell non-condensables partial pressure (TUB'). 4 e , 6 9 9 a - . . . .- . - - -. ~- -- -, .- ~ j 161-

j. 10. PLANT STATE RECOGNITION AND OPERATOR MITIGATING ACTIONS

, ~10.1 Introduction-1 i The accident signatures developed in Sect. 9 provide baseline infor-4 mation for six possible- sequences' leading to ~ core meltdown that might oc-l' cur during a Station Blackout. The' objectives:of this' chapter are to ex-amine these same sequences from the operator's standpoint in orde'r to- ~ identify and evaluate potential preventive and corrective actions keyed to the time windows available. l As discussed in Sect. 9, _ a number of plant- asfety systems would ' func-tion automatically during a Station Blackout. The automatic safety system responses include reactor scram,-vessel pressure control by SRV operation, and-level control through cycling of the HPCI system. Thus, assuming no }~ independent secondary failures, the reactor could be maintained in a safe stable state without operator action until DC power is lost at an assumed - four. hours into'the blackout.- Af ter the loss of DC power, operator action would be crucial in attempting to avert the impending core degradation. I 10.2 Plant State Recognition To enable the operator to better identify:and evaluate potential I ,_ mitigating actions during a Station, Blackout, the accident progression is categorized into a number _of plant states. The available time' windows for , each plant state are determined from the baseline information provided ~in . Sect. 9. The location of the plant states within each sequence is shown - in Fig.10.1, with the sequence chosen for the fission product transport

analysis of Volume 2. indicated by a dashed line.-

1 Plant State 1 (0-s). This is the initial state at the moment of loss of of fsite and- onsite power, and is common tc, all of the blackout sequences. l Plant State 2 (0-625 s). . This state represents the' period of. automatic: l- plant safety system response and is common to the six sequences considered  ; 4 in detail in this report. . { Plant State 3 (625 s-240 min). This state is'applic, le to sequences j TB' and TPB'.' It is characterized by the availability -of adequate de-

cay heat removal by relief valve operation and the HPCI and RCIC injection

systems, which 'are dependent on DC power from the unit battery.

j Plant State 3A (625 s-240 min). This state is applicable to sequences - T y B' and T yPB'. It is characterized by operator' action to control' level by operation of the RCIC system (only) and to rapidly depressurize the re-i actor vessel by remote manual' relief valve actuation. The loss of fluid through the relief valve exceeds the capacity of the RCIC system for a ~ short period of time, which causes 'a momentary core uncovery early in these j'. sequences. j Plant State 3B (625 s-28 min). This state is applicable to sequences TUB' I and IUPB'. In these sequences, the HPCI and RCIC systems are assumed to be inoperable from the inception of the blackout, because of equipment fail-

l. ure. This state is characterized by decreasing water level.

j Plant State 4 (240-295 min). This state is applicable to sequences TB' and . TPB'. This is the period immediately following the loss of DC power, I u_.____ - . _ _ . _ . . _ _ . . _ . _ .,. ~ .i . .u.u., & m ..u..24.4._. . < x .-- ._ia & u. .2. .. m .. .- - 4 . _.4.. _ - __ 4..~. ~ . . . . _ ., 4 a I

. t 4 4

_ ORht-DWG81-8163 ETD INITIATING EVENT . 8"V SYSTE M HPc6 AND MANUALSRv LOW PRESSURE ECCS LOSS OF OFFsaTE ONLY RCIC LOSS OF <se POWER *- DtPRESSURIZATION N E AUTOMATtC RCic DEPRE SSURllATION OPER ADLE - 07751TE OUTCOME '

OPE R ABLE - HPCL RCIC INOPE R ASL E SEFORE LP ECCS acPOWER RESPONSES OPE RA8(E WITHIN t$ m r OER RESTORED RESTORATICN I STATES 4 STATE $ t ORE MELT iTP8 ) 5 TATE 7 sTAM 4 l i I l STATE 3 t """"""""""" CORE MELT 178) I ! NO CC-RE OAMAGE F l l l l ST ATE 8A DAMAGE R,S' AND 7,Ps') 1 1 STATE 4A !517TD CORE MELT (7,8' AND T,78') PJ ATE 7A l ST ATt 3A . YES l A l NO CORE DAMAGE l 1 POTENTIAL CORE DAMAGE - , l = l 5 TATE 88 CO E DAMAGE N8) j . STATE 1 ' 51 ATE Sa E 1 CORE MELT ITUPB") STATETe } STATE 38 i ' I 4' V NO [ . STATE 68 POTENTIAL CORE DAMAGE Fig. 10.1 Operator key action event' tree. 2 I r e 9 , ,' 6 8 1 e - r11 163 power, characterized by the unavailability of the HPCI and RCIC systems and a steadily decreasing reactor vessel water level. Plant State 4A (240-305 min). This state is applicable to sequences T y B' and T yPB', and follows the loss of DC power. This state is char-acterized by a depressurized reactor vessel, unavailability of the HPCI and RCIC systems, and a steadily decreasing vessel water level. Plant State 5 (295-305 min). This state is applicable to sequence TPB' and is characterized by a stuck-open relief valve, so that the reactor vessel is depressurized. , Plant State 5B (28-38 min). This state is applicable to sequence IUPB' . This sequence involves the combined secondary failures of inoperative HPCI and RCIC systems and a stuck-open relief valve (SORV) from the inception of the blackout. It is characterized by a rapidly decreasing water level. Plant State 6 (295-355 min). This state is applicable to sequence TB' only. It is characterized by boilof f of the reactor vessel water inven-tory due to automatic relief valve actuation with the reactor vessel fully pressurized, and terminates with the beginning of core melt. Plant State 6B (28-70 min). This state is applicable to sequence TUB'. It is characterized by a pressurized reactor vessel, core uncovery, and the beginning of core melt. Plant State 7 (305-360 min). This state is applicable to sequence TPB'. It is characterized by a depressurized reactor vessel, unavailability of the low pressure ECCS injection systems, decreasing water level, core un-covery, and the beginning of core melt. Plant State 7A (305-360 min). This state is applicable to sequences T y B' and T yPB' is identical to state 7. Plant State 7B (38-360 min). This state is applicable to sequence TUPB' and is identical to state 7. Plant State 8 (305-360 min). This state is applicable to sequence TPB'. It is characterized by a depressurized reactor vessel and the recovery of AC power, which permits the restoration of normal vessel water level. Plant State 8A (305-360 min). This state is applicable to sequences T y 5' and T y PB' and is identical to state 8. Plant State 8B (38-360 min). This state is applicable to sequence TUPB' and is identical to state 8. 10.3 Operator Key Action Event Tree For sequences TUB' and TUPB', and af ter the loss of DC power in the case of sequences TB', TPB', T y B', and Ty PB', mitigating action by the operator is essential if core degradation is to be avoided and the plant brought to a safe stable state. Upon restoration of AC power, the operator's primary responsibility is to ensure vessel depressurization and operation of the low pressure in-jection systems. , The operator key action event tree is shown in Fig.10.1. 10.4 Operator Mitigating Actions In addition to continued efforts to restore electrical power, the recommended actions af ter the loss of water injection capability when 164 boilof f and core damage become inevitable include the use of portsble pumps, perhaps provided by fire engines, to flood the drywell in an at-tempt to preclude melt-through of the reactor vessel. If successful, this would keep the ' degraded core inside the vessel and prevent gross contain-ment failure and the corresponding major releases of radioactivity. For the sequences TUB' and TUPB', in which injection capability is assumed lost at the inception of the blackout, a drywell flooding rate of 3,200 A/s (50,000 gpm) would be necessary to flood the drywell to the level of the - reactor vessel core prior to reactor vessel melt-through. t 4 [ 2 165-11.. INSTRUMENTATION AVAILABLE FOLLOWING LOSS OF 250 VOLT DC POWER ~ Reactor vessel leve1 and pressureLcontrol can be maintained during a Station Blackout for as long as 250 volt DC power from the unit. battery 2 remains available, as discussed in Sect. 8. The instrumentation available . during this initial phase of. a Station Blackout was discussed in Sect. 5. It is the purpose of this section to discuss the instrumentation which , would remain operational af ter the unit battery is exhausted; this final l phase of a Station Blackout would constitute a Severe Accident because

there would be no means of injecting water into the reactor vessel to maintain a water level over the core.

In addition to the 250 volt unit battery system, there are two smal-ler battery systems which supply power to Control Room instrumentation and + alarm circuits. .The first of these is a 24 volt DC system which supplies power to the Source Range and Intermediate Range neutron flux monitors as well as radiation monitors for the off gas, RHR Service Water, Liquid Rad-waste, Reactor Building' Closed Cooling Water, and Raw Cooling Water sys-tems, none of which would be operational during a Station Blackout. The 24 volt batteries.for this system are designed to supply the connected loads for a period of 3 hours3.472222e-5 days <br />8.333333e-4 hours <br />4.960317e-6 weeks <br />1.1415e-6 months <br /> without recharging. Since the 250 volt sys-tem batteries are expected to last for a period of four to six hours, it-is unlikely that the 24 volt system would remain available af ter the 250

volt DC unit distribution system has failed.

. The second of the smaller battery systase is a 48 volt.DC power sup ply and distribution system for the operation of the. plant's communication

and annunciator systems. This system comprises three batteries, one of.

i .- which supplies the ' plant communications system while the- remaining- two batteries are for the annunciator system. However, the system design pro- , vides that the total station annunciator load can be supplied from one batte ry. The 48 volt DC system batteries are capable of supplying the connected loads for a period of eight hours without recharging. -This is well beyond the period of expected operation of the 250 volt DC systems and assures continued availability of the plant communications system. The ef ficacy of the annunciator system depends' upon the availability of the power supplies to the signal transmission ' systems of the various sen- - sors as well as the 48 volt DC system. Most of the alarm annunciator cap-ability will be lost when the unit battery fails. There is reason to believe - that plant preferred 120 volt AC single [ phase power would remain available for a significant period of time after ~ j f ailure of the unit ' battery. _ Plant preferred power is obtained from the plant 250-volt DC system during a Station Blackout by means of'a DC-motor, , AC generator combination. The plant battery is similar to each of the ! unit batteries, but would be more lightly loaded during a Station Black-out. The major loads on the plant battery are the turbo generator oil . pumps and seal-oil pumps .for the three Erowns Ferry Units. These pumps could be stopped within one hour af ter the inception.of a Station Blackout since the turbo generators will have stopped rolling and there is no jack- - a ing capability under Station Blackout conditions. l Assuming that plant preferred power does remain available during the period of core uncovery following the loss of the unit battery, two k i -- ~ . . . . . 166 sources of valuable information powered by this system and located outside of the Control Room would remain. The first is an' indication of the ten - peratures at various points within the drywell provided both by a recorder and an indicator meter for which the _ displayed drywell temperature can be selected by a set of toggle-switches. 4 The second luformation source provided by plant preferred power. consists of _ the tLaperatures at various points on the surface of the re-

• actor vessel as provided by 46 attached copper-constantan thermocouples.

Unfortunately, the . indicating range for the instruments reporting the thermocouple responses is only 315.6*C (0-600*F). Howeyer, when the in- , strument responses for the thermocouples attached near or on the bottom of the reactor vessel are pegged at the high end, the operator can be sure that the core is uncovered, and the only fluid within the vessel is super-heated steam. Both the drywell temperature indication and the reactor 4 vessel thermocouple recorder are mounted on panel 9-47, which is located

• on the back of the Control Room Panels.

Information concerning the reactor vessel water level would also be available outside the Control Room from the mechanical Yarway indication (which requires .no electrical power) located at the scram panels on the second floor of the reactor building; this would provide direct indication i of the reactor vessel water level over the range from 14.94 to 9.47 m (588 to 373 in.) above vessel zero. The lower boundary of this Yarway indicat-ing range is 0.33 m (13 in.) above the top of the active fuel in the core. There would be no indication of reactor vescel pressure af ter loss of the unit battery. However, it is expected that the. vessel pressure would a be maintained in the range of 7.722 to 7.377 MPa (1105 to 1055 psig)~ by repeated automatic actuation of a primary relief valve for as long as the

• reactor vessel remains intact.

It should be noted that the emergency lighting for the Control Room is supplied from the 250 volt DC system; af ter f ailure of this system hand-held lighting for the Control Room would be necessary. The door se-curity system is supplied by plant preferred power and would remain oper-able as long as the plant 250 volt DC system is functional. The area radiation moticors located throughout the plant are powered from the Instrumentation and Control buses and would not be operational from the inception of a Station Blackout. 4 e ,-r --- - . . . - - . . , . , - . , , . , . , ._.,-,,.,n.n -m-, n . -- , , - ---.-,..-------n-- -,-- - ,. , , -167

12. D4PLICATIONS OF RESULIS The purpose of this section is to provide a discussion of the pre-sent state of readiness at the Browns Ferry Nuclear Plant to cope with a Station Blackout. The discussion will include consideration of the avail-able instrumentation, the level of operator training, the existing emer-gency procedures,' and .the overall system design.

12.1 Instrumentation The availability of Control Room instrumentation during the period of a Station Blackout in which 250 volt DC power remains available has been discussed in Sect. 5. The most important parameters while injection cap-ability remains are the reactor vessel level and pressure, and these would be adequately displayed in the Control Room during this period. This in-strumentation available af ter. the loss of DC power was discussed in Sect, 11. The existing instrumentation at Browns Ferry Unit I which would be important during a Station Blackout is summarized below, with additional information concerning the power supplies. The instrumentation is discus-sed in the order given in reference 28, and is located in the Control Room unless otherwise indicated. - I. - C Te A. Core Exit Temperature ;- not available at Browns Ferry. B. Control Rod Position - the rod position indicating system is powered by the-120V AC unit preferred system, which would be available during a Station Blackout for as long as 250V--DC power remains available from the unit battery. C. Neutron Flux._ The Source Range and Intermediate Range moni-tors are powered by the 24 volt DC battery system, which is de- < signed to supply the connected loads for a period of 3 hours3.472222e-5 days <br />8.333333e-4 hours <br />4.960317e-6 weeks <br />1.1415e-6 months <br /> under the conditions of a Station Blackout. The Source and In-termediate Range detectors are withdrawn from the core durir ; power operation to a: point 0.61 m (2 ft) below the-bottom of the active fuel to increase their active life, and these detectors could not be reinserted during a Station Blackout. However, the i relative changes in neutron flux following reactor scram will produce a proportional change in Source Range meter (range: 0.1 ! to 106 CPS) response when the detector is f ully' withdrawn. i II. Reactor Coolant System i A. RCS Pressure . one channel of reactor vessel pressure powered l by the feedwater inverter and another channel powered by the. l unit preferred system, each with a range of 8.274 MPa (0 to 1200

. psig), would be available during a Station Blackout for as long L, as the unit battery continues to supply 250 volt DC power.

! B. Coolant Level in the Reactor two channels of reactor vessel level instrumentation, one each ~ powered by the feedwater in-l verter and the unit preferred system would be available during ' Station Blackout for as long as the 250 volt DC unit battery o I L I 168 system tenains functional.: ' These two channels each have a range ~ a of 1.524 M (0 to 60 in.) and provide indication of the reactor , vessel water level between 13.41 and 14.94 m (528 and 588 in.)- above the bottom of the vessel, a range which extends over the b upper portion of - the steam separators. - Additional level infor-nation would be.available outside of the Control Room, as fol-

lows

.

o Two channels of Yarway level indication over the range from I 9.47 'to 14.94 m (373 to 588 ir.) above the bottom of the ves-- sel are available at the backup control panel located in the - Backup Control Room. This instrumentation is powered fron

the unit preferred system; the lower limit of this indication 1 is 0.33 m (13 in.) above'the top of the active fuel in the i core.

o The Yarway -instruments are located at the scram panel on the

, second floor ~of the reactor building and mechanically display ! the reactor vessel water level over the same range' as that ' electrically transmitted-to the backup control panel. This level indication at the Yarway instruments requires no elec-trical power, and would remain availaole even after all DC power was lost. C. Main Steamline Flow -- two channels of steam flow instrumenta-i tion powered by the feedwater inverter would provide an indica-tion of total steam flow [ range: 2016 kg/s (0 to 16 x 10 6 , lbs/h)] during a Station Blackout while the unit battery remains . functional. D. Main Steamline Isolation Valves' Leakage Control System Pressure l - not applicable to Browns Ferry.- a E. Primary System Safety Relief Valve Positons. Including ADS - There is no provision for indication of the actual position of any primary relief valve, including those assigned to the ADS system. Under normal operating conditions, the operator can identify a leaking relief valve by means of the recorded tail-pipe temperatures available on charts behind the Control Room panels, or by. the recently installed (December,1980) acoustic valve monitors. Both of these. leakage detection systems would ,, be inoperable under Station Blackout conditions. l On the other hand, remote manual actuation of a relief l valve is accomplished by energizing the associated DC solenoid

operator, and each valve has lights on the control panel which

! indicate whether or not the solenoid for that valve is ener-gized. The capability for'this indirect indication of success-l ful remote manual valve actuation is maintained during the per- ! iod of a Station Blackout in which 250 volt DC power is avail-able from the unit battery. j' F. Radiation Level in Coolant -- not available at Browns Ferry. * , III. Containment i A. Primary Containment Pressure -- Drywell pressure instrumenta-tion with a range of 0.55 MPa (0-80 psia) powered by the *

unit preferred system will be available during a Station i

__ _ . , -_._._ ____ _ _ _ ___--___.._ _ _ __ _ _ _ _ - . ~ . - t 169 ' Blackout while the unit battery remains functional. A set of 12 vacuum breakers ensures that the pressure in the pressure sup-pression pool torus cannot exceed the drywell pressure by more than 0.003 MPa (0.5 psi). B. Containment and Drywell Hydrogen Concentration -- Instruments for this purpose ara provided for each Browns Ferry Unit, but would not be available during a Station Blackout. C. ' Containment and Drywell Oxgyen Concentration -- Instruments , for this purpose are provided, but as in the case of the hydro- . gen monitors, are powered from the Instrumentation and Control 4 buses which would not be available during a Station Blackout. D. Primary Contairment Isolation Valve Position . During a Sta-

tion Blackout, valve position indication is maintained for the Main Steam Isolation Valves (MSIV's) and for the systems which remain operable, i.e., the HPCI and the RCIC systems._ However, valve position indication for the primary containment isolation valves in the low pressure injection systems which are powered by sources not available during a Station Blackout would be lost.

E. Suppression Pool Water Level -- This instrument, powered by the unit preferred system, would indicate the water level in the torus with a range of _1.27 m (-25 to +25 in.) as long as the unit battery is f anctional. F. Suppression Pool Water Temperature -- This instrumentation would not be available during a Station Blackout. G. Drywell Pressure -- This is synonymous with " Primary Contain-ment Pressure" -- see "A" above. I H. Drywell Drain Sumps Level -- not indicated. The frequency of drain pump operation for each of the two 1.893 m3 (500 gal.) drain sumps in the drywell is indicated in the Control Room under normal conditions, but would. not be operable during a Sta-tion Blackout. I. High-Range Containment Area Radiation - not available at Browns Ferry. IV. Power Conversion Systems A. Main Feedwater Flow -- one channel of flow instrumentation with a range of 1008 kg/s (0 to 8 x 10f lbs/h) and powered - by the unit preferred system is provided for each of the three feedwater pumps. This instrumentation would remain operable 'as l long as the unit battery is functional. B. Condensate Storge Tank Level -- The condensate storage tank , level indication [ range: 9.75 m (0 to 32 f t)] is powered by the unit preferred system and would remain operational until the

• unit battery is exhausted.

i V. Auxiliary Systems A. Steam Flow to RCIC -- This instrumentation is powered by the Instrumentation and Control buses and would not be available during a Station Blackout. -., -,..rv,--,-,.r-,,-- , , , . - . - . - . , , - . , - - . - . - - - - . , , , ... ,~, .- .- ..,..----.----a , -- , 170 B. RCIC Flow -- The RCIC pumped flow is indicated over a range of 0.044 m3 /s (0-700 GPM) on the flow controller, which is powered by the unit preferred system and would be available as long as the unit battery is functional. VI. Radiation Exposure Rates -- The permanently-installed radiation monitors located throughout the plant are powered from the Instrumentation and Control buses and would not be operational during a Station Blackout. The available Control Room instrumentation is adequate to monitor the ' plant response to a Station Blackout during the estimated four-hour period 'uring which 250 volt DC power would remain available from the unit bat- , ce ry. Af ter the unit battery is exhausted, virtually all Control Room in-strumentation would be lost and water could no longer be injected into the reactor vessel to make up for that lost to the pressure suppression pool through relief valve actuation. Consequently, the reactor vessel water level would slowly decrease until the core became uncovered. The most important single parameter during the period of decreasing water level following loss of the unit battery is the reactor vessel level, which could be monitored at the Yarway instruments on the second floor of the Reactor Building as described in part III B above. In addi-tion, the drywell ambient temperatures and the reactor vessel surface tem-peratures would remain available as long as the plant preferred 250 volt-DC system is functional, as described in Sect. 11. 12.2 Operator Preparedness There is currently no specific training provided Browns Ferry Oper-ators in regard to plant response or required operator actions in the ev-ent of a couplete Station Blackout. . A Station Blackout casualty can be run at the TVA Browns Ferry Unit I simulator by selecting the pre programmed loss of-offsite power casualty and then having the simulator operator immediately punch the diesel gener-ator manual trips in the control room. Since the instrumentation in the simulated control room is powered from several diverse sources representa-tive of those at the actual plant, the resulting effect closely models a Station Blackout. This procedure was developed and tested by TVA simula-tor operating personnel as an effort in cooperation with this study, but has not been used for operator training. There is no Emergency Operating Instruction to cover a Station Black-out at the Browns Ferry Nuclear Plant, but there are Emergency Operating Instructions for the loss of individual subsections of the overall AC sys-tem. With two exceptions, the current level of operator training and the existing Emergency Operating Instructions are believed adequate to prepare the operator to cope with a Station Blackout during the initial phase in which 250 volt-DC power would remain available. The first exception in-volves the need to depressurize the reactor vessel within the first hour. . The Technical Specifications provide that the reactor vessel must be de-pressurized to less than 1.48 MPa (200 psig) if the pressure suppression pool temperature exceeds 48.9'C (120*F); this is based on a maximum per- . missible temperature of 76.7*C (170*F) for the pool so that complete con-densation of steam is ensured in the event of a LOCA. The Emergency 171 Operating Instructions and the Operator Training manuals establish this requirement, clearly.- During a Station Blackout, suppression pool temperature indication-is not available and the operator may be reluctant to depressurize the reac-tor vessel, knowing that suppression pool cooling is not operable. How-ever, depressurization is necessary for two reasons. The first is to pre-clude an excessive drywell ambient temperature by reducing the temperature ~ of the saturated fluid within the reactor vessel, as discussed in Sect. 3. The second reason is 'to provide additional time between the loss of injec- - . tion capability and core uncovery; as explained in Sect. 7, no coolant is J ost_ from the reactor vessel during the significant period of time re-quired for the reactor vessel to repressurize to the set point for auto-matic relief valve actuation. , With an early depressurization, a great deal of energy and coolant mass is dumped to the pressure suppression pool early in the Station , Blackout when the HPCI and RCIC systems are available to restore the lost coolant and maintain vessel level in the normal operating range. Should the Station Blackout continue to the point that the HPCI and RCIC systems f ail due to exhaustion of the unit. battery, a significant amount of time is required af ter injection failure for the reactor decay heat to add the energy required for vessel repressurization. Thus early reactor vessel , depressurization will provide valuable additional time for corrective ac-

i. tion before core uncovery.

The second exception to the current level of operator training cnd the Emergency Operating Instructions with regard to a Station Blackout in-

• volves the need~for the operator to reduce the load upon the 250 volt DC system as much and as quickly as possible to prolong the life of the unit-battery. There is a need for a procedure which lists all of the loads upon the unit battery, indicates whether or not each load is significant, the purpose of each load, and the consequences of removal.

Operator actions during a Station Blackout should be directed toward keeping the vessel level in the normal operating range, early vessel de-pressurization, and reduction of the battery load. All of these actions will increase the time available in which maintenance actions can be taken to restore AC power before core damage occurs. The recommendations of-this study concerning operator preparations can be summarized as:

1. Operator training and the Emergency Operating Instructions should ex-i plain that reactor vessel depressurization reduces the driving head for heat transfer into the drywell during a casualty, and provides a significant additional margin of time before core uncovery occurs once injection capability is lost. The present operator training implies that the only reason for depressurization is to satisfy the Technical Specification, written in anticipation of a LOCA.

2. The plant batteries are provided as energy-storage systems for tem-porary use when the normal AC power supply systems are unavailable.

l However, during a Station Blackout or other casualty in which the nor-mal AC supply to a battery charger is unavailable fcr a significant period of time, the operator is faced with the necessity of opening the power-supply breakers to the less-important loads to prolong the , availability of power to the more essential equipment.- It is recom-mended that the priority of the battery loads be established in ad-vance of the casualty, by means of a procedure which indicates the 2 P , y - y , .yy..yn,.-.- , - , , _ _w,.. . -.. - - , -,. 7 - .m , - ., .,y v-., . - , - - , gw-c , , , .a I 172 recommended order of removal of loads if the loss of AC power is per- , ceived to be long-term.' l

3. Recovery procedures should be developed to provide a detailed method '

for recovery of vital power supplies and equipment in a safe, effi-cient manner upon restoration of AC power following a Station Black-out. 12.3 System Design

• The existing system design provides sufficient instrumentation and

• equipment to maintain decay heat removal capability for several hours dur-ing a Station Blackout.- The -only questionable feature of the existing de-sign with regard to the Station Blackout sequence is the provision for automatic shif ting of the HPCI pump suction to the pressure suppression

- pool on high sensed pool level. As discussed in Sect. 8.1, this can lead to failure of the HPCI system because of the high temperature of- the pre-saure suppression pool water at the time the shif t in pump suction oc-curs. Separate provision is made for an automatic shif t of the HPCI pump suction if the normal condensate storage tank water source becomes ex-hausted. Thus it appears that the automatic high pool water level shif t must have '- n straight-forwardly based on a concern for the effect of high watet Aeeel in the pressure suppression pool. The basis is not given in the Technical Specifications, and it should be noted that there is no corresponding provision for the RCIC system. It is recommended that the , desirability of an automatic shif t in HPCI pump suction on high suppres-sion pool water level be reexamined and if found necessary, that the basis be included in operator training. , l l , 173 a

13. REFERENCES

1. Browns Ferry FSAR, Appendix F.

2. Nuclear Regulatory Commission Memorandum from H. R. Denton to Chair- -

man Ahearne, dated September 26, 1980:

Subject:

STATION BLACKOUT.

, 3. Browns Ferry FSAR, Paragraph 8.5.2.

1 4 R. L. Scott, " Browns Ferry Nuclear Power-Plant Fire on March 22, j 1975," Nielear Safety, 17-5, 592-411 (1976).

l 5. Browns Ferry PSAR-, Response to AEC Question 4.8.

6. Staff Report, Anticipated Tmnaiente Without Scram for Light Water Reactore, NUREG-0460, Vol. 2, p. XVI-74.

7. -Browns Ferry FSAR, Response to AEC Question 12.2.16.

I. 8. Continuous System Modeling Program III (CSMP III) Program Reference Manual, IBM Corporation, 4th Ed. (December 1975).

9. " Decay Energy Release Rates Following Shutdown of Uranium-Fueled Thermal Reactors." ANS-5.1, American National Standards Institute (1971).

10. R. T. Lahey, Jr. and P. S. Kamath, "The Analysis of Boiling Water Re-

'

• actor Long-Term Cooling," Nielear Technology, Vol. 49 l(June 1980).

! 11. J. F. Wilson, R. J. Grenda, and J. F. Patterson (A-C), " Steam-Water Separation Above a Two-Phase Interface," Tmne. Am #rel. Soc.,

i 4,356'(1961).

[ 12. R. T. Lahey, Jr. and F. J. Moody, "The Thermal Hydraulics of a Boil-ing Water Nuclear Reactor," American #selaar Society (1977).

13. R. O. Wooton and H.1. Avci, MARCH Code Description and User's Man-ual, Battelle Columbus Laborataories/USNRC Report NUREG/CR-1711 (Oc-tober 1980).

14. D. W. Hargroves and L. J. Metcalfe, CONTEMPT-LT/028: A Computen Progmm for Predicting Containment Pressure-Tempenture Response to a Loss of Coolant Accident, EG6G Idaho, Inc./USNRC ' Report NUREG/CR-0255 (TREE-1279) (March 1979).

15. Browns Ferry Emergency Operating Instruction No. 36, Loss of Cool-ant Accident inside Dryssett.

16. S. E. Mays, et al, Second IREP Status Report, Risk Assessment for Browns Ferry Nuclear Plant Unit 1, January 1981 (Preliminary).

174

17. Browns Ferry FSAR, Section 8.6.

18. D. D. Yue and W. A. Condon, " Severe Accident Sequence Assessment of Hypothetical Complete Station Blackout at the Browns Ferry Nuclear Plant," 1981 International ANS/ ENS Topical Meeting on Probabilistic Risk Assessment, Port Chester, N.Y. , September 20-24, 1981.-

19. Reactor Safety Study, WASH-1400. (NUREG-75/014), U.S. Nuclear Regu-

• latory Commission, 1975.

20. F. E. Haskin, et al, " Analysis of a Hypothetical Core Meltdown Acci-dent Initiated by Loss of Offsite Power for the Zion 1 PWR," NUREG/

CR-1988, Sandia National Laboratories,1981.

21. M. L. Corradini, " Analysis and Modelling of Steam Explosion Experi-ments," NUREG/CR-2072, Sandia National Laboratories, April 1981.

22. D. D. Yue, " Electric Penetration Assembly," U.S. Patent 4, 168, 394, September 1979.

23. American Nuclear Society Proposed Standard ANS-5.1, " Decay Energy Release Rates Following Shutdown of Uranium -- Fueled Thermal Reactors," October 1971, revised October 1973.

I 24 ANSl/ANS-5.1 (1979), " Decay Heat Power in Light Water Reactors,"

i.ugust 1979. "

25. W. B. Wilson, et al, " Actinide Decay Power," LA-UR 79-283, < Los Alamos Scientific Laboratory, June 1979. *

26. T. E. Lobdell, "REDYV01 User Manual," NEDO-14580, August 1977.

27. D. D. Chrj stensen, " Interim Browns Ferry RELAP4/ MOD 7, Station Black-out Calculation," DDC-2-80 Idaho National Engineering Laboratory, October 31, 1980.

28. Table 3, BWR VARIABLES, of Revision 2 to Regulatory Guide 1.97, "In-strumentation for Light-Water-Cooled Nuclear Power Plants to Access Plant and Environs Conditions During and Following an Accident" (December 1979).

29. Browns Ferry FSAR, Paragraph 5.2.1.

30. Safety Evaluation Report - Mark I Containment Long-Term Program, NUREG-0661, July 1980.

31. Three Dimensional Pool Swell Modeling of a Mark 1 Suppression System,

• EPRI NP-906, October 1978.

32. H. Nariai, I. Aya, M. Kobayaski, "Thermo-Hydraulic Behavior in a Model Pressure Suppression Containment During Blowdown," Topics in Tuo-mass Heat Imnsfer and Floo, pp. 89-98, Winter ASME Meeting, 1978.

175

33. W. G. Anderson, P. W. Huber, A. A. Sonin, "Small Scale Modeling of Hydrodynamic Forces in Pressure Suppression Systems," NUREG/CR-0003, March 1978.

34. B. D. Nichols, C. W. Hirt, " Numerical Simulation of Boiling Water Reactor Vent-Clearing Hydrodynamics," Ntel. Sci. A Engr., 73, 196-209.

35. D. Brosche, 74odel for the Calculation of Vent Clearing Transients in

, Pressura Suppression System," Nucl. Engr. 4 Des., 38,131-141, 1976.

36. J. _ S. Marks, G. B. Andeen, " Chugging and Condensation Oscillation Tests," EPRI NP-ll67, September 1979.

37. L. E. Stanford, C. C. Webster, " Energy Suppression and Fission Pro-duct Transport in Pressure-Suppression Pools," ORNL-Di-3448, April 1972.

38. M. Okazaki, " Analysis for Pressure Oscillation Phenomena induced by Steam Condensation in Containment with Pressure Suppression System,"

Niel. Sci. and Tech. J_6_, pp. 30-42, January 197 9.

39. J. S. Marks, G. B. Andeen, " Chugging and Condensation Oscillation,"

Condensation Heat Transfer, pp.93-102, 18th National Heat Transfer Conference, San Diego, California, August 1979.

40. w. Kowalchuk, A. A. Sonin, "A Model for Condensation Oscillations in a Vertical Pipe Discharging Steam into a Subcooled Water Pool,"

NUREG/CR-0221, June 19784

41. D. A. Sargis, J. H. Stubhmiller, S. S. Wang, "A Fundamental The rmal Hydraulic Model to Predict Steam Chugging Phenomena," Topic in Tuo-Maase Heat Transfer and floo, pp. 123-133, Winter ASME Meeting, 1978.

42. P. A. Sargis, et al. , " Analysis of Steam Chugging Phenomena, Vol.1,"

EPRI-NP-1305, Janauary 1981.

43. B. J. Patterson, " Mark 1 Containment Program: Monticello T-Quencher Thermal Mixing Test Final Report," NEDO-24542, August 1979.

44. " TRAC-PIA An Advanced Best-Estimate Computer Program for PWR LOCA Analysis," Los Alamos Scientific Laboratory, NUREG/CR-0665, LA-7777-MS, May 1979.

45. U.S. Rohatgi and P. Saha, " Constitutive Relations in TRAC-PIA,"

Brookhaven National Laboratt ry, NUREG/CR-1651, BNL-NUREG-51258, August 1980.

l I

176

46. " Application of the Coupled Fluid Structure Code PELE-IC to Pressure Suppression Analysis - Annual Report to NRC for 1979," Lawrence Livermore Laboratory, nUREG/CR-1179, UCRL-52733.

47. C. W. Hirt, B. D. Nichols, and 'N. C. Romero, "SOLA - A Numerical Solution Algorithm for Transient Fluid," Los Alamos Scientific Laboratory, LA-5852 (1975). ,

48. Browns Ferry Startup Test Instruction No.15, Nigh Pressure Infeo-

' tion System. *

49. Flou ~ Diagram Condensate Storage and Supply System, DMG No. 47W818-1 R10.--

50. Browns Ferry System Operating Instruction No. 64, Prina2ry Contain-ment Unit I, II, or III.

51. Browns Ferry FSAR Section 7.3, Printxry Containment and Reactor Vessel Isolation Control System.

52. Browns Ferry System Operating Instruction No. 73, High Pressure Coolant Injection System Unit I, II, or III.

53. Browns Ferry Startup Test Instruction No.14, Reactor Core Isolation Cooling System. ,

54 Browns Ferry Nuclear Plant Hot License Training Program Volume 4, Reactor Core Isolation Cooling System. ,

55. Browns Ferry FSAR Section 4.7, Reactor Core Isolation Cooling Sys-tem.

56. Browns Ferry FSAR Figure 4.7-3, Reactor Core Isolation Cooling Sys-tem Process Diagram.

57. Browns Ferry System Operating Instruction No. 71, Beactor Core Iso-Lation Cooling Unic I, II, or III.

58. B. J. Patterson, " Mark I Containment Program: Monticello T-Quencher Thermal Mixing Test Final Report," NED0-24542, August 1979 4

177 ACKNOWLEDGEMENT Mr. William A Condon made a significant contribution to this report by performing the MARCH code calculations and preparing the associated plots for the six station blackout accident sequences discussed in

• Chapters 9 and 10. This accomplishment is especially noteworthy because all MARCH computer runs in support of this project were by necessity

, made remotely on the CDC computer facilities at Brookhaven National Laboratory. Mr. Condon's work was in partial fulfillment of the re-quirements for the degree of Muster of Science with major in Nuclear Engineering at the University of Tennessee, and his lone hours _of dedicated effort are gratefully acknowledged.

4 1

1

V 179 8

APPENDIX A LISTING OF PLANT RESPONSE CODE B W LACP e

up b

181 sssCONTINU005 SYSTEM MODELING PROGRAM III v1M3 TRANSLATOR CUTPui sst LABEL --------------

---

6WR-LACP C00E------------- ------

LABEL CONTAINMENT MODEL WITH 1-N00E POOL i *----- --MACR 0 SUBPRCGRAM FCR POOL WATER T E M P C A L CU L AT I ON------------

M ACR0 KMC1.TI,WEI.MWI .LWi=WI TL TR LWL. LWR TkIO,FWI,FSVI .FTE!,FOI FS il

• MACR 0 TC CALCULATE MASS ENERGY BAL ANCES FOR SP WATER NCOE 1

• INPUTS :

• TL,TR= TEMPS OF WATER IN ADJACENT NCOES

,

• LWL LWR = LEVELS OF ADJACENT NCDES l

• TWIO=lNITIAL WATER TEMP CF NCOE I

• FWI= FRACTION OF TOTAL SP WATER VOL IN NCDE I tCONST ANT I

• FSVI= FRACTION OF SAFETY VALVE FLOW DIRECTED TO NODE I

• FTEI= FRACTION CF TURBINE EXHAUST . DIRECTED TO N00E I

• FDI= FRACTION OF PUMP DUSCHARGE DIRECTEC TC NCOE I

• FSI= FRACTION OF PDMP SUC TION TAKEN FROM NCDE I

• MASS BALANCE

• WSSPh=PUNP SUCTICN FLCWITOT AL FROM SP)

• WOSP W= PUMP 015 CHARGE F LOWI TOTAL TO SP)

• DUTPUTS:

• XNQ1=FRACTIN

• XNQI=FRACTCON CF STM CISCH TC NODE 1 NOT CUENCHED

• XNQIsFRACTION OF STM DI SCH TO NODE 1 NOT CUENCHED

• TI= AVG. WATER TEMP OF NOCE I (DEG-Fi

• WEI= RATE CF EVAPORATICN FRCM NCDE I ILB/SECl

• MWI= MASS OF WATER IN NCOE I (LBI

• LWI=W ATER LEVEL OF N00E I IIN)

• VWSP O=lN!TI AL TOTAL SP WATER VOLUMEICU.FT.)

MW IO= FWI *VW SP O

• t 6 3.9 .019 *TWIC I CMW I = l NT GRL t 0. , XQ I

• t FSV l *W SSV +FT E ! *WST E l *F INI-F S! *W SSP W-WEI)

, F I NI = WEGL+WE G R+F WI

• WCSPG+FD I* WOSP W MWI=MWIQ+CMWI VWI=MW I/ t 63.9 .019*T !!

LWI=AFGENISPLEV VWI/FWI)

• C ALC. QUENCH FRACTION FROM ASSUMPTION OF NO QUENCH WHEN NODE V APCR

• PRES $URE EQUAL TC TOTAL GAS PRES $URE P VAPl =NL FGENI SPF OT,TI )

XQI=L IMIT(0.,1. t PTSPG-PVAPI-DPQZl/0PQRI XhcI=tl.-XQ1)

• SPFOTtTIl= SAT. PRESS AS F OF TI

• DPQZ=0!FFERENCEtPSil CF PTSPG OVER PVAPI REQUIRED FOR ANY

• QUENCHIhG TO T AKE PL ACE

• DPZR=RAhGEIPSil CF PTSPG DVER PV API CVER WHICH GUENCHING

• GOES FROM O.0 TO 100 PERCENT

• EV APCR AT ICN R ATE 8 AS ED CN ECUN.13-33. RREITH'S ' HEAT TRANSFER'

• PS$PG=STM. PART PRESSURE ON SP G AS

• ASSPW= SURF AREA 0F SP WATERIFT.SQl WEI=tWEIN/WEIDa*CCMPARIPVAPI,PSSPG)

WEIN=ASSPh*FWI*HS!*12.3*tPVAPI-PSSPGl*PTSPGeVGSP W E I D= t PNSPG*2.-P VAPI I

• TG SPRe t MS SPG+ MN SPG)

HSI=5.83E-05*ttABSITI-TGSPil**.3331

.

• EQUALIZATION FLOWS FRCM ACJ ACENT NODEI SI

• FWE=EQ. FLOW CChSTANT IL8/SEC/IINCH LEVEL O!FFI I WEQL= F hE

• t LWL-LWI I W EQR= FW E

• t LWR-LW i l e

• ENERGY 8ALANCE

• HE AT TRANSF RATES TO METAL AND ATMCSPHERE NEGLEGIBLE

182

• WITH RESPECT TO MASS-HIXING ENERGY T RAN SF E R CMHwi = I hTGRL (0. , X41

• l F SV I *W S $v 8 HST+ hST E *FTE !*HS TE l+E 11 +E213 E 1I = ( T LK-32. l

• hEG L+ t TRX-32. l o w ECR+wML * ( T L-T i l +W MR * ( T R-T I)

E 21 = F W i *WC SPG *HF G SP+ F D I

• WO SP h* HD- F S I

• m S $ P we ( T I-32.1 -W E I

• 110 5.

TLX =T L *CCFP AR (LW L LW i l + T !

• COMP AR(LWI ,Lhl)

TRX=T R*C CPP AR 8 LhR, LW11 + T ! *CCMP AR (LW I, LW R )

• CALC OF dlXING FLOWStTHEY DID NOT APPE AR IN THE MASS 8ALANCE BECAUSE

• ZERO NET MASS TRANSFER IS ASSUMED WML=KNI X* SCRTI ABSI TL-T !l l

• WMR=KMix*SQRTI ABSI TR-Till TEMPERATURE CALCULATICN ASSUMES COhSTANT SPECIFIC HEAT =L.3 FOR W AT ER TI=32. + (ITWIO-32.l*MWIO + CMHhil/Mh!

EN DM ACR0 INITIAL

• -- - tNIT I AL IZ ATICN ' FOR REACTOR V ES S EL C ALCUL A T 10M----- -

• CCNSTANTS FOR REACTOR MCOEL

• ACOP=CCRE OUTLET PLENUM FLOW AREA (FT**21

• ACOR=CCRE FLOW AREA 4FT**2)

• ART = RISER TUBE FLOW AREAIFT**2B

• CPO= PRE-TRIP CORE PCWERITOTAL) (MWTH)

• EQX0= FRACTION CF WAY TO SATURATICM THAT STE AM CCNT ACT RAISES

• - INJECTION WATER IF DC LEVEL IS AT LEVEL GF JET PUMP SUCTION

• FFL ASH =FR ACT ION FL ASHED /SEC PER STU/LB ABOVE SATURATICN

• HCIO= INITIAL CCRE thlET ENTHALPY(BTU /L8)

• HINJih=ENTHALPY OF lhJECTION FLUI0tBTU/Let ,

• LBOT= HEIGHT CF STM SEP BOTTCMtFT)

• LDC0=INITI AL COwhCCMER LEVEL (FT l (HEIGHT ABOVE 80T. OF ACT. FUEL)

• LHE0ER= HEIGHT OF FW HEADER ABOVE 8CAFIFil

• LOP =2VG. LENGTH OF CORE CUTLET PLENUM (Fil ,

• LRT= AVG. LENGTH CF RISER TUBES (Fil (STANCPIPES)

• PCOR= CURE HEAT TRANSFER FERIPETERIFT)

• PO=1NITIAL REACTOR VESSEL PRESSUREIPSIA)

• RCICPX=hCMIN AL RCIC' FULL ~ FLCn(L8/SEC)

• TAULEN=$TABILITY TIME CCNSTahT FOR REGICN AVER AGE HE AT FLUX CALCISECl

• TCFUEL= TIME CCNSTANT TO ACCOUNT FOR RESIDUAL HEAT IN CORE FOR

• INITIALIZAT.CN CLCSELY FCLLOWING SCRAM

• 70= TIME AFTER TRIP THAT TRANSIENT INIT I ATES(SECl

• VOLP=LOhER PLENLM VOLUPE(FT**33

• VIDC= VOLUME OF ConNCOMER SETkEEN BOAF AhD PLMP OlFFUSER EXIT (FT**31

• WINJO=Ih!TIAL IkJECTION FLOW (LB/LEC)

• REFERENCE P ARAMETERS FOR NATLRAL CIRC LCSS CCEFF CALCULATICN

• CpR=R ATEC CORE PCWERIMWTHI

• HREF= REFERENCE CORE INLET ENTHALPY(BTU /L8)

• LOCR= REFERENCE CCWNCCMER LEVEL (I.E. NORMAL LEVEL AT STM SEP MICDLEI

• PRELR=REFEREhCE RELATIVE CORE PCWER

• PRR= REFERENCE RE ACTOR VESSEL PRESSLRE

• RH00=JENS ITY OF STE AMIL8/FT**3) IN R.V. AT INI TI AL PRESSURE =PC

• WGUESS= GUESS ON FLCWIL8/SECl FOR ITERATIVE SOLUTION FOR 11ITIAL FLOW

• WREF= REFERENCE F LCul L8/SEC)

• XREF=REF ERENCE FRACT ION AL QUALITY

• PARAMETERS FCUND IN FUNCTION SudPRCGRAMS -*

• YOIFF= JET PUMP v0L BELOW 80AF(FT**38

• VFREE= TOTAL STM. VOL. IN R.V.lLESS LO*ER PLENUM, CORE, CORE CUTLET PLENUM

.

• AND STM. SE P AR AT CRS ) AND MAIh STE APLINES TO ISOL AT ION V ALVESIFT**31

• VJET=VOL BE ThEEN JP SUCTION AND BOAF (F T**3) *

• V AN4= VOL BETWEEN TOP OF CORE OUTLET PLENUM AND JP SUCTICNIF T**3)

• - VSSOP=VCL BETWEEN TCP CF CO PLENUP AND 80TTCM OF STM SEPICU.FT.)

l

• WR A T ED= S TM. FL0w THUR 1 SRV IL8/SEC) kHEN PRESSURE = PRATED (PSIAI

• XKVGJ= EMPIR ICLE CONSTANT USED IN CALCULATION OF ORIFT VELOCITY l

l r

k

183

• XL1=0=ZERO PCINT FOR CC HEIGETIFT): C3INCODES WITH BOAF

• XL2= HEIGHT AT JET PUMP SUCTICNIFT)

• XL3= HEIGHT AT TOP OF CORE OUTLET PLENUMIFT)

• XL4= HEIGHT AT BOTTCM OF STEAM SEPARATORS (FT)

CCNSTANT ACCR=82., ACOP=234., ART =42.?

CONSTANT CPO=3293., EQX0=.5, FFLASH=.001 CONSTANT LHECER=23.4 LOP =5.58, LRT=1C.0 CONSTANT L80T=24.85 CONSTANT PCOR=5518., PRATED =1095.

, CONSTANT TAULEN=.75 TCFUEL=9.5 CCNSTANT VFREE=13000., VCLP=3350., V10C=192.,mRATE0=259.

• CARD TO INITI ALIZE AT ARBITRARY TIME POINT

• CCN$ TANT HC IO= 312.0 5, L OCC= 2 8.6 3, P O= 12 5. . T0 = 14 4 30. . RH OG 0 = . 2 7 8 8

• INITIAL CORE FLOh GUESS CEPENDS ON LCCO AND PRELO CONSTANT WGUESS=8310.

• REFERENCE PAR AMETERS FOR C ALCUL ATION OF NATURAL CIRCULATION

• FLOW RESIST ANCE COEFFICIENT CONSTANT CPREF=3293., PRELR=.32 PRR=1020.....

LOCR=27.58, XREF=.133, hREF=9111. , HREF=522.

• RUN CCNTRCL PARAMETERS CCNSTANT HINJIN=58. .PC 'C MX=82.98 76

• FUNCT lON SU8PROGR AMS

• C AV010( X I N , X CUT , W TOT AL, TS AT , RHOF. RP'JG)

• DKFUNI T )=P/PC A S A F UNCTION OF TIME AFTER SHUTOOWN

• QREGAVtBCTTCP,TOPeKAPPA,PEKAVG)=(AVG HT. FLUX FROM A TO 81/(CORE AVG)

• RHOTP(RHCF,RHOG,VC10)=2-PHASE DENSITY

• XL EN OC ( VCLOC ) =00 hNC CME R LEVEL AS FUNCTICN OF LICUID VOLUME

• VGIO(QU AL IT Y, TOT AL FLOW, FLOW ARE A. TSAT,RHOF.RHOG)= POINT VOID FRACTICN

• VOLOC(LCC)=CCWNCCNER HEIGHT ABOVE 80T CF ACT. FUEL

• STEAM TABLE CSMP INTERPOLATION FUNCTIONS

• VSATF(PRESS)= SAT FLb!O SPECIFIC VOL(FT**3/LB) 9 FUNCTICN VSATF=15.. 0167, 50.e.0173, 100. 01TT, 200. 0184,...

400.. 0193, 600.. 0201, 800.. 0209, 1000.. 0216, 1203.. 0223....

,1400.. 0231

• VSATG(PRE SS) =SA T GAS SPECIFIC VCL(F T**3/LB)

FUNCT ION V S A T G= 15. , 2 6. 3, 50.,8.51, 75. 5.81,100.,4.43 ...

150.,3.01.200.,2.29,400.,1.16, 600.. 77,800.. 569....

1000.. 446,1200.. 362.1400.. 302

• HSATF(PRES $URE)=S AT FLUID ENTHALPYl8TU/LB)

FUNCTION FSATF=15.,181.. 50. 250., 100. 298., 2C0. 355.....

400.,424., 600. 472., 800. 510., 1000.,543., 1200.,5T2.....

1400.,599.

• HS ATG(PRESSURE)= SAT GAS ENTHALPY(BTU /LB )

FUNCT ICN HS ATG=15. ,1151. , 50.,1174. , 100.,1187., 200. 1198.....

400. 1205., 600.,1204., 800. 1199., 1000.,1193., 1203. 1185.....

1400.,1175.

• TSATM(PRESSURE)= SAT MIXTURE TE MPE R ATURE ( CE G-F )

FUNCTION TSATM=15.,213., 50. 281., 100. 328., 200. 382.....

400. 445., 600.,486., 800. 518., 1000. 545., 1200.,567.....

1400.,587.

• VSCIENTHALPYl=SU8C00 LED FLUIC SPECIFIC VOL(FT**3/LB)

FUNCTICN VSC=21.4,.016, 121.. 0163, 221.. 0169, 272.. 0174,...

376.. 0185, 431.. 0193, 488.. 0203, 549.. 0217, 562.. 0221 ...

575. 0224

• CALCULATICN CF INITIAL s?fURATICN PRCPERTIES RHOF0=1. / ( AFGEN( VSA TF s P01 )

HFO=AFGEN(HSATF,PC)

HG0=AFGEN(HSATG,P0)

• T S AT0 =FUNGE N ( TS AT M,2, PC s

• SU8 COOLED WA TER OENSITY RHOSC0=1./(AFGENIVSC.HCIO3)

'184

• CALCULATION OF FLOW RESISTANCE CPa?8itlENT

• NEEDS REFERENCE CONDITIONS FOR 'ilunAL CIRCULATION CPREF,PRELR,PRR,XREF,

• WREF, HREF, LDCR

• RE TURNS: - LOSS COEFFICIENT: KLOSSU

• C ALCUL ATION OF REFERENCE DENSITIES AND SATURAIION PRCPERTiES HFR=AFGEN(HSATF,PRR)

TSATR =FUNGEN I TS AT P,2, PRR)

RHOFR=1./( AFGENIVSATF PRRll RbOGR=1./(FUNGEN(VSATG,2,PRRll e RHOSCR=1./tAFGENtVSC,HREFil QTOTR=CPREF*PRELR*94T.8

• THIS EXP FOR LSC ASSUMES AVG POWER = CORE AVG IN SC REGION LSCR=12.*bREF*(HFR-HREFl/GTOTR

• L8R=12.-LSCR RS C AV R= R HO FR / 2.+R F05CR / 2.

FUNCTICN SUBPROGR AM CAVOIO RETURNS AVERAGE CORE BOILING REGICN

• VOID FRACTION FRCM THE GIVEN INPUTS C A VR= C A VO ID i o . , XR EF . WRE F , T S A IR . RH uF R , RHOG R )

Rh0RR=RHOTP(Rb0FR.RHOGR CAVR)

R OCCR R=RSC AVR

• LS CR/12.+ RNCB R* L 8R/12.

R HOPR =R HO TP I RH OF R .RH OGR,0 P VR )

• FUNCTION VOIC RElbRNS POINT VOID FRACTICN FROM GIVEN INPUTS OPVR=VOIDtXREF WREF,ACOP,TSATR,RHCFR.RHOGRI R CR T R = RHO TP I R HOF R ,RHOGR , R TVR )

RTV R= VOI 0 t XR E F, WR E F , AR T , T SA TR , RHOF R , RHOGR )

RDP=LCCR*RH05CR/144.-RORTR*LRT/144. ...

RHCPR*LCP/144.-RCCORR*12./144.

KLOSSU= ROP / t WREF

• hREF )

• INITIAL RELATIVE POWER PRELO = ( OK FUN I T0 l + .94

• EX P t -T0/ T CFU EL i l QTOT0=PRELo*CPO*947.8 e

• CALCULATICN CF INITIAL TOTAL FLOW BY ITERATIVE SOLUT ION, W ITH

• STARTING FLOW GUESS INPUT BY USER

• WCIO=1PPLIWGUESS. 001,WCAL)

X00=tGTOT0/WCIO+HCIO-HF03/tHG0-HFoi X0=LIMITic.001,1.000,XOUI

• THIS EXP FOR LSC ASSUMES AVG POWER = CORE AVG IN SC REGION LSC0= LI M I T io . ,12. ,12. *W C IO * (HFO-HCI O S /GTOTo l L 80= 12.-L SC O

• C ALCUL AT ION OF INITI AL DENSITIES RSCAVC=RHCF0/2.+RHCSCO/2.

W800= WCI O* KO

• VOID AT BOTTOM AND TOP CF BOILING REGION CALC. 8Y OFVOIO

• FUNCTICN SUBPROGR AM, THEN AVERAGED TO GIVE CORE AVERAGE VOID CAVIN0=0FVOIDIACOR,WCIO,0..TSATO,RHOF0,RHOG03 CAVE XO=DF void ( ACOR ,WCI O-Wuc0, W800.T S A TO ,RHOF0,RHOGO )

C AVO = t CAV IN0 + CAVE X01/ 2.

RH080=RHOTPIRHCF0,RHOGO,Cav01 R OCOR 0=R SC A VO

• L SC 0 /12. + RH C80

• L 80/12.

R0P0=RHOTPtRb0F0,RHOGO,0PV01 OPVO= CFVO I0 t A COP, WCID -W800, W 800,T SA TO , RHCF 0, RHOGO )

RCRT0=RHOTP (RHOF0,RHOGO RTV01 RTVO=0FVOI0lART WCIO-WB00,W800,TSATO,RHOF0,RHOG01

• NET GRAVITY PRESS DRCPIPSil AVAILABLE FOR UNRECOV. OROP ACROSS CORE LDC X= LI M I T (0. , LOCR , LDCo l UNRECC=LDCX*RHOSC0/144. + (12.+LRT+ LOP-LOCXI*RHOG0/144. ...

ROR T0 *L RT/144.-R O PO

• LOP /144.-ROCOR C

• 12. /144.

OP0= LIMIT I.0001,1000. UNRECo l

• WCAL= SCR TIOPC*RHOB0/ t KLOSSU* RHC8Rl l

• ENO ITERATIVE LOOP e

• 00WNCOMER AND LOWER PLENUM INITIALIZATICN

• VOLUPE CF W ATER IN DCWNCOMER NODE GivEN SY FUNCTION

185

• VOLDCl ) AS A FUNCTICN OF W ATER LEVEL M DC0= VOLDCI LOC O ) *RH OSC O MHOC0=M0Co*hCIO MLPO=VOLP*RHCSCO M HLP C= ML P C

• HC I O G=4.T5* ART *RNCGO M TOT 0 = t L SCO* R SC AV0+ L B0*R HCB0 )

• A CCR + LC P

• ACCP

• ROPO + LRT

• AR T

• ROR T 0 + G

,

• PRESSURE CALCULATION INITI AL IZATION RHOSVR=1./FUNGENtVSATG 2, PRATED)

UC SR V= WR A TED / SQR T t PR A TED* RHOSVR )

MST0=tVFREE-VCLDCILDC03)*RHOGO 8 UMST0=MST0*tHGO-PO*144./t778.*RHOGO))

• - ----I NI TI ALI Z ATI ON FOR SUPPRESS ICN PCOL C ALCUL AT I CN----

• CONSTANTS FCR CONTAINMENT MODEL

• ADMET=HE AT TRANS ARE A BET OW MET AND CW ATMOStFT**2)

• APMET= HEAT TRANSF AREA BET SP MET AND SP ATMOS

• ASSPh= ARE A 0F POOL W ATER SURF ACE t SQ.FT. )

• BOWSPO*CU.FT/SEC/ PSI FLOW WHEN CCWhCCMERS CLEARED

• BSPDh0=CU.F T/SEC/ PSI CF FLOW kHEN VALVE OPEN

• COMET = MASS

• SPEC HEAT OF DW METALt BTU /DEG-F)

• CPAIR=NASS* SPEC NEAT OF AIR IN SP CHAMBER ROOM

• CPME T= MASS

• SPEC HE AT OF SP METAL IN CONT WITH GAS

• DM= AMOUNT OF STM. QUENCHEDIUNIFORMLY ARCUND POOL) IF THERE

• IS ELAPSED TIME BETWEEN NOMINAL ST ART AND INIT I ALILATION OF THf RUN

• DPQR=RAhGEtPSI) CVER WHICH CUENCH FRACTION GOES FROM 1 TO 0.

• WITH INCREASING VAPCR PRESSURE

-*

• DPQZ=M ARG IN t PS I) ABOYE SATURATION FOR COMPLETE QUENCHING

• FDI FSI=F9 ACTION CF PCOL C0 CLING DISCH ARGE AND SUCTIOy TO POOL NODE I

• FWE= FLOW BE TWEE N 5.P. N00ESILB/SEC) PER IhCH OF W ATER LEVEL DIFFERENCE

• FWI= FRACTION OF POOL WATER CONTAINED IN PCOL NODE I

• FSVI= FRACTION OF TOTAL SRV FLOW DISCHARGED TO POOL NODE I

• F TE!= FRACTION CF TURBINE EXHAUST DISCHARGED TO PCOL N00E I

• FLSPG. FLOWG= FRACTION OF ATM. LEAKED TO RX-BLOG. PER SECOND

• GCH= GAS CCNSTANT OF H2

• GCM= GAS CONSTANT OF P(MISC.)

• GCN= GA S CON S TANT OF N2

• GCS= GAS CONST ANT OF H2O VAP t PSI *F T**3/LB*0EG-R I

• HSRREF=CELTA TO REREFERENCE STM ENTHALPY FROM ASME STEAM TABLES

• TO PERFECT GAS EXP. THAT HAS H=0.0 AT 0 DEG-R

• HSTE=ENTH OF TURBINE EXHAUSTtBTU/LB)

• HJMDWO= INITI AL CW GAS HUMIDITY (P ER CENT I

• HUMSPO=IhlTIAL SP GAS HUM 10!TY (PERCENTI

• KMIX=EMPIRICLE CONSTANT FOR NAT. CIRC MIXING FLOW BET. POOL NODES

• LBASE=NCMINAL ST ARTING LEVEL OF POOLIIN. FROM INST. 0.3

• MHSPGO.MHDWG0=lNITI AL MASSES OF HILBS)

• MMSPGO,MMDWG0=INITI AL MASSES OF M (LBS)

• PDCVP= PRESS. DIFF NECESSARY TO CLE AR THE VENT

• PIPES FCR FLOW FRCM CW TO SP

• PTOWG0= INITI AL TOTAL PRESS CF OW GAS (PSIAI

• PT S PG0= IN IT I AL TOTAL PRESC OF SP GAS (PSIAl

• QRVHLO=REACTCRt+ PIPING) NEAT LOSSES (MW) FOR TEMP DIFF=DTRVHLtDEG-Fi

• TAUF0W= AVERAGE RESIDENCE TIME OF FCG'IN DWISEC)

• TAUFSP= AVERAGE RES. TIME OF FOG IN SP

• TAUVRV= TIME CONSTANT FOR SP VAC RELIEF VALVE TO LIFTISEC)

• TBASE=NCMINAL STARTIhG TEMP CF POOLICEG-Fi

• TBI!= PROVISION FOR BI ASING STARTING TEMP CF POOL NODE I

• T0=0!SCH. TEMPtFI 0F POOL COOLING FLOW

• TDMET0=Ih!TIAL DRYWELL MET AL TEMP (F)

• TGOWO=INITI AL Dh GAS TEMPT 0EG-F)

• TGSPO= INITI AL SP GAS TEMP (DEG-F)

• TPAIR0=th!TIAL SP CHAMBER ROOM TEMP

186

• TPME T0=INI TI AL TEMP SP METAL IN CONTACT WITH SP GAS

• VGDW= TOT AL FREE VOLUME OF DW IFT**3)

• VTSP=TCTAL FREE VOLUME OF SP (FT**3)

• WOLEAK= LEAK RATEIL8/SECl 0F SAT. WATER FROM RV TO DdVWELL

• WOSP k= DISCHARGE FLOW OF ' POOL C00LINGE L8/SEC)

• WSSPW=SUCTICN FLCW OF POOL COOL INGIL8/SEC)

• GEOMETRY AND PHYSICAL CHARACTERISTICS CCNSTANT ACMET=1.65E04,APMET=1.7E04,ASSPW=10860. g CCNST AN T BDhSPO=2500..BSPDb0=2000.

CCNSTANT CDPET=8.33E04,CPAIR=3200. CPMET=>.82E04 CCNS TAN T DPQR=2. DPQZ=2.

CONSTANT GC F= 5.3 61, GCM = . 24 3 6, G CN =. 3 8 29, GC S =. 59 5 5 '

CCNSTANT HSTE=915.,HSRREF=-854.5,PCCVP=1.75 CONSTANT TAuvnV=3.,TAUFDh=30..TAUFSP=15.

CONST ANT VGL W= 159000. .V T SP =2 67600.

• POOL NCDALIZATICh ..

CONST ANT FD 1= 1. .F Sl= 1. .F W1=1. ,F SV1=1. .F TE1 =1. .F hE=5000. M M IK =950.

• INITIALIZATICN CONSTANT DM=0. HUMDW0=18.4,HUMSP0=79.4.L8 ASE=1u.23 CONSTANT M F S P G0 = 0. . M HO WG 0 = 0. . M M S PG 0 = 0. , M M0 'aG 0 = 0.

CCNSTANT PTChG0=23.15,PTSPG0=23.65,T8ASE=169.1.T8I1=0.

CCNSTANT TG0 h 0= 26 8. , T GSPo =161. , T DMET0 =212., T PM ET 0 = 14 5.1, TP A I R O= 15 5.

• RUN CCNTROL CCNSTANT F LD hG=0. , FL SPG=0. . TD= 90. .CRVFLO =1. DTRV HL= 404.

CONSTANT WOLEAK=.68,WSSPW=C.,WDSPW=0.

HC=TC-32.

• INITIALIZATION CALC. FOR DW AND SP MASS AND ENEAGY 8AL ANCES

• ASSUME ZERO IN IT I AL HYDROGF,N AND MISC. GAS

• ---- FUNCTION TA8LES >

• FU NCTI ON SPLEV SP LEVEllINCHES FROM INSTRUNENT ZERC) AS A FUNCTION A

• FUNCTICN CF VOLUMEtCU.FTl

• FUNCTICN SPLEV=0.,-182. 20222..-134.,74646. -62.....

117344..-14. 130129.,0. 139300. 10.....

18 2 563. ,58.,242176. 130. 267611. 19C.

e

• FUNCTION STFOSVs SA T UR A T ION TEMP AS A FUNCTION OF SPECIFIC

• VOLUME r F UNC TI ON S TF O S V= 2. 8 3. 3 6 3.5.4.65 .3 24.1,7. 6 5,2 8 8.2.12.2. 2 5 7.6. . . .

20.1.228. 31.2.204.,44.7.185.6.76.4.160 5.158.9.129.6....

333.6.101.7,641.5,79.6.1235.,59.3

• FUNCTION SPFOT SATURATION PRES $URE AS A FUNCTION OF TEMPERATURE FUNCT ICN SPFOT= 59.3. 25,79. 6. 5.1C1 7,1.C .132. 9,2 4,. . .

160.5,4. 8.18 5.6,8. 5. 20 3.9 e 12.5. . . .

228. 20. 250.34,30.,288.2,56. 324.1,95. 363.5.160.....

381.8.200. 417.3.300. 444.6,400. 467. 500.,486.,600.....

503.,700.,525.,850. 544.6,1000. 556.,1100.,567.,1200.

• ---- PRELIMINARY CALCULATICNS SPOWO=NLFGEN( SPFOT, TGDWO)

$ PSP 0=hLFGEN(SPFOT TGSPD)

PSDWG0=(HUMDh0/100.)*SPOWO P S SP G0=I HUM SP 0 /100. )

• S P SPO =

PNSPG0=PTSPGO-PSSPGO PNO WG C= P TO WGO-P SD hGO MNS PGOsPN SPGO* t VTSP-VWSP01/ t GCN e t TGS PO +460. ) )

MNDWGO=Pt40WGO*( VGDWl / ( GCN* ( TG0 ho+ 460. ) )

187

• VWSPO=INITI AL TOTAL WATER VOL IN SP ,

MS SPG C = P S SP G0 * (V T S P-VWS P01/ ( GCS * ( T CS PO + 460. l l M SDWGC=P SDWGC* VGD W/(GC S* ( TGDWO+460. 8 )

e UMSPG 0= ( TGSPO +460. ) * ( MNS PGO * ( .2475 .1851 *GCH l +. . .

M S SPGO * ( . 4 5-4. 89/( TG SP O +460.1 .1851 *GC SI I 8 UMOWG0= ( T GOWO +460. l * (MN CW GO* ( .247 5 .1851*GCN I +. . .

M S 0WGO * ( . 4 5-4. 89/ ( T ODW 0 +460.1 .1851

• GCS I )

,

• DEPTH LBASE SPECIFIEC: INCHES FROM INST ZER0(4- Ita BELOW CENTERI '

• TBASE IS NOMINAL STARTING TEPP CF PCCL

• TBI(!) PROVIDES FOR BI ASING INITI AL TEMPS IN PULil-NCDE MODEL '

v!!=2.462*AII ,

A I I= ( L B A SE-4. l *S QRT ( 3 4 5 8 0 .-LB AS E *LB A S E + 8.*L B A S E l+ . . .

34596.*AR$1Nt (LBASE-4.1/186. I + 543*3.

NWII != FW 1*VII * ( 63. 9 .019* ( TB ASE +T8ill i MI!=MW11!

DTBASE=(1190.-(TBASE-32.ll*CM/(M11+0M) s T IC 1= TB A SE+D TB ASE+ Te l l

_VhSPO=(MWIII+FW1*CMl/(63.9 .019*TICII VGSPC=VTSP-VWSPO DYhAMIC

--0YNAMIC PORTION OF RE ACTCR VESSEL CALCULAT ION-

• - --RUN CONTRCLS FOR INJECitDd CONTROL .i LDCVZ=LDC*12.+216.

L00EL=REALPL(5CO.,2 00,LOCVZ)

RCICC=0.0 N,

', HPCID=0.0 W INJs RC ICMX * ( RCI CD+ 8. 33* HPCI D I

• ST EP ( 30. )

• ---------RUN CON TROLS FOR VE SSEL PRE SSURE CON TROL PCPEh=1120.

. PSHUT=1080. 1

• CALCULATION CF DYNAMIC SATURATION PRCPERTIES RHOF=1./( AFGCN(VSATF Pil RbOG=1./(FUNGEN(VSATG,2,Pil HF=AFGEN(HSATF,P) ,

HG= AFGEH l HSA TG ,P )

T S AT= FUN G EN ( T S ATM ,2, P )

• INTERMECIATE CALCULATIONS: AVERAGE HEAT FLUXES t

• 'D K F UN ' RETURNS DECAY HEAT AS A FbhCTICN OF TIME SINCE SCRAM

• ' QRE G AV ' RETURNS AVERAGE NORMALIZED POWER, GIVEN LOCATIOS OF 30P

• AND BOTTCM OF REGICN T=T0+ TIME TMIN=T/60.

\

PREL = (D KF UN ( T l +.94* EXP (-T/ T CFU EL ) I QTOT= PREL *CPO *947.8 QFCCR=GTOT/(12.*PC04)

AVM SC=QREG Av t 0. , LSCD )

AVMB= QREG Av t L SCDe L SC0 +LBo l QFSC=AVMSC*GFCOR QFB=AVMB*CFCOR y* '_

LS CC = R E A L PL (LS CO , T AUL EN ,L SC )

LSD=REALPL(LBO,TAULEN,L81

• 00WNCOMER ANNULUS CALCULATION

• NEEDS: C ENS IT I ES: RHCDC RHOLP

.

• FLCWS INTO DChNCOMER: WINJ, WRECIR, W INJ AS ,

• INJECTION ENTHALPY: HINJIN

• FLOWS OUT OF COWNCOMER: WCI, WFLDC, hFLLP -

• REACTCR VESSEL PRESSURE: P N

,o

\\~

188

• RETURNS: DCWNCCMER HEIGHTIABCVE BOTTCM CF ACTIVE FUEL) : LOC

• ENTFALPY INTO LOWER PLENUM: HOC

• INTERMEDI ATE CALCULATIONS MASS RATE CF ASPIRATION FROM STEAM SPACE

• DUE TO INJECTION FLOWi FLASHING RATE FRCM OC MAS $3 RECIRC FLOW R AI E EQXX= EQXO

• t LNEDER-LCC)/LHEDER EQX=LIMITto. 1.,ECXX)

HINJ F l =H I NJ IN + 1 HF-H I N JI N)

• E QX GP= 7 23*W INJ WINTCT=1NTGRLto.,WINJ) 9 G T= t W IN TO T

• 7. 4 81) /62.11 W I NJ AS= ( HF-H INJ IN ) *EQ X *W INJ/( HG-HF )

WFLCCX=MDC*FFLASH8(HDC-HF) g kFLDC= LIMIT (0.,9999.,WFLDCX)

• M ASS AND ENERGY BALANCESI61 15 A LOGIC SIGNAL TO STOP THE

• INTEGRATION kHEN LCC.LT.018ELCW 8CTTCM CF ACTIVE FUEL)

CMDC o !N TGR L ( 0. ,DMDC)

MCC=MCCJ+CMDC DMDC= 81 * ( bRE C IR +W INJ+W INJ AS-RHCCC * (WC I+W F LLP ) /RHOLP-WFL DC )+ WE XPC CMH DC = IN TGR L I O. , CMHOC )

MHOC=MHDC0 +CMHOC DMHCC=81*tHF *WRECIR+WINJ=HINJIN+HG*WINJAS ...

H DC* t WC 1+ hF LLPl

• RHCDC/ RHOLP-W FL CC

• P G)

H DC=M HDC /MDC RHOCC=1./ AFGEN (VSC.HCC)

• DC LEVEL IS CALCULATED BY FUNCTION SUBFROGRAM XLENOCIVOLUME)

VCC=MCC/RF0DC LOC =XLENDCtVDC)

. LDL= LIMIT (0..LOCR.LDC)

• LOWER PLENUM C ALCULATION

• NEEDS: SAME AS CCWNCCMER

,

• RETURNS: ENTHALPY AT CORE thLEftHLP) .

• INTERMEDIATE CALCULATIONS LOWER PLENUM FLASHING RATE, LP WATER VOL

• (VOLPW--USED FOR LOGIC CONTROL ONLY $1NCE THE LOWER PLENUM IS

• MODELED AS A CONSTANT VOLUME UNLESS CCWNCCMER 15 EMPTY) :

HLP=MHLP/MLP .

VOLPW=MLP/RPOLP RHCLP=1./AFGENIVSC,HLP)

X bF L L P= ( H LP-H F )

• MLP

• F FL A SH WFLL P=L IM IT (0.,9 99 9. , XWFLLP )

• LOGIC CChTROL n

X181=LCC .001 X2Bl=40LPW-1.01*VCLP 31N=NCR(X101,2281) al= NOT IBIN 4

• M ASS AND ENERGY BALANCES e

.DPLP=thCI+WFLLPl*(RHOCC/RFOLPlotB11-hCI-WFLLP+...

BIN *IWINJ+WINJAS)

CPLP=INTGRLlo.,0MLP)

. NLF=PLPO+CMLP U ' DMPLP =HDC *I WC I + WF LL Pl

• RHODC* B 1/RHCL P+ . . .

DIN #(WINJ*HINJIN+HGeWINJAS)-WCl*HLP-WFLLP*HG CPHLP=INTGRLID.,0PHLP)

PHLP=MHLPO+CMHLP

• CORE INL ET FLOW CALC.(FORCE SALANCE) t NEEDS CENSITIES RPOCC RHOLP,RSCAV,RNCB, ROP, RORT ,

• - REACTOR VESSEL PRESSURE

• P

• FLOW OUT OF BOILING REGION: h80

'*~ REGICN LENGTHS 2 LOC,LSC,LB

• RETURNS 8 FLOW INTO CCRE: WCI .

CMTOT=INTGRL(0. ,1WCI-WRECIR-WTOST+WFLLP))

MTOT=MTOT0+CPTOT WBO= 10FB *PCCR

• Lb D ) /l HG-HF )

189

^

- WRECI A=L IMITIC. aC000.,t WCI-WTOSTl *F RECIRl+ WCCLL WCCLL* LIMIT (0.,10000. tLINT-30.l*3000.1 FRECI R=LIMI TI C. ,10. ,t LINT-L8CT ]/ t LDCR-LBCT i l L L INT = L 5C D +L 8 + LOP T P +L R T TP WFLCP=0.0 WFL8R=0.0 WFLRT=0.0 MCOREM=12.*ACCR*RHOG MCPEM= LOP *ACCP*RHCG

' MRTEMstLRT+4.151*ARTokH0G WORAV=WC1G ASCAV=tRHCF+RNCLPl/2.

y WCIG=REALPLthC10,3.0,WCl3

• CALCULATE DEM51 TIES CAWIN=CFVCICIACOR,WCIG,0.0 75 A T,R HO F , R HOG )

C AVE x=0F VCID I ACCR,WCIG-W8C.W 80 .TS AT ,RHOF, RHOG )

- CAV= t C AVIN*C AVE s t /2.

AkOB=RHOTPIRbCF RNCG CAVI OPV=CFVOIDIACCF WCIG-W80,W80 . TS AT, R HO F,R FOG )

RHCOP=RHOTPERHCF,RHOG,0PV)

RTV= CFVOIDI ART,WCIG- W80 e b80 .TSAT RHOF.RHOGI RMCRT=RHOTPIRHOF,RHOG,RTV)

SUVIN=DFVCIDIACOR,WCIG WFLLP.TSAT,RHCF.RHCGI BUVEx=DFVOI0 t ACOR,WCIG-W60,WTOST.TSAT,RHCF RHOG)

BUAVG=l8UVIN+8UVExl/2.

RC8UB=RHOTPIRHCF,RHCG,80 AVG)

DELH1 =L IM If t 1.,400.. HF-HL P )

O L5C J = t-2. l

• t LSCX

• GF S C

• PCCR-W Cl *L IMIT io .,400., HF-HLP ) l/...

(RHOLP*ACCR*DELH11 LSCx= IN T GRL IL SCO, CL 5C XI LSC=LIMITto. 12. LSCAI MSC=RSCAV*LSCCoACCR 9 '

• FINO THE TWO PHASE LENGTNS AND REGION MASSES L8=LIMITto. 12.-LSCD,.8U)

• L 8U= t MTOT-MCOR EM +MSCLtUM-MCP EM-MR T EM-M SC l / t ACOR e t RHOS-RHCG l i LCOV= L I M I Tt o . 12. .L8COVU+ L 5CD I 4

L 8CCVU=L8U

• t R bC8-RHOG I / t R08U8-R HOG I M5CDUP=ACCR*Re.0G*LSCD M8=L8*RH08*ACOR MCOR=MSC+M8+t12.-L8-LSCDi*RHOG*ACOR LOPTP= L IMI T lo . , LOP, t MT OT-MCCR-MRT EM-MCP EM I / t ACDP e t R H00P-R H0G i l l MCP= t RH00P* LCP TP

• t LOP-LOP TP l

• RHCGl* ACCP MRT=MTOT-MOP-MCOR LRTTP=tMAT-MRTEMI/lARretRFORT-RHOGil MPCCatL8*ti.-CAVi*ACCR+LCPTP*tl.-CPVl*ACCP+...

LR T TP* t 1.-R TVI

• ARTl *RHOF

• TABULATE PRESS DROPS AND CALCULATE WCI l

• P C080 Ta l MCOR/ ACOR + MOP /ACOP +LR TLI M *RHCR T+ t 10.-LA T LI Mi

• RHCG I /144.

LRTLIP= LIMIT (0.,10.,LIN7-17.523 l'

L D551 N= L DL* R H00C/144. + t LDCR-LD L l

• RHCG/144.-PC080T X DUM= L IM I T i l . E-9,5. e LO 5 5 !N I WC1 =5 CRi t X DUM

• RH08/ t KLO55U*R HO BR i l WCID=REALPLthC10,10.,WCII l

j

• RE ACTOR VESSEL PRESSURE CALCUL AT!DN

!

• STEAM MASS BALANCE 9

, CMST=INTGALLO. DMST) l MST=MST0+CMST l DMST=WTOST*WEVAP+hFLDC-WEXPC-WCCNO-WINJAS-WSTC l , WTOST=L IM if t 0.,10CO. ,W80 +hf LLP-hPOO) 1 WCONC=0.0 WEVAP=0.0

WEXPC = LIM I T( 0. 100. 1 67E-04* M ST* t HG S T-HS Ti l i HGST=AFGENtHSATG.PST) l

190 WP00=PP00*DHF/(HG-HF)

HF D= R E AL P L ( HF 0,2. 0 eHF )

DNF=(PF-NFDl/2.

V$t=VFREE-VOC DVST= C ER IV t 0. , VS T )

VSVST=VST/MST

• VST= TOTAL VAPOUR SPACE VOLUME

• VSVST= SPECIFIC VOLUME OF STEAM ,

• TOTAL ENERGY B ALANCE CUMT0=INTGRL(0. 00MTO) =

UMT0=UMST0+CUPTO 00MT0=WTOST*HG+(WFLDC+WEVAPl*HG ...

W EX P C *HF-WCON D* hG- W IN J A S *H S T- W S TC *H S T-D V S T

• P* 144. / 77 8.

HST=UPTD/PST+PSTG*VSVST*144./778.

• STE AM PRES $URd C ALCULATED SY FUNCTICN SUBPROGR AM PFVH--STE AM PMESSURE

• AS A FUNCTION OF SPECIFIC VOLUME AND ENTHALPY

• PFVH REQUIRES AN INITI AL GUESS OF PRESSLRE, PSTG PST=?FVHtHST,%$VST,PSTG)

PSTG=REALPL(PO,2.0,PST)

P=PST

• SAFETY RELIEF VALVE MODEL--TWO VALVES MCOELED--WSSV= TOTAL SRV FLOW LOGREL=LUGV1+LOGV2 LOGV1=REALPL(0.. 62 LOGPV1)

L OGV 2= RE ALPL ( 0. . 62.LOGP V2 )

LCGoV!=RST(XIV1,X2V1.0.3 L CG P V2 = R S T ( X 1V2, X 2V2,0.1 X IV1=COMPAR ( P SHUT e P)

X IV2=COM P AR (P CPE N-25. , P )

X2Vl=CCMPAR(P,POPEN)

X2V2=COMPARIP,POPEN+25.) i W S SV = U CSRV

• SQ R T ( P S TG/ V SY S il

• LOGRE L

• WSTE=$ TEAM TURS EXHAUST FLOW (RCIC)

• ENTHALPY IS ASSUMED =915. BUU/LB WST E= ( RC I CD* ( .00 613*PS T + 1.11 +HP C 10 *( .0 38 7 +P ST + T. 8111

• STE P ( 30. 5

• WSTC 15 TCT AL STE AM FLCW RATE FROM REACTOR VESSEL TO CCNTAINMENT WSTC=hSTE+WSSV

• ----------CYNAMIC PCRTICN CF CCNTAINPENT CALCULATION-------------------

• CALLS TO SP WATER MACRO--ONLY ONE NECESSARY FUR THE SINGLE N00E PCCL MODEL XhC1 ,TP1 ,WE1 eMW1 ,LW1 =W(TP1,...

TP1 eLW1.LW1 ,TICL ,FW1 ,FSV1 eFTE1 eF01 ,FS1 i VWSP=MW1/(63.9 .019*TP11 MWSP=PW1 G WSP= 7.4 81* V h S P I

TWSPAV=(63 9-PWSP/VWSPl/.019 LWSPAV=AFGEA(SPLEV,VWSP)

• INTERFACE THE SP WATER TO THE SP GAS XhQDW=XNQ1 WSSVNG=WSSV8FSV1*XNQ1 WSTENC=WSTb*FTE1*XNQ1 WTESPW=WE1

• INTERFACE VARIABLES I RV TO SUPPRESSION POOL

• WSSV=STM. FLOW FkOM RV THRU RELIEF VALVES (L8/SCCI

• WSTE=STM FLOW FROM RV THRU TUR8INESIRCIC+HPCl3

• WHSV= HYDROGEN FLOW FROM RV TPRU RELIEF VALVES (L8/SEC) *

• HHM=HYDRCGEN ENTHALPY AT MIXED RV TEMP (8TU/L8I (REF TO 0. OEG Al

• HSTE=ENTHALPY OF IURBINE EkHAUSE(ASMEl

• TM= MIXTURE TEMP OF RV STEAM SPACE

191

• INTERFACE VARI A8LES : RV TO DRYWELI.

• WSDWR= STEAM FLCW, RELEASE TO ORYWELL(L8/SEC)

• HSQWR=ENTHALPY CF WSCWRIASME)

• WhudR= HYDROGEN F LOW, RELEASE TO DRYWELLIL8/SEC)

• HNDWR=ENTHALPY OF WMCWRtREF TO 0.0 DEG R)

• WMDhR= MISC. FLOW RELEASED TO DRYWELL

• HMOWR=EN THALPY OF kMDWR(0.0 DEG R3

= INTERF ACE V ARI ABLES

• SP W ATER TO SP GAS

• WSSVNg=TCTAL NCN-QUEhCHED RELIEF VALVE FLCW FROM RV TO SP VIA RV'S

• WSTENQeTOTAL NON-GUENCHED TURBINE EXHAUST FLOW

• WCSPG=TCTAL CONSECSATE FLOh, SPG TO SPW

,

• WTESPW=TCTAL EVAPOR ATICH RATE, SPW TC SPG

• HSTSPW=ENTHALPY OF STE AM(REF 0. AT 0. OEGI AT MASS-WEIGHTED SP WATER

• TEMP. TWSP

• HCSPG=ENTHALPY CF C0hCENSED SP STEAM (ASME)

• PASS 8ALANCES FCR SP GAS AND DW GAS

• INPUT FRGM OTHER CALCULATIONS

• VWSP=SP TIT AL W ATER VCL (FT**31 ( V AR I ABL E )

• WSSNQ= TCTAL NOA*GUEhCHED STEAP FLCW (LE/SECl

• WHSV= H FLOW FROM THE RV I THROUGH RELIEFSI

• hS0WR= FLOW P ATE CF STE AM RELE ASEC CIRECT TO DRYWELL

• kHOWR= FLCW RATE CF HYORCGEN RELEASEC TO CRYWELL

• WMOWR= FLOW RA TE 'E MI SC. RE LE ASED DIRECT TO DRYWELL

• DE F I NI T! CNS

• PSSPG= MASS STP IN SPG

• PHSPG= MASS H lh SPG

• MNSPGm NASS N IN SPG p

• PMSPG= MASS M IN SPG

• FLOWS = L8/SEC UNLESS OTHERWISE SPECIFIED

• WSSVNG= TOTAL NON-QUENCHED STM FROM SV'S ANT TUR8, EXHAUST

• kHSV= TCTAL HYDRCGEN FLOW FRCM RELIEF V ALVES

• DWSP= OE SIGN ATES FLOW FROM Dh TO SP

• SPOW= DESIGN ATES FLOW FROM SP TO DW

• SPL= SP GAS LEAKAGE

• 8= BULK FLOW RATE (FT**3/SEC)

• WCSPG= TOTAL CONDENSATE FLOW FROM SP GAS

• WCOWG= TOTAL C0hCENSATE FLOW FROM CW GAS

• SP GAS MASS BALANCE:

MSS PG= I NTG RL IM SSPGO. WS $VN Q +WSDWS P +WST ENQ+W TE SPW-WS SP CW-W SSP L-WCSPGl MH S PG = IN TGR L ( MH S PGO , WH S V + WHO WS P-WH S POW-h HS P L I MNSPGs I N T GR L I MN SP GO , WND W SP- WN S PL-WN S P0 w l PMSPG=INTGRL (MMSPGO WMDWSP-WMSPL-WM!PDW)

BSPL= VGSP *F LSPG WSSPL=8SPL*(PSSPG/VGSP) kHSPL=BSPL*(PHSPG/VGSP) kNSPL =8 SP L* ( MNSPG /VG SF y WMSPL=8SPL*lMMSPG/VGSP)

• BSPL=8ULK LEAKAGE FRCM SP GAS

• FLSPG=FR ACTION OF TOTAL SP GAS VOLUMES LEAKED PER SECOND

• VGSP= VOLUME OF SP GAS, VWSP=SP WATER VOLUPE VGSP=VTSP-VWSP

• NOTE: 0.5 PSI 15 NECESSARY TO OPEN VAC RELIEF VALVE BSPCWX= LIMIT (0.,1.E0o,BSPCWO*(PTSPG-PTDWG .5il BSPCh =RE ALP L 10. .T AUV RV,8S POWX )

. WSSPO W =o SPD h* ( MSSPG/VGSP )

W HS P CW= B SP OW 8 (MHS P G/VGSP )

WNSPCW=8SP0h*(PNSPGtVGSP)

WMSPOW =8 SPD h* (MMS PG/VGSP )

• BSPOW0* IFT**3/SECl/ PSI 0F PRESS DIFFERENCE hMEN

192 VAC RELIEF VALVE IS OPEN BDWSPX= LIMIT ( 0. ,1. E 06,8 D hSPO * ( P T DWG-P T SPG-P DCV P i l BCWSP=REALPL(0.,TAUVRV,8DNSPXI WSDWSP=8Dh5P*(MSDWG/VGOWa*ENCCW WHDWS P= 8 0W SP * ( PHOWG/V GDW )

WhDWSP=8DhSP*(MNDhG/VGDW)

WMD W SP =d D h SP * ( MMD bG /VGD W)

• BDWSP0= ( FT** 3/SEcl/ PSI WHEN DW PRESS I S GRE AT ENCUGH *

• TO CLEAR THE VEhT PIPE DOWNCOMERS

• PDCVP= PRESS DIFF. NECESSARY TO CLEAR TFE VENT PIPES W CS PG =W W CSP G + WV CS PG WVCS PG = ( M SSPG/ T AU F SPl * (1.-100./ HUMSP l

• COMP AR ( HUMS P ,100.1

• WWCSPG=RA TE OF COND. ON $PG HALL

• WVCSPG=R ATE OF COND. FROM SPG VOLUME

• DRYWELL MASS BALANCE:

MS DW G= IN TGRL (MSDWG0, W SD WR + h SSPO W-W SD W SP-W SD hL-hCDWG)

MHDW G a lN T GR L I PHCW GD , W PDWR + W HS P CW-W FDW L- WHD w S P )

M NDWG

• l N TGR L ( P h D WGD , WN SP D h- hND W SP-W NC w L )

MMDWG4INTGRL(MMDWGO,WMDWR+WMSPDW-WMDwSP-WMowLI WCDWG = W WCD WG + hvC DWG WVC DW G= ( M SD hG / T A UF D Wl * ( 1. -100. /HUMD W i

• C OMP A R ( HU MD W ,1D D . )

• HYDRCGEN AND MISC RELEASES SET TO ZERC

• FFRACT= FRACTION CF HCT LIQUID LE AK THAT FLASHES TO STE AM FFRACT=(AFGEN(HSATF,PSTS-AFGEN(HSATF,PTDWGal/...

(AFGEN(HSATG,PTCWGI-AFGEN(HSATF,PTDWGil WSDWR=WDLEAKeFFRACT HSDWR=AFGEN(HSATG,PTDWG)

WHDWR=0.

WMOWR=0. i H NO WR = 0.

HMDWR=0.

a

• WWCDWG= RATE OF COND. CN DWG WALL

• 4VCDWGAR ATE OF VOLUME COND. IN DRvhELL e

• ENERGY BALANCES FOR CW AND SP GAS

• INPUT FRCM OTHER CALCULATION $t

• HMOWR=ENTHALPY CF MISC. RELE ASED DIRECT TO CRf wELL

• HNDWR=ENTHALPY OF HYDROGEN RELE ASED DIRECT TO DRYWELL

• bHRV=ENTbALPY OF HYDRCGEN IN RE ACTOR VESSEL

• MSRV=ENTHALPY OF STEAM IN RE ACTOR VESSEL

• HSDMR=ENTHALPY OF STEAM RELE ASED DIRECT TO CRYWELL

• HSTE=ENTHALPY OF STEAM FROM TURBINE EXHAUST

• NOTE THAT THE ABOVE 3 ARE REF. TO ZERO AT 32.F WATER

• (ASME STP TABLESI, BUT ARE RE-REF TO ZERO

• AT ZERO DEG-R FOR CONTAINMEN T CALC.

• QVSPG,CVCWG= VOLUME HEAT SOURCE (FROM F.P.'S)

• QRVHLuME AT LCSS (THRCUGH INSULATICN) FRCM RV TO SP GAS

• QLSPG=HEA T LOSS FP.CM SP GAS TO SP LINER

• QLDhG= HEAT LOSS FRDM DW GAS TD DW LINER

• DEFINITICNSS

• TGSP=TEPF OF HCMcGENIZED SP GAS (DEF-Fi

• TODW= TEMP OF HCMCGEN! ZED DW GAS (DEG-Fi

• UM S PG= T O TA L INTERNAL ENERGY OF SP GAS (8Tul

• UMDWG=TCTAL INTERNAL ENERGY OF DW GAS (BTul

• MASS FLOh5 DEF1hED th MASS BALANCE SECTICN

• SP GAS EhERGY BALANCEI CUM 5PGalNTGRLIO.0,0UMSPG)

UMS P G=UM S P GO *CUMS P G

193 DUMSPG=EPSSP+EPHSP+EPNSP+EPMSP+GVSPG-QLSPC-MWORK+QSSPG EPSSP= W S SVNQ * ( H SSYNQ) + W STENQ* ( H ST E+ HS H REF l + . . .

W S DW S P * ( H S T GC W l-l W S SP DW + W S SP L l *H S TG SP =. . .

W CS PG 8 ( H F G S P + HS R R E F l +W T ES P W e HS T S P W MhDRKaDVOTPG*PTSPGo.1851 DVDTPG=(VGSP-VGSPDl/5.

VGS PC eRE ALPL (VGS PO S.eVGSP I

• WSSVNQ=TCTAL SRV FLCW THAT EXITS SURFACE OF POOL

• WSTENQ= TOTAL NDh-QUENCHED STM FROM TURBINE EXHAUST

• HSRREF= CELT A-H TO REREF. STE AM ENTHALPY

• HFGOW= SAT FLUID ENTH AT CW TEMP ( ASPEI

• HFGSP= SAT FLUID ENTH AT SP TEMP ( ASME) e

• MSSVNQ*ENTHALPY OF NON-QUENCNED STEAM ENTERING SP GASIREF TD 0.

• AT 0. CEG-R)

H S$V h C =H S T+H S R R E F HSTGSP=.458TGSPR-4.89 HSTSPW=.45*(TWSPAV*460.5-4.89 H STGD h=. 45* TGChR-4.89 HFGSP=TGSP-32.

HFGDW=TGDh-32.

E PHS P e W HLJ S P

• HH T G CW +W HSV* PHR V- ( WHSP DW +W H SPL l *HH TG SP HHTGSP=3.466*TGSPR-40.

HHT GOW= 3. 4 6 6* TG DWR-40.

EPNSP=HNTGDhabhCWSP-(WNSPCW+WNSPLl*HNTGSP HNTGSP=.2475*TGSPR HNTGCW=.2475*TGDWR EP M SP = HM TGD h* h MD W SP-( W MSP Dh+ W MS PL l

• HM TGS P HMT GS P= . 21

• T GSPR- 20.8 r H MT GCW = .21

• T G CW R-20. 8

• DRYWELL ENERGY B ALANCE CU MCW G= I NT GR L (D .O . DUM CW G I

• U MO kG =CUMDWG + U MD WGD DUMD W G=E P SD W +E PHOW+ E PND W+ E P MD W+ GV DWG+ . . .

QRV HL-CL OW G Q S$PGa( T W SP A V-TG S Pl

• A SSP h* 5.3 E-05 * ( ( A BS I TWS P AV-T GSP l l ** .33 )

QLSPGeQPMWET+QPMDRY QVDWG=0.

QVSPG=0.

QRVHL = QRV HLO

• 948. * ( T SA T- TGD W l /D TR VHL QLDWG=CDM EPSCWa t HS DWR +HSRR EF l *W SDWR+HSTG SP *WS SPD W-HS TGD h*( WSDW SP + W SD hLI . .

HFGDW*WCDhG e

EPHDW = HNDWR* WHDWR +HHTG SP o hHSPD h-HHTGO W* t hMD h SP+ kHOWL I E PND W.HN TG SP* hNSPCW-HN TGD h* ( WNCWSP+WNOW L )

E PMDW .HMD h R* h POk R + HM TG SP o h MS P C W-HM T GC W 8 ( W MDWS P+W M DW L I

• SDLUTION FOR Dh AND SP GAS TEPPER ATURES

• ITERATION IS NOT NECESSARY SihCE THE 5.N,HeAND M t

• ENTHALPIES ARE ASSUMED LINEAR WITH TEMPS

• H ( N21=.2475 7, H (H2 0l =.45T-4.89, H( M is. 21T-20.8. HtHl=3.4667-40.

T GSP R= ( UM SP G + 4.8 9 *M S SP G + 20.8 *MM SPG+ 4C. 0

• MH SPG l /. . .

' ( MNS PG* l .2475 .18 51 *GCN ) +MS S P G*( .45 .18 51 *GCS l + . . .

MMOWG*(.21 .1851*GCMl+MHSPG*(3.466 .1851*GChi )

T GDWR = ( UMD WG + 4. 8 9 *M SD WG+ 20. 8* MM D WG+ 40.

• MHD wG I /. . .

(MNCWG*(.2475 .1851*GCN)+MSCWG*(.45 .1851*GCSl+...

M MOWGe f . 21 .1851

• GCM l +MHOWG* (3.46 6 .18 51 *CCH l l

194 TG SP = T G SP R-460.

T GDW= T G0h R-4 60.

• TOTAL ANC PARTIAL PRESSUREStPSIA) CALC. FROM TEMPtFI P N SPG =MN SPG*GC h* TG SP R /VG S P PHSPG=MHSPG*GCH*TGSPR/VGSP PSSPG=MSSPG*GCS*TGSPR/VGSP PPSPG = MM SPG*G C P* T GS PR/VGSP PTSPG=PNSPG+PHSPG+PSSPG+PMSPG P NOWG = MN0 hGo G Ch* T GOW R/VGCW PHOW G=MH0 hG *GCH 8 TGDWR /VGQ W P S DW G=M S CW G* G CS *T GDWR /V GO W PPDWG=>MDhG*GCM*TGCWR/VGOW

• P TDWG = PN0 hG+ PHU WG + P SD WG+ P PO WG

• HUMIDITIES AhD CEWPOINTS H UMS P = 100.

• P S S PG/ hlF C EN E S P FO T . T GS P )

Hum 0W= 100*P50WG/NLFGENI SPFOT TGDW)

TC' aS P=NLFGEN tSTFCSV,VGSP/MSSPG)

TL t =0 ==NLFGEh t STF CSV,VGCW/MSDWG )

BChl=VGCW8FLChG WSDWL=6D WL* t M50WG /VGDW)

W H CWL = BOWL

• t M HDWG /V GD W )

W hCWL =80W L* ( PhCWG/V GCW )

WP0mL=80WL*IPP0mG/VGOW)

• CALCULATICN CF SP ANC CW WALL METAL TEMP ASSLMPTIchst

• 1.CCNSIDER CALY METAL SURF ACE IN CONT ACT WITH GAS

• 2.NO CONDENSING UNLESS ME TAL TEPP BELCW OEwPOINT

• STM.. CONCENS ATION ON WALLS IS AIR-LIMITED AND IS CALCULATED

• FRCM ECUh. I!!.8.26 CF MARCH M ANU ALI NU R EG/CR-17113 QPM=CPHORY+CPPWET-CPGAIR g QPMDRY= t T GSP-TPMET l *5 3E-0 5.* t t ABSI TGSP-TPPE Ti l**. 33 331* APPE T QPMWET=CCMPARITCEWSP.TPMETl*tTCEWSP-TPMET!*APMET*...

.0185*ttPSSPG/MNSPGl**.707)

TPMET=INTGRLt TPME TO,QPM/CPME T) =

QCM=CCMORY+GCPWET CDMOR Y= t TGDW-T CMETl *S.3C-5* t t ABS IT GCW-TOPER il * *.3 333 3 *ADMET QDMWE T= COMP AR I TDE hDW e TOME Tl e t T0EWCW-TOME il* ACMET *.. .

.018 5* t t M50WG/MNOWGl * *.70 T)

TCHET =INTGRLITDMET0 CDM/CCPET )

• CALC CF CCNDENSATICN RATE ON DW AND SP W ALLS. APPROXIM ATE VALUE

• CF 9CO BTU /L5 IS USED FOR (HG-HF)

WWCSPGaQPMWET/900.

W WCCW Ga QC PW E T/ 900.

• CALCULA T ION OF PCCL ROOM AIR TEMP QPAIR=CPGAIR+CPWAIR QP wAI R=5.3E-05* t t A85 t TWS P AV-T P AIR l l * * .33 33 )

• APM ET *t TWSP AV-TP AIR )

OPGA IR=5. 3E-5* t t ABSI TPMET-TP AIMll**.333)* APMET* t TPMET-T PAIR)

T P A IR= IN T GRL ( TP A IRO,QP A IR /C P A IR )

• NOTE THAT THE TERM APMET I S THE SAFE IN 8CTH EQU ATIONS

• BECAUSE IT IS FOR ONE HALF OF TOTAL PETAL SURF ACE AREA

• THESE CALCULATICNS ARE TC BE USED FOR AVERALL THERMO

• CONSERVATION CHECK O PS P= INT G R L lo . , W S T C l HE AT! h=I NTG A L IO. , CT CT )

• OMHR V= IN TGR L I O. , WI NJ

• H I NJ I N-W S T C* H S T )

LINTVl= LINT *12.+216.

PRINT LDCVZ,LSC LB P WTOST,WSTC TWSPAV,LWSPAV,GP,TGSP,TGOW,WCIO

• TIMER GELT =.25, FINTIM=2C.0CO, PRDEL=10.

METHCC RECT NCSCRT i

195 CALL DEBUGil.0.01 CALL CEBUGil.3600.1 CALL DEbuGile7600.1 CALL DEBUGile lC80C.I CALL CEBUG11.14400.1 CALL CEduGil 17999.1 END STOP

'P 1

9

197 APPENDIX B MODIFICATION TO MARCH SUBROUTINE ANSQ In the MARCH code, the calculation of decay heating is carried out in subroutine ANSQ. For this study, the calculation wac modified

, to include the actinide decay heat source in a BWR following a burnup of 34,000 mwd /T. A listing of the revised subroutine fol.'ows.-

Subroutine ANSQ (ANS. TIME. TAP)

C C ANSQ calculates the decay heat using ANS correla-C tions.

ANSl=ANS2=0.0 TVAR= TIME *60.0 IF (TVAR.GT 10.0) CO TO 10 ANSl=0.06950-0.001592*(ALOG(TVAR))

IF (TVAR.LT.3.) ANSl=1.

GO TO 40

' 10 IF (TVAR.GT.150.0) GO TO 20 ANS l= 0. 069241-0. 0069355 * ( ALOG (TVAR) )

GO TO 40

-20 IF (TVAR.GT.4.E6) CO TO 30 ANSl=4.0954E-2-2.8324E-3*(ALOG(TVAR))

GO TO 40 30 ANSl=4.6893E ~2-2.3800E-4*(ALOG(TVAR))

40 TVAR=(TIME + TAP)*60.0 IF (TVAR.GT.10.0) GO TO 50 ANS 2= 6. 950E-2-1. 592E-3 * ( ALOG (TVAR) )

GO TO 80 50 IF (TVAR.GT 150.0) GO TO 60 i ANS2=6.9241E-2-6.9355E-3*(ALOG(TVAR))

GO TO 80 60 IF (TVAR.GT 4.E6) GO TO 70 i ANS2=4.0954E-2-2.8324E-3*(ALOG(TVAR))

(- GO TO 80 70 ANS 2= 4. 6893E-3-2. 3800E-4 * (ALOG (TVAR) )

d0 ANS-ANSI-ANS2 RETURN END 9

l

[

[

- - m- J - - -- r- ----

199 b

i f

APPENDIX C i MARCH INPUT FOR ACCIDENT SEQUENCE TB' t

i

!e I -- - - - - . . . ._ .- _. , , , _ . . _ . - , . . . - , , - - -.

201 BROWNS ?EPRY CSB + HPCI/BCIC

$NLMAR ITRAN=1, IBRK=0, ISPR&=1, IECC=2, IBURN=0, NINTER=100, IP DTL= 7, IPLOT=3, 10=3, YOLC=416700.0,

• DTINIT=0.01, T&P=2.62E06, SEND

$NLINTL SEND STEEL CONCRETE DRYNELL1 DRYWELL2 CONC SHELLEISC STEELEISC CONC.

C

$NLSL&B Na&T=2, NSL&B=3, NOD =1,4,13, DEN (1) =4 86.924,157. 481, HC (1) =.19 3 7. 2 3 817, TC (1) = 25.0 01. 80024, ITL=1,1,2, IVR=1,1,2, F NN01=3,9,4, N AT1 = 1,2,1, MAT 2=1,2,1, e S AR E A= 186 8 4. ,5 358. 15 98 2. ,

I (1) =0. . 01,. 02 083, 114) =0. 01,. 0 3,. 07,.15,. 31, .63,1.27,2. 5, I(13) =0. . 01, . 03, . 062 5, TEEP =12*150.,4*95.,

SEND

$NLECC PDHIO=0.001, UNIO =3.11E04, P&CR0=0.001, AC50=3.11E04, PHH=1120.,

N H H 1 =-5 000. 0, PSIS=1120.,

WSIS1=0.0, PLH=1120.,

ULH1=0.0, STPHH=240.,

RUST 5=3.11E06, ECCRC=0.64, CSPRC=1.0, DTSU B = - 10 0. ,

, UTC&Y=100.,

l

, TRUST =95.0,

! $END e S NL BC1 l

SEND

$NLCSI SEND

202 SNLCOOL SEND SNLNACE NCU B = 2, NRPY1=2, NBPT2=1, ICECUB=-1, DT PNT= 2 0. 0, IDRY=-1,

• IVET=2, UPOOL=7.801E06, TP00L=95.0, +

TDRY=3.839E03, YTORU5=257700.0, NYs&I=5.146E05, BSNP=-2, MS5P2=2, NC&Y=-1, TCAY=4789.1, TFLR=15.0, AYBBK=292.0, CYBBK=4.04, TC (1) = 159000.0,257 700.0, ARE A (1) = 1. 63 99E03,1.098E04, H0 5 (1) = 0. 2,1. 0, TEMPO (1) = 150.,95.,

N=10, 5 S (1) = 1,1,1,3,3,2,2,2,2,2, NC (1) = 1,1,1,1,1,1,1,2,2,2, N T (1 ) = 1,2,3,- 7 -7, - 7, -7, -7,-7, -7, C1 (1) = 1. E6,1. E6,0. ,40 0. ,500. ,139. 7,18 9. 7,13 9.7,18 9.7,15 9.7, .

C2 (1) = 0. , 3. 333E S,7. 59106,5.92 97, . 5 83,5. 92 97, . 5 83,5.92 97, . 5 8 3, . 5 83, C3 (1) = 95. ,0.,1192. 5. 006 94,20. 97, . 0 06 94,20. 97,6. 94 E-4, . 083 3,6.94 E-3, C4 (1) =0. 0. 0.0,0. 0,14. 7,0. 0,14. 7,0. 0,14.7,14.7,14.7, KT (1,2) = 1,

• KT (2,1) = 1, STPECC=240.,

$END

$ NLBOIL NNT=37436, NR=35908, NDZ=50, ISTR=3, ISG=0, IRWA=1, IHR=1, WD ED= 3. 50 E 0 4, QZER0=1.1242E10, H=12, H0=28.,

DC=15.59, ACOR=104.833, ATOT=287.898, WATBB=97000.,

D=. 0 469 2, Dr=.04058,

• DH=0.056, CLAD =.005594, IOO= 8. 3 3 E- 0 6, RHOCU=68.783, IGOO=546.,

PSET=1050.0, CSRT=3380.,

FDCR= .5,

203 DPART=0.0208333, FZ BC B = 0. 05, FZOCR = 0. 0 8, FE051=0.1, WFE2=7992.,

TFE00=546.,

FULSG=0.0, PTSL=1000.,

, TC&Y=1210.,

ABRK=0.0, YBRK=45.0, e DT PMT B= 5. 0, DTPN=-5.,

YOLP=2.459E04, TO LS= 9. 6 38 E 03, WCST=3.11E06, F (1) = 0.1,0. 2 5, 0. 4 7,0. u s,0. 84,0. 96,1.13,1. 2 7,1. 2 95,1. 2 7,1. 2 4,1. 21,1.19 5, F ( 14 ) = 1.15,1.1 1,1. 0 8,1. 0 5 ,1. 0 3.1. 0 2,1. 016,1. 017,1. 0 5,1. 06,1. 0 61,1. 0 62, F (2 6) = 1. 0 61,1. 0 6,1. 0 5 9,1. 0 59,1. 06.1. 0 7,1. 07 5,1.0 95,1.11,1.12,1.185,1. 215, F (3 8 ) = 1. 2 5,1. 2 6,1. 2 4,1. 21,1.15,1. 0 9,1. 0,0. 8 7, 0. 7 6,0. 6 ,0. 41,0. 21,0.1, PF ( 1) = 1. 017,1. 0 87,1. 0 93 ,1. 0 95,1. 0 9 6,1. 0 9 4,1. 0875,1.12 8, 0. 9 6 65, 0. 4 0 8, TF (l) = 10 *0.1, TT=6*546.,

Cs= 18 24. ,7992. ,876 0.,2460. ,5550. 24900. ,

&H=740.,263.,9225. 400.,7000.,700.,

DD= 1. ,1. ,1. , .17, . 0 2,. 54 6, AR= 15 0. ,2 63. ,16 5. 0.,-10. ,-20. ,

SEND

$NLHEAD UZRC=140397.0, WFEC=30447.73, EUO 2= 361837. 0, BGRID=66750.,

. WHEAD=175927.08, DBH=20.915, THICK = 0. 52198, COND=8.0005, E1=.8, E2=.5,

$END

$NLHOT IHOT=100, DP=0.25, FLRMC=3360.,

SEND

$NLINTE C&YC=0.01524, CPC=1.30, DENSC=2.375, TIC =308.16, FC1=0.441, FC2=0.108, FC3=0.357, FC4=0.027, R B R=0.13 5,

[

R0=322.6, R=6000.0,

. HIs=0.2, HIO=0.09, WALL =1000.,

SEND

205 Appendix D PRESSURE SUPPRESSION POOL MODEL D.1 Introduction o The primary containment of each of the Browns Ferry Nuclear Plant units is a Mark I pressure suppression pool system. The safety objective of the Mark 1 containment is to provide the capability, in the event of an accident, to limit the release of fission products to the environment.29 The key to the safety objective of this system lies in the performance of the pressure suppression pool (PSP). The PSP is designed to rapidly and completely condense steam released from the reactor pressure vessel, to contain fission products released f rom the vessel, and to serve as a source of water for the emergency core cooling systems. If the PSP should fail to properly perform during an accident sequence, then there is a high probability that fission products will be released to the environment.

In the past few years, a major concern of the NRC, the BWR vendor, and the utilities who operate the early generation BWRs has been scenar-ios30 where it has not been conclusively shown that the PSP will meet the performance requirements stated abose. Specifically, there is a lack of information on condensation oscillations and the resulting loads they place on the PSP. In addition, there is the question of pressure suppres-sion pool thermal stratification and the resulting impact of this phenome-e non on the condensing ability of the PSP.

The problem of condensation oscillations

• includes the analysis of both loads which originate at the downcomer exit during a large break loss

' of coolant accident (LOCA) and loads derived from the safety relief valve (SRV) discharge during normal blowdown to the PSP. The condensation os-cillation problem involves coupling an analysis of the momentum trans-port within the PSP to a stress analysis of the torus walls.

The problem of pool thermal stratification is of interest because thermal layering of the water near a SRV discharge point can limit the ability of the PSP to condense the steam. Subsequently, this can lead to increased intensity of the condensation oscillation loads and possibly to overpressurization of the torus. Furthermore, the temperature distribu-tion in the torus determines (to some extent) the distribution of fission products in the FSP during an accident. Thus, an analysis of fission pro-duct transport to the environment will depend on an analysis of the tem-perature distribution in the PSP.

D.2 Purpose and Scope The purpose of this work is to study the dynamics of the Mark 1 pre-g ssure suppression pool (PSP). Knowledge of the response of the PSP is

• In the literature, when the condensation instability occurs outside the relief valve tailpipe it is termed " condensation oscillation." If it occurs inside the pipe, the phenomenon is called " chugging." Here, " con-densation oscillations" includes both of these effects.

206 desired for transients ranging f rom a normal SRV discharge to a full-blown, large break LOCA. The primary objective of this study is to im-prove the available BWR pressure suppression pool analysis techniques. A ne:ondary objective is to learn as much as possible about the complex phenomena involved in each of the transients (phenomena such as pool swell, condensation oscillation, chugging, and thermal stratification).

An analysis of the PSP dynamics will necessarily involve accounting for the two phase flow of steam jets into subcooled water and the ensuing

• transport of mass, momentum, and energy by the mechanism of condensation.

Bervase of the complex geometry of the PSP, (toroidal geometry plus large, ,

submerged, complicated flow obstructions ppis an air-water interface) the analysis tool used in this study will be a s: ate-of-the-art, 3-D thermal hydraulics code.

Pending completion of this work, it will be necessary to assume pool-averaged conditions f or accident analyses.

D.3 Description of the System The Mark I containment system consists of the drywell, the pressure suppression pool, the vent system connecting the dryuell and PSP, a con-tainment cooling system, isolation valves, and various service equipment.

Figure D-1 shows the arrangement of the drywell, PSP, and vent system within the Reactor Building.

The drywell is a steel pressure vessel with a spherical lower portion and a cylindrical upper portion. It is designed for an internal pressure s of 0.531 MPa (62 psig) at a temperature of 138'C (281*F). Normal environ-ment in the drywell during plant operation is an inert atmosphere of ni-trogen at atmospheric pressure and a temperature of about 57'c (135*F).

• The vent system consists of 8 circular vent pipes which connect the drywell to the PSP. The vent pipes are designed to conduct flow from the (

drywell to the PSP (in the event of a LOCA) with minimum resistance, and to distribute this flow uniformly in the pool. The vent pipes are de-signed for an internal pressure of 0.531 MPa (62 psig) with a temperature of 138'C (281*F); they are also designed to withstand an external pressure of 0.014 MPa (2 psi) above internal pressure.

The pressure suppression pool is a toroidal shaped steel pressure vessel located below the drywell. The PSP contains about 3823 m3 (135,000 f t3) of water and has an air space above the water pool of 3370 m 3 (119,000 f t3). Inside the PSP, extending around the circumference of the torus, is a 1.45 m (4.75 f t) diameter vent header. The 8 drywell vents connect to this vent header. Projecting down from the vent header are 96 downcomer pipes which terminate 1.22 m (4 f t) below the surface of the water. At 13 unevenly distributed positions around the PSP, discharge lines from the safety relief valves extend through the vent pipes and ter-minate in a T quencher device located near the bottom of the pool. Figure D-2 shows a cross section of the PSP and the relative locations of the 3 vent pipe, vent header, downcomer, SRV discharge line, and the T quencher, which has been rotated 90* for the purpose of illustration. Near the bottom of the PSP, a 0.762 m (30-in.) suction header (ring header) cir-

• cumscribes the torus and connects to the pool at four locations. At Browns Ferry, the RHR, HPCI, core spray, and RCIC systems are supplied from this header.

207 ORNL-DWG 79-7011 A ETD i ,-

C4g-l(m._v W 1 u u y; nf 4 y

- s g & :m -

M y W b DRYWELL

[~ W

\

o r Fy J(k' - REACTOR VESSEL I

fd _ , , oq.

s h J('

s

.O r]"j .

J

, 5 ,

H r 6 p .,,,,

/ ',.,

x U ih VENT PIPING t  ! ,1 ,- CTOR LDING (8 PIPES) % g } .j ,4 j Q- cm) -pi$*Y tj'/.

j- ,

} 1: 5 -( , y.:.

P SUPPRESSION 'Q' r er

{TM ,-

i.

Q[ T CHAMBER (  ;

E' ' ,

TOROIDAL \ K/ g ' 7, HEADER k\ tj

[, ,

p /  %'t DOWNCOMER "'^h'$/ 69' PIPES (96) ,4f Fig. D.1 BWR reactor building showing primary containment system enclosed.

The torus which contains the pressure suppression pool is designed to essentially the same requirements as the drywell liner, i.e., a maximum internal pressure of 0.531 MPa (62 psig) at 138 C (281*F), but neither the drywell nor the torus is designed to withstand the stresses which would be created by a significant internal vacuum. To ensure that a significant vacuum can not occur in the drywell, vacuum breaker valves aye installed, which will open to permit flow from the PSP airspace into the drywell whenever the suppression pool pressure exceeds the drywell pressure by more than 3447 Pa (0.5 psi). Additional vacuum breaker valves with the same setpoints are installed to permit flow from the Reactor Building into r the PSP airspace, to prevent a significant vacuum there.

D.4 Identification of the Phenomena s

The thermal hydraulic phenomena associated with a BWR pressure sup-pression pool are mainly those dealing with the PSP response during two

1 ORNL-DWG 81-8567 ETD MAIN VENT BELLOWS RING GIRDER

= - 31 ft 0 in. ID 4 f t 10 in. p VENT HEADER

"* r SAFETY RELIEF VALVE f

DISCHARGE LINE f 7 5 ft 10.5 in. 10 ft 10.5 in. ,/ \, Oin.&

k /f A

,-- .-4_

~ '

1 ft in' s .DOWNCOMER

4 ft 0 in., ,

.- ~

,~f-

/

18 ft 1 in.

Y \ s

[

I 2 ft 0 in. e i ** 16 f t 1 i"'

4 I VENT HEADER O C!!!!!!! I!!'

7 M'9U()

T-OUENCHER CTIOtt

" S /

LINE y

55 ft 9 in. TD AND DRYWELL

-4 e-t = 0.756 in,

• Bro n s Ferry Mark I containment Pressure suppression pcni, 4

s s &

• e

209 types of transients: (1) LOCA-related phenomena and (2) SRV-discharge phenomena.

A. LOCA - Related Phenomena Immediately following the pipe break in a LOCA, the drywell pressure 9 and temperature increase very quickly. The pressure increase forces water standing in the downcomer to accelerate rapidly into the PSP and impinge on the torus wall. Following the slug of water, air that was in the vent pipes and drywell is forced into the PSP. This forms a bubble of air at the downcomer exit which expands into the suppression chamber and causes the pool to swell. As the air bubble rises into the torus airspace, the water will experience a gravity-induced f allback and phase separation will again occur.

The pool swell transient described above lasts on the order of 3 to 5 s.30 It has been studied by several authors, both experimentally31-33 and numerically.34,35 The consensus is that pool swell impingement and drag loads induced during a LOCA are conservatively estimated and acceptable.

Immediately following the pool swell transient, an air / steam mixture will flow irto the PSP. Early in this process, when the mass flow rate is high, the injected steam condenses at an unsteady rate causing periodic oscillations in the pressure and flow. However, since the mass flux is high enough to maintain the steam / water interface outside the downcomer, the overall condensation proceeds at a regular rate. This phenomenon is

1. known as condensation oscillation. It is characterized by a steady, peri-odic variation in the pressure which forces local structures within the torus to vibrate in phase with the oscillations. Condensation oscillation has been studied experimentally 36,37 and analytically; 37,38 however, the basic driving mechanism f9 r the pressure resonance has not been iden-tified.39 When the air / steam flow through the downcomer decreases to the point where the condensation rate outside the pipe exceeds the steam flow exit-ing the pipe, the steam bubble collapses very rapidly. This results in a large drop in the steam pressure and the steam-water interface rushes up into the downcomer. Once there, the interface is warmed by condensing steam and the condensation rate begins to decrease. At some point, the steam pressure will rise, and the interface is pushed out of the downcomer to form an irregularly shaped bubble at the pipe exit. The bubble begins to collapse; and the entire process, known as chugging, repeats. Chugg-ing is characterized by rapid, irregular interface accelerations and pres-sure oscillations that cause large loads on the torus structure.

The chugging phenomenon is very similar to the condensation oscilla-tion problem. It has also been studied in detail, with analysis methods which range from manometer-like models that attempt to predict the gross motion of the interface, to probabilistic models that attempt to predict e internal chugging.40-42 The central problems which plague analysis of the chugging phenomenon are (1) high uncertainty in the basic condensation rates involved and (2) lack of understanding of the triggering mechanism for bubble collapse.

Both of the condensation phenomena (chugging and condensation oscil-lation) involve transient, stochastic, turbulent, two phase flow. Because

210 of the complexity of these problems, no accurate assessment has been made of the loads involved.30 Consequently, the analysis of PSP response to condensation oscillations must rely on data from experiments which model plant behavior. .Thus, there is a need for improvement of the analysis capability in this area.

B. SRV -- Discharge Phensmena ,

Wher e SRV actuates, water and air initially in the discharge line ,

l are immediately accelerated into the PSP. This results in air clearing loads much the same as the pool swell loads discussed earlier. These loads are of.no major consequence because they can be adequately " scaled" f rom the results of laboratory experiments.

Following the air clearing phase, steam is injected at high velocity l into the PSP. Experience has shown that, depending on the discharge de-vice used, condensation oscillations can occur as steam bubblea exit the pipe and collapse in the bulk fluid. As the steam bubbles collapse, severe pressure oscillations are induced on the surrounding structures.

l Current practice is to limit the severity of these oscillations by using a l T quencher device.43 The T quencher is a section of pipe (in the form of a "T") with holes strategically drilled in the arms to enhance local mixing.- The T quencher has been shown to be very effective in eliminating the severity of the pressure pulses, provided the local bulk fluid temper-ature is sufficiently low.

This is the factor that establishes a limit on the effectiveness of

• l the T quencher -- the local fluid temperature. If it becomes too high, the PSP will not be able to rapidly condense all- the steam. Some of it

~'

will escape into the air space above the pool and pressurize the torus.

When the suppression pool pressure becomes 3447 Pa (0.5 psi) greater than the drywell, the vacuum breakers will open and steam will escape inte the drywell (forcing the temperature and pressure up). As more steam is dis-charged to the PSP, the drywell and torus pressure will continue to in-crease. -

It is concern for this increase in drywell pressure (possibly resuit-ing in overpressurization) that has led to the establishment of strict procedures for the sequential ordering of SRV blowdowns to the PSP. Cur-  ;

rent practice involves sequentially venting SRVs on opposite sides of the torus; this prevents excessive local temperatures near any discharge point.

More recently, concern has been expressed over the long term response of the PSP during a prelonged Station Blackout. During part of this scenario, the operator loses manual control of the SRVs. The long term result is that a single SRV will continually open into the PSP. The local fluid temperature will monotonically rise, resulting in pressurization of the torus and possible condensation oscillations. The potential exists for rupture of the torus due to overpressure coupled with violent pressure oscillations.

For example, this potential exists in the particular case of a Sta-tion Blackout with loss of the HPCI and RCIC systems. At about 100 min into the transient, local pool temperatures at the discharge bay are post-ulated (c.f. Section 9) to be greater than 149'C (300*F). At that time, I

l

211 steam condensation oscillations are expected to accelerate due to the ex-cessive tenperature and the continuous discharge of superheated, non-con-densible gases (from hydrogen generated in the melting core) into the PSP.

These extreme conditions in the PSP yield a high pro'bability for rupture of the torus.

Along with the temperature rise assocated with SRV discharge comes a 0

gravity induced thermal stratification of the pool. This phenomenon is of interest because the layering tends to remain long after SRV discharge is c omp le te. Since the stratification remains for a significant period of

, time, the validity of the current design basis for Mark I PSPs becomes questionable. This is because a fundamental assumption in virtually all the transient analyses is that the pool is thoroughly mixed and at a uni-form temperature following SRV blowdown. The impact of thermal stratifi-cation on the performance of the PSP remains to be evaluated.

Thermal stratification is alsc uf interest because the temperature distribution in the PSP affects the fission product distribution in the t oru s. Should a breach of primary containmect occur, transport of the fis-sion products to the environm-et will depend, to some extent, on the the-rmal stratification in the torus. Thus, an analysis of fission product transport is coupled with the thermal stratification problem.

To the best cf our knowledge, no information is available in the open literature concerning systematic analysis of the thermal stratification problem. There io therefere the need for research in this area.

D.5 Pool Modeling Considerations e

Mcdel requirements s.

The prediction of suppression pool system behavior during the course of a Savere Accident is an exceedingly difficult task. The previously de-scribed phenomena comprise a complex set of physical processes for which few detailed analytical models currently exist. These phenomena can be roughly divided into two types, i.e., thermal hydraulic phenomena and fluid-structure interaction phenomena. The initial ef forts at ORNL are directed toward the development of the thermal hydraulic model. A brief survey was conducted to identify existing multi-dimensional thermal hy-draulic analysis codes which might be applied to the problem. The re-sults of :he survey are shoun in Table D1 which presents a brief summary of the candidate codes.

Due to the nature of the previously described phenomena, it was felt that any model employed for suppression pool analysis should possess the following characteristics:

1. Appropriate hydrodynamics

a. two phase

b. non-equilibrium

, c. free field format (equations and constitutive relationshipa independent of hydraulic diameter)

d. incorporate a non-condensable gas field

. e. employ multi-dimensional geometry (3 dimensions desirable)

2. Employ appropriate constitutive relationships

3. Utilize efficient solution techniques

4. Readily accessable to ORNL staff.

212 Table Dl. Candidate suppression pool analysis codes

= Code E.veloper Geometry T-H characteristics TRAC- .LANL 3D cylindrical 2 fluid

~(PIA, PD2) nonequilibrium .

COBRA-TF BPNL 3D cartesian 3 field nonequlibrium *

. COMMlX-2 ANL 3D 2 fluid nonequilibrium BEACON / Mod 2 INEL 2D cartesian 2 component vapor cylindrical- 2 phase spherical nonequilibrium Requirements 1.a and 1.b are imposed by the complex nature of the conden-sation processes involved in SRV discharge and LOCA blowdown transients.

Requirement 1.c is a result not only of the nature of the SRV discharge process, but of the geometric dimensions of the suppression pool. The en-ormous size and complex structural design of the suppression pool system virtually mandates that free field hydrodynamics be employed. - Requirement 1.d is imposed because significant amounts of non-condensable gasec will-be released from the fuel and generated in metal-water reactions during

• Severe Accidents. Requirement 1.e. is imposed by the nature of the hy-drodynamic phenomena and the complex structural relationship of the SRV discharge lines, vent downcomers, and ring header suction system.~ .Appro-priate constitutive relationships (requirement 2) are needed for closure of the hydrodynamic field equations. Many of the relationships employed in existing thermo-hydraulic codes may be invalid for use in _the present problem. Requirement 3 is imposed by the fact that it will be necessary to predict the behavior of the suppression pool for periods of _ time rang-ing from a few minutes to several hours. It is, therefore, important that all codes employ fast' solution techniques in order to minimize computing costs. The fourth characteristic is desired because the computer code selected for the analysis will require local modification as necessary to reflect the unique characteristics of the suppression pool problem.

We are unaware of any existing computer code or model which pos -

sesses all of the desirable' characteristics outlined above. The initial effort at ORNL is directed toward development of a pool model utilizing the TRAC-PIA 44 vessel module. TRAC (Transient Reactor Analysis Code)

- PIA is a best estimate computer code . developed at Los Alamos Na-tional Laboratory for analysis of large break PWR loss of coolant acci-dents. Although more recent vereions of TRAC are-available (i.e., PD2, PF1, and BD1), TRAC-PlA was chosen for this application because it is in-

• stalled and operational on ORNL's computer system.

While TRAC-PIA does employ a full three dimensional, two fluid,' non-equlibrium approach to reactor vessel hydrodynamics, it does not account '

for free flow or non-condensable gas fields. Additionally, the constitu-

' tive relationships employed in TRAC-PIA are based upon data which was i

I

213 originally developed for vertical pipe flow and may not be valid for ap .

plication to steam jet discharge phenomena.45 Though these limitations are significant, it is felt that our access to TRAC-PIA will allow us to implement any changes in the constitutive relationsh'ip package which right be necessary to reflect the characteristics of the suppression pool phenomena. LANL is currently modifying TRAC to include non-condensable gas field hydrodynamics (TRAC-PF1). ORNL's utilization of TRAC-PIA will allow us to quickly implement this modified version when it becomes avail-able.

, We have tentatively chosen the PELE-IC46 code for future use in an-alysis of the fluid-structure interaction within the suppression pool.

PELE-IC (developed by LLL) couples a two dimensional Eulerian fluid dy-namics algorithm to a Lagrangian finite element shell algorithm. The code can couple either a one dimensional or a lumped parameter description of compressible gases, and can employ either cartesian or cylindrical coordi-nates. PELE-lC employs the basic semi-implicit solution algorithm con-tained in the SOLA code.47 The movement of free surfaces is treated in a full donar cell fashion based on a combination of void fractions and interface orientation. The structural motion is calculated by a finite element method, from the applied fluid pressure at the fluid structure in-terface.' The finite element shell structure algorithm uses conventional thin shell theory with transverse shear and provides the fluid module with the resultant position and velocity of the interface. The code is capable of analyzing both vent clearing and condensation related phenomena.

• TRAC suppression pool model description A block diagram of the ORNL-TRAC suppression pool model is shown in Fig. D-3. The model is comprised of the TRAC VESSEL module which repre-sents the pool hydrodynamics, and two FILL and PIPE modules, which repre-sent the SRV discharge flow and T quencher assembly. A detailed view of ORNL-DWG B1-8150 ETD FILL FILL I

I f U l

PIPE PIPE I f U VESSEL l

Fig. D.3 Suppression pool model block diagram.

l l

I 214

the pool model is shown in Figs. D-4 and D-5. The pool is represented as 112 cells in cylindrical geometry, located at five axial, four azimuthal, and eight radial regions. SRV discharge is assumed to occur as shown in Figs. D-4 and D-5. Half of the SRV discharge flow is directed radially inward from the fifth radial zone, while the other half is directed ra-

dially outward. It should be noted that the T quencher is not modeled in a strictly physical fashion, since the discharge is assumed to occur through a single open ended pipe with a flow area approximately equal to '

that of the T-quencher device. This approach was chosen for the initial model due to the significant reduction in input preparation time and .

model complexity associated with this approach. The initial water level is assumed to be 4.47 meters above the pool bottom (i.e., at level 4 in ORNL-DWG 81-8151 ETD A

03 02 Rg o

R8 Rg

• -R l'OP VIEW R  ; ~R y R4 Rs

~

R6 k .

! R, = 12.3901 m R2 = 12.9311 m 3~

• 6 = 0.0165689 rad i

6 = 2.1053743 rad 1

4= 035 m 63 = d.1942798 rad i 5 = 16M26 m 64 = 6.2831853 rad 64 8i R6 = 19.2877 m ,

R7 = 20.4643 m l R8 = 21.0601 m

• Rg = 21.6011 m Fig. D.4 Supptession pool model ton view.

. , e * .- -

ORNL-DWG 81-8152 ETD T.

VESSEL R7 R8 Rg RR 3 2 R3 R4 R5 R6 CENTER Ev ll l l l l l l l ll l l l 1 I I i

- 8 54 *  !!!ji  !!!!s gi!!!

eh !!!i!!  ! si!!!!  !!!i!!!

s!!!!.! !!!!!!!!!!!!! !sts si!!! !!!!!!! !!!!!!!

!!!j!!!  !!!j!!! !!!j!!! !!!!!!!

'l!!l!! i  !!!!!!! ~

!!!lg!!jjj l,jjjj i jjj{9:ll jj l ig gjjj

. . . . . 4:

HALF CROSS SECTION VIEW AT AA

4:

.et E NO FLOW REGICN

.t:.: !!!)+s siis

. ,,4, .

!!!!!! !!!!!!! !!!)g$

Jig!!! i6.jjga.... ..... . . ...itg!!!!s .. . ..... ...isg!!!!!p!J j!!!s! [] INITlAL FLUID CELLS y

l l l l @ INITIAL VAPOR CELLS l l l 4- - - -l- - - -4 z3= 3.os m h; I I I I I ,g' -

i

, 2 = 1.68 m A

b -- g -- b -

Z = 1.37 m - -

7

..F  ! .. .

1 i

Fig. D.5 Suppression pool model cross section.

l 216 Fig. D-5) with all cells in level 5 occupied by vapor. Internal struc-tures (vent header, down comers, etc.) are not modeled.

Current status of suppression pool model Several preliminary computer runs have been made in which TRAC data base errors have been identified and eliminated. We are currently exper-iencing problems related to the limitations of TRAC-PIA in accounting for the simultaneous existence of steam (vapor) and perfect gas (nitrogen) '

within the vessel. The TRAC user must specify that either vapor or air properties be used everywhere within the code. In tenas of the pressure . l suppression pool, this means that the nitrogen gas (which is located in  !

the space above the pool) must be treated as subcooled vapor if one is to inject steam into the pool. It is also possible that the constitutive re-lationships within TRAC are producing instabilities in the solution due to their dependence upon hydraulic diameter and fluid property information.

We are currently working closely with the TRAC User Liaison Section at LANL in an effort to determine whether these problems can be overcome. In the event these problems prove to be insurmountable, we will re-evaluate the remaining code candidates and select an alternative program for the suppression pool thermal hydraulic analysis.

a S

l 217 APPENDIX E A COMPENDIUM OF INFORMATION CONCERNING THE BROWNS FERRY UNIT 1 HIGH-PRESSURE COOLANT INJECTION SYSTFli

• E.1 Purpose

• The High-Pressure Coolant Injection (HPCI) System is designed to en-sure adequate core cooling to prevent damage to fuel in the event of a loss of coolant accident that does not result in rapid depressurization of the reactor vessel. The HPCI System provides water to make up for that which is lost through steam generated by decay heat.48 E. 2 System Description The HPCI System (Fig. E-1) consists of a steam turbine driven booster pump - main pump combination and the associated piping and valves.

The booster pump can take suction from either the condensate storage tank or the pressure suppression pool. The HPCI pumped flow enters the reactor vessel feedwater line "A" via a thermal sleeve connection. A test line permits testing of the HPCI System at full flow while the reactor is at power, with the main pump discharge routed to the condensate storage tank.

• A minimum flow line connects the main pump discharge to the pressure sup-pression pool, as a means of ensuring a flow of at least 0.038 m3 /s (600 GPM) through the pumps. The normal setpoint for HPCI pumped flow is 18.93 m3 /s (5000 GPM).

The HPCI turbine is driven by steam extracted from main steam line "B" upstream of the main steam line isolation valves. The two primary containment isolation valves in the steam line to the HPCI turbine are normally open to keep the piping to the turbine at elevated temperatures to permit rapid system startup (within 25 s of receipt of an initiation signal). The normally closed DC-motor-operated steam supply valves up-stream of the HPCI turbine will open against full system pressure within 20 s after receipt of a system initiation signal. Signals from the con-trol system open or close the turbine stop valve. The turbine control valve is physically attached to the HPCI turbine and is positioned by the turbine governor as necessary to maintain the pumped flow at the level set by the operator, normally 0.315 m /s 3 (5000 GPM). The turbine exhaust steam is discharged to the pressure suppression pool. The turbine gland seals are vented to a gland seal condenser. A small water flow diverted from the booster pump discharge is used to cool both the turbine lubricat-ing oil cooler and the gland seal condenser. This cooling flow is re-turned, together with the gland seal condensate, to the booster pump suc-

/ tion. Noncondensible gases from the gland seal condenser are removed via a DC-motor-operated blower to the Standby Gas Treatment System.

A vacuum breaker line (not shown on Fig. E-1) is installed between the torus airspace and the HPCI turbine exhaust line. Its purpose is to prevent water from the pressure suppression pool from being drawn up into the turbine exhaust line as the steam condenses in this line following turbine operation.

218 ORNL-owG et-8591 ETD

' CONDENSATE

. STORAGE ELEVATION 577 ft- NORMAL WATER LEVEL ELEVATION 675 ft CONDENSATE MAIN STEAM l '

RETURN

() HEADER CoNDE NSATE (<)

SUPPLY HEADER THERMAL SLEEVE ELEVATION 574 ft F E E DWATE R SUPPRESSION =

POOL I hWATER LEVEL grg E LE V ATION 536 ft D

.A _-

SUPPRESSION TURBINE CHAMBER TOTAL HEADER MAIN PUMP LEAF 0FF PUMP Q ELEVATION 524 ft - - -

===== =f INE

^

AT NT SYS M BOoSTE R PUMP ]

7 il s li

~

-GLAND SE AL MINIMUM FLOW LINE _

CONDENSE R

__b

' LUBE 0:L SYSTEM TEST LINE , CooiER Fig. E.1 liigh pressure coolant injection system.

All components required for operation of the llPCI System are com-pletely independent of AC power, control air systems, or external coolind water systems, requiring only DC power from the unit battery. On loss of control air, the llPCI steam line drains to the main condensers will fail closed; this is their normal position when the llPCI system is in opera-tion.

The principal liPCI equipment is installed in the reactor building, at a level below that of the pressure suppression pool. The turbine pump as-  %

sembly is located in a shielded area so that personnel access to adjacent areas is not restricted during ilPCI System operation. The only operating component located inside the primary containment is the normally open AC-

• motor-operated inboard llPCI steam line isolation valve, which will remain open on loss of power.

219 E.3 HPCI Pump Suction The HPCI pump can take suction either from the condensate storage header or from t'ae pressure suppression pool via the suppression pool ring header. The normal lineup is for suction of the reactor grade water in the condensate storage header.

t Each Browns Ferry unit is provided with a 1419.4 m3 (375,000 gal-lon) condensate storage tad , which provides a water head to the storage header for that unit. The storage healer, which taps into the bottom of 3

the storage tank, feeds the suctions of the high pressure ECCS Systems, specifically, the pumps for the HPCI and the RCIC systems. The core spray pumps and the RHR pumps can be fed from this ' source, but are not normally aligned to it. All other demand for condensate storage tank water is fed via a standpipe within the tank; the standpipe height is,such that 511.0 m3 (135,000 gallons) of water is reservud for the ECCS Systems.

It is important to note that the local-manual opening of one normally locked-shut valve will cross-connect the Unit 3 and the Unit I condensate storage header; the opening of a second euch valve will cross-connect all three condensate storage headers.49 Thus the Unit 1 ECCS Sys: ems ha-re a minimum of 511.0 and a mm< imum of 1419.4 m 3 (135,000 to 375,000 gallons) available through normally open valves from the Unit I storage tank and these limits can be increased three-fold by manually opening two normally locked-closed valves. Two additional condensate storage tanks, each of 1892.5 m3 (500,000 gallon) capacity, have recently been installed at the o Browns Ferry Nuclear Plant. Thus it is unlikely that HPCI System opera-tion will ever be limited by the availability of condensate storage header water.

- The HPCI booster pump suction will be automatically shif ted from the condensate storage heeder to the suppression pool ring header if either:

1. The water level above the booster pump suction f alls to an elevation of 168.4 m (551 ft). This would mean that the condensate storage tank had completely drained and there remained just sufficient water in the condensate storage header to effect a transfer before losing net posi-tive suction head (NPSH).

or

2. The pressure suppression pool level increases to an indicated level of 0.18 m (+7 in). Since the normal pool level is maintained between 0.05 and 0.15 m ( 2 and 6 in.),50 this implies the addition of between 257.4 and 371.0 ad (68,000 and 98,000 gallons) of water to the pool.

The 0.76 m (30 in.) diameter suppression pool ring header lies parallel to and beneath the suppression pool. Water flow from the suppression pool to the ring header is via four 0.76 m (30 in.) diameter downcomer pipes spaced at irregular intervals around the torus. Each downcomer is capped with strainers at the torus end.

f The change ir. HPCI pump suction lineup is accomplished by the opening of two DC-motor-operated valves in the line from the pressure suppression pool to the HPCI booster pump suction followed by the closing of the DC-motor-operated valve in the suction line from the condensate storage header. A check valve in the line from the suppression pool prevents backflow from the condensate storage tank into the suppression pool during the change.

l l

220 Two pecssure switches are used to determine the water head from the condensate storage tank above the booster pump suction and two level switches monitor the suppression pool level. In either case, just one of the two available signals is suf ficient to initiate the shif t in HPCI booster pump suction. Once the booster pump suction has been automati-cally shif ted to the pressure suppression pool, the operator cannot re-position the valves back to the condensate header suction lineup. ,

The minimum required net positive suction head (NPSH) for the HPCI booster pump is 6.40 m (21 f t). This requirement is easily satisfied when suction is taken from the condensate storage header which is at an eleva- i tion of approximately 8.69 m (281/2 f t) above the booster pump center-line.

When the HPCI booster pump suction is shif ted to the pressure sup-pression pool due to a high indicated pool level of 0.18 m (+7 in.), the pool water level is 4.04 m (13.25 f t) above the booster pump centerline.

Therefore, the NPSH requirement of 6.40 m (21 f t) can be met as long as the temperature of the pumped water is below 85.0*C (185*F), assuming no containment back pressure.

E.4 System Initiation Either of two signals will cause an automatic start of the HFCI Sys-tem. These are:

1.

2.

Lou reactor water level [12.09 m (476 in.) above vessel zerol.

• High drywell pressure [0.12 MPa (2 psig)].

With either of these conditions, both the suction valve to the con-densate storage header and the minimum flow bypass valve will open if they -

were closed. The pump discharge valve between the main HPCI pump and feedwater header "A" will open, and the two test line isolation valves will close if they were open. The steam supply valve to the turbine will open.*

The DC-motor-driven auxiliary oil pump starts and as the oil pressure increases, the turbine stop and control valves open. Above 1800 rpm, the main oil pump which is driven on the turbine shaf t takes over and main-tains the oil pressure while the auxiliary oil pump shuts down. The mini-mum flow bypass valve closes automatically when the increasing HPCI System pump flow exceeds 0.076 m3/s (1200 GPM).

The time from actuating signal to full flow is less than 25 seconds.

The 3

turbine control system will act to maintain a pumped flow of 0.315 m /s (5000 GPM) into the feedwater line over a reactor pressure range of from 1.14 to 7.83 MPa (150 to 1120 psig). If desired, the operator in the Control Room can operate the system in the manual or automatic mode at a dif ferent controlled flow.

The HPCl turbine operates at between 0.746 to 3.356 MW, (1000 and 4500 horsepower) with a steam demand of between 6.17 to 23.18 kg/s (49,000  %

and 1Ei,000 lbs/h). System steam and pumped water flows at operating con-ditions are g tven in tabular form on Fig. 6.4.1 of the Browns Ferry FSAR.

• The two normally open primary containment isolation valves in the steam supply line will not reopen, if closed.

221 E. 5 Turbine Trips On an HPCI turbine trip, the turbine stop valve closes aad the mini-mum flow bypass valve to the pressure suppression pool closes to preclude drainage from the condensate storage header into the suppression pool.

The following conditions will cause turbine trip:

g 1. High reactor vessel water level [14.78 m (582 in.) above vessel zero].

2. High HPCI turbine exhaust pressure [1.14 MPa (150 psig)].

3. Low HPCI booster pump suction pressure [-0.381 m (-15 in.) Hg].

4. HPCI turbine mechanical overspeed (5000 rpm).

5. Any HPCI isolation signal.

6. Remote manual trip from Control Room.

7. Manual trip lever on the HPCI turbine.

All turbine trips except high reactor water level and HPCI isolation will reset automatically when the initiating condition clears. The high reac-tor water level signal can be reset manually, or will reset automatically when the reactor water level decreases to the low reactor water level HPCI initiation point, 12.09 m (476 in.) above vessel zero. The HPCI isolation signal must be manually reset.

E.6 System Isolation The Primary Containment and Reactor Vessel Isolation Control System initiates automatic isolation of appropriate pipelines which penetrate the Y primary containment whenever certain monitored variables exceed their pre-selected op' ational limits. The system is designed so that, occe initi-ated, automatic isolation continues to completion. Return ation after isolation requires deliberate operator action.gg normal oper-An automatic isolation signal for the HPCI System causes the inboard and outboard HPCI steam line isolation valves to close, trips the HPCI turbine, and closes the two motor-operated valves in the suction line from the pressure suppression pool. The inboard staam line isolation valve is AC-motor-operated, the outboard DC-motor-operated. The maximum closing time for these valves is 20 s.

The following conditions cause a HPCI System isolation signal:52

1. HPCI System equipment space high temperature. Since high temperature in the vicinity of the steam supply line or other HPCI equipment could indicate a break in the turbine steam supply line, an isolation signal is generated if this temperature exceeds 93.3*C (200*F). This temper-ature is sensed by four sets of four bimetallic temperature switches.

These 16 temperature switches are arranged in four trip systems with four tcaperature switches in each trip system. The four temperature switches in each trip system are arranged in one-out-of-two taken twice logic.

2. HPCI turbine high steam flow. Since high steam flow could indicate a break in the turbine steam supply line, an isolation signal is gener-ated if the measured steam flow exceeds 150% of design maximum steady state flow. .The steam line ficw is sensed by two differential pres-sure switches which monitor the differential pressure across a mechan-ical element installed in the HPCI turbine steam pipeline. The trip-ping of either switch at a differential pressure cb 0.62 MPa (90 psi) initiates HPCI System isolation.

222

3. Low reactor pressure. After steam pressure has decreased to such a low value that the HPCI turbine cannot be operated, the _ steam line is isolated so that. steam and radioactive gases will not escape from the HPCI turbine shaf t seals into the reactor building. The steam pres-sure is sensed by four pressure. switches from the HPCI turbine steam-line upstream of the -isolation valves. The switches are arranged in a one-out-of-two taken twice logic.' The set point -for this isolation ,

signal is 0.793 MPa (100 psig).

4. High turbine exhaust diaphragm pressure. A line tapping off the tur-bine exhaust line contains two rupture diaphrams in series, with the A space between vented' to the HPCI' equipment space through a flow re-stricting orifice. The diaphrams are designed for 1.138 MPa (150' psig). If the pressure in the space between the. diaphrams exceeds

~ 0.172 MPa (10 psig), a system isolation signal is generated.

5. Manual Isolation. If a HPCI initiation signal is present, the opera-tor can cause HPCI system isolation by pushing a control panel push button.

The low reactor pressure isolation will be automatically reset if re-actor pressure is restored; all other isolation signals seal-in and the operator must puch the HPCI auto isolation circuit reset push-button af ter the condition has cleared.

E.7 Technical Specifications

1. .The HPCI System shall be operable .

a. Prior to startup from cold condition

b. Whenever there is irradiated fuel in the reactor vessel and the reactor vessel pressure is greater than 0.945 MPa (122 psig), es- .

cept

c. If the HPCI System is inoperable the reactor may remain in opera-tion for a period not to exceed 7. days provided ADS, core spray, LPCI mode of RHR and RCIC are all operable.

If these conditions'are not met, an orderly shutdown shall be initi-ated and the reactor vessel pressure reduced to 0.945 MPa (122 psig) or less within 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br />.

2. HPCI testing shall be performed as follows:

a. Simulated automatic actuation test -- once per operating cycle

b. Pump operability -- once per month

c. Motor operated valve operability - once per month

d. Flow rate at normal reactor operating pressure -- once per three months

e. Flow rate at 1.138 MPa (150 psig) --_ once per operating cycle

3. Whenever HPCI is required to be operable the piping from the pump dis-charge to the last flow blocking valve shall be filled. Water flow from the high point vent must be observed monthly. (Purpose is to i prevent water hammer when the system is started.) .

-. . . m ,

223 APPENDIX F A COMPENDIUM OF INFORMATION CONCERNING THE BROWNS FERRY UNIT 1 REACTOR CORE ISOLATION COOLING SYSTEM F.1 Purpose The Reactor Core Isolation Cooling (RCIC) System is designed to en-sure that the core is not uncovered in the event of loss of all AC power.

The RCIC System provides water to make up for that lost through steam gen-erated by decay heat during reactor isolation.53 RCIC is a consequence limiting system rather than an ECCS System; the system design is not pred-icated on any loss of structure accident.54 F.2 System Description The RCIC System (Fig. F-1) consists v. a steam-turbine driven pump unit and the associated piping and valves. The RCIC pump can take suction f rom either the condensate storage tank or the pressure suppression pool and discharges to the reactor vessel feedwater line "B" via a thermal sleeve connection. A full flow system test line permits testing of the RCIC System while the reactor is at power, with the pump discharge roc.ed to the condensate storage tank. A minimum flow line connects the RCIC

1. pump discharge to the pressure suppression pool, as a means of ensuring a flow of at least 0.004 m /s3 (60 GPM) through the pump. The minimum flow bypass valve closes at a system flow of greater than 0.008 m3 /s (120 Gnt) . The normal setpoint for RCIC pumped flow is 0.038 m 3/s (600 GPM).

The RCIC turbine is driven by steam extracted from main steam line "C" upstream of the main steam line isolation valve. The two primary con-tainment isolation valves in the steam line to the RCIC turbine are nor-mally open to keep the piping to the turbine at elevated temperatures so as to permit rapid system startup (within 30 seconds of an initiation sig-nal). The normally closed DC-motor-operated steam supply valve just up-stream of the RCIC turbine will open against full system pressure within 15 seconds af ter receipt of a system initiation signal.

The RCIC turbine exhaust steam is discharged to the pressure suppres-sion pool. The turbine gland seals are drained to a barometric condenser, in which the steam is condensed by a water spray. The water spray is pro-vided by a flow diverted from the RCIC pump discharge to pass through the turbine lube oil cooler and subsequently form the spray. The condensate from the barometric condenser is pumped back to the RCIC pump suction. A vacuum pump removes the non-condensibles from the barometric condenser and inserts them into the pressure suppression pool.

A vacuum breaker line (not shown on Fig. F-1) is installed between the torus airspace and the RCIC turbine exhaust line. Its purpose is to prevent pressure suppression pool water from being drawn up into the tur-

. bine exhaust line when the remaining exhaust steam condenses af ter turbine operation and shutdown.

All components normally required for initiating operation of the RCIC System are completely independent of AC power, plant service air, and ex-ternal cooling water systems, requiring only DC power from a unit battery I

224

'ORNL-DWG81--8502 ETD '

' CONDENSATE e

STORAGE , .

TANK e

ELEVATION 577 ft NORMAL WATER LEVEL gEVATION 625 ft ,

MAIN STEAM

() CONDENSATE RETURN CONDENSATE SUPPLY (, ,) j, HEADER HEADER.

THERMAL-EVAbf4 5Y4 f:

.FE E DWATE R SUPPRESSION POOL I- --l WATER LEVEL gg E LEVATION 536 ft '

D Q

SUPPRESSION 2:N_ : - a CHAMBER - TURBINE],

HEADER -

}NMP - _ ~,

PUMP tC ELEVATIOE523 ft FF FROM TURBINE OIL -

, MINIMUM FLOW LINE UM CONDENSER ' \

CONDENSATE -VACUUM

[

_ SYSTEM TEST LINE , _

-PUMP TANK'

-N?

TO SUPPRESSION POOL'

• Fig. F.1 Reactor core isolation cooling system.

to operate the valves, vacuum pump, and condensate pump.55 On loss of control air, the RCIC drain lines to the main condensers will fail closed; this is their normal position when the RCIC System is in operation. The drain functions-of these valves is transferred to overseat drain ports in .

the turbine stop valves.

The principal RCIC equipment is installed in the reactor building at -

a level below that of the~ pressure suppression pool. The turbine pump .

assembly is located in a shielded area so that access to adjacent areas of the reactor building lis not restricted during RCIC System operation. The only operating component located within the Primary Containment is the l

. .- . -, . . . . -- - ., ~

, t 225

{

i normally-open AC-motor-operated RCIC inboard steam line isolation valve, s which will u main open on loss of power.-

i

F.3 RCIC Pump Suction t

The RCIC pump can take suction either from the condensate storage

_$_ header throuth a single normally-open DC-motor-operated -valve or ~ from the pressure suppression pool via the suppression pool ring header.

Each Browns Ferry Unit'is provided with a 1419.4 m3. (375,000 gal-

lon) condensate storage tank, which provides a water head to. the storage

, header for that unit. The storage header, which taps into the bottom of i the storaEe tank, feeds the suctions of the high pressure ECCS Systems,

specifically, the pumps for the HPCI and the RCIC systems. . The' core _ spray pumps and the RHR pumps can be fed from this -source, but are not .normally-

3. aligned to it. All other demand for condensate storage tank water is fed' via a standpipe within the tank;' the standpipe height is such that 511.01 m3 (135,000 gallons) of water is reserved for'the ECCS Systems.-

[,

There is no provision for an automatic shif ting of the RCIC pump suc-1 tion from the condensate storage header to the pressure suppression pool.

. However, if. condensate storage tank water is- unavailable for any reason,

! the control room operator can- shif t the RCIC pump suction to the pressure j suppression pool ring header. This is done by remote-manually opening the

two DC-motor-operated suction valves to. the pool ring header; when these tne valves are fully open, the suction valve to the condensate storage.

-1

~

header will automatically close.

The minimum required net positive ' suction head (NPSH) for.the RCIC j System is 6.10 m (20 f t), which is readily available with suction taken

' *- from the condensate storage tank. With suction from the pressure suppres-

• l -sion pool, the ~ required NPSH is available for suppression pool tempera-tures up to 85.0*C (185'F) with no containment back pressure.

~

F.4 System Initiation I

l An automatic start of:the RCIC' System is initiated by low reactor vessel' water level at 12.10 m (476.5 in'.) ~ above vessel zero. The single normally-closed valve in the steam supply line opens, and steam is ad-

~

mitted to the turbine. (The turbine stop valve and. control valve are nor-mally open when the RCIC System is in standby.) The barometric condenser vacuum pump starts, and the condensate pump will act to automatically con .

trol the water level in the condenser. The RCIC turbine speed and pump flow are automatically maintained by the steam flow controller. Turbine lube oil is supplied by a shaf t-driven oil pump. The single normally-closed valve in the pump discharge line to feedwater header "B" automati-cally opens,- and the minimum flow bypass valve closes automatically when 4 pump flow exceeds 0.008 m3 /s (120 GPM).

If the RCIC System is in an abnormal lineup when an initiation signal is received, the system will realign: .

. a. If the normally open ' pump discharge valve is closed, it will open.

b. If the normally open pump suction valve from the condensate storage header is closed, it will open provid da at. least one of the suppres-sion pool suction valves is not fully open.

226

c. If the normally closed valve in the system full flow test line, which returns RCIC pump discharge water to the pressure suppression pool, is open, it will close.

d. If the system logic mode is in test, it will automatically switch to automatic flow control to maintain turbine speed and pump flow.

The time from actuating signal to full flow is less than 30 seconds.

The turbine control valve is governed by a speed controller which compares ,

the measured (tachometer) turbine speed to a speed demand provided by the flow controller. In the automatic mode, the speed demand is established by camparison of the measured pump discharge flow to the set point signal,

• 3 which, though variable, is normally set at 0.038 m /s (600 GPM). In the manual mode, the operator positions a potentiometer to produce a direct speed demand signal to the speed controller.

The RCIC System is designed to provide a full flow of 0.038 m 3 /s (600 GPM) at reactor pressures from 1.138 to 7.826 MPa (165 to 1135 psia).

With the reactor vessel at the higher pressure, the RCIC turbine delivers approximately 0.373 MW (500 horsepower) with a steam demand of 3.53 kg/s (28,000 lbs/h). With the reactor vessel at 1.138 MPa (165 psia), the tur-bine delivers approximately 0.060 MW (80 horsepower) with a steam demand of 0.96 kg/s (7600 lbs/h).56 In addition to panel S-3 in the Control Room, a RCIC pump flow con-troller is also located at the remate shutdown panel for use in emergen-cies when the Control Room is not available.

F.5 Turbine Trips .

On an RCIC turbine trip, the turbine trip throttle valve (not shown on Fig. 1) closes. The valve in the steam supply line which was opened by the RCIC initiation signal remains open, as does the discharge valve to feedwater header "B". Thus, the RCIC System remains lined up for injec-tion with the exception of the turbine trip throttle valve.

The following conditions will cause the trip throttle valve to close:

1. High reactor vessel water levet (14.78 m (582 in.) above vessel zero].

2. Electrical overspeed at 110% rated speed.

3. Mechanical overspeed at 125% rated speed.

4. High RCIC turbine exhaust pressure [0.276 MPa (25 psig)].

5. Low RCIC pump suction pressure IO.381 m (15 in.) Hg vacuum].

6. Any automatic isolation signal.

7. Remote manual trip f rom the control room.

8. Local manual trip lever.

All turbine trips except the mechanical overspeed operate by de ener-gizing a solenoid valve which dumps oil, allowing a spring to close the

• rip throttle valve.

. When the condition causing the trip clears, the sol-noid is re energized ; howeve r, the valve must be ~anually reopened. The control room operator can reopen the trip throttle valve by running the

• motor operator to the "close" position which relatches the trip valve to the solenoid. The valve can then be opened by running the motor operator to " ope n". 54 ,

If the throttle valve is tripped by the mechanical overspeed trip, it must be manually reset locally at the turbine.57

227 F.6 System Isolation The Primary Containment and Reactor Vessel Isolation Control System iaitiates automatic isolation of appropriate pipelines which penetrate the primary containment whenever certain monitored variables exceed their pre-determined operational limits. The system is designed so that once initi-ated, automatic isolation goes to completion. Return to normal operation

] after isolation requires deliberate operator action.

An automatic isolation for the RCIC Syst em causes the normally-open

, inboard and outboard steam supply isolation valves to close, the turbine to trip, and the two RCIC suction valves from the pressure suppression pool to close (if they were open). The minimum flow bypass valve to the suppression pool is interlocked to close whenever the turbine is-tripped.

The following conditions cause a RCIC system isolation signal:

1. RCIC System equipment space high temperature. Since high temperature in the vicinity of the RCIC equipment could indicate a break in the turbine ste am supply line, an isolation signal is generated if this temperature exceeds 93.3*C (200*F). This setpoint is based on the calculated AT with a 0.001 m /s3 (15 GPM) steam leak in the space.

There are 16 temperature sensors arranged in four trip logics with four sensors in each logic. The 16 sensors are physically arranged in four groups with four sensors in each group. One sensor in each group is in each of the four one-out-of-two taken twice trip logics.

2. RCIC System high steam flow. Since high steam flow could indicate a leak in the turbine steam supply line, an isolation signal is gener-1 ated if the measured steam flow exceeds 150% of design maximum steady state flow. The steamline flow is sensed by two differential pressure

, switches which monitor the differential pressure across an elbow in-stalled in the RCIC turbine steam supply pipeline. The tripping of either trip channel at a differential pressure of 11.43 m (450 in.)

H 2O initiates RCIC System isolation.

3. Low reactor pressure. Af ter steam pressure has decreased to such a low value that the RCIC turbine cannot be operated, and there is no cool-ant spray to the barometric condenser, the steam line is isolated to protect against continuous gland seal leakage to the RCIC equipment space. The set point for this isolation signal is a reactor vessel pressure of 0.345 MPa (50 psig). The pressure is sensed by four pres-sure switches at the RCIC turbine steam line upstream of the isolation valve s. The switches are arranged in a one-out-of-two taken twice logic.

4. High turbine exhaust diaphragm pressure. A line tapping off the tur-bine exhaust line contains two rupture diaphrams in series, with the space between vented to the RCIC equipment space through a flow re-stricting orifice. The diaphrams are designed for 1.138 MPa (150 psig). If the pressure in the space between the diaphragms exceeds c 0.172 MPa (10 psig), a system isolation signal is generated.

5. Manual Isolation. If an RCIC initiation signal is present, the oper-ator can cause RCIC System isolation by pushing a control panel push-

. button.

All isolation signals are sealed in and must be manually reset after the condition causing them has cleared. A control panel pushbutton is provided for this purpose.

228 F.7 Technical Specifications

1. The RCIC System must be operable prior to startup from a cold condi-tion or whenever there is irradiated fuel in the reactor and the re-actor vessel pressure is above 0.945 MPa (122 psig).

If the RCIC System is inoperable, the reactor may remain in operation for a period not to exceed seven days if the HPCI System is operable I during such time.

If these conditions are not met, an orderly shutdown of the reactor i must be initiated and the reactor depressurized to less than 0.945 MPa (122 psig) within 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br />.

2. RCIC testing shall be performed as follows

a. Simulated automatic actuation test -- once per operating cycle

b. Pump operability - once per month

c. Motor -operated valve operability -- once per month

d. Flow rate at normal reactor operating pressure - once per three months

e. Flow rate at 1.138 MPa (150 psig) -- once per operating cycle

3. Whenever RCIC is required to be operable the piping from the pump dis-charge to the last flow blocking valve shall be filled. Water flow from the high point vent must be observed monthly. (Purpose is to prevent water hammer when the system is started.)

t b

o i

J

__ _ , , _ _ _ - , , _ . . < . - . . . - - , , , _ . _ . _ , - . 7

229 APPENDIX G EFFECT OF TVA-ESTIMATED SEVEN HOUR BATTERY LIFE ON NORMAL RECOVERY CALCULATIONS Section 7, Computet Prediction of Thermal Hydraulic Parameters for a, Normal Recovery, is based on the assumption that the 250 vde unit batter-ies will f ail in four hours under the conditions of Station Blackout.

Af ter review of the draf t results presented in Sect. 7, the Electrical En-gineering Branch at TVA performed a battery capacity calculation to deter-mine how long the Browns Ferry 250 vde batteries would last under these conditions, and arrived at an estimate of seven hours. This appendix ex-amines the impact of the newiv estimated seven hour battery failure time on the results and conclusi< s of Sect. 7.

Two basic conclusions are presented in Sect. 7.

1. If AC power is recovered any time in the first five hours, then a normal recovery will be possible if the 250 vde batteries have not failed.

2. If the 250 vde batteries f ail at four hours and the AC power remains unavailable then the time between battery failure and first uncovering of fuel will be at least three hours if the re-actor has bean previously depressurized as recommended by this re por t.

If the unit batteries were to last seven instead of four hours, there A

would be a slight lengthening of the time interval from battery failure to core uncovery due to the slightly lower decay heat. The most important question to be answered regarding the extension of the estimated battery failure time from four to seven hours is whether there is some other sys-tem failure caused by the increased temperatures and/or pressures during this period that would complicate or make impossible a normal recovery if AC power were restored.

To answer this question calculations similar to those shown in Figs.

7.1 through 7. 9 of Sect. 7 were performed. These new calculations start at an initial point four hours after station blackout and extend to seven hours. The batteries are assumed to last throughout this period and oper-ator actions are the same as detailed in Sect. 7.3.1, Normal Recovery-Assumptions.

Due to the elevated pool temperatures experienced during the period from four to seven hours, it has been necessary to modify the calculation of the fraction of relief valve (SRV) discharge that is quenched in the suppression pool. When the pool temperature is within its normal opera-tional range, the SRV discharge is completely condensed (quenched) in the cool water surrounding the submerged SRV discharge nozzle (T-quenche r) .

However, if the temperature of the water around the T quencher is suf fi-ciently close to saturation, then the condensation will be less than 100%

and some of the steam will reach the suppression pool atmosphere. Monti-cello tests 58 showed that an approximately 28*C (50*F) dif ference be-tween bulk and local pool temperature can exist during extended discharge through a single SRV. In this case, one would expect less than 100%

quenching to begin at a bulk pool temperature of about 74*C (165'F), cor-responding to 100*C (212*F) near the SRV discharge point (provided the

230 pool is at atmospheric pressure). For the purposes of the calculations reported in this appendix, the following provisions were made for the cal-culation of quench fraction in the pool:

1. The temperature of the water surrounding the T quencher during SRV discharge is assumed to be 28'C (50*F) higher than the bulk pool temperature.

2. If the vapor pressure of the water surrounding the T quencher is equal to or greater than the static pressure then 0% of the SRV .

discharge is quenched; if the vapor pressure is 5.0 psi or more below the static pressure then 100% of the discharge is ,

quenched. Variation of quench fraction is linear between these two points.

This should provide a reasonable urper limit estimate of containment pres-surization due to non quenched SRV discharge. Consideration of this ef-feet was not necessary for the zero to five hour results reported in Sect.

7 because bulk pool temperature does not exceed 82*C (180*F) during the first five hours of a Station Blackout.

Results of the four to seven hours af ter blackout calculations are shown in Figs G.1 through G.9. From these results it is concluded that system parameters remain within acceptable ranges during this period:

1. The 250 ydc batteries by TVA estimate last the full seven hours.

2. Reactor vessel level is within the normal control range, about 5.08 m (200 in.) above the top of active fuel.

3. Reactor vessel pressure is being controlled by operator action at about 0.69 MPa (100 psia).

4. At the seven-hour point, about 414 m 3(109,265 gal) of water '

have been pumped from the condensate storage tank, which had an assured capacity of 511 m3 (135,000 gal) before the blackout.

5. At the seven-hour point, average suppression pool temperature is about 92*C (198.2 *F), but this should not be a problem for the T quencher type of SRV discharge header piping.

6. Containment pressures would be elevated to about 0.23 MPa (33 psia), well below the 0.53 MPa (76.5 psia) design pressure.

7. Drywell atmosphere temperature is about equal to the 138*C (281*F) design temperature.

These results show that if power were recovered within seven hours of the Station Blackout a normal recovery would be possible.* The 92*C (198'F) bulk pool temperature af ter seven hours may cause elevated air temperature in the RCIC and HPCI spaces, but this would not be expected to lead to failure of the HPCI or RCIC, because the lube oil of both units is cooled by the water being pumped -- in this case, water from the conden-sate atorage tank which would not exceed 32*C (90*F). The periods of RCIC operation are very infrequent after the first four hours of Station Black-out, with about an hour between actuations.

The supply of control air for remote-manual operation of the SRVs is sufficient for the full seven hours. As discussed in Chap. 3 of this re-port, the accumulators provided for the six relief valves associated with the ADS system are sized to permit five operations or a total of 30 actua-tions. As illustrated in Sect. 7 and this appendix, less than 30 actua-tions are required during the first seven hours of the Station Blackout.

• With the assumption of no independent secondary equipment failures.

ZET C

q / B.

8 4

b s M t- E O 3 O N

lTP 4

e u

• 5 g O 2 D C

> m o z w 8

2 m

Q m m g-w 4 m W h--- y ,

O- -y m m ,_. 5 O m O n q J m 3 C M O z c g

> m L' \ Z o m

3 m

< / O -

$ >H6 -

I O

4 o.

a 8_!

m m

M m E c

^q b m, O

- n D

e )I H H' D(

z o m

a O m y m o Z 4 m n s 8 m a

g g-L  ? n g ,

n @

W .

1. 9 o C m "c z- n' b N f

• n m < g omH 3 a o nc y

-8 o W 3 m MO W

n N m 009 008 007 006 005 004 003 002 001 0 0011 0001

)aisp( ERUSSERP MAETS

, , , , , i _t i

i 6 5 4 3 2 1 0 8 7 e

)aPM( ERUSSERP MAETS l

l I

ORN L-DWG 81-8594 ETD S_ _

w 8- W 8- $'

8- N-

- _~

6 o.

3" =o3 8. -

o

$ 8- $

2 2 g.

I 6 6" a s- m 5 d d 8-N w S- m N~

N-

g. -RCIC TURBINE STEAM FLOW 81 - -NO STEAM FLOW FROM REACTOR VESSEL

g. N-

\ -

A

+1 SRV OPEN-> < 1 SRV OPEN - "

1 SRV OPEN  :

U u l l o- o .' . , , ,

210 270 300 330 360 390 420 TIME (min)

Fig. G.2 Vessel steam flow - four to seven hours after station blackout. l l

i q Q O 6

. _ _ _ _ - - . - . _ - - _ . = - . . . . -- . - . --..._c n

. s . 4 .-

.4 ORNL-DWG 81-8595 ETD 8

s s.

8

• TOP OF CONTROL ROOM INDICATION OVERSHOOT DUE TO HEATUP OF SUBCOOLED

~ WATER AFTER RCIC TURNED OFF g

/

m- U SRV SHUI S-h SRV SHUTS

. g N /

h* V h/ -RCIC OFF

_ _ _ _ _ _ . _ _ _ . . . _ _ ...____........ _ _._n .._______.. ___ ......____. .___....._____..

a a ..__h....

m m 3 "- -RCIC ON a

2

@3 a

.2 n

y BOTTOM OF CONTROL ROOM INDICATION U ha_

w w

> >g OPERATOR CONTROLS LEVEL BY REMOTE-MANUAL ACTUATION OF RCIC SYSTEM 8

e_

TOP OF ACTIVE FUEL

m. g n

m.

R 330 360 390 120 240 "70

,. 300 TIME (min)

Fig. G.3 Vessel level - four to seven hours after station blackout.

ORNL-DWG 81-8596 ETD M-M-

M- .

2-M-

0 $~ g n'

5. M- ~ . ~

@n- -

x S 3

3 8- u.g-OJ *-

g

u. O y m

aw w.

& g s-S b_

-- E 3_ n_ M-S-

e-RCIC RCIC

._ $- ON OFF ON OFF RCIC ON-w-

t n-240 2f0 $ 30 3b #

TIME (min)

Fig. C.4 Injected flow - four to seven hours af ter station blackout.

e an e

U*

0 D 2 4

T E /

7 9

5 j 8 1

8 G

W D 0 e

- 9 t L 3 a N s R n O e F d F

O n.

o ct CI u C mo ok R

0 rc

- 4 6 f a L - 3 l N Cb I

O Cn C Ro I i C yt R ba

/ n -

0(

)

n im ds e

pr t

3 3 E me ut MI pf T ( a ws or l u

? f o h

d 0

en t e 0

F F

3 cv ee O js C n I i o C t R l ar t u oo N Tf O 0 7 -

C I

2 C

R

5. )k Gn

- a

^ g

.t ie Fg

- a r

0 4

o 2

t 0

s

l. . e + k

}E23d>0 8 u 2 a< OH

- ~ ~

~

~

g b k g @ 8 8 O h

}c[23y> 0$82- < Og t

i

ORN L-OWG 81-8608 ETD E

- a

/

/

i -

-RCIC TURBINE EXHAUST ONLY bS X J g -RAPID INCREASE DUE TO QUENCHING OF g m- a SRV DISCHARGE AND RCIC TURBINE J J EXHAUST IF OPERATING 8

8

~

z g

- u m

9 ~~ h m

E

a. n I 3 8

o-

?

T-

'?

240 270 300 333 360 390 120 TIME (min)

Fig. G.6 Suppression pool level - four to seven hours after station blackout.

i

, , a , *' *

• L . o , ,

ORNL-DWG 81-8599 ETD k

o g SUPPRESSION CHAMBER

" ATMOSPHERE RAPID INCREASE DUE TO DIRECT MIXING WITH UNQUENCHED .

,o g. o' o STEAM FROM SRV DISCHARGE g p SRV CLOSED }

d O m

! m h- h

<8 <g SRV CLOSED -- - SRV OPENED _-

g g- f  % y a 2 2 w w u up -SUPPRESSION POOL g e o, #

~

SRV OPENED

$R- I d o o g LESS THAN 100%, QUENCHING-y y~ OF SRV DISCHARGE BEGINS /-CONDENSATION AND HEAT TRANSFER FROM ATMOSPHERE ARE DOMINANT

@8- @ m WHEN SRV NOT DISCHARGING e

& &B 3 g_ b B-

?-

8 390 120 R- 240 270 300 330 360 TIME (min)

Fig. G.7 Suppression chamber temperatures - four to seven hours after station blackout.

l l

l l

l

ORNL-DWG 81-8600 ETD 8'

g.

a n

G E o O.

y y DRYWELL DESIGN TEMPERATURE E

<~a. <M Ea /

m m l E E ............... ............... ................ .......... ... . . . . . . . . . . . . . . . ....__ ___

g 3

g w . -

7 l w w I

Ek I TEMPORARY DECREASE DUE TO DECREASE IN N h h DRYWELL ATMOSPHERE TEMPERATURE REACTOR COOLANT TEMPERATURE (CAUSED $

g g INCREASES GRADUALLY TOWARD BY DECREASE IN REACTOR PRESSURE -

< ~g, < REACTOR COOLANT TEMPERATURE SEE FIG. G.1) d w w dN 3 3 m m O O 8

8-8 240 270 300 330 360 390 420 TIME (min)

Fig C 8 Drywell temperatures - four to seven hours after station blackout.

- r

, e

r wue D 0 2

T 4 E

1 0

6 8

1 8

S G N W I G D GN 0 EI r L

t:

v B CH 9 3

t e

n P N f o U E a DU s IL Q S r U FN u B OIG o E h

%E R

U0 0B n 0 e S1 E 6 S G 3 ve EN R s S RA A E

O r PHT HC o

- t G

N7

) DI P SISS E AE D r RI OG RL V )

n u

UT . o SA G ENR im f SS IF RES, 0(

EN OHF 3 -

REE MWO 3 E PDE M s c I e

R N (S T r EO R u BC E s MFF s AOS N e A

H r p

C E S A NU R T 0 t n

OACT 0 3 e I

S R S EA m

E BE E n B i RN H M a PWD P A t n

_ U ON H o .

SDA C Ct N u O o I

S L L

0 7 9. kc S

E E 2 Ga l

_ R P

W g

.b Y

P R in U

S D Fo i

- t a

t s

U-0 4

_ 2 S R a ,R E o E

_=.a mEhe Hzwl(FZ8 . -

3 h N D' g x

{8* E g :an.gz $ Z :45o a

! , i I I  ;

! )' <

241 NUREC/CR-2182, Vol. T ORNL/NUREC/TM-455/Vf Dist. Category RX, 1S Internal Distribution 3 1. W. J. Armento 17. A. L. Lotts

2. T. E. Cole 18. T. W. Robinson, Jr.

> 3. W. A. Condon 19. H. E. Trammell

4. D. H. Cook 20. B. J. Veazie

5. W. B. Cottrell 21. C. F. Weber

6. W. G. Craddick 22. R. P. Wichner

7. G. F. Flanagan 23. D. D. Yue

8. S. R. Greene 24. Patent Office

9. R. M. Harrington 25. Central Research Library 10-14. S. A. Hodge 26. Document Reference Section

15. T. S. Kress 27-28. Laboratory Records Department

16. R. A. Lorenz 29. Laboratory Records (RC)

External Distribution 30-31. Director, Division of Accident Evaluation, Nuclear Fegulatory 9 Commission, Washington, DC 20555 32-33. Chief, Severe Accident Assessment Branch, Nuclear Regulatory Com-mission, Washington, DC 20555

34. Director, Reactor Division, DOE, ORO, Oak Ridge, TN 37830

35. Office of Assistant Manager for Energy Research and Development, DOE, ORO, Oak Ridge, TN 37830 36-40. Director, Reactor Safety Research Coordination Office, DOE, Washington, DC 20555 i

41-42. D. B. Simpson, Tennessee Valley Authority, W10C126 C-K, 400 Com-i merce Avenue, Knoxville, TN 37902 l

43-44. Wang Lau, Tennessee Valley Authority, W10C126 C-K, 400 Commerce Avenue, Knoxville, TN 37902

45-46. R. F. Christie, Tennessee Valley Authority, W10C125 C-K, 400 Com-l merce Avenue, Knoxville, TN 37902 l 47-48. J. A. Raulston, Tennessee Valley Authority, W10C126 C-K, 400 Com-merce Avenue, Knoxville, TN 37902 49-50. H. L. Jones, Tennessee Valley Authority, 253 HBB-K, 400 Commerce Avenue, Knoxville, TN 37902 51-52. Technical Information Center, DOE, Oak Ridge, TN 37830 53-552. Given distribution as shown under categories RX, 1S (NTIS-10)

+

l j *U.S. GOVERNMENT PRINTING OFFICE: 1981-7 4 082/291 l

T 2 ANRXIS 120555064215 g

US NRCADM DOCUMENT CONTROL DESK PDR 20555 DC HINGTON e

_J