ML20080U493

Evaluation of Natural Circulation Cooldown Test Performed at San Onofre Nuclear Generating Station

ML20080U493
Person / Time
Site:	Waterford, San Onofre, 05000000
Issue date:	01/31/1984
From:	ABB COMBUSTION ENGINEERING NUCLEAR FUEL (FORMERLY
To:
Shared Package
ML20080U490	List: ML20080U493 W3P84-0505, Forwards Evaluation of Natural Circulation Cooldown Test Performed at San Onofre Nuclear Generating Station. Boron Mixing Tests in Lieu of Natural Circulation Test Accepted by Updated SER & Sser 5
References
CEN-259, TAC-62848, TAC-62849, NUDOCS 8403020335
Download: ML20080U493 (136)

Text

{{#Wiki_filter:. CEN-259 AN EVALUATION OF THE NATURAL CIRCULATION COOLDOWN TEST PERFORMED AT THE SAN ONOFRE NUCLEAR GENERATING STATION COMPLIANCE WITH THE TESTING REQUIREMENTS OF BRANCH TECHNICAL POSITION RSB 5-1 Prepared for the COMBUSTION ENGINEERING OWNERS GROUP NUCLEAR pot ^tER SYSTEMS DIVISION E d S@ETEMS COMBUSTION ENGINEERING. INC. c?^* ?*$d

LEGM. NOTICE THIS REPORT WAS PREPARED AS AN ACCOUNT OF WORK SPON80 RED EY COASUBTION ENGINEERING, INC. NEITHER COMEUSTION ENGINEERING NOR ANY PERSON ACTING ON ITS SEHAI.P: A. MAKES ANY WARRANTY OR REPRESENTATION, EXPRESS OR lRFLIED INCLUDING THE WARRANTIES OF FITNESS FOR A PARTICULAR PURPOSE OR MERCHANTAtiLITY, WITH RESPECT TO THE ACCURACY, COtrLETENEffi, OR USEPULNESS OF THE INFORMATION CONTAINED IN THIS REPORT, OR THAT THE USE OF ANY INFORMATION, APPARATUS, METHCO, OR PROCESS DISCLOSED IN THIS REPORT MAY NOT INFRINGE MllVATILY OWNED RIGHTS,OR i - E. ASSUPES ANY LIASILITIES WITH RESPECT TO THE USE OF, OR FOR DAMAGES RESULTING PROM THE USE OF, ANY INFORMATION, APPARATUS, METHOO OR PROCESS Dem twm IN THIS REPORT. O b f a f a w-- ~~---r wm -m-n,,-w, ,,s,--m, ,- -. --,-a-.-,w,,-m,,--ww-:~-ep y-gen,-,,.,-a,wwm wws- ,._ww,, g

CEN-259 AN EVALUATION OF THE -NATURAL CIRCULATION C00LDOWN TEST PERFORMED AT THE SAN ONOFRE NUCLEAR GENERATING STATION [- COMPLIANCE WITH THE IESTING REQUIREMENTS OF BRANCH IECHNICAL POSITION RSB 5-1 PREPARED FOR THE COMBUSTION ENGINEERING OWNERS GROUP l l I NUCLEAR POWER SYSTEMS DIVISION JANUARY, 1984 COMBUSTION ENGINEERING, INC. i

.= p ABSTRACT This report provides an evaluation of sel*ected portions of the Natural Circula-tion Test pe-formed at Unit 2 of the San Onofre Nuclear Generating Station (SONGS) on '27 -July 1983. Results from this evaluation are presented which show that SONGS Units 2 and 3 comply with the requirements of Branch Technical Position RS8 5-1. In addition, the applicability of the SONGS test results to Waterford Unit 3 and to St. Lucie Unit 2 are confirmed. Finally, as a result of the SONGS data evaluation, information specific to the Combustion Engineer-ing designed System' 80 nuclear steam supply system is presented which can be referenced as part of a pre-test report for the boron mixing and natural cir-culation cooldown tests to be performed at Unit 1 of the Palo Verde Nuclear GInerating Station. l c 11

9 LIST OF ACRONYMS AND ABBREVIATIONS ADV Atmospheric dump valvt i AFWS- .- Auxiliary feedwater system - 'APS Arizona Public Service BTP Branch Technical Position C-E Combustion Engineering CEDM Control element drive mechanism CET Core exit thermocouple CPC . Core protection calculator CST Condensate storage tank CVCS Chemical and volume control system DNBR Departure from nucleate boiling ratio. EFWS Emergency feedwater system FAP Fuel alignment plate FP&L Florida Power & Light HFP Hot full power HZP Hot zero power HJTC Heated junction thermocouple HPSI High pressure safety injection LP&L-Louisiana Power & Light MSIV Main steam isolation. valve l NRC Nuclear Regulatory Comission l. NSSS Nuclear steam supply system FCV Presure control valve ~PLCS Pressurizer level control system j' ppm -Part per million -P Pressurizer pressure pzr. P/T Pressure / temperature PTS Pressurized thermal shock PZR Pressurizer RCP Reactor coolant pump RCS. . Reactor coolant system - RSB Reactor Systems Branch iii

RTD Resistance temperature detector RTP Rated thermal power RVGVS Reactor vessel gas vent system RVLMS Reactor vessel level monitoring system RVUH Reactor vessel upper head SBCS . Steam bypass control system SCE Southern California Edison SCS Shutdown cooling system SG Steam generator SI Safety injection SL1 St. Lucie Unit 1 SONGS San Onofre Nuclear Generating Station Tcold Cold leg temperature That Hot leg temperature T Pressurizer. temperature pzr rvuh Reactor vessel upper head temperature T UGS Upper guide structure VCT Volume control tank i l iv

L TABLE OF CONTENTS Secti on'. Title Page

1.0 INTRODUCTION

1 1.1 Purpose 1 1.2-Scope 2 1.3. -Background 2 2.0

SUMMARY

OF SONGS NATURAL CIRCULATION TEST 4 [ 2.1 Test Method. 4 2.2 -Summary of Test Results 9 3.0-BORON MIXING ANALYSIS 21 3.1 .Introouction 21 3.2 Boron Mixing Process 21 3.2.1 . Boron Delivery 22 3.2.2 Boron Dispersion in the RCS 25 I 3.2.3 Effect of Charging Fluid Temperature 32 L 3.3 l Boron Mixing Test Results 32 3.4 Conclusions 37 4.0' REACTIVITY ANALYSIS 51 l-II 4.1 Introduction 51 4.2 Conclusions 51 V . ~..--

TABLE OF CONTENTS (cont.) Section Title Page 5.0~ REACTOR VESSEL UPPER HEAD C00LDOWN ANALYSIS 54 5.1 Introduction 54 5.2 Analytical Results 55 5.2.1 Methodology 55 5.2.2 RVUH Temperature During the SONGS Natural 57 Circulation Test 5.2.3 3410 Class RVUH Cooldown 59 5.2.4 St. Lucie 2 RVUH Cooldown 59 5.2.5 System 80 RVUH Cooldown 50 5.3 Conclusions 61 6.0-COMPLIANCE WITH THE TESTING REQUIREMENTS 70 0F BTP RSB 5-1 6.1 Introduction 70 6.2 Compliance by the San Onofre Nuclear Generating 72 Stat-ion Units 2 and 3 6.3' Applicability of SONGS Test to Waterford 3 87 6.3.1 Introduction. 87 l 6.3.2 Comparison with SONGS Test I :.<ults 90 6.3.3 Conclusions 91 6.4 Applicability of SONGS Test to St. Lucie Unit 2 92 6.4.1 Introduction 92 .6.4.2 Comparison with SONGS Test Results 94 6.4.3 Conclusions 96 6.5' Compliance by the Palo Verde Nuclear Generating 98 Station Units 1, 2, and 3. vi w , - ~

TABLE OF CONTENTS (cont.) 1 ~ Section ,-Title Page 7.0 GENERIC OPERATOR GUICELINES 104 7.1 Introduction 104 7.2 Boron Mixing During N&tvral Circulation 104 7.2.1 Operator Actions 104 7.2.2 Bases for Operator Actions 106 7.3 Reactor Vessel Upper Head Cooldown During 108 Natural Circulation 7.3.1 Operator Actions 108 7.3.2. Bases for Operator Actions 112

8.0 REFERENCES

125 O vii

d

1.0 INTRODUCTION

1.1_ Purpose This report provides an evaluation of selected portions of the Natural Circulation Test performed at Unit 2 of the San'Onofre Nuclear Generating Station on 27 July 1983. Specifically, as requested by the NRC Staff and as stated in Reference 2, evaluations are included which deconstrate the ability to mix boron under natural circulation conditions with an estima-tion of the times required to achieve this mixing, an evaluation of RVUH cooldown rates both with CEDM cooling fans in operation and with CEDM cool-ing fans secured, an evaluation of feedwater usage with an assessment of the adequacy of the seismic Category I condensate supply, and an evalua-tion of the adequacy of.the safety-grade nitrogen supply to the atmospher-ic dump valves.- Generic operator guidelines are presented based upon the ~ SONGS data evaluation which are intended to enhance operator response to boron mixing during natural circulation and reactor vessel upper head cool-down during natural circulation. In addition, an evaluation is given of the effects on reactivity control of steam bubble formation in the reactor . vessel upper head. Information is presented based upon the San Onofre Natural Circulation Test as well as certain portions of cther tests in the San Onofre Natural Circulation Test Program which demonstrate that SONGS tMits 2 and 3 comply with the requirements of Branch Technical Position RSB 5-1. As such the l. information presented in this report, as well as information presented in l . Reference 12, satisfy the NRC Staff's requirement for a detailed post-test submittal following the San Onofre Natural Circulation Test of July 27. Since Unit 2 of the San Onofre Nuclear Generating Station is prototypical of pre-System 80. plants with respect to the requirements of 27P RSB 5-1, information'is presented whicn demonstrates applicability of the SONGS test.results to Waterford Unit '3 and to St. Lucie Unit 2. With-respect to the Combustion Engineering System 80 design, several differences exist from the pre-System 80 NSSS (particularly with respect . to _the size of the reactor vessel upper head) such that Arizona Public l-1

i Service plans to conduct their own boron mixing and natural circulation cooldown tests-on Palo Verde Unit 1. As suct,, the information contained in this report concerning the System 80 design is intended for use in a . detailed pre-test' report to the Staff as an evaluation to supplement actual. test procedures. -1.2 Scope The information contained in this report, as discJssed in the previous sec-tion, is applicable to the San Onofre Nuclear Geierating Station Units 2 and 3 (Southern California Edison), the Waterford Steam Electric Station = Unit 3 (Louisiana Powar & Light), the St. Lucie Plant Unit 2 (Flcrida Power & Light), and the Palo' Verde Nuclear Generating Station Units 1, 2, and 3 (Arizona Public Service).

1.3 Background

As part of its shutdown cooling system review, the NRC Staff has required that all systens. and components 'needed to take the plant from hot standby to cold shutdown must meet the four functional requirements stated in Branch Technical Position RSB 5-1. Essentially, these systems and compo-nents must be safety-grade and capable of operating assuming only onsite power available or only.offsite power available and with a concurrent single failure. Based upon system design reviews and supporting engineer- 'ing analyses, compliance with the requirements of BTP RSB 5-1 is summartz-ed in individual plant Safety Evaluation Reports. Since each of the four plants discussed in this report fall into the Class 2 catagory under BTP 1RSB 5-1, full: compliance with all aspects of the branch technical position is not required. ~ In order. to provide an actual demonstration of the capability of the Com-bustion Engineering NSSS to cooldown in the absence of forced circulation, a full natural < circulation cooldown test was dev..oped and performad at . Unit 2 of the San Onofre Nuclear Generating Station on 27 July 1983. This stest, along with information obtained during certain previous tests as dis-cussed in Reference 2, was designed to provide data in order to address 2. s w --w-, -, ,&+, v, w, -,---v.+ -.e-rw- ,e e- -r-, - - .-w.- www+-- -. - - - - -w %-me-- -e-.

specific concerns regarding boron mixing, reactor vessel upper head cool- ' down, Seismic Category I condensate supply, and ADV nitrogen ca'pacity. .Since SONGS Unit 2 is prototypical of pre-System 80 plants, test results along with a detailed post-test data evaluation from the SCE natural cir-culation cooldown test were referenced by Florida Power & Light and Louis-Liana Power & Light in showing compliance with RSB 5-1. The NRC found this acceptable pending submittal of a detailed report documenting applicabil-ity of the SONGS test, see Section 5.4.3 of Reference 13 and Section 5.4.3 of Reference 14. Because of differences with respect to the size of the reactor vessel upper-head, System 80 plants could not directly reference the SONGS natural circulation test in their licensing process. Therefore, Arizona Public Service will conduct its own boron mixing and natural circu-lation cooldown tests. The SCE natural circulation test and test results, however, will be referenced by APS as part of a pre-test report to supple-ment actual test procedures. 9 s I l l l-L p i 4 3 L

l 2.0

SUMMARY

OF SONGS NATURAL CIRCULATION TEST 2.1 Test Method The SONGS Natural Circulation Test, Sections 14.2.11.88 and 14.2.12.105 of Reference 1, was conducted over a time period of approximately twenty-four hours commencing with a plant trip at 10:20 AM on 27 July 1983 and ending with a full natural circulation cooldown and depressurizrtion to condi-tions that permit operation of the shutdown cooling system. Five specific objectives were stated for this tost as follows: 1. To measure plani: response to a total loss of reactor coolant flow while at 80% power. 2. To verify natural circulation following the trip from 80% power. 3. To determine that adequate boron mixing can be achieved under natural circulation conditions. 4 To demonstrate the ability to perform a natural circulation cooldoer. to shutdown cooling initiation conditions. 5. To evaluate' natural circulation flow conditions. Each of the above objectives was met during the performance of the Natural Circulation Test. In addition, information and data was obtained regard-ing the ability of the plant CPCs to automatically trip the reactor in re-sponse to a loss of forced circulation, feedwater requirements necessary to remove decay heat plus perform a plant cooldown, adequacy of the size of the safety-grade nitrogen supply to the ADVs, and RVUH cooldown under natural circulation conditions with the CEDM cooling fans in operation. Table 2-1 (p. 5) contains an cutline of the procedure used during the Natural Circulation Test along with a brief discussion of the major evol-utions. Section 2.2 below contains a summary of relevant test results. 4

Table 2-1 Procedure Outline SONGS Natural Circulation Test Procedure Outline Discussion I. Initial conditions. A. Plant. stable at 80 + 5% A. Adequate decay heat is of rated thermal power. available for natural circulation. B. All f our plant-protective system channels are in operation. 4 C. Steam bypass control system and bcth feedwater i -control systems are in automatic. ~ D. Pressurizer pressure and -level control systems are in automatic. E. Plant safety-related and E. All ' plant auxiliary and safety-auxiliary loads are related loads will be energized transferred to the reserve-throughout the initial auxiliary transformers. transient. Loss of offsite power will be simulated only during this test. The Loss of Offsite Power Test is desigrad to demonstrate plant behavior under an actual loss of offsite-power as required by Regulatory Guide 1.68, Rev. O. t II, Procedure. A. Reactor coolant pump A. During the initial transient . trip /na. ural ci. culation. 'and the subsequent 'boration, steaming will be accomplished via the steam bypass control system. The main feedwater system will be used to supply steam generators pre-trip; the auxiliary feedwater system will - be used to supply steam generators post-trip. . ~ - -.

Table 2-1 (cont. ) Procedure Outline Discussion

1. Trip reactor.

2. Establish natural circulation.

3. f tabilize plant and take data.-

4. Verification of natural circulation.

5. Measure power-to-flow rati o.

6.- Secure all but 1E 6. Only pressurizer heaters that pressurizer heaters. can be powered from the diesel gerarators will be used for the remainder of the test to simulate a loss of offsite power. ~B. Boron mixing demonstration.

1. Determine required 1.

Amount added will be the ' boron increase. greater of 100 ppm or that required for a proper shutdtwn margin at approximately 350*F. ' 2. Initiate boron sampling. 2. Sample locations and frequencies are designed to r, demonstrate adequate mixing and permit estimation of times - required to achieve such mixing.

a. Hot Leg #2 - 30 min.

E

b. Pressurizer - 60 min.

c. 'Velume control tank -

60 mi n.

d. Boronometer - 10 min.

3. Initiate boration.

a. Charge to loops using a.

Maximize mixing in the chemical a minimum of two and volume control system and l charging pumps, enhance boronometer readings. 6 x.

Table 2-1 (cont. ) Procedure Outline Discussion

b. Periodically operate b.

Boron mixing in pressurizer. heaters and auxiliary spray simultaneously.

c. Maintain pressurizer level 33 + 3%.

d. Acceptance criteria for adequate boron mixing - three successive hot leg samples within i 10 ppm, pressurizer and volume control tank within 1 10 ppm of reactor coolant system average.

C. Cooldown.

1. Ensure adequate feedwater 1.

Actual requirements to be a vailable. determined via careful analysis of post-test feedwater inventory data.

2. Initiate Hot Leg #2 boron samples and boronometer readin gs.

3. Transfer steam loads to 3.-

Nitrogen usage _to be evaluated atmospheric dunp valves. via careful monitoring of the Air to be supplied to nitrogen supply bottles. these valves via the safety-grade nitrogen supply bottles. - '4. Using botin atmospheric 4. Charge as necessary due to dump valves, initiate coolant contraction to t .cooldown. maintain pressurizer level. The concentration of the charging fluid to be maintained via the RCS auto mckeup system.

a. < 75'F/h r.

b. Feed steam generators

-via the auxiliary feedwater system. .7

r -- Table 2-1 (cont.) Procedure Outline Discussion

c. Stabilize plant after c.

Cooldown to a cold leg cooldown complete for temperature of about 350*F. approximately 30 min. D. Depressurization. D. The procedure will perform an early depressurization to check the accuracy of upper head steam bubble formation pre-dictions.

1. Maintain pressurizer level control system in automatic if possible.

2. Initiate auxiliary spray.

a. Monitor for reactor vessel upper head steam bubble for-mati on.

b. When steam bubble

b. -System pressure at the time of

-formation is steam bubble formation will be indicated, stop used to determine upper head depressurization prior temperature and thus cooldown to exceeding a pres-rate. surizer level of 40%, increase system pressure by 100 psi, and wait one hour before continuing.

c. Repeat Step II.D.2 until the plant is depressurized to shut-down cooling system entry pressure.

E. System restoration. E. Restart one reactor coolant pump and continue to take boron concentration-data in order to check for possible boron stratification. 8

2.2 Sumary of Test Results The SONGS Natural Circulation Test was initiated from a power level of 80% of RTP by manually tripping all four reactor coolant pumps. (Table 2-2 (p.10) lists the major sequence of events and Figure 2-1 (p.15) shows cold leg temperature and pressurizer pressure for the entire test.) An automatic reactor trip in response to securing all four RCPs was generated in less than one second by the core protection calculators on a low DNBR. RCS temperatures responded to this trip as predicted in Reference 2. Figure 2-2 (p. 16) shows hot leg and cold leg temperatures in Loop 1 for the first fifteen minutes following test initiation. Hot leg temperature decreased sharply following reactor trip as fission heat generation rapid-ly shut down in response to the inward motion of the control rods. Cold leg temperature decreased initially then stabilized after approximately five minutes as the SBCS acted to maintain steam generator pressure at approximately 1000 psia. After about one minute following trip, hot leg temperature decreased to a minimum of about 553'F then gradually increased as the speed of the reactor coolant pumps slowly decreased and the reactor coolant system went into natural circulation, Figure 2-2. Full natural circulation flow was verified twelve minutes following trip. The loop dif-ferential temperature at this time was 28'F which corresponds to a decay heat level of slightly greater than 1.5% of RTP. Both pressurizer pressure and pressurizer level, as shown in Figure 2-3 (p.17), responded to the plant trip as predicted. Level decreased rapid-ly from its initial value due to contraction in the RCS loops. (This con-traction resulted from the initial temperature drop shown in Figure 2-2.) The' decrease in level produced a corresponding decrease in system pres-sure. A minimum pressurizer pressure of about 1975 psia and a minimum pressurizer level of about 27% were reached one minute into the event. After reaching these minimums, pressure and level began to increase and eventually stabilized at their normal hot standby values. The increases in pressure and level were due to the following effsets: insurge into the pressurizer due to expansion in the RCS loops as the plant went into natur-al circulation, insurge into the presse-izer due to loop charging via the CVCS, and energy input from the pressurizer heaters. 9

f Table 2-2 SONGS Natural Circulation Test Sequence of Major Events Time Event Comment 10:20 AM Reactor coolant pumps Automatic reactor trip on low -(7/27/83) manually tripped. DNBR obtained in less than one seCond. 10:32 Full natural circulation Loop differential temperature flow verified. equal to 28'F corresponding to a decay heat level of slightly greater than 1.5% of RTP. 12:41 Commenced RCS loop boration Initial RCS boron concentration using two charging pumps was 470 ppm. Final anticipated concentration to be 697 ppm. 13:21 Required quantity of boron Secured adding concentrated boric added to RCS. acid to RCS. Boron concentration approximately 570 ppm and in-creasing as natural circulation mixing continues. 14:00 Boration completed. All added boron mixed in RCS loops. Hot leg boron sample equal to 693 ppm, average boronometer reading equal to 689 ppm. 20:33 Instr" ment air to ADVS ADVs placed on safety-grade secured. nitrogen supply in preparation to performing plant cooldown. Both MSIVs closed. 21:24 Commence RCS cooldown Safety-grade nitrogen supply to using ADVs. be monitored. 01:26 AM Plant entered Mode 4. .(7/28/83) 10

Table 2-2 (cont. ) Time Event Comment 01:36 RCS cooldown completed. Average cooldown rate equal to 48'F/hr. Cold leg temperature equal to 334*F, pressurizer pressure equal to 2287 psig. t 02:45 Manual operation of ADVs. Ability to manually operate ADVs by local handwheels as ordered from the control room demon-strated. 03:02 Commenced RCS depressuri-Fine control of auxiliary spray. zation via auxiliary spray. flowrate and hence depressuriza-tion rate maintained via throt-tling of main spray valves. Letdown in manual. i 06:03 Depressurization completed. SCS entry conditions achieved. RCS pressure equal to 374 psig. No detectable steam bubble formed in the RVUH. 06:35 Start RCP P004. Natural Boron concentration in RCS moni-circulation test complete. tored. No detectable concentra-tion changes noted following RCP start indicating that entire RCS was thoroughly mixed. h Y

.s

,. - ' ~'~ '_

• ?

. >-~~ {,. . -c f. 11 ~ x

' After the plant had stabilized following the establishment of natural cir-culation, preparations were made in order to comence boration.and conduct the boron mixing experiment.. An initial boration was made to the volume control tank in order to restore normal level following the plant trip and to' increase boric acid concentration. The increase in boron in the VCT was.necessary to prevent dilution in the reactor coolant system during subsequent evolutions as required by Section 7.1 of the plant Technical Specifications, Reference 3. A second boration, this time to the RCS loops, was then comenced using two charging pumps. (Refer to Figure 2-1 (p.15) and Table 2-2 (p.10) for the exact start time and duration for each of the major evolutions. conducted opring the Natural Circulation Test.) In order to provide the ~necessary concentration data, samples for chemical analysis were taken from the hot leg in Loop 2 and the pressurt-zer. -In addition, boronometer readings were obtained in the control room to check for trending and to serve as a backup to the manual samples. The initial RCS boron concentration was 470 ppm. Boric acid was added in order to obtain a final target concentration of 697 ppm. 4 An evaluat' ion and detailed description of the entire boron mixing exper-t iment is presented in Section 3.0 below. In summary, all required boric acid was added to the RCS~ using two charging pumps forty minutes after bor-ation'had been started. Loop concentration levels as indicated by manual ' sampling and boronometer readings began to rapi.dly increase within approx-i .'imately ten minutes of initiation of concentrated charging flow. By one hour and twenty minutes into the evolution, complete mixing in the RCS loops had been noted with the manual hot leg sample concentration equal to 693 ppm and the. average'boronometer reading equal to 689 ppm. Although not required in order to show compliance with the requirements of BTP PSB 5-1, mixing in the pressurizer was demonstrated by simultaneous use of the Class 1E pressurizer heaters and auxiliary spray. (hte that section 4.0 of this report contains a discussion of the effects of non-mixing in the pressurizer on reactivity.) -Following completion of the baron mixing experiment, preparations were 3 .made in order to perform a plant cooldown. These preparations included normal plant adrainistrative items as well as certain maintenance items which the operators elected to perform. As a result, hot standby condi-12 ... _ _ _ -, ~. -.

s m 1 x + ' tions were maintained in the RCS"as shown in Fi'gure 2-1 for approximately. seven and one half hours af ter horation before'comencing cocidown. Dar 3 ing this time frame significant coofinq took place in the' reactor vethel ] upper head due to the operation of the CEDM coelinh fans. (liefer to Sec-. ,O tion 5.0 below for a detailed evaluation cf1 RVUH cocidown during natural circulation with the CEDM cooling fans in seration: arid with tdc CED'A ' cooling fans secured.) Just' prior to cooldown, the nom.1al air supply to. the atmospheric dump /alves vak secured ind 4tt e safe.y-gr ade-nitrogen gas. bottles were placed in operati6n, see Figure 2 4 (p. 18) for ti simplified system schematic. The intent of this last evolution was to demo 1 strate the adequacy of the ADV nitroge'n supply.. As stated above, approximately seven and one ha)f hours after 'c'ompletion of the boron mixing experiment a symetric cooldown was commenced s,

1. e.,

a two steam generator cooldown using both atmospheric dump nives. ADV position was recorded and nitrogen usage was ' monitored in ardar to deter- ] mine the adequacy of the capacity of thel, safety-yeade sepoly ocetles. The plant was cooled from an average cold 14g temper 3ture of approximately 535'F to an average cold leg temperature' of.approximately 334*F in slight- ' ly greater than four hours. The RC2 was thus taken from Made 3, hot s standby, to Mode 4, hot shutdown, with an average cacidown rate of 48'F _s-per hour. Figure 2-5 (p.19) shows ADV position and accuatulator pressure during the cooldown portion of the test. The average rate of nitrogen usage was about 62 psi per hour with 'the average accumulator pressure de-creasing from approximately 960 psig to 'approximately 7F psig between.the w start and the completion of the cooldown.'?At that rate of usage, a full nitrogen accumulator, initial nominai pressure l'100. psig~, would take 17.7 - hours to completely discharge. sNote that the rate of nitrogin ' usage dur-ing a cooldown is high, i.e., a relatively large amount of' nnrogen is 're- ] quired to move an ADV to a new position in cogarisori to the small amount of leakage that is normally present with-no bovement. As a result, the 17.7 hour time to completely discharge 'an(accumulator, is, cov,ervative in ' that it does not consider the reistively 'small amount'of. nitrogen required to maintain the plant in hot standby when valve movemnt is miniraal. -(Seca tion 6.0 below contains a discussion of nitrogen usage by the atmospheric dump valves with respect to the requirements of Branch Technical positioni RSB 5-1.) 13 1 1

Following the completion of the cooldown (cold leg temperature equal to 537'F), the RCS was stabilized for one hour prior to commencing depressur-ization. During this time normal coalmunications were established between the ADV station and the control room and manual control of these valves via the manual handwheel, as di rected f rom the control room, was demonstra-ted. The plant was then depressurized to conditions that permit initia-tion of the shutdown cooling system using auxiliary spray. Figure 2-6 (p. 20) shows letdown flow, pressurizer pressure, and pressurizer level during the final portion of the plant depressurization. Pressure was re-duced rapidly and smoothly as shown. Pressurizer level was closely moni-tored for reactor vessel upper head steam bubble formation. The i'idi ca-tion of bubble formation in this test would have been a rapid increase in pressurizer level significantly greate than expected while operating g

y auxiliary spray. Level, as shown in Figure 2.6, exhibited only small and Y

} gradual changes. These changes are attributable to the plant operators K ;. f y.hp;f making small manua' changes in the letdown flowrate. The absence of steam h.. @N bubble formation in the RVUH, as discussed in Section 5.0 below, was due .A

i...

.,.~.: - to a combination of two factoes:

1) A relatively large cooling effect due

<t .,E Oq s. - to the CEDM cooling fans in operation. (These fans were left running wW ag [Q,@. during the er. tire natural circulation test in order to prevent damage to [.. the control element drive mechanisms. )

2) Maintaining hot standby con-

,. s se... ditions in the RCS for a relatively l~1g period of time, approximately y.p Cf .;A3-]; eleven hours, following trip prior to comencing plant cooldown. (During this time, as indicated in Section 5.0, the RVUH probably cooled to a f^..di.h) t temperature below that of the remainder of the RCS.) $s.: ,..'N 0'W Cnce a temperature and pressure were reached that would permit initiation fWh,,:i' t of the SCS, preparations were made to restore the plant and to canclude

.,.g the test. Pressurizer heaters were energized and plant pressure was in-g,a creased in order to restart a reactor coolant pun.p.

RCP P004 was restart-P <. ' Y N ,,pby*:% ed terminating natural circulation and terminating the test. Note that i boron concentration data via manual hot leg samples and boronometer read- (( .g.m 33z ing continued to be taken as discussed in Section 3.0 below, No varia- - jj__' # tions in RCS boron concentration were noted indicating complete mixing in l* all portions of the RCS by the end of the test. ~ T ' ~.. m.. V rg.% p ,\\.h. Y \\' fy& yi s 14

nessumummmer - - - .,,.7..39 ,, :. f.\\ < Yx; ,~.,.... y', gz 1Mummsa P gFg pzr (psia) Tcold (O F) ~ ~ w g [. 8 8 8 wr s .s. M* I l l l L.I I I I i 1 1 c M.f M g. Reactor trip D s 7.7, L;. p

c.

1. w,

y.u. %g .. u.m ~ M - Td % ' ' g. 2 ~ Commence boration I ;.t.M <.yc '- ? s i,. y>. f.g* ' 9(r,, to RCS loops 1.,,,.. ' w.'. % - f 2 64, c c.@t. 5 ?. ~Boration complete n E, w;:. .-l.,. .t. e,. np i.p . c .j me A og og .. e f]'? g4 p4i:;#

r a

+ G3 T. i, e 1 ,.:-Qi'n en pk..;.41. ta o On 8 } ?;

a. g :.:

9 zy t ne

:,,. y,y't.f: :., w-q to a,

'-p. m ,.-)_,.y 2C * . 4: 9 on .,;..c. &.... _. { ry a

q. 4. -

ca n -g. g g. n o, e V (g?..... _ p g . _v g. *'. u- .c w O, t.-t4 w s. N.J.h.:D, b Commence cooldown 3$ # . c w.: y.n!= s a: o e., 3.... y g;>.y a. .ww,w.. w- 'R).. .n. i._.-.. -g.x %N ee W . c. q.. o NO D 'O gO n

• 3 @

Q oa mm ec Cooldown complete 2 .4 . Commence depres-8 .suration m Depressuration com-l 8 plete - RCP restart, test complete

o L -s F ta m F Figure 2-2 RCS Temperature Response in Loop 1 for the E [ First Fifteen Minutes Following Plant Trip E During the EONGS Natural Circulation Test L P ? m 7 RCP trip followed by an automatic CPC trip W. -hy 590 P at time zero. .Y. - A :. ilt ?lN.?,0 g.y.. v Q g;% :~~.p, m

.ws -.

b. ,. ++.kp .'Y.,' A$f 2

.; t' L

580 h.; 5 m, ,a.,. w n [ h % f r . 3,9:gr.a.',,g ,4 r a m .9 .-.. 9-g ^ 'r y w ,,. ~ U .5 570 l .j. 4 w e

c. g.f g.s e

.v e: 3... p. -- M;,d1 t E y ',A m .y : - ' $; b 3 e

,,.7.L.E.

560

.[, f..^; ~

3, '{h,+,[; '**.I i.. 5 .h Q$- Mi M.' %j:g'- g ... 3 - e .: e S: - ?

g..

T 550 jg [ 9.:s+;y y E @ j';.E, i .h4k. 7 .kg..,;, r-1 I I I i i I s 'r ' '.. :/ W ', ', f :::~ A., 0 2 4 6 8 10 12 14 ...','.5' ..e ... i g., 4 w .: n 7 TIME (MINUTES) i,:y : 4@. ;. .m ~ ?, o jd:I'.. _' g p_ t = W M K- = i= 16 Kw - - -.-..-.i..

W...L. - A Figure 2-3 i ,Ay["1 1..' J-Pressurizer Pressure and Level Response J,, [;l! ~ for the First Fifteen Minutes Following Plant Trip During the SONGS Natural Circulation Test .,'r,+ g.- F. " W,6 r

59. 7,..,.

e,.

4,, g e

y a,

y 9. -v%

.

.g.,.g.,M l.s>.,. ih?M 'p. M 'i'M'~'*^ 50

f. %'

'.M Pressurizer level

.. -gAf%

h_..pQf 5 17 ' l T,. $ ^*

4 (>, !,: n p.y gg 1 M/, W A ". y* s%) 40 y 1 Q .:.. e,an h 2 .A%..y,a... .y- +.- 5 c.~,.:vy," - 30 K'*y@yh l l i l [ThM':. l l l l x >g. L.+..-

9.,7, 6;

.,. 3, ' '

g=, % lp d.b 2200 .q. ;. ;s. 9 'E.., % ; .;.}.' ~. -nV -} >,...,rph; 4, I a 5.n - ;', gg J MW*'- Pressurizer pressure ps-AS-m 4 :,{,[r.' 3 2100 --{- ~ 6, *.. u w'*k;.p y a, a, a s. ; :q 4t&. 5,. 2000 r t,. 'kM l l l I l l ',f.. 1 t t t 0 2 4 6 8 10 12 14 .a I $,.. 1'$p ,. f-TIME (MIWTES) lg.,i ..o b;,~J-{j; H,.5 l::N-~.- _o v.., ~ 3w.L.. : %f h'.4f-Q 17

i Figure 2-4 Simplified Schematic of SONGS Nitrogen Backup to the ADVs y'.. .?gS .(;. ' v.' ?- y. .-k,J ; h.i -/g. 'Y;M b ' '.f.~..f.fe ADV ADV cp '. g;,. n. l r.; ;- ; >y. '. y%'b 2-y. m. '.

:k.,%,M

g. &.

\\ / N / INST i1R . w .m. t %:. ? % '~,. -;j%'.7-PCV PCV 1 (FO) . wz., 4,' jf.t 7 g (FO) / % hf ) \\ _*; u - _ 7_ :.. n. J4 s..., :.- .T f, L.: wr/ g < r ;' %.' p( ALARM ALARM %. f. ] ' w -'. 'g. y:y., :.. <g. ..s .. x -p.. W., 'a (. t t. 441*1.p- : % 'A. J ' hp 4% $! 9: t, m.,s -T.' y... .A-( ) REGULATOR REGULATOR f (FO) (FO) O O NITROGEN NITROGEN ACCUMULATOR ACCUMULATOR O N) Note: Pressure control valves (PCV) are designed to open on loss of instrument air to the ADVs. 18

4 Figure 2-5 ADV Position and Accumulator Pressure During the Cooldown Portion of the SONGS Natural Circulation Test -z aN.y. '.w!;a '=2 >J O +r n P O*

-ev.. 4xT 75 g

[y.^"- u '. m 3. ' f-n,.: y=. 2o so r 4,- M Qx,.l l,:.^ E* .,;.y m cn ,a y. .,c ",.op o 25 l e c x j .x,,,. c. l A.g .a o I 4 ElllEESEE y'..: 7 'k?AE I,h ;;f a, s:v.

_ f;;~

> p_.; :_.. c 5. v,. ' A Accumulator MT212 .7J

,..,, h e Accumulator MT213 QR.p 1

,s. :, '. 3 >u.1. 'Y ; 1000

.K.. s. 3 is

@ a.c "g.g; v4

.. 'h.49 f$

R;.g'i s',( ;& G .../ v .. x, %' E' ,he' 900

g c:

2 2.~u e h

4 h'.

l y , y +/-.i ' [,%+.. I, c: .. ;. J :.. ^ S 800 .4 .'%,s*.'.'*,,, a p .g-..,. 3 .,]m. . }'.h 'F 2 . ~. y y 0 .W,, ? ,3 '3 i:s.. 5, 9 700 1 ') 8.' bl [' 3p u '*, . un. 1 2 3 4 5 TIME AFTER ISOLATING INSTRUMENT AIR (HOUBS) 19

Figura 2-6 Letdown Flow, Pressurizer Pressure, and Pressurizer Level During the RCS Depressurization Portion on the SONGS Natural Circulation Test 90 lLas 70- - v Z {30 c k 4 10 l I l l l -39 Q 37 ~ 'a ~ $. 35 ~ ~ i $ 33 m. 31 1 I I I i n c650 4 uso. Depressurization complete, S50 PZR heaters on in prepara-tion for RCP restart. 5 . m450 V - m g350 I I I I l 5:20 5:40 6:00 AM 20

3.0' BORON MIXING ANALYSIS 3.1 Introduction Two boron mixing experiments were conducted as part of the SONGS Natural Circulation Test Program. As required by Reference 4, the boron mixing experiments were designed to supply data in order to confirm that suf-ficient mixing of borated water added prior to or during cooldown can be achieved under natural circulation conditions and to permit estimation of th'e times required to achieve such mixing. The basic procedure demonstrat-ing mixing involved adding a predetermined amount of borated water to the reactor coolant system under specified test conditions. Appropriate samples at key sample locations were then taken at specified time ' inter-4 vals until adequate mixing was observed. By knowing the initial baron concentration and the amount of boron added at the beginning of the experi-ment, the expected final concentration was determined. Using this number coupled with the data obtained from the boron sample analyses, an estima-tion of mixing times was, calculated. On~ a microscopic level the process of baron mixing, i.e., the even distri-bution and dispersion of added boron throughout the reactor coolant system is quite complex. If the problem of confirming adequate boron mixing and of estimating the times required to achieve such mixing is approached on a macroscopic level then the tests and experiments designed to meet these objectives _and the subsequent analyses should be straight forward in nature. The intention of the following discussion is to consider the boron mixing process on a macroscopic level. In order to properly do this, component configuration and system features conducive to mixing will l l be elaborated upon. In addition, actual experimental and test data will ~ be presented in support of the overall mixing process as appropriate. 3.2 Boron Mixing Process. t

The process of mixing boron in the reactor coolant system, i.e., boration, should be divided into two discrete phases. The first phase is the deliv-ery phase. During delivery concentrated borated water is introduced or in-21 1

- - - - - - - - +

4 -jected into the RCS. Once boron'has been introduced, it must then be even-ly distributed and dispersed throughout.the primary system. Dispersion is essential if proper. reactivity control is to be maintained when plant temp- .eratures are reduced during cooldown. The second phase of baration is -therefore dispersion. The two phases of the boron mixing process are pre-sented separately in the subsections that follow. In addition, a subsec- -tion is included discussing the effect on mixing of variations in charging fluid ~ temperature.. 3.2.1 Boron Delivery Injection or delivery in a C-E NSSS is' accomplished by charging via the chemical and voluiae control system. Boron from a concentrated borated water source, either the boric acid makeup tanks or the refueling water tank, is injected directly into the reactor coolant system at the charging inlet nor.zles using positive displacemr.t type charging pumps. The exact location of the charging inlet nozzles is shown in Figure 3-1 (p. 38). Note that in the.St. Lucio, Waterford, and SONGS design.two inlet nozzles per plant are provided, and in the System 80 design only one charging in-let nozzle is provided. Each. charging nozzle penetrates the RCS cold leg pipe ~ 90' from the top of the pipe as shown in Figure 3-2 (p. 39). Two features of this mode of boron delivery are important to the mixing pro- . cess and to the requirements of RSB 5-1. The first feature of importance is the amount of. borated water that must be added to obtain the required shutdown margin for a plant cooldown and the time frame over which this boron =is added. The second feature of importance is the velocity of the charging fluid at the inlet nozzle as it penetrates into the reactor coolant. In order to take the reactor coolant system from hot standby temperatures (average coolant temperature typically greater than 350*F) to cold shut-down temperatures-(average coolant temperature typically less than 200*F), plant technical' specifications require boron to be added to the primary . system in order to maintain a proper shutdown margin. During the SONGS Natural Circulation Test, letdown was in operation throughout and boron was essentially batch added whtle in hot standby since letdown could be 22

used to control pressurizer level. (Refer to Section 3.3 below for test details.) In:the Branch Technical Position RSB 5-1 scenario, however, letdown will be. unavailable since it is not safety-grade. In addition, only a limited number of pressurizer heaters will be available. When borating the RCS during a loss of offsite power, careful consideration must be given to existing primary inventory, pressurizer level control, b and pressurizer pressure control. In addition, the amount of boron that must be added to maintain a proper shutdown margin will vary over core cy.cle. This _ variation is due to changes in the magnitude of such items as . moderator temperature coefficient and rod worth. As more boron must be added-toward the end of core cycle the net rasult, given a boric acid source of a specific concentration,-is that more solution must be added to the RCS. In a BTP RSB 5-1 scenario when letdown is unavailable and only limited pressurizer heater capacity is present, inventory control and thus pressure control are maintained by conducting boration concurrent with plant cooldown. In this manner concentra'ted boron can be added as part of the normal make-up necessary due to contraction, and thus pressurizer level and hence pressurizer pressure can be maintained fairly constant. (Refer to Reference 10 for a detailed computer simulation of a BTP RSB 5-1 scenario for System 80 showing _ the process of cooldewn with concurrent boration.) Further, calculations have been performed which show that sufficient boron can be: introduced into the RCS.and complete mixing will - occur during a concurrent cooldown and boration such that a proper shut-down margin can be maintained. - During. delivery, therefore, boron is injected into the reactor coolant i system of a C-E NSSS via the CVCS. This phase of the mixing process will be-smooth and continuous with all required boron being added concurrent with cooldown in the Branch Technical Position RSB 5-1 scenario. Gi ven ' the constant rate of boron addition, the fluid velocity at the charging - nozzles ca.7 be readily determined. Table 3-1 (p. 24) shows steady-state fluid velocities at the charging nozzles for SONGS Unit 2 as a function of the number of charging pumps with letdown in operation and with letdown secured. Note that the effect of letdown is to increase both fluid veloci-ties and temperatures at the charging nozzle, i.e., both the fluid veloc-l l ity and temperature with letdown in operation using a given number of l 23

Table 3,1 Fluid Velocities at the Charging Inlet Nozzle SONGS Unit 2 j Number of -Temperatu re ' Velocity at charging nozzle (ft/sec)* ' pumps (total at outlet to capacity) regenerative .at pump heat exchanger Single Both l'(44gpm)**- 90*F 6.8 3.4 ! 2 (88 gpm)** 90*F - 13.6 6.8 1(44gpm)I 445'F-8.1 4.0 2 (88 gpm)I 405'F 15.7 7.9 3(132gom)# 380*F 23.2 11.6 '* Velocity' is 'given for all flow diverted through one nozzle (single) and flow split between two nozzles (both).

• • No letdown.

Temperatures and velocities based upon steady-state test results .with. letdown (Attachment C, SONGS CVCS Integrated Test, 2HB-220-01). W 24

) charging pumps is greater than the corresponding situation without let-down. For comparison with the numbers in Table 3-1, the approximate vel-ocity of the fluid in cold leg during steady-state symetric natural cir-culation flow conditions and a core thermal output of 1.0% of rated thermal power is 1.4 ft/sec. Considering the limiting charging situation that was shown in Table 3-1, i.e., one charging pump with no letdown and flow to both inlet nozzles, the velocity of the charging fluid will be two and one half times greater than the velocity of the coolant in the cold . legs. This difference in velocity will ensure that the charging fluid, and hence boron, will penetrate substantially into the reactor coolant and that conditions whereby complete mixing can begin to take place will be established. Note that because of similarities in design, piping size, and piping configuration the above results and conclusions, although they ware obtained from San Onofre Nuclear Generating Station Unit 2 test data, are applicable to Waterford Unit 3, St. Lucie Unit 2, and System 80. For each of these plants the concentrated charging solution will penetrate sub-stantially into the reactor coolant during a natural circulation boration and immediately begin to mix. In sumary, the delivery phase of the mixing process is smooth and continu-ous due to boron addition via positive displacement type charging pumps. All boron is added concurrent with cooldown in a BTP RSB 5-1 scenario with the velocity of the charging fluid at the charging. nozzle such that sub-stantial penetration into the cold leg will occur and hence mixing with the coolant in the RCS can begin imediately. 3.2.2 Boron Dispersion in the RCS Once the boric acid solution has been in,iected into the reactor coolant, it must be evenly mixed throughout the RCS if proper reactivity control is to be maintained during cooldown. Referring to Figure 3-2 again, the con-centrated boron solution will penetrate 90* from the top of the cold leg pipe. Charging velocity at the inlet nozzle is high relative to the vel-ocity of the coolant in the cold leg during natural circulation such that the boron solution will penetrate substantial!" into the reactor coolant. Reynolds number calculations clearly indicate that flow through the cold 25 E

a i leg will be turbulent. - These two factors, penetration of charging fluid

into the. reactor coolant system and turbulent flow in the cold leg, will create the mechanism for substantial mixing of added boron as it flows from the charging inlet nozzle down the cold leg tows:d the reactor vessel inlet plenum and will prevent establishing the conditions necessary for l.

. boron.and. temperature stratification. Reference 5 contains experimental data which can be used to support the .' conclusion that substantial mixing of injected boron will in fact take s-place as it is carried toward the reactor vessel inlet plenum by the fluid in the reactor coolant system cold legs. The tests.in Reference 5 were designed to verify the mixing models used h Combustion Engineering in evaluating pressurized thermal shock events and,'as will be shown, certain of the results are applicable to the boron mixing process. Figure 3-3 (p. 40) shows the data from.one of these experiments, Test 42. The work . as funded by the Electric Power Research Institute and performed by w l Creare, Inc., on' a one-fifth scale-mixing facility. Each.of the points in Figure 3-3 are thermocouples with steady-state temperatures as indicated. The-test was designed to simulate the initiation of safety injection flow during natural circulation and the results show that mixing was approxi-4 mately:90% complete at the end of tha cold leg. The actual conditions ergloyed during the test and the scaled-up plant conditions to which they correspond 'are given in Table 3-2 (p. 27). ' Although,:as stated in Reference 5, data from a larger test facility are ._not yet available to confirm the scaling. rationale used to obtain the corresponding plant conditions in Table 3-2, several conclusions relevant to the boron mixing process can be drawn from the results in Figure 3-3. Using the highest nozzle velocity, for conservatism, in the absence of let-down from Table 3-1 (13.6 ft/sec) and typical results from actual natural -circulation testing, a table similar. to Table 3-2 can be compiled for the ' situation in which charging is used to add fluid to the reactor coolant system vice safety-. injection. This information is presented in Table 3-3 . (p..28). Since the. maximum charging flowrate (WCHG in Table 3-3) is . lower than the' safety injection flowrate (W ; in Table 3-2) and since 3 the ratio of loop flowrate to charging flowrate is greater than the ratio 26 -. - -. - ~

Table 3-2 Comparison of Test Conditions During Creare Test 42 with Corresponding Plant Conditions Plant Conditions During Corresponding Plant Parameter Test 42 Conditions '~' LOOP (lbm/sec) 2.7 450 n W ; (lbm/sec) 0.3 50 3 9 9 LOOP /WSI W TLOOP ( F)- 155 400 T ; (*F) 60 40 3 27 ... - ~..

Table ~ 3-3 Typical Plant Conditions During a ' Natural Circulation Boration Without Letdown Plant Parameter Typical Plant Conditions WLOOP (1bm/sec) 307 'WCHG (1bm/sec) 31 10 WLOOP /WCHG TLOOP.( F) 545 TCHG ('F) 90 e 5 28

N 4 of loop ;flowrate to safety injection flowrate, it can be concluded that the mixing reSults in Figure' 3-3 for safety injection envelop the situa-tion where borated water is introduced into the cold leg piping via the ~ -charging system. This conclusion 1s further supported by the configura-tion of the cold legs, i.e., the relative position of the charging inlet

nozzle to the safety injection nozzle. As seen from Figure 3-4 (p. 41)

Lthe charging inlet nozzle for SONGS Unit 2, Waterford Unit 3, St. Lucie U Unit 2,-and. System 80 is located at a distance from the reactor vessel inlet plenum that is comparable to the distance of the safety injection

inlet nozzle from the reactor vessel inlet plenum. Therefore, mixing of

. injected charging fluid should be at least t.s complete as the mixing demon-strated in the Creare mixing experiment in Figure 3-3. Finally, note that the charging inlet nozzle in the SONGS and System 80 design, Figtre 3-4, is actually located further from the reactor vessel inlet plenum than the safety injection n9zzle~ thus allowing for increased interaction with the coolant-in the cold leg and hence even more complete mixing of the charg-

ing fluid.

At this point in the discussion of the boron mixing process, the borated water that was added to the reactor-coolant system at a loop charging nozzle has. travelled down the cold leg to the reactor vessel inlet plenum. As was shown, thermal and hence boron mixing at the inlet plenum will be greater than 90% complete, i.e., the added boron will be 90% mixed with the' coolant in the cold leg at the in!et to the reactor vessel. This ' mixing will prevent the establishment of a definite colder. temperature lay-er and hence prevent stratification. Fluid from the cold legs then enters l the ~1nlet plenum, see Figure 3-5 (p. 42), where it travels down the down-comer and into the reactor vessel bottom. This region (i.e., reactor vessel inlet plenum, downcomer, and bottom). is essentially a common node where~ fluid from all four cold legs can interact and mix thus providing the means for boron to disperse into both RCS loops. In addition, note that-in the SONGS, St. Lucie, and Waterford design, an additional mixing mechanism is present.in that fluid entering the inlet plenum from the cold legs will gain a counter clockwise rotation as it flows into the downcomer region of the reactor vessel. This rotating motion results frcm the cold leg piping configuration as shown in Figure 3-6 (p. 43) and was actually r 29 L,a

observed at Southern California during pre-tore testing of the safety in-jection system by plant personnel standing above the uncovered reactor vessel. Thus, fluid rotation will enhance interaction in the downcomer between the fluid entering from all four of the cold legs. As was stated above, fluid rotation will enhance mixing in the reactor vessel downcomer in the SONGS,'Waterford, and St. Lucie design. In the System 80 design, however, the cold leg piping configuration is such that little or no fluid rotation is expect'ed to occur, Figure 3-7 (p. 44). Therefore, mixing between loops in the System 80 RCS will. result solely due to turbulent flow and interaction in the inlet plenum, downcomer and reactor vessel bottom. Note that Arizona Public Service plans to perform their.own boron mixing experiment similar to the San Onofre test during which they will demonstrate mixing of added boron under natural circula- ' tion and thus will demonstrate this common node interaction where boron from one cold leg is evenly distributed throughout the entire reactor coolant system. As can' be seen f rom the side views in Figure 3-5 (p. 42), two regions for . potential _ stratification or non-mixing of boron exist. These two regions are the reactor vessel upper head and the vessel bottom. In the reactor vessel bottom region there are two factors that will act to prevent strat- - i fi cati on. First, coolant leaving the downcomer region will be substant-ially mixed due to mixing in the cold leg and due to interaction in the inlet plenum and downcomer. This mixing, both thermal and boron, means that dense highly concentrated pockets or slugs of borated water will be eliminated prior to reaching the bottom of the reactor vessel.

Second, the' reactor vessel bottom is smooth and relatively open thus allowing for complete and unhindered interaction of all fluid entering the region. As can be seen from Figure 3-5, this is particularly true of the C-E pre-System 80 design.

In the System 80 design, instrument penetrations are provided for in the vessel bottom. However, these in-core instrument nozzles'are relatively few in number and well spaced such that adequate mixing will take place. (Note agair, that APS plans to conduct their own boron mixing experiment on Palo Verde Unit 1 in which complete mixing of all added boron will be demcnstrated under natural circulation conditions in the System 80 NSSS.) 30

~ -The second region where the potential for significant delays in boron mixing is present.is the reactor vessel upper head. This potential exists because~ there is no mechanism for rapidly introducing boron into the RVUH in the absence of forced circulation. With RCPs secured, mixing in the region takes. place through diffusion and a small amount of natural circula-tion induced flow. Although these mechanisms are relatively slow, mixing can be expected to occur over a period of several hours. As will be shown in Section 3.3-below, complete mixing of _added boron in all portions of the reactor coolant. system including the RVUH had taken place by eighteen -hours following boration in the SONGS Natural Circulation Test. Also, as will.be shown in Section 4.0, the relatively slow mixing of boron in the upper head does not present a problem with respect to reactivity control particularly in the event of RVUH steam bubble formation when the poten- -tial exists for flushing a large relatively dilute slug of water into the RCS.. The final area or region in the RCS' which will aid in the mixing process is the steam generator. As boron passes through the reactor vessel bottom, it flows up through the core and into the outlet plenum. Once in the outlet plenum, flow will split with' half traveling down Hot Leg 1 toward Steam Generator 1 and half traveling down Hot Leg 2 toward Steam Generator 2. Once in the steam generator, flow will pass up through the tube sheet and into the U-tubes. A's shown in Figure 3-8 (p. 45) U-tubes 'at the center of the tube bundle are substantially shorter than tubes on -the outside of the bundle. Since all U-tubes have the same diameter, the velocity of the fluid in each tube will be eq'fal. The result will there-fore be that an increment of coolant entering one of the shorter U-tubes ~ ill exit that _ tube and pass out of the-steam generator sooner than an in-w cr* ment of coolant entering one of the longer tubes. Eence, a mechanism - exists for elongating or spreading cut-of added boron in the actively flow-Ling regions of the reactor coolant system. In summary,' the process of boron mixing or-dispersion was discussed by con-sidering primary system component configuration supplemented with simple hydraulic calculations and experimental ' data where appropriate. Three mechanisms were found to be present that would ensure complete mixing of r 31

added boron and prevent stratification ir, all actively flowing portions of the reactor coolant system, i.e., all regions except the reactor vessel upper head. These mechanisms were the substantial mixing of baron with fluid in the cold leg; common node interaction in the reactor vessel inlet plenum, downcomer, ar.d vessel bottom; and elongation by the steam gener- 'ator U-tubes. Finally, even though the RVUH is relatively stagnant under natural circulation conditions, mixing in that region is expected to occur over a period of several hours due to diffusion and a small amount of natural circulation induced flow. 3.2.3 Effect of Charging Fluid Temperature As was shown in Table 3-1, letdown has two important effects on the charg-ing fluid. First, the use of letdown increases the velocity at the charg-ing inlet nozzle, and second, the use of letdown increases charging fluid temperatures. During the SONGS Natural Circulation Test letdown was not isolated and hence charging velocities and temperatures were relatively high..(Note that letdown remained in operation during this test in order -to minimize lthe thermal cycles that would b'e placed on the charging noz-zles if it were secured.) If letdown were isolated and hence charging velocity and temperature were relatively low, the probability of influ-encing.the mixing process through boron and temperature stratification - would be increased. Based upon the results of the mixing experiment that -were presented in Figure 3-3, however, stratificaticn will not occur. Referring again to Table 3-2, the corresponding plant safety injection temperature of 40*F was shown to be 90% mixed at the inlet to the reactor vessel downcomer. -The results of the Creare mixing test then onvelop the situation ~ where 90*r water at a much lower flowrate is being added to the cold leg via the_ charging system. Therefore, variations in charging temp-erature due to changes in letdown flow will have a negligible effect on boron mixing

3.3 Boron Mixing Test Results Two boron mixing experiments were _ conducted as part of the SONGS Natural

' Circulation Test Program in order to supply data sufficient to demonstrate 02

r. L [. , that adequate mixing of all added baron could be accomplished under nat-ural circulation conditions in the C-E NSSS, As will be discussed below the first experiment involvfd adding a concentrated boron slug to the reactor _ coolant system during natural circulation with the reactor crit-icil to simulate decay heat. The progress of the slug was then followed -on the plant reactivity computer as it transited the primary loops. The second experiment involved adding 'a predetermined amount of boron to the RCS during natural circulation following a plant trip from 80% power and then monitoring primary system boron concentration at discrete locations as mixing occurred. The results from both experiments combine to demon-strate that complete mixing of borated water added prior to and during cooldown can be achieved under natural-circulation conditions and to allow a calculation of the times required to achieve this mixing. ~ The first experiment used to obtain boron mixing data was performed early in the SONGS Natural Circulation Test Program following the low power nat-ural circulation tests. (Refer to Reference 7 for a comp'lete discussion -of the SONGS Low Power Natural Circulation Tests.) The basic procedure involved adding a concentrated ten gallon slug of boron using one charging pump to the reactor coolant system at or,e of the loop charging inlet noz-zles. 'The progress of tne slug was then followed on the plant reactivity computer as it transited the primary loops. Exact plant conditions at the time of the test were as follows: 1. R'eactor critica; at approximately 1.8% of rated thermal power to simulate decay heat. 2. All four reactor coolant pumps secured, i.e., natural ci rculation.

3..- - Reactor coolant system pressure approximately 2250 psia.

4 Cold leg temperature approximately 515'F. 5. Hot leg temperature approximately 548'F. 6.- Steam generator pressure approximately 771 psia. 33 b-

b 7. Group 6 control element assemblies greater than 60 inches withdrawn. - All other control element assemblies full out. [ 8. Concentration of boron slug approximately 9.5 weight percent boron or apprnximately.16,600 ppm. 9.~ Initial loop boron concentration less than 600 ppm. Figure 3-9_(p. 46) shows the behavior of the plant reactivity computer fol-lowing the addition of the ten gallon slug of borated water. For refer-ence, Figure 3-10 (p. 47)'shows the location of the detectors used by this computer. Note'that the signal from both detectors is averaged and that this average signal is then used to plot reactivity. 4 Several important pieces of information can be obtained by looking at the 4 general reactivity trends in Figure 3-9 which support the mechanisms for - boron mixing discussed in the previous section of this report. First, with each pass of the boron slug through the core a negative reactivity - spike resulted; from these spikes a loop transit time.of 4.8 minutes could be directly measured.- Next, with each pass through the core the slug of horon became more and more diluted, i.e., the slug of boron became more ~and more elongated and mixed as seen by noting that each successive neg-ative reactivity spike is smaller and broader than the one before. Thi s . mixing took. place due to mixing in the cold leg and interaction in the reactor. vessel-inlet plenum, downcomer, and bottom. Elongation took place 'due to the effect of different steam generator U-tube. lengths. Finally,

the boron slug was substuntially mixed by the time it made its fourth pass through the core approximately 17. minutes after it was first added., There-fore, complete mixing of-the slug can be expected within five or six loop transit times.

The second experiment which was used to obtain boron mixing data was per-formed as part of the last test in the SONGS Natural Circulation Test Program, i.e., the Natural Circulation Test from 80% power. Following l plant trip and the establishment of natural circulation, an initial bora- . tion was made to the volume control tank in order to restore normal level 34 - 2

. and;to-incre6se boric acid concentration. The increase in boron in the l

VCT was necessary to prevent dilution in the reactor coolant system during

_ subsequent evolutions as required by Section 7.1 of the plant Technical . Specifications, Reference 3. The ntxt boration, this time to the RCS ~ loops, was then made using two charging pumps. In order to provide the necessary concentration data, samples for chemical analysis were taken - from the-hot leg in loop 2 and the pressurizer as shown in Figure 3-11 -(p. 48).. In addition, boronometer readings were obtained in the control room to check for trending and to serve as a-backup to the manual sam- . ples. ' Several additional borations were made to the reactor coolant system as discussed below in order to prevent dilution during makeup operations. The results of the boron mixing experiment are shown graphically in Figure 3-12 -(p. 49) and in Figure 13 (p. 50). Figure 3-12 contains data from - the hot -leg samples'and the boronometer for the first five hours of the mixing cxperiment showing the first and second RCS borations. Figure 3-13 . contains data from the hot leg, boronometer, and the pressurizer for the entire natural circulation test. Sample frequencies and locations were . chosen to enhance data collection and provide maximum information with respect to boron' mixing. As can be seen from Figure 3-12, the intial RCS boron concentration was 480 ppm. Following the initial boron addition to the VCT, the first RCS boration was commenced with the ~ operators adding enough boron for an expected final concentration of 697 ppm. RCS boron ' concentrations 1as indicated by both the manual sample:, and the boronometer - readings began to increase shortly.'after the commencement of boration to ~ the loops and continued to rapidly. increase as seen from Figure 3-12. 4 Note that due to the delay time' associated with the boronometer (approx- ' imately ten minutes), these readings-will be less.than actual loop concen-trations following a boration until complete mixing has occurred. As con-cluded from the results in Figure 3-9, complete mixing of an added slug of - boron can be expected to occur'in five to six loop transit times during - natural circulation. At the time' that this. experiment was conducted, the decay: heat level that existed in the core was 0.8% of rated thermal power which corresponds to a loop transit time of approximstely six minutes. Therefore complete mixing in the primary system should have been seen in i 35-

about' thirty-six minutes following the addition of the final increment of boron. As seen from Figure 3-12, this is exactly the behavior that was observed. The first and largest RCS boration was commenced at 12:41 with all boron being added to the plant by 1:21. Then, approximately 39 min-utes later at 2:00 boron levels had stabilized with the manual hot leg sample concentration equal to 693 ppm and the average boronometer reading equal to 689' ppm. Several subsequent small borations were made as indicat-ed in Figure 3-12 and in Figure 3-13 as part of normal RCS makeup. Note that boron concentrations increased smoothly and rapidly with each add-ition as expected. Although not required for reactivity control (as discussed in Section 4.0 below) or to meet the requirements of Branch Technical Position RSB 5-1, the ability to mix boron in the pressurizer using auxiliary spray was demonstrated. As shown in Figure 3-13, boron concentration in the pres-surizer gradually increased over a period of many hours. This increase resulted from'small additions to the pressurizer via the auxiliary spray system which were made by plant operators throughout the entire natural . circulation test. Two final pieces of information were obtained during the SONGS Natural Circulation Test which are important with respect to the boron mixing process. First, not only was the ability to mix boron while in hot stand- .by demonstrated, but the ability to borate concurrent with cooldown was also shown. As seen in Figure 2-1 (p.15), the plant cooldown was commenc-ed _at approximately 9:20 PM on July 27 and completed four hours later at-1:30 AM on July 28. Referring to Figure 3-13, a third boration to the RCS was conducted during this time as part of ncrmal makeup due to contraction in the loops. As can be seen from the hot leg samples during this period, boron concentration increased smoothly and rapidly. The final piece of information relative to the mixing process was obtained at the very end of the test at 6:30 AM as shown in Figure 3-13. At that time one of the four reactor coolant pumps was restarted. Concentration data continued to be taken in order to monitor for non-mixing or stratification. No changes in boron concentration were observed indicating complete mixing of all added boron in all portions of the RCS by the end of the test. 36

c I 13;4 ' Conclusions-t I, Based upon empirical test data, the ability to completely mix added boron and pr' vent stratification in all actively flowing portions of the reactor coolant system has been demonstrated. Mixing under natural circulation conditions is rapid,- and because of system configuration, mixing is inde- ' pendent of charging: fluid temperature. Three mechanisms exist which en-sure complete and rapid mixing as follows:

1) Substantial mixing with

--fluid in the cold leg.- 2) Common-node interaction in the reactor vessel inlet plenum, downcomer,~and vessel' bottom.- 3) Elongation by the steam generator U-tubes. In addition, fluid rotation in the downcomer will further enhance interaction and mixing in the pre-system 80 design. Adequ-Late mixing of added boron, as demonstrated by empirical test data, can be achieved in five to six. loop transient times. Based upon the range of possible fluid velocities that could exist due to differences in decay / heat lev'els under subcooled natural circulation flow conditions, this mix- ^ ing can be expected to occur by sixty minutes following addition. Final- .ly(, boron mixing in' the relatively. stagnant reactor vessel upper head _ region of-the RCS in the pre-System 80 design has been demonstrated to occur within eighteen hours following boration, t' 4 4 i 9 37

l Figure 3-1 Reactor Coolant System Showing Charging Inlet Nozzles i am.i iut '7' og to f tl'o,0 s A. EAcTom /_4 Ilt "55n s I 4' O J - STEAM s f a s'[ k'y h* g=/e= m I i 4) / '6 t i ur i Puw as. rs MasP % 1A Parssuerzta Typical of SONGS 2&3, Waterford 3, and St. Lucie 2 PAG 90UA12tM

OPuta, osas.In. ta 2T Pyaar g

le 2A l i \\ , - g >croa 9t o... m.,,,, a ?! \\ N i System 80 l l I i 38

Figure 3-2 C-E NSSS Charging Inlet Nozzle Configuration a 15" CHARGING FLOW >-2.6* I.o. = 'l COLD LEG PIPING CROSS SECTION. SEE FIGURE 3-1. l 39 ._..____a_._,_-.._.

Figure 3-3 I Fluid Temperature Distribution for CREARE Mixing Test 42 2.14 GPM TiO0s n1xto = 145.8oF \\ 60 F 1.75" 5.62" 1:n; w s c^ . :.4 e 5+10 l. ~ e 14 II4 143+5 145 M 4 n 146g3 (M3 347 q,46143 160 1464>q,144 3 120 22.4" ~ 145 d) f f I t 160 il146 146qi 120 2" a g g g g 0 20 40 60 80 1 TIME (sec) i I. I 40 l - ~ - -

m. s s. t Figure 3-4 C-E NSSS RCS Cold Leg Configuration Charging nozzle to inlet plenum / x l\\ m-m_ A Y\\ SI nozzle to inlet plenum SONGS LFAL SL2 SYS 80 X 173" 136" 135" 191" Y 151" 145" 165" 164" / L 41 =..

Figure 3-5 Reactor Vessel Arrangement -CONTROL R00 0 RIVE 5 o o ,a CONTRgN#ft e l' "'% ; F INSTRUMENTATICN I "C"# W ,Af ") ~i' N0ZZLE / N g l) d d-{l a 3 i l 'l coNTad 'll (LEMENT )L* = A55fM8LY ar ::: CEA FU - C00t0L A55EM8LY $ $ I ll. Wim0RYAN I l N0 NORAWN J( '. 1 7 f (A UG5 BARREL --.j iI } t l N'!D < CALANDRIA AF' [ f [ g fsfg{ml "~ l l M l l-N0ZZLE {!D fg?}D INtET N0Z2J k l OUitET N0ZZLE {70Vb[ Il id Fu $$tMgty f j T i1 ii ! 1.i +4i. g l i ll Alb4NT PLATE i gg. , e CORE gw. l", 0 0.51 MI di ]--- -5HROU3 o.s! mi i ACTIVE ii:i;, y"' FUEL A55EM3LY CORE ACTIVE i l 8R L k$EMBLY TH I 7"-CCRE SHROUD + 7-,- -,, v r = b il\\ (jp11g -[5 W l'aucmar qi Y \\'p 'l0*S*I"I ' fh7NTATl0N ['$f Typical pre-System 80 System 80 Design 42

Figure 3-6 Top View of Reactor Coolant System Showing Downcomer Fluid Rotation Due to Cold Leg Angle of Penetration SONGS 2 & 3, SL2,-Waterford 3 i .MM <im. /< e-[at / N. .r-l i. )n : : e r i i t. j =. ;. v.: ,i g

1 g

7, = -.. \\p// 4' i. / -5D1-7) A. .W v. 7 a 4 43

r-Figure 3-7 Symmetry of System 80 Cold Leg Penetrations \\ o , so - ~ 30 n' s A. ) .:yto;;; 3 ,;; 3, ,,, t q.i r r.,. u .y .2 / ? ~~g. 30o 300 ~ '44 ~

Figure 3-8 Relative Length of Steam Generator U-tubes EM I . j -d -i l l 5 AN [! A k l -,n I' ' \\\\ .ll m, T Outer U-tube .1 M, l ) i \\y Inner U-tube 6 . : i Ni ,i l l l l, l l l I l l l l i ^ 1 all NlMN SMR a s i s s SONGS LP&L SL2 SYS 80 Outer Tube length 949" 949" 904" 985" Inner Tube Length 554" 554" 528" 550" l 45

Figure 3-9 Heactivity Following Addition of Ten Gallon Boron Slug to One Loop of Reactor. Coolant System During SONGS Low Power Natural Circulation Test 4 TEN GALLONS CONCENTRATED BORON SOLUTION ADDED TO SUCTION OF CHARGING PUMPS AT TIME ZERO. +5 h. Th!RD SECOND Pass z FIRST Pass y PASS i v C 0 am 3 FOURTH PASS t y 5 a a: -5 4.8 i MINUTES i I i l l l l l l 0 2-II 6 8 10 12 lit 16 18 TIME (MINUTES) i

llI,

,ll REDRO C ER i Y J t TR ur IE pe VT R nt IU I u TP mAy p CM M hm AO to EC ET R Sg 0 iC2 8 1 w 2 yt r .o C nti . X)f, N A oin P. # ,x. - 0., P .,O M eivU 0 r ti U P u atS f,y\\. g ccG i MeO aN i g3 [_L 6~ F RS r oe = th ,/ ct e 2 to et D R s O"- T R R C5 _ - /C O O A5 T T f" R C R E E E T Z T I E R D R E E D ,I U ,h / S A S I ER = .o P iN _':c \\ f. b P M ( U 4 t r O.!n.. P s ~ . $7(z. \\ /.e'/ ik RO T A R i Gm D ll l lll!llll

Figure 3-11 Simple Locations.for SONGS Boron Mixing Test i ~ N PZR SA5f?LE LETDOWN TO BORONOMETER 4r h O o v o o SG2 SG1 \\ I o u L_ / \\ HOT LEG SAMPLE O o Note: Sample tap on surge line located as close to PZR as possible. O e 48

k Figure:3-12 Concentration Data for.the First Five Hours of the SONGS Boron Mixing Experiment 4 1 j A Hot leg-samples a O Boronometer readings fgepa .,/s o O g5 n.g.o g* ~ v O O i p O J L' it g v 2nd RCS z boration 3 O H 600 54 W o 8. n N z Q Start 1st RCS lat RCS boration i O g boration complete 500 V A 00 0 o 000000 I I i i j 12:00 PM 2:00 4:00 7/27 j

09 BORON CONCENTRATION (ppm) m m m m m a y a o s 2 to e o s 2 o o o o o o o o l l O, a e* g, _d . o. NK = = cs First RCS boration M o o o e a w e n ,O m - $$0 w e 3m c p o e Second RCS boration Ea o

• .p o

s = E E j a I > e M o e 4 ; y$ l EE h 3# -em W r og g e m E'8,w* -W 1 CE U a oo 3 'jO my a li O @[ + Third RCS boration yy O "M a to > g o @K g O m a a O o e,g O

2:o to -

= Fourth RCS boration M g*o e e A p a ) fp e l > o o RCP restart 3 1

4.0 REACTIVITY'ANALYSI_S l i 4.1 Introduction j Current Emergency Procedure Guidelines, Reference 11, require that suffi-cient boron be added to the reactor coolant system prior to or during a I natural circulation cooldown in order to maintain proper technical specifi-t cation shutdown margin. These guidelines require that enough boron be add- ' f ed to the primary loops to borate the entire RCS water mass including the pressurizer in order te prevent a possible. loss of shutdown margin due to outsurges from the pressurizer. (As an alternative, bcron can also be add-ed to the pressurizer via the auxiliary spray system.) Since the RVUH is relatively stagnant under natural circulation conditions, boron mixing in that region is.a slow' process dependent upon diffusion and a small amount of natural-circulation induced flow. As a result, the potential exists in j the event of. a steam bubble formation in the upper head to flush a rela-

tively large and relatively dilute slug of borated water into the reactor coolant system. This -slug can then be' carried through the RCS loops and into the reactor thus adding a momentary spike of positive reactivity to the core.

As part of the SONGS natural circulation cooldown test data evaluation, an analysis was performed to investigate the effect on reactivity of steam bubble formation in the RVUH. - The' analysis involved simple reactivity bal-ances for each of the reactors under -investigation. Three cases were exam- ~ .ined as-follows: Scram from hot full power, scram from hot zero power, and cooldowa from hot. standby. The results are shown in Table 4-1 (p. 52). . Note that in each case xenon was assumed to be completely decayed away ~ thus adding a significant positive reactivity component to the values in Table 4.1. Also note that the size of the water slug flushed from the RVUH was; assumed to be such that the core became entirely deborated. - 4.2 : Conclusions Based upon the results' presented in Table 4-1 for System 80, no return to criticality will occur provided all control rods are fully inserted follow-51

Table 4-1 Reactivity Balance Table

• i System 80 3410 St. Lucie'2 l

Tcchq1 cal Specification Shutdown 6.00% 5.15% 5.00% Margin Assumed Volume of Slug (cubic feet) 1350 900 600 A. Scram From HFP: Net Scram Worth -17.50% -11.00% -11.00% Worst Stuck Rod (9350*F) +5.40% +1.56% +1.56% Loss In Rod Worth During Cooldown +4.64% +3.22% +3.22% Equilibrium Xenon Defect +2.86% +2.97% +2.78% -Temperature Defect (HFP-350*F) +5.15% +5.01% +4.53% Core Reactivity After Dilution +0.55% +1.76% +1.09% 9350*F' B. Scram From HZP: Net Scram Worth -16.50% -10.50% -10.50% Worst Stuck Rod (9350*F) +5.40% +1.65% +1.56% Loss In Rod Worth During Cooldown +3.64% +2.45% +2.45% 'Equilibr.ium Xenon Defect +2.86% +2.97% +2.78% Temperature Defect (HZP-350*F) +2.52% +2.3G% +2.15% Core Reactivity After Dilution -2.08% -1.13% -1.56% 9350*F C. Cooldown From Hot Standby: Initial Shutdown Margin -6.00% -5.15% -5.00% Equilibrium Xenon Defect +2.86% +2.97% +2.78% Temperature Defect (HZP-350'F) +2.52% +2.30% +2.15% Core Reactivity After Dilution -0. 6 2% -0.12% -0.07% 9350*F

• All-values based on first cycle conditions. Reactivity is given in terms of. % ao.

e 52

ing the reactor scram. In the event that the most reactive control rod ~ remains stuck out follow'ing scram, no return to criticality will occur pro-vided steam bubble formation occurs prior to 30 hours or complete mixing of boron has occurred in the RVUH by 30 hours. Note that the 30 hour period is based upon xenon decay since the equilibrium xenon defect shown ~ on Table 4.1 begins to decrease thy shutdown margin only after 30 hours. For the 3410 Class plants an'd St. Lucie Unit 2, no return to criticality will occur provided there are no stuck rods and steam bubble formation occurs prior to 30 hours or complete mixing of boron has occurred in the RVUH by.30 hours. This 30 hour period, as previously stated, is based upon xenon cecay_ and the equilibrium xenon defect influence on shutdown ma rgi n. - A The results presented in Table 4-1 are based upon conservative assumptions particularly-with respect to mixing in the RVUH and the size of tne water slug' flushed from that region during bubble formation, i.e., zero mixing was assumed in the upper head and the entire core was assumed to be comJ plately 'deborated as a result of bubble formation. As demonstrated during the SONGS boron mixing test, see Section 3.3 above, mixing in the upper head region does occur over a period of several hours due to diffusion and - a small-amount of natural circulation induced flow. In addition, the re-sults in Tablu 4-1 envelop the situation where relatively dilute boron is introduced into the RCS due to outsurges from the pressurizer since the . volume of fluid in the pressurizer is smaller than that assumed to be flushed from the upper head. The conclusion then is that the reactor will remain subcritical at all times during a natural-circulation cooldown even though a'.certain amount of boron dilution is possible due to pressurizer outsurges or RVUH steam bubble formation.

l. -

53

l 5.0 REACTOR VESSEL UPPER HEAD COOLDOWN ANALYSIS 5.1. Introdur. tion As was stated in Section 2.2 above, the depressurization portion of the SONGS natural circulation cooldown test proceeded smoothly and rapidly with no indication of reactor vessel upper head steam bubble formation. Analytical predictions, however, submitted as part of the pre-test report of Reference 2 indicated that RVUH steam bubble formation was likely to occur and thus extensive training was given to plant operators on proper handling of such an event. In this respect the computer simulations con-tained in Reference 2 were conservative in that they understated the heat losses from the RVUH in order to show the plant operators the NSSS re-sponse to bubble formation. These heat losses, and the cooling in the upper head which was produced, resulted from the operation of the control element drive mechanism cooling fans. (Refer to Figure 5-1 (p. 62) for a typical CEDM cooling fan arrangement.) In the actual SONGS Natural Circulation Test two factors combined which resulted in a substantial cooling of the RVUH and therefore prevented steam bubble formation during the plant depressurization. (Both of these factors will be discussed in detail in the subsequent sections.) The first factor was a relatively large cooling effect due to the CEDM cool-ing fans in operation. Note that these fans were left running during the entire natural circulation test in order to prevent damage to the rod drive mechanisms. It is also interesting to note that the CEDM cooling fans were operating at St. Lucie Unit I during their cooldown event of 1980, Reference 6, in which a steam bubble was formed. As will be shown, this difference between the SONGS Natural Circulation Test and the St. Lucie Unit i natural circulation cooldown event, i.e., a steam bubble was formed in the St. Lucie RVUH and not in the SONGS RVUH, was due to sig-nificant differences in the time frames over which the plant depressuriza-l tions were performed since the cooling effect of the CEDM blowers for both plants is roughly identical. The second factor which contributed to the substantial cooling in the upper head region of the RCS during the SONGS test was the relatively long period of time, approximately eleven 54

v l I!i hours, during which the plant was maintained in hot standby conditions following trip prior to connencing the cooldown. As will be discussed below,'the RVUH cooled during this time to a temperature below that of 'the remainder of the RCS. In the sections that follow, previous analyses will be presented which show reactor vessel upper head cooldown in the absence of CEDM cooling fans. Since the CEDM cooling fans,sre not safety-grade, credit cannot be taken for-their operation in a BTP RSB 5-1 scenario as explained in' Sec-tion 6.0 below. For comparison, the results will be presented of recent ' analyses performed subsequent to the SONGS natural circulation cooldown test which-show the cooling effect of the CEDM blowers. Note that during the SONGS test on Unit 2, the heated junction thermocouple RVLMS was not installed. -Consequently, dire,ct temperature indication in the upper head was unavailable, and the exact fluid temperature in that region was un-known at the end of the test following plant depressurization, 6:03 AM in Figure 2-1 (p.15). An upper bound, however, was determined for this temperature based upon the final pressurizer pressure. This upper bound, along with data from the St. Lucie Unit 'l cooldown event of Reference 6, was used to benchmark the methodology used to model the effect of the CEDM blowers on the RVUH cooldown. 5.2 Analytical Results 5.2.1-Methodol ogy. Under natural circulation conditions, heat removal from the reactor ves-I sel upper head can occur via a number of mechanisms. These mechanisms in-clude the following: 1. Heat transfer through the RVUH metal and insulation to the containment. 2. Heat transfer to the ' containment through the control element drive mechanisms. 3. ~ Circulation of cooler loop water up into the RVUH. 55 ,,,--e,,#, -,~m- .n.,, ,,.,.,.,,e-.-m-es. m _,..-u.m,v,.,.,,--,4 -rm,...n-,.r

l 4. Heat conduction down fra the upper head to the reactor vessel outlet plenum through the upper guide structure. 5. Heat conduction down through the dome and reactor vessel metal. Although all of the above contribute to the cooling of the upper head, the mechanism which will dominate, i.e., the one which will have the larg-est effect, will depend primarily on whether or not the CEDM cooling fans are in operation. Previous analyses contained in Reference 8 considered the fourth mechanism above to be the major heat removal path with CEDM blowers off. In that work the RVUH was treated as a single uniformly mix-ed node with the upper head metal assumed to be at the same temperature as the upper head water. The heat transfer via conduction from the upper -head metal and water through the UGS to the outlet plenum was based upon calculations using the two dimensional heat conduction code in Reference 9. Although conduction through the upper guide structure plate was the dominant cooling mechanism, natural circulation flow through the UGS and - flow into the upper head as a result of upper head coolant contraction were also modeled. The results of these analyses are contained in Sec-tion 5.2.3, Section 5.2.4, and Section 5.2.5 below for comparison with 4 recent evaluations of RVUH cooldown with CEDM cooling fans in operation. When the CEDM cooling fans are operating helping to remove heat as shown in Figure 5-1, the dominant mechanism for cooling of the upper head under natural circulation is the heat loss from the control element drive mech-anisms. In order to determine a realistic value for this heat loss, field data was used based upon the St. Lucie Unit'l cooldown event of Reference 6. The exact temperature in the upper head at the time bulk boiling occurred was determined using pressure data. From this tempera-ture and the initial RVUH temperature (assumed to be equal to the hot leg f temperature due to forced circulatica from the RCPs), a CEDM cooling ?ffect was calculated. Upper head cooling rates were then determined for other plants using a simple scaling process based upon the numbca of I -CEDMs. l l l 56 e.

In addition to the CEDM cooling effect, recent analyses have looked close-ly at RVUH flow patterns under a variety of' conditions. When the driving potential'ex1_sts, flow from the outlet plenum enters the RVUH through the control element assembly shrouds and holes in the UGS plate and exits through the core suppcat barrel key ways, Figure 5-2 (p. 63). This flow is the result of natural circulation induced flow and shrinkage induced flow due to cooling of the water in the upper head. When the water in the dome region of the RVUH is cooler than the water in the outlet plenum (as would be the. situation if the plant were maintained in hot standby for,an extended period of time with the CEDM cooling fans on), good mix-ing of the fluid in the ' dome region will occur with warmer circulating fluid in the lower regions due to establishing convective cells, i.e., natural bouyancy effects. This situation is shown in Figure 5-3 (p. 64) and was modeled using a single uniformly mixed RVUH node in which upper head' metal and water temperature are a sumed to be We same. When the water in the dome region of the RVUH is warmer than the water in the outlet plenum (as would be the-situation during a plant cooldown with 7 CEDM cooling fans on), mixing of fluid in the dome region with cooler circulating water in the lower regions will be less effective since'the convective cells shown in Figure 5-3 will not exist. For this situation, a two region RVUH model was employed as shown in Figure 5-4 (p. 65) with each region assumed to be uniformly mixed. The boundary between the two regions then moves upward as the watcr in the dome cools and shrinks due cto CEDM heat removal. As a conservatism, no mixing across this interface-is allowed. :RVUH metal and water are assumed to be at the same temper-i ature with a certain amount of heat conduction through the upper guide -structure to the cooler outlet plenum modeled. -5.2.2 RVUH Temperature During the. SONGS Natrual Circulation Test Using the methodology and models discussed above (CEDM cooling fans in ~ operation), a simulation of the SONGS natural circulation test was per- ,7 formed in order to show the _ temperature behavior in the reactor vessel . upper head. The results are shown in Figure 5-5 (p. 66). Also shown in this figure is the saturation temperature corresponding to the pressuri- . zer pressure at each point in the test. Initially, cooling of the fluid 57 {.

~. i lI' .in the reactor vessel upper head was dominated by the heat loss through the control element drive mechanisms with the CEDM cooling fans removing this heat as shown in Figure-5-1. As a result, upper head temperatures ' fell quite. rapidly at first, approximately 25'F per hour, until they de-creased below that of the RCS hot leg. As the cooling fans continued to remove heat, convective cells were established between the warm circulat-ing water predicted in the lower regions of the RVUH and the cooler water in the upper stagnant dome region. (This situation with the convective cells shown is depicted in Figure 5-3.) By approximately twelve hours -after trip, a steady-state condit; ion existed as seen from Figure 5-5 in which the heat loss from the RVUH via the contrcl element drive mechan-isms was balanced by the heat input from the convective cells shown in Figvre 5-3 and the fluid temperature in the upper head stabilized. At approximately twelve hours after trip as shown in Figure 5-5 the plant cooldown was comenced. As RCS hot leg temperature decreased below that of the RVUH the convective cells shown in Figure 5-3 no longer existed and the. upper head continued to rapidly cool via CEDM heat losses. Note that at the conclusion of the test following the plant depressurization, almost 100 degrees of subcoolieg existed in the upper head. In contrast to the SONGS Natural Circulation Test, a steam bubble was formed in the RVUH during the St..Lucie cooldown event described in Refer-ence 6. This' difference was due to the time over which the plant depres-surization was conducted at St. Lucie such that the heat removed by the CEDM cooling fans was less than the amount required to prevent reaching saturation conditions in the upper head. Looking closely at the exact g - time frames and sequence of events from Reference 6, RCS depressurization was comenced within one hour of plant trip by the Florida Power & Light operators so that when saturation conditions were reached in the RVUH, - the CEDM cooling fans-had been removing heat.for less than four hours. E During the SONGS test, as seen in Figure 2-1 (p.15), the CEDM cooling l: fans had been removing heat from the upper head for over sixteen hours before the plant depressurization was even commenced. As a result, l-enough heat had been removed from the SONGS upper head by the end of de-pressurization to prevent reaching saturation conditions. l 58

L' I 5.2.3 3410 Class RVUH Cooldown Using tne. methodology and models discussed in Section 5.2.1 above, an evaluation was performed of the reactor vessel upper head cooldown under natural circulation renditions for the 3410 Class plants. This evalua- . tion included an analysis of upper head cocidown with CEDM cooling fans in operation and an analysis of uoper head cooldown with CEDM cooling -fans secured. The results are shown in Figure 5-6 (p. 67) and are applic-able to both the San Onofre Units 2 and 3 and the Waterford Unit 3 de-si gn. As can be seen, the CEDM blowers are predicted to have a substan-tial cooling effect on the RVUH. These fans are not available, however, during the Branch Technical Position RSB 5-1 scenario. Therefore, cool-ing.of the RVUH in this situation would be a slower process than that observed during the SONGS natural circulation cooldown test. Section 6.2 below contains a detailed discussion which shows compliance of SONGS Units 2 and 3 with the requirements of the branch technical position and Section 6.3 below contains a discussion of the applicability of the SONGS test to Waterford Unit 3. As will be presented, sufficient seismic Category I condensate is available to first cool the plant to a hot leg temperature of less than 350*F (shutdown cooling system entry tempera-ture) and then maintain this temperature until the. upper head has cooled 3 -sufficiently to allow plant depressurization to SCS entry pressure with-out-forming a RVUH steam bubble. Consequently, comniiance with the re-quirements of BTP h3B 5-1 for the 3410 Class plants is a relatively straight forward process which can be accomplished without the loss of subcooling in the upper head. 5.2.4 St. Lucie 2 RVUH Cooldown Using the methodology and models discussed in Section 5.2.1 above, an f evaluation was performed of the reactor vessel upper head cooldown under natural circulation conditions for St. Lucie Unit 2. This evaluation included an analysis of upper head cooldown with CEDM cooling fans in operation and an analysis of upper head cooldown with CEDM cooling fans secured. The results are shown in Figure 5-7 (p. 68). As can be ::cen, l ~ the CEDM bituers are predicted to have a substantial cooling effect on L !l' 59

f l l the RVUH. These fans are not available, however, during the Branch Tech-nical Position RSB 5-1 scenario. Therefore cooling of the RVUH in this situation would be a slower process than that observed during the SONGS natural circulation cooldown test. Section 6.2 below contains a detailed discussion which shows compliance of SONGS Units 2 and 3 witn the re-quirements of the branch technical position and Section 6.4 below con-tains a discussion of the applicability of the SONGS test to St. Lucie Unit 2.. As will be presented, sufficient seismic Category I condensate is available to first cool the plant to a hot leg temperature of less than 350*F (shutdown cooling system entry temperature) and then maintain this temperature until the upper head has cooled sufficiently to allow plant depressurization to SCS entry pressure without forming a RVUH steam . bu bbl e. Consequently, compliance with the requirements of BTP RSB 5-1 is i-a relatively-straight forward process for St. Lucie Unit 2 which can be accomplished without the loss of subcooling in the upper head. 4 5.2.5 System 80 RVUH Cooldown 4 { Using the methodology and models discussed in Section 5.2.1 above, an evaluation was performed of the reactor vessel upper head cooldown under { natural circulatica conditions for System 80. This evaluation included an analysis of upper head cooldown with CEDM cooling fans in operation and an analysis of upper head cooldown with CEDM cooling fans secured. The results are shown in Figure 5-8 (p. 69) and are applicable to the l 'Palo Verde Units 1, 2, and-3 design. As can be seen, the CEDM blowers are predicted to have a substantial cooling effect on the RVUH. These fans are not available, however, during a Branch Technical Position RSB 5-1 scenario. Therefore, cooling of the RVUH in this situation will be a l sinwer process than will be observed during the Pa?o Verde natural circ-ulation cooldown test. (Note that during the APS Natural Circulation Test the CEDM blowers will remain energized throughout the entire test to prevent equipment damage.) Because of the size of the System 80'RVUH, 1.e., approximately 2000 cubic feet as compared to approximately 1000 cubic feet in the pre-System 80 design, and because of the time frame required to cool the upper head in the absence of void formation, a reactor vessel upper head steam bubble is intentionally formed in the 60

System 80 plant in crder to show compliance with BTP RSB 5-1. Intention-ally forming a steam bubble in the upper head along with concurrent use of the reactor vessel gas vent system shortens the plant depressurization process and allows shutdown cooling system entry conditions to be attain-ed prior to depleting available seismic Category I condensate inventory. Note that Reference 10 contains a computer simulation of a strict BTP RSB 5-1 scenario for System 80 showing the process of steam bubble formation in the RYUH with suosequent use of the RVGVS to aid in plant depressurization. 5.3 Conclusions An evaluation of the reactor vessel upper head cooldown under natural cir-culation conditions was performed using the methodology and models discus-sed in Section 5.2.1 above. This evaluation included an analysis of up-per head cooldown with CODM fans in operation and an analysis of upper head ccoldown with CEDM fant secured and was performed for the C-E 3410 Class plants (SONGS Units 2 and 3 and Waterford Unit 3), St. Lucie Unit 2, and for System 80. The results are shown in Figure 5-6, Fioure 5-7, and in Figure 5-8 and indicate that the CEDM fans have a significant cool-ing effect on the RVUH. - For the situation in which the CEDM cooling fans are secured, as would be the case in a Branch Technical Position RSB 5-1 scenario, cooldown of the upper head in the absence of steam bubble forma-tion in that region would require a hold period of approximately 15 hours in the pre-System 80 design and a hold period approximately 50 hours in the System 80 design following cooldown to SCS entry temperatures. There-fore, as indicated in Section 6.0 below, compliance with the requirements of BTP RSB 5-1 for pre-System 80 plants is a relatively straight forward process which does not involve forming a steam bubble in the RVUH, and compliance'with the requirements of BTP RSB 5-1 for System 80 involves forming a steam bubble in the upper head in order to expedite the depressurization process. 61

i Figure 5-1 Typical CEDM Cooling Fan Arrangement = I f CEDM cooling fans Cavity cooling s s system i e' i a -l s. N, / / 'h ( / u / / .1 4 / + 1 / / / / 62

Figure 5-2 Reactor Vessel Flow Paths Under Natural Circulation Conditions Alignment keys (4) / _N /i \\ \\ kj (uas u s g 7 C ,o - = ___-_-e L s FAP / 6 t6 I ,/,- a [ I 63

i Fiqure 5-3 RVUH Flow Paterns with CEDM Cooling Fans Running and Hot Standby Conditions in the RCS SCEDM W b' o q u b l a u I \\ tr o a UGS) l \\ --J'i W i b i lj -~ _~ I Ib y FAP 1 U s p* a s9 .p l 1 t J I 64

Figure 5-4 Thermally Stratified RVUH During RCS Cooldown with CEDM Cooling Fans Running m S CEDM l; 9 Region 1 Thermally s'. ratified layer _ Region 2 L-n s / -4s 6 b j I 4 i i k 4 sh'p plf 89 e p f j ', L 1 65

Figure 5-5 RVUH Temperature During the SONGS Natural Circulation Test l l 650 Tpzr. 6 \\ \\ T 550 hot g . fa. w C / ~ g Trvuh E EF 450 .350 4 8 12 16 20 TIME AFTuR TRIP (HOURS) 66

a e t 1 0 i 3 N 5 i 2 a n ) gw S no R id Ms U rl Dn 0 O i uo rEa 2 l l Dos eCf ( Ct t e n at g P. 3 rna wu n I - uol o i t R 5 tiP Hh l o T 2 at Ut o 'h e ras Vi o R r els Rw c [ E u pua T g ncl F n i erC s 5 A i F Ti n 1 C0 r a E r 1 eWf M el4 tD I t a3 aEg T ar wCn Wu i t ih l l Ha Ut o \\ UN Vi a [\\ V Rwc R' 0 i 1 i 5 L 0 0 o 0 0 o 6 5 3 Cv$EaMkI JE m"

Figuro 5-7 RVUH Water Temperature During a i-Natural Circulation Cooldown St. Lucie 2 j 600 k i i n i . m k. v 500 RVUH water without CEDM cooling fans 6 I s4 RVUH water 400 with CEDM 1 cooling fans i gn ( hot I 300 l l l l l l i 5 10 15 20 25 30 TIME AFTER TRIP (HOURS) 4 1

0 g 6 4 0 l 5 a ) Ms n' Dn t S gw rEa o R no id eCf h U 0O I rl t 4H [ at g uo ( Do wu n C o i P e Hh I 8 rn Ut Vi s R uo0 5 ti8 Rw n, T r a at emf. R s ram tD E r el e aEg T u put wCn 0F l g mcs i 3A i ery Hh l F tis Ut o E C Vi o M r I R w c-el T ta ar Wu t Xa UN V 0 p, \\ i R 2 0 l 1 0 0 0 0 0 0 0 0 6 5 4 3 ,A v $ob$m@Nb Dm mo i i; i l; .} l

i

6.0 COW LIANCE WITH THE TESTING REQUIREMENTS OF BTP RSB 5-1 6.1 Introduction The design requirements for the residual heat removal system as given in Branch Technical Position RSB 5-1 state that all systems which must be used to take the reactor from normal operating conditions to cold shut-down shall satisfy the following: i l

1. The design shall be such that the reactor can be taken from normal operating conditions to cold shutdown using or.ly safety-grade systems. These systems shall satisfy General Design Criteria 1 through 5.

2. The system (s) shall have suitable redundancy in components and features, and suitable interconnections, leak detection, and isolation capabilities to assure that for onsite electrical power system operation (assuming offsite power is not available) and for offsite electrical power system operation (assuming onsii;e power is not available) the systec function can be accomplished assuming a single failure.

3. The system (s) shall be capable of being operated from the control room with either only onsite or only offsite power available.

In demonstrating that the system can perform its function assuming a single failure, limited operator action outside of the control room would be considerad acceptable if. suitably justified. '4. The system (s) shall be capable of bringing the reactor to a cold shutdown condition, with only offsite or onsite power available, within a reasonable period of time following shutdown, assuming the most limiting single failure. 70

i Detailed review of the various systems required to bring the plant to cold shutdown have been performed by the NRC and the results can be found in the individual plant Safety Evaluation Reports. Based upon this re-

view, the Staff has concluded that the requirements of BTP RSB 5-1 have been met._ The Staff also required that an actual demonstration in the form of a plant test or tests be conducted to confirm the conclusions l-that-the branch' technical position is indeed met.and to address certain l.

additional conccens. With respect to the testing in support of BTP RSB 5-1, tho Staff requires that adequate tests with supporting analyses be _ performed to confirm that sufficient mixing of borated water added prior

to or during cooldown'can be achieved under natural circulation condi-tions and to. permit estimation of the times required to achieve such mix-ing. In addition, the Staff has requested specific evaluations based upon actual test data with respect to feedwater usage and the adequacy of the seismic Category I condensate supply, reactor vessel upper head cooldown under natural circulation conditions, and the adequacy of the safety-grade nitrogen supply to the ADVs.

The Combustion Engineering nuclear steam supply systems can be divided essentially into two' categories with respect to BTP RSB 5-1, pre-System 80 and System 80. The most significant distinction between these two categories is the relative size of the reactor vessel upper head, i.e., the volume of the RVUH in the ' pre-System C0 design is approximately 1000 cubic feet with the volume of the Sys'.em 80 upper head being about twice I 'as = large or 2000 cubic feet. Because of this difference, as will be detailed below, the basic method or scenario which will be followed in p . order to meet the requirements of the branch technical position is differ-l 'ent for the two. plant categories. Specifically, in order to meet the Staff's requirements with respect to RSB 5-1 (these requirements being that-the plant can be maintained in hot standby for four hours followed .by a cooldown to cold shutdown conditions with a loss of offsite power L . and an-assumed single failure)-intentional formation of a steam bubble in the upper head in the pre-System 80 design is not requi;'ed. To meet the Staff's requiremen's with respect to RSB 5-1 with the System 80 design, i however, a steam bubble must be intentionally fewd in the upper head in i order. to expedite RVUH cooldown to allow initiation of the shutdown cool-L ._,_-4 ..,_,___._.,_,,__..,_.,,.,.._,,.,,___.___m.

'ing system. Section 5.2 below explains in detail how the process of ob-taining cold shutdown is-accomplished for SONGS Units 2 & 3, these plants being prototypical of pre-System 80 plants, and Reference 10 explains in detail how the process is acconplished for System 80. 6.2 ~ Compliance by the San Onofre Nuclear Generating Station Units 2 and 3

The analyses and testing in support of Branch Technical Position RSB 5-1, as detailed in Reference 2, was met by Southern California Edison through l-the performance of essentially five-discrete tests.- Each of these tests L

was assembled into a single program, the SCE Natural Circulation Test

Program, and approved for performance by the NRC Staff. Since signifi-cant safety conferns were raised with respect to performing a single test

-which demonstrated a strict BTP RSB 5-1 scenario, an approach was develop-ed which utilized relevant portions and data from a number of tests which

then could be brought together ~and referenced in demonstrating compliance with the branch technical; position. The first three of the five tests from which data was obtained by SCE were performed during the week of 5 September 1982 on SONGS Unit 2 and were termed the "A" series tests.

' These tests were conducted with RCPs secured and the reactor critical in _or' er'to simulate decay heat. Information that was obtained during these d tests and used_to address the Staff's concerns regarding BTP RS8 5-1 is detail'ed in Table 6-1 (p.'73)'. The remaining three tests from which d6ta was obtained were performed in July of 1983 and were termed the "B" series tests. These tests were performed with RCPs secured using actual decay heat. (Again, refer to Table 6-1 for a list of the information relative to BTP. RSB 5-1 that was obtained during the "B" series tests.) i I-Reference 2 listed several specific items which the NRC wanted addressed during the SONGS natural circulation cooldown test and as part of the . post-test data evaluation. - These items included a demonstration of the ability to mix _ boron under natural circulation with an estimation of mixing times, an evaluation of RVUH cooldown rates, an evaluation of feedwater usage with an assessment of the adequacy of the seismic Category I condensate supply, and an evaluation of the adequacy of the safety-grade nitrogen supply to the atmospheric dump valves. In order to 1 72

Table 6-1 Natural Circulation Test Program San Onofre Nuclear Generacing Station Units 2 and 3 Information Obtained With Test Respect to BTP RSB S-l' Natural Circulation Demonstration Demonstrate basic natural circulation -(Test A1) capability of plant. Flow and thermal-hydraulic data for exact natural circu-tL lation flow calculations. Boron mixing information. Natural Circulation at Reduced Demonstrate natural circulation sensi-Pressures (Test A2) tivity to pressure variations. P res - surizer ambient heat loss information, auxiliary spray efficiency data. L Natural Circulation with Reduced Demonstrate natural circulation with one H:at Removal Capacity (Test A3) steam generator isolated. Boron mixing information. Loss of Offsite Power Demonstrate ability to maintain plant in -(Test 81 & B2) . hot' standby following actual lossaof offsite power. Natural Circulation Test. Demonstrate plant response to a total (Test B3) loss 'of flow. Ability to mix boron and perform a cooldown under natural circula- [ tion conditions. Data on RVUH cooldown, l feedwater usage, and nitrogen usage by the ADVs. Demonstrate manual operation of ADVs. r 73

i properly address each of these items and to show'that San Onofre Units 2 and 3 can comply.with the requiraments of the branch technical position, -operator actions necessary to take the plant from normal operating condi-tions to cold shutdown will be discussed. All systems and components which will be used are safety-grade.and moet the functional requirements -of BTP RS8 5-1. In addition to operator actions, plant response to the event will_be discussed and relevant test data with supporting analysis will be presented as appropriate. To begin the discussion of the process of taking the plant from normal -operating conditions to cold shutdown, the plant is assumed to be operat-ing at 100% of RTP at normal system temperature and pressure. (Referto Table 6-2 '(p. 75) for a sumary of the operator actions required to per- -form this evolution.) Further, the plant is assumed to have been operat-ing at this power level for approximately six hundred days in order to build up the maximum amount of decay' heat possible. Offsite power is lost at time zero; the plant trips and goes into natural circulation. (Note that Section 5.4 of Reference 2 contains the results of a full scope computer simulation of a strict BTP RSB 5-1 scenario showing initial plant response to the event.) Main feedwater to the steam gener-ators will be lost upon loss of offsite power at the beginning of the event. Emergency feedwater flow will be automatically initiated when 1 generator level reaches approximately 23% (narrow range). At this point, plant. operators.will take' manual control of the EFWS-and slowly recover -level in order to prevent overcooling. Also,' manual control of the atmosphric-dump valves will be taken and hot standby conditions will be . maintained. Finally, as part of the initial response to the loss of off-site power, control signals will be sent to start both emergency diesel generators. One of these generators is assumed to fail, however, disabl-ing one entire emergency 1E power train. ._ Atithis point in the event,_ the operators have taken all actions neces-sary to stabilize the plant in hot standby and to establish natural circu-lation. 'The fact.that-the reactor coolant system is indeed operating under subcooled natural circulation conditions will be verified and, .according to the requirements of BTP RSB 5-1, hot standby conditions will j 74

Table 6-2 Basic Operator. Actions Necessary to Comply With BTP RSB 5-1 at SONGS 2 8 3 ' Event Operator Action Discussion I. Loss of offsite power, 1. Take manual control of AFWS.- 1. Limit SG refill following plant trip. trip to prevent overcooli;q. gj 2. Take manual control of ADVs. 2. Prevent further lifting of Monitor safety-grade nitrogen secondary safety valves. supply. Control SG pressure at hot zero power. 3. Establish and verify natural 3. Criteria for verifying natural circulation in the RCS. circulation as follows: o Loop AT iess than normal full power ALT. O Cold leg temperatures constant or decreasing. 0 Hot leg temperatures stable or decreasing, o No abnormal differences between hot leg RTDs and CETs.

Table 6-2 (cont.) Event l Operator Action Discussion I. Loss of offsite power, 4. Maintain 20*F of loop 4. Ensure subcooled natural plant trip - cont. subcooling via charging and circulation flow. available pressurizer heaters. ? II. Four hour hot standy

1. LMaintain SG level via manual 1.

Ensure subcooled natural j period. nr automatic control of AFWS. circulation flow. 9 Monitor CST level. o> 2. Maintain pressurizer level 2. Letdown is not available. via manual control of CVCS. 3. Maintain SG pressure via ADVs. ] Monitor safety-grade nitrogen supply. i 4. Control RCS temperatures to 4. 50ron concentrations will be j . maintain hot standby monitored and proper shut-j conditions. down margin will be maintained ) 5. Maintain 20*F loop subcooling 5. Ensure subcooled natural via charging and available circulation flow. i pressurizer heaters, i l =

Table 6-2 (cont.) Event Operator Action Discussion II. Four hour hot standby 6. Monitor RVUH temperature 6. RCS pressure should be period - cont, via HJTC system. maintained above saturation pressure in the upper head for entire evolution. 7. Determine boren requirements for cooldown. III. Flant cooldown 1. Commence plant cooldown 1. Cool down at maximum rate at 175*F per hour via possible to maximize RVUH ADVs. Monitor safety-cooldown and boron mixing. grade nitrogen supply Minimize ADV movenent in order to conserve nitrogen. 2. Observe plant pressure / temperature limits. 3. Charge as necessary to 3. Minimize pressurizar level and maintain PZR level due hence pressure transits since to contraction. only limit heaters are avail-able.

Table 6-2 (cont.) Event Operator Action Discussion III. Plant cooldown - cont. i 4. Secure available pressurizer 4. Allow RCS pressure to decrease heaters. ' Energize only as due to ambient heat losses 3 necessary to maintain 20 F from the pressurizer. This subcooling in RCS loops and will minimize the amount of RVUH. auxiliary spray required to perform the final plant de-pressurization. 9* 5. Borate concurrent with 5. See No. 3 above. l cooldown as part of normal system make-up. 6. Sample for boron concentration 6. Ensure mixing ana proper shut-down margin. 7. Maintain SG 1evel via manual 7. Ensure adequate heat sink. or automatic control of AFWS. 8. Monitor RVUH temperature 8. RCS pressure should be via HJTC system. maintained above saturation pressure in the upper head for entire evolution. i 9. Ccol down to a hot leg 9. SCS entry temperature. temperature of 350 F or less. i

+ Table 6-2 (cont.) Event Operator Action Discussion IV. RVUH cooldown period. 1, Maintain hot leg temperature 1. Allow RVUH to cool down at 350*F or less for to a temperature less than approximately 12 hours. the saturation temperature corresponding to SCS entry pressure. 2. Maintain SG pressure via 2. Minimize ADV movement to ADVs. Monitor safety-grade conserve nitrogen. If nitrogen supply. nitrogen supply sheuld become a depleted, operate ADVs via manual handwheels. 3. Maintain SG 1evel via 3. Ensure adequate heat sink. manual or automatic control of AFWS. Monitor CST level. 4. Maintain 20*F of loop 4. Ensure subcooled natural subcooling via charging and circulation flow. Continue to available. pressurizer heaters. allow RCS pressure to decrease via ambient losses from the pressurizer. Maintain subcooling in both RVUH and RCS loops. 5. Monitor RVUH temperature via 5. Follow upper head cooldown HJTC system. progress.

Table 6-2 (cont. ) ' Event Operator Action-Discussion IV. RVUH cooldown period - 6. Maintain pressurizer level cont. 'via manual control of CVCS., V. Plant depressurization. 1. Following RVUH cooldown, 1. PZR cooldown to be commence RCS depressurization limited ta technical via the auxiliary spray specification limit. system. Monitor for void .o formation. 2. Lower system pressure to the point where the SCS can be initiated. I l 3. Observe plant pressure / temperature limits. 4. Maintain SG level via 4. Ensure adequate heat sink. manual or automatic j control of AFWS. 5. Maintain SG pressure via manual control of ADVs. l l l

, _ Table 6-2 '(cont. ) Event Operator Action Discussion V. Plant depressurization - 6. Maintain 20*F of loop 6. Ensure subccoled natural cont. subcooling via charging circulation flow, and available pressurizer heaters. t VI. SCS initiation. 1. Initiate shutdown cooling .co system. s 4 2. Continue cooldown to cold shutdown. 3. Observe plant pressure / temperature limits. 4 l 4 i s l

i be maintained for the next four hours. During this period, steam genera- -tor pressure and hence RCS temperature will be maintained by manual control of the atmospheric dump valves. Steam generator water level will r -be maintained by the auxiliary feedwater system operating in either manual or automatic. Following the four hour hot standby period, a plant cooldown to a hot leg temperature of less than 350*F will be commenced. The cooldown rate will be controlled at less than or equal to 75'F per hour by regulating the ADVs-and system inventory will be maintained by charging via the CVCS. _ Sinct ly limited IE pressurizar heater capacity (150 kw from the operable emergency power train) is available, plant operators will charge as necessary to maintain pressurizer level during the cooldown due to system contraction. Tnis action will minimize pres-sure changes due to changes in pressurizer level..In addition, boration l will be accomplished concurrent with the cooldown as part of normal inven-tory make-up. Simultaneous boration_ and cooldown are necessary again to minimize pressurizer level transients since letdown is not available. As was shown in Section 3.0 above. -boration can be accomplished concurrent with cooldown and boron mixing times are such that complete mixing of added boron can be accomplished within approximately sixty minutes. Once the RCS hot leg temperatures have decreased to less than 350*F, SCS entry temperature, the plant cooldown will be stopped. Loop temperatures will then be stabilized and maintained at shutdown cooling system entry conditions until the upper head has cooled to less than 438'F. As was .shown in Section 5.2.3 above, the time required to cool the upper head in the absence of CEDM blowers will be approximately 15.5 hours after the start of cooldown, assuming the hot leg temperature was cooled to 330*F. l - Following this hold period, the plant can be depressurized and placed on

shutdown cooling without forming a steam bubble in the reactor vessel upper head. The depressurization will be accomplished in essentially two steps. First, the available pressurizer heaters will be secured shortly H

after the start of the plant cooldown and then energized only as neces- . sary to maintain approximately 20*F of subcooling in the upper head. Pressure will. therefore decrease slowly during this period due to ambient . losses from the pressurizer. Second, the final depressurization required to lower plant pressure to the point where the SCS can be initiated will 82

L be? accomplished using the auxilicry spray systen once the upper head has cooled to less thar. 438'F. Note that in the unlikely event the auxiliary spray system is not available, the final plant depressurization could be accomplished using the safety-grade reactor coolant gas vent system. A _ simplified drawing of this' system is~ shown in Figure 6-1 (p.101) and a comparison of the depressurization rates achievable using auxiliary spray and using the pressurizer vent is shown in Table 6-3 (p. 84). Once the plant has been cooled down and depressurized to the point where the shut- - down cooling system has been placed in operation, core decay heat removal and continued plant cooldown to cold shutdown is accomplished by removing heat at the shutdown cooling heat exchanger with heat removal at the steam generators no longer required. As stated above, four specific items were identified by the NRC to be addressed in the follow-up evaluations to the SONGS natural circulation test. These items were an evaluation of feedwater usage, RVUH cooldown, B safety-grade nitrogen supply to the atmospheric dump valves, and a boron mi xi ng _ analysis..The results of the boron mixing analysis were presented in Section 3.0 above. As was stated in that section, complete mixing of boron 'in the RCS can be expected to occur within approximately sixty minutes following addition. The results of the RVUH cooldown analysis were-presented in Section 5.0 above. For the situation where CEDM blowers are inoperable, as would be the situation during a BTP RSB 5-1 scenario.- the RCS can be taken from hot standby to cold shutdown con-ditions without forming a steam bubble in the upper head. With respect to feedwater usage, one of the objectives of the SONGS natural circulation cooldown test was to provide data sufficient to demonstrate-the adequacy of the seismic Category I condensate supply. Current plant technical ~ specifications' require that 424,000 gallons of. water be available at all times during Mode 1. Of this amount, 80,000 gallons is potentially unavailable due to piping configuration such that a total of 344,000 gallons can be used in order to meet the requirements of the branch' technical position. Further, the bases section of the SONGS technical specifications states that condensate storage require-ments are specified such that the plant can be maintained in hot standby 83

Table 6-3 Comparison of Depressurization Rates Using the Auriliary Spray System vs the Pressurizer Vent System for SONGS Units 2 & 3 Initial RCS Pressure Depressurization Rate (psi / min) ~ '(psia) Auxiliary Spray (1) PZR Vent (2) 2000 53 10 1500 46 8 1000 33 5 500 18 3 1.' Initial pressurizer level 33"., auxiliary spray flowrate equal to 44 gpm at 120*F, i.e., letdown not available.

2. Initial pressurizer le. vel equal to 33%, pressure vent flow area equal to 7/32 of an ine.h.

84

-a y conditions for twenty-four hours with steam discharge to the atmosphere following a loss ~ of offsite power. Note that these requirements are more strict than those of BTP RSB 5-1, i.e., more condensate is required to maintain hot standby for twenty-foar hours with a loss of offsite power than to maintain hot standby for four hours followed by a cooldown to cola shutdown with a loss of offsite power and an assumed single failure. Figure '6-2 (p.102) shows feeowater usage based upon condensate storage ~ tank level for the entire SONGS natural circulation test. As shown, approximately 140,000 gallons of water from the CST was required over the twenty hour period during which the test was conducted. This water was - required to accomplish the following three functions:

1) Decay heat removal during the entire test. 2) Sensible heat removal during the plant cooldown.

3) Inventory makeup in the steam generators as part of operator actions to recover level following the initial plant trip. The analysis to demonstrate the adequacy of the condensate supply must there-fore be' performed in essentially three parts. First, a breakdown of the water required for each of the above functions must be made. Second,

-this breakdown must then be adjusted upward to reflect differences in actual test decay heat level and the maximum decay heat level that could exist based upon the standard decay heat curve. Finally, the adjusted -values must then be added and the results compared to the 344,000 gallons of available condensate. A breakdown of the three basic functions for which water was supplied to the steam generators is shown in Table 6-4 (p. 86). Itc.i 1 in that table represents the difference in the steam generator inventory between 80% . power and hot zero power, Item 2 represents the feedwater that would be -required to cool the entire RCS metal and water mass from a cold leg temperature of approximately 545*F to a cold leg temperature of approx- 'imately 334*F taking into account ambient losses, and Item 3 was obtained ~ by subtracting Items 1 and 2 from the total feedwater usage of 140,000 gallons. The second step in the process now is to adjust the feedwater required to remove decay heat, Item 3 in Table 6-4, upward to reflect the -difference in tha actual test decay heat levels and the standard decay heat curve. Figure 6-3 (p. 103) shows the graph of decay heat level 85

n-. Table 6-4 Breakdown of Condensate Usage During the SONGS Natural Circulation Test Function Water Usage 1. Inventory makeup as part of 16,000 gallons steam generator level recovery. 2. Sen*ible heat removal during 24,000 gallons plant cooldown. 31 Decay heat removal during 100,000 gallons

• entire test.

"'Itein 3 was 'obtained by subtracting Items 1 and 2 from the total feedwater usage of 140,000 gallons. I 3 e 86 . ~ - -

versus time. Curve I in that figure is a plot of standard decay heat and Curve II'is a plot of the actual decay heat levels that existed during p the SONGS Natural Circulatics Test based upon measure loop differential ' temperatures. Integrating the area under Curve I from 10 minutes to 1200 minutes and dividing by the area under Curve II from 10 minutes to 1200 -minutes yields a multiplication factor. of 2.06, i.e., over twice the amount of heat would have to be removed following a trip from 100;. of RTP - with full-decay heat than was removed during the actual natural circula-tion test. _ The third step then is to calculate the final adjusted feed-water usage. This is done in Table 6-5 (p. 88). Note that the result. 246,000_ gallons, is 98,000 gallons less than the total condensate avail-able as specified in Reference 3. l The final item identified by the NRC Staff for post-test evaluation was the adequacy of the safety-grade nitrogen supply to the atmospheric dump valves. As previously seen, valve position and accumulator pressure during the cooldown portion of the SONGS test is _ presented in Figure 2-5 (p. 19). At the ' average rate of usage noted during this portion of the - i test,17.7 hours would be required to completely discharge an accumulator charged to an initial nominal pressure of 1100 psig. Since nitrogen usage is greatest during a cooldown.. i.e., a relatively'large amount of nitrogen is required to move an ADV to a new position in comparison to the small amount of leakage that is normally present with no movement, enough nitrogen capacity should be available to perform an entire BTP RSB .5-1 evolution. Note in the event that the nitrogen supply should become ~ depleted, manual local control via manual handwheels is possible as demonstrated during the SONGS test, see Section 2.2 above. i: I 6.3 Applicability of SONGS Test to Waterford 3 <6.3.1 Introduction During the course of the NRC Staff's review of the Waterford Unit 3 shut-down cooling system, see-Section 5.4.3 of Reference 14, Louisiana Power & ' Light was requested to demonstrate how the requirements of Branch Techni-cal Position R58 5-1, " Design Requirements of the Residual Heat Removal I 87 L

Table 6-5 Adjusted Condensate Usage ' Functi on* ' Water Usage Item 1 16,000 gallons Item 2 24,000 gallons t

Item 3 (100,000 x 2.06) 206,000 gallons Total 246,000 gallons IRefer to Table 6-4.

? I. I l-p- i 88

System," have been met. Specifically, LP&L was asked to demonstrate that the plant could be brought to the point of SCS initiation in less than 30 hours using only seismic Category I equipment, assuming the most limiting single failure, and with only onsite or only offsite power available. Louisiana Power & ' Light, in its response to the above request, identified tne systems which would be used to meet the Staff's requirements. Follow-ing its review of this information and satisfactory resolution of all out-standing issues identified in Section 5.4.3 of Reference 14, the NRC Staff concluded that the capability to achieve cold shutdown for Water-ford 3 meets Branch Technical Position RSB 5-1. The Staff did note, however, that LP&L's planned natural circulation test did not include _ provisions for demonstrating adequate boron mixing when forced circula-tion is not present. Instead, LP&L referenced the boron mixing tests to 'be performed.at San Onofre which was found to-be acceptable due to design similarities. The Staff did require that Waterford submit a review of the SONGS test and demonstrate acceptability and applicability. ' As was stated in the introduction above, see Section 6.1, Combustion Engineering nuclear steam supply systems can be divided into two cate-gories with respect to the requirements of Branch Technical Position RSB 5-1. ' These categories are pre-System 80 and System 80 with the distinc-tion beingLthe relative size of the reactor vessel upper head, i.e., the -volume of the RVUH in the pre-System 80 design being approximately 1000 cubic feet with the volume of the System 80 upper head being about twice as_large or 2000 cubic feet. Accordingly, the _ procedure that would be employed by Waterford Unit 3 to meet RSB 5-1 would be identical to that employed by-San Onofre. In other words, following plant cooldown to SCS initiation te:nperature, system pressure would be maintained relatively high for a fifteen' hour. hold period until the RVUH had cooled sufficient- -ly to depressurize the RCS and place the plant on shutdown cooling with-out forming a steam bubble in the upper head.

Since the ability to cooldown the C-E NSSS under natural circulation con-ditions has been demonstrated three times on two different plants (once during the SONGS' Natural Circulation Test and twice at St. Lucie Unit 1) 89

s and since the scenario that is used by Waterford Unit 3 to meet the re- 'quirements of BTP RS8 5-1 is straight forward and does not involve form- 'ing a steam bubble in the reactor vessel upper head, the arguments for applicability of the SCE test to Waterford will concentrate on the simi-larity in design between the two plants with respect to four specific -areas. These areas are the four areas for which information was request-ed of 3CE in Reference 2 and include boron mixing, reactor vessel upper . head cooldown, feedwater availability, and nitrogen supply to the atmospheric dump valves. ._6.3.2 Comparison with SONGS Test P.esults Section 3.0 above contains the analysis of boron mixing under natural circulation conditions including a discussion of the mechanisms for mixing and an evaluation of the boron mixing test data from the SONGS Unit 2 test. Since both plants are of the 3410 Class with identical RCS piping configuration, the three mechanisms which were found to effect boron mixing in the San Onofre NSSS, as presented in Reference 2, will function-in an identical manner to ensure complete mixing for Waterford. ..These mechanisms are the substantial mixing.of boron with fluid in the cold leg; common node interaction in the reactor vessel inlet plenum, -downcomer, and vessel bottom with fluid rotation in the downcomer; and elongation by the steam generator U-tubes. Therefore, the results of the boron mixing evaluation along with the estimation of boron mixing times contained in Section-3.0 are directly applicable to Waterford 3. Section 5.0 above contains the detailed analysis of RVUH cooldown under ~ snatural circulation conditions both with CEDM cooling fans in operation Jand with CEDM cooling fans secured. The results are shown in Figure 5-6 (p. 57) and are applicable to both SONGS and Waterford since the size and configuration of the RVUH for both plant: are identical. Hence, as pre-viously stated, Waterford 3 would comply with the requirements of BTP RSB ' 5-1 in a manner identical to SONGS in that following plant cooldown system pressure would be maintained relatively high for a fifteen hour hold period until the RVUH had cooled sufficiently to depressurize the RCS ~and place the plant on shutdown cooling without forming a steam ~ bubble in the upper head. 90

=. i Section 6.2 above contains the evaluation of feedwater usage during the SONGS Natural' Circulation Test in which it was shown in Table 6-5 that a total of-246,000 gallons of water was required to meet RSB 5-1. Since l the scenario esployed to meet the branch technical position requirements is identical for both plants and since the cooldown and depressurization rates of the two plants are similar, the results of the SONGS feedwater avaluation will envelop Waterford if the water requirements at Waterford for decay heat l removal and_ sensible heat removal are equal to the require-ments at San Onofre. This is in fact the case since the total RCS metal L-mass and the total RCS volume for Waterford and SONGS are equal. In addi- , tion.since the rated thermal output of both plants is the same, the feed- ~ water required to remove decay _ heat will be identical. Therefore Water-ford 3 will also require.246,000 gallons of water to meet BTP RSB 5-1. , Note that-this volume is over 250,000 gallons less than the total combin-ed seismic Category _I water available from the condensate storage pool and the wet cooling tower. basins. . Finally with respect to nitrogen usage by the atmospheric dump valves, .'enough safety-grade nitrogen should be available to perform the entire BTP RSS 5-1 evolution.. (The Waterford design, as stated in Section 10.3.1 of Reference 15, provides enough nitrogen capacity for thirty-six - hours of valve operaton.) Note, however, that nitrogen usage is very much dependent upon operator actions to position the ADVs and therefore can not be predicted in advance. Should the nitrogen supply become depleted over a pericd of many hours, manual local control via manual ' handwheels is possible as stated in Section 5.4.3 of Reference 14. r L6.3.3

Conclusions:

Following its _ review of the shutdown cooling system and satisfactory e-- solution of all issues identified in Section 5.4.3 of Reference 14, the NRC Staff concluded that the capability.to achieve' cold shutdown for i: Waterford Unit 3 meets Brench Technical Position RS8 5-1. In demonstrat- ) ing ;1ts ability to cooldown under natural circulation conditions-and in particular its ability to properly mix boron in the RCS, Waterford 3 referenced tha natural circulation _ cooldown test which was subsequently 91

performed at San Onofre in July of 1983. Based upon an identical RCS piping configuration and an identical RVUH configuration, as stated above, the SONGS test results are directly applicable to Waterford 3. 6.4 Applicability of SONGS Test to St. Lucie Unit 2 .6.4.1 Introduction During the course of the NRC Staff's review of the St. Lucie Unit 2 shut-down cooling system, see Section 5.4.3 of Reference 13, F.lorida Power & Light was requested to demonstrate how the requirements of Branch Techni-

cal Position RS8 5-1.. " Design Requirements of the Residual Heat Removal System," have been met. Specifically, FP&L was asked to demonstrate that the plant could be brought to SCS initiation in less than 36 hours using only seismic Category I equipment assuming the most limiting single fail-ure, and with only onsite or only offsite power available. The Staff re-quested that FP&L demonstrate that the seismic Category I auxiliary feed-water system has sufficient inventory to maintain the plant at hot shut-down conditions for four hours, and then cool down to the point where core decay heat could be rejected by the shutdown cooling system.

In add-ition, supporting analysis was requested which would provide the follow-ing:-

1. Confirm that adequate mixing of borated water added prior to or during cooldown.can be achieved under natural circulation conditions. The analysis must include an estimate of the times required to achieve such mixing.

2. Confirm that the cooldown under natural circulation conditions can be achieved within the limits specified in-the smergency operating procedures.

Florida Power & Light, in its response to the above request, identified the systems which would be used to meet the Staff's requirements. The most limiting single failure was also identified, and FP&L showed in its response that the plant could be placed into cold shutdown within the 92

' required time limit following reactor trip using only safety-grade equip. . ment with a loss of offsite power and a concurrent single failure. Su bse-quently, however, the Staff required that St. Lucie Unit 2 actually demon-strata its ability to cooldown using natural circulation in the primary j system including the adequacy of boron mixing during this mode. Florida Power & Light referenced the natural circulation cooldown and boron mix-ing tests be conducted at Unit 2 of the San Onofre-Nuclear Generating ~ Station as being applicable to St. Lucie 2. The Staff found this accept-9 able'pending a favorable evaluation.of the SONGS test results and the sub-mittal of a report by FP&L documenting this applicability. As was stated in the introduction above, see Section 6.1, Combustion Engineering nuclear steam supply systems can be divided into two cate-gories with respect to the requirements of Branch Technical Position RSS 5-1. These categories are pre-System 80 and System 80 with the distinc- . tion being the relative size of the reactor vessel upper head, i.e., the

volume of the RVUH in the pre-System 80 design being approximately 1000 cubic feet with the volume of the System 80 upper head being about twice as large or 2000 cubic feet. Accordingly, the procedure that would be employed by St._ Lucie Unit 2 to meet RSB 5-1 would be identical to that 2

employed by San Onofre. In other words, following plant cooldown to SCS initiation temperature, system pressure would be maintained relatively high for a fifteen hour hold period until the RVUH had cooled sufficient- ~ ly to depressurize the RCS and place the plant on shutdown cooling with-t out forming a steam bubble in the upper head. Since the ability to cooldown the C-E NSSS under natural _ circulation con-- i ditions has been demonstrated three times on two different plants (once durin'g the SONGS Natural Circulation Test and twice at St. Lucie Unit 1) and since the scenario that is used by St. Lucie Unit 2 to meet the re-quirements of BTP RSB 5-1 is straight forward and does not involve form- .ing a steam bubble in the reactor vessel upper head, the arguments for applicability of _ the SCE test to St. Lucie will concentrate on the simi-larity in design between the two plants with respect to four specific areas. These areas are the four areas for which information was request-ed of SCE in Reference 2 and include boron mixing, reactor vessel upper head cooldown, feedwater availability, and atmospheric dump valve usage. l ~ 93-L

b l- - 6.4.2 Comparison with SONGS Test Results Section 3.0 above contains the analysis of boron mixing under natural circulation conditions including a discussion of the mechanisms for mix-ing and ac evaluation of the boron mixing test data from the SONGS Unit 2 test. Based upon a comparison of RCS piping configuration, the three mechanisms which were found to effect boron mixing in the San Onofre NSSS. as presented in Reference 2, will function in an identical manner to ensure complete mixing for St. Lucie. These mechanisms are the sub-stantial mixing of boron with fluid in the cold leg; common node inter-action in the reactor vessel inlet plenum, downcomer, and vessel bottom with fluid rotation in the downcomer; and elongation by the steam genera-tor U-tubes. The RCS piping configuration comparison showed that the hot Lleg piping diameters and the cold leg piping diameters are identical, the hot leg piping lengths and configurations and the cold leg piping lengths and configurations are identical, and the steam generator configurations are identical. In addition,.a comparison of cold leg fluid velocity, loop transit. times, and Reynolds numbers between SONGS and St. Lucie was made as shown in Table G-6 (p. 95). Based upon the piping configuration similarities and the flow calculations shown in Table 6-6, mixing of boron in an identical manner as that observed during the 50 HGS Unit 2 boron mixing test will be observed for St. Lucie Unit 2. Section 5.0 above contains the detailed analysis of RVUH cooldown under natural circulation conditions both with CEDM cooling fans in operation and with CEDM cooling fans secured. - As shown by comparing Figure 5-6 (p. 67) and Figure 5-7 (p. 68), the cooldown rate of the SONGS upper head is almost identical to the cooldown of the St. Lucie upper head. This similarity is due to the comparable size of the two reactor vessel upper heads. Hence, as previously stated, St. Lucie 2 would comply with the requirements of BTP RSB 5-1 in a manner identical to SONGS in that follow-ing plant cooldown system pressure would be maintained relatively high -for a fifteen hour hold period until the RVUH had cooled sufficiently to depressurize the RCS and place the plant on shutdown cooling without form-ing a steam bubble in the upper head. e 94

b Table 6-6 Comparison of Plant Parameters Relative to Boron Mixing iPirameter SONGS 2 St. Lucie 2 -Fluid velocity in cold leg 9 1% RTP.. 1.4 ft/sec 1.3 ft/sec ' Loop transit _ time 9 1% RTP. 5.8 min 6.3 min P Cold leg Reynolds 6 6 . Number.- > 10 > 10 ' 4 95

Section '6.2 above contains the evaluation of feeduater usage _during the ~ SONGS Natural Circulation Test in which it was shown in Table 6-5 that a total'of 246,000 gallons of water was required to meet RSB 5-1. Since the scenario employed to meet the branch technical position requirements is identical for both plants and since the cooldown and depressurization - rates oflthe two plants are similar, the results of the SONGS feedwater evaluction will envelop St. Lucie Unit 2 if the water requirements at St. Lucie for decay heat removal and sensible heat removal are less than the requirements at. San Onofre. This-is in fact the case as shown in Table 6-7 (p. 97). Since the total RCS metal mass and total RCS volume for St. Lucie are smaller than that of SONGS, less water would be requir-ed for sensible heat removal tnan indicated in Item 2 of Table 6-5. In addition, since the rated thermal output at stretch power of the St. Lucie core is approximately 20% smaller than the SONGS core, the feed-water required to remove decay heat will be correspondingly less than in-dicated in Item 3 of Table 6-5. A single calculation involving the ratio of metal masses and system volumes and the ratio of decay heat levels -indicates that the water required for St. Lucie to meet BTP RSB 5-1 is approximately 46,000 gallons less than that needed by SONGS, or a total of 200,000 gallons. Note that this volume is over 100,000 gallons less than the total condensate available as specified in the plant Technical-r ~ Specifications. Finally with respect to the atmospheric dump valves, the St. Lucie 2 design utilizes DC motor operated valves capable of being powered from - tafety-grade vital buses. Therefore, operation of tne ADVs will be avail-able at all times from the control room during a BTP RSB 5-1 evolution. . Note that in the unlikely event that operation from the control room is i' not possible. manual local control via manual handwheels is possible as stated in Section 5.4.3 of Reference 13. 6.4.3

Conclusions:

Following its review of the shutdown cooling system, as stated in Section D 5.4.3 of Reference 13, the NRC' Staff concluded that St. Lucie Unit 2 could be placed into cold shutdown within 36 hours following reactor trip 96 = 7 9 y yem ,grr-ap% w ,-4 ew awy.-c.-wg,_-,___.,a. -.gg- .--W%w,.y --mee p4-wpe--er'e--Te'sw.--mweem-+----w-rweg-D e- 'survTq m =--

• r-*,w-e,.-.g,_,y,qg,-;,

~ Table 6-7 Comparisen of Total RCS Metal Mass and Volume t l i l SONGS 2 St. Lucie 2 L Total RCS 4.9 x 106 lbm 4.4 x 10 ~ 6 lbm Metal Mass 3 3 . Total RCS-10,212 ft 9,458 ft ' Volume

• B s.

,e am-i t. t-t 97 'T v -t v+- ,-we e-- y w-g aw w,,v.,--e r e vve-, +- -e----< ---ww ---er,wvww,--e ww, ve,e--- ---*vgr e- ,www-T, we--w--v-e v =---e-

using only safety-grade equipment with a loss of offsite power and a con-current single failure. In demonstrating its ability to cool down under natural circulation conditioc'. and in particular its ability to properly mix boron in the RCS, St. Lucie referenced the natural circulation cool-down test which was subsequently performed at San Onofre in July of 1983.- Based.upon similarities in RCS piping configuration and upon similarities in RVUH configuration, the SONGS test results, as stated i [ above envelop (and are therefore applicable to) St. Lucie Unit 2. l - 6.li. Compliance by the Palo Verde Nuclear Generating Station Units 1, 2, and 3 With respect to th6 requirements of Branch Technical Position RS8 5-1, Combustion Engineering designed nuclear steam supply systems, see Section 6.1 above, can be divided into essentially two categories, pre-System 80 and System 80. The most significant distinction netween these two cate-gories is.the relative size of the reactor vessel upper head with the volume being approximately 1000 cubic feet in the pre-Systen: 80 design and approximately 2000 cubic feet in the System 80 design. For Palo Verde, then, therre are two items to be considered because of the relative-ly-large RVUH. First, APS will conduct its own natural circulation test in order to demonstrate, among other things, cooling of the upper head; and second, the basic scenario employed to meet RS8 5-1 is different for the System 80 design than for the pre-System 80 design. Current plans are for Arizona Public Service to conduct the required I tests in order to obtain data sufficient to demonstrate compliance with BTP RS8 5-1, i.e., tests will be conducted in order to demonstrate the ability to adequately mix boron and to demonsts ate the ability to perform l .a cooldown to SCS initiation temperatures and pressures under natural cir-culation conditions. - (For exact details refer to the specific test proce-dures to be supplied by APS at a later date.) The boron mixing test will be performed in a similar manner as the boron mixing test performed at. SONGS Unit 2 in July of 1983. During this test a pre-determined amount of boron will be added to the Palo Verde RCS following a reactor coolant pump trip and samples at key sample locations will be taken in order to ' shoe adequate mixing and to permit an evaluation of the mixing times. 98

The natural circulation test at Arizona will be performed in order to obtain information sufficient to demonstrate the ability to cool down the RCS including the reactor vessel upper head in the absence of forced ci rculation. In addition, data for evaluation will be obtained from the APS natural circulation test in order to address specific concerns regarding seismic Category I condensate storage capacity, atmospheric dump valve operation, and reactor vessel upper head temperatures. As was stated in Section 5.2.5 above, the basic procedure that would be j employed by Palo Verde in order to conduct a BTP RSB 5-1 evolution would I involve intentionally forming a steam bubble in the reactor vessel upper head. In this manner the upper head can be cooled sufficiently fast to allow the RCS to be depressurized and placed on shutdown cooling within the available seismic Category I condensate supply. A detailed full scope computer simulation of an RSB 5-1 evolution for System 80 is con-tained in Reference 10. In that analysis the reactor vessel gas vent system was used to preferentially direct charging fluid to the RVUH fol-lowing bubble formation in order to effect bubble collapse and thus cool the upper head. Note, as shown in Reference 10, the entire evolution took 10.5 hoars and a total of 220,000 gallons of condensate wa: required. As a final item, two important pieces of information are presented in this report which serve as pre-test predictions to the APS baron mixing and natural circulation cooldown tests. During the Arizona baron mixing test, as stated in Section 3.2 above, three mechanisms are present which will ensure complete mixing of all added boron and prevent stratifica-tion. These mechanisms ara the substantial mixing of boron with fluid in L the cold leg; conson node interaction in the reactor vessel inlet penum, downcomer, and the vessel bottom; and elongation by the steam generator U-tubes. Because of the similar reactor coolant system design,1-e., similar RCS piping sites and piping configurations, an identical mixing behavior as that demonstrated at Southern California and shown in Figure 3-12 (p. 49) should be observed at Palo Verde. In other words, complete mixing of added boron should be observed within six loop transits times following injection into the reactor coolant system. The second piece of 99 ~._

information that serves as a pre-test prediction is the RVUH cooldown - g,,-,; 4-. J. ..e curve shown in Figure 5-8 (p. 69). As can be seen, the CEDM cooling fans 4;0.f{,v]. have a large cooling effect on the upper head during natural circulation. P : .- ~.? (. t f:- Note that during the APS test, unlike the San Onofre natural circulation t; / g; ;]., .:.+- test, the K)TC reactor vessel level monitor system will be in operation '. '.:.;,q;f Q and exact upper head temperature data will therefore be available. g'3f ,' M G

\\,.y. 7

-.y:. e y. ..,.-.'I* .,'Q: y.',

1. .,,,.'4,

., g ;).:., 7. ,-[

• TW?~.

.~,s . h.. ";. y. q..,, -yg. .g 'N ., ;u.Q.c. -.

c.. c

? 8 sf. }, ?.,

m..ny

,..'.'.s ~.... p ' ;* W 4 p. 3. g._ I,p, ' h 4 +j,is

l.,.. '...

.17' ;. '4V i.',' '] f :N %: ,. '...: 7[

4.

x,.

l Y.,.'yl.,% ..y,, 'l. S

• .;,.n' p g '~;.3.j ll':

..#3.(i

j. g

?. jn,.t.,: t.$ Q. *,

t_y,

..,. > e..' > W, ~ + - ):, k V 100

""""""""""""""""158 _* * 't ; le _ _ s. Qy< h,: ,1..' a ,, ~4.fs. ' a , ',. ~.~ C e '; t ev ..7, g g e g r+-44..g

th,f[- QAW' a

.@'zt.+4.% 5.g $e 1 oaa o" pgg >c u; v.j 'lf 1,,"f Q ',f-nu O cy o e, 4. s,

1. .:2 -

fm, w t A k Ak ' f:] i~if. ': k M.-: 1.1

.+

t. ih 'y,* * - g e,- . ;wl:~.; ~._tng* m. p O, .-: c mm-, - p gg .r. *g, 1-q &g -bg-'4, i o . j' M A t-,'.. g Q ? v p .. M +*;9t. < w 1 S. -(.,. +. ' ',, b , c t.,i, se. t 7~ ie .D, . t '.,i tD % - %( 4 % O -vv g. b Q q;; ~g h{ " c, 2 + .s Y ',eW 60 4,a e M -: y. % g - j ~ *. k. e c.^, r : x,., ; m m m m m .. + - 's*',. y

. &w :?$'N-v;;

. F y 'a n rf y;*': k}g .yy G L_>. w ys y-g.... M W p.,4 .:sm Os i i g5 '- g a - -j. V m ? t. k rn 4 '%. E . :-f. = ,,.i - 1 } u, -= 1 T T,2 _ Wg*,.. - -s - p . Nt- _' $.$ b-j / /(g .o bC b od C uo N O.C g Eh e o2 kM bW g %D Em C OW b kQ C. N> 101

Kier n -Me ..}l4i '-;an; r$2

y.,..g $ p

• 9
• x y ~}.ca;,

4 *I, y,. yyq f-7'. . ~9 Figure 6-2 .f4b,lL., , e; - Feedwater Usage Based Upon CST Level '.](d/i;>J. J y....u During the SONGS Natural Circulation Test i .-'9> a -.; y _m h, 4,: ...'.- ~ ..n l M, >.M ' 'y ' !,/ t;:% s c.2/ j., - 7,,,.A 140 / /

4., n. a.

.9 +1.. :.\\ r~ ~ +, %yg g.#p,- ?gd,+;h - Q. :. ' i.F'. ; .A~3g.f x , ' 7 :.. 120 g.g ?. :'yNA: ' >.l4 ,4 ~,. p : q:. x Q k ;.. . ; u g.. s _,,...y -: y 4y y'y..y^ g., % - .g.r *> :T-100 - + s-

n.es;.s., y} ;

.3s.....y 7, ';.;.. .. ~: s -.-- O ar oo-a 80 cn 2:q a o 60 t / 40 20 5 10 15 20 TIME AFTER TRIP (HOURS) 102 1

- m, '. m.. p.y y .. "' ; 'y %, e ,,6.,mw., 4 1 p. _',',.'n,y- 'i-O ': dh3 .. 'i ^ -..~g.$,-[<.V ?l k M al 2 ' 4l ' !

u. x - c i Q

Qk -w d (b. - * '.f*.4. U >n .,i s -w e yy g g g 3.'

• ',V 4. - >

5-*Ds*.'.' %1 d Z ? 1,,.,, 'et O 9 O ~..

• +,

2 n Co M o R'", 3,-. Q) C U1 o >. aZc W O Ed n;...;d " T... ?. - o >sCO !, ? '.. y y % m,';.,.. :_- O n c

r.. dr, -

.. -)' O Q O,,, t m g g., %ea y ~ .$.. 5 : r". j +* n 4@y.,< .c ;. r, y g W M, bC M, " p.' eq k r' '$., fm. c yeu 9 .u ., ; ' 'A' d 'f Ji*,. - g a k 5e v"yb' e u a oce e M CO <tU m c .:.E e 2%,,p C ar. o C e.t - l),. E' t I s '.; J.-; g o n a { 'yf* = '. *KT. CO O m m 0e W ~ ' - g 2 Ea -. i.(." 6'y'an T gg g g U

3 w g

+ 1.c $hA.. E*

• k u

Z m a a = %.pkg*e e 4 j - 5 (,Q U y G .a 4f+..E-c4 v .E f,Q + 'i t-sen.[-Q L J y :e's.t l.' n kO o 1 0 M o d W >M o as Ea c wk k 43 C 3 0a W 's to :: W 2; ra. W C0 0C Ca3 QZ 2 CO M co E* @k@ o OA ~ O + 0M C e 2 Q o ~ - C N 1 I I O C O C N w H C (d1H JO %) Ifidifl0 7VMH3H13 HOD 103 N

t J f m i did 7.0 GENERIC OPERATOR GUIDELINES J. # Jg;({E j:f[y . 3. ; I 7.1 Introduction f" 'ipf.:L, < f .M.M i..D.Mtf<$ E The following generic operator guidelines were generated based upon the A '.%. c % ' h SONGS Unit 2 natural circulation cooldown test and the subsequent post-1 test analysis contained in this report. They are provided to supplement ';. g )._ F the guidelines contained in Reference 11, Combustion Engineering Emer-a..g;y *.:g [. gency Procedure Guidelines, in order to improve operator response to

• v.n s:

.s s boron mixing and reactor vessel upper head cooldown in the absence of

• v.JU N a

s y' N"%s... h; forced circulation. (Note that bracketed items denote niaces where plant ' Q: A 9 + a%g E-specific information should be supplied.) s3 y .rs S k. g ' ,h f 7.2 Boron Mixing During Natural Circulation Mc, 9g . pa. m Thc*:Of.?g s n, v.... i 7.2.1 Operator Actions Qr. % Qe .;~~ k,jr, The recovery actions described in Reference 11 for a loss of forced -- U .. i.k?W n 9 C circulation have been initiated. A natural circulation cooldown followed M. 3W. w [ by a plant depressurization is to be performed in order to establish f conditions necessary to initiate shutdown cooling. In addition to the E actions specified in Section 9.0 of Reference 11, the following B guidelines are provided concerning boration in the absence of forced circulation: (Refer to Figure 7-1 (p.119) for a flow chart of the boron Fl -M h mi xing gui delines. ) mN.g* X 7 9 a L 1. Borate the plant in accordance with Technical Specffication p.y. y t ;, Q - .... g limitations. t-- [ a. If letdown is not available borate concurrent with cooldown. In b. If letdown is available borate prior to cooldown. Ensure mi xing k-of boron added prior to cooldown by performing one of the following (listed in order of preference): E i b 1. Representative manual loop sanple readings within -~ E [10] ppm of target value. [ E 104 E

.*1 d .E. g iv j'.?.* ,.[t'.-* M* . ',Q:;.,.:: ?.g ii. Borenometer readings within [10] ppm of target ,,.1. ii :.5:. ;94 L j[. 7 y ;..p... .n ya1ue. g.w y.g:j .f.:$3;,._J y iii. If bc onometer readings and manual samples are not available, wait [60] minutes following addition [p_.. ;x3 sj. : %[. {.; Q prior to consnencing cooldown. ..%.f';1 2. To avoid RCS boron dilution and loss of shutdown margin by I-[.h..].d]-. + .,., fc. .w. pressurizer outsu(ge during the cooldown, perf orm one of the 3 p. g,7. 7; following steps ('.isted in order of preference): .)kh. :, -(. ( , p.. w q:. a

ag 5, a.

Calculate and add sufficient boron to the RCS to raise the entire N 'cM '". '. 3.7 .?. . s. g RCS (including the water mass in the pressurizer) to cold D,,2.,[.t %... shutdown conditions. [] ',f.g} - g.. J.4.} :.'. b. If letdown is available, use auxiliary spray to increase and N (, maintain pressurizer boron concentration within [50] ppm of RCS b:p..%. -,.. :, q . q..... concentration using heaters to control pressurizer pressure. p.' A...::r.- ~. x..- s:;.p .; ; y;

1. p-lf _V..:.:

%Ss. .;* 4 ; '*C - i. c. if letdown is not available, use auxiliary spray and pressurizer

y, s heaters to control pressurizer pressure and increase RCS boron

f. y qr; C:

concentration to [50] ppm greater than that required for minimum ,f;h ) *.) shutdown margi n. S../d:#.S?"

, : y gg%c.u;-

.. M. n L f.WZs.. i'b 3. Perform RCS cooldown at maximum rate possible within plant Technical 3 Speci fication limitatians. memwasom.- M le...i.+ %g.,,:fC.I ,igi. JJ: 4. Verify mixing of boron added during cooldown via 1 cop samples or Qisk. :(.!' boronometer readings no more than [10] ppm Ins than target value by Wy;ff g:$"y Mr, wv. w. one hour fol1owing addition. ...R.:g. l ; g - f.a J t, ,.d.2$$, Q;.1. e.. 7.4 a. If added boron is properly mixed, continue cooldown. Y 'p ji: 5-?:s. fl. 5, b. If added boron is not mixed, stop cooldown and stabilize plant

• s, r.

, : s..,. until added boron has properly mixed, i.e., loop samples or

3g :?5 *4F baronometer readings no more than [10] ppm less than target

- M W $.!t a

_. p \\g !lg;.f i value. . big y.l. .;- r. ~. 4 ..:.g$'..,?_. .l 105 .y. z > kM. '> _C(f.

[ 7.2.2 Bases for Operator Actions The following baces are intended to provide the operatar with the 'I information necessary to be able to understand the reasons fer, and the consequences of, the action taken to borate in the absence of forced v;gl ci rculation : ,. ~. f; 1. Borate plant in order to maintain shutdown margin specified in plant j Technical Specification. W- = r ; - a. If letdown is unavailable, boration should be performed concurrent with cooldown as part of normal inventory makeup due to contraction. Pressurizer level will be maintained via charging in order to enhance pressure control by minimizing level transients. b. If letdown is available it is preferable to borate the plant li prior to commencing cooldown. Pressurizer level will be ^ ' 1 i maintained via either manual or automatic control of the PLCS in F order to enhance pressure control by minimizing level {$f.] $', transients. Verify mixing of added boron via loop samples or (( [ boronometer readings, if possible, prior to cooldown. If sample yg readings or baronometer are not available, the [60] minute hold +f,.i)g^i; v. p[? q e+9 V period (which is based upon empirical boron mixing data) is Vg ;,g-. designed to ensure complete niixing of added boron in all [,, ' .'.$m.ty s a - g 'J- / circulating portions of the RCS. f &i.,L$.l 2. During the cooldown, shrinkage of RCS inventory due to cooling may 4., cause outsurge of pressurizer fluid. Since this fluid is not

.p.

.M ;. y directly borated by charging flow, it may be at a lower boron y) concentration than the RCS loops and therefore may dilute the loops ., j].f.'~.y ' (h .[p.; and the vessel somewhat. In order to avoid this possible loss of ...,.%. v.f-ll shutdown margin, perform the following actions (listed in order of preference): Kyg j.7- % @ h..\\ $..'.? ;.O : ::. O n.. (,Y,[,$' 7 %lo.%y 106 g g ; s. ??.$;f..e, .nx? '

e g

• ;*.N

' h '% . - E:

p. 5 -

,.f ' $ r jd. 'A ' ~ ' .'N*^ W*'+' ' WY~ 'Y ' * ^' . 5. ~ , ?. h.,l-: '&.. s J. ...'.g. i ..4.3 '. M, '$4f* Y 4, 4 ' -(*., 5 e. 4 ~.. g-

l A.

"%.f, % v

.k '. *

's.* .,- w,,...e t 4. _ Wa. x

x.... a ',

t .-w .. yy. k =. p.

g.'p.9

-s ..A. k.,.k.

4. - -,'

?. ;;,*.?ge 3 'l~s. ~ " 'g,., + (.g. -. 4.. .g p-T... .,, ' f p.- t(.. g, _..,4 y. .t-p. I. A 0., 3..... e. ,r q *<. ~y '4.. 1 s: -.. ...s,5 /.. -..$ ".e. ,. x..r. y p; :. 4 y

y '...-

-

is.. % .y T',', p:;.... v.

r

[.

> :s - +* *, ',i 9.. u

.;,?

.r,. 4,, ap ,:._..f--. u n.

'..wo. 35g

.~

e ~

f.. W.. :

t - so,:,,;p...k , +- .[ ..m - 4.,. N*.

d. ' '.., ' 7 )'. d...,

.5 1. ,..:~ -..?- .r

• ..=,. *s

.s i.. - g '.?*....:,.- s' $f ' . ?j.Y,.s.

{

a. Q,, l s g. ' ;. 'f., ' ;;.;' e ;: 'd {... -.. ?g, s,'" e %.w, s. ' - <d e .t.- ,. -^s4'.*~4',,- ,g. < +.,. -. ",. s'.. '. ;/, ' ',., ' l:.'. .h g <* '.gV'O = 3: e y - [> es-v' ' n,c -]._.*]- %.y,f.'*'.N... .+n s. ,3,

• e

<-, P f' .'I ..Q

• l *. ~

y. a... ;. ..f* i. Q' g*.a . ys

c. m ;

n; ,e. h ,}.' Q.' '** )'. h NksY,h. y iT - p + e.- I<*,* i,4 \\;. ,+ W % *'T .6 w 9 ..4< a g.... g M 4,s *. .4

• • ,.A'

':,r

f' ---'.'.'.

.

a s', ~.s;' ,f .:5 ..n' .. s. s v J + ,. a.?.. - /3 se,;..,-... ;

.+*.4.

Q. *-.. t, '.: .,.e

<;, A - -

s,. .'.,*.'....'W.O.,i 7. T'.A. a.' .h ..,9 g<,. '.. ' hs' .t .Y.? ' ' &..*,~l ?- ' c ..t

f. ~.,..r-..

~.g. ~- c y, ..*s. th-....c - n,., ".. '

• 1.

b N +.hi h 'h .' '.x - ,i v.r, -- k [ W. ..M 2, ,e,',r.... .e.4 't'.., =.,'# g+

n..

a I..

%..h. .h* [ $ A '-9 , h./.i. ', g', - 'o i- ',9 i s., i. .'h., @ % -- .w ... Ws .h.. ' a-. A.-.z. l-3,. -h..'. % ~.';..-.. * ,- Q -' . 1 th.. ,1 c. '.l. (- e

L 7.3 Reactor Vessel Upper Head Cooldown During Natural Circulation 7.3.1 Operator Actions The recovery actions described in Reference 11 for a loss of forced circuiation have been initiated. A natural circulation cooldown followed by : plant depressurization is to be performed in order to establish conditions necessary to initiate shutdown cooling. In addition to the actions specified in Section 9.0 of Rcference 11, the following guidelines are provided concerning RVUH cooldown in the absence of forced circulation. (Refer to Figure 7-2 (p.122) for a flow chart of the RVUH cooldown guidelines.) i I-1. Comence an RCS cooldown in accordance with Technical Specification h Limitations. Cool down at maximum rate possible. Cool down by h performing one of the following (listed in order to preference): i a. If the condenser and turbine bypass system are available, o commence the cooldown using the turbine bypass system and [ main k or auxiliary] feedwater. t b. If the condenser or the turbine bypass system are not available, commence the cooldown using the atmospheric dump valves and [ main or auxiliary] feedwater. ...x .M 2. During the cooldown, monitor the available condensate inventory and ..).f.- w[.M:; h replenish from alternate sources as required. a. . pyyy $g.((h 3. If a steam generator was isolated, cool the isolated steam generator jk as necessary to prevent isolated loop void formation by (listed in orderofpreference): Oh,5 E. E Dy i Wifs a. Starting at least one OCP to force cool the SG. b. Feed and bleed the isolated steam generator with normal feedwater .M

it.4j

. if supply and steam generator blowdown. my b 108 l-

[95N f[;d M.I[ c I ?: [N' D: D D.'M E '9 *[E -9 "# [ '#?'* f(.pfhj C'N - _ y y,:: t ; s _w ~m

a,.,. : ::,.

hkA

.. ; a. y. y* g r h,s;;s, y. 11, g t.% ' y..: yQ. p ), i... 3 ..,e. ? c. Feed and steam the isolated SG to the condenser or to atmospnere. c,,7 -y...i f '. ?. 4h < F.%.c M f vg g ^ p

• g

,j 4. During the cooldown if letdown is not availat,le, secure the Tm.II ib d...,;J a J.7.:,. J.. pressurizer heaters and allow pressure to decrease via ambient losses .-.hg,'.f%. 7 i-,. } while maintaining pressure within the acceptable post accident 'rg &; ? -p c.. pressure / temperature limits. If letdown is available, pressurizer u?.J ?. 1 ? D,? heaters need not be secured, however. post accident %, g~v. g t i..,

f..p..y -

c g . V, pressure / temperature limits are to be maintained. Maintain 20 F of I e- ,e %L subcooling in the RVUH. Concrol RCS pressure by A tG 7:.,,3i :-f ?,' .S.p. sfv:Q 'n. ;' h .V. +.. ..h y. .W Z X44 + a. Controlling RCS heat removal via the steam generators. .g* }{c? C q.N 4l;....,..< t and p., h. ffej.l.I: % 5.i D? 7 b. Controlling RCS pressure using (listed in order of preference), [f14Qfp %7i-4.,y q n, n.sr. .n 4.y $ W:O 2,. M d.w.,.. T T- ( i. Pressurizer heaters and auxiliary spray. h r. [,. y y My. ',$;. Q. g e.. r 1 -..,,.m. h ) ii. Charging and letdown. y& c%p%y.?<4g:

g. 7

z. _ I'.

p' .A t f.( T e iii. HPSI pumps. A:..h. y.a : - n. 3. ,}s.m :. m:. '.p? % W, py., -s :.

x,--

> p..SM.. j'g 4

iv. [ Pressurizer fill 'and drain method.] C-8 o..y q y -a. yp ..,, i;* g ', ,.s..v : K; M. y.%,',:.<: T pg' 5. Secure cooldown and stabilize plant at a hot leg temperature of f J:. ;g;i[ N' D(& $ b D [350 F]. + r.f i.h . pg.%. g': o -.. - 6. Based upon plant status and available condensate inventories, if a -gg.g,. J..'7.;

• ?.

[15] hour hold period can be sustained, perform Step 7 through Step 7DM:C _%. r,1 1 : 1 . CIS. + -..M 11. If a [15] hour hold period cannot be sustained, perform Step 12 .; - n.s. d through Step 16.

h. - t & *: 4..

. M,1.-.

A, n.y..

c

. c.- ip;4. 4 v.,- .p g;J'~ . [s +:n. '.'i n 7. Maintain plant conditions for [15] hours or until RVUH temperature is 9 )jNE. ; -0 3. less than the saturation temperature corresponding to SCS entry ..4,. 'y(: Q,.L i.Qu.L. pressure. .% :f x %;;.. ..e 3 .j;:Q( 4 ( <.,,. - ..,.i l,f'}(f, 4.* - 7.j ] ' l ?.,' '.+ 1-l p . 3;g % ? y;7, 109 g ::, gr3..l,." ;> R.fL % n.3;amn mwxmmn w u m m n nE m:

• i y

8. If auxiliary spray is available, depressucize plant to SCS entry 1., 5.Y m % p( ^ pressure using auxiliary spray. If auxiliary spray is not available. {, depressurize plant to SCS entry pressure as follows (listed in order gfty,.;ye,, of preference): 'j:.$ {5 / (( , Q :.;.9.- M...,i &M M ? # ' @ e. a. Perform a pressurizer fill and drain if letdown is available. g.. ';;4 ~. *, y. m. . / b. Operate the pressurizer vent system, i-M - N.g, ./'. y,.;. . q,;.g 7 s. 9. Maintain 20 F of subcooling in the RCS loops. Monitor for RCS - m. 7 0. N,. 3 voi di ng. Indications of voiding are any of the fo' lowing parameter Q{'[;;.:./ g n changes or trends:

d. J. A f. '

f. {A,$'-J;,N..> .t d,- f :.e a. Letdown flow greater than charging flow. . R : ', p* J ..f.h.+. ' ~ ' p,f \\ s,.. b. Pressurize % vel increasing signif.cantly more than expected while operating auxiliary spray. n... ,h mmm c. RVLMS indicates void formation. 4 ' A, t. ',.-W t.4 .., o .i gig. r[

10. If void formation occurs, eliminate the void as follows:

4.;.~y ;. y ^%p,.t .g' f';r; a. Isolate letdown. 4.i{4 - 1 tu s' v;f. '.," b{.,q:f z b. Stop the depressurization, jJ iM b?,i', .,..m.- T Q M 5. c. Increase RCS pressure by approximately 100 psi within the limits .{ r-k' - Qyp ',,4 of the post-accident pressure / temperature limit curves by ..jj operating pressurizer heaters (preferred method) or HPSI and w " f '2^. / r. chargiag pumps (alternati te method). J + v,e m W.... r. . 4., > d. Operate the reactor vessel head vent if necessary to help clear 9<.f.),47; w%a.s s > gases from the RVUH and help collapse upper head voids. gc

a. ' 'S

M.}(.[,aq 4/{ bM. 11 If system pressure is < [400] psia, initiate shutdown cooling. If system pressure is above SCS entry pressure, repeat Step 8 through Step 10 as necessary to reduce pressure. 110

-_.__.-w__...__u_.u_.w._____.____._______m_;____ 2 pg. y.3 Ay;q;...x w 4 i.; > % ;.y.

12. Perform a rapid cooldown of the RVUH per Step 13 through Step 16.

'.<y .y c ...m g g4".r. 4'g..

13. If auxiliary spray 1,s_ available, depressurize plant to intentionally a

k. a su. : 4 m

form a steam bubble in the RVUH using the auxiliary spray system. If J auxiliary spray is not available, depressurize the plant to .':$,.;gdfWh,} i intentionally form a steam bubble in the RVUH using the pressurizer 0~,( #j,.).) ff... y v. vent system or performing a pressurizer fill and drain if letdown is c,..,7 y, h.h r 4-s = available. F. d.fi Q /.'.

. a...

16.1:' d: [ _s

14. Monitor for RVUH voiding. During void formation, allow actual s[ W @ J -

s t., n*. h M s, m E; pressurizer level to increase to approxmately 807.. Indi cati ons 1p fA

. a

.. f.,.' of voiding are any of the following parameter changes or trends: f ghjs ec.- a.. q ;~. in x .,,e .n. > m y a. Letdown flow greater than charging flow. lY,$gS,.& 4.. l n..,. b. Pressurizer level increasing significantly more than expecteJ 'k -[ e while operating auxiliary spray. [ ki-h,.:Q,,f,, gy.~,., . $Q,- c. RVLMS indicates void formation. N.[ h ; ,. a.dfp..$& t-l

15. When RVUH voiding occurs, eliminate the void as follows:

.YNt a bQ;%g -c M-p Ati m. 6A.. w.- y,,... v " d a. Isolate letdown.

.N y

-) 4:j,q ::*L..l b. Stop the depressurization, i:.3c'. v.M.f Q (i h.

9 ;.,.,

:

.9..

w. E -.* >.h!' c. Increase RCS pressure by approximately 100 psi within the limits .. = s s P of the post-accident pressure / temperature limit curves by h,. ?/_. ks,, .s E operating pressurizer haaters (preferred method) or HPSI and Q.c7.<fy .a:

• aE; charging pumps (alternative method).

c.j,rM'q%,.;T 7-q ~,_; : aq 5. @j$.![Q$ r. d. Operate the reactor vessel head vent if necessary to help clear 7l %cjt c.., E-gases f rom the RVUH and help collapse upper head voids. A.,..:. d > ~S

3..:.t.y

},l~.Q ~ a

f. 6 '~ &o.

16. If system pressure is f [400] psia, initiate shutdown cooling.

If J :.? 3.. %. = system pressure is above SCS entry pressure repeat Step 12 through % 'j g g.g n., 4 Step 15 as necessary to reduce pressure. M..$4.-]f i 111 = L

e-:Q 7.3.2 Bases for Operator Actions The following bases are intended to provide the operator with the information necessary to be able to understand the reasons for, and the consequences of, the actions taken to enhance RVUH cooldown-1. The RCS cooldown should be commenced at the maximum rate possible in accordance with Technical Specification limits by performing Step a or Step b below (listed in order of preference): a. The RCS is cooled down by feeding the steam generators with [ main I or auxiliary] feedwater and discharging steam using the turbine bypass system. This method can on!y be implerented if the condenser is available. b. If the condenser is not available, an RCS cooldown should be commenced using [ main or auxiliary] feedwater and dumping steam using the atmospheric steam dumo valves. Using atmospheric dump l valves to cool down a steam generator causes a depletion of condensate and, therefore, it could be more limiting than using y the turbine bypass system. i Cooling down at the f astest rate possible will maximize natural 7 = circulation flow which will enhance boron mixing. In addition, the differential temperature between the RVUH and the RCS loops wi'l be maximized which will enhance upper head cooldown. m 2. Throughout the cooldown, the available condensate inventory should be monitored and replenished from available sources to provide a source for a secondary heat sink. Condensate inventory requirements should = be determined according to the condensate inventory curves. E xample of alternate sources of condensate are ron-seismic tanks, fire mains, i lake water supplies, portable tanks, etc. Plant ipecific alternate sources of feedwater should be identified and cited in the procedure. ? 112 -I

-~ W. o T g 3. Equipment malfunctions may require that one steam gerierator be i ..p {g isolated continuously from the RCS, as a heat sink (i.e.,all l I l feedwater and steam flow in and out of that steam generator p s. stopped). During forced flow conditions when one steam generator j;gi must be isolated as a heat sink, sufficient heat transfer occurs to M.k maintain the isolated steam generator at the same relative -- T

h.,

4,.b tem erature as the operating RCS loop. However, with no RCPs operating, conditions can be generated which will stop natural -; f fs { p( $v yV L. ..,.. &. y circulation flow through the isolated steam generator and RCS loop, .. g&[ @ *,g#N M leaving those components in a hot stagnant condition. This condition ' d.', + m by itself will not necessarily affect core cooling via natural g circulation in the unisolated steam generator. As long as reactivity 43 T2P control, RCS pressure control, RCS inventory control, and RCS heat ,b -,...y-removal are properly maintaine ' in the operating loop, sufficient 'Q..W natural circulation flow will be maintained through the core and h N[ J@(3.p. .; v.. y h,; ;w{' y operating loop. However, a hot isolated steam generator presents a problem when .].- trying to depressurize the RCS, e.g., to initiate shutdown cooling. y,~ 7. n,5

• 7 Depressurization of the RCS below the isolated steam c arerator's

'. c2 p' s. saturation pressure could quickly void large portions of the isolated RCS loop which could cause the isolated steam generator to act like a pressurizer and prevent further depressurization to the shutdown cooling initiation pressure. Thus, an isolated steam generator should usually be cooled dov.n along with the RCS. The preferred method of cooling an isolated steam generator is to start any RCP, if one is available. Forced reactor coolant circulation through an isolated steam generator will provide adequate heat transfer to maintain the isolated steam generator's temperature approximately the same as the operating steam generator's temperature. RCP restart criteria must be met to restart an RCP. Another method, if possible, is to feed and bleed the isolated steam generator using normal feed supply and steam generator blowdown. This method permits cooldown by regulation of the feed and drain 113

r i i h b E' rates. Draining and then refilling is not preferred since the transient from this process is difficult to ccatrol. Plant specific i.l-r procedures should be developed for this process. An isolated steam ? S';f -Q,ra:.g - y _ i {. p @ :yT% [ generatt. mighc. also be steamed to the condenser or to atmosphere in .!9.c ... > c 7*6 order to cool it down. F

p r g.~a, 4 ;y- -;: -

4 If letdown is not available, the potential exists for filling the g.7 pressurizer solid during system depressurization through inventory ?. W. ?,7 l'. CJC = E tch ' ;* ; addition via auxiliary spray and system pressure expansion. f; [- Therefore, securing heaters and allowing pressure to slowly decrease g f{ ]. via pressurizer ambient heat losses during the cooldown will minimize a b kii h the amount of auxiliary spray required to reduce system pressure to .M M[% ib-P SCS initiation conditiors. 3 - wc;3f g Subcooling should be maintained in RVUH in order to prevent forming a

• ~O C 4

{ steam bubble in that region prior to completing cooldown. E The bases for maintaining the RCS pressure and temperature within the acceptable range of the P/T limit curve during the cooldown is that { it will allow the operator to ensure adequate core cooling by k maintaining minimum subcooling and will also require operator actions y (such as termination of HPSI or charging flow) which prevent excessive repressurization of the RCS. Excessive repressurization I (i.e., > [200 F] subcooling) may, esult in reactor vessels stresses in the range of concern for pressurized thermal shock. Maintaining proper subcooling will minimize the chances of void formation in areas of low flow. The operator ha-two basic methods K { to maintain RCS pressure and temperature within the acceptable range of the P/T curve. These methods are:

1) Control RCS heat removal, i.e., cooldown rate; and 2) Control RCS pressure using pressurizer E

heaters and spray, charging and letdown, and [ pressurizer fill and Ei drain]. The operator will choose which method or combination of methods to be b used based on existing plant conditions, as no two events are likely u4 E

% mew-. :-w-amws - -.n @ A / ;, f ^.,L ' ' :..g, 5 " Q Z V]c. to follow the same scenario. For example, if the main condenser was . +' .% y; f g [. %..- a s.. not available and the only method for PCS heat removal is the atinospheric dump valves, than the choice would be to remove RCS neat .J., v f at the rate consistent within technical specifications and the ,p, }fhk I atmospheric dump valve capacity. Pressurizer pressure would be . v.:. - e controlled by use of pressurizer heaters and auxiliary spray Tjiy/ J. W7*te v, 'g (preferred) or by maintaining the required HPSI or charging pump flow rate to maintain the RCS pressure within acceptable P/T curve limits. On the other hand if the main condenser is available the preferred method would be to control the RCS heat removal at a rate allowed by technical specification limits using the turbine bypass valve. At i the same time RCS pressure would be controlled by using auxiliary spray (preferred) and pressurizer heaters or a combination of HPSI i and charging pumps to obtain a RCS pressure within acceptable P/T j j curve limi ts. ~ i b As many variables will exist, the operator must use judgewnt based P on the existing plant conditions as to the best method to maintain the RCS within the desired P/T curve limits to minimize PTS concerns and provide for adequate core cooling. 5. Cool plant to shutdown cooling system entry temperatures. 6. Based upon plant status and available condensate inventories, the plant will be stabilized at SCS entry te@erature and held there until the RVUH has cooled, or a rapid cooldown of the RVUH will be $., llg.:Py performed. T'[ T K . 't. A v. et,, 7. The hold period is designed to allow the RVUH to cool so that a l system depressurization to SCS entry pressure can be performed % f /.. " without reaching saturation conditions in the upper head. [Fi f teen] kk. E.k., hours is the approximate time necessary for upper head cooling to y '} take place in the absence of CEDM cooling f ans. Note that if CEDM cooling f ans are operating than the hold period will be approximately [6] hours, }- ?lN.; . T l . $ / e,,4 (. l.

• U 'Y Iy((.

8. Depressurize to shutdown cooling system entry pressures. L 9. ihe information provided in this step and the next deals with voids. [ Operational in*ormation is provided for use in detectirig and eliminating voids. There are certain RCS conditions for which the k presence of voids is acceptable. In other words, it is not always imperative that the operator take measures to remove voids, if they are detected. Voids may be allowed to remain as long as the core and y RCS heat removal and the RCS inventory safety functions are being E satisfied. The continual assessment of these safety functions via f the safety function status check of Reference 11 will verify their fulfillment. Anytime it is found that either the heat removal or inventory control safety functions are not being satisfied, or l voiding is causing the RCS to remain pressurized above the SCS entry [ f pressure when SCS operations is desired, an attempt at elimination of a g the voiding must be made. E f The operator should monitor for the prasence of voids. Voiding in E the RCS may be indicated by the following parameter changes or trends: Y l [ a. Letdown flow greater than charging flow. b. Pressurizer level increasing significantly greater than expected ~ while operating auxiliary spray. I c. The following test can be performed by the operator to determine I the presence of voiding within the reactor coolant system. r F F i. Stabilize pressurizer level and pressure (i.e., place PLCS f and PPCS in manual). ii. Start an additional charging pump ta demonstrate that h press:;cizer level responds as expected: !r L [ increase of [2] inches / min, per charging pump h (approximately). E r [ 116 m E

iii. Activate pressurizer heaters and demonstrate that the pressurizer pressure instrumentation responds as expected: ...y.3j j- .q. y.qy1y. increase of [15] psi / min (approximately). s c N A p_ p pp.E'q Q&y$%. lY, iv. Activate pressurizer auxiliary spray and demonstrate that . 4% ?mW.;:.%:. the pressurizer instrumentation responds as expected.

p%.. Md. ' g,'-

k.3 ?.. (h tv.,y $ e.::. decrease of [26] psi / min (approximately). s ,k) 47.-.;ihD , '2 i If pressurizer parameters respond within the above criteria a.nd subcooling is within the limits of the post-accident pressure / temperature limit curves, then significant voiding does not exist. El,RMilem If pressurizer parameters do not respond within the above criteria, j .w then voiding is indicoced. Jy 7 (f,

10. Void eliminatinn proceeds as follows:

.g. n... :... . ?,)s <W. # $:q-a. Letdown is isolated (or verified to be isolated) to minimize k'h.,jy$ ..s.9:. 7, further inventory loss. A . '? Nd:L5 9.s g b. The dences urization is stopped to prevent further growth of the ,gg y v,ai d. J Kd Eaum Mll c. Repressurizing the RCS by approximately 100 psi within the limits l .gl.,_ g of the post-accident pressure / temperature curves will collapse %, b.. -p a .};,[ the vold. Repressurizing has the ef fect of filling the voided >E'I I.N.;3-[ p k portion of the RCS with cooler loop fluid which will remove heat r i .A- # 'NA f rom this regi on. Limiting repressurization to approximately 100 m $h l psi will facilitate subsequent depressurizations. This pressurizing may be accomplished using the pressurizer heaters J O, y-Q '.L f r y, (preferred method) or the HPSI system, starting and stopping .( W r; charging pumps or throttling HPSI pumps (alternative method). { ~}p s, 4 .s .no g.. a s.' S c 1, t ".7. '..;.' ' '& l j: . i4;, :-{.4% ' [. 11I .M./M I. [ !' =

f.y .;.p;L s..%3 7. G. yiq f-:,.. fc._ ~. :. y.y l A :y;, y v., 9 _. ., ;._t.., m.:*-; &g.n. pm'.g.a.. q.: ~.z : ;hk. c. v at ., 4 f.. , '.g',l e.t. = q.; wp 2 i {. / .. pt'f i, ...1<?, ' 5;;,, W.s v : e44. -. f. a... * > o j hlU

g. s _* g& u 4,!,..

4.y 5 _.. %% 9 i... -

1. pi%. - -

1 ~ NM [; y. ..2 d. The reactor vessel head vent system may be operated to help clear . A.

f. J.

s< +s s 8 5as been placed on shutdown cooling to control system inventory and gWcn.,7 L .q: wy. s, a y eliminate the void. . ; g.l-l 3:.e.; u;.,. su:v. v J....;~; s, + 6h .k 2. tM k*

J

/ 1 ga93 .* n - y .f[q, A. ' 'f '.+.; ' 1 [ e y. .$f 11g y$% P -{, ?p.+. ; Q:. g (. 7 x, q.. 9a .p.; .g ~ a d, ' .. ' W..h 'f %i j.," i , ' ( ( ' ,1,,,.L., 5., y - ]f g.",, t.. '. ;

• e,'

s j. s. w. .1,'.. 47^' ' ' 7.. '.' " - 81 gi f. ,. ' '(t, y

p i M -l Figure 7-1 bq r_ l [ Flow Chart for Boron Mixing Guidelines k u ~ k parw in NO 'fE 5 + ,g aratzlized Car'idrfien. 4_ L r i. 4 U 15 MC) terdeu>o NO audslak:k.* ^ 4 1r W best. per L w.<4: L Spr.cA cAfico brwitatteris eB m m 1r hg ir MsCal f SpeciWcarieo M '"'

• T L

via N ual !804 #N # i.3.m c.aondwo 4 g eo m .a = macin3s em n== P than De] ppm le w tarep p v alve. o r' n-m w p, =- g w _:.uv cet ava.m, .N' B e t Cwo] h: E T4Me.achen m,nv=s ww bees:$dg men i occ'kstge g-1f Tc N S C h'C O T-beed diluha: duc.% PLR. Outsertge h-g' l% Q ~ l (b Q) (b Q) I + 119 A96 ):

f )-[~ di E

-Qg 8.. ' ,4 ?

I Figure 7-1 (cont) Y65 m N* xd'w: WOL. W $I Ii.e more b. i Wgl"a s-peor to or donm ccenu sd Oc. meved o+hio O nt. D bikto( Gddre 120 1

---

inimm

e...< b.:.

4. u,,, 9. +

,A ~ Figure 7-1 (cont) !..{ f,,t f: + 3. _ ( },, k,'i -)' '.*,. e" ; g k , y,. ,' e, 4 /.' t u . d Q.;, a y' ~,. Ny >4., I'.y

s. ;<; n.

.s % ;. t - es k 5 l @W

,ys. s,'. t. -

e Jb j . + ) .l .f s. - f n %. : h e ....k..-%. e m b.." '.g: r " SPQOliK. NES pterw urmi..E.-, MO w - -. -- ',v; ,Th.. boreri Nh1rWte. Ned ? w _ ;,.', s...;. + ;..,, a.. 54* ? d-k%.T.. * ' y. .. - I CentTove -P',.4. f* Pear 4rugg, 'r g. ' .h

• n.,*.

? e f, *,.. -..,_ e ... m, ~;' ' 4,'.". * ' [$Y. ~ ~; '1 -: 1: -% ld /. x 24fr-tv4 ~?-l%'.~ t 4,* +($,- t ', %t. '., f,'#' ' 4, .I ,-v,. ' ? _....s,l%. > l.p %. * *.&. , -W /4 v '6 ,.. ?.11~.

y.,

py..., ,e ,.. $. ~'. Y ' .- t mmsg . - lf? q.? A 4. d 6* '? t 4,h [ '.I f)w y,_. A.~'. e'. A h< ,, y _ ' k ?, '. ,.: te ;;': x ..w. y,.3. ...l.'f. -e .G g

h V ',.; '.

.g. % am ..m k u%.- e .-. c, q. : s, t +m. m 121

'".=ir ,_r Figure 7-2 Flow Chart for RVUH Cooldown Guidelines y he r*M p in assereer==. In,3s9M TerhrtsCel ' ~ Spes.s6c&tten Lime

• amens.

( neJnta. c at manewm rtif., pm bn. n Monetcreud,lable. c.ondesuast. irwetWor and ^ es r* rr* ArY i.s N If a 54 toes 2 is>leted, c. col downi3> W 56. ' = 15 h/o letdem>d YM wiave.) Mfk bTdO wew prescre., wrthen m ~ alkre prv.sSET-t+T GCff#E to ort _re.ase-ve post amoent e/r u n,.a ao e ,, n.w ~ ~ of Wi% &O* F SW'T in +ne. Rc.5 in tne. Avt:H? \\ cops P V3 I RVOH. g 122 am a

Figure 7-2 (cont) RP 'Y drid StabeleZC., piant er a c+ L [350*Fj No Ne.s = %ntain ptsnt ecedit, ens fbr-MXI%Uk h:vs er ei d RVuH y erut, cco crf ttv %VV H. tesyven2rure. CCrnCS@n3% sc_s e-w pre ssure.. 1P 1r NES NO NF.S O qv sy 123

Figure 7-2 (cont) - ~ L2 W dh 4) W 3 o o Depm.sson'xe. D4mcarize. n usorg me. q

Wxe, M*

S 3N#"

• 3F #

t o (trm a sw irntrmerww per4semerg b.**st in tt* h r m a. a m+ii Deptssunsc. Rvutt us.n3+h Sum mc. g gn u.s g 4 Prt e.spne'1 in twe. auxiliaq gf C'~ M '* EM N'T te a Fzx Siu e w m =1% draio rf 50fty %m is SWtm available.. 4 v v Hanntain 2DeF Maintasn,1caF ~ cf 9 W iirg of %hmkg Y in tts%5 leefL in the.Hc.s Han1cr 4br-leeps. Moniten-RVuH ved & AVOH ibrmanen. veed 4t:rmation. _x 1 aa,se.m w waxe to 80%. Sep NG rarico tedKated, 3 rep cmiapse. tm deptsunratien um d. Collage,1tt., void. g la c A 9 g l l ir 1 Ente.y-Swn ccoling. d 124 I --i iieium

8,0 ~ REFERENCES 1. " Final Safety Analysis Report for San Onofre Nuclear Generating Station, Units 2 & 3," Docket No. 50-361/362, Amendment, 30 July 1982. -2. " Natural Circulation Test Program, SONGS Unit 2, Natural Circulation Cooldown," CEN-201(S) Supplement No.1 January,1983. 3. " Technical Specifications, San Onofre Nuclearf Generating Station, Unit No. 2," Docket No. 50-361, NUREG-0741, Appendix A to License Number NPF-10, February,1982. 4. " Guidance for Residual Heat Removal," Regulatory Guide 1.139, U.S. ~ Nuclear Regulatory Commission, May,1978. 5. " Verification of C-E Mixing Model for Pressurized Thermal Shock Events with CREARE-1/5 Scale Data," CENPSD-204, September, 1982. 6. " Analysis and Evaluation of St. Lucie Unit 1 Natural Circulation Cooldown," NSfC-16/INP0-21, december,1980. 7. " Natural Circulation Test Program, San Onofre Nuclear Generating Station Unit 2, Safety Evaluation," CEN-201(S), April,1982.

8.. " Natural Circulation Cooldown, Task 430 Final Report," CE-NPSD-154,,

October, 1981. 9. " LION, Temperature Distributions for Arbitrary Shapes and Complicated Boundry Conditions," KAPL-M-6532, July 27,1966c

10. Letter from A. E. Scherer to D. G. Eisenhut dated 12 August 1983, LO-83-074,

Subject:

Natural: Circulation Cov.t :<a Reanalysis for CESSAR-F. 125

f

11. " Combustion Engineering Emergency Procedure Guidelines," CEN-152 Rev.'01, November, 1983.

12. " San Onofre Nuclear Generating Station Unit 2 Startup Report to the Nuclear Regulatory Commission," Docket Number 50-361, for the period ending 8 August 1983.

13. " Safety Evaluation Report Related to the Operation of St. Lucie Plant Un'. 2," NUREG-0843, Docket No. 50-389, December,1981.

14. " Safety. Evaluation Report Related to the Operation of Waterford Steam Electric Station, Unit No. 3," NUREG-0787, Docket No. 50-382, July, 1981.

15. " Final Safety Analysis Report, Waterford Steam Electric Station Unit No. 3, Louisiana Power & Light," Amendment No. 34, January,1984.

e 126

.A .A_ ~ aa .--a .-6, k Coll 8USTION ENGINEERING, INC. 1888 PROSPECT Hill RD. WINDSSR, CT 86095 i 1 ( l I .....--}}