ML20236Q905

From kanterella
Jump to navigation Jump to search
Non-proprietary,Rev 0 to Torus Temp & Pressure Response to Large Break LOCA & MSLB Accident Scenarios
ML20236Q905
Person / Time
Site: Vermont Yankee Entergy icon.png
Issue date: 04/27/1998
From: Fago C, Song L, Yaung W
MPR ASSOCIATES, INC., VERMONT YANKEE NUCLEAR POWER CORP.
To:
Shared Package
ML19325F690 List:
References
VYC-1628(NP), VYC-1628(NP)-R, VYC-1628(NP)-R00, NUDOCS 9807210113
Download: ML20236Q905 (120)


Text

_ _ _ _ _ _ _ _ _ _ _ _ - - - - - . _ _ _ _ _ - - - - _ _ _ _ _ _ . - - - _ _ _ _ _ _ _ _ - . . - - - - - - - - - - - - - - - - - - - - - - - . --- - -

b ORIGINAL: PAGE 1 of /f63 PAGES FOR INFORMATION Rev.1: PAGE 1 of PAGES Rev. 2: PAGE 1 of PAGES Rev. 3: PAGE 1 of PAGES QA RECORD?

X., YES D

T[]N RECORD TYPE NO. 13.C16.036 13.C09.001 l NO /p Safety Class /P.O. NO. (if applicable) WRF 98-0079-00 YANKEE NUCLEAR SERVICES DIVISION CALCULATION / ANALYSIS FOR Tm Torus Temperature and Pressure Response to Large Break LOCA and MSLB Accident Scenarios PLANT VY CYCLE N/A CALCULATION NUMBER VYC-1628 Rev. O PREPARED BY' REVIEWED BY APPROVED BY SUPERSEDES

, /DATE /I) ATE /DATE CALC /REV. NO.

/' gM' VYC-1290 R/0

/ V y/p,f/fr VYC-1290 R/l o f[#h E. F. Goodwin VYC-1290 R/2 hl$8 S ^ik,19s Y $7, ORIGINAL , / ( . R. Chap an l

<as-y l L. . Song MPR Associates N* N&G

3. S. Hsich 4 27-t'2 i

KEYWORDS GOTHIC 50e, RELAPSYA-B1 A. Wetwell. Torut. LOCA (Loss of Coolant Accident). LPCI (Low Pressure Coolant Iniection). CS (Core Snray), MSLB (Main Steam Line Break),

ECCS (Emerrency Core Cooline System). NPSH (Net Positive Suetion Head)

! L'5C5 ((kreSrn ay (-4 Mre-1)

COMPUTER CODES: RELAP5YA-B1 A. GOTHIC 50e, EXCEL 5.0. Fh <Av r v -1 C6 EQUIP / TAG NOs.: P-10-1 A. B. C. D: P-46-1 A. B SYSTEMS: RHR, Core Spray. Primary Containment

REFERENCES:

See pares 14.38,62.69.325.430.568.(14Iett ie 4, Ichi FORM WE-103-1 9807210113 980710 Revision 4 PDR ADOCK 05000271 '

P PDR 8

Details of indwidual contnbutions are after the Table of Contents.

n n a n n s t-T A MV qy* 7 n U F L\ L [. L tm r

&3 NW.NfJGDWL Y afG$/ss) t - - _ _ .______-____-____--_ ]

i TABLE OF CONTENTS l

1.0 Objective and Background . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Non-l 1.1 ECCS Pump NPSH, Analytical Perspectives . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . Non-P l 1.2 ECCS Pump NPSH, Regulatory Perspective . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Non-Pr

t. 1.3 ECCS Pump NPSH, Plant Historical Pers l 1.4 References . . . . . . . . . . . . . . . . . . . . . . . .pective . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

1

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Non-Proprietary 2.0 Method and System Description . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Non Pr{

2.1 System Description . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Non-Pro l

2.2 LBLOCA and MSLB Event Description . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Non-Proprie l 2.3 Suppression Pool Temperature and Wetwell Pressure Phenomenolo 2.4 Method Description . . . .. . . . . . . . . . . . . . . . . . . . . . .,. . . . . . . . . . . .gical Perspective .

2.4.1

. . . . . . . . . . . . . . . . Non-Proprietary )

Modeling Issue - Treatment of Feedwater . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Proprietary l 2.4.2 Modeling Issue - Single Node Vessel . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Proprietary 2.4.3 Modeling Issue - Break Reverse Flow . . . . . . . . . . . . . . . . . .

2.5 Code Description . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Propri

! . . . . . . . . . . . . . . . . Non-Proprietary 2.6 Method and Model Benchmarking . . . . . . . . . . . . . . . . . . . . . . . .

2.7 Model Description . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Non-Propr

. . . . . . . . . . . . . . . . . . Non. Proprietary 2.8 Calculation Organization and Presentation . . . . . . . . . . ,

2.9 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Non-Proprieta

. . . . . . . . . . . - . . . . . . . . . . . . . Non-Proprietary 3.0 Evaluation Cases and Scenario Description . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Non-Propr 3.1 Major Sensitivity Parameters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Non-Proprie )

3.1.1 Offsite Power . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Non-Proprietary 3.1.2 Single Active Failure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Non-Proprietary

, 3.1.3

' Post-Accident Containment Cooling Mode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Non-Proprietary 3.1.4 Containment Spray . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Non-Proprietary 3.1.5 RHR Service Water Flow . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Non-Proprietary )

3.1.6 Additional Heat Sources and Sinks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . I

~

3.1.7 Additional Uncertainties . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Non-P 3.2 Limiting ECCS Operation Cases . . . . . . . . . . . . . . . . . . . . .. . . . . . .. .. Non-Proprietary . . . . . . . . . . . . . . . No l

3.3 Individual Analysis Case Description . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Non-Proprietary 3.3.1 Run 0 - Base Case Specified By Vermont Yankee . . . . . . . . . . . . . . . . . . . . . . . . Non. Proprietary 3.3.1.1 Run 0 - Feedwater For VY Specified Base Case . . . . . . . . . . . . . . . . . . . . . . Non-Proprietary 3.3.2 Run 1 - DC-1 Single Failure (Table 3.1, Case 2) . . . . . . . . . . . . . . . . . . . . . . . . . Non-Proprietary 3.3.3 Run 2 - Recirc Discharge Line Valve Sensitivity . . . . . . . . . . . . . . . . . . . . . . . . . Non-Proprietary 3.3.4 Run 3 - RHR Heat Exchanger Failure (Table 3.1, Case 1) . . . . . . . . . . . . . . . . . . Non-Proprietary 3.3.5 Run 4 - ECCS Flow Sensitivity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Non Proprietary 3.3.6 Run 5 - RHRSW Flow Sensitivity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Non-Proprietary 3.3.7 Run 6 - Sho t-Term Torus Temperature Sensitivity . . . . . . . . . . . . . . . . . . . . . . . Non-Proprietary

, 3.3.8

' Run 7 - Feed Rate Sensitivity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Non-Proprietary 3.3.9 Run 8 - RHR Pump Failure Sensitivity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Non-Proprietary 3.3.10 Run 9 - Minimum Overpressure Sensitivity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Non-Proprietary 3.3.11 Run 10 - Main Steam Line Break . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Non-Proprietary 3.3.12 Run 11 - FSAR Benchmark . . . . . . . . . . . . . . . . . . . . . . .

3.4 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Non-P

. . . . . . . . . . . . . . . . . . . Non-Proprietary 4.0 Assumptions, inputs and Initial Conditions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Proprietary 4.1 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Proprietary l

CDF. VYC-16-lEP Wy 8.1998 7

l = 5.0 Mass and Energy Release . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Prop l

' 5.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Proprie l

5.2 Methodology for Mass and Energy Release Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Proprieta l -5.2.1 Compliance with NRC's Standard Review Plan for Mass and Energy Release Calculation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Proprietary 5.3 Analysis and Calculation Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Propriet 5.3.1 Initial Conditions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

5.3.2 Steady State Calculation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. .. .. ... .. .. Proprietary . . . . . . . . . . . . Pro i 5.3.3 i

Transient Model Development and Calculation Results . . . . . . . . . . . . . . . . . . . . . . . . Proprietary 5.3.3.1 Run 1 - Recirculation Pump Suction Line DEG LOCA . . . . . . . . . . . . . . . . . . . . . Proprietary l 5.3.3.2 Run 10 - Main Steam Line Lar l 5.4 Sensitivity Cases . . . . . . . . . . . . . . . . . . . . .ge Break LOCA . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Proprietary 3.4.1 Run 0 - Base Case Specified by Vermont Yankee . . . . . . . . . . . . . . . . . . . . . . . . . . . . Proprietary

5.4.2

' Run 2 - Recirc Discharge Line Valve Sensitivity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Proprietary 5.4.3 Run 3 - RHR Heat Exchanger Failure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Proprietary l 5.4.4

' Run 4 - ECCS Flow Sensitivity . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Proprietary 5.4.5 Short Term Torus Temperature Sensitivity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Proprietary i 5.4.6 Feedwater Flow Rate Sensitivity . . . . . . . . . . . . . . . . . . . . .

l 5.5 Benchmark . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Pro

. . . . . . . . . . . . . . . . . . . . . Proprietary 5.6 S um many . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

l 5.7 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Pr

. . . . . . . . . . . . . . . . . . . . . . Proprietary i

1 Microfiche Listing for Section 5 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Proprietary 6.0 Responses of the Primary Containment . . . . . . . . . . . . . . . . , , . . . . . . . . . . . . . . . . . . . . . . . . . . . Propnetar 6.1 Development of the GOTHIC Base Model input . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Propnetary i

, 6.2 Run0 - Vermont Yankee Base Case . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .e. . Propn. tary i

, 6.3 Run t - DC ,1 Sm, gle Failure (Table 3.1, Case 2) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Propnetar 6.4 Run2 - Recirculation Discharge Line Valve Sensitivity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Proprietary i.

6.5 Run3 - RHR Heat Exchanger Failure (Table 3.1, Case 1) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Propnetary l 6.6 Run4 - ECCS Flow Sensitivity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .e . Propn. tary l

6.7 Run5 - RHRSW Flow Sensitivity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Propn. e tary L

6.8 Run6 Short Term Torus Temperature Sensitivity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Proprietary 6.9 Run7 - Feed Rate Sensitivity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Propn. e tary 6.10-Run8 - RHR Pump Failure Sensitivity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Propnetary 6.11 Run9 - Mm,i , mum Pressure Sensitivity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Propnetary 6.12 Run10 - Main Steam Line Break . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Propnetary i 6.13 u Summary of Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Propnetary 6.14 List of Microfiche Dayfiles for Run0 through Run8 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Proprietary 6.15 Re ferences . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Proprietary 7.0 Results and Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Non-Proprietary 8.0 QA Compliance Documents . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Proprietary WE-103 Evaluation of Computer Code Use . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Proprietary NED NED Analysis Process Checklist . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Prr ietary NED NED WE-103 Review Checklist . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Proprietary WE-103 Calculation / Analysis Review . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Proprietary Appendix A Comparison of Analysis Method to SRP Guidelines . . . . . . . . . . . . . . . . . . . . . . . . . . Proprietary Appendix B Input Assumpt:ons Memorandum, D. E. Yasi to J. R. Chapman . . . . . . . . . . . . . Non-Proprietary t _

cnr.vve.16-twr May s.1998

- -_ _ _ _ - _ - _ _ - - - - - - - - - - _ - - - - - - - - - - - - - - - - - - -- - - - - ./

,w .

~f f . .

Appendix C =

GOTHIC Primary Containment Model Development . . . . . . . . . . . . . . . . . . . . . . . . . . Proprietary Appendix D Model and Method Benchmark Against FSAR DBA Containment Response . . . . . . . Proprietary

' Appendix E - RHR Heat Exchanger Model Benchmark . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Prop Appendix F Data Transfer Between RELAP5YA and GOTHIC . . . . . . . . . . . . . . . . . . . . . . . . . . . Proprietary Appendix G ~ Decay Heat Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Proprietar Appendix H L RELAPSYA LOCA Model Development . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Proprietary H.14 . Initial Stored Energy [ PROPRIETARY] . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . GE - Proprietary Appendix I L - (Not Used)

.' Appendix J Condensate and Feedwater System Availability . . . . . . . . . . . . . . . . . . . . . . . . . . Non-Proprietary Appendix K Calculation of RHR Flow Splits to the Heat Exchanger . . . . . . . . . . . . . . . . . . . . Non. Proprietary Appendix L -

Transmittal Memorandum & Margin Assessment Memorandum . . . . . . . . . . . . Non-Proprietary

.j Appendix M; GOTHIC Run0 - Model Input and Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Proprietary Appendix N GOTHIC Run i - Model Input and Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Proprietary

~ Appendix O . GOTHIC Run2 - Model Input and Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Proprietary

~

- Appendix P - GOTHIC Run3 - Model Input and Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Proprietary

. Appendix Q GOTHIC Run4 - Model Input and Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Proprietary a

Appendix R - GOTHIC Run5 - Model Input and Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Proprietary

. Appendix S GOTHIC Run6 - Model Input and Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Proprietary Appendix T GOTHIC Run? - Model Input and Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Proprietary

{

Appendix U-p GOTHIC Run8 - Model Input and Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Proprietary l

l

~

l i

i CDr.VYC 16-l.WP May B.1998 j

/

_ _ _ - - - _ - - _ _ - - - _ - _ _ - _ - - _ _ _ - _ - _ _ _ _ __-- - _ - - _ - __ - - _ _ - = _ =

Torus Temperature and Pressure Response to Large Break LOCA cnd MSLB Accident Scenarios VYC.1628 Rev. O Pige 2 TABLE OFCONTENTS Preparer Reviewer I

' 1.0 Objective and Background . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 CDF EFG )

1.1 ECCS Pump NPSH. Analytical Penpectives . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8 (

l.2 ECCS Pump NPSH. Regulatory Penpoetive . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9 13 ECCS Pump NPSH. Plant Historical Perspective . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11 1.4 n.r. ............................................................. u 2.0 Method and System Description . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15 CDF EFG 2.1 System Description . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15 2.2 LBLDCA and MSLB Event Description . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17 23 Suppression Pool Temperature and Wetwell Pressure, Phenomenological Perspective . 21 2.4 Method Description . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24 2.4.1 - Modeling Issue - Treatment of Feedwater . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26 2.4.2 Modeling Issue - Single Node Vessel . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27

- 2.43 ~. Modeling Issue - Break Reverse Flow . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38

. 2.5 Code Description . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30 2.6 Method and Model Benchmarking . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31 ,

2.7 Model Description . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . 32 2.8 Calculation Organization and Presentation '. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37 2.9 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 8 3.0 Evaluation Cases and Scenario Description . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39 CDF EFG 3.1 Major Sensitivity Parameters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39 3.1.1 - Offsite Power . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . 40 l

3t' 3.1.2 Single Active Failure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40 3.13 Post-Accident Containment Cooling Mode . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41 3.1.4 Containment Spray . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42 3.1.5 . RHR Service Water Flow . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42 3.1.6 - Additional Heat Sources and Sinks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42 l 3.1.7 ' Additional Uncertainties . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42 l 3.2 Limiting ECCS Operation Cases . . . . . '. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44 l 33 Individual Analysis Case Description . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47 l- 33.1 Run 0 - Base Case Specified By Vermont Yankee . . . . . . . . . . . . . . . . . . . . . . . 49 L ' 33.1.1 Run 0 - Feedwater For VY-Specified Base Case . . . . . . . . . . . . . . . . . . . . . 50 L 33.2 Run ! - DC-1 Single Failure (Table 3.1, Case 2) . . . . . . . . . . . . . . . . . . . . . . . . 52 l 3.33 Run 2 - Recire Discharge Line Valve Sensitivity . . . . . . . . . . . . . . . . . . . . . . . . 53 33.4 Run 3 - RHR Heat Exchanger Failure (Table 3.1, Case 1) . . . . . . . . . . . . . . . . . 54 l 33.5 - ~ Run 4 - ECCS Flow Sensitivity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55 {;

1-33.6 Run 5 - RHRSW Flow Sensitivity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 56 33.7 = Run 6 - Short-Term Torus Temperature Sensitivity . . . . . . . . . . . . . . . . . . . . . . 57 33.8 Run 7 - Feed Rate Sensitivity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58 4 33.9 Run 8 - RHR Pump Failure Sensitivity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59 33.10 Run 9 - Minimum Overpressure Sensitivity . . . . . . . . . . . . . . . . . . . . . . . . . . . . 60 33.11 Run 10 - Main Steam Line Break . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61 33.12 Run 11 - PSAR Benchmark . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61 3.4 References . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62 ,

. j 4.0 Assumptions, Inputs and Initial Conditions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63 CDP EFG 1 4.1 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 69 GF.VVe4HV Apr5 25. 8998 4

, Torus Temperature and Pressure Response to Large Break LOCA and MSLB Accident Scenarios .

i: - , VYC.1628 Rev. 0 Page 3 t 5.0 Mass and Energy Release . . . . . . . . . .' . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 70 WSY LS/MB

' i 5.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 70

' l ' 5.2 Methodology for Mass and Energy Release Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . 72 5.2.1 Compliance with NRC's Standard Review Plan for Mass and Energy Release Calculation . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . 74 N 5.3 Analysis and Calculation Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 75 5.3.1 Initial Conditions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . 75

.3 <

5.3.2 Steady State Calculation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77 5.3.3 - Transient Model Development and Calculation Results . . . . . . . . . . . . . . . . . . 90 5.3.3.1 Run 1 - Recirculation Pump Suction Line DEG IDCA . . . . . . . . . . . . . . . 92 H 5.3.3.2 Run 10 - Main Steam Line Large Break LOCA . . . . . . . . . . . . . . . . . . . . 132 L '54 SensitivityCases ......................................................151-5.4.1 Run 0 - Base Case Specified by Vermont Yankee . . . . . . . . . . . . . . . . . . . . . . 151 -

5.4.2 . Run 2 - Recirc Discharge Line Valve Sensitivity . . . . . . . . . . . . . . . . . . . . . . . 197

'L 5.4.3 Run 3 - RHR Heat Exchanger Failure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 213 l

.' 5.4.4 - Run 4 - ECCS Flow Sensitivity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 231 5.4.5 Short Term Torus Temperature Sensitivity . . . . . . . . . . . . . . . . . . . . . . . . . . . 249 5.4.6 Feedwater Flow Rate Sensitivity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 267 l' 5.5 Benchmark . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 284 5.6 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 23 5.7 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 325 l Microfiche Listing for Section 5 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 327 6.0 Responses of the Primary Containment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 332 LYS JSti l 6.1 Development of the OOTHIC Base Model Input . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 334 JSH

$... - 6.2 Run0 - Vermont Yankee Base Case . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . WSY. 360 5

  1. ' 6.3 Runi - DC-1 Single Failure (Table 3.1, Case 2) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 366 WSY 6.4 Run2 - Recirculation Discharge Line Valve Sensitivity ' . . . . . . . . . . . . . . . . . . . . . . . . 373 WSY.

6.5 Run3 - RHR Heat Exchanger Failure (Table 3.1, Case 1) . . . . . . . . . . . . . . . . . . . . . . . 380 WSY 6.6 Run4 - ECCS Flow Sensitivity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 393 WSY t-

[.  : 6.7 Run5 - RHRSW Flow Sensitivity . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 400 WSY 6.8 Run6 - Short Term Torus Temperature Sensitivity . . . . . . . . . . . . . . . . . . . . . . . . . . . . 405 WSY

. 6.9 Run7 - Feed Rate Sensitivity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 412 WSY j 4 6.10 Run8 . RHR Pump Failure Sensitivity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 419 WSY l 6.11 Run9 - Minimum Pressure Sensitivity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 424 WSY
l. 6.12 Run10 - Main Steam Line Break . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 425 WSY 1 6.13 L Summary of Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 426 WSY 6.14 List of Microfiche Dayfiles for Run0 through Run8 . . . . . . . . . . . . . . . . . . . . . . . . 428 WSY

' 6.15 References . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 430 JSH/WSY

. i

' 7.0 Results and Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 432 CDP EFG

' 8.0 QA Compliance Documents CDF EFG l

WE-103 Evaluation of Computer Code Use . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 437 NED.6 - NED Analysis Process Checklist . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 438 NED NED WE-103 Review Checklist . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 440

. WE.103 Calculation / Analysis Review . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 443 j

t Appendix A Comparison of Analysis Method to SRP Guidelines . . . . . . . . . . . . . . . . . . . . 521 CDF LS i

Appendix B - Input Assumptions Memorandum, D. E. Yasi to J. R. Chapman . . . . . . . . . . . 529 DEY N/A i

ar.vieseemP Are25Jove t

l /

. _ _ - _ _ _ _ _ _ _ - _ _ _ _ - _ . _ - _ _ - - _ _ _ _ _ _ . _ _ _ - _ J

Appendix C GOTHIC Piimary Containment Model Development . . . . . . . . . . . . . . . . . . 543 CDF EFG

- WE-103 3 - Calculation / Analysis Review . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 569 ppendix D - Model and Method Benchmad Against FSAR DBA Containment Res LYS

. WE.103 3 - Calculation / Analysis Review . . . . . . . . . . . . . . . . . . ....... . . . . . .754. . . . . . .ponse . 577 JSH Appendix E - RHR Heat Exchanger Model Benchmark . . . . . . . . . . . . . . . . . . . . . . . . . . . . 760 CDP JS WE-103 Calculation / Analysis Review . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 770 Appendix F - Data Transfer Between RELAPSYA and GOTHIC . . . . . . . . . . . . . . . . . . . . . 810 CDF EFG WE.103 3 - Calculation / Analysis Review . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 856 Appendix G Decay Heat Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 859 CDF EFG i

I Appendix H RELAPSYA LOCA Model Development . . . . . . . . . . . . . . . . . . . . . . . . . . . . 891 WSY LS/MB H.14 Initial Stored Energy [ PROPRIETARY] . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1001 KES GBP AppendixI (Not Used)

Appendix J Condensate and Feedwater System Availability . . . . . . . . . . . . . . . . . . . . . . . 1068 RTF N/A (Not a QA h=% forinformation only)

- Appendix K Calculation of RHR Flow Splits to the Heat Exchanger . . . . . . . . . . . . . . . . . 1082 KRR EFG WE.103 Calculation / Analysis Review . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1126

h Appendix L . Transtnittal Memorandum & Margin Assessment Memorandum . . . . . . . . . 1128 CDP EFG i Tj Appendix M = GOTHIC Run0 - Model Input and Results . . . . . . . . . . . . . . . . . . . . . . . . . . . 1163 ' LYS JSH a Ippendix N GOTHIC Runt - Model Input and Results . . . . . . . . . . . . . . . . . . . . . . . . . . . 1215 LYS WSY Appendix 0 GOTHIC Run2 - Model Input and Results . . . . . . . . . . . . . . . . . . . . . . . . . . . 1274 LYS WSY Appendix P GOTHIC Run3 - Model Input and Results . . . . , . . . . . . . . . . . . . . . . . . . . . . 1334 LYS WSY Appendix Q GOTHIC Run4 - Model Input and Results . . . . . . . . . . . . . . . . . . . . . . . . . . . 1393 LYS WSY Appendix R . GOTHIC Run5 - Model Input and Results . . . . . . . . . . . . . . . . . . . . . . . . . . . 1452 LYS WSY
Appendix S GOTHIC Run6 - Model Input and Results . . . . . . . . . . . . . . . . . . . . . . . . . . . 1460 LYS WSY Appendix T GOTHIC Run7 - Model Input and Results . . . . . . . . . . . . . . . . . . . . . . . . . . . 1519 LYS WSY Appendix U GOTHIC Run8 - Model Input and Results . . . . . . . . . . . . . . . . . . . . . . . . . . . 1574 LYS WSY CDF- Carl D. Fago, DE&S - Bolton KRR - Ken R. Rousseau, DE&S - Bolton DEY- Dan Yasi, Vermont Yankee LYS -Ling Yu Song, MPR Associates EPG - Ed Goodwin, DE&S -Bolton MB - Mad Bolander. Computer Simulation Associates i JS-Julien Shirkov, DE&S - Bolton RTF - R. Tom Femandez, Independent Contractor  !

' JSH - Jen Sheng Hsich, DE&S - Bolton WSY - Moses Yeung, DE&S - Bolton i KES - Kevin St. John, DE&S - Bolton GBP - Gary Peeler, Nuclear Engineering Technology l ar.vycewr wnam

.Y

l Torus Temperature and Pressure Response to Large Break LOCA and MSLB Accident Scenarios VYC 1628 Rev. 0 PJge 5 1.0 Objective and Background This calculation has three objectives. Thefirst objective is to develop a methodologyfor calculating the long term suppression pool temperature and torus pressure response to a large break loss ofcoolant accident (LBLOCA) and main steam line break (h!SLB). The second objective is to analyze the suppression pool temperature and torus pressure response to a variety ofLBLOCA and ofSLB scenarios for use in calculating the limiting e*nergency core cooling system (ECCS) pump netpositive suction head l (NPSH). The third objective is to determine the peak suppression pool temperaturefollowing specific LBLOCA events in support ofthe replacement ECCS suction strainer design basis and toform part of the

{

basisfor the technical specification on torus temperature. j

{

The Vetmont Yankee containment is a General Electric Mark I pressure suppression containment. It l consists of a drywell which contains the reactor pressure vessel, recirculation loops and support equipment, a wetwell which contains the suppre::sion pool and a vent system which connects the drywell and the wetwell. The vent system is arranged so that during a LOCA or MSLB steam, air and water in the drywell is directed to the wetwell through the suppression pool. '

l The primary containment provides the capability, in the event of the postulated loss-of-coolant accident, to limit the release of fission products to the plant environs so that off site doses would be within 10CFR100 l limits. One of the methods used to provide this capability is to direct the steam from the LOCA through

]'

the vent system'to the suppression pool thus condensing the steam and limiting the subsequent peak pressure below the containment design pressure.

Additionally, the suppression pool provides a source of water to the ECCS pumps for long term core cooling. The ECCS pumps of concern are the core spray (CS) and residual heat removal (RHR) pumps.

These pumps, in combination, ensure the delivery of sufficient cooling water flow providing post-accident core and containment cooling through the RHR heat exchanger. Thus, CS and RHR pump operability is significant to the safety of the plant. One of the primary design factors affecting pump operability is available versus required NPSH. The available NPSH depends on the post-LOCA torus water level, the suppression pool temperature, the pressure in the toms and the head loss through the pump suction piping.

The available NPSH is being re-evaluated, using, in part, the results of this calculation, in order to addre, 2 uncenainties not previously analyzed, provide a clearer design basis and identify potential margin improvement opportunities. The required NPSH is determined by the pump manufacturer and is provided on the pump curves.

The VY FSAR [1.1) contains information on previous containment analyses. The pressure and temperature response of the containment to a design basis accident (DBA) LOCA is contained in Section 14.6.3.3 of the VY FSAR. The FSAR containment analysis consists of two pans, a shon-term peak pressure analysis and a long-term pressure and temperature response analysis. The shon-term peak I

pressure analysis contained in the FSAR is the DBA analysis contained in the Plant Unique Load l

CDP.YYC.464 %7 Aprg 21. l998

_ _ _ _ _ _ _ _ _ _ _ _ - _ _ _ - - )

Torus Temperature and Pressure Response to Large Break LOCA and MSLB Accident Scenarios VYC-1628 Rev.0 -

Page 6 Definitions Report (PULDR). [1.2] The long-term pressure and temperature response analys's is the original analysis performed for original plant licensing. As well as the DBA LOCA, containment pressure and temperature responses to other accidents such as { small break LOCAs, MSLBs, Appendix R transients, anticipated operational transients (AOTs) and safety

         . relief valve (SRV) transients are required to be analyzed to ensure all containment design criteria are met.

The analysis initial conditions and results provide the basis for the Technical Specification on initial torus temperature and the limits on suppression pool water volurne. The lory-term containment original design basis was a double ended guillotine (DEG) suction line break detailed in *SA R 3ection 14.6.3. The initial suppression pool temperature assumed in this original analysis was 90*P resulting in a limiting maximum suppression pool temperature post-LOCA of about 166T. General Electric performed the initial containment analyses for Vermont Yankee. As a result of the Mark I Containment program, the analyses were redone in the late 70's and early 80's. The Plant Unique Lead Definition (PULD) report analyzed large, intermediate and small break LOCAs for the purpose of analyzing the torus loads. The cunent short-term peak pressure design basis analysis as described in the l FSAR is from the Plant Unique load Definition (PULD) report. License Amendment 88 dated June 6,1985 increased suppression pool initial temperature from 90T to 100T. 'Ihe application for this amendment focused on changes related to SRV blowdown. More recent i. Q' examinations revealed that this amendment impacted a number of other issues for which assessments and analyses were not part of the original submittal to the NRC. EDR 94 05 was issued to address these j deficiencies. BMO 96 05 was written tojustify temporary operation at an initial suppression pool l temperature of 907 pending final resolution of the deficiencies identified in the submittal for License Amendment 88. i As a result of the efforts to address the deficiencies, it was determined that the design basis for suppression pool temperature is not entirely clear, did not address all potential uncertainties and may not have been sufficiently conservative. In addition, several margin improvement opportunities were discovered. Thus, ( it was desired to initiate a complete long-term containment analysis which would address the previous l deficiencies as well as potential margin improvement. To this end, this calculation presents a methodology I I for calculating the long term suppression pool temperature and torus pressure response to a LBLOCA and MSLB. This calculation isjust one part of a comprehensive containment analysis effort. Separate analyses are planned to follow to address other postulated accident or safe shutdown scenarios such as { small/ intermediate break LOCAs, small/ intermediate steam line breaks, SRV transients,10CFR50 l Appendix R (Fire Protection), use of the attemate cooling system and station black-out. This calculation presents a methodology that improves on the cunent peak suppression pool temperature design basis methodology [1.3][1.4]. This calculation's methodology consists of a LOCA analysis or.nc wr .

                                                                                                                                                  *nt. im e                                                                                                                                                                 ,
                  - Torus Temper:ture and Pressure Response to Large Break LOCA rad MSLB Accident Scenarios VYC-1628 Rev.0                                                                                                                            Page 7 calculation of break mass and energy release based on the RELAP5YA-BI A [1.5] computer cocle and a
   +

containment response method using the GOTHIC 5.0e [1.6] computer code. Input assumptions have been chosen to ensure the calculated suppression pool temperature and pressure are conservative when used for calculating NPSH available for the ECCS pumps. As well, conservative input assumptions have been used to ensure the calculated peak suppression pool temperature is maximized. In support of the third objective, this analysis determines the peak suppression pool temperature for a range of specific cases in support of the bid specification for the replacement ECCS suction strainers. NRC Bulletin 9643 tasks utilities with addsessing high post-LOCA debris loading on their suction strainers.

                                                                                                                                                                        )

VY has committed to installing new ECCS suction strainers as a result. The peak suppression pool L.@&ure cases that are analyzed consider these potential debris loading scenarios in determining the potential limiting p'eak suppression pool temperature scenarios. In support of the bid specification, the highest peak suppression pool temperature, following a LBLOCA or MSLB, is calculated for: any scenario where only one RHR pump is running in an RHR train. any scenario where two RHR pumps are continuously running in an RHR train, and

                   =

any scenario with s core spray pump operating.' k1 i l l 8 The limiting suppression pool temperature for the last case is the higher peak suppression pool temperature of eit her of the prior two cases since any postulated scenario always has at least one core spray pump operating. Howver, the cases ax presented in this order to be consistent with the bid speciGcation. C3F.YVCl6ELINP Ap80 21,1998

Torus Temperature and Pressure Response to Large Break LOCA cnd MSLB Accident Scenarios VYC-1628 Rev. 0 P:ge 8 1.1 ECCS Pump NPSH, Analytical Perspectives 1

                'Ihe value of available NPSH for the ECCS pumps is required to ensure it exceeds the required NPSH to assure continued post-accident pump operability. NPSH,           is calculated from the following equation:

NPSHo, = (P,,,,, - P,,,(Tg))

  • 144 g' + h,,,,, - h f,e,% (1,1)

P8 i where: NPSH, m = the NPSH at the pumpinlet datum, ft Pw = the pressure above the water surface, psia P.,(Tg) = the vapor pressure of the suction water source, psia Tg = water source temperature, 'F { g, = unit conversion constant,32.174 ft-lbm/lbf-s2 l g = gravitational acceleration constant,32.174 ft/s2 p = water density, Ibm /ft' i

h. = the static head of the water column of the source above the pump inlet datum, ft
                                . h,%             = the friction and form losses through the piping and components from the water source to the pump inlet datum, ft I

] / The values calculated in this analysis are Pa and Tg; h can be calculated using output (torus water volume) from this calculation; h,m has some fluid temperature dependence. Available NPSH is usually calculated based on two time domains, short-term and long-term. The short-tenn time domain is characterized as that time prior to any assumed operator action to throttle pump flow. The long-term time I domain is characterized as that thne after any assumed operator action to throttle pump flow. Pwand Tg anc be highly interrelated in that the pool temperature and RHR heat exchanger effectiveness l determine the containment spray temperature which strongly influences torus pressure. And the pool { temperature is strongly influenced by the heat exchanger effectiveness which is fluid flow and condition I dependent. It is seen that the minimum NPSH occurs for the lowest total pressure and the highest pool temperature given constant values of static head and piping and component friction and form losses. A more detailed perspective on suppression pool temperature and wetwell pressure is provided in Section 2. Additionally, due to the potentially different inputs to NPSH during the each time domain, short and long-term, the suppression pool temperature nd pressure in each time domain is ofinterest. This calculation does not determine pump NPSH directly but the results can provide input to an NPSH calculation. j cor.vre.a.+wr vs n. im l

                                                                                               -                               J

I Torus Temperature and Pressure Response is Large Break LOCA and MSLB Accident Sc:narios VYC-1628 Rev. O Pige 9

               ' 1.2      ECCS Pump NPSH, Regulatory Perspective

'y ' The design and construction of Vermont Yankee meets the intent of the General Design Criteria (GDC) published in Appendix A of 10 CFR 50. The description of how the intent of the criteria is met is contained in Appendix F of the Vermont Yankee FSAR. The GDCs that specifically apply to this calculation are GDC 35 - Emergency Core Cooling,38 - Containment Heat Removal and 50 - Containment Design Basis. The calculation of peak suppression pool temperature relates to each of these GDC. The

   ,             results of this calculation are consistent wi.h VY FSAR Appendix F.

The guidelines for acceptable design and analysis to show that the applicable GDC are met is provided in NUREG 0800, Standard Review Plan (SRP) which post <iates the VY license. While VY has not committed to the SRP for containment analysis, the SRP is used as a reference and guideline for evaluating the appropriateness of method and modeling techniques us-A in this calculdion. The applicable ama8== criteria in the SRP have been evaluated with respect to this calculation; the evaluation is contained in Appendix A. Specific regulatory guidance with regard to calculation of pump NPSH is provided in NRC Regulatory Guide 1.1 (Safety Guide 1) [1.7]. Reg. Guide (RG) 1.1 requires that adequate NPSH be provided for the ECCS and containment heat removal system (CHRS) pumps. More specifically, RG 1.1 requires that the ECCS and CHRS pumps be designed with adequate NPSH assuming maximum expected temperatures of

  ,j pumped fluids and no increase in containment pressure from that present prior to postulated LOCAs (i.e.

no credit for accident-induced containment overpressure). Cunently, VY does not credit any wetwell l . pressure above atmospheric [1.8]. This calculation provides suppression pool temperature and wetwell pressure. The methodology l developed, with appropriate inputs, is sufficient to determine a conservative minimum NPSH margin given l- the constraints of RG 1.1. Additional NRC documentation has been reviewed for potentialimpact on this calculation. Some of these , documents relate directly to calculation of NPSH while others relate to input considerations. The calculation methodology and models are consistent with the advice, comments and wamings contained in these documents. These documents include:

              . NRC Information Notice 96-39: Estimates of Decay Heat Using ANS 5.1 Decay Heat Standard May Vary Significantly NRC Informuion Notice %-55: Inadequate Net Positive Suction Head of Emergency Core Cooling And Containment Heat Removal Pumps Under Design Basis Accident Conditions GF. VVC 164 WP                                                                                                Apr8 tl.1998

Torus Temperature and Pressure Re.sponse to Large Break LOCA rnd MSLB Accident Scenarios

                                                       . VYC 1628 Rev, O Pige 10
 ~

NRC Information Notice 97-27: Effect ofInconeet Strainer Pressure Drop on Available Net Positive j Suction Head NRC Information Notice 97-78: Crediting of Operator Actions in Place of Automatic Actions and Modifications of Operator Actions, Including Response Times NRC Generic I.etter 85 22: Potential for Loss of Post LOCA Recirculation Capability Due to Insulation Debris Blockage NRC Generic Letter 97-04: Assurance of Sufficient Net Positive Suction Head For Emergency Core Cooling and Containment Heat Removal Pumps NRC Bulletin 96 03: Potential Plugging of Emergency Core Cooling Suction Strainers By Debris in Boiling Water Reactors Analysis guidance is also contained in three applicable ANS standards. These standards are not specifically implemented unless otherwise stated in the body of the calculation because the methods are either already implemented or are bounded by the guidelines presented in the SRP. However, at a minimum, they provide additional guidance to analytical methods. The three applicable standards are: ANS 5.1 - Decay Heat Power in Light Water Reactors i.

                                                 ' ANS 56.4 - Pressure and Temperature Transient Analysis for Light Water Reactor Containments ANS 56.5 - PWR and BWR Containment Spray System Design Criteria                                                                                                                   4 l

I l i 1 l l EIIF.YTC 864WP Apre 28. f 998

Torus Temperature and Pressure Response to Large Break LOCA and MSLB Accident Scenarios VYC-1628 Rev. 0 Page 11 1.3 ECCS Purnp NPSH, Plant Historical Perspective

  )

This section provides a summary of the cunent plant design basis regarding ECCS pump NPSH and peak torus updure as well as the historical path to that design basis. The current design basis does not take I L credit for any torus pressure above thst present prior to the accident (nominally atmospheric pressure).

               *Ihus the minimum available ECCS pump NPSH nominally occurs when the suppression pool temperature is at a maximum. Other considerations, such as strainer, piping and fitting frictional and form losees may be less for higher suppression pool temperatures. However, the dominant factor involving temperature is generally the fluid vapor pressure. [1.9)

The current basis for showing that adequate NPSH is available to the ECCS pumps -consists of several interrelated calculations. Table 1.1 provides a concise description of the calculations that constitute the current ECCS NPSH design basis. Table 1.1 - ECCS Pum ) NPSH Design Basis Calculations Calc # Title Description VYC-1388 Rev. 0 Core Spray NPSH Evaluation Provides a suppression pool temperature (Apptoved 5/16/95) versus flow curve which corresponds to adequate NPSH for the core spray pumps assuming 14.7 psia torus pressure and no debris loading on the suction strainer.

 )               VYC-1389 Rev. 0           Residual Heat Removal (LPCI)NPSH          Provides a suppression pool temperature (Approved 5/19/95)        Evaluation                                versus flow curve that corresponds to adequate NPSH for the RHR purnps assuming 14.7 psia torus pressure and no debris loading on the suction strainer.

VYC-808 Rev. 2 Calculation of NPSH Margin forRHR Calculates the NPSH margin for the ECCS (Approved 12/6/95) and Core Spray Pumps at Design pumps assuming design suction strainer debris Flowrate Including Fibrous Insulation loading given a suppression pool temperature Debris on Intake Strainers from a and a system flow rate. Design Basis LOCA VYC-1290 Rev. 2 Vermont Yankee Post-LOCA Torus Calculates the peak suppression pool (Approved 8/8/96) Temperature and RHR Heat temperature for a LBLOCA (to be compared Exchanger Evaluation to the allowable temperatures in VYC-1388, 1389 and 808) l VYC-1442 Rev. O Peak Torus Temperature for Calculates the peak suppression pool (Approved 9/6/96) Intermediate and SmallBreak LOCA temperature for an IBLOCA and SBLOCA (to be compared to the allowable temperatures in l VYC-1388.1389 and 808) i i GF . YTC-44.WP AprB 28.1998 j

Torus T...,, ews and Pressure Response talarge Break thCA and MSLB Accident Scenarios

           - VYC-1628 Rev. 0                                                                                                      Page 12 Table 1.1 - ECCS Pum > NPSH Design Basis Calculations
 't Calc #                               Title                                     Description M                     VYC-1493 Rev. 0               Peak Torus Temperature Sensitivity to       Given an increase in service water (Approved 9/26/96)              Service WaterTemperature and Flow         temperature, this calculation determines the Rate                                       required increase in service water flow which will compensate for the increased service       I water temperature with respect to maintaining the naak torus temperature below 176'F.

VYC-1529 Rev. O VY WetwellTemperature Response Calculates the suppression pool temperature (Approved 10/18/96) to Appendix R Reactor Shutdown response for a variety of App. R safe l shutdown scenarios for use in determining  ; adacuate NPSH to the ECCS pumps. VYC-1034 Rev. 0 Evaluation of Torus Temperature Provided the documentation that VY met the (Approved 12/6/91) Transients for the 1631 MWth Uprate intent of NUREG-0783 with regard to torus I temperature limits during SRV transients. VYC-886 Rev.1 Station Blackout Documentation Provided the documentation for torus heat up (Approved 5/12/95) Analysis l following a station blackout. Oriainal Desion' Bacic 1 The long-term containment original design basis was a DEG suction line break detailed in FSAR Section l

 }'

14.6.3. The initial suppression pool temperature assumed in this original analysis was 90'F resulting in a l limiting maximum suppression pool temperature post-LOCA of about 166'F. Key assumptions relative to limiting long-term containment temperature response in this original analysis include (see Appendix D for funher discussion of PSAR long-term containment analysis): [1.9][1.10][1.11][1.12] l 1. Ims of Normal Power

2. Only the suppression pool acts as a heat sink in the primary containment
3. Operation of one RHR cooling loop including one RHR heat exchanger, one RHR pump and one RHRSW pump. l
4. May-Witt decay heat curve and sensible energy, developed by General Electric, using 1665 MWth as
          . the initial reactor power.
5. Initial vessel blowdown of 210.5e6 BTU, this energy included:

e energy from vessel metal structures from normal operating temperature (nominally,

                                        , saturation temperature at operating pressure) to 281*F (structures initially in contact with steam were assumed to remain at their initial temperature) e all vessel initial water inventory including inventory in attached piping.

1 Recie IJndatetK'haneet The SER to License Amendment 88 dated June 6,1985 allowed suppression pool initial temperature to be increased from 90*F to 100*F. The application for this amendment focused on changes related tc, SRV ar.vvem w w n. im

                                                                              .                                                                I a             _ _ _ _ - - _ - - -                          -                                                                                   J

Torus Temperature and Pressure Fesponse to Large Break LOCA and MSLB Accident Scenarios VYC-1628 Rev.0 Page 13 blowdown. Closer examination revealed that this amendment irbpacted a number of other issues for which 3 assessments and analyses were not part of the original submittal to the NRC. EDR 94-05 was issued to address these deficiencies. VYC-1290 Rev. O was written in July 1994 to determine the post-LOCA suppression pool temperature assuming 5% tube plugging of the RHR heat exchanger and 100*F initial suppression pool temperature. This analysis used assumptions similar to the original design basis (the decay heat curve was modified to ANS 5.1 - 1979 with 2a uncertainty included) but used a new methodology based on RELAPSYA mass and energy release to 10 minutes post-accident, GOTHIC containment response resulting from the RELAPSYA mass and energy release up to 10 minutes and a spreadsheet energy balance subsequent to 10 minutes. Subsequently, PAC 96-02 (and associated BMO 96-05) was initiated documenting concems among which included a determination that the assumption of Loss of Normal Power may be non-conservative for torus temperature analyses due to the potential for relatively hot feedwater addition post-LOCA. As a result of the PAC, Standing Order #19 was issued to limit suppression pool temperature to 90'F. I i 2 VYC-1290 Rev. 2 was written in July 1996, in support of the conclusions of BMO 96-05, to evaluate post-LOCA suppression pool temperature with the additional assumption of conservative feedwater 3 addition. The initial suppression pool temperature used in this evaluation was 90'F. Additionally, also in l 3 support of the conclusions of BMO 96-05, VYC-1457 Rev. O was written in August 1996 to evaluate post-1 LOCA suppression pool temperatures for break sizes other than large break. This conservative analysis identified that small break LOCAs may be limiting with respect to peak suppression pool temperature when feedwater addition is factored into the analysis. l l ER 98 0025 was written on 6 January 1998 to address other potential discrepancies, non-conservatism and/or errors discovered in previous torus temperature analyses as a result of the development of this l calculation. The issues described in ER 98-0025 are addressed in this calculation. Other design basis analyses have been performed with respect to toms temperature evaluation: VYC-1529 Rev. 0 was written in September 1996 to evaluate the torus temperature response (and thus ensure adequate ECCS pump NPSH) during various Appendix R fire events. SRV transients to meet the intent of NUREG-0783, Suppression Pool Temperature Limits for BWR Containment were reviewed in VYC-1034 Rev. O. VYC-1290 Rev. I addressed sensitivities in the Rev. O analysis to changes in RHR now or RHRsW now to the RIIR heat exchanger but did not otherwise modify the design basis. mr.vyct M wr l Apr6121. 499s

Torus Temperature and Pressure Response to Large Break LOCA and MSLB Accident Scenarios

VYC 1628 Rev. O Page 14 1.4 - References 1.1 Vermont Yankee Final Safety Analysis Report (FSAR), Revision 14 r 1.2 NEDO-24581 Rev.1, Mark I Containment Program Plant Unique I. cad Definition Vennont Yankee Generating Station, April 1981 1.3 Calculation VYC-1290 Rev. 0, Vermont Yankee Post-LOCA Torus Temperature and RHR Heat t

Exchanger Evaluation, appmved 1 August 1994 1.4 Calculation VYC-1290 Rev. 2, Vermont Yankee Post-LOCA Torus Temperature and RHR Heat l Exchanger Evaluation, approved 8 August 1996 1.5 R. T. Fernandez, et. al., "RELAP5YA - A Computer Program for Light-Water Reactor System Thermal Hydraulic Analysis," YAEC-1300P, October 1982, Revised June 1993 1.6 GOTHIC Containment Analysis Package, Version 5.0e, Users Manual, NAI 8907 02 Rev. 6 l 1.7 NRC Regulatory Guide 1.1 (Safety Guide 1), Net Positive Suction Head For Emergency Core Cooling and Containment Heat Removal System Pumps, dtd November 2,1570 l 1.8 Calculation VYC-808 Rev. 2, Calculation of NPSH Margin for RHR and Core Spray Pumps at

                                                                                                                              ~

Design Flowrate Including Fibrous Insulation Debris on Intake Strainers from a Design Basis LOCA approved 6 December 1995 l l

g.  ; 1.9 Calculation VYC-1762 Rev. O, Determination of Basis for FSAR Figure 14.6-7,14ss-of. Coolant Accident Suppression Pool Temperature Transient, approved 30 March 1998 1.10 ' Memorandum THSAG-VY 98-009, C. D. Fago to J. R. Hoffman, dated 10 February 1998

!- 1.11 General Electric Memorandum PACE-7, Pressure Suppression Primary Containment System l Topical Report, May 31,1966 1.12 General Electric Design Specification 22A1265, Rev.1, Vermont Yankee Primary Containment, September 18,1969 i l. I CDP.VVC-lHM Apr8 21.1998 l

1 Torus Temperature end Pressure Response to Large Breck LOCA and MSLB Accident Scenarios VYC-1628 Rev. 0 , Page 15

  .            2.0      Method and Systern Description
   .i This section provides a description of the analysis methodology, plant systems modeled and analysis models which result from the methodology and system discussions. First, a brief overview of the events to be analyzed as well as the systems to be modeled is provided. Second, the method of analysis is described as well as the codes used and theirjustification.

2.1 Systern Description Models of several different systems are used in this analysis. Understanding of system operation following a LBLOCA or MSLB is critical to the understanding of the phenomena analyzed herein. These brief system descriptions are intended only to provide a very rudimentary understanding of the systems involved and are not intended to be all inclusive or detailed. j The major systems involved in this analysis are:

               .        reactor vessel and coolant system
               .        reactor feedwater system
               .        primary containment a        low pressure coolant injection / residual heat removal system
    )          .        core spray system
               .        RHR service water system Figure 2.1 provides a schematic of the interface of these different systems. During a LBLOCA, the reactor vessel will depressurize and reactor coolant will be discharged to the primary containment. The reactor feedwater system may continue to deliver hot feedwater to the reactor vessel at the discretion of the              i operators dependent on availability and procedural guidance. Upon the proper actuation r:;aal, the LPCI and core spray system pumps will start. After the reactor vessel pressure drops below the pressure permissive, the injection valves for these two systems will open allowing water from the suppression pool          i l

to be delivered to the reactor vessel. After the operators are able to diagnose the situation the RHR service water pumps can be manually started allowing heat removal through the RHR heat exchangers. This aids in core cooling and removes heat from I the primary containment delivering it to the ultimate heat sink. The RHR system may be realigned allowing direct water retum to the suppression pool after passing through the RHR heat exchanger. Altemately, the operators may continue with the RHR system injecting into the vessel in LPCI mode through the RH,R heat exchanger. ! "Ihe operators may be directed by procedure to initiate drywell or wetwell spray. Drywell and wetwell spray is initiated manually by realigning valves which divert RHR flow from the RHR heat exchanger to spray headers in the drywell and wetwell. CDF .WC-164 WP ArwS 28.1998

T h T o O , g s. s ora

                            .                                                                                                                sp                     re
                                                                                             ~                                           ,_.                        p r-
                                                                                                                                              .p
                     %                                                                                         w %                            .m
                                                                                                                                              .u cP S

e t t o w i- r

                  .r                                                  *                              -                           (

oof tr so C e

                                                                                                                                         ,                                n
                 .e-n w

s w

                                                                      *                      ~

t o i.

                                     .c r,
                                                        *s                              W l,         j t

A,,"*  % , n e

                                                                                                                                                                              'r i

_ *t ' j , T m gf\ - N . *., n O e

                                                                 ,% '       m i

e 2 u . - ' - . t T.

                          ,    l                  i                ll                               _ .l: * , . ,                                               n i.

4 o m u

                                        .. G.c

{ C g ,.* . v ,- y m gII ,; t

                                                                                                                                         ~.w yI         -

h , * . ' .  : , r e

                                                    .S 4
 .                                -                                         o                                     .,.                                            m u

g. t P i r s

                         %w%                  m s-,r s

m., . i. t

                                                                             ,/

mo . p L e m "~'

                                                                        ..            o                .

c u o p l s s s t. w t.,. e r n t. u

                                                                                . e w
                                     ;                                                c                .                              _-

o w, w ii c. e m "' .

                                                                                                                                           ,                             h s

M

                                      .                                 N       t t                    u                              _-

s t.. n 7Y x%ro x . I i

                                                                                                                                                               ,' C        e
a. -

r r

                                      .                                                                      '     {                                               P t
                                                                  >                                                                        n>
                                                                                                                   }

L M X ,k

                                                                                                                                                ,d        @        /

o w Rt

                                                                                  ,                           &              g3 n            ^

Hf

                                                                                                  -                 z-                                    ,f       R o H                                                         e
                                                                                                                                                , k
                                                                                                                                      ,'                                    n
                                                                                                                                                ,d.                         o H                    ,-
                                                                                                                                                           .             t m                                       , e
                                                                                                                                                 ,,,,l     %

o Co V

                                                                                                                                                 ,, o.
                                                                                                                          ;                    nF,l        w L

M r-

                                                                                                                             ,I 7

E a.8 ! gyi l I

Torus Temperature and Pressure Response to Large Break LOCA cnd MSLB AccidInt Sc;narios VYC-1628 Rev.0 Page 17 All of these systems are modeled to the extent necessary as dictated by the individual scenarios analyzed and the objectives of the calculation. 2.2 LBLOCA and MSLB Event Description This section provides a summary description of the LBLOCA and MSLB events that are generally common to the scenarios analyzed in this calculation. The inputs and assumptions used in this description are included only for descriptive clarity and are provided in more detail in Sections 4,5 and 6. The basic event is a design basis LBLOCA which is described in the VY FSAR (Section 14.6.3); some portions are repeated here for clarity and identifying the differences (the DBA LOCA in the FSAR assumes offsite power is lost at the beginning of the accident). The MSLB is similar to a LBLOCA and its characteristics will be detailed following the LBLOCA. Table 2.1 provides a typical sequence of events for these accident scenarios. LBLOCA Two-phase critical flow at the break locations is quickly established after break initiation. The reactor is rendered suberitical due to void formation in the core as the system rapidly depressurizes. High drywell pressure initiates a scram in less than one second. The main steam isolation valves (MSIV) are shut at their fastest rate (3 see) following a reactor vessel low-low level signal with no credit taken for signal 4, response time. Additionally, the recirculation pumps trip on the reactor vessellow low level signal or MG under-frequency if offsite power is lost. Reactor vessel blowdown is essentially complete at 40 seconds and is completely depressurized by about 90 seconds. Either reactor vessel low-low level or high drywell pressure actuates the core standby cooling systems. The RHR system is aligned to the reactor vessel in LPCI mode and the core spray system is energized. When the pressure permissive setpoint is reached the LPCI and CS pumps inject water directly I to the core. I Feedwater flow increases as the vessel pressure decreases. The feedpumps reach mnout condition and feed flow is ma'ximized. The temperature of the incoming feedwater decreases as energy is removed from the feedwater heaters. The energy in thc feedwater heaters continues to decrease as feedwater removes energy. Steam supply to the shell side of the feedwater heaters is terminated due to the MSIV c*osure. The feed system is non-nuclear safety grade and its behavior is conservatively treated by non-mechanistically terminating feedwater addition after all the hot feed is added to the system. Hot feed is defined as that feedwater which would contribute to an increase in peak suppression pool temperature. The vessel is refilled by the injection from the core standby cooling systems and the feedwater cddition. l The vessel and core are cooled by this injection flow. The injection flow transfers the core and vessel sensible heat and the core decay heat to the suppression pool via break spillage and the drywell and vent system. cw.vmw w n. im l j

Torus Temper-ture and Pressure Response to Large Break LOCA cnd MSLB Accid:nt Sc:narios VYC-1628 Rev. 0 Page 18 By 600 seconds (10 minutes) the operators are expected to have diagnosed the problem and completed

   .)             align:nent of the RHR system to a containment cooling mode (either suppression pool cooling or injection mode cooling). The mode of cooling chosen by the operators is subject to their ability to assess reactor vessel water level. If reactor vessel water level is above the top of the active fuel (TAF) and the vessel level indication can be declared reliable, the operators will initiate suppression pool cooling (OE-3101). If the reactor vessel water level is not recovered above TAF or the vessel level indication cannot be declared reliable, the operators are directed to continue injection into the vessel (OE 3101 and OE-3102). They are also directed to place an RHR heat exchanger on line (OP-2124). This cooling mode is called " injection mode cooling"in this analysis.

At some time after 10 minutes, the operators may reduce core spray pump flow to ensore adequate NPSH is maintained for these pumps. The flow is assumed to be reduced to between 3000 to 4000 gpm. Guidance to throttle ECCS pump flow is given in plant procedure ON-3164 [2.1) and OE-3101 [2.2). Containment and vessel cooling is maintained for the duration of the event. At some time after containment cooling is initiated, the heat addition to the suppression pool will equal the heat removal by the RHR heat exchangers. At this point the peak suppression pool temperature is reached and the pool will begin cooling from that point forward provided the cooling conditions are not altered.

 ,                 Table 2.1 - Typical Long-Term Accident Sequence for LBLOCA3 Event                                      Time (sec)

LOCA or MSLB starts (normal power lost if assumed), torus temperature > 90*F 0 Drywell high pressure, reactor vessel low-low level <1 Begin ECCS injection to core 26 Broken loop empties ofliquid 30 Minimum vesselliquid inventory 40 Intact loop refilled 60 Vesseldepressurized 90 Vessel refilled to 2/3 core height (maximum level given large recirc suction line break) 200 Feedwater addition ends 300 Containment cooling initiated 600 ECCS pumps manually throttled > 600 Peak suppression pool temperature -16000-20000 3 All times are approximate and will vaty dependmg on the specirac scenano. 03F. YYC-16-9M Aprd 23. 8WM l

Torus Temperature and Pressure Response to Large Break LOCA cnd MSLB Accident Scenarios VYC-1628 Rev. 0 Page 19 MSLB The MSLB sequence of events are similar to the LBLOCA sequence of events with the exception that vessel inventory loss and depressudzation occurs at a slower rate. These exceptions are due to no liquid flow through the bmak (except for any carryover that may occur) and a smaller break area than a DEG recire suction line break.. Cdtical flow at the break locations is quickly established after break initiation. The reactor is rendered subcdtical due to the void formation in the core as the system rapidly depressudzes. High drywell pressure initiates a scram in less than one second. The main steam isolation valves (MSIV) are assumed to shut at their fastest rate following a main steam line high flow signal (about 3 seconds). This maximizes the energy left in the system allowing more i energy to be transferred to the suppression pool. The recirculation pumps tdp on the reactor vessel low-low level signal or MG under-frequency if offsite poweris lost. l The reactor vessel is depressurized at a slower rate than a LBLOCA. 'Ihe core standby cooling systems are i actuated by either reactor vessel low-low level or high drywell pressure. The RHR system is aligned to the reactor vessel in LPCI mode and the core spray system is energized. When the pressure permissive setpoint is reached, the LPCI and CS pumps inject water directly to the core. Feedwater flow increases as the vessel pressure decreases. The feedpumps reach runout condition and feed flow is maximized. The temperature of the incoming feedwater decreases as energy is removed from { the feedwater heaters. The feedwater addition is non-mechanistically terminated when all hot feed is injected into the vessel similar to the LBLOCA event. I The vessel is refilled by the injection from the core standby cooling systems and the feedwater addition. The vessel and core are cooled by this injection flow. The injection flow transfers the core and vessel sensible heat and the com decay heat to the suppression pool via break spillage and the drywell and vent system. By 600 seconds (10 minutes) the operators are expected to have diagnosed the problem and completed alignment of the RHR system to a containment cooling mode (either suppression pool cooling or injection mode cooling). The mode of cooling chosen by the operators is subject to their ability to assess reactor vessel water level. Reactor vessel water level is above the TAF due to the break location (main steam line). If the vessel level indication can be declared reliable, the operators will initiate suppression pool cooling. If the vessel level indication cannot be declared reliable, the operators will continue injection into the vessel and commence injection mode cooling. ar.vyc.u w w zum l J

Torus Temperature and Pressure Response to Large Break LOCA cnd MSLB Accident Scenarios VYC 1628 Rev.O Pige 20 Containment and vessel cooling is maintained for the duration of the event.' At some time after iS- ' - containment cooling is initiated, the heat addition to the suppression pool will equal the heat removal by l' the RHR heat exchangers. At this point the peak suppression pool temperature is reached and the pool will begin cooling from that point forward provided the cooling conditions are not altered. l l g i l 3 I ? a . _.,, m u. .. E n ' L: _ _ - _ _ - _ _ _ - _ _ _ _ _ - _ _ _ - - - _ _ . _ - _ - - - _ - - - - - - - - - - - - - - - - - - - - - -

Torus Temper ~ure and Pressure Response ts Largs Break LOCA cnd MSLB Accident Scenarios

              ' VYC-1628 R;v. O Pege 21 2.3      Suppression Pool Temperature and Wetwell Pressure, Phenomenological
h. Perspective The suppression pool temperature following a LBLOCA or MSLB is determined by the mass and energy

{ added and the mass and energy removed from the suppression pool. Therefore, a time-dependent mass and energy balance on the suppression pool is sufficient to calculate the time-dependent temperature response j of the pool. Likewise, a time-dependent mass and energy balance on the wetwell air spar. :s sufficient to calculate the time dependent temperature (and pressure) response of the wetwell but with the added complication of the pressure dependence on the available gas volume for the mass of vapor contained in that volume. This section provides an overview of the relevant phenomena that can impact the suppression pool and wetwell vapor space mass and energy balances. The peak suppression pool temperature occurs when the rate of energy addition to the suppression pool is equal to the rate of energy removal. The magnitude of the suppression pool temperature is determined by integrating all the mass and energy addition and removal rates from the beginning of the transient. There are three main heat addition mechanisms to the suppression pool: initialreactor vessel blowdown feed and bleed from the reactor vessel pump heat from low pressure ECCS pumps s

   )

Heat is introduced to the reactor vessel by three mechanisms: a core decay heat ' extemal hot feedwater addition a heat transfer from hot structural components An additional heat addition to the containment is the energy added to the pumped fluid by the low pressure ECCS pumps. The containment can be considered as a closed system with the energy entering the pump shaft as rotational energy being added to the system. Through frictional and form losses, the mechanical energy will be transferred to thermal energy throughout containment. Heat removal from the vessel occurs as injected ECC water is heated and spills out the break cooling the vessel. Heat is removed from the suppression pool (or containment in the case ofinjection mode cooling) by the RHR heat exchanger. These mechanisms are primarily input driven as opposed to phenomena driven. In other words, the inputs and assumptions chosen for evaluation will primarily determine the resulting peak suppression pool temperature. The integration of the important mechanisms result in the time-dependence of suppression pool temperature. However, these mechanisms are highly inter-related. Two examples of the complicated interactions are provided. t GIF.VYC-lH WP Apr9 22.1998

          . Torus Temperature and Pressure Response to Large Break IDCA rad MSLB Accident Scenarios VYC-1628 Rev. O                                                                                            Pige 22
1. Increased short-term feed flow introduces energy into the system but the break flow is also V increased and the convective heat transfer from hot structures in the vessel is affected due to the changes in liquid distribution in the core and vessel.
2. Increased long-term RHR flow, when in injection cooling mode, increases the break mass flow rate therefore increasing the rate of energy addition to the suppression pool. However, the ireeess RHR flow increases the heat removal rate of the RHR heat exchanger due to the increased efficiency from the higher convective heat transfer coefficients arising from the increased velocity (flow) through the heat exchanger.

The effects of phenomena occurring early in the event can have an effect on the system response later in the event. Operator action and single failure assumptions lead to a clear division between the short-term and the long-term as will be shown in Section 3 and Section 4. 'Ihus, events occurring in the short-term can have an effect on the long-term resuks. Sensitivity analyses are performed to assess the integral effects of the impoitant variables consistent with the above set of equations. This is further detailed in Section 3.3. and Section 6. The phenomena important to wetwell pressure are best illustrated by examining the first two terms of the NPSH equation, Eq.1.1. These term can be recast as: AP = (P,,,,,-P,,,(Tg)) (2.1)

 .0 P,,,, = P,,, + P.,,                                                       (2.2)

AP = P,,,l r, + P,,,(T,,,,) - P,,,(Ty,) (2.3)

                                                        =

T,,,,,, T,,,,, - ATmmx (2.4) where: P. = the partial pressure of the non-condensible gases in the wetwell assuming air temperature is equal to spray temperature P,,,, = the partial pressure of the water vapor in the wetwell air space, at the expected 100% RH, tnis would conespond to P at the vapor space temperature T,,,,, = wetwell spray. temperature ATamx = RHR-side temperature drop across the RHR heat exchanger From the AP component, Eq. 2.3, two major variables are suppression pool temperature, T,,,i and wetwell [ ~air partial pressure, P,. In order to determine a conservative, minimum, NPSH,m, it is seen from Eq. 2.3 that the minimum T,,,,, will yield the lowest non-condensible gas partial pressure as well as the lowest vapor partial pressure. Similarly, the highest pool temperature will result in the highest pool vapor I- temperature and a lower NPSH . Yet T,,,,, and T,,,, are related, as shown in Eq. 2.4, because the wetwell spray is delivered from the suppression pool through the RHR system, including heat exchanger. 1 Therefore, in order to minimize NPSH,m, the coldest spray consistent with the highest pool temperature l car.vve.w w 4 ,e n ms j

                                                                                                                    .______-_________________A

[? ! Torus Temperature and Pressure Response to Large Break LOCA cad MSLB Accident Scenarios VYC-162S Rev.0 Page 23 should be considered. Due to the non-linear nature of the P. vs T. curve and the dependence of ATaunux

 )           'on the RHR heat exchanger performance, it is not immediately clear what service water conditions would lead to the minimum NPSH,,,                  .

Containment initial conditions are also important variables. The highest initial suppression poal temperature will yield the highest peak suppression pool temperature. A maximum peak temperature also results from a minimum water mass for a given energy content, T = d + T,. As for containment

                                                                                                               " 'r pressure, the initial containment volume and initial air mass are key input parameters. During the everit.

the pool temperature is high and the containment pressure is low late in the event. If the drywell and ) . wetwell atmospheres are at the same temperature, spray temperature, and pressure as might be expected during periods of containment spray, the containment pressure is given by (for simplification, the vent system is ignored for this discussion): (m,,,,,,, + m,,g,)RT_ p* , (2.5) (veivniv# + vdrpeell) I where m,,,,,,,, = wetwell air mass m,,,,,,, = drywell air mass R. = gas constant V,,,,,,,, = wetwell air space volume V,,,,,,, = drywell air space volume In order to minimize the containment pressure, the spray temperature is minimized, initial air mass is minimized (since no air is introduced during the event) and the containment volume is maximized. Since the wetwell and the drywell have different initial conditions and maximizing the containment volume results in an increase in air mass, it is not sufficient to maximize each individual volume nor minimize each individual mass. "Ihe mass and volume terms in Eq. 2.5 represent the average air density in containment. Since the air mass does not change during the course of the event", the initial conditions should be chosen to minimize the initial average air density of the system. The key phenomena detailed for suppression pool temperature will determine the modeling techniques and inputs to determine the limiting peak suppression pool temperature. While the objectives of this calculation do not include, calculation of limiting wetwell pressure, the phenomena important to determination of wetwell pressure is provided for possible future use. There is some slight containment leakage bounded by the Technical Specification limit of 0.s weight percent per day at 44 psig. For the purposes of this discussion,this slight leakage is ignosed. I car.vic.is.e w AereIt. toss l 1 i l

Torus Temperature and Pressure Response to Large Break LOCA and MSLB Accident Scenarios VYC-1628 Rev.0 Page 24 2.4 Method Description

 ,)

The containment response methodology consists of two distinct elements identifiable by the computer code each element requires (specific details of the two codes used in this calculation are provided in Section 2.5). The two elements are:

a. LOCA ' mass and energy release calculation using the RELAPSYA BI A code and a plant model derived from the curent NRC-approved LOCA licensing analysis per 10CFR50 Appendix K. The hitC Standard Review Plan (SRP), Section 6.2.1.3 accepts the mass and energy release analysis for postulated LOCA events for use in containment analysis if the analysis complies with the relevant requirements of GDC 50 and 10 CFR 50 Appendix K para. LA. In general, these requirements assure that the approved Appendix K evaluation model is used with appropriately conservative inputs for containment analysis.

This calculation modifies the NRC-approved VY LOCA Appendix K model with inputs chosen to conservatively calculate suppression pool temperature and wetwell pressure and meets many of the requirements detailed in the SRP. I

b. Containment calculation using the GOTHIC 5.0e code. GOTHIC was developed under EPRI and nuclear industry guidance specifically for containment analyses. GOTHIC rs used in this calculation to perform the dynamic mass and energy balance on the containment. It has been validated against a selected matrix of separate effects and integral tests to evaluate the available modeling choices [2.3] Additional

{ l benchmarking is included in this calculation t'or the purpose of validating GOTHIC 5.0e and the developed containment model for use in calculating long-term containment response for NPSH and peak suppression pool temperature evaluation. Details of the benchmarking is provided in Section 2.6. To implement this methodology, the calculation has specific steps leading to completio t:

1. identify potential limiting LBLOCA and MSLB scenarios based on a detailed examination of the potential single failures and availability of offsite power,
2. develop a base analysis model for analyzing the identified scenarios,
3. benchmark the analysis models to assure adequacy of both the models and the method, i
4. analyze a base case scenario using the tested models and method, j

i

5. analyze other scenarios and sensitivities to assist in understanding of the phenomena and scenario l

l variabilities.

                        'Ihe scenario development process includes examming previous design basis calculations for post-IDCA torus temperature and ECCS single failure criteria analyses as well as evaluation of balance of plant (BOP) systems, specifically feedwater and condensate. After a LBLOCA or MSLB the operators, based on CIIF.YVC-lH WP                                                                                                                                                  AprE II. I988
                                                                                                                                                                                 )

Torus Temperature and Pressure Response to Large Break LOCA and MSLB Accident Scenarios VYC 1628 Rev. O Page 25 procedural guidance and equipment availability, may use feedwater in order to assist in mitigation and

 -3' recovery from the accident. Altemately, the feedwater system may continue to inject relatively hot fluid into the vessel if offsite power is not lost and no action is taken to prevent its addition. The feedwater system has the potential to add significant amounts of mass and energy to the primary containment. While beneficial from the perspective of ensuring the safety of the core, this can contribute significantly to the post. accident heatup of the suppression pool. The methodology developed and used in this calculation assumes conservative feedwater injection from the perspective of maximizing suppression pool temperature.

De mass and energy release from the reactor vessel to the primary containment will be mechanistically calculated using a detailed RELAP5YA model of the reactor vessel with a coupled feedwater system model. This detailed model will calculate the mass and energy release until the reactor pressure vessel (RPV) and ECCS conditions stabilize to quasi-steady state and the RPV and drywell pressures equalize and the core power output is essentially decay heat (about 80 seconds). This mass and energy release is input into a GOTHIC model of the Vermont Yankee primary containment. This containment model calculates the containment response to the mass and energy release to the quasi-steady state time. After the quasi-steady state condition is reached, a less detailed vessel model is required due to the quasi. steady state nature of the RPV. The mass and energy release will be calculated using a simple vessel model with appropriate decay heat and passive metal structure heat transfer with the GOTHIC code (Section 2A.2 contains more detailed information on the simple vessel model and its justification). The containment response to the mass and energy release is based on simple mass and energy balances on both the drywell and wetwell. The GOTHIC code is used to calculate this mass and energy balance. The GOTHIC code has the capability to model heat removal from the suppression pool using a dynamic heat exchanger model as well as modeling the RHR and Core Spray system interaction with the reactor vessel. I 1

          . Large breaks and MSLBs with various failures of mitigating equipment and with and without off-site AC power available are considered to determine the limiting event and conditions. Input paiaw.sca will be at their conservative limiting value within the operating range, including uncertainties where applicable.

Several types of single failures are evaluated - failures that produce degraded containment heat removal (e.g. loss of an RHR heat exchanger) and failures that degrade ECC injection capability affecting the rate at which heat is transferred from the vessel to the suppression pool (e.g. loss of a core spray pump). The rate of mass and energy addition to the vessel from the feed and condensate system is also considered. Based on the results in VYC-1290 Rev. 2, the initial analysis is a double ended guillotine break at the location of the recirculation pump suction with off-site power available. This break location results in significant mass and energy being transferred to the containment from the vessel. De availability of off-site power allows the operators to continue hot feedwater injection into the vessel in an effort to maintain the core cooled. The hot feedwater contdbutes to the temperature increase of the suppression pool because the temperature of the incoming feedwater is higher than the temperature of the suppression pool. A er.vve.mr ace n.ms

Torus Temperature and Pressure Rtsponse to Larg Break LOCA and MSLB Accident Scenarios VYC 1628 Rev.O Page 30 2.5 Code Description

   -)?'
               ' Tyco codes are used in this analysis - RELAPSYA-B1 A and GOTHIC 5.0e. RELAPSYA is used to calculate the short-term core and reactor vessel response and mass and energy release resulting from the i

postulated large break. GOTHIC is used to calculate the long-tenn break mass and energy release and the containment response for the entire event. RELAP5YA provides a consistent, integral analysis capability of the system and core response to LOCA events and otherplant transients. Extensive assessments of RELAP5YA compared to many separate effect and integral test results have been performed. The Vermont Yankee model used in Section 5 of this calculation is based on an approved 10CFR50 Appendix K ECCS evaluation model. The RELAPSYA ' assessments and the licensed model establish the viability of the RELAP5YA code and model to predict 3 complex thermal-hydraulic phenomena such as those encountered in the reactor system analysis of the { LOCA events contained in this calculation. [2.11][2.12] GOTHIC is a general purpose thermal-hydraulic code for design, licensing and operating analysis of nuclear power plant containments and other confinement buildings. [2.13] GOTHIC solves the conservation equations for mass, momentum and energy for multi-component, multi-phase flow. The phase balance equations are coupled by mechanistic models for interface mass, energy and momentum transfer. The interface models allow for the possibility of thermal non-equilibrium between phases and unequalphase velocities.

               - A simplified set of conservation equations are used in this calculation. All volumes in the GOTHIC l                models are lumped parameter (as opposed to subdivided, multi-dimensional) volumes (see Section 6 and Appendix C). As such, the mass and energy balances are maintained among the volumes and the

, interconnecting junctions pass the flow from one volume to the next. As the fluid flow in the GOTHIC models are primarily (post-blowdown) gravity-dominated flows and the piping systems for the connecting systems are not modeled explicitly, the momentum balance across the junctions is of secondary importance. The GOTHIC models serve to solve the simple energy balance between the drywell, wetwell, vent system and reactor vessel and the liquid and vapor phases in each location. The code developers, Numerical Applications Inc. (NAI), have performed various assessments of GOTHIC compared to a variety of separate effect and integral test results. [2.3] Specific benchmarks of the models used in this l calculation have been performed. A discussion o' those benchmarks follows in Section 2.6. 'Ihe GOTHIC assessments and model benchmarks establish the appropriateness of the GOTHIC code and models to predict the containment response to the LOCA events contained in this calculation. o OF.YTC-16-9.WP Apr$ II.1998 I r E-- _ _ - - - - -

Torus Temperature and Pressure Response to Large Break LOCA and MSLB Accident Scenarios VYC 1628 Rev. O Pcge 31 2.6 Method and Model Benchmarking A set of benchmarks have been perfonned specifically to assess the adequacy of the methods and models used in this calculation. Dese benchmarks include: 1. Separate effects benchmarks for the RHR heat exchanger and RELAP feedwater model are performed. He assessment of the RHR heat exchanger model against the plant surveillance

                   ._ criteria is included in this calculation (Appendix E). The assessment of the feedwater model against plant trip data from April 1997 is contained in calculation VYC-1660 [2.5].

Dese benchmarks provide the basis forjudging the adequacy of the RHR heat exchanger model and the feedwater model. De benchmarks show that both of these models are conservative and account for relevant uncertainties and are, therefore adequate. 2. Comparison of the integrated method against the DBA analysis in the VY FSAR (Appendix D). De case corresponds to a LBLOCA without offsite power ava!!able. This benchmark provides a total method and model comparison against a previously NRC-approved analysis. The methods and models employed in the original FSAR analysis are substantially different from those employed in this calculation. His results in a benchmark that does not,in and ofitself, show the adequacy of the models and methods employed in this calculation but the comparison to a previously NRC-approved analysis is required by the NRC. De separate effects benchmarks as well as the mass and energy balances performed during the analysis provide the necessary assurance that the long-term containment response is being adequately calculated for the given inputs. It has been shown that the long-term containment response to a IDCA is input driven (Section 2.3). Therefore, the methods and models documented in this calculation are considered adequate to calculate the long-tenn' containment response to a large break LOCA. Because the long term containment response to a MSLB is similar to that of a large break LOCA, the methods and models are considered adequate for calculating the long-term containment response to a MSLB. 8

                           *1ong-te.m" as used here refers to > 30 seconds post-accident afar the guessure effects of vent clearing have concluded.

CDP.YYC.M e.WP A,,8 22.1958

                                                                                                                                        , ~ ~ ~ ~   )

Tonis Temperature and Pressure Response to IArg2 Break LOCA cnd MSLB Accident Scenarios VYC 1628 R:,v.0 P ge 32 2.7 Model Description In order to model the events just described as well as the systems that participate in those events, several models are developed for this analysis: a reactor vessel, core and feedwater model using RELAP5YA and a primary containment and associated system model using GOTHIC. j l The RELAPSYA core and vessel model nodalization is shown in Figure 2.2. The development and further justification of the modelis described in Section 5. Based on the assessments performed to support

                                                                                                                               )

approval of this model by the NRC for LOCA analysis [2.11], the RELAP5YA core and vessel model are I judged to adequately predict the phenomena resulting from a LBLOCA and MSLB for containment analysis. The RELAPSYA feed train modelis shown in Figure 2.3. Its development is described in detail in , VYC-1660 Rev. 0 [2.5]. The interface of the feedwater model with the core and vessel model is described in Section 5. Based on the assessment contained in VYC-1660, the feedwater model provides a bounding high estimate of the heat transferred to the reactor vezel after a LBLOCA or MSLB. The GOTHIC containment modelis shown in Figure 2.4. Additionally shown is the relationship between the GOTHIC containment model and the VY primary containment. He model development is detailed in Appendix C. In order to capture the phenomena necessary to assess long-term suppression pool t temperature and torus pressure, the drywell, vent and torus are modeled separately. The fluid systems that can significantly affect the rate of energy addition, rate of energy removal or rate of energy transfer between the drywe!!, wetwell and reactor vessel are modeled as needed. Figure 2.5 shows a sample GOTHIC containment model with the system models developed in Section 6. He system models are developed assuming the system flows are quasi-steady state, i.e. wherm the system flows are pump driven, a constant pump flow rate is assumed since the differential pressure acmss the pump is relatively steady after the initial vessel blowdown is complete very early in the accident sequence. Based on the specific scenario assumptions, the system flow rates are known apdon. Therefore, complicated piping models are not required and the simple system models used adequately model a particular system's response during a LBLOCA or MSLB. l

      . . vve.. ..,

w u. . L_-__________ _ _ _ - - _ _ _ _ _

                                                                                                                                                                                                                                                                                 -- ]

l Torus Temperature and Pressure Response to Imge Break LOCA and MSLB Accident Scenarios VYC-1628 Rev. O P:g2 33 Soe

                                                                                                                                                                                                                                    +4d f G2 l 8(PCs
                                                                                                                                                                                                                                    $59
                                                                                                                                                                                                                                    +4d G59 l ADS
                                                                                                                          'WMott1 Atty crost, 240                                                   *SS
                                                                                                                          %Nof/14t# o#(y                                                                                                   -+l 655 l SRv 553
                                                                                                                                                                                 +

gag ->4d 653 l SRV l 234 l 250 / 551

                                                                                                                                                                                 +                             542                 HH 651 l SRv I          232              i ese ei,                                     544 256 226                         254                                 .

260 220 262 3 t + L E l 264 l 210 l 264 l qY

                                                                                                                                                             +                                   +                                                    M9 N 20e                  \\                     n                   3 I 550 f*
  • 1_

26e + 2ss a 3 TURB o l 72o M 722  ; f ff + h 548 wcs i 28 l . . . . I26eI j 547 ass

                                                                                                                                                           +         &       &&&                  +           r l270               ,                    270l                r,                            N
'+ w J 274 8 y l274l h
                                                                                                                                              ~                  Z I'                                        ?         2

[ 5

                                                                                                                                                                            ' ~            E l 400 l l 700 l t
                                                                                                                                                                ?             c12l ? l27sl                                    FD M N
                                                                                                                                                     278 V                          %
                                                                                                                                                                                                                                                                                      )
                                                                                                                                                                                 +                                                                      Ncv /ce/-                     )

28o 9 lclol kafo //<afr<

                                                                                                                                      .        302             V y               +       V                352           _                               m e/e/
                                                                                                                                      $                             s L2=_I $ Ias($2*o                                                  n 2         '\ me l        i l                we
                                                                                                                                           "                                                          /

_c-+1 33o-i }J + + T-1 3eo.i h 5 E

                                                                                                                                                          %d +* '                                              ;

5 l h 742 qd' j y22 M a

                                                                   "                                                      l 7e i                                                                                             I vm I                   .
                                                                                                                           &c4               g        y                                                                    tecte                  g v          f E                                                                        ^ 802                                                                                                      4-a                                                                                  614

{ -

                                                                                                                                                                                                               ~                                      -
                                                                                                                                                                                                                                                                                      ]

5 b l 8 cic -

                                                                          % 112                                                10                                                                                                          362-313 308                                                                                    358 GROKEf4 LOOP (A)                                                                                                      If 4 TACT LOOP (B)

Figure 2.2 RELAPSYA Core and Vessel Model Nodalizatien cor.vve is-, we

                                                                                                                                                                                                                                                        *' m "'

( . . . , .

m

                       -                                          u            3 P             8 2                          d               -

2 e 2 4 5 e 4 F

                                        -    1
                                        - 2            ee                      2 cL                      8 1

d4 uH o I 2 2 seB 5 5 se/ 4 eLA 2 8 8 P eI-2 2 2 5 5 4 hd2 ioe-eE 1 8 HFI -

                                                     ' -                       2 9                                 5 n            1                                 4 o

i t C4 c e

                                                                               >35           c r                                                      4   eL
                   .           t      8                                                   rH n            x     1 5                  u     I i

E 4 5 seB - b 9 se/ r n eLA 6 re v5 54 u e 4 T e - 5 5 P.I - P S t - 7

                                        -    1         .c eL 5          4
                                                                                          .d3 oeE e-F 9           7 Q4               e vH   I LFI 7'

1 5 1 5 seB ee/ n

                                                                               >55 1                                    rLA           o                4 6           8     6 6              P eI-V v1 5     1 5

1 4 hdl i t c D oe- a T i eE r HFI t 6 x 5 v., -- 5 E 4 1 Q4 m e e l G D75 c 4 eL 4 rH 1 9 u I 4 5 seB 9 se/ elA 8 8 n

                                          - 3 6

5 v5 85 5 4 reI Pw-d4 i t o

                                          -   1                                           we-         a oeE z Q4                                                 LFI        i l

a D9 d

                                                                                  - 54                o 2

1

                                                                                  -                 N l

4 e d 8 o 6 4 M i n a r D1 T 1 c d 1

                                                                                   -- 64  eL           e Fe 4

rHI _ 3 u 6 eeB ne/ A 4 5 _ l 6 elA Y d e 5 V v6 29 2 6 reI Pw- S P o 4

                                                                                           .d5 A

_ e M D e-1 8 i nn T oeE L LFI t i !. E 2 8 se. i p

                                                               .               D36                   R 4
                                                   ,   xi.                             4 i    . EPV e

6 e 6 s 3 4 n 2 e e V nd r D in u T oo g nC i F - 9 [g g" i

l Torus Temperature and Pressure Response to Large Break LOCA and MSLB Accident Seccarios VYC-1628 Rev. O P;ge 35

                                                                                                                                                                }
                                                                                                                          >3L                                                 '

3 g 1 - q 3

                                                                                                                                                                                         +-

e g u <. < :.;;#, p,sy .. af bb b

                                                                                                                                                    -             ivii ,0;p, 3
                                                                                       ~
   .                                                                                   H I

il[

                                                                                                                                                                                ..i              /3
                                                                                                                                                                           /                       w 3

1 z b 1 x ET 6 g li ~

                                                                                                                                                                                      -              c is c

s a 2- 0 "I W I 5 8

                                                                                         !!!lllllllllll[va i           e                 e  e    e              c                                :                          -          -

e I EI  !

                                                  "     -       y    a                                   E                                E                          E                  g      g    go
   ' Oll'* YYC BM.wr Apr8 28.1995 l

I _ - _ --- _ - - - 1

Torus Temperature and Pressure Response to Large Break LOCA and MSLB Accident Scenarios

 )          VYC-1628 Rev.0 Page 36 J

runt, Torus Temperature Response (BLOCA Apr/2of9810:39:58 GOTHIC Version 5,0(QA)-e - October 1996 LONGTERM - Long Term Containment Response to LB LOCA g _Drywell_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ ,

                                                                    ;                                         i Break M&E       F       10 i                                          I

(< 80 sec) I F  ! l Feed Flow 1 (> 80 sec) I 1 1._ _ , l I I I Ves l l gli1NN1lI H l , ,1y l l l

 -n                            Core Spray (> 80 sec)               1 l-- _@_ j Break (> 80,Isec) i
                               & LPCIInjection (80-600 see)        i                                         I I                                         I I                                         I 4U"                          !

l s l se l l l l l I i I __a g______, i 2 0 ,t_ 1 Vent I i I I I Wetwell i I H I l " 1H

                                              -@           l F -s-(l._______.           ___a
                                                   !)

a RHR in Suppression TF-ECCS Suction _ g~~ Pool Cooling (< 80 see) (> 600 sec) Figure 2.5 GOTHIC Containment and System Base Model Nodalization l me.we iaw w n. im

Torus Temperature and Pressure Response to large Break LOCA and MSLB Accident Scenarios

               . VYCal628 Rev. O                                                                                        Page 37 2.8       Calculation Organization and Presentation 7
   ~

1his short section pmvides a "roadmap" to the calculation which assist in understanding how the different pieces fit together. The first sections of the calculation, which have already been presented, provide the background information and problem description as well as a description of the analysis method and philosophy. Section 3 provides a detailed explanation of the scenarios that were chosen for analysis as well as the justification for choosing those scenarios. The major uncertainties in the analysis are also identified. Section 4 provides a list and justification of the major assumptions and key initial conditions and plant p-.iime. Not all analysis input parameters are listed in this section,just selected input parameters. Section 5 provides the documentation for all the mass and energy release calculations performed with RELAP5YA. Section 6 provides the documentation and results for all the containment responses determined using GOTEUC. Section 7 provides a summary of the key results and conclusions of the calculation.

 '.?)

The appendices provide supporting documentation and are referred to in the body of the calculation where appropriate. I (Iir. Wyc.144.wr Apr5 21.1998 (

Torus Temperature and Pressure Response to Large Break LOCA and MSLB Accident Sc:narios

                                     . VYC-1628 Rev. 0                                                                               Peg 2 38 2.9     References 2.1     VY Procedure ON 3164 Rev.1. ECCS Suction Strainer Plugging 2.2     Emergency Operating Procedure, OE-3101 Rev.12. RPV Control Procedure 2.3     GOTHIC Containment Analysis Package, Version 5.0e, Qualification Report, NA18907 09 Rev. 3 2.4     Calculation VYC-1290 Rev. 2. Vermont Yankee Post-LOCA Torus Temperature and RHR Heat Exchanger Evaluation, approved 8 Augnst 1996 2.5     Calculation VYC-1660 Rev. O, Vermont Yankee Feedwater Heater Train RELAPSYA Model Development, approved 17 October 1997 2.6 -   Vermont Yankee Final Safety Analysis Report (FSAR), Revision 13 2.7     Calculation VYC-1290 Rev. 0, Vermont Yankee Post-lhCA Torus Temperature and RHR Heat Exchanger Evaluation, approved 1 August 1994 2.8     Emergency Operating Procedure, OE-3102 Rev.11. Alternate level Control Procedure 2.9     Calculation MYC-1784 Rev. 0, CONTEMPT LT-028 Mass & Energy Addition Input Generation, approved 13 November 1995 2.10     Calculation MYC-1819 Rev. 0, Maine Yankee Containment Pump NPSH Calculation, approved 29 December 1995, A .

2.11 R. T. Fernandez, et. al., "RELAPSYA - A Computer Program for Light-Water Reactor System Thermal Hydraulic Analysis," YAEC-1300P, October 1982, Revised June 1993 2.12 Calculation VYC-1421 Rev. O, VY Cycle 19 LOCA Verification, approved 23 August 1996 2.13 Calculation YC-351 Rev. I A, GOTHIC Containment Analysis Package, approved 18 December 1997 I l aw.muw mu.sm c _ _ _ >

f ( ) Torus Temperature and Pressure Response to Large Break LOCA end MSLB Accident Scenarios l VYC-1628 Rev. 0 Page 39 3.0 Evaluation Cases, Inputs and Assumptions, Scenario Description l l >; The calculation's objectives have in common the need to determine the suppression pool te:nperature and l torus pressure for a variety of LBLOCA scenarios. As described in Section 2, the results can be used to assess the adequacy of ECCS pump NPSH. Based on the NPSH equation, Eq.1.1, the highest suppression pool temperature (given a torus pressure and suppression pool surface elevation) results in the minimum i NPSH available. The scenarios examined in this calculation were chosen to conservatively bound the l maximum potential suppression pool temperature after an LBLOCA. They have not been selected to necessarily minimize or maximize wetwell pressure. I Two matrices of potential scenarios were developed. The first matrix was developed based on capturing the peak post-accident suppression pool temperature. The second matrix was developed from the output of an independent and parallel effon undenaken during the ECCS suction strainer replacement project. The two matrices provide a diverse set of potential accident scenarios for examination. Using these matrices and sensitivity evaluations (see Section 3.1) the scenarios which can result in the highest peak suppression pool temperature can be determined. From these case matrices, the input models for RELAPSYA-B1 A and GOTHIC 5.0e were developed and were used to calculate the suppression pool temperature and wetwell pressure response. The analysis oflong-tenn suppression pool temperature and torus pressure is an energy balance through

  ~ ];           time between the reactor vessel, the drywell and the wetwell consistent with the event descriptions provided in Section 2.2 and the specific case scenarios that will be outlined in this section. The input parameters and assumptions which can significantly affect the suppression pool temperature and wetwell pressure are discussed first in order to assist in providing a basis for determining potentially significant single failures and scenarios to be analyzed.

3.1 Major Sensitivity Pararneters Many parameters can affect suppression pool temperature and wetwell pressure. Section 2 summarized the analytical details of suppression pool temperature and wetwell pressure and that discussion is used as a basis for qualitatively assessing the potential effects of the significant parameters. One result from the analytical discussion was the observation that a higher peak suppression pool temperature generally results , in a lower NPSHm. Assumptions and input conditions, which are conservative with respect to peak suppression pool temperature, are used in the input models funher described in Sections 5 and 6. In addition, a set of sensitivity parameters has been identified to assess the effect on the mass and energy release rates, heat removal from the containment and the long-term mass and energy distribution within the primary containment. These sensitivity parameters provide part of the basis for the development of the l case matrix. l i CDP. YYC-4&4 W Apr0 21 em

Torus TGmpercture and Pressure Response to Large Breck LOCA and MSLB Accident Scencrios VYC 1628 Rev. O Page 40 3.1.1 Offsite Power J

                . The availability of offsite power impacts the mass and energy release. With offsite power available the reactor feed pumps remain powered allowing the operators to continue injection of feedwater during the LBLOCA or MSLB. Additionally, the availability of offsite power may affect the availability of mitigating equipment after an event.

Feedwater addition may be either conservative or non-conservative with respect to peak toms temperature depending on the temperature of the incoming feedwater.. Any feedwater addition at a temperature above the peak suppression pool temperature would be expected to increase peak suppression pool temperature. Conversely, any feedwater addition at a temperature below the peak suppression pool temperature would be expected to decrease peak suppression pool temperature. Based on an expected peak suppression pool temperature of approximately 175'F, addition of feedwater above this temperature increases the peak temperature. Addition of feedwater below this temperature reduces the peak temperature. Only feedwater addition above 175'F will be included.'

               ' Modeling of feedwater addition will be done in two ways. The first is to introduce feedwater mass and energy with similar assumptions used previously. [3.1] This is referred to as "non-mechanistic feedwater".

We second approach is to introduce feedwater mass and energy via a mechanistically coupled feedwater I model. This '  ?.ned to as " mechanistic feedwater". 3 Availability of mitigating equipment will be determined based on the availability of offsite power and the single failure assumptions made for a given scenario. The availability of mitigating equipment can have a significant impact on the results due to the potential for varying the heat removal from the reactor vessel and subsequent transfer to the wetwell. 3.1.2 Single Active Failure - l The suppression pool temperature response is a function of the initial pool temperature and the subsequent rate of heat addition from flow through the break, heat addition from the by operating ECCS pumps and heat removal via the RHR heat exchangers. Single active failure assumptions can change the number of l operating RHR or CS pumps affecting the pump heat rate and rate of ECCS injection into the vessel. The ! rate of ECCS injection controls the rate of energy removal from the vessel and, therefore, energy addition  ; to the suppression pool. Single active failure assumptions can change the number of RHR heat exchangers  ! in service or the flow through the heat exchangers and, therefore, the energy removal from the suppression

                                                                                                                                                                                                                          )

pool. Derefore, the results of the single active failure analysis developed in VYC-1511 [3.2) is used in l assessing potentially limiting failures.  ! While the peak torus temperature will vary depending on the specific scenano, the 175'F feedwater cutoff mill be adequate for this calculation as the effects of any minor variation in the feedwater added will have no significant impact on the results. ' See Section 6.13 for numericaljustirmation of the significance of feedwater cutoff temperature variations around 175*F. G F.VVe-864 WP ' AM82kIWU I J j

Torus Temperature and Pressute Response to Largs Bre:k LOCA and MSLB Accident Scenarios VYC-1628 Rev.0 Page 41 3.1.3 - Post-Accident Containenent Cooling Mode t- ' There are three significant potential RHR alignments allowed by procedum that car. be used post accident to cool the containment. 'Ihe first two are torus cooling and containment spray' whereby at least one loop of the RHR syistem is realigned, after adequate core cooling is achieved, to take a suction on the suppression pool and discharge through the RHR heat exchanger back to the suppression pool or to the drywell and/or wetwell spray headers.' The other method (called injection mode cooling in this analysis) recognizes that OP-2124 [3.3] instructs the operators to direct RHRSW flow through the RHR heat exchanger as soon as practicable after LPClinitiates and OE-3102 [3.4] may require continual vessel injection with all available ECCS. These two different RHR cooling modes will result in different post-IACA suppression pool temperatures due to the differences in the rate of energy removal from the vessel

                                     - and, therefore, energy addition to the suppmssion pool.

l OE-3104 [3.5] calls for the operators to " operate all available torus cooling using only those RHR pumps

                                     . not required for adequate core cooling". In actuality, one definition of " adequate core cooling" is one core spray pump injectir.3 at design flow. [3.6] However, if the operators are basing ajudgement of " adequate                                                     4 core cooling" on vessel level alone and are not convinced that adequate core cooling is being achieved or                                                     i could be continued with redirection of an RHR pump to torus cooling (given that a large enough break has occurn6d that vessel level indication may not be reliable or given that the break is a large suction line                                                      I break and that safety-related systems alone may not be able to restore vessel level above the top of the
  )                                      core), OE-3102 directs them to continue their attempts to flood the vessel. Operators would eventually attempt to initiate containment flood-up if the core cannot be covered. Containment flood-up would add relatively cold water from sources external to primary containment into the primary containment. This cold water addition would most likely arrest any torus heatup and increase NPSH,                                               by both lowering
j. the torus water temperature and increasing the suppression pool static head. However, containment flood-up may not be guaranteed (given a single failure) with safety-related systems. Therefore, this action is not credited in this analysis.
                                       'A core spray line break would result in a special case of requiring injection mode cooling to assure continual makeup to the vessel. As this calculation is only addressing large break LOCAs, a core spray line break is not explicitly within the scope of this analysis and will be addressed in subsequent analyses.

[3.7] > i l For this analysis only torus cooling is evaluated. Injectioa mode cooling will be addressed as a sensitivity during follow-on calculations. Torus cooling is taken as an explicit assumption in Section 4 and reiterated in the conclusions (Section 7). i I Containment spray initiation is considered " torus cooling" for the purposes of this discussien. The difference in the heat transfer phenomena from the reactor vessel to the wetw!! is not sufficiently different from torus cooling to warrant separate discussion in Qis section. at.vyc.66.ewr Apre 22. swo l l"  ; i . _ _ _ _ _ _ _ _ - - - - _ _ . ___ J

Torus Temperature and Pressure R sponse t2 Larg: Break LOCA and MSLB Accident Scenarios VYC-1628 Rev. 0 . Page 42 3.1.4 Containment Spray i Spray headers in both the wetwell and the drywell are supplied from the discharge of the RHR pumps after the flow has exited the RHR heat exchanger. These sprays are manually initiated based on guidance pmvided in OE-3103 [3.8] and implemented in OP 2124. Initiation of containment spray will reduce the wetwell and drywell pressure and temperature due to steam condensation on the cold spray drops and due

              'a the reduction in temperature of the non-condensible gases. The lowest wetwell pressure and air space
              .emperature would be expected based on the coldest spray temperature. The coldest spray temperature is l              expected if the RHR heat exchanger is operating at its highest efficiency with the coldest service water l              temperature, maximum RHRSW flow, minimum RHR flow, and no tube plugging in a clean heat                           i exchanger. Additionally, the coldest pool temperature would result in a lower spray temperature.

Howe;ver, NPSH margin may be increased for a cold suppression pool due to the reduction in water vapor pressure. Therefore the effects of cold spray flow is studied by performing a sensitivity reducing long-term containment pressure (Run 9). 1 3.1.5 RHR Service Water Flow j i

'Ihe RHRSW flow rate impacts the post-LOCA suppression pool temperature and the post-LOCA l containment pressure. A decrease in the RHRSW flow will reduce the suppression pool heat removal rate and, thus, increase the post-LOCA peak suppression pool temperature and vice-versa. Therefore, the j, effects of RHRSW flow and temperature will be studied by performing a sensitivity on RHRSW flow (Run 5). .

i i 3.1.6 Additional Heat Sources and Sinks The RHR and CS pumps transmit energy to the fluid in the fonn of kinetic energy and pressure increase as well as thermal energy from pump inefficiencies. As the single failure assumptions can affect the number ! of pumps running and therefore the pump energy added to the fluid, the energy added by the pumps is modeled and considered in assessing scenarios to be analyzed. The effects of pump heat due to various single failures is accounted for in the set of runs performed. Additionally, the water that is initially in the ECCS piping acts as an additional heat sink. Credit is taken for this additional heat sink for those ECCS trains that continue operating long-term. 3.1.7 Additional Uncertainties Major uncertainties are addressed by analysis to determine their effects on the results. Some specific l uncertainties analyzed are: OlF.VTC.16 9M Apre 22.19M

Torus Temperature and Pressure Response to Large Break LOCA cnd MSLB Accident Scenarios VYC.162,8 Rev. O Page 43 l , closure of the recirculation line discharge valve in the broken loop; a sensitivity is performed [. assuming the valve remains open due to the valve's potential inability to close under a high differential pressure; however, there is no guarantee that the valve will not close. This uncertainty is studied in Run 2. l post-LOCA reactor vessel water level; the base analyses assume that the water inventory within the reactor vessel and core after a LOCA and refill will be based on the jet pump inlet elevation; l however, this assumption is based on all ECC flow exiting the core and vessel with only a small elevation head required to maintain the required discharge flow rate from a DEG suction line break; other break sizes may result in a higher elevation head required to discharge all incoming ECC flow and a resulting larger mass of water in the reactor vessel. This uncertainty is studied in Run 10'. I if offsite power is available and an RRU fails, the accompanying room heatup may result in loss of l pumps operating in the room with the failed RRU. However, it can be postulated that the pumps l do not fail in the time fmme of this analysis (<10 hours run time) due to exceptionally robust I pumps or cool initial conditions in the room. Therefore, a sensitivity is performed to assess the differences if the room cooler does not fail (Run 8). These major uncertainties and parametric variations illustrate the dependence of the results on both ECCS, 4> RHRSW and feedwater system operation. The next section will discuss in more detail how the uncertainty in offsite power and limiting single active failures will lead to the actual cases to be examined. i

                                                                                                                                              )

i l l l l l l l i i ar.vyc.i a wr w 22.

  • l 1 j

Torus Temperature and Pressure Response to Large Break LOCA and MSLB Accident Scenarios VYC 1628 Rev.0 Page 44 3.2 - Limiting ECCS Operation Cases he purpose of this section is to identify a set oflimiting ECCS cases for this analysis. Two sets of cases with varying system operations are developed for two of the three calculation objectives related to determining the long-term containment response. The two objectives are:

1. Analyze the suppression pool temperature and wetwell pressure response to a variety of LBLOCA and MSIE scenarios for use in calculating the limiting ECCS pump NPSH;
2. -

Determine the peak suppression pool temperature following specific LBLOCA events in support of the replacement ECCS suction strainer design basis and to form part of the basis for the technical specification on torus temperature. The two resulting sets of system operation cases are independent of break location or type. In other words, given the offsite power availability and single failure assumptions, the system mode of operation is defined. For each system mode of operation, a large break LOCA or MSLB can be analyzed. The VY LOCA Single Failure Analysis, VYC-1511 [3.2], was examined and single failures consistent with maximiring post-LOCA torus temperature were extracted. The single failures are documented in Table 3.1 along with the effects on the ECCS and the probable effects on the feed system. In support of the replacement ECCS suction strainer bid specification a set of evaluation cases was developed and is contained in Table 3.2. ! It is not expected that all of the cases identified in Tables 3.1 and 3.2 will require a full analysis. As the

            . analysis process is relatively linear, as case results become available and a trend with regard to the sensitivity of the results to system configuration is identified, it will become apparent that some cases are
non-limiting with respect to peak suppression pool te;nperature. Those cases identified as non-limiting do not require full analysis.

De criteria used to choose the cases identified in Tables 3.1 and 3.2 are based on a variety of single failures resulting in different ECCS and feedwater system operation combinations. De base scenario for evaluation is the large break LOCA, a double. ended guillotine rupture of the recirculation piping at the l pump suction. (A sensitivity case will be analyzed based on a MSLB inside the pdmary containment.) Loss of offsite power is not assumed apriori but is considered as a sensitivity. In all cases examined HPCI and RCIC are assumed to not be available either due to possible single failure or due to the rapid depressurization of the reactor vessel resulting in no turbine drive steam. A single active failure of a component is assumed with subsequent equipment availability consistent with the ECCS single failure analysis cited previously [3.2]. A preliminary feedwater and condensate system failure modes and effects analysis (FMEA) was performed to support determining the potential effects of the ECCS single failures on the feed and condensate system e n . w e. u w wn=

Tonas Temperstme and Pscssure Response to Large Break LOCA and MSLB Accident Scenarios VYC-1628 Rev.0 Page 45 (included in Appendix J, unverified and for information only). The preliminary FMEA identified that these could be a wide range of post-accident feed and condensate system operation modes resulting in a wide range of potential post-accident feedwater flow rates. Therefore, post-accident feed flow rate will be examined as a sensitivity to identify bounding feed flow conditions. Operator action is credited to initiate containment cooling by 10 minutes. Other operator action is assumed based on the requirements of a specific sensitivity analysis. For instance, the operators may need to throttle the core spray and/or RHR pumps at some point in the accident to ensure adequate NPSH is maintained. For this particular sensitivity case a specific operator action, to throttle the core spray and/or RHR pumps at a certain time, will be assumed. t 4 dar vic.aseer Apro 22. swa I

I l . o,g ;; 9 QEE *o,Oy p3)N U $E5QgQgCC>hRpSE8

                                                                                                                  -                          i                             .

<

os pol poC- as c t t P - - - - - - - - I I 2I 22 1 1 22 ( pa l mfDb o s s r io n TT SL TT SL TT SL TT SL e n TT TT TT TT TT TT rmmWcii eu b e oio ct nw s s i a r SL SL SL SL SL SL T pS fRoa ume l S t g=ikeid i a r luh t m nCRa eP sv a o oe c s + S I Sthi eeie t u S i C C C = hbi actos C M T ewt ypnl n t k ) B M E ) L, doufo t r a x e o y m (eni d si a I B I e p r - r o ) mb P I o I 2 p I I I I I I I I sI el pe oc - Ao - o CI C CC dCd nmr I I I P C CC CC P 3 I C PP CC PP CC CC C nP oo ul u c de 1 RL L P PP PP e P PP PP P o L m e pi c a & e L LL LL LL LS L r s L LS l f 3 He LL s s b L s s e I e le d h _ pps s pps nCh idUeet s s a pp s s s s s r le Ri c pp pp pp pp pp r i T pp pp b a e mm uu pp mm uu mm uu mm uu c e mm mm mm mm mm mm uu pp uu pp 0 gEd 0 6i nI eeRe Rf o f - T R ( pp pp pp R pp uu pp uu pp uu pp uu < ait ni i f h t, h h s t is ( ( i t n a 22 1 1 I O 2O O 22 22 gnri 22 22 20 mvid e ewibne 1 - - - - - - - - - - - - - - n v - - - - - - r TT TT TT TT TT TT TT TT TT TT epa r i ess oh t Tme isoaa wei t SL SL SL SL SL SL SL SL SL SL t s _ t r opo u pI . t f I .ercmay _ e h s s s ol p S = iigof n m p m I p yi s nh n t i o e e r X H a P) B u =Cml a lor TP ede lgul m SLpa uno cFT u oh s / F O A u P _ H ni 7 ia 2 e O e I P R 7 _ SF R 1 - U lg 2 S S H U H R C D C-D R R b S CL D (w N / n o P U C R R 1 I R ) ) ) ) 1( 2( 34 ( ( _ e e s s a s _ a 3 6 8 e C i 2 3 4 C 0 1 / 2 4 / 5 / 7 to N 0 12Iw* {Wi Torus Temperature and Presswe Response to Large Break IACA and MSLB Accident Scenarios VYC 1628 Rev. 0 - Page 47 3.3 . Individual Analysis Case Description .a - This section pmvides a description of the individual cases that are analyzed and in the sequence they are analyzed. Specific documentation is provided for operator actions consistent with the design basis assumption of no caedit for operator action prior to 10 minutes and actions consistent with the emergency  ; operating procedures (EOPs). Several assumptions are listed here to aid in understanding of the following run descriptions; Section 4 contains additional inputs and assumptions. The following assumptions are consistent for all cases: HPCI and RCIC are unavailable (due to rapid vessel depressurization there is insufficient time for these systems to have any significant impact on the results) Broken loop recire discharge line valve may or may not close (closure of the valve in the broken loop under the differential pressure imposed during the initial stages of the accident is not assumed . in the LOCA analysis; a sensitivity is performed to assess the impact) RHR is initially aligned normally (i.e. not aligned for suppression pool cooling) Operators will initiate containment cooling in toms cooling Feedwater is injected at a constant (maximum) flow rate regardless of the effects of the single failure (for any single failure at least one condensate pump is available to continue hot feed injection) The primary reference for ECCS availability given certain single failures is VYC-1511 Rev. O, Vermont Yankee ECCS Single Failure Evaluation (for LOCA). The set of cases as well as sensitivities analyzed in this calculation and in the order of analysis is given in Table 3.3 (more detailed discussions of the specific inputs for each case are provided in subsequent sections). - The base case presented here (identific i as Run 0) is a case that was developut in conjunction with a Containment Task Force assembled by Vermont Yankee management. [3.9] The base case represents a starting point for VY management to make comparisons with the current FSAR long-term containment analysis as well as comparisons with methodologies used by the rest of the industry. It is not intended to be the limiting or bounding case. The values for various parameters identified in the following case descriptions are based on the inputs and assumptions detailed in Section 4. No other references are provided in the following sections for the parametric values. The reason for putting values in the case descriptions is for ease of analysis for the multiple preparers. i e an.we m me w n. im . . . . . - . _ _ _ _ _- - _ _ _ _ _ _ _ _ _ - _ _ _ _ - - _ - -A Torus Temperature and Pressure Response to Large Break LOCA r.nd MSLB Accident ScInarios VYC-1628 Rev.0 Page 48 .,, Table 3.3 -Index of Analysis Cases i Run Title RELAP BriefDescription' Req.?' O VY-Specified Base LBLOCA with DC-1 failure, normal power available ECCS Yes Case flows at nominal values, non-mechanistic feed, Ref. 3.9. I DC-1 Failure Table 3.1 Case 2, LBLOCA with DC-1 failure and normal power Yes available, ECCS flows at minimum values, broken loop recire discharge line valve shuts 2 Recire Discharge Line Run 1 but the broken loop recire discharge line valve remains Yes Valve Sensitivity open 3 RHR Heat Exchanger Table 3.1, Case 1. LBLOCA with failure of MOV-89A and Yes Failure normal power available, ECCS flows at minimum values; recire discharge line valve shuts 4 ECCS Flow Sensitivity Run 2 but the ECCS pumps are running at maximum expected Yes flow when injecting into the RPV 5 RHRSW Flow Run 4 but the RHRSW flowis at 2850 gpm No Sensitivity 6 Short-Term Toms Run 3 but with inputs selected to maximize the short-term (-10 Yes Temperature Sensitivity minutes) suppression pool temperature 7 Feed Rate Sensitivity Run 3 but with feed rate increased Yes 8 RHR pump failure Run 4 but the operating RHR pump in the room with the No { sensitivity inoperable RRU does not fait due to room heat up j 9 Minimum Overpressure Run 4 but with inputs chosen to minimize overpressure while No Sensitivity maximizing torus temperature, containment sprays modeled 10 Main Steam Line Break Run 4 with recirc discharge valves shutting normally but MSLB Yes instead of LBLOCA 11 FSAR Benchmark Details provided in Appendix D Yes I 8 i Unless otherwise stated: j + the feedwater addition is assumed to be mechanisticaDy determined f e the socirc discharge valves in both loops is assumed to close normally + torus cooling is assumed to be initisied at 10 minutes l

  • core sprey pumps are assumed to be throttled to minimum design flow at 10 minutes i Additional information is contained in the detailed case descriptions.

This column identifies if a RELAP5YA run is required or if the run uses previous RELAP5YA results, cor.vve.sawr Aem 22. twa . Torus Temperature and Pressure Response to Large Break LOCA tad MSLB Accident Scenarios VYC 1628 Rev. O Peg 2 49 - 3.3.1 . Run 0. Base Case Specified By Verrnont Yankee A' This case description is taken from Ref. 3.9. Event Description '!he postulated event to be analyzed is a double-ended guillotine break of the largest recirculation piping, nominally on the suction' side of the recirculation pump. Coincident with this break is a single-active - failure of the DC-1 electrical bus. Offsite poweris assumed to remain available (resulting in allowing feed addition). InitialConditions ' The following initial conditions are used (justification for these initial conditions are provided in Section 4): Reactor Power 1625 M Wth Toms Water Volume 68,000 ft8 Torus WaterTemperature - 90 7 - River Water Temperature 85T - iDrywellPressure 1.7 psig Other inputs and initial conditions consistent with Section 4. + 4 ECCS Performance ' The ECCS performance during this accident is assumed as follows considering the single failure: RHR Pump Operation - one RHR pump operating in each train at runout flow for the first 10 minutes; one pump transitioned to torus cooling at 10 minutes and the other pump is assumed be secured consistent with the limiting FSAR case. CS Pump Operation - one CS pump operating at mnout flow for the first 10 minutes; CS pump is throttled to design flow at 10 minutes - RHR heat exchanger performance calculated assuming 5% tube plugging and design fouling RHR pump flow is assumed to be 7400 gpm at runout and 7000 gpm when in toms cooling CS pump flow is ' assumed to be 4600 gpm at mnout and 3000 gpm at design flow. ' RHRSW flow is assumed to be 2700 gpm Other ECCS performance parameters consistent with Section 4. L . car wc-m.+we ' Apsis. ms Torus Temperature and Pressure Response to Large Break LOCA (nd MSLB Accident Scenarios VYC-1628 Rev. 0 Page 50 Other Assumptions The following additional assumptions are used: Core decay power calculated according to ANS 5.1 - 1979 with 2a uncertainty applied Hot feedwater b added consistent with VYC-1290 Rev. 2 assumptions as to the amount of hot feed (Section 3.3.1.1 provides a more specific description) Volume of water contained in the ECCS piping is credited, Ref. [3.10) 3.3.1.1 Run 0 - Feedwater For VY-Specified Base Case The methodology for determining the amount of feedwater addition in VYC-1290 Rev. 2 was to sum up the mass of feedwater in the feed piping above a reference temperature as well as adding the energy of feedwater system metal components above the reference temperature. The feedwater addition for the VY-specified base case uses the same methodology (Section 3.3.1) but the reference temperature is taken as 175'F (Section 3.1.1). The nodalization of the feedwater addition using the new reference temperature is summarized in Table 3.4 and described in the following paragraphs. De feedwater components upstream of isolation valves 96B and 27B with temperatures above the . reference temperature of 175'F tabulated in Table 1 from VYC-1290 Rev. 2 are listed in Table 3.4. The  ? information on the component temperature, fluid mass, fluid enthalpy, fluid energy and metal mass are also included. Given the components' temperature and metal mass, using the heat capacity for metal components listed in VYC-1290 Rev. 2,0.12 BTU /lbm *F for canton steel, the energy of the metal component above the reference temperature of 175'F can be calculated from: q = me,(T - T,,,). The feedwater c'omponents are combined into nine nodes and assigned to the nodes based on the component temperatures. The mass in each node is the sum of the components comprising that node. 'Ihe enthalpy of each node is the sum of the energy (fluid and metal) energy in each node divided by the mass in that node. The result is the two right hand columns of Table 3.4 providing the mass and enthalpy for the - nine nodes. l l l. The RE.AP model used for the mass and energy release in VYC-1290 Rev. 2 did not include all the heat structures in the feedwaner piping from the RPV to talves 968/278. Thus, these metal components were accounted for in the tabulation in VYC 1290 Rev. 2. However, the RELAP model used in this analysis includes those hat structures. Therefore, they are excluded from the feedwater tabulation. Me MMM Agf( $1, jM _ _ _ - _ _ _ _ _ - _ _ _ _ _ _ _ _ _ _ .-_ ._. b y mmmm b b b b b m mmmm lp A A A A b b b b a n A A A M ht UUUU TTT7 U T UUUU n BB B B TTTT E B BBBB d 2 3 3 3 7 1 0 5 0 n 3 5 6 8 2 8 9 0 8 a 4 4 3 3 3 2 2 2 1 s s M a b l mmmm b l b bl b m b mmmm b b b 1 l l l l l l a 7 4 0 1 8 8 21 4 0 d 09 0 9 6 7 6 4 o 8 4 7 7 6 7 61 2 4 N 4 51 0 4 1 2 1 1 3 1 3 2 3 2 7 5 U 7 B 6 5 6 6 6 5 4 5 5 5 5 5 5 5 6 5 5 7 0 0 0+ 0+ 0+ 0+ 0+0 +0 +00+ + 0 0 0 0 0 0 0 l t n+E+ a r E ee 6 5 EEEEEEEEE 0 1 6 4 9 4 0 7 9 + + + EEE 9 7 0 + + + + EEE E 6 4 t 3 MnE 0 8 1 1 0 2 34 2 7 5 75 50 2 39 3 32 4 2 6 7 0 1 1 1 2 1 1 1 2 3 l s 4 7 s 8 7 3 2 9 4 3 9 1 3 a 2 3 7 8 4 9 1 5 9 8 3 M l 6 2 1 1 0 0 9 4 0 0 0 6 2 0 2 5 5 0 5 3 3 0 5 6 7 9 7 6 0 0 0 0 4 0 a 4 7 0 7 0 4 9 8 8 0 0 4 7 6 2 0 0 7 4 0 8 8 3 8 3 6 7 6 8 9 t e 4 9 7 8 3 3 1 2 1 9 0 1 4 40 U M 1 1 T B 6 5 6 6 6 5 S 6 6 6 5 6 6 6 6 6 6 7 y 0 + 0+ 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 d + + + + + + + + + + + + + + + + u rmEE EEEEEEEEE EEE EEE E i ~ l e 9 2 2 1 0 1 3 0 4 8 0 4 8 1 1 9 0 1 F En 36 0 2 2 7 1 1 7 2 7 9 1 3 0 32 4 4 7 5 6 4 6 3 3 2 1 3 3 2 2 5 7 4 5 y 4 4 p 8 8 8 4 4 12 2 2 2 2 2 2 2 7 l 7 8 7 7 7 7 7 7 5 5 5 8 30 a 7 7 3 9 3 7 7 7 7 7 7 5 5 5 ht 4 4 1 8 5 n 2 9 8 6 6 6 6 6 6 6 6 6 3 9 6 3 3 3 2 2 2 2 2 2 2 2 2 2 2 2 1 1 m n E b l i s o s 1 2 1 6 7 9 7 4 9 7 1 5 3 1 3 t a 2 2 6 0 35 7 0. 3 5 2 2 5 5 7 2 2 i aM z 7 1 0 9 3 0 1 9 0 9 0 6 7 13 9 5 8 8 5 8 6 3 5 5 6 6 1 0 1 1 6 4 5 3 0 3 l d 6 1 4 7 7 5 5 2 7 3 8 2 7 6 2 4 5 ai d lu 2 2 5 0 4 12 1 1 0 4 1 4 8 8 1 7 5 3 1 1 2 I 1 1 2 3 2 1 oF 2 N t r e  ; m6 . 4 4 6 9 1 1 6 6 1 9 31 8 8 8 8 8 8 1 1 1 1 1 9 9 9 6 0 6 0 5 5 5 3 0 7 a d . ww3 . 7 7 3 5 2 3 3 3 2 2 2 2 2 2 9 9 9 9 9 9 9 2 9 2 9 2 6 3 9 2 2 1 e e F T. 1 e - 1 s t - a E 1 3-C r 2- - r e E 1 4-e t E r d r s H e a t E a o r n d p e H r n H B t H t o n ea m o H t d B o C ioH u t le er er t e n 7 t - P n d d o Gio 2 / 1 Xtce o t o I a a 1 t B o n n rr i c t c t t o e e 3- _ e pS e  % E I t n n o od d e e t p HH ) E S- i v r t 2-- CC - -ae ae 3( mte le u l t r e Y V b Hb EXXHH s P I nI n l l a :e a : : : : : p b Hb l l a V :ela e en e n en e n e n m e e e l l l - i n 1 - n 2-iiiii u n n n - n - 3 iL i iii 4 E Le E lL L LL P L - 4 - LLL - - s e e e e e e d e e ep E e E l a 3 p r p r p ip ip ip ip ee p r t o Pp Pip iP t t tt e l e i P i HP HP P P P P F i i HP H T i b a 0 I 2 3 4 5 T 1 2 3 4 5 6 7 8 9 1 I 1 1 1 1 16 8.h!t i =. l Torus Temperature and Pressure Response 19 Large Break LOCA cnd MSLB Accident Scenarios VYC 1628 Rev.0 Page 52 3.3.2 Run 1 - DC-1 Single Failure (Table 3.1, Case 2) 3, This case represents a LBLOCA (DEG break of the suction side of a recirculation loop) with no loss of offsite power (I.OOP). De break is assumed to be on the.A loop cor.sistent with LOCA modeling methods (a break in B loop would be symmetric). The single failure is assumed to be a failure of the DC-1 bus. Equipment operability is in accordance with Table 3.1 Case 2. ECCS flows are assumed to be at their minimum values. RHR System Short Term Operation - two RHR pumps are available for LPCI mode of operation, one in each loop; the other two RHR pumps are not available as a result of the DC-1 failure; the flow from the LPCI pump aligned to the broken loop is injected into the vessel as the recire discharge valve is assumed to . shut; the other pump injects into the vessel through the intact recirculation loop. RHR System lAng Term Operation - at 10 minutes the operators have aligned the RHR system in the . intact loop to torus cooling. In the realignment the RHR heat exchanger bypass is closed and the RHRSW L pump is staned. De broken loop RHR pump fails due to failure of the associated RRU. CS System Operation - one core spray pump is available and injects water into the vessel for the duration of the transient; the other core spray pump is unavailable due to the DC-1 failure. The operatoris assumed l to throttle the core spray pump to design flow at 10 minutes. I l Feed & Condensate System Operation - offsite power is available and only two condensate pumps (no l feedpumps) are available for feedwater injection due to the DC-1 failure; the operators continue to allow l feedwater injection through the condenskte pumps as one of the pitferred water sources (OE-3101 and OE-3102); feedwater injection begins when vessel pressure is reduced sufficiently below the head capacity of the condensate pumps (for modeling purposes, feedwater is assumed to continue injecting immediately after the break occurs as the vessel pressure is reduced almost immediately); feedwater injection terminates when the temperature of the feedwater injected into the vessel reaches 175'F. Key Run Data Short-Term RHR Pump Flow -Intact Loop 5640 gpm I 1 -Broken Loop 5640 gpm (injecting) ' Ieng-Term RHR Pump Flow 6400 gpm(torus cooling) Short-Term CS Pump Flow 3708 gpm l i Long-Term CS Pump Flow 3000 gpm RHRSW Flow 2700 gpm  ; . i SF. Yte.464WP Apri 21.1998 l j l Torus Temperature and Pressure Response to Large Break LOCA cnd MSLB Accident Scenarios VYC 1628 Rev. 0 Page 53 3.3.3 Run 2 - Recirc Discharge Line Valve Sensitivity i l * 'Ihis case provides an analysis of the sensitivity of the results to broken loop recire discharge line valve failure. Run 1 is analyzed bat instead of the broken loop recire discharge line closing, the valve is assumed to fail open. Themfom, the Run 1 RHR system operation is changed such that any injection into the broken loop is spilled out the bmken loop to the drywell instead ofinjecting to the vessel. Key Run Data Shon-Tenn RHR Pump Flow -Intact Loop 5640 gpm - Broken Loop 5640 gpm (spilling) long-Term RHR Pump Flow 6400 gpm (torus cooling) Shon-Tenn CS Pump Flow 3708 gpm Long-Term CS Pump Flow 3000 gpm RHRSW Flow 2700 gpm l l s I ) l l car.vyc e 4.w Aps 22. eves ! l l- Toms Temper ture and Pressurs Response to Large Break LOCA and MSLB Accident Scenarios . VYC-1628 Rev.0 Page 54 - 3.3.4 Run 3'- RHR Heat Exchanger Failure (Table 3.1, Case 1) Q This case represents a LBLOCA (DEG break of the suction side of a recirculation loop) with no loss of offsite power (LOOP). The break is assumed to be on the A loop consistent with LOCA modeling methods (a break in B loop would be symmetric). The single failure is assumed to be a failure that prevents the heat removal capability of one RHR heat exchanger (e.g. MOV-89 A/B fails shut). ECCS flows are assumed to be at their minimum values. RHR System Short Term Operation - all four RHR pumps are available for LPCI mode of operation; the flow from the two LPCI pumps aligned to the broken loop is injected into the vessel due to the closure of the recire discharge valve; the other two pumps inject into the vessel through the intact recirculation loop. RHR System Long Term Operation - at 10 minates the operators are assumed to have realigned the RHR train in the broken loop for torus cooling. In the realignment one RHR pump is secured, the RHR heat exchanger bypass is closed and the associated RHRSW pump is started. The intact loop RHR heat exchanger is inoperable as a result of a single active failure. The operators continue to allow the remaining two RHR pumps'to inject into the RPV as allowed by OE-3101 and OE 3102 for additional core cooling. !. In addition, the opastors are assumed to have throttled the RHR pumps to design flow. l CS System Operation - two core spray pumps are available and inject water into the vessel for the duration of the transient. At 10 minutes, the operators are assumed to have throttled the pumps to design flow. Feed & Condensate System Operation - offsite power is available as are the two running feedwater pumps j and three condensate pumps; the operators continue to allow feedwater injection as one of the preferred - water sources; feedwater injection terminates when the temperature of the feedwater injected into the i vessel reaches 175'F. l l Key Run Data l- Short-Term RHR Pump Flow -Intact Loop 11104 gpm l- -Broken Loop 11104 gpm (injecting) l Long-Term RHR Pump Flow -Intact Loop 12800 gpm(injecting)" i

l. - Broken loop 6400 gpm (toms cooling)

Short-Term CS Pump Flow 7416 gpm long-Term CS Pump Flow - 6000 gpm RHRSW Flow 2700 gpm-II 12800 gym = 2

  • 6400 spm ; this is an arbitrary minimum long-term RHR now using Iwo pumps in one train.

GF.VYC te 4.WP Aprd 23. IW8 C________ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ . _ _ _ _ _ _ _ _ _ _ _ _ ._ Torus Temperature and Pressure Response 13 Large Break LOCA and MSLB Accident Scenarios VYC-1628 Rev. 0 P ge 55 3.3.5 Run 4 - ECCS Flow Sensitivity *!his case provides a sensitivity of the results to ECCS flow rate. Run 1 is analyzed but instead of the ECCS ppmps delivering the minimum expected flows, the pumps are assumed to deliver the maximum expected flows. An exception to this is the flow through an RHR heat exchanger that is actively temoving i heat as the operator sets the RHR flow through the heat exchanger using procedural guidance. This limits the sensitivity to investigating the effects of the increased flushing of heat from the vessel and drywell to the suppression pool due to the increased ECCS flow. Key Run Data (equipment operation is based on Run 1) { Short-Tenn RHR Pump Flow -Intact Loop 7400 gpm - Broken loop 7400 gpm (injecting) I.ong-Tenn RHR Pump Flow 6400 gpm (torus cooling) j Short-Tenn CS Pump Flow 4600 gpm long-Term CS Pump Flow 4000 gpm RHRSW Flow 2700 gpm 1 l a,. -. - ai-i L______----_---____----._--__.----.-----_--------------------- - - - - - - - - - - Torus Temperature and Pressure Response to Large Break LOCA and MSLB Accident Scenarios VYC-1628 Rev.0 Page 56 3.3.6 Run 5 - RHRSW Flow Sensitivity This case provides a sensitivity of the results to RHRSW flow rate. Run 4 is analyzed but, instead of using the minimum RHRSW flow, an increased RHRSW flow is assumed based on potential plant configuration and/or procedural changes. This increases the rate of heat removal through the RHR heat exchanger. This case does not require a recalculation of the RELAP5YA mass and energy release. (No additional RELAP run is required because RHRSW flow is assumed only after 600 seconds). Key Run Data (equipment operation and flows are based on Run 4) Short-Tenn RHR Pump Flow -Intact Loop 7400 gpm - Broken loop 7400 gpm (injecting) long-Term RHR Pump Flow 6400 gpm(torus cooling) Short-Tenn CS Pump Flow 4600 gpm lang-Term CS Pump Flow 4000 gpm RHRSW Flow 2850 gpm I l . cit.VYe-964.WP Apr8 ti,19H Torus Temperature and Pressure Response t) Largi Break LOCA cnd MSLB Accident Scenarios VYC-1628 Rev. 0 Pege 57 3.3.7 Run 6 - Short-Term Torus Temperature Sensitivity his case provides an examination of the highest suppression pool temperature early (<!800 see) in the event, it is anticipated that the operators will throttle core spray pump (and possibly RHR pump) flow at some time after a large break LOCA has occurred. This action will assist in preserving (or re-gaining) NPSH margin. In order to provide additional information to assist in determining the minimum required time for pump throttling this sensitivity uses inputs which would tend to result in a higher suppression pool temperatum early in the event. These inputs may not be the same as those chosen to maximize the peak suppression pool temperatum. A RHR heat exchanger failure similar to Run 3 will be analyzed with parametric inputs selected to maximize early heat addition. Dese include the assumption that the recire discharge valve in the broken loop shuts nonnally and ECCS flows are at their maximum values. In addition, no throttling of the injecting ECCS pumps is assumed so as to provide an assessment of the length of time the operators have available to throttle the ECCS pumps. Key Run Data Short-Term RHR Pump Flow -Intact Loop 14200 gpm l -Broken Loop 14200 gpm (injecting) Long-Term RHR Pump Flow -Intact Loop 14200 gpm(injecting) - Broken Loop 6400 gpm (torus cooling) ) Short Term CS Pump Flow 9200 gpm Long-Term CS Pump Flow 9200 gpm RHRSW Flow 2700 gpm i I, CIIF.WW-lH wr Aptd 23. AVY6 Torus Temperature and Pressure Response la Large Break LOCA r_nd MSLB Accid nt Sc:narios VYC-1628 Rev. O Page 58 3.3.8 . Run 7 - Feed Rate Sensitivity ' .This case determines the sensitivity of the toms temperature response to the feed rate during the course of feed water addition. The single failures that have been examined may have an effect on the feed and conden:ste pump availability. Since the feed rate can affect the rate at which hot fluid is flushed from the vessel to the suppression pool this sensitivity seeks to assess the dependence of the results on feed rate. Based on review of the feed and condensate pump hed vs flow curves and design point, it is postulated that the flow rate through the feed system could be double the normal feed rate. Therefore, double the 3 normal feed rate is arbitrarily assumed for the sensitivity. Key Run Data (equipment operation and flows are based on Run 3) Feed Rate (do' u ble normal feed rate) 3666 lbm/sec Short-Term RHR Pump Flow - -Intact Loop 11104 gpm - Broken Loop 11104 gpm(injecting) Long-Tenn RHR Pump Flow -Intact Loop 12800 gpm (injecting) -Broken Loop 6400 gpm (torus cooling) Short Term CS Pump Flow 7416 gpm . Long-Term CS Pump Flow 6000 gpm o I I (IIF.VVC-MWP Apr# 2 t.1998 ) Torus Temperature and Pressure R:sponse to Large Break LOCA and MSLB Accidat Scenarios VYC-1628 Rev.0 Page 59 3.3.9. Run 8. RHR Pump Failure Sensitivity This case determines the sensitivity of the torus temperature response to RHR pump failure of a DC 1 or - DC-2 failure. Run 4 postulates that the RHR pump operating in the room with the failed RRU (due to the DC-1 single failure) ceases to operate at 10 minutes post LOCA based on the VY LOCA single failure analysis. [3.2] However, it is not ' guaranteed that the RHR pump will fail as the room may be cooler than initially assumed or the pump may be more robust than assumed in the single failure analysis. Given the energy an RHR pump can potentially add to the containment in the form of thermal energy due to pump losses or conversion of the pump's work to thermal energy due to system friction and form losses, it is not clear that the assumption of RHR pump failure due to room heatup is conservative. Therefore, this sensitivity assumes the RHR pump continues to operate. Run 4 includes consideration of maximum ECCS flow rates which was determined to be limiting with respect to a DC-1 or DC-2 failure. No additional RELAP run is sequired because the pump failure in Run 4 is assumed only after 600 seconds. Key Run Data (equipment operation and flows are based on Run 4) Shost-Term RHR Pump Flow -Intact Loop 7400 gpm -Broken Loop 7400 gpm Long-Term RHR Pump Flow -Intact loop 6400 gpm(torus cooling) - Broken Loop 7000 gpm (injecting)'2 Short-Term CS Pump Flow 4600 gpm Long-Term CS Pump Flow 4000 gpm RHRSW Flow 2700 gpm l l l i i l l 1 12 RHR pump assumed throttled at 10 minuses to ensure continued adequase NPSH available to the pump. an.nc.uw w aem Tonas Temperature and Pressure Response to Large Break IDCA and MSLB Accident Scenctios VYC.1628 Rev. O Pye 60 3.3.10 Run 9 - Minimum Overpressure Sensitivity The previous sensitivity analyses have focused on examining the events and phenomena that can have an effect on the suppression pool temperature. This sensitivity examines the impact of containment spray on wetwell pressure. Inputs are chosen to minimize containment pressure while maximizing suppression pool temperature. The most limiting run is used as the basis with a minimum value for SW temperature and a minimum RHR' flow through the RHR heat exchanger given service water temperature being used to obtain the maximum temperature difference between the suppression pool temperature and the spray temperature. No additional RELAP run is required because heat removal in the most limiting case is assumed only after 600 seconds. I Key Run Data (equipment operation and flows are based on Run 4) { Short-Term RHR Pump Flow -Intact leop 7400 gpm -Broken Loop 7400 gpm leng-Tenn RHR Pump Flow 6400 gpm(spray cooling) i Short-Tenn CS Pump Flow 4600 gpm j ' Ieng-Term CS Pump Flow 4000 gpm RHRSW Flow 3000 gpm I ) RHRSWInlet Temperature 33*F '8 Note that scheduling restraints prevent completion and full documentation of this sensitivity. The documentation and results will be provided in a supplemental addition to this calculation. The conclusions drawn in this calculation reflect the fact that this sensitivity is not completed. l 88 337 is used vice 327 as a precaution to ensure that the low temperature does not create a dif0cuhy in the GOTHIC run. 1hc difrerence in the resuhs between 327 and 33F mill be negligible and the conclusions drawn as to the behavior of j corr.ninment under very cold spray conditions uill be valid. l m.vveae.ew Apni n. im J Torus Temper:ture and Pressure Response to Large Break LOCA rnd MSLB Accident Scenarios VYC 1628 Rev. O Pzgi61 3.3.11 Run 10 - Main Steam Line Break This case provides an evaluation both of other break sizes and of other break locations. A double-ended guillotine rupture of a main steam line is chosen to represent this evaluation. The single failure assumption, equipment availability and inputs are the same as Run 1. Run 1 is chosen as it is expected that both recirc discharge line valves will shut as the break is not in a recirculation loop. ECCS flows are chosen at their maximum values in anticipation that this will be shown to be limiting from the results of Run 4. Key Run Data Short-Term RHR Pump Flow -each Loop 7400 gpm . 1 Iong-Term RHR Pump Flow 6400 gpm (torus cooling) Short-Term CS Pump Flow 4600 gpm Long-Term CS Pump Flow 4000 gpm RHRSW Flow 2700 gpm , Note that scheduling restraints prevent completion and full documentation of this sensitivity. The documentation and results will be provided in a supplemental addition to this calculation. The conclusions j[, drawn in this calculation icflect the fact that this sensitivity is not completed. 3.3.12 Run l'1- FSAR Benchmark This run represents a case similar to that contained in the FSAR Chapter 14 long-term containment analysis. Details of this case are provided separately in Section 5 (for the mass and energy release) and Appendix D (for the containment response). l I i l l .. ,_, - , . . . . 4 I t Torus Temperature and Pressure Response to Larg) Break LOCA rnd MSLB Accident Scenarios VYC-1628 Rev. O Pags 62 3.4 References ..y 3.1 Calculation VYC-1290 Rev. 2, Vermont Yankee Post-LOCA Torus Temperature and RHR Heat Exchanger Evalunion, approved 8 August 1996 3.2 Calculation VYC-1511 Rev. O, Vermont Yankee ECCS Single Failure Evaluation (for LOCA), approved 7 February 1997 3.3 VY Procedure OP-2124 Rev. 39 Residual Heat Removal System 3.4 VY Procedure OE-3102 Pg.1 Rev.12 Pg. 2 Rev.14, Pg. 3 Rev.14, Altemate Level Control 3.5 VY Procedure OE-3104 Rev.13, Torus Temperature and level Control 3.6 VY EO,P Study Guide, Rev. 6 Approved 19 December 1997 3.7 Work Release Form 98-0079-00, Torus Temperature Analysis 3.8 VY Procedure OE-3103 Rev.14, Drywell Pressure, Temperature and Hydrogen Control 3.9 Memorandum OPVY 98/53-A, J.R. Hoffman to C. D. Fago & B. C. Slifer dated 17 February 1998, Input for Base Case Containment Analysis 3.10 Calculation VYC-1756 Rev. O, Estimate of Water Volume and Weight of Pipe for Residual Heat Removal and Core Spray Piping System, approved 2/23/98 4 l l sar.vve.m4w Ap s s.sous Ws MDCbv.O 7.0 ~ Results cnd Conclusi:ns This section provides a summary of the calculation results as well as conclusions and key assumptions [ which require confirmation prior to validation of the results of this calculation. Additionally, as will be L shown in the summary of results, the bounding combination of sensitivities has not been analyzed in this calculation nor have all sensitivities been documented. Subsequent supplemental additions to this calculation addressing missing sensitivity documentation and the combination of sensitivities providing the limiting peak torus temperature are being performed subsequent to this calculation. ' This calculation has documented the development of a long-term post-LBLOCA containment analysis methodology for Vermont Yankee, a BWR with a Mk I primary containment. The method uses a RELAP5YA vessel and core model for short-term mass and energy release calculation and a GOTHIC containment model for calculation of the containment response as well as a GOTHIC vessel model for long-term mass'and energy release. The method and models were benchmarked and validated as being adequate for conservatively calculating the long-term containment response to a large break LOCA. ! A set of cases with differing single failure assumptions and input sensitivities were analyzed to explore the l post-LOCA torus temperature response. The GOTHIC portion of the planned MSLB and minimum pressure sensitivity were not comple'.ed due to time limitations. The RELAP portions of both of these cases were completed. The documentation of the cases that were not completed will be included in follow-on analyses planned to address the bounding combination of sensitivities. l ne bounding combination of senstivitities has not been analyzed in this calculation. Specifically, Run 6, ! _ maximum ECCS flow, and Run 7, maximum feed flow, are not combined to determine the bounding short-term or peak torus temperature. The results of these analyses provide information to assess the bounding combination of sensitivities for planned follow-on supplementary additions to this calculation. As well, these results provide suppression pool temperature responses which can be used to determine the replacement ECCS suction stminer NPSH performance. The methods and models can be used to analyze other scenarios as required or requested by Vermont Yankee. De transmittal memorandum for completion of this calculation is provided in Appendix L I ew.nc.uw wnm t_________________________._______________ _ _ . _ _ Torus Temperature and Pressure Response to Large Bresk LOCA and MSLB Accident Scenarios Page 43h VYC 1628 Rev 0 7.1 Summary of Results and Observations Table 7.1 provides a summary description of the cases analyzed in this calculation with key equipment operability per ms as'well as a summary description of the ECCS flows and peak torus temperature results. The cases are divided according to their basic single failure assumption. The results from the previous design basis torus temperature analysis, VYC-1290 Rev. 2 is provided for comparison. Common assumptions for each case include (except where specifically addressed otherwise): = Breakis on Loop A 2700 GPM RHRSW flow through the active RHR heat exchanger 6400 GPM RHR flow through the active RHR heat exchanger Both recirculation line discharge valves shut normally Mechanistic feedwater model using nominal normal feedwater flow rate ECCS pump heat added for the operating ECCS pumps Suppression pool initial conditions of 90'F and 68,000 ft8 water volume 1593 MWth (+2% uncertainty) initial reactor power ANS 5.1 1979 decay heat (+2a uncertainty) Several observations have been made based on the results of this analysis relative to the previous large break torus temperature analyses: The DC bus single failure is not limiting with respect to torus temperature as was assumed in VYC-1290. Comparing the results from Run 3 and Run 1, the RHR heat exchanger failure is . more limiting. The RHR heat exchanger is limiting because it flushes the maximum energy from the vessel. As well, the number of operating pumps is maximized consistent with the EOPs which adds the maximum heat from the operating pumps. The method used in VYC-1290 Rev. 2 was conservative for calculating peak torus temperature L (though not conservative for the short-term temperature) given the inputs and single failure assumptions that were analyzed. Thus, the method of directly adding the hot feedwater at ten minutes may not have been entirely conect as the effects of additional flushing of hot water from the vessel to the suppression pool were not accounted for.

The issues presented in Vennont Yankee ER 98-0025, which identified that the previous torus

'l ! ' temperature did not identify the limiting single failure, pump heat, etc., was correct and conservative, i i CDF . VYC,N-9M AprilD. tWp o k s s s s s s s o k s s s e a i me P 0 4 0 9 5 6 5 0 1 $ 0 0 2 5 8 5 5 7 2 e a me iP 5 3 7 4 3 6 8 2 5 e T 4 1 8 1 8 1 7 1 7 6 7 -. T. 9 1 9 9 n e 1 1 1 1 1 s r - e p ' w m m m Tr e r Taek 7 7 7 T 7 7 T . r ek T T T gt 1 3 7 7 2 5 8 T. Tage 2 3 3 e 4 3 5 5 6 5 8 r 1 1 1 t 7 7 7 7 7 7 7 te 8 8 8 W a I mP 1 1 1 1 1 1 1 a mP I 1 1 s W I s u r r u o T n n ei )n T T T 7 7 T 7 T o m) en r T 7 T Tm t r 4 3 5 4 8 9 6 8 0 1 0 0 Tim t 7 1 2 2 4 2 o10 1 1 r h( 5 1 5 1 5 1 5 1 6 6 6 o1 0 5 1 6 6 1 1 1 h( 1 1 S . S mm yy mm pp mm yp mm pp mm pp mm pp 9la mp t 9la mp t 9la pm 9 t gg gg gg sg sp gg o g I o g 1 o g I I t Ct Ct C m 00 00 00 00 00 00 00 00 00 00 00 00 Cm0 P L y4 0 e r P p4 m0 1m0 0 p 0 P r )n m )n L 00 00 04 04 L p e i 37- 37 36 36 04 46 04 46 g6 u r i e 2 6 1 2 g6 4 2 Tm g01 99 99 99 99 99 99 &090 I 0 l i a Tm g0 &09 0 &09 4 & 0@0 F 8 n o> SR SR SR SR SR SR S1 R 1 S 8R S 3R S L( CH CH CH CH CH CH CiH e Im(> CI H C2H C 21 1 R 1 R 1 R I R iR I R 1 R l g 2 R 2 R s I 1 ~ I i i 1 1 n i 1 i s o w S wo l r l P e e mm mm mm mm g F r S u CC pp m yc a. yy pp pp mm py n S sh n e ci ah n c a ci c a l ss gm gg sg gg gg a C a ar ar p i 00 h C met pm met m a B 00 8 p 0 s 80 04 00 00 00 00 00 00 c E g p pm pm F 64 7 3 0 76 64 64 64 x g p g e m )n 47 4 35 47 47 47 E m )n 8 g 0 g 8 g l ri e 99 965 99 ri 0 4 0 0 0 g Tm 99 99 99 t a e Tm 7 10 3 6 4 20 71 n t 0 SI S SI SI SI Sl e 0 3 i S r 1 o - h0 CC P Ce C CP C PC C PC C C I f H tr 1 o - 9i I 914 9 1 S( I L I C P 1 L 1 L 1 L 1 I R h0 S( Se C S9 S C@ I 1 2 2 2 I C I C L H 2CP 2C P 2C D 2 R L L - - 4 4 t s s l t l u s r e u e t s a e R d w ) U R e e n R e r f e e R r t u ic yo p d u ) w t a t s t i s ile a lo r i n ivla ) a r f e a it i ) m f e n p h s f w p ht p s c ne lo g tyo i n v i e e Se la f 0 w m iv tc m e v n 5 ym e i je T n v e io 8 2 it o T e t in s n i s eo s n l a gr tc v or s k r eS u a Vah e it n t lu SC _ r Ch j n iywis t ins i r E i a eC o t e c gs yi v o n ep o F r T ewi s ri ad i S t t l T r tuE. tyl

  • f a e e h ivc inW i

s f Seut m )e f e g a x r - o Br u r u c e t i C o e a iv 2 el l si t sE eS r pa u r n a pm i d t y r v e ei kf a n ia F ic De r S e xa SR n wH FaHo ire p l y r h c mh et is n a R n p a x Ti w e ale) wtm lor p Ro s e o i E Sw n 0 Ygn lg i o mus n mtu lou Ft u t t al o wt t t _ e 9 tnsino n i a r _ n 2 itp i S l un P b Wb ub e n H e eb T 3 lo u 1 o P4 u n mms ce S 4 u I _ 1 S F3 S C- I uk S R t - Y rCu C co e r Cn uRnHo HonRni t H r n ou d n e o V eDsa D V( R( b C E( R R(R R( Ro c R hR S( cR F( 1 b 1 7 7 'e 0 t 2 4 5 8 e e s 3 6 7 h l n n n n n n l a n n n u u u u u u b C u u u c R R R R R R a R R R T T $ 4i,E3 Toms Temperature and Pressure Response to Imge Break 1DCA and MSLB Accident ScenariosPage M VYC 1628 Rev.0 7.2 Kes Assumptions and Limitations 1 Three key assumptions regarding operator actions were made in this calculation. These assumptions require some degree of validation prior to use of this calculation as a design basis. They are:

1. . Torus cooling is active at ten minutes after the start of the event.

This is based on the operators providing at least one RHR train in torus cooling (including wetwell and/or drywell sprays) with toms cooling active no later than 4' 10 minutes after the start of the event. Current emergency operating procedures and training would allow the operators to continue RHR operation in LPCI mode with an RHRSW pump operating through an RHR heat exchanger (i.e. injection mode cooling). - Additionally, subsequent supplemental additions to this calculation are planned to address the use ofinjection mode cooling.

2. The operators will provide at least 2700 gpm of RHRSW flow to the active RHR heat exchanger when in torus cooling. Current procedural guidance identifies 2700 'gpm as a maximum value. Considering potential instrument uncertainties and operator bias to preclude exceeding a maximum limit, the RHRSW flow could be less than 2700 gpm.
3. The core spray pump flow is throttled to between 3000-4000 gpm at 10 minutes after the P

start of the event. . This is based on the operators taking action to throttle the CS pumps to 3000 - 4000 gpm Ul no later than 10 minutes post-LOCA. Cunent procedural guidance directs the operators to throttle the low pressure ECCS pumps if there are indications of cavitation (inadequate - NPSH). However, this procedural guidance may be insufficient to ensure flow is  ; throttled to within this range. RecommaaA** ions have been provided (Memorandum ' THSAG-VY 98-030) for revised cautionary statements in the EOPs that would preclude or deter operation of the ECCS pumps under wetweli pressure and suppression pool temperature conditions that could lead to inadequate NPSH and ensure that the CS pumps are throttled to within the appropriate range. There are several limitations of this model/ method and the accompanying analyses; l

1. The initial fuel stored energy is based on a Cycle 19 core with GE-9 fuel. The stored energy does not necessarily bound stored energy from a core with GE-13 fuel'(or a mixed core).

' 2. [ This model, while it does calculate long-term containment response has not been 4 i adequately benchmarked for use in predicting conservative drywell peak pressure response. 1 . l r_______.-_______---_ _ _ _ _ _ _ _ . - - _ Torus T .p.e and Pressure Response to large Break LOCA cnd MSLB Accident Scenarios Page W :> VYC-1628 Rev. 0 7.3 Procedural Compliance Notes 1

All outstanding EDRs and CRs have been reviewed for impact on this calculation. .The following CRs were detennined to have some potential relevance to this calculation. The disposition of the CR with respect to this calculation is included:

CR Description Disposition 98 0007 Use of the 1979 version of the ANS Decay Heat Use of ANS 5.1 - 1979 is specifically Standard in detennining the hot leg recirculation justified in this calculation. switchover time instead of the 1971 version. 98-0009 Potential discrepancy in modeling of energy Specific energy deposition distribution deposition between PROSSTEY-2 and assumptions have no impact on the RELAP5YA computer codes. results in this calculation as long as the totalenergyis conserved which was confirmed in this calculation. Relevant SER conditions, specifically for FROSSTEY-2, have been met in the preparation of this '~ calculation (see Appendix H, Section H.14). No other SER conditions are applicable in this calculation. ,, This calculation, by itself, does not require a change to the Vermont Yankee FSAR or Technical Specifications. To support use of this calculation in the design basis of VY, the operating procedures for the RHR system as well as the EOPs will require revision to validate the assumptions detailed in Section '7.2. ' Relevant WE-100 Table 1 input considerations were considered in the development of this calculation. i 4 I l an . v,c.u . l e n. - 1 l i ! Torus Temperature and Pressure Response to Large Breik LOCA and MSLB Accident Sc:narios Page 521 l - VYC-1628 Rev. 0 - !

  • Appendix B j I I MEMORANDUM l'

L . VERMONT YANKEE DESIGN ENGINEERING - BOLTON To J.R. Chapman Date December 29.1997 +e& l From D.E. Yasi Group # OPVY 491- 97 ** Subject Design Engineering and Operations Input to NED's W.O.# Conrainmaar Analysis l Re-issued 1/12/98 with new memo nunber i

References:

1. Memo: LOCA-VY97-054
2. Memo:LOCA-VY97-086 (attached)

Discussion: A working meeting was held in Vernon on Wednesday 10/29/97 to review key operational and design inputs and assumptions for the Containment Analysis. Duke's Nuclear Engineering Department (NED) is preparing an updated Containr'nent Analysis in support of the ECCS Strainer Replacement EDCR, Bulletin 9643, and BMO l 9645 resolution, and a Technical Specification amendment to change the torus i yer. Lire operating limit from 100*F to 90*F. In Reference (2) NED provides an updated and consolidated request for input. , Attachmant I responds to Reference (2) for use as design input to the on-going l4 . containment analysis. It represents integrated input from VY Design Engineering and

    ?)             VY Operations Department. Attachment 1 is intended to meet WE-103 " design input"                    gg
 .-                requirements. Where follow-on actions are required, the responsible department and                   '

l agreed upon due date are identified and AP-0028 commitments will be assigned by the Technical Support Department. Any such commitments not completed prior to NED ' l- calculation issue, shall be noted in the conclusion section and highlighted in transmittal l i memo. The NED calculation will require revision after completion of the commitments and prior to its use as the calculation of record for Vermont Yankee. It is ax=*M that all open items will be completed prior to startup from the 1998 refueling outage. Reference-(2) is provided as Ateachmaat (2) for information. L d b, Paul A.Rainey-Consultant" Fluid Systems En ineering a l -

                                                                       'Gedge'J. Hengede -

PrincipalEngineer,I&C Design Engr [ ~ Mames Kritzer EOP Coordinator- Operations Dept j i

                                                                                                                                )

Torus Temperature and Pressure Response to Large Bre*k LOCA (nd MSLB Accident Scenirios Page 3 1 VYC-1628 Rev. 0 Appendix B

  )
                                ' Page 2 Oc4d Dar.iel E. Yas'
                                      .                                                                                                                        Nuclear Se      Manager en     ~

A dn1*

                            .                                                                                                                                  Kevin Bronson Operations Dept Manager cc:      J.R. Lynch J.R. Hoffman                                                                                                             .

C.Fago _ R..McCullough , R.G. January J.H. Callaghan EJ. Betti B.C. Slifer J. Pulford

                                                                                                                                                                                                      .                 1918 O

i h

 +                                                                                                                                                                                                8 9

L___________-__--_____- __

lllljlllljj!i' llli l :l -l ,l 11

                                                                           ~

g$. i

                                        -I                 g g [j y6. 85h3E[                                    6                   > gg1"2 $if                                        E8 F  5_.
                                                       'o
                                 <9 ;W;                                                                                                                                                            ;

dD2m e yo e i nl l e C- G gs S s r s e ht mhR r s r i ht s Yw Vo C-Y=F tool l a o s g' t n o, t eR H R o f l Vewl > (e. t ndi seo e e es mG e e mG f oU f mmhc i tmi,fW v i r dHoRh) t t e i s v ,fF i rS nR g s ob i l vf et e n ts J t a mredn eos . d r eh do wg6l 8 edn( r reo iR) s de/l9 S rW 8 e y m m m

                                                                                                                                                                 %Siso
                                                                                                                        .                                                               t r ml   l    i r ce                                              ice                    l                  l i

n no u1 iuol u l l g vef r Coiuo w es qt el l o r4 / w qt a ic1 i wu/d3 ww n i. t e gU p GGr esv vah s i n oi 8 Grs nom t snRl wESsaiom) t 2 0F Saiu) t Hu XmoMF u8 GeG Sr FnS F Ss o Ghm) p8 ii xrRa e m. eam t aui61 mf ( 2. eam 8 0- oF Eee rd aui6 (. 290 g 56 :e gat t i n eh d l 1 t l i t l 1 0 ngee Pl i cx5 a1 2xe 2icx5 34e3 P2 a70P27/ t t t APina Sicm/PadXPi l 2 S miH s Si nca m 2/ l a1 S1 r S1 23onh t 1 NEcto en n " htafn e t ei s B t e t hd i e e2 2 e i s snTe n w r r de . o , S i e e ce)i.gdl so dte u u7h t edet u s Y t ar5l nf aa ed s s t aCo t aolar .h V r o 6 n rd8aurnoeh pn . ameYt 4 p pV ed pt idvt 8 i wn3a3 ebrd o d ua l 0gmrl i i t c sgeen ar _ 0- d ,1 hwo* l p 1 0eept ni a vip l 7 e l f g w c P 3 0 Pp eeo s 9 t s gnS l oewirdwS E7s u a srdt i niYmf z i e c i V, g m ed sre IC g I O 2 r )n be l p dt oo o ibWrum i t Y- s l s 1V o n t 0micl 2s i uta a E. w . oe2 0 rt id el Stsp l TA

                              -                  o                         ct a1                                          t nwei a    oyd no      u l

gi R n g

      .       NC l

f n sl up r e 96in i 54 nd et xe l t i 4l oh awHi0 t u4Ri20 l l EO o o c P s m ins w A, er 2fWcs o21 1 d m2m w3o $. ML i t c t fT i i sea 2e yty Ho e diSwi h h ), 8 0 p Pt a O c )w t a fOm*0 r n o rPuet 0 _

r
  ~                                            j                          er                             w987                                      ni n e.                     io(ef                                                      i edeet                                        iA Cm l

Ae it mi a pondeo n c T.c4C1 y7 8 r nbPed l r .piw0 u n TM up r e er et or Y3 dtuiaOc e nta a d me3 t r r o T o mor xdigmuYedaCV2 e c

                                                                                                                         /

eewEuct cso( e a pt At i xp ap msapxxgcVey r1 ol l l l Fmdon i i s r i. va s g w s e n e sn m ae u m b, e e itn a( pb nmn9 s Pw ri g s m10g a0pi f gmi t f no l o o hera i x apn mgmlo e po7 nr u 0 m r, g p ar d p iomd1s5yenge t g vie t 0 ept s e pt s ,7 m gi turio0s a mu r0 cm d gi i t C-4 at aie r x e a o ac R e s6 Re8 e0enut r s 0nY e r p ep a c/00 owr0 e7ei p2 d Y9 D7omton0oV 0 pi i7c O o (m x + 7e a 2 H o > %in e l g g n n w o i s i t nd W S mp pmHd R cs )e oun l f x a Reg t n a g g Ro ce Hnge ' m 0 0 mar Rarinm nge i s t t 7 0 0 y. nwar eo t u 6 4 7 6 4 die n nga e cr dl if pn es r t wau 8 il c 9 cor cgoi S R co a cwgc anrd s r Y HS t s sl oga c ion t tf a sa e RC n or gs t e t V ouf progtae n i t pe pine .

                                                                                                                                                                                                                 \>

m a e s u m: r s i et nr em d (p a i t pu deot puer r u ca nd y qf e omocu i vme i r e nu r or v a o r uponrs t cocc P R ol Cv a Ppofi r ucr Ppapa m e em. 3 e t - g s y o a t P SI N1 2 3

                                                                                                                                                                                                          ~

(

r X H) ( t f o l l e s l , it d8 si9s/ ts e r ) l i s e . mW t a w p5 n el l i79 f98 ws syt e rS hyf o0 t 1

                                                                                                                 /

n R i

                                                           )l          a           f l

22 mi,2/ wse1 Dar 5 t 1 oH f iWheG aosvC4/ Eanw i r n nl a N lo) cR eSt S nivY S (Xf98 l f l o v R g( F . oHi n. i o a eVP SHW5 wy t / t Cet rl t yO N RS/Gi / 1 t

                                                                                                    . O Ren t t e e 8

2 Oh pob(r d s BHR3 SiseSdemm l b 0 ei s D R H S PaiFs t s e 0 4. U i y 5 es R N 6. t a t r pc i r

7. e w i q rou ul /

P 3PRqa3i nPv wF3 eDPc eo3 ePcl cl ASRr aSr e n( eSavSiar . nl e s e d - ne . - 0 d s , n9 ot a 3t n e0 e yer , to a8 ddep f e s.b 5 t n . aue sl de l i l d dL . 0 p e t s a un 0 o yudPm-n ead ar ogaannteG a i nna coI ovSesl i l

                                                                       ,        bli p        l 0, o     f       l g     pmis                 d e Q pR 8

nwP 6lusaAt 3pb ym pt sn t r n u 6 i t mwd SuA rCS t f ot eo aUeS d d 8 r eI I s si t i e 0- ae n0l ps l f purr en v n s s ioa rF , beg d .i n l er3 af t e0 uoiaee t r m. u M- u ot o ret t l umt s t as0 m n7 5, w d e s r r a Y s t a i uee ne opx i s a0i a6hceny r c ts e e ed 0, t c d e. w at ss nh h ge 1V vietnau r s s r uis6ole8 n e,i e pi t qv . C ba hb Slev x

                  -             i 0r                             i TA                             i c e             ct a               s5 e p      e                                                       w yh          e
 , .                              o ;e v m tooa                                   s8                                   eel            ri t a ce                  n NC                  r p    gr ed dd                                   a27 l

el EO oa ne s pe nX l c d /9 r r

                                                                                                                     ) evf         eoaimTsly vm                    r ML                 t h          ,        ng                      nn1                                7.         bl uet al had   i t            u r a Ho :              d      c yt m      I ae n s H, ). oa3 i

i ,/ 3 ( o s pisl eo e _ ' e e rien pd gts c a02

                                                                              . t                                          t un         nc at vt e ort Cm l

nuas.e sd ne u7y01 cd e e o ng y imr s pf oa l ri - i Ae l d r t si pnt grl t d TM l ee cys cei t w m,l c2 a ,b Snoia at i m8 p wcona h atdemnaraI v n7 0u r i T l s r e a r a c0B s At o ef _ l gr p uncf dg cpedf o e u t aoa( d oi22 s5N e5 e Ttoal t e r,a shCf is e e s n s n a en X pl r l a epivf2 r t gnebedo l nwYd e o p e o i_ h cmHegeat, si o 7- t t iins r a vectVl u t O pewueojt m fe c:d s e p ;g suuRn s r r u ul t e isC nw s t s s ee R eion t r sHa s a h s v3 eoY iiMcos3 xh c s o oh r c RPfiRcae1 SfV l E'IBaHaf PNs t . 7 9 r e t e eO r g a oM. a w m B ;y osr e . p s r r ufi d o r oe ei s rdw _ i e r awn t t u p n 3i m t r f pl oot n 0 o (tCe i t n o r _ _oc f 0 rp i

                                                                                                                                          .m

_ n 5, p m )e ru s

                        -         o                                                                                  7 al vt     i i

6sl is cvin aat t r a i _ e t t s u me e e n i r mtyenird m ui t f a q n nul an s a r a e o RC olono1 Cvac4 P m 4 g t e s m. yteo t n a P S I N3 c o . 4

f E T ! 3 " f a s { a E X- { > g g p a y g i y g g s - - t w

                     <h h                  ,
                                                                                                                                                                          %w
 ,H
                                                           . y t l F

z sf bt ere n 3 r t id i r foo Fut n4te e2 a P c dt m1 d y S :e denu t t 2 i Pu p c.I od o mOg .mNac o s c n r a a ' Cdm ase i w f ot ut n8bt

                                                            ).

9t i o 8 pynio/n nty 2 eta /eiona1 i 0 U=m m3mtat 0 1 u rSurrbe r r

                                  - S             sf opt ic P

AS P nnOn mii(iancu t sl t y )

                                                                                                             .       g                             .

nl eev t F w n '

                                                                                                                                     ,           w         R e_i F                           0               ooye i

r) o ki9m0

                                     , mi                                                                          l omp             l         . C nt a                                            9'.          .t l el tu            f                 f n RD bn eto v et md                     bnf al g                                                                   g    g i r oCE 6

at t s ni onint c mgn' t ni ois u yi meb a t l l a bmi.r8bt ei i u e xqou9 r4e r e u r l ei01 n n0 r s c u3eejn 0l l i t iDn t eEid A.weisnwunF a memihenc r l 8 s eio=fiini e m ei e p d a 4 so Si P r ws cc :< a h ndmt l l r

                                    . l                                i 0                     a   d                                      t 7-                    r    a    r v o nr2 nt      t n  t ei r oyu    ar emae1 c2r                       ee aI          mbe not ogr          a (y ed w ifgi f i

r t 9 fl t p yIi , t t nr n u e n o ph ga nar at i t t nwsieco n Y- i mspeeimngusf gl ot 0 mot nd urhiaa et atni fo d e l pe c 1 V- .i emix siat nf p e 7. a hc e nrfm ro4t6 o A monoc nam 0st r i c e t n phecr a u gb l ms -y TA NC _l l a s r cis3ms t m  :+c-i t uo s ar ui r t e u n m m0l impe b l i t n ucehm ion ni o retot ahs us i0iw l EO i i pc ei wpanlasl Twt i ML yooptemt r sgc t x i l aitt f n t o ori e t s0l

                                                                                                                        ). onna0wn ei                       o     .

H :o arf S na e 0 eh s aur r eo o r ri n (a eetaycr i l umf epwn2 d a yt l oi4 e f mgm3ntna/ r sib t Cm h aaga r s,O A e l o7t l a r - Ae t e p9 c =p wt earf f t t mitnhmrem7 TM i ao rt a nT apsspnpe e yf 9 o sr nHireo1 t erf7 T p n h o *s ip arOeo s e

                                                                                .ct de StS(

i u adRhi r p Smta in oPl etequ iyY C-pe Atoe olat t ut mse) t en t i wSnCo t i e r p P. snr s t c e nnya omo ei gD t oNyr s n s psleeo mt et of nd p C 0 f oe(sEdl j i i t l r el e s gd bgr i n aV, o p s na o ss piyt ededh eostadautu1 nt rPr ul et o use g ort i r eC d ectat aenete20 l t gguoutquria c4 i l r nsn uf ntetownat r t e sFl a R Re0nmm0ieFPt 9 a i a 9 IiSnacr l l eci uuRAhNboit v ci r eit t se nh de n sen7 eu amaDU9 i sut

                ~                           1 t
                                    .s i

s o i t p msur d n wa - i l l o ob - d at l f y - e ia cl u t i yt n n yi t a . i l an r p A t n o c ni a o r SMC ocO Ff ounLk

                                                    .                                              r 0e e sr e     t 9 tam      t Cna   oe t

e s e rmi pr s r mir rt b ufior ep i a m a a r qnpm e o pe f ce ni g od rna P RCat Cil m

         ,3 e

t-e t e s y m. eo P S t I N5 6 z ll l

~ - ~ W. Y ~ S

  • f' ~
                                            .sr                        re ""

- s o edi vr ** t n t i n

  • e m pm io t a" '

t i m ph s d'P m oc U

  • i

- o s eh . ) o t Ciwe8u8C 8 v r 2 e R50)a/E t 1 I 0 r 4 1 e 4 1 0pS 2 P AS( POeOP*g 7PmP7 t ( S'

                                                                                                )

d e s g ht e %o - o l' o n r s ur 0g m d oio t f b o l a u ce 0 d o r t oD p 6 l ut r f at oie n oiy r rl o of )ev :esbo 1 (l u f

                                                                                                              . s 1      w1 nW or      o hmpul l

l uwev odeflo f t feiC s si o cdaOi r t l f s oayz e a se f. sl C (Noa _ 6 l t c _ 8 sl nag y' l yel vcvln ainr e m oe p( pR 0-7 is acla yi P toieS nLW( .ma al er odixtc u l i r mav eiea a rs se l t . 9 ay nl . dedli uhfee d r t Y aae nr s )' t agr oewimmr s n eyoi ot otno s ed 1V - u vF e. dei ca i t een a ae5 t r r ed h) eh ep eat a rem t , _ TA r r u t n c y a ," r t  : r e r s e oy NC n mhee t 1 t pb /- r eFd a e nol a nis i c sanyt l i yOpt ra r s a al a .deps f EO i aamlei(+ p ee5 pulvgs r l

        'ML                             t                   l mbl 8u   3 mc           oi nFt                                                            s aue e ac vs t r enlabdst erdoton c8l y )w nt                 y ei H o:                        o5 n e)4 t          l wd pr s enemie s
    .CmAe                             h ef o p5       p 0 i W p eeW ;w apid m t     a                                       o toon s

pi I umim2 en1 u1 r TM l i a ue rP s0eSo cSg yc s t t cf i c ns mCba2 T o noi n ea xt h ba r oi Ps sn P l it r x nf oime Lo oO 1 At htr vt a aeOu ecWeeoEaf t c; s e e un t ennSumf r i ac ohnp r5 n s c gW (m e a f t t d Rn nr r o T.Vi i so8 o p s o nnie p cdvc oes cdeues i ml l i r vnppe oraseH eedeR Oeo r ptaot n.oret r l e P ten g2 uRui4 1, s e e e r otsri l R e n eh el ra meefm i1ofOcmh plu pnnoieio2 no8 RWh Ser - t pi snu t t t mc1 t R

                                                 .o                                           H t

s R s , n f e o ov Mia na l d W.l t i ov t s

                  .e u               S vu                                                     i s s n                 nsl gn t c ay
                                                              .                                 o p ap i

t ii cc y ' n o saye egt n l ab

                  .c                dant                    s                                   mX

~ ~ s Feii as 5 r t oH r r e t 8 tureo n nB a cc s mN uewr en e t e m e r pur n i 5 m i u f n6 a r qvmd e e e nss o0 a RRt aa C1 P

               -    m s.s t

e s y eo m. _ r, .S t I N7 5

_v . - ;s r g _ t t o n n a e uf o9 8 l l i) _ t i misC/ s moo / 1 _ m ns3 mroyoP mruuS r . o rfO _ Cf et of r 8 P a A (n mCo _ 2 0 9 11 frOi t Li a _ i PAnBi t AOcLIo n d t g e a n f 'f t c d i o ol a d i r _ u Aet n e l o t l oi n e te o o ar r f . sCta eb o0 r r r _ t c1 o m d n moi oLnilsd t s t t no t e n n i i i s n r on i e t i s uma p c oi o m k n o oh e a r - 6 r n aioht s t e mtza o 8 0-ae s h re g csf s s d a h r, t hh ei r

r. f r

e 7 o wb r nsstsenlo i i e m n//b u iPFos p 9- t a gelgoyuin r nr ol n ai b e h00ss t 08l e i s ), Y s9 w1sto d ar pe i r i el ac ai t F l y2 1V pol oor s a nme0 l i l a 5 csFoneo urt si _ l n o1n0 w 0 udd n

                                                                                                                                .       1 TA                     e                                                        ).i e6          o     ee s eaC s

. 'v NC EO ht 4o0 e uFs met s,ts> r

                                                      .t a e n

u r g 5 et aywc mrar e t nmt ee l n u rY ee uV s-ML 2t 0 nslieb mia l r oi n t wu a rulub a( t ut abai o omimdfno 1 2l e1 t l i r s t HCm :o epi t l t _ i Pbs s n nf w aeu o el di oax uin l t Ae Ol ad i ee a s o ci go e uf mr e s x r oyyn uaa mar en TM d a n cRr stat rr u n comp pmmc a v i o t e a xAoe s8 T o eS Fdt a sr j- 9 p e a ,ai d y sus sc At l s.lar nop e s hoirsnna t e ar pt oR r - e 4l 0a eF r mmo e r t e  ; 1 e ueeif esm)w..rf gtatems s n s 3t g e n s ei x o n a at t htuid ei s mpt rera nP l ed p oEir t esi t nr et i aOl e s n s e sp Oih p n m h uomstekp ri u nufntu r r o s suoetgn E r yl ps w o, ep R e e RPtoe t e1 w a (t mCts mev0is t s oi oa l A ( Ah er s t t a oC er pd AA g l s n a - teRi l tu )e uHo n o l cl aba iRc l a mes h u mheve st i n nt r d rehc ote sr to - i t t ur n a aa s ca ah cf co ts ht t a et d e r t n a _ t oh cAn od e i ewcr ,pn e a r e u sCg s n aAeo s xw pnOi l a rCdd s ooL von o t _ l n t i s o rmtphei s R L (a oc - i t c uf e i q nu et r e mt m Skta egV r A e ost sf y s orerP r RCaas FbtaR t o a r g OI e m.. p eo t N1 2 L l!

                   '                                '                                      i!l      !l     ll     !j1  4j)ll

. $.ggEa[ygap . . E> Ia EEw >Bta $iE IO5". < , *

                                                                  ;         g8 in l                       '
  • w 0-1$w 3

c t s . n e t m d _ o ' m o C 8 2 ' 0 _ 0- _ P A n ae . i seh w savTo yh . lei sf i a rl Abnyt a o . pi C s ioi l r n . _ 6 swOl tcs ps y e sLea

                                                                          )

4 8 t s w. ae v 2 _ 0-7 i dl vee hl d eu r u 1 2 eat fi op rt P! P Y

               -                mvytoir        i edl   vnn y      .

g O, mae e- n 1 V- ir u pd o a go7 oi 7 l wsedgkl l

                                                  =

J TA , ceno2 NC e h ipl yly o w si c 1, y u. re6 _Y. EO ML T. yev apak

                                 ,           l l

er u sd go2 ssic 7 1, N Ho : mit a an se l t pe c t 5 Cm Ae a nml re9a7 e vg e unid >s u2 t ri q 1, . TM nesbt3 e s ie _ T At o mo neth s o g1 0e a7 d 4 r e ncloni e3 so1 ur 2 s n eiayc s t l t nhEsf D a e nnen ceOrdW o t r .pe p s o oieso pcnde c nsi C r e t f cl 3keur u e i i v R s q :e RFdav2toeca/aoeR t r e e f i v o s l t - at s 5 r. ni i o s e eo r,p vy 5, t yv i i aaiAm wv ipaCu t rt d i nsvOmo l r eLnt i e u teles a nBin i t n iw dte ec oLmer f m n _ o c m wso af na on i t

                     -        i mdeno nt         oio  t n

. s n t s ml r aaicc it .  : e r e p pid w emeA l _ i t o uf qn i rC c e oy iys u pO A RCdaatysoL r r t o a 8 e g a r e p eom. - P OI N3 t

s p u s g c B nR iH%i L) k.- ht g e t d a r S m p;- t n .f/ oN me i wa h n rRn m a go o) t e mteiD e s sB y d r u ec g n e r 3 k gewWt i r t gRl u8 a6 1 i n l pl 7 e a u 4 n a/ n mn8coa h c d) e 2 mt oinHi4 i ni t aRinS t o ct 32 sPs)spFncS r n oa1 F 4 o iP pigo/ r e CAe1n/pO dO, r t

                                                                                                                          ,                           / p n

8 Qm( O8 de f8 orD 1 4 1

                                                                                                                                       %uar            se r g o         o ye

(. 2 1 i A n A 2P1 u n. es 0 4 tac 2. 4 309 439B 0 f di b n OS 5 ed mp o P Ai O A / AO5( 5.0 M2 i s e eA oo AOocLO( 1 S e SO@ci e r v pa t r s ee r n n n g hA t t. g g pd g i i ueJ i sd oC e i n s eye s t r i ;. n; d a en bOe f t n n e t i e r d ad dt af iLh e u ns e .p l r e s, t p. d epy ar d . . h n ta a e m.hw ns hr Se l i . t oe t r o ht toso e n o oi l eu or ea t t

                                                                            .fbw o                         ei t          desect h atn os x l

i t t sd 6 8 i i f c t oFomx r tnsA euQ t d a nl u u c mndi r on s s c oi e y t en a m g, n 0 f u h 0a F. tCo i 6 h po nMlmmd r e s n d a . ipa e e l a en 7 si nw1 t rptn a t e cRfne a ,C or mn ut a s ptut r m ci ni n t pi 9 me7 i t - st n yl yaa u spD csi nuiqnn aa Y- t eim6 l pme eo h r i el eoi s auB gR 0 d a ni s nt _ 1V mdtekt gic al d vio i a vt e1 r er ai nhtai n vk u u ed e c r eri 3

                                                                                           /

nnH nre n 1 cf r n efrAC a nmsl e r e ea _ ii i t t d l deu2q1 bmd e enRiubcyo d) l EO moxt cor o,hrl uir i l wor y e ser et ne mu r et nisd e h s i4 s2 ML Hs pat e c ndb gue glyowa n1 Ho : S n pC e rp P o

t. I et
                                                                          -           l               nd ae    u  mal n      apet                       otem.o2o r

o c P y. NieaOb i

e. y.api un3 coh c c at iov t l spwWl ep s OE Cm t p d vLot o . r t

doMisa n l i v nu r _ Ae TM t i rs sdei6hP a

                                                                                          ,D5 c                e io                    os               n w s, 5 p Px e

a os a se i - h eO+ t t e eeE1 u s. b pr et n s s ps a T o moban ucaee o k dp )m e cg / ss e pr At e pt u mb ngDeaN1 r ntuny n d el ol l Dsi c a s pc yeh t i mt sD(f Er yoeel a n w0 l 0 s nmabta xi r *I 4b o e ,

                                                                                                                                                                              .i, s                                   gi0 nfioob er
                                                                                                                                                              )

_ n s n pu4r r o l o n e8 o r a s3 i g8 ueps m pie wiuroHp g0 s mB mg9 i p oS p( an9 t s t pt t - s s r p @i p e l s m w eri nmel d y o n5 areRrRoi3 a n1 aa/ r I R e m.a e o u .s uo piao R C p ;c b pf Ot rfb l r an s a eh0 c arc 5 ORoRac( a pH pH ppoO e rl oE ec4 p o y O rb _ r r f o f o - d e e-A d eHHy RR e nC1 o0C r O e nRRl l a l nL l eni ae ha k i l t t i t a n ai tn oni nt t a n - t eor r e r t e op)s s i p d oh eb e u pt oen ps em ot t n f o og fotouu np otns r a i t n i n o t co aintol s t comh t o _ c pcwo t ai pc1 0 b t ma op ma ht _ s n o t sirl e s of e t s ir s s ap t t o (a i g wn u s a yt q sere ar u er i i t c n s pua e min nn A e s p pi s r RAoS m Aopt u r r t o 9 a r e p eom. , O iN4 t 5

Torus Temperature and Pressure Response to Large Break LOCA and MSLB Accident Scenari:s Page G ' VYC.1628 Rev. 0 Appendix B

m. .
     }

AdacS w f 2 MEMORANDUM 0~Mfes) YANKEE ATOMIC-BOLTON (* " Dare November 10.1997 l Ta 8 6 r Ms Group # LOCA-VY97-OS4 os6 { W.C.# 1 From C.D.F=ea R-- - =; R&,M For Innuts and A=en.. tr.: to _ LM.S.# ~

  • Subject _ the Or *r.: tht=*.-mr.t Analvnis Mie #

ma.c un_r ~ =<.caur r ,4me a .sc- r c mw ra sa. l r,. c.aas. .- <a ss passan.a 7'.t. Feest niora M. ret Mvunweanor es. Her os.Ms. Stitxt. REFERENCES 1. LOCA-VY 97-035," Request for Assistance'on Tofus Tr f-eture and ECCS Pump NPSH Calculation Inputs," dated June 2,1997 l ' l 2. LOCA-VY 97-054, " Solicitation of Comments for ECCS Pump NPSH Calculation , I Inputs," dated October 7,1997

SUMMARY

    )
 >              During various meetings and discussions it has been identified that several inputs to the c post-IDCA torus %.6 4 calculation need additional clarification or review to ensure g -

Le accuracy and reduce uncertainty. To date, NED has been unable to either resolve the l' l surrounding several of the inputs or obtain accurate valuesdfor other inputs. This memorm , provides a summary of the current support needs g@ed to limit thie = :-- --i l ons j torus.tensperature analysis. Psevious support needs were idatiMM in Ref.1. This  : mem includes those support needs as well as expanded requirements as a result of the on-g effort. -

                *Ihc =*=n hd list ofinppt and assumption support noods id=dhd herein are specific to the Whilemanywillalsobeapplicabletothefollow-ontorus'fr+-- .

breakLOCAmhdadan it mt-1. dan (smallbecakLOCA,smallsteamlinebreak,SRV; =='*. AWivRtrans en s, etc) it is anticipated ithat addit' onal support will be sequired for these follow-on impact of the various inpets and assumptions has been identified in ReE 2. In order to support the timely completion of the LBLOCA torus =rm : e analysis, this l information is i@d by 30 November 1997. l' \ Ubbh3 . I C.D.Fago,Nudear Engineer D , . , LOCA Analysis Group U"M[~@J N atej* Nuclear Engineering Department ., , g3

       ..                                                                                                                                                                             l
                                                                                                                                   \ .V...

e*

                                                                                                                                          'W *" ^ ' ....r ~~~            -

l

                                                                                                                                     ,;Q& Y      , , :$ {. i t ,

1 l l l - --- -- _ _ - _ _ _ _ _ _ _ _ _ _ _ - _ _ _ _ _ _ _ __

Torus Temperature and Pressure Response to Large Break LOCA and MSLB Accident Scenarios Page 5d VYC-1628 Rev.0 Appendix B 1 November 10,1997 ()., Page 2

                                .             oc:                      J. R. Chrpman   M. P. IeFrancois                       K. R., Rousseau l

P. A.Bergeron W. S. Yeung l R. K. Sundaram M. A. Sironen - J; Lynch E.J.Betti R. G. January G. Hengerle J.H.('mHapan I, i 1 l r

       ')

i l

  • e 1

i i l s t l ::c - e t

                                                                                ..                                                                                     g

Torus Temperature and Pressure Response to Large Break LOCA and MSLB Accident Scennios Page M VYC 1628 Rev.0 , Appendix B

  ;A I.

ovember 10,1997

 - 5 ,, .                  88* I ATTACHMENT                                                                                               :. n  t

{ sv.e== r_. _.m . 1

1. Conthm that tbc .. I- $

RHR and CS flow rates, 7400 gpm and 4600 gym, respecdvely, tdaariflad in ==arnar=ndn= VYS 98/97, dated 28 August 1997, " Evaluation of Maximum @" Flows forBOCS StrainerRg':=-  :", are adequate as design input values to a WB-103 calW~- 2. Current narninal RHR flow through an RHR imit e- '=ger with one RHR pump operating is 7000 spm. Given that the operators use instrmnanrarian to addeve this flow and tha . the operators' instructions are to limit flow to less than 7000 spm through an RHR heat ^ ., exchanger, an mt.:-y band around the nominal flow rate is required. 3. Current namin=I RHRSW flow through an RHR beat =L= seris 2700 gpm. BMO 97-27 Rev. I states that the uncertainty band around the nominal value could be 4/- 150 gym. T ,

 ;;                 temperstme results are highly sensitive to RHRSW flow and previous diman== tan has identified

,;g.< the possibility ofensuring that the low analytical value of2700 gpm is supported by proced g A naminal RHRSW flow and an miaay band around the nominal is required. The low I uncertainty value should include naneidaration ofplant procedures that limit RHRSW flow to less than the namin=1 value.

4. t "Ibe Tdataal S;=!"- =' -, torus water volume limit is 68,000 ft'. Torus *==.aa=*=e
                - results me highly sensitive to initial water mQi. It is proposed to have an analytical limit                 !
                = of67,500 ft3 ofwater volume. It is ragw that this analyticellimit be reviewed in light o issues raised in BMO 9741 and the cunent seapom* t projects to ensee the analytical limit ca sW i
               . 5.
                              *(be analytical limit forinitial torus **=p==*== is proposed to be 90T.      d     It is M , / J that this analyticel limit be seviewed in light ofthe issues raised by G. Hengerle and the current setpoint projects to ensure the analyticel limit can be supported.

l 6.~ > It is anticipated that the operators will be instructed to throttle core spray flow to a I specified flow range at some time after tig LOCA. In support of potential operator actions requimdte ensure the core spraypumps have adequate NPSH, the mQ band for core spray

               - flowin the range of3000 to 4000 gpm based on the =' sa: # -Gon the operators would use to i , ,. . .                   ,

j verify post-accident core spray flow is reqd4 's " l 7.

                            'Ibe design service waterinlet %. Luc is 857(VYS 55/91," Review ofService Water Design Temperatwe," dated 3/27/91). Because this is the design temperature; it will be taken as 3
                                    .                                                                                                   l I

I Torus Temperature and Pressure Response 12 Large Break LOCA and MSLB Accid:nt Scen:rios 'PJge N/ VYC-1628 Rev. 0

         .                                                                                                                              Appendix B
 .s, .
     .: g
 .y Novembe 10,1997

. L.. . , Page 4 the analytical value. It is requested that this analytical lirait be reviewed against plant procedures and the setpoint project to ensure that the appropriate uncertainty is applied. 8. He flow split between the RHR heat --ma-er and the bypass is being calenI= tad by NED. To assist in this analysis, the normal position (i.e.100% open, 85% open, etc.) needs to be confinned. '

                         .        e O
                                                                                                                                                                   ' O
 .,e,                                                                                                                              *
 +

9 O 4 l O L e i .. ( l l I I

r Tes Tanperature and Pressure Response to Large Break LOCA and MSLB Accident Scenctios Pa 5YL

                 ! VYC-1628 Rev. O
             ~

o n. j'

  • November 10,1997 .

Page5 - b.! OpemtorActions assumptions need to be confinned is. order to valid

                                                                                                                                          'these 1.

l is aligned fortoruscooling. Operator action is assumed such that ten

                                                                                                                                                               )

2.' .

                                                                                                                                                                )

directed to cooldown theplant at a rate not to ex ,

                                                                                                                                   , normal
                                                                                                                                                                \

{' i actual

                      -is  required forthetarget analyses.     (and expected achievable) cooldo                                              .      e eve 3.

i-Operator action is taken to initiate drywell and wetwell spray afte i . at the direction of OE 3103TD,W/r-12 and DW/P-8). This action is immediately as the faster response time is conservative e enniaing the for the purpose . j of the transient. Again, this is conservative with t , r ' y~ i 1he following additional operator actions are currently s s andare assumed in , being ** relative to theirimpact on the analysis. "Ibey are and included her ' l l'

                   . as a " heads-up" to the potend=1 need for pla the analysis.           -                                                                                                                   i
                  - 4.

i assumed action, as well as the targetor flow rate, action will be assumed for no sooner than ten minutesaepost-LOCA an to mavi=i a the assumed time for this operator action.

 ;                5.                                                                          ;

operating. The assumed time for this action will be i

                - (ten minutespost-LOCA).

4

                                                                                            .                                                                   j
. l
                                  .                                                                                                                    i

Torus T:mperature and Pressure R:sponse 13 Large Bre:k LOCA and MSLB Accident Scenarios Page /c M VYC-1628 Rev. 0 Appendix J1 FOR INFORMATION ONLY - UNVERIFIED APPENDfX J: CONDENSATE AND FEroWATER SYSTEh! AVAILABILITY NOTE: THIS APPENDIX HAS NOT BEEN INDEPENDENTLY VERIFIED AND IS INCLUDED FOR INFORMATION ONLY. Table of Contents Pace J1. Objecti ve and Background . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Jl ...................

12. Evalu at io n A pproa ch . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .Jl. . . . . . . . . . . . . . . . . . . .

J3. Assumptions and Inputs . . . . . . . . . . . . .. . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .J2. . ... . . . . . . .. J4. JS Results....................................................................................................... J2 Co n cl u sio n s . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .33. . . 16. References................................................................................................. ........... ............... J3 J1. Obiective and Backcround The objective for this evaluation is to identify the impact of offsite power availability and loss, together with several different LOCA ECCS single failure assumptions, upon the condensate and feedwater system capebility to inject coolant during LOCAs. Figure J l shows a schematic diagram of the condensate and feedwater syste.ns. This evaluation supports VYC-1628, Rev. O, Suppression Pool Temperature for ECCS NPSH. A previous suppression pool temperature analysis, contained in VYC-1290, Rev. 2 (Ref. Ji], identified that VY feedwater fluid mass and energy, which might be transponed to the torus during LOCA events, can increase the peak suppression pool temperature. His is particularly significant if(a) offsite power is available (OPA) for the reactor feed and/or condensate pumps and their support equipment,(b) single failure assumptions (SFA) allow the feedwater and support equipment to be available, and (c) the operators are directed to continue feedwater injection

     ~      by the emergency procedures. If a loss of offsite power (LOOP) occurs, then the additional feedwater mass and energy added to the RPV is limited to that which could enter by boiling or flashing during the late blowdown, refill and long term cooling phases of the postulated LOCA cases. Bus, this evaluation is focused on identifying LOCA events which tend to maximize the transport of hot feedwater mass and energy into the RPV while minimizing transient thermal energy removal from the suppression pool by the RHR HXs.

His evaluation draws upon information in VYC-1511, Rev. O, VY ECCS Single Failure Evaluation For LOCA, [Ref. J2] to account for ECCS single failure effects on the feedwater system. His is particularly significant for the cascading efrects and results of a DC-1 and a DC-2 single failure assumption with offsite power available. J2. Evaluation Annroach J2.1 ne scope of this evaluation applies to the period from the LOCA initiation until after the peak suppression pool temperature is reached and the pool temperature is continuously trending downward toward ambient conditions. Preliminary analyses indicate this period can range between about 2 to 6 hours. J2.2 Consider that all relevant power production systems and engineered safeti systems are in their normal operating mode or normal standby alignment per plant operating procedures for rated power. J2.3 Identify major equipment required for the reactor feedwater system to function at rated power. Identify their l dependency (s) on support equipment (e.g. AC and DC power, component cooling, lube oil, station air systems). ,ldentify the power supplies for the major feedwater system and support system equipment. J2.4 Evaluate feedwater and support system equipment availability for the categories of OPA and LOOP for cases where the assumed LOCA single failure does not affect (DNA) the feedwater and support , system availability. J2.5 Evaluate feedwater and support system equipment availability for the cases of a DC-1 or a DC-2 single ! failure coincident with OPA and with LOOP. ne other LOCA ECCS single failures evaluated in VYC-1511 1 l do not affect the availability of the feedwater and its support systems. L

Torus Temperature and Pressure Response to Largs Break LOCA and MSLB Accident Sctnarios P:ge / ON VYC-1628 Rev. 0 Appendix J2 FORINFORMATION ONLY - UNVERIFIED J2.6 Summarize the reactor feedwater system availability to inject fluid into the RPV for the six cases evaluated above (OPA/DNA, OPA/DC-1, OPA/DC 2; LOOP /DNA, LOOP /DC-1, LOOP /DC-2). J3. Assumptions and innuts J3.1 Initially, the station auxiliary load is powered by the Main Generator and Unit Auxiliary Transfonner T-2 [Ref. J7]. He LOCA rapidly causes a Hi DW Pressure signal that leads to a Scram and Turbine Trip. He signal also actuates an automatic fast transfer to the Startup Transfonners. Ifno offsite power is available, then the transfer is blocked. De LOOP is sensed on Buses 3 and 4, and the emergency diesci generators start [Ref. J8]. . J3.2 De evaluation assumes that any postulated single failure assumption and/or any loss of offsite power are concurrent with the initiation of the postulated LOCA event. If offsite power is available, it continues to be available throughout the LOCA event. Likewise, if c'fsite power is lost, it continues to be lost throughout the LOCA event. J3.3 He DC 1 and DC-2 single failure assumptions are identical to those two assumptions devel ped and used in the VY ECCS Single Failure Evaluation for LOCA, and described in Attachments B1 and B2 of that calculation [Ref. J2]. For the Distribution Panel DC-1 SFA, power is lost to the 4KV Bus I since the failure mode assumes unsuccessful fast transfer of power from the Unit Auxiliary Transfonner T 2 to Startup Transformer T-3A. No credit is taken for Battery Charger BCl 1 A nor the Swing Charger CAB. However, Ref. J16 directs the operators to attempt a transfer ofcontrol power for selected equipment including Bus 1 to DC-2. Likewise, for the Distribution Panel DC-2 SFA, power is lost to the 4KV Bus 2 since the failure mode assumes unsuccessful fast transfer ofpower from the Unit Auxiliary Transformer T-2 to Startup Transformer T-3B. No credit is taken for Battery Charger BCl-1B nor the Swing Charger CAB. . However, Ref. J17 directs the operators to immediately attempt a transfer of DC-3A to DC-1. 33.4 A review of the other LOCA ECCS single failures (besides those in A3.3 above) evaluated in Ref. J2 shows that they do not affect (DNA) the reactor feedwater system's availability to inject fluid mass and energy into the RPV. These other single failures include UPS-I A, UPS-1B, RRU 7, RRU-8. They also include a failure modes and effects assessment ofindividual remote operated valves, pumps, active check valves in the ECCS. 13.5 Short terni refers to the early accident period that includes RPV blowdown, lower plenum refill, and core reflood to re-establish core cooling. From a design basis LOCA analysis view, this occurs in approximately the first 10 minutes of the event. No operator actions are assumed to be initiated during this period (first 10 minutes) other than to diagnose, monitor, and verify that early automatic actions have occurred. J3.6 Long term refers to the post accident period including sustained core cooling with fuel clad temperatures near or below Tsat associated with th. RPV pressure. His occurs after core cooling is re-established in the short tenn and enntinaes until post accident recovery is initiated. J3.7 No credit is given for operator actions that would repair failed equipment during the LOCA events. J3.8 Background infonnation for Tables J1 and J2 was guided by the excellent VY system summaries, assembled j by Richard Turcotte and Kevin Burns, in Reference J18. His is especially true for information used from Section 3.2.30: Feedwater Pump Trip, and Section 3.2.25: TBCCW System of Ref. J18. In addition, the VY FSAR and many current Control Wiring Diagrams (CWDs) were used as information sources). l i I l I k i

1 Torus Temperature and Pressure Response to Large Bre~ k LOCA cnd MSLB Accident Sc:nrrios Page /o70 VYC-1628 Rev. 0 Appendix J3 FOR INFORMATION ONLY - UNVERIFIED j J4. Results l he reactor feedwater and condensate systems are described in VY FSAR Section 11.8, Condensate and Reactor Feedwater Systems [Ref. J3]. The major equipment for these systems is identified on the P&lD," Flow Diagram - Condensate, Feedwater & Air Evacuation Systems [Ref. J4]. The condensate and feedwater system pump and valve lineups for normal operations are described in OP-2170 Rev.17 [J5] and OP-2172 Rev. 20 [J6], respectively. The feedwater and condensate active components together with their support equipment are summarized in Table JI. References for their power supplies and support systems are tabulated in the right hand column of Table J1. f During normal plant operation, the electrict! power for most of the active components (e.g. pumps, valves) in the reactor feedwater and condensate systems is provided by the 4KV Bus 1 or 4KV Bus 2. The electrical supply for these buses is described in FSAR Section 8.1: Station Electrical Systems [J7] The electrical loads for these buses is described in FSAR Section 8.4: Station Auxiliary Power System [J8]. Note that FSAR Figure 8.4-2," Balance of l Plant 4160 Volt Auxiliary One Line Diagram" provides an overview of these loads. More detailed electrical load information for Bus 1 and Bus 2 is provided on the Control Wire Diagrams (CWDs) cited in the summary on the next page. This summary also identifies the lower tier electrical equipment that supplies power to so_me active components in the condensate and feedwater systems. For LOCAs with Offsite Power Available (OPA), electrical power is assumed to be continuously available to the 4KV Bus I and 4KV Bus 2, subject to the DC-1 or DC-2 single failure assumptions discussed in Item J3.3 above. For LOCAs with Loss of Offsite Power (LOOP), electrical power is los to both the 4KV Bus 1 and 4KV Bus 2, and therefore their lower tier loads. Emergency diesel power is subsequently supplied to 4KV Bus 3 and 4KV Bus 4. 4KV Bus 1 (CWD B-191301. Sh. 301/R7 & 302/R8) 4KV Bus 2 (CWD B-191301. Sh. 304/RI & 305/R6) Circulating Water Pump PS-1 A Circulating Water Pump PS-1B Condensate Pump P2-1 A Circulating Water Pump PS-lC Reactor Feed Pump PI 1 A Condensate Pump P2-1B Reactor Feed Pump PI-1B Condensate Pump P2-IC Station Service Transformer T6-1 A Station Service Transformer T71 A 4SOV SWGR Bus 6 - 480V SWGR Bus 7

                                     -      MCC-6B, MCC-68                                       -  MCC-7B, MCC-7C Station Service Transformer Til-1 A 480V SWGR Bus 11 MCC-10B Table J2 summarizes the condensate and feedwater support system availability states as a consequence of a postulated LOCA, offsite power availability, and ECCS single failure assumption. This table is obtained by applying the assumptions and inputs from Section J3, together with support system descriptions in References J9 to J14, to the condensate and feedwater power supplies and support equipment.

Table J3 summarizes the condensate and feedwater equipment availability states as a consequence of a postulated LOCA, offsite power availability, and the ECCS single failure assumption. This table is obtained by applying the availability =tates identified in Table J2 to the appropriate equipment listed in Table 11. Table J4 provides a simplification of Table J3, where only the dominant, active components are listed. t 1 ( l

Torus Temperature and Pressure Response 13 Large Break LOCA cnd MSLB Accident Scenarios Page / > 7/ VYC-1628 Rev. 0 Appendix J4 FORINFORMATION ONLY- UNVERIFIED J5. Conclusions Table J4 provides a summary of the Condensate and Feedwater System availability to inject for the cases evaluated. These cases are summarized below in the order ordecreasing condensate injection capability for a LBLOCA, i.e.: Offsite Power Available with a DNA, or DC-1, or DC-2 single failure assumed. Loss of Offsite Power with a DNA, or DC-1, or DC-2 single failure assumed. J5.1 OPA/DNA: All equipment is initially available. His includes 3 condensate pumps,2 feedwater pumps, the FWCS, FWRVs, and the RFP Hi Level Trip. All bypass valves for recire to the condenser are closed. HPCI is also available, and will affect level for small and intermediate breaks. Injection will continue, subject to the FWCS, operator actions, and the available inventory in the reactor feedwater system and main condenser. CST and the Hotwell Makeup valves are available. During a I BLOCA, the feedwater pumps might accelerate to their runout condition. It is not clear at this

                                                                                                                          )

time whether this could lead to a RFP Low Suction Pressure : rip. If so, the RFPs would trip off; injection l would resume by the 3 condensate pumps once the RPV pressure decreased below about 400 Dsia. { J5.2 OPA/DC-1: Two condensate pumps are initially available to inject to the RPV once the RPV pressure decreases below about 400 psia. Their flow will be diminished by the failed open condensate bypass valve. { All 3 feedwster pumps have failed. The feedwater min flow bypass valves for recire to the condenser are { failed closed. HPCI is not available. The Hi Level Trip for the RFP has failed but is of no consequence since they have failed. The FWCS and FWRVs are available for reactor feedwater control. Hotwell emergency l i makeup is available. l J5.3 OPA/DC-2: One condensate pump and 2 feedwater pumps are initially available to inject to the RPV. All bypass valves for recire to the condenser are failed open. HPCI is available. The Hi Level Trip for HPCI and the RFPs has failed and is inoperable. The FWCS and FWRVs are available for reactor feedwater control. FSAR Section 11.8.3 describes the case where 2 condensate pumps can support 2 feedwater pumps. However, I condensate pump, together with its bypass to the MCH, will not be sufficient to provide the 150 psig suction pressure for the two feed pumps for a LBLOCA where the RPV depressurizes rapidly. One condensate pump provides about 3.49 Mibm/hr at their runout condition, i.e.130% of design flow. Two feedwater pumps have a design flow of 7.0 Mlbm/hr. Thus, the two reactor feed pumps will trip on Low Suction Pressure. Injection would resume by the Icondensate pump after the RPV pressure decreased below about 400 psia. Some flow will bypass through the failed open condensate and feedwater bypass valves to recirc to the main condenser. 1 1 J5.4 LOOP with a DNA, or DC-1, or DC-2: All three condensate pumps and all three feedwater pumps fail. Flow l Control Valve FCV-102-4 has failed open so there exists a path for condensate fluid to return to the relatively ' cold main condensers by drainage or natural circulation if the head differences support this flow. The three Feedwater Control vales fail as is, then drift open in a matter of minutes based upon discussions with VY Project and Plant personnel. The RFP Minimum Flow Bypass Valve for each reactor feedwater pump remains normally closed for the DNA and DC-1 single failures. This blocks a return path to the main  ; condenser for those sceaarios. However, each of these valves Fails Open for the DC-2 single failure, and l allows a drain path to the main condenser for that scenario. l

Torus Temperature and Pressure Response to Large Break LOCA and MSLE Accident Scen:rios Prge /u72 VYC-1628 Rev.0 Appendix JS FORINFORMATION ONLY-UNVERIFIED

36. References JI.

Calculation VYC-1290, Rev. 2: Vennont Yankee Post-LOCA Torus Temperature and RHR Heat Exchanger Evaluation,7/1I/96

12. Calculation VYC 1511, Rev. 0: VY ECCS Single Failure Evaluation for LOCA,2/7/97 J3. VYNPS FSAR, Section 11.8, Condensate and Reactor Feedwater Systems, Rev.13 J4. Drawing G191157 Rev. 44, Sh.1: VYNPS Flow Diagram-Condensate, Feedwater & Air Evacuation
         - Systems J5. Procedure RP 2170, Condensate System, Rev.17                                                                                                                3 J6. - Procedure RP 2172, Feedwater System, Rev. 20                                                                                                                 {

J7. VYNPS FSAR, Section 8.1: Station Electrical Systems, Rev.12 { ' J8. VYNPS FSAR, Section 8.4: Station Auxiliary Power System, Rev.12 J9. VYNPS FSAR, Section 8.4.5.1: 4160 V Switchgent, Rev.12 J10. VYNPS FSAR, Section 8.4.5.2: 480 V Buses, Rev.12 ( Jl1. VYNPS FSAR, Section 8.4.5.3: 120/240 V instnamentation Distribut' ion System, Rev.12 J12. VYNPS FSAR, Section 8.6: 125 V DC System, Rev.12 J13. VYNPS FSAR Section 10.10: Turbine Building Closed Cooling Water System (TBCCW) , Rev.13 114. VYNPS FSAR Section 10.14: Station Instrument and Service Air Systems, Rev.13 J15. Letter, WP Murphy (VYNPC) to USNRC, Vermont Yankee Response to NRC Generic Letter 88-14: Instrument Air Supply System Problems Affecting Safety-Related Equipment", BVY 89-17, dated February s 16,1989 J16. Procedure ON-3159: Loss of DC 1, Rev.1, Issued 5/6/97 317. Procedure ON-3159: Loss of DC-1 and DC-3, Rev.1, Issued S/6/97 318. Vennont Yankee Individual Plant Examination (VY IPE), December 1993, VYNPS, Vernon VT 4 I i 4

Torus Temperature and Pressure Response to Large Break LOCA and MSLB Accident Sc;narios Pzge / 4 .5 VYC-1628 Rev. O Appendix J6 FORINFORMATION ONLY- UNVERIFIED TABI.E JI: FEEDWATER & CONDENSATE EottienttxT & PowEn Sttret its Component Power &

References:

VYNPS SYSTEM Name Sunnort Sources Control Wirine Diacrams B-191301 Condensate System -64 MCH A Outlet Valve MOV SB64-1 A MCC 7C Sheet 514, Rev. 7 MCH A Outlet Valve MOV SB64-IB MCC-7C Sheet SIS, Rev. 5 MCH-B Outlet Valve MOV SB64 IC MCC-7C Sheet 516, Rev. 6 MCH-B Outlet Valve MOV SB64 ID MCC-7C Sheet $17, Rev. 6 Condensate Pump P2-1A Bus-1, DC-1, TBCCW Sheet 530, Rev. 6 Condensate Pump P21B Bus-2, DC-2, TBCCW Sheet $31. Rev. 6 Condensate Pump P2-1C Bus-2, DC-2, TBCCW Sheet 532, Rev. 6 Cond Recire Valve AO FCV-102-4 Station Air, Notes 1,2 Sheet 530, Rev. 6, Dwg. 5920-4719 Cond Recire Sol. Valve FSO-102-4 120V DC Vital; Note 1 Sheet 530, Rev. 6 LPH Train A Inlet MOV64-23 MCC-6B Sheet 533, Rev. 5 LPH Train B inlet MOV64-21 MCC-6B Sheet $34, Rev. 4 LPH Train A Outlet MOV64-25A MCC-6B Sheet 535, Rev. 4 LPH Train A Outlet MOV64 25B MCC-10B Sheet 536, Rev. 7 LP Heater Bypass MOV64-19 MCC-6B Sheet $37, Rev. 8 Hotwell Emerg M/U MOV64-31 MCC-7B Sheet $19, Rev. 8 Feedwater System - 63 RFP Lo Suct Press Trips PSL102-20A,B,C(pump motors) (Sheets $50,551,552) HPCI/RFP Hi LevelTrip RPV Hi Level DC-lC+DC-2C+DC 3A Sh. 550B Rev.4, Sh.1451 Rev.23, Sh.1455 Rev.18, Note 6 FWCS Power & Logic 120V AC Vital, DC-3A Sheets 500 Rev 17,501 Rev 13,550B Rev 4 Rxt Feed Pump PI 1 A Bus 1, DC-1, TBCCW Sheets 550 Revl7,550A Rev 6,550B Rev 4 Rxt Feed Pump PI-1B Bus 1, DC-1, TBCCW Sheet $51, Rev.14 Rxt Feed Pump PI-IC Bus-2, DC 2, TBCCW Sheet 552, Rev.17 RFP instruments Note 3 120V DC VitalInstr Sheets 550,550A,550B RFP Aux. Lube Pumps P57 IA,1B,1C MCC-6B, Note 4 Sheets 556,557,558, all Rev. 4 RFP Recire Valve FCV102 2A 125V DC 3A Sheet 559 Revl, Dwg Gl91372,1/5, Rev 40 RFP Recire Valve FCV102 2B 125V DC-3 A Sheet 559 Revl, Dwg Gl91372,1/5, Rev 40 RFP Recirc Valve FCV102-2C 125V DC-3A Sheet 559 Revl, Dwg Gl91372,1/5, Rev 40 Rxt Feed Pump Discharge MOV63-4A MCC-6B Sheet 553, Rev. 7 Rxt Feed Pump Discharge MOV63-4B MCC-6B Sheet 554, Rev. 7 Rxr Feed Pump Discharge MOV63-4C MCC-10B Sheet 555, Rev. 6 Fdw Reg Valve inlet-A MOV63-1I A MCC-78 Sheet 520. Rev. 6 Fdw Reg Valve Inlet-B MOV63-11B MCC-68 Sheet 521, Rev. 6 Fdw Reg Valve Inlet-C MOV6310 MCC-7B Sheet 522, Rev. 5 FWCS Valve (55%) AO FCV612A 120 AC Vital, Station Air Sheet 502, Rev. 8 FWCS Valve (55%) AO FCV6-12B 120 AC Vital, Station Air Sheet 502, Rev. 8 Fdw Reg Valve (10%) AO FCV6-13 120 AC Vital, Station Air Sheet 501, Rev.13 HPH inlet A MOV63-6A MCC-6B Sheet 509, Rev. 4 HPH Inlet B MOV63-6B MCC-10B Sheet $10, Rev. 6 HPH Outlet A MOV63-7A MCC-10B Sheet SI1, Rev. 4 ! HPH Outlet B MOV63-7B MCC-10B Sheet 512, Rev. 5 HPH Bypass Valve MOV63-5 MCC-10B Sheet $13 Rev. 8 Fdw Recire Valve MOV63-22A MCC-6B Sheet 545, Rev. 6 Fdw Recire Valve MOV63 22B MCC-10B Sheet 546, Rev. 6 FC = Flow Control, MO = Motor Operated, AOV = Air Operated, FWCS = Feedwater Control System Note 1: Fails open on loss of any Condensate Pump per CWD Sheet $30. Rev. 6, or loss of air. Note 2: Copes-Vulcan Valve. Serial D-600-16D, Direct Action Actuator with air to close/ spring to open. Note 3: Loss of 120V Vital Instrument Bus will cause Low Flow Instruments to trip all Reactor Feed Pumps. Note 4: RFP Aux. Lube Oil Pump must be running to complete startup circuit for a Rxt Feed Pump from standby. i L

I Torus Temperature and Pressure Response to Large Break LOCA and MSLB Accident Scenarios Page /C7Y VYC-1628 Rev. O Appendix 37 FOR INFORMATION ONLY- UNVERIFIED TAntiJ2: CONnENSATE & FErnwATER SUPPORTSYSTEkl AVAlt ARII.lTYSTATES OPA OPA OPA LOOP LOOP LOOP SUPPORTSYSTEnt / Sinele Failure > DNA DC-1 p_Q-1 DC-1 DC-2 DNA.

4 KV POWER IJ91 I

SWGR Bus I Avail Fail Avail Fail Fail Fail SWGR Bus 2 Avail Avail Fail Fail Fail Fail 480 V AC POWER fJ101 MCC-6B, MCC-68 (from 4KV Bus 1) Avail Fall Avail Fail Fail Fail MCC-10B (from 4KV Bus 1) Avail Fail Avail Fail Fail Fail MCC-7B, MCC-7C (from 4KV Bus 2) Avail Avail Fall Fail Fail Fail 120/240 V AC VITAt.UPS lJIII ~ 120/240 V AC Vital UPS Distr. Panel Avail Avail Avail Avail Avail Avail 120 V Instr. Subpanel VAC-A Avail Avail Avail Avail Avail Avail 120 V Instr. Subpanel VAC-B Avail Avail Avail Avail Avail Avail 125 V DC POWER fJI21 DC-1, DC-1C Avail Fail Avail Fail Avail Avail DC-2, DC-2C Avail Avail Fail Avail Fail Avail DC-3A (nonnally from DC-2) Avail Avail Fail' Avail Fail' Avail Note *: DC-3A is normally fed by DC-2 via DC-3; it will be manually transferred to DC-1 per ON-3160 if

  • possible.

TBCCW SYsTEnt fJ13.J181

   . Either Pump P58-1 A or P58 1B together with either TBCCW Heat Exchanger E22-I A or E22-1B can provide the support functions of Condensate Pump Motor Cooling and Reactor Feed Pump Seal Water & Lube Oil Cooling.

Therefore this function is provided subject to (1) manual restart ofone pump following LNP after emergency power , has been restored to the associated emergency bus (4KV Bus 3 or Bus 4), and (2) failure assumptions since the i support equipment for the Feedwater & Condensate Systems is classified as non nuclear safety. J 4 KV Bus 3. MCC-8C, Pump P58-I A Avail Fail Avail Fail Avail Avail 4 KV Bus 4, MCC-9C, Pump P58-1B Avail Avail Fail Avail Fail Avail TBCCW System Avail Avail Avail Avail Avail Avail STATION INSTRUhf ENT AND service AIR SYSTEntS IJ t d. JIS. JI81 Four, parallel air compressors discharge air to two separate receivers for the Instrument Air System and the Service Air System. The main power supply for each compressor is listed below. Control power for each compressor is supplied from the same MCC that supplies the motive power. Thus, this function is provided subject to (1) manual restart of at least one compressor following loss of normal power (LNP) after emergency power has been restored to the associated emergency bus (Bus 3 or Bus 4), (2) failure assumptions since the equipment for the Feedwater & Condensate Systems is classified as non-nuclear safety. 4 KV Bus 1, MCC-11B, Compressor Cl-ID Avail Fail Avail Fail Fail Fail 4 KV Bus 2, MCC-7B, Compressor Cl-lC Avail Avail Fail Fail Fail Fail 4 KV Bus 3, MCC-8C, Compressor Cl-1 A Avail Fail Avail Fail Avail Avail 4 KV Bus 4 MCC-9C, Compressor Cl-1B Avail Avail Fail Avail Fall Avail Station Instr. and Service Air Systems Avail Avail Avail Avail Avail Avail _ - _ _ _ _ _ - _ _ _ _ _ _ _ _ _ a

_ . _ _ _ _ _ _ _ _ - - - - - - _ - - - - - - - - - - - - - - - - - - ~ ~ ' Torus Temperatu're and Pressure Response to Large Break LOCA and MSLB Accident Sosnarios Page /Mf VYC-1628 Rev. 0 . Appendix 38 FORINFORMATION ONLY- UNVERIFIED TABLEJ3: CONDENSATE & FEEDWATER EotflPMENT AVAILABILITY STATES I L Component Power OPA OPA OPA

                                                  ' Condensate System - 64 Name                                                                                                               LOOP LOOP LOOP Source    DEA            DC-J.,                                     DC:1                RC,,J. RC-1 DNA,
                                                 . MCH-A Outlet Valve         MOVSB641A MCC-7C                  NO             Avail FO                                                       FO       FO      FO MCH-A Outlet Valve         MOV SB64-1B MCC-7C                NO.            Avail FO                                                       FO       FO      FO MCH-B Outlet Valve         MOV SB64-lC MCC-7C                NO             Avail 'FO                                                      FO       FO      FO MCH-B Outlet Valve         MOV SB64-ID ' MCC-7C              NO             Avail FO                                                       FO       FO      FO Cond Suction Header     30" C-3 L                                                   Condensate Pump            P2-1A                  4KV Bus 1 On              Fail                                      On                   Fail     Fail    Fail Condensate Pump            P2-1B                  4KV Bus-2 On              On                                        Fall                 Fail     Fail    Fail Condensate Pump            P2-IC                  4KV Bus-2 On              On                                        Fail                 Fail     Fail    Fail CP Discharge Header       24"C-8

[ Intercondenser(2) ' ~ ! Steam Packing Exhaust E24-IA Cond Demin System . L Condensate Line 24" C-18 l Cond.Recire to MCH FCV 102-4 Notes I,2 NC- FO FO FO FO FO l HotwellEmerg M/U MOV64 31 MCC-7B NC Avail FC , FC FC FC LPH Train A Inlet MOV64-23 MCC-6B NO FO Avail ' FO FO FO ! LP Heater ES 1A l Condensate Line 20" C-25 l LP Heater E4-IA l Condensate Line 20" C-27 LP Heater E3-1A-l LPH Train A Outlet MOV64-25A L MCC-6B NO FO Avail FO FO FO Condensate Line 20" C-29 LPH Train B Inlet MOV64-21 MCC-6B NO FO Avail FO FO FO LP Hester ES-1B Condensate Line 20" C-26 LP Heater E4-1B Condensate Line . 20" C-28 LP Hester E3-1B LPH Train A Outlet MOV64-25B MCC 10B NO- FO Avail FO FO FO Condensate Line 20" C-30 LP Hester Bypass MOV6419 MCC-6B NC FC Avail FC FC FC RF Pump Suction Header 24"C-30

                                               . Note I: Fails open on loss of any Condensate Pump per CWD Sheet $30, Rev. 6, or loss of air.

l Note 2: Copes Vulcan Valve, Serial D-600-16D, Direct Action Actuator with air to close/ spring to open. Note 5: Nonnal status positions, detennined from References A4, A5 and A6, are recorded for OPA/DNA. FC = Flow Control, MO = Motor Operated, AO = Air Operated, FWCS = Feedwater Control System FAl/DO = Fall As Is/ Drift Open, FC = Fail Closed FO = Fail Open, NO = Nonnal Open, NC = Normal Closed q l _ _ _ _ _ _ _ - _ _ _ _ _ _ _ _ _ - _ _ _ _ _ - _ _ _ _ _ _______-___ U

Torus Temperature and Pressure Response to Large Break LOCA and MSLB Accid:nt Sc;nari:s Page /OM VYC-1628 Rev. O Appendix J9 FOR INFORMATION ONLY -UNVERIFIED tam J3: CONorNs ATE & FrrowATrn EctfirMENT AVAti Antt ITYSTATES (CONTINUrn) Component Power OPA OPA OPA LOOP LOfir LOOP Feedwater System - 63 Name Source DNA DC-1 DC-2 DC-1 DC-2 DNA. HPCI/RFP Hi Level Trip DC-lC+DC-2C+DC-3A Avail Fail Fail Fail Fail Avail FWCS Power & Logic 120V AC Vital +DC-3A Avail Avail Fail

  • Avail Fail' Avail Rxr Feed Pump PI-1A 4KV Bus 1 On Fail On Fail Fail Fall RFP Min Flow Bypass FCV-102 2A Air, DC-3A NC NC FO' NC FO' NC Rxr Feed Pump Disch MOV63-4A MCC-6B NO FO Avail FO FO FO Rxr Feed Pump PI-1B 4KV Bus-1 On Fall On Fall Fail Fall RFP Min Flow Bypass FCV-102-2B Air, DC-3A NC NC FO* NC FO* NC Rxr Feed Pump Disch MOV63-4B MCC-6B NO FO Avail *:0 FO FO Rxr Feed Pump PI-IC, Note 4 4KV Bus-2 Standby Fail Fail Fail Fall Fail RFP Min Flow Bypass FCV 102 2C Air, DC-3 A NC NC FO' NC FO' NC Rxt Feed Pump Disch MOV63-4C MCC-10B NO FO Avail FO F6
                                                     ~                                                 FO RFP Discharge Header        24"FDW-1 FR Block Valve Inlet-A MOV63-11 A            MCC-7B           NO     Avail    FO       FO       FO      FO FWCS Valve (55%)            AO FCV6-12A AC Vital, air         NO     Avail    FAI/DO Avail      FAl/DO FAI/DO FR Block Valve Inlet-B MOV63-11B             MCC-68           NO     FO      Avail FO           FO      FO FWCS Valve (55%)            AO FCV6-12B AC Vital, air         NO     Avail    FAl/DO Avail      FAl/DO FAl/DO FR Block Valve Inlet-C MOV63-10              MCC-7B           NO     Avail    FO       FO       FO     FO FWCS Valve (10%)            AO  FCV6-13      AC    Vital, air NC     Avail   FAI/DO Avail       FAl/DO FAl/DO HPH Inlet Header            18" FDW-10 HPH Bypass Valve            MOV-63 5         MCC-10B          NC     FC      Avdl FC            FC     FC HPH Bypass to FDW           10" FDW-1i HPH Inlet A                 MOV63-6A         MCC-6B           NO     FO      Avail FO           FO     FO HPH Outlet A                MOV63-7A         MCC-10B          NO     FO      Avail FO           FO     FO HPH Outlet Train A          18" FDW-14 HPH Inlet B                 MOV63-6B         MCC-10B          NO     FO      Avail FO           FO     FO HPH Outlet B                MOV63-7B         MCC 10B          NO     FO      Avail    FO        FO     FO HPH Outlet Train B          18" FDW 15 FDW Line Cross-tie          10" FDW-11 FDW Train A to Rxt          16" FDW 14 FDW Train B to Rxt          16" FDW-15 Note 3: Loss of 120V Vital Instrument Bus will cause Low Flow Instmments to trip all Reactor Feed Pumps.

Note 4: RFP Aux. Lube Oil Pump must be running to complete startup circuit for a Rxr Feed Pump from standby. Note 5: Normal status positions, determined from References A4, AS and A6, are recorded for OPA/DNA. Note 8: DC-3A is normally fed by DC-2 via DC-3;it will be manually transferred to DC-1 per ON 3160 if possible. FC = Flow Control, MO = Motor Operated, AO = Air Operated, FWCS = Feedwater Control System FAl/DO = Fail As is/ Drift Open, FC = Fail Closed, FO = Fall Open, NO = Normal Open, NC = Normal Closed

Torus Temperature and Pressure Response to Largs Break LOCA and MSLB Accident Scenarios PJge /077 VYC-1628 Rev. 0 Appendix J10 .} FOR INFORMATION ONLY- UNVERIFIED I TABIE J4: SUkthf AltY OF CONDENSATE & FrrinvATER EOUIPhf ENT AVAILABILITY STATES l l Component Power OPA OPA OPA LOOP LOOP LOOP Condensate System - 64 Name Source s DNA P_Cd P_Cd P_Cd PS DNA. Condensate Pump P2-1A 4KV Bus-1 On Fail On Fail Fall Fail Condensate Pump P2-1B 4KV Bus-2 On On Fail Fail Fail Fail i Condensate Pump P2 IC 4KV Bus-2 On On Fail Fall Fall Fail Cond. Recire to MCH FCV-102-4 Notes t,2 NC FO FO FO FO FO Hotwell Emerg M/U MOV64 31 MCC-7B NC Avail FC FC FC FC LP Heater Bypass MOV64-19 MCC-fB NC FC Avail FC FC FC Component Power OPA OPA OPA LOOP LOOP LOOP Ecedwater System - 63 Name Source DNA PC-1 P_Cd DC-1 DC-2 DNA. HPC1/RFP HiLevelTrip DC-IC+DC-2C+DC-3A Avail Fail Fail Fail Fail Avail FWCS Power & Logic 120V AC Vital +DC-3A Avail Avail Fail' Avail Fail' Avail Rxr Feed Pump PI-1A 4KV Bus-1 On Fail On Fail Fall Fail RFP Min Flow Bypass FCV-102-2A Air, DC-3A NC NC FO' NC FO' NC Rxr Feed Pump PI-1B ' 4KV Bus-1 On Fail On Fail Fail Fall RFP Min Flow Bypass FCV-102 2B Air, DC-3A NC NC FO' NC FO*

                        ,                                                                                                   NC Rxt Feed Pump             PI-IC            4KV Bus-2       Standby Fail            Fail  Fail     Fail              Fail RFP Min Flow Bypass       FCV-102-2C       Air, DC-3A      NC      NC              FO'   NC       FO'               NC FWCS Valve (55%) -        AO FCV612A AC Vital, air         NO      Avail           FAl/DO Avail FAl/DO FAl/DO FWCS Valve (55%)          AO FCV612B AC Vital, air         NO      Avail FAl/DO Avail FAl/DO FA!/DO FWCS Valve (10%)          AO FCV6-13      AC Vital, air    NC      Avail- FAI/DO Avail FAl/DO FAl/DO HPH Bypass Valve          MOV-63-5        MCC 10B          NC      FC              Avail FC       FC                FC Note 1: Fails open on loss of any Condensate Pump per CWD Sheet 530, Rev. 6, or loss of air.

Note 2: Copes-Vulcan Valve, Serial D-600-16D, Direct Action Actuator with air to close/ spring to open. Note 5: Normal status positions, determined from References A4, A5 and A6, are recorded for OPA/DNA. Note 8:

  • DC-3A is normally fed by DC-2 via DC-3, but can be manually transferred by a switch to DC-1.
      - Acronyms:

MCH = Main Condenser Hotwell, M/U = Makeup, FWCS = Feedwater Control System FC = Fir iv Control, MO = Motor Operated, AO = Air Operated FAl/DO = Fail As Is/ Drift Open, FC = Fail Closed, FO = Fail Open NO = Normal Open, NC = Normal Closed

                                                                                                                                         \

( - - __- -- _ - - - - - - -

                                                                                                                          )

i Torus Temperature and Pressure Response ta Large Break LOCA and MSLB Accident Scenarios Page / o 28 VYC-1628 Rev. 0 l Appendix 311 FOR INFORMATION ONLY - UNVERIFIED Review Comments and Responses i j Review Comments from Carl Faco. 5/22/97: Cl. Clarify the failure mode of the Reactor Feed Pump (RFP) Hi Level Trip identified in Tables Al, A3 and A4. RI. Failure of DC-lC or DC-2C or DC-3A power prevents the RFP Hi Level Trip for all 3 pumps from occurring. I l I reviewed and verified the failure mode for the RFP Hi Level Trip with George Hengerle, VYP I&C. This trip, actually one for each RFP, is supported by the logic for the HPCI,a_nd RCIC level transmitters. Logic circuits are shown on CWD B 191301, Sheets $50B, Rev.4, Sh.1451, Rev. 23 and Sh.1455, Rev.10. Power for the logic circuits are provided by DC-1, DC 2 and DC-3A. Failure of any one power supply will cause the HPCI, RCIC and RFP trip functions on high reactor vessel level to not occur. More detailed information { below is taken from the VY IPE, Section 3.2.30: Feedwater Pump Trip on High Level (FT), prepared by R. T. { Turcotte and K. Burns [Ref.J18].

      "Feedwater Pumn Trin Description                                                                  ~

{ An illustration of the feedwater pump trip logic is shown on Figure 3.2.30A [Ref. Jl8]. The Feedwater Pump Trip (FT) on high reactor water level is designed to prevent overfilling the vessel should the Feedwater Control System malfunction. The FT on high level requires two-out-of-two logic from the HPCI and RCIC level transmitters. The bistable trip units send a trip signal when the reactor level reaches a height of > 177" above the top ofenriched fuel. The 177" FT setpoint is set at the same level as the turbine trip and reactor scram to avoid spurious trips. f t On a high-level output signal, the HPCI and RCIC sla"e output relays energize the FT relay. Once this relay is energized, its contacts close, energizing the trip coils in each of the three feedwater pump circuit breakers, immediately tripping the pumps. The FT signal, however, does not remain sealed in. When in " auto" mode, the previously running pumps may start agahi, iflevel subsequently decreases below the actuation setpoint within five seconds ofits initiation. Iflevel stays above the high level setpoint for greater than five seconds, the pumps will not restart and would have to be started manually if desired. Interfaces / Dependencies The dependency matrix for the feedwater pump trip logic is shown on Figure 3.2.30B [Ref. J18]. The FT depends on de power for level sensing, logic, and actuation. The 24 V de power from both Panel A and Panel B is required to power both the HPCI and RCIC level transmitters and trip units. These level transmitters are used in the ECCS actuation signal for low reactor level. The 125 V de power from both DC-IC and DC-2C is required to power the ECCS output relays. The 125 V de power from DC-3 A is also required to power the FT relay, while DC-1 and DC-2 supply power to the pump trip coils on FT relay energization." C2. Clarify the failure mode of the Feedwater Control System (FWCS) identified in Tables J1, J3 and J4. R2. The FWCS depends on the 120 VAC Vital and DC-3 A Buses. The 120VAC Vital Bus is considered a very reliable power supply because of redundant and diverse sources to power it. However, a DC-3A failure will de-energize power to the FWCS. This eliminates control signals to the FW Reg Valves. They " Fail As Is", due to air operated locking valves that attempt to hold them in their current position. Several sources (Chris Hansen, Jeff Chizever, George Hengerle, VYNPS Operator Van Bowman) have said the locking valves work for a short time (few minutes), but slowly lose their air so that the FWRVs drift open from ;he FAI position.

Torus Temperature and Pressure Response to Large Break LOCA and MSLB Accident Scenctios Page /u 'f VYC-1628 Rev. 0 Appendix J12 FOR INFORMATION ONLY - UNVERIFIED C3. Review the " Loss of DC Power" emergency procedures to determine possible operator responses. R3. I have reviewed ON 3159: Loss ofDC-l. Inote the following: a. Operator Actions, Note I reaffirms the feed pump high level trip is inoperable, b. Action 2 directs the operator to monitor reactor vessel 'evel closely and secure feed pumps as necessary. For the DC-1 failure (OPA and LOOP), the feed pumps have failed, so the action is consistent. c. Action 5, upon conditions to be satisfied, directs the operators to swap Bus I,3 and other bus control power to DC-2 per specified steps. Ifsuccessful, this would restore most all of the important functions. I have reviev.ed ON-3160: Loss of DC-2 and DC-3. I note the following: a. Auto Action 4 reaffirms that the reactor feed pump recire (b> pass) valves fail open. b. Operator Action 2 states " Direct an operator to transfer DC 3A to DC-1 immediately." This would restore failed equipment in Table J3 and J4 marked with an *, and commented on in Note 8. Review Comments (written and oral) from Liliane Schor. S/22/97: C4. Clarify the failure mode of the Resctor Feed Pump (RFP) Hi Level Trip identified in Tables AF. A3 and A4. R4. See response to C1 above. C5. The Reactor Vessel Hi Level Trip is normally available to trip HPCI on high level. Is it available for the DC-1 or the DC-2 failures? ,) RS. No, HPCI auto isolation on Hi Rxt Level requires both DC-1 and DC-2. Failure of DC-1 fails HPCI. Failure of DC-2 can fail HPCI to trip on Hi Rxt Level, and overfill. This could flood the steam line and fail the HPCI turbine. See Ref. A2, Calculation VYC-1511, Rev. 0: VY ECCS Single Failure Evaluation for LOCA. 2/6/97, pages B3 and B4 of 16 in Attachment B. C6. Clarify the failure mode of the Feedwater Control System (FWCS) identified in Tables Al, A3 and A4. I R6. See response to C2 above. i C7. Table J3 and the colored schematic diagrams for the VY Feedwater/ Condensate System show the available ' feedwater and condensate pumps, valves, bypass valves and flow paths for the six cases evaluated. Clarify the bypass flow paths for the six cases evaluated. For example, Case OPA/DC-1 shows that 2 condensate pumps and no feedwater pumps are available. Will some flow bypass through the condensate bypass valve (FCV-102-4) or through the feedwater bypass valves (FCV-102-2A,-2B,-2C)? Please add the colored schematic diagrams to the memo since they make it much easier and quicker to see the main flow paths. R7. Table J4 and Table R3 below (both simplified from Table A3) summarize the availability of the feedwater and condensate pumps, together with valves that control the main feedwater injection flow path and the bypass i flow paths. In all cases, the main flow path is available, subject to sufficient pressure to open check valves. I The colored schematic diagrams will be added to the memo since I agree it is easier to see the main flow paths. l C8. Please review the current reactor feed system model in the RELAPS input deck, and provide comments and recommendations on this model for LBLOCA and SBLOCA conditions. R8. The assumptions for feedwater injection are appropriately conservative for OPA/DNA, and may be overly conservative for the OPA/DC 2 case since no leakage is allowed for min flow bypass to the main condenser. 1

Torus Tsp.e e and Pressure Response to Largi Break LOCA and MSLB Accident Scenarios Page /CED VYC-1628 Rev.0 Appendix J13 FOR INFORMATION ONLY-UNVERIFIED l _ TABLk R3: SUnthlARYOFCONDENSATE & FEEDWATER EOt'IPhfENT AV { l-Component OPA OPA OPA SYSTEM LOOP LOOP LOOP Name DEA DC-], DC-1 Dfd DC-1 DNA. l Condensate System Condensate Pump P2 1A On Fail On' Fail Fail Fail

            . Condensate Pump           P2-1B                On      On        Fail

! Fail Fail Fait Condensate Pump P2-IC On On Fail Fail Fail Fail Cond. Recirc to MCH FCV 102 4 NC FO FO FO FO FO. l Feedwater System HPC1/RFP HiLevel Trip Avail Fail Fail ail Fail Avail

FWCS Power & Logic Avail ' Avail Fail * . Avail Fail
  • Avail

! Rxr Feed Pump PI-IA On Fail On Fail Fail Fail RxrFeed Pump PI-1B On Fall On Fall Fail fail Rxr Feed Pump PI-IC Standby Fail Fail .- Fail Fall Fall l RFP Min Flow Bypass FCV-102 2A NC NC FO* NC FO* NC RFP Min Flow Bypass FCV-102-2B _ NC NC FO' NC FO* NC l RFP Min Flow Bypass FCV-102-2C NC NC FO' NC FO* NC FWCS Valve (55%) AO FCV6-12A NO Avail FAl/DO Avail FAI/DO FA!/DO FWCS Valve (55%) AO FCV6-12B NO Avail FAl/DO Avail FAI/DO FA!/DO FWCS Valve (10%) AO FCV6-13 NC Avail FAI/DO Avail FAI/DO FAl/DO Acronyme; LAO = Air Operated valve DNA = Other ECCS Single Failures that "Do Not Afrect" Feedwater/ Condensate System availability FAl/DO = Fail As Is/ Drift Open FC = FailClosed

           , FCV = Flow Control Valve FO = Fail Open FWCS = Feedwater Control System LOOP = Loss ofOfTsite Power MCH = Main Condenser Hotwell MO = Motor Operated M/U = Makeup
          - NC = Normal closed NO = Nonnalopen OPA = Offsite Power Available l

4 e

      +

t

R W8 1 E K E g A - S A - M N I

                          -                                     V E                                                                                                                L D 6-N O E                           jW OWIA                                                                            L E

C W T O

  • N]

H s/ 1 I u , c n 8, S o n 3 e3 t 4 1

                                                                                                                                        -8 W

U C

                                                                                                                             ]            v o

A , 8 C ' P l 2 2; 2

                                                                               ;              y-Y,             Y                 W             E T                               i A
         '                                                            Ai             e                 cS                               -

t i- N w T V2P mb2 D 5 p P E

  • C
                                        ,                                                                                         7:                                              E A
                                                                ]r.             '

N O C A 5 C - 7g a i u 2e E S c RE J E Y ' w A ;'u 00 L V H* M Y:e s c K T n o w a

                                                                                                                                                                       . es EL t- W Y L

A R U S8 S ENR

                                                                                       -                                              R et t

T O E TI R AA P ER G H R N IR E E Z WHT TO T O ET KE AIM L JC CT $ L E AS A E N WJ PU E R T AE A D E S Y E MH N N IS . E - TR AX O M R SI EE C E U SA

   /                                                                                A                              T                                                                              S    R S                   D                                                          S    EH 4

ET g RAl A P ER 7g wH T 0 L "N g* ;p

  • S D hg b Y R S C U

C - S W Y R I S M < c i O R L C F R C L . 4 C L L - C W 0 1 LE ED TT O R 4

                                                                               -P 8E 2R E

8 1 1 D OV W l P

  • 6 2

V YA F 1

                                                                                                                                                         ?                        :,

R D - E V[ V > D o E

                                                                        - V M

R C m E D DI IS ST NU 2-2o - t 6W 2 6 V UE CT wS T l U

                                                                                                                     $S8 SR EEN RTI F,3     g {u S              PAA L

8 8 8 I O A 2 V.3 6 8 HHT ER a< B 8 E S S rH wE 9 2 8 2 6 9 V Mm V 7 2 0 1 M wo w 0 V u

                                                                                                                                                                                      -             1 I
                                                                                                                                                                                                      -     6 2

oLM _ E l y Y V, Y P y Y LF  : L a

                                                       . '                                      '                                                                                                            't 0

m<s R

                                                                                                                                               -                                                                 g

_ O^  :  : T - C A [ 3 A E wE A b A' T 8-

                                                                        -                      'B 6

E R 1 e R rN 9 0 9 V 2 2 U C O I L 2- - V

                                                     -             1N V-                           V
                                                                                                                      $SA SR F

T V V EEN , l A A

                                                                        -W
               '                                                                                              l       RTI                                                            4 A

OE } - 1 PAA . V - 6 ER o l 2 MIE HHT AE u P - OCT C 2R g A V

                                                                                                                                                                                ,f[ .

RPS FHT IH 1 V 1 1 2T - O C V 0 C

                                                                                                                --                                                 9

(Ub Z. YANKEE NUCLEAR SERVICES DIVISION CALC. NO. VYC-1628 REV 0 DATE / 4 f[/ 7 TITLE Torus Temperature and Pressure Resoonse to Laroe Break LOCA and MSLB Accident Scenarios PREPARED BY KRR REVIEWED BY PAGE K- OF APPENDIX K - CALCULATION OF RHR FLOW SPLITS TO HEAT EXCHANGER The purpose of this attachment is to determine the flow splits between the RHR heat exchange and the 20" bypass line around it. Since the f:ow splits are a function of the total flow rate, and can vary for situations such as bypass valve position and temperature, this attachment will include the development of a spreadsheet program to determine the flow splits. K.0 Method of Sofution The method of solution for this calculation is straight forward fluid mechanics for parallel pipe flow using the modified Bemou!Ii equation. The friction resistances and form losses will be derived from either standard industry information (Crane Technical Paper No. 410) or from component specific information where appropriate. The specific component information includes: RHR Heat Exchanger E-14-1 A & B Pressure Drop Gate Valves RHR-23A,B and RHR 28A,8 design information (to be able to use Crane generalized formulae)

              - Globe Valve RHR 65A,B component specific information - particularly the Cy versus position.

The solution to the problem follows the following process:

1. Review the PalD drawings and the Isometric drawings to determine the piping and component lengths, diameters, elevation,
2. Generate a group of simultaneous equations (modified Bemout!!) to calculate pressure drops at key positions along the piping route as illustrated in Figure K-1.
3. Create a spreadsheet program using Quattro-Pro 6.0 for Windows 3.1 on an Intel Pentium 166 processor, to iterate on a solution to these simultaneous equations (the need for iteration versus analytical solution stems from the friction factor variable with Reynolds Number). Verify the spreadsheet.
4. Perform sensitivity analysis to identify the critical and non-critical parameters.
5. Run the spreadsheet program over a range of total flow rates (if necessary) with a range of bypass valve percent open positions and create a result data table.

K.1 Review of P&lDs and Isometrir Drawinog The following drawings were reviewea to provide information to this calculation: Drawino Nomber hey Date Description EBASCO G 191172 49 3/95 Flow Diagram (P&lD)- RHR System Wi-RHR-Part 11 Sh 3/5 0 2/87 Piping Isometric - RHR Reactor Building (Loop A) W-ISI- 5920-9288 3 9/85 RHR Part 7 - (Piping Isometric - Loop B) W 5920-11957 0 7/17/97 EPRI Program Valve Intemal Dimensions - V10-65A,65B W 5920-2635 N/A N/A Valve intemals V10-23A,B, V10-28A,8

The first step was to compare the two isometric drawings for Loop A and Loop B to ensure that the piping is

! symmetric. With the exception of some minor piping length differences, the two loops are identical. Since

                                                                                                                                                                        /<J G YANKEE NUCLEAR SERVICES DIVISION                                 CALC NO. VYC-1628              REV o          DATEh/2i87 TITLE Torus Temperature and Pressure Resnonse to Laroe Break LOCA and MSLB Accident Scenarios PREPARED BY           KRR                         REVIEWED BY                                  PAGE        K-     OF APPENDIX K - CALCULATION OF RHR FLOW SPLITS TO HEAT EXCHANGER the key resistances are in the valves and the heat exchanger, then the development of the model for either the A loop or the B loop will yield virtually identical results (within the necessary accuracy for this calculation).

For the remainder of the calculation, I will calculate the flow splits for the A loop (VYl-RHR.Part 11 Sh 3/5). Table K 1 summarizes the information gleaned from the isometric drawings as well as the loss coefficients determined from either the Crane paper or by calculation on the notes page after the table. Note that for the piping run lengths, the radii of the 90' elbows are removed as is appropriate via Crane. K.2 Simultaneous Ecuations Devclooment

                                                                                                                                                            ~

Based on the information in Table K-1, we can now create a series of simultaneous equations to solve for the flow splite. The four equations based on the modified Bemoulli Equation are: d2 g'2 P, - P,,, = (0.240 + 2.71f20,,) + - (0.464 + 2.13fa,,) [g.1) A,*n,2g,p, A *,,2g,p, P, ., - Pi-2 * (Ku-xcm) s'* 16.0

                                                                                                                           *P2                                   [K-2)

A 's,2g,p, Ec 144 0 2 P,., - P,,, = (0.884 + 6.01/g,,) *'2 g 6.917

                                                                                                           + (1.025 + 1.947f20")                   - P3 A,*s,2g,p,                                         A ',2g,p,          Ce 144 0 P3 - P,,, = (1.920 + 12.20f,o,, + Kmy)                               2           g 9.083 A ' .2g,p, ,p,Ce 144 0 Finally, from the continuity equation:

dro m " di+d2 [K-5] l As we can see above, the equations require the following known variables: mrors = Total System Flow Rate (ibm /sec) pi = Density of RHR fluid entering split (ibm /ft') - also bypass flow temperature p, = Average density of fluid in heat exchanger (Ibm /ft8) p, = Fluid temperature exiting heat exchanger (Ibm /ft') Km = Heat exchanger loss coefficient

(c.Ut YANKEE NUCLEAR SERVICES DIVISION CALC. NO. VYC-1628 REV O DATE # 447 TITLE Torus Temperature and Pressure Resoonse to Laroe Break LOCA and MSLB Accident Scenarios PREPARED BY KRR REVIEWED BY PAGE K- OF > APPENDIX K - CALCULATION OF RHR FLOW SPLITS TO HEAT EXCHANGER Kmyg = Bypass valve loss coefficient A, = Flow area for 16" pipe (ft') Ay = Flow are a for 20" pipe (ft') These variables w Il be put into the spreadsheet program as inputs which are allowed to be varied. K.3 Creation of Spreadsheet Procram RHRSPLIT.WB2 - OUATTRO-PRO Spreadsheet Proaram Tabte K-2 provides a printout of the spreadsheet program RHRSPLIT.WB2 developed to perform the calculations described earlier. In order to run this spreadsheet using QUATTRO-PRO Version 6.0, the input requirements are marked by the shaded areas. As described earlier, the input requirements are: Total Flow Rate (gpm) Inlet Temperature - Upstream of Split (*F) Intet Density (Ibm /ft') 7 Intet Viscosity (Ibf-sec/ft" x 10 - the number directly from the ASME steam tables) Outlet Temperature - Upstream of Split (*F) Outlet Density (Ibm /ft') Outlet Viscosity (Ibf-sec/ft" x 10'- the number directly from the ASME steam tables) Heat Exchanger Loss Coefficient Bypass Valve Loss Coefficient in addition, the T term as a function of Reynolds number is a variable. For the purposes of this spreadsheet, we have approximated the curve from Crane Page A 25 in the approximate range from 1 x 10 5 to 9 x 10' (using extrapolation of test data). A separate tab on the spreadsheet program called " friction" provides the verification of the fit equations by comparison to ASME. The fit equations are: (16") = -0.002191(log Re)8 + 0.010600(log Re)8 - 0.017740(log Re) + 0.022710 (20) = -0.002378(log Re)8 + 0.010600(log Re)2 - 0.016880(log Re) + 0.021750 The fit verification tab " friction"is listeo in Table K-3. In order to verify the spreadsheet rhrsplit.wb2, a hand calculation of the spreadsheet verification listed in Table K-2 is provided on the fo!!owing pages - including a verification of the friction factor fit data. The hand verification demonstrated that the spreadsheets is working correctly and can now be used for production.

14 find as it tv2. s YANKEE NUCLEAR SERVICES DIVISION CALC. NO. j QC-/MP REv. O DATE II f 85f9~) TITLE PREPARED BY K v4 ' REVIEWED BY PAGE E- OF_ M FAD S P r FT t/ c Ew n c A Ts o n (l of 1)

    ' H X b er n a l d.r # ( IG " , t i s *f )                                      -

C 'd- A = f VA g, -

                                                            +awR 6c                ,

(94 9. c 2me- ) ( '3~ 2 <c3) _= 3 <.i; e c ( l.tc e W ) ( p t.sg )( t. o r-?)(3 2.17) (3c. to sc) aavo , e y;.ci , lh he y.o I s ( te ", i.t r. q '- ) (H9.t7)('.L'E) tt  : 33.ca E .c-

                                  'd (                                                                        D 1 '* ' C-) V
                                                            ~(i. 2 e v ) c 10. 7s )( t. c i-2)be- t 7)         .

MReynalls # (Zo ", tcf or) n (4091,2.1)(',';*b a Sl# 9 0 F[

                                                ~

I 6 /' 7 9 T l " (l .ici) ( st.95)( /.o (-7)f 5 Z.I7) 20.> a . c c % << l Hk 4.,,,y, d ( zo , icg*r ) ( Hs,n)( y ) # 28:97 Cr

                                  ,                                                                           I'O'SI Wb "

( ; , q y,) ! t!.95,.'Ic'(*1)iIY.11) Mr ' Aryuids ( 20 , / 4 g, 4 " t )

                                                                                / S. 6 q                   ,

(i N9. I7)( 7 ) _

                                                                                                                       ?6,//,. E f (I,.9c3 )(Co.7J)(/.0E
                                                                             /               /)(37./7)

(3(*# 5 0 )' l

A,qcos d o r l< Ms YANKEE NUCLEAR SERVICES DIVISION CALC. NO. b- K M REV. O DATE Il IT 97 l ( TITLE PREPARED BY KN REVIEWED BY PAGE OF Sf'tuosacGT \/ s ici ri o re - a ( t. . + .<( ) R eso K1 - f*1a u F l. a it,sr; > TaTAt. f t.a w (yfm)- 2 1 9 4 1. 0 g p* m - z upa r To tat flow (l h /sr.) - (% 199I.o f ( 6 0.94 '#*~hr 3

                                                                                                                                                                           ,        y y, c,9 01 *  ifN)          ?                                               (29,7.oh Ny   ft.o R       ( /k^ /JE)          :                    $ 19, t7          / 6 4, /gec,                        - 7 t e r , y r ,/

3, p< n T:,a (lunian) . .a o p y. pp lu a, /3n -- 7t ra , /4 To Tn f t.* w r 9 4 6. 9 7 1 zo '):1.23 = 2977,09 ' (,onjer/A J l'e # 0*

 .,                                                                                                                                               %O                                               {
                                                                                                                                                                                                   }

nu sa  %,, - n e o o r - E w- 1.3 ' . .

                                                                                                              ~ (919. 77
  • fa ~ ],, c (, 190 4 2,18 { .ot t 3)h (g,ggf (,y,y 2,,y)( c o .9 a Mf 44) _
                                                                                                            ~

( w,5tF) _ (941.F7)z

                                                .:.(o,1c1 + z. /3 (.on C)                                       -

( n. .i Hr,1 '< ) u (/,247) #(t)(5Z /7)l0# idl#Y m o.401 P 1id v ( o , com 6.s-do'4 f n (94S.87)t

-# ~ h.: a h, c 2d (1, z c rf M (y2, /1)(4 /,07)ll44) ;

[l( ,2 ] -

                                                                                                                          /6 %                                   / 6. 3 2 P #31d e p. 4#,w >                                              -

[, e. w <a

                                     #'                                                                                                                               r e s r<0/ f o.96 & C f. 2 4
                                       & ,        (o-         -
                                                                  = l l. o 1 (d1.07)
                                                                                                                                                                                                   )

A ,.,as s>< < ,a. , YANKEE NUCLEAR SERVICES DIVISION CALC.NO. b '/C1i REV. O DATE fI

TITLE PREPARED BY M REVIEWED BY PAGE OF spinoo rre c i v s . r.

taecan 1 or a > g m P, , , - %,, - ( o. rr< 4 c.oico.oer.) 7 ,, m y c o o 7. o ) o r. z e ( s o a z z.33 (ic 'l nr.0) >- , ( H 9.17 '

                            + ( I. o pf 4- I. *M1 ( .0 ' ; 3 b
                                                                                                 ,.,y ( i t. t s) ( i s o

( g, ,, ,, , , ) t, 1. u ) ,( U '. ..:" - 3 c.9#7 = -

f. C G

(6Ll'N{sM (-t.rq) 13WA 33 Loo /* - (K~9] 7 (7en.23)2. y,a . Pa- (!.9m+iz.zo(.otft)91Ac)7 { t.f c.n?(c)(5t.i?)((0 90)('1') _ q,o c3 % _ /f,360 8 M' 4 (6 0.90 ) /11 - {/g',3gf f y 't.i ) St Ject an *% e s e cc I c. a l a 4 Een s , +he ,$preadsherf l [ .$ . Ve e t (Ied G scl Co A be a SC d b o'er 1%e' a .IJs f tans. I I c..<dl1rles l 7F< <e Fa ;,n Fa e is F <. E q, <> & r'. - s Der sh tM i& ms of ssfat l5 the (# n$ f ek Q c lu e C c> c [a e c f i~ of G yrs Ida M. One co n once as il.y o f ds te t Web e3 j'ovi oI t4 thea % . {-I e a e soe , to .Syre el +N perocetr, et [it 9s)eJ on. {L y ee let 3 ot. u " h - ca a 3 4.e s. [ved .

         - T 2;. to le       Ic  -3                                                                    '

f T4 Ot i/s I '. d .17 u=> c.4 c V-e ct .e I c /.1< t-2

                                                                                                                    .__.-____---__j

A,ya,.4i< K  ;;, .z, YANKEE NUCLEAR SERVICES DIVISION CALC.NO. V d-'C'2I REV. 6 DATE U//f/92 i TITLE - PREPARED BY K A*(- REVIFWED BY PAGE OF S?CffvsHCCr \lct:ACAr< w (.5hrer7 9 0 4~ 4) Tu ve. : h th<'l i-h e. C:1 w<> fr ofe I.,. -l- p u + (~. De ms% spmd svet ~ +a k L oa, c n 9 A - T,, ,,, j I t ",T = ,oozi91(Ie,3 F C, /c) o.oto c c o -( l .3 3 c. i.)

                                                                                  ~

0 .et2 7 4 a ( t.3 3c.is) 4 0. 0 z 2 ? * = ,p y - _ [ T ,,4 7 . u ) . oo 2.3 7t (.Ia3 2r,17)# 4- c.o io c oo ( /q Zt.s?)* oILet ( l.3 Z. r,91) + , o 2 i ? f -

                                                                                ~

s ,otz3 v l _ l h (li, e , zo ", Ed r , o o z 3 7 E I lu 3 L6 6seceo /.3 t)3 (I.) ( /. 7 /)

  • l-l
                                                                                        ,O/CPV { lag (1,3ll &                        6L/73                  ,Q tli /

l l_ h e's c &

  • h hee b h. t> he w los o',,

o Q<. & ,cro-o p IV. b f *1c d5 hoc f ff4AS$II, M B 2. ' g .) yt e" t (e i el .

                                                                                                                                                                                )

J [ i  ! l j

icVi YANKEE NUCLEAR SERVICES DIVISION CALC. NO. WC-1628 REV O DATE TITLETorus Temperature and Pressure Resoonse to Laroe Break LOCA and MSLB Accident Scenarigs_ PREPARED BY KRR REVIEWED BY PAGE K- OF APPENDIX K - CALCULATION OF RHR FLOW SPLITS TO HEAT EXCHANGER K.4 Sensitivity Analvses We have verified the spreadsheet in the previous section. In this section, sensitivities will be performed to identify the critical and non-critical parameters. The specific sensitivities are on: Fluid Temperatures Heat Exchanger Loss Coefficient Total Flow Rate Bypass Valve Loss Coefficient - ( l K.4.1 Iemoerature Sensitivity The spreadsheet has sufficient detail to model temperature differences expected throughout the s,; lit which would be created by the operation of the heat exchanger. Using the verification case (Table i K-2) as a baseline case, we can set all the temperatures equal at 165'F for our first sensitivity. Next { we can set the temperatures all at 100*F, then set the temperatures at 165'F (inlet) and 100*F (outlet), and finally set adjust the flow rate from 2181 gpm to 21813 gpm to bound the range of expected operation. As seen below, between a total flow rate of 5000 gpm to 21813 gpm, the variation in flow split flow is less than (0.15). Below 5000 gpm, the flow split decreases with the large temperature difference from inlet to outlet. Therefore, the temperature (heat exchanger performance) impact on the split flow is considered negligible within the flow range from 5000 gpm to 22000 gpm. Deviation (%) l from Baseline ) Sensitivity Table Bypass % (Table K 2) J Baseline Case 121941 gpm K2 31.91 N/A Tout = Tin = 165'F l 21941 gpm K-4a 31.88 -0.03 1 Tout = Tin = 100*F 1 ~21941 gpm K-4b 31.88 -0.03 I Tout = 100*F l Tin = 165'F l 21941 gpm K-4c 31.97 +0.06 Tout = 100*F l Tin = 165'F l 2194 gpm K-4d 31.19 -0.72 Tout = 100*F 1 Tin = 165*F i 14000 gpm K-4e 31.96 +0.05 Tout = 100*F 1 Tin = 165'F i 10000 gpm K-4f 31.95 +0.04 i Tout = 100*F l Tin = 165*F l 5000 opm K-4g 31.83 -0.08 l l K.4.2 Heat Excha.noer loss Coefficient The heat exchange loss coefficient has been shown to vary from 9.622 to 4.724. The impact of this variation is a subject of sensitivity. To address, three cases will be run to compare to the baseline case. These include a case using the baseline K,,m with a 2181 gpm flow rate, a case with the i baseline flow rate with the 4.724 K, ., and a case with the reduced K,,m and2181 gpm flow o___--------__---------  ;

                                                                                                                                                         / o fc.

I YANKEE NUCLEAR SERVICES DIVISION CALC. NO. VYC-1628 REV O DATE TITLE Torus Temperature and Pressure Resoonse to Laroe Break LOCA and MSLB Accident Scenarios l PREPARED BY KRR REVIEWED BY PAGE K- OF i APPENDIX K - CALCULATION OF RHR FLOW SPLITS TO HEAT EXCHANGER rate. As seen below, the heat exchanger flow increases from 32% to 38%. Therefore, the heat exchanger pressure drop is a significant consideration for determination of the flow split. However, as will be shown later, the sensitivity is not as significant as for the bypass valve position. For the purposes of this analysis, we will conservatively assume the fu!! loss coefficient derived from the manufacturer's specifications derived at 10 psid. This will create a conservatively low estimate of heat exchanger flow fraction versus bypass valve position. Heat Exchanger Deviation (%) Loop Flow from Baseline ' Sensitivity Table Fraction (%) (Table K-2) i Baseline Flow Case - Ks xest, = 9.622 K2 31.91 N/A { l 2194 gpm Flow Case - K xe,en = 9.622 K-Sa 31.67 -0.24 Baseline Flow Case - K xe,en = 4.724 K-5b 38.12 +6.21

                                                                                                                                                                 ]

2194 ppm Flow Case - K-n = 4.724 K-Sc 37.81 +5.89 j l K.4.3 Total Flow Rate Sensitivity I By necessity, the previous sensitivities included a variation in flow as well as the primary variable of interest. A comparison of the impact in flow only is provided below. As can be seen, between 5000 gpm and 21812 gpm, the impact on flow only has less than a 0.15% impact on flow fraction. l 1 Therefore, we can present the RHR flow split as a fraction independent of total flow rate. For the { purposes of our analysis of heat exchanger flow fraction versus bypass valve position, we will continue to base the split fraction on the 21812.9 gpm total flow rate with the heat exchanger temperature change from 165'F to 100'F. This will provide sufficient accuracy to cover a range in total flow from 5000 gpm to 21612 gpm. Heat i Exchanger Deviation (%)  ! Loop Flow from Baseline Sensitivity Table Fraction (%) (Table K-2) Baseline Case - Kgxe,en = 9.622 K-2 31.91 N/A 2194 gpm Flow Case - Kaxe,en = 9.622 K-Sa 31.67 -0.24 Tout = 100*F l Tin = 165'F l 2194 g; m K-4d 31.19 -0.72 Tout = 100'F I Tin = 165'F l 14000 gpm K-4e 31.96 +0.05 Tout = 100 F 1 Tin = 165'F l 10000 gpm K 4f 31.95 +0.04 Tout = 100*F l Tin = 165'F l 5000 opm K-4g 31.83 -0.08 i I i i [ _ _ _ . _ _ _ _ _ _ _ _ _ _

                                                                                                                                                                  /C'f)

YANKEE NUCLEAR SERVICES DIVISION CALC. NO. VYC-1628 REV 0 OATE TITLE Torus Temperature and Pressure Resoonse to t_arce Break LOCA and MSLB Accident Scenarios PREPARED BY KRR REVIEWED BY PAGE K- OF APPENDIX K - CALCULATION OF RHR FLOW SPLITS TO HEAT EXCHANGER K.4.4 Bvenss Valve Ooen Position Sensitivity Using the sensitivities performed previously, we can get a reasonably conservative (low) estimate of the flow split fraction by using the base case as the full open bypass condition, and performing a series of runs varying bypass valve position from 100% open to 0% open. These are summarized below: Heat Exchanger Deviition (%) Loop Flow from Baseline Sensitivity Table Fraction (%) (Table K 2) 100% - Valve Open Position - Baseline K-2 31.91 N/A 90% - Valve Open Position - K = 4.234 K-6A 32.27 +0.36 80% - Valve Open Position - K = 4.321 K-6B 32.42 +0.51 70% - Valve Open Posl tion K = 4.505 K-6C 32.73 +0.82 60% Valve Open Position - K = 4.701 K-GD 33.05 +1.16 50% - Valve Open Position - K = 5.019 K 6E 33.56 +1.65 40% - Valve Open Position - K = 6.354 K-6F 35.52 +3.61 , 30% - Valve Open Position - K = 11.297 K-6G - 40.97 +9.06 20% - Valve Open Position K = 25.403 K-6H 49.68 +17.97 10% - Valve Open Position - K = 101.61 K-6I 65.91 +34.00 0% - Valve Open Position - K = = N/A 100.0 % +68.09 l K.4.5 Summaryof Sensitivities Based on the sensitivities in the previous section, it became apparent that the key parameters defining the flow split are the heat exchanger loss coefficient and the bypass valve position. Since the other parameters have been demonstrated to have a relatively insignificant impact on the flow split, we can present the results simply as a table of RHR heat exchanger flow percent versus bypass valve position, as described in the final 1 table of the previous section. Figure K 2 illustrates the data in the previous table. This information. which uses the hiaher estimate for the heat exchancerloss coefficient (K3 xw = 9.622). is considered to be a conservativefv low crediction of the RHR heat exchancer flow solit versus bvoass va've oosition adeouate for use in the containment calculation. This flow split is valid over a range in total flow rate from 5000 gpm to 22000 gpm and over a temperature range from approximately 70*F to 195*F (based on sensitivity / judgement explicit temperatures 30*F).

                                                                                                                                                                          )

l 1 1

A,op.sLix d WT YANKEE NUCLEAR SERVICES DIVISION CALC. NO. NC-/[JJREV, D DATE //!/ 7 7 TITLE ' Tc , , _3 - 7 , , ,, PREPARED BY K 4/f REVIEWED BY PAGE og Fo o re r< - / t Q,-1. cc 2c's_- d/

                                                                                                                                                 !v      ,

Th V io -z e n,'{

                                                                                                                                                                                          /\
                                                                                                                                                                                     #*'            B n l G S~A E s <-*
  • g.. y _ , &
                                                                                                                                                                                                    ~
                                                                                                                                                                                                         'f k
  • M * ' g , , , .
                                                                                                                                                                           ,c.  .             A               va flto M bif1 C.

St 44148#g 6:*4-134 s

                                                                                                                                                           $f                                                                                                              Et- 2M 3' Q %. ',c -                                                                                                           ih f,<s-14 0.s7e c I

f I t I

(46) YANKEE NUCLEAR SERVICES DIVISION CALC. NO. VYC-1828 REV O DATE TITLE Torus Temnerdure and Pressure Roemnse to Laroe Break LOCA and MSLB Accident Scenarios PREPARED BY KRR REVIEWED BY PAGE K- OF APPENDIX K - CALCULATION OF RHR FLOW SPLITS TO HEAT EXCHANGER Figure K 2 VY RHR Flow % vs. Bypass Valve % Open 100 0 k

            ,0   \                                                                                                                                                                    -
                   \

80

                     \
         ?

E" \ 3 h e s 60 \

         *                \
                            \

I '" ' ( 40 e y x m , g -- -- ,- ,, s ) l 20 l L 10 i-i 0 0 20 40 60 80 100 Bypass Valve Position (% Open) mm l

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

j&( YANKEE NUCLEAR SERVICES DIVISION CALC. NO. VYC-1628 REV O DATE TITLE Torus Temperature and Pressure Resoonse to Laroe Break LOCA and MSLB Accident Scenarios PREPARED BY KRR REVIEWED BY PAGE K- OF. APPENDIX K - CALCULATION OF RHR FLOW SPLITS TO HEAT EXCHANGER Table K 1 RHR Heat Exchanger Piping Arrangement Heat Exchanger E-14-1 A Sheet 1 of 5 Diam. Length K or Temp Center Component (In) (ft) (ft/D) (*F) Elevation Height Reference Heat Exchanger Branch - P,- P,., Tee (flow through) 19 N/A 0.240 165 225'3" O' VYl-F1HR-113/5, (20f,) Crane A-29 20' RHR 3C (Sch 30) 19 4292 2.71f 165 225'3* O' VYl-RHR-113/5 Reducer (20*-16*) 1525 9 0'84 .36021 165 225'3" O' VYl-RHR-113/5, Crane - A-26 Gate Valve V10-23A 15.25 N/A 0.104 165 225'3" 0" VYi-RHR-113/5, with flow area equal to (8f,) Crane- A 29 pipe ID - F = 1.0 DWG 5920-2635 16" RHR-3C (Sch 30) 15.25 2.708 2.13f 165 225'3" O' VYl-RHR-113/5 Total Resistance P. P,., = [20*)(0240 + 2.71f) + [16*](0.464 + 2.13f) Heat Exchanger Branch - P,., - P,4 Heat Exc. E 141 A 15.25 N/A 158 225'3" 16'0* VYl-RHR-113/5, 9.622] GE Design Spec 21A1036 Total Resistance - P,., - P , = (KW Heat Exchanger Branch - P,,- Pu 90' Elbow (16' radius) 15.25 1.05(r/d) 0260 149 241'3' 0" VYi-RHR-113/S, (20f,) Crane - A-29 16' RHR-SA 15.25 3.417 2.67f 149 241'3" O' VYi-RHR-113/5 90' Elbow (16* radius) 1525 1.05(r/d) 0260 149 241'3" O' VYl-RHR-113/5, (20f,) Crane A-29 16' RHR 5A 15.25 425 3.34f 149 234'4' 6'11' VYi-RHR-113/5 (Ref. Elev. 241'3*) Gate Valve V-10-28A 1525 N/A 0.104 149 237'2" O' VYl-RHR-113/5, with flow area equal to (8f,) Crane - A 29, pipe ID D = 1.0 DWG 5920-2635 I 90* Elbow (16' radius) 15.25 1.05(r/d) 0.260 149 241'3" O' VYl-RHR-113/5, (20f,) Crane A-29 Reducer (16" 20") 19 90**i 305 149 234'4" O' VYl-RHR-113/5, Crane - A-26 20*-RHR-SA 19 3.083 1.947f 149 234'4' 0' VYl-RHR-113/5

luS T YANKEE NUCLEAR SERVICES DIVISION CALC. NO VYC-1628 REV O DATE TITLE Torus Temperature and Pressure Resoonse to Laroe Break LOCA and MSLB Accident Scenarios PREPARED BY KRR REVIEWED BY PAGE K- OF APPENDIX K - CALCULATION OF RHR FLOW SPLITS TO HEAT EXCHANGER Table K 1 RHR Heat Exchanger Piping Arrangement Heat Exchanger E-14-1 A Sheet 2 of 5 Diam. Length K or Temp Center Component (in) (ft) (fUD) ('F) Elevation Height Reference Tee (Branch) 19 N/A 0.72 149 234'4* 0" VYl-RHR-113/5 (60f,) . Total Resistance Pu - P = [16"](0.884 + 6.01f) + [20"](1.025 + 1.947f) Sypass Branch - P., - Pu Tee (branch) 19 N/A 0.720 165 225'3" 0" VYl-RHR-113 '5, (60fr) Crane - A-29 20"-RHR-6 19 5.583 3.53f 165 225'3" 0" VYl-RHR-113/5 90* Elbow (20" radius) 19 1.05(r/d) 0.240 165 225'3" 0" VYl-RHR 113/5, (20f,) Crane - A-29 20"-RHR-6 19 6.625 4.184f 165 225'3" 9'1 ' VYl-RHR-113/5 (Ref. Elev. 241'3") 90' Elbow (20" radius) 19 1.05(r/d) 0.240 165 234'4" 0" VYl RHR-113/5, (20f,) Crane A-29 20"-RHR-6 19 2.433 1.5361 165 234'4" 0" VYl-RHR-113/5 Globe Valve RHR 65A 19 N/A 4.066W 165 234'4" 0" VYl-RHR-113/5, DWG 5920-11957 20" RHR-6 19 4.667 2.95f 165 234'4" 0" VYl-RHR-113/5 Tee (branch) 19 N/A 0.720 165 234'4" 0" VYl-RHR 113/5, (60fv) Crane - A-29 Total Resistance - P. - P = (1.920 + Kyyn + 12.20f) l I I I I I I Note that the loss coefficients for the heat exchanger (KM and the bypass valve (Kvgn) have been left as variables in the equation. This will be incorporated into the spreadsheet in that manner to allow the user to manipulate the heat exchanger loss coefficient as well as the bypass valve Cywhich varies with stem position. i t

(c.9 L. \ YANKEE NUCLEAR SERVICES DIVISION CALC. NO. VYC-1628 REV 0 DATE TITLE Torus Temperature and Pressure Resoonse to Laroe Break LOCA and MSLB Accident Scenarios l 1 PREPARED BY KRR REVIEWED BY PAGE- K- OF l APPENDIX K - CALCULATION OF RHR FLOW SPLITS TO HEAT EXCHANGER j Tabte K-1 Notes / Calculations from Piping Arrangement Table K-1 (Sheet 3 of 5) [1] Reducers are assumed to have a 8 of 90'(Crane page A 26 for gradual contractions / enlargements) [2] Using equation of Page A 26 of Crane: 0.5(1 - sine g, , i 15.25' 19.0 , ')$ 2

                                                                           = 0.360
                                                                                                  ~
                                                       ' 15.25 '
                                                       , 19.0  j

{

                                                                                                                         )

[3] To back out the heat exchanger loss coefficient, we use the rated conditions from the heat exchanger manufacturer's specification (VYC-1290 - Spec. 21 A1036AE Rev 4 - sheet 5 of 6): Design Conditions: Flow = 3,500,000 lbm/hr (7000 gpm) = 972.2 lbm/sec Temp in (shell) = } 165'F

 ...        Temp out (shell) =        148.7'F Temp ave (calc) =         165 + 148.7 / 2 = -157*F Pressure Drop (psid) = 10 psid Density ave. (p) =        61.05 lbm/ft8 (assuming approximately 50 psia pressure - reasonable                 !

estimate) l ( Using the modified Bernoulli equation: ' K= U rh* Relating the loss coefficient to the 16" piping, the flow area is: A = n(15.25')'/4(144) = 1.268 ft' Plugging in the design conditions above: (10 E)(144 N)(1.268 ftr)r(2)(32.17 Ib,-sec*)(61.05 lbm) It* ft* ft3 Kn.xcuta * *9M (972.2 lbm/sec): l I

I l lvf7 l l YANKEE NUCLEAR SERVICES DIVISION CALC. NO. VYC-1628 REV O DATE TITLE Torus Temperature and Pressure Resoonse to Laroe Break LOCA and MSLB Accident Scenarios PREPARED BY KRR REVIEWED BY PAGE K- OF APPENDIX K - CALCULATION OF RHR FLOW SPLITS TO HEAT EXCHANGER Table K-1 Notes / Calculations from Piping Arrangement Table K-1 (Sheet 4 of 5) A second method for calculating the heat exchanger loss coefficient is to use the value calculated using the HTRI information from VYC 1290 Page 159 (Letter Senior Engineering to YAEC - Evaluation / Analysis Service for RHR Heat Exchangers, dated July 22,1994). From the HTRI output, the calculated pressure drop for the heat exchanger is 4.91 psid at the rateu conditions. Therefore, the loss coefficient based in the HTRI method is: - Kwn (HTRI) = (4.91/10.00)

  • 9.622 = 4.724

[4] Assuming 90' gradual expansion, relative to the 20" piping, from page A-26 of Crane: g_ 15.25)2 K* l9'0 2 ' = 0.305 15.25' ( 19.0 j [5] Example calculation of K for Valve RHR-65A. From VY Drawing 5920-11957, the full open Cy for 20" valve (19' ID) V 10-65A, B is 5344. Using the equation from page 3-4 of Crane: i Kym,c = 891(19")' = 4.066 (5344)2 l l l l l l \

                                                                                                                                                   /c S t .

YANKEE NUCLEAR SERVICES DIVISION CALC. NO. VYC-1628 REV O DATE TITLE Torus Temocrature and Pressure Resoonse to Laroe Break LOCA and MSLB Accident Scenarios PREPARED BY KRR REVIEWED BY PAGE K- OF APPENDIX K CALCULATION OF RHR FLOW SPLITS TO HEAT EXCHANGER Tabte K 1 Notes / Calculations from Piping Arrangement Table K-1 (Sheet 5 of 5) Using (F4 equation above, and the Cy data from drawing 592C 11957, we can get the loss coefficient as a functica of valve percent open: Valve Percent Cyfrom Equivalent K . Open DWG 5920- (using equation (%) 11957 above) 0.0 0 - 10.0 1069 101.610 20.0 2138 25.403 30.0 3206 11.297 40.0 4275 6.354 50.0 4810 5.019 60.0 4970 4.701

                                                            ;               70.0                    5077                 4.505 80.0                    5184                 4.321 90.0                    5237                 4.234 100.0                   5344                 4.066 Table K-2 Spreadsheet Verification Run i

1

M V%a. E F fun f fpd o's X 25.Nov.97 Torus Ternperature and Pressure Response to Large Break LOCA and MSLB Scenarios Prepared by:_KRR Reviewed by: Pa

                                                              ===================================================================================================                                                                                              ======

Table: K 2 1 / ,, . _

                                                                                                 ; Case Desc: Baseline Case - Bypass Valve 100% Open . Tin = 165 F iTout = 148.4 F 100% Max Flow
                                                                                               ' Vermont Yankee RHR Piping Flow Split Calculator RHR Loop Containing Heat Exchanger E 14-1A Filename: RHRSPLfT.W82 Inputs                                                                                                                          Constants ID (in) Ares (ft2)

Total Flow Rate in = 21941.00 9pm 16" Pipe 15.25 1.268 6ntet Temperature = 165.00 Dog F ' 20" Pipe ' 19.00 1.969 inist Density = 60.90 ItwWft3 Inlet Vlecosity = 61.95 lbf-secAt2 x 10^7 Hz Outlet Temp = 148.40 De9 F Hu Outlet Density = 61.24 lbrWit3 Outlet Viscosity = 90.75 ef-secst2 x 10^7 H-XCHER Loss Coef = 0.622 - SP Valve Loss Coef = 4.066 , Calculations

                                                                                                                                                                                                                                                                      ~~-

Resuits . Hest Exchcnger T-inlet f T-outlet f gpm Ibm /sec  % Reynolds No. (E-5) .16" 36.10 -: 0.0125 32.60 0.0125 Total Flow in = 21941.00 2977.09 Reynolds No.(E 5).20" 28.97 0.0123 _26.16 0.0123 Hx Flow Rate = 7000.50 949.87 31.91 Bypass Line . Bypass Flow Rate = 14940.50 2027.22 68.09 Reynolds No. (E-5)- 20" 61.84 - :0.0119 H ~KCHER Density = 61.07 lbmNt3 DP Convergence = -5.39E-05 Heat Exchanger Loop Bypass Loop I l = Protected Cell Pin . P1 1 (psid) = 0.600 Pin Pout (psid)= 15.359 P14 - P1-2 (psid) = 16.322 l l = Unprotoded Cell P12. Pout (psid) = 1.563 Pin . Pout (psid) = 15.359352 T l1 e l r

y g-u..n I t. AgreJ i's (( WC-1628 Rev.0 - 16-Nw97 Torus Tervperature and Pressure Response to Large Break LOCA and MStB Soonanos Prepared ty_KRR Renamed tY Page Table K-3 4 ' -.RHRSPUT.WB2 (Frkeion Tab) The purpose of this page is to demonstrude the widty W s :hird order fit of . Crane Page A 25 for 16" and 20" piping Iri the Reynolds neber range from appreuimetelylis104 to Sr104 The thw orderas proposed are: f(16" pipe)= 0.002191 (log Rea 3) + 0.010000 (log Rea2) + 0.017740 (log RE) + 0.022710 f(20" pips)= 0.002378 (lag Re*3) + 0.010000 (log Re a2) + 0.016880 (log RE) + 0.021750 Reynolds f(16") f(16") Difference (%) f(20') f(20") Difference (%) w cree R crane R

                                                    $ 0.0147..                                                  0.0147                                                             0.28                                                                                               0.0143          0.0143       0.13 8 0.0130                                                    0.0137                                                              1.29                                                                                              0.0136          0.0134       1.47                                                            .,

10 0.0132 0.0134 -1.36 0.0129 0.0131 -1.49 30 0.0125 0.0126 0.22 0.0123 0.0123 0.16 50 0.0124' O.0124 -0.18 0.0120 0.0120 0.05

                                           " x10*5 l

L x< e L [_ __ _ ._ _ _ _ . _ _ _ _ _ . _ _ _ _ _ _ _ _ _ . _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ .___

f-k)b

                                                                                                                    & n'ndt[ k run 25-Nov-97 Torus Temperature and Pressure Response to Large Break LOCA and MSLB Scenarios .

Prepared by:_KRR

  .........................................................................=..................ge Reviewed by*_                                                                     Pa Table:           ~K 4alET Case Desc: Bypass Vald.100% Op'en; Tin = 165 F. Tout = 165 F - 100% Max Flow Vermont Yankee RHR Piping Flow Split Calculator RHR Loop Containing Heat Exchanger E 14-1 A Fliename: RHRSPLIT.WB2
  %                                                                 Constants ID (in) Area (ft2)

Total Flow Rate in = 21941.00 ppm 16" Pipe 15.25 1.268 Inlet Temperature = .165.00 Dog F 20" Pipe 19.00 1.969 Inlet Density = 60.90 lbmm3 Inlet Vscosity = 81.95 lbf-seem2 x 10^7 Hz Outlet Temp = ,, 165.00 Dog F Hz Outlet Density = *

                                    - 60.90 lbmm3 Outlet Vocosity =        s          81.95 lbf-seem2 x 10*7 H-XCHER Loss Coef =                 9.622                                                                       .

BP Valve Loss Coef = 4.066 Calculations Results Heat Exchanger ' T inlet f T-outlet f gpm ibm /sec  % Reynolds No. (E-5) .16" 36.06 -0.C125 36.06 0.0125 Tota! Flow in = 21941.00 2977.09 Reynolds No. (E-5) . 20" 28.95 0.0123' 28.95 0.0123 Hx Flow Ra'.e = 6993.96 948.98 31.88 Bypass Une Bypass Flow Rate = 14947.04 2028.11 68.12 Reynolds No.(E-5) 20" 61.8G A' O.0119 H XCHER Density . 60.9 lbmnt3 DP Convergence . 0.000139 Heat Exchanger Loop Bypass Loop l l = Protected Cell Pin - P1 1 (psid) = 0.599 Pin Pout (psid)= 15.369 P1 1. P1-2 (psid) = . 16.312 l l = Unprotected Cell

. P12. Pout (psid) =                  1.542 Pin . Pout (psid) =       15.3693379 I

i

                                                                                                                                          )

1 i

                                                                                                                                                                                                                              \lyc.-ic.)y             Hu A y c.x m ic run                                                                                                                                                                                                                             254ev4,7 Torus Ternperature and Pressum Response to Large Break LOCA and MSLB Scenarios Prepared by:_KRR                                                                         Reviewed by:                                                                                                                          Page s           e....... 6....ra.........                                  .........................................................................s........

Toble: K-4b - Case Desc: Bypass Valve 100% Open . Tin = 100 F . Tout = 100 F .100% Max Flow Vermont Yankee RHR Piping Flow Split Calculator RHR Loop Containing Heat Exchanger E 14-1A FNoname: RHRSPLIT.WB2 inputs Constants ID (in) Ares (ft2) Total Flow Rate in = i21941.00 gpm 16" Pipe 15.25 1.268 Inlet Temperature = '3 P 100.00 DogF 20" Pipe 19.00 1.969 Inlet Dansky = inist Viscosty = W.142.30'ItW 62.00 sec#t2 lbmnt3 x 10^7 - Hz Outlet Temp = '. K100.00 Dog F l Hz Outlet DensNy = - Lo 62.00 lbmNt3 ( OutistViscosity =  ; J142.30 lbf-sec#t2 x 10^7 H-XCHER Loss Coef = , '. 9.622 ..

. I SP Valve Loss Coef = " 4.066 l

l Calculatens Resuits l- Heat Exchanger T-inlet f T-outlet f gpm Ibm /sec  % Reynolds No. (E-5) .16" 21.15 ' O.0127- 21.15 0.0127 Total Flow in = - 21941.00 3030.86 Reynolds No. (E-5) . 20" 16.97 ~0.0126 16.97 0.0126 Hz Flow Rate = 6995.18 966.29 31.88 Bypass Line Bypass Flow Rate = 14945.82 2064.57 68.12 Reynolds No. (E-5) . 20" 36.27 1:.'.0.0122' H-XCHER Density = 62 lbmnt3 DP Convergence = 0.0004007-Heat Exchanger Loop Bypass Loop l- l = Protected Cell Pig - P1 1 (psid) = 0.611 Pin. Pout (psid) = 15.653 P1 1. P12 (psid) = 16.610 l l = Unprotected Cell P12 - Pout (psid) . 1.568 Pin . Pout (psid) = 15.6532639 l 1 I O h

HC)

                                                                                                                                                                                                                                                                                }3 y P d It N t
run 25-Nov-97 l l Torus Temperature and Pressure Response to Large Break LOCA and MSLB Scenarios j Prepared by,,,,,KRR Reviewed by
                                  ....................................................................................................ge                                                                                                                                                            Pa l                                                                            Table:                                                                     .K-4c l '.r.

Case Desc . Bypass Valve 100% Open -Tin = 165 F. Tout = 100 F.100% Max Flow Vermont Yankee RHR Piping Flow Split Calculator RHR Loop Containing Heat Exchanger E-14-1A Filenarne: RHRSPLIT.WB2 inputs Constants I ID (in) Area (ft?) Total Flow Rate in = 21941.00 gpm 16" Pipe 15.25 1.268 Inlet Temperature = .165.00 Dog F 20" Pipe 19.00 1.969 Inlet Density = ' 60.90 lbm/ft3 inlet Viscosity = 81.95 lbf-seem2 x 10^7

Hx outlet Temp = .100.00 Dog F Hx Outlet Density = . 62.00 lbrr./ft3 Outlet Viscosity =  ; 142.30 lbf-seem2 x 10^7 H-XCHER Loss Coef = ' ' 9.622 -

BP Valve Loss Coef = 4.066 Calculatens Results Heat Exchanger T inlet f T outlet f gpm abm/sec  % ( Reynolds No. (E-5) 16" 36.17 0.0125 20.83 0.0127 Total Flow in = 21941.00 2977.09 Reynolds No. (E-5) . 20" 29.03 ' 'O.0123 16.72 0.0126 Hx Flow Rate = 7014.83 951.81 31.97 Bypass Line . Bypass Flow Rate = 14926.17 2025.27 68.03 Reynolds No.(E 5).20* 61.785 :. . _0.0119 H-XCHER Density = 61.45 lbm/ft3 DP Convergence = 2.63E-06 Heat Exchanger Loop Bypass Loop l l = Prutected Cell Piri - P1 1 (psid) = 0.603 Pin Pout (psid)= 15.337 P1 1 P1-2 (psid) = 16.344 l l = Unprotected Cell I" P12. Pout (psid) = 1 610 Pin . Pout (psid) = 15.3373369 l l f I l l l l L l l l i

V yc - tu n (IC*' Ay .,Ji < IC tun 25-teov-91 Torus Temperature and Pressure Response to t.arge Break LOCA and MSLB Scenarios Prepared by:_KRR Reviewed by:

                                       .............s...................r...................................................................ge                                                                                                         Pa Table:          K-4d Case Desc: Bypass Valve 100% Open. Tin = 165 F. Tout = 100 F 10% Max Flow Vermont Yankee RHR Piping Flow Sptit Calculator RHR Loop Containing Heat Exchanger E-14-1A Filename: RHRSPLIT.WB2 inputs                                                                                       Constants 10 (in) Ares (ft2)

Total Flow Rate in = 2194.00 9pm 16" Pipe 15.25 1.268 Inlet Temperature =  : 165.00 Dog F 20" Pipe 19.00 1.969 Inlet Density = 60.90 lbm/ft3 Inlet Viscosity = 81.95 lbf-sec/ft2 x 10^7 Hz Outlet Temp = - 100.00 Dog F Hz Outlet Density = 62.00 lbm#t3 OutletViscosity = 142.30 lbf-sec/ft2 x 10^7 H-XCHER Loss Coef = S9.622 , BP Valve Loss Coef = 4.066 Calculations Results Heat Exchanger T-inlet f T-outlet f opm Ibm /see  % Reynolds No. (E 5) .16* 3.53 0.0158 2.03 0.0182 Total Flow in = 2194.00 297.70 Reynolds No. (E-5) . 20" 2.83 0.0161 1.63 0.0186 Hx Flow Rate = 684.21 92.84 31.19 Sypass Line Bypass Flow Rate = 1509.79 204.86 68.81 Reynolds No. (E-5) . 20" 6.25" , 0.0138 H-XCHER Density = 61.45 lbm/ft3 DP Convergence = -6.01E 05 Heat Exchanger Loop Bypass Loop l l = Protected Cell Pin . P1 1 (psid) = 0.006 Pin . Pout (psid) = 2.959 P1 1. P12 (psid) = 6.918 l l = Unprotected Cell P1 Pout (psid) = ' 2.965 Pin . Pout (psid) = -3.95937383 l l 0 f

l( -l O 5 5 l((,;Q 377 m e run 25-rgov.97 Torus Temperature and Pressure Response to Large Break LOCA and MSLB Scenarios Prepared by:_KRR Reviewed by . s a n = = = = == s a n a ss= == = = a s a ss e ss ma s s = ==== = == == ======. . .................. sssss s=== = === == == == === = == = ===Pag

                                                                                                                                 ===.. ..

Table: K de . . Case Desc: Bypass Valve 100% Open . Tin = 165 F - Tout = 100 F - 14000 gpm Total Flow , Vermont Yankee RHR Piping Flow Split Calculator RHR Loop Containing Heat Exchanger E-141A Filename: RHRSPUT.WB2 l inputa Constants ID (in) Ares (ft2) Total Flow Rate in = 14000.00 gpm 16* Pipe 15.25 1268 Inlet Temperature = 165.00 Dog F 20" Pipe 19.00 1.969 Inlet Density = 60.90 lbmnt3 Inlet Viscosity = 81.95 lbf-secnt2 x 10^7 Hx Outlet Temp = 100.00 Dog F Hx Outlet Density = . 62.00 lbmnt3 Outlet Viscosity = 142.30 lbf-sec/r:2 x 10^7 H-XCHER Loss Coef = 9.622 - BP Vafve Loss Coef = 4.066 Calculaterts Results l Heat Exchanger T-inlet i T-outlet f gpm Ibm /sec  % Reynolds No. (E-5) .16" 23.07 0.0127 13.29 0.0131 Total Flow in = 14000.00 1899.61 Reynolds No. (E-5)- 20" 18.52 0.0125 10.67 0.0130 Hr Flow Rate = 4474.87 607.16 31.96 Bypass Line Bypass Flow Rate = 9525.13 1292.43 66.04 Reynolds No. (E-5) . 20" 39.42 - 0.0121-H XCHER Density = 61.45 lbm/ft3 DP Convergence = 6.598E-07 Heat Exchanger Loop Bypass Loop I i = Protected Cell Pin . P1 1 (psid) = 0.246 Pin . Pout (psid) = 8.526 P1 1. P12 (psid) = 10.700 i i = Unprotected Cet P1-2. Pout (psid) = 2.420 Pin - Pout (psid) = 8.52556858 I I' l

                                                                                                                                                        \}%-ICA                     lic .

ypycn x - l< l run 25-Nov-91 Torus Ternperature and Pressure Response to Large Break LOCA and MSLB Scenarios Prepared by:__KRR Reviewed by:

         ..................................................................................................ge                                                            Pa i                                                                Table:                K-4f
. Case Desc- Bypass Valve 100% Open Tin.=.165 F Tout = 100 F- 10000 gpm Totat Flow Vermont Yankee RHR Piping Flow Split Calculator -

RHR Loop Containing Heat Exchanger E 141A Faenome: RHRSPLtT.WB2 Inputs Constants ID (in) Area (ft2) Total Flow Rate in = 10000.00 gpm 16" Pipe 15.25 - 1.268 j inlet Temperature = 165.00 Dog F 20" Pipe 19.00 1.969 l Inlet Dens 8y = 60.90 lbmnt3 inlet Viscosity = ) l' 81.95 lbf-secat2 x 10^7 Hz Outlet Temp = 100.00 Dog F Hx Outlet Density = - 62.00 lbmNt3 l Outlet Vsoosity = 142.30 lbf4ec#t2 x 10^7 t- H XCHER Loss Coef = , 9.622 SP Velve Loss Coef = 4.086 l l- Calculations Results i Heat Exchanger - T-intet f T outist f gpm Ibm /sec  % Reynolds No. (E-5) 16" 16.47 0.0129 9.49 0.0135 Total Flow in = 10000.00 1356.86 Reynolds No.(E 5)-20" 13.22 0.0128 7.61. 0.0135 Hx Flow Rate = 3194.52 433.45 31.95 Bypass Line Bypass Flow Rate = 6805.48 923.41 68.05 Reynolds No.(E 5) 20". 28.17' ; 0.0123 l H-XCHER Density . 61.45 lbmNt3 DP Convergence = 2.09E 05 l Heat Exchanger Loop Bypass Loop l. I 1 = Protected Cell l Pin - P1 1 (psid) = 0.125 - Pin Pout (psid)= 6.233 P1 1 - P1-2 (psid) = 8.801 l 1 = Unprotected Cell l~- i P12- Pout (psid) = 2.693 Pin Pout (psid)= 6.23331294 l

      .(

1 t 9

                                                                                                                                                                                                                                                           ~f      k          llL'4
                                                                                                                                                                                                                                                   /kgt'n d        tX    lC' run 25-Nov.97 Torus Temperature and Pressure Response to Large Break LOCA and MSLB Scenarios Prepared by:_KRR                              Reviewed by-
               ......................................................................................................ge                                                                                                                                            Pa Table:              K-4g                    .

Case Dese: Bypass Valve 100% Open Tin = 165 F -Tout = 100 F 5000 gpm Total Flow Vermont Yankee RHR Piping Flow Split Calculator RHR Loop Containing Heat Exchanger E-141A Filename: RHRSPLIT.WB2 inputs Constants ID (in) Ares (ft2) Total Flow Rate in = $000.00 gpm 16" Pipe 15.25 1.268 inlet Temperature = 1 165.00 Dog F 20" Pipe 19.00 1.969 inlet Density =  ; 60.90 lbmnt3 Inlet Viscosity = ' 81.95 lbf-secnt2 x 10^7 Hx Outlet Temp = - 100.00 Dog F Hx Outlet Density = 62.00 lbmnt3 Outlet Viscosity = 142.30 lbf-secnt2 x 10^7 H-XCHER Loss Coef = 9.622 '- BP Valve Loss Coef = 4.066 Calculations Results Heat Exchanger T-inlet f. T-outlet f gpm Ibm /see  % Reynolds No. (E-5) 16" 8.21 0.0137 4.73 0.0149 Total Flow in = 5000.00 678.43 Reynolds No.(E 5) 20" 6.59 0.0137 3.79 i 0.0151 Hx Flow Rate = 1591.55 215.95 31.83 Bypass Line Bypass Flow Rate = 3408.45 462.48 68.17 Reynolds No.(E-5) 20* 14.11 - f'0.0127 H-XCHER Density = 61.45 lbmnt3 DP Convergence = 6.703E-05 Heat Exchanger Loop Bypass Loop l l = Protected Celt Pin - P1 1 (psid)

  • Pin - Pout (psid) = 4.442 P11 P12 (psid) = 0[031 7 318 I I = Unprotected Cell PI 2 Pout (psid)= 2,907 Pin - Pout (psid) = 4 44193661 l

i i l l L

                                                                                                                                                                                                                             '$ D S b.            llCA*-

C Ad i t ) run . . 25.Nev.91 l Torus Temperature and Pressure Response to Large Break LOCA and MSLB Scenarios ! Prepared by: KRR Reviewed by: ' Pag

                                               ...................................................................................................r.e.....

1 i Table: K-5a':PG . Case Desc: - Bypass Valve 100% Open k Tin =,165 F - Tout = 148.7 F- 10% Max Flow KHX = 9.622 l Vermont Yankee RHR Piping Flow Split Calculator RHR Loop Containing Heat Exchanger E-14-1 A Filename: RHRSPLIT.WB2 l Inputs Constants 10 (in) Area (ft2) Total Flow Rate in = 3: 2194.00 gpm 16" Pipe 15.25 1.268 inist Temperature = Q 165.00 Dog F 20" Pipe 19.00 1.969 Inlet Density a g 60.90. ItwnNt3 inlet Vascosity = '*

                                                                                                                            .5 81.95'Itd sec4L2 x 10^7 l                                              Hz Outlet Temp =                                                                          148.70 De9 F
Ha Gullet Donsky = 61.24 lbm4t3 l Outlet Vascosky = 90.75 lbf-secNt2 x 10*7 I H.XCHER Loss Coef a t i

N;M ' 9.622 .. BP Valve Loss Coef = 4.086 i' l Calculations . Results ! Heat Exchanger T inlet f T-outlet f gpm Ibm /sec '  % l Reynolds No.(E-5) 16* 3.58 b 0.0158' 3.24 ' :0.0161 Totaf Flowin = 2194.00 297.70 l Reynolds No. (E 5)- 20" 2.88 ' O.0160 2.60 ' O.0164 Hx Flow Rate = 694.84 94.28 31.67 l Sypass Line Bypass Flow Rata = 1499.16 203.41 68.33 l Re fnolds No. (E-5)- 20" 6.20 C 0.0138 l H XCHER Density . 61.07 lbmNt3 DP Convergence . -7.22E 05 l. Mat Exchanger Loop Bypass Loop l l = Protected Cell Pin P11 (psid) = - 0.006 Pin Pout (psid)= 3.958 l P1 1 - P12 (psid) = 6.880 l l = Unprotected Cell _.- P1 Pout (psid) = ' -2.928 Pin Pout (psid) = - 3.9577077 l g. f i e ___ __.___._i_ _ _ _ _ _ _ . _ _ _ _ . _ _ _ _ _ _ . _ _ . _ _ . _ . _

                                                                                                                                                           \/ Vc -I C 4 /sof
                                                                                                                                                               /
  • icJ ia N l run ' .

!- 25-Nov.97

      - Torus Tornoerature and Pressure Response to Large Break LOCA and MSLB Scenarios J         Prepared by: ,KRR                                                             Reviewed by-
                                       .                 ....................s-------.                 .................................            ...=...................Page_.

Table: K-5td '-r Case Desc Bypass Velve 1.00% Open '-Tin = 165 F -Tout = 148.7 F - 100% Max Flow KHX = 4.724 Venmed Yankee RHR Piping Flow split Calculator RHRLoop Containing Heat Exchanger E 14-1A Filename: RHRSPLIT.WB2 . Inputs Constants 10 (in) Area (ft2) ' TotalFlow Rate in = i' 940.00 gpm 16" Pipe .15.25 1.268 inist Temperature = ^ .165.00 Dog F t 20" Pipe - 19.00 1.969 inist Density = ,60.90 kmNt3 inlet Viscosity =  ; 81.95 lbf-secNL2 x 10^7 Hu Outlet Temp = ~ 148.70 Dog F Hu Outist Density = u ._. 61.24 kmNt3 Outlet Viscosity = p*j90.75 lbf-secNt2 x 10^7 H XCHER Loss Coef = '4.724 SP Velve Loss Coef = "4.066 I Calculehens Results Heat Eschsnger Tintet i T-outlet i gpm Ibm /sec  % Reynolds No. (E-5)- 16" 43.13 'O.0125 38.95 0.0125 Total Flow in = 21940.00 2g76.95 Reynolds No. (E 5) 20" 34.62. 0.0122 31.26 0.0123 Hx Flow Rate = 8364.46 1134.94 38.12 Bypass Line Bypass Flow Rate = 13575.54 1842.01 61.88 ' Reynolds No. (E-5) 20" 56.19 N.10.0119. H-XCHER Density = 61.07 lbmNt3 OP Convergence = 0.0002123 heat Exchanger t. cop Bypass Loop

                                                                                                                                             'l-        l = Protected Cell -

Pin -Pt 1(psid) = 0.857 . Pin - Pout (psid) = 13.352 i P11 P12 (psid) = 13.470 l l = Unprotected Cell - ! P1-2 Pout (psid)= 0.974 Pin Pout (psid)= 13.3525209 i I l I i

l. l I

I l t- . I i J

                                                                                                                                  & -Il D                      IIIa run p>J O N 25-taov.s?

Torus Ternperature and Pressure Response to Large Break LOCA and fASLB Scenarios Prepared by:_KRR Reviewed by: i ass......ss........s.sa.................s...........................s.....za......m========Page ..... .. Table: K-Sc Case Desc: Bypass Valve 100% Open . Tin = 165 F. Tout = 148.7 F - 10% Max Flow KHX = 4.724 Vermont Yankee RHR Pping Flow Split Calculator

                                      .             RHR Loop Containing Heat Exchanger E-14-1A Filename: RHRSPLIT.WB2 tnputs                                                              Constants
                                                        ,                                             ID (in) Area (ft2)

Total Flow Rate in = , 2194.00 gpm 16" Pipe 15.25 1.268 Inlet Temperature = 165.00 Dog F 20" Pipe 19.00 1.969 Inlet Density = 60.90 lbmm3 Inlet Viscosity = ,81.95 lbf-seem2 x 10^7 Hx Outlet Temp = 148.70 Dog F J Hx Outlet Density = 61.24 lbm#t3 Outlet Viscosity = .90.75 lbf-seem2 x 10^7 H.XCHER Loss Coef = , 4.724 s. BP Valve Loss Coef = '4.066 Calculations Results Heat Exchanger T-inlet f T outlet f spm Ibm /sec  % Reynolds No. (E-5) 16" 4.28 0.0152 3.86 .0.0155 Total Flow in = 2194.00 297.70 Reynolds No. (E-5)- 20" 3.43' O.0154 3.10 0.0157 Hx Flow Rate = B29.48 112.55 37.81 Bypass Line Bypass Flow Rate = 1364.52 185.15 62.19 Reynolds No. (E 5).20" 5.65 0.0140 H.XCHER Density = 61.07 lbmm3 DP Convergence = -5.32E-06 Heat Exchanger Loop Bypass Loop Pin . P1 1 (psid) = 0.009 Pin. Pout (psid) = 3.938 ) P1 1 - P12 (psid) = 6.851 1 1 = Unprotected Cell P12 - Pout (psid) = 2.922 Pin. Pout (psid) a 3.93783875 i l 1 I

dD jjy fun g g, A K 25-Nov-97 Torus Temperature and Pressure Response to Large Break LOCA and fASLB Scenarios Prepared by:_KRR Reviewed by:

                          ....................................................................................................ge                                         Pa Table:          K 6a ,

Case Desc: Bypass Valve 90% Open - Tin = 165 F - Tout = 100.0 F 21941 gpm Total Flow Vermont Yankee RHR Piping Flow Split Calculator RHR Loop Containing Heat Exchanger E 14-1A Filename: RHRSPLIT.WB2 inputs Constants ID (in) Ares (ft2) Total Flow Rate in = , ' 21941.00 9pm 16" Pipe 15.25 1.268 Inlet Temperature = ^ 165.00. Dog F 20" Pipe 19.00 1.969 inlet Density =  : 60.90 lbm/ft3 Inlet Viscosity = 81.95 lbf-sec/ft2 x 10^7 Hz Outlet Temp = 100.00 Dog F Hx Outlet Density = 62.00 lbm/ft3 Outlet %scosity = 142.30 lbf-sec/ft2 x 10^7 H-XCHER Loss Coef = _ 9.622 .. BP Valve Loss Coef = 4.234 Calculations Results Heat Exchanger T-inlet i T-outlet f gpm ibm /sec  % Reynolds No.(E-5) 16* 36.51 0.0125 21.02 0.0127 Total Flow in = 21941.00 2977.09 Reynolds No. (E-5) 20" 29.30 0.0123 16.87 0.0126 Hx Flow Rate = 7079.55 960.60 32.27 Bypass Line Bypass Flow Rate = 14861.45 2016.49 67.73 Reyncids No. (E-5) - 20* 61.51 * .0.0119-H-XCHER Density = 61.45 lbm#t3 DP Convergence = 1.261E-06 Heat Exchanger Loop Bypass Loop I I = Protected Cell Pin - P1 1 (psid) = 0.614 Pin Pout (psid)= 15.550 P1 1 - P1-2 (psid) = 16.521 l l = Unprotected Cell P1 Pout (psid) = 1.584 Pin - Pout (psid) = 15.5502423

f ~{(}E lll ,o a :N ( /( 5 run 25.Nov-97 Torus Temperature and Pressure Response to Large Break LOCA and MSLB Scenarios Prepared by:_KRR Reviewed by:

                             ..................................................................................................... age                                P Table:          K-Sb Case Desc- Bypass Valve 80% Open - Tin = 165 F Tout = 100.0 F - 21941 gpm Total Flow Vermont Yankee RHR Paping Flow Split Calculator RHR Loop Containing Heat Exchanger E-14-1A Filename: RHRSPLIT.WB2 Inputs                                                                        Constants ID (in) Ares (ft2)

Total Flow Rate in = 21941.00 gpm 16" Pipe 15.25 1.268 inlet Temperature = 165.00 Deg F 20" Pipe 19.00 1.969 { i inlet Density = 60.90 lbm/ft3 j inlet Vrscosity = 81.95 lbf-sec/ft2 x 10^7 Hx Outlet Temp = .100.00 Deg F ) Hx Outlet Density = 62.00 lbm/ft3 Outlet Viscosity = 1C.30 lbf-sec/ft2 x 10^7 H-XCHER Loss Coef = 9.622 . { BP Valve Loss Coef = 4.321 l Calculations Results Heat Exchanger T-inlet f T-outlet f gpm Ibm /sec  % Reynolds No. (E-5)- 16" 36.68 0.0125 21.12 0.0127 Total Flow in = 21941.00 2977.09 j Reynolds No.(E 5) 20" 29.44 0.0123 16.95 0.0126 Hx Flow Rate = 7112.50 965.07 32.42 Bypass Line Bypass Flow Rate = 14828.50 2012.02 67.58 Reynolds No. (E-5)- 20" 61.37 0.0119 H-XCHER Density = 61.45 lbm/ft3 DP Convergence = -5.36E-06 Heat Exchanger Loop Bypass Loop l l = Protected Cell Pin - P1 1 (psid) = 0.620 Pin - Pout (psid) = 15.659 P1 1 P1-2 (psid) = 16.611 I l = Unprotected Cel' Pt 2 Pout (psid)= 1.571 Pin Pout (psid) = 15.6594113 1 i i e I  ! ! I l

                                                                                                                \/ yc_- / L 2 Y           //ts run gud Ex               X 2bNov-97 Torus Temperature and Pressure Response to Large Break LOCA and fASLB Scenarios Prepared by:_KRR                        Reviewed by:
   . ...................           ...........................................................................Page              ........

Table: K-6c Case Dese: Bypass Valve 70% Open -Tin = 165 F - Tout = 100.0 F - 21941 gpm Total Flow Vermont Yankee RHR Piping Flow Split Calculator RHR Loop Containing Heat Exchanger E-14-1A Filename: RHRSPLIT.WB2 inputs Constants ID (in) Area (f12) Total Flow Rate in = 21941.00 gpm 16" Pipe 15.25 1.268 inlet Temperature = 165.00 Deg F 20" Pipe 19.00 1.969 Inlet Density . 60.90 lbm/ft3 Inlet Viscosity = 81.95 lbf-sec/ft2 x 10^7 Hx Outlet Temp = 100.00 Deg F Hx Outlet Density = 62.00 lbm/ft3 Outlet Viscosity = 142.30 lbf-sec/ft2 x 10^7 H-XCHER Loss Coef = 9.622 - BP Valve Loss Coef = 4.505 Calculations Results Heat Exchanger T intet f T-outlet f gpm tbm/sec  % Reynolds No. (E-5)- 16" 37.03 0.0125 21.32 0.0127 Total Flow in = Reynolds No. (E-5) 20" 21941.00 2977.09 29.72 0.0123 17.12 0.0126 Hx Flow Rate = 7181.00 974.36 32.73 Bypass Line Bypass Flow Rate = 14760.00 2002.73 67.27 Reynolds No.(E 5)-20" 61.09' O.0119 H-XCHER Density = 61.45 lbm/ft3 DP Convergence = -4.78E 05 Heat Exchanger Loop Bypass Loop I I = Protected Cell Pin - P1-1 (psid) = 0.632 ' Pin - Pout (psid) = 15.888 P1 1 - P12 (psid) = 16.800 l = Unprotected Cell

 "                                                                                                I P12 - Pout (psid) =

1.544 l Pin Pout (psid) = 15.8879149 l I I i 1 1 i 1 i ? l h t

                                                                                                                                                                                            '&                 ll}) b f        FAdhk

[- run -

25.Nov.97 Torus Temperature and Pressure Response to Large Break LOCA and MSLB Scenarios Prepared by_KRR Reviewed by-j ====.................................... .............. ............. ............................ age P l

) Table: K4d: l Case Desc: Bypass Valve 60% Open - Tin.= 165 F - Tout = 100.0 F 21941 gpm Total Flow , ! Vermont Yankee RHR Piping Flow Split Calculator RHR Loop Containing Heat Exchanger E-14-1A i Filename: RHRSPLIT.WB2 inputs Constants ' 10 (in) Area (ft2) i Total Flow Rate in = 21941.00 gpm 16" Pipe 15.25' 1.268 intet Temperature = 16:i.00 Dog F 20" Pipe 19.00 1.989 inist Density = 60.90 lbmnt3 inist Viscosity = . 81.95 lbf secnt2 x 10^7 I Hr outlet Temp =_ 100.00 Dog F Hz Outlet Density = > 7 62.00 lbmnt3 Outlet Viscosity = 142.30 lbf-sec4t2 x 10^7 H-XCHER Loss Coef = .. - 9.622 - Sp Valve Loss Coef = ' ' 4.701 ! Calculations Results Heat Exchanger T-Inlet f T-outlet .f gpm Ibm /see  %

Reynolds No. (E-5)- 16" 37.40 . 0.0125- 21.54 0.0127 Total Flow in = - 21941.00 2977.09 i

Reynolds No.(E 5)-20" 30.02 ~ C.0123 17.29 0.0126 Hx Flow Rate = 7252.24 984.03 33.05 Bypass Line Bypass Flow Rate = 14688.76 1993.06 66.95 Reynolds No. (E-5) 20" 60.79 0.0119-H XCHER Density = 61.45 lbmnt3 DP Convergence = -5.29E-05 Heat Exchanger Loop Bypass Loop I 1 = Protected Cell Pin P1 1 (psid) = 0.644 Pin - Pout (psid) = 16.128 P1 1 P1-2 (psid) = 16.999 l l = Unprotected Cell P12 Pout (psid) = -1.516 l^ Pin Pout (psid)= 16.1279138 l I l l.

g a)Y IU> run y, ,< J h N 25-Nov-97 Torus Temperature and Pressure Response to Lar9e Break LOCA and MSLB Scenarios Preparedby: KRR Reviewed by: j

                  .................................. ..............                                                                ................................................Page          ........

Table: . K-6e " { Case Desc- Spas Valve 50% Open . Tin = 165 F. Tout = 100.0 F 21941 gpm Totat Flow i Vermont Yankee RHR Piping Flow Split Calculator RHR Loop Containing Heat Eachanger E-14-1 A l Filename: RHRSPLIT.WB2 .! Inputs Constants 10 (in) Ares (ft2) I TotalFlow Rate in = P ' 21941.00 gpm 16* Pipe 15.25 1.268 Inlet Temperature = y 165.00 DogF 20" Pipe 19.00 1.969 I inlet Density = W. X 60.90 lbmilt3 Inlet Viscosity = '

                                                           ~ 8'i.95 Itd4ecitt2 x 10^7 Hu Outlet Temp =                         100.00 Dog F Hz Outlet Density =                       62.00 Itwntit3 Outlet Viscosity =            ,         ' 142.30 Iti-socitt2 x 10^7 '

H XCHER Loss Coef = 9.622 - DP Valve Loss Coef = 5.019-

                                                                                                                                                                                                              )

Calculations Results Heat Exchanger T-inlet i T-outlet f gpm Reynolds No.(E 5).16" Ibm /sec  % 37.97 1 0.0125 21.87 0.0127 Total Flow in = 21941.00 2977.09 Reynolds No. (E-5) 20" 30.48 0.0123 17.55 0.0126 Hx Flow Rate = 7364.24 999.23 - 33.56 Bypass Line Bypass Flow Rate = 14576.76 1977.86 66.44 Reynolds No. (E 5).20* 60 33 '10.0119 H-XCHER Density = J_ I, .W./ft3 DP Convergence = -0.00012G

                                                           ~ ^ ~

Heat Exchanger Loop Bypass Loop I l = Protected Cell Pin P11 (psid) = 0W - Pin Pout (psid) = 16.510 P1 1. P14 (psid) = 7D6 I I = Unprotected Cell P1-2. Pout (psid) = 1.470 l Pin . Pout (psid) = - 16.5100171 l l I l - - _ _ _ _ _ _ _ _ _ _ _ _ = _ _ _ _ _ _ . . _ _ _ b

C.-!l 0 {flL,

                                                                                                                            .c'eA d E K Y run                                  .

Torus Tergrature and Pressure Response to Large Break LOCA and MSLB Scenarios 25.Nev.97 Prepared by: ,KRR Reviewed by:

    ==.. .............................. ........................                            ....................................Page     ........

Table: K 6f '- Case Desc: Bypass Valve 40% Open . Tin = 165 F - Tout = 100.0 F - 21941 gpm Total Flow Vermont Yankee RHR Piping Flow Split Calculator RHR Loop Containing Heat Exchan9er E-141 A Filename: RHRSPLIT.WB2 inputs Constants 10 (in) Ares (ft2) l- TotalFlow Rate in = - 21941.00 gpm 16" Pipe 15.25 1.268 inlet Temperature = 165.00 Dog F 20" Pipe 19.00 1.969 inist Density = '- '60.90 lbmtit3 l Inlet Viscosity = - 81.95 lbf-seclit2 x 10^7 Hx Outl6 Temp = 100.00 Dog F Hz Outlet Density = - 62.00 lbm/ft3 Outlet Viscosity = s > 142.30 lbf-sec/ft2 x 10^7 H-XCHER Loss Coef = ' ' ' 9.622 - SP Valve Loss Coef = 6.354 i' Calculations Results Heat Exchanger Tinlet f T outlet f gpm Ibm /sec  % Reynolds No. (E-5)- 16* 40.18 'O.0125 23.14 0.0127 Total Flow in = 21941.00 2977.09 Reynolds No.(E-5) 20" 32.25 0.0122 16.57 0.0125 Hz Flow Rate = 7792.75 1057.37 Bypass Lins 35.52 ;i Bypass Flow Rate = 14148.25 1919.72 64.48 Reynolds No. (E-5) . 20" # l 58.56 : 0.0119 H-XCHER Density = 61.45 lbm!ft3 DP Convergence = 1.271E 08 Heat Exchanger Loop Bypass Loop I I = Protected Cell Pin . P1 1 (psid) = 0.744 Pin - Pout (psid) = 18.026 P1 1. P12 (psid) =

  '                                     18.572                                                          l      l = Unprotected Cell                    i l'   P1-2. Pout (psid) =                    1.290 Pin Pout (psid)=          18.0259163 l

i r [ l

                                                                                                                                                        )

1

                                                                                                                    .                                   I h                                                                                                                                                       )

((l l fun g e , 4 tc l 25 tJov-97 Torus Ternperature and Pressure Respmse to Larg(p Break LOCA and MSLB Scenarios Prepared by:_KRR Reviewed by:

                                          ....................................................................................................Page                                                                                      ........

l Table: K4g Case Desc: Bypass Valve 30% Open . Tin = 165 F . Tout = 100.0 F. 21941 gpm Total Flow Vermont Yankee RHR Pping Flow Split Calculator RHR Loop Containing Heat Exchanger E-14-1 A Filename: RHRSPLIT.WB2 inputs Constants ' 10 (in) Ares (ft2) TotalFlow Rate in = 21941.00 gpm 16" Pipe 15.25 1.268 Inlet Temperature = 165.00 Deg F 20" Pipe 19.00 1.969 inlet Density = 60,90 lbm/ft3 inist Visootity = 81.95 lbf-secM2 x 10^7 Hx Outlet Temp = 100.00 Dog F Hx Outlet Density = 62.00 lbm/ft3 Outlet Viscosity = 142.30 lbf-secM2 x 10^7 H-XCHER Loss Coef = . 9.622 BP Valve Loss Coef = 11.297 l Calculations Results l Heat Exchanger T-inlet .f T-outlet f gpm - Ibm /see  % Reynolds No. (E-5) .16" 46.35 : 0.0124 26.69 _ 0.0126 Total Flow in = Reynolds No. (E.5) 20" 21941.00 2977.09 37.20- 0.0122 21.42 0.0125 Hx Flow Rate = 8988.66 1219.64 Bypass une 40.97 Bypass Flow Rate = 12952.34 1757.45 59.03 Reyno:ds No. (E-5) . 20" 53.61i 0.0120 H.XCHER Density = 61.45 lbm#t3 DP Convergence = 4.073E.C7 Heat Exchanger Loop Bypass Loop l l Protected Cell Pm .P1-1 (psid) = 0.989 Pin . Pout (psid) = 22.710 P1 1. P12 (psid) = 22.453 l = Unprotected Cell l l _. P12. Pout (psid) = 0.732 Pin . Pout (psid) = 22.7099442 i l I

(' llIb fun f i'kt X Y 25-twov.97 Torus Temperature and Pressure Response to urge Break LOCA and MSLB Scenarios Prepared by:_KRR Reviewed by-

                             ====.....====..................===....==.........................===..=..=....=========..==.......Page                         .....=..

Table: K4h Case Dese: Bypass Valve 20% Open . Tin = 165 F . Tout = 100.0 F. 21941 gpm Total Ficw Vermont Yankee RHR Piping Flow Split Calculator RHR Loop Containing Heat Exchanger E-14-1 A Filename: RHRSPLIT.WB2 inputs Constants ID (in) Area (ft2) TotalFlow Rate in = 21941.00 opm 16" Pipe 15.25 1.268 inlet Temperature = 165.00 Deg F 20" Pipe 19.00 1.969 Inlet Density = 60.90 lbm/ft3 Inlet Viscosity = 81.95 lbf-sec/ft2 x 10^7 Hx Outlet Temp = 100.00 Deg F Hx Outlet Density = 62.00 lbm/ft3 Outlet Viscosity = 142.30 lbf-sec/ft2 x 10^7 i H-XCHER Loss Coef = ,

                                                              ,9.622                                                                          -

BP Valve Loss Coef a 25.403 j Calculations Results j Heat Exchanger T-intet f T-outlet f gpm ibm /see  % 1 Reynolds No. (E-5)- 16" $6.43 0.0124 32.50 0.0125 Total Flow in = 21941.00 2977.09 l Reynolds No.(E.5) 20" 45.29 0.0121 26.08 0.0123 Hx Flow Rate = 10943.57 1484.89 49 88 ' Bypass une Bypass Flow Rate = 10997.43 1492.20 50 12 Reynolds No. (E-5) 20" 45.52' O.0121-H-XCHER Density = 61.45 lbm/ft3 DP Convergence = -0.000466 Heat Exchanger Loop Bypass Loop i i = Protected Cell Pin . P1 1 (psid) = 1.466 Pin . F out (psid) = 31.804 P1 1. P12 (psid) = 29.989 l 1 = Unprotected Ceu P12 Pout (psid)= 0.349 Pin . Pout (psid) = 31.8039902 l i I l

Qq} run f a d 't A l' 25-Nov.97 Torus Ternperature and Pressure Response to Large Break LOCA and MSLB Scenarios Prepared byLKRR Reviewed by-

                                                     ....................................................................................................Page                    ........

Table: K41 Case Dese: Bypass Valve 10% Open - Tan = 165 F- Tout = 100.0 F - 21941 gpm Total Flow Vermont Yankee RHR Piping Flow Split Calculator RHR Loop Containing Heat Exchanger E 141A Filename: RHRSPLIT.WB2 inputs Constants ID (in) Ares (ft2) Total Flow Rate in = 21941.00 ppm 16" Pipe 15.25 1.268 Inlet Temperature = 165.00 Dog F 20" Pipe 19.00 1.969 Inlet Density = . 60.90 lbrnNt3 inlet Vrscosity = - 81.95 lbf-secNt2 x 10^7 Hz Outlet Temp = ' 100.00 Dog F Hz Outlet Density = 62.00 lbm#t3 Outlet Viscosity = 142.30 lbf.sec#t2 x 10^7 H-XCHER Loss Coef = 9.622 - BP Valve Loss Coef = 101.61 Cak;ulatons Results Heat Exchanger T-inlet f T-outlet f gpm Ibm /sec  % Reynolds No. (E 5) 16* 74.57 0.0123 42.95 0.0125 Total Flow in = 21941.00 2977.09 Reynolds No. (E-5)- 20" $9.85 0.0119 34.47 0.0122 Hx Flow Rate = 14461.44 1952.22 65.91 Bypass Line Bypass Flow Rate = 7479.56 1014.87 34.09 Reynolds No. (E-5) 20" 30.96 0.0123 H-XCHER Density = 61.45 lbm/ft3 DP Convergence = 0.0003072 Heat Exchanger Loop Bypass Loop I l = Protected Cell Pin -P1 1 (psid) = 2.558 Pin- Pout (psid) = 52.660 P11 - P12 (psid) = 47.272 l = Unprotected Cell l P12 Pout (psid)= 2.830 Pin - Pout (psid) = 52.6604374 I l l l m._ _ _ _ _ . _ _ _ . _ _ _ _ _ .. _ _ _ _ . _ _ _ _ _ _ _ _ _ . -

i llN

                                                   }/fy (f L              (YCL/                ,3
                                                                                                    \

1 UTILITY PRODUCTS DIVISION f M2 f

                                                                                                .,  1 ENGINEERING CO                     h   I l

5701 S. Eastem Avenue  ! Los Angeles, Califomia 90040 { P.O. Box 54940 l-July 22, 1994 1.os Angeles, Califomia 90054 ) Y:nkee Atomic Electric Company . 580 Main Street OOlton, MA 01740-1398 Attention: Mr. Aubrey Doyle Senior Systems Engineer Vermont Yankee Project . R ference: Vermont Yankee Nuclear Power Corp. P.O. 94-556272-00 i Senior Job No. 94-0504, Rev. 1

Subject:

EVALUATION / ANALYSIS SERVICE FOR RHR HEAT EXCHANGERS

                                                                                                    )

G ntlemen: l In cccordance with the requirements of the referenced purchase order, transmitted he.rewith is an original and one copy bound copy of the final Thermal Performance Report for the subject project. As you can see, this is Rev. 1 and includes corrections for small typographical errors that were caught, as well as some minor ^ ch2nges to the duties in Appendices C thru H. vcry truly yours, h,j, /gI 1 i S I R EN EERING COMPANY gf  ; H Wightman, P.E I l Director , Marketing Services HW/dc sun.sn V y j L 2,. f ces Lud Zabel gudi x k P 3 K-Telephone: (213) 726-0641 Telefax: (213) 726 9592

           \l9C.- I c ?2                 -

l/)f(lf3 112 t K Age 4 t/y'6 -/294 $1NIOR'

             )                                                 ENGINEERING CO VERMONT YANKEE NUCLEAR POWER STATION RHR HEAT EXCHANGERS THERMAL PERFORMANCE REPORT Submitted To VERMONT YANKEE PROJECT REFERENCE PURCHASE ORDER #94-556272-00 SENIOR JOB    94-0504, REV. 1 i

By b (l Mr l SENIOR ENGINEERING COMPANY Los Angeles, California l June 1, 1994 7 Prepared By: N.R.Gainsb ro, P.E. Approved By: _

                                                    'H. Wi   tman, P.E.
                                           'W
c. .r= :_ . fm l&.g
                                                                                                                                                        . . _ _ _ _St,
                                                  ;      ,8C.=lle.M AyK                                                   t 3            K-                               -

GENER AL h ELECTR10 atoute rowca tousewrut cutas. nut HEAT EXCHA A SHEET Y YM b O/ (CONTAINMENT SPRAY OPERATING CONDITION)SPEC. 21A1036AE, Rev. 4 y/P PR*TECT REQUISITION _ _ _ _ VERMONT YANKEE 296-79900 M( ' pace ~ ~S OF 6 se .tse .f dai, INTERMITTENT 1 Stae foem No. Nee. Se. Fe. Type _ 2 Na a f U..., 2 Conneefed in j b iere eer uni _, __ .. she r,k,,er_tini, surf,ee ., shett 7 Performance of One Unit u.ie w, I I ,m m, 4 remin, or Inhib. Demin. In.,id Cirseteeeg . j RIVER WATER 5 3" I

                                                                                             .tetel Flutd Enferine -                                     t/hr         !                                                   ~~l e                                                                          I      veper                                           ~ g/h,            [

7 3. 5_K LOW __(7000 rom) j Uc.id s."h, i 1.35 7100-(2700 gpm) i Steem _ , , , _ t/br  ! 1 N.a.cead...e bre. .,_ ffk, I 80 Field Vaporised er Camdensed I/kr

                                                                                                                                                                                                                          ]

33 __

                                                                                            .1*ma m crad.eenr1                                          ukr l3                                                                            Geev4y                                                                                                                   ___

I is se @ ' F. Design Vincesity I 14 e, @ _ . r. _. jMaleet'for Weight I 15 - 8.T.11./l IPesifie Hees le l g

                                                                                                                                                                                                             .T.U./ t 365 8 T.M           Leoe ne Heat                                                        '

s T,1) /t er af

                                           .                      . . , ,
  • r. _ Temperseurein._, _ _ t 65 .;,

in 11/S.7 /P1/w'/ F# F. Temperseure Ov, i 127.2

  ,           q,                           h                                Asis            one,6s Preinre                                                      I
                                                                                                                                                                                                           .p, _

2g / h Number of Pepes I PilG 3: / 5 ft (v.fecity _ ($0ffMUM) 8 -cu-hi/20 S t. Stl. 23 . l . . alh xy/sec.' 4,s..;. o,.p ,eri/.a u s _ ~ i _,,js,,, llo

         ,23                ( w.Jou)                                                                                                                                                  _

_ g

                                                                                                                                                                                                           /s ,.in.

p,,g,, p,, ,,, _ 'O.002_ 24 H:st Eseb. g ed . I.T.U/Hr. 5 7 5 a 1(6 u.T.o. (Corrected) _ 25 frari ter b g . Se, vie. Cf n. Construction 453 24 s/s, in oe. ,. r,euvre __ _ 27 CQPE //5q. in Test Press.<a .,,, _}50 g/sq. s., CODE g/s e. In. N MO ' F, Desinn,Inmeersevre 350 __ 3, 2f T4bes _ No. O.D. SWG Min. Ave. Length -. aQQ _ Plesh 20 skit _ lb. O.D. Thicinen 31 58til Co.or. Fleellag Head -- 32 Cheneef Channel Cevee - 31 dbe Sheets.$tettenary_ _ -_ Heating 34 Beff!H .Ces,es - Type ,,, Thickness Spectag Il telfle.Long _ TrPe Thfeiness ja Tal,e $wppceti Thiel. ness 37 1 lees 30 Conce etiene . XheN . in Ct Seeles - _ __ Coerosten Allowence. Shen Ende 39 c,annettaan rhann.f. fm Out , _ $erles Carr=.1== Affmer==*a . fuba Wa 40 Code Reev' empts ASME .

  • Teme Cteu di speelficatione 12, Watehts. fe h R_ tvedle _ ._ peg .( W,,,, __,

i REMARK 5: Mew ($.R.) if steen eefieved and (X.R.) if eedtegreeked 4i NOTEL MANUALLY OPERATED TUBE DRAIhWCATED ON FLOATING HEAD. 4s 44 O . _ . , . . - - (HI 91

                                                                                                                                                                           ...                        // 2 3
                                                                                                                                     ~    - _N C** I                           17                                                             ff{f0iY
                      ,            g                    ..

RES M AL HEAT REMOVAL % GENE.RALhELEDTRIC i Atc..ic rowse soueeutwr cuaerua g( y 7 EXCHANGER DATA SHEET (SH'fI. W. nPFu tw i A1036AE, Rev. 4_~ PRC*ECT \ VERMONT YANKEE) % _ ndE

                                                                                                                                                                 ~ ~

j REOLII SITION h_ __ 296 .)),600 PAGE 6 OF Final i { 5:r-ice of Jett _  : \ TNTERMITTANT ltem No. i 2 sr. w.. se. 54 rype Cea.e.ted I. n s u, .. j i 2 wr... .., uait 7 skethant unir ufen mer s5en Pedermance of One Unit c6-w m- l

                                                                              !                                                                                 ra. m.                                :

4 Domin tralized 1Jarer ,lHold Cirevisied _ River Water j s _ Totet Fivid Entering t/kr _ ,_ 4 Ye per f/he 1 7 {0,00gpm

                                 ,,,                                               tr e,1d            .                         s/he              _2ggpm                                             ,

s steem - f/kr . 9 Nen.ceedense blo s g/hr to Hvid Vep.cised er Condensed f/br j 11 _ steam credene,J f /br I l I2 _ Grevity l 13 cp @ . ' F. Desiga Viscosity gp @ e r, i 14 Meteculer We*ght l 15 B.T.U./l specific Heat g.T.U./t ta 8.T t)Ji Laten Heat I i _T.Uj r 87 if,5 _ ' F. Temperetv,e s. 75 q, 18 9 _26 ' F. Temperefere Ove .,8 9. 6

  • F.

9 pstG 0eer viag,fresivre fsto ) 3 _ l Number of Penes i 21 [ 5 F4/s.g. .!. city 8 Cu Ni/20 St. Stl. Ft./sec. I 22 5 10 p/ss.t. Pressur Drop per Unit Mei 10 t/sg. in. 23 _ Iroutind Ydter 1 (later) 24 H2e- Eachanged. I.T.U/Hr. ,,,, _,, b . 20 x 106 y,y,o, ge,,,se,,g) j 15 fraed.e int < _. serviu O ~ Cte en y/d/Yd Construct. ion 24 LSO s/5e. In.l Design Preuvre 350 t/sq.4 27 CODE t/sq. la. Test Pressere CODE t/sg. la. 2s 350 ' F. - Deifen Temme+ure 1 3 5 0___ ' F. 29 Txb:s No. O.D. SWG , Mig Avg , Leng!,h. fitch . A Q,O , 30 Shell __ l.0. C.D. Thietness 31 Skou Cover- Fleeting Heed 32 Chernet Channel Cover 33 dbe sheet..Itettenery - Fleetlag 34 te W s. Creae Trpe - Thickness Speelag 35 Be W .Leag Trpe Thienness 36 Tibe Supperi . Thicinesa. 37 &olets - 14 Connections . s ts#. in - Out ,, serres . Corresten Aftewence.5ket Side 39 c.aame%,n rkennet. in out series carra.t.. Alt...... . Tuk. tvJ. 40 C.Je Reqw emente AsME Teme C(ess Al Specifiut*ess .

   *2 Wilghts. Ess6 Sken                             _

sendie FvR of Weier. RCMARK1: Lie 4 ($ R.) If stress relieved ead (XI.] If redlegrePhed l 45 44 47 jut 2 3 1968

F

.c

K V VC-/( M p w_ - - c h~g.fj'? 1121 y,a,c K SOUTHWESTERNv ENGINEERING .- -COMPANY g ,,, .... .-y[1 .yExcN ANocr. saccaricArson ~ . - . . ..,. succ ......,

                                                                                                                                                                                                                 /

,- i i y p SW-

                             *         "8*
                                                "5
                                       #" 0""*".".
                                                                                                                                                                               .ros No.
           -                                                   ___ GENERAL.Er5 CTRIQ,._C.QMPANY 4                                                                                                                                                KLYNO.                    27453 6

_ 'L ANT LOCATION g;f,g,g/hQ. _1.7.199.r.t,0nAvenue,,, NouSan R Y Jose. Calif. . NO,,.,,,,, H,1,0,2 7 urnvice or uN ypJ[lf DArc 10/lT/76~ l ,e 3f;,fg ggy,i4 RESIDUAL HEAT EXCHANGERl/V/ arIM NO. i N_o. UNITE 2 SNEU.5/UNar } CONN.IN

                                                                                                                               -SERIES
                                                                                                                                     " msm vser                                sunrAcc/sNELL 5 365 i
                                                                                                                                                                                                                     ~

L

  • PARALLE1 SURFACE UNIT 3,,_163~1 i
. i.

PERFORMANCE s cu. eeoc OF ONE UNIT . CONTAINM r 6. ceacuyrso vuec ..e

                       -ii                                                 _ . .              _ _

me i. ri.u o enrememo s/ma. t DEMIb1ERALIZED EATER i

                        ,e vAeon                                                                             3,,500,,,000                                             RIyER WATER t-                                                                                        n/ ..

1,._350.000 is uouco */aa

                       "                    ***                                                                         ,,,_3,1 @ ,_000 15
                                                                                        "'"*-                                                                                                J.[j50,000 NON.CoNOEN.A.LE.                   _ _.. $/N
                                                                                       - R.,             . . . ..                              . . -

le __ _ flu O WAreafECD Ost CoNDEN CO S /N R.l _

                       '7 erEAN CoNoENeED                                       3/ a.

is snavirv - u ouso - le - 90000* VisCOeffv aCENTIPOrets) _.

                   - so                                                                                                                         ,. C# e           ' r.

WCLteuLAa WCleHT-VAeone - CP e El .- ePatCarsC HEAT . e aa ._varreu. cona. .Ea'~rTrE' ~ ~ ~ ~ ' ' ~ " " ~ as vene==Avunesw ~

  • r.
        ,          .u          raucc=Atuac our                                                                                                 165
  • r. 85 as oesursws casseune 148.7 esio 127,2

, as munesa or rasses [ 'se ar. m up "rtesee ~ I 2 \ Qc,acuac enoe ) yhrit# _ rea pNry 4.35 as me: .M is Aeca ou ruac oni - _ _,,, y to )v/srft. 5 so usAr esenanoc WO0ds;

                                                           -:.r.u.fusw. 57.100.000                                                                                                                             .0021 at t=Ausreet mAvs4 oravice                                                                                                                          u.r.o.eeenacercon                  47,1 225                             cLsAw ,_             513              Torat.rs,uuwe Ava:LAsoc as                                                   .
                                                                                                                                                                                                              .0021 CONSTRUCTION j

'- ss [sessoh reessunc seso. in. sa vest recesunc 450 350

                                                                     ~__ s/so.ew.

so _oreiw vener=Arvar

  • r. .

3.zy. counscrious, ease. . . - " ' - 3J0 .

                                                                                                                                                                                     ...       ..-:350              ' "
                                                                                                  ** 16"                           o. ur                 16"             sa er _ nAraws si recess                                                                                                       _ . . , _

12" out 12" so - vecti mo. 1.664 0.o.

                                                                                           -i seh. 40 W E.                                                   Sch. 40 W. E.

so Slg" awa 20 AN. *ve ' _.,_ ocuom 4 24Su emoss sacroctres Sollt gegenental SPACsWe LONG. AprW TYPE: . estem 7/8 M 4e , cowst=uereew-4.uc scetTow . . ei Webs cA.ewru. vewAca.As* C _ 3.L150 lbs. cuev. eccc. 21 A 1036 ':O N o . da *uu.ar **Tsa - 40. 000 lb.araavaa'c auwa*.* 1;5 000' 1bs.

             -40         starie mettsvttipAnts tsARICsm d##6===AD'USRAPhr# PARTS MARWre (NRI FART SEATERiAL                      ' TetM           ComR. AL.            GASKETe

,,,,,_,_ 44 vus*8~ MATCRIAL

           ***           **"to T, a.

304 SS A-249 e g_ g o,iggcl.A 3W cuaw-coven Steel Jacketed Asbestos

  , ; 88                F'T* rA                                                                           __ .1                      caAN.- -F's. TA
                                                  (",sAi/d",SS Clad                                            11        -
                                                                                                                                     **LA.--788 T.s.
                        **" . . ' . ' -             C S.fYR I                                         '
                                                                                                                                     "*u. covca                     Steel Jaeketed Asbeatos
        , 44          '. sun.Lcovta                 C
                                                                                                                 .1 steel _ Jacketed Asbestos
            -43 _ vues suecoare                                                                                  .1                  P'T* "<^o

_.E_fXR1 Steel Jacketed Asbeatos ae m aterLes Q, s. 3/8' . Amas

                                                                                                                                   **'   Carbornastic   eshall SocHen coating dr.E. Tak.= h e V
 ;ceg

__ caA=sco . C.S.ee .1

   , , ". Os           caA*.sovs=                 c. s. * *                                                    .1 1",,._J.04 Ss drain valve on floatiner head
      . ce __ rce us. cow.-
     . . c4
c. s. * * " .2 Wald tubes 'to tubesheet (3/4 t) -
                .,   __towe. ny,, rete            .          .

Shell ReHef Valve.Consolf 1975c.1"x.1"se:

             * * .UP
                                                . C, S.
          . os e . , a.. . .
                                                  .                                          L,, , ,a                          iBoth R aHelief Qb_sm.R                 ei VeValve-Con h'* * - #-'s oll/1910
                                                                                                                                                                                        '- ' - - ~ ' '

Lc. 4"x6".

l234S678901234Sa7890123436789012343678901 234567990t2 C0C0O00001 1111I l 1112222222222333333333344444444443S3

        - e9291 s0537
  • 0I E94 E271170I S908DT DT E
                                     *N0LR13                                                R12943                        LE101IE                                                IE                                             L
        - e . 42
  • I0AU . . U0 .469 LD450UL UL O F 1.TT06 1 T .4824 A 20GT QT H
        - es45 800. .

S N O 1 223 RR000IU I U E

            #      9= =                                  T                                                                EE        00LG                                         LO                                             L6 1RF               O                                                                                    VV          . .                                                                                       F5
                //E                  I                   F                                                                OO        00                                                                                          F1 41F                  T            2OLS6                                                                                         T106S450                             T069                                       A0 35S7 E . .N0 CD5 EN9TE . .

LS6 L0S0980T L96020 C E L 09S062

                                                                                                                                                          . . .       .29 E072 L . .
                                                                                                                                                                                                       '                       B .

0 C26At RE . NH31 E9 .286 PL3 T S21S04 T270. O N A M 3R 5T0 . O 1 = R 0ES23 C C R H .2S63 S0 222 ( A9 T3 E U O i1 1 U3 G

  • E T

R TE - EE T eM7. .

  • B O tA RASPLTE e T1 023457 T094 *U F S EH&( ZEL
  • E 007910 E064
  • T R- I0 FP0 ZLT * '.OOO)
  • L . . . .33 L . .

EWW 9 BA . ONU

  • ONNN . G E N S31S03 N27O. IL FO0R2 MG1 *NIO eNN MSN7CE I i1 1 I1 . L NE 3

A

  • sO SSSIEI7NC E)WS I F0 O .R
  • ILDDD( CL .AN I H DII = T  :
  • 4 STTLLL NU&TA ( S0 IWW C  : Ae S1 RCDOOO AOSST I I I) ) ) I I ) 0 S// TT0SD . EENNNN TF IS .CI .22 .CI 8O0
      - LWW                          AEePN8                                                 TRAYYY                        S         SI                 NESNCC                        l    NES                                ET0 LOO 2 EB .OE                                                                   ERREEE                        I         ES*ONI SPIEE                                     lI SP                                   C -

ELL 9 H OR MOPRRR S RE*NO( / t (SS A( /t NE4. HFFS3 D AC EE3 R* I T I T AL

   %        S-XXO0 E N       ,

CO4 W7 R AL TEE TRRR9G U .NL

                                                                                                                                             *DT NI F

( EFF TT D A F t RD AN 0 I = D RD . PL ELYLT0IASES L// R EU 1 .T9 I UN0 A LTAA 2 LIE O ZSS ( LB

                  .FC                S                   SI2                                  .W             NULM                    UTLLP                          ZLL                                                      C S-           LEA                  LLtSW                                                  T        TTIO                 R          ONZE                           0( (                                                        L B
  • EOR 4 LAeE .LNN GE FEZNH N LL
  • sVCF0 ETsRS3 HLI IEENHL9 RONC P e .S P AE t

e 0 HO .PS1 AOOLLITL6LENAA ORS E O RH T2o MW O .

  • RPPDDL E .AF HE RE- L R TS5 hTcMLO0. ST0 W I eOIL= e eR4 sC eEDDNNUeH4TFLC sVI I UUOsS1 OI L T DDV .

NT 2 Z D E7 MO8 s e eOMMBDFe TDETARYEURN 0 RYE AT6 t.N C U s M F F A HA ETR E NETR I - .

                                                                                              ..                                             S       RTI UTE                              TI U                               DE0

_s D0 O3900O se 600 90 1 ) TEECSHLL sEBMOSCZL EECS BMOS eF L sSE - UALE oF .AUDRI0 . . 60 0 5F eLMALEIZE t IEU0T .U 76 21 .0 9 - nNUIERE0H TIER sA

7. eI NDVPHNS eDVP eB
        ~    STO EAI0O
        ~ RWL5 3 1

0 21 2 2 2 0E

                                                                                                                . C D

U 1 ( - 090 0OO) S .&0 2 440000

         -   TR         1///// //                                                       /        7            =               -              05 .            1      NL            &L1 0                5         633303 ES                         1094                              .         7        9                 )          -              0S&               .        L           . .72                .       6 .234 DVN                 00 . 23                                               2                    1 7              -                .S&           t          E        3C . .                1         1960 .

. LIE ORB N 30

                                         .6 44 g            ./6 504 09 09 4

4 3 5 O. N S H 2 O1 09 00

                                                                                                                                                                                                                           . 1

.MC'C-I 91 21 S9 0 . 0o EI E8

                                                                                                                                                                         /            T                             )

H O I

       .-                                                            7yA                /

9 t)

                                                                                                                . t          S E
                                                                                                                                             ) ) )

L .TL ZNAt I

                                                                                                                                                                                                       )

T) )T) O . .AS BN'NRE _ s .t- 36.- t 1 G 2L ZIL& l N . 0E D0 O74O00 e ee 6 1 00 13

                                                                                          .                  )

3 9 A S NTL IFE 0(Pt N 2 A! It N I Lt It L ( ( 0R E . M DRI0 . . 000 6 0 . S *( ( H /3T1 D t ED 5G IEU0T 99 .05o2 )

  • 0O E
  • S A6R1 HER 7E
                          .U           44                                  6 .2s2                F*          0t              M            *          /       E .O                 T)                             SVE 9D

.E - STO LAI0O 11 410s ) *=0)

  • 2 R0P/ CA P OTS .(

S 3 0e) RO* .7 G T A P1 PE34 A RMES0 A~ELWL0 3 1 . FHE I 1S N N F U5 ( R47 L EEME N . C HN 0 / D D( I O ( S3e A24 R PRAN R

9. N A D 2sT E s I M1 E

.~ 2U( EL 1 3 I SIS 9 . MN 97 TT CO) 4 R T EA NSD2eUT20. UDA V O eTSDC . I T eNE NT . ~ TEE

          - ODS               N   00 . C1 5S
                                          .&                       .                  AFBENC0 T           CE/0 A

W ADRF MIAR OOE/

                                                                                                                                                                             *CC43 P                      sUDEH!A
                       .3////uM./
          - H                 I        61                                             AR6NRT0(                                           R           AI RRH9NE(                                         E        OUDTt P 1                                              DHEEEO                                  R          OLL               .              3OL                           L     NCTI               H OE SLCE
          -                                                                              / 0RFH1 = (                          O           FLA .OELR1IF .                                                F S                                                                                  EU   .EF(                               F           NETFEPAA7TFO123F                                                    IBTTLTR E-                                                                                                                                    IHOFDYEE3AAN                                                   A     TUCUAIU Q_ C I         )) )                         )           ) ) S      CTiFI                         )

L- R F UNBt FD3I 1 A S STET TSR MB B ATPOWFT F- H

                        /
                                   ~FF^

tCO- PE

                                                   !           E                   tat 2t AMLDDE/3L 01T                   E G

N O R O M R F- IT12OSMODF .8000S ON7e06I A- B EE= T / RAE RF9E A A- L D D > _ t_ I/ F FOUIEU( . D S TE .EANNNO 0532 FI1032N 1

          - s                          (f                                             FTFRT                   Ot              S           CC            0YE               UIE              3776               NA47 e         0                                       &           RR2RCI      UA)                                   E           U             9        R        3        S          . . .           I L .8                .8

.S- _W_ HOTEACTRS t

                                                                                                                    )         M           R       L              T        /E               9990,                  P09             1

.T- e 0 s 0 N - O / TFP/EAEE F T E T5 YL . 222 E 21 . . EA S- - 1 I I F L s Le UC TARLASEMF NPRPL t M A S N L L EEF N- RFL DFD R U R C R O AELS /AD T T A u. SFHLE/ PEF I LT 8 ( ALDMTA D G C RRGE TTB SI I I) ))) 9 . U BD A R O

                                             / /

E Lu I

                                                                              - Y - RCEE E/TTN(

D T M O R L AUNP0NE PTAP I We S . ANMNM

                                                                                                                                                                                                 .  .  . e s

TT . FFN) - 1 OS D P A A A_e T EEVDA A L P L O sLPII II s a t ( I (I 5 N8 ..HE S NE N V GcT R E Om EAN T FFCOOLLE) OSA EULME RCA R (

                                                                                                                      =

E E E H SE 11OTP I NESJNeELt AINI O

  • tL.L EGGGS t t t S HH /

TTTT4 GGEU0 . T - - C O OT O I N- .m C RST LR U S P TLROT AHNNNE .I WSR IA P o .AE T D eYS APTI A RSIIIN NNEO3 TT TR T . OI MEI *FCVs( T - *TELT SSM A/CCCM EEH(

.LL                                CEE RM                                   LFS*E                  Oe                M
  • ILNT NI PSAAAC LLS wLUoEDW I

ARR D I SE*O . e= - eLREENLAT EPPPI EOL o ESoMNO RUU FNR* CYES E LEHIEAPS SSSSH E rRIA HEsAOL FTV ET A TV R V -  ; ESSRMENE S AL T P YLITAE vUTT SReNCF AA ERG LUI O - MSS OESE I Rr PA TE TLT RM u TRR Ui GTN LDT T T - G T DDL MEE C A I A C C C - LELNSLH $RTEL ACL -- t IA GPP O RT s RTE A E - ALSLIRLG ESTELF EREAAE

e. IUUT I MME L EAU EAF F F - MESEPIEI PONLTF BEFTEB LLO EEE RE VEO VEF - F - EHAHMAHE YRENUA U9FORU WTT7 V AHF OHE F E - TSPSIPSW TCCI Og TOETAT s .. F F.T aE.1234S676tSt23456789012345678901 2343678901 23436789Ot2 nI - C000O000Of t 1 1 11 111 1 222222222233333333334444444444SSS f -

ll?L CALCULATION / ANALYSIS REVIEW FORM CALCULATION NO. VYC-1628 REVISION NO. O Page Ar6 um K l COMMENTS RESOLUTION Appendix K Section K.1 states that the selection of the A loop will Words added to text to state that GE yield results "with in the necessary accuracy for this Specification is conservative and will bound calculation" compared to the B loop. What is the expected losses. necessary accuracy? Section K.3 identifies portions of the spreadsheet as Spreadsheets re-run with shading tumed on. shaded areas. The shaded area are not differentiated in a photo-copy, will they be differentiated when the calculation is filmed as a permanent record? The reference isometric drawing apparently contains This enorin the drawing has no impact on an error in the vertical dimension of approximately 10 results. All cases run from a common inches for the section of piping examined. Is this elevation reference point. (VY Design significant to the results of the calculation? engineering notified of apparent error.) The physical location of point Pout is not clear. Equations modified to include branch losses. I Equations K-3 and K-5 include losses up to the tee where the two paths recombine. Would it be more l appropriate to consider Pout at the exit of the tee (through the branch) thus accounting for the losses in the tee itself?

    'The friction loss in the vertical section is not included    Friction losses were included in development.

in the development of equation K-3. A similar Reviewer concurred.

    ~

condition exists in the development of equation K-4.

 . The table of pipe lengths for the bypass branch does       Pipe lengths included accounting for elbow not reconcile with the references isometric drawing.        radii. Reviewer concurred.

Total Flow is expressed in gpm and since the Statement added to identify gpm is inlet gpm. conversion to lbm/sec is dependent on density the

      " location" of that flow rate must be identified. Is there a relationship between the flow definition used here and a flow measurement in the plant?

Table K-1 identified spring hangers as Tee's for Done. All cases re-run. losses. These should be removed. Identify method (s) of review: Calculation / analysis review O Altemative calculational method O Qualification testing Resolution By^: /brw/Mb "/"I 7 Pieparer/Date Comments Continued on Page: ^'W Concurrence with Resolution Nydm. "/#/M l f/ Reviewer /Date i FORM WE-103-3 Revision 4 1 ( L  !

                                                                                                                         //77 CALCULATION / ANALYSIS REVIEW FORM CALCULATION NO.                               VYC-1628   REVISION NO.         O                     Page AntAdie K COMMENTS                                  RESOLUTION The calculation does not verify the calculation of the     Verification added.

density used for the heat exchanger loss. (Average of inlet and outlet densities.) s a- i Identify method (s) of review: [ Calculation / analysis review D Altemative calculational method

        . O Qualification testing Resolution By": t MIb */O 2 Preparer /Date Comments Continued on Page:     N+

Concurrence with Resolution 86~/o4 4/.2#97 V Reviewer /Date  ; i FORM WE-103-3  ! Revision 4  ! I

l l Torus Temperature and Pressure Response 12 Larg2 Break LOCA cnd MSLB Accident Scenarios Page ' t ,e VYC 1628 Rev.0 AppcNid. MEMORANDUM

     /

DE&S-BOLTON To L R. Hoffman Date Anril 27.1998 Group # THS AG-VY 98-063 f From C. D. Fago W.O.# Subject Completion of Long-Term Containment Analysis I.M.S.# Model and Method File # 98-063-1.WP REFERENCES

1. DE&S Calculation VYC-1628 Rev. O. Torus Temperature and Pressure Resporlse to Large Break LOCA and MSLB Accident Scenarios, approved 04/27/98
2. Service Request PM-2219, Complete the Evaluation of Suppression Pool (Tonis)

Temperatures

3. Service Request PM-2488, Torus Temperature Issues
4. Work Release Form 98-0079-00 Torus Temperature Analysis
5. Memorandum C. D. Fago to J. R. Hoffman, THS AG-VY 98-064, Torus Temperature N Margin Assessment for Confirmatory Analyses, dated 04/27/98
6. Memorandum D. E. Yasi to J. R. Chapman, OPVY 491-97, Design Engineering and Operations Input to NED's Containment Analysis, dated 12/29/97

SUMMARY

I j

                                                                 ~

A long-term containment analysis model and method has been developed for Vermont Yankee using the RELAP5YA and GOTHIC codes. Using this model and method, a set oflarge break LOCA scenarios were analyzed to evaluate shon and long-term peak torus temperature. The  ! model, method and scenario analyses are documented in DE&S calculation VYC-1628 Rev. 0 [Ref.1). The analyses were requested to support a revised Technical Specification for initial torus temperature, margin assessment and design basis reconstitution [Ref. 2,3 & 4]. j DISCUSSION VYC-1628 Rev 0 [Ref.1) documented the development of a long-term post-LOCA containment analysis methodology for Vermont Yankee, a BWR with a Mk I primary containment. The method uses a RELAP5YA vessel and core model for the shon-term mass and energy release calculation and a GOTHIC containment model for calculation of the containment response as well as a GOTHIC vessel model for long-term mass and energy ' release. The method and models were benchmarked and validated as being adequate for conservatively calculating the long-term containment response to a large break LOCA. I i k____

Torus Temperature cnd Pressure R:sponse to Large Break LOCA and MSLB Accidtnt Sc:n:rios Page //5 VYC-1628 Rev. 0 Appendix 1.'

                                                                                 ~

J. R. Hoffman April 27,1998 Page 2 A set of cases with differing single failure assumptions and input sensitivities were analyzed to examine the post-LOCA toms temperature response. The calculation does not necessarily present the limiting toms temperature response based on requests from Vermont Yankee staff to examine additional scenarios. These additional scenarios have been addressed via an engineering assessment [Ref. 5) and the completion of the confirmatory analyses will address the limiting case. These results provide suppression pool temperature responses which can be used to determine the replacement ECCS suction strainer NPSH performance. The methods and models can be used to analyze other scenarios as required or requested by Vermont Yankee. RESULTS Tables 1 and 2 provide the results for the nine cases analyzed. Both the peak shon-term and _ peak long-term temperatures are provided. The peak shon-term temperature occurs at 10 minutes after the stan of the accident coinciding with the assumed operator action to throttle the low pressure ECCS pumps. The time of peak long-term temperature is provided. As stated previously, based on the sensitivities examined, the bounding peak toms temperature has not been determined with this set of cases.

     ~

Common assumptions for each case include (except where specifically addressed otherwise): W Breakis on Loop A 2700 GPM RHRSW flow through the active RHR heat exchanger

  • 6400 GPM RHR flow through the active RHR heat exchanger
  • Both recirculation line discharge valves shut normally Mechanistic feedwater model using nominal normal feedwater flow rate ECCS pump heat added for the operating ECCS pumps Suppression pool initial conditions of 90'F and 68,000 ft8 water volume
  • 1593 MWth (+2% uncenainty) initial reactor power ANS 5.1 - 1979 decay heat (+2a uncertainty)
  • Note that these are analytical inputs and that plant operability limits will need to be established such that the analytical input values remain valid considering uneenainties (see
                    " Assumptions", below). [Ref. 6]

1 L,____________._._ _ . _ _ _ _ . _ _ _ _ - _ - - _ - - - - - - - - - -

             >ygQ*.$00        .

Tjw . f c e e e c e e c c o k e e e e e e f e e e s s s s s s o k s s s e a 5 5 0 0 5 ea 5 e me iP 9 6 0 1 3 9 0 2 8 5 5 7 2 e iPme 3 7 4 3 6 8 2 r 5 u T 8 1 8 1 7 1 7 1 6 1 7 1 r tu T 9 1 9 1 9 1 ta r a r ep e p m m m e r F F F F F F m r F F F e Tr Taek ek 3' . 7' . 7 2 '5 8 Tr Tage 2 3 3 ge 3 5 5 6 5 8 1 te nP 7 7 7 7 7 7 te nP 8 1 8 8 1 a o 1 1 1 1 1 1 a o 1 1 1 Ws L Ws L u u r o m) r e in F F F F F F r o m r )n F F F T ' * * *

  • T e '

Tm Tm i 5 8' . 6 0 0 0 7 t 4 9 8 t 1. 2' . i 4 r 0 5 1 l 1 r 0 2 2 o1 5 5 6 6 6 o1 6 6 6

                        'l h (                      1          1         1           1          1 h (        1              1                1 S                                                                                                      S mm pp gp mm pp gg m

p g ppp g mm mm gg pp gg

                                                                                      @la mp t o    g
                                                                                                                                   @ al I

t top m g

                                                                                                                                                  @la I

t o gp m @lta mp I o g I t C Ct Ct m )n ri 00 00 00 00 00 04 0 0 0 0 0 4 00 00 04 00 00 04 Cm0 P Lp 0 m r )n e i L pg P m00 P L p m 0 0 P m00 L p Te m 4 37 36 3 6 4 6 46 g64 s Tm 2 6 2 g 4 6 2 g6 4 0 1 0 ic 0 0 0 g 1 n o> SR

                                          @@ S@R@

SR SR SR

                                                                   @@ @@ &00                 0 i

t v g0 n1 o

                                                                                                                                   & 0@   8
                                                                                                                                                  &04 @ &0@              8 S1 R                                         S 8 R          S 3 R            S 8R CH                                                                                              L (>

i L( CH C HR CH RCH I CI H i t C 1 H C2 H C1 H s I R I R 1 1 R I R s n s 2 R 2 R 2 R w lo I I 1 I I 1 S e w o I i i l F d F S mm pp m pc

a. m pp m m pc
a. m pc
a. m a. n S a.

c pami n a. m in a.mi n ( s C pc a C c pa c pa ic C 00 gg gm 80 gg gm gm gm e r C mgt r mgt r mgt r iv E t 8 p 0 p 0 p 0 p E p4h p0h p4 h 00 0 g 04 0 g 0 g m )n 64 7 76 6 6 0 g 6 lu g0 1 c a g0 c g0 c i ri 47 3 04 3 5 4 00 4 00 4 0 i a m )n 8 i e 0 2 a 4 e 8 1 1 c a it e 0 ri s n Tm 0 @@ @5 6 @ @ @7 @7 @7 4 4 4 F e Tm 0 I 7 01 6 0I 7 e t r SI S S I S S S e t 0 3@ 4O 3@ S d o 1 h0 S( CC P C@ C C C@ I P I C@ I C@ lg n o r h0 1

                                                                                                                                  @      I C       @C     I
                                                                                                                                                                  @C    I n

I L I PC l L 1 C 1 C 1 C I i S( S P SP SP r I I P P P S C L CL CL L L L L r 2 4 24 e e 24 r 2 2 2 2 g lu n f a c h F ) n c x c e ig r e p E ) n ty o d t w i ta w )n e i s l e a o S v l

                                                                                )          i             c                                             l d p eo i

it f ia ) m f a I I f

                                                                                                                                                                        )

1 e s s e w p ht n te

      -                         f l                  n v               o         g                    R                                           tyo                     a C                             ici   a              e   l l

f 0 i w I it v c i r D t f is S a n 5 I ie w n e v e v 8 tymo R t j o s, al v e io 2 e is n n t eh ca v l a g t c s iv o s, r ni l u s r e t lu eS d s e it i c ae e C m. gr V e hc a tyin j ityi w i r ) it n n iprteae s l u s l e t ia SC eC f e e R T en s oh c a gis r S v o iG e e i T F r tuE e a e ad ivc it S mp R r e a tyl b 0 B ed n. si r u h c re iC t sW n e u o 0 g r e a

x. i v o u u po n

v er u e li isic isE eS r

                                                                                                            .                      a              pm             i d e                   ke ila ic    r    a        De r        n x e a        SR        l Rt    s      v                        h                mh t

ish R nf e r F np S wH iaHe F R e c x et nt 8 a e Yl p e io o w m lor p Ru n E Ti w ei S w 2 go lg t o t Ft u t i 8 mt lal lou mut n t 6 n n 2 a r u wt u 1 tnis lo o n i S u n F b c e I Wb S 4 ub o P 4 c 6 1 l e eb T 3 lob C m1- e 1 r k S - I F3 ic RnT e br o C Cn Rn HuR t rCk c - t C R r n d n Y e C u Hu R H ou e u V V(b Dr D R( E( R R( R( R R Y R h S( R F(eR

     -                                                                                               V-1 e                                                                                     2 ic           s        0               t          2          4           5         8                  e              e          3 a         n                n         n          n           n          n                               s n             6               7 b

l a n - n c C R u R u R u u u u b C u u u R R R a R R R T T FI

Torus Temperature rnd Pressure R:sponse 12 Larp Breck LOCA and MSLB Accident Scenarios VYC 1628 Rev. O Page_ fig Appendix L J. R. Hoffman April 27,1998 ' Page 4 ASSUMPTIONS A variety of assumptions were required to be made during the course of the evaluation. These assumptions were made after consultation with, and with the concurrence of, Vermont Yankee staff. The results of this analysis are contingent on the following assumptions:

1. A minimum RHRSW flow of 2700 gpm to the RHR heat exchangers is provided -

within ten minutes after a LOCA event.

2. A minimum long-term RHR flow of 6400 gpm to the RHR heat exchangers is, provided within ten minutes after a LOCA event.

L

3. Core Spray pump NPSH limits are provided in plant procedures which result in long-term core spray pump flow of a minimum of 3000 gpm and a maximum of 4000 gpm.

SERVICE REQUEST COMPLETION . Completion of VYC-1628 Rev. O represents partial completion of the following service requests. These service requests will be considered complete upon completion of all planned confirmatory analyses (Ref. 4 & 5). PM-2219. Complete the Evaluation of Suppression Pool (Torus) Temperatures PM-2488 Torus Temperatureissues SR-2890 Incorporate Instrument Uncertainties Associated with Torus Water Temp. Ind. Into the Containment Analysis (ER 97-0716) SR-3005 Feedwater Temperature > FSAR/ Operating Procedures SR 3006 Heat ExchangerThermal Performance CONCI'USIONS The long-term torus temperature calculation, VYC-1628 Rev. O, has been completed. The results of this calculation do not necessarily represent the bounding peak torus temperature, L and follow-on confirmatory analyses are required to document the bounding case. [Ref. 5] 0

                                                       ' Completion of this calculation represents partial completion of WRF 98-0079-00 [Ref. 4] and

( the associated service requests. 1 f-

l Torus Temperature and Pressure Response to Large Break LOCA cnd MSLB Accident Scenarios VYC-1628 Rev. 0 Page f]t { Appen n I. 1

1. R. Hof(man i April 27,1998 Page 5 j RECOMMENDATIONS Validation of the above three assumptions is required prior to this calculation being used in the design basis.

Prepared f/37/53 C. D. Fa'godngineer Date Thermal Hydraulic and Safety Analysis Group Nuclear Engineering Department I Reviewed' in 4/27/98

                                                             \\J.E}Metcalf                                                                   'Date                    .

Polestar Applied Technology f i Approved -- d!2-7!Y

  • J. . ChaQneral Manager 'Dat/

clear Engineenng Depanment Action Taken (if any) i Recommendation Accepted Recommendation Denied Comments: Signed: Date: cc: DE&S VY- Bolton VY-Brattleboro VY-Vernon S. P. Schultz J. Riley D. E. Yasi D. A. Reid K. E. Bronson P. A. Bergeron M. F. Kennedy J. Lynch D. Leach K. Oliver L Schor M. A. Sironen J. H. Callaghan R. Wanczyk J. Kritzer K. J. Burns t1, E. Ter,mx R. G. January G. Sen M. P. leFrancois P. A. Rainey T. Silko W. S. Yeung E. J. Betti - J. S. Hsieh E. F. Goodwin B. C. Slifer __--_ j

i i i Torus Temperature and Pressure Response to Large Break LOCA and MSLB Accident Scenarios VYC 1628 Rev. 0 Pags M

                                                     ,                                                   Appendix L MEMORANDUM i

I DE&S-BOLTON

                                                                                                                                              \

To J. R. Hoffman Date April 27.1998 Group # THSAG-VY 98-064 From TFD. Fago ' W.O.# Subject Torus Temocratum Margin Assessment for LM.S.# Confirmatory Analyses (Reissue to Correct Small File # 98-064-1.WP J Break Marcin Assessment) l I l REFERENCES

1. Memorandum C. D. Fago to J. R. Hoffman, THSAG-VY 98-063, Completion of Long-Term Containment Analysis Model and Method, dated 04/2768  ;,-
2. DE&S Calculation VYC-1628 Rev. O, Toms Temperature and Pressure Response to Large Break LOCA and MSLB Accident Scenarios, approved 04/2768 ,
3. Memorandum S. J. Ganthner & M. P. I.cFrancois to J. R. Chapman, THS AG-VY 98-068, Engineering Evaluation of Abnormal Operational Transients for Peak Torus Temperature, )

dated 4/24S8 (Attached) pagigssi j 4 PURPOSE This memorandum provides an engineering assessment of the maximum expected peak torus , temperatures for use in supporting revision to BMO 96-05. This assessment is based, in part, on  ! engineering judgement and represents preliminary information from on-going analyses. DISCUSSION l Memorandum THSAG-VY 98-063, dated 04/2768 (Ref.1), documented the completion of a l peak toms temperature methodology development and analysis effort for large break LOCAs, VYC-1628 Rev. 0 [Ref. 2]. The results demonstrate that the peak torus temperature is less than 185'F for the analyzed scenarios, including scenarios that include assumptions much more conservative than the original design basis assumptions. The analysis determined the maximum torus temperature occurring in the first ten minutes post accident as well as the peak toms temperature occurring approximately five hours after the event (depending on the specific scenario examined). The ten minute time frame is ,, defined by the time at which operator action is credited to take action to establish containment cooling and throttle ECCS pumps in accordance with EOi ,,uidance. To support a full assessment of the bounding tems temperatures and a revision to BMO 96-05, the analyses documented in Reference 1 must be supplemented to assure that I

I L I j Toms Temperature and Pressure Response to Large Break LOCA and MSLB Accident Scenarios VYC-1628 Rev. O Page //N

Appendix L i
1. R. HofEman April 27, I998 Page 2 le other events do not exceed short-term and long-term torus temperature limits. These other l events that add heat to the suppression pool include:

Lfm l ,, .1) limiting large break LOCA and main steam line breaks i 2) small break LOCA and small steam line breaks (i.e. breaks smaller than double-ended guillotine recirculation or steam line breaks)

3) Appendix R safe shutdown scenarios
4) station blackout safe shutdown -

l 5) AOTs involving SRV operation 1 The' assessment of the peak torus temperature for these confirmatory events will be l characterized by a margin term to be added to the Vermont Yankee-specified base large break LOCA analyzed in Reference 2. The peak torus temperature for the Vermont Yankee-specified base case large break LOCA is 173.3'F as documented in VYC-1628 Rev. O. W f l Including the margin term, a short-term maximum torus temperature of 164'F and a peak ' torus temperature of 185'F has been determined by a conservative engineering assessment to bound the short and long-term peak toms temperatures for the other events listed above. The j engineering assessment of the margin is attached.  ; The engineering assessment for AOTs involving SRV transients is attached. Reference 3. !C Other events involving SRV operation are included in the other assessments (small break LOCA, Appendix R events, station blackout). ", 4 i ASSUMPTIONS j The analysis on which this margin assessment is based, Reference 2, made several key assumptions. These assumptions are detailed in the conclusions of Reference 2 and reiterated

in Reference 1. Results in Reference 2 and this engineering assessment are contingent on the validation of the assumptions prior to using these results in operability assessments. These - -

assumptions were made after consultation with, and with the concurrence of, Vermont Yankee staff.

1. A minimum RHRSW flow of 2700 gpm to the RHR heat exchangers is provided; except l after a LOCA if offsite power is available at which time a minimum of 4000 gpm is ,

l- provided. I 1 i' . 2. : A minimum long-term RHR flow of 6400 gpm to an RHR heat exchanger is provided l using one RHR pump, e.

3. Core Spray pump NPSH limits are provided in plant procedures which result in long-term f core spray pump flow of a minimum of 3000 gpm and a maximum of 4000 gpm. I l

n l ________._______A

Toms Temperature and Pressure Response to Large Break LOCA and MSLB Accident Scenarios Page //J( VYC 1628 Rev. O Appendix L T J. R. Hoffman

              ' April 27,1998 Page 3 CONCLUSION                                                                                                                                                                    ~

Based on review of completed torus temperature analyses and issues presented in ER98-0025, the peak torus temperature following either a LOCA, Appendix R fire event, station blackout event, or SRV transient event is bounded by 185'F. For a large break LOCA, the maximum torus temperature expected within 10 minutes is 164'F. This evaluation is based on analyses completed to date, preliminary results from on-going analyses and engineering assessments. As such this evaluation may be used for operability determination purposes as allowed by NRC Generic Letter 91-18. Prepared IMs C. D. Fago, L4gineer ate Thermal Hydraulic and Safety Analysis Group Nuclear Engineering Department N b~' Reviewed hI f/2'i/'i2  : J."E. Metcalf Date yolestar Applied Technology Approved V 4A 2 # J. . Chapman, Gheral Manager 'Dat/ uclear Engineenng Department ATTACHMENT: THSAG-VY 98-068 cc: DE&S VY- Bolton VY-Brattleboro VY-Vernon S. P. Schultz D. E. Yasi D. A. Reid K. E. Bronson J. Riley J. Lynch D. Leach K. Oliver P. A. Bergeron J. H. Callaghan R. Wanczyk J. Kritzer M. F. Kennedy R. G. January G.Sen L.Schor P. A. Rainey T. Silko K. J. Bums E. J. Betti ^^' M. A.Sironen M. P. LeFrancois W. S. Yeung i J. S. Hsich

   '...             E. F. Goodwin                                                                                                                                 -

B. C. Slifer l I t I I  ! l E ______ __ --

l Toms Temperature and Pressure Response to Larg3 Break LOCA and MSLB Accident Scenarios - Page //A VYC-1628 Rev. O Appendix L F L R. Hoffman April 27,1998 Page 4 t Engineering Assessment of Peak Torus Temperature Margin Addition - u

1. Introduction
2. Limiting Large Break LOCA
3. Small Break LOCA
4. App. R Events
5. Station Blackout
6. References
1. Introduction To support a revision to BMO 96-05 an assessment of the peak torus temperature for a set of design basis events is made. Using a detailed and improved methodology, the peak torus . . ,

temperature following a large break IDCA has been determined to be 173.3*F for a base case . "# with parameters specified by Vermont Yankee [Ref.1, Section 3.3.1.1]. Whereas this base case does not capture the peak torus temperature for the limiting large break LOCA nor necessarily bound all small and intermediate break LOCAs/ steam line breaks or safe shutdown events (e.g. App. R events, SRV transients), a conservative engineering assessment of the potential peak torus temperature is made for these other events. The assessment is based on evaluation of the issues identified in Vermont Yankee ER 98-0025 [Ref. 2], existing calculations of record, hand-calculated verification, and preliminary analyses. O 2. Limiting Large Break LOCA Peak torus temperature has been analyzed for a set oflarge break LOCA scenarios [Ref.1]. In addition, other large break LOCA scenarios, including main steam line breaks, with assumptions made to allow greater operational flexibility (and which have the potential to be more limiting) are currently being documented. Factors which are being addressed by these additional large break LOCA scenarios are: canbined worst case feedwater flow rate and ECCS flow rates

                                    =

injection mode cooling

                                    =

increased fuel stored energy due to GE-13 fuel Run 3 of Reference I resulted in a short-term (10 minute) temperature of 161.7*F and a long-term peak temperature of 181.2*F. Run 3 represents a large break LOCA with offsite power. available (feedwater continues to inject) and one RHR heat' exchanger fails. Run 6 and Run 7, sensitivity cases based on Run 3, show that increased feed flow and increased ECCS flow, . , respectively, result in increased torus temperatures. Table I provides the results and brief to " "' l scenario information for these three runs,3,6 & 7. Additionally, Reference 2 provided an f engineering assessment of the impact ofinjection mode cooling. Table 2 provides the (. summation of these impacts relative to the Run 3 results. It should be noted that increased ! feed flow, increased ECCS flow, and injection mode cooling all serve to increase peak pool ie temperature by increasing the stored energy removal from the vessel. Therefoce, it is conservative to treat these effects as additive. 4

- g 1 t Ea $ 8 ait tf E" Een .

                 ?e* F .                                                                                    >!IaaD5a u skN
                                                                                                                       > .E'
  • t
             " N. m%'I

- >b.G" s Tku f c e c o k e s e e s s ea 5 4 8 ' me iP 3 7 3 6 2 5 e r T 9 1 9 1 9 1 u i. y . c r m r T e ek Ta T 2 T 3 7 3 r ge 1 1 te nP 8 8 1 8 a o 1 1 1 W l s u r o m T re )n i 7 7 T Tm t 7 1 2 2 2 4 or 0 6 6 6 h (1 1 1 1 S elam top elam op t 9 lam I g I g I top s Ct Ct Ct Pm0 Lp0g 4 Pm0 L p0 4 P m00 m )n r Lp 4 e i 2 6 2 g6 2 g6 0 Tmg0 &09 0

                                                                           &09 0
                                                                                           &09 0
                                     .                          8                 4                 8 v             n1 o>        S 8R             S 3R            S 8R e           L(         C1 H             C2HR            C1 H R                        2        R I

2 I 2 R I 8 s 2 w 6 l o 1 F S sh n C sh n ah n C e ci a a e ci a a c a cia Y C E met pm r met pm r met r V sp g p pm sy m )n 8 s 0 g 8 g m ri 0 4 0 0 0 4 o r e Tm 7 3 0 1 6 0 4 2 7 3 1 0 f t 0 1 4 s r 1 o- 91 91 91 1 l t h0 S( Se C Se C S9 C u s 2C I 2C I 2C I e P P P L L R 4 4 L 4 e r

                    .~

t u a r e p

                                                                                )

m w e lo T f n ) s yo te u r it ti r a vc o ie t j lo w T e r ins n i f t u eS d c l i SC e e e F a eC r f l e r aE t ic S e iyb

e. xa. u t

g f n pm vo ' , o h a

c. h i

it d sh y c x et nt r E Ti w ei S w a t mt wtu a r u m e eb lob m H T 3 F3 t u R r n d n ou r S H R hR S( e u eR F( - I l e 3 6 7 b n n n a u u u R R R T I! lli lI I 1 l I l

Torus Temperature and Pressure Response to Large Break LOCA and MSLB Accident Scenarios Page1/Ja VYC-1628 Rev. 0 l Appendix L J. R. Hoffman

                .          April 27.1998 Page 6 l

This assessment also presents the impact of a potential increase in RHRSW flow above 2700

                                                                                                                                              .~

GPM used in Reference 1. When offsite power remains available, as is the case for Run 3, the " RHRSW flow would be increased to a minimum 4000 gpm thmugh an RHR heat exchanger.

                   ' Based on preliminary analysis, if RHRSW flow was increased to 4000 gpm, the peak torus temperature could be reduced by about 4.2*F. 'Ihere would be no impact on the short-term temperature as RHRSW flow only affects the analysis after 10 minutes.

l ' Loss of offsite power (LOOP) would reduce the amount of feedwater that would be injected

                  . into the vessel due to the loss of power to the feedpumps. However, some hot feedwater could                                         {
                  . be injected into the vessel through flashing and potential carryover as the vessel depressurizes -

below the saturation pressure of some feedwater. This effect has not been mechanistically .,

                  . ' assessed. Therefore, a bounding approach would be to include only the hot feedwater present                                       '

in the feed piping system as was done for the VY-specified base case. ER98-0025 assessed . j the impact of the additional hot feedwater due to no LOOP as +2.2*F. This, then is subtracted - M. . from the results that include mechanistic feedwater addition. An additional column is presented in Table 2 assessing the conservative long-term impact of LOOP on peak torus temperature. The short-term impact of LOOP is not further assessed as the impact with no  ? LOOP is bounding. LOOP is significant only if the additional RHRSW flow, mentioned I above,is credited. ' , g An additional evaluation is provided for the potential additional stored energy due to the i intmduction of the GE-13 fuel for the cycle 20 reload. An evaluation was perfonned using i bounding Cycle 20 fuel volume-averaged temperature for GE-13 fuel. The stored energy  ! increase due to the OE-13 fuelis less than 1.65 mBTU. Distributed into a 4.8 mlb pool, the

potential temperature increase due to this energy addition would be about 0.3*F.

1 Table 2 - Limiting Large Break LOCA Peak Torus Temperature Issue STImpact LTImpact LTImpact No LOOP No LOOP IDOP t Feedwater Flofv Sensitivity (A between Run 7 and Run 3) 4.7'F 4,1*F -2.2*F ECCS Flow Sensitivity (A between Run 6 and Run 3) +0.5'F +0. l *F +0. l *F Non-Limiting Containment Cooling Mode (injection mode) N/A 4.9'F +0.9'F ' Cycle 20 reload with GE-13 fuel . 4.3*F +0.3*F d.3*F ( Net PctentialIncmase above Run 3 +1.5'F +1.4*F -0.9'F

    ~~
                                                                                                                                               ';f.*

l Net Peak Torus Temperatum 163.2*F 182.6*F 180.3'F

                                ' Potential Credit for RHRSW Flow of 4000 gpm N/A                        -4.2'F    N/A
  ,                ' Peak Torus Temperature w/RHRSW Flow Credit                                163.2*F   178.4'F
  <,                                                                                                               180.3'F Required Margin to Bound VY Base Case (173.3'F)                              N/A       +5. l *F' +7.0*F I
 =                  _ _ _ _ - - _ _ _ - _ _ _ _ _ _ _ _ _ .

Torus Temperature and Pressure Response to Large Break LOCA and MSLB Accident Scenarios YYC-1628 Rev.0 Page f)f_ Appendix L

      ^

J. R. Hoffman

                                         .. April 27,1998 Page 7
3. Small Dreak LOCA
       ) . . -

Peak torus temperature for LOCA events smaljer than large break has been analyzed

  • previously with apeak torus u .

w.iore of175.4'F. (Ref. 3] Analyses are underway to re-analyze the peak torus temperature following a small break IDCA usingthe new methods and models developed in Reference 1. In lieu of new analyses, the issues identified in ER98-0025 are conservatively evaluated for their potential impact on the peak torus temperature followi a small break IDCA.- The aseecamert is made relative to the limiting DC-1 single failure case (Run 4 - 176.2 'F peak) which has long-term equipment operability similar to that expected after a small break LOCA (long-term one RHR pump and one CS ' pump). The issees that are enumerated in ER98-0025 are applicable to large and small break IDCA torus temperature analysis; however, the numerical impact is different between large and smaill P break LOCAs. Table 3 provides a list of the issues identified in ER98-0025 and the assessment of theirimpact on the small breakIDCA analysis. Following th- table a '~"> description is provided of the impacts of each issue on the small break LOCA analysis. The previous assessment addressed the impact of the ER 98-0025 issues relative to the VY-i specified base case. However, the VY-specified base case does not account for instrument i uncertainty on RHR flow through the RHRHX while in torus cooling. Therefore, a re- ' assessment is made relative to the limiting DC-1 single failure case (Run 4 - 176.2*F peak) - which has long-tenn equipment operability similar to that expected after a small break (long-

     .{                               term one RHR pump and one CS pump).

t

                                  . As stated for the limiting large break LOCA case, RHRSW flow may be increased to
                                    .4000 gpm when offsite poweris available. If offsite power is available, which would be the                                  ,

case when significant hot feedwater could be added, then the increased RHRSW flow would significantly reduce the peak torus temperature (potentially by up to 4.2*F based on preliminary analyses), resulting in a peak temperature of 175.2*F. As with the limiting large break LOCA, the effects of LOOP are assessed.

                                - loss of offsite power (LOOP) would reduce the amount of feedwater that would be injected '
                               ' into the vessel due to the loss of power to the feedpumps. However, some hot feedwater could be injected into the vessel through flashing and potential carryover as the vessel depressurizes below
                                   ===*=eedthe saturation pressure of some feedwater. This effect has not been mechanistically                                             ,

Therefore, a bounding approach would be to include only the hot feedwater present in the feed piping system as was done for the VY-specified base case. ER98-0025 assessed -

                                                                                                                                                                           }

the impact of the additional hot feedwater due to no LOOP as +2.2*F. This, then is subtracted

                                                                                                                                                                           )

from the results that include mechanistic feedwater addition. ' An additional column is . i

                             ' presented in Table 3 assessing the conservative long-term impact ofIDOP on peak torus                                             MR ;

teruer .ture. LOOP is significant onlyif the additional RHRSW flow, mentioned above, is credited.~ j q*- l

                                                                                                                                              .                             I 1

d i__ ____ ____ i____________________._________.__.___________

Torus Temperature and Pressure Response to Large Break LOCA and MSLB Accident Scenarios Pags //Yo VYC-1628 Rev. 0 Appendix L i

1. R. Hofiman April 27,1998 Page 8
 '1 Table 3 - ER 98-0025 Issue Impacts on Small Break LOCA Peak Torus Temperature b "..". "

Issue Impact . Impact - NoLOOP LOOP SmallBreak LOCA +1.3*F +1.3 *F Feedwater Under-estimated N/A -2.2*F Non-Conservative Single Failure Assumption N/A N/A i ECCS Pump Heat NotIncluded +0.2'E +0.2'F FROSSTEY-2 Enor N/A N/A Non-Limiting Containment Cooling Mode N/A N/A  !.iM*" Core Spray Line Break +1.6*F +1.6'F Cycle 20 reload with GE-13 fuel d.3*F 4.3*F - , o 3 Net Potential Increase above Run 4 (176.2'F) +3.4*F +1.2*F , ([', Net Peak Torus Temperature 179.6*F 177.4*F pl Potential Credit for RHRSW Flow of 4000 gpm -4.2'F N/A  : Peak Torus Temperature v./RHRSW Flow Credit 175.4*F 177.4*F Required Margin to Bound VY Base Case (173.3*F) +2. I'F +4.1 *F I Small-Break LOCA - VYC-1442 Rev. O concluded that a small break LOCA results in a peak torus temperature greater than a large break LOCA. The increase attributed to the small break

                                                       - LOCA is 1.3*F (difference between VYC-1290 Rev. 2 results and VYC-1442 Rev. O results).

Non-Conservative Single Failure Assumption - ER 98-0025 indicated that increased ECC

                                                       ' injection resulted in an increased peak torus temperature due to the additional flushing of hot
                                                       - fluid from the reactor vessel to the torus. For a small break LOCA, the ECC injection is limited to that which can be removed through the break. As the break size decreases, the L                                                           amount of ECC injection which can be sustained decreases. For the limiting small break, L                                                           which results in the longest delay time before vessel blowdown, the ECC flow is limited to i

one core spray pump which will ensure that a pressure will be maintained in the vessel. Therefore, the single failure assumption issue, while applicable for large break LOCA, is not applicable to small break LOCAs.' -Oa. ECCS Pump Heat Not Included - The ECCS pump heat inchded in Run 4 is applicable to small break LOCAs. For a small-break LOCA the number oflong-term operating ECCS 4'" pumps would be fewer as fewer pumps would be required to ensure adequate core and

                                                     - containment cooling. Based on the previous small break LOCA peak torus ternperature results                                                                         j

_ (VYC-1442 Rev. 0), the smaller the small break, the higher the peak temperature because of the delay in reactor vessel blowdown and the resulting decrease in integral RHR heat . i

Torus Temperature and Pressure ?- ; _ to Large Break LOCA and MSLB Accident Scenarios VYC-1628 Rev. 0 .

  • Page] Ni Appendix L L R. Hoffman April 27,' 1998 '

Page 9

                                                                                                       ~
 =

exchanger effectiveness.. Therefh, for the very small breaks with only one RHR pump n ,. o operating in long-term t' rus cooling and one core spray pump providing long-term vessel ,. make-up, the ECCS pump heat would be limited to the contribution from these two pumps'. Therefore, the long _ term pump heat credited in Run 4 is appropriate. J i t

        '.'          - The short-term ECCS pump heat contribution could potentially be from all six low pressure ECCS pumps, albeit while on minimum flow. These additional short-term pumps are not accounted for in Run 4 and would not be present in the long-term since the operators are directed to secure unnecessary pumps and limit the pump operation on minimum flow. The
                    . pump heat added by these extra short-term pumps is bounded by their full-flow pump beat.
                    ' Over the ten minutes of their operation (two RHR and one CS pump), they would contribute L      '                1.1 mBTU which, applied to the 4.8 mlb suppression pool, would increase temperature by about 0.2*F.                                                                                                                                                                   )
                                                                                               ,                                                                                       p. .  .

j yg

                   . FROSSTEY-2 Error -The net increase in energy due to the FROSSTEY-2 error has been l                      accounted forin the results for Run 4. Therefore, no additional penalty is applicable.

1

Non-Limiting Contamment Cooling Mode ; The cunent EOPs direct the operator to place -  !

RHR trains not needed for adequate core cooling into torus cooling. For the limiting small '

                   , break LOCA, the operating core spray pump is adequate to provide reactor vessel level above the top of the active fuel (TAF) satisfying the EOP requirements for adequate core cooling.                                                                     gq

{ i < Therefore, the operators would not need to use injection mode cooling as described in ER 98-0025 and this issue is not applicable to small break LOCA. ~ [ Core Spray Line Break - A special case of a small (actually intermediate) break LOCA is a l' core spray line break.' The operating core spray pump may not provide any substantial l ! injection into the vessel. For this case, the operators would be required to continue injection via an operable RHR pump. There are two potentially limiting scenarios for the core spray line break.' The operators would continue to operate the broken loop core spray pump since they would not have indication that the pump was not providing any core spray flow. However,.the reactor vessel level would not be maintained as the level would dmp due to J continued steaming. Therefore, the operators could choose to inject with an operable RHR pump on the side with the non-operable RHR heat exchanger, leaving the other RHR train in i t torus cooling. The other possibility would be that there are no additional operable RHR I pumps and the operators would choose to inject into the vessel via the operable RHR pump and RHR heat exchanger (injection mode cooling).

                ' Therefore, the two possibilities, which would increase torus temperature, are:
1. : one RHR train in torus cooling, the other RHR train maintaining vessel level using one
                                                                                                                                                                                      %f pump;
                - 2.

3, one RHR train in injection mode cooling with no other RHR pumps operating. For the first case, the additional pump beat would increase torus temperature adding l 14.9 mBTU over the six hours to the peak torus temperature. Added to the 4.8 mlb pool this i:

                . would increase the temperature by an additional 3.l'F. However, for the second case, if I

Toms Temperature and Pressure Response to Large Break LOCA and MSLB Accident Scenarios Page //vz-VYC-1628 Rev. O Appendix L i u I J. R. Hoffman ' April 27,1998 Page 10

                                          . injection mode cooling was used, this could increase the torus temperature by 0.9'F based on                                                             m ..

similarity with injection mode cooling usage during a large break LOCA (ER98-0025). The limiting scenario is the first case with the additional 3.1'F temperature increase.

                                                                                                                                                                                                   .       3 U"
                                        ~
                                          - This temperature increase would be mitigated by the additional heat removed by the RHR heat exchanger as the temperature across the heat exchangeris incrementally increased. Because the torus temperature profile becomes similar to a large break after the reactor vessel blowdown (just delayed in time), the mitigative effect of the increase in heat removal for a small break would be the same as for a large break LOCA.' Based on the evaluation provided in ER 98-0025 the temperature increase would be mitigated by approximately half. Thus, the toms temperature would be increase by about 1.6'F for the additional RHR pump.

Cycle 20 :eload with GE-13 fuel - In order to bound the Cycle 20 reload, the potential .. temperature increase as a result of the introduction of GE-13 fuel is included. The basis for this increase is discussed in the limiting large break section. WM 3 4 l I 1

                                                                                                                                                                                                      < y:3
                                                                                                                                                                                                     .b: $$ '

a

  \ ..
                                                                                                                                                                                                              'l

_ - _ _ - - _ _ _ - _ - _ _ _ - _ _ - _ _ _ _ - - _ _ _ _ _ - - - _ - _ _ _ =- -_ sToms Temperature and Pressure Response to large Break LOCA and MSLB Accident Scenarios Pags /W.3 VYC-1628 Rev. 0 : Appendix L  ; p. J. R. Hoffman'

                                               ' April 27,1998 Page11
     ,     y w-                            : 4.       App. R Events
                                             ' Peak torus temperature for a variety of App. R events has been analyzed previously with a peak torus temperature of <176*F. [Ref. 4] Analyses are planned to re-analyze the peak torus                           I temperature for the limiting App. R events. In lieu of completed analyses, the issues identified S             -

in ER98-0025 are evaluated for their potential impact on the peak torus temperature following an App. R event.-

                                          - The cunent App. R analysis analyzes three different scenarios (control room / cable vault fire, fire with high pressure makeup systems available, and intake structure fire) with a variety of sensitivity cases for each scenario [Ref. 4]. Of the three scenarios, the first scenario is bounded by other App. R events due to the 100*Finitial torus temperature assumption -              .                  .,
whereas the maximum allowable initial torus temperature is 90*F. The second scenario has ,u five sensitivity ces, one of which is limiting. Considering the ER98-0025 issues, the other ' 24 sensitivity cases would continue to be bounded by the limiting sensitivity case as the impact of j l

all issues e,ould generally apply equally. 'Ihe third scenario has three sensitivity cases, one of which is limiting. Considering the ER 98-0025 issues, the other sensitivity cases would ' continue to be bounded by the limiting sensitivity case as the two non-limiting sensitivity - cases have fewer ER 98-0025 issues that apply (e.g. non-conservative feedwater addition). Therefore, only the most limiting case for each of Scenario 2 and Scenario 3 need further . assessment. lr i

s. -

The issues that are enumerated in ER 98-0025 are applicable to the limiting App. R torus

l. . temperature results but the numerical impact is different. Table 3 provides a list of the issues L identified in ER98-0025 and their assessment ofimpact on the limiting App. R casas.

L Following the table, a description is provided of the impacts on the App. R. tJrus temperature analysis.. An additionalimpact on the App. R analysis is also assesri. The decay heat used in the App. R. torus temperature analysis is cycle dapaadaat. The decay heat for cycle 20 has been determined to be slightly in excess of the decay heat used in the App. R. torus l temperature analysis. Therefore, this increase in decay heat is included in the a:wssment. Table 4 also includes an assessment of the conservatism used in the App. R. Torus temperature analysis. Specifically, the analysis assumed that the reactor vessel liquid

                                         -inventoryis entirely saturated at the start of the event. However, there is actually substantial
                                                                                                                                                                       )

subcooling of the liquid.' The impact of this conservatism is included in the assessment. j i The assessment of the App. R events is divided into'two parts. The first part addresses the L impact on the limiting case for Scenario 2 and the second part addresses the impact on the limiting case for Scenario 3. The scenarios are substantially different such that the impacts are ~ not necessarily the same for each scenario.

                                                                                                                                                               >       1

_.____.___.__m. - _ - _

Torus Temperature and Pressure Response to Large Break LOCA and MSLB Accident Scenarios Page /N Y VYC-1628 Rev. O Appendix L J. R. Hoffman April 27,1998 Page 12 ( Table 4 - ER 98-0025 Issue Impacts on App. R Event Peak Torus Temperature w" . a Issue Impact on- Impact on Scenario 2 Scenario 3 Feedwater Under-estimated +2.2'F +5.8'F Non-Conservative Single Failure Assumption N/A N/A ECCS Pump Heat Not Included 44.1*F N/A FROSSTEY-2 Error & Cycle 20 reload with GE-13 fuel +1.0*F +1.0'F Non-Limiting Containment Cooling Mode N/A N/A Decay Heat 40.2*F 40.I'F b 't@< VesselInventory Conservatism -4.2*F -4.2*F Net Potential Increase above App. R Reference +3.3 *F +2.7'F Net Peak Torus Temperature 183.2*F 181.0*F ' - Required Margin to Bound VY Base Case (173.3'F) +9.9'F

                                                                                                             +7.7*F                  gJ
                                                                                                                                                \

4.1 App. R Scenario 2 l

                                                                                                                                                )

App. R Scenario 2 results from a fire which results in the loss of one train of ECCS. The operators achieve safe shutdown using the available high and low-pressure injection systems. RCIC is assumed to be lost and the shutdown strategy uses feedwater to maintain reactor vessel level prior to manual depressurization. After manual depressurization, alternate shutdown cooling (with a low pressure ECCS pump injecting into a full reactor vessel and water being returned to the torus through an open SRV) is initiated to complete the shutdown l to cold conditions. Feedwater Under-Estimation - The App. R torus temperature analysis included an assumption of feedwater addition based upon all hot feedwater initially in the feed piping being used to maintain vessel level. A detailed feedwater model which included modeling of the energy on the secondary side of the feedwater heaters showed that the potential hot feedwater addition was much greater than assumed previously.  ! 1 Non-Conservative Single Failure Assumption - ER 98 0025 indicated that increased ECC injection resulted in an increased peak torus temperature due to the additional flushing of hot , fluid from the reactor vessel to the torus. The App. R scenario was prescribed by VY staff ( which included any applicable single failure assumption. No change in the App. R scenario ) development has been discovered. Therefore, the single failure assumption issue is not L '" applicable to the App. R scenario.

  • E i

t i Toms hperahue and Pressure Response to Large Break LOCA and MSLB Accident Scenarios Page M r VYC-1628 Rev. 0

        .                                                                                                       Appendix L J. R. Hoffa an
                 ^

April 27.1198 L Page 13 ECCS Pump Heat Not Included - The ECCS pump heat addressed in ER 98-0025 included i,... long-term operation of two core spray and three RHR pumps. For the App. R events the number oflong-term operating ECCS pumps would be fewer based on the scenario description. The App. R event results in only one RHR pump operating in long-term torus cooling and one core spray pump pmviding long-term vessel make-up, the ECCS pump heat

            . would be limited to the contribution from these two pumps.

Based on one RHR and one CS pump operating long-term, the ECCS pump heat addition is 19.9 mBTU over five hours (approximately the time of peak toms temperature). Distributed into 4.8 mlb of water (initial inventory plus feed inventory) this heat addition represents about 4.l*F in torus temperature. His potential temperature increase would be mitigated by the resultant increase in heat removal through the RHR heat exchangers. However, this increase in heat removal is conservatively ignored for this assessment, t FROSSTEY-2 Error - The fuel volume-averaged temperature'of the Cycle 20 core has been calculated. [Ref. 5) This calculation conects the FROSSTEY-2 error cited in ER 98-0025 as - well as recalculating the fuel temperatures for the new core and new GE-13 fuel. An i assessment of the change in stored energy from that assumed in the App. R analysis and the i new core has been made. It was determined that the new core would contain approximately 5 mBTU more energy than that assumed in the previous analysis. This additional energy, when

            , applied to a 4.8 mlb suppression pool results in a temperature increase of approximately 1.0'F.              N

{_', This potential temperature increase would be mitigated by the resultant increase in heat

             . removal through the RHR heat exchangers. However, this increase in heat removal is conservatively ignored for this assessment.

Non-Umiting Containment Cooling Mode - The App. R scenario that was analyzed is prescribed by the VY staff arid the results were used to develop safe shutdown strategies. As such, the use of injection mode cooling is not a mode of operation for this scenario. Therefore, this issue is not applicable to the App. R scenario.

           ' Decay Heat - The App. R torus temperature analysis was conducted using decay heat .

assumptions applicable for Cycle 19 and cycle-specific input data. For Cycle 20 a new decay heat curve has been developed [Ref. 6] and represents a slight increase in decay heat due to 1 different cycle operation and the introduction of GE-13 fuel. Based on the difference in

            . integral decay heat over the course of five hours (time of peak torus temperature), there is a difference of 1.1 mBTU. His translates to about a 0.2*F increase in torus temperature for a 4.8 mlb suppression pool and assuming no mitigation due to increased heat removal capability due to the increased temperature.

I

           ' Vessel Inventory Conservatism - The reactor vessel initial conditions used in the App. R
          ~

analysis are significantly conservative in that the liquid in the vessel is assumed to be at saturated conditions rather than the actual subcooled conditions. This represents a conservatism of about 20 mBTU.' Distributed into 4.8 mlb of water, this conservatism

    ..-       represents 4.2*F. This conservatism is removed by subtracting it from the other ER 98-0025
           ' issues. Note that this conservatism would be reduced when applied to the peak torus temperature by the resultant decrease in heat removal through the RHR heat exchanger.

However, this conservatism represents a complete offset of the pump heat consideration

j;' Torus Tesaperature and Pressure W n to Large Break LOCA and MSLB Accident Scenarios . YYC-1628 Rev. 0 Page //4 Appendix L L l ' i R. Hoffman . l April 27, I998. l - Page 14 y , . l - detailed above which also did not account for the difference in heat exchanger heat removal. y,

j. Therefore, the increase and decrease in heat removal result in a net zero difference. '

[ "w

                                       ' 4.2 App. R Scenario 3

,~ App. R Scenario 3 results from an intake structure fire which results in the requirement to use

                                                                                                                                                                          )

!- the Alternate Cooling System to achieve safe shutdown. It is assumed that the Altemate . I

                                      . Cooling System can be aligned within two hours after which shutdown cooling is used with the Altemate Cooling System providing the Ultimate Heat Sink. Feedwater is assumed to be I

used to maintain reactor vessel level prior to manual depressurization at one hour. Ievel 1 l continues to be maintained using feedwater until event termination (two hours). L ' Feedwater Under-Estimation - The App. R toms temperature analysis included an assumption !ME l ~of feedwater addition based upon dl hot feedwater initially in the feed piping being used to !~ maintain vessel level. A detailed feedwater model which included modeling of the energy on , i i the secondary side of the feedwater heaters showed that the potential hot feedwater addition - was much greater than assumed previously. Based on the difference calculated for the short-term response for a large break LOCA, the feedwater difference for this case is identical as no l ' heat removal occurs during this scenario. W (' ~ Non-Conservative Single Failure Assumption - ER 98-0025 indicated that increased ECC injection resulted in an increased peak torus temperature due to the additional flushing of hot l fluid from the reactor vessel to the torus. The App. R scenario was prescribed by VY staff which included any applicable single failure assumption. No change in the App. R scenario  !

j.  :

development has been discovered. Therefore, the single failure assumption issue is not i applicable to the App. R scenario. ECCS Pump Heat Not Ir.cluded -The ECCS pump heat addressed in ER 98-0025 included

                                      ' long-term operation of two core 3 pray and three RHR pumps. For this App. R event there are no long-term operating low pressure ECCS pumps..

FROSSTEY-2 Error - The fuel volume-averaged temperature of the Cycle 20 core has been calculated. This calculation corrects the FROSSTEY-2 error cited in ER 98-0025 as well as

                                     - realentering the fuel te+4 ares for the new core and new GE-13 fuel. An assessment of
                                   . the change in stored energy from that assumed in the App. R analysis and the new core has 7 - been made. It was determined that the new core would contain approximately 5 mBTU more g                                        energy than that assumed in the previous analysis. .This additional energy, when applied to a L

4.8 alb suppression pool results in a temperature increase of approximately 1.0*F. .

                          ]

Non-Umiting Containment Cooling Mode - The App. R scenario that was analyzed is prescribed by the VY staff and the results were used to develop safe shutdown strategies. As

                                   - such, the use of injection mode cooling is not a mode of operation for this scenario.
    ..-                                 Therefore, this issue is not applicable to the App. R scenario.                   .,

e Decay Heat - The App. R torus temperature analysis was conducted using decay heat

                                   ' assumptions applicable for Cycle 19 and cycle-specific input data. As described previously,                              j
           ' Toms Taparabue and Pressure Response to Large Bmak LOCA and MSLB Accident Scenarios                                                     Page BV7
           . VYC-1628 Rev. 0
         ,                                                                                                                                        Appendix L    .

J. R. Hoffman ' April 27,1998 Page 15 the difference in integral decay beat over the course of the two hour transient is about

                                                                                                                                                                  .,        y
             'O.5 mBTU. This translates to about a 0.1*F increase in toms temperature for a 4.8 mlb                                                                  "-~~

suppression pool. Vessel Inventory Conservatism - The reactor vessel initial conditions used in the App. R analysis em significantly conservative in that the liquid in the vessel is assumed to be at saturat.ed conditions rather than the actual subcooled conditions. This represents a conservatism of about 20 mBTU. Distributed into 4,8 mlb of water, this conservatism represents 4.2*F, This conservatism is removed by subtracting it from the other ER 98-0025 issues. hi.i@ i

     .f.

i O' i -

=
5. W* j f
                                                                     -            _           a_ __, _    _.- - _ _ _ _ _ . - ___..____._- __ - _ --

Torus Temperature ;.nd Pressure Response to Large Break LOCA and MSLB Accident ScenariosPage_/g VYC-1628 Rev. O Appendix L J. It Hoffman ' April 27,1998 Page 16

5. Station Blackout gg Analysis for pool heat up following a station blackout has been performed previously using an initial suppression pool temperature of 100*F with a resultinP pcd tempt rature of 167.0*F for the limiting scenario (Ref. 7]. The issues identified in ER98 325 are evaluated for their ^

potential impact oa the peak torus temperature following a station blackout event. Due to the limited applicability of the ER98-0025 issues to the station blackout analyses and the significant margin available, no follow-on re-analysis is planned. Table 4 provides a tabulstion of the ER98-0025 issues and their effects on the station blackout analysis. The conclusion is, based on the initial torus temperature of 100'F and the low peak temperature, the station blackout analysis is bounded by all other analyses addressed in this evaluation. Table 4 - ER 98-0025 Issue Impacts on Station Blackout Event Peak Torus Temperature Issue Impact f Feedwater Under-estimated N/A I Non-Conservative Single Failure Assumption N/A g' i ECCS Pump Heat Not Included + 1.8'F { FROSSTEY-2 Error +0.5*F Non-Limiting Containment Cooling Mode N/A Cycle 20 reload with GE-13 fuel +0.3 *F t Net Potential Increase above Limiting SBO +2.6F Net Peak Torus Temperature 169.6F j Required l\fargin to Bound VY Base Case (173.3*F) N/A Feedwater Under-Estimation - The station blac':out recovery assumes the use of HPCI or RCIC to maintrin vessellevel. Non-Conservative Single Failure Assumption - The single failure appropriate to the station  ! blackout analysis, failure of one RHR heat exchanger was taken in the analysis. The issue l raised in ER98-0025 with respect to the single failure assumption is not applicable as any  !.i ss j l required makeup capability from low pressure ECCS pumps is well within the ability of one i pump to provide and as a result multiple pumps would not be used. ECCS Pump Heat Not Included - The energy to drive the HPC1/RCIC pump comes from the

         - -   vessel (steam) and thus is not an external energy source for the vessel and containment control volume. One RHR pump is provided in torus cooling. Over the six hours for the limitmg scenario, this could result in an additional 15.3 mBTU distributed in a pool of 4.216 mlb of water, this would represent a 3.6*F temperature increase. However, this increase would be                                               i           j l

c- -_ _ _ _ _ _ _ _ _ _ . _

Torus Temperature and Pressure Response 13 Large Break LOCA and MSLB Accident Scenarios Page /NT VYC-1628 Rev. O Appendix L J. R. Hoffman April 27,1998 Page 17 mitigated by the inemased heat remcval due to the increased temperature. Based on experience, the integrated effect would result in only a 1.8'F peak pool temperature increase. FROSSTEY-2 Error -The net increase in energy due to the FROSSTEY.2 error remains the same as assessed in ER 98-0025. The impact on station blackout event peak torus temperature would be approximately the same as that for large break LOCA peak torus temperat'ure. Non-limiting Containment Cooling Mode - The station blackout scenaric was prescribed and did not involve any potential for injection mode cooling as discussed in ER98-0025. Therefore, this issue is not applicable to station blackout analysis. Cycle 20 reload with GE-13 fuel - In order to bound the Cycle 20 reload, the potential , temperature increase as a result of the introduction of GE-13 fuelis included. The basis for this increase is discussed in the limiting large break section. J. R. Hoffman April 27,1998 Page 18

6. References
1. DEAS Calculation VYC-1628 Rev. 0, Torus Temperature and Pressure Response to Large Break LOCA and MSLB Accident Scenarios, approved 27 April 1998
2. Vermont Yankee Event Report, ER 98-0025, initiated 6 January 1998
3. Calculation VYC-1442 Rev. 0, Peak Torus Temperature for Intermediate and Small Break IDCA, approved 6 September 1996
4. Memorandum L Schor & D. E. Yasi to M. J. Marian, LOCA-VY 96-044, Torus Temperature Evaluation for Appendix R Scenarios, dated 18 October 1996
5. DE&S Calculation VYC-1763 Rev.0, Fuel'Ihermal Performance for Appendix R and Containment Analysis (Cycle 20 / GE-13), DRAFT
6. DE&S Calculation VYC-1777 Rev. 0, VY Appendix R Analysis for Cycle 20 w/ Fiche, DRAFT
7. Calculation VYC-886 Rev.1, Station Blackout Documentation Analysis, approved 12 May 1995 Prepared VI48B C. D. Fago, Bligineer # '

Date Thermal Hydraulic and Safety Analysis Group l Nuclear Engineering Department l Reviewed iA h' 1/tt/# J. E.1Metcalf Date Polestar Applied Technology I (.

Tonas Temperature and Pressure R:sponse to Largs Bre k LOCA cnd MSLB Accident Scenarios

          . WC-1628 Rev. 0 -                                                                                                                                           Page/fJD
   .                                                                                                                                                                      Appendis L h TfacJ W I                              MEMORANDUM "p>I 'T6Af=-# 18'OW
                                   .      DUKE ENGINEERING AND SERVICES - BOLTON To J.R. Chanman                                                                                                                         Date Aor1124.1998 Group # 'ISHAG-VY 98-068 From S.J. Ganthner/M. P. IzFrancois                                                                                                     W.O.#

Subject ~ Enstneering Evaluation of Abnormal OoerationalTransients I.M.S.# for Pamir Toms Temperature File # VY98-068

References:

1. Work Release Form 98-0106-00. Torus Temperature Analysis - SRV Transient ~s, dated March 11,1998.
2. WC-1290, Vennont Yankee Post-LOCA Torus Temperature and RHR Heat
                         ' Exchanger Evaluation. August 1,1994.
3. EDCR 97-423. RHR and Core Spray Strainer Replacement. March 1998. f
4. WS 137/97, 'IYansmittal of Draft Proposed Change No. xxx TS 3.7.A.1 Suppression i

Pool Temperature Limits, draft dated April 22,1998. l j

                 - 5. WC-1660 Revision O. Vermont Yankee Feedwater Train RELAP5YA Model
    • Development. October 17,1997.

g

                                                                                                                                                                                         ~
6. OP 2124. Residual Heat Removal System. Revision 39, dated April 26,1996.
7. NUREG-0783, Suppression Pool Temperature IJmits of BWR Containments, dated July 1981.
8. Letter. USNRC to BWR Owner's Group, Transmittal of the Safety Evaluation of General Electdc Co. Topical Reports: NEDO-30832 Entitled
  • Elimination of Limit on BWR Suppression Pool Temperature for SRV Discharge with Quenchers' and . {

NEDO-31695 Entitled 'BWR Suppression Pool Temperature Technical Specification Limits", dated August 29,1994. i Executive Summary: An engineering evaluation of Abnormal Operational Transients (AOTs), as identified in FSAR Chapter 14, involving the exhaust of pressure vesselinventory through the  ! Safety / Relief Valves (SRVs) was performed for Vermont Yankee (per request of Reference 1) to enable assessment of peak bulk suppression pool temperatures for the limiting AOT. ' j, 'lhe transient ton 2s temperatures can directly impact temperature based design criteria such as those for RHR and Core Spray pump NPSH. Other transients and accidents including large, intermediate and small break LOCA. Appendix R and station blackout f- scenarios have been assessed elsewhere for peak torus temperature requirements. 'Ihe t evaluation summarized in this memo has received a peer review to ensure that the approach is appropdate and the results are conservative. The remaining tasks under the Reference 1 request involve completion of sensitivity analysis and documentation for this

     -Torus Temperature and Prissure Response to Large Breik LOCA and MSLB Accident Scenarios          Page,//fL
    ' VYC-1628 Rev. O Appendix L J.R. Chapman THSAG W 98-068 .

April 24, 1998 Page.2 .-

          =-F+T.4 evaluation per Vermont Yankee and Duke Engineering Appendix B                                  ,

requirements.' The SRV exhaust AODs analyzed for the torus temperature assessment, which employed the REIRAN thermal hydraulics code and Vermont Yankee reload licensing model, applied a number of conservative assumptions relative to avnetM reactor system and operator performance. Rese assumptions are consistent with or more conservative than assumptions in the large break IDCA analysis which has provided the basis for an acceptable peak torus temperature of 185 F or less. . Of key interest are the assumptions of 85F service water, a torus operating temperature of 90 F. consideration of degraded RHR heat avehanger performance and the new torus suction strainer design. To confirm ._ performance of the REIRAN modelin predicting torus temperatures, a limited compadson ~

       - of results of an SRV exhaust event to results reported for a peer plant was made. ne dedvation of the limiting SRV exhaust ACT for peak torus temperature involved establishing base cases avn~tM to provide limiting results and performing a number of
        . sensitivity analyses for confirmation. Since all sensitivity analysis for the limiting case were
         .not complete a case with bounding assumptions has provided the basis for our conclusion.

fit is concluded that for the bounding AOT involving SRV exhaust, the peak bulk - suppression pool temperature remains below 180 F which is within the large break LOCA ,, based torus temperature limit of 185 F. A detailed summary of the engincedng evaluation is contained in the next section. H. Discussion: g Introduction to SRV Discharge AOT Evaluatinn

         ' Evaluation of the long term suppression pool temperature in response to abnormal
       ' operational transients (AOrs) involving exhaust through the Safety / Relief Valves (SRVs) was performed to allow assessment of peak torus temperatures. De analysis was performed to confirm that applicable containment design criteria continue to be met in                   1
        ' light of RHR heat exchanger as found performance, the new torus RHR/ Core Spray suction                  I strainer installation and revisions to torus allowed (Tech Spec) temperatures (References 2.
3. and 4). The AOT evaluation summarized in the memo provides input to the overall ,

i evaluation of containment design criteria. 1 l' . Abnormal operational transients are events which are awpetM to occur at least once during the lifetime of the plant. At the time of the irdtiating event or operator error, the

       ' plant is assumed to be operating in any licensed configuration with all required equipment
       . operable.1. e., the plant is not operating under any limitation for continued operation. In addition to the irdtiating event an additional single failure, which results in the greatest             !

challenge to applicable cdteda, is postulated to occur for the purpose of analysis. De

j. ' analyzed transient combining the worst _ initiating event and single failure resulting in the i b ' greatest challenge to the applicable design criteria is referred to as the limiting AOT. )

J Abnormal Operational *nanelants- Anolicable Design Criteria { l Abnormal operational transients with the expected frequency of at least once dudng the l l; . hfetime of a plant are required to meet specified acceptable fuel design limits, to maintain  ! core cooling, reactor vessel and piping integrity, and to maintain primary containznent l integrity. Analysis of the abnormal operational transient events as reported in the FSAR I I i= .

Torus Temperature end Pressure Rssponse is Large Break LOCA and MSLB Accident Scenarios

         . VYC-1628 Rev. 0_

Page g Appendix L i' J.R. Chapman

        ~THSAG VY 98-068
        -April 24, 1998 Page 3                  .

more specifically address the fuel and core cooling cdteda and reactor vessel integrity since ,. most AOTs do not result in release to containment. he SRV exhaust analysis summarized in this memo was performed to confirm that the peak bulk suppatssion pool temperature that would result from any AOT involving SRV exhaust was less than ISS F. Eis peak temperature value is consistent with requirements established by assessment of thCA. Appendix R and station blackout scenados, nis temperature limit has considered the design of the new torus suction strainer design. Rese events do not challenge fuel specified acceptable design hmits for fuel or cladding failure and or the pressure hmits defined for maintenance of reactor vessel , and piping integrity (ASME code requirements). After the original design and licensing of Vermont Yankee, the potential for containment structural damage as a result of SRV discharge was identified as a result of the condensation stability performance of SRV steam quenchers at a foreign BWR. Industry and regulatory eKorts established applicable criteria and specific analysis requirements for

              , postulated events involving SRV discharge to ensure containment integdty in 1981
              ' (Reference 7). Subsequent BWROG efforts were successful in eliminating consideration of the SRV condensation stability issue on an generic basis for plants with SRV quenchers of the design currently installed at Vermont Yankee (Reference 8). Rus, the criteda and specific analysis requirements established for the condensation stability issue are not applicable. However, due to the similarity in the types of events, the assumptions made for mitigative systems and operator actions were ofinterest in developing the specific event              _

scenarios evaluated for peak torus temperature. *

            ~

Selection of Sneelne SRV Discharse AOTs for Evaluation The specific events Svaluated were selected based on a review o' f the FSAR Chapter 14 AOT

           ' with the potential for significantly increasing the temperature of the suppression pool. De FSAR AOTs are identified in Table 1, with indication whether the event could result in
            . discharge to the torus. Of these events, the Generator Wip and the Turbine Trip events which may result in release to the containment for a very limited Ume are far from limiting
            ^ and were'not evaluated. The Main Steam IJne Isolation Valve Closure was not analyzed since the MSIVs will be reopened by the operators to reestabush reactor coohng via the.             ,,

condenser, resulting in short term blowdown to the torus only. Two events resulting in significant heat addition to the suppression pool were identifled from the AOT events

            .' reported in the FSAR: the Stuck Open Relief Valve (SORV) event which adds significant
energy addition to the suppression pool and the loss of Auxiliary Power (LAP) event which i results in the unavailability of the main condenser as an alternative heat sink to the suppression pool.
                             ~

Single Failures Analysis of AOTs requires the inclusion of the single failure of an active device which would l< hamper mitigation of the transient.' ne failure of passive devices, such as a spdng safety

            ' valve or a multiple failure of a redundant safety grade device (e.g. a reactor protection
            . system function) are not required. ne inclusion in the analysis of the failure of a passive device or a redundant safety grade device has been carried out in certain cases in the past to provide a bounding assessment of a transient or group of transients. The MSIV closure for analysis of ASME code vessel pressure requirements is an example, in this case, the
 ;             failure of the MSiv position switch scram function is assumed as the single failure to

l ' Torus Temperature and Pressure Response to Large Bre;k LOCA cnd MSLB / ccident Sc;narios Page //ff VYC-1628 Rev. 0 { Appendis L J.R. Chapman THSAG VY 98-068 April 24, 1998 Page 4 . l l provide a bounding assessment of overpressure events which may challenge vessel integrity. _ Single failures were considered for the SORV and IAP events, which would either reduce l the capability of mitigative systems to remove energy from the systern, or result in the addition of more energy to the suppression pool. He primary means of removing heat from I the system are the main condenser and the RHR Heat Exchangers (HXs). He amount of i heat added to the suppression pool is dominated by whether an SRV is open and how i much energyis avaDahe from the vessel. Energy may also be added to the torus from the l RHR and core spray ,xanps. o For the SORV event two long tenn heat removal systems are available, the main condenser and the RHR HXs. The first potentially Emiting single failure considered for the SORV event is the loss of an RHR HX which reduces the heat removal capabihty of a mitigative system. De second potentially limiting single failure considered was an assumed single failure that makes the main a,ndenser unavailable throughout the duration of the event. - l Assuming that the condenser is unavailable after its initial isolation is conservative since I the operator will have a number of alternative power sources and system configurations by which the condenser can be re-estabushed. No additional single failure was identifled for the SORV event which could increase the heat added to the suppression pool. The SORV event with the assumed single failure of the main condenser as a heat sink results in the suppression pool being the only heat sink for the vessel in this event. i The IAP event results in loss of the main condenser as a heat sink due to the assumed loss of the means by which to re-estabush condenser vacuum. For the IAP event the first potentially limiting single failure considered was the failure of an RHR HX which reduces the capability of the remaining mitigative systems to remove heat from the system. De second potentially limiting single failure considered was a safety / relief valve staying open after opening to control the pressure of the isolated vessel. The SORV as a single failure would then result in an increased energy addition rate to the suppression pool. No other i single failure was identified for the IAP event which would result in either less capability to remove heat from the system or an increase in energy addition (or rate of addition). Limitina Cases Selected De resulting four AOT events identified as resulting in potentially limiting suppression pool temperatures are listed in Table 2 by initiating event. postulated single failure and the I available, remaining and assumed mitigative systems. Key Assumptions The following additional assumptions were made for the purpose of simulating the SRV exhaust events: The initial operating power is 1664 MWth, the design limit of the turbine. *n11s value is l 104.5% of the licensed power level of 1593 MWth. 1 ne initial torus operating volume is at its minimum allowed. 68,000 ft' [ Technical Specifications 4.7.A.e) .- 1 J

Torus Temperature and Pressure Response 13 Large Brnk LOCA and MSLB Accident Scenarios VYC-1628 Rev. 0 Pagefjfy Appendix 1. ( . 4i . . JJ.R. Chapenan THSAG W 98-068 April 24; 1998 Page 5

  • Re service v.ater temperature is the maximum allowed per the FSAR, 85 F. . -

he heat removal capability of each RHR HX was assumed to be 6.56E+5 BnJ/hr F. his reflects the as tested performance with tube plugging (Reference 3). . He enthalpy of the feedwater (when credited) is maximized to provide a conservative ' estimate of stored energy in the feedwater (Reference 5). If operators align core spray or RCIC sucuon to the torus instead of the condensate storage tank, the enthalpy of the torus water assumed is the bulk temperature calculated by J RETRAN.. Work done on the steam from the vessel used to drive the HPCI and RCIC pump turbines

                                                              .which exhausts to the suppression pool is neglected l Re enthalpy of the flow injected into the vessel by HPCI and RCIC , which uses the                             -

Condensate Storage Tank as a source, (when credited) is 58.10 BTU /lbm, based on the marimum expected temperature of 90F. ' No heat loss to the containment atmosphere and structures is assumed. We stored energy of the fuel, reactor vessel, intemal structures and p! ping is modeled to maximize energy to the suppression pool.

                                                                                                                                                                            .g An open SRV will reclose at a pressure of 65 psia due to its mechanical characteristics and will reopen at 115 psia.1                                                                                     ,

When available. shutdown cooling may be esitered after the vessel pressure is less than 115 c psia provided the vessel pressure is maintained below 115 psia ( Reference 6). He operator is assumed to take actions consistent with Technical Specification and i , L operating procedures The values of specific parameters (1. e., suppression pool temperature) which are assumed to trigger operator action are the analytic values calculated in the simulation. Specific operator actions credited include the following: j' When the suppression pool temperature exceeds 90F, the operator is assumed to !- initiate action that will result in operation of available RHR HX in the suppression pool cooling mode in ten minutes. When the suppression pool temperature exceeds 110F, the operator is assumed to manually scram the reactor. When the suppression pool temperature exceeds 120F. the operator is assumed to depressurize the reactor vessel at a rate not to exceed 100F/hr (Sensitivity studies - were perfonned to analyze the effects of slower cooldown rates). i L i

[ Torus Temperature and Pressure Response t) Large Brc~ k LOCA cnd MSLB Accident Scerurios I VYC-1628 Rev. 0 Page/fff Appeh L l' . J.R. Chapman THSAG VY 98-068 April 24, 1998 Page 6 . For the SORV event with an assumed single failure of one RHR HX. the operator is

              - assumed to take 20 minutes to re-estabush the main condenser as a heat sink and open the main turbine bypass.

For the transition to the shutdown cooling mode from the suppression pool cooling mode with only one RHR HX available, the flushing stage is deleted and the transition is assumed to take 16 minutes. Evaluation Code and Models , The RETRAN code was selected for use in the simulation of the transients based on its general capabilities of modeling thermal-hydraulic conditions over a wide range of pressure and temperature. The RETRAN code has been used for analysis of VY FSAR Chapter 14 transients involving changes in vessel pressure and vessel inventory such as the Generator load Rejection Without Bypass. "Ibrbine Trip Without Bypass and Main Steam Line Isolation Valve Closure. A review of NRC Safety Evaluation Reports for both the generic - approval of the code and the Duke Engineering and Services BWR transient analysis methodology was performed to ensure applicable restriction and cautions were applied to the model inputs and the code was used within its range. The model prepared for use in the analysis of the four cases identified in Table 2 was based on the Vermont Yankee Reload RETRAN licensing model. Components specifically included for analysis of the SRV exhaust events included: a volume representing the suppression g pool, heat exchangers representing the RHR HXs. addition of heat slab material to represent vessel structural materials, additional volumes and junctions to represent main turbine bypass valves, and modification of the control system representing the feedwater system such that the enthalpy of the vessel inventory make-up could be specified as a function of the integrated mass injected. For the LAP with a manually controlled cooldown rate, junctions and control system components were also introduced to facilitate the simulation of manual operation of SRVs. RETRAN Model Benchmark The appilcation of the RETRAN code and model as developed for the simulation of the SRV exhaust events was verified by performing an evaluation that would closely parallel the conditions evaluated and reported for another BWR/4 for Case 2b. The peak bulk suppression pool temperature reported for the reference BWR/4 plant was 140F at 6515 seconds. For the purpose of reasonably matching the conditions as evaluated for the reference BWR/4 plant the suppression pool volume was increased to obtain the same power level to suppression pool initial volume ratio as the reference plant. Also, the area of the SRV was increased to produce about 122.5% of the choked flow obtained for the SRV flow area and the service temperature was assumed to be 67F. It was assumed that the heat removal capability of the RHR HX of the reference BWR/4 plant was 200 BTU /s*F. Other potential differences that could impact the comparison of the results which were not specifically evaluated include decay heat assumptions and heat addition to the vessel from operation of l ECCS components.

                                                                                                            )

O Separate simulations were performed for comparison to the reference plant results with and without trip of the recirculation pumps since it was not clear from the description of j c l i

Torus Temperature end Pressure Response to Largs Break LOCA cnd MSLB Accident Scenari:s s VYC 1628 Rev 0 Page,/43 Appenda 1. 19 ,

    .J.R. Chapman-THSAG W 98-068 April.24, 1998
    .Page ~1-                                      .

l the event whether the recirculation pumps were assumed to have tripped or not. For the l simulation with the recirculation pumps assumed to be operating the peak bulk . i> suppression pool temperature obtained was 145F at about 5800 seconds and for the simulation with recirculation pumps assumed to trip a peak bulk suppression temperature of 143F at about 4600 seconds was obtained. As noted above, the peak temperature reported for the reference BWR-4 plant was 140 F. Rus,it is concluded from the - l benchmark to the reference BWR-4 that the RETRAN model has provided a reasonable and } L conservative pr= den of peak toras temperatures and was adequate for analysis of Vennont Yankee.- l SRV Discharge AUT Case Results Descriptions of the base cases and the sensitivity analyses run for the SORV and 1 AOTW are provided below. Base Cases

        ' n==a 1A - SORV with Imsa of 1 RHR HX
  • 1
                                                                                                             ~
        . Se SORV with loss of one RHR HX results in a peak bulk suppression pool temperature of about 163F. After a mild depressurization in response to the SORV opening the turbine contal system recovers pressure and power returns to 98.2% of the initial power level in
        ;less than 20 seconds. We operator initiates actions to establish suppression pool cooling due to the heat addition to the suppression pool from an initial temperature of 90F. The            g operator is assumed not to scram the reactor until the suppression pool reaches a -

temperature of 110F which occurs about 360 seconds after the SORV opens. After reactor scram the MSIVs close in response to the depressurization caused by the SORV. Suppression pool cooling begins at about 600 seconds and the operator is conservatively assumed to open a single main turbine bypass valve at 1200 seconds after opening the MStVs and re-establishing condenser vacuum. Depressurization of the vessel continues after 1200 seconds by both the SORV and the main turbine bypass valve.

       ' At about 4000 seconds the vessel pressure has decreased to 115 psia and the pool temperature has increased to about 155F. At about 8000 seconds with the pool temperature at about 160F it is assumed that the operator initiates the switch over to l shutdown cooling and terminates pool cooling for the next 1000 seconds. At about 9000 seconds,just before shutdown cooling increases the rate of depressurization of the vessel, the SORV recloses as vessel pressure drops below 65 psia and a peak pool temperature of less than 163Fis reached.

Cagg IB - SORV with tons of Main Condanner he SORV event with the main condenser unavailable through the event progresses the same as the SORV event with loss of one RHR HX through reactor scram. After reactor scram it is assumed that the MSIVs close due to the rapid depressurization caused by the SORV and the operator is unable to open the MSIVs causing the main condenser to remain Unavailable throughout the remainder of the event. At about 600 seconds into the event

      , suppression pool cooling begins with two RHR HX available. At about 7600 seconds a g          suppression pool temperatuar of about 165F is reached at a vessel pressure of 133 psia.

l g-

             ' Torus Ternperature and Pressure Response is Large Bre .k LOCA rnd MSLB AccidInt Scenarios

_ VYC-1628 Rev.0 P:geffG Appendh L J.R. Chapman-lTHSAO VY 98-068

            ~ April 24, 1998 Page 8~                                                  .

After reaching a vessel pressure of 115 psia and a pool temperature of slightly less than ' 164F at about 10000 seconds, the operator is assumed to initiate the switch over to shutdown cooling of one of the two operating RHR HX. De temperature of the pool rises to temperature ofless than 167F before the combined impact of shutdown cooling and the i SORV reduce the vessel pressure to less than 65 psia at 12000 seconds. At a vessel j pressure ofless than 65 psia, the SORV recloses and the heat addition to the pool is

                                                                                                                                         ]

tenninated as shutdown cooling continues to reduce vessel pressure. , emma 2 A -IAP with r== of one RHR HX

                          ' De IAP event with loss of one RHR HX and feedwater for vessel inventory control results in                  a a peak bulk suppression pool temperature of approximately 180F. Ris is a boundfrig case                J for SRV exhaust A014 since it represents the worst combinadon of transient initiator and              j single failure and it has been additionally assumed that feedwater supplies vessel inventory            1 control throughout the event. De feedwater system would be inoperable due to the Joss of                 !
         ..                       auxiliary power. De feedwater enthalpy is higher than that of either the CST or the torus              I 4
              ' at any time in the transient. In comparison, the peak temperature reached if the inventory untrol systems are assumed to be taking suction from the CST is about 172F at about 17500 seconds. - Results for the case where the torus would provide inventory from
                         - the suppression pool are not yet available, but are expected to be in the range of 175F.

After reactor scram and isolation of the vessel due to MSIV closure the SRV with the lowest l w pressure setpoint periodically cycles to maintain vessel pressure. De periodic cycling of g this SRVgradually raises the temperature of the suppression pool to 120F at about 6000 { i seconds while suppression pool cooling is assumed to be initiated at about 600 seconds. l

                        ' After the suppression pool temperature reaches 120F the operator is assumed to initiate                          I

' actions which result in a cooldown rate of 100F/hr by manually controlling a SRV He I

                        - cooldown rate of 100F/hr results in a relatively slow vessel depressurization compared with the cases which involve a SORV. .De peak suppression pool temperature occurs more than four hours after the event begins due to the limited heat removal capability of the single available RHR HX and the lower suppression pool temperatures during earlier
                           . portions of.the event, o

(?mes 2B - LAP with Failure of SRV Onen For this event a single SRV is assumed to remain open when the SRVs open on high vessel pressure after the assumed reactor scram and closure of the MSIVs. he uncontrolled depressurization of the vessel through the SRV adds heat directly to the suppression pool E but there are two RHR HX available for pool cooling. A local temperature of slightly greater l

,X than 156F is reached at about 8400 seconds at a vessel pressure of about 125 psia. After reaching a vessel pressure of 115 psia and a pool temperature of slightly less than 156F at about 10000 seconds, the operator is assumed to initiate the switch over to shutdown cooling of one of the two operating RHR HX. With one RHR HX in the pool cooling mode and the other in shutdown cooling mode about 1000 seconds later, the temperature of the pool rises to temperature ofless than 159F before the combined impact of shutdown
                        ' cooling and the SORV reduce the vessel pressure to less than 65 psia at about 11500
  • seconds. At a vessel pressure ofless than 65 psia, the SORV recloses and the heat
                      ~addition to the pool is terminated as shutdown cooling continues to reduce vessel pressure.                         !
   ,                        he peak value is greater than the value calculated for the reference BWR/4 plant evaluations discussed above primarily due to the differences in the assumed serylce water
                      . temperature, the heat removal capability of the available RHR HX and operator actions.

1 _ _ _ _ - _ _ _ _ _ _ _ _ _ _ _ - _ - _ - --_ l

l Torus Temperature cnd Pressure Response t) Largs Break LOCA and MSLB Accident Scenirios Page a m VYC-1628 Rev. O Appendix L i J.R. Chapman i THSAG VY 98-068 l April 24, 199J Page 9

  • Sensitivity Studies Several perturbations of the most limiting SRV exhaust event. Case 2A, were evaluated to identify the most limiting results. ne perturbations which were evaluated considered allowed operator actions that alter how the event is tenninated. i. e., the termination of heat addition to the suppression pool ne alternative! operator accons considered were the switching from the suppression pool coohng mode to the shutdown cooling mode of the RHR system, maintaining a lower cooldown rate and trying to continue cooling below a vessel pressure of 65 psia.

The sensitivity studies used the Case 2A scenario where CSTwas assumed to provide inventory for HPCI, RCIC and Core Spray. His scenario results in a peak bulk suppress,on j pool temperature of about 172F at about 17500 seconds. and occurs when the operator .j maintains a cooldown rate of 100F/hr until the vessel pressure is just above 65 psia and  : selects to switchover to shutdown cooling. For this scenario. the operator is assumed to l suspend cooldown of the vessel and maintain vessel pressurejust above 65 psia which I avoids closure of the SRV due to low pressure. The switchover is calculated to be initiated when the suppression pool temperature is about 170 F at about 16500 seconds. During the 16 minutes that it is assumed to take the operator to switchover to shutdown coohng i the temperature of the suppression pool increases to 172F as the operator maintains pressurejust above 65 psia by manual operation of the SRV. g L' Re first sensitivity t;onsidered assumes the operator tries to cooldown below a vessel pressure of 65 psia which results in closure of the SRV due to the mechanical characteristics of the SRV at a suppression pool temperature of about 170F. The SRV is assumed to remain closed until the vessel pressure recovers to 115 psia due to decay heat. The increase in suppression pool temperature that occurs as the SRV re-opens at 115 psia untilit re-closes again at 65 psia is less than the reduction that results from the operation of the RHR HX in the pool cooling mode and a secondary peak bulk suppression pool temperature ofless than 169F occursjust prior to the re-closure of the SRV on low vessel pressure, i A second sensitivity was also considered assuming an operator controlled cooldown rate of  ! 80F/hr. At this reduced cooldown rate, the peak bulk suppression pool temperature was { negligibly less than the 100F/hr case at a vessel pressure of 65 psia which was calculated e , to occur at about 19000 seconds. Due to the later occurrence oflower peak temperature it was concluded that neither a switchover to shutdown cooling nor attempting to cooldown p '. below 65 psia would be more limiting than for a cooldown rate of 100F/hr since the decay heat rate wpuld be less and less heat would be added to the suppression pool. j However, it was observed that at the time the vessel pressure reaches 115 psia, the maximum pressure for estabitshing shutdown coohng. the suppression pool temperature for an assumed cooldown rate of 80F/hr is about 165F at about 16000 seconds and is greater than the temperature of about 163F at about 14000 seconds for an assumed ' cooldown rate of 100F/hr. To confirm that the suppression pool temperature reached as a result of switching over to shutdown cooling once a vessel pressure of 115 psia was 3 reached with a cooldown rate of 80F/hr was not Ilmiting an additional scenario was i considered. his third sensitivity assumed that switchover to shutdown cooling occurs

                                       . when the vessel pressure reaches 115 psia at a cooldown rate of 80F/hr and that the operator. in addition to terminating suppression pool cooling during the switchover, also                   l l                                                                                                                                                       l l

i

( Torus Tems erature and Pressure Response to Large Break LOCA (nd MSLB Accident Scenarios -- Pagejg5(

                       . VYC-1628 kev. O Appendi .
                  .J.R.. Chapman THSAG VY 98-068 April 244 1998 Page 10:                '*

conservatively continues vessel cooldown at 80F/hr. For this scenado, a peak bulk , suppression pool temperature of about 170F at about 17000 seconds was calculated. An angn== ing evaluation of the limiting Abnormal Operational Transients resulting in SRV exhaust to the suppression pool for the purpose of determining peak bulk temperatures

                          ' was pe formed. Based on peer review, it has been determined that treatment of assumptions and inputs including assumed initial conditions, use of the RETRAN code.
  • op-.h actions, single failures, etc. is conservative. Companson of the RETRAN code ind model to another BWR-4 analysis of torus temperature provides confidence that the analytical results from REIRAN for Vermont Yankee are appropriate. It should be n6ted that the REIRAN code and model are fully approved for ACT analysis. In part, it was the latent of this aafnaa ing evaluation to provide confidence that REIRAN was correctly
                            . applied, within its caa=Mties and restdctions. We peak suppression pool temperatures evaluated for SRV exhaust scenados are less than 180F. Rus, based on this engineedng evaluation, it is concluded that AOTs resulting in reactor coolant discharge to the suppression pool, in consideration of the new stainer design, reduced RHR HX
_ pe fonnance, and proposed operating temperature limits of 85 F and 90 F for service water and the torus, respectively maintain acceptable torus temperatures for the bounding transient based on an upper limit of 185F based on the large break LOCA analysis.
                         - ne angla== ing evaluation summanzed in this memo has not been fully completed or -

i w' documented in a calculation at this time._ However, it is our bestjudgement based on the g bounding case described previously, a benchmark to a peer BWR and peer review of this - work that our conclusions are valid and the peak predicted torus temperature will not be above 180 F. He task to complete the work and finish documentation is presently underway. A separate transmittal will be provided when the calculation is completed and

' approved.

SAFETY EVALUATION I , This memorandum provides an engineering evaluation of a podion of the safety analysis for

maintenance of the torus integdty during transients and accidents and is therefore safety related.

I i L e l I I

Torus Tcrnperature and Pressure Response to Large Break LOCA cnd MSLB Accident Scenari s VYC-1628 Rev. 0 Page/Lc.p Appendis 1. J.R. Chapman THSAG VY 98-068 April 24, 1998 Page 11 .

                                            ~

Prepared By: 1 N

                                                                                                              ,S.J.G g", Consultant                    / Da(e
                                                                                                               'Ihermal Mydraulics and Safety Analysis Nucicar Engineering Department Prepared By:               -

h #/- Y"O8 M. P. IIF[ancols. Lead Engineer Date Thermal Hydraulics and Safety Analysis Nuclear Engineering Department -- Approved By: b kou~J du #!-l'[-f[ M. F. Kennedy, Manager Date Safety Assessment Nuclear Engineering Department cc. P. A. Bergeron K. R. Rousseau N. Fujita J. Pappas C. D. Fago D. M. Kapitz J. E. Metcalf J l i I

l Torus Temperature and Pressure Response to Large Break LOCA cnd MSLB Accident Scenarios VYC-1628 Rev. 0 . Pagej/Q Appendix I. J.R. Chapman THSAG W 98-068 i April 24, 1998 Page 12 .. TABLE 1  ! FSAR AOT Review for SRV Diseharae l FSAR Transient Section SRV Discharge? 14.5.1.1 GeneratorTdp ( 14.5.1.2 Yes ' Turbine ' nip 14.5.1.3 Yes Main Steam Line Isolation Valve Closure Yes 14.5.2.1 loss of Feedwater Heater 14.5.2.2 No Shutdown Cooling Malfunction . No 14.5.2.3 Inadvertent Pump Start 14.5.3.1 No i Continuous Rod Withdrawal Dudng Power Range Operation No 14.5.3.2 Continuous Rod Withdrawal During Reactor Startup 14.5.3.3 No - Control Rod Removal Error During Refueling No 14.5.3.4 Fuel Assembly insertion Error During Refueling { 14.5.4.1 No Pressure Regulator Failure l 14.5.4.2 No Inadvertent Opening of a Relief Valve or Safety Valve Yes j 14.5.4.3 loss of Feedwater Flow 14.5.4.4 No loss ofAuxiliary Power Yes 14.5.5.1 Recirculation Flow Control Failure - Decreasing Flow 14.5.5.2 No Trip of One Recirculation Pump No { 14.5.5.3 Trip of Two Recirculation Pumps 14.5.5.4 No Recirculation Pump Seizure No 14.5.6.1 { Recirculation Flow Control Failure - Increasing Flow No 14.5.6.2 Startup of an Idle Recirculation Pump 1 14.5.7.1 No I ' loss of RHR Service Water Flow No 14.5.8.1 Feedwater Controller Failure - Maximum Demand No 14.5.9 loss of Habitability of the Main Control Room 10.5.2 No I Fuel Pool Cooling - Safety Objective No i L )

Torus Temperature and Pressure Response to Large Break LOCA rnd MSLB Accident Scentrios P:ge //4L VYC-1628 Rev. O Appendix 1. J.R. Chapman THSAG VY 98-068 ') April 24, 1998

         ,Page 13                                            ,

Table 2 I SRV E=8= ===t L 20s D- "^ 2=d in Poten*a.ts, ria.:::_, g..,,,ug.a pas;

                                                                                                                                                                     ~

Temperatures 4 Mitigative , Initiating Available Case Postulated - Remaining Mitigative { Systems Event Mitigative . . Single Mitigative Systems Systems Failure . Systems Assumed 1A RHR HX HPCI PW HPCI (1 train) RCIC Turbine RCIC FW- Bypass FW hrbine Cose Spray htt>tne Bypass Shutdown - Bypass Core Spray Cooling Core Spray - Shutdown RHR HX .

                                                                              ~ Shutdown                               Cooling        (1 tr'ain) .                   -

Cooling SRVs

                                                         .         Stuck        ,SRVs                                  RHR HX Open                                                (1 train)                                           t Relief RHR HX        1B      MSIV            PW             FW yg                             Valve .

Isolation (2 trains) HPCI Core Spray HPCI - (SORV) - RCIC Shutdown . y RCIC Core Spray N Cooling . ' PW Shutdown RHR HX

                                           . Turt>lne                                                                 Cooling -      (2 trains)

Bypass- SRVs Core Spray RHR HX Shutdown (2 trains) Cooling j SRVs 2A- RHR HX HPCI RCIC*

                                                                                 - HPCI                (1 train)      RCIC -         Core Spray
                                                                                 ' RCIC                               RHR HX -       Shutdown Pool.                                      Core Spray                              (1 train)      Cooling
                                        . RHR HX                               Shutdown                                              SRVs (2 trains)          1.oss of       Cooling                                              RHR HX Offsite          SRVs                                              (1 train)

Power 2B RHR HX SORV HPCI RCIC (2 trains) RCIC Core Spray . RHR HX Shutdown

                                                        '                                                             (2 trains)     Cooling RHR HX (2 trains)

[- "For the purposes of providing a bounding case Case 2A has been analyzed assuming the

                     . availability of feedwater for inventory control throughout the event. 'Ihis assumption provides vessel inventory makeup with an enthalpy which far exceeds that of the available sources, the Condensate Storage hnk (CS11 and the torus inventories, i-f 9

I- l I-t .. . . i}}