ML20091C178

From kanterella
Jump to navigation Jump to search
Cycle 16 Core Performance Analysis Rept
ML20091C178
Person / Time
Site: Vermont Yankee File:NorthStar Vermont Yankee icon.png
Issue date: 02/10/1992
From: Paul Bergeron, Sironen M, Slifer B
VERMONT YANKEE NUCLEAR POWER CORP.
To:
Shared Package
ML20091C171 List:
References
YAEC-1844, NUDOCS 9204030098
Download: ML20091C178 (98)


Text

-

myy:,-ey >t,- , _

.a n<,, q u,,+-u : -

[![

+ - .

p?l i: YANKEE ATOMICLELECTRIC/ COMPANY!

L e <. <

to ,

e4 1<

., , s :.

.:'.) U j g '.. -'

e -;

)

. . , . 4, tr- Verm:nt Yankee Cycle 16 s.;

Core Performance Analysis Repott

+ ~  ::

-t b

o l

-i

, < ., 'i

'4.

1.;

3 .

e h%}

YANKEIE

~ -.

1 J g x r

'i,--

.A t-f,,.,,

(<-

-i 9204030098 920331 PDR ADOCK 05000271 p PDR c- -%

IE& ...-

s

.__ . ._.m..__ . -_ . . - __ _ _ - _ _ . . . . .- . . . _. _ ._

Q'.

Prepared by: -

( M 41Xi4L> d!7 /fel M..A. Sirotfen '(Dat e)  :;

Nuclear Engineering Coordinator Approved by: 88<F 7 4//c) A f R. f. Cacciapoytf, Manager '

(Dat'e)

Re6ctor Physig.Y Group Approved by: . M ^eM @b P. A. Bergeron, fanager (Date)

Transient Anal is Group Approved by: M W N b R. K. Sundaram, Manager (b at'e )

LOCA Analysis Group Approved by: GceNJ A A4 2Mit B. C. Slf fer, Director (Date)

Nuclear M.gineering Department Yankee Atomic Electric Company Nuclear Services D! vision

'580 Main S r a*.

Bolton, Massachuse ;; 01740 11-WPP40/10

++*W T %s.- y em-o..t-ee aem, e -r.---g.*.wr-7-r 1c++w -m _g.-W -

73-r y.- s-- m-m--.m - - -


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

D7SCL A7MrH or iytrot:SIBIL7'rY

, This document was prepared by Yankee Atomic Electric Company

("Yankve"). The use of information contained in this document by anyone other than Yankee, or the organization for which this document

.; was prepared under contract, is not authorized and, with respect tn any unauthordred use, neither Yankee nor its officers, directors, agents, or employees assume any obligation, responsibility, or liability or make any warranty or representation as to the accuracy or j completeness of the material contained in this documsnt. i i

l I

3 1

l l

4 i

t b

i 1

t

-iii-WPP40/10

,,. - -,_..;_..__....__.._. _ _ . _ - . _ _ _ . _ , _ . . - . _ _ - - . . _ . - _ . - . _ , _ _ . , . _ _ _ . ._.__.a

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

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

1 4 APSTRACT This report presents design information, calculational results, and operating limits pertinent to the operation of Cycle 16 of the Vermont Yankee Nuclear Power Station. The1e include the fuel design 4

and core loading pattern descriptionsi calculated reactor power distributions, exposure distributions, shutdown capabflity, and reactivity datas and the results of safety analyses performed to justify plant operation throughout the cycle.

l l

)

s I

i l

h b

t i

=-iv- -

WPP40/10 r w d , A ,, L ,. ,,,, -- w ,. e r,rn,w ---,,-w,.e,v r,,,<-r.n-~,,

l TA91T,or coHTrNTS [

M ],

DISCLAIMER OF RESPONSIDTLITY . . . . . . . . . .- . . . . . . . . iii j

-ABSTRACT- . . . .. . . . . . . . . . . . . . . . . . . . . . . . Av ,

TABLE OF CONTENTS . . . . . . . . . . . . . . . . . . . . . . . . V LIST Or TIGURES . . . . . . . . . . a . . . . . . . . . . . . . . vii LIST OT-TABLES . -. . . . . . . . . . . . . . . . . , . . . . . . . ix  !

ACKNOWLEDGEMENTS . . . . . . . . . . . . . . . . . . . . . . . . . x  !

l

1.0 INTRODUCTION

. . - . . . . . . . . . . . . . . . . . . . . - . 1 h I

2.0 RECENT REACTOR OPERATING HISTODY . . . . . . . . . . . . . 2 2.1. ' Operating History of the current Cycle . . . . . . . . 2 2.2' Operating History of Past Applicable Cycle . . . . . . 2 5

3.0 RELOAD CORE DESIGH. DESCRIPTION

. . . . . . . . . . . . . . 5 3.1 Core Tuol Loading . . . . . . .- . . . . . . . . . . . 5

-3.2 Design Reference Core Loading Pattern . . . . . . . . -5'  !

3.3 Assembly Exposure Distribution . . . . . . . . . . . . 5

4.0 TUEL MECHANICA1JAND THSRMAL DESIGN , . . . . . . . . . . . .- 0 4.1 Mechanical ~ Design . . . . . . . .- . . . . . . . . . . 8 .l 4.2 Thermal Design .-. . . . . . . . . . . . . . . . . . 8 4.3 Operating Experience . . . . . . . . . . . . . . . . . -

9 5.0' NUCLEAR DESIGN . . . . . ' - . . . . . . . . . . . . . . . .. 15 51 Core Power Distributions . . . .. . .

.- . . . . . . .. , 15-

~5.1.1 Haling Tower Distribution .-. . . . . . . . . . 15 5.1.2 Rodded Depletion Power Distribution . . . .- . .- 15 i

. 5.2 Corn Exposure' Distributions- . . . . .- . . . . . . . . 16- 3 5.3 Cold Shutdown Margin . . . . . .. . . . . . . . . . . . - -16  :

5.4 Maximum K. for the Spent Tuel Pool . . . . . . . . . . 17 6.0 THEPF.AL-HYDRAULIC DESIGN . . . . . . . . . . . . . . . . . 26 6.1 Steady-State Thermal Hydraulics . . . . . . . . . . . 26 6.2 Reactor Limits Determination . .' . 4 . . . .- . . . . . . 26-  ;

-7.0 ABNORMAL OPERATIONAL TRANSIENT ANALYSIS . . . . . . . . . . 28  ;

.7.1' Pressurization) Transient Analysis . . .- . . . . . . . . 20-

.y.

WPP40/10 1

g

I TABtr OF TONTENTS  !

(Continued) f EARA l 7 1.1 Hethodology . . . . . . . . . . . . . . . . . . 28 ,

7.1 2 Initial Conditions and Assumptions . . . . . . 29 t 7.1.3 One-Dimensional Cross Secsions and Kinetics .

Parameters . . . . . . . . . . . . . . . . . . 31- l 7.1.4 Pressurization Transients Analysed . . . . . . 32 7.2 Pressurization Transient Analysis Results . . . . . . 33  !

7.2.1 Turbine Trip Without Bypass Transient (TTWOBP) . . . . . . . . . . . . . . . . . . . 33 7.2.2 Generator Load Rejection Without Bypass  !

Transient (GLRWOBP) . . . . . . . . . . . . . . 33 ,

7.2<3 Loss of Teedwater Heating Transient *

(LOFWH) . . . . . . . . . . . . . . . . . . . . 34 i 7.3 Overpressuritation Analysis Results . . . . . . . . . 35 7.4 L0 cal Rod Withdrawal Error Transient Results . . . . . 35 7.5 Hisloaded Bundle Error Analysis Results . . . . . . . 38 t 7.5.1 Rotated Bundle Error . . . . . . . . . . . . . 38 7.5.2 Hislocated Bundle Error . . . . . . . . . . . . 39 8.0 DESIGN BASIS ACCIDENT ANALYSIS . . . . . . . . . . . . . . 72 8.1 control-Rod Drop Accident Results . . . . . . . . . . 72 8.2 Loss-of-Coolant hacident Analysis . . . . . . . . . . 73 8.3 Refueling Accident.Results . . . . . . . . . . . . . . 74 7 9.0 CORE COMPONENT 00ALIFICATION PROGRAM- . . . .. . . . . . 78  ;

9.1 Siemens Nuclear Power Tuel Assemblies . . . . . . . . 78 10.0 STARTUP PROGRAM .. . . . . . . . . . . . . . . . . . . . 80 REFERENCES . . . . . . . . . . . . . . . . . . . . . . . . . . . 81 APPENDIX A . . . . . . . . . . . . . . . . . . . . . . . . . . . 83

.. -vi-WPP40/10 t

LIST OF FIGURES j i

Number Tit 1e EJLQ2 ,

3.2.1 VY Cycle 16 Design Reference Loading Pattern, Lower Right Quadrant . . . . . . . . . . . . . .. . . 7 4.2.1 VY Cycle 16 Core Average Gap Conductance Versus Cycle Exposure . . . . . . . . . . . . . . . . . . . . 13 l

.4.2.2 VY Hot Channel Gap Conductance for GE8h8HD Versus Exposure . . . . . . . . . . . . . . . . . . . .. . . 14 ,

5.1.1 VY Cycle 16 Haling Depletion, EOTPL Bundle  !

Average Relative Powers . . . . . . . . . . . .. . . 19 ;

5.1.2 VY Cycle 16 Haling Depiction, E0rPL Core Average Axial Power Distribution . . . . . . . . . . .. .. . 20 5.1.3 VY Cycle 16 Rodded Depletion - ARO at EOPPL, Bundle Average Relative Powers . . . . . . . . . . . . 21 5.1.4 VY Cycle 16 Rodded Depletion - ARO at EOFPL, Coro Average Axial Power Distribution . . . . . . . .. . . 22 5.2.1 VY Cycle 16 Haling Depletion, EOFPL Bundle 23 i Average Exposures . . . . . . . . . . . . _ . . . . . . .

$.2.2 VY Cycle 16 Rodded Depletion, EOrPL Bundle

, Average Exposures . . . . . . . . . . . . . . ., . . 24 5.3.1 VY Cycle 16 Cold Shutdown AK in Percent Versus >

Cycle Exposure . . . . . . . . . . . . . . . . ... . 25 >

7.1.1 riow Chart for the Calculation of ACPR Using the RETRAN/TCPYA01 codes . . . . . . . . . . . . .... . 44

.7.2.1 Turbine Trip Without Bypass, EOFPL16 Transient Response Versus Time, " Measured" Scram Time . .. . . 45 ,

7.2.2' Turbine Trip Without Bypass, E0rPL16-1000 mwd /St Transient Response Versus Time, " Measured" Scram .

Time . . . - . . - , . . . . . . - . . . . . . . . . . . . . 46 7.2.3 Turbine = Trip Without Bypass, EOFPL-2000 mwd /St Transient Response Versus Time, " Measured" Scram.

Time . . .... . . . . .. . . . . . . . . . . . . . . 51 9

?

-vii-WPP40/10 l

,, - , , . . . , . . . . . _ .._ _ . ,,...,.-__..__,,.m___#r,,_

I l

LIST Or PfGUkr$

(Continued) ,

&,Mpi Tit 1e E,0,qr, 1,2 f, thqt/pt" Load Rejection Without Bypass, EOPPL16

,:.nsient Response Versus Time, " Measured" Scram i f1 ae . . . . .. . . . . . . . . . . . . . . . . . . . 54 ,

i

" Ji - b wenerator Load Rejection Without Bypass,  !

EOTPL16-1000 mwd /St-Transient Response Versus

  • Time, " Measured" Scram Time . . . . . . . . . . . . . 57  ;

- 7.2.6 Generator Load Rejection Without Bypass, ,

EOTPL16-2000 mwd /St Transient Response Versus  !

Time, " Measured" Scram Time . . . . . . . . . . . . . 60 7.2.7 Loss of 100*r reedwater Heating, BOC16 (Limiting I Case) Transient Response Versus Time . . . . . . . . . 63 f 7.3.1 MSIV Closure, Flux Scram, E0rPL16 Transient Response Versus Time, " Measured" Scram 71me . . . . . 65 7.4.1 Reactor Initial Conditions and Transient Summary I for the VY Cycle-16 Rod Withdrawal Error Case 1 . . . 60 7.4.2 Reactor Initial Condition 1 and Transient Summary l for the VY Cycle 16-Rod Withdrawal Error Case 2 .. . 69 y VY Cycle-16 RWE Case 1 - Setpoint Intercepts

) 4.3 Determined by the A and C Channels . . . . . . . . . . 70 .

f 7.4.4 VY_bycle16RWECase1-SetpointIntercepts

' Determined by the B and D Channels . . . . . .. . . . 71 8.1.1 First Four Rod Arrays Pulled in the A Sequences . . . 76 8.1.2 First Four Rod Arzays Pulled in the B Sequences . . . 77 t-t y

f

-viii-WPP40/10. i

LTST or TABLES Number J.$,,1,lg E,0qa 2.1.1 VY Cycle 15 Operating Highlights . . . . . . . . . . . 3 2.2.1 VY Cycle 14 Operating Highlights . . . . . . . . . . . 4 3.1.1 Assumed VY Cycle 16 ruel Bundle Types and Numbers . . 6 3.3.1 Design Basis UY Cycle 15 and Cycle 16 Exposures . . . 6 4.1.1 Nominal ruel Muchanical Design Parameters . . . . . . 10  ;

4.2.1 VY Cycle 16 Gap Conductance Values Used in i Transient Analyses . . . . . . . . . . . . . . . . . . Il e

4.2.2 Peak Linear Heat Generation Rates Corresponding '

to Incipient Pvtl Centerline Melting and 1% 4 Cladding Plastic Strain . . . . . . . . . . . . . . . 12 5.3.1 VY Cycle 16 rwu values and Shutdown Margin 1 Calculation . . . . . . . . . . . . . . . . . . . . . 18 5.4.1 VY Cycle 16 Maximum cold }L of any Enriched Segment . . . . . . . . . - . . . . . . . . . . . . . . 18

- 7.1.1 VY Cycle 16 Summary of System Transient Model Initial Conditions for Transient Analyses . . . . . . 41 7.2.1 VY Cycle 16 Pressurization Transient Analysis  ;

Results . . . . . . . . . . . . . . . . . . . . . . . 42 +

7.3.1 VY Cycle 16 overpressurization Analysis Result. . . . 43 7.5.1 VY Cycle 16 Rotated Bundle Analysis Results . . . . . 43  !

7.5.2 VY Cycle 16 Mislocated Bundle Analysis Results . . . . 43 6.1.1 VY Cycle 16 Control Rod Drop Analysis Results . . . . 75 9.1.1 Nominal ANT-IX ruel Mechanical Design Parameters . . . 79 A.1 Vermont Yankee Nuclear Power Station Cycle 16 MCPR Operating Limits . . . . . . . . . . . . . . . . 84 A.2 MAPLHGR Vercus Average Planar Exposure for BD324B . . 85 A3 MAPLHGR Versus Average Planar Exposure for BD326B . . 86-A.4 MAPLHGR Versus Average Planar Exposure for BP8DWB311-100Z . . . . . . . . . . . . . . . . . .. . 87 A.5 MAPLHGR Versus Average Planar Exposure for BP8DWB311-11GZ . .. . . - . . . . . . . . . . . . . . . 88

-ix-HPP40/10

i l

ACKNOWLTDGEMENTS j The author and major contributors would like to acknowledge the contributions to this work by the YAEC Word Processing Center. Their }

assistance in preparing this document As recognized and greatly I

- appreciated. ,

b I

j i

t k

1

.x.

WPP40/10

- , , , ,,,,.--,v. . , _ , . , - - - . . . . ~ ~ . _ ,

1.0 INTRODUCTTON This report provides information to support the operation of the Vermont Yankee Nuclear Power Station through the forthcoming Cycle 16.

In this report, Cycle 16 will frequently be reiorred to as the peload Cycle. The preceding Cycle 15 will frequently be referred to as the Current Cycle. The refueling between the two will involve the

, discharge of 126 1rradiated fuel bundles and the insertion of 128 new fuel bundles. The resultant core will consist of 120 new fuel bundles and 240 irradiated fuel bundles. The General Electric Company (GE) manufactured all the bundles, except four qualification fuel bundles manufactured by Siemens Nuclear Power (SNP), formerly known as s- Advanced Nuclear Fut1s (ANF). Some of the irradiated fuel war also present in the reactor in Cycle 14. This cycle will frequently be 2

referred to as the Past Cycle.

This_ report contains descriptions and analyses results pertaining to the mechanical, thermal-hydraulic, physics, and safety aspects of the Reload Cycle. The analyses assumed _the reload core contained all

-GE bundles. This is justified-because the SNP bundles-were designed to match the GE bundles- Section 9.0 describes the Reload Cycle Core Component Qualification Program and its impact on the analyses. The MAPLHGR and MCPR operating limits calculated for the Reload Cycle are given in Appendix A. These limits will be included in the Core Operating Limits Report.

l P&P40/10

2.0 RECrNT FEACTOR OPERATI!!G HISTORY 2.1 Operat ino Hi st ory of the Current Cvele The current operating cycle is Cycle 15. To date, th6 Current Cycle has been operating at, or near, full power with the exception of sequence exchanges, a one short repair outage, four scrams, and a coastdown to the end of cycle. The operating history highlights and control rod .equence exchange schedule of the Current Cycle are found in Table 2.1.1.

2.2 Operatina Hitterv of Pest Applicable Cvglg

~

The irradiated fuel in the Reload Cycle includes some fuel bundles initially inserted in Cycle 14. This Past Cycle operated at, or near, full power with the exception of sequence exchanges, a short repair outage, two scrams and a coastdown to the end of cycle. The operating history highlights of the Past Cycle are found in Table 2.2.1. The Past Cycle is described in detail in the Cycle 14 Summary Report [1].

ejo WPP40/10

. _ . . _ _ _ _ _ . _ _ _ _ , - . . ~ . . - _ _ _ - . _ . _ _ _ _ . _ . _ _ _ . _ . - . . - , _ . _ _ . _ . _

TABLE 2.1.1 i

VY Ovele 15 Operatino Hichlichts i

, Beginning of Cycle Date October 14, 1990 End of Cycle Date March 7, 1992'

' l Weight of Uranium An-Loaded (Short Tons) 72.95 Beginning of Cycle Core Average Exposure" (HWd/St) 10809 End of rull Power Core Average Exposure" (HWd/St) 20069 End of Cycle Core Average Exposure" (HWd/St) 20930' Number of Fresh Assemblies 128 1

Number of. Irradiated Assemblies 240 i Control Rod Sequence Exchange Schedule: '

Sequence M I.L92 .T2

  • December- 15,-1990- A2-1 B1-1 January 28, 1991- B1-1 B2-1 ,

March 12,.1931 -B2-1 Al-1 ,

May 19, 1991 Al-1 A2-2 July 21, 1991 A2-2 B1-2 September 14, 1991 B1-2 B2-2 November 23, 1991 82-2 Al-2 F

v Projected dates and exposures.

According to the Plant Process Computer.

6TP40/10

T A91 P 2 . 2 .1 VY Cvele 14 Operatino Richlichtc Beginning of Cycle Date April 8, 1989

- End of Cycle Date August 31, 1990 Weight of Uranium As-Loaded (Short Tons) 73.94 Beginning of Cycle Core Average Exposure' (mwd /St) 9195 End of Full Power Core Average Exposure' (mwd /St) 18343 End of Cycle Core Average Exposure * (mwd /St) 19642 Number of Fresh Assemblies 136 Nurber of Irradiated Assemblies 232 Control Rod Sequence Exchange Schedule:

Sequence E.a,Lt f.I.9.0 T.2

-June 3, 1989 A2-1 B1-1 July 29, 1989 B1-1 A1-1 Septerber 23, 1989 Al-1 B2-1 November 18, 1989 B2-1 A2-2 January 6, 1990 A2-2 B1-2 March 21, 1990 B1-2 Al-2 May 19, 1990 Al-2 B2-2 July 7, 1990 B2-2 A2-3 L

According to the Plant Process Computer.

WPP40/10

3.0 RELOAD CORE DESIGN DESCRIPTION 3.1 Core rue) Leadina The Reload Cycle core will consist of both new and irradiated assemblies. All the assemblies have bypass flow holes drilled in the lower tie plate. Table 3.1.1 characterizes the core by fuel type, batch size, and first cycle loaded. A description of the fuel is found in.the GE Standard Application for Reactor ruel[2].

3.2 Desian Reference core Loadino Pattern The Reload Cycle assembly locations are indicated on the map in

-Figure 3.2.1. For the sake of legibility only the lower right quadrant is shown. The other quadrants are mirror images with bundles l

of the same type having nearly identical exposures. The bundles are identified by the reload number in which they were first introduced into the core. If any changes are made to the loading pattern at the time of refueling, they will be evaluated under 10CTR50.59. The final i

loading pattern with specific bundle serial numbers will be supplied in the Startup Test Report.

3.3 Assembly Exposure Distribution The assumed nominal exposure on the fuel bundles in the Reload Cycle design reference loading pattern is given in rigure 3.2.1. To obtain this exposure distribution, the Past Cycle was depleted with -

the SIMULATE-3 mode 1[3),[4] using actual plant operating history. For the Current Cycle, plant operating history was used through March 20, 1991. Beyond this date, the exposure was accumulated using a best-astimate rodded depletion analysis to End of Full Power Life (EOPPL) followed by a projected coastdown to End of Cycle (EOC).

  • Table 3.3.1 gives the assumed nominal exposure on the Current Cycle and the Beginning of Cycle (BOC) core average exposure that results from the shuffle into the Reload Cycle loading pattern. The Reload Cycle EOFPL core' average exposure and cycle capability are provided.

v&P40/10

_ _ . ~ . . . _ _ _ _ _. _ . _ . . _ . , . _ .__ _ __

TABLE 3.1.1 Assumed VY Cvele 16 ruel Bundic Tvoes and Numbers ruel Reload Cycle Number Desias.st,ipjl Desionation Loaded of Bundles Irradiated BD3263 R13A 14 68 BD324B R13B 14 44 BP8DNB311-104Z R14A 15 64 BP8DWB311-11GZ R14B 15 64 New BP8CWB311-10GZ R15A 16 40 i

BP8DWB311-11GZ R15B 16 88

[

TABLE 3.3.1 Desian Basis VY Cve3e 15 and Ovele 16 Exposures

  • Assumed End of Current Cycle Core Average 20.872.6 GWd/St Exposure with an Exposure Windot- of 1 600 mwd /St[5]

-Ass'umed Beginning of Reload Cycle Core Average 11.22 GWd/St Exposure Haling Calculated End of Full Powe'.' Life Reload 20.46 GWd/St Cycle Core Average Exposure Reload Cycle Full Power Exposure Capobility 9.24 GWd/St

~

. According to SIMULATE-3.

6-WPP40/10

_._.m R13B RitB R138 R16B R13A RitB R13A RISB R14A R14A R13A 20.378 0.000 20.614 0.000 21 256 0 000 21.947 0.000 11.100 12.758 22 482 R16B R14A RitB R14A R168 R14B R1$8 AldB R15A RIAA R13B 0.000 9.006 0.000 11.307 0.000 12.102 0.000 12.710 0.000 12.982 22 885 R13B R1tB R13B R15B R13A RitB R14A R16A R14A R14A R138 20.672 0 000 20 431 0.000 21.663 0.000 13.213 0.000 12 94, 13 094 2$ 499 R168 R14A R188 RIAA R1EB R148 RICA R16A R14B R12B l 16 0.000 11397 0.000 9.104 0.000 12.285 0.000 0.000 12.923 23 608 R13A R1EB R13A Ri$B R13A R16B R14B R148 R13A 14 21.047 0.000 21.967 0.000 22.313 0.000 12 691 12.645 22 471 R160 R14B RitB R148 R16B R140 R15A R14B R13B 12 0.000 12 062 0.000 12.298 0.000 12.467 0.000 12.877 22.t99

, ~

R13A R1tB R14A R16A R148 R16A R14B R13A R13A 10 22.159 0.000 13246 0.000 12.590 0.000 12.520 22.026 25.379 R168 R148 R15n A16A R140 R14B R13A R13A 08 1 0.000 12.674 0 000 0.000 12.661 12.850 '21.923 25202 R14A R15A R14A R148 R13A R13A r.13B 06 11.135 0.000 12.791 12.963 22.433 23.095 2C.430 R14A R14A R14A R138 poott n reti,ers GNA?!cN 12.809 12.861 13.079 23 535 R13A B0326B R13B B03243 R13A R13A R138 .... .. BUNDLE ID R14A BP20WB311-10CZ R14B BP8DWB311-llG: .02 22.504 22.635 25.476 -... 800 EXPOSURE (GWD/ST) y [ $

23 25 27 29 31 33 35 37 39 41 43 LIGURE L2.1 VY Cycle 16 Demian Reference Loadino Pattern, Lower Richt Ouadrant 7

WPP40/10

4.0 FUrt MrcHANicAL AND THrup>st DrsTGN I

4.1 Mechanical Desian Most of the fuel to be inserted into the Reload Cycle was fabricated by GE. The major mechanical design parameters are given in Table 4.1.1 and Reference 2. Detailed descriptions of the fuel rod mechanical design and mechanical design analy+es are provided in Reference 2. These design .alyses remain valid with respect to the Reload Cycle operation. Mechanical and chemical compatibility of the fuel bundles with the in-service reactor environment is also addressed in Reference 2.

4.2 Thermal Desion The fuel thermal effects calculations were performed using the FROSSTEY-2 computer code [6). The FROSSTEY-2 code calculates pellet-to-cladding gap conductanco and fuel temperatures from a combination of theoretical and empirical models which include fuel and cladding thermal expansion, fission gas release, pellet swelling, pellet densification, pellet cracking, and fuel and cladding thermal conductivity.

The thermal effects analysis included the calculation of fuel temperatures and fuel cladding gap conductance under nominal core steady state and peak linear heat generation rate conditions. Figure 4.2.1 provides the core average response of gap conductance. These calculations integrate the responses of individual fuel batch average operating histories over the core average exposure range of the Reload Cycle. The gap conductance values are weighted axially by power distributions and raditlly by volume. The core-wide gap conductance values for the RETRAN system simulations, described in Sections 7.1 through 7.3, are from this data set at the corresponding exposure statepoints.

The gap conductance values input to the hot channel calculations (Section 7.1) were evaluated for the limiting fuel bundle type as a function of the assembly exposure. The calculation assumed a 1.4

.g.

WPP40/10

chopped cosine axial power shape with the peak power node running at the maximum average planar linear heat generation rate (MAPLHGR) limits [7). Figure 4.2.2 provides the hot channel response of gap conductance for the limiting bundle type. In rigure 4.2.2, " planar exposure" refers to the exposure of the node operating at the MAPLHGR limits. Gap conductance values for the hot channel analysis were extracted from the figures using the limiting bundle exposure of any minimum critical power ratio (MCPR) limiting bundle within the exposure interval of interest. The SIMULATE rodded depletion (Section 5.1.2) provides predictions of both limiting MOPR and the associated bundle exposure for the entire cycle.

Table 4.2.1 provides the core average and hot channel gap I conductance values used in the transient analyses (Section 7.1). Fuel rod local linear heat generation rates (LHGR) at fuel centerline incipient melt and 1% cladding plastic strain as a function of local axial segment exposure for the peak gadolinia concentrations used in the Vermont Yankee fuel bundles are shown in Table 4.2.2. Initial conditions assumed that fuel rods operated at the local segment power level of the maximum allowable LHGR prior to the power increase. --

4.3 Operatino Experienge All irradiated fuel bundles scheduled to be reinserted in the Reload Cycle have operated as expected in Past Cycles of Vermont Yank'ee. Off-gas measurements in the Current Cycle indicate that a fuel rod failure has occurred. Vermont Yankee is planning to identify i and discharge the failed rod (sf during the outage.

s 1

WPP40/10

..,_ _ . ___ _ ,._..--.____ -. _ .. . _ ._.._ .._,_ --.._. _ _ ,_.-. _ . - . . ~ . - . _ . _ . - _ -

. . ___._..._.._._..___.__._.__._.._m _ . _ . . . _ _ . . _ . _ . _. ..m___...- _ . _ _ _ _ _ . . . - . _ . . _

i TABLE 4.1.1

'l-  !

Nominal ruel Mechanical Desian Parameters

+

Irradiated Fuel Tvoes New ruel Tvr>e ruel Bundle' Twice-Burned Once-Durned Bundle Types BP8X8EB GE8XBNB- GEBX8HB Vendor BD324B 6 BP8DWB311-10GZ s BP8DWB311-10GZ s  !

Designation BD326B BP8DWB311-11GZ BP8DWB311-11GZ Initial 3.24 & 3.26 3.11 3.11 '

Enrichment,w/o Um

' Rod Array .BXB 8X8 8X8 t

l'uel Rods per 60 60 60 Bundle Outer ruel Channel. .

Material Er-4 Zr-2 Zr-2 Wall Thickness, 0.000 0.080 0.080 i inches  !

o ,

e

  • . Complete bundle, rod, and pellet descriptions are found in Reference 2. . ,!

t WPP40/10 i

'l

1 TABir 4.2.1 .

i VY Ovele 16 Gap Conductance Values Used in Transient Analyses i

l Ovele Exoouure Core Averaae Hot Channel Hot Channel Gap Stateooint Guo Conductance Bundle Exposurg Condu ct a n ce'. !

(BTU /Hr-Pt 8 *P) (mwd /St) (DTU/Hr-Ft 8 *P)

BOC 3120 7930 4900 EorPL-2000 mwd /St 4670 6576 5010 t E0rPL-1000 mwd /St 4890 10005" $000" i EOTPL 5000 10005" $0B0" l

}

l 5

?

i s

e

)

Hot channel gap conductance values are derived' for the BP8DWD311-10GZ fuel type because it is conservative compared >

to the other fuel types. l Corresponding to the maximum encountered in the. exposure range.

WPP40/10 4

. I

?Mit t 4 . 2 2, Peak LLr1Nr Hee.t Generation Bates Correspondino t o Incituent Tel Centerline Meltina and 1% 02nddino Plastic Strain' DLtQ lyng Qrctg.g h 0.w/o Gd>Q, " 0 w/o Gd,23 GIlxPrB fMWd/Eti g g g g,

  1. N/ft) ,$W/ft) (kW/ft) (kW/f t ).

BD324D 0 74.0 24.0 21.0 23.0 and 35,000 24.0 24.0 20.0 20.0 .l BD326B 20,000 24.0 16.0 19.0 12.0 l

ruel type gnosure 0,0 w/o GAR, 4.0 w/o c1,g, ,

GE8XBNB (mwd /bt) g g g g  ;

(kW/ft1 (kW/ft) (kW/ft) (kW/ft)

BP8DWB311- 0 24.0 24.0 21.0 23.5 21GZ {

and 25,000 24.0 24.0 20.0 20.0  :

DP8DWD311-50,000 20.5 13.5 16.5 11.0 10GZ -

i l

i 2eak linear heat generation rates shown are minimum bounding values to the occurrence of the given condition.

WPP40/10

., e-.. .Iam:,,- ,w ..w,. , n ..,n ,

.,..,,,,_..._,,p. ., ,,..m...,, . , , , . , , , , , - . , ,emn, em_- .e,..- ,. ~ --s er.-,,e

. . - , . . - - - . - . . _ . - . . . . _ - , _ . . - ~ _ - - - _

i W Cycle if *,o's Awap Gap Ctrdxave m__

d M.

F 1 f":

4 4No.

L ')

t,

u. ~

4 m-r i

1 M i i . . . , , , ,

0 ,1 2 8 4 6 6 7 e i to Cr.h (gan (0*t17)

F PIGURE 4.2.1 VY Cvelo 16 Core Averace Gap Conductance Versus Cvele Expop,y,Lg I

-2 3 -

WPP40/10 I

l l

Wib:Chave! Gap Condxtaw br GE8MB em r

go . p:  ;~0-:  : : e *+-:  ; _ :,

f I

00-l '

20-d 1

M0 -

l <

0

-1 1M, c , > i i e $ 10 il N tl Parat (gos.re @C$7)  !

FIGURE 4.2,2 VY Hot Channel Gao Conductance for GE8X8tiB Versus Exposure

~14-WPP40/10 s

-- ,,-.u- , - . n , ., - , . . , . , . . ,.,w , - e

5.0 NUCLEAR DES 2CN 5.1 Core Power Distributions The Reload Cycle was depleted using SIMULATE-3 to give both a rodded depletion _and an All Rods Out (ARO) Haling depletion.

5.1.1 Halina Power Distribution The Haling depletion serves as the basis for defining core reactivity characteristics for nost transient evaluations. This is primarily because its flat power shape has conservatively weak scram characteristics.

The Haling power distribution is calculated in the ARO condition.

The Haling' iteration converges on a self-consistent power and exposure l i

distribution for the burnop step to EOrPL. In principle, this should provide the overal) minimum peaking power shape for the cycle. During I the. actual cycle,' flatter power distributions might occasionally be achieved by shaping ~with control rods. However, such shaping would leave.underburned regions in the core which would peak at another point in time. Figures 5.1.1 and 5.1.2 give the Haling radial and axial average power distributions for the Reload Cycle. ,

5.1.2 Rodded Deoletion Power gigt-ibucioh The rodded depletion was used to evaluate the mislocated bundle error and the rod withdrawal error because it provides the initial rod patterns and more realistic' predictions of initial CPR values. It was ,

also used in the rod drop worth and shutdown margin calculations  ;

because it burns the top of the core more realistically than the i Haling depletion. The rodded depletion also provides the hot channel bundle exposures for the gap conductance calculation.

I To generateLthe rodded depletion, control rod patterns were developed which give critical eigenvalues at several points in the cycle and peaking similar to the Haling calculation. The resulting patterne were frequently more-peaked than the Haling, but were below 15-

>&P40/10'

--w i *=-*+-w mm,, ,y~ --y - , .g ,y g- > r mwe -www- wm e - we--eew~ -is,wm+ -e v e e & w + = r ' Nw'+m*5 e-e.mem n mWmeem' 't+*- F ov

oxpected operating limits. Howevor, ce stated above, the underburned regions of the core can exhibit peaking in excess of the Haling ,

peaking when pulling ARO at EorPL. rigures 5.1.3 and 5.1.4 give the l

ARO at E0rPL power distributions for the Reload Cycle rodded  !

depletion. Note, in rigure 5.1.4, that the average axial power at ARO i for the redded depletion is more bottom peaked than the Haling (rigure '

5.1.2). The rodded depletion would result in better scram '

characteristics at EOPPL.

r 5.2- Core Exposure Distributions  !

l The Reload Cycle exposures are summarised in Table 3.3.1. The  ;

projected BOC radial exposure distribution for the Reload Cycle is  ;

given in Figure 3.2.1. The Haling calculation produced the EOPPL radial exposure distribution given in ragure 5.2.1. Since the Haling power shape is constant,.it can be held fixed by SIMULATE-3 to give  ;

the exposure distributions at various mid-cycle points. Boc, 7 EOPPL-2000 mwd /St, EorPL-1000 mwd /St, and EorPL exposure distributions west usad te develop reactivity input for the core wide transient analyses. '

The rodded depletion differs from the Haling during the cycle because the rods shape the power differently. However, rod sequences are swapped frequently and the overall exposure distribution at end of

. cycle is similar to the llaling, rigure 5.2.2 gives the EorPL radial exposure-distribution for the Reload Cycle rodded depletion.

5.3 Celd shutdown Marain Technical Specifications (B) state that, for sufficient shutdown  !

margin, the core must be suberitical by at least 0.25% AK + R (defined below) with the strongest worth control rod withdrawn. Using SIMULATE-3, a search was made for the strongest worth control rod at '

various exposures in the cycle. This is necessary because rod worths change with exposureoon adjacent assemblies. Then the cold F*r: with

'the strongest-rod out was calculated at BOC and at the end of each control rod sequence. Subtracting each cold K,,, with the strongest rod out from the cold critical K,: defines the shutdown margin as a WPP40/10 i

function of enposure, rigure 5.3.1 shows the results.

The cold critical K ,, was defined as the average calculated critical F(,, minus a 954 confidence level uncertainty. Then all cold

- results were normalized to make the critical K,,, equal to 1.000.

Because the local reactivity may increase with eet>osure, the shutdown margin (SDM) may decrease. Te account for this and other uncertainties, the value R /s calculated. R is defined as R plus Rs.

3 R is the difference between the cold FL,, with the strongest rod out at 3

BOC and the maximum cold FL,, with the strongest tod out in the cycle.

R, is a measurement uncertaints in the demonstration of SDM associated with the manufacture of past control blades. It is presently set at

.07% AK. The shutdown margin results are summarized in Table 5.3.1.

5.4 _ Maximum K_ for the soent ruel Pool i

Section 5.5E of the Technical Specifications requires that the K.

for any bundle stored in either the new fuel vault or the spent fuel pool not exceed '1.31 to ensure compliance with the T.,, safety limit of

. 0.95. The bundles used in the Peload Cycle do not exceed the specifications in Section 5.5E, as shown in Table 5.4.1. These values are obtained from CASMO-3Gl9).

5 WPP40/10

IAE!.E 5.3.1 i i

VY Cvele 16 K.,, Values and Shutdown Harain Calculation j Cold critical K.er 1.0000 i

B OC K.r - Cont rolled With Strongest Worth Rod Withdrawn .9849  !

Cycle-Minimum Shutdown Margin Occurs at BOC With I Strongest Wortu Rod Withdrawn 1.51% AK  !

Ru Haximum Increase 4n Cold Ywer With Exposure 0.00% AK i

TABLE S,4,1 VY Cvele 16 Maximum Cold Y_ of env Enriched Seement '

.r Bundle Type Maximum V%

BD324B 1.20 BD326B 1.20 BP8DWB311-10GZ 1.20 BP8DWB311-11GZ 1.20 l i

?

h WPP40/10

, - . - - - - . - - , . . , . , . - . , -_ . , _ . . , _..-.._.___..,._.._...__....-__.-._...,-._..._z,.; , . . - - - - _ _ _ . - - _ - . _ . _ . . . -

i l

R130 R16B R13Q R1EB R13A R16B R13A Q150 RioA Rt4A R13A

' 22 1.110 1.392 1.146 1.393 1.118 1.347 1.051 1.235 0.989 0.762 0.434 1

)

RISB RIAA R160 R14A R1EB R14B R1$8 R14B R16A R14A R138 l 20 1.391 1.208 1.411 1.254 1.383 1.204 1.311 1.083 1.077 0.733 0.403 R13B R15B R138 R16B R13A R16B R14A R16A R14A R14A R130 2

18 1.144 1.411 1.162 1.400 1.116 1.351 1.144 1.192 0 902 0.654 0 349 R16B Ri4A R158 RIAA. R168 R14B R15A R15A R140 A13B 1C 1.393 If 54 1,401 1.271 1.364 1.175 1.254 1.113 0.786 0456 R13A R168 R13A R168 R13A ' R158 R148 R14B R13A 14  ;

1.121 1.383 1.116 1.365 1.071 1.276 1.050 0.888 0.578 R16B R1*B R168 R148 R158 R148 RISA R148 R13B 12 1.347 1.205 1.352 1.176 9.277 1.067 1.072 0.754 0.446  ;

R13A R158 R14A R15A R140 R15A R148 R13A R13A 10 1.049 1.311 1,143 1154 1.052 1.073 0.823 0.551 0 332 R1tB R148 R15A RISA R148 RidB R13A Ri3A 08 1.234 1.083 1.192 1.114 0.888 0.753 0.552 0.354 R14A R15A R144 R146 R13A R13A R130 06 0.988 1.077 0.904 0.786 0.578 0.444 0.332 R14A R14A R14A R130 T'Jtt Cts!GN A':' Ten

~

Ptac1E 12 0.761, 0.7U 0.65.5 0.457 R13A BD3265 R138 20324a R13A R13A R13B .. m.BUNDLEID -R14A BPeDWB311-10Gt R14B BPBCWB311-110t J2 0.433 0.402 0.350 .. - EOFPL RELATIVE POWER R15A BPBDWB311-10Gt R13B BP$DWB311-110t i

23 25 27 29 31 33 35 37 39 41 43 l

FIGURE 5.1.1 1.

VY Cvele 16 Halina Deolet ion. E0FPL Bundle Averace Relative Powers i

i

.WPP40/10 l

, . , , - , - - - , , . . - - ~ - . , - . - - , , , . . , - . . . . . . , - , . _ , . . . - . _ . . . , , _ , . . . - - _ , - - ,.-..,n

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

4 I

t 1.5 --  :

1,4 __ ._ _ _ _ - _._ __ __ _ _ . _ _ _ _ ._ _ _ _ _ ,

1.3 - - - - =- - - - - - -- - =- --- - -- - -- - - - -

1.g + - - - . = --

-_: _ - x  : ; - - - -

p ~  %

1,1 __ _ _ p) -

_ _ _ = N  % _ _ _

i.0 __ _ - - -

_ _ ._ _ _ _\ __ _ = __

E o.g . _ _ _ _. ._ _ _ _ _ _ _ _ _ i. - _ _ .-

N ,

08- - - - - - - -- - - - - - - - - --- - - --- -

0.1 - - -- ~~ ~ ~ - -- > <- - -- -

~= ~

- c a06-g - -- - - - - - -- - -- -- - - - - -

0.5- - -- - - ~ ~ - - - - - ~ ~ - - -- - -- =

0,4- - - -- - - - - -

== - - - ~ ~~ - - - - - -

0.3 - - -- - - - --- - - - - - - - - - - - - - -- - - -

i .

0.2= - - - - -

--- 1 .

0.1 - =- - - =-- -- - -= - - - - - - - -

00- --- - - - = = -

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 00 21 22 23 2f 25 PLANT AND SIM 3 AXIAL NODES (TOP 25)

FIGURE $.1.2 VY Ovele 16 Italino Depletion, EOPPL Core Averaae Axial ,

Power Distribution WPP40/10

. _ - . . . _ _ _ _ . _ , . .m._ .. . - - - . . _ _ _ _ .- . . . .- _ _ . .. .- _.

1 I

i R138 R19B R138 R15D R13A R15B R13A R1&B R14A R14A A13A 22 1.094 1.348 1,121 1.401 1.148 1.385 1.076 1.242 0.986 0.752 0.430 R158 R14A R158 R14A RiSB R14B R158 R14B R15A R14A- R13B 20 1.352 1.068 1.188 1.247 1.417 1.225 1.328 1.087 1.073 0.724 0.309 R13B R158 R138 R15B R13A R158 R14A R15A R14A R14A R13B 18 1.126 1.190 0.995 1.398 1.144 1.375 1.156 1.195 0.898 0.648 0.345 R158 R14A R158 R14A R158 R148 RISA RISA R14B R138 i

1E 1.410 1.254 1.402 1.285 1.39i 1.196 1.271 1.118 0.785 0.455 R13A R158 R13A R158 R13A R158 R14B R14B R13A

_14 1.159 1.425 1.149 1.400 1.100 1.303 1.061 939 0.579 R158 R14B R158 R148 R158 R14B 915A R148 R138 1.402 1.235 1.385 1.203 1.307 1.084 1.087 0.756 0447 R13A R15B R14A R15A R148 RISA- l R149 R13A R13A 1.085 1.340 1.1 C2 1.279 1.068 1.089 0.831 0.554 0 334 R10B R140 RISA R15A R14B R149 R13A R13A 00 1.253 1.095 1.203 1.124- 0.893 0.7% 0.555 0.356 R14A R15A R14A R148 R13A R13A R130

~ ~

0.992 1.080 0.905 0.789 0.581 0.446 - 0133 R14A R14A R14A R13B Bt w n t' n ers m n u m

~

0.755 0.729 0.652 0.457 R13A BD3265 R13b BD324B R13A R13A R138 ... BUNDLEID R14A BPBDWB311-10G:

R14B BP8DW3311-11G:

0.431 0.400 0.34L ,. .... EOFPL RELATIVE POWER R15A BP80WB311-10G R15B BP b FE 311-11G i

i s i 23 25 27 29 31 33 35 37 39 41 43 FIGURE 5.1.3 VY C/cle 16 Rodded Depletion - ABO at EOFPL, Bundle Averace Relative Powers WPP40/10

. ~ . - - . . - - _ - - . . _ _ - , - .-

1.5 1.4= -- ---

1.3- .= - -

12-- -

,w

/

2,-

1.1 - -

1.0 -

K o.g -

'N \

f \\

E 0.8 =

s h 01 - f

$.6-g 0

- f-- - - - - - -

h '

- 0.5 - -

0,4 - 8-I 0.3 -

O2- - - - - - - -

\ ,

0.1 - -

0.0 - j - -

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 23 21 22 23 24 25 PLNJT AND SIM 3 AXtAL NODES (TOP-25)

FIGDFE 5.1.4 VY Cvele 16 Rodded Depletion - ARO at EOFPI.,

Core Averace Axial Power Distribution 6

WPP40/10 i

l l

R138 R158 R138 R158 RI3A R158 R13A RISB R14A RIAA R13A

. 30.637 12.794 31.198 12.801 31.542 12.380 31.625 11.344 20.192 19.755 26 476 R158 R14A R15B R14A RISB R14B R15B R14B R15A R14A R138 20 12.781 20.894 s2.968 22.824 12.707 23.171 12.044 22.658 9,889 19.715 26.608 R13S R15B R13B R158 R13A R15B R14A R15A R14A R14A RISB 18 31.238 12.965 31.077 12.867 32.139 12.416 23.716 10.940 21.225 19.105 28.716 R15B R14A R158 R14A R158 R14B- 815A R15A R148 R130

~

12.802 22.619 12.875 20.774 12.532 23.086 g 11.515 10.222 20.150 27.624 R13A RISB R13A R158 R13A R15B R148 R14B R13A 14 31.362 12.712 32.241 12.539 32.168 11.728 22.340 20.803 27.786 R158- R14B R15B R14B R158 R14B R15A R148 R138 i 12 12.380 23.140 11420 23,103 11.730 22.271 9.846 19.802 27.017 R13A R158 RIAA R15A R14B R15A R14B R13A RM 10 31.810 12.042 23.744 11.518 22.256 9.847 20.084 27.101 28.440 RI5S R14B R15A R15A R148 R148 R13A R13A 08 11.337 22.625 10.941 10.223 20.818 19773 27.005 28.457 R14A R15A R14A R148 R13A R13A R13B 66 20.209 9.888 21.090 20.187 27.752 27.182 28.500 R14A R14A R14A R138 etmott .to. rtTL Drs!GNA?!C.N j 19.793 19.602 19.094 27.755 R13A BD326B

-- l R13B BD324B

- BUNDLE ID R14A BPSDNB311-10GZ R13A R13A R138 BP8DWB311-11G R14B .02 25.491 26.338 28.709 . .. EOFPL EXPOSURE (GWD/ST) y ((~.[ $

23 25 27 29 31 33 35 37 39 41 43 FIGURE 5.2.1 VY Cvele 16 Haline Depletion, EOFPL Bundle Averace Exposures h?P40/10 l l

l l

I

.---;,+ ~--m , y- g-- p

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

R138 R10B R13B R15B R13A R158 R13A R158 R14A R ?.1 A R13A 29.887 10.473 30.003 11.027 31.038 11.756 31.988 11.569 21.170 20.751 27.173 RISB R14A R158 R14A R158 R148 R158 R148 R15A R14A R138 20 10.329 19.555 10.987 22.037 11.499 23 542 12.134 23.539 1r387 20.630 27.246 R13B R158 R138 R15B R13A R158 R14A R15A R14A R14A R139 29.999 10.926 30.201 11.412 31.852 12.249 24 418 11.272 22.159 19,951 29.310 R15B R14A R15B R14A R15B R148 - R15A R15A R14B R138 16 10.879 21.920 11.338 20.482 11.840 23.513 11.566 10.475 20.974 28.448 R13A R158 R13A R158 R13A R158 R148 R148 R13A 14 30.703 11.327 31.852 11.793 32.276 11.543 22.967 21.586 28.463 R158 R148 R15B R148 R158 R14B R1CA R148 R138 12 11.366 23.341 12.108 23.400 11.441 22.717 9.830 20.476 27.562 R13A. R15B R14A RISA R148 R15A l R14B R13A R13A 31.879 11.942 24.317 11.401 22.768 9.793 20.629 ,?7.660 28.891 R158 R148 R15A R15A R148 R14B R13A R13A n8 11.375 23.365 11.136 10.370 21.533 20.418 27.553 28.880 R14A R15A R14A R148 R13A R13A R13B 06 21.074 10.261 21.929 20.940 28.382 27.696 28.946 R14A R14A R14A R138 ptvett 1q rt'r. ers GNA?:cN 20.713 20.448 19.881 28.338 R13A BD32(B l R13B BD324B R13A R13A R138 BUNDLEID 'B 3 27.144 26.933 29.267 ..... EOFPL EXPOSURE (GWD/ST) yfy3fff$

23 25 27 29 31 33 35 37 39 41 43 FIGURE 5.2.2 VY Cvele 16 Rodded Depletion, EOPPL Bundle Averace Exposures WPP40/10

1 l

l l

3.0 2.9 : 1 2.8 :  !

2.7 - - -

2.6 2.5 I /

2 4 --

! /

2.3 g2  ? / -1 2.1  ! /

2.0 1.9 i f 71.8 ,

$ 1.7 - y/

$ /!  !

~ 1.4  ! -

1.3 l f 12 1.1 l - Minimum Shutdown Margin 1.0 Te$ncalSpedicahon Limit 0,9 l

OA ,

0.7 O.6

[

O.5 0.4 -

l

,,,l.... .... ........

g,3 _ ........ 2... .... ... .... .... ,,,, ........ ... ,,,, ,,,,, .... ...

0.2 --

I 0.1 0.0 ,

I 0 1 2 3 4 5 6 7 8 9 10 ,

Cyde Eposure(GWdSt)

FIGURE 5,3,1 VY Cycle 16 Cold Shutdown AK in Percent-Versus Cvele Exoosure WPP40/10

l 1

l 6.0 THERMAL-HYDRAULIC DES:GN I l

6.1 - Steadv-State Thermal Hydraulics Core steady-state thermal-hydraulic analyses for the Reload Cycle were performed using the FIBWR(10),(11),(12) computer code. The FIBWR code incorporates a detailed geometrical representation of the complex flow paths in a BWR core, and explicitly models the leakage flow to the bypass region and water rod flow. The FIBWR geometric models for each GE bundle type were benchmarked against vendor-supplied and plant thermal-hydraulic information.

Using the. fuel bundle geometric models, a power distribution calculated by SIMULATE-3 and core inlet enthalpy, the FIDWR code calculates the core pressure drop and total bypass flow for a given total core flow. The core pressure drop and total bypass flow predicted by the FIBWR code were then used in setting the initial conditions for the system transient analysis model.

6.2 Reactor Limits Determination The:cbjective for normal operation and anticipated transient events is to maintain nucleate boiling. Avoiding a transition to film boiling protects the fuel cladding integrity. The Fuel Cladding Integrity Safety Limit (FCISL) for Vermont Yankee is a Critical Power Ratio (CPR) of 1.07 (2). CPR is defined as the ratio of the critical power (bundle power at which some point within the assembly experiences onset of boiling transition) to the operating bundle power. Thermal margin is stated in terms of the minimum value of the Minimum Critical Power Ratio (MCPR) which corresponds to the most limiting fuel assembly in the core. Both the transient (safety) and normal operating thermal limits, in terms of MCPR, are derived with the GEXL-Plus correlation [13), with appropriate coefficients representative of the Reload Cycle's hot assembly fuel type.

The Reload Cycle fuel has a Linear Heat Generation Rate (LHGR) limit of 14.4 kW/ft for all bundle types. The basis for the Maximum LHGR (MLHGR) limit can be found in Reference 2.

WPP40/10

l The Reloud Cycle fuel has Average Planar Linear Heat Generation l Rate (APLHGR) limits shown in Appendix A. The Maximum APLHGR (MAPLHGR) values are the most limiting composite of the fuel mechanical analysis MAPLHGRs and the LOCA analysis MAPLHGRs. The fuel mechanical design analysis, using the methods in Reference 2, demonstrate that all fuel rods in a lattice, operating at the bounding power history, meet the fuel design limits specified in Reference 2.

The transients described in section 7.0 were analyzed to verify that design criteria in the mechanical design analysis methods was not exceeded during the transient. The LOCA analysis is described in ,

Section 8.0.

l WPP40/10

7.0 ABNORMAL OPERATIONAL TRANSIENT ANALYSIS 7.1 Pressurization Transient Analysis Transient simulations are performed to assess the impact of certain transients on the heat transfer characteristics of the fuel.

~

It is the purpose of the analysis to determine the MCPR operating limit, such that the FCISL is not violated for the transients considered.

7.1.1 Methodoloav The analysis requires two types of simulations. A system level simulation is performed to determine the overall plant response.

Transient core inlet and exit conditions and normalized power from the system level calculation are then used to perform detailed thermal-hydraulic simulations of the fuel, referred to as " hot channel calculations." The hot channel simulations provide the bundle transient ACPR (the initial bundle CPR minus the MCPR experienced during the transient).

The system level simulations are performed with the one dimensional (1-D) RETRAN model[14),[15],[16). The hot channel calculations are performed with the RETRAN[17),[18] and TCP YA01 [19) , [11) , [15 ) computer codes. The GEXL-Plus correlation [12]

is used in TCPYA01 to evaluate critical power ratio. The calculational procedure is outlined below.

The hot channel transient ACPR calculations employ a tuo-part

, process, as illustrated by the flow chart in Figure 7.1.1. The first part involves a ser3es of steady-state analyses performed with the FIBWR, RETRAN, and TCPYA01 computer codes. The FIBWR analyses utilize a one-channel model-for each fuel type being analyzed, with bypass and water rod flow also modeled. The steady-state FIBWR analyses were

. performed at several power levels with other conditions (i.e., core pressure drop, system pressure, and core inlet enthalpy) held constant. The FIBWR code tasult is an active channel flow (AP) and bypass flow (BPF) for each active channel power (AP).

PPP40/10

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

The FIBWR conditions for channel power, channel flow, and bypass flow were then used as input to steady-state RETRAN/TCPYA01 hot

channel calculations. Other assumptions are consistent with those in-the FIDNR analysis. The Initial Crit ical' Power Ratio -(ICPR) is the l key result-for each steady-state RETRAN/TCPYA01 analysis. These  !

results allow for the development of functional relationships, ,

describing AP e.s-a function of ICPR, and AF and BPF as functions of AP for each fuel type. These relationships are used in the iterative process used during the trans_ont calculations as described below and shown in Figure 7.1.1.-

The second part iterates on the hot channel initial power level.

This is necessary because the ACPR for a given transient varies with Initial Critical Power Ratio (ICPR) . However, only the ACPR corresponding to- a transient MCPR _ equal to the FCISL lirait (i.e., 1.07

+ ACFA = ICPR) is appropriate. The approximate constancy of the ,

ACPR/ICPR ratio is useful in these iterations. Each iteration requires a RETRAN hot channel run to calculate the transient enthalpies, flows, pressure and saturation properties at each time step. These are required for input to the TCPYA01 code. TCPYA01 is then-used to calculate a CPR at each time step during the transieni, from'which a transient ACPR is derived. The hot channel model assumes a chopped cosine axial power shape with a peak / average ratio of 1.4. <

As noted in Section 6.1, analyses for the Reload Cycle included benchmarking the FIBWR model against vendor-supplied thermal-hydraulic information. -Therefore, the FIBWR results of AP and BPF for a given AP and core pressure drop are passed directly to RETRAN. As shown in Figu.e 7.1.1, the current iterative process involves a single loop.

7.1.2 Initial Condit ions and Assumntions The initial conditions for the Reload Cycle are based on a reactor power level of 1664 MW o which includes a 2% calorimotric uncertainty on the reactor power level of 1631 MWu. The assumed

. Reload Cycle analysis reactor power bounds the currant licensed power level of 1593 MWu. The reactor core flow is assumed to be 100%. The core axial power distribution for each of the exposure points is based WPP40/10

l on the 3-dimensional SIMULATE-3 predictions associated with the generation of the reactivity data (Section 7.1.3). The core inlet i enthalpy is set so that the amount of carryunder from the steam separators and the quality in the liquid region outside the separators is as close to zero as possible. For fast pressurization transients, this maximites the initibl pressurization rate and predicts a more i aevere r.outron power spike. A summary of the initial operating state used for the systen, simulations is provided in Table 7.1.1.

i Vermont Yankee operators adjust core flow during the cycle for short-term maneuvering. During this type of operation, core flow may be as low as 87% while at 100% power. To ensure the safety analysis bounds these conditions, transients are reanalyzed at the limiting f exposure statepoint (limiting in terms of an increase in ACPR) et 1664 -

HWu power and 87% flow. These analyses are performed at both the

" Measured" and the "67B" scram times. The ACPR penalty (defined as the difference in ACPR) generated during this reanalysis is applied to the applicable transient ACPR results.

Assumptions specific to a particular transient are discussed in the section describing the transient. In general, the following assumptions are made for all transients:

1. Scram setpoints are at Technical Specification [7] limits.
2. Protective system logic delays are at equipment specificatien limits.
3. Safety / relief valve and safety valve capacities are based on Technical Specification rated values. '
4. Safety / relief valve and safety valve setpoints are modeled as being at the Technical Specification upper lhnit. Valve responses are based on slowest specified response values.
5. Control rod drive scram speed is based on the Technical Specification limits. The analysis addresses a dual set of scram speeds, referred to as the " Measured" and the "67B" WPP40/10

scram times. " Measured" refers to the faster scram times

.given in Section 3.3.C.1.1 of the Technical Specifications.

"67B" refers to the slower scram times given in Section 3.3.C.l.2 of the Technical Specifications.

7.1.3 One-Dimensional Cross Sections and Kinetics Parameters The one-dimensional (1-D) cross sections and kinetics parameters are generated as functions of fuel temperature, moderator density, and scram. The method [20) is outlined below.

A complete set of 1-D cross sections, 1-D kinetics parameters, the axial power distribution, and the kinetics parameters are generated from base states established for EOFPL, EOFPL-1000 mwd /St.

EOFPL-2000 mwd /St, and BOC exposure statepoints. These statepod-*_

are characterized by exposure and void history distributions, control i rod patterns, and core thermal-hydraulic conditions. The latter are consistent with the assumed system transient conditions provided in 4 Table 7.1.1.

The BOC base state is established by shuffling from the previously defined Current Cycle endpoint into the Reload Cycle loading pattern. A criticality search provides an estimste of the BOC critical. rod pattern. The EOFPL and intermediate core exposure and void history distributions are calculated with a Haling depletion as described in Section 5.2. The EOFPL state is unrodded. As such, it is defined sufficiently. However, the EC'?L-1000 mwd /St and EOFPL-2000 mwd /St exposure statepoints ~ require base control rod patterns. These are developed to be as " black and white" as possible,

, That is, beginning with the rodded depletion configuration, all control rods which are more than half inserted are fully inserted, and all control rods which are less than half inserted are fully withdrawn. If the SIMULATE-3 calculated parameters are within operating limits, then this configuration becomes the base case. If the limits are exceeded, a minimum number of control rods are adjusted a minimum number of notches until the parameters fall within limits.

Using this method, the control rod patterns and resultant power distributions minimize the scram reactivity and maximize the core 31-K?P40/10 l

ovorage moderator density reactivity coefficient. For the events analyzed, this tends to maximize the transient power response.

At each exposure statepoint, a SIMULATE-3 initial control state reference case is run. A series of perturbatior cases are run with SIMULATE-3 to independently vary the fuel temperature, moderator temperature, and core pressure. All other variables normally ass'ciated with the SIMULATE-3 cross sections are held constant at the reference state. To obtain the effect of the control rod scram, another SIMULATE-3 reference case is run with all-rods-in. The perturbation casec described above are run again from this reference case. For each control state, a data set of kinetics parameters and cross sections is generated as a function of the perturbed variable.

There is a table set for each of the 27 neutronic regions, 25 regions to represent the active core and one region each for the bottom and top reflectors.

7.1.4 Pressurization Transient s Analvred Past licensing analysis has shown that the transients which result in the minimum core thermal margins are:

1. Generator lond rejection with complete failure of the 1 turbine bypass s.atem.
2. Turbine trip with complete failure of the turbine bypass system.
3. Loss of feedwater heating.

The "feedwater controller failure" (maximum demand) trar.sient is not a limiting transient for Vermont Yankee, because of the plant's 110% steam flow bypass system. Past analyses have shown this transient to be considerably less limiting than any of the above for all exposure points. The events reported herein are limiting; no other transients would produce more restrictive MCPR operating limits for the 2eload Cycle. Brief descriptions and the results of the transients analyzed are provided in the following section.

WPP40/10 l

l

-7.2 Eressurizafion Trangient Analysis Results The transients selected for consideration were analyzed at exposure points of EOPPL, EOFPL-1000 mwd /St, and EOFPL-2000 mwd /St 4 with the exception of the loss of feedwater heating transient which l was evaluated at EOPPL-1000-mwd /St, EOPPL-2300 mwd /St, and B00. The transient.results reported in Table 7.2.1 correspond to the limiting bundle type in the core. The MCPR limits in Table 7.2.1 are calculated by ddding the calculated ACPR to the FCISL. The worst ACPR for the pressurization-transients include the 0.01 adjustment to allow for the exposure window of 1600 mwd /St on Current Cycle and the exposure uncertainty on the Reload Cycle.

7.2.1 Turbine Trio Without Bvoass Transient (TTWOBP)

The transient is initiated-by a rapid closure (0.1 second closing time) of the turbine stop valves. It is assumed that the steam bypass valves, which normally open to relieve pressure, remain clcsed. A reactor protection system signal is generated by the turbine stop valve closure switches. Control rod drive motion is conservatively assumed to occur: 0.27 seconds after the start of turbine stop valve motion. The ATWS recirculation pump trip is assumed to occur at a setpoint of 1150 psig dome pressure. A pump trip time delay of 1.0 second is assumed to account for logic delay and M-G set generator field collapse. In simulating the transient, the bypass piping volume up to the valve chest is lumped into the_ control volume upstream of the turbine stop valves. Fredictions of the salient system parameters i at the three exposure points are shown in Figures 7.2.1 through 7.2.3 l-for the " Measured" scram time analysis.

7.2.2 Generator Load Reiection Without Evoass Transient (GLRWOBP)

The transient is inf tiated by a rapid closure . (0.3 seconds L closing-time) of the turbine control valves. As in the case of the l turbine trip transient, the bypass valves are assumed to fail. A reactor protection system signaltis generated by the hydraulic fluid pressure switches in the acceleration relay of the turbine control system. Control rod drive motion is conservatively assumed to occur i

I

'~

WPP40/10 t-i 1

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

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

0.28 seconds after the start of turbine control valve motion. The same modeling regardit.g.the ATWS pump trip and bypass piping is used as'in the turbine trip simulation. The influence of the accelerating main turbine generator on the recirculation system is simulated by specifying the main turbine generator electrical frequency as a function of time for the M-G set drive motors. The main turbine generator frequency curve is' based on a 100% power plant startup test and is considered representative for the simulation. The system model predictions for the three exposure points are shown in Figures 7.2,4 through 7.2.6 for the " Measured" scram time analysis.

7.2.3 Loss of Feedwater Heatino Transient (LOFWH)

A feedwat'er heater can be lost in such a way that the steam extraction line to the heater is shut off or the feedwater flow bypasses one of the heaters. In either case, the reactor will receive cooler feedwater, which will produce an increase in the core inlet subcooling, resulting in a reacto. power increase.

The response of the system due to the loss of 100'r of the feedwater heating capability was analyzed. This represents the

-maximum expected feedwater temperature reduction for a single heater or group of heaters that can be tripped or bypassed by a single event, j- Vermont Yankee has a scram setpoint of 120% of rated power as part of the Reactor Protection System (RPS) on high neutron flux. In this analysis, no credit was taken for scran on high neutron flux, thereby allowing the reactor power to reach its peak without scram.

This approach was selected to provide a bounding and conservative analysis for events initiated from-any power level.

- The transient response of the system was evaluated at several exposures during the cycle. The transient evaluation at BOC was found to be the limiting case between BOC to EOFPL-1000 mwd /St. The results l of the system response to a loss of 100*F feedwater heating capability evaluated at BOC as predicted by the RETRAN code are presented in Figure 7.2.7.

1

$&P40/10 i-i i

I I

7.3 overoressurization Analysis Result s Compliance with ASME vessel code limits is demonstrated by an analysis of the Main Steam Isolation Valves (MSIV) closing with failure of the MSIV position switch scram. EOFPL conditions were analyzed. The system model used is the same as that used for the l

transient analysis (Section */ .1.1) . The initial conditions and modeling assumptions discussed in Section 7.1.2 are applicable to this simulation.

The transient is initiated by a simultaneous closure of all MSIVs. A 3.0 second closing time, which is the Technical Specification minimum, is assumed. A reactor scram signal is generated on APRM high flux. Control rod drive motion is conservatively assumed to occur 0.28 seconds after reaching the hign flux setpoint. The system response is shown in Figure 7.3.1 for the

" Measured" scram time analysis.

The maximum pressures at the bottom of the reactor vessel calculated for the " Measured" scram time analysis and for the "67B" scram time analysis are given in Table 7.3.1. These results are l

within the ASME code overpressure design limit which is 110% of the vessel design pressure. Vermont Yankee's design pressure is 1250 psig so the maximum pressure limit is 1375 psig.

7.4 Local Rod Withdrawal Error Transient Results The rod withdrawal error (RWE) is a local core transient caused by an operator erroneously withdrawing a control rod in the continuous withdrawal mode. If the core is operating at its operating limits for MCPR and LHGR.at the time of the error, then withdrawal of a control rod could increase both local and core power levels with the potential

,- for overheating the fuel.

There is a broad spectrum of core conditions and control rod patterns which could be present at the time of such an error. For most normal situations it would be possible to fully withdraw a control rod without exceeding 1% clad plastic strain or violating the WPP40/10 l

i

FCZSL.

To bound the most severe of postulated rod withdrawal error events, a portion of the core MCPR operating limit envelope is specifically defined such that the cladding limits are not violated.

The consequences of the error depend on the local power increase, the initial MCPR of the neighboring locations and the ability of the Rod Block Monitor (RBM) System to stop the withdrawing rod before MCPR reaches the FCISL.

The most severe transient postulated begins with the core operating according to normal procedures and within normal operating limits. The operator makes a procedural error and attempts to fully withdraw the maxinum worth control rod at maximum withdrawal speed. -

The core limiting locations are close to the error rod. They experience the spatial power shape transient as well as the overall core power increase.

The core conditions and control rod pattern are conservatively modeled for the bounding case by specifying the following set of concurrent worst case assumptions:

1. The rod should have high reactivity worth. This is provided for by analysis of the core at several exposure points around the core peak reactivity. The test patterns are developed with xenon-free conditions. The xenon-free condition and the additional control rod inventory needed to maintain criticality exaggerates the worth of the withdrawn control rod when compared to normal operation with normal xenon levels.
2. The core is initially at 104.5% power and 100% flow.
3. The core power distribution is adjusted with the available control rods to place the locations within the four by four array of bundles around the error rod as close to the operating limits as possible.

WPP40/10

l l

4. Of the many patterns tested, the pattern with the highest ~

ACPR results is selected as the bounding case.

The RBM System's ability to terminate the bounding case is

-evaluated on-the following bases:

1. Technical Specifications allow each of the separate RBM channels to remain operable if at least half of the Local Power Range Monitor (LPRM) inputs at every level are operable. For the interior RBM channels tested in this analysis, there are a maximum of four LPRM inputs per level.

One RBM channel averages the inputs from the A and C levels; the other channel averages the inputs from the B and D levels. Considering the inputs for a single channel, there are eleven failure combinations of none, one and two failed LPRM strings. The RBM channel responses are evaluated separatoly at these eleven input failure conditions. Then,

-for each channel taken separately, the lowest response as a function of error rod position is chosen for comparison to the RBM setpoint.

2. The event is analyzed separately in each of the four quadrants of the core due to the differing LPRM string physical locations relative to the error rod.

l 4

Technical Specifications require that both RBM channels be operable during. normal operation. Thus, the first channel calculated to intercept the RBM setpoint'is assumed to stop the rod. To allow for control system delay times, the rod is assumed.to move two inches

. after the intercept and stop at the following notch.

- The analysis is performed using SIMULATE-3.- Two separate cases are presented from numerous erplicit SIMULATZ-3 analyses. The reactor conditions and case descriptions are shown in Figures 7.4.1 and 7.4.2.

Case 1 analyzes the bounding event with zero xenon at the most

reactive point in the cycle'for the worst case abnormal rod pattern configuration. Case 2-is the worst of the 104.5% power conJitions modeled with more normal control rod patterns and equilibrium xenon.

- 3 ", -

WPP40/10

The transient results, the ACPR and maximum Jinear heat generation rate (MLHGR) values, are also shown in Figures 7.4.1 and 7.4.2. Tho ACPR values are eva3uated such that the implied MCPR operating limit

- equals PCISL + ACPR. This is done by conserving the figure of merit (ACPR/ICPR) shown by the SIMULATE calculations. The use of this method provides valid ACPR values in the analysis of normal operating states where locations near the assumed error rod are not initially

.near the MCPR operating limit.

Case 2 is the worst of all the rod withdrawal transients analyzed from 104.5% power, full flaw and normal rod pattern conditions. Case 2 is bounded by case 1 by at leact 0.02 ACPR margin to assure that the '

exposure uncertainties on the Current Cycle and the Reload Cycle are accounted for.

The Case 1 RBM channel responses are shown in Figures 7.4.3 and 7.4.4. They also show the control rod position _at the point where the weakest RBM channel response first intercepts the RBM setpoint. For

-this same bounding case, the operating limit ACPR envelope component versus RBM sotpoint is taken from Figure 7.4.1. The same figure shows tne resultant-LHGR assuming the limiting bundle is placed on the operating limit of 14.4 kW/ft prior to the withdrawal. The limiting bundle MLHGR demonstrates margin to the 1% plastic strain limit given the low exposure af the limiting bundle. High exposure bundles which have low 1% plastic strain limits are never limiting.

7.5 Misloaded Bundle Error Analysis Result s 7.5.1 Rotated Bundle Error The primary result of a bundle rotation is a large increase in local pin peaking and R-factor as higher enrichment pins are p3 aced adjacent to the surrounding wide water gaps. In addition, there may be a small increase in reactivity, depending on the exposure and void fraction states. The R-factor increase results in a CPR reduction, while the local pin peaking factor increase results in a higher pin LHGR. The objective of the analysis is to ensure that, jn the worst possible rotation, the LHGR and CPR safety limits are not violated

- WPP40/10 l

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

with the most limiting monitored bundles on their operating-limits.

To analyze the CPR response, rotated bundle R-factors as a function of exposure are developed by adding'the largest possible AR-factor resulting from a rotation to the exposure dependent R-factors of the properly oriented bundles. Using these rotated bundle R-factors, the MCPR values resulting fror a bundle rotation are determined using SIMULATE. This is done for each control red sequence -

throughout the cycle. The process is repeated with the K-infinity of the_ limiting bundle modified slightly to account for the increase in reactivity resulting from the rotation. For each sequence, the MCPR for-the properly oriented bundles is adjusted by a ratio necessary to '

place the corresponding _ rotated CPR on its TCISL. The maximum of these adjusted MCPRs is the' rote.ted bundle operating limit.

i To determine the MLHGR resulting from a rotation, the ratios of the maximum rotated bundle local peaking factor to the maximum properly oriented bundle local peaking are determined for the expected range of exposure and void conditions. The maximum of this ratio is applied to the LHGR operating limit of 14.4 kW/ft. This maximum rotated bundle LHGR is,-in hddition, modified to account for the poonible reactivity increase resulting from the rotation.

The results of the rotated bundle analysis are given in Table 7.5.1. Comparing Table 7.5.1 to Table 4.2.2, there is sufficient margin to the 1% plastic strain limit.

7.5.2 Mislocated Bundle Error

, Misloading a high reactivity assembly into a region of high neutron importance results in a location of high relative assembly averaga power,- Since the assembly-is-assumed to be properly oriented t (not rotated), R-factors used for the misloaded bundle are the standard values for the fuel type.

The analysis uses multiple SIMULATE-3 cases to exmnine the  !

I effects of explicitly mislocating every older interior assembly in a quarter core with a-fresh or once-burned assembly. Because of WPP40/10 l ,

L:

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

symmetry, the gesults apply to the whole core. Edge bundles are not examined because they are never limiting, due to neutron Icakage.-

1 The effect of the successive mislocations is examined for every )

-control rod sequence throughout the cycle. For each sequence, the i MCPR for the properly loaded core is compared to the MCPR of the misloaded core at the misloaded location. The MCPR for-the properly loaded core is adjusted by a ratio necessary to place the mislocated

~ assembly on the FC1'SL, The maximum of these adjusted MCPRs is the m'.slocated bundle operating limit. +

The results of the mislocated bundle analyais are given in Table 7.5.2. Comparing Table 7.5.2 to Table 4.2.2, there is sufficient-margin to'the 1% plastic strain limit.

1 l.

!~

l-l WPP40/10 l:

t

TABLE 7.1.1 VY Cvele 16 Summarv of System Transient Model  !

Initial Conditions for Transient Analyses I

Core Thermal Power ' (MWa) 1664.0 Turbine Stemn Flow (10'11&/hr) 6.75 Total Core Flow (10'11N/hr) 48.0  !

Core Bypass Flow (10'Itg/hr)

  • 6.2 Core Inlet Enthalpy (BTU /lb,) 523.2 Steam Dome Pressure (psia) 1034.7

. Turbine Inlet Pressure (psia) 985.7 Total Recirculation Drive-Flow (10'lb./hr ) 23.7 Core Plate Differential Pressure (psi) 20.2 Kurrow Range Water Level (in.) 162 Average ruel Gap Conductance (See Sect.)n 4.2)

I l

I Includes water roc flow.

kmP40/10 I

k i

i

i l

TABLE '/.2.1 l i

l VY Cvele 16 Pressurization Transient Analysis Resulty l

Peak Prompt Peak Average Power Heat Flux Exposure (Fraction of (Fraction of Operating Transient St atenoint, Init ial Value) Initial Value) ACPR' MOPR Limits Turbine Trip 'EOFPL 2.954- 1.213 0.14 1.21 without Bypass, EOFPL-1000 1.872 1,086 0.06 1.13 "Heasured" Scram '

Time EOTPL-2000 1.211 1.000 0.02 1.99 Turbine Trip EOTPL 3.301 1.259 0.17 1.24 Without Bypass, EOFPL-1000 2.2e5 1.144 0.10 1.17 "67B"~ Scram Time EOFPL-2000 1.605 1.049 0.05 1.12 Generator Load EOTPL 2.868 1.192 0.12 1.19 Rejection goppt.1000 1.990 1.069 0.05 1.12 Without Bypass,

" Measured" Scram EOFPL-2000 1.086 1.000 0.01 1.08 Time Generator Load EOFPL' 3.484 1.257 0.15 1.?2 Rejection EOFPL-1000- 2.584 1.141 0.08 1.15

~

Without Bypass, "67B" Scrum Time EOFPL-2000 1.612 1.024 0.03 1.10 Loss of-100*F -EOFPL-1000 1.145 1.146 0.11 1.18 Feedwater EOFPL-2000 1.144 1.145 0.11 1.18 Heating BOC 1.147 1.148 0.12 1.19 The worst ACPR for TTWOBP and GLRWOBP -includes a 0.01 ACPR adjustment.to allow for the exposure window of 1600 mwd /St on

-Current Cycle and the exposure uncertainty on the Reload Cycle.

WPP40/10 l

I M E 7.3.1 VY Cvele 16 Overpressurization Analysis Result s I

i Maximum Pressure at Reactor

,O'- Conditions vessel Bottom (esial-

" Measured" Geram Time 1253 "67B" Scram Time 1280 TABLE 7.5.1 VY Cvele 16 Rotated Bundle Analysis Results Operatino MCPR Limit Maximum Attainable LHGR (kW/ft) 1.21 19.86 4

TABLE 7,5,2 VY Cyclg J ,Mi'1004sted Bundle Analysis Resulte Operatino MCPR Limit Maximum Attainable LHGR ~ (kW/ft) 1.17 18.09 WPP40/10- .

l

Part I j j- Part II l 'FIBWR andl l

- l:

RETRAN/ Estimate Initial

.' S t te Transient Analysis l l l- l-

! .Funct1onal Rel ati ons hi ps RETRAN/TCPYA01 l

r AP - f(ICPR)

..-AF = f(AP).

[l' -

Transient Hot

. Chantiel Analysis l 8PF - f(AP) l l

t______': _ _ _ _ _ _ _'

Estimate New ICPR1 as:

ICPR 1 - 1.07 ACPRo ICPRo No Transient FCISL7 Yes STOP FIGURE 7.1.1 Flow Chart ~for-the Calculation of ACPR Usina the RETRAN/TCPYA01 Codes hTP40/10'

. _ . . . . . _ . . ~ ,

--! i

.51, _

1 l

' ' ' ' ' ' ' ' ' ' f 150-

^

8.0-

?

~

- 5.0-i 1.25-7- kl as

,' 's h.4.0- <

4 ,

3' > 4 y e 5, 2.

4 l '

o. E ,
l. . -

E ',

  • g.co ,

f:

_3.0-

' ./

a b - 's-

a. : 2 ' -

O * *

'2-g- p g O. g g. O \

S 4 g z

.. Z k 's s 0.75- 's, -

^

- CORE INLET FLOW 's,

- i. o .

........ AVG. HEAT FLUX - .

i t-f..

-0.0- , . . . O'50-0G 0.5  : 1.0 1.5 2.0 2'.5 3.0 O.0 0.5 i.0 i.5 I.0 d.5 3.0

TIME ( SEC ) 11ME ( SEC )

i-l:-

l l

l l

-FIGURE 7,2,1

. Turbine Trio Without Bvoass. EOPPL16

[:

' Transient Response'Versus Time, " Measured" Scram Time 1

WPP40/10

e l i

4 1500.0 . 2000.0-

. ........a.. + .. , .

i

} lk 1500,0- I  !. j! ..

l  ! l!.  !\

1400.0- j! ,,, 11 jk ,,

1,  :

.fg 1

es-1000.0'  ! .n.

it I !'.':1  !

3 5..

i ['

" l .f 3 .] 1, I

  • . {d I (( k j,' $!

y l 1300.0 Q $00.0- l , jt {f

.. 1 3 $

f 5 .

0.0- -

W '

3 h 1,

{-

O-t1200.0-o -500.0- *;

l _ __

~

-1000.0- .

i 110 0.0-3500.0 S/R V ALVE (NEG.) . .

.. . . .... FEED W ATER

................ VESSEL STEAM OUTLET

(' 1000.0-0.0 1.0 -

2.0 3.0  ;

4.0 5.0 6.0

l. -2000.0 0.0 1'.0 2'.0 3'.0 4'.0 5'.0 - 5.0 i.

wt ( ste ) Tiut ( ste )

FIGURE-7.2.1 (Continued)

Turbine Trip Without Bypass, EOPPL16 Transient-Resoonse Versus Time, " Measured" Scram Time g

~

WPP40/10 l

l -.

- t , 6-~, ., . . - .r- , ,,o., - . . _ < , ,w - m- < . , -

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

ce  ;

l i

9

' ' ' ^ ' '

2.0- ,

- 1. 8 -

1.6 -

1.4 -

l 1.2 -

, 1.0 -

i -- * '

m 0.8-m.

g 0.0-0.4-h t

80.2- '....

0.0 ..,....

i ...

-o.2- s, '-

. s

-o.4 - ,

-0.6 -

E

-0.0 t,

- 1.0 - j

-1.2 -'

g

~'*'

. . . . . . . . - SCilAM

-1.e- ' . . . . . . . . . 00f%'LER

- 1. e . -. MOD $RATOR i

-2.0 , , , , ,

0,0 0.5 1.0 1.5 - 2.0 2.5 J.0

' TIME ( SEC )

I I '_

FIGURE 7.7.1 (Continued)

Turbine Trio Without Bvoass, EOFPL16 j' Transient Response Versus Time, " Measured" S. im Time WPP40/10' l

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

^' ' ^ ' ' ' ^ ^ ' ' ' ' ' '

4.0 1.50 5.0 - -

1J8- .

td

- 4.0 - ,

2- .'

Z ,#

Z

' s

- 1.00- , , s .

h. s I-3.0-i ,

a r s

' ~' , 's z

o

- .g -

E 8 t,,'s -

N s r

@ s.0- -

s 0.75- \s .

. ,,e . CORE INLET F10W \

........ AVG. HEAT Ft.UX s,'s, .,

I l

1 l

! 0.0 , , , . .

. 0.00 , , , , ,

0.0 0.5 . 1.0 13 2.0 2.5 3.0 0.0 0.5 i0 1.5 2.0 2.5 3.0 TIME ( SEC ) TIME ( SEC )

1:

i-FIGURE 7.2.2 Turbine Trio Without Bvoass, EOPPL16-1000 mwd /St l

I- Transient Response Versus Time, " Measured" Scram Time i

I WPP40/10 i-

, . .. . . . . . - . . . . _ . _ . . .. .. ~ _. . -. . . . . - - _ - - . _ . - . ..-

' ^'^ ^' ' ' ^^ ^' ^ ' ' ' ^^

1900.0 2000.0

]********.'.....,,

r 1600.0-E ..

f. '

E" 1400.0--

I

' ^ 1000.0- h

?

n O t 6,/

1F v a f 4 800.0- ,j 1300.U. .Q

= 1 i 9 0.0- --

- l Irpfl.0- d i}

b -800.0- -

y

!,[l 8 8 i

~~

I'5 1000,0- -

l 1101.o 0

)

gsee.g. S/R VALVE (NEG.) .

.. ..... FEED WATER e

................ VESSEL STEAM OttiLET 1000.04 - ,- -2000.0-0.0 1.0 s.O J.0 4.0 s.O e.0 n.o i.e i.o ~ i.e ~ i,o .ie e.g 11ME ( SEC ) TIME ( SEC )

t

! FIGURE 7.2.2 (Continued)

, Ai T tine Trio Without Bvoass, EOTPL16-1000 mwd /84 l Iranf: tent Pesponse Versus Time, " Measured" Scram Titne

~"~

meno 4,p

" ..-_,.m.-. , _ _ _ _ . , . _ , .,_ __

-..c I

2.0 a f

~

t. e -

t.s .

t.4 -

tJ.

f 1.0

/

m 0. s -'

O.e- .

o.4 -

l ,

8 c.2- l

"^

-0.3 q".............................."""'"',,,...-

ha -o.4- t s

~o.5 \

-o.s- s,

~ t.0 - 's,

  • t.2 - 's '

iTOTA

.~.... . ' CRAM

-t' ................ Dp PPLER

-t s. .-

MQDERATOR

-3.0- .

0.0 0.5 1.0 1.s o 3.s 3.o llMC ( SEC )

FIGUPE '7.2.2 (Continued)

Turbine Trio Without Evnass. EOTPL16-1000 mwd /St Transdent Response Versus Time, " Measured" Scram Time WPP40/10 r

l 6.0' l.50-5.0 825-ta h 4.0

< r .

~Y '

o 'd 3o 6 i oo. -,

N5 u. s t2

  • s r,i z  :. 's.,',' ',

s' o 's m @ ',

y 20 g '.,.

E 's, on 's,

,o _ CORE lilLET [f 0,W

_..\ ........ AVG. IIEAT FLUXN,,

's

..,m 0.0- , , , , i 040 , , , , ,

00 03 10 1$ 20 2,$ 3.0 00 0$ l0 I$ 20 2.6 30 llME ( SEC ) 11ME ( SEC )

PIGURE 7,7,3 2,urpine Trir Without Bypaso. E0rPL-2000 mwd /St Transient Response Versus Time, " Measured" Scram Time WPP40/10

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

t l

1500.0-

  • a e , , [

2000 0- j r

-}-aa........,,,..

j j,j '***

1500 0 il

$  !! A 1400.0 '

Y l\ $\

n lsis il 6

,,b \ l1 11 1000.0- /.

i af l I '

!. .[ hi. .,J '.

^  :

t d~- O N I k,lI h# $00.0- jt t 1300 0 Q s

l.
  • ' v ,.  :

8i m " .

p!

w fJ '-

2 ' 4 {g

-o 1100.0 ot _}

O -500.0 .$ E.

2

)

o . !j: '-

Y l,$  ?

h I

-1000.0  :.

1100 0  ! ,

~1500.o S/R V Al.VE (11E0.) - 6

-....... ITED WATER

. .. ... .. VESSEL STEAM OUTLET 1000.o. ~2000o. '

in M io. . i.o . i0 io ,'n 08 80 20 10 lo i.o e.o TIME ( SEC ) TIME ( SEC )

I FIGURE 7.2.3 (Continued)-

Turbine Trio Without Bypass, EOTPL-2000 MNd/St Transient Response Versus Time. "Me g red" Screm Time t i1PP40/10 t

, . . - - - - , -,- r-r,-,-.,vv- ,,-.,,,.,a,,-&,~..,,&,,---,-.-,,,,-,,--,-,,v.-rn,-~ , ,

i

)

y,o n 1.6 14 17

1. 0 n fi.6 u- o0

.) 0.4 8 or

/

oo .. ,

\"-

h -c.2

's, h -o 4-

,f, -o o ',

  • -0.0 '  !

t t

-10 ',

- 1'2 t

t- 101 L

~'d

.....'i. SCRA A

-16 . . .\ 00PPL'b

-te 'i,MODERA R 0.0 0.5 1.0 15 2.0 2.5 3.0 llME ( SEC )

1 FIGURE 7,2,3 (Continued)

Turbinri Trip Without Bypass, EDFPL-2000 mwd /St IJ atis t ent genonse versus Timn, " Measured" Scram Time

~ ~

WPP40/10

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

' ' ' ' ' j

' '^^- ' '

160-8.0-l 60-t:6 51 l . 4.0 < lg s

.) 8 i f t h'

  • i co- , , \ ,

3.0- .

k . , s

  • o s ,

e 't '~~.

m 9 ,

s (C 'g

O - g'g 5 m

} s, l s,$. '

0.75 CORC lillET FLOW 'N s .

I t .o . ,

... .... AVO. HEAT TLUX I

+

-i

. i 0.50 ,

Jo !

00-i6 2o 2.5 3.0 oo c.6 10 1.6 2.0 2.s 00 0.5 io TIME ( SEC ) IIMC ( SCC ) l l

r.

t 1

FIGURE 7.2.4 Generator Load Reiection E'ithout hvoses, EOFPLi6 l Transient Response Versus Time. " Measured" Scram Time r

~

WPP40/10-i

-- i.i.,-,..._._,_,.

4 a

' ' ' i ^^ ' ' ' ' ' '

C00 0- #000.0

!i 1500 0 .

ji ,

i 1 I% j\1 140 G.0 - I, I I '\ I1 l"a\

i

-m

.f. -

Q.

m O

1000 0- \

I j

l

{l

} I, 1

}

/

,)

l\ '

-l I

v u

y 600.0 f.'

f il j t g 11 4

< L

'1300 0' 2 #

g/

3

'l' v i

-k.

lI L  ;

g, 1100.0 h.

'O. ,

o g 600 0- )  ;

t .

4 -1000.0 +

110 0.0 i

.,3no o S/R VALVE (NEG.)

....... FEED WA1ER

................ VESSEL STEAM OUILET .

t

.1000.0 . , , , . = -

.~20000 , , ,. , ,  !

O,0 10 2.0 3.0 40 6O 30 0.0 1,0 2.0 3.0 4.0 6.0 8.0 - i TIME ( SEC ) llME ( SEC )

{

t

?

4 4

I

TTGURE 1.2.4 (Continued)  ;

Generator Load Reietrion Without Dvriass, EOPPL16 Transiunt }<esponse Versus Time, "Maasured" Scram Time ,

l.

t 1

' WPP40/30 i

4 t

i j.

~

,,w.,9 ~h-.,,..m-,.,,,,,3..,,-,..,yg._ ...,_,,w,,m.+,w_,,.u_w.,,,,_,m_,. , , . . . . ._,,,..,,,.-.mcw,.,r.,,,,,.,,,.--,y

l l

P i

^ ' ' ' ' ' '

2.0-16 16 .

1.4 -

1.2 - .

f

. 10-n o.e .

}@ oe h c4- .

80.2- ...

- u .K. . . . . ..

- -o,2- 1, '*

~04- s g 1 y =06 \ l l~ E .o.a. 's ,

s

-t o ',

i a

- 1.2 10% A / t

~' SCIMM'

. t.s - . , . . . . . . . . . . . . . DOPh;LER 3

-1e . MODipA10R l

-2.0 , , , , ,

0.0 0.5 1.0 16 2.0 2.5 3.0 TIME ( SCC )

t l

FIGURE 7.2,4' }

(Continued)  ?

Generator Load Reiection Without Bvnass, EOPPL16 Transient Response Versus Time. " lieasured" Scram Timo {

l^ -i WPP40/10 w- e w r vw - v->-,--c.a-,%.=-,3,---#w,.vr,wv,si-, ,r,vw,-,=---.ew'www ,,w,m y ww,~m-3,- n =,%eg- 7 sy,-y-.w,,.%.reywess-,.- m-.+,.-+o w p w -- w w . -v. r -e 9 -er--qp es n -e-w my% p - e r- g-

e i >

,3, so 60 128 (J

[5 40 B 4.

y "s R .3 z 5 ,,

o D ,

6 3.0

$ 100

's tr -% s o , s, g 's ,/ '

,,. f'.-

n: { N

@0 1

~.

0.7 b s N

CORE HJL.ET T1.0 t0 ,N

..\./ ....... AVG. HEAT flux -

00- ~ .

10 7.0

, 'T 25 3.0 o s0-0.0 0.6 40 tb 2.0 26 3.0 0.0 0$ 1F

.11ML ( SEC ) 11ME ( SEC )

FICURE 7.2.5 Generat or 1 oad Feies, tion Wit hout Dvness, E0FPL16-1000 mwd /St 7,Iansient FocooD.ge Versus Time. "Mensured" Scram Titne i

~ ~

I1W 40/10 1

1500.0 20000-a.

! il 1500 0 i 1400 0 [g D

lj ,

^ 5000 o f f "i t i

.I . lS 3 1'

m I ;j \! .s\

1 1 f O 3 h '

i .' i; ij v

1300.0  %

s S00.0 i[l' -

f k,7 I'

o J

" i 00- -4 h 1200.0 y j ,

O g -600.0 j ,; 1

. . . N

' O *t000 0 1100 0

.i$on.o S/R V ALVE (NEG.)

........ FEED WA1ER 4 ..... ....... VLSSEL STEAM OUILC1 1000 0' , , , -2000.0-do 8,(

i 00 to 20 30 4' 0 b,0 6.0 00 t' 0 2' O .$ .0 4'.0 TIME ( SCC ) TIMC ( SCC )

rictmo 7.2.s (Con *,inued)

Generator 1. cad Petection Without E vris s o , 20rPL16-1000 mwd /St Transdent Fesr>onse Versus T1me, " Measured" Serem Titne

~ ~

WFP40/10

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

y_n. n n .n > ,

1. -

/

l.4 -

'Nj 12- l 10 n 0.0- -

$ 06

-) 0.4 - .

. 0.2

0.0 . - .. , , , . ,

. -02

.o.e -

o

  • W

-0e i

-00 's

's t,

s TOTAL

' ........ SERAM

-ta .. . . . .. . DQPPLCR

-t a- . MODERATOR t

~2.0 , , , , ,

00 0.5 1.0 1.5 2.0 2.r, 3.0 TIME ( SEC )

PIGURE '7. 2. 5 (Continued)

Generator Load Reiection Without Bypaso, EOPPI16-1000 mwd /St 17Ansient Response Versus Time, " Measured" Scram Time WPP40/10

-l e,o >

t.50 -

i 40-128-id f2' 4.0-g .

4 2 T r

z O

- @ s.o k .00-1 I ,

d b '. s z z ,,

o s' s, i r: ,

te o o 2.0 <

z k .,

's' 0,75 .

i s, t

t.0- CORC INLET f LOW '


. . AVG. IfEAT TLUXN,. ,

i Od 0 50 , , , ,

0'.0 d,$ 3' O l'.3 2'.0 2',3 3.0 8 88 '0 ' 2.0 l.S 3.o TIME ( SEC ) ilME ('SSEC )

FIGURE 7.2.6 Generator to,ad Reiection Without Bypass,-EOFPL16-2000 mwd /St -

Transient Response Versus T$mo, " Measured" Scram Time

~60-WPP40/10 l-

.-, ,. . , . - . . . . .. . . - . - . , -+ , -.. - ,, , . . -

1500.0 20000- ^*'

i  :,

1500.0 1 !I i '. .

1400.0-  !

b ,

i l ..

si  ;  : .

m 10e0 0 1

.  : : r-5 m i i  ! ', t :f ':\.

v E o La

! I i :

l i::. 'l i!!l i ta Q 500.0 , 1 t! u et 1300.0 y 0 9 i

,  : i

!  : pl 0 -

E D~  ! h to F-  :

1200.0- (I a g 500.0- i .i .

s o 'f

$ d 3 \ -1000.0 110 0.0

_ ,39o o . S/R V ALVE (NEG.)

. ...... IEED WA1ER

, . . . . . . . . .. VESSEL STEAM OUILET 1000.0 . , , . . -2000.0- , , , , ,

00 1.0 2.0 3.0 4.0 5.0 6.0 00 1.0 20 3.0 4.0 5.0 6.0 TIME ( SEC ) llME ( SEC )

[J GUR E ~7 . 2 , 6 (Continued)

Generator Load Reiection Without Bypass, EOFPI,16-2000 mwd / t Transient Responue Versus Time, " Measured" Scram Timo WPP40/10

. .. i. . . .

,g.

18 16 14-12 1.0 n Os

'al 0 6 -

0.4-h .

O o 0.2 /

  • 1 ...

v on _. . /. . . - ...

& -0.2 N

o

-04 i gr) -O n - ',

U

-0.6 i,

-10 ',

-t2 '

- TOTA

-14

__....'i_ SCRAl. '

-t6 ,

. g DOPPLLR

-te , _ _ _ . i' MODERAT R i

-20 1 4 i i 1 00 05 10 1.5 2.0 2.5 3.0 TILW ( SEC )

FIGURE ~1.2.6 (Continuec.)

Generator Load Redection Withnyt Bvonso, EOTPL16-2000 mwd /St Transient Fesponse Versus ilma, " Measured" Scram Time WPP40/10

' i ' ' 2500 0 * ' ' ' '

u o-i 125 1250.0 to n 4 d m

> ...g,.;;>.,.,...----~~....._,.......,s3 a . = - -

E .,

.y a)

E ,00 .%  ; 2000 0 m

o Ia-f ;y tr G_ o 4 d E

0.75 1150 0 TEED WATER CORE ltlLET FLOW -. ...__ VESSEL STEAIA OUILET

........ AVG. IIEAT FLUX

... ... NORM. IJEUTROH POWER 0.50 , , , , , -f 1500 0 i , . , ,

0.0 2h 0 50.0 75 0 10 0.0 125 0 150.0 00 2h 0 $0.0 7b.0 100 0 12 5. 0 15 0,0 llME ( SEC ) TIME ( SEC )

FIGURE ~1.2.~1 Loss of 300'r 5'eedwater Heatino, BOC16 (Limitino Case)

Transient Response versus Tirne WPP40/10

---. --___.__m._____.___m__.__._ - __m_---_m_-_--_-_____m__ _ _ . _ _ _ _ __--__. .--- - __

l 1.0 100.0-O . 8 --

90 0 06-04 m 80 0 3

p.--~~-

~

0.2 - - - - - -

00- m ----------------

g g _ , _, .

y so o O Z Q - 0.4 - a v o 50 0

-0 6 -  ! g- -~,

g$-0.e-M 'O O o V O

a:

-t.0 J

30 0 k'$

O

- 1. 4 ' '

TOTAL

-1.6- -------- SCRAM DOPPLER 10 0

- 1. e

. MODERATOR

- 2.0- i . . - - 00' > > > > -

0.0 25.0 50 0 76.0 100.0 125.0 15 0.0 0.0 25.0 b0 0 76 0 100 0 126 0 t$0.0 ilME ( SEC ) 11ME ( SEC )

FIGURT 7.2.7 (Continued)

Loss of 100er Feedwat er Hnating, BOC16 (Limitino Caso)

Transient Response Versus Time WPP40/10

60-

^

150 - *-*'

S.O 125 bl 6 4.0 3 n y l, 's e ., 's z 'S ,

O b ,

i i

Ib 3.0 100 -4%" 's

% u.

O s

Id

's 7 s i 9 's a s 2.0 {y- \

It i, 0 7b s, s N go ./ CORE INLEI flQW \

........ AVG. llEAT flux, 0,0- , , , , i 0.50- , , i i .

00 10 2.0 3.0 4.0 S0 60 00 10 20 30 40 S0  : 8 .'0 TIME ( SEC ) lilK ( SEC )

f,JIEURE *1. 3 1 MSIV Closure, Flux Scram, EOPPL16 Transient Resonnt.e yerous Jime, " Measured" Scram Time WPP40/10

^* '

i 1500 0-20000-1500 0 ',

1400 0- '.

I I,  :: ' * ' ' ' ' . "

^ 1000 0 ', i *( -

1

^ I

  • O '

v t0 k

bl '

1300 0' .

L5 s i s .  :

$ 0 5 'a

  • s 00- N-l3 1200 0

-1,00.0 0 4 3 O

-1000 0-1100 0

-tL00.0 S/R V ALVE (llEG )

.. . . . .. . [ [ED W AT ER

, , , , , , , , VESSEL S1EAM OUILET 1000.0- , , , , i -2000.0- - i 4.0 S' O 8.

00 1.0 2.0 J0 4.0 SO 60 00 10 $0 3.0 TIME ( SEC ) lluE ( SEC )

FIGURr 7.3.1 (Continued)

MSIV Closure, Flux Scram, F.0FPI,16 Trancient Fesponse Versus Time, " Measured" Scram Timo,

- G i> -

WPP40/10

10 18 .

10 14 .'

12 to ,

.I m OB

@ 0.s .

S" \

O*1 v

co - --

h -0.7 \ ' . .

$ -os 't o ,

g -os i, '

E -0.0 5,

-t o i

  • hg

~'#

TOTAk

.......- SCRAb,!

-'8 .. . .. . DOPPLj R

-te M00ER(,10  :

', I

-2 o , , , ,

0.0 10 2.0 30 4.0 $.0 6.0 11ME ( SEC ).

Elc2URE 7,3,1 (Continued)

MSIV Closure, Flux Scram, TOTPL16 Tra n s {e_nt Response Versus Time, " Measured" Scram Time WPP40/10

41 39 l

6 5 6 I

3R 38 38 31 6 0 to 0 6 27 - 38 38 23 - 4 20 0 20 4 19 - 38 38 15 6 0 20 0 6 11 38 38 07 '6 4 6 _

03 02 06 'O'4 18 22 46 30 J4 48 o2 1664 MWt Initial MCPR - 1.243 Core Thermal Power =

- 48 Mlb/hr initial MLHGR - 14.40 kw/f1 Core Flow

- 1042 psia RWE Control Rod - 30 15 Core Exit Pressure Zero Xenon TRANSIENT

SUMMARY

RBM Rod MLHGR.

Sgtpoint Position ACFR ikwltt) 10 0.12 15.1 104 12 0.16 15.4 105 12 0.16 15.4 106 14 0.21 16.6 107 16 0.25 17.8 108 FIGURE 7,4,1 Reactor Initial Conditions and Transient Sumary for the VY Cvelo '6 Rod Withdrawal Error Case 1 WPP40/10

At I

n <2 42 35 31 24 12 12 24 n._

23 - 12 8 8 12 19 _

16 24 12 12 24 11 OL , gg gg 03 02 00 'O '4 18 22 26 J4 48 d2 Core Thermal Power = 1664 MWt initial MCPR - 1.530 Core Flow - 48 Mlb/hr initial MLHGR - 11.5 kw/ft Core Exit Pressure = 1042 psia RWE Control Rod - 26 23 Equilibrium Xenon TRANSIENT

SUMMARY

RBM Rod MLHGR.

Sotoolnt Position ACPR ikw/ft) 104 18 0.10 14.8 105 20 0.11 15.0 106 20 0.11 15.0 107 24 0.14 15.3 108 34 0.18 14.7 TIGURE 7.4.2 Peactor Initial Conditions and Transient Summary for the VY Cycle 10 Fod Withdrawal Error Case ?

WPP40/10

1 i

i i I t I i g,4$ ; . . . . .

~ '

(3 140 inttnament f alkves 1.35 - -

1.3- - -

1.25 - - -

0

" ~

^

1.2 -- ~N- -

1,15 - -

1.1- -

1.05 - [ '

1. All Intercepts are determined l>y the ELD Channel.

1.0 * - . c ~ r- i , , ,

0 8 10 24 32 40 48 ERROR ROD POSITIOf1

}'IGUBE 7. 4,3 VY CvyJg 16 Ri'J_ Case 1 - Setpoint Intercept s Determined My the A and C Channels WPP40/10

u a i i i i i i i a i i i =

I '4

n ilo Instrument Ialic -

i 1.35-l

  • 1.3 -

un tg _s

  • b 1.25 - -

ct 1.2 -

f to E i

s. Assiis g j$.

> " -4+ 4-1,1-

/ /'/

ses

/ ,,-

l. RIiti f.ettioint Intercept is marked

  • 1.05-

, . /_ ,,,, with (.),

2. Roil is stotiped at notch following two inches of Free Rod Hotion.

1.0 * - - - . .

i 0 8 16 24 32 40 48 ERilOfl(10D POSillOfJ FIGURE 7.4.4 5 VY Ovele 16 RWE Case 1 _ Set oqint Int e,,t :ent s Determined by the H and D Channels 11-WPP40/10

8.0 J]rSIG!! PASIS ACCIPrfiT ANALYSIS 8.1 Control Pod Dren Accident Pesult s The control rod sequences are a series of rod withdrawal and banked withdrawai instructions specifically designed to minimize the worths of individual control tods. The sequences are examined so that, in the event of the unco >pling and subsequent free fall of the rod, the incremental rod worth is acceptable. Incremental rod worth refers to the fact that rods beyond Group 2 are banked out of the core and can only fall the increment from full-in to the rod drive withdrawal position. Acceptable worth is one which produces a maximum fuel enthalpy less than 280 calories / gram.

Ovme out-of-sequence cor.ttol rods could accrue potentially high worths. However, the Rod Worth Minimiter (RWM) will prevent withdrawing an out-of-sequence rod, if accidentally selected. The RWM is functionally tested before each startup.

The sequence in the RWM will take the plant from All Rode In (ARI) to well above 20% core therral power. Above 20% power even multiple operator errors will not create a potential rod drop situation above 280 calories per gram [21),[22),[23). Below 20% power, however, the sequences must be examined f or increment al rod worth.

This is done throughout the cycle using the full core, xenon-free SIMULATE-3 model.

Doth the A and B sequences were examined at various exposures throughout the cycle. For startup, the rods are grouped, as shown in rigure. 8.1.1 and 8.1.2, and are pulled in numerical order. All the rods in one group are pulled out before the pulling of the next group begins. The rods in the first two groups are individually pulled from full-in to full-out. Beyond Group 2, the rods are banked out using procedures [2 4 ), [2 5) which reduce the rod incremental worths.

The potentially high worths that occur in the pulling of the Group 1 rods are ignored because the reactor is suberitical in Group

1. Therefore, if a rod drops from any configuration in the first WPP40/10

group, its excess reactivity contribution to the Rod Drop Accident (RDA) is zero. Successive reloads of exially zoned fuel have extended this suberiticality situation to the second group as well.

The second group of rods was examined using the analysis procedure [26). Relatively few control rod configurations were found to be critical. For conservatism, " critical" was defined as the SIM0 LATE-3 average cold critical K. , minus 1% AK (reactivity anomaly criteria). The few potentially critical configurations in Group 2 contributed less excess reactivity to the RDA than subsequent configurations ir Grcup 3.

Most pre-drop cases ir Group 3 are critical. Therefore, the ,

entire dropped rod worth cc ibutes toward the RDA excess reactivity insertion. The method used to evaluate Group 3 involved pulling Groups 1 and 2 out and banking Group 3 to varying positions. The types of cases examined included:

1. Bankad positions 04, 08, 12, and 48 (full-out).
2. Group 3 rods pulled out of sequence, creating high flux regions.
3. Xenon-free conditions, both cold moderator and " standby" (i.e., 1020 psia).
4. Group 3 rods dropping from 00 (full-in) to the appropriate banked position.
5. Stuck rods from previously pulled Group 1 or 2 dropping from 00 to 48.

The highest worth results, ptesented in Table 8.1.1, fit under the bour. ding analysis in References 21 through 23, 8.2 Loss-of-Coolant Accident Analysis The results of the complete evaluation of the loss-of-coolant v&P40/10

accident for Vermont Yankee!27) provide the required support for the operation of the Reload Cycle. The LOCA analysis performed in accordance with 10CrR50, Appendix K, demonstrates that the HAPLl!GR values comply with the ECCS limits specified in 10CTR50.40. The MAPLilGR limits for a.'.1 the fuel types in the Reload Cycle, as a function of average planar exposure, are provided in Appendix A. Only the limiting MAPLHGR limits f or the zoned fuel are provided in Appendix A. HAPLHGR lindts exist for each lattice type and are specified in the process computer.

8.3 Pefuelino Accident Rooults If any assenbly is damaged during refueling, then a fraction of the fission product inventory could be released to the enviroranent.

The source tenn f or the ref ueling accident is the maximum gap activity within any bundle. The source term includes contributions from both noble gases and lodines. The calculation of maximum gap activity is based on the MAPLl!GRs, the maximum operating fuel conserline temperatures, and maximum bundle burnup.

The fuel rod gap activity for the Reload Cycle is bounded by the values used in Section 14.9 of the PSAR(28).

WPP40/10

TABIE B.1.1 VhCvelo 16 Cont rol Rod Dron Analysis Popult e Maximum incremontal Rod Worth 0.61% 6K Calculated Cold, Xenon-Free Bounding Analysis Worth for 1.30% AK Enthalpy Less than 200 Calories per Gram (21),[22),[23)

WPP40/10

43 ' 3 39- 2 1 1 2 35 -

4 3 4 3 4 31 1 2 2 1 27 - 3 4 3 4 3 23 - 1 2 1 1 2 1 19 - 3 4 3 4 3 15 1 2 2 1 11 4 3 4 3 4 07 2 1 1 2 03 3 I I i 02 OG 10 14 18 22 26 30 34 38 42 FIGURE 8.1.1 First Four Rod Arrave Pulled in the A Secuences WPP40/10

43 3 3 39 - 2 1 2 35- 3 4 4 3 31 2 1 2 1 2 27 - 3 4 3 3 4 3 23-1 2 1 2l 1 19 - 3 4 3 3 + 3 15 - 2 1 2 1 2 11 - 3 4 4 3 07- 2 1 2 I

03 - - -

3 3 1 1 1 02 06 10 14 18 22 26 30 34 38 42 FI6dRP 8,1,2 First Four Rod Arrays Pulled in the B Secuences WPP40/10

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

1 i

' 9.0 CORE COMPONENT OUAL7FTCATION PROGRAM 9,1 Siemens Nuclear Power Fuel Assemblies Vermont Yankee has four ANF-f..-3.04B-EGE Qualification Fuel Assemblies (QFAs) [29) to qualify this bundle type for uso as a t potential reload bundle. The bundles were loaded into Cycle 15 and have been irradiated for one cycle. The ANF-IX Qi As were manuf actured by Siemens Nuclear Power Corporation (SNP), formerly known as Advanced Nuclear Fuels (ANF), and designed to match the GE BP8DWB311-10GZ bundles neutronically and thermal-hydraulically. However, they differ from the GE bundles in the following ways: 1) the average bundle enrichment is lower, at 3.04 w/o U-235; 2) the fuel pins are smaller in diameter and their numbers are higher, at 72; and 3) a large squaro l

inner water channel is used rather than a large round water rod. The l major mechanical desiga parameters are given in Table 9.1.1 and SNP's Generic Mechanical Design Report (30).

The bundles will be located at 05-22, 39-22, 05-24, and 39-24.

These locations are expected to be nonlimiting with respect to MCPR, MAPLHGR, and MLHGR for the entire cycle during steady-state operation.

The bundles will be monitored during the cycle to assure that they remain nonlimiting during steady-state operation. The use of the QFAs

-does not significantly affect the safety analysis described in Section 7.0[31). '" mecific calculations were also performed to show that the analyn.c - #ection 7.0 bounded the ANF-IX OFAs. Therefore, the ANF-IX CFA- cn .e monitored as a GE bundle with conservative adjustmen.1 . ne R-factor tables.

W7P40/10

TABLE 9.1.1 Nominal ANP-IX Fuel Mechanical'Desion Parameters Fuel Bundle * ,

-Bundle Types- ANF 9X9-IX Vendor Designation ANP-IX-3.04B-EGZ Initial Enrichment, w/o U,n 3.04 Rod Array 9X9 Fuel Aods per Bundle 72 Outer Fuel Channel r

Materia 3 Zr-2 Wall Thickness, inches 0.080 k

p t

Complete bundle, rod,. and pellet descriptions are found in References 29 and 30, WPP40/10 i

l

}-

80.0 S?ARTUP PROGRAM' Following refueling and prior to vessel reassembly, fuel assembly pesition and orientation will be verified and videotapea by underwater televi$1on.

.The Vermont Yankee Startup Program will include process computer u data checks,1 shutdown margin -demonstration, in-sequence critical measurement, rod scram tests, power distribution comparisons, TIP reproducibility, and TIP symmetry checks. The content of the Startup.

Test Report will be similar to that sent to the Office of Inspection  ;

and Enforcement in the pautf32).

I m

i, tai 40/10 e

~w

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

REFERENCES

1. C. Chiu, Ve rmont Yankee Cvele 14 Summary Peco rt , YAEC-1773, May 1991.
2. General Electric Standard Application for Reactor Fuel (GESTARII), NEDE-24011-P-A-10, GE Company Proprietary, February 1991, as amended. l
3. A. S DiGiovine, J. P. Gorski, and M. A. Tremblay, SIMULATE-3 Validation and Verification, YAEC-1659-A, September 1988.

4, R. A. Woehlke, et al., MICBURN-?/CASMO-3/ TABLES-3/ SIMULATE-3 Benchma rkino of Ve rmont Yankee Cveles 9 throuch 13, YAEn-1683-A, March 1969.  ;

5. Report, B. Y. Hubbard, et al., End-of-Full-Powtr-Life Sensitivity Study for the Revised BWR Licensino Methodoloov, YAEC-1822, October 19 91.
6. VYNPC Letter to USNRC, " Vermont Yankee LOCA Analysis Method: FROSSTEY Fuel Performance Code (FROSSTEY-2)," FVY 87-116, dated December 16, 1987.
7. Loss of coolant Aeeldent Analysis for Ve rmont Yankee Nuclear Power Station, NEDO-21697, May 1990, as amended. '
8. Appendix A to Operating License DPR-20 Technical Specifications and Bases for Vermont Yankee Nuclear Power Station, Docket No. 50-271.
9. A. S. DiGiovine, et al., CASMO-3 Validation, YAEC-1363-A, April 1988,
10. A. A. F. Ansari, Methods for the fnalysis of Boiline Water Reactors:

Steady-State Core Flow D2 st ribut ion Code (FIBwR), YAEC-1234, December 1980,

11. A. A. F. Ansari, R. R. Gay, and B. J. Gitnick, FIBWR: A Steady-State Core Flow Distribution Code for Boiline Water Reactors - Code Verification and Qualification Report, EPRI NP-1923, Project 1754-1 Final Report, July 1981.
12. USNRC Letter to J. B. Sinclair, SER, " Acceptance for Referencing in Licensing Actions for the Vermont Yankee Plant of Reports: YAEC-1232, YAEC-1238, YAEC-1299P, and YAEC-1234," NVY 82-157, Septenber 15, 1982,
13. General Electric Company, GEXL-Plus Correlation Application to BWR 2-6 Reactors GE6 throuch GE9 Fuel, NEDE-31598P, GE Company Proprietary, July 1989.
14. A. A. F. Ansari and J. T. Cronin, Methods for the Analysis of Boiline Water Reactors: A Systers Transient Analysis Model ( R ET RAN ) , YAEC-1233, April 1981.
15. USNRC Letter to R. L. Smith, SER, " Amendment No. 70 to Facility License No. DPR-28," dated November 27, 1981,
16. V. Chandola, M. P. Lerrancois, and J. D. Robichaud, Application of One-Dimensional Kihetics to Boilino Water peactor Transient Analysis Methods, YAEC-1693-A, Revision 1, November 1989.
17. Electric Power Research Institute, RETRAN - A Procram for _One-Dimensional Transient Thermal-Hydraulic Analysis of Complex Fluid Flow Systems, CCM-5, December 1976.

WPP40/10

= . .

t I

18. USNRC Lettor to T. W. Schnatz, SER, " Acceptance for Referencing of I Lic3nsing Topical Reports: EPRI CCM-5 and EPRI NP-1850-CCM," September 4, 1984.
19. A. A. F. Ansari, E, J. Burns, and D. K. Deller, Met hods f or t he AnalE.2.

of Boilina Water deactorst Transient Critical Power Ratio Analvshs

_(RETRAN-TCPYA01), YAEC-1299P, March 1982.

20. J. T. Cronin, Method for Generation of ono-Dimensional Kinetics Dat a for RETRAN-02, YAECa1694-A, June 1989.
21. C. J. Paone, et al., Rod Drep Accident Analysis for Laroe Boilina Water Resetors, NEDO-10527, March 1972 "
22. R. C. Stirn, et al., Rod Drop Accident Analysis for l aroe Boilino Water Beact ors Addendum No. 1, Multiple Enrichment Cores with Axial Gadolinium, NEDO-10527, Supplement 1,= July 1972.
23. R. C. Stirn, et al., Rod Drop Accident Analysis for Laroe Boiline Water Reactor Addendum No. 2 Exposed Cores, NEDO-10527, Supplement 2, January 1973.
24. C. J. Paine, Banked Position Withdrawal Scovence, NEDO-21231, January 1977,
25. D.-Radcliffe and R. E. Bates, " Reduced Notch Worth Procedure," SIL-316, Novenber 1479.
26. M, A. Sironen, Vermont Yankee Cyclo 14 Core Performance Analysis Report, YAEC-1706, October 1988.
27. General Electric Company, Loss-of-Coelart Accident Analysis for Vermont Yankee Nuclear Power Station, NEDO-21697, August 1977, as amendedt NEDE-21697, Supplement 1, November 1987; and NEDE-21697, Supplement 2, May 1990.
28. Ve rmont Yankee Nuclear Power Station Final Safety Analysis Repo rt ,

December 1991.

29, M. E. Garrett, K. D. Hartley, M. H. Smith, Verment Yankee 9X9-IX Oualification Fuel Assembly Desion Report, Meehanical, Thermal-Hydraulle, and Neut ronie Desion, ANF-90-034 (P), Revision 0, SNP Company Proprietary, March 1990.

30. Siemens Nuclear Power Company, Generic Mechanical Desions for Advanced Nuclear Fuels 9X9-IX and 9X9-9X BWR Reload Fuel, . ANF-8 9-014 (P) , SNP Company Proprietary,-May 1969.
31. M. E. Ga'tre t t , K. D. Hartley, and M. H. Smith, Ve rmont Yankee Qualification Fuel Assemb1v Safety Analysis Report, ANF-90-048, May 1990.
32. L. A. Trerrblay Letter to USNRC, " Cycle 15 Startup Test Report," BVY 91-10, January 23, 1991 WPP40/10

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

APPENP,1)LA CALCULATEp OPERATING MMITS The MCPR operating limits for the Reload Cycle are calculated by:

adding the calculated ACPR to the FCISL. This is done for each of the analyses in Section 7.0 at each oi the exposure statepoints.- For an exposure interval between statepoints, the highest MCPR limit at '

either end is assumed to apply to the whole interval.

Table A.1 provides the highest calculated MCPR 31mits for the Reload Cycle for each of the exposure intervals for the various scram speeds and for the various rod block lines. These MCPR operating limits are valid for operation of the Reload Cycle-at full power up to 9845 mwd /St'and for operation during coastdown beyond EOFPL.

l Tables A.2 through A.5 provide the maximum calculated MAPLHGR limits'for all the GE assembly types in the Reload Cycle.

l WPP40/10 l

l

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

TABLE A,1 Vermont Yankee Nuclear Power Station Ovele 16 MCPR Operatino Limit s Value of "N" in Average Control Cvele Exposure Ranoe MCPR Operating PBM Ecuation(1) Rod Scram Time Limit (2) . (3)

Equal to or 0 to 9845 mwd /St 1,32 42% better than L.C.O. 3.3.C.1.1 Equal to or 0 to 9845 mwd /St 1.32

'better than L.C.O. 3.3.C.1.2 Equal to or 0 to 9845 mwd /St 1.28 41% better than L.C.O. 3.3.C.1.1 Equal to or 0 to 9845 mwd /St 1.28 better than L.C.O. 3.3.C.1.2 Equal to or 0 to 9845 mwd /St 1.23 5 40% better than L.C.O. 3.3.C.1.1 Equal to or. O to 8241 mwd /St 1.23 L C.1.2 8241 to 9845 mwd /St 1.24 (1). The Rod Block Monitor (RBM) trip setpoints are determined by the equation shown in Table 3.2.5 of the Technical Specifications.

t (2) The current analysis for the MCPR operating limits does

, not include the 7X7, 8X8, 8X8R or P8X8R fuel types. On L this basis, .if any of these fuel types _ are to be reinserted, they will be evaluated in accordance with the 10CFR50.59 to ensure that the - above limits are bounding for these fuel types.

(3) MCPR operating limits are increased by 0.01 for the l single loop operation.

WPP40/10 L

l

3 4

l LMLElal -1 1

MAPLHGR Versas Avernae Planar Exposure for BD324B Plant: Vermont Yankee Fuel Type: BD324B Averaoe Planar Exoosure MAPLHGR Limits (kW/f t )

,QiWd/St) Two-Loop operation sinole-Loon operation

  • 200.0 11.22 9.31 3,000.00 11.83 9.81 8,000.00 12.69 10.53 10,000.00 12.80 10,62 15,000.00 12.74 10.57 20,000.00 12.05 10.00 l- 25,000.00- 11.39 9.45 35,000,00 10.12 8.39 s :-- :00.00 8.46 7.02 50,000.00 5.99 4.97 L

MAPLHGR limits for single-lcop operation are obtained by multiplying the two-loop operation MAPLHGR limits by 0.83.

WPP40/10 i

l

.:. . . . ~ . . , . - . ...- . ~ . . . . . _ _ - . -

f  ;

I H TABLE'A.3 1

MAPLHGR Versus Averaae Planar Exoosure for BD326B u Plant: Vermont Yankee Fuel Types -BD326B hyerace Planar Exoosure MAPLHGR Limit s (kW/ft)

(mwd _/ S t ) Two-Loon Operation Sinole-Loop operation' 200.0 11.26 9.34 3,000.00 11.72 9.72 8,000.00 12.76 10.59 10,000.00 12.90 10.70 15,000.00 12.82 10.64 20,000.00- 12.12 10.05 25,000400 11.44 9.49 35,000,00 10.15 8.42 45,000.00 8.63 7.16 50,000.00 6.17 5.12 s

MAPLHGR limits for single-loop operation are obtained by multiplying the two-loop operation MAPLHGR limits by 0.83.

FTP40/10 i-

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

1

~ TABLE'A.4 l l

MAPLHGR Versus Averace Planar Exposure for BP8DWB311-10GZ Plant: Vermont Yankee Fuel Type: BP80WB311-10GZ Averace P3anar Exposure MAPLHGR 4.mit s (kW/ f t )

(mwd /St) Two-Looo Operation Sinole-Loon Operation

  • 200.0 11.00 9.13 6,000.00 11.92 9.89 7,000.00 12.11 10.05 8,000.00 12.34 10.24 L 10,000.00 12.83 10.64 12,500.00 13.00 10.79 20,000.00 12.24 10.15 25,000.00 11.55 9.58 45,000.00 8.76 7.27 50,740.00 5. !1 4.90 9

l MAPLHGR- limits for single-loop operation are obtained by multiplying the two-loop operation MAPLHGR limits by 0.83.

! .WPP40/10 l'

I

1 TABLE A.5 MAPLHGR Versus Averaae Planar Exposure for BP8DNB311-11GZ_

Plant Vermont Yankee Fuel Type: 3P8DNB311-11GZ Averace Planar Exposure MAPLHGR Limit s (kW/ft) '

(mwd /St) Two-Looo operation Sinole-Loop operation *'

200.0 11.00 9.13 ,

6,000.00 11.92 9.89 7,000.00 12.11 10.05 8,000.00 12.34 10.24 10,000.00 12.83 10,64 12,500.00 12.90 10.70 15,000.00 12.81 10.63 35,000.00 10.24 8.49

-45,000.00 8.76 7.27

, 50,740.00 5.91. 4.90 MAPLHGR limits- for single-loop operation are obtained by multiplying the two-Joop operation MAPLHGR limits by 0.83.

D&P40/10