ML20056F834

From kanterella
Jump to navigation Jump to search
Cycle 14 Core Performance Analysis
ML20056F834
Person / Time
Site: Maine Yankee
Issue date: 04/30/1993
From: Paul Bergeron, Rousseau K, Michael Scott
YANKEE ATOMIC ELECTRIC CO.
To:
Shared Package
ML20056F828 List:
References
YAEC-1864, NUDOCS 9308310184
Download: ML20056F834 (152)


Text

{{#Wiki_filter:m. m m x ,p + ~ .. ~,

y

^ i >

5 5'.
',J ;

__v' >;n_. e ___. ' - 'jy s ); 4 JYANKEE ATOMIC ELECTRIC.CDMPANY I + l'. 1 [q ?-:q -~ - I a

f..,

3 pf ^ .'j .] 1 3 s-f A f .1q t 7., 51 s a ~ ~, 9 n ANKEE ~ / w u- + .777 t f' ; K. ' < e

..j r e

') 4 -e ',, f.. (,' ' 4 g

.x'

[1.'. 'I J'_ !;;:e & &s' Y p;: + 9308310184 930825 P PDR ADOCK 05000309 ..e1 m, s-PDR j;.3, p 1 M,. vem es u

-p g

~ ' ' ' fi; ~

5 m N MAINE YANKEE CYCLE 14 CORE PERFORMANCE ANALYS5 April 1993 Major Contributors: Reactor Physics Group Transient Analysis Group G. M. Solan K. R. Rousseau M. E. Cote S. P. Crofton J. W. Keats V. M. Esquillo M. C. Menard S. Peterson K. B. Spinney R. W. Sterner l P. A. Theriault l S. Van Volkinburg l LOCA Analysis Group R. C. Harvey Q. A. Hague K. E. St. John L Schor R. P. Smith G. Swanbon Yankee Atomic Electric Company Nuclear Services Division 580 Main Street Bolton, Massachusetts 01740

DISCLAIMER OF RESPONSIBILITY This document was prepared by Yankee Atomic Electric Company (" Yankee"). 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 to any unauthori7ed 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 completeness of the material contained in this document. N .iii.

-0 MAINE YANKEE CYCLE 14 CORE PERFORMANCE ANALYSIS April 1993 Major Contributors: Reactor Physics Group Transient Analysis Grou2 G. M. Solan K. R. Rousseau M. E. Cote S. P. Crofton J. W. Keats V. M. Esquillo M. C. Menard S. Peterson K. B. Spinney R. W. Sterner P. A. Theriault S. Van Volkinburg LOCA Analysis Group R. C. Harvey Q. A. Haque K. E. St. John L Schor R. P. Smith G.Swanbon Yankee Atomic Electric Company Nuclear Services Division 580 Main Street Bolton, Massachusetts 01740 /

APPROVAL 9 Prepared by: < u su(( N d o-m ^~ f 3 ~K.'R. Fousseau, Senior Nuclear Engineer Date ~ Transient Analysis Group b-4fl5/93

p. M. S61an, Ilhncipal Engineer Date Reactor Physics Group 0

3h3 R. C. Harvey, Senior Ngear Engineer 'Dat6 LOCA Analysis Group Reviewed by: hd lA 4 //$k3 M. W. Scott, Nuclear $ngineering Coordinator bafe Nuclear En eering Department et + ~ r Go8>. /W2 ~s (d. DiStefano, Senior Kadiological Engineer / Dats Radiological Engineering Group f Approved by: C -- - /3!98 P. A. Bergeron, Marpfger Date Transient Analysis Trou-0 'l /3f93 a{ tactor Physics GroupQ. Cacciapo/ti, Manager ' Dale ll4 w, [s a. + '/l' \\l'A R. K. Sundaram, Manager ~ Date LOCA Analysis Group /1$ V' ~ t P. S. Liftlefield,' Manager Date Radiological Engineering Group b3 3 e

p. Slifer, pirector

' Dlte i Nuclear Engmeering Department .ii. I

DISCLAIMER OF RESPONSIBILITY j This document was prepared by Yankee Atomic Electric Company C' Yankee"). 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 to any unauthorized use, neither Yankee nor its l officers, directors, agents, or employees assume any obligation, responsibility, or liability or make any warranty or representation as to the accuracy or completeness of the material contained in this document. -iii-

ABSTRACT This report presents design and analysis results pertinent to the operation of Cycle 14 of the Maine Yankee Atomic Power Station. These include core fuel loading, fuel description, reactor power dist ributions, control rod worths, reactivity coefficients, the results of the safety analyses performed to justify plant operation, the startup test program and the Reactor Protective System (RPS) setpoints assumed in the safety analysis. The analysis results, in conjunction with the startup test results, the RPS setpoints and Technical Specifications, serve as the basis for ensuring safe operation of Maine Yankee dunng Cycle 14. I l l ej\\%

TABLE OF CONTENTS Page APPROVALS.......... . 11 DISCLAIMER OF RESPONSIBILITY................................... iii ABSTRACT................... ...................................iv TAB LE OF CONTENTS............................................... v LIST OF TABLES................................................ viii LIST OF FIGURES.................... .........................xi

1.0 INTRODUCTION

1 2.0 OPERATING HISTORY....... 4 2.1 Cycles 1 and 1A. 4 2.2 Cycle 2......... 4 2.3 Cycles 3 and 4... 4 2.4 Cycles 5 and 6. 5 2.5 Cycles 7 and 8... 5 2.6 Cycles 9 and 10 5 2.7 Cycles 11 and 12......... 6 2.8 Cycle 13.. 6 3.0 RELOAD CORE DESIGN... 9 3.1 General Description... 9 3.1.1 Core Fuel toading...... 9 3.1.2 Assembly Enrichment and Burnable Poison Loadings................ 9 3.1.3 Core Loading Pattern..................................... 10 3.1.4 Assembly Exposure History............................ 10 3.1.5 CEA Group Configuration.... .........................11 3.2 Fuel System Design... 12 3.2.1 Fuel Mechanical Design. 12 3.2.2 Fuel Thermal Analysis.................................... 13 3.2.3 Thermal-Hydraulic Design..... 14 4.0 PHYSICS ANALYSIS.... 30 4.1 Fuel Management.......... 30 4.2 Core Physics Characteristics...................................... 30 4.3 Power Distributions............. 30 4.4 CEA Group Reactivity Worths....... 31 -v-l -,,w.+-

TABLE OF CONTENTS (continued) fage 4.5 Doppler Reactivity Coefficients and Defects........................... 31 4.6 Moderator Reactivity Coefficients and Defects.......................... 32 4.7 Soluble Boron and Burnable Poison Reactivity Effects.................... 33 4.8 Kinetics Param eters............................................. 33 4.9 Sa fety-Related Characteristics...................................... 34 4.9.1 CEA Group Insertion Limits................................... 34 4.9.2 CEA Ejection Results........................................ 34 4.9.3 CEA Drop Results and Post-CEA Drop Restrictions................. 35 4.9.4 Available Scra m Reactivity.................................... 36 4.9.5 Shutdown Margin Requirements. 38 4.10 Pressure Vessel Fluence........................................... 39 4.11 Methodology and Methodology Revisions............................. 41 5.0 SAFETY ANALYSIS............................ 69 5.1 General.... 69 5.1.1 Initial Operating Conditions. 70 5.1.2 Core Power Distributions.. 72 5.13 Reactivity Coefficients....... 73 5.1.4 Shutdown CEA Characteristics................................. 74 l 5.1.5 Reactor Protective System Setpoints and Time Delays............... 75 5.1.6 Pla n t Modifica tions.......................................... 76 5.1.7 Summary of Significant Changes in Cycle 14 Reload Safety Analysis.... 76 5.2 Events Analysis Review........................................... 78 5.3 Anticipated Operational Occurrences for which the RPS Ensures No Violation of SAFDL's..................................... 79 5.3.1 Control Element Assembly Group Withdrawal..................... 80 5.3.2 Boron Dilution............................................. 81 5.3.2.1 Dilution During Refueling.............................. 81 5.3.2.2 Dilution During Cold, Transthermal and Hot Shutdown with RCS Filled................................ 82 5.3.2.3 Dilution Dunng Cold, Transthermal and Hot Shutdown with Drained RCS Conditions..................... 82 5.3.2.4 Dilution Dunng Hot Standby, Startup and Power Operation.... 83 5.3.2.5 Failure to Borate Prior to Cooldown 84 -Vi-

TABLE OF CONTENTS (continued) Pace 5.3.3 Excess Load Incident 84 5.3.4 Loss of Load Incident........................................ 85 5.3.5 Loss of Feedwa ter Inciden t.................................... 66 5.4 Anticipated Operational Occurrences which are Dependent on Initial Overpower Margin for Protection Against Violation of SAFDL's....... 87 5.4.1 Loss of Reactor Coolant Flow.......... 88 5.4.2 Full Length CEA Drop.. 88 5.5 Postulated Accidents..... 91 5.5.1 Steam Line Rupture.. 91 5.5.2 Steam Generator Tube Rupture.. 92 5.5.3 Seized RCP Rotor. 93 5.5.4 CEA Ejection. 94 5.5.5 Loss of Coolant 94 5.5.5.1 Introduction and Summary..... 94 5.5.5.2 Large Break LOCA Analysis............................ 95 5.5.5.2.1 Break Spectrum Analysis. 96 5.5.5.2.2 LOCA Limit Calculations...................... 96 5.5.5.3 Small Break LOCA Analysis...... 96 5.6 Methodology and Methodology Revisions. 97 6.0 STARTUP TEST PROGRAM........................................ 129 6.1 Low Power Physics Tests....... ..................... 129 6.2 Power Escalation Tests ........................ 130 6.3 Acceptance Criteria...... ........................130

7.0 CONCLUSION

.............................. 133

8.0 REFERENCES

............ 134 -vii-

l LIST OF TABLES Number Title Page ~ 1.1 Significant Changes for Cycle 14 3 2.1 Operating History Summary....................................... 7 2.2 Fuel Assembly Types by Cycle.............. 8 3.1 Cycle 14 Assembly Description 16 3.2 Cycle 14 Core Loading........................................... 17 3.3 Mechanical Design Features of Maine Yankee Cycle 14 Fuel.............. 18 3.4 Cycle 14 Ratio of Maximum Radial Relative Pin Powers - Maximum in Reinserted Fuel Types to Maximum in Core. 19 3.5 Cycle 14 Bounding LHGR Limits for FCM SAFDL and Ratios by Fuel Type. 20 3.6 Cycles 3,13 and 14 Thennal-Hydraulic Parameters at Full Power........ 21,22 4.1 Cycles 3,13 and 14 Nuclear Characteristics............... 42 4.2 Cycles 13 and 14 CEA Group Worths at HZP.. 43 4.3 Cycles 13 and 14 Core Average Doppler Defect. 44 4.4 Cycles 13 and 14 Core Average Doppler Coefficient..................... 45 4.5 Cycles 13 and 14 Moderator Temperature Coefficients 46 4.6 Cycles 13 and 14 Scrammed Moderator Defect with Worst Stuck CEA 47 1 4.7 Cycles 13 and 14 Kinetics Parameters 48 4.8 Cycles 13 and 14 CEA Ejection Results from Full Insertions............... 49 4.9 Cycles 13 and 14 CEA Drop Results at BOC.......................... 50 4.10 Cycles 13 and 14 CEA Drop Results at EOC......................... 51 4.11 Cycles 13 and 14 Dropped CEA with Power Level Restriction - M ost Limiting Peaking Cases..................................... 52 -Viii-

LIST OF TABLES (continued) Number Title Pace 4.12 Cycle 14 Available Scram Reactivity................................. 53 4.13 Cycle 14 Required Scram Reactivity... 54 4.14 Cycles 1 through 14 Relative Pressure Vessel Fluence Comparisons......... 55 4.15 Maine Yankee Physics Methodology Documentation.................... 56 5.1 Statistical DNBR Limits (SDL's) used in Safety Analysis......... 98 5.2 Maine Yankee Safety Parameter Assumptions .................... 99-102 5.3 Cycle 14 - Incidents Considered 103 5.4 Cycle 14 Safety Analysis - Summary of Results... 104-107 5.5 Cycle 14 Required Initial RCS Boron Concentrations to allow Margin to Criticality for Dilutions from Shutdown Conditions with the RCS Filled (15 minutes) and the RCS Drained (30 minutes)..... 108,109 5.6 Summary of Boron Dilution Incident Results for Cycle 14............... 110 5.7 Summary of Boron Requirements for Refueling Conditions for Cycle 14.... 111 5.8 Cycle 14 Nominal Scram Reactivity Worths Required to Prevent a Return to Power during a Steam Line Rupture Accident... 112 5.9 Cycle 14 CEA Ejection Accident Results............................ 113 5.10 Comparison of Thermal Margin for Limiting Cycle 14 Power Distributions to FSAR Design Power Distribution................................ 114 5.11 Reactor Protective System Trips Assumed in the Cycle 14 Safety Analysis... 115 5.12 Scram Reactivity Assumed in Cycle 14 Safety Analysis................. 116 5.13 Cycle 14 large Break LOCA Analysis Results........................ 117 5.14 Cycle 14 Small Break LOCA Analysis Results........................ 118 5.15 Maine Yankee Safety Analysis Methodology Documentation........... 119,120 -ix-

LIST OF TABLES (continued) Number Title Pace 6.1 Cycle 14 Startup Test Acceptance Criteria........................... 132 l l .x.

LIST OF FIGURES Number Title Pace 3.1 Cycle 14 Fuel Pin and Burnable Poison Shim Assembly Locations........ 23,24 3.2 Cycle 14 Assembly Loading Pattern................................. 25 3.3 Cycle 14 Calculated Assembly Exposures - BOC (0 hBVd/Mt)............. 26 3,4 Cycle 13 Burnup Distribution by Assembly - INCA vs. Predicted near 6,000 mwd /Mt Cycle Exposure.............................. 27 3.5 Cycle 14 CEA Group Locations.................. 28 3.6 Cycle 14 Bounding LHGR for Fuel Centerline Melt SAFDL versus Local B urnup........................................... 29 i 4.1 Cycle 14 Assembly Relative Power Densities HFP, ARO, BOC (500 mwd /Mt). 57 4.2 Cycle 14 Assembly Relative Power Densities HFP, ARO, MOC (6,000 mwd /Mt) 58 4.3 Cycle 14 Assembly Relative Power Densities HFP, ARO, EOC (15,000 mwd /Mt)... 59 4.4 Cycle 14 Assembly Relative Power Densities HFP, Group 51n, BOC (500 mwd /Mt) 60 4.5 Cycle 14 Assembly Relative Power Densities HFP, Group 51n, MOC (6,000 mwd /Mt).... 61 4.6 Cycle 14 Assembly Relative Power Densities HFP, Group 5 In, EOC (15,000 mwd /Mt) 62 4.7 Cycle 14 Allowable Unrodded Radial Peak versus Cycle Average Burnup.... 63 4.8 Cycle 14 Moderator Temperature Coefficient Upper Limits versus Power Level........ ........, 64 4.9 Cycle 14 Power Dependent Insertion Limit (PDIL) for CEA's.............. 65 4.10 Cycles 13 and 14 Maximum Peaking versus Dropped CEA Worth from Specified Power Levels............ 66 .xi-

--*ev+ LIST OF FIGURES (continued) Number Title Ppge Cycle 14 Shutdown Margin Equation and Required Scram Reactivity....... 67 4.11 4.12 Cycle 14 Required Shutdown Margin versus RCS Boron Concentration...... 68 5.1 Allowable 3-Loop Steady-State Coolant Conditions.................. 121 5.2 Design Axial Power Distributions................................. 122 5.3 Cycle 14 Normalized Reactivity Worth versus Percent CEA Insertion - BOC Scram at HFP and HZP... ............................... 123 5.4 Cycle 14 Normalized Reactivity Worth versus Percent CEA Insertion - ........ 124 l EOC Scram at HFP and HZP... l l 5.5 Cycle 14 Thermal Margin / Low Pressure Trip Setpoint - A versus Excore Symmetric Offset .. 125 3 5.6 Cycle 14 Thermal Margin / Low Pressure Trip Setpoint - QR versus Fraction of Rated Thermal Power......................... 126 2 5.7 Cycle 14 Symmetric Offset Trip Function - Three Pump Operation ... 127 5.8 Cycle 14 Linear Heat Generation Rate (LHGR) Limits versus Core Height... 128 -xii-

1.0 INTRODUCTION

This report provides justification for the operation of Maine Yankee during the next fuel cycle, Cycle 14. The report describes the fuel mechanical, thermal-hydraulic, physics and safety analysis aspects of Cycle 14 operation, including a description of the startup test program. Significant changes for Cycle 14 are summarized in Table 1.1. The Cycle 14 refueling willinvolve the discharge of 73 assemblies, the insertion of 72 new fuel assemblies, and the reinsertion of one Type M assembly. The new fuel assemblies, designated Type T, are fabricated by Combustion Engineering (CE) and are similar in design to the CE fuel provided for recent cycles. The one Type M fuel assembly, previously irradiated in Cycles 8 and 9, was provided by Exxon Nuclear Corporation (ENC). The Type S and T mechanical designs are essentially identical and differ from the previous CE fuel by incorporation of a debris-resistant lower end-fitting design. The ENC fuel design is similar, but not identical to, the previous CE designs. Small differences exist in the mechanical and hydraulic characteristics of the fuel types, as discussed in Section 3. The core design for Cycle 14 includes an increase in the fresh fuel average enrichment to 3.91 w/o U-235, using two fuel pin enrichments in the assembly designs. The increased assembly average enrichments of the assembly designs are below the Technical Specification enrichment limit. Cycle 14 is a continuation of the low-leakage core designs for vessel fluence reduction, as initiated in Cycle 7. The reload core design is discussed in detail in Section 3. The physics analysis results are presented in Section 4. The results are based on a nominal end-of-Cycle 13 exposure of 14,000 mwd /Mt, which includes power coastdown beyond end of full-power life. The safety analysis results are presented in Section 5. Explicit allowances are included in the inputs to the thermal-hydraulic and safety analyses to justify a range of end-of-Cycle 13 exposure from 12,000 to 14,000 mwd /Mt, and a maximum end-of-Cycle 14 exposure which is defined by the bumup limits of the fuel mechanical design evaluations. The prtnary system operating conditions evaluated for Cycle 14 are unchanged from those evaluated for Cycle 13 in (75). The conditions are a rated core thermal power of 2700 MWt, a _.

steady-state operating pressure of 2225 to 2275 psia, and a maximum indicated core inlet temperature of 551.3'F. A flexible core inlet temperature range from 500 to 551.3*F is justified in the safety analysis. In addition, operation is allowed over a pressure range from 2075 to 2225 psia by imposing a limit on the maximum core inlet temperature at lower pressures to preserve Departure from Nucleate Boiling (DNB) margin. This assures that DNB performance is the same for all possible limiting temperature and pressure combinations. These ranges of conditions are consistent with the power uprate submittal in (62) and bound expected coastdown operation. The secondary system operating conditions evaluated for Cycle 14 are expanded from those evaluated for Cycle 13 in (75). Specifically, the assumption on the range of overall heat transfer coefficient (UA) across the steam generators is expanded to cover a minimum full-power steam generator dome pressure of 755 psia (actual) at the maximum core inlet temperature. This change is discussed in further detail in Section 5. The methods used in these analyses are in accordance with those described in (4-14,66,67,73, 77, 78, 80). These methods have been approved by the NRC for use on Maine Yankee in (15-18, 68, 69, 74, 79, 83). Methods used in safety-related analyses for the fuel mechanical design evaluations are based on CE and ENC generic models which have received prior approval by the NRC. The physics analysis utilizes the advanced nodal methods, which were described in (66, 67) and approved by the NRC in (68, 69). These methods were first applied to Maine Yankee in portions of the Cycle 13 physics analysis in (75). The fuel thermal performance analysis utilizes the FROSSTEY-2 methodology, as described in the documents in (73) and approved by the NRC in (74). The Reactor Protection System (RPS) setpoint analysis utilizes a Statistical Combination of Uncertainties (SCU) methodology, as described in (77, 78, 80) and approved by the NRC in (79). The small break LOCA analysis is updated to incorporate the RELAP5YA computer program, as approved by the NRC in (83). The impacts of the methods changes for Cycle 14 are discussed in the appropriate sections of this report. i l

i I TABLE 1.1 l l Maine Yankee Slenificant Chances for Cycle 14 l Describedin Caterorv Chance for Cvele 14 Section(s) Plant No hardware modifications impacting safety 5.1.6 Modifications analysis assumptions Core Design 72 fresh assemblies witn increased average enrichment, 3.1.1, using two fuel pin endchments in the assembly designs 3.1.2 Initial Increased range of overall heat transfer coefficient 5.1.1, Operating (UA) across the steam generators to cover an increased 5.1.7 Conditions range in steam generator pressure at full power

  • Reduction in nominal core inlet temperature at hot 5.1.1, zero power from 532 to 525"F 5.1.7 Analysis Increased uncertainty on the low pressure portion of 5.1.5, Assumptions the TM/LP trip function' 5.1.7 i

Analysis Fullimplementction of advanced nodal physics metr >ds 4.11 Methods Use of FROSSTEY-2 fuel thermal performance code 3.2.2, methodology 5.1,5.1.7 Use of Statistical Combination of Uncertainties method-5.1, l ology in the Reactor Protection System setpoint analysis 5.1.7 Use of the RELAP5YA computer cc,de in the Small 5.1, Break LOCA analysis' 5.1.7 Cycle-Allowable Unrodded Radial Peaking 4.3 l Dependent l Operating Power-Dependent Insertion Limits for CEA's 4.9.1 i Limits i Required Shutdown Margin 4.9.5 RPS Symmetric Offset LCO bands and TM/LP and SO13 5.1.5, Setpoints setpoints 5.1.7 i

  • Changes evaluated for Cycle 13 operation, subsequent to Cycle 13 CPAR analysis in (75) i

2.0 OPERATING HISTORY The operating history of Maine Yankee consists of fourteen cycles designated as Cycles 1,1A, j and 2 thmugh 13. The significant operating conditions and durations of the cycles are defined i in Table 2.1. The fuel assembly types loaded by cycle are given in Table 2.2. 2.1 Cycles 1 and 1 A The initial core consisted of unpressurized, low density fuel designated as Core 1 design fuel assemblies (Types A, B, and C). Cycle 1 power operation was restricted due to leaking fuel assemblies and rodded operation for moderator temperature coefficient control. Cycle 1 A operated after the leaking fuel assemblies from the initial core were replaced with fresh fuel j designated as Replacement Fuel (Type RF) assemblies. The mechanical design of the Type RF assemblies was essentially the same as the Core 1 design. The significant design difference was the pressurization of the fuel rod with helium sufficient to prevent creep collapse of the fuel rod cladding and improve gap heat transfer. All fuel performed successfully during Cycle 1A. ) 2.2 Cvde 2 i Cycle 2 consisted entirely of fresh assemblies designated as Core 2 design fuel (Types D, E, and F). Mechanical design changes were made in comparison to the Core 1 design fuel. These comprised pre-pressurization, higher fuel density and smaller diameter pellets. The design changes were discussed in (19). The Core 2 design fuel performed successfully. Subsequent to Cycle 2 operation, burnable poison shim failures were discovered in the Type E assemblies. j Corrective action consisted of replacement of all Type E shims with water-filled zircaloy rods prior to reinsertion in subsequent cycles. i ] 2.3 Cycles 3 and 4 Cycle 3 consisted of fresh fuel assemblies of Core 2 design (Types G and H) and Replacement Fuel assemblies (Type RF) reinserted from Cycle 1 A. All fuel perfonned successfully during Cycle 3. Late in Cycle 3 operation, a licensed power uprate from 2440 to 2630 MWt was I i t

i obtained. Plant operation was limited to 97% power due to secondary plant limitations. Cycle 4 consisted of all fuel assemblies of the Core 2 design. Minor design changes for the fresh (Type { D fuel were discussed in (20). New fuel and once-burned fuel assemblies from Cycle 2 were l inserted and the Type RF fuel discharged. A small number of leaking fuel assemblies were I discovered near End-of-Cycle (EOC). i 2.4 Cycles 5 and 6 f Cycles 5 and 6 consisted of fuel assemblies of the Core 2 design and fresh assemblies designed by ENC (Types J and K). A discussion of the ENC design assemblies was provided in (1,23). Five leaking Type H and I assemblies for reinsertion in Cycle 5 were repaired by replacement of failed and suspect fuel rods with fresh, low enrichment Core 2 design fuel (34 rods) or water-filled zircaloy rods (10 rods). All fuel performed successfully during Cycles 5 and 6. l 2.5 Cveles 7 and 8 Cycles 7 and 8 consisted almost entirely of ENC fuel. One Type E assembly of the Core 2 design with Cycle 2 exposure was inserted in the core center location. The fresh ENC fuel (Types L and i M) represented an increase in enrichment to 3.30 w/o U-235. The Cycle 7 design was the first low-leakage, low-fluence core design. The Control Element Assembly (CEA) group configuration for Group 5 was changed, providing increased CEA worth and creating two subgroups,5A and 5B. Plant operation at 2630 MWt was achieved in Cycle 7 with the addition of a steam 4 riven main feedwater pump. Minor fuel design changes for the Types L and M fuel were discussed in (12). All fuel performed successfully during Cycles 7 and 8. ] 2.6 Cycles 9 and 10 i Cycle 9 consisted of fresh fuel assemblies designed by CE (Type N) and reinserted assemblies designed by ENC (Types L and M) at 3.30 w/o U-235 enrichment. One Type E assembly with Cycle 2 exposure was inserted in the core center location in Cycles 9 and 10. The Type N fuel design was discussed in (21). A small number of leaking fuel rods were discovered during Cycle 9 operation. Cycle 10 consisted of fresh (Type P) fuel and reinserted (Types L, M and N) fuel _

TABLE 2.1 Maine Yankee Operating History Summary - Core Power Level - Date of Maximum Cycle Power Licensed Operated Burnup Cycle Escalation ) (MWt) (%) (NBVd/Mt) a 1 11/ 8/72 2440 50-80 ) 10,367 2 1A 10/12/74 2440 80 ) 4,492 G 2 6/29/75 2440 100 17,365 3 6/11/77 2630W 93 11,105 I 4 8/28/78 2630 97(0 10,500 l 5 3/17/80 2630 97(0 10,799 6 7/20/81 2630 97CO 11,585 7 12/12/82 2630 100 12,483 8 6/20/84 2630 100 12,504 9 10/25/85 2630 100 14,424 l 10 6/18/87 2630 100 12,675 11 12/16/88 2700* 97-98W 13,786 12 6/30/90 2700 100 15,364 l 13 4/19/92 2700 100 14,000W m Date of initial phasing onto power grid j m Power restrictions due to leaking fuel and rodded operation for moderator temperature coefficient control. Primay system pressure decrease to 1800-2000 psia for leaking fuel. W Licensed power increase on 6/20/78 from 2440 MWt/2100 psia to 2630 hnVt/2250 psia operation during Cycle 3 m Power restriction due to secondary plant limitations W Licensed power increase on 7/10/89 from 2630 to 2700 hnVt operation during Cycle 11 W Estimated 2 i i

t TABLE 2.2 ' Maine Yankee Fuel Assembly Types By Cycle Assembly Fuel Enrichment Meci nical Number of Fuel Assemblics by Cycle Type (w/o U-235) IAW Type 1 1A 2 3 4 5 6 7 8 9 10 11 12 13 A 2.01 CE-Core 1 69 57 IJ 2.40 CE-Core 1 80 24 C 2.95 CB-Core 1 68 64 RF 233 CE-RF 2 RF 1.93 CE-RF 70 65 D 1.95 CE-Cere 2 69 E 2.52 CE-Core 2 80 12 61 1 1 1 1 1 1 9' F 2.90 CE-Core 2 68 68 12 G 2.73 CE-Core 2 32 32 32 11 3.03 CE-Core 2 40 40 40 I 3.03 CE-Core 2 72 72 72 J 3.00 ENC 72 72 72 K 3.00' ENC 72 72 72 72 72 72 8 8 L 330 ENC M 330 ENC 72 72 64 1 1 1 N 330 CE-Core 2 72 72 64 P 350 CE-Core 2 72 72 72 8 72 72 68 Q 3.70 CE-Core 2 R 3.70 CE-Core 2 72 72 S 70 CE-Core 2 68

3.0 RELOAD CORE DESIGN 3.1 General Description 3.1.1 Core Fuel Loading The core of Maine Yankee Cycle 14 consists of 217 fuel assemblies of the type and quantity detailed in Table 3.1. Assembly Type M is the core center assembly, fabricated by ENC, with r irradiation exposure from Cydes 8 and 9. Assembly Types P, R and S were introduced in Cydes l 10,12 and 13, respectively. Assembly Type T is fresh fuel introduced in Cyde 14. Assembly Types P, R, S and T are fabricated by CE and designated Core 2 design fuel. There are 72 fresh Type T assemblies inserted in Cyde 14, increased from the 68 fresh Type S assemblies inserted in Cycle 13. As in Cycle 13, eight Type P assemblies, inserted in Cycles 10 through 12, are reinserted on the core periphery for their fourth cycle of exposure, providing additional fluence reduction for Cycle 14. The Types M and P fuel initial enrichments are 3.30 and 3.50 w/o U-235, respectively. The Type i R and S fuel initial enrichment is 3.70 w/o U-235. The Type T average enrichment is 3.91 w/o U-235. The total number of fuel rods by assembly type for Cyde 14 is given in Table 3.1, and the core loading by fuel type is given in Table 3.2. l 3.1.2 Assembly Enrichment and Burnable Poison Loadines 1 l The Type M, P, R and S assemblies utilize a single fuel pin enrichment in the assembly designs. The Type T assemblies utilize two different pin enrichments in the assembly design. Fuel pins l of 4.20 and 3.50 w/o U-235, used in assembly designs with 0,4 or 8 burnable poison pins per assembly, are shown in Figure 3.1. The lower enrichment fuel pins in the vicinity of the guide i tubes result in lower local peaking factors for these assemblies. The assembly average enrichments and total number of fuel rods by assembly type are detailed in Table 3.1. As required by Technical Specifications, the radially-averaged enrichment of any axial enrichment zone within each fuel assembly in Cyde 14 is below 3.95 w/o U-235. i 9 i

The burnable poison shim locations in the assemblies are also shown in Figure 3.1. All barnable r poison shims are composed of B,C in A10. The loadings of the burnable poison shims, 2 3 expressed in milligrams of Boron-10 per inch of active shim length, are given in Table 3.1. The CE and ENC design shim irradiation integrity has been demonstrated by the performance of t these designs over many cycles of operation. i 3.1.3 Core Loadine Pattern The fuel assembly locations designated for Cycle 14 are given for the first quadrant in Figure 3.2. They are given relative to the previous locations of the Type M assembly in Cycle 9, the Type P assemblies in Cycle 12, and the Types R and S assemblies in Cycle 13. The appropriate rotation index, relative to the previous assembly position in the core, is also given for each assembly. The loading and rotations of the other quadrants are such that rotational symmetry exists with respect to the quadrant boundary lines. The Cycle 14 loading pattern incorporates a low-leakage design, achieved by placement of fresh fuel assemblies in selected core interior locations and burned fuel assemblies on the core edge. The Cycle 14 loading pattem is similar to the Cycles 7 through 13 low-leakage loading patterns. The benefits of such a core design are: 1) Reduced irradiation exposure to the reactor pressure vessel, thus reducing the rate of irradiation embrittlement, 2) Extended cycle full-power lifetime due to reduced neutron leakage, 3) Preferred fuel rod power and exposure histories for fuel performance and mechanical integrity (i.e., higher relative powers at lower burnups), 4) Improved stability to axial xenon oscillations near EOC, and 5) A less severe moderator defect with cooldown at EOC, providing greater shutdown margin for cooldown transients. t 3.1.4 Assembiv Ilxposure Historv The calculated exposure history of the Cycle 14 fuel assemblies at Beginning-of-Cycle (BOC)is 10-t t

given in Figure 3.3. The exposures are based on an expected cycle length of 14,000 mwd /Mt for Cycle 13 and the achieved cycle length of 15,364 mwd /Mt for Cycle 12. Table 3.2 gives BOC average exposures by fuel type. The BOC average exposure for the core is approximately 17,390 mwd /Mt. The exposure history of the assemblies utilized in the analysis is demonstrated to be l accurate by comparison with incore detector data. Figure 3.4 is a comparison of predicted and j actual assembly burnup data during Cycle 13. The excellent agreement demonstrates a high I confidence in the prediction of the core depletion behavior. i 3.1.5 CEA Group Confismration The Control Element Assembly (CEA) group configuration for Cycle 14 is unchanged from Cycle

13. Figure 3.5 shows the CEA group locations in the quarter core. All CEA's contain five (5) full-strength fingers. The CEA group configuration consists of:

i o Five (5) regulating CEA groups (Groups 5, 4, 3, 2 and 1), each comprised of single CEA's, in which each CEA is attached to one drive shaft, and Three (3) shutdown CEA groups (Groups C, B and A), each comprised i o of dual CEA's, in which two diagonally-adjacent CEA's are attached to i one drive shaft. The first regulating group inserted is Group 5. The Group 5 configuration consists of: Nine (9) CEA's, designated Subgroup 5A, which are scrammable and o contribute to the Available Scram Reactivity (ASR), and l o Four (4) CEA's, designated Subgroup 5B, which are non-scrammable and do not contribute to the ASR. These four CEA's were added to Group 5 for local power distribution control, beginning in Cycle 7. The CEA groups are subject to the insertion limits described in Section 4.9.1. As in Cycle 13, j l Subgroups.5A and 5B are independently moveable and not directly connected as a single CEA l group. As such, their movements are admmistratively controlled for positioning as a single CEA group. To accommodate this movement, the physics input to the Reactor Protective System (RPS) setpoint analysis includes power distribution cases sufficient to justify a 15 step difference in insertion between these subgroups, as allowed by Technical Specifications. In addition, each subgroup's movement is also subject to the CEA group insertion limits. 3.2 Fuel System Desien l 3.2.1 Fuel Mechanical Desien The fuel assemblies are designed to maintain mechanical, material, chemical, and thennal-hydraulic compatibility with all other fuel and structures in the reactor. Table 3.3 lists l 4 the mechanical design features and vendors of all fuel types in Cycle 14. The detailed fuel assembly descriptions and mechanical design criteria for the reinserted fuel are described in (1, l 21, 23-26, 63, 70). The mechanical design of the fresh Type T fuel, described in (76), is essentially identical to the j Type S fuel supplied by CE for Cycle 13. These fuel types are similar to the Types P, Q and R fuel currently in use in Cycle 13. Relative to these fuel types, the Type S and T fuel designs include the following improvements: o A debris-resistant lower end-fitting and lower retention grid design, and I o A debris-resistant lower end-cap design, for both fuel and burnable poison rods. i In order to maintain acceptable free plenum volume in the fuel rod design, a lower volume plenum spring design is used and the nominal active fuel length is decreased from 136.7 to 136.25 inches. These design changes are addressed in the mechanical design evaluations in (70, l 76). Thermal-hydraulic differences between fuel types are addressed explicitly for Cycle 14 in Section 3.2.3. a o

The Type M fuel, supplied by ENC, will achieve exposures higher than the original design analysis and has been reanalyzed to demonstrate compliance with the appropriate design criteria at these higher exposures. These analyses are documented in (26) and employ the methods described in (27), which have been reviewed by the NRC staff in (28). I 3.2.2 Fuel Thermal Analysis T The thermal effects analysis encompasses a study of fuel rod response as a function of power and exposure history. An enveloping fuel rod power and burnup history is evaluated, which bounds the maximum power and burnup rods of each fuel type in each cycle. A detailed evaluation of each fuel assembly's maximum power and bumup rod histories is performed each cycle to verify the enveloping history used in the analysis. 1 The fuel thermal analysis for Cycle 14 is changed to the FROSSTEY-2 fuel performance methodology, which is described in the documents in (73) and approved by the NRC in (74). The FROSSTEY-2 code calculates pellet-to-clad gap conductance from a combination of I theoretical and empirical models which predict fuel and cladding thermal expansion, fission gas release, pellet swelling, pellet densification, pellet cracking, and fuel and cladding thermal conductivity. The Specified Acceptable Fuel Design Limit (SAFDL) for Fuel Centerline Melt (FCM) is expressed as a Linear Heat Generation Rate (LHCR) limit as a function of local burnup. The UO melting temperature as a function of local bumup is provided in Technical 2 Specifications. j A bounding LHGR limit for the FCM SAFDL, versus local bumup, is shown in Figure 3.6. This i figure bou.. " the FROSSTEY-2 calculated results for the enveloping fuel rod power and burnup history. With local burnup, the LHGR limit decreases by a ratio of (19.6/22.0), or 0.89, from fresh to maximum local burnup conditions. These LHGR limits are similar to those reported for the previous cycle in Table 3.4 of (75), which were based on the GAFEX computer code (29). In Table 3.4 of (75), the ratio of the LHGR limits for highly bumed fuel (Type Q at EOC) to fresh fuel (Type S at BOC) was (19.6/23.5), or 0.83. i Ratios of the maximum radial relative pin power for the reinserted fuel types to the maximum ! [ t

l in the core are provided in Table 3.4. Since the LHGR limit is shown to decrease by a factor in the range of 0.8 - 0.9 with local burnup, any fuel types with ratios in Table 3.4 which are less than 0.8 will not be limiting. Clearly, the Types M, P and R fuel are not limiting, since the maximum ratio for these fuel types is 0.708. For the Types S and T fuel, the maximum local l burnup of each fuel type at BOC and EOC, and its associated LHGR limit from Figure 3.6, are shown in Table 3.5. The ratios of the Type S to the Type T LHGR limits are also calculated. The LHCR limit ratios for the Type S fuel in Table 3.5 are greater than the power ratios in Table 3.4. Thus, maintaining the FCM SAFDL's for the fresh Type T fuel assures that those of the reinserted Type S fuel are met since: 1) The fresh fuel contains the core-wide maximum power pin during the cycle, and i i 2) The ratio of maximum pin power in the reinserted fuel to that of the I fresh fuel is always less than the ratio of LHGR limits, indicating that the fresh fuel is always closer to the FCM SAFDL than the reinserted i fuel. Adequate margins to the FCM SAFDL's are demonstrated for all reinserted fuel types by the use of ratios in Tables 3.4 and 3.5, for the stated rodded conditions and times-in-life. Similar ratios are also maintained under transient conditions, such as CEA drop or withdrawal. 3.2.3 Thermal-Hydraulic Desien Steady-state and transient Departure from Nucleate Boiling Ratio (DNBR) analysis of Cycle 14 has been performed using the COBRA-IIIC computer program (30), in the manner described in-(4,5) and as described below. The models reflect the intended Cycle 14 coolant conditions, power distributions, inlet flow maldistribution, assembly flow redistribution due to differences in hydraulic characteristics and the specific geometry of the fuel assemblies. A COBRA-IIIC model was used to determine hot assembly enthalpy rise flow factors. This model explicitly represents each fuel assembly in the one-eighth (1/8) core in the specific l l

l location it will reside for Cyde 14 operation. Differences in hydraulic characteristics between the standard CE, the debris-resistant CE, and the ENC fuel assemblies are explicitly modeled. The inlet flow maldistribution imposed on this model is based on the results of flow j measurements taken in scale model flow tests of the Maine Yankee reactor vessel reported in (31) and the FSAR (32). The hot assembly enthalpy rise flow factor for the fresh CE fuel was calculated to be 0.962 for l ' bottom-peaked power distribudons and 0.971 for top-peaked power distributions. The introduction of the debris-resistant design for the Type S and T fuel has resulted in an increased flow penalty for Cycles 13 and 14, relative to previous cycles, due to a higher inlet spacer loss coefficient. This penalty has diminished from Cyde 13 to 14, as more fuel of this design is inserted. A 0.950 enthalpy rise flow factor is conservatively applied to the ENC assembly. These factors are applied to the inlet mass velocity in the hot channel model in predicting DNB performance. 1 I The potential effects of fuel rod bow on thermal-hydraulic performance have also been evaluated for Cycle 14 operation. Using the channel closure correlation in (33), the maximum channel gap i closure due to fuel rod bowing for the CE fuel assembly with the highest burnup during Cyde l 14, a Type R assembly, was calculated to be 26E Tests performed at Columbia (34) indicate that I a degradation in DNB performance is not experienced until channel closures exceed 50E Therefore, no penalty is required for fuel rod bow considerations. i Allowances for manufacturing tolerances on rod pitch and clad diameter for the ENC assembly, _ I if considered in the most adverse situation, would result in a maximum channel closure of about 1 10E Using the methodology of (35), the maximum channel gap closure due to fuel rod bowing for the ENC fuel assembly during Cyde 14 was calculated to be less than 345 Therefore, no penalty is required for fuel rod bow since the channel dosure resulting from rod pitch, bow and clad diameter considerations for the ENC assembly during Cycle 14 will be less than 50% A list of the pertinent thermal-hydraulic design parameters, used for both safety analysis and for generating RPS setpoint information, is presented in Table 3.6. The list also indudes the l corresponding thermal-hydraulic parameters for comparison to Cycles 3 and 13 (36,75). a 1

TABLE 3.1 Maine Yankee Cycle 14 Assembly Derription Cycles Number Initial Number Initial Number Number Number of of Fuel Assembly of Shim mg B-10 of of Total of Total Assembly Exposure Rods per Average Locations per inch Assemblics Shim Fuel Desinnation History Assembly w/o U-235 in Assembly in Shims in Core Locations Rods M-8 8,9 168 3.300 8 23.8 1 8 168 P-0 10,11,12 176 3.500 0 8 0 1,408 T R-0 12,13 176 3.700 0 36 0 6,336 R-4 12,13 172 3.700 4 31.4 12 48 2,064 R-8 12,13 168 3.700 8 31.4 20 160 3,360 S-0 13 176 3.700 0 20 0 3,520 S-4 13 172 3.700 4 31.4 28 112 4,816 S-8 13 168 3.700 8 31.4 20 160 3,360 T-0 Fresh 176 3.914 0 8 0 1,408 T-4 Fresh 172 3.907 4 31.4 28 112 4,816 T-8 Fresh

68 3.900 8

31.4 36 288 6,048 Core Totals 217 888 37,304 i l l

TABLE 3.2 j Maine Yankee Cvele 14 Core Loadine I Uranium per Uranium Exposure Assembly Number of Assembly Total at BOC' i Type Assemblies (kt U) (ke U) (mwd /Mt) i M-8 1 363.8 364 31,474 P-0 8 388.7 3,110 34,762 R-0 36 390.8 14,069 31,739 l t R-4 12 381.9 4,583 36,034 l R-8 20 373.0 7,460 36,039 j i S-0 20 389.5 7,790 14,794 i S-4 28 380.6 10,657 16,977 S-8 20 371.8 7,436 18,914 i T-0 8 389.5 3,116 0 T-4 28 380.6 10,657 0 i T-8 36 371.8 13,385 0 l l Core Totals 217 82,627 17,390 " i a Based on End-of-Cycle 13 at 14,000 mwd /Mt Core average exposure is based on a uranium-mass weighing of the fuel type exposures l l 4 I l

TABLE 3.3 Mechanical Desien Features of Maine Yankee Cycle 14 Fuel Type M Tvne P Tvoe R Type S, T Fuel Vendor ENC CE CE CE Fuel Assembly Overall length 156.718' 156.718 156.718 156.750 Spacer grid size (maximum square) 8.115 8.115 8.115 8.115 t Number of zircaloy grids 0 8 8 8 Number of inconel grids 0 1 1 1 Number of bimetallic grids 9 0 0 0 Fuel rod growth clearance (cold) 1.300 min. 1.600 1.600 1.600 Fuel Rod Active fuel length 136.70 136.70 136.70 136.25 Plenum length 8.8 8.375 8.375 7.413 Clad OD 0.440 0.440 0.440 0.440 Clad ID 0.378 0.384 0.384 0.384 Clad wall thickness 0.031 0.028 0.028 0.028 Pellet OD 0.370 0.3765 0.3765 0.3765 Pellet length (37) 0.450 0.450 0.450 Dish depth 0.008 0.021 0.021 0.021 Clad material Zr-4 Zr-4 Zr-4 Zr-4 Initial pellet density 94.0 % 94.75 % 95.25 % 95.25 % Initial fill gas pressure (37) (24,53) (63) (70,76) Poison Rods Overall rod length 146.500 146.322 146.322 146.472 Active poison length 122.70 122.70 122.70 122.70 Clad OD 0.440 0.440 0.440 0.440 Clad ID 0.378 0.384 0.384 0.384 Clad wall thickness 0.031 l'.028 0.028 0.028 Pellet OD 0.353 0.362 0.362 0.362 Clad material Zr-4 Zr-4 Zr-4 Zr-4

  • Alllength dimensions are in inches CE - Combustion Engineering ENC - Exxon Nuclear Corporation TABLE 3.4 Maine Yankee Cycle 14 Ratio of Maximum Radial Relative Pin Powers -

Maximum in Reinserted Fuel Types to Maximum in Core I Rodded Conditions HFP, Equilibrium Conditions of ARO Ratio of Maximum Radial Relative Pin Powers at Cycle Exposure (mwd /Mt) Fuel Regulating BOC MOC EOC Type Groups Inserted 500 6,000 15,000 M ARO 0.609 0.647 0.654 Group 5 0.460 0.471 0.483 Groups 5 + 4 0m 0.237 0.243 9 P ARO 0.348 0.383 0.441 Group 5 0.322 0.355 0.430 Groups 5 + 4 0.327 0.373 0.445 1 R ARO 0.685 0.692 0.690 Group 5 0.687 0.708 0.696 Groups 5 + 4 0.643 0.663 0.671 S ARO 0.940 0.902 0.842 Group 5 0.959 0.919 0.888 Groups 5 + 4 0.957 0.947 0.902

TABLE 3.5 Maine Yankee Cycle 14 Boundine LHGR Limits for FCM SAFDL and Ratios by Fuel Type t Bounding Bounding Ratio of Time Local LHGR for LHGR to in Fuel Burnup FCM SAFDL Type T Life Type (mwd /Mt) (kW/ft) LHGR i BOC S 23,600 21.19 0.963 T 0 22.00 1.000 EOC S 44,700 20.29 0.965 T 28,200 21.03 1.000 l F t t l l i

TABLE 3.6 Maine Yankee Cycles 3,13 and 14 Thermal-Hydraulic Parameters at Full Power General Characteristics Units Cycle 3 Cycle 13 Cycle 14 Total Heat Output MWt 2630 2700 2700 10' Btu /hr 8976 9215 9215 Fraction of Heat Generated in Fuel Rod 0.975 0.975 0.975 Pressure psig 4 Nominal 2235 2235 2235 Minimum in Steady-State 2185 2060 2060 Maximum in Steady-State 2285 2260 2260 Design Inlet Temperature (steady-state)

  • F 554 548 - 556m 548 - 556m Total Reactor Coolant Flow (design) 10' lb/hr 134.6 135.7 - 134.2 135.7 - 134.2 Coolant Flow through Core (design) 130.7 131.6 - 130.2 131.2 - 129.7 Hydraulic Diameter (nominal channel) ft 0.044 0.044 0.044 Average Mass Velocity 10' lb/hr-ft2 2.444 2.460 - 2.433 2.453 - 2.425 Pressure Drop across Core (design flow) psi 9.9 11.3m 11.55 Total Pressure Dn>p across Vessel (based on nominal dimensions and design flow) psi 32.4 33.75 33.90 Core Average Heat Flux" Btu /hr-ft 178,742 184,072 184,702 2

Total Heat Transfer AreaW ft' 48,978 48,811 48,644 i

._= - TABLE 3.6 (Continued) Maine Yankee Cycles 3,13 and 14 Thermal-Hydraulic Parameters at Fuil Power General Characteristics Units Cycle 3 Cycle 13 Cycle 14 [ Film Coefficient at Average Conditions Blu/hr-ft'"F 5,640 5,840 5,824 Maximum Clad Surface Temperature 'F 656 657 657 Average Film Temperature Difference 31.7 31.5 31.7 G Average Linear Heat Rate of Rod kW/ft 6.03 6.21 6.23 5 Average Core Enthalpy Rise Btu /lb 68.7 70.8 71.0 Calculational Factors ENC Q ENC G ENC g Engineering Heat Flux Factor (* 1.03 1.03 1.03 1.03 1.03 Engineering Factor on Hot Channel Heat Input (0 1.03 1.03 1.03 1.03 1.03 Flow Factors: Inlet Plenum Non-Uniform Distribution 1.05 1.05 1.05 1.05 1.05 Rod Pitch, Bowing and Clad Diameter (0 1.065 1.00 1.M5 1.00 1.065 m Temperature range assumed in safety analysis, except the containment pressure analysis (62). O Irrecoverable pressure losses only, incorporating recent fuel vendor flow test measumment results for the zinaloy spacer grid loss coefficient -G Includes allowance for axial shrinkage due to fuel densification (O Factors which am statistically combined in the Cycle 14 analysis (77-80) m m .-4 -.---m...- .....,..~,...,---..~...~...,_-,.m.

Figure 3.1 Maine Yankee Cycle 14 Fuel Pin and Burnable Poison Shim Assembly Locations Types M, P, R and S Fuel 00000000000000 00000000000000 00000000000000 00000000000000 OO 000000 OO 00O000000n00 OO 000000 OO OOVOOOOOOUOO 00000000000000 00000000000000 00000000000000 00000000000000 OOOOOOAOOOOOO 000000n000000 OOOOOOVOOOOOO OOOOOOUOOOOOO 00000000000000 00000000000000 00000000000000 00009000000000 OO 000000n00 OO 000000 OO OO OOOOOOUOO OO 000000 OO 00000000000000 00000000000000 00000000000000 00000000000000 0 Shim Assembly 4 Shim Assembly Types P-0, R-0 and S-0 Types R-4 and S-4 00000000000000 00000000000000 000000090A00 OOVOOOOOOVOO 00000000000000 00000000000000 000000n000000 OOOOOOUOOOOOO 00000000000000 00000000000000 000000000A00 OOUOSOOOOUOO 00000000000000 00000000000000 8 Shim Assembly Types M-8, R-8 and S-8 O Fuei Pin 9-Shim Pin, 31.4 mg B-10/ inch Figure 3.1 (continued) Maine Yankee Cycle 14 Fuel Pin and Burnable Poison Shim Assembly Locations Type T Fuel 00000000000000 00000000000000 00000000000000 00000000000000 00 000000 00 00A000000A00 OO 000000 00 OOVOOOOOOVOO 00000000000000 00000000000000 00000000000000 00000000000000 000000Q000000 000000Q000000 OOOOOOUOOOOOO OOOOOOUOOOOOO 00000000000000 00000000000000 00000000000000 00009000000000 00 000000 OO OO OOOOOOOOO 00 000000 OO OO OOOOOOVOO 00000000000000 00000000000000 000@O000000000 00000000000000 0 Shim Assembly 4 Chim Assembly Type T-0 Type T-4 00000000000000 00000000000000 OO 000000 OO OO 000000 00 00000000000000 00000000000000 000000 000000 000000 000000 00000000000000 00000000000000 OOOOOOOOO OO OOUOSOOOO 00 00000000000000 00000000000000 8 Shim Assembly Type T-B O - Fuel Pin, 3.5 w/o U-235 0 -- Fuel Pin, 4.2 w/o U-235 9 - - Shim Pin, 31.4 mg B-10/ inch 1 Figure 3.2 Maine Yankee Cycle 14 Assembly Loading Pattern FUEL q TYPE M M-8 62 Cycle 14 Location 28 Cycle 9 Location 0 Rotation Index* R-0 1 P-0 2 11 1 P l P-0 2 Cycle 14 Location 2 2 1 Cycle 12 Location 2 Rotation index* R-0 3 T-0 4 T-4 5 T-8 6 R-4 7 16 36 R R-0 1 Cycle 14 Location 3 1 S 11 Cycle 13 Location 2 Rotation Index-R-0 8 T-4 9 S-4 10 S-0 11 S-8 12 T-4 13-33 6 4 19 T T-0 4 Cycle 14 Location 0 0 0 3 R-0 14 T-8 15 S-8 16 T-8 17 R-8 18 T-8 19 R-8 20 ) 42 17 27 28 O 1 0 0 R-0 21 T-4 22 S-8 23 S-0 24 R-8 25 T-4 26 R-4 27 S-4 28 23 31 15 35 43 26 1 3 2 3 3 3 i T-0 29 S-4 30 T-8 31 R-8 32 T-8 33 R-0 34 S-0 35 S-4 36 46 50 10 5 13 0 1 3 0 0 T-4 37 S-0 38 R-8 39 T-4 40 R-0 41 S-8 42 S-4 43 S-4 44 29 49 30 53 9 40 R-0 45 0 0 1 0 0 0 36 2 T-8 46 S-8 47 T-8 48 R-4 49 S-0 50 S-4 51 R-0 52 T-8 53 I 48 51 37 22 44 P-0 54 1 1 0 0 0 45 2 R-4 55 T-4 56 R-8 57 S-4 58 S-4 59 S-4 60 T-8 61 M-8 ' 62 59 58 26 56 40 28 1 0 2 0 3 0 Clockwise multiple of 90 degrees relative to previous cycle of insertion indicated. :-4

1 J Figure 3.3 l Maine Yankee C cle 14 i Calculated Assembi Exposures BOC (0 MW /Mt) Assembly Type and Core Position R-0 1 P-0 2 Assembly Average Exposure 33091 34761 R-0 3 T-0 4 T-4 5 T-8 6 R-4 7 32013 0 0 0 35974 R-0 8 T-4 9 S-4 10 S-0 11 S-8 12 T-4 13 28804 0 15780 13350 18986 0 R-0 14 T-8 15 S-8 16 T-8 17 R-8 18 T-8 19 R-8 20 29543 0 18839 0 36557 0 35716 R-0 21 T-4 22 S-8 23 S-0 24 R-8 25 T-4 26 R-4 27 S-4 28 31950 0 18822 15816 35756 0 36122 18878 T-0 29 S-4 30 T-8 31 R-8 32 T-8 33 R-0 34 S-0 35 S-4 36 0 15779 0 35652 0 32278 15726 18733 T-4 37 S-0 38 R-8 39 T-4 40 R-0 41 S 42 S-4 43 S-4 44 0 13354 36514 0 32236 18986 15392 18898 R-0 45 33019 T-8 46 S-8 47 T-8 48 R-4 49 S-0 50 S-4 51 R-0 52 T-8 53 0 18938 0 36010 15731 15382 32714 0 P-0 54 34761 R-4 55 T-4 56 R-8 57 S-4 58 S-4 59 S-4 60 T-8 61 M-8 62 35973 0 35716 18878 18732 18898 0 31474.

Figure 3.4 Maine Yankee Cycle 13 Burnup Distribution by Assembly INCA vs. Predicted Near 6000 mwd /Mt Cycle Exposure Assembly Type and INCA Location O-4 8 P-8 21 INCA Assembly Exposure (mwd /Mt) 36180 40323 Predicted Assembly Exposure (mwd /Mt) 36031 40391 Percent Difference - 0.4 0.2 0-0 15 S-0 31 S-0 11 S-4 25 R-4 4 33351 5858 7028 6743 25602 33245 5848 6917 6675 25466 -0.3 -0.2 - 1.6 - 1.0 - 0.5 O-4 16 S-4 33 R-0 13 R-0 28 O-4 7 S-4 20

  • Maximum Exposure 38540 6643 23367 24536 40566 7920 38394 6606 23407 24240 40421 7940 Octant Location 10 0,4 0.6 0.2

- 1.2 -0.4 0.3 i Measured 41807* Predicted 41930* S-0 34 R-0 14 S-8 30 O-8 10 S-8 24 O-0 3 % Difference 0.3 6900 23666 7976 41807

  • 8009 37959 6928 23158 8000 41930
  • 7985 38179 0.4

- 2.1 0.3 0.3 -0.3 0.6 0-0 32 O-4 12 S-4 27 R-8 6 R-8 19 35528 40105 8008 27055 26783 35390 40297 8122 27336 26677 -0.4 0.5 1.4 1.0 -0.4 R-0 29 O-O 9 R-8 23 R-4 2 l Fuel Exposure (MWiMt) Percent Type INCA Predicted Differene 20220 37632 26779 26540 20218 37510 26892 27246 M-8 36802 37024 0.6 0.0 -0.3 0.4 2.7 P-8 40323 40391 0.2 0-0 36163 36161 0.0 R-0 26 R-4 5 R-0 18 04 38848 38786 - 0.2 20616 26905 23460 O-8 41807 41930 0.3 20647 27438 23650 R-0 23048 22903 - 0.6 0.2 2.0 0.8 R-4 264BB 26897 1.5 R-8 26890 27027 0.5 O-O 22 S-8 1 ~ S-0 6534 6492 - 0.7 37691 7892 i S-4 7214 7249 0.1 38048 8041 S8 7972 8002 0.4 0,9 1,9 + Core 24107 24122 0.1 M-B 17 Absolute Average 0.72 Standa d Deviation-0.96 370 0.6 c Percent Difference CA 27-

Figure 3.5 Maine Yankee Cycle 14 CEA Group Locations R-0 1 P-0 2 Regulating Shutdown Dual CEA Groups CEA Groups 5 (SA and 58) C 4 8 R-0 3 T-0 4 T-4 5 T-8 6 R-4 7 3 A C 1 R-0 8 T-4 9 S-4 10 S-0 11 S-8 12 T-4 13 A C 5A R-0 14 T-8 15 S-8 16 T-8 17 R-8 18 T-8 19 R-8 20 5A A 3 R-0 21 T-4 22 S-B 23 S-0 24 R-8 25 T-4 26 R-4 27 S-4 28 A 2 T-0 29 S-4 30 T-8 31 R8 32 T-8 33 R-0 34 S-0 35 S-4 36 C A 5B* T-4 37 S-0 38 R-8 39 T-4 40 R-0 41 S-8 42 S-4 43 S-4 44 R-0 45 C 2 B 4 T-8 46 S-8 47 T-8 48 R-4 49 S-0 50 S-4 51 R-0 52 T-8 53' P-0 54 1 3 B R-4 55 T-4 56 R-8 57 S-4 58 S-4 59 S-4 60 T-8 61 M-8 62 5A 4 5A

  • Non-Scrammable CEA Locations (Subgroup 5B) i Figure 3.6 Maine Yankee Cycle 14 Bounding LHGR for Fuel Centerline Melt SAFDL versus Local Burnup 23

..j. .......g J.

.... 4..i.....(...

.4... .....g...........4.... .4... ..c., COORDINATES l. i... ]1i;...,.;.. 7 -*- i i i 1i ti i 1 i.- - j-g '.77-r + - -

7 f

t : p ) i i, .._ ',s _.i...!,. i .I...I-Local i i , 3. ! i i ..g.. .. 9. g.7 7 ..e._7. i i. _. gIi ..'..a..l q....i. .. _.i..._.._. f, ....g..* T~ Bumup LHGR . T....i.. q.. ..... 7., ..7.. 9.....p.......4... ..e... _,.,..1 l (. e j .....!.:..: i if! I. l 4. I i !i1 ! ' i ! I k k. 0 22.0 i' i t i 29 21.0 N. i i. p;.j ,. s 1..!.,1.. L,,I. i l 2 i m... ..i.. t...:...i. 4..., ..p.. .. p.. 60 19.6 ..!.. '.. i .t !. 4..+. 4 . ii ......g... g 9 f } a .!...$. 1.. I .........p. ....4... .4. i .,! 9 4_ .. q.....' -..e...q. ....i..,:... ll { [ i... I..;..,!.! i i l1 ..a:...l' a. w.. -._. ' f.4....i. 4 ' l' iji 1.. . :.4 . !p .i.- l l' i, j ,I...! l g .7; ..,. !...I I I .i... 's i 4-...t . 9.. ..yf.a .. i,, I..... ..j..]. ...j.. .h=~ . 4 ..j.. ..h. ..h.., 6... 9.. ' i.. i.,.. g;.. i. ...i......,. : ;; ..{:... ^.i.- ....9.. ............ ;.....4.., p 4.. ....;..q... ....p......... .? g 4 iil i 'I i! i 1 ' i Jili ' ' iiljj ...,i..,. 4 i..... . 4'.4.. 4 .[.... !. L.....i. ..._.,I i.., - i Iili! 5 i ! !1 21

  • {

.D 4 i 4 .J. . j. ..f a ..p. 4........+.....o.. ..p. ...3... ......4.. .p. ...p.........}._ .+.....;... _.i.. l }' 1 i i I. .. '. '. ' ~. ' iiI : i l.

4..t, '*; -'.~i,... '

'.'.'.1'"':

  • ~+"!

' ' * - ' + - ' ' ' ~ + ~ - + - * - s ' ~ '. '. '. ' ' - 1i --*-'*I*!! '!. l ' '.' " r' T-',.... '---'..I.'.l.......... .....t...!. : ....;.,... 1 i ~ 1: ....i.4 .4.....

1. 7.i.....

.I,ji 'iiiii i...+...i _ i_ F 44 jn .. p.j. - g ...j;jit, ..p. p.4. 4..;..j.y q... ......;..J. j... 4..j.... ...{... ..9., ' q .....a ..i.. j.. ...;.... 4.. g ..u.. f'I ... _4 4 +.. _ ,e 4,,T,,,_ i Z ... 4. a..f.a. ...,1.. ,,p,,,,,, { ^ , _g,, ,,,,,,f,4,,, I t, i ~ ! '! '......... '- i }'i I.... l,l: l l8 - 4 ji a l._j...,,j... +..,1.4 4.

  • I-l 4.!

i J .~ I...i _ i i ~ Ii I Ql I 20

I f

! i 3 1 i. q~.

7. p.. !. 9..

i ..p 9 4....j..g q. ...a... .q. . 9 q.., 9.4, ...p.. !.... 3. q!. 9. 9.. t..... ...p..i. .9...... ...p ...i....., t

t i

.....m..... I..;... d..; N .. 4. 4....... 6... .4..........!.. ..L. li - ....4....a 4... i i i ... l..l 1 !!; ...,_,i...i._. l... .,.. i i....!.i ' t... !....... :.4 i_l i !.-.i.t! !.{ ~ ) ....1......1 1 1.... ......! m.i i

i' i

i..- _.i _ ... 7..i q:. i ..7 1 1..p: '. g. i ! 4 3 ..Ll.. L i1 4 4. L4.L .i.. -.4 4 4 4 ..L 444 . l. 4.+. 4 ......i...!. 4

i

!...i.... p!.. ..!. 4.. i ...i...i... ...1..- ,.1.......I~i i J. 4..... _ ' ~ ..p.& ........i.) 1 ..r.. .I ' I - i 2 .J. .i l .j .J..... 4.......... 4 &..;.. 8 ! I ......t.l...L.. .4... .4. i ' { i. I 1; I .......i_j. i..... ! _.i.... !1 ! ... '.i.!...... 4......:..[..[ ij' .i.. 19 1 i e i i 4 ii! ' ii ! i ' i t! Ii i! l ! l !i ( l ! i

n. 4 4.1..

.a.. J4 4 a .d4 .j.. l 4.ij ...i.1..t.. .4.1.L 4 ..I....j.. . 4.li.[....j.. ~ ..i..L .j. .i ~ ...l I 4....1 4 ll l } i ), iI ' Ii i ! ' j' 4. 4 +.4.. f k.

4. !......e.'

I!I: i ^ ii)'!'. f... ...4. 4....... f }. l. 4.. _.....:... ..,.k.. I i; .....3. 4..,... . {ll ~1 7,. y

i.,:. }.4 4 _

.4i i ! i l lIi: ...,.4 ,1 t-'+'-+i i ......74 1 ..,. I.7 4.. 1 . 7,i f,. q!- 4.. a _{'...y.. s 3 p. 4..p:.. 4 3 q. 2 .p,I.. ,.r.t......: .4...(,... .7 1 .... 4.. i..q;... .g. 2 .4 -.a..q...i...t.. a:..q..._.. 5.. i...:...:......... -l-.... ,...l..,._. ..l..,............i.- - i i.,t. ;. ........,...!.i..' ...'..._...e!.... 1i. ...i..... 1 ' l*! : I.. i i

._ ~ i-

'i: i.j ... _.i.i ...p..; .a 3..a... 4... - l pa j t 4 q jr { 18 0 10 20 30 40 50 60 Local Bumup (KMWD/MT) .=- 4.0 PHYSICS ANALYSIS 1 4.1 Fuel Manacement Maine Yankee Cycle 14 consists of irradiated and fresh fuel assemblies, as descr. bed in Section I 3.1.1. The core layout is given in Figure 3.2. The assembly enrichment and burnable poison I loadings are described in Section 3.1.2. A low-leakage loading pattern is employed, as described in Section 3.1.3. For an end-of-Cycle 13 at 14,000 mwd /Mt, Cycle 14 is expected to attain a full power lifetime of approximately 13,100 mwd /Mt. I 4.2 Core Physics Characteristics i The primary nuclear characteristics of Cycles 3,13 and 14 are given in Table 4.1. Parameters are typically compared for nominal Hot-Zero-Power and Hot-Full-Power (HZP and HFP) conditions at Beginning and End-of-Cycle (BOC and EOC). The Cycle 13 and 14 core physics characteristics are similar, but differ due to the following core design changes: l i o Increased fuel enrichment in the fresh fuel, j o Increase in the number of fresh assemblies from 68 to 72, and o Increase in the number of burnable poison shims. ll l The changes in the physics characteristics for Cycle 14 are discussed in the following sections. All Cycle 14 physics results are based upon the advanced nodal methods, which were described j in (66,67) and approved by the NRC in (68,69). I 1 l j 4.3 Power Distributions l 4 Assembly relative power densities for Cycle 14 at HFP, equilibrium xenon conditions are presented for unrodded and rodded (CEA Group 5 inserted) configurations at Beginning, l Middle, and End-of-Cycle (BOC, MOC, EOC). Figure 3.5 shows the locations of the CEA groups. j The unrodded power distributions at BOC, MOC and EOC (500,6,000 and 15,000 mwd /Mt) are l presented in Figures 4.1 through 4.3. The rodded (CEA Group 5 inserted) power distributions i ) : i r

I i a at BOC, MOC and EOC are presented in Figures 4.4 through 4.6. For all of these conditions, the j maximum radial peakings for Cycle 14 are less than Cycle 13's. The maximum unrodded and rodded radial peakings are 0.6% and 4.9% lower than Cycle 13's, respectively. Comparisons of measurements to the allowable unrodded radial peaking (with uncertainties) j versus cycle exposure are required by Technical Specifications to ensures peaking will not exceed i t the values used in safety analysis. The limits in Figure 4.7 are based on predictions which ] change each operating cycle. The maximum radial peaking is shown to occur early in cycle life. j ] In the quarter core, the core power distributions are slightly octant asymmetric due to the t burnup gradients across the octant boundary assemblies. The quarter-core analysis uses rotational symmetry boundary conditions, which are consistent with the symmetry of the actual I~ full core loading. r 4.4 CEA Group Reactivity Worths The CEA group configurations were shown in Figure 3.5. The CEA group worths at HZP are i presented in Table 4.2 for Cycles 13 and 14. In general, the CEA group worths for Cycle 14 are i i slightly less than Cycle 13's for the regulating groups, and slightly greater for the shutdown groups. The total worths in the All-Rods-In (ARD configuation are comparable. The differences between the cycles are the result of power distribution changes near the CEA locations. CEA ) 1 group reactivity worths are verified by the startup test program and the associated acceptance j criteria. ^ l 4.5 Doppler Reactivity Coefficients and Defects 1 The fuel temperature, or Doppler, components of reactivity are presented in Tables 4.3 and 4.4 2 for nominal conditions in Cycles 13 and 14. The total core average Doppler defect from 4,000"F l is given in Table 4.3 and the core average Doppler coefficient in Table 4.4. The defects and f coefficients in Cycle 14 are smaller than Cycle 13's, due primarily to the increase in the fresh fuel importance from the increase in number of fresh assemblies. Uncertainties of 25% are conservatively applied to the coefficient and defect values prior to transient analysis, except for. 1 ,1

h CEA Ejection. The CEA Ejection methodology utilizes the core average unrodded Doppler defects from Table 4.3 and a Doppler weighing factor technique, as described in detail in (13). An uncertainty of 15% is applied to the Doppler defects for CEA Ejection, prior to transient analysis. These uncertainties and methods are unchanged for Cycle 14. j ~ l 4.6 Moderator Reactivity Coefficients and Defects The Moderator Temperature Coefficients (MTC's) at nominal operating HFP and HZP critical boron conditions are presented in Table 4.5 for Cycles 13 and 14. The MTC's in Cycle 14 are 1 l more positive (or less negative) than Cycle 13's, due primarily to the increase in the fresh fuel importance from the increase in number of fresh assemblies. l The MTC Technical Specification limits are shown in Figure 4.8 and are unchanged for Cycle 14. Prior to transient analysis, an uncertainty of 0.5 x 104 Ap/"F is conservatively applied to calculated MTC values, or, if more limiting, the MTC Technical Specification limit with an additional allowance of + 0.1 x 104 Ap/'F is used. The MTC uncertainties and methods are l unchanged for Cycle 14. The Moderator Density Defect (MDD) curve used in the LOCA analysis [ infers specific MTC values in the operating range and are consistent with the Technical Specification limits. The startup test program demonstrates the validity of the ana'ysis values and assures that the limits will not be exceeded. The MDD appropriate to the scrammed-less-worst-stuck CEA configuration is given in Table 4.6 for Cycles 13 and 14. This MDD curve yields a conservative moderator reactivity increase versus temperature or density, while accounting for the effects of loss in total CEA worth and the worst stuck CEA. The worst stuck CEA is typically a dual CEA from one of the shutdown CEA groups. This calculation is performed at BOC, high soluble boron and EOC, no soluble boron conditions, since the MDD is strongly dependent upon soluble boron concentration. A power level and boron-concentration-depen ient required shutdown margin is developed, as discussed in Section 4.9.5. For cooldown trandents from HZP, the Cycle 14 MDD's are less than Cycle 13's, which is consistent with the chmges in MTC's. The Cycle 13 MDD results were based on the original physics methods reference report in (9) and utilized two-dimensional models. The Cycle 14 MDD results are based on the advanced nodal methods and utilize three-dimensional models. An uncertainty of 15% is applied to the MDD values in cooldown transients from HZP, which is unchanged from Cycle 13. For the two-dimensional methods applied in Cycle 13 and previous cydes,25% was applied to the MDD values in cooldown transients from HFP, where the additional 10% represented the reactivity l component for moderator redistribution effects. For the advanced nodal methods, the moderator f redistribution effects are explicitly accounted for in the calculations, as witnessed in the larger MDD from HZP to HFP in Table 4.6 for Cyde 14 relative to Cyde 13. As such, an uncertainty l 1 of 15% is applied to the MDD values in cooldown transients from HFP in Cycle 14, which is the same as the uncertainty applied for cooldown transients from HZP. l 4.7 Soluble Boron and Burnable Poison Reactivity Effects The soluble boron and burnable poison shim reactivity effects are shown in Table 4.1 for Cydes 13 and 14. The critical boron concentrations and inverse boron worths for Cycles 13 and 14 are similar. There are more burnable poison pins in Cycle 14 and a comparably larger burnable, l poison reactivity worth at BOC. The additional burnable poison is required for reactivity and MTC control due to the increase in fresh fuel enrichment and the number of fresh assemblies. 4.8 Kinetics Parameters The total delayed neutron fractions and prompt neutron generation times for Cycles 13 and 14 i are presented in Table 4.1. The values are comparable and the differences reflect the effects of { exposure and power weighing. Table 4.7 details the delayed neutron fractions and lifetimes by I delayed neutron group for Cycles 13 and 14 at HFP, All-Rods-Out (ARO) conditions. Kinetics l parameters for HFP and HZP conditions, both unrodded and rodded, are calculated for use in transient analysis cases, as appropriate, and a 10% uncertainty is applied in a conservative manner. These uncertainties are unchanged for Cycle 14. l l l 4 l 1

~ i' 4.9 Safety-Related Characteristics l 2 4.9.1 CEA Group insertinn Limits a The CEA group insertion limits are provided in Figure 4.9 and are referenced by Technical 4 Specifications. The Power Dependent Insertion Limit (PDIL) for CEA's provides for sufficient f Available Scram Reactivity (ASR) at all power levels and times in cycle life. It also specifies the l 7 ) allowable CEA configurations for analysis of dropped, ejected, and withdrawn CEA's. l t The CEA group insertion limits are changed for Cycle 14. Compared to Cycle 13's, the limits l 1 are unchanged at HFP and HZP, and are slightly less restrictive at intermediate power levels. + Within the specified overlap sequence in Figure 4.9, the CEA group insertion limits allow for. I 4 o 14% insertion of Group 5 (155 steps withdrawn) at HFP, to r I j o 100% insertion of Group 5 (0 steps withdrawn) and 40% insertion. 3, of Group 4 (108 steps withdrawn) at HZP. [ The CEA group insertion limits are less restrictive at intermediate power levels due to decreased l core peaking, especially with CEA Group 5 insertion, as discussed in Section 4.3. i i i 4.9.2 CEA Eiection Results 1 i e The calculated worths and planar radial maximum 1-pin powers resulting from the worst ejected i CEA's at BOC and EOC are shown in Table 4 8 for Cycles 13 and 14. No credit is taken for t j feedback effects in these calculations. The calculations assume full insertion of CEA Group 5 at ] HFP and Groups 5+4 at HZP, which are conservative relative to the allowable insertions at these power levels from the CEA insertion limits in Figure 4.9. The range of allowable core inlet t temperatures is examined to determine the worst cases. In general, the Cycle 14 values are l ] higher than Cycle 13's and recent cycles, and are comparable to those witnessed in Cycles 9 and j i 10, as presented in (54). The increases result primarily from the higher fresh fuel enrichment, r j and the correspondingly higher powers in the region of the pre-ejected CEA. The uncertainties t and methods for CEA Ejection, discussed in (13), are unchanged for Cycle 14. i i ] 34-s

4.93 CEA Dron Results and Post-CEA Dron Restrictions The calculated worths of the most limiting dropped CEA's for Cycles 13 and 14, with the resulting maximum 1-pin radial powers, are given in Tables 4.9 and 4.10 for BOC and EOC. Since Cycle 4, this analysis has utilized a local pinwise Doppler feedback methodology, which was verified in (38) by a special at-power CEA drop test performed during the Cycle 4 startup physics tests. The CEA drop analysis and uncertainties are unchanged for Cycle 14. The calculations are performed for all CEA drops at 20% increments in power level. CEA drops l from ARO, Group 5, and Groups 5+4 inserted are considered, with conservatively assumed CEA insertion limits with power level. CEA drops from ARO and Group 5 inserted are the most j important due to the higher power levels permitted in these CEA configurations. The CEA drop results in Tables 4.9 and 4.10 are compared for Doppler feedback conditions of 80% of rated thermal power. The worths and increases in maximum 1-pin peaking due to a dropped CEA are comparable between Cycles 13 and 14. i Detailed separate envelopes of maximum percent increase in radial peaking versus reactivity l worth of the dropped CEA are calculated for various power levels and presented in Figure 4.10. In general, the peaking increases are similar to Cycle 13's, which are also presented in the figure ( s for the 100% power case. In the design analysis for dropped CEA's, the radial peaking increases [ l in Figure 4.10 are combined with the most limiting radial and axial peaking allowed by the symmetric offset limits to obtain total peaking for the given power level. As in Cycle 13, this ] peaking, increased by 10% for uncertainties, is accommodated in the transient analysis. f 1 j Analysis for post-CEA drop operation was first presented in (12) to determine the required rate of power level reduction which the design analysis method, described above, would support. The results indicated that the following actions, which are unchanged for Cycle 14, are required i to maintain the core within the limits of the design analysis following a dropped CEA: 1) Decrease thennal power by at least 10% of rated power within one-half hour, . [

f 2) Decrease thermal power by at least 20% of rated power within one

hour, 3)

Maintain thermal power at or below this reduced power level, and 1 4) Limit CEA insertion to the maximum allowable insertion level corresponding to the pre-drop thermal power. I The power reductions described above assure that proper limits are maintained for operation up to four (4) hours post-drop. The plant Technical Specifications reflect these restrictions. Similar calculations were performed for Cycle 14 to quantify the peaking increases under these power level restrictions. The highest of the CEA drop peaking increases with continued operation are presented in Table 4.11 for Cycles 13 and 14. The peaking increase ;esults for Cycles 13 and 14 { are similar. These peaking increases with continued operation, as well as the instantaneous l ] peaking increases, are accommodated in the licensing analysis for CEA drops. I y i 4.9.4 Available Scram Reactivity i j The Available Scram Reactivity (ASR), from both HFP and HZP conditions at BOC and EOC, is tabulated in Table 4.12. Allowances for the worst stuck CEA and the Power Dependent Insertion Limit (PDIL) for CEA's are included. The CEA programming allowance corresponds d to the loss in ASR due to insertion of all CEA's a maximum of 3 inches (4 steps) from the fully withdrawn position, as permitted by Technical Specifications. Relative to Cycle 13, the ASR with } uncertainties at EOC is reduced for Cycle 14 by 0.39%Ap at HFP and 0.34%Ap at HZP. This is due primarily to the decreased scrammable CEA worths in Cycle 14. The increase in the number { of fresh assemblies in Cycle 14 contributed to the reduced ASR. The four (4) additional fresh assemblies reside in the CEA Subgroup 5B locations, which are non-scrammable. The higher r relative powers in these regions result in lower relative powers near the remaining CEA's, which l are scrammable. As a result, the scrammable CEA worths in Cycle 14 are lower. The required scram reactivity at the HZP condition is determined from the requirements of the l Steam Line Rupture (SLR) analysis in Section 5.5.1 and the other safety analyses in Section 5. l The required scram reactivity at HZP must be sufficient to prevent a return-to-cnticality following the most limiting SLR event from HZP. It also must be greater than assumed in other J 36-I v i

r i safety analyses from HZP. The ASR at HZP, from Table 4.13, must be greater than the required scram reactivity at HZP. In addition, the required scram reactivity at HZP, when added to the additional ASR provided by the CEA insertion limits versus power from Figure 4.9, must be I sufficient to prevent a return-to-criticality following a SLR event from any power level. It must also be greater than the value assumed in other safety analyses from at-power conditions. f r l The SLR analyses are performed from both HFP and HZP conditions, as discussed in Section { 5.5.1. They explicitly account for the moderator defect as a function of moderator density, and Doppler defect as a function of fuel temperature, with the uncertainties stated in Sections 4.5 and l 4.6. The SLR analysis explicitly includes an evaluation of the impact of operating between the Technical Specification minimum and maximum cold leg temperatures of 500 and 551.3'F, f respectively. Other safety analyses are also performed from both HZP and HFP conditions. The CEA insertion limits are designed to provide increased ASR with increased power level. s The SLR analysis provides the minimum required worth in CEA's for cooldown events from HFP and HZP conditions to maintain suberiticality. In addition, other safety analyses have implicitly assumed minimum required worth in CEA's, as stated in Section 5.1.4. The minimum l required scram reactivities are compared, in Table 4.13, to the ASR from Table 4.12. The table demonstrates that, in each condition and time in cycle life, the ASR is greater than the required scram reactivity for nominal HFP and HZP conditions. This is true for the range of allowable core inlet temperatures from 500 to 551.3 F, for both HFP and HZP conditions. The ASR's, with uncertainties, are compared in Table 4.13 to the values assumed in the analyses. A 10% uncertainty component is included in the determination of the minimum required worth in CEA's for the SLR analysis, as part of the Statistical Combination of Uncertainties (SCU), described in (14). These uncertainties and methods are unchanged for Cycle 14. Compliance with the startup test criteria on CEA worths demonstrates the ASR in Table 4.13. As such,it also demonstrates ASR in excess of the required scram reactivities. 1 l e

i 1 4.9.5 Shutdown Margin Recuirements l Shutdown margin (SDM) is defined as the sum of: l l t 1) the reactivity by which the reactor is subcritical in its present condition, and i 2) the reactivity associated with the withdrawn scrammable CEA's, less the reactivity associated with the highest worth withdrawn scrammable CEA. i For a critical reactor, SDM must be maintained by sufficient ASR. The required and available scram reactivity comparison in Table 4.13 is the result of calculations which demonstrate adequate SDM by bounding all the crit: cal operating conditions. Adequate SDM exists, provided the CEA insertion limits and assumptions inherent in them are fulfilled. These assumptions are: l l 1) the ASR calculations, l 2) the operability of all scrammable CEA's, and f r 3) the CEA drop time to 90% of full insertion in less than 2.7 seconds. The minimum required worth in CEA's for the SLR analysis is calculated at typical full-power BOC and EOC conditions, corresponding to soluble boron conditions of 978 and 0 ppm, respectively, for Cycle 14. The boron concentration determines the magnitude of the MDD and has the most direct impact on the minimum required worth in CEA's. The result is that the f minimum required SDM can be expressed as a function of soluble boron concentration in the Reactor Coolant System (RCS). The SDM requirement is shown in Figure 4.12 and referenced in Technical Specifications. The f equation representation in the figure allows for calculation of *.he minimum required SDM for any RCS boron concentration and power level. For Cycle 14, the SDM requirements are decreased by 0.6%Ap at 0 ppm, and are unchanged at high soluble boron concentrations. This darcase at 0 ppm is primarily the result of the decreased MDD with cooldown, as discussed in Section 4.6. i 1

The SDM representation is demonstrated, in Figure 4.11, to bound the required scram reactivities j of Table 4.13 from both HFP and HZP conditions. In addition, Table 4.13 shows that there is sufficient ASR to meet the SDM requirements at these conditions. Based on the discussion in Section 4.9.4, meeting the startup test criteria on CEA worths demonstrates the calculated ASR with uncertainties and, thus, demonstrates compliance with the required SDM. The minimum required SDM is given in Figure 4.12 and referenced in Technical Specifications to provide a well-defined requirement as a function of key plant parameters. Under normal { operating conditions, the CEA insertion limits provide assurance that the requirement is met. In the event of an inoperable or slow CEA, Technical Specifications and applicable procedures assure that the requirement is met. h 4.10 Pressure Vessel Fluence A program for reduction in pressure vessel fluence has been in place since Cycle 7 to address Pressurized Thermal Shock (I'TS) concerns. The Cycles 7 through 14 core designs have been a progression towards lower-leakage loading patterns, with particular emphasis on the high ~ fluence area from 0 to 10* from a perpendicular line to the core shroud flats. The core shroud flats are the core boundary lines defined by assembly locations 1 and 2 (or 45 and 54) in Figure l 3.2. The program for fluence reduction has been detailed in (39,40), with target fluence reductions for Cycles 7,8 and subsequent cycles, relative to the Cycle 6 fluence rate as a reference. The [ Cycle 6 "out-in" fuel management provided relative fluence levels in the 0 to 10" region which were similar to the fluence history accumulated from Cycles 1, I A, and 2 through 5. Given these 4 target fluence reductions, the materials assessment submitted in (58) and updated in (72) concludes that the circumferential weld seam between the middle and lower shells will not reach the PTS screening criteria until well beyond the expiration of the current plant license. The fluence reductions are shown in Table 4.14. The Fluence Reduction Factors (FRF's)in the table are the fluence rate reductions, relative to the Cycle 6 average fast fluence rate (greater then J 1 MeV), at the stated azimuth's peak axial location on the surface of the pressure vessel. The , a

inverse of the FRF is the fraction by which the fluence rate is reduced relative to the Cycle 6 average fluence rate. The target fluence reductions in (39) for Cycles 7 through 14 are compared f to the actual core design fluence reductions, obtained by a Fluence View Factor (FVF) weighting technique of the average quarter-assembly powers for the cycles. The FRF calculations have been improved from those provided previously in (75). The improved f calculations were performed in detail for all cycles, including those prior to Cycle 6. The FRF l calculation improvements include the following-l l Change from vendor-generic to plant-specific FVF's, based on detailed adjoint o neutron transport calculations for Maine Yankee, i t o inclusion of neutron spectrum hardening in the FVF's, which is a function of local t burnup and the resulting higher fraction of plutonium nuclides fissioning, I Normalization of the quarter-assembly, cycle average powers to assembly average o incore measurements for Cycles 1 through 12, and l 1 o Conversion of local powers to neutron source strength, which is a function of local burnup and the resulting changes in fissioning nuclide distribution. The conclusion in Table 4.14 is that the cumulative FRF target to end-of-Cycle 14 has been achieved and exceeded for both the 0 and 10" azimuthal angles. This conclusion is unchanged I from the previous results reported in (75), with the current cumulative FRF's within 2% of the } previously reported factors for both azimuths. At the critical circumferential weld at 0*, the as-designed cumulative FRF due to low-leakage fuel management is 1.29, or a cumulative fluence reduction 13 78% of the Cycle 6 accumulation rate. This is achieved by a Cycle 14 FRF of 1.93, or a cycle fluence rate reduction to 52% of the Cycle 6 accumulation rate. Similar FRF's are f expected for future cycles to meet and exceed the targets set forth in (39). i [ i i ) i ? P

l 9 4.11 Methodoloev and Methodology Revisions A summary of the reference reports and supplemental documentation for the application of j physics methodology to Maine Yankee since Cycle 3 is given in Table 4.15. The reference ^ physics methodology reports are (4,9,66,67). All of the physics analysis for Cycle 14 utilized the advanced nodal methods, which were described in (66,67) and approved for use on Maine Yankee in (68, 69). i 4 l i i

9 I TABLE 4.1 Maine Yankee Cycles 3,13 and 14 I Nuclear Characteristics ~ Cycle 13 ) Cvele 14m Characteristics Cycle 3 ) 0 G Cycle Length at Full Power (mwd /Mt) 10,200 12,300 13,100 Core Average Exposure at BOC (mwd /Mt) 7,000 17,800 - 17,400 E Reactivity Coefficients at ARO (10dAp/'F) Moderator Temperature Coefficient l HFP, BOC -0.34 -0.79 -0.61 f HFP, EOC -1.98 -2.68 -2.42 Fuel Temperature Coefficient HFP, BOC -0.130 -0.128 -0.124 HFP, EOC -0.137 -0.141 -0.142 Kinetics Parameters at ARO Total Delayed Neutron Fraction (py HFP, BOC 0.00611 0.00616 0.00610 HFP, EOC 0.00517 0.00516 0.00517 Prompt Neutron Generation Time (104 seconds) HFP, BOC 29.3 23.6 21.5 l HFP, EOC 32.3 28.3 27.0 Control Elements Assemblies Number of CEA's - Full /Part Length 77/8 81/0") 81/0") l Total Scrammable CEA Worth (%Ap) HFP, BOC 9.18 9.66 9.09 HFP, EOC 9.56 10.59 9.91 Burnable Poison Number of Rods - Fresh / Total 64/756 272/760 400/888 Total Worth at HFP, BOC (%Ap) 1.4 1.6 2.8 Critical Soluble Boron at BOC, ARO (ppm) HZP, No Xenon, Peak Samarium 1,075 1,464 1,434 HFP, Equilibrium Xenon, Peak Samarium 782 1,080 978 Inverse Boron Worths (ppm /%Ap) HZP/HFP, BOC 84/89 109/115 105/111 HZP/HFP, EOC 74/79 88/ 93 83/ 87 i m Cycle 3 nominal conditions: HFP at 2440 MWt,2100 psia,542'F core inlet; HZP at 525*F m Cycle 13 nominal conditions: HFP at 2700 MWt,2250 psia,550"F core inlet; HZP at 532*F m Cycle 14 nominal conditions: HFP at 2700 MWt, 2250 psia,550 F core inlet; HZP at 525'F W Four full-length CEA's (Subgroup 5B) are non-scrammable in Cycles 13 and 14 i

TABLE 4.2 Maine Yankee Cycles 13 and 14 CEA Group Worths at HZP Worths (%Ap)' Cycle 13 Cycle 14 BOC EOC BOC EOC Shutdown CEA Groups Groups C + B + A 6.09 6.27 6.26 6.46 Reculatine CEA Croups Group 5 1.30 1.39

1..6 1.52 Groups 5 + 4 1.60 1.74 1.49 1.79 Groups 5 + 4 + 3 2.45 2.76 2.28 2.73 Groups 5 + 4 + 3 + 2 2.98 3.41 2.82 3.34 Groups 5 + 4 + 3 + 2 + 1 3.89 4.30 3.59 4.26 All CEA Groups Groups 5 + 4 + 3 + 2 + 1

+C+B+A 9.98 10.57 9.85 10.72 Cycle 13 HZP values at 532'F, Cycle 14 HZP values at 525'F l P TABLE 4.3 Maine Yankee Cycles 13 and 14 Core Averace Doppler Defect Doppler Defect ( x 104 Ap ) Fuel Resonance Cycle 13 Cycle 14 Temperature l

  • F BOC EOC BOC EOC 4000 0

0 0 0 3500 41.9 46.2 40.9 42.8 3000 86.3 95.1 84.3 88.3 2500 133.8 147.4 130.9 137.2 2000 185.3 204.0 181.6 1903 1500 241.8 266.2 237.5 248.9 1000 305.8 336.4 300.5 314.7 800 334.3 367.7 327.8 343.0 532 376.1 413.5 525 368.8 385.7 300 416.8 458.2 405.0 424.0 i 200 436.4 479.6 421.1 441.2 100 457.3 502.6 438.4 459.7 68 464.6 510.5 444.3 465.9 i i ;

I TABLE 4.4 Maine Yankee Cycles 13 and 14 Core Averate Doppler Coefficient i Doppler Coefficient ( x 104 Ap/ *F ) i Fuel Resonance Cycle 13 Cycle 14 i Temperature

  • F BOC EOC BOC EOC 100 0.218 0.239 0.179 0.191 200 0.202 0.222 0.173 0.184 300 0.189 0.206 0.167 0.177 l

t 400 0.177 0.194 0.162 0.171 525 0.155 0.1 63 ~ 532 0.166 0.181 800 0.147 0.162 0.142 0.149 1000 0.138 0.151 0.134 0.140 1500 0.120 0.132 0.118 0.123 2000 0.108 0.119 0.106 0.111 2500 0.099 0.109 0.098 0.103 3000 0.092 0.101 0.091 0.096 3500 0.086 0.095 0.085 0.090

i TABLE 4.5 t Maine Yankee Cycles 13 and 14 Moderator Temperature Coefficients Conditions: HFP and HZP, ARO, Critical Boron Concentrations MTC ( x 104 Ap/ 'F ) Cycle 13 Cycle 14 j Conditions BOC EOC BOC EOC i i HFP Equilibrium Xenon, Samarium -0.79 -2.68 -0.61 -2.47 i HZP' No Xenon" +0.11 -1.32 +0.34 -1.06 I i 1 1 i ) Cycle 13 HZP values at 532*F, Cycle 14 HZP values at 527F BOC values at peak samarium, EOC values at equilibrium samarium 46-I a l

i i TABLE 4.6 l Maine Yankee Cveles 13 and 14 I Scrammed Moderator Defect with Worst Stuck CEA Moderator Defect ( x 104 Ap )' Average Cycle 13 Cycle 14 i Moderator Temperature BOC EOC BOC EOC 'F 1200 ppm 0 ppm 978 ppm 0 ppm 576 (HFP) -52.3 -137.1 574 (HFP) -59.3 -163.4 i 532 (HZP) 0.0 0.0 525 (HZP) 0.0 0.0 i 500 10.5 70.4 19.1 53.6 ~ 450 32.9 163.3 52.3 140.7 l 400 60.0 237.3 79.0 204.7 350 88.6 295.5 99.8 250.0 300 115.8 340.5 115.2 281.1 250 138.3 374.8 125.7 302.3 l 200 153.0 401.2 131.8 318.1 150 157.0 422.4 134.0 332.9 f 100 146.9 441.1 132.9 351.2 i 68 131.7 453.0 130.6 366.7 1 Moderator defect at a constant 2250 psia for the specified temperatures

i

i TABLE 4.7 Maine Yankee Cveles 13 and 14 Kinetics Parameters Conditions: HFP, ARO, Critical Boron Cycle 13 Cyc!e 14 Delayed Time in Neutron Effective Lifetime Effective Lifetime 4 d Cycle Life Group Fraction (sec ) Fraction (sec ) BOC 1 0.00018 0.0126 0.00020 0.0128 2 0.00129 0.0305 0.00125 0.0316 ^ 3 0.00117 0.1174 0.00113 0.1220 4 0.00239 0.3146 0.00242 0.3234 5 0.00083 1.1802 0.00089 1.4054 6 0.00029 3.0233 0.00022 3.8709 TOTAL 0.00616 0.00610 EOC 1 0.00014 0.0127 0.00015 0.0128 2 0.00111 0.0304 0.00108 0.0314 i 3 0.00097 0.1202 0.00096 0.1250 4 0.00197 0.3213 0.00200 0.3295 5 0.00072 1.2014 0.00078 1.4143 e 6 0.00024 2.9924 0.00019 3.8153 TOTAL 0.00516 0.00517 i 1 ?

1 I TABLE 4.8 t Maine Yankee Cycles 13 and 14 CEA Eiection Results from Full Insertions Cycle 13 Cycle 14 BOC EOC BOC EOC Maximum 1-Pin Radial Peak HFP Group 5 In, 3.59 4.05 4.18 4.55 Ejected 5 (INCA Location 20) HZP Groups 5 + 4 In, 6.04 6.05 5.88 6.07 Ejected 5 (INCA Location 20) Maximum Eiected Worth (%Ap) i HFP Group 5 In, 0.290 0.376 0.358 0.449 Ejected 5 (INCA Location 20) HZP Groups 5 + 4 In, 0.418 0.543 0.467 0.593 Ejected 5 (INCA Location 20) t l ' i

TABLE 4.9 Maine Yankee Cycles 13 and 14 CEA Drop Results at BOC CEA Group Dropped Dropped CEA Worth Maximum 1-Pin Positions CEA ( %Ap ) Radial Power

  • Before Drop Type Cycle 13 Cvele 14 Cycle 13 Cycle 14 r

ARO A 0.118 0.129 1.76 1.77 ARO B 0.154 0.140 1.72 1.72 ARO C 0.108 0.119 1.75 1.75 ARO 1 0.067 0.067 1.66 1.67 Group 5 In A 0.109 0.124 1.87 1.91 Group 5 In B 0.180 0.152 1.86 1.89 I Group 5 In C 0.110 0.120 1.88 1.91 Group 5 In 1 0.061 0.059 1.82 1.80

  • Pre-Drop Maximum 1-Pin radial powers:

Cvele 13 Cycle 14 ARO 1.541 1.563 Group 5 In 1.727 1.693 Post-Drop Maximum 1-Pin radial power at 80% of full power conditions i

TABLE 4.10 Maine Yankee Cycles 13 and 14 CEA Drop Results at EOC CEA Group Dropped Dropped CEA' Worth Maximum 1-Pin Positions CEA ( %Ap ) Radial Power

  • i Before Drop Type Cvele 13 Cycle 14 Cycle 13 Cvele 14 i

ARO A 0.126 0.131 1.72 1.71 ARO B 0.169 0.154 1.69 1.65 P ARO C 0.109 0.114 1.70 1.68 ARO 1 0.071 0.075 1.63 1.61 Group 5 In A 0.120 0.128 1.86 1.79 Group 5 In B 0.189 0.172 1.85 1.77 Group 5 In C 0.112 0.121 1.85 1.77 Group 5 In 1 0.065 0.072 1.76 1.68. r

  • Pre-Drop Maximum 1-Pin radial powers:

Cycle 13 Cycle 14 i ARO 1.504 1.488 Group 5 In 1.659 1.566 Post-Drop Maximum 1-Pin radial power at 80% of full power conditions -

TABLE 4.11 T Maine Yankee Cveles 13 and 14 Dropped CEA with Power Level Restriction Most Limitine Peakine Cases CEA Drop from Time Maximum Permitted Percent Increase in Power Level Post-Drop Power Level Maximum 1-Pin Peaking (%) (Hours) (%) Cycle 13 Cycle 14 100 0.5 100 12.14 11.97 1.0 90 13.95 14.49 2.0 80 16.65 18.18 3.0 80 18.28 19.86 4.0 80 20.01 21.09 90 0.5 90 12.72 12.74 1.0 80 14.72 15.32 2.0 70 17.64 19.11 3.0 70 19.15 20.72 4.0 70 22.21 21.94 80 0.5 80 13.45 13.50 1.0 70 15.34 16.19 2.0 60 19.03 20.03 3.0 60 21.47 21.56 4.0 60 22.61 22.71 70 0.5 70 14.27 14.45 1.0 60 16.30 17.25 2.0 50 20.51 21.52 3.0 50 23.03 22.98 4.0 50 25.26 24.00 60 0.5 60 14.92 15.44 1.0 50 17.12 17.79 2.0 40 21.83 23.42 3.0 40 24.24 24.62 4.0 40 26.65 25.44 5.

TABLE 4.12 Maine Yankee Cycle 14 i Available Scram Reactivity Worths (%Ap) l l BOC EOC HFP HZP HFP HZP Scrammable CEA Worth") 9.09 8.83 9.91 9.65 Stuck CEA Worth 1.34 1.29 1.56 1.65 C PDIL CEA Worth ) 0.15 1.55 0.21 1.80 CEA Programming Allowance 0.06 0.10 0.06 0.13 Available Scram CEA Worth - Nominal 7.54 5.89 8.08 6.07 G - With Uncertaintles ) 6.79 5.30 7.27 5.46 l 0) Total CEA worth less non-scrammable CEA worth (four Subgroup 5B CEA's) G) PDIL CEA insertion limit for HFP is 14% of Group 5 inserted PDIL CEA insertion limit for HZP is 100% of Group 5 and 40% of Group 4 inserted i G) Uncertainty factor of 0.9 -

r TABLE 4.13 Maine Yankee Cycle 14 Reautred Scram Reactivity T ~ Worth (%Ap) l for Time in Cycle Life and Soluble Bo on Concentration i' BOC EOC 987 ppm 0 ppm Parameter HFP HZP HFP HZP 4 t Minimum Required Worth in CEA's Assumed - Steam Line Rupture Event (Section 5.5.1)* 4.06 2.77 6.58 4.07 - Safety Analyses (Section 5) 5.80 3.80 5.80 3.80 Required Scram Reactivity " 5.80 3.80 6.58 4.07 Available Scram Reactivity with Uncertainties 6.79 5.30 7.27 5.46 (Table 4.12) Shutdown Margin Equation (Figure 4.12) 5.80 3.80 7.00 5.00 t Excess above Required Scram Reactivity - Available Scram Reactivity with Uncertainties 0.99 1.50 0.69 1.39 - Shutdown Margin Equation 0.00 0.00 0.42 0.93 1 An uncertainty factor of 0.9 is applied to the nominal minimum required worth in CEA's i for the steam line mpture event, from Table 5.7, for comparison to the Available Scram j Reactivity with uncertainties. This uncertainty component is statistically combined with the other uncertainty components to derive the nominal minimum required worth in CEA's, as discussed in (14). Maximum of either the minimum required worth in CEA's assumed for the steam line j mpture event or other safety analyses in Section 5. 54 i

t TABLE 4.14 i Maine Yankee Cveles 1 throuch 14 t Relative Pressure Vessel Fluence Comisarisons i Total Effective Fluence Reduction FactorsW at Azimuthal Angle Full-Power from Perpendicular to Core Shroud Flats Years ( EFPY's ) 0* 10" Cycles to EOCm Tarceted Desiened Tarceted Desiened 4 1 0.86 1.21 1.19 1A 1.23 1.21 1.22 2 2.66 1.08 1.08 3 3.59 1.01 1.03 4 4.47 0.95 0.97 i 5 5.37 0.96 0.96 6 6.34 1.00 1.00 1.00 1.00 7 7.36 1.02 1.07 1.28 1.18 i 8 8.38 1.35 1.35 1.51 1.36 9 9.55 1.35 1.41 1.51 1.44 10 10.60 1.35 1.73 1.51 1.61 I 11 11.75 1.35 1.66 1.51 1.56 12 13.04 1.39W 1.72 1.55W 1.57 13 14.22W 1.39 1.89 1.55 1.63 14 15.48W 1.39 1.93 1.55 1.66 Future Cycles 1.39 1.55 Cycles 1-14 CumulativeW 1.17 1.29 1.24 1.27 i Based on 2,700 MWt full power operation m W Inverse of fractional cycle average flux relative to Cyde 6 W Increase in fluence reduction factor target for Cyde 12 and later cydes due to power uprate from 2630 to 2700 MWt W Estimated cycle lengths of 14,000 and 15,000 mwd /Mt for Cydes 13 and 14, respectively m Inverse of the EFPY-weighted fractional cyde average fast fluxes of each cycle 1 TABLE 4.15 Maine Yankee Physics Methodoloey Documentation Supporting Application ) ~ Description of Methodology Documentation Reference in Cycle t i Reference Reports l Reactor Physics Methods YAEC-1115 9,15 3 Reactor Protective System Setpoint Analysis YAEC-1110 4,15 3 Advanced Nodal Cross Section Validation YAEC-1363-A 66,69 13 i Advanced Nodal Methods Validation YAEC-1659-A 67,68 13 Extension of Fine Mesh Diffusion Theory and PC No. 64, 20 4 Nodal Physics Methods to Reactivity Parameter Section 4.8 Calculations and a Change in the Nodal Neutronic WMY 78-102, 38 Coupling Model Attachment B l Introduction of Local Pointwise Doppler Feedback PC No. 64, 20 4 Effects in Two-Dimensional Pinwise Diffusion Section 4.8 { Theory for Dropped CEA's and Special CEA Drop WMY 78-102, 38 Test at 50% Power for Method Verification Attachment C j Uncertainty Applied to Moderator Reactivity Defect YAEC-1259, 43 6 from Hot Zero Power reducM from 25 to 15% Section 4.7 Doppler Defects for CEA Ejections calculated YAEC-1324, 12 7 with explicit Pre-Ejected Local Power Weighing Section 4.10 and Uncertainty reduced from 25 to 15% s CEA Ejections calculated from Partial CEA YAEC-1479, 3 9 Insertions Section 4.11 i Moderator Density Defect for LOCA Analysis YAEC-1479, 3 9 calculated using Fine Mesh Diffusion Theory Section 4.11 i Augmentation Factors Eliminated as a YAEC-1573, 54 10 Power Spike Penalty Section 4.11 6 ; t I

Figure 4.1 Maine Yankee Cycle 14 Assembly Relative Power Densities HFP, ARO, BOC (500 mwd /Mt) Assembly Type and Core Posithn - R-0 1 P-0 2 Assembly Average Relative ower 0.302 0.293 r Assembly Peak Fuel Rod Palative Power 0.577 0.532 Assembly Peak Channel F. elative Power 0.558 0.513 R-0 3 T-0 4 T-4 5 T-8 6 R-4 7 0.408 1.004 1.142 1.071 0.688 0.870 1.442 1.500 1.425 0.892 O Summary of Maximum Powers 0.832 1.389 1.455 1.368 0.875 Power of Value Location R-0 8 T-4 9 S-4 10 S-0 11 S-8 12 T-4 13 Assembly 1.384 31 0.436 1.105 1.208 1.265 1.193 1.380 Fuel Rod 1.529 31 0.865 1.445 1.349 1.436 1.300 1.528 Channel 1.481 31 0.832 1.392 1.304 1.397 1.244 1.469 R-0 14 T-8 15 S-8 16 T-8 17 R8 18 T-8 19 R-8 20 0.432 1.025 1.120 1.382 0.959 1.329 0.928 0.859 1.333 1.260 1.528 1.000 1.487 1.007 0.825 1.274 1.228 1.480 0.991 1.433 0.992 R-0 21 T-4 22 S-8 23 S-0 24 R-8 25 T-4 26 R-4 27 S-4 28 0.408 1.104 1.120 1.128 0.938 1.347 0.877 1.026 0.870 1.445 1.260 1.222 0.998 1.462 0.937 1.129 0.832 1.392 1.228 1.190 0.976 1.408 0.930 1.079 T-0 29 S-4 30 T-8 31 R-8 32 T-8 33 R-0 34 S-0 35 S-4 36 1.004 1.208 1.384

  • 0.940 1.287 0.921 1.113 1.101 1.442 1.349 1.529*

1.001 1.408 1.048 1.218 1.222 1.390 1.305 1.481

  • 0.979 1.361 1.034 1.190 1.158 T-4 37 S-0 38 R-8 39 T-4 40 R-0 41 S-8 42 S-4 43 S-4 44 1.143 1.266 0.961 1.350 0.922 1.081 1.149 1.146 R-0 45 1.501 1.436 1.002 1.465 1.050 1.188 1.265 1.255 0.303 1.456 1.398 0.992 1.410 1.036 1.125 1.211 1.189 0.578 T-8 46 S-8 47 T-8 48 R4 49 S-0 50 S-4 51 R-0 52 T-8 53 0.559 1.071 1.194 1.331 0.879 1.114 1.150 0.936 1.325 p.0 54 1.426 1.301 1.488 0.939 1.220 1.266 0.995 1.454 0.293 1.369 1.245 1.434 0.933 1.192 1.212 0.981 1.407 0.532 R-4 55 T-4 56 R-B 57 S-4 58 S-4 59 S-4 60 T-8 61 M-8 62 0.513 0.688 1.380 0.928 1.02C 1.101 1.146 1.325 0.898 0.892 1.528 1.007 1.129 1.222 1.255 1.454 0.936 0.875 1.469 0.992 1.079 1.158 1.189 1.407 0.907 --

Figure 4.2 Maine Yankee Cycle 14 Assembly Relative Power Densities HFP, ARO, MOC (6,000 mwd /Mt) Assembly Type and Core Position R-0 1 P-0 2 Assembly Average Relative Power 0.328 0.329 Assembly Peak Fuel Rod Relative Power 0.604 0.569 Assembly Peak Channel Relative Power 0.589 0.553 R-0 3 T-0 4 T-4 5 T-8 6 R-4 7 0.420 0.968 1.116 1.101 0.718 0.846 1.365 1.421 1.398 0.890

  • Summary of Maximum Powers 0.814 1.323 1.387' 1.354 0.880 Power of Value Location R-0 8

T-4 9 S-4 10 S-0 11 S-8 12 T-4 13 Assembly 1.382 31 0.466 1.110 1.167 1.210 1.176 1.371 Fuel Rod 1.485 56 0.876 1.415 1.302 1.340 1.277 -1.485 Channel 1.450 31 0.848 1.366 1.233 1.308 1.213 1.442 R-0 14 T-8 15 S-8 16 T-8 17 R-8 18 T-8 19 R-8 ~ 20 0.461 1.072 1.123 1.381 0.950 1.340 0.931 0.870 1.349 1.245 1.482 0.985 1.462. 1.001 0.842 1.297 1.193 1.450 0.970 1.428 0.980-R-0 21 T-4 22 S-8 23 S-0 24 R-8 25 T-4 26 R-4 27 S-4 28 0.420 1.109 1.122 1.124 0.939 1.337 'O.882 1.033 0.846 1.415 1.245 1.211 0.992 1.425 0.938 1.127 0.814 1.366 1.193 1.185 0.965 1.386 0.923 1.071 T-0 29 S-4 30 T-8 31 R-8 32 T-8 33 R-0 34 S-0 35 S-4 36 0.968 1.167 1.382* 0.941 1.308 0.924 1.096 1.088 1.366 1.303 1.483 0.995 1.394 1.026 1.195 1.197 1.323 1.233 1.450* 0.967 1.360 1.017 1.174 1.136 T-4 37 S-0 38 R-8 39 T-4 40 R-0 41 S-8 42 S-4 43 S-4 44 1.117 1.210 0.952 1.338 0.925 1.087 1.146 1.134 R-0 45 1.422 1.340 0.986 1.427 1.027 1.186-1.260 1.238 0.329 1.387 1.308 0.971 1.388 1.019 1.120 1.197 1.177-0.605 T-8 46 S-8 47 T-8 48 R-4 49 S-0 50 S-4 51 R-0 52 T-8 53 0.590 1.101 1.177 1.341 0.884 1.097 1.146 0.946 1.354 P-0 54 1.399 1.279 1.463 0.941 1.196 1.261 0.996 '1.445 0.329 1.355 1.214 1.429 0.925 1.175 1.198 0.985 1.409 0.569 R-4 55 T-4 56 R-8 57 S-4 58 S-4 59 S-4 60 T-8 61 M-B 62 0.552 0.718 1.371 0.931 1.033 1.088 1.134 1.354 0.921 - 0.890 1.485* 1.001 1.127 1.197 1.238 1.445 0.958 0.880 1.442 0.980 1.071 1.136 1.177 1.409 0.927 > l l

1 Figure 4.3 Maine Yankee Cycle 14 Assembly Relative Power Densities HFP, ARO, EOC (15,000 mwd /Mt) l Assembly Type and Core Posit!on R-0 1 P-0 2 Assembly Average Relative Power 0.386 0.403 Assembly Peak Fuel Rod Relative Power 0.679 0.655 Assembly Peak Channel Relative Power 0.666 0.640 R-0 3 T-0 4 T-4 5l T-8 6 R-4 7 0.450 0.947 1.124 1.180 0.783 0.846 1.273 1.368 1.405 0.924 O Summary of Maximum Powers 0.818 1.250 1.336 1.376 0.916 Power of Value Location R-0 8 T-4 9 S-4 10 S-0 11 S-8 12 T-4 13 i Assembly 1.381 31 0.516 1.129 1.i14 1.151 1.147 1.367 Fuel Rod 1.483 56 0.912 1.354 1.213 1.249 1.217 1.483 Channel 1.444 56 0.886 1.328 1.167 1.235 1.174 1.444 R-0 14 T-8 15 S-8 16 T-8 17 R-8 18 T-8 19 R-8 20 0.511 1.141 1.104 1.381 0.947 1.359 0.939 0.905 1.358 1.188 1.458 0.977 1.464 0.997 0.873 1.324 1.150 1.435 0.963 1.431 0.982 R-0 21 T-4 22 S-8 23 S-0 24 R-8 25 T-4 26 R-4 27 S-4 28 0.450 1.128 1.104 1.100 0.945 1.324 0.885 1.013 0.846 1.353 1.188 1.177 0.988 1.435 0.938 1.079 0.818 1.327 1.150 1.159 0.967 1.395 0.933 1.041 T-0 29 S-4 30 T-8 31 R-8 32 T-8 33 R-0 34 S-0 35 S-4 36 0.947 1.114 1.381

  • 0.046 1.335 0.918 1.044 1.029 1.273 1.213 1.458 0.989 1.404 1.022 1.141 1.101 1.250 1.167 1.435 0.968 1.381 1.014 1.129 1.059 T-4 37 S-0 38 R-8 39 T-4 40 R-0 41 S-8 42 S-4 43 S-4 44 1.124 1.151 0.948 1.325 0.918 1.042 1.082 1.069 R-0 45 1.368 1.249 0.978 1.435 1.023 1.105 1.175 1.149 0.386 1.336 1.235 0.964 1.395 1.015 1.063 1.127 1.110 0.679 T-8 46 S-8 47 T-8 48 R-4 49 S-0 50 S-4 51 R-0 52 T-8 53 0.666 1.180 1.148 1.360 0.886 1.044 1.083 0.931 1.361 P-0 54 1.405 1.218 1.464 0.940 1.142 1.175 0.971 1.438 0.403 1.376 1.175 1.431 0.933 1.130 1.127 0.963 1.411 0.654 R-4 55 T-4 56 R-8 57 S-4 58 S-4 59 S-4 60 T-8 61 M-8 62 0.640 0.783 1.367 0.939 1.013 1.029 1.069 1.361 0.938 0.924 1.483*

0.997 1.079 1.101 1.149 1.438 0.970 0.916 1.444

  • 0.982 1.041 1.059 1.110 1.411 0.949 Figure 4.4 Maine Yankee Cycle 14 Assembly Relative Power Densities HFP, Group 5 in, BOC (500 mwd /Mt)

Assembly Type and Core Position R-0 1 P-0 2 Assembly Average Relative Power 0.319 0.287 Assembly Peak Fuel Rod Relative Power 0.608 0.529 Assembly Peak Channel Relative Power 0.576 0.510 l R-0 3 T-0 4 T-4 5 T-8 6 R-4 F 0.441 1.119 1.239 1.046 0.585 0.949 1.598 1.642 1.455 0.703

  • Summary of Maximum Powers 0.907 1.541 1.592 1.406 0.697 l Power of Value Location R-0 8

T-4 9 S-4 10 S-0 11 S-8 12 T-4 10 Assembly 1.431 31 0.358 1.113 1.310 1.353 1.109 0.812 Fuel Rod 1.643 37 0.693 1.493 1.474 1.574 1296 1.023 Channel 1.594 37 0.673 1.453 1.435 1.531 1.243 0.950 R-0 14 T-8 15 S-8 16 T-8 17 R-8 18 T-8 19 R-8 20 0.354 0.580 1.033 1.429 1.011 1.335 0.861 0.688 0.907 1.256 1.620 1.072 1.505 0.930 0.668 0.842 1.241 1.564 1.061 1.442 0.910 R-0 21 T-4 22 S-8 23 S-0 24 R-8 25 T-4 26 R-4 27 S-4 28l 0.441 1.112 1.033 1.036 0.860 1.370 0.947 1.122 l 0.950 1.494 1.257 1.152 0.992 1.520 1.004 1262 0.908 1.453 1.242 1.123 0.981 1.465 0.986 1217, T-0 29 S-4 30 T-8 31 R-8 32 T-8 33 R-0 34 S-0 35 S-4 36' 1.120 1.310 1.431

  • 0.863 0.777 0.893 1.245 1268 1.599 1.475 1.622 0.995 0.981 1.009 1.383 1.416 1.542 1.437 1.565 0.984 0.906 1.005 1.352 1.343 T-4 37 S-0 38 R-8 39 T-4 40 R-0 41 S-8 42 S-4 43 S-4 44' 1.240 1.355 1.013 1.373 0.894 1.154 1.307 1.319 '

R-0 45 1.643

  • 1.575 1.074 1.523 1.011 1.322 1.451 1.446 0.319 1.594
  • 1.533 1.063 1.468 1.008 1.264 1.392 1.365 i

0.609 T-8 46 S-8 47 T-8 48 R-4 49 S-0 50 S-4 51 R-0 52 T-8 53 0.577 1.046 1.110 1.337 0.949 1.247 1.308 1.037 1.396, p.0 54 1.456 1298 1.508 1.008 1.384 1.452 1.129 1.608 ; 0.287 1.407 1.245 1.444 0.989 1.353 1.393 1.114 1.557 ' O.529 R-4 55 T-4 56 R-B 57 S-4 58 S-4 59 S-4 60 T-8 61 M-B 62 0.510 0.585 0.812 0.861 1.122 1.268 1.319 1.396 0.629 0.703 1.023 0.930 1.262 1.416 1.446 1.608 0.755 0.697 0.950 0.910 1.217 1.343 1.365 1.557 0.703 ; Figure 4.G Maine Yankee Cycle 14 Assembly Relative Power Densities HFP, Group 5 in, MOC (6,000 mwd /Mt) Assembly Type and Core Position R-0 1 P-0 2 Assembly Average Relative Power 0.349 0.328 Assembly Peak Fuel Rod Relative Power 0.634 0.570 Assembly Peak Channel Relative Power 0.612 0.554 R-0 3 T-0 4 T-4 5 T-8 6 R-4 7 0.453 1.079 1213 1.081 0.619 0.921 1.516 1.551 1.425 0.718 O Summary of Maximum Powers 0.886 1.469 1.515 1.377 0.713 Power of Value Location R-0 8 T-4 9 S-4 10 S-0 11 S-8 12 T-4 13 Assembly 1.428 31 0.376 1.109 1264 1.298 1.098 0.800 Fuel Rod 1.602 61 0.693 1.455 1.422 1.473 1.276 1.023 Channel 1.565 61 0.678 1.420 1.358 1.436 1.218 0.942 R-0 14 T-8 15 S-8 16 T-8 17 R-8 18 T-8 19 R-8 20 0.372 0.590 1.026 1.427 1.007 1.356 0.874 0.688 0.919 1.228 1.572 1.054 1.484 0.945 0.672 0.845 1.200 1.533 1.040 1.446 0.923 R-0 21 T-4 22 S-8 23 S-0 24 R-8 25 T-4 26 R-4 27 S-4 28 0.453 1.108 1.026 1.026 0.861 1.364 0.960 1.139 0.921 1.455 1228 1.132 0.979 1.491 1.011 1.272 0.886 1.420 1.201 1.111 0.970 1.450 0.990 1.220 l T-0 29 S-4 30 T-8 31 R-8 32 T-8 33 R-0 34 S-0 35 S-4 36 1.080 1.264 1.428* 0.863 0.779 0.898 1.234 1.262 1.517 1.423 1.573 0.982 0.983 1.003 1.352 1.396 1.469 1.359 1.534 0.972 0.903 0.996 1.327 1.327 T-4 37 S-0 38 R-8 39 T-4 40 R-0 41 S-8 42 S-4 43 S-4 44 1.214 1.299 1.009 1.366 0.899 1.165 1.308 1.309 R-0 45 1.552 1.473 1.055 1.493 1.005 1.326 1.450 l 1.427 0.349 1.516 1.437 1.042 1.452 0.998 1.263 1.373 1.352 0.635 T-8 46 S-8 47 T-8 48 R-4 49 S-0 50 S-4 51 R-0 52 T-8 53 0.612 1.081 1.099 1.358 0.963 1.235 1.309 1.048 1.420 P-0 54 1.426 1.278 1.486 1.015 1253 1.451 1.135 1.602 0.328 1.378 1.220 1.448 0.992 1.328 1.374 1.122 1.565 0.569 R-4 55 T-4 56 R-8 57 S-4 58 S-4 59 S-4 60 T-8 61 M-B 62 0.554 0.619 0.800 0.874 1.139 1.262 1.309 1.420 0.626 0.718 1.023 0.945 1272 1.396 1.427 1.602* 0.755 0.713 0.942 0.323 1.220 1.327 1.352 1.565* 0.702 1 , l

Figure 4.6 Maine Yankee Cycle 14 Assembly Relative Power Densities HFP, Group 5 in, EOC (15,000 mwd /Mt) Assembly Type and Core Pos! tion R-0 1 P-0 2 Assembly Average Relative Power 0.419 0.413 Assembly Peak Fuel Rod Relative Power 0.721 0.672 Assembly Peak Channel Relative Power 0.707 0.658 R-0 3 T-0 4 T-4 5 T-B 6 R-4 7 0.488 1.070 1.242 1.182 0.694 0.927 1.433 1.495 1.432 0.774

  • Summary of Maximum Powers 0.896 1.405 1.459 1.391 0.770 Power of Value Location R-0 8

T-4 9 S-4 10 S-0 11 S-8 12 T-4 10 Assembly 1.431 31 0.409 1.124 1217 1250 1.083 0.788 - Fuel Rod 1.561 61 0.710 1.403 1.334 1.386 1.223 0.992 Channel 1.528 61 0.696 1.377 1.288 1.365 1.185 0.911 R-0 14 T-8 15 S-8 16 T-8 17 R-8 18 T-8 19 R-8 20 0.405 0.602 0.998 1.431 1.013 1.389 0.889 0.704 0.890 1.168 1.557 1.047 1.498 0.959 0.690 0.816 1.156 1.520 1.038 1.466 0.946 R-0 21 T-4 22 S-8 23 S-0 24 R-8 25 T-4 26 R-4 27 S-4 28 0.488 1.124 0.998 0.996 0.860 1.352 0.969 1.124 0.928 1.403 1.168 1.085 0.985 1.488 1.014 1220 0.896 1.377 1.156 1.074 0.976 1.446 1.003 1.176 T-0 29 S-4 30 T-8 31 R-8 32 T-8 33 R-0 34 S-0 35 S4 36 1.071 1.217 1.431

  • 0.862 0.768 0.882 1.175 1.196 1.433 1.334 1.558 0.987 0.946 0.987 1251 1.287 1.405 1.288 1.521 0.978 0.868 0.981 1236 1239 T-4 37 S-0 38 R-8 39 T-4 40 R-0 41 S-8 42 S-4 43 S-4 44 1242 1.251 1.014 1.353 0.882 1.107 1231 1230 R-0 45 1.495 1.386 1.048 1.489 0.988 1230 1.333 1.310 0.420 1.459 1.365 1.039 1.447 0.982 1.185 1.278 1.263 0.722 T-B 46 S-8 47 T-8 48 R-4 49 S-0 -50 S-4 51 R-0 52 T-8 53 0.708 1.183 1.083 1.390 0.970 1.176 1232

'1.022 1.404 p.0 54 1.432 1224 1.499-1.017 1.252 1.333 1.087 1.561 0.413 1.392 1.186 1.466 1.004 1.236 1278 1.077 1.528 0.672 R-4 55 T-4 56 R-8 57 S-4 58 S-4 59 S-4 60 T-8 61 M-8 62 0.658 0.694 0.788 0.889 1.124 1.196 1.230 1.404 0.612 0.774 0.992 0.959 1.220 1287 1.310 1.561

  • 0.754.

0.770 0.911 0.946 1.176 1.239 1263 1.528* 0.698 i t

1 NOTE: 1. THIS CURVE INCLUDES 10% CALCULATIONAL UNCERTAINTY

2. F; - Fl x 1.03
3. MEASURED F; SHOULD BE AUGMENTED BY MEASUREMENT i

I UNCERTAINTY (8%) BEFORE COMPARISON TO THIS CURVE. i 1.82 2 ..t.-..a. 4 ...e 3.r..r,;r ** - -F + - + + - + + - -e t---+- - + - -- 4---t --t

  • - + - + -e-i--+ -r r*

4;.... 1.81 -...... p...p..n.[.

4..p.._.pq.{.; i...,_ 4. +.
j. 4.+...l..a.[..4. p..

..:.,.4 - 4..j L.p.a...i..i.,+ t .;.t . 4.+ ... j... .. +...,. p.

p.... +

- * + --i*+. -

  • i-F* * -

- + " - . ; +J -.t 2 9... +- - T' -. -*9,. 4 + l - 4*+*- ..-+j 4 ' -*,4.*,-- +- l.44.+. } * -+4- '4.

      • ..p.-...4,44

' *&4..4 +- ..-4..a,.4,.. .f. l' .l. 44._. .... 4 a... 4 4...... 4. 4. .. y 4_ 4 5.p+ .a ).. 9,.. p....- ..p.-. 4 .+,.4...,

4. 4.a.4 p.p.4

.+.;., .p. i l..,.. .I, ..p 4... a.4 ..4... p+,.4 4.+.p 4. +a.... 4 4 4.. 44.-+. 4a...... .p..... .a 1.80 .1...q.,.$+ . 9. .g.., -.p+,9..t.+p9 ..p 4 .. + +. . 4,. o++g COORDINATES (KMWD/MT, F',) -+47.,.*.p .p.,.. + H" + "' 4..,.... ..... -,..+ ., +. . +. .u 9_...e.. 1.79 ' : +:. (0.00, 1.732) (0.50, 1.732)

+.j......,.,.&.

. p.... J.t..p....,...p 4.p.p... -..4...,...9 .. +... +. + a. (1.00, 1.729) (2.00, 1.719) a..,.. a :. 4.a.4 . :...9 a..,i.. ....! {. a.j.. 4 4 4..,; ...x .g....p.- ..i.,4.e 1.78 w (4.00, 1.698) (6.00, 1.683) ..q..., ....a 4... + q.4 e... 4 4 ,.. 4 . +.,s. .r.a. T. i 4 . 4 -. 4....i (8.00, 1.673) (10.00,1.677)

4. H. 4,....i+ i,.

.!.1. i,..... F..i..i...- ..=.:..:.. ,i. i , 1 77 (12.00,1.684) (14.00,1.680) -4 _4; 4. ..g..... w p. 4.+.. (16.00,1.680) ..,.3.. .4.... 4 p.. t -+a+4, +.4+m ..q. ...9.. .p.

4. o.

. 4.;. 4-Lu 1.76 ~ ~ . 3 9,. ..4.e,,.....i.e....,,...s.+i, ~~ y ..p 4., ....c.... ..g u.9.. 4. 4...,.. ..;,.... 9..1 3.. .g g. t,. ..p.., + + + ..,.......a.... ..,. +,..,.va 9 44.+. .j. 9.. q443 <. 4 4., 94 4.. g..44 .l. 4 4 -+g.4 9 4.p. 4 ..,.y.. Q. ..j 4, .a.+, e

3..

-4 4.a.+.

4...

h 1.75 ....g .4...,.. 4.,,. a.a. . +.. ....,.....g._..,. 4 f.+,..4,.. .p.,,.}... 4... 4pl 4,

q v4 6 }

441. 3.f.g., f.. .. j.. u 9.- 3 ..g.y. p,.,. ..9.. . p.,9.......p. 9 ..g p. .7.g p.i. 9 a j.. i4-4.;g.... 4-41 ....,4,.. .9...,. .9 9,. .,4 O 4 ..p,.. i 1... .+ ...*t,- {4* -.' *4-- 4 ai, y 4 44 4..,

4.i a 1.74 w 9. 2 y.+.9
4..,9.

. 9,9 + 9, .p. 9 + 9.y.... yq.. 9.,..j..,....q... ..t., 444., 9. q.. ,9 +. 9,., a(, .9. + ..j.y. 4a..g..... 1..j - 9. +.j. -.,4.,.4 ...g4 ..(..;..p.. 9,.4.4,..[..+4. 4.. l.4 44 4 4.., ..;.. 4. .g. 4...+4. .g. 4 4 .f 44 4..! j. 9,. 49, 4... }.. f...,. 2 1.73 ( ..p. 9. 9

9...

.e ....s. 4 g.....,...........,.,.. + .. a... .. 9 .x ..9. . 4 9.. .a..;...4e I .+4.... q

4.,. 4 5.,.,.

p, p.. r !._9 ?..!.t. 9. +.. A +,}.i..*.t i.. l { 1 *+ ..+t.e { !,e4 y ..+.,9...... j1.....+t.t9... 4....a_ .i.. +3..t 9...+... y 4.,.9..g... + 4... a. ...i.1.e. ..{ ...++,=.9....e.. i.1 ) <. i.9 q..- i.> t .,9.t.... 4 ..l. 9.e 94. 9 +.5.+. ,...;.... + +...... 4. -.,4 4..p 4 + 7..

9. a,.p, j.

pp .p.. 9 +4 s 4 . +.. ......,9 X 1.72 ... 9,9,a.4,. 9. 49...+.q...,..4, .g. 9.e...7... ,..,.+9 ..+ 9 9 3,..,.,..{.+. .., a. 3... .t +.p 4 .p,. 49,a.. 4.m ..a...

9. +

...~.p.p ,. 9 4.. ,4 4.,. 4..... +. ..p...,..

9. 4..

., 9 9.,.. y., 9 4...{... 4.q ..,.4., ..,.f.,.,..+.;.., - +,.. + j.... 4...., 1.,.j.... :. 3.,. 9af, ..,. pq, e+;j.,.. 9.q a. 9.+.. 7....9 4 .,4 4 u ... pg, a 9, ..{..,.,- ~ i ].. 1.71 .+.p..... .p ...... +... .;m. 4., 9... 9.. 9 .p 9... t, ... +. 9. 9.,..., 4.al..... 9...e +....., .j 9 a.y..,., 4 p.,., 9 9 .p 4..,..,., +e+ .p.. +4...... a..;. 9..... + +... 4 4.pp, 9 4 p....p 4+. 9. 4 .#, q. +. p. +. H. q 4 +...~ pt.+. 4.. .... 444',... +;.p. . 4.-+;. 4 4., ..,. 4

q...f44

. 4. 4..p. 4. .g4 4....p 4... 99 4..[+4 4... 1.70 3 .. +,.. ..+., .. y ,n y. ..l49.. e.p 4,. ,..p,. ..n .aj, ..,. 9,.,..p - u4 + ;... +.p 4....g ..,+.e,.4. q-l 44. ..m.y m 9.e..;. +e.+. + +. ..j -{ 4..p..ll e.l.i. ..,..{,.+.9.- .+ a e.. i t I+ ((+! ..p.i,..i. ; j i 4, t..i,..t ... 17.....:.4..,...i..+...l.. 9., ..,...t.p. 4 4. 4 i,-..i. t..f,. 9.t y.f,....Fi.4 4 Fi t a ,.9.i...l..- s..p,.~... p 4p 1.69 p m.p...... - -y a.

j. 4.p 4..+ _

9..++.. .,9 .p 4.. ;9.a . 9,, ..+.,4, ..i.. .,. 4 .;. +.. . +.9....9 4+..y ...9...q..

4...p..... +a

.l. ,.e+. 4.;..., 4.4-,

9..+ 4, 9..q mg...g 94..4

..4,..p 94 ..4.,- q,, p3...,,., 4 4,.+, ,. +.. {3 +, +,. .pq...... a,,.w+4., ..,q. 94 .i... 4 1.68 a

m...,,. y

..,,..,.43. ,,.,3 g3 - - 3 ., 4..y g...g...

4. :,

yq.. - , g. i gg 3..,g 3.g,, .a ....,.. + p .i.. .4... 9..,

3. 4.v..

1 67 .. 44.j..;...j 4.44 4.,u p.p. 4...q.... p 3..,.p.,....4.,..p. 4 ..g. .. 4.,..p'{,.... 44 4 +pp,4494.. .,4..p. g4a ..g p. .4 .q. 4.e a. 3 44 .p......q,.- + p...p..,..p., . a..g 4 a. .,...a....

2..p.. 4..p.

....44...4.... . 3.,.4 ...- ' + + - t 6+-! f.t,.+t -..+q..p..F..a*.,- .+.p... 4..+; ~ ,. ;.+. +pl u.4 . 4 4... a+ +:.,. 4. 4 +4.,, . 4,., ...,9 m[ 4 ..,H. .+ y4 4p + ++ y. ++.),.. 4 - + v.+ ..p,4 ..a.... p,. a ..,. m..-. 4 y. y4

4 41

+o a4 m. .u+ ..u a un ..p.4 .m e 1.66 n u u n u s 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 CYCLE AVERAGE EXPOSURE (KMWD/MT) i i 5 Allowable Unrodded Radial Peak MAINE YANKEE Figure Versus Cycle 14 4.7 Cycle Average Burnup 63-

a l 06~ 4 !...;. 4....l...4 i. .a. 4..; {... 4...;. 7..,

.... 4,

L..i.4. 4 .. 4.4 ;...,4. 4.J..a.., j..}: 4..

.I t
1:

i ... 3.,i t *....'..+{..t,.4....,.D,..*..i..' '..t...i...**.1.p....n.a..y,.*.'. i: -- { i 3 g..q..,. .*4 M... f....+...y+..#..'. f....T *t... -[-- M-M .._g.4.. ..... +.... 4..{. !. 4....j.j. 4 4.....j 4 qp. 9... p ..l. 4. 4. 4.. . 4 4.. -.4..&..[ p-..y..p..j. 4 ..[ j. 4 4.. 4..j..p..p.. j..

  • 1
4. 4..i...i....&..i..

..J...i. 4 4 ..i. 4 .J J 4 ..p.p.4. 4.. .J. ;. 4...p.. '.t,! - F..! UNACCEPTABLE OPERATION 1 1., .t.....

1...

...t... g : ..a. . 4. -i.4. i.... j...p.4.. j..q.. ..b..f i-j.... 4...;.J. p.,, 4. ; p.; . t !.4,......; ; ; ;

4. ;. 4.. p.,,;

.p. ..p.;. 4 4.....q p..j.... 4..j...j...).J .j 4 a.......).4 ;.....p.g..j...g..;..j p., j. j. 4.., j. g 4 4....p.4 4 p., -f + d.+.,*.., ......If.f.- ~ 4- + 1.. ..4...+.!

,..t. -+{ +.t,.t.-t- -+t....~~

.r-+-+-t.-- **: ". 4*t 4-'- + '. -+ - + - I.,:.. .,.,3.+... l ....+.y. 9....q..q........,.......8.. p r....1 }..1... ...t...:.4 9, ,! g q. # I.. .a...,3

r...

I g . 5 9 p4...,.. t-POWER MIQ ...,.......~...9.,. ..,...a., g m.,. s. , I..q., ..,.8......4.4..,a..,. 3 ...i.j.....e... 3. 9 ,t..a... ^ t a..,

i

..:...e .a. .,s a.* ),..

B >....

. 4_9... y Q

e f

,..,..p.4.. ..,. 9, g .l.....,..{..,..f' ' t**i^- - --t" I g gg

*
g i t j :

I t }p..9.. . g...- -9..p C. .....e.-;..,.....;..; 54. _ .1...>. 4.. 2..1.a. t_i. 4, a,..t. p 5..i.4. 4,....t...p.4 ;....: 4 u. w ..I..,t...!.. . 9..,*...L.1........t.. ..t....t.........I ....i. 4.. 100 0.12 ...4.... ...;....,.......g...............q...t.... ..a.....g ;...,....:.... . I..,j ....q......' _,.4

9..

O c ..i. :,.4.4 3 4.. O W ^ L.?... ?.. L. 4.. 4. i..... 4. ?..J...J..,..L.. :. 4... .J...,i. 4.. 4... . 4. 4.._.. ..?...LJ.. 4...,..i......4..4,.. ..;.4.. 4...L,.. g . 3. 3 ) ..'. 4.4 j... 4. 4..l, . 4..i.......h....L.4.....

4. 4..i...

. 4. ;..J. 4 ..p /. 4. 4 .. 4. 4.J....4....,...i.., 4 j J 4.. .E O, ..:p.....+. ....4.y.. . 9.,: 1... 2.. ...b...... .......,t .q,.. 4..,e .....,...4. .p .a..a.....+.. .,.4 y ......a ...,2.a. .a........... a..q . 9.. .a.. .a.,....a..a..,... ,..q a.

., 9...

r-g i e... .....,.:....{......;...,.., .,. 9 ..p.. .......p..i..,....p.9. i. ...q.,. 9 4 . 9.. C < 0.3 .. 9.g .q.a,.:... .. 9. .. p......,7.... .. 7..j j.. p...p. j... j y p p... t ,,,j j 1....., 4.......... q.. ..:..a.., .a... ..p..,....I.., ,i E L,J ~......... .9...v....

...4..

......,.....,.....t....,i..,:...c .4.. 4 r g

1. 9

..a.. 1 o j...s 9 i .. 9. 9 ..9.... 9...- .9... 9.. :..7... ;,...p9. 9 ..p ,t..a g:

9. 9..g

.9... y H b,e 44

4.....

4 4._ 4..:._.

4...a....4....

.;. 4. 4..; 4.4..L.4...,. 4.J. 4j a ...4..._- 4._..,.. ..,...,....a.......i..% ,..J..J. 4. 4.. .....J... C i ..i ..i...i....,.

i i

O O .9...- 9..'... t.......'....=...... '.' '. + 'a'... a................ 9.......' 9. 9..i .. +.., '... i i I g o,,,, .. 9.. r- < l ..a...,.9...7....4... .4...... 9. .p.. C ...4.... ..... 4. ...p.4,.. ..... 4. 4

4. 4...;.

...a....... a..a..,..... 4 b..l .a.., 4 ..i.q..q.4.... 4.4 .4.4 4.4.y. ....a... 4 -.i...:...i....,.:.9...a.., Lj

........3....:........

-.......I... .,..,.a....

j O

1., i... ..(. : .4..,- y.9...... 9.. -.....9.. ......,........,...,?... i: i:. O i :

i
. 9. 9

,9..,... . 9. 9.. .g.....,s... p..p.9 ..j..q. 9. 9 .....p. I .4 .t,...r, r..,..T.., r...J', L," T.., t..... +.. '."..t..r.... ACCEPTABLE OPERATION t .,s.. y ..4.,.- a.... ..p.... 9. 9 7 ........i..., ....i...p,., .p.9.... 1.. 9 9..-. 9 9..s. p..

... i p.,.

..pq.....a. 9. - ..i..,.+... ..,.....p.

y.,....,..

4 . +... .y. 9... p.... O.1 ...p.4... ..4.....

4...:

4.; .4.. j...i.4 4 ..,b.L.a,..; ..,i...:.... 4..&.4 4 .4..L.; - 4444....l.4.4..i... .. 4 4..,T. .. L.:...,:...;.. ...i...i......... ...E.+;...: .4..5... ...a...... .......a. 4... I ) .4..1..1 ..4...e..r ....,.?.. .I..; y; .I.

;i 1

g ...,3... ...p.q.a.. ,4 ...?....:. .-..6.....1.. 9..g...,. 9.. ...i.j......s....a.,:...i..1... ....s...... ........3.. 9 4..a.;......e....... p., g 3... p ..i t t

; 3 1

1 p.9. 9-

g. 9. 9..p.
9. 9....a...

1 9 .. i. 4.... 4 .. 4. 4..; 4 4 #.. 4 4..,, p.. ..+..p,..... 4... . 4.-

. e.4 4...q,4 4.J

. 4.... i... . 4,.,i J '.. ..b...b.4. 4 . 4.. 4... p $a ..,1.......- I : : . 3 : : t :...I 4.. ..+,......1.. ..,I.l..a i a.I..t...9...!.......3.......- .y.a. t, 4 8 p.,.

i 3 *
...,.....e...
: l l

. f : ..+.4..a......a..a.q'...4..a..:..,'.....:.....,..&... -,s..,t..i.. .a..,t .. p... 9. 9..

9. 9 g...

....g-..7.,. -.g.......!.. 8 .......,..i... . q,.a... 9.. .. 2.q: :. _p4.a a.......j.... 9. -.7.. i :, i : : .i t :.. g 4... e.. 94.. ... 9. 4. p,.3... 0.0 ) 0 10 20 30 40 50 60 70 80 90 100 POWER LEVEL (% OF RATED POWER) b 1 .i 1 L 2 1 Moderator Temperature Coefficient Upper Limits MAINE YANKEE Figure Versus j Cycle 14 4.8 Power Level l 1 i II e j,

M A CN 2y,,_gO2 gwg,, d>wa y g gwo o3wC . O b$9g yg.,. Qm_ I j y E cl 8 C eY ',.ig1: q :.,

11

. 44 !!j .. 4K4 . 444. 4444

7. jq i.

7. .p y41; p 44.j4 yt. ' pj 'yi,.7: -r!: r A 4.+7 j t. .j44 1 H4!54 4 j. N j

i,.

,g 4j .pH y'4 ,. 4 4 q4 1 l41h. K 0 E 8i{ . kC Ow( 3 r"Oc<5Q. y+ py 44 .444 f. 9 .+ ] .sg L E 4i - #y $ gc031m.,.mp Lm-5 4 a 4. d.. j4?.4 a j,{4 .py{. 14 p; yj44 .4 p. !.44 ..4 !t l: [ m t' 0 N e . 4 . jj 7 liL. 7.i 4j;4 p y 11 4 ii.l ri 7 - q411 Sjaj !4j 14 441. 4444 j4j. 4 ;p 4 +; 8i ,8 ,8 mo co *mo,oe f ( 8 ,5,oo oo,Mo,eo .i p144

7. }.44

] . 4 ;.j Nh. d yi 7 444 y}. 4 f I.ll 41 4i 1i 1 s. g'4 }. 7yj 4;A _}44 1 4 h 4 p 4. 44;.. y41: !4j 8 r3 8 ,8,eo mo vMo voo ,F g} 1t. f 8j} } ..y.y 44.j44. 4444 3 Pi M Re s ,N voo mo 'mo,mo - w.. 41 j444 i y...- O h 7-4 p. 4.j ?.4 4 [i. ]4 8 ,O.mo mo oo,mo 3F p J. 44 4.44 ri

7. F j

.i o 4 0 ii ,8,oo mo,'eo,oo . - g.. 1 y .q4 7jj4 p,gy.... 14_ j1.- i:

[44 d

y444 .} i+ j J. 3 8i 8 yi j 8 veo mo "oo,eo ...41 !jl 8 P !4{ .;g 14j4 , q. 1 .;y ,. - 44 N r; 4. 4l44 4 i o $ vmo mo "eo,oo 8 . h. 9; 1.-. 14 4. 44... 4 i 4. {. w U ri 4}1

F y.4 e

l 8 D,=6.m mo voo,eo

7. 4 7j44 p{.p{ f!h5

)lf' T .c.. 41 p4L. 4 i. 4 J y -l, . 1 r 3 y+

7. y44 D

0 8 S 8 vE -mo,oo,oo . M.. h p +- ,5

l 7jj4 4 p,.

. 44 ! +f 4 j;. i. 4 e i I..11 4_ 1s } oe mo vmo,oo 3 1

7. i 7j44

,3 .4 i1 J f e Fi h ir.. 7 7.4 7 }..,j .qj .,. j 6 o n .j . { 11 .j41F ( .. 44_ y44. 7j44 5 r e !4l

4. }..

P d 3 rt;. { 7i,.4 y i,., +Tr+, +p. 41 C D A % n j 11 . 4a. _ yT'3st; y3y+ 7444 y. q44 t E I t .~. 4f s' ni 3 f+. A )L qf,q 'o. 4 f.! .}. 44L q l' l !1ojj tti+ .ti.gy.+ 7.4. 44 p.44 p j q f. .{44 7 T' {. :. A44 444. j444 s W trpj gipj 3,, i 7 1.1 e Ty

. A.44 j44.

y444 3 r r {..' f. +g...y:yty,

y j i.

!l.J. 4 r ti M 8 'p{. Q.. o + .j. ytiL'~y7'1L~",' y 4 ?. T{.-y j L "c".. . 1... l. n O .gpp 4444 44. ..4 yi,. !4} . 4 .y j. . 4 A. 7. 4., j4.Q p. L d Pry. l .i*. ..l .g

y1I'T,

.g.l - 4 . 44 . if yt. r j-4. +}. p784t, 777

  • i Pt;{

pi, { .q. L. 11.4L.1.u 1 m l ti 8 gt.. y a4 tpj4 j. 1;.' [d . P ~ -.- 3t" g, . ~ ? {. 3J.j 8 4 ty 4.4, 444.. j444 tt y 7rgj:3. h'., 1 t[' 77 .. 41 5 p.,1y i.

4. j _ 4. 4. 4.. p. j 4. 4 Pt3j. 9. fd...

pt' 8. t', p.jyi

7. j44 T"j44 4

.}-j j l y-.. ,fg.i! iL 4L.1. pp}i. q jj ~ -. 4 p.ypt gj44 ,3 h p. {. t. 9 <oowg % m. $4ooW(<C-DJW 7 1 WC<s OZ 1.. y".r _ 444?. tr.j Q1Wg<F{ I'. 7 4

4. J.*.}

r y4 .ti gp7i

3. y.j T 4 f.

.[i 74 j4 ++ lj1 .\\ 4j_ 44 ;..

.f.i. i _ [

4,_4pj4 rt74 ;. T7. j. yt77: ,il! y: yp. t ~.. 2 i 4 J. 2 8I ..s. q ) C Ae 5 E 0MR _$ $o ~ "8 -8

  • 8 4

g .g ~g o 8 ~g ~8 8 F ogiLN AT 4gi 2 9u re a g "8 8 8 o S Oo "g O f3gm 0CO .lj 1lili I l.l,

Figure 4.10 Maine Yankee Cycles 13 and 14 Maximum Peaking vs. Dropped CEA Worth from Specified Power Levels d 1 i t a f i a ii1 a o 1 i i I t i e I a iI I i t t I ! I 46 - 44 : Powe 42 - cycle 14 Envelopes Points of intersection Leve: 40 - 0,35.9291 (%) ~ 6,23.481' i 38 -- 4*18.338'l 5,16.256'l 0.128,14.5835 ^ 36 - 0.132,13.365[ ~ 0 34 - c) u O O 32 - ~ .c i .E

c. 28 -

u c) 3: 26 - O 0 ~ 20t E 3 2 .E_ 22 - X 20 - (U2 40i ~ C 5gg j h 60 : 14 - joo g .E 12 - 10 - 8-l 6- -- Cyd 12 at 100% Power 4 - Cycle 14 2-0 0.00 0.02 0.04 0.06 0.08 0.10 0.12 0.14 0.16 Dropped CEA Worth (% Ap) Figure 4.11 Maine Yankee Cycle 14 Shutdown. Margin Equation and Required Scram Reactivity 8.0 7.5. { 7.0 - 6.5{.. ' HFP, P = 100% 6.0 - p...........b,, c 5.5 - 5.0 - 3 5; 4.5 - 5 HZP, P = 0% C0 c3 D 1 4.0 ~^.... 'E h....'.%,,.... 8 $3.5-c. 3.0 - ....g 2.5. 2.0 - I Legend 1.5.' Shutdown Margin Equation Required Scram Reactivity (Table 4.13) G"- ~~O Steam Line Rupture Event 1.0 - e......e Other Safety Analyses 0.5 - i 0.0 0 100 200 300 400 500 600 700 800 900 1000 1100 1200 1300 1400 1500 1600 1700 1800 RCS Soluble Boron Concentration (ppm) ' i

l l l 8 2.. j.. . 4..[. [. .. i....!... j.. ...4... ..j... .. [.. j. _4.. l. .. ; l. q...j..j... j.

. p. i.. !.

i;! i i i i

i

. {'. 7.,i-9g. 9 7 9....9.. 3 ....i..t....-,.. .I. _,i..... 4 - ....i... ;i........i.. ....i...,8......,t.. .........,!....l.. 9 l .......,i.....i.. ... +.......... 4..p....+. .4...,........ ! l l ! ! ;..i..!.

i..i i_.._.!
..i.i _

.!.!.; i.._.i _..i.i _ ;..._.i.i.i.i..!iii _.i i. i ! 7_ . i 'I i i i i ! I i! it i! ' I ! t .I I ! ! I ii! i! l 1 i. i. .. i. ;.. i.. ...t. - ...[.....j......,i.... :..,i. 3.. .. i. i 1 . 9.. .4 i I...i..g . )...

4. -

4 = %......t i i i....,.. i l.....1.. 6 p. joo i ;, ; .l....,..,,.... i..;.g y . i ! 1 1 i ! ! i 1 8 ..f -+ ? p SQ . : - I ~ 7 p t 3 f r 8 8 i t i l 1 i 3 g l d. 4-it e-(f.. 4.. !-....i. -.i..1.. 4.J.?_.d ..!.i...j. 4.. - i i i P = 60 i - .t 4_I i.. 4 iii i iii i .._ _1. 4...[... i_4. I i i i gg i.!. i... ......p_z i_{ -. { i_1 p. { }.;. .i {.{ +~ P = 40 4"-f- ~f-f-h-f-2 --i

  • 1, j.;.

. i.. q;.......... ..ig.j.j. *.

i..

1 i. i 3 i i.. {...i O [..i.., .. j.....i...l-i g. - > i - p a i i c: 4 2 . i i

1. ",-

i i r i 1, i 3 P= 0 z ...i.....i.. . \\.... i i > i ii' ' i i i 0 r- --r--- 1 o l. . j. !.. !,. I. !.ji. !...i 1,. I i - ...-.....:........&...t...... i i! - i - - i - - i! - ii' i i i i ! i i - i -

  • 1 m3

,.. :... !..1 .J.i.Li.. a ...!...i. LJ.. .J...:... i...L. 4...i.. t..;.. _i.. :... ..t. ...;_.....t..

i
ii

- i. .. i....!...;. ...i.. :. 3.. i 2: i i.

i. i

..i.. i _.3....i...i. w i _,.4.. .,...e ~ i ' i a i m._ O j .3 .. j.. cr 2 SDM - 5.00 - 0.00300 C + 0.0200 P i I> when C is less than 400 PPM Y-1 e ...j..' i SDM = 3.80 + 0.0200 P ...j. when C is greater than or equal to 400 PPM ,i - 1 2 1 where 3= 4-f+1-SDM is the required shutdown rnargin in percent reactivity -H[-N: f "t"i. ~..*~t " 1 1. i Q. : C is the RCS boron concentration in PPM . g.. .j , Lj. P is the power level in percent of rated power 0 i 0 200 400 600 800 1000 1200 1400 1600 1800 I ACTUAL RCS BORON CONCENTRATION (PPM) Required Shutdown Margin MAINE YANKEE Figure Versus C cle 14 4*12 Y RCS Boron Concentration 68 \\

5.0 SAFETY ANALYSIS 5.1 General l r A review of the safety analysis for operation of Maine Yankee during Cycle 14 is presented in this section. The parameters which influence the results of the safety analysis are listed in Table 5.2. Values are provided for the Reference Safety Analysis, Cycle 13 and Cycle 14. The ~ Reference Safety Analysis for Maine Yankee consists of the Cycle 3 power uprate analysis to { ) 2630 MWt (36), the Cycle 11 power uprate analysis to 2700 MWt (62), or any safety analysis 4 completely reanalyzed since the power uprate. Table 5.4 lists the Reference Safety Analysis for j each event. j j The safety parameters may be divided as follows: 1) initial operating conditions,2) core power i distributions,3) reactivity coefficients,4) shutdown CEA characteristics, and 5) Reactor Protective System (RPS) setpoints and time delays. A discussion of the differences between Cycle 13, Cycle i 14, and the Reference Safety Analysis values for the parameters listed above is contained in l: Sections 5.1.1 through 5.1.5. i t For Cycle 14, several changes in the safety analysis methods were implemented. The methodology changes affecting the safety analysis are the use of the FROSSTEY-2 methodology s i in the fuel thennal performance analysis, the use of the Statistical Combination of Uncertainties i methodology to derive the RPS setpoints and LCOs, and the use of the RELAP5YA computer j program in the small break LOCA analysis. Section 5.6 summarizes all of the safety analysis [ methodologies approved for Maine Yankee and the chronology of their incorporation into the j safety analysis. The impact on the safety analysis resulting from the incorporation of these three l t methodology changes are described as follows. 1 1 0 The incorporation of the FROSSTEY-2 methodology (74) in the fuel performance evaluation l introduced a slight reduction in the peak Linear Heat Generation Rate (LHGR) corresponding j to the fuel centerline melt (FCM) SAFDL as a function of local exposure (Figure 3.6) as compared to equivalent calculations for previous cycles using the CAPEX methodology. Relative to the i previous methodology, this is more conservative. ! l

For Cycle 14, the use of the Statistical Combination of Uncertainties methodology provided significant margin improvement to the Specified Acceptable Fuel Design Limits (SAFDI2) for Minimum Departure from Nucleate Boiling Ratio (MDNBR) and fuel centerline melt. The Cycle 14 safety analysis continues to evaluate the MDNBR using the YAEC-1 CHF correlation (44) which was approved for Maine Yankee in (45) with a SAFDL of 1.20. However, for Cycle 14, this limit is statistically combined with the fuel physical uncertainties and penalties, analytical uncertainties, and plant hardware uncertainties to create a statistical DNBR limit (SDL). A l similar statistical combination of these uncertainties has been incorporated into the safety analysis evaluations of the FCM SAFDL. This statistical combination of t neertainties (SCU) methodology has been submitted and approved for Maine Yankee (77-79, The statistically combined uncertainties, probability distributions and final SDLs are provided.' (80). For Cycle 14, the final SDLs are listed in Table 5.1. j 4 The Small Break LOCA (SBLOCA) was re-analyzed during Cycle 13 using the RELAP5YA computer program, which was approved by the NRC for SBLOCA analysis (83). The analysis also included sensitivity studies to address Cycle 14 specific input parameters. The results of the analysis continued to support the conclusion of previous cycles, that the Large Break LOCA is the limiting LOCA event for Maine Yankee. 1 5.1.1 Initial Operatine Conditions i i i The initial conditions assumed in the safety evaluations considered in this section are listed in i Table 5.2. These conditions are conservative with respect to intended Cycle 14 operation. l Uncertainties used to account for measurement errors associated with plant instrumentation include-4 1 a) 2% allowance for calorimetric error in core thermal power. j ) b) 4*F allowance for measurement error on RCS temperature. c) 25 psi allowance for measurement error on RCS pressure. Technical Specification limits on core inlet temperature and pressure during power operation are illustrated in Figure 5.1. These limits are based on preserving DNB overpower margin for all 3 F l

? l i possible combinations of temperature and pressure incluc ng those experienced under coastdown t conditions. The preservation of DNB overpower margin is accomplished by reducing the ~ allowable core inlet temperature when operating at lower pressures. This ensures that the i minimum DNBR reported for each of the incidents considered remains conservative for operation at the lawer system pressures. Hot and cold leg RTD response times, which affect the AT power input to the RPS functions, were also considered in the analyses. The safety analysis for Cycle 14 continues to evaluate plugging up to 250 tubes in each steam generator for all events. Explicit allowances are made in the inputs to the Cycle 14 thermal-hydraulic and safety analyses j to justify a range of end-of-Cycle 13 exposure from 12,000 to 14,000 mwd /Mt, and a maximum end-of-Cycle 14 exposure beyond end of full-power life (i.e., coastdown) which is defined by the [ burnup limits of the fuel mechanical design evaluationr. i For Cycle 14, the safety analysis assumptions have been expanded to account for a degradation, in the steam generator overall heat transfer coefficient (UA) which would allow full power operation at the maximum core inlet temperature, with a steam generator pressure as low as 755 i psia (actual)in the steam generator dome. The impact of the reduced steam generator pressure and/or UA assumptions were specifically addressed for the CEA Withdrawal, the Loss of Load, the Steam Generator Tube Rupture, and the small break LOCA analyses. The lower steam _ ~i generator pressure and/or UA assumptions were determined to be conservative relative to the other events. For Cycle 14, the safety analysis has considered a potential reduction of the nominal HZP core inlet temperature from 532"F to 525"F to provide pressure margin between the HZP steam i generator pressure and the lowest setpoint main steam safety valve. Since the safety analysis is evaluated over the range of temperatures specified in Figure 5.1, this change does not affect i the safety analysis. I . I i

l i i i 5.1.2 Core Power Distributions The power distribution in the core, and in particular, the peak heat flux and enthalpy rise, are i ~ of major importance in determining core thermal margin. The procedure used in the safety analysis was to first, define the initial conditions (inlet temperature, power, pressure, CEA insertion, and axial power distribution) and then, through analysis, ensure that sufficient initial overpower margin is available to prevent the violation of the acceptance criteria for each incident analyzed. 2 This procedure is continued for Cycle 14. If the available overpower margin is not sufficient for the set of initial conditions, new power distributions are selected either by modifying the symmetric offset limiting condition for operation (S/O LCO) or by modifying the allowable CEA insertion limit versus power until it is demonstrated that sufficient margin exists. i ] As a starting point, the safety analysis assumes the FSAR design power distribution (F, = 1.68 I and FN H = 1.49) shown in Figure 5.2. Because the cycle-specific " flyspeck" power distributions 7 have been evaluated using SCU for this cycle, the DNBR overpower margins are assessed relative to the SDLs defined in Table 5.1. The SDLs defined for the flyspeck cases are derived by statistically combining both hardware and analysis uncertainties (77-80). To provide a l consistent comparison of the relative severity of the limiting flyspeck distribution to the FSAR l design shape, Table 5.10 provides a comparison of ic. limiting 100% cycle-specific flyspeck case to the design FSAR case. Since the FSAR distribution is not subject to several of the calculational l uncertainties used to derive the SDL for the cycle-specific flyspeck case, this comparison l incorporates the uncertainties in a deterministic manner for both cases with a direct comparison to the DNBR SAFDL of 1.20. The results presented in Table 5.10 demonstrate that the limiting power distribution predicted for Cycle 14 is more limiting than the FSAR design distribution with respect to DNBR and overpower margin. For Cycle 14, the Power Dependent insertion Limit on the CEAs has been modified slightly. Table 5.10 demonstrates that, with respect to initial MDNBR and overpower margin, the limiting full power Cycle 14 predicted power distribution bounds the limiting power distributions at i 4 lower power levels. These comparisons assume that the CEAs are inserted to the Cycle 14 i f f e r

specific PDIL defined in Figure 4.9 within the S/O LCO band defined in Figure 5.7. Overpower margin, presented in Table 5.10 as Po-Po, is defined as the percent rated thermal power margin between P, the power level at which the MDNBR for a given power distribution would equal j o the SDL, and P, the maximum initial power level allowed by the PDIL. Because of variations o in the subchannel location in which MDNBR is predicted at nominal conditions versus limiting conditions, this is a more precise indicator of relative DNB margin between power distributions than initial steady-state MDNBR. Hence, thermal margins calculated using the Cycle 14100% power PDIL power distributions are conservative for Cycle 14. Transient power peaking associated with the CEA Drop and the CEA Ejection events for Cycle [ 14 are compared with reference values in Table 5.2. The effect of differences between Cycle 14, Cycle 13, and the Reference Safety Analysis for the CEA Drop and CEA Ejection are discussed in Sections 5.4.2 and 5.5.4. 5.1.3 Reactivity Coefficients The transient response of the reactor system is dependent on reactivity feedback effects. In particular, these are the moderator and fuel temperature reactivity coefficients. Nominal values for each of the above feedback coefficients are given in Sections 4.5 and 4.6. Variations in the above parameters will influence each transient in a different manner. Therefore, the effect of the difference in reactivity coefficients is discussed on an event-by-event basis. For Cycle 14, the allowable positive values for MTC in the power range are detailed in Figure 4.8. The analyses, limited by a positive MTC, with the exception of the CEA ejection analysis, 4 were conservatively performed at HFP conditions with MTC equal to +0.5 x 10 Ap/*F and ] bound the values in Figure 4.8. The CEA Ejection analysis for Cycle 14 (Section 5.5.4) assumes i 4 the most positive of either the predicted MTC with +0.5 x 10 Ap/'F uncertainty added or the value specified in Figure 4.8 with an additional allowance as described in Section 4.6. Events limited by a negative MTC are discussed in their respective sections. 1 The effective neutron lifetime, delayed neutron fractions, and decay constants are functions of fuel burnup and the fuel loading pattern. The Cycle 14 kinetics parameters are compared to the 73 1 i i i

corresponding Reference Safety Analysis values. Small differences that are experienced from cycle to cycle have an insignificant impact on the response of the plant for all transients, except the CEA Ejection. In the CEA Ejection, the power response is sensitive to the ratio of the ejected l ~ CEA worth to the effective delayed neutron fraction. An evaluation of this event for Cycle 14 1: pravided in Section 5.5.4. i 5.1.4 Shutdown CEA Characteristics The negative reactivity insertion following a reactor trip is a function of the acceleration of the ~' CEA and the variation of CEA worth as a function of position. The safety analysis considers this function in three separate parts: 1) the CEA position versus time,2) the normalized reactivity worth versus rod position, and 3) the total negative reactivity inserted following a scram. r The CEA position versus time assumed in the Reference Safety Analysis was provided as Figure 4.2 in (36). This curve reflects a conservative rod insertion time of 3.0 seconds. This curve is ] based on results from plant measurements and is not expected to change from cycle to cycle. Furthermore, CEA drop times are measured at each refueling as part of the startup test program to verify this assumption. A conservative normalized reactivity worth versus rod position was calculated for various operating conditions in Cycle 14. The normalized reactivity worths versus CEA position calculated for Cycle 14 are shown in Figures 5.3 and 5.1. i Values assumed in the Reference Safety Analysis and for Cycle 13 for the total negative reactivity inserted following a scram are given in Table 52. Table 5.12 provides a listing of the specific scram reactivity assumed for each of the reload safety analyses for Cycle 14. The required values ' of scram reactivity specified in Table 4.13 bound those assumed in the safety analysis supporting operation of Cycle 14. i , l l \\ 1 i

-_~ _ M d 5.1.5 Reactor Protective System Setr>oints and Time Delays 4 The reactor is protected by the Reactor Protective System (RPS) and Engineered Safeguards Features (ESF). In the event of an abnormal transient, the RPS is set to trip the reactor and prevent unacceptable core damage. The elapsed time between when the conditions at the sensor j meet the trip setpoint and when the trip breakers open,is defined as the instrument delay time. The values of the trip setpoints and instrument delay times used in the Reference Safety Analysis are provided in Table 4.7 of (36). The setpoints assumed for Cycle 14 are given in Table 5.11 and s Figures 5.5,5.6, and 5.7. As indicated in (36) and (62), the Reference Safety Analysis assumes no credit for the high rate of change of power trip function. This remains unchanged for Cycle 14. Credit is taken for the functioning of the Variable Overpower (VOI'T), Thermal Margin / Low Pressure (TM/LP) and i l Symmetric Offset Trip Systems (SOTS) in several areas. First, these trips are credited in limiting i the initial power distributions considered in setting the SOTS setpoints as a function of power level. Second, these trips are also credited in limiting the power increase and power distribution c changes possible during the CEA Group or Subgroup Withdrawal, the Excess Load, and the CEA l Drop transients, as discussed in Sections 5.3.1, 5.3.3 and 5.4.2. Credit is also taken for the i j functioning of the VOPT in the analysis of the CEA Ejection transient, Section 5.5.4. [ i i The TM/LP and SOTS setpoints are cycle dependent. They are derived from the predicted core f behavior as described in (4). The Cycle 14 setpoints for the TM/LP and SOTS are presented in Figures 5.5,5.6, and 5.7. a t i j For Cycle 14, the SOTS setpoints have been expanded slightly for power levels above 50% Rated [ l Thermal Power (RTP). This increased operating space was made available by the increased margin realized from the use of the Statistical Combination of Uncertainties methodology. Each j of the affected events was specifically reanalyzed to ensure that the expanded SOTS and f cozresponding S/O LCO bands, for the turbine in normal and IMP 1N mode, continue to provide I f protection against exceeding the acceptance criteria. The TM/LP trip generated for Cycle 14 has not changed significantly and credit of this function to initiate the reactor trip in the analysis of s j l 1 i h E i ~, .=

. _ =. _ __ _ h i 4 the Steam Generator Tube Rupture (SGTR) accident, Section 5.5.2, continues to be a conservative assumption. f i For Cycle 14, the reported uncertainty on the low pressure portion of the TM/LP trip function i has increased from116 to132 psid. This increase in uncertainty was calculated as part of an I i ongoing RPS setpoints review program implemented a t Maine Yankee which involves a complete review of all the RPS setpoints and their uncertainties. The program identified this change in the Cycle 13 analysis assumption and the uncertainty change was addressed prior to the Cycle i 13 startup. The increase in the setpoint uncertainty has no significant impact on the Cycle 14 l safety analysis. i 5.1.6 Plant Modifications f i In addition to the cycle to cycle safety parameter changes related to the core redesign and l 1 t j operating space, modification of plant hardware features (e.g. system piping modification, valve j stroke times, control system modification, etc.) can potentially affect the safety analyses. Each i of these changes is independently reviewed as part of the design change package procedure. e d This section serves as an informational section identifying key design changes as they pertain to the safety analysis results. For Cycle 14, there are no significant plant modifications which affected the safety analysis j i results. 4 i i 5.1.7 Summary of Sienificant Chances in Cycle 14 Reload Safety Analysis i 4 i Significant changes which were implemented between the Cycle 13 reload safety analysis and ) the Cycle 14 reload safety analysis are summarized as follows. i 1. The FROSSTEY-2 computer code was incorporated into the fuel thermal performance analysis methodology. This change resulted in slightly more restrictive LHCR limits corresponding to the FCM SAFDL These limits are used in the safety analysis to ensure that the FCM SAFDL continues to be met. 2 j, I h' l

\\ l 1 2. The Statistical Combination of Uncertainties methodology was incorporated into the evaluation of the SAFDLs for each of the safety analysis events. This methodology provided some margin improvement which allowed for an increase in the SOTS i setpoints above 50% RTP, as well as improvements to the margin to the SDL for i several events. 3. The use of the RELAP5YA computer code in the small break LOCA methodology for Cycle 14 demonstrated that the results of the small break LOCA continue to be i bounded by the large break LOCA. Specific small break LOCA results are provided in this report. 4. To account for potential degradation in the steam generator overall heat transfer coefficient (UA), the safety analysis assumption on UA was expanded to cover full i power operation at the maximum core inlet temperature, witn a steam generator [ pressure as low as 755 psia (actual)in the steam generator dome. The impact of the i reduced steam generator pressure and/or UA assumptions were specifically f addressed for the CEA Withdrawal, the Loss of Load, the Steam Generator Tube Rupture, and the small break LOCA analyses. For the remaining safety analysis f events, a higher value for UA and/or steam generator pressure was determined to be more limiting. l l 5. The uncertainty on the low pressure portion of the TM/LP trip function has increased from +16 to +32 psid. This increase in uncertainty was calculated as part of an ongoing RPS setpoints review program implemented at Maine Yankee which involves a complete review of all the RPS setpoints and their uncertainties. The program identified this change in the Cycle 13 analysis assumption 'and the uncertainty was addressed prior to the Cycle 13 startup. The increase in the setpoint uncertainty has no significant impact on the Cycle 14 safety analysis. l 6. The SOTS setpoints have been expanded slightly for power levels above 50% RTP. This increased operating space was made available by the increased margin realized from the use of the Statistical Combination of Uncertainties methodology. Each of 77

the affected events was specifically reanalyzed to ensure that the expanded SOIS and corresponding S/O LCO bands, for the turbine in normal and IMPIN mode, q continue to provide protection against exceeding the acceptance criteria. l 7. The TM/LP trip is cycle-speciiic and has changed slightly. However, full power TM/LP setpoints have not changed significantly and credit of this function to initiate the reactor trip in the analysis of the Steam Generator Tube Rupture (SGTR) accident, Section 5.5.2, continues to be a conservative assumption. i e 8. The safety analysis has considered a potential reduction of the nominal HZP core inlet temperature from 532 F to 525"F to provide pressure margin between the HZP steam generator pressure and the lowest setpoint main steam safety valve. Since the safety analysis is evaluated over the range of temperatures specified in Figure 5.1, this change does not affect the safety analysis. i 4 ] 9. The Power Dependent Insertion Limit on the CEA's was modified slightly. This j change was explicitly considered in the safety analysis. l 5.2 Event Analvsis Review I Each transient and accident considered in (62) has been reviewed and/or re-evaluated for Cycle

14. The incidents considered are categorized as follows:

l f 1) Anticipated Operational Occurrences for which the RPS ensures that no violation of 1 Specified Acceptable Fuel Design Limits (SAFDL) will occur. i 2) Anticipated Operational Occurrences for which sufficient initial steady-state overpower margin must be maintained in order to ensure acceptable results. ? i 4 i 3) Postulated Accidents. I 4

i Table 5.3 provides a summary of the reload analysis events considered organized by category. Those events that required a complete reanalysis for Cycle 14 included: 1) Boron Dilution 2) Excess Load 3) Loss of Coolant Flow 4) Seized Rotor 5) CEA Ejection Other events that required a partial reanalysis or review included: 1) CEA Withdrawal 2) Loss of Load 3) Loss of Feedwater 4) CEA Drop 5) Steam Line Rupture 6) Steam Generator Tube Rupture 7) LOCA A description of the Cycle 14 evaluations performed for these events is provided in Sections 5.3 through 5.5. A summary of the results for Cycle 14 is presented in Table 5.4. 5.3 Anticipated Operational Occurrences for which the RPS Ensures No Violation of SAFDLs j Each event described in this section is classified as an Anticipated Operational Occurrence { 3 (AOO). An AOO is an event of moderate frequency which does not involve any release of radioactivity from the reactor fuel to the reactor coolant and does not involve any breach of the reactor coolant system boundary. The results of the Cycle 14 analyses demonstrate that the incidents in this category do not violate the fuel SAFDLs, thus ensuring fuel integrity. The integrity of the reactor coolant system boundary is ensured by demonstrating that the transient primary or secondary coolant system pressure limits are not exceeded for these events. l Setpoints were generated for the TM/LP and SOTS which include the changes in power distributions associated with Cycle 14. Thus, protection against violation of the SAFDLs continues to be ensured by the RPS. Sections 5.3.1 through 5.3.5 review the Anticipated Operational Occurrences for which the RPS ensures no violation of the SAFDLs. ~ 5.3.1 Control Element Assembly Grour> Withdrawal i The Reference Safety Analysis for this event demonstrates that the most severe CEA Withdrawal transient occurs for a combination of reactivity addition rate and time in core life that results in the slowest reactor power rise to a level just below either the VOPT or the TM/LP trip. This combination of parameters maximizes the core thermal heat flux and core inlet temperature and v results in the minimum DNBR. l t The Reference Safety Analysis for the 2700 MWt uprate (62) considered parametric analyses at l full power for Moderator Temperature Coefficient (MTC) and Reactivity Addition Rate. The l 4 4 4 ] ranges evaluated were +'J.5 x 10 to -3.0 x 10 Ap/"F and 0.0 to 0.7 x 10 Ap/sec. As indicated l 4 ) in Table 5.2, the Cyce 14 predicted value of MTC with uncertainty, is -3.18 x 10 Ap/*F. ( Reference (36) showed the MDNBR to occur at an MTC of -2.9 x 10 Ap/*F for this event, with 4 more negative MTC resulting in higher MDNBR. c Table 5.2 shows a higher maximum rate of reactivity addition for Cycle 14 for the limiting full CEA group withdrawal. The increased full CEA group withdrawal reactivity addition rate resulted in an overshoot of power above the RPS trip setpoint greater than the normal.3% f overshoot allowance used to evaluate the SAFDLs and was analyzed explicitly with the I increased overshoot. For the subgroup 5B withdrawal, the maximum reactivity addition rate did not result in an overshoot of power greater than the 3% overshoot allowance. The' analysis I determined that, because of more severe excore decalibration effects and the localized radial peaking increases, the subgroup 5B withdrawal transient remained limiting. l The MDNBR for a CEA Group Withdrawal transient for Cycle 14 occurs from the subgroup 5B withdrawal assuming the CEAs are initially positioned at the corresponding insertion limit. l Specific CEA Subgroup Withdrawal transient related uncertainties and decalibration effects have d i ; 4 m x-u .-n,.,,

i been incorporated into the derivation of the TM/LP trip setpoints to ensure that this trip will occur prior to reaching the DNBR SAFDL. Hence, the MDNBR for this event is greater than the TM/LP SDL defined in Table 5.1. l The limiting CEA Withdrawal event with respect to fuel centerline melt (FCM) for Cycle 14 is a Subgroup 5B Withdrawal initiated from approximately 81% power with the CEAs initially inserted to the corresponding PDIL The peak RCS pressure for a CEA Group Withdrawal is listed in Table 5.4 as less than the ASME design overpressure limit of 2750 psia. 5.3.2 Boron Dilution i The Boron Dilution incident was addressed in (36,47,48) and the FSAR (32). An inadvertent dilution of the RCS was considered under a variety of plant conditions which could result in either an inadvertent power generation or loss of shutdown margin (SDM) if sufficient time is E not available for the operator to take corrective action. i l A complete re-evaluation of this incident was performed for Cycle 14 addressing events f postulated during refueling, shutdown, startup, hot standby and power operation conditions. I Table 5.6 presents a summary of the acceptance criteria and analysis results for Cycle 14. 5.3.2.1 Dilution Durine Refueline ) Assumptions made in the Cycle 14 evaluation for dilutions during refueling are consistent with those made in (36) and (47). The limiting dilution in (36) was based on the maximum capacity of the CVCS via the normal makeup and letdown flow paths (200 gpm each). The limiting ) dilution event in (47) was based on the maximum flow of the Primary Water Makeup System (250 gpm). Both analyses assumed letdown flows equal to the dilution flow rates and minimmn reactor vessel water volumes of 2599 ft3 (volume below lower lip of reactor vessel nozzles). Hence, the Primary Makeup Water System dilution is the limiting dilution under refueling conditions. r - -.

u 1 Additional refueling situations were evaluated for the normal refueling condition during which the reactor vessel is filled to l' below the vessel flange (5200 ft ). Table 5.7 summarizes the 3 boron requirements for both conditions during refueling operatiens. l v For the limiting condition where the reactor vessel level is below the hot leg nozzles, it is concluded that if the reactor vessel boron concentration is maintained above the limits specified I in Table 5.7, it would require a continuous dilution at the maximum rate for 30 minutes to achieve an inadvertent criticality. Table 5.7 provides limits for conditions with the two most reactive CEAs withdrawn or ARI during Cycle 14 refueling. This is sufficient time for the i operator to acknowledge the audible count rate signal and take corrective action to cut off the source of the dilution. l f j 5.3.2.2 Dilution Durine Cold, Transthermal, and Hot Shutdown with RCS Filled i l 1 Dilutions during cold, transthermal, and hot shutdown were addressed in (48). The limiting dilution is via the CVCS (200 gpm), and the RCS is assumed to be filled (no credit taken for pressurizer volume). The highest worth CEA is assumed to be stuck out of the core, the loop i 2 stop valves open and either RHR or an RCP operating. In order to address potential partialloop f i 4 i stagnation during RHR operation, portions of the RCS loops, as well as the reactor vessel upper 4 head, were not credited as mixing volumes. Required minimum RCS initial boron concentrations j to allow 15 minutes margin to criticality are listed in Table 5.5, along with the boron i concentration required to meet the Technical Specification 5%AK/K subtriticality requirement j 1 for shutdown conditions. The boron concentrations required by the Technical Specification j 5%AK/K subcriticality requirement conservatively bound those required to meet the 15 minute l requirement for margin to criticality during boron dilution events. L 5.3.2.3 Dilution Durine Cold, Transthermal, and Hot Shutdown with Drained RCS Conditions 4 Dilutions during shutdown conditions with the RCS pa rtially drained were addressed in (47) and l (48). In order to conservatively bound any partially drained configuration with one or more 1 reactor coolant loops isolated, the assumption is made that only the portion of the reactor vessel below the lower lip of the nozzle is filled. With the exception of the CEA of highest worth, 5 i 2 ! 4 1 - i

l l l l which is assumed to be stuck out of the core, and a maximum 1%AK/K of CEA Group Withdrawal, all CEAs are assumed to be inserted in the core. The limiting dilution in this situation is Case No. 3 of (47). The required initial RCS boron concentrations to allow 30 minutes margin to criticality during drained RCS conditions are given in Table 5.5. This margin is used to bound mid-cycle i " refueling" situations where the reactor vessel head may be removed to perform maintenance I operations. Administrative procedures ensure that the higher of the 30 minuNs margin to l criticality or the 5%AK/K suberiticality boron concentrations in Table 5.5 is usrA during drained RCS conditions. Thus, a minimum of 30 minutes margin to criticality is ensm ed for the limiting boron dilution event from drained conditions. 5.3.2.4 Dilution During Hot Standbv, Stactup, and Power Operation The assumptions made for boron dilution events during hot standby, startup, and power operation in (36) remain the same for Cycle 14. The hot standby critical boron concentration j with uncertainty is lower for Cycle 14 than for Cycle 13,1796 versus 18S4 ppm. The results for Cycle 14 using Figures 4.3-4 and 4.3-5 of (36) are summarized below: Maximum Reactivity Insertion Rate Dilution at Hot Standby 9.50 x 104 Ap/sec 4 Dilution at Power 9.72 x 10 Ap/sec The consequences of events with respect to margin to the SAFDLs for these small reactivity addition rates are bounded by the results reported in Section 5.3.1 for the CEA Withdrawal incident. Based on the maximum reactivity addition rate it would take approximately 49 minutes of continuous dilution at the maximum charging rate to completely absorb a 3.2%AK/K shutdown margin. Because of the available alarms and indications, there is sufficient time and information to allow the operator to take corrective action. i

5.3.2.5 Failure to Borate Prior to Cooldown s i Because of the large negative MTC at EOC, a decrease in RCS temperature adds reactivity to the reactor core. Consequently, during the process of cooling down the RCS for refueling or repairs, l I boration is used to compensate for this reactivity addition. The failure to add boron during cooldown was evaluated on the basis of the following j assumptions: [ (a) The MTC is the most negative value expected with all CEAs in the core, including uncertainties. j i (b) The initial shutdown margin (SDM) is 3.2%AK/K at an average temperature of l 550'F (a more conservative condition than the nominal 525*F). This SDM is I conservative relative to the requirements defined in Figure 4.12. (c) The RCS temperature is reduced at the rate of 100 F/hr, the maximum cooling rate { permitted. In order to make the reactor critical from these initial conditions, the average coolant temperature must be reduced from 550 F to about 456*F. This temperature reduction requires approximately 56 minutes to accomplish. This is sufficient time for the operator to diagnose the condition and take the necessary corrective action. 5.3.3 Excess Load incident e An Excess Load incident is an event where a power-energy removal mismatch leads to a decrease in the RCS average temperature and pressure. When the MTC is negative, ) unintentional increases in reactor power may occur. Thus, the Excess Load incident as reported in (62) was analyzed over a wide range of power levels and negative MTCs to determine the minimum margin to the FCM and DNBR SAFDLs. .l d J M-J i

X -m r s-4 ~e +44 Ar a +- m l i. .I i j The most severe consequences from an Excess Load event result from an inadvertent opening of the Steam Dump and Bypass System near full power or an inadvertent opening of the turbine .r admission valves without credit for the turbine load limiters at lower power levels. At various I power levels, initial power distributions are obtained from the limits of the S/O LCO band in j Figure 5.7. From any initial power level, the Excess Load transient is assumed to start at the S/O LCO band and terminate when the AT power or excore power signals reach the S/O trip i and/or VOlilimit. The excore power signal is decalibrated by the transient-induced cooldown l j in the downcomer and the AT power signal lags slightly behind the power excursion. Hence, l the peak power achieved during this transient can overshoot the RPS trip setpoints. The maximum power varies as a function of MTC, so a wide range of negative MTCs are considered. 4 j For Cycle 14, the excess load incident was completely reanalyzed. The peak power and closest i approach to the FCM SAFDL was determined to occur at MTC values expected mid-cycle. Mid-l cycle MTCS are more limiting because the rate of power increase, with moderately negative MTC values, allows for maximum deposition of heat in the fuel prior to the reactor trip. The closest l approach to FCM corresponds to an event initiated near 93% RTP from the negative edge of the symmetric offset band resulting in a power increase to the VOPT setpoint. i i The minimum DNBR occurs towards the end of cycle for all initial power levels because the i strongly negative MTC provides a faster rate of power increase while minimizing the beneficial ] DNBR effects due to the transient-induced cooldown of the RCS. The MDNBR for the most ? limiting Excess Load event is greater than the transient SDL defined in Table 5.1 and corresponds to an event initiated near 93% RTP from the positive edge of symmetric offset band resulting in a power increase to the VOPT setpoint. 5.3.4 Loss of Lord Incident l The typical causes of a Loss of Load event are a station separation from the grid, a turbine trip, or an electrical generator malfunction while the plant is at power. The subsequent closure of the turbine generator main steam stop valves causes a large mismatch between the power input to the primary system and the secondary system heat removal capacity. The principle concem of I the Loss of Load event analysis is to ensure that the RPS setpoints, combined with the i i I

overpressure protection devices (i.e., primary and secondary code safety valves), prevent the peak transient RCS and secondary pressures from exceeding the ASME design overpressure limit l of 110% of their respective design pressures. System parameters which have a major influence upon the severity of the pressure excursion are [ ] the initial conditions for power level, RCS pressure, pressurizer level, steam generator UA, steam generator pressure, primary and secondary relief valve capacities and setpoints, high pressurizer ) pressure trip setpoint, and moderator temperature coefficient. 1 For Cycle 14, the Loss of Load event was completely re-analyzed. All of the system parameters identified above remained the same as the Reference Safety Analysis with the exception of the l 4 initial steam generator pressure which was assumed to be 755 psia in the steam generator dome. j The steam generator UA continued to be conservatively assumed as that corresponding to a clean steam generator, even though the RCS volume was conservatively reduced to account for j 250 plugged tubes per steam generator. ) The peak transient RCS pressure and secondary pressure, as noted in Table 5.4, were j demonstrated to remain below the ASME design overpressure limit of 2750 psia. A COBRA-IllC analysis with peaking consistent with the 100% power PDIL was performed to determine the MDNBR for Cycle 14. The MDNBR for this is greater than the transient SDL defined in Table 5.1. t 5.3.5 Loss of Feedwater incident i -t l A Loss of Feedwater transient occurs as a result of a loss of the main feedwater supply to the steam generators while the plant is at power. The principle concem for the loss of feedwater event is that the minimum level achieved in the steam generators does not drop below the level j adequate for adequate heat removal both for forced and for natural circulation conditions in the ] event of a coincident loss of offsite power. Since the transient results in a partial loss of 4 I j secondary heat removal, the event is also evaluated to ensure that the RPS setpoints, combined l with the overpressure protection devices (i.e., primary and secondary code safety valves), i

l 1

i 1 4

i i prevent the peak transient RCS and secondary pressures from exceeding the ASME design I overpressure limit of 110% of their respective design pressures. System parameters which have a major influence upon the determination of the transient minimum steam generator level as well as the RCS and secondary system pressure excursions l are the initial conditions for power level, RCS pressure, pressurizer level, steam generator UA, steam generator pressure, primary and secondary relief valve capacities and setpoints, high . pressurizer pressure trip setpoint, and moderator temperature coefficient. l i For Cycle 14, the Loss of Feedwater event was partially re-analyzed. All of the system i parameters identified above remained the same as the Reference Safety Analysis except the initial l steam generator pressure which was evaluated at 755 psia in the steam generator dome. i For a Loss of Feedwater transient from full power with the single failure of one emergency l 1 feedwater pump, the steam generator collapsed liquid level reaches a minimum of 31% of the i tube bundle height 16.8 minutes after the low level trip occurs assuming operation of the RCPs. j Without RCPs operating, the steam generator collapsed liquid level reaches a minimum of 37.2% j of tiie tube bundle height 8.3 minutes after the low level trip. These le.>els ensure adequate heat sink throughout the transient for both forced flow and natural circulation conditions (71). i Peak RCS pressure for the Loss of Feedwater transient, with the lower initial steam generator pressure, remains bounded by the peak Loss of Load transient pressure at less than 2750 psia. 1 A COBRA-IIIC analysis with peaking consistent with the 100% power PDIL was performed to i determine the MDNBR for Cycle 14. The MDNBR for this transient is greater than the transient SDL defined in Table 5.1. 5.4 A_nticipated Operational Occurrences Which are Dependent on Initial Overpower Marcin for Protection Acainst Violation of SAFDLs The incidents in this category rely on the provision of adequate initial overpower margin to l ensure that they do not result in violation of the SAFDLs. These incidents are reviewed here, i -87 ]

with the parameters listed in Table 5.2,in order to demonstrate that the incidents of this category j do not violate the SAFDLs, primary system pressure limits, or site boundary dose limits j i (10CFR100) under Cycle 14 conditions. 5.4.1 Loss of Reactor Coolant Fiow l I A Loss of Reactor Flow transient can occur as a result of the ! css of electrical power to one, two or all three Reactor Coolant Pump motors. The principle concern for the Loss of Reactor Coolant Flow transient is that the operating restrictions on power level, symmetric offset, PDIL. pressure ) and temperature, provide sufficient initial overpower DNB margin to provide protection from the SAFDL on MDNBR. 1 i 1 1 Results of the Loss of Coolant Flow analysis are sensitive to initial overpower DNB margin, rate { of flow degradation, low reactor coolant flow reactor trip setpoint,available scram reactivity,and l MTC. The assumptions pertaining to MTC, low reactor coolant flow trip setpoint, and rate of coolant flow degradation remain the same as in the Reference Safety Analysi:: for this event. { l t For Cycle 14, the Loss of Reactor Coolant Flow transient has been completely re-analyzed. There l t are ns significant deviations in the safety parameters affecting the Loss of Flow transient from j the Reference Safety Analysis. However, the incorporation of the SCU methodology, which defines a specific SDL for the Loss of Flow transients, dictates the need for a complete reanalysis l r to establish a new Refererice Safety Analysis for Cycle 14. l The lintiting event is a complete (three pump) Loss of Flow transient. The MDNBR for the i complete loss of flow from 100% power using the limiting 100% power PDIL power distribution for Cycle 14 is greater than the Loss of Flow SDL defined in Table 5.1. l 5.4.2 Full Lencth CEA Drop a l The drop of a fulllength CEA results in a distortion of the core power distribution and could lead to the violation of SAFDL. As discussed in Section 5.1.2, the S/O LCO band is designed .l i 9, r i e

to restrict permissible initial operating conditions such that the SAFDL for DNB and FCM are I not exceeded for this incident. The Reference Safety Analysis (62) of this incident identified the limiting transient as one r initiated from near full power. To cover all potentially limiting conditions, the CEA drop for Cycle 14 was evaluated from power levels ranging from 0% to 100% of 2700 MWt. Power distributions used in the evaluation of DNBR and proximity to FCM were selected at each power level from the limiting cases within the S/O LCO band. The initial increase in peaking as a I function of dropped CEA worth for Cycle 14 is given in Figure 4.10. The value for the maximum iw;rease in peaking for any dropped CEA from Figure 4.10 was conservatively applied j at each power level considered (Section 4.9.3). 1 The CEA Drop analysis also considers the inu .J peaking which results from xenon i redistribution for the period of time during which operation with a dropped CEA is allowed by the Technical Specifications, as described in Section 4.9.3. The increase in peaking from Figure l 4.1D was conservatively augmented by the increase in peaking due to xenon redistribution at i subsequent points in time, assuming operation consistent with the power level reductions required by the Technical Specifications. The margins to the SAFDLs were then determined for f the limiting power distributions within the S/O LCO band at any point in time following a drop, assuming the CEAs to be inserted no deeper than the PDIL associated with the pre-drop power level. l The limiting full length CEA drop with respect to DNB was identified to occur at the point where the minimum worth CEA resulted in the maximum increase in peaking (62). Thus, for conservatism, the plant response assumed in the Cycle 14 evaluation considered a worth of 0.13%Ap (Figure 4.10). The results of the DNB evaluation for Cycle 14 indicate that the MDNBR occurs 0.5 hours after a full length CEA drop from 100% power. At this time, with conditions at the positive edge of the S/O LCO offset band, the MDNBR is greater than the transient SDL defined in Table 5.1. The limiting fulllength CEA drop with respect to FCM is one initiated from power distributions at the edge of the S/O LCO band at each power level. The maximum allowable steady-state, 1

f linear heat generation rate is limited to ensure that the maximum post-drop linear heat f generation rate does not violate the limiting centerline melt SAFDL. These limits are reflected 1 in deriving the S/O LCO band. The safety analysis of the CEA Drop event assumes that control of the turbine admission valves is performed manually. Following the drop, core power initially decreases due to the negative j ] reactivity associated with the inserted CEA. The relatively high steam demand from the constant turbine throttle valve setting reduces the RCS temperature resulting in positive reactivity i feedback. For conservatism, it is assumed that there is sufficient positive reactivity to return to i l core power to its pre-drop level. l i During operation with turbine admission valve control in Impulse IN mode UMPIN), the [ r i admission valves automatically react to changes in steam flow or pressure to maintain constant impulse pressure at the inlet to the first stage of the high pressure turbine. Thus,if the inlet I steam flow and pressure were to drop, as they do immediately after a CEA drop, the throttle valves would open further in an attempt to restore the impulse pressure to its set value. While f operating in the IMPIN mode, the potential exists for the transient core power to exceed the pre-drop power level, due to either a single failure of the IMPIN control logic or overshoot of the controller during the return of core power and SG pressure for the transient steam demand. l This could cause core power to return to values higher than the pre-drop level. The potential [ power overshoot is limited to a maximum of 10% above the initial power level by the VOPT i setpoint. 4 1 l The allowable range of plant operation following a CEA drop, with respect to post-drop core t power distributions, is determined by verifying the acceptability of continued power operation within the S/O LCO band with respect to the DNB and FCM SAFDLs. Any decrease in margin l due to power levels returning above the pre-drop level affects the S/O LCO. Consequently, a 1 suitably conservative S/O If0 operating band for the IMPIN operating mode, which protects j both the DNB and FCM SAFDLs, has been developed by lowering the normal S/O LCO by an j amount equal to the maximum potential increase in post-drop power level (determined by the VOPT setpoint). Both the normal and IMPIN S/O LCO bands are shown in Figure 5.7. , i i I i,

1 5.5 Postulated Accidents The incidents in this category were previously analyzed in (3,12,20,36,43,46,49,50,62). For the conditions in these reports, it was demonstrated that each of these incidents met the appropriate acceptance criteria. Each of these incidents have been reviewed and results for Cycle 14 are reported in the following sections. 5.5.1 Steam Line Rupture The system analysis cod e, RETRAN-02 MOD 2, was used in the most recent complete Steam Line i i Rupture (SLR) analysis (62) to predict the consequences of a double-ended guillotine break in i the main steam line coincident with a single failure. The limiting single failure was determined to be a feedwater regulating valve failure in the faulted steam generator failing in the open position. The goal of this analysis is to ensure that the core does not return to a critical state following the initial reactor scram. Adequate margin to subcriticality is demonstrated if the available scram reactivity combined with the time-dependent boron worth from safety injection is larger than the reactivity due to moderator and Doppler defects at all times during the accident. This is conservative with respect to the actual acceptance criteria, which require that fuel damage be of sufficiently limited extent such that the core will remain intact with no loss of core cooling capability, and the calculated off-site doses not exceed the values of 10CFR100. A system analysis was not required for Cycle 14 because none of the thermal-hydraulic characteristics have changed, ensuring that the thermal-hydraulic response predicted in the Reference Analysis (62) remains valid. As in previous cycles, potential variations in boron concentrations throughout the HPSI system have been considered. Specifically, the piping between the HPSI pumps and the cold leg injection points is assumed to be completely deborated. The boron concentration in the piping between the RWST (1720 ppm) and the HPSI pumps is assumed to be as low as 1370 ppm.,

r Table 5.8 gives the nominal scram reactivity necessary to avoid recriticality for HFP and HZP cases at BOC and EOC, along with the nominal available scram reactivides for Cycle 14. The most limiting case for Cycle 14 was HFP at EOC. The minimum margin for the HFP at EOC case is 0.77Ep. Since the nuclear uncertainties in the available scram reactivities have been statistically combined in the SLR analysis, the available scram reactivities listed in Table 5.8 are 'I nominal values as determined in Table 4.12. The required scram reactivities calculated for Cycle q 14 are used to determine the shutdown margin requirements. 5.5.2 Steam Generator Tube Rupture l The analysis of the SGTR event performed in the Reference Safety Analysis (36) was reviewed for its applicability to Cycle 14. The primary system thermal-hydraulic response is mamly a function of the initial system pressure and the time of reactor trip. The nominal primary system operating pressure of 2250 psia remains unchanged from the value assumed in the Reference Safety Analysis. However, a gradual degradation in the overall steam generator heat transfer I coefficient (UA) has been observed since the initial plant startup, which has reduced the steam F generator pressures. To bound the lower steam generator operating pressures, the SGTR event i was partially re-evaluated during Cycle 12 assuming a minimum steam generator operating pressure of 755 psia in the steam dome, which was lower than the Reference Safety Analysis value of 857 psia. The analysis results, assuming operation with the lower steam generator pressure, resulted in an increased integrated break flow from the RCS to the secondary system. i In addition, a conservatively small UA was used in the analysis to maximize RCS temperatures. Thus, the fraction of the primary coolant which was determined to flash directly to steam increased. h The thermal-hydraulic portion of Reference Safety Analysis credited a reactor trip at the thermal margin trip setpoint. Since the TM/LP trip setpoint has not significantly changed for Cycle 14, the time of reactor trip determined in the Reference Safety Analysis remains valid. t To evaluate the radiological consequences of the SGTR event with the reduced steam generator pressure, a conservative penalty on integrated break flow and flashing fraction was applied to the Reference Safety Analysis thermal-hydraulic results. These results were incorporated into 1._-

1 the SGTR radiological evaluation to determine the impact on offsite consequences. The results j of this evaluation demonstrated that 1 release increased slightly due to the greater integrated l 2 break flow and higher flashing fraction determined from the thermal-hydraulic evaluation. However, the results remained below the acceptance criteria of 10CFR100. l An additional SGTR sensitivity analysis was performed under coastdown conditions, during which the RCS is operated at lower temperatures and power levels, and the steam generators . may be operated at lower pressures. A conservative minimum steam generator pressure of 500 psia in the steam dome was assumed. The increased differential pressure between the RCS and j the steam generator resulted in an increase in the integrated break flow for the SGTR. However, i operation at the lower temperature conditions reduced the amount of coolant which was determined to flash directly to steam in the steam generator. This resulted in a decrease in the 1 release. The results of this evaluation demonstrated that the SGTR event initiated at full 2 power conditions bounds the SGTR event under coastdown conditions. i i For Cycle 14, the SGTR results remain below the acceptance criteria of 10CFR100. l 5.5.3 Seized RCP Roter The radiological consequences of the Seized Rotor accident are sensitive to the number of pins predicted to experience clad damage combined with the primary to secondary leakage rate. Conservative inputs for the core power distribution, radial peak pin census, assumed rate of flow degradation, reactor coolant low flow trip setpoint (time of reactor scram), and the MTC were used in the Seized Rotor analysis to predict the number of fuel rods with a calculated MDNBR less than the Loss of Flow SDL defined in Table 5.1. i The Seized Rotor accident was completely reanalyzed for Cycle 14. The fraction of fuel predicted to undergo DNB during a Seized Rotor accident was determined assuming full power initial conditions, the power distribution consistent with 100% power PDIL, and the pin census for Cycle 14. A conservative MTC of +0.5 x 10-4 Ap/ F was used to bound intermediate power I Any pin which was predicted to experience DNB was conservatively assumed to cases. experience clad failure. The results for Cycle 14 are included in Table 5.4. i i

Radiological release analyses show that the release limits of 10CFR100 will not be exceeded even if all pins experiencing DNB fail. Therefore, the predicted consequences of a Seized Rotor accident during Cycle 14 are acceptable. 5.5.4 CEA Fiection i i The consequences of a CEA Ejection accident are most sensitive to ejected CEA worth, effective delayed neutron fraction, and post-ejected peaking. Specifically, the severity of the transient increases for higher ejected CEA worths, smaller delayed neutron fractions, and increased post-ejected peaking. A complete re-analysis of the CEA Ejection accident was performed for Cycle 14 consistent with the approved methodology (13). i A summary of the results from Cycle 14 for the HFP and HZP cases is presented in Table 5.9. All of the cases investigated resulted in a radially averaged fuel enthalpy of less than 280 cal /gm at any axial location in any fuel pin. The radiological release results in off-site doses less than 10CFR100. 5.5.5 Loss of Coolant 5.5.5.1 Introduction and Summary a Large break Loss-of-Coolant Accident (LOCA) calculations for Maine Yankee were performed for Cycle 5 through 9 using the Yankee Atomic Electric Company (YAEC) WREM-based generic PWR ECCS evaluation model (11). The Cycle 5 calculations (1) consisted of a complete break spectrum analysis and the calculation of cycle-specific limits. The analyses performed in Cycles 6 through 9 demonstrated that the Cycle 5 break spectrum results remained applicable. The results of that analysis were then used to calculate the LOCA limits for each cycle. For Cycle 10, and subsequent analyses, the YAEC LOCA methodology was modified to include a more complete spectrum of possible power shapes. This change required that a break spectrum analysis be performed for four different power shapes. The results of the break spectrum analysis was then used to calculate the LOCA limit curve for Cycle 10. Detailed i ? discussions of the break spectrum analysis and the LOCA limit calculations for Cycle 10 are i provided in (55) and (56). The Cycle 10 LOCA analysis served as the Reference LOCA Analysis for Cycle 11 at 2630 MWt. l To evaluate the operations of Cycle 11 at the uprated power level of 2700 MWt, the break spectrum analysis and LOCA limit calculations were performed using Cycle 11 inputs with a 2700 MWt core power level. In Cycles 12 and 13, the break spectrum analysis and LOCA limit calculations were performed using cycle specific inputs using the same methodology as the Cycle 11 uprate analysis (62). For Cycle 14, the LOCA input parameter values are,ery similar to Cycle 13. Hence, the cycle-specific differences were addressed via sensitii ty analyses. The results showed that Cycle 14 [ 4 operation would be bounded by increasing the Cycle 13 Peak Cladding Temperature (PCT) values by 10"F. The main cause for this increase was due to changes in the reactor kinetics and reactivity parameters. With this increase in PCT,10CFR50.46 criteria continue to be met for Cycle 14 for operation within the linear heat generation rate limits specified in Figure 5.8. 5.5.5.2 Larste Break LOCA Analvsis 4 To determine the LOCA limit at an axial location in the core, an axial power shape with its peak { at the specified axial location is used in the LOCA analysis. If the results of the analysis meet l the 10CFR50.46 acceptance criteria, the Peak Linear Heat Generation Rate (PLHCR) is the derived limit at the specihed location. The process is repeated for several axial locations so that a curve through the established PLHCRs forms the axially dependent LOCA limit cun'e. i l The Cycle 13 LOCA limit curve which is applicable to Cycle 14 is shown in Figure 5.8 and was j generated by calculating the limits at four axial locations: 52%,65%,73% and 85% of the core height. To calculate the limits at the four specified locations, the limiting break for each limiting power shape was identified. A brief discussion of the analysis follows. a 4 ; - i \\

J 5.5.5.2.1 Break Sr>ectrum Anaivsis A break spectrum analysis was performed for Cycle 13 to determine the limiting break sizes at the four axial locations specified above. This analysis was found to be applicable to Cycle 14. l ~ ~ As in Cycle 13, the fresh fuel assemblics inserted in Cycle 14 core are the CE debris-resistent j design and have identical hydraulic characteristics. The combined effects of Cycle 14 reactor input parameters used in LOCA analysis were determined to have a very smallimpact on results 4 l of Cycle 13 break spectrum analysis. Therefore, the limiting break sizes determined on the Cycle l t 13 analysis are applicable to Cycle 14. i 5.5.5.2.2 LOCA Limit Calculations i The PLHGR values used in the break spectrum analysis are the LOCA limit values. These limits 4 i for each axial power shape are provided in Table 5.13. The Cycle 14 peak cladding temperature, cladding oxidation values, and hydrogen generation results demonstrate compliance with Appendix K criteria with additional safety margin. The PLHCR values are plotted as a function of core height in Figure 5.8. The LOCA limit is assumed to be linear between the analyzed points, and is specified at values less than or equal to the points shown in Table 5.13. These j assumptions are discussed in detail in (57). l

l i

5.5.5.3 Small Break LOrA Analvsis 1 P The Small Break LOCA (SBLOCA) was re-analyzed during Cycle 13 using the RELAP5YA l computer code, which was approved by the NRC for SBLOCA analysis (83). The Cycle 13 re-analysis included consideration of the expanded assumptions on initial steam generator pressure and UA. The analysis includes a spectrum of cold leg breaks ranging in size from 0.10 ft to 0.35 j 2 2 ft, using the four axial power shapes from the large break LOCA analysis, with the corresponding PLHCR LOCA limit values. The analysis also includes sensitivity studies to j address Cycle 14 specific input parameters. j i l The results of the analysis for each axial shape that bound Cycle 14 operation are presented in l Table 5.14. The calculated cladding temperature, cladding oxidation values, and hydrogen ' i i l 1 1

i i generation results demonstrate compliance with 10CFR50.46 acceptance criteria with additional 2 safety margin. The most limiting case is the 0.15 ft break with a 73% peaked shape and a l PLHGR of 14.10 kw/ft. i i Calculated peak cladding temperatures are lower than those predicted in the current large Break l LOCA analysis. Thus, the large break LOCA analysis continues to provide the limiting LOCA i scenario for Maine Yankee. 5.6 Methodoloev and Methodolocv Revisions l t A summary of the reference reports and subsequent methodology changes applied to the safety an:: lysis methods for Maine Yankee since Cycle 3 is provided in Table 5.15. For Cycle 14, the reload safety analyses have incorporated three first time applications of methodology changes. l These include the fuel performance analysis using the FROSSTEY-2 computer code (74), the Statistical Combination of Uncertainties methodology (77-80), and the small break LOCA (SBLOCA) analysis incorporating the RELAP5YA computer program (83). t i I s

-t TABLE 5.1 Statistical DNBR Limits (SDLs) Used in Safety Analysis t LIMIT APPLICATION VALUE' SDI To be used to evaluate DNBR concerns 1.57 TRANS1ENT for all transients which do not involve RCS flow reduction 7 \\ SDI ujgy To be used to derive the TM/LP equation 1.59 q SDLop To be used to evaluate DNBR concerns 1.64 i t for events which involve a reduction in RCS flow j l i i

  • These values were derived for Cycle 14 using the methodology described in (77-80).

l l [ 4 I e i i i i

TABLE 5.2 Maine Yankee Safety Parameters Assumptions Cycle 3 Cycle 13 Cycle 14 including including including Uncertainties Uncertainties Uncertainties Parameter Units - Planar Radial Peaking Factor 1.68W3) 1.75 1.69 Group 5 Inserted to 100% PDIL 1.72W - Axial Peak for Shape Resulting in MDNBR at 100% RTP 1.42G 1.50 1.40 - Augmentation Factors 1.0 to 1.067 ) NoneUU) NoneUU) 0 - Moderator Temperature Coefficient 10 Ap/"F 0 to -2.74 + 0.88" U to -3.39 +1.2000 to -3.18 4 - Ejected CEA Worth UOC, HFP %Ap 0.210 0.409 0.412 BOC,HZP %Ap 0396 0569 0.537 EOC, liFP %Ap 0.230 0.441 0316 EOC, IIZP %Ap 0.544 0.646 0.682 - Ejected CEA 3D Peak UOC,IIFP 5.53 6.67 6.47 BOC, HZP 1332 13.57 13.42 EOC, HFP 5.59 639 6.89 EOC,HZP 14.08 15.09 2239 - Dropped CEA Integral Worth %Ap 0 to 030 0 to 0.20 0 to 0.20

TABLE 5.2 (Continued) Cycle 3 Cycle 13 Cycle 14 Including Including Including Parameter Units Uncertainties Uncertainties Uncertainties - Dropped CEA Integral Figure 4.4-1 of Figure 4.10 of Figure 4.10 Radial Peak Reference 3 Reference 75 - Power Level (including 2% uncertainty) MWt 2683 2754 2754 - Maximum Reactor Coolant Inlet Temperature "F 554 548 - 55602) 548 - 55602) - Reactor Coolant System Pressure psia 2200 - 2300 2050 - 2300 2050 - 2300 - Reactor Coolant System Flow Rate 10 lbs/hr 134.6 134.1W-135.3W 134.1W-1353W 6 W - Axial Power Distribution Symmetric Figure 63-1 Figure 5.7 of Figure 5.7 Limit Offset of Reference 3 Reference 75 - Power Dependent Figure 4.9 of Figure 4.9 of Figure 4.9 Insertion Limit Reference 20 Reference 75 - Initial Steady-State

  • 1.793W 1.764W Minimum DNB Ratio YAEC-1 1.977 1.012<6) j,774 )

- Maximum Possible Rate W 4 4 4 j of Reactivity Addition Ap/sec 0.7x10 1.26x10 1.51x10 l l l l L .~.

1 TABLE 5.2 (Continued) Cycle 3 Cycle 13 Cycle 14 Including including Including Parameter Units Uncertainties Uncertainties Uncertainties - Steam Generator Dome Pressure Full Power Conditions psia 877 877 - 758 877 - 755 - Steam Generator Level (Narrow Range) 66 77 - 48 77 - 48 Steam Generator (SG) Tubes Plugged /SG 250 250 250 - Minimum Required CEA Worth .y Assumed in Safety Analysis %Ap BOC, HFP 4.0 5.73 5.80 13OC,1-17.P 2.0 3.20 3.80 EOC, IIFP 5.7 7.50 658 680 EOC,HZP 2.9 5.17 4.07 . ~

..~. r TABLE 5.2 (Continued) l [ Notes l 1) Applies only in fuel centerline melt calculations. i 2) With limiting cycle dependent power distribution as limited by the associated cycle's symmetric offset pretrip alarm. Power level refers to conditions allowed by PDIL for that cycle. 3) Values shown in Reference 12 did not include uncertainty. N = 1.49, F = 1.68). Includes deterministic application 4) FSAR design power distribution (F3g 2 of 2% calorimetric power uncertainty and 3% allowance for maximum tilt allowed by ) Technical Specification 3.10. l 5) Based on Reactor Coolant System pressure of 2200 psia, and temperature of 556 F. 6) Based on Reactor Coolant System pressure of 2050 psia, and temperature of 548 F. j 7) Groups 5 and 4 inserted to PDIL at 100% per Cycle 3 PDIL. 8) EOC, HFP steam line break assumed 6.5%Ap. i 9) For CEA group withdrawal transient. 10) Augmentation factors were removed in Cycle 10 (54). ) 11) Used in the HZP, BOC CEA Ejection evaluation only. j 1 a 12) Temperature range assumed in the safety analysis except the containment pressure analysis (62) which assumed 551.3"F. 1, o l l

i

-102-i i

i r TABLE 5.3 i Maine Yankee Cycle 14 - Incidents Considered i A. Anticipated Operational Occurrences for which the RPS ensures no violation of SAFDLs: t i 1. Control Element Assembly Group and Subgroup Withdrawal 2. Boron Dilution 3. Excess Load 4. Loss of Load 5. Loss of Feedwater B. Anticipated Operational Occurrences which are dependent on Initial Overpower Margin for protection against violation of SAFDLs. 1. Loss of Coolant Flow s 2. Full Length CEA Drop C. Postulated Accidents: l i 1. Steam Line Rupture l 2. Steam Generator Tube Rupture

3. - Seized Rotor 4.

CEA Ejection i 5. Loss of Coolant l 1 V -103-

i TABLE 5.4 Maine Yankee Cycle 14 Safety Analysis Summary of Results incident Section Criteriaum Reference Safety Analysis Cycle 130 Cycle 140 Cycle 11 CEA Withdrawal 5.3.1 SDL = 1.590 MDNUR 21.20 MDNBR >1.20 MDNBR21.59 RCS pressure RCS pressure RCS pressure RCS pressure <2750 psia <2750 psia <2750 psia <2750 psia FCM SAFDL Not exceeded Not exceedal Not exceeded Cycle 13 Boron 53.2 Subcritical: Subcritical: Subcritical: Subcritical: Dilution Sufficient time Refueling-30 min. Refueling-30 min. Refueling-30 min. .L for operator Startup-64 mins Startup64 minutes Startup-67 minutes action Critical: Bounded Critical: Bounded Critical: Bounded Critical: SDL = 1.57G by CEA withdrawal by CEA withdrawal by CEA withdrawal Cycle 11 CEA Drop 5.4.2 SDL = 1.570 MDNBR = 1.29 MDNUR = 1.23 MDNBR = 1.77 FCM SAFDL Not excealed Not exceeded Not exceeded Cycle 13 Loss of 5.4.1 SDL = 1.640 MDNBR = 1.26 MDNDR = 1.26 MDNBR = 1.81 Coolant Flow Cycle 11 Seized Pump 5.5.3 10CFR100 13.9% of rods with 10.0% of rods with 11.6% of rods with Rotor MDNBR less than 1.2 MDNBR less than 1.2 MDNBR less than 1.64 Radiological dose Radiological dose Radiological dose within 10CFR100 within 10CFR100 within 10CFR100

TABLE 5.4 (continued) Incident Section Criteria ReferenceSafety Analysis Cycle 13 Cycle 14 Cycle 13 Excess Load 5.3.3 SDL = 1.570 MDNBR = 1.22 MDNBR = 1.22 MDNUR = 2.03 Cycle 11 Loss of Load 5.3.4 SDL = 1.57G MDNBR = 1.60 MDNBR = 1.40 MDNUR = 1.97 RCS pressure RCS pressure RCS pressure RCF pressure <2750 psia <2750 psia 2750 psia <2750 psia Cycle 11 Loss of 5.3.5 RCS pressure Peak RCS pressure Peak RG pressure Peak RG pressure Feedwater <2750 psia <2750 psia <2750 psia <2750 psia SDL = 1.57G MDNBR = 1.52 MDNBR = 1.43 MDNBR = 2.02 h Cycle 11 Steam Line 5.5.1 10CFR100 Fuel rod integrity Fuel rod integrity Fuel rod integrity Rupture is maintained since is rnaintained since is maintained since reactor does not reactor does not reactor does not return critical return critical return critical Cycle 3 Steam Generator 5.5.2 10CFR100 Radiological dose Radiological dose Radiological dose Tube Rupture within 10CFR100 within 10CFR100 within 10CFR100 Cycle 13 CEA Ejection 5.5.4 10CFR100 Radiological dose Radiological dose Radiological dose l within 10CFR100 within 10CFR100 within 10CFR100 Max. clad failure = 10.9% Max. clad failure = 10.9% Max. clad failure = 11.8% l { i

TABLE 5.4 (continued) Incident Section Criteria Reference Safety Analysis Cycle 13 Cycle 14 LOCA 5.5.5 10CFR100 Cycle 11 10CFR50.46 Radiological dose Radiological dose Radiological dose within 10CFR100 within 10CFR100 within 10CFR100 Results meet acceptance Results meet acceptance Results meet acceptance criteria of 10CFR50.46 criteria of 10CFR50.46 criteria of 10CFR50.46 Cycic3 Steam Line 10CFR100 Radiological dose Reference analysis Reference analysis Rupture Outside within 10CFR100 unchanged by unchanged by Containment Cyde 13 reload Cycle 14 reload Cycle 3 Feedwater 10CFR100 Bounded by steam line Reference analysis Reference analysis a Line Rupture rupture unchanged by unchanged by Outside Cycle 13 reload Cycle 14 reload Containment Cycle 3 Containment Peak pressure Peak pressure less Reference analysis Reference analysis Pressure less than 55 psig than 55 psig unchanged by unchanged by containment Cycle 13 reload Cycle 14 reload design pressure Cycle 3 Fuel llandling 10CFR100 Radiological dose Reference analysis Reference analysis incident within 10CFR100 unchanged by unchanged by Cycle 13 reload Cycle 14 reload

TABl.E 5.4 (continued) Incident Section Criteria Reference Safety Analysis Cycle 13 Cycle 14 Cycle 3 Waste Gas 10CFR100 Radiological dose Reference analysis Reference analysis System within 10CFR100 unchanged by unchanged by Failure Cycle 13 reload Cycle 14 reload Cycle 11 Spent Fuel 10CFR100 Well within Well within Well within Cask Drop 10CFR100 limits 10CFR100 limits 10CFR100 limits Cycle 3 lbdioactive 10CFR100 Radiological dose Reference analysis Reference analysis Liquid Waste within 10CFR100 unchanged by unchanged by System Leak Cycle 13 reload Cycle 14 reload s Notes

1) The specific SDL criteria are defined in Table 5.1 as specified in (80).
2) The SAFDL on MDNB~. for cycles prior to Cycle 14 is 1.20. The Safety Analyses for these cycles do not incorporate the Statistical Combination of Uncertainties methodology in (77-80).

e +--er. _,, +i .w- ,-.-wrm --vvy,,-i- -, -,-. - + *.e wit, c. -iv eee --w-+me ---*.-----,.s -,u-* c-w-- e ,-uw s er-- wsm-w - - e - * - =-r ,_m--_ - + -

i TABLE 5.5 i Cycle 14 Reauired initial RCS Boron Concentrations i to Allow Marcin to Criticality for Dilutions from Shutdown Conditions with the RCS Filled (15 minutes) and RCS Drained (30 minutes) CEA Reautred initial Concentration (ppm) l Conflcuration Temperature. *F RCS Filled RCS Dmined 5%Ak/k W

ARI, BOC 550 J60 1-53 1092 500 1( 25 1514 1144 300 1038 1537 1193 68 1095 1495 1180 EOC 550 0

0 0 500 0 0 0 300 33 47 167 i 68 101 137 216 ARI Less 1 Stuck CEA, BOC 550 1129 1708 1245 l 500 1175 1735 1281 300 1307 1828 1392 68 1329 1814 1405 l EOC 550 0 0 41 i 500 0 0 156 l 300 223 311 350 i 68 292 399 405 ARI w/ Withdrawn Group. BOC 550 1275 1929 1382 500 1324 1955 1421 300 1368 1913 1448 68 1349 1842 1422 i EOC 550 18 28 196 500 135 200 294 300 324 454 446 68 370 505 475-ARI Less 2 Stuck CEAs.(2) BOC 550 1494 2260 1583 500 1539 2273 1620 300 1567 2191 1634 f 68 1530 2089 1593 EOC 550 165 249 334 500 279 412 429 l 300 458 641 573 68 495 675 594 -108- ] i

4 i TABLE 5.5 (Cont'd) Cycle 14 Reauired Initial RCS Boron Concentrations to Allow Marcin to Criticality for Dilutions from Shutdown Conditions with the RCS Filled (15 minutes) and RCS Drained (30 minutes) CEA Reoutred initial Concentration (ppm) I Configuration Temperature. *F RCS Filled RCS Drained 5%Ak/kW ARO. BOC 550 2113 3197 2152 500 2058 3040 2099 i 300 1884 2635 1930 68 1762 2406 1810 i EOC 550 725 1096 848 500 754 1113 866 300 755 1056 849 68 711 971 797

1) Margin of subcriticality required by Technical Specifications for shutdown conditions.
2) The critical boron concentrations for 2 stuck CEAs bound those for a withdrawn CEA group.

Therefore, the calculations for 2 stuck CEAs also bound intermediate combinations of 2 or more CEAs withdrawn during CEA rod drop testing during situations where the rods being dropped are from the same CEA group. 1 -109-l i

i I TABLE 5.6 l Summary of Boron Dilution Incident Results for Cycle 14 Minimum Required Minimum Time Acceptance Operating Technical Specification to Absorb SDM CriteriaW 1 Mode SDM (%AK/K) (Minutes) (Minutes) ( Refueling 5 30 30 j Cold Shutdown Filled RCS 5 >15 15 Drained RCS Sm 30W 30m l i Hot Shutdown Filled RCS 5 >15 15 Drained RCS 5m 30m 30m Startup 2 W/2 CEAs Withdrawn 3.2 ) 6F 15 i ARI 3.2W 95G 15 t Hot Standby 3.2W 50W 15 i Power Operation 3.2W 49W 15 i Failure to Borate Prior to Cooldown 3.22 56

  • 15 I

1) 30 minutes margin is used to provide sufficient margin for drained conditions where the head is removed. These are classed as " refueling" conditions in the Technical Specification. Margin quoted is for initial boron concentrations administratively required for these conditions. 2) Minimum value for shutdown margin specified in Technical Specification. l 3) Time to absorb minimum specified 3.2%AK/K shutdown margin. 4) Cooldown rate assumed to be 100 F/hr. 5) Time span between event initiation and criticality. j -110-i 1 a

I TABLE 5.7 i Summary of Boron Reauirements for Refueline Conditions for Cycle 14 f i i i Reouired ) Initial Concentration (ppm) i U RCS Filled to RCS Filled to 1 ft CEA Bottom of Hot Below Reactor Vessel 5%Ak/k ) f C Conficuration Lee Nozzles Flance (Normal Refueline) (w/uncertaintv) l 2 CEAs Withdrawn 1616 1393 1525 f ARI 1375 1185 1344 i b (1) The required boron concentration is the maximum of the initial boron concentration defined by the boron dilution analysis for the particular CEA configuration and RCS level combination, and the 5%Ak/k requirement. All boron concentrations include a 2%Ak/k [ uncertainty allowance. (2) Boron concentration corresponding to 5%Ak/k margin of subcriticality required by Technical Specifications for shutdown conditions. -111-

i [ TABLE 5.8 I Cycle 14 Nominal Scram Reactivity Worths Reauired to Prevent a Return to Power Durine a Steam Line Rupture Accident I Nominal Scram Reactivity (%Ap) l I Case Recuired_ Available t l BOC, HFP 4.51 7.54 BOC, HZP 3.07 5.89 EOC, HFP 7.31 8.08 l EOC, HZP 4.52 6.07 b f 6 l t r I t [ l -112-i i i I f i

l l i TABLE 5.9 i a Cvele 14 i CEA Eiection Accident Results i BOC EOC l HFP HZP HFP HZP Percent of Rods that Suffer Clad 11.81 0.0 11.61 0.0 Damage (Radial Average Enthalpy Above i 200 cal /gm), % i Percent of Fuel Volume Exceeding 4.52 0.0 4.53 0.0 Incipient Melting Criteria l (Enthalpy greater than 250 cal /gm), % j i i i 1 1 -113-1 l

TABLE 5.10 Comparison of Thermal Marcin for Limitine Cvele 14 Power Distributions to FSAR Desien Power Distribution SCU MDNBR Evaluation Compared to SDL of 1.57 Power Power YAEC-1 ) YAEC-1 ) 0 0 Level Distribution MDNBR M 100 (2)(3) 2.279 27 i 93 (2)(3) 2.388 29 t 81 (2)(3) 2.598 34 64 (2)(3) 3.031 37 r 56 (2)(3) 3.276 39 Deterministic MDNBR Evaluation Compared to SAFDL of 1.20 Power Power YAEC-1") YAEC-1W i Level Distribution MDNBR M 1 100 (2)(6) 1.761 21 100 FSAR(6) 1.764 22 100 (5)(6) 1.608 17 100 FSAR(5)(6) 1.793 24 i

1) Compared to full RCS flow SDL limit of 1.57 from Table 5.1 l
2) Limiting Cycle 14 power distribution within symmetric offset pretrip alarm band plus uncertainty for indicated power level.
3) At 2225 psia,554 F - remaining uncertainties statistically combined.

l t

4) Includes allowances for 2% calorimetric power uncertainty,3% tilt, and 10% physics radial peaking uncertainty on non-PSAR cases.

i S) Cycle 13. -{ i

6) At 2200 psia,556*F - remaining uncertainties deterministically applied.

-114-j 1

f I t TABLE 5.11 Reactor Protectivy System Trips l Assumed in the Cvele 14 Safety Analvsis 1 Delay Setnoint Uncertainty Time (Sec) l High Neutron Flux 106.5 % 15% 0.4 i 5 Low Reactor Coolant Flow 93 % 12% 0.65 High Pressurizer Pressure 2400 psia 122 psi 0.9 Low Steam Generator Pressure 500 psia 122 psi 0.9 Low Steam Generator Water Level 35% NR +10 in 0.9 I Low Pressurizer PressureW 1850 psia 32 psi 0.9 { Safety injection Signal 1600 psia 122 psi (2) l Thermal Margin Trip (TM/LP) Figures 5.5 (3) 0.9 l and 5.6 l Symmetric Offset Trip (SOTS) Figure 5.7 (3) 0.9 i Variable Overpower Trip (VOPT) Q + 10%W 5.5% 0.4 t i .i l l i 1 l

1) Low limit of thermal margin trip.
2) See specific accident for time delay assumed for safety injection delivery.
3) The uncertainties associated with these trips were statistically combined using the Statistical Combination of Uncertainties method in (80).'
4) Q = Initial indicated power level in percent thermal or nuclear power.

-115-

TABLE 5.12 Scram Reactivity Assumed in Cycle 14 Safety Analysis Scram Reactivity Wran) BOC EOC Event HFP HZP HFP HZP CEA Withdrawal 5.73 5.73 Boron Dilution 3.20 3.20 Excess Load 5.60 3.80 5.80 3.80 Loss of Load 5.73 Loss of Feedwater 5.73 Loss of Coolant Flow 5.73 CEA Drop CEA Ejection 5.80 3.80 5.80 3.80 Steam Line Rupture

  • 4.06 2.77 6,58 4.07 Steam Generator Tube Rupture 4.00 2.00 5.70 2.40 Seized Rotor 5.73 Loss of Coolant Maximum Assumed in Any Event 5.80 3.80 6.58 4.07 An uncertainty factor of 0.9 is applied to the nominal roquired scram reactivities assumed for

( the Steam Line Rupture event from Table 5.8 for comparison to the available scram reactivities with uncertainties assumed for the other events. This uncertainty component is statistically combined in the Steam Line Rupture analysis with the other uncertainty components to derive the nominal required scram reactivities for that event as discussed in 04). u -116-

TABl.E 5.13 Maine Yankee Cvele 14 12rce Break LOCA Analvsis Results Axial LOCA Peak Cladding) Cladding Power Limiting Limit Temperature OxidationO Shape Break (kW/ft) ("F) (% T' ekness) 52.08% 1.0C 15.78 1992 3.6 64.58 % 0.6G 14.35 2073 4.1 72.92 % 0.6C 14.10 2117 4.7 85.42 % 0.8G 12.80 2034 3.7 i

1) These PCT values are 10 F higher than Cycle 13 values due to cycle-specific input differences.
2) These are Cycle 13 values. An insignificant change is expected for Cycle 14 associated with a 10 F increase in PCT values. Hence, the values will continue to be much less than the 17%

oxidation lir at of 10CFR50.46. Less than 1% hydrogen generation is predicted for all cases I analyzed. 1 117-

TABLE 114 Maine Yankee Cycle 14 Small Break LOCA Analysis Results Axial Limiting LOCA Peak Cladding Cladding, Power Break Limit Temperature Oxidation 2 Shape (ft ) (kW/ft) ( F) (% Thickness) 52.08 % 0.15 15.78 1820 4.1 64.58 % 0.15 14.35 ISOS 4.0 72.92 % 0.15 14.10 1887 6.5 85.42 % 0.15 12.80 1739 2.9 Less than 1% percent hydrogen generation is predicted in all analyses. P W -118-

TABLE 5.15 Maine Yankee Safety Analysis Methodolocv Documentation Supporting Application Description of Methodolocv Documentation Reference in Cycle Safety Analysis Methods - YAEC-1202 1 5 Reference Reports YAEC-1160 11 5 YAEC-1132 36 3 YAEC-1099P 10,15 3 YAEC-1101 8,15 3 YAEC-1102 5,15 3 YAEC-1103 6,15 3 YAEC-1104 64,15 3 Reactor Protective System YAEC-1110 4,15 3 Setpoint Analysis - Reference Report Justification for Operation YAEC-1148 46 3 with a Positive MTC Modification of MSLR FMY-81-662 49 6 methodology to use Attachment A RETRAN 01 MOD 3 instead of FLASH-4 Change from use of W-3 YAEC-1296P 44,45 7 CHF Correlation to YAEC-1. Corresponding Reduction in SAFDL Limit from 1.30 to 1.20 Modification to MSLR YAEC-1447 14,18 9 methodology to use RETRAN 02 MOD 2 and BIRP as well as statistical combination of uncertainties Changes to the CEA ejection YAEC-1464 13,17 9 methodology to account for localized Doppler feedback weighting coefficients and post +jected peaking reductions -119-

TABLE 5.15 (Cont'd) Maine Yankee Safety Analysis Methodolocy Documentation Supporting Application Description of Methodology Documentation Reference in Cycle Modification to LOCA MN-86-141 57 10 methodology to include injection AP and multiple axial power shapes Modification to LOCA NMY-88-084 61 11 methodology to include i enhanced steam cooling model Modification of RPS YAEC-1642P 77-80 14 setpoints methodology to include statistical comb- { ination of uncertainties Modification of the Fuel MN-93-26 74 14 Perfonnance methodology using the FROSSTEY-2 computer code i Application of RELAP5YA to NYR 88-283 83 14 Small Break LOCA Analysis Modification of MSLR and YAEC-1752 81,82 Future. CEA Ejection methodologies cycles to incorporate STAR 3-D space-time analysis code. Provides MSLR return-to-critical capability i l l l -120-l f I l

2300 UNACCEPTABLE OPERATION ..+- -.4-...-... i e...... ....4... .....4.. ..4... P. ....,..4.. 9 ;....i.. .....4... ...;....;..i... ... f._ i f i ! ! i ...,...k.. ..,..s v, s.. 4..,, ' ~ i j l l' [ I I l l [ ' ..4..4...a ...).. i. ;. 4... ... i....;...;... 2250 c i i ...j i ! l . 4.. 4... ,........i.. ' :.. j.. 4... .. 4.. 4... * ... 4....j... I 4... ..4.. ....j...... ..4...... ..f... ..4... .l..... e t i g ACCEPTABLE OPERATION )....... 4... b ........a. .E i, ' ' ! f, i i o 2200 m .....t m ) c) i . t E + COORDINATES "+ '+" "f"..-f ~ + + - (500,2275) E 4..-.... !.. m. o (551.3,2275) e ^ (551,3,2225) - t~'i ~ t x (.O - 2150 (544,2075) -E (500,2075) .. p....;.. O (,) ... i...a.. 6. t i .. 1.. ...i... h.. cO I 1 ..4... j 2100 ! i ...... d.. ... ;...m 4 ...A... ..4... ,..4.. ....4... v ' 4 .......i........ ...;.....4.. ..........i..d.. g 1 -++ UNACCEPTABLE OPERATION ~+m- + t 2050 l l l 490 500 510 520 530 540 550 560 Nominal Cold Leg Temperature (indicated), deg F MAINE YANKEE Allowable 3 Loop Steady State Figure Cycle 14 Coolant Conditions 5.1 -121-

S4 2 s. g I i ~ i j l i 1 l 1 I .e lI = i '\\ m j I 1 \\ I I I \\ 1 \\ I i I 6 I I . e. E I I E 3 e i 1 o l -e 8 l .= o m I za e s. e. 3 c 3 l s e g .y c ac m 1 o m C - g M,.E Y\\ ~ - 523 o. m o n q N m W C

  • g _e _

C K.N z <.2 >[tcYAwrtt b9UII r u: rne ca. Des.ign Axis: ner Distribut. ion s.: -122-

FIGURE 5.3 MAINE YANKEE CYCLE 14 NORMALIZED REACTIVITY WORTH VS PERCENT CEA INSERTION BOC SCRAM AT HFP AND HZP 1.0 0.9 - 0.8 - 0.7 - 8 3 0.6-CsFO @ 0.5 - c Su Y 0.4 - E g 0.3 - 0.2 - HFP HZP 0.1 - 0.0 0 10 20 30 40 50 60 70 80 90 100 PERCENT CEA INSERTION -123-

FIGURE 5.4 MAINE YANKEE CYCLE 14 NORMALIZED REACTIVITY WORTH VS PERCENT CEA INSERTION EOC SCRAM AT HFP AND HZP 1.0 0.9 - 0.8 - 0.7 - E 8 l S 0.6 - l b 1 ?o @ 0.5 - c: B B< 0.4 -

s Oz 0.3 -

0.2 - HFP HZP 0.1 - 0.0 0 10 20 30 40 50 60 70 80 90 100 PERCENT CEA INSERTION 1 -124-

WHERE: ODNB = A)* OR$ TRIP AND PVAR = 2041.72 ODNB + 173TC - 10053.0 TC = COLO LEG TEMPERATURE, F 1.40 i ... i.. 9. t ... ^ I _. i +... l .......!....!.. 9 ... 9. m .. 9...... 9 I i _r+ ...q._ y ..i... a. .i....... ..._....i...... i.. 1.35 t, ....... +..., .+ .l-- ....9.... t 4. _. ~_b _~ j. .i_ i_ I ! .4_... l l ....!..:..I ..l.._... ~.......-,!.. .3... o .. 4...a.. 9.. .... 9 4... ....,. 4 a.... i_ [. 1.30 ~ l } r i r ~ l ... ;.... _; ___ _ __.L _i ...I.. i . i.. i ,....,..9.. ..p... y ,..9..

i. a.

.L.. ._ ;.i... ..4.... 9. . 4... ;.._.. A (+)=0.54457(S.O.)+1.01939 1.25 [...i.q... 4 l. .. 2 . 4. r .......9...7 y,_... 1... i 4 ..4......i.,.... I ......;...;...l ...l........ ...i.A.;_ Y -. ;..j _ ._4 t , 1.20 4 4. 2. t. ..4... .a._.,....... ._2 t. ........ j J........ i 4_ . _ _1 4._. 9_ ..a.4.. 4.. ...j... ..J... . i...; ;.. .. 4.. 1.15 I A (-) -0.25843(S.O.)+0.99080 ...J......; 4...p. l 3 ..k i ? I ... i.. ...i......... ...[.. A.

4...

i m..... ...f... 4. ,. 4. I,.. f 4...&... I I _...... _.. _,3_ 9 ......i.. ... i.. 4... .;i. .........L., i... L.,...i.. 4...i j ...i.. l .. _ +... -+ l l 1 4.. .. 4.. 4_.i... .....j.. 4. 1.05 h I i .i ..?...:.., l .. +....... .4.. -.......,i... ...p... i r l. l .y.. 9 i

7..

.i...... ...[. .....,.i... J T ...o.......... .... 4.. - -g _..., h d II 1.00 -0.5 -0.4 -0.3 -0.2 -0.1 0.0 0.1 0.2 0.3 0.4 0.5 Excore Symmetric Offset Y = A*((U-L)/(U+L))+B 3 MAINE YANKEE Triermal Margirttow Pressure Trip Setpoint Figure Cyck 14 (A versus Excore Symmetric Offset) 5.5 3 -125-

WHERE: ODNB = A ' OR$ 3 TR'P AND P AR = 2041.72 ODNB + 17.9TC - 10053.0 TC = COLD LEG TEMPERATURE, F 1.2 c ...... 4. 1.0 ~ 4 ..~.. ..._.....7.. 0.8 f'- ~ 7... 4... C 0.6 - ../. __._ O ._./ i ... +.. .. 7 ...a. (1.20,1.2000 ) (1.00,1.0000) .. 4.. ( 0.91,0.9800) 0.2 - g ( 0.66,0.8000 ) (0.10.0.1738) 0.0 0.0 0.2 0.4 0.6 0.8 1.0 1.2 Fracuon of Rated Thermal Powcr MAINE YANKEE Thermal Margirt' Low Pressure Trip Setpoint Figure 5.6 Cycle 14 (OR versus Fraction of Rated Thermal Power) 3 -126-

i 120 _-.} ..1........... ..4.._._..._ ...p.. .._.A e _.... I ...l.._._.1._.. ._1..4.......... ~ .........4.... . d, ...._-.~._i.... .......d... 110 g _. J... ._ * * - = =. * =. - ' --*** -+ .._.h.. d _ J.t _ - -. _, _ ~ _ - -. _....

7...

.. 7.. ._.....7_ -..g j .~ TRIP LIMIT 100 i, i, (-0.5000,15.0 ) .._.7........_ .+ b _. - I--.. ....4. .. f _..h... .... _.+ {.Q,$QQQ,$Q,Q ) 7 90 l (-0.2500,100.0) . 1 3_;, .~ t ( 0.0000.110.0) . +.._. +- - - -. - _ < ..r- .-y_- -.1-- j; 2.._; (0.2000,100.0) 80 .-.4 (0.5000.50.0 )

.-_.-...~~

i m (0.5000,15.0 ) o +_ 3; .. 4.....;... -... _. 4_.4. 4.... l a 4....; o ..a...,. U-70 J. y..- LCO ..a g .L...._.. [ .h.._.._ 1 " +- l ; +- (-0.4000,20.0 ) E -t-o e & 60 (-0.4000,50.0 ) 1-- _.j.. a..... e.-........... ... -. _.. ' (-0.1500,100.0) j.- .o ...(.. ~. o .p a .........7.... (.0.0100,105.5) C 50

l...

o l (0.1000,100.0) _;~ .. q.. (0.4000,50.0 ) Q ...l.. 1 I b.- ..._......j... ....-. 4.. a.. (0.4000,20.0 ) 40 { _7,.. _-t... ' ' IMPIN MODE LCO ...e e. 30 l l .._i.-.. ...._;... l... ...J l (-0.4000,20.0) .. q.._ ...p..y... .--...) (-0.4000,40.0) _p_. 9. ._.+--y..- 20 ...........j.._. (.0.1500,90.0) (-0.0100,95.5) +Y ...T . ~ ._ a.. .....7....... ~ " " ' ~ ' ~ "~' (0.1000,90.0) g._t_ i-- - t..- (0.4000,40.0) 10 - -t- -t... i ..i_.... .. 4 4 ._.4 (0.4000,20.0) 4 ...,...._4 ...p.. .....-.r ...i....a........_. j. ...e....-.4... -..-+-4.-..-e ...i._4. I _1._.i _. f.._ j.i... l

a..J...-.j_

..-..i.. e.... 0 i -0.6 -0.4 -0.2 0.0 0.2 0.4 0.6 Excore Symmetric Offset Y = A*((U-L)/(U+L))+B 3 MAINE YANKEE Symmetric Off set Trip Function Figure 5.7 Cycle 14 Three Pump Operation -127-

/ p 20 ~ i ...1 - 1..y._i i 8 ..........l..'...'. l...!.. i.~.. . T.. ...;....7. 7 . 7.. 7.. - -.7 18 g _ j.. . 4.. j.. .....L... ...L.! .. l... ..!...j.. ! L1..t......j 4... [.. .4... g a. . j.. 4 + UNACCEPTABLE OPERATION -.f. .4-4 16 %;,'..l. ! ......]..... \\ ...j... (f) .l......i ,'c ... i. - 5 .T. ...i.. . [.. .. q... ...i.. _.i..... [.. ...i.. ., j. %.L, J t i. -4.- .i i i i i 2 3y 4 .+. 3 ...l.... [. A....j.. ...l... .. L..;.. + ~ O l i . j.. . 4... i .17.. y .p. M 12 .._... I.~ ACCEPTABLE OPERATION ~ w H 10 ..!.{. '..{..; h E .. +.... 4.. .. 9.. ...j.... 4... .. +. Z t +- - + - -t "- r +1 +- ~ ~ - -+- O ~t~ + + + 1~ + i-8 . COORDINATES ...}......;.. .. 4 .+. ( 0, 15.78) ~ .*l' I. a,. ~1~ g 7 4 w z 6 (16, 15.78) .i-g . 7.. O . I.. ' ( 65, 14.33)

!.- i F

4 (73, 14.09) ~ ~ 1. .t i w i (( 93, 85, 12.84) .L -1 4 1.. ~+ ~b,- If~ I 4 . i.. -.+. 11.88) y f.. (100, 9.57) T t -I~ tt w Z .. j.. .. i.. .... ].. 4 0 O 10 20 30 40 50 60 70 80 90 100 CORE HEIGHT (%) Linear Heat Generation Rate (LHGR) Lirnits MAINE YANKEE Figure Versus Cycle 14 5.8 Core Height -128-

6.0 STARTUP TEST PROGRAM ) The startup test program includes low power physics and power escalation tests for the purposes of: 1) Verifying that the core is correctly loaded and there are no anomalies present which could cause problems later in the cycle; 2) Verifying that the calculational model used will correctly predict core behavior during the cycle. The low power physics tests are conducted at a power levelless than 2.0% of rated full power with a primary system temperature and pressure of approximately 525 F and 2250 psia, respectively. 6.1 Low Power Physics Tests The following reactor parameters are measured at the low power conditions: 1) Critical boron concentration is detennined at unrodded and, if required, selected rodded positions. 2) The integral worth at the hot zero power condition of CEA Groups 1,2,3,4 and 5 in the non-overlap condition. The total of the worths of these groups must be within 10% of the predicted value. If this condition is not met, then Groups B and C will be measured and the sum of the worths of all measured groups must be within1107c of predicted. 3) The isothermal temperature coefficient is measured by trending moderator temperature and reactivity changes. The measurement is performed at unrodded and,if required, a rodded condition. -129-

l 4) CEA drop times are measured by monitoring reed switch voltage for position indication versus time. All scrammable fulllength CEA drop times are measured. 6.2 Power Escalation Tests i Power escalation tests ensure the performance of various primary and secondary plant systems. Plant parameters are stabilized and test data taken at approximately 48% and approximately 100% of rated power. The following plant parameters are evaluated at the above power levels, or as indicated: 1) Core radial power distributions at essentially unrodded conditions at the above power levels are determined using the fixed incore detector system. 2) Isothermal temperature coefficients, if required, are derived at 48% power by partially closing the steam turbine governor valves, which increase reactor coolant system temperature. The result is a change in moderator temperature and power level, from which the coefficient is inferred. l 6.3 Acrerttance Criteria Acceptance criteria for the prediction of key core parameters are defined in Table 6.1. The acceptance criteria for Cycle 14 are unchanged from Cycle 13. The permissible deviations from predicted values are selected to insure the adequacy of the safety analysis. In these tests, the nominal measured value is compared to the nominal calculated value, the latter corrected for any difference between the measurement and calculational conditions. If the criteria in Table 6.1 are met, verification is obtained that the core characteristics conform to those assumed in the safety analysis. In particular, compliance with the shutdown margin Technical Specification is demonstra ted by the CEA worth and drop time measurements, provided all trippable CEA's remain operable. -130-l

If the initial criteria in Table 6.1 are not met, additional measurements, as prescribed by the table, are performed. In addition, any deviations are evaluated relative to the assumptions in the safety analysis for the given core parameters. Such deviations and their evaluations are reported to the S'.aff. A startup test summary report will be submitted within 90 days of the completion of the startup test program, in accordance with Technical Specifications. 9 3 -131-

TABLE 6.1 Maine Yankee Cvele 14 Startup Test Acceptance Criteria Measurement Conditions Criteria 1. Critical Boron Hot zero power, near Measurement within 11%Ap Concentration all-rods-out of predicted value 2. CEA Group Worths - Hot zero power, CEA Total worth within 110% Regulating Groups 1+2+3+4+5 in the of the predicted value non-overlap condition 3. CEA Group Worths - Hot zero power, CEA This measurement is not Shutdown Groups B+C+1+2+3+4+5 required if the criteria in in the non-overlap Measurement (2) is met. If condition the criteria in Measurement l (2) is not met, the total worth of all CEA groups measured must be within 110% of the predicted value i 4. Isothermal Hot zero power, near ITC measurement within 1 5 x 10%p/*F Temperature all-rods-out 0 Coefficient at of predicted value and the HZP MTC is in the acceptable region specified in Figure 4.8 5. Isothermal At or slightly below 50% This measurement is not Temperature power, near all-rods-out required if both criteria Coefficient at in Measurement (4) are met. 50% Power If either criteria in Measurement (4) are not met, the MTC mus* be in the acceptable region specified in Figure 4.8 6. Control Rod Drop Operating temperature Drop times to 90% insertion e Times no greater than 2.70 seconds 7. Radial Power At or slightly below Each assembly average Distribution 50% power, near all-power within 10% of rods-out predicted value 8. Tilt Monitoring 5-48% rated power, near Tilt trends to less than 3.0% for Symmetry all-rods-cut, tilt is for greater than 50% power Verification monitored at 5% power operation, as indicated by the intervals relative changes in excore or incere detector readings. -132-M- ___________m____

7.0 CONCLUSION

The results of analyses presented herein have demonstrated that design criteria as specified in the FSAR will be met for operation of Maine Yankee during Cycle 14. Table 5.4 summarizes the results of each incident analyzed, including the Reference Cycle result and the appropdate design limit. This table illustrates that Specified Acceptable Fuel Design Limits (SAFDL) on DNB and FCM, the primary coolant system ASME code pressure limit, and the 10CFR100 site boundary dose limits are not violated for any of the incidents considered. 1 i -133-

P

8.0 REFERENCES

1. YAEC-1202, Maine Yankee Cycle 5 Core Performance Analysis, P. A. Bergeron, et al., Attachment A to Maine Yankee Letter to USNRC, WMY 79-143, dated December 5,1979 l 2. Maine Yankee Letter to USNRC, WMY 77-75, dated August 1,1977. 3. YAEC-1479, Maine Yankee Cycle 9 Core Performance Analysis, April 1985. 4. YAEC-1110, Maine Yankee Reactor Protection System Setpoint Methodolozy, P. A. Bergeron, D. J. Denver, September 1976. 5. YAEC-1102, Maine Yankee Core Thermal-Hydraulic Model usine COBRA IIIC, R. N. Gupta, June 1976. 6. YAEC-1103, Maine Yankee Core Analysis Model usine CHIC-KIN, R. N. Gupta, September 1976. l 1 7. YAEC-1068, GEMINI-II - A Modified Version of the GEMINI Computer Procram, T. R. Hencey, April 1974. 8. YAEC-1101, Maine Yankee Plant Analvsis Model usine GEMINT-II, P. A. Bergeron, June 1976. 9. YAEC-1115, Application of Yankee's Reactor Physics Methods to Maine Yankee, D. J. Denver, E. E. Pilat, R. J. Cacciapouti, October 1976. 10. YAEC-1099P, Maine Yankee Fuel Thermal Performance Evaluation Model, P. A. Bergeron, February 1976 (Proprietary). I 11. YAEC-1160, Application of YANKEE WREM-BASED Generic PWR ECCS Evaluation i Model to Maine Yankee, July 1978. ] 12. YAEC-1324, h[aine Yankee Cvele 7 Core Performance Analysis, September 1982. 13. YAEC-1464, Modified Method for CEA Eiection Analysis of Maine Yankee Plant, December 1984. 14. YAEC-1447, Application of RETRAN-02 Mod 2 to the Analysis of the MSLB Accident at MYAPC T. D. Radcliff, M. P. LeFrancois, September 1984. 15. Letter, R. W. Reid (USNRC) to R H. Groce (MYAPC), Amendment No. 29, Cycle 3 Operation and Safety Evaluations for GEMINI-II and GAPEX, dated May 27,1977. 16. Letter, R. W. Reid (USNRC) to R. H. Groce (MYAPC), " Review of Application of Yankee WREM-Based Generic PWR ECCS Evaluation Model to Maine Yankee (3-Loop Sample Problem)," dated January 17,1979. -134-4 t i

17. USNRC Memorandum, L S. Rubenstein to G. C. Lainas, " Safety Evaluation Report of YAEC-1464, Maine Yankee Modified Method for CEA Ejection Analysis," dated June 20, 1985. 18. USNRC Letter to Maine Yankee, NMY 85-166 from E. J. Butcher,

  • Safety Evaluation of theMaineYankee AtomicPowerCorporation(MYAPC)ReportYAEC-1447, Applications of RETRAN02/ MOD 02 and BIRP to the Analysis of the MSLB Accident at MYAPC,"

dated October 2,1985. 19. Maine Yankee Letter to USNRC, WMY 75-28, Proposed Change No. 27, " Maine Yankee Core 2 Performance Analysis," dated March 27,1975. 20. Maine Yankee Letter to USNRC, WMY 78-62, Proposed Change No. 64, " Maine Yankee Cycle 4 Core Performance Analysis," dated June 26,1978. 21. Letter, E. L Trapp (CE) to R. T. Yee (MYAPC), M-CE-R-011, " Maine Yankee Batch N Reload Fuel Design Report," dated April 3,1985. 22. Maine Yankee Letter to USNRC, WMY 77-87, dated September 22,1977. 23. XN-NF-81-39P, Generic Mechanical Desien Report for Enon Nuclear Maine Yankee 14 x 14 Reload Fuel Assemblies, C. A. Brown, June 1981. 24. Letter, R. S. Freeman (CE) to R. T. Yee (MYAPC), M-CE-R-118, " Maine Yankee Nuclear Fuel Design Report - Update for Batch P," dated October 28,1986. 25. XN-NF-81-85, Mechanical Desien Report Supplement for Exxon Nuclear Maine Yankee XN-1 throuch YN-4 Extended Burnup Procram, November 1981. 26. XN-NF-86-94(P), Mechanical Desien Report Supplement for Exxon Nuclea r Maine Yankee XN-3 and XN-4 Extended Burnup, September 1986. 27. XN-NF-8206(P), Revision 1, Qualification of Enon Nuclear Fuel for Extended Burnup, M. J. Ades, June 1982. 28. Safety Evaluation of the Exxon Nuclear Company Topical Report, XN-NF-86-06(P), " Qualification of Exxon Nuclear Fuel for Extended Burnup," July 1986. 29. XN-73-25, GAPEX: A Computer Procram for Predictine Pellet-to-Claddine Heat Transfer Coefficients. August 1973. 30. BNWL-1695, COBRA HIC: A Dicital Computer Procram for Steady-State and Transient Thermal-Hydraulic Analysis of Rod Bundle Nuclear Fuel Elements, D. S. Rowe, March 1973. 31. Combustion Engineering Report TR-DT-34, The Hydraulic Performance of the Maine Yankee Reactor Model, June 1971. -135-

32 Maine Yankee Atomic Power Station Final Safety Analysis Report (FSAR), 33. CEND-414,The Evaluation and Demonstration of Methods for Improved Fuel Utilization, October 1983. 34. E. S. Markowski, L. Lee, R. Biderman, J. E. Casterlin, "Effect of Rod Bowing on CHF in PWR Fuel Assemblies", ASME paper 77-HT-91. 35. XN-75-32 (NP), Supplement 2, Computation Procedures for Evaluatine Fuel Rod Bowine, July 1979. 36. YAEC-1132, Tustification for 2630 MWt Operation of the Maine Yankee Atomic Power Station, P. A. Bergeron, P. J. Guimond, J. DiStefano, July 1977. 37. XN-NF-79-52, Maine Yankee Reload Fuel Desien Report / Mechanical Thermal-Hydraulic I and Neutronic Analyses,1979. l 38. Maine Yankee Letter to USNRC, WMY 78-102, " Maine Yankee Startup Test Report," dated j November 15,1978. 39. Maine Yankee Letter to USNRC, MN-83-76, " Reactor Vessel Pressunzed Thermal Shock (FI'S)," with Enclosures A, B, and C, dated April 22,1983. 40. Mairie Yankee Letter to USNRC, MN-84-88, " Reactor Vessel Pressurized Thermal Shock (FIS)," with Enclosures A, B, C, and D, dated June 1,1984. 41. Maine Yankee Letter to USNRC, MN-86-69, " Augmentation Factor Removal," dated May 20,1986. 42. USNRC Letter to Maine Yankee,," Augmentation Factor Removal," with Attached Safety Evaluation Report, dated June 20,1986. 43. YAEC-1259, Maine Yankee Cvele 6 Core Performance Analysis, Attachment to MYAPC Letter to USNRC, FMY-81-65, Proposed Change No. 84, dated April 28,1981. i 44. YAEC-1296P, DNBR Limit Methodolotv and Application to the Maine Yankee Plant, J. j Handschuh, January 1982 (Proprietary), Attachment to YAEC Letter to USNRC, FYR-82-41, MN-82-78, dated April 8,1982. 45. USNRC Letter to Maine Yankee, NMY 83-62, ' Topical Report YAEC-1296P, DhTR limit Methodology and Application to the Maine Yankee Plant," dated March 9,1983. 46. YAEC-1148, Justification for Operation of the Maine Yankee Atomic Power Station with a Positive Moderator Temperature Coefficient, P. J. Guimond, P. A. Bergeron, April 1978. 47. Maine Yankee Letter to USNRC, WMY 78-2, " Evaluation of Potential Boron Dilution Accidents," dated January 5,1978. -136-m

48. Ma!ne Yankee Letter to USNRC, MN-82-53, " Boron Dilution During Hot and Cold Shutdown (Mode 5 Operation)," dated March 18,1982. 49 Maine Yankee Letter to USNRC, FMY 81-162, Attachment, Cycle 6 MSLB Analysis, dated October 29,1981. 50. YAEC-1396, Maine Yankee Cycle 8 Core Performance Analysis, January 1984. 51. Maine Yankee Letter to USNRC, WMY 77-87, dated September 22,1977. - 52. " Acceptance Criteria for Emergency Core Cooling Systems for Light-Water Cooled Nuclear Power Reactors", Federal Register, Vol. 39, No. 3, Friday, January 4,1974. 53. Letter, R. S. Freeman (CE) to R. T. Yee (YNSD), " Maine Yankee Nuclear Reload Fuel Design Report - Update for Batch Q," M-CE-R-214, dated May 24,1988. 54. YAEC-1573, Maine Yankee Cycle 10 Core Performance Analysis, December 1986. 55. Letter, G. D. Whittier (MYAPC) to A. C. Thadani (USNRC), " Maine Yankee LOCA Analysis," MN-87-15, dated February 24,1987. 56. Letter, J. B. Randazza (MYAPC) to Director Nuclear Reactor Regulation, USNRC, MN-87-18, dated February 24,1987. 57. Letter, G. D. Whittier (MYAPC) to A. C. Thadani (USNRC), MN-86-141, dated November 10,1986. 58. Letter, G. D. Whither (MYAPC) to USNRC, " Update of Assessment 10CFR50.61 Fracture Toughness for Protection Against Pressurized Thermal Shock," dated December 6,1988. 59. YAEC-1648, Maine Yankee Cycle 11 Core Performance Analysis, July 1988. 60. USNRC Letter to Maine Yankee, NMY 88-147, " Issuance of Amendment to Facility Operating License No. DRP-36," dated September 27,1988. 61. USNRC Letter to Maine Yankee, NMY 88-84,

  • Evaluation of Maine Yankee Steam Cooling Model for Large Break," Tac. No. 65463, dated June 28,1988.

62. YAEC-1662, Lustification for 2700 MWt Operation of Maine Yankee for Cycle 11. December 1988. 63. Letter, R. S. Freeman (CE) to R. P. Barna (YNSD), " Maine Yankee Nuclear Fuel Design Report - Update for Batch R," M CE-R-298, dated October 31,1989. 64. YAEC-1104, Maine Yankee Plant Accident Analysis Model Usine FLASH-4, November 1976. -137-

65. YAEC-1713, Maine Yankee Cvcle 12 Core Performance Analysis, December 1989. 66. YAEC-1363-A, CASMO-3G Validation, A. S. DiGiovine, et al., April 1988. 67. YAEC-1659-A, SIMULATE-3 Validation and Verification, A. S. DiGiovine, et al., September 1988. 68. Letter, A. C. Thadani (NRR) to G. Papanic (YNSD), " Acceptance for Referencing of Topical Report YAEC-1659, SIMULATE-3 Validation and Verification," NVY 90-32, dated February 20,1990. 69. Letter, A. C. Thadani (NRR) to G. Papanic (YNSD), " Acceptance for Referencing of Topical Report YAEC-1363, CASMO-3G Validation," dated March 21,1990. 70. Letter, J. M. Betancourt (CE) to R. P. Barna (YNSD), " Maine Yankee Batch S Design Report," M-91-076, dated November 1,1991. 71. CE-NPSD-154, Natural Circulation Cooldown, Combustion Engineering, July 1981. 72. Letter, S. E. Nichols (MYAPC) to USNRC, " Update of PTS Assessment to Address the Revised PTS Rule (10CFR50.61)," MN-91-151, dated October 28,1991. 73. Letter, J. R. Hebert (MYAPC) to USNRC, "Use of the FROSSTEY-2 Methodology for Reload Analyses," MN-93-26, dated March 5,1993. 74. USNRC Letter to Vermont Yankee, NVY 92-178," Vermont Yankee Nuclear Power Station, Safety Evaluation of FROSSTEY-2 Computer Code (TAC No. M68216)," dated September 21,1992. 75. YAEC-1830, Maine Yankee Cvele 13 Core Performance Analvsis, December 1991. 76. Letter, J. M. Betancourt (CE) to R. P. Bama (YNSD), " Maine Yankee Batch T Fuel Design Report," M-93-001, dated January 25,1993. 77. YAEC-1642P, Maine Yankee RPS Setpoint Methodolocy Using Statistical Combination of Uncertainties, Volume 1, Prevention of Fuel Centerline Melt, March,1988. 78. YAEC-1642P, Maine Yankee RPS Serpoint Methodoloev Usine Statistical Combination of Uncertainties, Volume 2, Departure from Nucleate Boiline Limits, March,1988. 79. Letter, E. J. Leeds (NRR) to C. D. Frizzle (MYAPC), " Statistical Combination of Uncertainties Methodology for Maine Yankee Operating Limits", NMY 90-074, dated September 20,1991. 80. YAEC-1642P, Maine Yankee RPS Serpoint Methodolocv Usine Statistical Combination of Uncertainties, Volume 3, Bases for the Uncertainties Included in the Statistical Combinations, March,1993. -138-

81. YAEC-1752-A, STAR Methodolorv Application for PMTs. Volume 1, September 1990 and Volume 2, October 1990. 82. Letter, E. H. Trottier (NRR) to C. D. Frizzle (MYAPC), " Topical Report YAEC 1752, Space and Time Analysis of Reactors (STAR) Computer Code /' NMY 91-097, dated December 9,1991. 83. Letter, A. C. Thadani (NRR) to G. Papanic (YNSD), " Acceptance for Referencing Topical Report YAEC-1300P, Volumes 1,2 & 3, RELAP5YA: A Computer Program for Light Water Reactor System Thermal-Hydraulic Analysis," NYR 88-283, dated October 14,1988. .I -139- _ - _ _ _ - _ _ _ _ _ _ _}}