ML20247N395
| ML20247N395 | |
| Person / Time | |
|---|---|
| Site: | Peach Bottom  |
| Issue date: | 05/16/1989 |
| From: | Hesse S, Olson A, Waldman J PECO ENERGY CO., (FORMERLY PHILADELPHIA ELECTRIC |
| To: | |
| Shared Package | |
| ML20247N393 | List: |
| References | |
| PECO-FMS-0006, PECO-FMS-6, NUDOCS 8906050337 | |
| Download: ML20247N395 (208) | |
Text
{{#Wiki_filter:- _ _ _ _ _ _ _ _ 8 h . I METHODS FOR PERFORMING BWR RELOAD SAFETY EVALUATIONS Prepared By: . / 7. 5/l6l79 A. M. Olson, Engineer-Supervisory Date Fuel Management Section
/[M M M(/jp Prepared By:' f. P. Waldman, Engineer Date Fuel Management Section
/ ee - T /(, 39 Reviewed By:
S. R. Hesse, Engineer-Supervisory 'Da'e t Fuel Management Section Reviewed 3y: / W. Y Le/, SupeWising-Engineer c, A# 6Date88/ Fuel Management Section Approved By: da b*j L. F. Rubino, Superintendent Date Fuel Management Section Operating Licenses DPR-44 and DPR-56 Philadelphia Electric Company Nuclear Support Division 2301 Market Street Philadelphia, PA 19101
'8#06050337'890530 PDR ADOCK 05000277 P PDC
~ ---%_-.m. _m _____, _ __ _ _ _ _ _ _ _ _ _ _ _
DISCLAIMER 6 This document was prepared by Philadelphia Electric Company and is believed to be true and accurate to the best of its , knowledge and information. This document and the informa-tion contained herein are authorized for use only by Philadelphia Electric Company and appropriate subdivisions within the U.S. Nuclear Regulatory Commission for review purposes. With regard to any unauthorized use whatsoever, Philadelphia Electric Company and its officers, directors, agents, employees, and contractors assume no liability or make any warranty or representation with regard to the contents of this document or its accuracy or completeness. o lJ p U s q.
- a. -
t i i . . .
Acknowledgment This report represents the combined efforts of a number of Philadelphia Electric and contractor personnel. The authors would like to acknowledge the efforts of Mr. Vincent DeMasi of NUCOMP, Inc., for providing technical direction in its planning and preparation. Additional acknowledgments ,are extended to Mr. Greg Storey, Mr. James Tusar, Mr. William Gassmann, and Ms. Lesley Andres of the Fuel Management Section for their contributions in the preparation of the necessary technical data presented in the various sections. Final thanks go to M cs . Jennifer Dixon and Ms. Michele Polizzi for providing the stenographic services required in the typing and layout of the manuscript. ii
Abstract Philadelphia Electric Company (PECo) has developed methods for performing reload design and licensing evaluations of Boiling Water Reactors. This report provides a description and an integrated qualification of these methods and establishes a basis of confidence in PECo's ability to evaluate conservative plant operating limits (e.g., MCPR), Selected FSAR abnormal operational transient events have been evaluated to establish the licensing basis for Peach Bottom Atomic Power Station (PBAPS) using PECo methods. This licensing basis defines the evaluations to be performed for each core reload / cycle to determine plant operating limits. These methods, although qualified here for PBAPS, are applicable for analysis of other Boiling Water Reactors of similar designe l l iii
'N Table of Contents Pace ,
Disclaimer i Acknowledgment li Abstract iii Table of Contents iv List of Tables viii . List of Figures ix 1.0 Introduction 1-1 , 2.0 Description of PECo Computer Models and Methods 2-1 Employed in BWR Reload Design and Safety Evaluations 2.1 PECo Computer Code Sequence 2-1 . , , 2.2 Linkage Between Steady-State and Transient 2-13 Analysis 3.0 Qualification of RSE Methodology '3-1 3.1 Verification of Steady-State to Transient Model 3-1 Linkage: 3-D to Point Model Comparisons 3.1.1 Transformation of Feedback Variables 3-2 6 3.1.2 Void and Doppler Reactivity Verification 3-3 3.1.3 Treatment of Control Rod Worth 3-4 3.2 Verification of tteady-State to Transient Model 3-5 Linkage: 3-D to 1 D Model Comparisons 3.2.1 Power Distribution Calculation 3-5 3.2.2 Control Rod Worth Comparisons 3-8 l 3.2.3 Transformation of Feed.back Variables 3-9 3.2.4 Void and Doppler Reactivity Verification 3-10 iv - e
s Table of Contents (Cont.) Page 4.0 Reload Salety Evaluation Procedures for FSAR 4-1 Abnormal Operational Transients 4.1 Increase in Reactor Pressure Events 4-2 4.1.1 Generator Load Rejection Without Bypass 4-2 4.1.1.1 Description of GLRWOB Event 4-2 4.1.1.2 GLRWOB Analysis Assumptions 4-3 4.1.1.3 GLRWOB System Wide Response 4-4 4.1.1.4 GLRWOB Event Evaluation 4-15 4.2 Decrease in Cure Cool' ant Temperature Events 4-34 4.2.1 Feedwater Controller Failure (Max. Demand) 4-34 4.2.1.1 Description of FWCF Event 4-34 4.2.1.2 FWCF Analysis Assumptions 4-35 4.2.1.3 FWCF System Wide Response 4-36 4.2.1.4 FWCF Event Evaluation 4-47 4.2.2 Loss of Feedwater Heating 4-50 4.2.2.1 LFWH Analytical Assumptions 4-50 4.2.2.2 LFWH Analytical Methods 4-51 4.2.2.3 LFWH Event Evaluation 4-52 4.3 Reactivity and Power Distribution Anomaly Events 4-56 4.3.1 Continuous Rod Withdrawal Error During Power 4-56 Operation 4.3.2 Fuel Loading Error - Rotated 4-59 4.3.2.1 RBLE Analytical Assumptions 4-59 4.3.2.2 RBLE Analytical Methods 4-61 4.3.2.3 RBLE Event Evaluation 4-62 4.3.3 Fuel Loading Error - Mislocated 4-64 v
Table of Contents (Cont.) Pace 4.4 Increase in Reactor Coolant Flow Rate Events 4-65 4.4.1 Recirculation Flow Controller Failure 4-65 4.4.1.1 Description of RFCF Event 4-65 4.4.1.2 RFCF Analysis Assumptions 4-66 4.4.1.3 RFCF System Wide Response 4-67 4.4.1.4 RFCF Event Evaluation 4-68 4.5 Decrease in Reactor Coolant Flow Rate Events 4-78 4.5.1 Two Recirculation M-G Set Trip 4-78 4.5.1.1 Description of Two M-G Trip Event 4-78 4.5.1.2 Two M-G Trip Analysis Assumptions 4-79 4.5.1.3 Two M-G Trip System Wide Response 4-79 4.5.1.4 Two M-G Trip Event Evaluation 4-81 4.6 Decrease in Reactor Coolant Inventory Events 4-90 4.6.1 Loss of Feedwater Flow 4-90 4.6.1.1 Description of LOFW Event 4-90 4.6.1.2 LOFW Analysis Assumptions 4-91 4.6.1.3 LOFW System Wide Response 4-91 4.6.1.4 LOFW Event Evaluation 4-93 4.7 ASME Vessel Overpressure Protection Events 4-101 4.7.1 Main Steam Isolation Valve Closure with 4-101 Position Switch Failure 4.7.1.1 Description of MSIVC Event 4-102 l 4.7.1.2 MSIVC Analysis Assumptions 4-102 4.7.1.3 MSIVC System Wide Response 4-103 4.7.1.4 MSIVC Event Evaluation 4-105 vi
Table of Contents (Cont.) Pace 5.0 Reference Cycle Analysis Summary 5-1 5.1 Peach Bottom 3 Cycle 7 Licensing Analysis 5-1 6.0 References 6-1 Appendix: PEco Supplemental Reload Licensing Report A-1 vii
List of Tables Number Title Pace 3.2.1 Kappa-Sigma Fission Adjustment Factors 3-8 3.2.2 Comparisons of 3-D vs 1-D Void Reactivity 3-11 Worths 3.2.3 Comparisons of 3-D vs 1-D Doppler Reactivity 3-13 Worth For a 20% Increase in Power 4.1.1 Response Surface Fitting Coefficients for 4-19 GLRWOB S 4.1.2 Summary of RETRAN Model Uncertainty for a 4-30 Generator Load Rejection Without Bypass 4.1.3 Statistical Adjustment Factors and Scram 4-33 Speed Adjustment Factors for the PBAPS GLRWOB Event 4.2.1 Summary of RETRAN Model Uncertainty for a 4-48 Feedwater Controller Failure 4.2.2 Pesponse Surface Fitting Coefficients for 4-49 FWCF at EOC 4.2.3 Statistical Adjustment Factor and Scram Speed 4-49 Adjustment Factor for the PBhPS FWCF Event 4.2.4 PB3C7 LFWH Event Sensitivity Results 4-53 4.2.5 LFWH Event Exposure Sensitivity Study 4-54 4.2.6 LFWH Event Power / Flow Sensitivity Study 4-55 4.3.1 Rod Withdrawal Error Results 4-57 l Peach Bottom 3 Cycle 7 1 4.3.2 Peach Bottom 3 Cycle 7 RBLE Results 4-62 viii
List of Figures Number Title Pace 2.1.1 PECo BWR Analysis Computer Code Sequence 2-11 3.2.1 PB3C7 EOC Haling 3-13 3-D vs. Unadjusted 1-D Power Distribution 3.2.2 PB2C7 EOC-2000 3-14 3-D vs. 1-D Rod Worth Initially Rodded Core 3.2.3 PB2C7 EOC 3-15 3-D vs. 1-D Rod Worth Initially Unrodded Core 3.2.4 PB3C7 EOC-2000 3-16 3-D vs. 1-D Rod Worth Initially Rodded Core 3.2.5 PB3C7 EOC 3-17 3-D vs. 1-D Rod Worth Initially Unrodded Core 4.1.1 PB3C7 EOC GLRWOB 4-7 Vessel Steam Flow 4.1.2 PB3C7 EOC GLRWOB 4-8 Steam Dome Pressure 4.1.3 PB3C7 EOC GLRWOB 4-9 Core Reactivity Components 4.1.4 PB3C7 EOC GLRWOB 4-10 Neutron Flux 4.1.5 PB3C7 EOC GLRWOB 4-11 Core Avg. Heat Flux 4.1.6 PB3C7 EOC GLRWOB 4-12 Core Inlet Flow 4.1.7 PB3C7 EOC GLRWOB 4-13 Reactor Water Level 4.1.8 PB3C7 EOC GLRWOB 4-14 Feedwater Flow 4.1.9 Statistical Assessment of Transient RCPR 4-17 Limits ix
List of Figures (Cont.) Number Title Eagg 4.1.10 Response Surface Evbluation Box Matrix 4-19 4.2.1 PB3C7 EOC FWCF 4-39 Feedwater Flow 4.2.2 PB3C7 EOC FWCF 4-40 Vessel Steam Flow 4.2.3 PB3C7 EOC FWCF 4-41 Steam Dome Pressure 4.2.4 PB3C7 EOC FWCF 4-42 Core Reactivity Components 4.2.5 PB3C7 EOC FWCF 4-43 Neutron Flux 4.2.6 PB3C7 EOC FWCF 4-44 Core Avg. Heat Flux 4.2.7 PB3C7 EOC FWCF 4-45 Core Inlet Flow 4.2.8 PB3C7 EOC FWCF 4-46 Reactor Water Level 4.3.1 Limiting RWE Rod Pattern 4-58 Peach Bottom 3 Cycle 7 4.3.2 Rotated /NonRotated Differential Reactivity 4-63 P8DRB299-3G5.0/4G4.0 Bundle Unrodded, Core Average Void 4.4.1 PB3C7 EOC RFCF 4-69 Jet Pump Flow 4.4.2 PB3C7 EOC RFCF 4-70 Core Inlet Flow l 4.4.3 PB3C7 EOC RFCF 4-71 l Core Reactivity Components 4.4.4 PB3C7 EOC RFCF 4-72 Neutron Flux 4.4.5 PB3C7 EOC RFCF 4-73 Core Avg. Heat Flux 4,4.6 PB3C7 EOC RFCF 4-74 Vessel Steam Flow x
List of Figures (Cont.) Title Page Number PB3C7 EOC RFCF 4-75 4.4.7 Steam Dome Pressure PB3C7 EOC RFCF 4-76 4.4.8 Reactor Water Level PB3C7 EOC RFCF 4-77 4.4.9 Feedwater Flow PB3C7 EOC Two MG Trip 4-82 4.5.1 Recire Drive Flow PB3C7 EOC Two MG Trip 4-83 4.5.2 Core Inlet Flow PB3C7 EOC Two MG Trip 4-84 4.5.3 Neutron Flux PB3C7 EOC Two MG Trip 4-85 4.5.4 Reactor Water Level PB3C7 EOC Two MG Trip 4-86 4.5.5 Vebbel Steam Flow PB3C7 EOC Two MG Trip 4-87 4.5.6 Steam Dome Pressure PB3C7 EOC Two MG Trip 4-88 4.5.7 Core Avg. Heat Flux PB3C7 EOC Two MG Trip 4-89 4.5.8 Feedwater Flow PB3C7 EOC LOFW 4-94 4.6.1 Neutron Flux PB3C7 EOC LOTW 4-95 4.6.2 Core Avg. Heat Flux 4.6.3 PB3C7 EOC LOFW 4-96 Vessel Steam Flow PB3C7 EOC LOFW 4-97 4.6.4 Steam Dome Pressure 4.6.5 PB3C7 EOC LOFW 4-98 Core Inlet Flow xi
List of Figures (Cont.) Title Epagg Number 4.6.6 PB3C7 EOC LOTW 4-99 Feedwater/HPCI Flow 4.6.7 PB3C7 EOC LOFW 4-100 Reactor Water Level 4.7.1 PB3C7 EOC MSIVC A-106 Vessel Steam Flow 4.7.2 PB3C7 EOC MSIVC 4-107 Steam Dome Pressure 4.7.3 PB3C7 EOC MSIVC 4-108 Core Reactivity Components 4.7.4 PB3C7 EOC MSIVC 4-109 Neutron Flux 4.7.5 PB3C7 EOC MSIVC 4-110 Core Avg. Heat Flux 4.7.6 PB3C7 EOC MSIVC 4-111 Core Inlet Flow 4.7.7 PB3C7 EOC MSIVC 4-112 Reactor Water Level 4.7.8 PB3C7 EOC MSIVC 4-113 Feedwater Flow xii
1.0 Introduction This ' document is the last in a series of six reports prepared by Philadelphia Electric Company (PECo) for the purpose of qualifying BWR reload design and licensing methods. Previous PECo submittals to NRC in this' regard are listed as follows:
- 1) PECo-FMS-001, " Steady-State Thermal Hydraulic Analy-sis of Peach Bottom Units 2 and 3 Using The FIBWR Computer Code", February 1985 [1].
- 2) PECo-FMS-002, " Method for Calculating Transient Critical Power Ratios for Boiling Water Reactors (RETRAN-TCPPECO)", November 1985 [2].
- 3) PECo-FMS-003, " Steady State Fuel Performance Methods Report", July 1987 [3].
- 4) PECo-FMS-004, " Methods for Performing BWR Systems Transient Analysis", September 1987 [4].
- 5) PECo-FMS-005, " Methods for Performing BWR Steady State Reactor Physics Analyses", January 1988 [5].
These earlier submittals have described and qualified PEco analytical capabilities in each of the individual disciplines inherent to the BWR reload design and licensing process. The subject report, " Methods for Performing BWR Reload Safety Evaluations", describes and qualifies how these individual disciplines are implemented by PECo in the overall scope of BWR plant licensing. In this regard, particular attention has been given to the analysis of abnormal operational transients (AOTs)1 as defined in chapter 14 of the Peach Bottom Final Safety Analysis Report (FSAR) [6). Based on the FSAR approach, AOT events are classified into six event categories: 1) reactor 1 Transient events of moderate f requency are described as " abnormal operational transients" ( AOls) in the PBAPS FSAft. More recently, these events have been descrioed as
- anticipated operational occurrences" (LOOS).
1-1
pressure increase events, 2) core coolant temperature decrease events, 3) reactivity anomaly events, 4) core coolant flow increase events, 5) core coolant flow decrease events, and 6) reactor coolant inventory decrease events. A seventh event category has also been included in recent reload licensing submittals to evaluate ASME vessel overpressure protection events. PECo has reanalyzed the most severe event (s) in each of these seven categories, using the methods described herein. The results of these analyses for Peach Bottom 3 Cycle 7 are summarized in section 5.0, and as such constitute a reference cycle analysis for comparison to similar results derived for future reload designs. Conservatism in the event analyses is treated in a manner similar to that used by the current fuel vendor. In the case of the limiting rapid pressurization transients (i.e., generator load rejection without bypass and feedwater controller failure), detailed parametric studies have been performed, the results of which are used in the statistical evaluation of these events. To this end, generic Statistical Adjustment Factors (SAFs) have been determined for each of these transients, and are ultimately applied to the nominal p cycle-s'ecific licensing calculations. his technique derivec conservative plant operating limits at a 95/95 statistical confidence level. In the analysis of all other events, a conservative deterministic approach is followed. In these events, conservative values are coincidently used for the sLnsitive input parameters in the event analysis, thereby directly ensuring conservative analytical predictions for the plant operating limits. Finally, the Appendix to this report contains the format for a generic PEco reload licensing supplement which will be utilized to report the results of design and licensing 1-2
~
/
calculations for future cycles.. The use of this supplenent should standardize the reload licensin'g submittal process, ." thereby facilitating both the preparation and review of the important data. W s T t' a 4 s WP
- h 1
d' 4. D t b B 1-3 ..
),
_______m __________-______.____________m_ _ _ _ _ _ _ _ _ _ _ _
2.0 Description of PEco Computer Models and Mcthods Employed in BWR Reload Design and Safety Evaluations 2.1 PEco Computer Code Sequep::e Figure 2.1.1 depicts the integrated computer program sequence used by Philadelphia Electric Company (PECo) in the analysis of BWP. fuel and the reactor NSSS. With the exception of FROSSTEY and TCPPECO, all primary analysis and linkage codes were originally supplied to PEco by the Electric Power Research Institute (EPRI) as part of EPRI's Reactor Analysis Support Package (RASP) [7). The package includes BWR core physics, fuel performance, and thermal-hy-draulic programs, allowing for a wide range of steady-state and transient analysis applications, including fuel management', core design, calculation of plant Technical Specifications limits, and reload core licensing. The codes in the PEco sequence have been extensively benchmarked against plant data and higher order calculations via studies performed or sponsored by NRC, EPRI, national laboratories, 'and a number of electric utility companies. Documentation for much of this work is in the public literature, anc) has been previously reviewed by NRC in support of methods report cubmittals for other BWRs. Boiling water reactor analysis techniques embodied in these programs are state-of-the-art, and have a well established basis for application to the aforementioned tasks. With the exception of the SIMTRAN-E program, PEco has described and presented qualification information for all of the codes in the PECo sequence via previously submitted methods reports [1, 2, 3, 4, 5]. Additionally, a brief description of each of the PECo programs is presented here, including references to previous qualification efforts, where appropriate. 2-1 l
MICBURN [8] is a. one dimen.cional, cylindrical geometry pin cell depletion code which PECc employs in the development of 25 energy group microscopic gadolinium cross sections for a Gd2 03 loaded fuel pins. The program generates a table of effective gadolinium pin cross sections as a function of an equivalent total Gd-155 and Gd-157 number density over the expected burnup range. This gadolinium library is written to an external file for later use by the CASMO-1 single assembly lattice physics code. The version of MICBURN used by PECo remains unchanged from the original program developed by STUDSVIK ENERGITEKNIX for EPRI under research project 118-1. The program was originally reviewed and approved by NRC for Yankee Atomic Electric Coropany (YAEC) on Docket No. 50-271 [9] and is currently being reviewed by NRC for PECo applications on Docket Nos. 50-277 and 50-278 [5]. Co CASMO-1-PE_C_o [10] (hereafter referred to as CASMO-1) is a multi-groop, two dimensional transport theory lattice physics program developed by STUDSVIK to model the depletion of LWR fuel. For PECo applications the code is used to generate two-group average macroscopic cross sections and instrument response factors for input to the three dimensional rea: tor analysis program, SIMULATE-E. Neutron-ics data are developed as a function of the following BWR primary nodal parameters:
-Exposure History (E)
-Exposure Averaged Relative Moderator Density (VH)
-Control Rod Presence (CT)
-Instantaneous Relative Moderator Density (U)
-Fuel Temperature (TF)
-Moderator Temperature (TM)
-Xenon Concentration (CXE)
-Boron Concentration (CB).
2-2
'F i
CASMO-1 is also employed in the generation of few-group - cross section data for input to the PDQ-7-E fine mesh diffusion theory program. PECo's version of CASMO-1 is fundamentally the same as the original STUDSVIK ENERGITEKNIK AB program [11, 12] as ; originally reviewed by NRC for YAEC on Docket No. 50-271 [9] and is currently being reviewed by NRC for PECo applications . ! ]' ' on Docket Nos. 50-277 and 50-278 (5). n' NORGE-B-PEco [13] (hereafter referred to as NORGE-B) is a ' PECo enhanced version of the original EPRI NORGE-B code ; [14], and provides an automated data link between CASMO-1 - and SIMULATE-E. The program accesses data files as s - generated by CASMO-1 and prepares a SIMULATE-E input deck consisting of two group macroscopic partial cross section data. Two dimensional interpolating tables and lower order - polynomials are used to represent cross section dependencies .. on local fuel conditions. Fuel conditions are characterized - by parametric variations in fuel exposure, control rod presence, instantaneous relative moderator density, exposure averaged relative moderator density (void history), fuel > temperature, boron concentration, moderator temperature and - ' xenon number density. Polynomial fits for fission product yields and neutrons per fission are also generated by the , program. , , The fundamentals of the NORGE-B code have previously been e approved by NRC for Pennsylvania Power and Light (PP&L) on Docket Nos. 50-387 and 50-388 [15]. The applit.ation of the : 1 NORGE-B methodology to PECo core modeling techniques is , currently being reviewed by NRC on Docket Nos. 50-277 and 50-270 [5]. ::: 4
. .. 3 2-3 .
L
3 SIMULATF-E-PECo (16) (hereafter referred to as SIMULATE-E) is a three dimensional nodal analysis program used by PECo -e 4 in the nimulation of BWR cores in a state of nuclear and thermal-hyd'raulic equilibrium. It is an EPRI/PECo modified version of Yankee Atomic Electric Company's original SIMULATE program [17). The code models the reactor core as a matrix of neutronical.'.y coupled nodes, each of which is representative of a six inch axial segment of one BWR fuel assembly. At , , each node, SIMULATE-E accesses CASMO-1 generated two-group macroscopic cross section data (r, .r..r,.vr .Krf) as processed f by NORGE-P. Each of these cross sections are expressed in terms of the nodal fuel conditions, which are listed within the NORGE-B code description. The neutron source distribu-tion is then evaluated by solving the Modified Course Mesh Diffusion Theory (i.e., PRESTO) equations (other neutronics models are also available in the code). The fast and thermal neutron flux distributions are then inferred from the neutron source based on the steady-state balance of the slowing down and thermal capture reaction rates. Self-con-sistent nuclear and thermal-hydraulic solutions are assured
~'
by performing iterative calculations for the neutron flux . and steam void distributions.
+ ,
The FIBWR computer program has been incorporated into . SIMULATE-E to model BWR two-phase flow phenomena. In-chan- y nel and bypass flow distributions are calculated based on a detailed pressure drop analysis. FIBWR is called at the 5 ,, beginning of each void iteration. After completion of the . FIBWR flow balancing calculation, nodal qualities are -
~
determined from the most recent evaluation of the thermal - power distribution. At each node, the in-channel relative , moderator density is finally calculated using the Zolotar-Lellouche void quality profile fit model (i e. EPRI void correlation). Converged power and flow distributions 2-4 '
are ultimately used to predict margins to Technical Specifications operating limits, including Minimum Critical Power Ratio (MCPR), Linear Heat Generation Rate (LHGR), and Maximum Average Planar Linear Heat Generation Rate (MAPLHGR). The fundamental nuclear and thermal-hydraulic formulations in PECo's version of SIMULATE-E remain unchanged from those .' originally reviewed by NRC for YAEC on Docket No. 50-271 [18]. PECo/EPRI enhancements to the program are currently
~
being reviewed by NRC on Docket Nos. 50-277 and 50-278 [5]. COPHIN [19] 3s an EPRI sponsored code which provides an automated data link between the CASMO-1 multi-group s transport theory code and the PDQ-7-E/ HARMONY general geometry fine mesh diffusion theory sequence. The program accesses CASMO-1 nuclear data ffles and generates virtually complete PDQ-7-E/ HARMONY input decks for the same lattice, including few-group cross section tables, geometry input, mesh specifications, etc. The COPHIN qualification for application to design calculations was undertaken in report . PECo-FMS-005 [53 The COPHIN computer code is not directly applicable to the subject of this report (reload safety .- i evaluation) but is described here for completeness. PDO-7-E [20, 21] is a general geometry fine mesh neutron - diffusion theory program. The related HARMONY program models the depletion of fuel. PEco applies PDQ-7-E to the evaluation of the local (pin) fission rate distribution within a group of adjacent BWR fuel assemblies. Local peaking factors are then calculated accounting for the effects of flux gradients produced by control rods and/or , discimilar neighboring fuel assemblies. These accurate fine mesh solutions are ultimately used to benchmark the PINUP local peaking facter production code. 2-5
PECo's PDQ-7-E/ HARMONY is an EPRI released version of the PDQ-7 fine mesh diffusion theory program which is currently in widespread use throughout the industry. It is an enhanced version of an earlier Argonne National Laboratories release of the code [22), and has not been altered by PECo. The PDQ-7-E qualification for application to design calculations was undertaken in report PECo-FMS-005 [5]. The PDQ-7-E computer code is not directly applicable to the subject of this report (reload safety evaluation) but is described here for completeness. PINUP is a PEco developed production code used in the approximate evaluation of the 2x2 assembly geometry fuel pin fission rate distribution and associated local peaking factors. PINUP reads single assembly geometry (infinite lattice) fuel rod fission rate data from either CASMO-1 or l PDQ-7-E, and assembly power data frem either SIMULATE-E or , PDQ-7-E. Fuel rod fission rates in multi-assembly geome-l tries (non-infinite lattice) are then calculated by the program based on flux reconstruction techniques. The PlNUP qualification for application to design calculations was undertaken in report PECo-FMS-005 [5). The PINUP computer l code is not directly applicable to the contents of this l report (reload safety evaluation) but is described here for completeness. RWEASY is a SIMULATE-E post-processor which was developed by PECo in order to simplify the data reduction procers associated with control rod withdrawal error analysis. The program uses SIMULATE-E calculated Local power Range Monitor (LPRM) readings to predict Rod Block Monitor (RBM) responses as a function of error rod notch position and failed LPRM string conditions. The code then generates a thermal limit summary edit which lists MCPR, delta-MCPR and MLHGR as a 2-6
function of fuel type and error rod position. The qualification of RWEASY Rod Withdrawal Error methods was discussed in detail in report PECo-FMS-005 [5). SIGMA-PEC2 and TOPS are also SIMULATE-E post-processors. They rice used to perform straight forward statistical comparisons of SIMULATE-E predictions of Traversing Incore Probe (TIP) readings and in-core measured TIP flux trace data. These comparison statistics are ultimately used to infer uncertainties in SIMULATE-E predictions of nodal and assembly integral powers for reload design application. The application of SIGMA-PECo/ TOPS to the evaluation of SIMULATE-E nodal and assembly integral power uncertainty statistics has been discussed previously in report PECo-FMS-005 [5). The SIGMA-PECo and TOPS computer codes are not directly applicable to the subject of this report evaluation), but are included here for (reload safety completeness. SIMTRAN-E-PECo [23) (hereafter referred to as SIMTRAN-E) is a PECo enhanced version of the EPRI sponsored SIMTRAN-E code which was developed by Energy Incorporated under projects 1761-17 and 1761-14 as part of EPRI's Reactor Analysis Support Package [7). Its function is to radially collapse three dimensional cross section data generated by the CASMO-1/ SIMULATE-E sequence, perform perturbation calcula-tions on that data, and ultimately develop a series of l polynomials which represent primary thermal feedback mechanisms for input to the RETRAN-02 one dimensional l kinetics model. I PECo modifications to SIMTRAN-E include the incorporation of: 1) edits of core average delayed neutron parameters ( tr . A ) . 2) a kappa fission cross section adjustment option, 3) 2-7
a cross section normalization technique to account for differences between the SIMULATE-E and RETRAN-02 ther-mal-hydraulics models, and 4) the direct accessing of delayed neutron data from CASMO-1 data files. However, the fundamental radial collapsing (cross section, delayed neutron fraction, neutron velocities, etc.), perturbation, radial leakage correction, and polynomial fitting procedures in PECo's version of SIMTRAN-E remain identical to those included in the EPRI released version of the code. The EPRI version is described in GPU Nuclear Corporation Report TR-033-A [24] and has been reviewed by NRC on Docket No. 50-219 for the Oyster Creek Nuclear Generating Station. The qualification of SIMTRAN-E for PECo applications will be treated in detail in section 3.0 of this report. RETRAN-02-PECo [25) (hereafter referred to as RETRAN or RETRAN-02) is a complex, one dimensional, thermal-hydraulic transient analysis computer program. It is a variable nodalization code requiring the user to input a control volume / flow path network / heat slab model for the system to be analyzed. The program includes a variety of neutronics, thermal-hydraulic, heat transfer, system component, and special purpose models, as described in the EPRI RETRAN-02 documentation. The development of RETRAN-02 was sponsored by EPRI under RASP. Since that time, the code has been the subject of extensive verification / qualification efforts by EPRI and its contractors, NRC, and the nuclear utility community at large. The program has been approved by NRC for reference in licensing applications to the extent specified in the generic RETRAN-02 Safety Evaluation Report [26). Further-more, RETEAN-02 (including PECo minor modifications) has been approved by NRC for PECo applications [4). 2-8
FROSSTEY [27] is an LWR fuel rod steady-state thermal performance program. The code is designed to evaluate fuel rod pellet to cladding gap conductance, temperature distribution, dimensional changes, fission gas release, internal pin pressure, and stored energy as a function of fuel rod operating history and power level. For PECo Reload Safety Evaluation (RSE) applications, FROSSTEY analyses will be limited to calculating hot-channel and core average gap conductances for use in transient analyses. For each axial segment of a fuel rod, FROSSTEY calculates values for the thermal parameters involved by solving the one dimensional, radial, steady-state temperature dependent heat transfer equation with thermal-hydraulic boundary conditions specified by the user. Burnup dependencies are modeled and updated at each exposure step. For each exposure step, the code calculates the rod fission gas release and gap conductivity using the values calculated for each axial segment. FROSSTEY remains unchanged from the version reviewed and approved by NRC for YAEC on Docket No. 50-271 [28]. Qualification of the code for PEco applications was demonstrated in report PECo-FMS-003, which is currently being reviewed by NRC on Docket Nos. 50-277 and 50-278 [3]. TCPPECO [20] is used by PECo in the evaluation of steady-state and transient Critical Power Ratio (CPR) in BWRs. A system transient response is initially determined using the RETRAN-02 program. The system response, in turn, provides boundary conditions for a RETRAN-02 hot-channel - model. Output from the hot-channel calculation including nodal enthalpies, mass flow rates, bundle pressure and saturated liquid and vapor enthalples are ultimately input to the TCPPECO code. TCPPECO then utilizes the NRC 2-9 2-l _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ . _ _ _ _
approved, fuel vendor critical quality versus boiling length correlation to calculate CPR at each time step during a transient by an iterative procedure. An initial CPR estimate is made and the boiling length and local quality are recalculated based on this estimate. The critical quality from the critical power correlation is calculated based on the recalculated boiling length and compared to the revised local quality. This process continues until the revised quality is equal to the revised critical quality, and the estimated CPR which produces this condition is the converged CPR. The CPR is in this way calculated at each time edit. TCPPECO is a PECo modified version of the YAEC code TCPYA01. PECo modifications to TCPYA01 have been previously described in report PECo-FMS-002 [2]. The RETRAN-02/TCPYA01 methodol-ogy was originally reviewed and approved by NRC for YAEC on Docket No. 50-271 [30], and has been approved by NRC for PECo applications on Docket Nos. 50-277 and 50-278 [2]. FIBWR [31] is a computer code developed for the steady-state thermal-hydraulic analysis of BWRs. The program evaluates the two-phase flow / void distribution within the reactor core by solving the steady-state one dimensional equations of continuity, momentum and energy. 1 FIBWR was developed by the Yankee Atomic Electric Company and was made available to PECo through EPRI. The original FIBWR qualification and verification report has been published as a separate EPRI document [32]. FIBWR was originally reviewed and approved for YAEC on Docket No. 50-271 [33], and has been recently approved for PECo reload licensing applications on Docket Nos. 50-277 and 50-278 [1]. 2-10
F'GURE 2.1.1 ~ PECo BWR ANALYSIS COMPUTER CODE SEQUENCE [_ l PRMARY ANALYSIS CODE
'~
L....~~~ ] A m m CooE MICBURN LNKAGE CODE (GAD X. SECTIONS) U COPHIN m CASMO- W CO I ' NGLE ASSEMBLY PECO fATTICE PHYSICS)
+
PDO-7-E/ HARMONY i NORGE-B ; (MULTIPLE ASSEMBLY PECO LATTICE PHYSICS) . ',
.......Y PINUP ' SIMULATE-E-PECO
' > (WITH FBWR) "
L . .(LOCAL PEAK!NG..F,ACJpgs,( ,,,j (3-D SIMULATOR) l~ ~ ~ ~R ~ ~55 SE ~ ~ ~ ~ h 8 s' l.(ROD WITHDRAWAL....EggR(,,,,j , SIGMA-PEbO/TbPSI s s
- i. . . .(STATISTICS)............e P .
FROSSTEY SIMTRAN.E (FUEL PERFORMANCE) PECO 1 FIBWR m RETRAN (STAND ALONE * (SYSTEM TRANSIENT THERMAL. HYDRAULICS) ANALYSIS) _ VU m RETRAN TCPPECO - ; (HOT-CHANNEL) (THERMAL MARGtN) 2-11
l 2.2 Lin' cage Between Steady-State and Transient Analysis l l PECo reload safety evaluation methods employ the RETRAN one dimensional space-time kinetics model for the analysis of rapid pressurization transients (e.g., generator load l rejection, turbine trip, feedwater controller failure, and main steam line isolation valve closure). Conversely, the RETRAN point kinetics model is used by PECo for the analysis of slower events (e.g., loss of feedwater flow, two recirculation M-G set trip, and recirculation flow controller failure). Reactor kinetics input for the RETRAN point kinetics model consists of tables relating core average void reactivity to core average moderator density, core average Doppler reactivity to core average fuel temperature, and core average scram reactivity to axial control rod insertion distance. In additier., the core average six group delayed neutron parameters (B.A) are required. The point kinetics model reactivity parameters are generated by the 3-D steady-state physics code, SIMULATE-E. The delayed neutron parameters are obtained from CASMO-1. The calculational procedures used for this purpose have been described and qualified in the PEco physics methods report [5). The methodology associated with the RETRAN one dimensional space-time kinetics model is more complicated, and requires the generation of two-energy group macroscopic cross sections for each axial elevation within the core. This is accomplished by an automated procedure utilizing both the SIMULATE-E and SIMTRAN-E computer programs. To this end, SIMTRAN-E uses the three dimensional, two-group flux and cross section distributions predicted by SIMULATE-E to perfom a flux weighted averaging of these cross sections in both of the transverse (X and Y) directions. This process generates a set of average, one dimensional, axially dependent cross sections which, when used in the RETRAN one 2-12
i-l' ! i dimensional diffusion theory model, reproduce the SIMULATE-E ; core average axial flux and power distributions to a high degree of accuracy. The cross section collapse is performed by SIMTRAN-E using the equations presented in the SIMTRAN-E u::.er refere- e manual [23). In addition, SIMTRAN-E also generates the required core average six-group delayed ) ne'2 tron parameters (B. A). In - order to accurately represent reactor scram conditions, the cross section collapsing equations are solv 2d at a number of' intermediate control rod insertion densities between the beginning of scram (typically all rods out)'and the end of scram (all lods in) states. For licensing applications, PECo executes a series of computations consisting of separate three dimensional SIMULATE-E cases at 0 ft, 1.ft, 3 ft, 6 ft, and 12 ft control rod bank insertion distances. The first of these cases (0 ft insertion) is solved with thermal feedback turned on, while the remainder of the cases assume no thermal feedback (i.e., power distribution is fixed and taken from the first case). .Next, a SIMTRAN-E case is executed which reads the SIMULATE-E fast and thermal flux distributions'at each control rod insertion state. At each control rod state, perturbation calculations are performed by changing the thermal-hydraulic feedback variables, i.e. the fuel temperature, T, F and the relative moderator density, U. The effect of each of these perturbations on each of the macroscopic cross sections is determined using the standard SIMULATE-E cross section formulation. The perturbed three dimensional cross section distributions are then collapsed to one dimensional core average axial distributions using the cross section collapsing equations discussed previously. The unperturbed three dimensional flux distributions are used for this l l 2-13
purpose. Finally, at each axial node, the collapsed one dimensional macroscopic cross sections are related- to changes in Ty and U using a polynomial fit representation: r = Co + C laE + C JU 2 + C 34UJE + C4W 2 + C5 g2ag A set of polynomial coefficients is generated by SIMTRAN-E to represent the perturbed and unperturbed one dimensional core average cross sections for each of the control rod states. This data, collectively, is written by SIMTRAN-E to an . e::ternal file for input to RETRAN. In order to accurately account for the effects of core radial leakage in the one dimensional diffusion theory solution, SIMTRAN-E calculates a fast group radial buckling parameter, B12(K), at es_ch axial node, K. The equations used for this purpose are described in the SIMTRAN-E user reference manual [23). For each axial node, polynomial coefficients for B12 (K) are written by SIMTRAN-E to the same external file used for linking the cross section data to RETRAN. As a verification of the numerical accuracy of these cross section collapsing techniques, SIMTRAN-E performs a fine ) mesh one dimensional diffusion theory calculation for the I fast and thermal neutron group fluxes. This calculation is performed with the equations emp3cyed by RETRAN for this same purpose. The top and bottom axial reflector regions are explicitly represented in this calculation using a set of user supplied cross sections. Adjustments to the reflector region Di coefficients are made in SIMTRAN-E to better match the results of the equivalent axial albedo assumption used in SIMULATE-E. In this regard, zero flux boundary conditions are imposed at the top and bottom boundaries of the reflector regions. The SIMTRAN-E predicted fine mesh fast and thermal flux distributions are 2-14
]
)
u further processed into the form of an axial nodal power distribution. The latter is directly compared to the SIMULATE-E core average axial power distribution, and difference statistics are generated. A typical comparison between the SIMULATE-E and SIisTRAN-E (and hence, RETRAN) core average axial power distributions yields a maximum nodal difference of less than 4%, with the RMS of the nodal differences being less than 2%. In addition to the linkage techniques included in the EPRI distributed version of SIMTRAN-E, PECo has incorporated some additional code changes into SIMTRAN-E to improve the agreement between the 3-D (SIMULATE-E) and 1-D (RETRAN) models. These changes are further described in section 3.2 of this report, and involve:
- 1) the use of Krf adjustment factors to further improve the agreement between the 3-D model and 1-D model core average axial power distributions, and
- 2) a transformation of the nuclear feedback thermal-hy-draulic variables (TF and U), in order to better match the 3-D model (SIMULATE-E) and 1-D model (RETRAN) predictions of reactor system respense to changes in reactor core fuel temperature and pressure.
2-15
3.0 Qualification of RSE Methodology PECo's steady-state physics methods used in the calculation
. of reactivity parameters have been qualified in an' earlier submittal to NRC [5]. PECo's transient analysis methods have likewise been qualified using a "best-estimate" model to predict actual plant transients and test data [4). In order to qualify PECo methods for application to safety analysis, tvo demonstrations remain, namely:
- 1) the linkage. between the physics and transient analysis models preserves the reactivity. effects and power distribution predicted by the physics model, and
- 2) the safety analysis methods include appropriate assumptions to ensure a conservative evaluation of operating limits with a high degree of confidence.
The first of these requirements will be addressed in the present section. Section 4.0 will deal with Reload Safety Evaluation (RSE) procedures implemented by PECo to satisfy the second requirement. 3.1 Verification of Steady-State to Transient Model Linkage: 3-D to Point Model Comparisons The RETRAN point kinetics model requires input tables for core average reactivity parameters (i.e., void, Doppler, scram reactivities). These parameters are generated using the three dimensional (3-D) core simulator code, SIMULATE-E, and are designed to accurately predict the transient ~ kinetics behavior of the reactor core. The 3-D simulator is the most appropriate choice for the source of the point kinetics perameters as it has been extensively qualified against plant operating data (5). This section verifies the , accuracy of the generation and application of the point kinetics parameters. 3-1 1
3.1.1 Transformation of Feedback Variables The one dimensional core average axial power distribution calculated by _ SIMULATE-E is input directly to ' the RETRAN point kinetics model. Likewise, the SIMULATE-E calculated point kinetics values for void and Doppler reactivities'are input. to RETRAN in the form of tables of reactivity versus the feedback variable. The feedback variables used by the RETRAN point kinetics model are core average moderator density for void reactivity and core average fuel temperature for Doppler reactivity. The 3-D simulator also utilizes these same feedback variables on a nodal basis. However, the models within the 3-D simulator calculate slightly different core average feedback values than the corresponding RETRAN values. Due in part to differences in the averaging techniques used by RETRAN and SIMULATE-E, the core average moderator density and fuel temperature for identical reactor system conditions are not identically predicted by the two codes. However, the differences between the SIMULATE-E and RETRAN core average feedback variables are small (less than 2.0%). To preserve the 3-D model reactivity feedback effects in the RETRAN-02 model, the following procedure is performed. In the point kinetics case, SIMULATE-E and RETRAN cases are executed in order to predict changes in the core average l feedback variables as caused by identical changes in core-wide reactor conditions (e.g. core thermal power or total core flow). A transformation of feedba::k variables then substitutes the RETRAN core average values of moderator density and fuel temperature for the SIMULATE-E values in \ l the void and Doppler reactivity tables input into the RETRAN point kinetics model. This transformation results in the preservation of the 3-D model reactivity characteristics for a given change in system parameters in the point kinetics calculation. The transformation yields a small improvement 3-2 l
L and-is' performed primarily to maintain consistency with the 1-D kinetics application, where the effect is more significant. The 1-D transformation is described in section L 3. 2. - 3.1.2 Void and Doppler Reactivity Verification The ability of the RETRAN point kinetics model to accurately l predict core response using the void and Doppler reactivity tables was . verified' by comparison to predictions from the 3-D model. To accomplish this, .a flow increase transient was simulated with both the RETRAN point- kinetics and the-SIMULATE-E 3-D steady-state models. First, a series of SIMULATE-E and RETRAN cases was executed to generate the void and Doppler reactivity tables for use in RETRAN. This series of cases represented the initial conditions ' for the transient, followed by instantaneous -incremental step changes in core flow, and instantaneous incremental step changes in. core fuel temperature chosen to bound the range of'the feedback variables during the transient. SIMULATE-E void and Doppler reactivity predictions were tabulated as a function of the RETRAN code predictions of average in-core moderator densit; and fuel temperature for each case executed. Next, using this kinetics input, the flow increase transient was simulated with the RETRAN point model, allowing reactivity feedback to occur, until a new reactor equilibrium state was predicted at the end of the transient. These new reactor conditions (power, flow, subcooling, pressure) were then input to a SIMUIATE-E branch case with thermal feedback. This case was allowed to ' converge to a new k-effective value with a fixed power shape and fission product distribution, as assumed by the RETRAN point kinetics model. The k-effective difference predicted by SIMULATE-E between the initial and final equilibrium states was small. This result indicates that RETRAN and SIMULATE-E model predictions of changes in core thermal i
)
3-3
m= i J power differ by_'less than 3.4%. 'This verifies that the methods used to develop _the void and Doppler point kinetics tables in'RETRAN reproduce the SIMULATE-E predictions to a high precision. 3.1.3 Treatment of Control Rod Worth The treatment of control rod worth is very straightforward. The 3-D simulator static rod worth is input to the RETRAN-point kinetics model as a table of reactivity versus ' rod position. The use of the static rod worth'to model a scrara is conservative compared to a dynamic rod worth treatment since . the' offect of delayed neutrons is not incorporated. Because 3-D model scram worth predictions are input directly into the RETRAN point kinetics model, preservation of control. rod reactivity effects is ensured in the linkage between the two models. No further verification of the control rod worth used in the point kinetics model is required. i , { 1 t 3-4 ; I
m,
\ ,
) 3.2' Verification of Steady-State to Transient Model Linkage: 3-D to.1-D Model. comparison A general description of.the generation of one dimensional' l (1-D) reactor-kinetics parameters (cross sections) using the SIMTRAN-E computer code was presented earlier in section 2.2 of this report. These cross sections'are based on the three a dimensional; core simulator code, SIMULATE-E, and are designed. to- accurately predict the transient kinetics. behavior of the reactor core. As stated previously, the 3-D. simulator is the'most appropriate choice for the source of the cross' section data. Thus, it is important to demonstrate that RETRAN calculations utilizing SIMTRAN-E processed 1-D- cross sections reproduce the reactivity characteristics of 'the 3-D simulator accurately. This section verifies the . accuracy of the generation and application of the 1-D cross sections. 3.2.1 Power Distribution Calculation i It is-desirable that the 1-D base cross sections reproduce the 3-D simulator average axial power distribution as accurately as possible while also preserving the critical eigenvalue. The 3-D to 1-D collapsing techniques ' used by SIMTRAN-E are . designed. to meet this goal. However, differences ' in .model assumptions between the 3-D simulator and SIMTRAN-E limit the agreement between the 3-D and 1-D calculations. In order to account for radial leakage effects in the 1-D-calculation, SIMTRAN-E evaluates the core average fast group
' radial- buckling parameter, B12(K), at each axial node, K. !
Polynomial coefficients representing B12(K) are passed from SIMTRAN-E to RETRAN, together with the analogous polynomial
' coefficients for .the collapsed 1-D macroscopic cross sections. When used in the RETRAN one dimensional diffusion !
equations, the SIMTRAN-E collapsed nuclear parameters result ; i 3-5 i i I _ _ _ _ _ _ _ _ _ _ . _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ . _ _ _ _ _ _ . . _ _ _ _ _ _ _ _ . _ _ _ _ _ _ _ _ _ . _ _ _ _ _ . _ _ _ _ _ . _ _ ._d
l 1
.in an ' axial power distribution which is in reasonable agreement with - ' that - predicted by SIMULATE-E (differences less ' than 3%) at'all fuel nodes not.on the core boundary.
With regard to the core boundaries, SIMULATE-E does not model the axial core reflector regions explicitly as unfueled' nodes (as SIMTRAN-E does) but uses an albedo method L to calculate. axial leakage. SIMTRAN-E requires cross h sections. to be input for the reflector regions. This limitation prevents the calculation of the fast group radial buckling term for the top and bottom fueled nodes. SIMTRAN-E assigns values from the adjacent fueled nodes to these locations. SIMTRAN-E has several optional solution methods available to further improve the 1-D power distribution calculation. One method iteratively adjusts the radial buckling term in all fueled nodes until the average nodal power error meets a given convergence criteria. This method produces excellent agreement between the 3-D and 1-D power distributions but unacceptably biases the critical eigenvalues. Other methods include. adjusting the radial buckling at only the top and bottom fueled nodes and/or adjusting the axial reflector fast diffusion coefficients (D1) to'be consistent with the SIMULATE-E albedos. These methods yield improved (but not exact) 1-D power distribution predictions without introduc-ing significant biases in the critical eigenvalues. In order to match the 3-D simulator power distribution as accurately as possible, PECo has added a separate option to the GIMTRAN-E code which adjusts the kappa-sigma fission cross sections of the fueled nodes. This option, in conjunction with the methods already available in SIMTRAN-E, is used to match the 3-D power distribution as accurately as possible without biasing the critical eigenvalues. The overall procedure used is as follows: 3-6
4 i i) The SIMTRAN-E options to adjust the fast diffusion ' coefficient- (D1) of the axial reflectors -and the radial-- leakage of the top and bottom fueled nodes are chosen.
- 2) The axial reflector thermal group absorption and fast group scattering cross sections may be adjusted to further improve the power distribution calcula-tion.
l
- 3) Upon. completion of 1) and 2) above,-the differences between the SIMULATE-E and SIMTRAN-E core average axial power distributions are small (less than 3%) .
However, for further improvement, the kappa-sigma-fission cross sections in the RETRAN 1-D solution are adjusted to exactly match the 3-D core average - axial power distribution.
' Figure'3.2.1 presents a comparison of the 3-D average axial power. distribution to the 1-D power distribution without the kappa-sigma fission cross section adjustment for PB3C7 at end-of-cycle. The agreement is very good, indicating that in spite of the code differences discussed above, SIMTRAN-E can accurately reproduce the 3-D average axial power distribution.
It should be noted that the kappa-sigma fission cross sectio:. 9ustments affect'the power distribution only. The two-group neutron flux distributions are not affected, thus, the nodal reactivity. characteristics are not biased (this will be further demonstrated in the following sections). Additionally, the adjustments made are small in magnitude. A. list of the nodal absolute maximum and average adjustments made in the analysis of Peach Bottom 2 Cycle 7 (PB2C7) and Peach Bottom 3 Cycle 7 (PB3C7) at end-of-cycle (EOC) and end-of-cycle minus 2000 MWD /ST (EOC-2000) exposure condi-tions.is presented in Table 3.2.1. 1 I 3-7 I i --.;;_._-._._- __ _ -_- -_._:_- ~ _ _ _ . . .-_--_--..___--_______-__._____--_-_______.._..___a
r.; V; Table ' 3.~2.1 ) k Kappa-Sigma Fission Adjustment Factors . . Statenoint May(%i Ay_cif1)_ P2C7 EOC - 2000' 2.34' 0.68 P2C7 EOC 2.31 0.84 P3C7 EOC - 2000 2.22 0.69 i P3C7 EOC 2.55 0,92 l Average. 2.36 .0.78 3.2.2 Control Rod Worth Comparisons The ability of the 1-D cross section model to reproduce the 3-D simulator' control rod worths is '.important in order to accurately' predict- transient core kinetics response. To-test' ' the - RETRAN 1-D multiple control state rod model, a c'mparison o was made-between the 3-D simulator static control
. rod worths and the'l-D model predictions for PB2C7 and PB3C7 at EOC and EOC-2000 exposure conditions. For each
.statepoint, the control rod bank was inserted into the core in one foot intervals to determine the rod worths. Tigures i- 3.2.2Lthrough 3.2.5 illustrate the results of the analysis.
'The agreement'in the predicted rod worths is excellent. The
.overall mean percent difference (1-D minus 3-D) and root-mean-squara (RMS) percent ' uncertainty are -0.43% and 1.38%, respectively. Over the first six feet of rod insertion, the mean percent' difference is -2.69%, indicating that~the 1-D control rod representation is conservative over the most important range of rod motion. This analysis verifies that' the 1-D control rod model accurately reproduces, with a slight conservative bias, the 3-D simulator control rod worths.
3-8
r 3.2.3' Transformation of Feedback Variables l The thermal-hydraulic nuclear feedback variables used by the RETRAN-1-D kinetics model are nodal average fuel temperature and nodal normalized. moderator density. The'3-D simulator also utilizes these feedback variables. 'However, the models used by the.3-D simulator to calculate these variables are somewhat different from the models employed.by RETRAN. Thus, given identical core conditions of power, flow, pressure, subcooling, and average axial power distribution, RETRAN would.not.necessarily reproduce the average axial moderator density or fuel temperature distributions predicted by the 3-D simulator. More importantly, RETRAN represents the core as.a single channel with the average core power. (and power distribution) and flow. The 3-D simulator models- each channel individually. Thus, for a given change in reactor system conditions (e.g., increased reactor pressure or power), the average change in the nodal values of the feedback variables at each axial level (as calculated by the 3-D simulator) may not agree with the change in the feedback , variables of the average node (as calculated by ' RETRAN) , even if the thermal-hydraulic feedback models in each code are identical. It is necessary, therefore, to transform the feedback variables to equivalent RETRAN values when fitting the collapsed 1-D cross sections rather than using the 3-D simulator values. This transformation preserves the 3-D reactivity characteristics for a given set of system perturbations (e.g., pressure, power) in the 1-D calcola-tion. The methods used to perform the transformations are similar to those described for the generation of point kinetics, but applied on a nodal basis as opposed to a core average basis. 3-9
.- ;}
~
3.2.4 Void and Doppler Reactivity Verification
'To demonstrate the accuracy of the 1-D cross section generation methods, a series of comparative calculations
.between the 3-D simulator and RETRAN were performed. These calculations were designed to ascertain any uncertainties in the 1-D reactivity components (void & Doppler) created when collapsing the 3-D cross sections.
'To verify that ' RETRAN accurately reproduces the 3-D void reactivity, predictions of instantaneous void reactivity worth at~various conditions were made and compared to the corresponding 3-D ' simulator values. Because the typical licensing transient . analyzed with 1-D kinetics results in insertion of'the control rods during a reactor pressuriza-tion, some of the conditions chosen included both an
- c. increase in reactor pressure and insertion of the control rods to reflect actual reactor conditions during a transient event (e.g. GLRWOB). These conditions, and the results of the 3-D and 1-D predictions, are presented in Table 3.2.2 for PB2C7'and'PB3C7 at EOC and EOC-2000 exposure conditions.
Based on the twelve cases analyzed, the overall mean percent difference (1-D minus 3-D) and root-mean-square (RMS) percent uncertainty are -0.2% and 2.6%, respectively, indicating that the 1-D cross section collapsing methods introduce very litt.le bias or uncertainty in the void reactFrity of the 3-D mode). 3-10
;, g.y - -
3 y; Table 3.2 2 Comparisons of 3-D vs 1-D Void Reactivity Worths, Pressure Control- Void Worths Increase Rod Inser- 3-D- 1-D Statepoint~ (PSI) ' tion (ft) (4K/K) * ( AK/K)
- PB3C7 50- 0 4.446 4.323 EOC~ .50 1 4.692 4.606 100 3 9.639 9.591 PB3C7 50 0 4.458 4.163 EOC-2000- 50 1 4.745 4.570 100 3 9.622 9.758 PB2C7. 50 0 4.553 4.586 EOC 50 1 4.827 4 862 100 3 9.962 10.108 PB2C7 '50' O 4.494 4.388
' EOC-'2 0 00 50 1 4.885 4.840 100 3 9.962 10.323 Mean void Worth:* 6.357 6.343 Mean Difference:* -0.014 STD Difference:* 0.165 (Mean-Difference /Mean) x 100% : -0.22%
(STD Difference /Mean) x 100% : 2.60%
- All values expressed in units of 10-3 JK/K
'To verify that the 1-D cross sections reproduce the 3-D Doppler reactivity, comparisons were made between the
~ SIMTRAN-E 1-D Doppler reactivity edit and the corresponding
- 3-D - simulator calculation for a 20% increase in reactor
}
power. The results of this analysis are presented in Table g 3.2.3. The 1-D cross sections conservatively underestimate the 3-D Doppler reactivity by approximately 5.0%. s i I i l. i l \ l 3 3-11 , j:
~
mvmm - ,
, ,4
~. J9 m ';
- p. ;
+
MJ Table 3.2.3 s ' Comparisons of 3-D vs l'-D Doppler Rome.tivity Worth For a 20% Increase in-Power m Doppler Worth
t.
3D- '1D. Statepoint (4K/K) * . (dK/K)
- I9 * .
PB3C7 - EOC.; -1.578 -1.500. PB3C7 EOC-2000 -1.696 -1.586: PB2C7 EOC', -1.>510 -1.466' PB2C7 EOC-2000 -1.671 -1.576
'Meanl Doppler Worthi* -1.614 -1.532 :j Mean Difference:* 0.082 l
.STD Difference:*' 0.025 i
-(Mean Difference /Mean) x 100% : -5.08%
'(STD Difference /Mean)~. x. 100%.: 1.55%
- All values-expressed'in units of 10-3 JK/K The results of.these analyses demonstrate that the 3-D to 1-D collapsing techniques utilized by PECo reproduce the void and Doppler reactivity components predicted by the 3-D
- simulator with very little bias'or uncertainty.
i l: 3-12
1ill 0 1 5
. 1
- . p
- o T
0 i2 1 D - N _- - 1 _ - . G O _ N EDI I T : : , . - i. 4 - . : . : 0) 9 n
- 1. L T U - - .
( i ASB R R_
- 2. _-
3 H UR I E E_ t h J _ W W_ g E CDS T _
- O O_ i P .
e RO AI P_ D .
. H U E ND D __
G7U - 3 1_ e r
.R I
F C E 0 o 3s . . . . ,'. C BvW 6 PDO P 3 - _h -
,0
.L :
- . - . . . : : . i 3
/ -
- . m -
t o t
. o B
l
,0 _
5 0 5 0 1 1 g 0
- )
u o R c) O a > ,_O O & Yu
, C! !
i l' / -
/, <
e,
' . . E 2 i8
. - 0 _.
. E _
T
. A _ .
L N_ U A_ R_ M T_ i7
- . . I E_
. /,
. S R_
. E .
H R - .
- . , :- :- , . .' , - . . ; 6 T
0 RO - 0OC 2.0WD. . . 22 DD E /.. .
. ,5 2
3C O D EOR C 1 O RE D R . U7 .
,4 G01Y L F 2sL I
BvA P I
-i8 DT .
2. I 3 3N . * : :...! I . .
. i:.j:::f! .. i4 2
- ,O 0 o, o, 0 -
g 1 7 0 - i1 2 -
. ~ '
~ . .
0 0 0 o . o, 0 4 3 2 i 0 n m g v e c c O c_ O + 0 IM O
- ,Eg
- Y;
- e .
E
. 8
- . - 2 _.
0 E T ._ A N_ .
- . L A ._
U
- - . R M T _. ,7
. I E __
- S R_
- - j E' -
H R
' .,6 TO -
RC - OD 3 - .
.CWE - - .
,5
- 2. O D DD 2
3E E7RO O 1 . RC DNR U2 U ~ G B1 ~
,4 I
F P sY L . . '
,8 vL -
~
~
DA
~ . .
I - . . 3I T
- - . : I ,3 N
I
. . . ' ;:::::::- ,4 .
,2
.O 0 0 o -
0 .
,' . ,1 1
4 0 2 1 p
. _ - - - - - ~ . .
,0 0 0 0 Q 0 4 3 2 0 QVeo) coco %O o._~O+-.
,ig lt L 1l it
.f. r i
1 l _ / - - l r
- - E 2 ,8
/ ,
- f f
E 0 t
- - T
/
- A N L A e -
- U R M T
'r,v ,7
- - I E
- S R
- - 7, -
/' -
H E
/ - ,5 T R 0 RO -
0OC 4.0WD 22 E -
- ,5 'M DD -
2 3C O D
. . 1 -
EO V' . . - RE R O . . - DR U7 .
,4 GC1Y L .
./:s I
F 3sL . BvA
~. 8 -
P I DT .
'r p
I . . - 3N .
! ! ,3 I . - -
{.;. ,. * * ::: 4 -
,2
-.O ,
0 o, o, 0 - 1 4 7 0 - . ,1 2 1 [0
' ~ ~ - ' - - . - ' '
0 0 0 0 0 4 3 2 1 0 QkvoQC0IO t0IM 0. o& . ui e r l 1 l
- 1 - ~
r - - r
- f - -
r -
- i ~
~
', - E _
,8 i
~ 2_
- - ~ - 0 E .
/
- - T _
A _ N_ L A
~ U R_
M
~
- T __ ,7 I
E_
- ,- S R __
l , E - H R
,6 TO - - -
RC - ~ 5 OD - OD D 2.CWE ,5 D 2 - 3E E7RO O f 1 - RC DNR t U3 . ~ GB1 U
. - ~
~
~
,4 I
F P sYL - : '
* ' .~
- - ;8 vL .
DA I T ,. - . - ,
.: ,3 3I -
N I
;: l .,..::.*.:- .l4
- ,2 oO -
0 0, 0 0 - 1 4 7 0 -
,1 2 1
. . - - - - . . - . . ,O 0 0 0 0 0 4 3 2 1 0
)
Qv DUCOro +0Ixo;0G . a uic
4.0 Reload Safety Evaluation Procedures for FSAR Abnormal Operatic 7al Transients This 'section describes the methods used by PECo to analyze abnormal . operational transients (AOTs) to determine their consequences and to evaluate the plant capability to control ( or accommodate such occurrences. Results of PECo analyses for the limiting events are also presented for the reference cycle, Peach Dottom 3 Cycle 7. As such, the analytical results are ' representative of the general system response expected for the various events. The postulated transients are divided into seven categories based on the type of disturbance initiating each event. The categories examined here are:
- 1) Increase in Reactor Pressure Events
- 2) Decrease in Core Coolant Temperature Events
- 3) Reactivity and Power Distribution Anomaly Events
- 4) Increase in Reactor Coolant Flow Events
- 5) Decrease in Reactor Coolant Flow Events
- 6) Decrease in Reactor Coolant Inventory Events
- 7) ASME Vessel Overpressure Protection Events Utilizing the analytical methods previously described herein, the most severe event (s) in each category has (have) been quantitatively analyzed as reported in the following sections. Each event analyzed (except for category 7) is assumed to be classified as an incident of moderate frequency (i.e. frequency between 0.05/ plant-year and 1.0/ plant-year [34]). Thus, each event is reviewed with respect to the minimum critical power ratio (MCPR) safety criteria. The results of these analyses establish the l licensing basis for future PECo reload licensing applicati- i I ons at PBAPS (section 5.0). )
i 1 ' l 1 4-1 a
4.1 Increase in Reactor Pressure Events Transient events which result in an increase in reactor
. pressure challenge the Reactor Coolant Pressure Boundary (RCPB). Increasing pressure also results in the collapse of core - voids, thereby increasing core reactivity and power which threatens the fuel cladding integrity due to overheating (i.e. CPR). The most severe event in this category is the generator load- rejection without bypass.
(GLPWOB) which is described below. 4.1.1 Generator Load Rejection Without Bypass A generator load rejection is initiated by a significant loss of electrical load on the~ main generator, resulting in the rapid closure of the turbine control valves. The steam bypass system is subsequently assumed to fail. Accumulated plant operating data indicates that this event can be categorized as an infrequent incident with a frequency of
-0.0036/ plant-year [35). The basis for this categorization is currently under review by the NRC. However, NRC continues to classify the event as one of moderate frequency. A description and analysis of this event using PECo methods is presented in the following sections.
i 4.1.1.1 Description of GLRWOB Event i A loss of. generator electrical load without bypass at high power (>30% - NBR l ) results in the following plant transient 1 sequence:
- 1) Electrical load is lost ( t- 0. 0 sec). Turbine-Gener-ator begins to accelerate. Recirculation pumps also begin to accelerate. The recirculation M-G sets power cupplies are in-house.
1 I l 1 NBR s Nuclear Boller Rated 4-2 , 1
m _ _ - - - _ _ --.- i 4 2): Turbine-Generator power-load unbalance (PLU)' devices trip to initiate turbine control valve. (TCV) fast closure (t=0.0 sec).
- 3) Turbine bypass valves fail to operate (t=0.0 sec).
- 4) TCV~ fast closure initiates reactor protection system (RPS) actuation (t=0.03 sec).
~5) TCVs closed-(t=0.075 sec).
- 6) . Reactor control rods begin insertion (scram).
(t=0.28 sec).
- 7) _ Reactor pressure increases to the recirculation M-G set high pressure-trip setpoint (t=1.45 sec).
- 8) Reactor pressure increases to safety / relief - valve setpoints: (t=1.5 sec). Safety / relief valves open and discharge to suppression pool terminating the pressure increase.
4.1.1.2 GLRWOB Analysis Assumptions The following sections describe tha evaluation of the GLRWOB
, event which is typically the limiting JCPR event for PBAPS at' EOC. Due to the time dependent variation of the axial power distribution during this event, the RETRAN 1-D kinetics option is ' utilized. With the exception of the following; conservative inputs, the evaluations to be presented were performed utilizing the best-estimate RETRAN model, described in- PECo-FMS-004 [4]. The use of these conservatism in the evaluation of the GLRWOB event results in an increase in the predicted ACPR of 0.035. This represents an increase of 20% above the RETRAN best-estimate value (4CPR=0.173) . The conservative assumptions utilized in the' evaluation are:
1)_ Scram setpoints at Technical Specifications limits. o
- 2) RPS logic delays at Technical Specifications limits.
- 3) - Relief- valve capacities at minimum ASME specified values.
i 4-3 )
e -_
- 4) Relief valve setpoints and response characteristics.
L > at Technical . Specifications setpoint limits'.and maximum specified response times..
- 5) Main steam isolation valves, turbine control valves, I? turbine stop valves, and turbine bypass valves stroke times at minimum -(or maximum) of~ design values.
1 -6) - No ; credit taken - for recirculation M-G set - shed- on generator load rejection. Additional conservatism'to account for uncertainties in the
~
CRD scram speed, reactor thermal power, and the RETRAN-model will' be ' addressed in a statistical manner in section 4.1.1.4. 4.1.1.3 GLRWOB System Wide Response The closure of - the TCVs terminates the steam flow to the turbine very rapidly (t=0.075 sec). This causes a rapid
' pressure increase in the steam lines upstream of the TCVs which then propagates at sonic . velocity to the' reactor
. vessel (in 0.3 seconds) resulting in a rapid pressurization there. The TCV closure also initiates'a reactor scram. The resultant pressure wave is' reflected several times between the reactor. vessel and the closed TCVs. This oscillatory behavior is seen in the response of the vessel steam flow illustrated in Figure 4.1.1, but is . not as evident in the steam - dome pressure response (Figure 4.1.2) due to the J
-capacitance effect of the reactor vessel steam dome region.
The- flow oscillations and pressure increase are mitigated i by the safety / relief valves when they open at 1.5 seconds. ] The peak reactor pressure is 1217 psig , well below the RCPB safety limit.of 1375 psig l. i 1 The limiting ASME vessel overpressure event is the MS!VC with position switch failure. This event is evaluated independently in section 4.7. 4-4
--- . _ - = - -
The rapid pressurization in the reactor vessel inserts a large ' positive reactivity due to a rapid decrease in core void' content. The total reactivity change is initially negative due- to the anticipatory reactor scram (t=0.28 sec) .
- However, the positive void reactivity due to the pressure
. increase overcomes ' the . initial negative scram reactivity, causing the total reactivity to increase and peak at a value of : $0.707l (t=0.88 sec). The subsequent ' reduction in the
. positive void reactivity insertion after ths'first pressure oscillation- and the increase in the negative scram reactivity after 0.88 seconds. combine to rapidly reduce the total core reactivity and terminate the nuclear transient.
The . transient scram, void, Doppler and total core reactivities ~are~ illustrated in Figure 4.1.3. The transient. core power (neutron flux) is illustrated in Figure 4.1.4. Initially, the power decreases slightly due'to
.the negative scram reactivity. The power then increases repidly to a peak value of 395% NBR at 0.90 seconds'as the total c o,r e reactivity increases. The power decreases thereafter as the total reactivity decreases and the core is shut down.
The transient core average fuel clad surface heat flux response is illustrated in Figure 4.1.5. Initially, the heat flux decreases due to the decrease in reactor power but then. increases rapidly as ' the core power increases. The heat flux attains a peak value of 127% NBR at 1.05 seconds and 'then decreases as the reactor power excursion is terminated. . The core inlet flow response is illustrated in Figure 4.1.6. The oscillatory behavior of the pressure wave is evident in the flow response. The trip of the ; recirculation M-G sets l- on high reactor pressure at 1.45 seconds results in a l gradual reduction in core flow as the recirculation pumps l 4-5 1
L. l.
'and M-G sets coast down. The reactor water level response is illustrated in Figure 4.1.7. The water level drops initially due'to the collapse of core void and reduction of i
]
feedwater flow. The feedwater flow (Figure 4.1.8) decreases initially due to increasing reactor pressure and the loss of low pressure steam to the feedwater turbines with the l ' closure of the TCVs. The feedwater flow recovers when the L feedwater turbine high pressure steam admission valves open at 2.8 seconds. Reactor water level subsequently recovers i (not simulated). To simulate the high pressure steam admission valves, a q model modification was required. This modification was
. developed to address an item specified in the transient methods report (PECo-FMS-004) SER [4] with regard to the use of the feedwater system model for licensing purposes.
However, it should be noted that this modification of the feedwater system model has no impact on the prediction of MCPR for the GLRWOB event due to the difference in timing of the MCPR (t=1.05 sec) and the opening cf the high pressure steam admission valves. I 4-6
W W__ - O O_ . _ . L L_ - F F_ - 5 M F __ - i ' . A E_ I - .;t E L_ - T E_ - S R_ - 3 _il1 . i4 BW OO - WLF -
- 1. R 1.L M -
4 GA ECE _ 3 T - R OS UEL G I 7E - FCS 3 S - BE PV . 2
. . . .: .;2 I i
- j .
l ,l ,,t l' V . 1
. - ~
. - - - . a . - 0 0 0 Q ,
0 0 0 0 Q 0 0 5 0 S 5 1 1
%DZMvg9u.Eay+V 1 C
)
Alq
E - R 4 B U - OS - WS - 2.R E 1.L R ' 4GP ECE - ROM - UEO - GD I 7 FC M 3 - BA - - - . P TE '
~
~
9 t : . S -
,1
. _ m .
- ~ W 0 0 0 0 O. .
0 0 0 0 0 0 5 0 5 1 1 2-
- ) )
- 0 O GL _.
m.m.va) m .- Q aL 3 m m b'm
7 - s
. R
- - . E ML.
- - - L
. P A A.P D T R. .
I 5
- - O C. O O T S_ D V S -
T - N - E - 4 N * - B O - OP - 3 WM - f
.R O -
1. 4GL C
/
EC Y 3 T - ROI - UE V -- I - . - G7 T I FC C - - - 3A - BE - PR -
,2
,:1
- - , ,.. : , l
-E - -
R - O - C u,,** ) ,,,, ,--
~ -
y- -
-o. -
s14 s 1 t hi.
- g I
1 1
\
g s 1 \ g - ,
- \
~ - ,
\
't - \ 1
- N -
O', r o s -
- ~ -
1,,,t -
- N -
l,*,,, s -
- \ -
,,,l \ -
- - . - _.,- - - _ - - ~ - -
0 5 0 5 0 5 7 0 7 5 1 0 0 0 1
^Wvb>+.OOgE
.. (
se r pl 1I 1ll ,I'
i5
. 4 - . ' ,4 B
O - WX -
- 4. R LU - .
- 1. L F 4G N -
EC O ROR - U G7 U ET - I - - FCE SN - B P - 9 D
- - ,1 j
1 8 . ,0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 4 3 2 1 DZ mZDZMv C 2L C O1e.- a L.O l l
s
- 4
- g5 4
- E E
X . . e
~
g4 BU - OL - I 5 WF -
.RT - -
1LA 4GE
- 5 EC H
- - g3 ROG -
UE G V - E A - I 7 - FC E E 3R B
- I P O C .4 . . . . , 9 ' -
, , t a 5 . 4 9 5 0
- - F
- - 5 5
- g1
- - _ ~ - _ _ - . - '
g 0 0 0 g 0 0 0 5 0 5 1 1 uD - 003 mCCZMv X3_ . xl-"
- 4 B -
OW - 0 WO
.R L -
- 1. L F 4GT E - .
ECL - RON - UEI . G7 E I FC R -
- 3O -
BC - P .
- . ' . ,2
,1 0 0 9 0 0 0 9 0 4 2 9 8 1 1 3 mEnZMv)o.oLoO
( c L a1 aw ,fiIL
~ .
L - . BE . OEV . WL
- 7. R R .
- 1. L E .
4G T EC A R OW . UER G I 7O . FCT . 3C . BA . PE R l ! i2 _ ,1
- a - - ~ . -
0 0 0 0 . 0 0 0 0 _ 4 2 2 _ -I
%x(NILLL ZO_* .J...
xI l l
,5 4 - . - . .
,4 B -
OW - WO 8 - -
.RL -
1L F 4GR EC E T - - 3 ROA - UEW G - I 7D - FC E - 3E - BF - P ,
. . : ;2
,1
- a m w ~ _ ~ - ~ - - ,.
0 0 0 0 0 0 0 0 5 0 5 1 1 L O -. O 5a0 0 a.
%DZMv N L-C L
>lr
g - -- - -
,I L j 4.1.1.4 GLRWOB Event Evaluation The methods used by Philadelphia Electric Company to calculate transient critical power ratio. (CPR)' have been previously described and qualified in PECo-FMS-002 [2].. The . time-dependent reactor power and thermal-hydraulic boundary conditions (lower plenum, upper plenum and core bypass volume pressures) from the RETRAN GLRWOB system run were l used to drive a RETRAN/TCPPECO hot-channel CPR calculation j to the safety limit CPR 1. The JCPR was calculated to be 0.21 (initial CPR was 1.28) and the corresponding RCPR .(4 CPR/ICPR) was calculated to be 0.16.
The overall objective of this evaluation was the conservative determination of the plant Operating Limit Minimum Critical Power Ratio (OLMCPR). The OLMCPR is designed to prevent the violation of the Safety Limit Minimum Critical Power Ratio (SLMCPR) in the event an abnormal operational transient occurs. Thus, a high degree of confidence in the determined OLMCPR is des 4 ad. To that end. a statistical assessment was performed to establish the uncertainties associated with the evaluation of the transient CPR and to determine the probability of violating the SLMCPR. The statistical ascessment was based on the use of transient response RCPR probability distributions which quantify the uncertainties associated with key transient input parameters. The developed RCPR probability distribu-tions were then used to determine the value of RCPR (and i thus OLMCPR) which resulted in a 95% probability at a 95% confidence level (95/95) of not exceeding the SLMCPR in the -I event an anticipated operational transient occurs. 1 For purposes of this evaluation a safety limit CPR of 1.07 was chosen. This is the value originally used by the fuel vender for 8x8 0-tattice fuel. The saf ety limit CPR is dependent on the specific fuet design and is provided by the fuel vendor. 4-15 1
-The -RCPR probability distribution was constructed by repeatedly evaluating the GLRWOB event using sets of random -
values from statistical distributions of key transient input parameters. To generate a statistically - significant number-of values for the RCPR probability distribution would have required a large number 'of RETRAN/TCPPECO evaluations. In order to implement the statistical approach on a practical basis, a response surface technique, which provides an approximation of the RCPR probability distribution,- was used. The response . surface technique resulted in' an analytic equation which accurately defined the RCPR probability distribution with fewer evaluations. The' flow of the overall statistical assessment process is displayed in ~ Figure 4.1.9. The transient input parameters selected for the assessment were those that were determined to- be most significant with regard to the prediction ' of transient CPR, which are:
- 1) _ Control Rod Scram Speed.
- 2) Initial Core Thermal Power.
- 3) Model Uncertainties.
The statistical distributions of the control rod scram speed data and core thermal power data used in the statistical assessment were obtained from the reactor vendor [36). - i l 4-16 _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ )
Figure 4.1.9 Statistical Assessment of Transient RCPR Limits ESTABLISH STATISTICAL DISTRIBUTION OF KEY INPUTS
- 1. CRD Speed
- 2. Core Thermal Power
- 3. hiodel Uncertainties V
RETRAN/TCPPECO RUNS A1ADE TO DEFINE RESPONSE SURFACE V REGRESSION ANALYSIS TO DEFINE RCPR RESPONSE SURFACE RCPR - F(CRD,P) V hiONTE CARLO ANALYSIS PERFORhiED TO DEVELOP RCPR HISTOGRAM F 0.1 0.2 RCPR V PROBABILITY DENSITY FUNCTION DEVELOPED TO DETERMINE 95/95 RCPR 4-17 L _ __ _ .. _ __
N l q f The form of the response surface equation ' utilized in the
, statistical assessment is as follows: .
l RCPR = [ Ao + A1(P) +A2(P)2 + A3(CRD) + .. A4 (P) (CRD)
+A5 (CRD) 2 + RSU] x RMU ( 4 -l') - q where:
1 P= Random-value of initial power (%NBR) minus nominal I value_(100%) i CRD = . Random value of time (sec) to 20% CRD insertion minus nominal value (.694 sec) RSU = Random response surface fitting uncertainty RMU = Random fractional model ur':ertainty Aj = Response surface fitting coefficients (i = O-5) A response surface was developed for the GLRWOB event at End-of-Cycle (EOC) conditions (all control rods out) and at a - mid-cycle (EOC-2000 MWD /ST) exposure point (with som2 control rods still inserted in the core). The fitting coefficients for.both conditions were developed by executing nine RETRAN/TCPPECO evaluations using a box-matrix tech-nique. To this end, each of the key input parameters (CRD speed, core thermal power) was varied over a range of plus or minus three standard deviations as illustrated in Figure 4.1.10. In some instances, where a change of three standard deviations in a parameter resulted in no transient increase in RCPR, a smaller deviation was used. A least squares fitting technique was then used to determine the fitting coefficients for each cycle exposure condition. Table 4.1.1 presents the resulting coefficients (A1) and the associated
. fitting uncertainties (RSU) for each statopoint.
4-18
Figure 4.1.10 Response Surface Evaluations Box Matrix Core Power Stande.rd Deviation
-3 0 +3
-3 X X X CRD Speed Standard 0 X
- X Deviation
+3 X X X
* - Nominal Calculation X - Uncertainty Calculations Table 4.1.1 Response Surface Fitting Coefficients for GLRWOB GLRWOB GLRWOB EOC EOC-2000 Ao 0.1624 x 100 0.3324 x 10-1 A1 0.1139 x 10-3 0.2906 x 10-2 A2 0.6806 x 10-4 0.4222 x 10-4 A3 0.3632 x 100 0.7280 x 100 A4 0.2951 x 10-2 0.1756 x 10-1 A5 -0.1215 x 101 -0.2363 x 101 Fitting Uncertainty 0.0004 0.0004 1
4-19 ~~--__-_--___-___- >
'The determination of the overall model uncertainty fraction (RMU) 'is described in the following sections. The model uncertainty was obtained by performing parametric studies of the GLRWOB at end-of-cycle conditions. These studies were performed to assess the impact of model inputs for which values are . uncertain. Key _ input parameters. were indepen-dently varied over their respective 95 percentile uncertainties (i.e. =2o) to evaluate their- individual contribution to the model uncertainty. The individual uncertainties were then statistically combined to develop'an estimate of- the overall model uncertainty. Only the uncertainties that resulted in an increase in the RCPR were censidered in 'the statistical evaluation. The input parameters included in the studies fall into four categories. These are: 1) reactor core nuclear model, 2) reactor core thermal-hydraulics. models, 3) recirculation
-system model, and 4) steam line model. A detailed discussion of the impact of each of the uncertainties . on predicted RCPRs is presented in the following sections.
4.1.1.4.1 Determination of Nuclear Model Uncertainties Four input parameters were investigated in the uncertainty studies of the reactor core nuclear model. These consisted of the three nuclear reactivity components (void, scram, and . Doppler) and the prompt moderator heating fraction. The ) uncertainties associated with each of these parameters and their impact.on the GLRWOB evaluation are discussed here. ) There are several contributors to the overall void reactivity uncertainty as utilized by the RETRAN 1-D nuclear model. These are: 1) uncertainty in the predictions of the lattice physics code, 2) uncertainty in the fitting of the lattice physics cross sections to polynomial form for the 3-D simulator, 3) uncertainties due to the collapse of 3-D l 4-20
r cross sections to radially averaged 1-D cross sections, and i
< 4) uncertainties in the fitting of the collapsed 1-D cross sections to polynomial form for RETRAN. The uncertainties associated with the first two items are discussed and quantified in PECo-FMS-005 [5). The combined uncertainty'of the last two items is discussed and quantified in.section 3.2. Combining these quantities in a statistically conservative manner yields an overall ' uncertainty (95%
confidence) of 21% in the 1-D void reactivity. To assess the effect of this uncertainty on RCPR, the base 3-D simulator cross sections were modified to result in a 21% increase in the void teactivity and the RETRAN 1-D cross section file was regenerated. The GLRWOB event was then reevaluated with the new cross section file. The increase in void reactivity resulted in a higher peak power, heat flux and RCPR. The result of this evaluation is presented in Table 4.1.2. As with the void reactivity, there are several contributors to the overall scram reactivity uncertainty. These are: 1) uncertainty in the predictions of the lattice physics code,
- 2) uncertainty in the fitting of the lattice physics cross sections for the 3-D simulator, 3) uncertainties due to the collapse of 3-D cross sections to radially averaged 1-D cross sections, 4) uncertainties in the fitting of the collapsed 1-D cross sections to polynomial form for RETRAN, and 5) a conservatively assumed 5% adder to account for potential control rod depletion effects. The uncertainties associated with the first two items are discussed and quantified in PECo-FMS-DOS [5]. The combined effect of items 3 and 4 in discussed in section 3.2 and indicates a conservative bias exists in the 1-D scram reactivity model.
No credit was taken for this conservative bias. Combining the above quantities in a statistically conservative manner yields an overall uncertainty (95% confidence) of 11% in the scram reactivity. To assess the effect of the 11% scram ' 4-21
uncertainty on the GLRWOB RCPR, the RETRAN rod cource mixing. fraction was reduced from 1.0 to 0.89, and the GLRWOB was then' reevaluated. A slight increase in peak power and RCPR was observed. A second evaluation was performed in which the' control rod scram speed was adjusted to result in an 11% reduction in scram reactivity.at the 5% and 20% control rod insertion points. This evaluation yielded a more conserva-tive . result (i.e. larger RCPR) and was used in the determination of the overall model uncertainty. The result of this evaluation is presented in Table 4.1.2. The contributors to the overall Doppler reactivity uncertainty are: 1) uncertainties in the predictions of the lattice physics code, 2) uncertainties in the fitting of the lattice physics cross sections to polynomial form for the 3-D simulator, 3) uncertainties due to the collapse of 3-D cross sections to radially averaged 1-D cross sections, and
- 4) uncertainties in the fitting of the collapsed 1-D cross.
sections to polynomial form _ for RETRAN. The uncertainties associated with the first two items above are described and quantified in PEco-FMS-005 [5]. The calculations presented in section 3.2, for uncertainties 3 and 4, indicate a 5% conservative bias in the 1-D Doppler reactivity for which credit is taken. Combining these quantities in a statistically conservative manner yields an overall uncertainty (95% confidence) of 15% in the Doppler reactivity coefficient. The base 3-D simulator cross sections were modified to result in a 15% reduction in the Doppler reactivity and the RETRAN 1-D cross section file was regenerated. The GLRWOB event was then reevaluated with the new cross section file. The decrease in the Doppler l reactivity resulted in a higher peak power, heat flux and RCPR. The result of this evaluation is presented in Table 4.1.2. 4-22
The prompt moderator heating fraction accounts for 'that portion of the total fission ' energy wl:1c.' is deposited directly and. essentially instantaneously into the fluid
' moderator via gamma heating. In the PBAPS RETRAN model, the prompt moderator heating fraction is a function of moderator density. The nominal core average value for the in-channel prompt moderator heating fraction at rated core conditions is 0.02. Based on fuel vendor data [37],'the prompt heating fraction is larger than 0.016 with 95% certainty. The core average prompt heating fraction was reduced to this amount and the GLRWOB event was then reevaluated. The decrease in prompt heating resulted in an increase in peak power, heat flux and RCPR. The result of this evaluation is presented in Table 4.1.2 4.1.1.4.2 Determination of Thermal-hydraulics Model Uncertainties Seven parameters were investigated in the uncertainty studies related to the reactor core thermal-hydraulics models. These consisted of: 1) core pressure drop, 2) core bypass flow, 3) core average gap conductivity, 4) hot-channel gap conductivity, 5) core bulk void model (algebraic slip), 6) core subcooled void model, and 7) fuel L pellet radial power distribution. The uncertainties associated with each of these parameters and their impact on ,
the GLRWOB evaluation are discussed below. i The initial core pressure drop utilized in the GLRWOB i evaluation is determined by the steady-state thermal-hy-draulics code FIBWR. The methods used to determine this value are described and qualified in PECo-FMS-001 [1]. Based on the data presented, the core pressure drop uncertainty (95% confidence) was determined to be bounded by 1.5 psi at rated reactor conditions. The core pressure drop l was reduced by 1.5 poi and the GLRWOB was reevaluated. The 1 4-23
Ly ~] j I Ih .)
.1 l
. reduction of the pressure drop required a compensating ]
reduction in recirculation drive flow to maintain the pressure distribution in the reactor vessel. The decrease
~in core pressure drop resulted in an increase .4n peak core power and heat flux, resulting in a slight increase in RCIB. ]
The result of this evaluation is presented in Table 4.1.2. The initial core bypass flow is also determined by the j faady-state thermal-hydraulico code FIBWR. . Based on the data presented in PEco-FMS-001 [1], the core bypass flow uncertainty (95% confidence) was determined to be 1.1 Mlbm/hr (10%) at rated core conditions. The initial core 1 bypass flow was increased 1.1 M1bm/hr and the G'LRWOL was { reevaluated. The increase in core bypuss flow resulted in a j slight decrease in peak power and heat flux. However, the core in-channel flow at the time of MCPR was also reduced slightly, result.ing in a slight overall incresse in RCPR. The result of this evaluation is presented in Table 4.1.2. i The core average and hot-channel gap' conductivities utilized in the GLRWOB evaluation were calculated with the FROSSTEY fuel performance code. FROSSTEY has previously been described and qualified in PECo-FMS-003 [3]. Based on the data presented, the core average and hot-channel gap conductivity uncertainties (95% confidence) were conserva-tively estimated to be 14.2% and 17.2%, respectively. For the core average gap conductivity uncertainty evaluation, the core average gap conductivity was decreased by 14.2% and the GLRWOB was reevaluated. The decrease in gap conductivi-ty resulted in an increase in peak core power and heat flux with a consequent increase in RCPR. In the case of the { 1 hot-channel gap conductivity uncertainty study, the nominal GLRWOB system wide results were used to drive a new hot-channel evaluation with the gap conductivity increased by 17.2%. The increased gap conductivity resulted in a
)
4-24 1 1
l
- snaller fuel thermal time constant and an increase in the peak' heat flux, with a consequent increase in RCPR. Results of both evaluations are presented in Table 4.1.2.
The PECo RETRAN core model utilizes the algebraic slip and subcooled void model options. Uncertainties for both of these models were' determined based'on the data presented in EPRI NP-2246-SR [38). The uncertainty in the subcooled void model was estimated to be 6' 0 % of the total subccoled void fraction. Likewise, the uncertainty in the bulk boiling void fraction was estimated to be 0.05. The subcooled void model input parameters were adjusted to achieve a reduction in the subcooled void content of 60% and the GLRWOB event was reevaluated. The decrease in subcooled void resulted in a slight reduction in the peak core power but an increase in the peak heat flux and RCPR due to broadening of the power peak. The algebraic slip model inputs were adjusted to achieve a reduction of 0.05 in the core average void fraction in the bulk boiling region of the core. This is equivalent to a 10.7% increase in the drift flux parameter (Co) at approximately 0.5 void fraction. The increase in the drift flux parameter resulted in an increase in the peak core power and heat flux, with a consequent increase in RCPR. Results of both evaluations are presented in Table 4.1.2, The fuel pollet radial power distribution is assumed to be uniform (flat). In actual operation, it is peaked sharply to the outside of the pellet due to plutonium buildup and self shielding effects. The energy deposited in the fuel pellet by the fission process is transferred to the coolant by conduction through the pellet material (UO 2). When the power deposition is peaked higher near the edge of the fuel pellet, the average distance the energy (as heat flux) must traverse is smaller, resulting in ar< effectively smaller j fuel thermal time constant. A smaller thermal time cohstant 4-25
results in . f aster thermal feedback- (for power excursion events such as the GLRWOB)- . which tends to mitigate the magnitude of the power excursion and thus reduces the change in the transient critical power ratio. Therefore, a uniform radial pellet . power distribution is conservative for the . limiting licensing transients. 4.1.1.4.3 Determination of Recirculation System Model Uncertainties Five parameters were investigated in the uncertainty studies of the recirculation system models. These consisted of: 1)
-jet pump efficiency, 2) jet pump inertia, 3) recirculation loop inertia, 4) steam separator inertia, and 5) steam separator pressure drop. The uncertainties associated with each of the parameters and their impact on the GLRWOB evaluation are discussed below.
Development of the PBAPS jet pump model inputs is described l in. PECo-FMS-004 [4]. To determine the uncertainty in the jet pump model input parameters, which establish the jet pump M-N characteristics, the uncertainty in the plant data used to establish the input was examined. Based on the uncertainty in the plant data, the uncertainty (95% confidence) in the jet pump model M-N efficiency was estimated to be 9%. The jet pump model inputs (j et pump
. suction and discharge loss coefficients) were adjusted to result in a 9% increase in the M-N efficiency and the GLRWOB event was reevaluated. The increase in jet pump efficiency required a compensating reduction in recirculation drive flow to maintain the pressure distribution in the reactor vessel. The increase in the jet pump efficiency resulted in less attenuation of the pressure wave as it traveled from I
the steam dome through the downcomer region and the jet 1 pumps into the lower plenum and finally the core. This 4-26
resulted in an increase in peak core power and heat flux, and a consequent increase in RCPR. The result of the evaluation is presented in Table 4.1.2. The bounding values for the jet pump and recirculation loop inertias were conservatively assumed to be a factor of three in either direction from.the nominal. Thus, the jet pump inertia was reduced by 67% (a factor of 3) and the GLRWOB was- reevaluated. The decrease in inertia resulted in a slight decrease in peak core power.and heat flux. However, the decrease in heat flux was offset'by a decrease in core inlet flow at the time of peak heat flux, resulting in an overall slight increase in RCPR. Similarly, the recircula-tion loop inertia was also reduced by 67% and the GLRWOB was reevaluated. This resulted in a very slight increase in the peak core heat flux and a consequent increase in RCPR. Results of both evaluations are presented in Table 4.1.2. The steam separator inertia is provided by'the manufacturer. Thus, there is insufficient data to establish a precise estimate for the uncertainty in the separator inertia. For the purposes of this study, however, it was conservatively assumed that the separator uncertainty is 90% of the nominal value. The separator inertia was varied over this range and the .GLRWOB was reevaluated to determine the value of uncertainty that resulted in the largest increcse in the transient 4CPR. This was determined to be a 30% reduction of the nominal value. The reduced separatcr inertia resulted in an increase in peak core power and heat flux and a consequent increase in RCPR. The result of this evaluation is presented in Table 4.1.2. The steam separator pressure drop is obtained from a vendor l l equation relating pressure drop to separator inlet flow and quality. Comparison of the pressure drops predicted by the vendor equation to those measured for each of the three p 4-27
J I L turbine trip tests [39] indicate that the average measured ' j pressure drop is approximately . 70% of the ' predicted value. _ This.is illustrated as follows. 4 Measured vs Predicted Steam
-Separator Pressure Drop !
Predicted Measured. Case .Ips.il (psi) TT1 8.3 7.4 TT2 7.3 4.5 T.Tl 1.i.2 1, i Average 8.43 6.10 The' steam separator pressure drop was reduced by 30% and the GIRWOB was reevaluated. The reduced pressure drop required a compensating reduction in the recirculation drive flow to maintain the pressure distribution in the reactor vessel. The decrease in . separator pressure drop resulted. in a reduction (-0.005) in the transient RCPR. Thus, the use of the vendor equation to obtain the separator pressure drop is conservative. 4.1.1.4.4 Determination of Steam Line Model Uncertainties i Three parameters were investigated in the uncertainty studies of the steam line model. These consisted of: 1) steam dome volume, 2) steam line pressure drop, and 3) steam line volume / inertia. The uncertainties associated with each of these parameters and their impact on the GLRWOB evaluation are discussed below. The steam dome vclume uncertainty was conservatively estimated to be. 3.0%. The steam dome volume was reduced by 3.0% and the GLRWOB was reevaluated. The reduced volume 4-28 i
resulted in a slightly higher pressurization rate and an increase in the peak core power, heat flux and RCPR. The result of this evaluation is presented in Table 4.1.2. Comparisons between predicted and measured steam done pressure in PECo-FMS-004 [4] indicate a steam line pressure drop uncertainty (95% confidence) of approximately 20%. The steam line model loss coefficients were reduced to the result in a decrease in the pressure drop of 20% and the GLRWOB event was reevaluated. The reduced pressure drop resulted in an increase in the pressurization rate and an increase in peak core power, heat flux, and a consequent increase in RCPR. The result of this evaluation is presented in Table 4.1.2. The steam line diameter utilized to calculate the ,sts::, line flow area and volume in the PBAPS RETRAN model has been established to be 1.3% less than the minimum value indicated by the steam line material specification as supplied by the vendor. It is concluded, therefore, that the RETRAN steam line model volume and inertia parameters are set to conservative input values. Thus, no parametric sensitivity studies were necessary for these parameters. The change in RCPR (JRCPR) determined for each model uncertainty is presented in. Table 4.1.2. The overall model uncertainty was estimated by taking the square root of the sum of the squares of the individual uncertainties (" propagation of errors" technique). The overall uncertain-ty was thus determined to be 0.030 RCPR (18.3% of the i nominal value) at the 2a level. This value forms the basis of the term RMU in the response surface equation (egn. 4-1). l 4-29 f l I o - --
l Table'4.1.2 Summary of RETRAN Model Uncertainty for a Generator Load Rejection Without Bypass. Parameter Uncertainty 4RCPR I. -Nuclear Model A) Void Reactivity +21% .020 B) Scram Reactivity -11% .013' C) Doppler Reactivity -15% .008 D) Prompt Heating -20% .007' II. Core T-H Model-A) Core Pressure Drop -1.5 psi .002 B) Core Bypass Flow +10% .002 C) Core Avg. Gap Conductivity -14.2% .006 D) . Hot Channel Gap Conductivity +17.2% .004 E) . Core Bulk Void (Slip) Co + 10.7% .002
'F) Core Subcooled Void -60% .003 G) -Fuel Pellet Radial Power conservative N/A Distribution III. Recirculation Model A) . Jet Pump Efficiency +9% .005 B) Jet Pump Inortia -67% <.001 C)- Recirculation Loop Inertia -67% . <.001 D) Separator Inertia -30% .005 E) Separator Pressure Drop conservative N/A IV. Steam Line Model A) Steam Dome Volume -3% .003 B) Steam Line Pressure Drop -20% .008 C) Steam Line Volume / Inertia conservative N/A
-Total (JRCPR) .030 l
t 4-30 I L_________ _----_- _ _
4.1.1.4.5 Determination of Statistical Adjustment Factors and Technical Specification Operating Limit Minimum Critical Power Ratios Utilizing . the response surface methodology set forth in equation 4-1, a Monte Carlo analysis was performed to determine a 95/95 confidence estimator (RCPR95/95) for the licensing calculation of RCPR. To this end, the Statistical Adjustment Factor (SAF) is defined such that: SAF = RCPR95/95 - RCPRnominal (4-2) where RCPRnominal is the nominal RCPR (JCPR/ICPR) calculated by the transient analysis. The SAF is used to adjust the nominally calculated RCPR values when calculating the OLMCPR. This assures a 95% probability at a 95% confidence level of not exceeding the SLMCPR in the event an anticipated operational transient occurs. The OLMCPR95/95 is calculated as follows: SLMCPR OLMCPR95/95 = , (4-3) The Technical Specifications operating limit for MCPR (OLMCPR T echSpec) is a function of measured scram time and is determined from the following general equation: OLMCPR T echSpec = OLMCPR95/95 + r(aOLMCPR) (4-4) with: f T '"# T# r= ; 05751 (T3 - r, j The OLMCPR T echSpec is commonly referred to as the option B limit for values of r,vo s t, (T-0) and as option A for r,y,-r, (r - 1). The option B limit, which is less restrictive, is 4-31
s i. available .to ~ plants which- demonstrate scram speed i compliance. -1f scram - speed compliance is not' demonstrated (i.e. r , < r ,ve sr a ), the OLMCPR T echSpec is' determined -by linearly interpolating between the option B and option A values using equation 4-4. The interpolation is based on
~ the relative difference between the average measured. scram time (rm), the technical specification upper conformance
- limit on scram time to 20% control rod insertion (r, = 0. 9 0 L sec), and the adjusted analysis mean scram time to 20%
control rod insertion (r,), with: I a f r,T (4-5) T ~ ac " E. s \ n j 7,- + 1.6S a (4-6)
. where r, is the time to 20% insertion for control rod "i", n is the total number of surveillance rod tests performed to date in the cycle including the N number of control rods
' tested at the beginning of cycle, u is the nominal time to 20% control rod insertion utilized in the statistical assessment (0.694 sec) and o is the associated scram time standard deviation (0.016 sec). The term oOLMCPR is a scram speed adjustment factor which accounts for the sensitivity of the event to the difference in scram speeds used as the b a s i s o f r , a n d r ,. It is defined as the difference between the OLMCPR95/95 calculated for the event using the Technical Specifications scram speed limit (67B, r ,ye - r ,) and the -)
OLMCPR95/95 from the nominal (r ,c - r,) calculation. The Monte Carlo analysis as described above was performed utilizing the BSAFE [40] computer code, which has been previously reviewed by NRC as part of a similar submittal
)
4-32
)
i l
)
1 made by the Tennessee' Valley Authority [41]. The BSAFE code . uses. random _ values from -the following four normally distributed variables: t
- 1) . Initial core power (P, %NBR) _i
- 2) Time to 20% scram insertion (CRD, sec)
- 3) Response surface uncertainty (RSU, JCPR/ICPR)
- 4) RETRAN model uncertainty (RMU, Fraction) l A prediction for RCPR is then derived for each set of random variables using the response _ surface equation 4-1. A I Probability Density Function (PDF) for RCPR is generated by repeating this sampling technique for some large number (e.g., N=200,000) of trials. Numerical integration of the PDF function yields the desired 95/05 estimator for RCPR.
Table 4.1.3 presents. the statistical adjustment factors (SAFs) ar.d scram speed adjustment factors (40LMCPRs) that have been derived using this methodology for the GLRWOB event. Results are tabulated for the EOC and EOC-2000 MWD /ST exposure statepoints. Table 4.1.3 i i Statistical Adjustment Factors and Scram Speed Adjustment Factors for the PBAPS GLRWOB Event Statenoint SAF AOLMCPR EOC 0.026 0.08 EOC-2000 0.038 0.15 l 4-33 I i 1 l
s i 1 4.2 Decrease in Core Coolant Temperature Events. Transient events that result in a decrease in core coolant l
' temperature lead to increases in core reactivity and power )
which . threaten fuel cladding integrity due to overheating I (i.e. CPR).- The two most severe events in this category are 3 the feedwater controller failure (max. demand) event (FWCF) i and the loss of feedwater heating event (LFWH), which are l described.below. 4.2.1 Feedwater Controller Failure (Max. Demand) The feedwater controller failure (FWCF) event is postulated to occur as the result of the failure of a control device which results in an increase in feedwater flow up to the maximum system capability. This event is considered to be an incident of moderate frequency. 4.2.1.1 Description of FWCF Event A reactor feedwater controller failure results in the following plant transient sequence.
- 1) Feedwater controller failure results in maximum feedwater demand (t=0.0 sec).
- 2) Feedwater flow increaseu to maximum. Reactor water level increases to high level (L8) setpoint. Main turbine and feedwater turbines are tripped (t=20.51 sec).
- 3) Turbine stop valve (TSV) closure initiates reactor protection system actuation (t=20.53 sec). TSVs are closed and turbine bypass valves start to open.
- 4) Reactor control rods begin insertion (scram)
(t=20.78 sec). i
- 5) Reactor pressure increases to recirculation M-G set high pressure trip setpoint (t=22.59 sec).
4-34 i
- 6) Reactor pressure increases - to ' safety /rel'ief l valve setpoints. Safety / relief valves - open and discharge-to. suppression. pool- terminating- the pressure.
' increase (t=23.05 sec).
- 7) . Reactor low . water ' level (L2) . initiates HPCIS/RCICS (not- simulated) . - Water. level stabilizes and ' recov-ers.
l; A summary of ' the reactor water level associated with each level setpoint (L2 through LB) at 'PBAPS is presented 'below. - Level Setooint Function L2 -48 in. HPCI/RCIC/Recirc. M-G Trip- .. L3- 0 in. . Scram"(10.'5 in.-above bottom of skirt) L4' +6-in. ADS-Permissive L5' +17 in. Low-Level Alarm L6 -+24 in. Nominal Water Level L7 '+29 in. High Level Alarm L8 +45 in. Turbine-Trip /F.W. Turbine Trip-4.2.1.2 FWCF Analysis Assumptions The evaluation of the FWCF event is performed utilizing the analysis ' assumptions described in section 4.1.1 for the GLRWOB event, including the use of the RETRAN 1-D kinetics option.and the development 'of a 95/95 confidence estimator
'for RCPR.
In addition to the above, the assumptions with regard to two parameters which are important for the FWCF event . are as-follows. The feedwater system maximum runout capability (approximat ely 130% NBR) is based on the actual plant control configuration and is consistent with the current vendor licensing basis. The initial reactor water level is assumed to be at the low level alarm setpoint (L5) to allow more time for the core heat flux to increase during the first phase of the event prior to the turbine trip. This ' assumption is also consistent with the current licensing basis. , 4-35
1 l 4.2.1.3 FWCF System Wide Response The- feedwater controller failure event results in two distinct _ transient phases. The-first phase.is characterized by - a gradual increase in reactor power as the. increased feedwater flow (Figure 4.2.1) results in increased core l subcooling. ;The increase in vessel steam flow that accompanies the increase in reactor power is insufficient to 1 offset the increase in' feedwater flow and reactor water level begins to rise. The second transient phase of ' the event begins when the reactor water level reaches the high. level (L8) trip setpoint and the main turbine and the feedwater turbines are tripped. The main turbine trip results in the closure of the TSVs which terminates the steam flow to the turbine very rapidly (in 0.1 seconds). This causes a rapid . pressure increase in the steamlines upstream of ' the TSVs which is partially mitigated by the rapid' opening (in 0.3 seconds) of the turbine bypass valves. The TSV closure also initiates a reactor scram. The pressure increase propagates at sonic velocity to the reactor vessel (in 0.3 seconds) and causes a rapid pressurization there. The pressure wave is reflected several times between the reactor vessel and the closed TSVs. This oscillatory behavior is seen in the response of the vessel steam flow illustrated in Figure 4.2.2, but is not as evident in the steam dome pressure response (Figure 4.2.3)~ due to the capacitance effect of the reactor vessel steam dome region. The flow oscillations are mitigated when the safety / relief valves open at 23.0 seconds. The peak reactor vessel pressure is 1165 psig, well below the RCPB safety limit of 1375 psig. The rapid pressurization in the reactor vessel inserts a large positive reactivity due to the rapid decrease in core void content. The total reactivity change after the turbine trip is initially negative due to the anticipatory reactor l 4-36
scram (t = 20.8 sec). However, the positive void reactivity due to the pressure increase overcomes the initial negative scram reactivity causing the total reactivity to increase and peak at a value of $0.31 at 21.5 seconds. The subsequent reduction in the positive void- reactivity insertion after the first pressure oscillation and the increase.in the negative scram reactivity.after 21.5 seconds combine to rapidly reduce the total core reactivity and terminate the nuclear transient. The transient scram, void, Doppler 'and total core reactivities are illustrated in Figure 4.2.4. The transient core power (neutron flux) response is illustrated' in Figure 4.2.5. The power increases.to 107% NBR during the first' phase of the transient. After the turbine' trip, the power initially decreases slightly due to the negative scram reactivity. The power then increases rapidly to a peak value of 153% NBR at 21.5 seconds as the total core reactivity increases. The power decreases thereafter as the total reactivity decreases and the ccore is shut down. The transient core average fuel clad surface heat flux response is illustrated in Figure 4.2.6. The heat flux increases gradually as the power increases during the first phase of the transient. After the turbine trip, the heat flux decreases initially due to the initial decrease in reactor power but then increases rapidly as the core power increases. The heat flux attains a peak value of 109% NBR at 21.6 seconds and then decreases as the reactor power j
-excursion is terminated.
The. core inlet flow response is illustrated in Figure 4.2.7. The core flow does not change appreciably until after the turbine trip. The oscillatory behavior of the pressure wave is evident in the flow response. The trip of the 4-37
= . _ _ ._ .
l ,. ,i q
'l 1
recirculation M-G set on high reactor pressure at 22.6 seconds results in a gradual reduction in core flow as.the 3
- i. recirculation pumps coast down. The high reactor water level trip (LB) of the feedwater turbines results in a rapid decrease in water level (Figure 4.2.8). 'The difference !
between the actual and sensed levels is due primarily to a i l " difference in their respective "zero reference" points. Eventually the low 2evel (L2) trip setpoint is reached, initiating actuation of HPCIS and RCICS. This- portion. of the event is not simulated. It is demonstrated in the LOFW analysis (section 4.6) that HPCIS alone is sufficient to maintain adequate core coverage after reactor scram. During the first phase of the FWCF transient, an oscillation is observed in the core reactivity (Figure 4.2.4) and core power (Figure 4.2.5) between approximately 11-12 seconds. The cause of this oscillation is a code limitation (described in RETRAN problem reports 99, 265, and 266) associated with the core boiling boundary crossing a junction. A boiling boundary crossing may cause a step change in the void content of the volume upstream of the junction. When this occurs in conjunction with the use of 1-D kinetics, a step change in core reactivity on the order of 8.0E-5 JK (1.4 cen+s) can occur. Because the total core reactivity is small at this time, the result is the observed power oscillation. Because this oscillation is relatively small (it has no significant impact on core average heat flux) and it occurs early in the transient prior to the time of MCPR, it has no impact on the prediction of MCPR. 4-38
0 2 FW -
,5 CO -
- D .
1 1.W L -
- 2. F F 4 CR -
EO E T RE A UW7 GCD I - F3 E .0 BE - 1 P F - I 9
,5
. . ~ . ~ - ~ ' .
,0 0 0 0 0 0 0 0 0 5 0 5 1 1 D
ZCZMvN ' u O3ao$
*l !i
~
- ~
- ~
I h - 1 [i'r ',' 0 2 W - FO -
. CL 1 . - . , h ! :. :- ' . * ; ;5 2 F - ~ 1 .W -
2.F M - 4 C EA - EO T - RE S U - 7 L - GCE I F3B S -
. .01 P$
E - V - W O W_ O_ L_ O W_ - L L_ F_ F F_ S M F S_ A IE __ A_ E L_ P T E_ Y_ - S R_ B_
. D 6 : l
- l5
_ - . . " . 0 _._0 0 0 0 0 0 0 0 0 0 5 0 5 5 1 1 mKmZMv k EOe@ Ah
,0
- - 2 E -
R U F S 3 CS .:.: . : . ';
.. b e . . : . .51 .WE -
2.F R - 4O P EOE - REM U 7O - GCD I F 3 -
.0 BM
- - 1 PA -
E T - S -
- . ,5
- . . . . . - . . - - ,0 0 0 0 0 0 0 0 0 5 0 5 1 1 OmO_v m & eL3mmE o-
,{-
/, e
- - l l)
,I' k' i s
/ - 1 i i I I I
'f 8,' - N \ 1 lt \
,S' ,, \ -
FE 1 i \ - 0
~ ,2
- ~ -
S _R T E__ - N M_ L -
~ _
A_L_ T R P PD~ A ~ - E C_ OO I
~
~
O N T S_ D V
~
~
O
- ~ _ -
F P ~
' CM -
5
~_
~ - 1
~_
- 4. W O -
- ~_ -
- 2. F C -
~
~ -
4CY -
~
~
EO T - REI - - N'_ U 7I V G CT -
~"
F3C I BA
~
~
,0 1
PE ~=-
- ~- -
R
- ~- -
~ -
E -
~
R ~ - O - C -- -
,5
~ ~ ~
- - . - 0 0 5 0 5 0 5 7 0 7 5 1 O 0 0 1 n*v> ,_ + O O 0 G y1u
l 0
- 2 F -
CX -
,5
- 5. - - - 1 WU -
2 F FL 4CN - EO RE O R - - - U 7T - G0U - F0E I - -
.01 BN -
P - 9 { ,5 d e u ~ - ' . - _ . S - ,0 _ 0 0 0 0 0 0 0 0 0 0 - 0 5 0 5 1 2 1 D mWCZMv X2u. COLs-)OZ - 6Su L : ,I
,0
- 2 X
U - - FL - - CF , -
. 4 6 5 1
6
.WT 2FA E
4C H EO. RE G U 7V GCA I F 3E - - ,0 B R - 1 P O - C -
' - T g l , 4 ,5
,0
_ _ _ a 0 0 0 0 0 0 0 0 5 0 5 1 1
.O $
mtrDZMv X3_ C . xi6*
M i - l f] j - 0 i 2 FW - -
. CO -
?
. .51 7,W L -
F 2.F 4 C TE EOL - RE N U 7I G CE - F3R I
,0 BO 1
PC -
' E U .
5
,0 0 0 0 0 0 0 0 0 4 2 0 8 1 1 1 ECZMv D OC gL0
\
- t i
\
1
\
1 t
)
\
s I
'T',ttt s,\ '
' 'i '
t ' 4 1 h -
#m
' 0
'
- i2 s '
- ~
J - L *a E
- F V '
- Y, E
8 CL , ' e i51
.W R - /, *
- 2. F E
- 4 CT
/, '
EO A '
~ ' *
. REW ' s U 7R / - GC O - ' I ' - F 3 T -
/, - ;0 BC '
1 PA d- .
/ -
E e- - R s- - n- - l a e- s t u S- - - c R- - s, A W- - s i5
/-
* , e
- l s
. - . - ~ *
~ - - 0 0 0 0 0 0 0 0 0 6 4 2
) - )
nU q)3UCG 0>QJ . b1$
r U 4.2.1.4 FWCF. Event Evaluation The time-dependent reactor' power' and- thermal-hydraulic boundary conditions from the RETRAN FWCF system run were used to drive a RETRAN/TCPPECO hot-channel transient CPR ! . calculation to the safety limit CPR 1. The JCPR was calculated to be 0.06 and the corresponding RCPR (JCPR/ICPR) was calculated to be 0.05.
'The statistical assessment! process previously described in section 4.1.1~for the GLRWOB event was repeated for the FWCF event. at end-of-cycle (EOC). The results 'of the model uncertainty studies. which determine the value of RMU (fractional model- uncertainty) for equation 4-1 are presented in Table ' 4. 2.1. The associated responce surface
' fitting, coefficients for the-FWCF event at EOC are presented in Table 4.2.2. The resultant statistical adjustment factor (SAF) and scram speed adjustment factor (JOLMCPR) are presented in Table 4.2.3.
For the EOC-2000 MWD /ST statepoint, it was determined that the FWCF event is bounded by the LFWH event (section 4.2.2) even when analyzed using the Technical Specifications limits for. scram speed (67B, r u-r). a Thus, the FWCF event at EOC-2000 MWD /ST is not a limiting event and will not be reevaluated on a cycle-by-cycle basis. 1 i For purposes of this evaluation a safety limit CPR of 1.07 was chosen. This is the value originally used by the fuel vender for BXB D-lattice fuel. The safety limit CPR is dependent on the specific fuel design and is provided by the fuel vendor. 4-47 i l
f' d 'I-- Table.4.2.1 Summary of RETRAN Model Uncertainty for a Feedwater Controller Failure
.]
Parameter 9.DGeltainty -JRCPR
-I.' Nuclear Model A); Void Coefficient +21% .025
'B).-Scram Reactivity -11% .013 n' C)' Doppler Coefficient -15% .005
.D) Prompt Henting -20% .002 II. Core.T-H Model.
A) Core. Pressure Drop -1.5 psi .004 B) . Core Bypass Flow -10% .001 C) : Core Avg; Gap Conductivity -14.2% .004
' D) ' Hot Channel Gap Conductivity +17.2%' .001 E) Core Bulk Void (Slip) .Co + 10.7% .001 F)' Core Subcooled Void -60% .005 G) Fuel Pellet Radial Power conservative N/A Distribution III. Recirculation Model
. A) ' Jet Pump Efficiency +9% .005 B) Jet Pump. Inertia -67% .007 C) Recirculation Loop Inertia -67% .002 D) Separator Inertia +30% .001 E) Separator Pressure Drop conservative N/A 151.- Steam Line Model A) Steam Dome Volume -3% .003 f
B) Steam Line Pressure Drop -20% .010 l C) Steam Line Volume / Inertia conservative N/A l 1 J Total (4RCPR) .033 1 1 1
- l l
1 4-48 I I
- p p;
5 g h Table 4.2.2 i Response Surface t . Fitting Coefficients for FWCF at EOC
. Ao = 0.5021 x 10~1-
= -0.8889 x 10-4 H -A l A2= 0.1481 x 10~4 A=
3 0.2597 x 100 A4=- 0.1372 x 10-1 A5'= 0.1230 x 101 Fitting Uncertainty = 0.0031 Table 4.2.3 Statistical Adjustment Factor and Scram Speed Adjustment Factor for the PBAPS FWCF Event i Statenoint SAF JOLMCPR EOC 0.030 0.07 4-49
-+
4.2.2 Loss of Feedwater Heating The loss of feedwater heating (LFWH) event is a system
. transient characterized' by a gradual decrease in reactor l ' feedwater temperature. A loss of feedwater heating occurs-as the result of a reduction in extraction steam flow to one L or- more feedwater heaters. The reduction in feedwater temperature results in a decrease in core inlet enthalpy and a consequent increase in core power.
l The feedwater heating system . design is such that- the temperature reduction under any single failure condition is limited to less than 100 0F. Recent plant operating data indicate that this limit is more likely to be less than 60 0 F [5]. The probability of this transient is considered to be sufficiently low to categorize it as an infrequent event. However, due - to the lack of a sufficient data -base, this event is analyzed as an incident of moderate frequency. A description and analysis of this event are presented in the following sections. 4.2.2.1 LFWH Analytical Assumptions The following conservative assumptions are utilized in the evaluation of the LFWH event. Each assumption enhances the overall severity of the event. All other reactor parameters , are assumed to be at their nominal values.
- 1) A bounding value of 100 F is assumed for the loss of 0
feedwater heating.
- 2) No credit is taken vor APRM high neutron flux (120%)
scram.
- 3) core power distribution is assumed to be fixed. No credit is taken for power redistribution effects.
I 4-50
4.2.2.2 LFWH Analytical Methods The loss of feedwater heating event is a slow transient which begins and ends with the reactor in a condition of thermal equilibrium. Thus, it is evaluated using the steady-state three dimensional core simulator code, SIMULATE-E. The evaluation of the LFWH event and the determination of its severity are based upon a comparison of the initial and final reactor conditions. To determine the initial condition, a SIMULATE-E case is executed at the pre-transient statepoint (rated reactor power and flow, equilibrium fission products, rated feedwater temperature) with a representative control rod pattern at the beginning-of-cycle. The post-transient condition is then determined by executing SIMULATE-E in an iterative power search mode with the feedwater temperature reduced by 100 0F. The code is directed to reconverge on the critical eigenvalue from the previous case, with an identical rod pattern, fission product inventory and core flow. The expected increase in feedwater flow, which results from the increase in the core thermal power, is considered in the iterative power search. The calculated value of MCPR at the end of the transient is then compared to the initial value to determine the severity of the event. The use of the 3-D simulator for the analysis of the LFWH event is currently under review by NRC on Docket Nos. 50-277 and 50-278 [5]. This approach is consistent with that currently employed by the fuel vendor and approved by NRC I (34). 4-51
E i h 4.2.2.3 LFWH Event Evaluation Utilizing. the methods previously . described in .section 4.2.2.2, a LFWH analysis was performed for Peach Bottom 3 Cycle 7 (PB3C7).. The JCPR from this analysis was determined to be 0.13. To ensure that conservatism exists in the analytical methodology, several sensitivity studies were p'erformed by PECo to quantify the change in . LFWH event severity with respect to analytical assumptions ' and reactor ' parameters which may vary appreciably during normal operation, or which are capable of significantly changing the severity of the event. The analytical assumptions presented in section 4.2.2.1 were examined to quantify their individual effects on the severity of the LFWH event. The reactor power increase during the event was not sufficient to reach the APRM.high neutron flux (120%) scram setpoint. Thus, the assumption of no credit for a reactor scram has no impact on the severity of the event for this cycle. The assumptions of a bounding 1000 F feedwater temperature reduction and a constant core power distribution were previously examined in PECo-FMS-005 [5). The results of the evaluations are summarized here in Table 4.2.4. As indicated in the table, the conservatism inherent in these assumptions are significant.. I 4-52 i
.-n Table 4.2.4 ..
PB3C7 LFWH Event sensitivity Study Results i-Analysis Change Assumotion JCEE in eCEB
. Nominal 0.132 --
600F Loss of Feedwater Heating 0.064 -0.048 variable Core Power Distribution 0.076 -0.056 The LFWH event was found to be sensitive to two other reactor. parameters, namely, the cycle exposure and the
' initial core power / flow condition. The sensitivity of the LFWH event severity with regard to cycle exposure was examined by' reevaluating the event at several cycle exposure conditions- from beginning-of-cycle (BoC) to end-of-cycle (EOC), with all other initial conditions and modeling assumptions as noted.in sections 4.2.2.1 and 4.2.2.2. The results of this study are presented in Table 4.2.5 for both Peach Bottom 2 Cycle 7 (PB2C7) and PB3C7. The LFWH event is shown to be a weak function of cycle exposure, with the maximum severity occurring in both cases at a mid-cycle exposure point. This result is expected, as the core void coefficient of reactivity is greatest in magnitude near the middle of the cycle but is otherwise not a strong function of. cycle exposure. The nominal LFWH evaluation is performed ht BOC exposure conditions. Thus, the potential uncertainty in the predicted JCPR due to exposure dependencies is expected to be less than 0.02.
4-5'3
Table 4.2.5 1: , LFWH Event Exposure Sensitivity Study Cycle Exposure PB2C7 PB3C7 (GWD/T) 4GB 42B ' O.0 0.138 0.132-3.0 0.144 0.135 5.0 0.150 0.139 7.0 0.151 0.132 EOC 0.138 0.130 The final sensitivity study examined' the change in LFWH. event . severity . with reactor power / flow conditions. These conditions were chosen to correspond to the peripheries of the reactor operating map. (with the highest allowable power for a given core flow). In addition, the 104.5/100 q condition was chosen to examine the sensitivity of a potential reactor power uncertainty. These studies were performed at the cycle exposure point at which the event was found to be most severe. All other initial conditions and modeling assumptions were as described in sections 4.2.2.1 and 4.2.2.2. The results of this sensitivity study are presented in Table 4.2.6 for both PB2C7 and PB3C7. With the exception of the 85/61 power / flow condition, the LFWH event was shown to be a weak function of reactor power / flow conditions, with the maximum severity occurring at the highest power / flow ratios. This result is expected, as the core void coefficient of reactivity increases slightly in Thus, magnitude as the core average void content increases. the potential uncertainty in the predicted 4CPR due to power / flow conditions is expected to be less than 0.01. The difference in the predicted ACPR for PD3C7 at the 85/61 power / flow condition versus the nominal calculation was 0.037 (0.043 aCPR for PB2C7). This value is significant to the conservatism inherent in the with respect 4-54
l aforementioned analytical assumptions. However, the low core flow (61% NBR) condition associated with this L statepoint places the event in the operating regime where flow-biased multipliers (Kf factors) are applied to the operating limit MCPR (OLMCPR). As will be shown in section 5.0, the PB3C7 minimum OLMCPR during the cycle is 1.27. At 61% NBR core flow, the Kg factor is 1.067 (based on the 102.5% flow line). Thus, the minimum effective OLMCPR at' this core flow is 1.355, effectively increasing the margin to the safety limit CPR by 0.085. AS indicated in Table 4.2.6, this increase in margin outweighs the 0.037 JCPR increase at the 85/61 power / flow statepoint. Therefore, this statepoint need not be considered when evaluating uncertainties in the prediction of LFWH JCPR. Table 4.2.6 LFWH Event Power / Flow Sensitivity Study Margin Margin PB2C7 to PB3C7 to Power / sLMCPR Flow
- opf_B SLMCPR 49_EB 0.091 0.169 0.116 85.0/ 61 0.194 0.142 0.058 100.0/ 87 0.160 0.040 0.151 0.049 0.137 0.063 104.5/100 0.135 0.065 100.0/100 0.151 0.049 0.146 0.054 0.134 0.069 100.0/105
- Numbers refer to percent of rated The results of these sensitivity studies indicate that the uncertainties introduced by the analysis of the LFWH event at BOC and at rated power and flow conditions are small in relation to the larger conservatism resulting from the assumptions of a 1000F decrease in feedwater temperature and no core power redistribution. Therefore, it has been established that the analytical methods put forth in this j section provide a reliable and inherently conservative means for the analysis of the LFWH event.
4-55 l J
/
t 4.3 Reactivity and Power Distribution Anomaly Events Reactivity and power distribution anomaly events are those events which result in local positive reactivity insertions c.s the result of control rod or fuel handling errors. The events evaluated in this category are:
- 1) Continuous Rod Withdrawal Error During Power operation
- 2) Fuel Loading Error - Rotated Bundle (RBLE)
- 3) Fuel Loading Error - Mislocated Bundle (MBLE) 4.3.1 Continuous Rod Withdrawal Error During Power Operation The Rod -Withdrawal Error (RWE) event is a localized power anomaly event which results from the erroneous continuous withdrawal of a high worth control rod from the fully inserted position. The subsequent insertion of reactivity causes a spatial redistribution of power, as well as an overall power increase both of which result in a loss of margin to the critical power ratio safety limit. The event is terminated when either the rod is fully withdrawn, or the Rod Block Monitor (RBM) system inhibits further rod motion.
The PEco steady-state physics methods report [5] describes the PECo SIMUIATE-E based RWE methodology and associated qualification. For illustrative purposes, RWE results for l Peach Bottom 3 Cycle 7 are presented here. 4-56
L,. g. o, TABLE 4.3.1
. ROD WITHDRAWAL ERROR'RESULTS
- - PEACH BOTTOM 3' CYCLE 7 Rod position
.(Feet Withdrawn)' ACEE 4.5- .0.14 5.0 0.16 5.5 0.18
-6.0 0.20 I 9.0 0.27 10.0~ 0.28 12 '. 0 '
O.33 Rod Block Reading Rod Position (% Initial) (Feet Withdrawn) 104 4.7 105 5.2 106 5.8 107 8.5 108 9.5 109 10.0 i 110 12.0 f.- 1 L 4-57
7 e FIGURE 4.3.1
. Limiting RWE Rod Pattern
' Peach Bottom 3. Cycle 7 02 06 10. 14 18- 22 26 30 59 06 10 55 44 26 26 51 06 06 02 47 44: 30 30 ,
30 43 06 06 10. 14 39 .26' 30 35 .10 02 14 00 31 26 30 Notes: 1. Rod pattern is quarter core symmetric. Upper left quadrant shown on map.
- 2. Number indicates the number of notches withdrawn out of 48. Blank is a fully withdrawn rod.
- 3. Error rod is (26,35).
4-58 1 l.. _____hmm____.mu-_ _ _ .____.___m_ _m__m._m. -.____s- - _ _ _ _ _ - . . _u__ _ _ _ _ _ _ _ m -
4.3.2 Fuel' Loading Error - Rotated The Rotated Bundle Loading Error (RBLE) is a highly localized power anomaly event which results from the loading of a.. single fuel assembly in an improper orientation (i.e., operator fuel loading error). Furthermore, the rotated bundle is -assumed to be positioned in an unmonitored location while a symmetrically loaded (correctly oriented) monitored assembly is operated on thermal limits. Because of the relatively low probability of the RBLE, no other events or equipment failures are assumed to occur concurrent with the fuel loading error. In the case of a typical D-lattice BWR (e.g., Peach Bottom Units), an erroneously rotated loading (180 degrees) results in: 1) an overall increase in assembly power, and 2) a radial redistribution of the local pin powers across the assembly, relative to a correctly oriented assembly. If sufficiently pronounced, these effects could result in the rotated assembly approaching the MCFR safety limit for extended periods without being detected by plant systems. Therefore, although the RBLE event is of sufficiently low probability that it is rigorously classified as an accident, transient limits for critical power ratio will be applied as the figure of merit. 4.3.2.1 RBLE Analytical Assumptions The following conservative assumptions form the basis of Philadelphia Electric Company's RBLE methodology:
- 1) The fuel assembly is erroneously rotated 180 degrees relative to its design orientation as a result of the fuel handling operation. For D-lattice plants such as the Peach Bottom units, a 180 _ degree rotation 4-59
l q I results in the higher enrichment fuel rods being i positioned next to the wide-wide water gap, thus 1 maximizing the reactivity of the rotated assembly. ! l
- 2) The fuel lcading error is not detected through the f core reload verification process. {
j L 3) The error assembly is loaded in a location which is I not explicitly monitored by the in-core monitoring i system. Furthermore, the symmetrically loaded assem- {i bly which is explicitly monitored is operated at the licensed MCPR operating limit throughout the course , of the cycle.
- 4) The core is operated at nominal licensed conditions.
No other events or equipment failures occur concurrent with the RBLE.
- 5) The vide-wide water gap adjacent to the 180-degree rotated assembly is assumed to be of a constant width over the full axial length of the assembly when evaluating the effect of the rotation on assembly overall reactivity. Actually, a 180 degree rotated assembly.would be tilted toward the wide-wide gap due to the positioning of the channel fastener and spacer buttons relative to the top grid. The constant gap assumption overpredicts neutron thermalization in the wide-wide gap adjacent to the high enrichment pins, tending to overpredict the overall reactivity of the rotated assembly.
- 6) A chopped cosine, midp3ane peaked axial power distribution is conservatively used in the evaluation of MCPR.
- 7) Rotated bundle variable gap R-factors are supplied by the fuel vendor for use in the determination of the assembly critical power ratio (CPR). These R-factors are representative of the average effect of a variable width gap (i.e., tilted assembly assumed) at conservatively high void conditions. A 0.02 JCPR penalty will be applied to PECo RBLE evaluation to account for a non-conservative systematic bias identified by the fuel vendor when evaluating CPR for rotated assemblies. This penalty is consistent with the fuel vendor's current licensing basis as approved by NRC [34].
4-60 J t
4.3.2.2 RBLE Analytical Methods PECo RBLE evaluations are performed using a combination of lattice physics [5], thermal-hydraulic [1], and transient CPR [2] analytical methods. First, the CASMO-1 LWR fuel assembly burnup program is executed for a given fuel assembly lattice at core average hot, unrodded conditions for both the nominal and 180 degrees rotated orientation. Differences in lattice Ka values between the nominal and rotated cases are evaluated at a series of exposure points over the anticipated fuel burnup range. A RETRAN/TCPPECO hot-channel evaluation of the rotated case is performed next. Using the rotated R-factor assembly, together with the assumed chopped cosine axial power distribution, a search is performed for the critical bundle power which forces the calculated CPR to the safety limit. FIBWR derived solutions for in-channel flow are used in this power search iteration process. A second RETRAN/TCPPECO hot-chan-nel sequence representative of the nominal bundle orientation is then executed. In this case, the power level of the assembly is reduced from that of the rotated case by an amount proportional to the maximum K= ratio between the rotated and nominal CASMO-1 cases as evaluated over the fuel assembly burnup range. As in the rotated hot-channel evaluation, channel flow for the lower power, nominal orientation hot-channel case is determined with FIBWR. The nominal orientation R-factor is used in this hot-channel calculation, while the chopped-cosine power shape is held constant. The RBLE ACPR for the fuel type is defined to be the difference between the rotated and nominally oriented MCPR. The procedure is repeated for each fuel design in the core. The 0.02 4CPR rotated R-factor penalty is finally added to the limiting fuel design ACPR to establish the cycle specific RBLE MCPR. 4-61
I f t, I i 4.3.2.3 RBLE Event Evaluation PECo has applied the . RbLE analytical technique described herein to the evaluation of the Peach Bottom 3 Cycle 7 reference core. Figure 4.3.2 depicts the ratio of rotated to non-rotated reactivity as a function of lattice burnup for the PB3C7 RBLE limiting fuel type. The peak differential ~ reactivity between the nominal and rotated cases was conservatively-used to derive the RBLE JCPR. PECo ACPR results for the reference cycle RBLE evaluation are reported in Table 4.3.2 along with values generated by ! the fuel vendor es reported in the Peach Bottom reload licensing submittal [42). PECo results are in excellent agreement with the current Peach Bottom 3 Cycle 7 licensing basis. This agreement, when considered in conjunction with the aforementioned conservative RBLE modeling assumptions, serves to qualify the PECo RBLE licensing analysis methodology. TABLE 4.3.2 PEACH BOTTOM 3 CYCLE 7 RBLE RESULTS Corrected ICPR MCPR 4.C.EB aCPR* Fuel Vendor Results [42) 1.20 1.07 0.13 0.15 PECo Results 1.20 1.07 0.13 0.15
- 4CPR corrected to account for non-conservative bias in variable water gap R-factors as provided'by the fuel vendor.
.I i
4-62
- Y - T I V I _ T - C A EEI D RLO - . DV L - - AU NE -
.41 TBG I - -
N A - E 0R - 2.R4E 3 EGV 4.F4A F/ EI E - . - R D 50. R UDG O - GE3 I T-C, F A9D - . - T9E - O2D RB D - ' - . :::: [7 NR O OD R - N 8
/P N D U .
E T - A T - O - R 0 8 6 g, 2 0 O 0 n 0 0 8 o;e:::EE1o2w<OvxCO-8 m:$e3 o2m<O D
' ,i ,t !l
4.3.3 Fuel Loading Error-Mislocated The Mislocated Bundle Loading Error (MBLE) event is a low probability, localized power anomaly event which results from the loading of one fuel assembly into an improper location. This results in an actual' loading , pattern which is inconsistent with the reload licensing basis. Because an MBLE would require multiple independent errors and procedural ' violations by plant personnel, the event has a low probability of occurrence. Therefore, no additional abnormal events or equipment failures are assumed to occur simultaneously. The MBLE evaluation is performed to demonstrate that the critical power ratio safety limit would not be exceeded during otherwise normal operation. The PECo steady-state physics methods report [5] describes the MBLE methodology and qualification. A generic MBLE evaluation has been performed in this earlier report which demonstrates that the MBLE is a non-limiting JCPR event provided that the reload fuel assembly average enrichment does- not exceed 3.5 w/o U-235. PEco results for the reference cycle (Peach Bottom 3 Cycle 7) are presented in the report to support this conclusion. The assumptions used as the basis for this evaluation will be monitored to ensure their continued applicability. o 4-64
4.4 Increase in Reactor Coolant Flow Rate Events Transient events which result in an increase in reactor coolant flow lead to an increase in core reactivity and power (due to core void collapse)- which threatens the fuel cladding integrity due to overheating (i.e. CPR). The most severe event ir this category is the recirculation flow controller failure which is described below. 4.4.1 Recirculation Flow Controller Failure The recirculation flow controller failure (RFCF) is postulated to occur as a result of the failure of one of the two M-G set speed controllers (i.e. fluid coupler). Failure of the master flow controller can also result in a recirculation pump speed increase. However, the speed control loop contains error signal limiters that prevent a rapid speed increase. The failed fluid coupler is assumed to accelerate to full speed demand at a bounding rate (25% of full speed per_second). The most severe event results when the reactor-is initially operating at minimum pump speed on the 100% load line. This corresponds to approximately 57% NBR core power and 43% NBR core flow. This event is classified as an incident of moderate frequency. 1 1 4.4.1.1 Description of RFCF Event The recirculation flow controller event is initiated at minimum recirculation pump speed on the 100% load line and results in the following plant transient sequence.
- 1) One M-G set speed controller (fluid coupler) fails (t=0.0 sec).
- 2) Reactor APRM high flux (120% NBR) scram is actuated (t=2.3 sec).
- 3) Turbine control valves begin to close to maintain pressure (t=4.5 sec).
- 4) Reactor attains new steady-state with decay heat pcwer only (t > 100 sec) (not analyzed).
4-65
t. I. 4.4.1.2 RFCF Analysis Assumptions The analysis of the RFCF event is performed in a deterministic (i.e. conservative) fashion to result in the most : limiting minimum critical power ratio. The conserva-tive assumptions utilized in the analysis address the uncertainties in key input parameters that. influence the predicted value. of MCPR. These assumptions .are listed below.
- 1) Scram setpoints at Technical Specifications limits.
- 2) RPS logic delays at Technical Specifications limits.
- 3) Control rod scram speed at Technical Specifications limits.(67B).
- 4) Safety / Relief valve capacities at minimum ASME specified values.
- 5) Safety / Relief valve setpoints and. response charac-teristics at Technical Specifications setpcints limits and maximum design specification responsei times.
- 6) Bounding value of failed scoop tube insertion rate (25%/sec) is assumed.
- 7) Bounding values of core average and hot-channel gap conductivity are assumed.
- 8) Conservative values of core void, Doppler, and scram reactivity functions (point kinetics) ' are assumed.
- 9) No credit taken for flow biasing of APRM scram (120%
NBR scram setpoint assumed). The use of a deterministic approach to analyze this event
.results in a bounding value of predicted MCPR and precludes the necessity of a statistical analysis such as that performed in section 4.1.1.
l l l 4-66 !~
4;4.1.3 RFCF System Wide Response The failure of the M-G set speed controller results in a rapid increase in the jet pump flow of the failed loop. The associated increase in the jet pump discharge pressure is sufficienti to overcome the available driving head to the jet pumps of the normal loop and results in the total loss of
-jet pump flow in that loop '(in fact a slight flow reversal is observed). This is illustrated in Figure 4.4.1. The net effect is'a rapid increase in the core inlet flow. (Figure 4.4.2). The increasing core flow inserts positive void reactivity (Figure 4.4.3) due to a decrease in core. void content and results in a rapid increase ir core power (Figure 4.4.4).. A reactor scram on APRM high flux occurs (t=2.3 sec) and eventually terminates the nuclear transient.
The increase in core - power results in a large increase in core average heat flux (Figure 4.4.5) and reactor vessel steam - flow. (Figure 4.4.6) which then decrease after the reactor scram. The increase in core power results in a slight increase in reactor pressure (Figure 4.4.7), however, the RCPB is not challenged. After the reactor scram, the pressure begins_to fall rapidly. The collapse of the core voids due to the increase in core flow and the subsequent reactor scram results in a rapid drop in reactor water level (Figure 4.4.8). Feedwater flow (Figure 4.4.9) increases in response to the level drop. Eventually, the reactor stabilizes at a new steady-state condition with decay heat power only. This portion of the event is not analyzed as it is not critical with respect to the RCPB or reactor fuel integrity. The prediction of reverse jet pump flow through the normal loop results in the use of the jet pump model outside of its normal range of validity as specified in the RETRAN SER [26]. To evaluate the possible impact of the predicted flow 4-67
- [
,f .)
l-i s reversal, a ' reanalysis of the RFCF was performed in which .) j
< the reverse loss coefficients of the jet pumps were adjusted
~ to conservatively preclude the occurrence of reverse flow.
It was determined'that due-to the timing and small magnitude of the flow reversal, there was no significant'effect on the L predicted MCPR. . The typical limiting licensing events-(i.e. GLRWOB and FWCF) do not result in reverse jet pump flow, therefore this restricted use of the jet pump model outside of its normal range of validity has no significant impact on the determination of conservative plant operating limits. 4.4.1.4 RFCF Event Evaluation The time-dependent reactor power and thermal-hydraulic boundary conditions from the RETRAN system run were used to drive a RETRAN/TCPPECO hot-channel transient calculation to the safety limit CPR. The calculated JCPR was 0.29 (initial-CPR was 1.36) which is larger than that predicted for the GLRWOB at end-of-cycle (0.21). This result would appear to make the RFCF event more limiting. However, the low initial core flow (43% NBR) assumed when analyzing this event placee it in the operating regime where flow-biesed multipliers (Kf factors) are applied to the. operating limit MCPR (OLMCPR). Based on the results from the GLRWOB evaluation in section 4.1.1, the PB3C7 end-of-cycle option B OLMCPR is 1.32. At 43% NBR core flow, the Kf factor is 1.14. (based on the 102.5% flow line) . Thus, the effective OMCPR at this core flow is 1.50. Comparing this to the predicted ICPR (1.36) for the RFCF event indicates that the Kg factor provides a margin of 0.14 to the MCP3 safety limit. At higher reactor core flows, the Kf factor becomes smaller. However, the RFCF results become correspondingly less severe at higher initial core flows. Thus, the RFCF event is conservatively bounded by the GLRWOB event and will not be evaluated on a cycle-by-cycle basis. 4-68
6 P
- O P-
- - O O-
- L -
O- - L L- - F _ - A _ -- D- - 1 CW _- M R E- -
.FO __ -
O L- - IA-4.R FL ~
~
~
N F-4CP ~-
~
EO M ,
~-
~ - ,4 RE ~ ~
U U ~
~-
GC 7P - I F 3T y - BE - P J '
'e,
/
- i,
- e -
,2 d*',
I " . t i
/
- . . . - - . = - 0 0
0 0 0
.~0 0 0
O 0 0 2 8 4 4 1
- a m0cZMv bo k OEaL +0O C
>imw
M - - 6 FW - CO 2 .FL 4.R F - 4C ET - EO L -
- 4 i4 RE N -
U7I - GCE I F 3R BO PC - 4 4 s 2 m u
. - ~ - - - _ _
i e 0 *
,0 0 0 0
0 0 0 5 0 5 0 0 7 5 2 1 mTDZMv C %o k OLOO
,14
1 I ii1ll l1 8 R-- E. - L M. L. . A A- P. D. . R P T O T CO S- D l-O- V. . S
~ ,
,6 T .
/ .
N . E . N
'f' O .
FP . CM 3 .F O .'s ' .
)
- 4. R C - _ .
\
c 4C
\
\
\
t e S EO Y
. \
. \
T -
._) i 4(
REI 'e . N A e U7IV s .
' - N N
m GCT
. i s .
T F 3C I
/
i .
. s .
BA i .
/.
f N PE a'
. s R .
m' . E . s, R .
/
,~
O / . w
. i .
C .
. .* ,2
. . . " . ~
. . . 0 -
0 Q 0 Q 0 2 j 0 j 2 mMvh5._s0 ooh'
.' C A ,u-
, i!
,6 F -
4.C FUX . 4.R FL . 4C: N EO O
,4 RE R .
U 7T GCU I F 3E BN P
- ,2
- . ^
. e ~ 0 0 .
0 o _ 0 0 0 0 0 0 0 0 3 2 3 rn mCcZMV X3g COh3oZ
*bu i[! ( l ,'
t
- i6 X
U - __ FL - 5 CF -
.F T 4.R EA -
4C H - EO.
- i4 RE G UV 7
GCA F 3E B - PR - O - C
,2 u
f _ . m - . _ - - - - _ _ _ - . ,0 0 0 9 0 0 0 5 9 5 O 0 7 5 2 1 mW0zgvx 0 3 E 4ODI i
>Lu l l l' !
~ -
~
^
- ~
* - ,6
-W -
FO - CLF 6 .F -
- 4.R M 4CA -
EO E
; 4 T .
RE S U7 L - - GC E I F 3 - B S P S E - V - 4 * . S 9 ,2
- h M e . . . - ' ~ . a - "
,0
- ~ -
0 0 0 0 0 0 5 0 5 0 0 7 5 2 _ 1
. ) -
w o _L E OQ+-- mWDZ C u,E
, .i - ...
i6 E ~
~
R - U - - FS CS
- 7. F E
- 4. R R 4CP '
EOE '
.',4 REM -
U 7O - - GC D - FS I - - BM - PA - E - - T - S -
. :. ,z
. . - O o .
0 0 0 0 0 0 0 0 0 4 2 2 4
- n. .
) .
n .Ctr vt n n )L Q-tt a m._ tQv Q BIN
- i6
'L E
V . FE CL 8
.F 4, R R E
4CT . E' OA ' .
. ,4 REW .
U7 GC R I O F 3T B C P A E . R .
. li 2
- - . > 0 0 0 0 0 0 0 0 0 4 2 2
) l
%zV1LL(EI z O $ 0!.
> 1y
!l I
6 FW - CO 9 - -
.FL -
- 4. R F - -
4CR - - EOE T
- ,4 RE A U
GC 7W - D I - - F 3E - B - P E F - - i2
. W W d
e
- ~ .
~ . -
. . . 0 0 0 0 0 0 0 5 0 5 0 0 7 5 2 1
^rDZMv t C $ LO+0kcDO-L o
s 4.5 Decrease in Reactor Coolant Flow Rate Events
' Transient events which result in a decrease in reactor coolant flow threaten the fuel cladding integrity Jue to-inadequate heat removal (i.e. CPR). The most severe event in this category is the two recirculation M-G set trip which is described below.
4.5.1 Two Recirculation M-G Set Trip I The two recirculation M-G set trip is postulated to occur due to a simultaneous loss of all AC power to both M-G set drive motors. The loss of.AC power can occur as.the result of an automatic trip function (actual or spurious) or a manually ' initiated trip. This event is categorized as an incident of moderate frequency. A description and analysis of this event are presented in the following sections. 4.5.1.1 Description of Two M-G Trip Event
.A two recirculation M-G set trip results in the following plant transient sequence:
)
- 1) Trip of both M-G set drive motors initiated (t=0.0 sec).
- 2) Reactot high water level (LB) setpoint initiates main turbine and feedwater turbine trips (t=12.95 sec).
Turbine stop valve (TSV) cloaure initiates ranctor j 3) protection system actuation (t=13.02 sec). TSVs are - closed and torbine bypass valves start to open. Reactor control rods begin insertion (scram) j 4)
-(t=13.27 sec). 'i 1
- 5) Reactor low water level (L2) initiates HPCI/RCIC systems (not simulated). Water level stabilizes and recovers.
l 4-78 _ - _ _ _ _ - _ _ _ _ _ _ _ _ \
- - - - ~ -. - - - . ____
l: l .. 4.5.1.2 Two'M-G Trip Analysis Assun.ptions The analysis of the two recirculation M-G set trip event is
~
performed in a deterministic fashion as described in section 4.4 to result in the most limiting change in minimum critical power' ratio. The conservative assumptions utilized
.in'the analysis are identical-except for the-following.
i i 1) Reactor initial power of 102% NBR.
- 2) . Bounding scoop, tube insertion rate does not apply.
4.5.1.3 Two'M-G Trip System Wide Responst The trip of the two M-G set drive motors . results in a gradual decrease in recirculation pump. speed and. flow. The coastdown rate is governed by the rotational inertia of the M-G sets and the recirculation pump frictional and hydraulic-
. torque characteristics. The transient recirculation' drive flow is illustrated 'in Figure 4.5.1. The-reduction'in the
. recirculationidrive flow results in a' gradual reduction' in core flow (Figure 4. 5.2) . The reduction in reactor core flow . results in' an increase in core void content. As a result of negative void reactivity feedback, the core power decreases '(Figure 4.5.3). As the core power decreases, the void content equilibrates and a new steady-state ' power is- -
approached. The increase in core void content also results in a water level increase in the reactor downcomer.(Figure 4.5.4). Tuo difference between tim actual and sensed leve]s is due primarily to a difference in their respective "zero reference" points. At 12.95 seconds, the high water level (LB) trip setpoint is reached resulting in a main turbine trip and a feedwater turbine trip. The main turbine trip j results in the closure of the TSVs which terminates the steam flow to the turbine very rapidly (in 0.1 seconds). This causes a rapid pressure increase in the steamlines upstream of the TSVs which is partially mitigated by the 4-79
rapid opening (in 0.3 seconds)~of the turbine bypass-valves. The TSV closure also initiates a reactor scram. The resultant pressure wave propagates at sonic velocity to the reactor vessel.and causes a rapid pressurization there. The pressure' wave is reflected.several times between the reactor vessel and the closed TSVs. This oscillatory behavior is seen in the response of the vessel steam flow illustrated in Figure 4.5.5, but is not evident in the steam dome pressure response (Figure 4.5.6) due to the capacitance.effect of the reactor vessel steam done regien. The flow oscillations are mitigated when the reactor steam production rate drops below the bypass system capacity. The peak reactor pressure is. 1068 psig which is well below the RCPB safety limit of 1375 psig. The rapid pressurization in the reactor vessel inserts positive reactivity due to a decrease in core. void content. However, the positive reactivity is overcome by the negative -scram reactivity and the core is shut down. Thus, the pressurization results in no appreciable increase in core power. The transient core average i'uel clad surface heat flux response is illustrated in Figure 4.5.7. The heat flux decreases gradually as core power decreases during thu first phase of the transient (0-13 seconds). The heat flux then decreases more rapidly after the turbine trip and reactor scram. ! The high reactor water level trip of the feedwater turbines results in a rapid decrease in feedwater flow (Figure 4.5.8) and water level. Eventually, the low level (L2) trip setpoint would be reached, initiating HPCIS and RCICS. This portion of the event was not simulated since it is 4-80
( l l demonstrated ' in.- the ' LOFW analysis (section 4.6) that HPCIS alone'is sufficient to maintain _ adequate core coverage after a reactor scram. 4.5.1.4 Two M-G Trip Event Evaluation The ' time-dependent reactor power and thermal-hydraulic l- boundary conditions from the system run_were used to drive a RETRAN/TCPPECO hot-channel transient CPR calculation. No
-decrease in the CPR was predicted for the nominal-calculation. A sensitivity study was performed in which the M-G set coastdown time constant was reduced by a factor of two to increase the rate of the core flow reduction. This analysis also resulted in no decrease in the CPR. Thus, the two recirculation M-G set trip event is conservatively bounded by the GLRWoB event and will not be evaluated on a cycle-by-cycle basis, i
1 4-81
!1 4 0
2
- ~ -
~ .
~~ .
' .51 P
I R - - - T W - O - - - 1 GL -
.MF -
)
E - c
- 5. O V 4
e E WI
- - S TR k
- U 4 .0 (
R D 1 e UC - m GOC - i T ER I - - - F I 7C CE - - 3R - B - - P -
. 0 9 ,5 e
m o D - - _ - _ - _ _ O u . e ~
,0 .
0 0 0 0 0 0 5 0 5 0 0 7 5 2 _ 1 _
) .
mEmZ v 5O- G> LO
,5N _
y:g(L[;ill -
i! ll
= -
n g 0 2
& 6 9 4 4 9 - , .
- .
- 4 9 * # 8 . 4 .51 P
I R - TW - 2.G O 5 MLF
)
. c 4 .O T
. e WE
. S
.0 (
E R TL 4 * : e 9 * ;' .' * .
- . 4 . t . 4 N
1 e U CI . m GOE i T I F E R 7O CC 3 B - P
- . : ,.:. 5
,.t . f 4 4 9 9 t 4 94 0 S$ O . 4 t
- - - . . - . ,0 0 0 o, 0 0 0 o 0 2 8 4 1
CD ACCZMv3ou L loo
^
.Imu
, . - & ,5
. 1 P
I R . T . 3 GX .
.MU 5 L . -
4 .OF . E WN i0 R TO I' 2
. - 1 U C TR GOU .
I E E . F . - 7N - C . 3 . B . P 4 S l . 9 i5 m u . u s
. . - 0 0 0 ,
0 0 0 0 0 0 2 g 1 . WwZMv X3Eot+ OZ f I
s s
. - s
- s
- s
\
s
- s
\
,s
-s.
\
s - s
\ -
g
. s -
. s s
s s N - s s N
- . ... ,5
.~ -
s . - P L I s u - s RE - TV - E - GL -
- 4. M R -
5.O E 4 - T , .. - E W A T - : : :.... l
.a* . . . ' .
R W $. UCR e - GO I EO d_ e
- p -
F .T .- s __
- 7C .
- n_
e_ l CA .. -
- a S.._
SE u
. - t c R_
B P R .- A W_
- :..:..=.i5 p
. . . - ~ - -
0 Q 0 0 0 Q 0 0 6 4 2 ns n)q c_ o C_ V __ O > OJ_
,t x i o (f
~
~
5 : . -
.r ~ , : . . ,5 1
N P I
- ' i f
,'si',
RW - TO - 5 GLF - 5
.M -
+
M 4 .O A E WE
,0 R TTS
.- - ' * ' - . ' . 4 1
UC GOL W EE I -
- WO F S
- ' O L 7 S - L F F
CE - ' M S S - 3 ' B V A A . E P T Y P s B .
- .. ' . 4 . 4 5 : .e
- i5
. - - . - - _ - ' =
- ~
0 0 O 0 0 0 0 0 0 0 2 8 A 4 1 rECD Z *> v 3 O._k E o e._ .
>1 'e t'!1
~
~
~
- - 9 . ' 6
- - , 1
/ -
I PE - RR - TU S 6
- G S -
5
.ME R
4 .O P E W-T E ,0 4 -
' * ~ . . ' .
R M UCO - GOD - I E - F 7M - CA 3E - BT - PS - t - : ' : ;5
- a - .
. . - O 0 0 0 0 0 0 0 0 4 4 8
~ -
Om_v@$.xeL3( O nA ( u_Q AL~
. S 4 . - . i.E t : 9 %$ . ,5
. - 1 P
I RX . TU L GF . 7.M T . 5 A . 4 .OE E WH
.. - ,0 R T G
- . 4 4 1
UCV GO I E A F E 7 . - C R . SO . BC . P - . 5
. . ' . . . - 0 4 I * . S$
i5
== . _ - . . . -
,0 0 Q ,
0 0 0 g 0 0 2 g 4 1 AgDZMv X3- MQg3 - C
>Imm
.4 - . ' - . . .5 P
I R - T W - 8 GO -
.ML 5 -F 4 .O R E WE
- - ,0 TT
- ,A . 4 , ' . . . . < , . : '
R A - UC GOW - I F ED - E - - 7E - CF 3 B l P - -
' * . . , . . G e . !
- a . - - . , - ,0 0 0 0 0 0 0 0 O 2 8 4 1
WDZMv 3 C u O+O 3xeeg
.I=e 11
4.6 Decrease in~ Reactor Coolant Inventory Events Transient events that result in a decrease in reactor coolant inventory threaten the fuel integrity as the available coolant- becomes less able to remove the heat generated in the core. The most severe loss of- coolant transient is the ' loss of feedwater flow event which is described below. 4.6.1 Loss of Feedwater Flow A' loss of feedwater flow'(LOFW) results in a condition where the' mass flow out of the reactor vessel (i.e. steam flow) exceeds the mass flow into the reactor vessel ~ 1eading to a decrease in the coolant -inventory available to remove heat from the core. .The~1oss of.feedwater flow can be initiated by the failure of the feedwater control system or by a spurious trip ' o1 the feedwater turbines. This event is categorized as an incident of moderate frequency. A' description and analysis of this event is presented in the following' sections. 4.6.1.1 Description of LOFW Event A loss of feedwater flow event results in the following plant transient sequence.
- 1) Trip of all three feedwater turbines initiated (t=0.0 sec).
- 2) Recirculation M-G set speed runback initiated on coincident low water level (L5) and low feedwater flow (20% NBR) (t=6.5 sec)
- 3) Reactor low water level (L3) initiates reactor scram (t=7.8 sec).
- 4) Reactor low water level (L2) initiates recirculation M-G set trip and initiates HPCIS (t=16.9 sec).
- 5) Reactor water level stabilizes and recovers (t>60 sec) with adequate core coverage.
4-90 I I
\
l 1 4.6.1.2 LOTW Analysis Assumptions .;
]
l The analysis .of. the. loss' of feedwater flow event. is .q 1 performed in a. deterministic fashion as described in section. 4.4 to result in the most limiting loss in reactor water inventory. The conservative assumptions utilized- in the analysis are ' identical to those . used in the. RFCF event, except for.the'following.-
- 1) Reactor initial power of 102 %NBR
~
- 2) Bounding scoop tube insertion rate does not apply 4.6.1.3 LOFW System Wide Response The trip of the-three feedwater turbines results in a rapid coastdown-(in 8.0 seconds) of the feedwater flow. The loss of feedwater flow leads to a rapid decrease in reactor water level as the result of'a large steam-feedwater mass flow imbalance. :The coincident decrease in. reactor level and feedwater flow initiates a recirculation flow runback'.at 6.5' seconds.which is quickly followed by a reactor scram trip on low level (L3) at 7.8 seconds. The- recirculation flow runback initiates a gradual decrease in core power due to an increase in core void content. The subsequent reactor scram then quickly shuts the core down (Figure 4.6.1). The resultant decrease in core power leads to a gradual decrease
.in core-average fuel clad surface heat flux (Figure 4.6.2),
vessel steam flow (Figure 4.6.3), and vessel steam dome pressure (Figure 4.6.4). The steam dome pressure stabilizes as.the turbine electro-hydraulic control system responds to the decreasing pressure by stroking the turbine control valves closed in an effort to maintain pressure. The reactor scram also results in the collapse of core voids which contributes further to the decrease in reactor water level. At 16.9 seconds, the low water level (L2) trip setpoint is reached which initiates a recirculation M-G set l 4-91
b g g trip and results.in a gradual coastdown in core flow (Figure
~
4.6.5). The' L2. trip setpoint also initiates the HPCI and RCIC ; systems.. No . credit is taken for the RCICS actuat'en (HPCIS only credited in the analysis). The feedwater and-HPCI' system transient. flows are illustrated in Figure 4.6.6. The reactor scram . and subsequent HPCIS actuation reverses
'the initial' vessel mass f. low imbalance. This results in'the stabilization and eventual recovery of the reactor water level' (Figure 4.6.7) with more than adequate margin to core it- uncovery. The-lowest water level achieved is 90 inches (7.5 feet).above the top of active fuel.
The LOFW event results in a large loss of liquid inventory in the downcomer. region. Therefore, to prevent the upper downcomer volume.from becoming single phase during the event analysisl, volume (and hence inventory) is transferred from the . middle downcomer to the upper downcomer in the model. This is not a change in model nodalization per se, as no volumes are-'added or deleted from the model and no volume boundaries are changed. The overall downcomer volume and initial: liquid inventory are conserved. As a result, the upper downcomer volume has sufficient initial liqaid inventory to predict the transient without becoming single phase. In addition to the previously described model modification, a slight model nodalization change was required to analyze the'LOFW event. Initial calculations with the nominal PBAPS RETRAN model resulted in anomalous behavior when the falling mixture level in the upper downcomer volume resulted in the steam separator volume becoming single phase (all steam). In an attempt to maintain liquid inventory in the steam separator volume, the separator liquid exit junction (junction 10) orientation was changed from horizontal to 1 The RETRAN pressurizer modet is used to represent the upper downcomer. The RETRAN SER [263 restricts the use of the pressurizer model to two phase conditions. 4-92
,1 y
, b xy ' ? ,F
}f l J . vertical . ; This' . change helped to maintain liquid inventory ) in / the' , steam separator volume; but- resulted ;in large steam flow. rates (more' than -expected .from normal carryunder effects) through the. separator liquid. exit - junction. To ,a . counteract this phencmenon, a - control-transfer . volume was added: to the .model. . The two phase flow from the ' steam :
' separator ' liquid exitL . junction was directed into .the
' control-transfer volume.. Control logic . was ; then used to extractsthe excess - steam flow and redirect it: back'. to ' the
~
vesselc steam dome. region while conserving mass -and' energy. The remaining . steam flow due to carryunder.and the~1iquid flow-was? directed to the' middle downcomer. The additional Lvolume Lwas ' arbitrarily . sized at ~ 100 ft3, . Sensitivity. studies.have indicated that the results-of the_ analysis are insensitive to.'the' volume size. This model change is.a one
~ time application J not expected to be . used when performing.
standard reload licensing calculations. 4.6.1.4.LOFW Event Evaluation The . time;' dependent- reactor power and thermal-hydraulic boundary conditions from the system run were used to drive'a No
~RETRAN/TCPPECO hot -channel . transient CPR calculation.
decrease in-the CPR was predicted for the LOFW event. Thus, the LOFH event is~ conservatively bounded by the GLRWOB event, and will not be evaluated on a cycle-by-cycle basis. Furthermore, it has been demonstrated that the system response (HPCIS - actuation) to the - loss of feedwater flow provides adequate margin to core uncovery and thus maintains fuel- integrity.. The lowest reactor water level achieved during the event is 90 inches (7.5 feet) above the top of active fuel. y 4-93
~
_ ~ ~
~
W F X
- 1. O LU 6 - .
.L F -
4 CN - EO O - RE R - UTT . GCU . I - . F3 E - B N - P -
. 9 9 .:-
l :
. .: .9 . T $ e 9 b.
., m - - ~ . .
0 0 o, 0 0 0 0 0 2 8 4 1 1O C Z Mv X3 . C I J. Os_.e 3 D Z
*3oO t
~ -
~ -
0
,0
- - 1
~ -
~ -
0 X . .
- 8 U -
- WL F F 2.O T -
6.LA E 4C H EC . 0 6 RE G - U 7V GCA I - F 3E - B P R O - C 4 : . .
. T T 4 h : . !
0
- 4
~ -
~ -
- - ,0 2
i g
- . . ~ - _ . - -
,0 0 0 0 0 0 0 0 0 2 8 4 1
mxuZMv c X3 L .OO3 - i Aieu r i'
~ - .
~ -
. - . - ' - a . . .
W - WO F L 3 F - -
.O -
6LM 4C A EO T E - RE S U 7L - - GC E I F 3S - B S - P E - - V . .D : . h 6: . 4 8 e -l
- e O e 5 D 9 e:. 9
. - o . ^ ~ - . - ,0 0 0 0 0 0 0 0 0 2 8 4 1
Ea O12 E OO.M CcZMv . AI* *
,t ,'
,~- ,;r
,,l1,i
- 1 0
2 1
=. - - .
0
- 0
- 1 E -
0 R . ...... ..... .. . ? 8 U - WS - FS - 4.O E -
)
6.L R - c e 4CP
- S EOE -
0( 6 REM - e U 7O - m GCD I i T F 3 - - BM - PA E - T ..................: l...
- ... .:...::i:..0
.... ....... l 0
4 S -
- ,0 2
= " 0
- o. 0 0 0 o 0 0 0 3 6 9
~ - -
^.mv Q m.Z OL3mM? Q-
*1eN I,!
0 0
. . 1 N
D
. . i0 8
WW FO 5 . . 6L.OLF . 4C T E EOL . 0 6 RE N . U7 I . GCE I F 3R . BO . PC .
, R 4 . I 6 s : . 5 0 0
. . 4 0
i
. - - 0 0 Q ,
o . 0 0 Q o 0 2 1 g a g DC Z gv ho L a. Loo xiO=
0
,0
- - 1 W -
,0 8
O - WL F F
- 6. O I -
6L C R _. P
. - E _
4CH
- T _ -
- A _ -
EO/ _ ,0 W _ RE R E D E I_ 6 U 7T - C_ - E P_ - GCA I F H_ - F B3W - P D E - E . t t t t ,
-: .= .:.Oa ' ,*: ,
F - - n 4 j
- ,0
' 2
,i'
.f
_ - ~
~ - . 0 0 0 0 0 0 0 0 0 2 8 4 1
m1DZMv 3o-C r o' DOa3 . moo.L I*
/
0
- 0
- 1 L - .
0 E - 8
' V -
WE - FL -
- 7. O R - .
e 6.L E b m u q -
- p a -
g 4 CT - p m EO A e l a e s e 0 REW - u s a m e 6 U 7R - t c m u e ne e A - GC O I FST - BC - PA - E .
- 0 R
- -;: b : : R : 4 4
- ~
- ~
- i0
- 2 61 -
t
\
\ -
r
- - - ~ . ~
. . 0 0 0 0 0 0 0 0 0 4 4 8 mMOC.oc._.O>OI -
>i-l
p- , p m" j , g p {' 4'.7 ASME-Vessel Overpressure Protection Events h Transient- events which result in an increase in reactor pressure challenge the Reactor Coolant Pressure Boundary (RCPB)'. ' Primary system overpressure protection.is provided by a pressure relief system which is-designed to preclude an uncontrolled release of fission products. At PBAPS, the f pressure relief system is composed of 11 piped (to the torus) safety / relief valves and 2 unpiped' safety valves and is designed in compliance with Section III, Nuclear-- L Vessels, of the.ASME boiler and pressur'e' vessel code.- To. assess,the pressure relief system's ability ; to satisfy the requirements of the ASME code, the peak reactor vessel pre'ssure for the most - severe overpressurization event is
-compared to the most restrictive maximum pressure design requirement of the code. The most . severe event in this category is a simultaneous closure of all main steam isolation valves ("JIVs) with failure of the direct position switch scram signal.
4.7.1 MSIV Closure with Position Switch Failure The MSIV closure event is postulated to occur as the result of an inadvertent MSIV closure signal due to a single system malfunction or operator error. Section III of the ASME code defines four categories of conditions when evaluating overpressurization events: 1) normal, 2) upset, 3) emergency, and 4) faulted. An upset condition is defined as i any deviation from normal caused by any single error or f malfunction. An emergency condition is defined as having 'f
" ... a low probability of occurrence ... . Reactor vendor analyses have determined the MSIV closure event to be of
' moderate frequency, thus, falling in the " upset" category. ]
To make the MSIV closure event more severe, it is assumed that the direct position switch scram signal fails. The imposition of this additional failure results in this event 4-101
O i becoming a low probability occurrence, thus, application of I the " emergency" limit (1.2 x 1250=1500 psig) is appropriate. However, PECo conservatively applies the " upset" limit (1.1 x 1250=1375 psig), which is more restrictive. Because - the MSIV closure with position switch scram-failure event'has a low probability of occurrence, it is not considered when determining critical power ratio limits. 4.7.1.1 Description of MSIVC Event An MSIV closure event'with position switch failure results in the following plant transient sequence:
- 1) All MSIVs begin to close (t=0.0 sec).
- 2) MSIV position' switch scram fails.
- 3) Reactor APRM high flux (120% NBR) scram actuated (t=1.47 sec).
- 4) Reactor pressure increases to recirculation M-G set high pressure trip setpoint (t=2.50 sec).
- 5) Reactor pressure increases to safety / relief valve setpoints. Safety / relief valves open and discharge to. suppression pool terminating the pressure increase (t=2.59 sec).
- 6) Reactor low water level (L2) initiates HPCIS/RCICS (not simulated). Water level stabilizes and recov-ers.
4.7.1.2 MSIVC Analysis Assumptions The analysis of the MSIV closure event was performed in a deterministic (i.e. conservative) fashion to result in the most limiting predicted peak vessel pressure. In addition to the conservative application of the code " upset" limit instead of the applicable " emergency" limit, other conservative assumptions utilized in the analysis address 4-102 L-
_ _ _ _ _ = _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ L 'the uncertainties in key input parameters that influence the predicted value of peak pressure. These assumptions are J listed'below.
- 1) Reactor initial power of 102% NBR.
- 2) Scram setpoints at Technical Specifications limits.
3)- RPS logic delays at Technical Specifications limits.
- 4) Control rod scram speed at Technical Spec 4.fications limits (67B).
- 5) Safety / Relief valve capacities at minimum ASME specified values.
Safety / Relief valve setpoints and response charac-
~
6) teristics at Technical Specifications setpoint limits and maximum design specified response times.
- 7) An MSIV closure characteristic which results in a very conservative closure rate of 90% per second vs.
a maximum nominal value of 25% per second. The analysis of the MSIVC event is performed utilizing the 1 RETRAN 1-D kinetics option. However, the use of a g-
. deterministic approach to analyze this event results in bounding values of predicted-peak reactor vessel pressures and precludes the necessity of a statistical analysis such as that performed in section 4.1.1.
4.7.1.3 MSIVC System Wide Response The closure of the MSIVs results in the termination of steam flow through the valves. This causes a rapid pressure . I increase in the.steamlines upstream of the MSIVs which then j
< ' propagates at sonic velocity to the reactor vessel and causes a rapid pressurization there. The pressure wave is reflected sevt. hi timca between the reactor vessel and the closed MSIVs. This oscillatory behavior is seen in the vessel steam flow response illustrated in Figure 4.7.1, but is not evident in the steam dome pressure response (Figure 1
4-103 a 2_-_ . _ _
4.7.2) due to the relatively~small size of the oscillations-and the capacitance effect of the vessel steam dome region. The flow oscillations' and the pressure increase are mitigated when the safety / relief valves open at 2.6 seconds. The rapid pressurization in the reactor vessel inserts a large positive reactivity due to the rapid decrease in core void content. The positive reactivity causes a .large increase in core power. The core power quickly reaches the reactor APRM high flux (120% NBR) scram setpoint and the
- control rods are inserted. The negative scram reactivity eventually overcomes the positive void reactivity and the nuclear transient is terminated. The transient scram, void, Doppler, and total core reactivities are illustrated in Figure 4.7.3.
The transient core power (neutron flux) is illustrated in Figure 4.7.4. The power increases rapidly to a peak value of 270% NBR at 2.1 seconds as the total core reactivity increases. The power decreases thereafter as the total reactivity decreases due to control rod insertion. A small secondary peak is observed at 2.85 seconds as an increase in the positive void reactivity temporarily overcomes the negative ceram reactivity. The transient core average fuel clad surface heat flux response is illustrated in Figure 4.7.5. The heat flux increases rapidly as the core power increases and attains a peak value of 127.8% NBR at 2.3 seconds and then decreases as the reactor power excursion is terminated. The increase in heat flux enhances the reactor vessel pressurization caused by the closure of the MSIVs. The core inlet flow response is illustrated in Figure 4.7.6. The oscillatory behavior of the pressure wave is evident in the flow response. The trip of the recirculation M-G sets 4-104
i - on high'roactor pressure at 2.5 seconds results in a gradual reduction in core flow. as the - recirculation pumps - and M-G sets.-coast down. The reactor water level response is illustrated in Figure 4.7.7. The water level drops rapidly
-due to the collapse of core = voids and the . reduction ' of
-feedwater flow.
The feedwater flow (Figure 4.7.8) decreases due to increasing reactor pressure and the loss of driving steam'to
'the feedwater turbines with the closure of the MSIVs.
Eventually the low level (L2) trip setpoint is reached, initiating HPCIS and' RCICS. This portion' of the event is not simulated. It.has been previously demonstrated in the LOFW analysis (section 4.6) that HPCIS:alone is sufficient to maintain adequate core coverage after reactor scram.' 4.7.1.4 MSIVC Event Evaluation The peak reactor vessel pressure attained during the MSIV closure event was 1262 psig, well below the RCPB safety limit ~of 1375 p'sig. Thus, it has been demonstrated that the PBAPS RCPB can safely accommodate the challenge ~of the most severe overpressurization event. To insure that sufficient margin to the RCPB safety; limit is maintained, the MSIV closure event will be analyzed on a cycle-by-cycle basis.
/
4-105
' l i
W W-O O- - L L- - F F- - M F- - 5 A E- . E IL-T E-S R- - 4 W CO . - VL -
- 1. I F .
- 7. SMM 4 A - .
E CE . R OT
. - ,3 U ES l l
;>r .
G7L I CE l' F 3S BS - PE . V .
, . : . ,2
,1 m - -
m
~ ~ . - . . -
,0 a
0 Q 0 0 0 Q 0 0 _ 5 Q 5 _ 1 j ZMv)[ Ec2w bij ll
,5 E -
R 4 CU V S I S -
- 2. S E -
7.M R - 4 P - - C E OE
- 3 R EM U
O - G CD I 7 - F 3M - B PE A T : . : 1 ' '
- .* ' : ,::-l
- l l . . : - .
2 S -
.1
- . - _ _ _ - _ - - . , 0 0 0 0 0 0 0 0 0 0 0 0
_ 3 2 1 OmO_v o M._ e Su 5 n n D? .
- ( t
*iO
/
R. . E.. _- M . L- _ -
/
L , - A AP R. P .- D, l T O T C. O S- D e& V_ i S - T N _ E - N - - 4 O
~
/ -
CP
- V -
3IS M - O - 7. 4 M.C . E C-Y
- - \
OT
\
\' i3 R EIV
\
~
U ' - - N GC 7I T i
/ ' -
- N N
F3 C I ' - - - \
'r, - -
\-
\
A
- \
B E ' -
\ -
P ' x - R 's, -
.h]~,N
,' .- l
- i2 E -
R /,9t 9
- N -
- 1 -
O - i
%N C
] -
i1
- ~ ~ '
- ' - . - _ - ~
0 0 5 0 5 0 5 7 0 7 5 1 0 0 0 1
)
mNvy+.>,OOGO'
*l g.
e
,5
,4 C
VX I
- 4. S LU
- 7. M F .
4 . E CN R OO
. . ,3 U ERT GC I
7U . . F 3E BN . P
- ,2
,1
- . ~
~
. . . 0 o .
0 , 0 0 0 0 0 0 3 2 r mCmZMv X2L COL *-- Z
*18 l
i5 X -
,4
;U CL VF -
5I -
. ST -
7.M E A 4 E CH O.
- ,3 R EG U
GC 7V I A F 3E BR PO - C - 3 4 : ,2 i1
. W _
a m
. . M # 0 0 0 0 0 0 0 0 0 5 0 5 1 1 EmzMvX2u_&O$
> la m~ l
,5
,4 CW V
6I O .
.SL F
- 7. M .
4 T . E CE OL
- . ,3 R
U ENI G 7 CE . F3 R I BO .
~
PC i ,2
,1
. . . ~ - . . e a
,0 0 0 0 0
0 0 0 0 4 1 2 1 o t 8 2CZMv O e.' Lg
i5 L i4
- E CV -
V LE I
- 7. S R -
- 7. M E 4-CT E OA - 3 R E W -
U G CO 7R I F 3T
- BC PA E -
2 R B B i1
. m i0 0 0 0 0 0 0 0 0 4 2 2 wI6LLjEl t
Z v $0 ui "
I1, w 6
- 5 4
CW V - 8 I O -
.S L -
)
7.M~F - c e 4 R E C E -
- S
( OT
- - i3 -
R EA e U - G 7W - i m I CD - - T F 3E - BE - - PF - 9 2 1 -
. m - - -
i ,0 0 0 0 0 0 0 0 0 2 8 4 _ _ 1 _ ATDZMv hoc L C 03-a Om I:" m m _ u
!s 5.0 Reference Cycle' Analysis Summary
. - The results of the analyses described in section 4.0 define l- the' licensing basis for future ' PECo Peach Bottom reload safety' evaluations. .The licensing basis defines those abnormal operational transient's which are typically limiting.
or near-limiting with regard to thermal margins ~(i.e. MCPR) and overpressure protection, and require reevaluation' on a cycle-byi-cycle basis. Of the ten events analyzed in section 4.0, six were determined to be limiting or near limiting. These were:
- 1) .GLRWOB at EOC and EOC-2000
- 2) FWCF'at EOC
- 3) LFWH
- 4) RWE
- 5) RBLE
- 6) MSIVC The other events analyzed were determined to be conserva-tively bounded by those listed above, and do not require reevaluation on a cycle-by-cycle basis. Future plant
. modifications that have the potential to- affect the licensing basis will be monitored and evaluated for their impact.
5.1 Peach Bottom 3 Uycle 7 Licensing Analyses The analyses presented in Section 4.0 were . performed using reactor conditions for Peach Bottom Unit 3 Cycle 7 (PB3C7).
- As the first application of . PECo reload safety evaluation methods, the PB3C7 evaluation will be referred to as the reference cycle analysis. Subsequent licensing evaluations can be compared to the reference cycle (and other previous licensing evaluations) for consistency and reasonableness.
5-1
The' following pages' contain a summary of PECo-generated j results for- this reference cycle in a standard ' reload- ] licensing report format.- A' similar report format will be utilized -: in future reload licensing'.submittals and appears.'-
]
in generic form in the Appendix. -It is noted,Lthat the CPR ! 1 operating limits reported here are' based on a CPR safety-limit of'1.07. More recent.. statistical ' arguments. presented by the' fuel vendor to NRC have reduced the'.CPR. safety limit for these.- D-lattice fuel types. . Future PECo- licensing-submittals will utilize appropriate' values for the CPR safety limit in this regard'. i 5-2
- - _ _ _ _ _ _ _ - _ - _ _ _ _ _. _ _ _ _ - _ _ _ ______________________________-__________b
. .__;=,_-_ .
g se., , , , , .
- e. .
.c.. i
'L
't.5 j ..-
PEco Supplemental Reload Licensing Report. .,
.a
[ay Nuclear Unit: Peach Bottom Unit 3 Cycle / Reload No.: ' Cycle 7 / Reload.6 Reference Cycle Analysis For-RSE Methods. Report Date: May, 1989 Y l l 5-3 =_ _ ,
E' o i k: ^ .\1
)
PLANT-UNIOUE ITEMS -l
.1. r 1 i
L
-Analysis Conditions: Appendix A
- 2. RELOAD' FUEL BUNDLES Bundle Name (Fuel Tvoe) Cycle Loaded Number Irradiated P8DRB299 5 196
'P8DRB284H- 6 56 P8DRB299 6 224 PBLTA1 6 2
'PBLTA2 6 2 New BPSDRB299H 7 140 BP8DRB299 7 11A
. Total -764
- 3. REFERENCE CORE LOADING PATTERN I Hominal previous cycle core average exposure at end of cycle: 20348 MWD /ST Minimum. previous cycle core average
. exposure at end of cycle from cold shutdown considerations: 20348 MWD /ST Assumed reload cycle core average exposure at end of cycle: 19533 MWD /ST l
Core loading pattern: Figure 1 5-4 b _ _ _ _ _ _ _ _ - _ _ . _ _ - _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
k }; ' ifl . . 4 4 .' CALCULATED CORE EFFECTIVE MULTIPLICATION AND CONTROL SYSTEM WORTH 'NO VOIDS. 20 DEG. C
.Beginning-of Cycle, k-effective Uncontrolled. l.,123 Fully Controlled 0.972
. Strongest Control Rod Out 0.990 R, Maximum Decrease in Cold Core-Shutdown Margin with Exposure into Cycle, Delta k 0.0
- 5. STANDBY LIOUID CONTROL SYSTEM SHUTDOWN CAPABILITY
,.a Shutdown Margin (Delta k) ppm (20 DEG. C, Xenon Free)
, 660 0.034
- 6. RELOAD-UNIOUE NUCLEAR DYNAMIC PARAMETERS 1 Void Fraction (%) 40.2
' Void Coefficient (C/%V) -19.45 Scram Worth (3 ft insertion) ($) -0.53 1 These-values derived at EOC exposure conditions.
5-5 w-_=-_____-_-_
- 7. RELOAD-UNIOUE HOT-CHANNEL TRANSIENT ANALYSIS INITIAL CONDITION PARAMETERS Exposure: BOC7 to EOC7-2000 MWD /ST Bundle Fuel Peaking Factors R- Power Bundle Flow Desian Radial Axial Factor (MWT) (1000 lb/hr) ICPR BP/P8x8R 1.75 1.40 1.051 7.386 111.2 1.11 Exposure: EOC7-2000 to EOC7 MWD /ST Bundle Fuel Peaking Factors R- Power Bundle Flow Desian Radial Axia l __. Factor (MWT) (1000 lb/hr) ICPR BP/PSx8R 1.53 1.40 1.051 6.450 115.5 1.28
- 8. SELECTED MARGIN IMPROVEMENT OPTIONS Transient Recategorization: No Recirculation Pump Trip: No Rod Withdrawal Limiter: No Thermal Power Monitor: No Improved Scram Time: Yes (Option B)
Exposure Dependent Limits: Yes Exposure Points Analyzed: 2 i 5-6
p.L
.9 . OPERATING FLEXIBILITY OPTIONS
. Single-Loop Operation: Yes N Load Line' Limit:
No Extended Load Line Limit: No (See' Appendix A)1 Inc.reased Core Flow No-(See Appendix A)1
. Flow Point Analyzed: ---
Feedwater Temperature Reduction: No (See Appendix A)1-
' ARTS Program: No Maximum Extended Operating Domain: No
~
- 10. CORE-WIDE TRANSIENT ANALYSIS RESULTS Exposure Range: BOC7 to EOC7-2000 MWD /ST Flux Q/A JCPR JCPR Transient (%NBR)- (%NBR) BP/P8x8R 'LTA2 Ficure GLRWOB- 277 109 0,04 ---
2 Exposure Ranga: EOC7-2000 MWD /ST to EOC7 Flux Q/A JCPR- JCPR Transient (%NBR) (%NBR) BP/P8x8R LTA 2 Fioure GLRWOB 395 127 0.21 --- 3 FWCF- 153 109 0.06 --- 4 Exposure Range: BOC7 to EOC7 Flux Q/A JCPR JCPR Transient (%NBR) (%NBR) BP/P8x8R LTA2 Ficure LFWH 117 117 0.13 --- N/A l 1 Planned for future reloads . 2 LTA bundles are bounded by BP/P8X8R results 5-7 L -----m
,e e
11..~ LOCAL ROD WITHDRAWAL ERROR (WITH LIMITING INSTRUMENT FAILURE) TRANSIENT
SUMMARY
Litniting Rod Pattern: Figure 5
-Rod Block Rod (Feet JCPR ACPR
.' Readina l Withdrawn)' BP/P8X8R LTA l 104 5.0 0.16 ---
105 5.5 0.18 --- 106 6.0 0.20 --- 107 8.5 0.26 --- 108 9.5 0.27 --- 109 10.0 0.28 --- 110 12'.0 0.33 --- Set Point Selected: 106
- 12. CYCLE MCPR VALUES Non-Pressurization Events Exposure Range:. BOC7 to EOC7 BP/P8X8R LTA 1 Loss of 100*F Feedwater Heating 1.22 ---
Fuel Loading Error 1.22' Rod Withdrawal Error 1.27 --- Pressurization Events 2 Exposure Range: BOC7 to EOC7-2000 MWD /ST Option A ODtion B BP/8Xa.B LTA l BP/8X8R LTAl GLRWOB 1.30 --- 1.15 --- 1 LTA bundles are bounded by BP/P8X8R results 2 Statistical adjustment factors applied to pressurization events are documented in the PECo Reload Safety Evaluation report PECo-FMS-006, dated May, 1989. 5-8 _ = . _ _ _ - _ _ _ _
l r' , , i.,,
. Pressurization Events.(cont.).
}.-.
. Exposure. Range: EOC7-2000 MWD /STLto EOC7-Oction A~ ,
Oction B' BP/8X8R- LTAl 'BP/8X8R LTAl,' FWCF- 1.23 --- 1.16 1.40 ' --- 1.32 --- GLRWOB. .i.
' 13. OVERPRESSURIZATION' ANALYSIS
SUMMARY
Steam Line Vessel Pressure Pressure ' Plant "ransient (osial (osia) Resoonse MSIV Closure (Flux Scram) 1225 1262 Figure 6 14 . . LOADD*G ERROR RESULTS Variable Water Gap Disoriented Bundle ~ Analysis:- Yes
'i 2 -
Fvent 49_EE Rotated Bundle Loading Error. 0.15-
- 15. CONTROL ROD DROP M YSIS RESULTS Control rod drop analysis results are provided by the fuel vendor. For Peach Bottom 3 Cycle 7, the vendor calculated Resultant Peak Enthalpy, HSB, was 264.6 cal /gm. j
( i i l i
~.
1 LTA bundles are. bounded by BP/P8X8R results ] 2 Includes.0.02 penalty for variable water gap R-factor uncertainty. ) 5-9
4 4 , I '
'16..' STABILITY' ANALYSIS RESULTS p
GE SIL ~)80 - recommendations have been - included in the : plant operat'ing . procedures - and. Technical . Specifications; there . fore,'no stability analysis -is; required. NRC approval for-
- s deletion' of a. cycle-specific stability analysis- is-documented in NEDE-24011-P-A-US.
L. 17 . - LOSS-OF-COOLANT ACCIDENT RESULTS, Loss-of-coolant accident results are provided by the ' fuel vendor. See " Loss-of-Coolant Accident Analysis for Peach Bottom' Atomic Power Station Unit 3", General Electric Company,_ December 1977 (NEDO-24082, as amended). 5-10
1 I I so 5a emEma0m0[Ds [DDE)DE)DE)D[g[DegJ[aOg]g] ss [0DE][DlDB8EJJ DD[0[DE][DCOD[DD_0EJ 54 [DE7@Dk)OOCDO[DEDF9CDDG@[D[DE) 5 [DEJbbYbbbbYbbbbbb-hbkbbbbbbE]EJ
- [DBbBbBb E BbM M M M M EbbBb0
- BBBbBbBbBbBbBbBbMMMMMMBb e-emessememememme 3 e - DE0 [DED !D5]F90 [DE] DO [D(0 Do rum [DCD rem D@ [De DO 00 34 ((b b-b b B D0 32-60[U0@@DO 3o - DO DO D[D@[DOFO[DO 000@0 D[D[D000[@DO @[D@@@[DO[ DEDO B0[D00000 D5][D(0DE)D0[DTD99900D[D 2
- [DBbBbBb E BbM M M M M EbbBb0
26e --{@{&(((( D(0 (D5][DO@{@ [D0 m TJ DO [& {@ & [@ [@ [@[@ {@ 24 - [DE0[DO[D(0[DO [DGJ D900 Drl]E8 [DO CDE [DE [D90M] OJD 22 - [DO D@ @O!D@ @@[DO f@ DO f@ [DO[D@[DO D@ @O D[0 ) zo - DCO[DCD [DE0 CDCO DI[DE DGJ [DCO [De [DEC EDE [DCO[De [DCD CD0 te - O@@@@E980E[DO@E[D@@@[D@[D@@@[D@OECD is 14 [DDMFIDE[! DEO D[0[Dr s$[D D00M]Dt0[D8 DCa[D e sm
][DE[DE[DO[DiO[DO[D%[DOO@OL >2 ExD Drat 0 mmD[00s]Dm De tdt 00s Do mmo s '
10 [Df0ET0[D80E0E9@@@[D@[DEFCOGO e e [D[Dr0[@0mD[DesD[cEeDCODt0D(0[a [DD@ [DOED@DOf@[DOD@9[D[0 4 i 0m010E00100001000CD III99999999999999l 1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 3133 35 37 39 41 43 45 47 49 51 53 55 57 59 FUEL TYPE i A = BP8DRB299H D = PBLTA1PBLTA2 B = P8DRB284H E = BP8DRB299 7 = P8DRB299
) C = P8DRB299 Figure 1. Reference Core Loading Pottern 5 - 11
7 i d[ rs ,' 1 e 3 COTtE FLOW 1 PRESSUI,E RISE (pt) 3 EELIEF VALVE FLOW
't NEUTRON FLUX 2 SAFETY VALVE FLOW 4 BYPASS VALVE FLCW 2 NEAT TLUx -
1 150 200-- r ; 1 1 f
!, 1
{.300, [ 2@ - f >-
.E
\.. ..i E
a i o .o (. M E' 0 s ,, y' i 100 -
$ 50 - ~t .
w l
-3 -
0 4 6 4 6 0 2 0 2 TIME (SEC) TIME (SEC) 3 TURB:NE STEAV " LOW I VObC- 3 SCRAM 1 LEVEL (in) 2 DOPPLER - 4 TOT AL
- 2 VESSEL STEAM FLOW - 4 FEEDWATER FLOW N .t i
1-3 1-
- l /
n e '* - s , p.
-E:
- a. o o- S o 0- f u
- l. 'y -
i u . -
- t ho %1
~
h W h i y .3 .. 4 . 100 .: 4 6 0 2 4 e
.0 2 TIME (SEC) TWE (SEC)
Figure 2 I Plant Response to GLRWOB at EOC7-2000 5-12
)
- Y c..
'. I NEUTRON' FLUX. . 3 CORE FLOW . 1 PRESSULE RISE (psi) 3 RELIEF VALVE FLOW - )
2 HEAT FLUX . 2 SAFETY VALVE FLOW 4 BYPASS VALVE FLOW 150 300 1 .
- p. .1 -} l
~
E 100 . .. d. . .}... S 200 . . . .
- Ol . , - W
- g. ) ,
2 b -
- j. ,-
l' b i - ' e3 ! 3 W. ,
...j -...
M E 300 N
,4g
.-E ' $0 1_
i . s I 1 ;f
! -i i 0
0
^ / ,
O 2 4 E O. 2 4 6 TIME (SEC) TIME (SEC) 1 LEVEL (in) 3 TURBtNE STEAu FLOW 1 VOID 3 SCRAM 2 VESSEL STE/M FLOW 4 FEEDWATER FLOW 2 DOPPLER 4 TOTAL' 209 2 l [. '
\
~
1 a'
; G 1-i ,
'100- - -- 1- - < '
s 5 i L c. O 2 0- - I i- -- g w-5 0 k-g '. "N '
$. O y M .1 ;
t i I ,
-100 -2 0' 2 4 6 0 2 4 E TIME (SEC) TIME (SEC)
Figure 3 Plant Response to GLRWOB at EOC7 5-13
e--
,1
,;, i 1 NEUTRON FLUX - 3 CORE FLOW 1 PRESSURE RISE (psi) 3 RELIEF VALVE FLOW 2 SAFETY VALVE FLOW 4 BYPASS VALVE FLoy;- -- l
' 2 HEAT FLUX 150 -!
' 150 - -
i
!1 i
; i -
1
'~
S.100 : E 100- -1, U l U
.g -
g b . b -l T i n. ? 5 .5 ;
'E' i M .
l g 30 -, ...,. ..L g 50 .;... _.
- i. ; l -
N i - 4 1 1 0 , 0 ' ...;; . 0 10 20 30 0 10 20 3: TIME (SEC) TIME (SEC) 1 LEVEL (in) 3 TUR6tNE STEAM FLOW 1 VOID 3 SCRAM 2 VESSEL STEAM FLOW 4 FEEDWATER FLOW 2 DOPPLER- .4 TOTAL 150 2
. - 1 i
! I j
g 1
. l ..
r v E 100 2' - - ti {W l1 E
" 5 o.
f i a --- o g .o 4
- 7 l
; o ;
E u : D 50- Y- l 6 m 1 ' 1 1 7 I : 4 3
-2 0 ,
? ,
0 10 20 30 0 10 20 !{ TIME (SEC) TIME (SEC) Figure 4 Plant Response to FWCF at EOC7 5-14 l
L,
,o l.
02, 06 10. 14 18 22' 26' 30 59' 06 10 l p , 55 44 26 26
~
! 51 06 06 02
-47 44 30 30 30 43 06 06 10 14 39 26 30 35 10 02 14 00 31 26 30 Notes: 1. Rod pattern is quarter core symmetric.
Upper left quadrant shown on map.
- 2. Number indicates the number of notches withdrawn out of 48. Blank is a withdrawn rod..
- 3. Error rod is (26,35).
FIGURE 5 Limiting RWE Rod Pattern ' Peach Bottom 3 Cycle 7 5-15
= - - - - - _ - _ _ - - - _ _ .
1 NEUTRON FLUX 3 CORE FLOW 1 PRESSURE RISE (psi) 3 RELIEr VALVE Flow 2 HEAT FLUX 2 SAFETY VALVE FLOW 4 BYPASS VALVE FLOW
*t". 330 i
1
$ 100 $200-E E a u
$ .o
- . - i 5 5 e e g 33 _
g300 ...; . , . . . 1
; g
- 2 e E C 2 4 5 T!vE (SEC) TivE (SEC) 1 LEVEL (in) 3 TURB1NE STEAv FLOW 1 VOID 3 SCRAM 2 V'.SSEL STEAu FLOW 4 FEEDv. AiER rLOW 2 DOPPLER 4 TOT AL 2:* --
2 g 1-v
.. e '
' ^ ~
% C a -
, 3 4 L C.-
O '
$ G U
3 ,
^
b t E 4 6 C-g g _3 ..,.. . . . i
-100 -2 0 2 4 6 0 2 4 (
TIME (SEC) TIME (SEC) Figure 6 Plant Response to MSIVC at EOC7 5-16 - ~ - - _ . -
l j' , . APPENDIX A.' M . ANALYSIS' CONDITIONS. This~section'shall provide a discussion of the increased-core flow,' final feedwater temperature reduction, feedwater-heater-'out-of-service, and extended. load line. limit analyses ~ and the' conditions under which they are applicable.. This section will-be completed in future reload submittals which. analyze.these conditions. 5-17
t. 1 V '; q, y c. < m
- i. < ,
l 6.0 References L! l
- 1. -Young, K. R. and S. A. Auve, " Steady-State Thermal j
Hydraulic' Analysis of' Peach Bottom Units 2 and 3 Using The.FIBWR Computer Code", PECo-FMS-001,-Philadelphia Electric. Company, February 1985.
- 2. Young,1K. R. and S. A. Auve, " Method for Calculating Transient Critical Power Ratios for Boiling Water Reactors (RETRAN-TCPPECO)", PECo-FMS-002, Philadelphia Electric-Company, November 1985.
- 3. Buckley, J. F., " Steady-State Fuel Performance Methods-Report", PECo-FMS-003, Philadelphia Electric Company, July 1987.
- 4. Olson, A. M., " Methods for Performing BWR Systems Transient Analysis", PECo-FMS-004, Philadelphia Electric Company, September 1987.
- 5. Hesse, S. R., " Methods for Performing BWR Steady-State Reactor Physics A9alyses", PECo-FMS-005, Philadelphia Electric Company, January 1988.
- 6. Philadelphia Electric Company, ~ " Updated Final Safety Analysis Report - Peach Bottom Atomic Power Station Units 2 and 3", Vols. 1-9, Rev. 6, J&nuary, 1988.
- 7. Engel, R. E., et al., "The Reactor Analysis Support Package", EPRI NP-1761, Electric Power Research Institute, Final Report May 1986.
- 8. Ahlin, A. and M. Edenius, "MICBURN - Microscopic Burnup in Gadolinia Fuel Pins", Chapter 7, Part II, Advanced Recycle Methodology Program System Documentation, AB Atomenergi, Studsvik for the Electric Power Research Institute, November 1975.
- 9. Pilat, E., " Methods for the Analysis of Boiling Water Reactors Lattice Physics", YAEC-1232, Yankee Atomic Electric Company, December 1980.
- 10. Ahlin, A., M. Edenius and H. Haggblom, "CASMO-1: A Fuel Assembly Burnup Program, User's Manual", rev. ed. S. R.
Hesse, PECo-FMS-CCM-003, Philadelphia Electric Company, August 1987.
- 11. Ahlin, A., M. Edenius, and H. Haggblom, "CASMO - A Fuel Assembly Burnup Program, Users Manual", AE-RF-76-4158, Studsvik Energiteknik, June 1978 (Studsvik Proprietary).
I 6-1 l l 1
) = _ _ _ = _ - _ - - _ _ - - . - -
I' A
~12.IAhlin,LA., "CASMO - A Puel Assembly Eurnup Program, Programmer'sJManual"; Studsvik/RD-78/9, Studsvik.Ener-p :giteknik, June 1978 (Studsvik Proprietary).
- 13. Cobb, W. R., B. S. Singer and B. L. Darnell, "NORGE-B Code Description", rev. ed. W. G.. Lee, PECo-FMS-CCM-004, l Philadelphia Electric Company, September 1987.
1 14.'Cobb, W. R., B. S. Singer and B. L. Darnell, "NORGE-B Code Description", RP976-3, Scie.nce Applications'Inc. for the Electric Power Research_. Institute, July 1983.
- 15. Dyszel,.A., et al., " Qualification of Steady State Core' Physics Methods for-'BWR Design and. Analysis",
PL-NF-87-001-A, Pennsylvania Power &; Light, April 28,
'1988.
16.. VerPlank, D. M., " SIMULATE-E: A Nodal Core Analysis Program for~ Light Water. Reactors, Computer Code User's
. Manual",-rev. ed. S.1R. Hesse, PECo-FMS-CCM-001, Philadelphia Electric Company, August 1987.
- 17. VerPlank, D. - M. , " Manual for the Reactor Analysis ProgramLSIMULATE",LEPRI-RP710-1,. Yankee Atomic Electric CompanyJfor'the Electric Power Research-Institute, August 1978.
'18. VerPlank,-D..M., " Methods for the Analysis'of Boiling Water Reactors Steady-State Core Physics", YAEC-1238, Yankee Atomic Electric Company, March 1981.
19.'Mosteller,,R. D. and R. S. Borland, "COPHIN Code
. Description", EPRI NP-1385 (Project 1252-3), Science Applications Incorporated for the Electric Power Research Institute, April 1980.
- 20. Poetschat, G., " Abbreviated PDQ-7/ HARMONY Users Manual with EPRI-ARMP Modifications", Chapter 9A, Part II, Advanced Recycle Methodology Program System Documenta-tion,RG.R.P. Consulting for the Electric Power Research Institute, March 1983.
- 21. Rothleader, B., "PDQ-7/ HARMONY User's Manual" Chapter 9B, Part II, Advanced Recycle Methodology Program System Documentation, Science Applications Inc. for the Electric Power Research Institute, March 1983.
- 22. Brown, A. W., T. A. McClure and R. J. Wagner, " Summary of PDQ-7 (IBM-360-370 Version) Input Data Requirements and Operating Procedures", ANCR-1061, Aerojet Nuclear Company, March 1972.
6-2 l
y E 1 .7
- 23. McClure, J. A., G. C. Gose and D. J. Denver, "SIMTRAN-E:
A SIMULATE-E to RETRAN-02 Datalink", rev. ed. J. P. L, Waldman, PECo-FMS-CCM-002, Philadelphia Electric Compa-ny, September 1987. 24.'Cabrilla, D. E., et al., " Methods'for the Generation of Core Kinetics Data'for RETRAN-02", TR-033-A, General L Public Utilities, May 1988.
- 25. McFadden, J. H., et al.,'"RETRAN: A Program for Transient Thermal-Hydraulic Analysis of Complex Fluid Flow Systems", Vols. I-III, Rev. 3, EPRI NP-1850-CCM-A, Energy Incorporated for the Electric Power Research Institute, June.1987.
- 26. Letter, A'. C. Thadani (NRC) to R. Furia.(GPU-Nuclear Corporation), October 19, 1988,.
Subject:
Acceptance For Referencing Topical Report EPRI-NP1850 CCM-A, Revisions 2 and 3 Regarding RETRAN-02/ MOD 003 and MOD 004.
- 27. Schultz, S. P. and K. E. St. John, " Methods for the Analysis of Oxide Fuel Rod. Steady-State Thermal Effects (FROSSTEY), Code /Model Description Manual", YAEC-1249P, Yankee Atomic Electric Company, April, 1981.
- 28. Schultz, S. P. and K. E. St. John, " Methods for the Analysis of Oxide Fuel Rod Steady-State Thermal Effects (FROSSTEY), Code Qualification and Application",
YAEC-1265P, Yankee Atomic Electric Company, June, 1981.
- 29. Philadelphia Electric Company. "TCPPECO User's Guide",
PECo Software Development Record No. 5021, August 1985. 30.-Letter, D. B. Vassallo (NRC) to J. B. Sinclair (YAEC',,
Subject:
1 Acceptance of RETRAN-02/TCPYA01 Methodology, September 15, 1982.
- 31. Gitnick, B. J., R. R. Gay, R. S. Borland and A. F.
-Ansari, "FIBWR: A Steady-State Core Flow Distribution Code for Boiling Water Reactors - Computer Code User's Manual", EPRI-NP-1924-CCM (Project 1754-1), Yankee i Atomic Electric Company for the Electric Power Research i Institute, July 1981. i
- 32. Ansari, A., R. R. Gay and B. J. Gitnick, "FIBWR: A Steady-State Core Flow Distribution Code for Boiling Water Reactors, Code Verification and Qualification i Report"., EPRI NP-1923, Yankee Atomic Electric Company for the Electric Power Research Institute, July 1981.
6-3
.,e
- 33. Ansari, A., et al., " Methods for the Analysis of Boiling '
i Water ReactorsLSteady-State Core Flow Distribution", YAEC-1234, Yankee Atomic Electric Company, December 1980.
- 34. " General Electric Standard Application for Reactor Fuel Supplement for the United States,": General-Electric Company Licensing' Topical Report, NEDE-24011-P-A-US, General Electric Company, September 1988.
- 35. Philadelphia Electric Company, " Final--Safety Analysis Report, Limerick Generating Station Units 1 and 2", Vols 1-19, Rev 56, January 1989.
- 36. Letter, J. S. Charnley (GE) to H. N. Berkow (NRC),
January 16, 1986,
Subject:
Revised Supplemental
-Information Regarding-Amendment II to GE Licensing Topical Report NEDE-24011-P-A.
- 37. General Electric, " Qualification of the One-Dimensional Core Transient Model for Boiling Water Reactors", Volume 1, NEDO-24154, General Electric Company, October, 1978.
'38. Lellouche, G.'S. and B. A. Zolotar, " Mechanistic Model for Predicting.Two-Phase Void Fraction for Water in Vertical Tubes, Channels,.and Rod Bundles", EPRI-NP-2246-SR, Electric Power Research Institute, February 1982.
- 39. Carmichael, L. A. and R. O. Niemi,'" Transient and Stability Tests at Peach Bottom Atomic Power Station Unit 2 at End of Cycle 2", EPRI NP-564, General Electric Company for Electric Power.Research Institute, June, 1978.
- 40. Letter,,F. Burrow (TV.A) to W. G. Lee-(PECo), February 13, 1987,
Subject:
BWR Statistical Adjustment Factor Evaluation-(BSAFE) Program. i
- 41. Forkner, S. L., et al., "BWR Transient Analysis Model Utilizing the RETRAN Program", TVA-TR81-01-A, Tennessee i Valley Authority, April 7, 1983.
- 42. Dennison,.D. K., " Supplemental Reload Licensing )~
Submittal for Peach Bottom 3, Reload No. 6", 23A4685 (Revision O), General Electric Company, April 1985. l 6-4 L
.) L t
i ', 9 APPENDIX ' _i' ;
, PEco. Supplemental Reload Licensing ' Report :
1 Nuclear Unit: Peach: Bottom Unit'- Cycle / Reload No.': Cycle - / Reload-'- Date:. -- , ---- .- !l i,
- 2 l
i L II' A-1 I
,1
\
~ %. ,
.y Kg W
h;; .
,e : q .y.
.1. PLANT-UNIOUE ITEMS,
.A Analysis Conditions: ~ Appendix A 2'. RELOAD FUEL BUNDLES Bundle Name-(Fuel'Tvoe) Cycle Loaded Number Irradiated
- , ~New t
Total 764,
- 3. REFERENCE CORE LOADING PATTERN Nominal previous cycle core average exposure at end of cycle: ----- MWD /ST
' Minimum previous cycle core average
. exposure at end of cycle from cold shutdown considerations: ----- MWD /ST Assumed reload cycle core average exposure at~end of cycle: ----- MWD /ST I .. , Core loading pattern: Figure 1 1
1 i A-2
-w; , . . .
;;7 3 ',
;[f ' \ +4'- 2
>> . . . _a Y
' .; ' >T .
af 47- ~ CALCULATED CORE EFFECTIVE' MULTIPLICATION AND CONTROL u ' SYSTEM-WORTH - NO VOIDS. 20 DEG. C ,
- . ;b . ,
r].P ; Beginning-fof' Cycle', . k-ef fective. M Uncontrolled -----
~-----
.l Fully ' Controlled.'
. Strongest' Control Rod.Out A
g. RifMaximum Decrease in' Cold.. Core
~
Shutdown Margin with 2xposure into Cycles. Delta-k 5 .- STANDBY LIOUID CONTROL' SYSTEM SHUTDOWN CAPABILITY Shutdown Margin (Delta k) ppjg. (20 DEG.-C.' Xenon Free) 1' 6 ', EELOAD-UNIOUE'NUCLET.R DYNAMIC PARAMETERS Void Fraction (%) --- Void Coefficient (c/%V) ---
- i. Scram Worth (3 ft insertion) ($) ---
1 These values derived at EOC exposure conditions. 9: A-3
- _ _ _ _ _ _ _ _ .__.,__.m.__
- 7. RELOAD-UNIOUE HOT-CHANNEL TRANSIENT ANALYSIS INITIAL CONDITION PARAMETERS Exposure: BOCx to EOCx-2000 MWD /ST Bundle Fuel' Peaking Factors R- Power Bundle Flow Desian Radial Axial Factor (MWT) (1000 lb/hr) ICPR XXXXX ---- ---- ---- ---- ---- ----
xxxxx ____ ____ ____ ____ ____ ____ Exposure: BOCX to EOCx-2000 MWD /ST with Increased Core Flow Bundle Fuel Pecking Factors R- Power Bundle Flow Desian Radial Axial Factor (MWT) (1000 lb/hr) ICPR XXXXX ---- ---- ---- ---- ---- ---- XXXXX ---- ---- ---- ---- ---- Exposure: BOCx to EOCx-2000 MWD /ST with Feedwater Ternperature Reduction and Increased Core Flow Bundle Fuel Peaking Factors R- Power Bundle Flow Desian Radial Axial Factor (MWT) (1000 lb/hr) ICPR XXXXX ---- ---- ---- ---- XXXXX ---- ---- ---- ---- Exposure: BOCx to EOCx-2000 MWD /ST with Extended Load Line Limit Bundle Fuel Peaking Factors R- Power Bundle Flow Desian Radial 3xial Factor fMWT) (1000 lb(hrl ICPR XXXXX ---- ---- ---- XXXXX ---- ---- ---- ---- A-4
1 1-1 Exposure: EOCx-2000 to EOCx MWD /ST l Bundle Fuel ~ Peaking Factors R- Power Bundle Flow Desian Radial Axial Factor (MWT) (1000 lb/hr) ICPR l t. L XXXXX ---- ---- ----- ---- ---- ---- l- XXXXX --"- ---- ---- ---- ---- ---- p Exposure: EOCx-2000 to EOCX MWD /ST with Increased Core Flow Bundle Fuel Peaking' Factors R- Power Bundle Flow Desian, Radial. Axial Factor (MWT) (1000 lb/hr)- ICPR XXXXX ----- ---- ---- ---- ---- ---- XXXXX. ---- ---- ---- ---- ---- ---- Exposure: EOCx-2000 to~EOCx MWD /ST with Feedwater Temperature Reduction and Increased Core Flow Bundle Fuel . Peaking Factors R- Power Bundle Flow ;
- Desian- Radial. ~~ Axial Factor- (MWT) '(1000'lb/hr) ICPR xxxxx ---- ---- ---- ---- ---- ---- 1 xxxxx ---- ---- ---- ---- ---- ----
. Exposure: EOCx-2000 to EOCx MWD /ST with Extended Load Line I Limit Bundle i Fuel- Peaking Factors R- Power Bundle Flow I Desian Radial Axial Factor (MWT) (1000 lb/_hr) ICPR XXXXX ---- ---- ---- ---- ----
xxxxx ---- ---- ---- ---- ---- ---- l l 1 1 A-5 l _1_12_.________ _ _
Q~, , a
- 8. -SELECTED MARGIN-IMPROVEMENT' OPTIONS-Transient:Recategorization: : '--
l Recirculation Pump Trip:.. Rod Withdrawal. Limiter: -- Thermal Power Monitor: -- Improved Scram Time: --
-Exposure Dependent Limits: --
, Exposure Points' Analyzed: --
- 9. OPERATING FLEXIBILITY OPTIONS 4
Single-Loop Operation: -- Load Line Limit:
-Extended Load Line Limit: - --
(See Appendix A) Increased ~ Core Flow -- (See' Appendix A) Flow-' Point Analyzed: -- Feedwater Temperature Reduction: -- (See Appendix A)-- ARTS Program: -- Maximum Extended Operating Domain: --
)
m e__.i-O--__-__ ...-.l.. - - . . -- .2 .
r i . 4 - 10. CORE-WIDE TRANSIENT-ANALYSIS RESULTE y':
' Exposure' Range: BOCx to EOCx Flux. Q/A 4CPR ACPR
- Transient- (%NBR) .(%NBR) XXXXX XXXXX F_icure -
.N/A-LFWH
- Exposure Range:-BOCx to EOCx-2000 MWD /ST Flux Q/A 4CPR 4CPR ..
Transient (%NBR) (%NBR) XXXXX XXXXX Fiaure. GLRWOB --- --- --- --- 2 Exposure Range: BOCx to EOCx-2000 MWD /ST with Increased Core Flowl
~'
Flux Q/A ACPR -4CPR: Transient- (%NBR) (%NB11). XXXXX XXXXX 'Fiaure GLRWOB --- --- --- --- 3 l Exposure Range: BOCx.to EOCx-2000 MWD /ST with'Feedwater' Temperature' Reduction and~ Increased Core--
.Flowl-Flux' Q/A 4CPR' 4CPR Transient- (%NBR) (%NBR) XXXXX XXXXX Ficure-GLRWOB --- --- --- --- 4 Exposure Range: BOCx to EOCx-2000 MWD /ST with Extended Load Line Limitl Flux Q/A JCPR ACPR Transient (%NBR) (%NBR) XXXXX XXXXX Ficure GLRWOB --- --- --- --- 5 1 See Appendix A A-7
ik,( ' l 7 ;c : Exposure Range: EOCx-2000 MWD /ST to EOCx Flux Q/A" 4CPR ACPR l Transient (%NBR)- (%NBR) XXXXX XXXXX Ficure t I a GLRWOB --- --- --- --- 6 FWCFl' --- --- --- --- 7 Exposure Range: EOCX-2000 MWD /ST to'EOCx with Increased Core Flow 2 Flux _. Q/A' JCPR 4CPR Transient -(%NBR) (%NBR) XXXXX XXXXX Ficure GLRWOB --- --- --- --- 8
' Exposure' Range: EOCx-2000 MWD /ST to EOCx with Feedwater Temperature Reduction and Increased Core Flow 2 a
Flux Q/A JCPR JCPR Transient (%NBR) L%NBR) XXXXX XXXXX Eig_un GLRWOB' --- --- --- --- 9 Exposure _ Range: EOCx-2000 MWD /ST-to EOCx with Extended Load Line Limit 2 Flux Q/A 4CPR ACPR Transient (%NBR) (%NBR) XXXXX XXXXX Ficure GLRWOB --- --- --- --- 10 4 1 1 This event is bounded by the GLRWOB event at all operating , f conditions for all Cycle x exposures. 2 See Appendix A. 1 I A-8 1
)
,s 1
l
..- 11. LOCAL ROD WITHDRAWAL ERROR (WITH LIli,ITING INSTRUMENT FAILURE) TRANSIENT SUMMARK Limiting Rod. Pattern: Figure 11'.
. Rod' Block Rod (Feet- JCPR JCPR Readina .i Mithdrawn) XXXXX XXXXX-4, .:: ;
~
Set Point Selected: - - - - 12.. CYCLE MCPR VALUES Non-Pressurization Events-Exposure Rangei BOCX to EOCx~ XXXXX XXXXX-
. Loss of 100'F Feedwater Heating . --- --
Fuel. Loading Error --- --- Rod Withdrawal Error- --- --- N > 1 A-9 _ _ _ _ _ _ _ _ _ _ 1
Pressurization Eventsl Exposure Range: BOCx to EOCX-2000 MWD /ST Option A Option B XXXXX XXXXX XXXXX XXXXX "GLRWOB --- --- Exposure Range: BOCx ' to EOCx-2000 MWD /ST - with- Increased Core Flow 2 Option A Option B XXXXX XXXXX XXXXX XXXXX GLRWOB --- --- --- Exposure Range: BOCx to EOCx-2000 MWD /ST with Feedwater Temperature Reduction and Increased-Core Flow 2 Option A Option B XXXXX XXXXX XXXXX XKXXX GLRWOB --- --- Exposure Range: BOCx to EOCx-2000 MWD /ST with Extended Load Line. Limit 2
~
Option A Ootion B XXXXX XXXXX XXXXX XXXXX GLRWOB --- ---
)
i
.l 1' Statistical adjustment factors applied to pressurization I events are documented in the PECo Reload Safety Evaluation report PECo-FMS-006, dated May, 1989. .2.See Appendix A )
A-10 l 1
Li p. Pressurization Events (cont.) Expcsure Range: EOCx-2000 MWD /ST co EOCx Option A Ootion B XXXXX XXXlX XXXXX XXXXX FWCF1' GLRWOB Exposure Range: EOCx-2000 MWD /ST to EOCx with Increased Core Flow 2 Option A Ootion B ZXX_KK XXXXX XXXXX XXXXX GLRWOL Exposure-Range: 'EOCx-2000 MWD /ST to EOCx with Feedwater Temperature Reduction and Increased Core Flow 2,3 Option A Ootion B XXXXX XXXXX XXXXX XXXXX
-GLRWOB Exposure Range: EOCx-2000 MWD /ST to EOCx with Extended Load Line Limit 2 Option A Option B XXXXX XXXXX XXXXX XXXXX
--- ~~~ -~~ ~~~
GLRWOB l I l 1 This event is bounded by the GLRWOB event at all operating conditions for all Cycle'x exposures. 1 2 See Appendix A. 3 These limits bound operation with Feedwater Heaters out-of-service (FWHOOS). I I A-11 i
. - _ _ . . _ _ _ _ _ _ _ _ -. __ _ _ _ . . _ _ _ _ _ _ _ _ _ _ . _ . _ _ - _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ __________..__.___________________________J
7.-,. ,
'13. OVERPRESSURIZATION ANALYSIS
SUMMARY
Steam Line Vessel Pressure Pressure Plant Transient (osic) (esia) Response MSIV Closure
.(Flux Scram) --- --- Figure 12
- 14. " OADING ERROR RESULTS Variable Water Gap Disoriented Bundle Analysis: Yes l
Event 4EP.E Rotated Bundle Loading Error ---
- 15. CONTROL ROD DROP ANALYSIS RESULTS Banked Position Withdrawal Sequence (BPWS) has been
-implemented at the Peach Bottom Atomic Power Station Unit x;
.therefore, the Control Rod Drop Accident Analysis is not required. NRC approval for BPWS is documented in NEDE-24011-P-A-9-US, September 1988.
- 16. STABILITY ANALYSIS RESULTS GE SIL-380 recommendations have been included in the plant operating procedures and Technical Specifications; there-fore, no stability analysis is required. NRC approval for deletion of a cycle-specific stability analysis is documented in NEDE-24011-P-A-US.
1 Includes 0.02 penalty for variable water gap R-factor uncertainty. A-12
- 1. ,
.j i
i l l'i 17. LOSS-OF-COOLANT ACCIDENT RESULTS Loss-of-coolant ~ accident results are provided - by . the - fuel' vendor. See o " Loss-of-Coolant Accident Analysis for-fPeach
'Bettom Atomic -Power. Station Unit 3", General-~ Electric
' Company, December 1977 (NEDO-24082, as amended)..
l 1 A-13
[ i 1 I So QDODOLCLOLODOD 5e 000D0D0D000D0000 Se DQDODQD0p0D00000D0000 10 0D DD QD DL 0 4 54 000D0D00 52 QDQDQDODQDOLODQDOLQDQD So 0000D00000000000D00000D000 4e Q00000000D0D000000000D0000 46 0D0D000D0D0D000D000000000D00 44 O00000000000000D000D0D0000 42 -- D00 Q00p0 D000 D000 D000 D0 D0 D0 D0 D0 D0 D0 40 - Q00D00Q0000000000D 000000000D Q0 s e - D0000000000000 D000 D00000 D000 D0 s345 --D0000000D0D0000 QD QD 000000000D 0p000 100D000D10CDQD D00000 D0 D000 32 - DO DO DO QO DO DO DO DO DO DO DO DD DD Dn Qo a o - DD D07D DD 707D OTO DO OTO 60 GD DD OTO CTO DO 2e-D000D000D00000D0D0D00000D0D0D0 26 - 07070 GD 700T0707D 0T070700T 70707070 24-D000D000D00000D0D00000D0D00000 22 - TO OD DD 70 DD DD DD I__ D OD 700D DD DD 700TO 20-D000000000D0D0D0D000000000D0D0 18-DDDDDDDDGDGDDDDDGDDDDDCDDDDDCD 15 0000E000010D00J 00Q0001000D00 14 70D0000T0000T0000T0000000DD070 32 0000D00000000D0000D0D0D000 ~ to - 700TOOTO700DDOTODOODETOD0
- b b b b b b bbbb 70 b 0 4
000000D00000D000 2 00D0D0D0000000 1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 3133 35 37 39 41 43 45 47 49 5153 55 57 59 FUEL TYPE ; l A=__________ D = __________ l B=__________ E = __________ C=__________ F=__________ { Figure 1. Reference Core Loading Pottern A-14
1 PRESSURE RISE (psi) 3 RELIEr VALVE FLOW 1 NEUTRON TLUX 3 CORE FLCw 2 SAFETY VALVE FLOW 4 BYPASS VALVE FLOW 2 HEAT FLUX 300 150 i S $oo. ...... . , . $ 200- i-E E a a o $ E E d u , 5 5 , 6 60- , 6 100-0 0 0 2 4 6 0 2 4 6 TIME (SEC) TIMF. (SEC) 1 VOID 3 SCRAM 1 LEVEL (in) 3 TURBINE STEAM Flow 2 DOPPLER 4 TOTAL 2 VESSEL STEAM FLOW 4 FEEDWATER FLOW 2 200 3 1-E 100- $ 2 0 E 5 b $O 0-r u z U C 5 E
- t. 0-y 5
a .1 - 2 100 4 6 2 4 S 0 2 0 TIME (SEC) TIME (SEC) Figure 2 Plant Response to GLRWOB at EOCx-2000 A-15
3 CORE FLOW 1 PRESSURE RISE (psi) 3 RELIEF VALVE FLOW 1 NEUTRON FLUX 2 SAFETY VALVE FLOW 4 BYPASS V ALVE FLOW 2 HEAT FLUX t50 300 h 100 - [200- < E E a a o o E E w w d 8
$ 60- ' 4- $100-0 0 4 6 0 2 4 6 0 2 TIME (SEC) TiuE (SEC) 3 TURBINE STEAM FLOW 1 VO.D 3 SCRAM 1 LEVEL (in) 2 VESSEL STEAM FLOW 4 FEEDWATER FLOW 2 DOPPLER 4 TOTAL 200 2 G
1- 4
$100- {
w E
- 5 b !i 0- -
o U z M t 5 E E 0- -+- g 0m . , . . . . . . _ l
-100 , -2 0 2 4 6 0 2 4 6 TIME (SEC) TIME (SEC)
Figure 4 Plant Response to GLRWOB at EOCx-2000 with Feedwater Temperature Reduction and increased Core Flow A-17
3 CORE FLOW 1 PRESSURE RISE (psi) 3 RELtEF VALVE FLOW 1 NEUTRON TLVX 2 SAFETY VALVE FLOW 4 BYP ASS VALVE FLOv. 2 HEAT FLUX 150 300 {ggo. . { poo . 4 4 a
=
$ o E E U
Y 100 - i-
$ 50 -
0 0 4 6 0 2 4 6 0 2 TIME (SEC) TIME (SEC) 1 LEVEL (in) 3 T'JRBINE STEAM FLOW 1 votD 3 SERAv 2 VESSEL STEAM FLOW 4 FIEDWATER FLOv. 2 DOPPLER 4 TOTAL 200 2 G 1-m
$100- ('
C
= 5 5 5 0-a
- U z
U C 5 E l b 0- - g 0 m .
+~
3 l l !
-100 <
-2 ,
2 4 6 0 2 4 6 0 TIME (SEC) TIME (SEC) Figure 5 Plant Response to GLRWOB at EOCx-2000 with Extended Load Line Limit A-18
1 NEUTRON FLUX 3 CORE FLOW 1 PRESSURE RISE (psi) 3 RELIEr VALVE FLC A 2 HEAT rLux 2 SAFETY VALVE FLOW 4 BYPASS VALVE FLOn 160 300 1-h100-
$200-tior 7 a:
E E d u E 5 6 50- ,6 100- - 8 0 0 0 2 4 6 0 2 4 6 TIME (SEC) TIME (SEC) 1 LEVEL (in) 3 TURE:NE STE AM rLOV. 1 VO!D 3 SCRAM 2 VESSEL STEAM FLOW 4 FEEDW ATER FLO A 2 DOPPLER 4 TOT AL 200 2 2 1-
$100- {W E
- 6
$ $ 0-o U
2 M t E i E b 0- - - - g 6 m .1 - - I 100 2 0 2 4 6 0 2 4 6 TIME (SEC) TIME (SEC) Figure 6 Plant Response to GLRWOB at EOCx A-19
1 NEUTRON FLUX 3 CORE FLOW 1 PRESSURE RISE (psi) 3 RELi'if VALVE FLOW 2 HEAT FLUX 2 SAFETY VALVE FLOW 4 BYPASS VALVE FLCV. 150 150
$100- $100-Q Q w a $ o E E O : 0 $ ! 5 6 50- -~
S 50- - i f i a 0 0 0 10 20 30 0 10 20 3:' TiuE (SEC) TIME (SEC) 3 TURBINE STEAM FLOW 1 VOID 3 SCRAM 1 LEVEL (in) 2 VESSEL STEAM FLOW 4 FEEDWATER FLOW 2 DOPPLER 4 TOTAL 150 2
! Q v
1- e
$100- - -
{ G W
" o CL L
o 8 0- + L-- U E . l w i l 8 y l { So. .. I . g 5 e - i 0 ,
-2 0 10 20 30 0 10 20 3 '.
TIME (SEC) TIME (SEC) Figure 7 Plant Response to FWCF at EOCx A-20
1 NEUTRON FLUX 3 CORE FLOW 1 PRESSURE RISE (psi) 3 RELIEr VALVE r.OW 2 HEAT Flux 2 SAFETY VALVE FLOW 4 BYPASS VALVE FLOW 150 300 g 100- 8 200-E E m a o 5 E E U $ i-60- $100-0 0 2 4 6 0 2 4 6 0 TIML (SEC) TIME (SEC) 1 LEVEL (in) 3 TURBINE STEAM FLOW 1 VotD 3 SCRAM 2 VESSEL STEAu FLOW 4 FEEDWATER FLOW 2 DOPPLER 4 TOT AL 200 2 2 1-m g 100- { w 7
= 5 5 i 0-o o
E U t 5 E 6 0- D 6 " a -100 -2 2 4 6 0 2 4 6 0 TIME (SEC) TIME (SEC) Figure 8 Plant Response to GLRWOB at EOCx with increased Core Flow A-21
1 NEUTRON rLux 3 CORE FLOW 1 PRESSURE RISE (psi) 3 RELIEr VALVE FLCW 2 HEAT FLUX 2 SAFETY VALVE FLOW 4 BYPASS VALVE FLOW 150 300 8 100- + ' o 200-d > W a & b b e .
, s -'
6 60- '- i 6 100 - o o O 2 4 6 0 2 4 6 TIME (SEC) TIME (SEC) 1 LEVEL (in) 3 TURE!NE STEAM rLOW 1 VOID 3 SCRAM 2 VESSEL STEAM FLOW 4 FEEDWATEP FLOW 2 DOPPLER 4 TOTAL 200 2 3 1- -
$ 100 -
q {
" '5 b i o 0-O E
d t 5 E 6 0- G l 6 e .1 - - i
-100 -2 0 2 4 6 0 2 4 6 TIME (SEC) TIME (SEC)
Figure 9 Plant Response to GLRWOB at EOCx with Feedwater Temperature Reduction and inct eased Core Flow A-22
s 1 PRESSURE RISE (psi) 3 RELIEF VALVE FLOW t NEUTRON FLUX 3 CORE rLOW 2 SAFETY VALVE FLOW 4 BYFASS VALVE FLOW 2 HEAT FLUr 150 300 L
$ 100 - ; $200- ;
2 ! 2 e a o
' o e- '
5 5 W W
$ 60- - $100-0 0 4 6 0 2 4 6 0 2 TIME (SEC) TIME (SEC) 3 TURBINE STEAM FLOW 1 VOID 3 SCRAM 1 LEVEL (in) 2 DOPPLER 4 TOTAL 2 VESSEL STEAM FLOW 4 FEEDWATER FLOA 200 2 2 1-
$100- y Q W
" 5 5 I 0-
- o o
E u t 5 E 6 0- - 3 5 e .3 ... . 100 2 4 6 0 2 4 6 0 2 TIME (SEC) TIME (SEC) Figure 10 Plant Response to GLRWOB at EOCx with Extended Load Line Limit A-23
02 06 10 14 18 22 26 30 59 55 51 47 43 39 35 31 Notes: 1. Rod pattern is quarter core symmetric. Upper left quadrant shown on map. FIGURE 11 Limiting RWE Rod Pattern Peach Bottom - Cycle - P A-24 _
1 NEUTRON FLUX 3 CORE rLow 1 PRESSURE RISE (psi) 3 REllEF VALVE FLOW 2 HE AT FLUX 2 SAFETY VALVE FLOW 4 BYPASS VALVE Flow 150 300 4
$100-
$200-E E a m
-o 5 E E u u L 50 -
a b 100-(
- 0. c
; 4 C 0 2 4 E O
TIME (SEC) TIME (SEC) 3 TURB:NE STE AM FLCw 1 V0;D 3 SCRAM 1 LEVEL (in) 2 VESSEL STEAM FLOW 4 FEEDw TER FLOA 2 DOPPLER 4 TOT AL 200 2 f g 1 . , . y
$100-7 E'
-m 3
$ $ 0-o U
E u t 5 E b 0- g 6 _1 .. . l l
-100 -2 , ,
4 6 0 2 4 6 O 2 TiuE (SEC) TiuE (SEC) Figure 12 Plant Response to M2 VC at EOCx A-25
APPENDIX A ANALYSIS CONDITIONS This section shall provide a discussion of the increased core flow, final feedwater temperature reduction, feedwater heater out-of-service, and extended load line limit analyses and the conditions under which they are applicable. This section will be completed in future reload submittals which analyze these conditions. i t A-26
e
- PHILADELPHIAELECTRIC COMPANY i
{ i 3
- ,_ m u ua w u m uua.w w w.s m s.u w as 6
l mjy i r F=- - o ( RachBottom Atomic PowerStation
. . _ . _ . _ . . . . . . . . . . . . . . ..}}