ML19310A233
| ML19310A233 | |
| Person / Time | |
|---|---|
| Site: | Saint Lucie  |
| Issue date: | 02/29/1980 |
| From: | ABB COMBUSTION ENGINEERING NUCLEAR FUEL (FORMERLY |
| To: | |
| Shared Package | |
| ML17208A685 | List: |
| References | |
| CEN-123(F)-NP, CEN-123(F)-NP-PT03, CEN-123(F)-NP-PT3, NUDOCS 8006060329 | |
| Download: ML19310A233 (100) | |
Text
CEN 123(F)-NP STATISTICAL COMBINATION OF UNCERTAINTIES PART 3 i
FEBRUARY,1980 E
POWER SYSTEMS j
COMBUSTION ENGINEERING. INC.
80 060 6 OM3
d t
0 LEGAL NOTICE THIS REPORT WAS PREPARED AS AN ACCOUNT OF WORK SPONSORED BY COMBUSTION ENGINEERING, INC. NEITHER COMBUSTION ENGINEERING NOR ANY PERSON ACTING ON ITS BEHALF:
A.
MAKES ANY WARRANTY OR REPRESENTATION, EXPRESS OR IMPLIED INCLUDING THE WARRANTIES OF FITNESS FOR A PARTICULAR PURPOSE OR ME RCHANTABILITY, WITH RESPECT TO THE ACCURACY, COMPLETENESS, OR USEFULNESS OF THE INFORMATION CONTAINED IN THIS REPORT, OR THAT THE USE OF ANY INFORMATION, APPARATUS, METHOD, OR PROCESS DISCLOSED IN THIS REPORT MAY NOT INFRINGE PRIVATELY OWNED RIGHTS;OR B. ASSUMES ANY LIABILITIES WITH RESPECT.TO THE USE OF,OR FOR DAMAGES RESULTING FROM THE USE OF, ANY INFORMATION, APPARATUS, METHOD OR PROCESS DISCLOSED IN THIS REPORT.
e C
C
=
G e
G
6 CEN-123(F)-NP STATISTICAL COMBINATION OF UNCERTAltlTIES METHODOLOGY PART 3:
C-E CALCULATED DEPARTURE FROM NUCLEATE BOILING AND LINEAR HEAT RATE LIMITING CONDITIONS FOR OPERATION FOR ST. LUCIE UNIT I t
ABSTRACT The three parts of the Statistical Combination of Uncertainties (SCU) report describe a method for statistically combining uncertainties involved in the calculation of the limits for the Reactor Protection and Monitoring Systems (RPS).
Part 1 of the SCU report describes the application of these new methods for the development of the Local Power Density (LPL'.
Thermal Margin / Low Pressure (TM/LP) Limiting Safety System Settings (LSSS's).Part 2 describes the statistical basis for a revised Departure from Nucleate Boiling Ratio (DNBR) corresponding to the Specified Acceptable Fuel Design Limit (SAFDL).
This part of the report, Part 3, describes the methods used to statistically combine uncertainties for the C-E calculated departure from nucleate boiling (DNB) and linear heat rate (LHR) limiting conditions for operation (LC0's).
4 Descriptions of the probability distributions of the LCO-related uncertainties and the stochastic simulation techniques developed for this program are presented.
The total uncertainties presented in this report are expressed in percent over-power (Pfdn, Pfdl) units for the DNB and LHR LC0's respectively, at the 95%
probability /95% confidence level limit.
Since the Required Overpower Margin (ROPM) is used to determine the LCO's, studies performed to determine the sensitivity of R0PM to these uncertainties 4
are also discussed.
H, S
i i
TABLE OF CONTENTS Chapter Page Abstract i
Table of Contents 11 List of Tables y
List of Figures vi Definitions of Acronyms and Abbreviations viii 1.0 Introduction 1-1 1.1 Purpose 1-1 1.2 Cackground 1-1 1.2.1 Protection and Monitoring System I
1.2.2 Previous Uncertainty Evaluation Procedure 1.2.3 Design Basis Event Transient Analysis Evaluation 1.3 Report Scope 1-2 1.4 Summary of Results 1-3 1.5 References for Section 1.0 1-3 2.0 Analyses 2-1 2.1 General 2-1 2.2 Objectives of Analyses 2-1 3
2.3 Analysis Techniques 2-1 2.3.1 General Strategy 2.3.2 DNB LC0 Stochastic Simulation 2.3.3 LHR LCO ~ Stochastic Simulation 2.4 Analyses Performed pa 2.4.1 DNB LC0 Uncertainty Analysis 2.4.1.1 Simulation Module SIGMA 2.4.1.2 ASI Uncertainty Simulation 2.4.1.3 Processing Uncertainty Simulation 2.4.1.4 Overpower Calculations with Respect to DNB LC0 2.4.1.5 Combination of Uncertainties 3
2.4.2 LHR LC0 Uncertainty Analysis 2.5 References for Section 2.0 2-7 ii
TABLE OF CONTENTS (continued)
Chapter -
Page 3.0 Results and Co'nclusion 3-1 3.1 Results of Analysis 3-1 3.1.1 DNB LC0 3.1.2 LHR LC0 3.2 Impact of Statistical Combination of Uncertainties 3-3 3.2.1 Impact on Margin to Limits 3.2.2 Impact on Consequences of DBE's Appendix A.
Basis for Uncertainties Used in Statistical Combination of Vi. certainties Al Axial Shape Index Uncertainties A-1 A2 Measurement Uncertainties A-3 A3 Monitoring System Processing Uncertainties A-3 A4 Reference for Appendix A A-5 B.
Summary of Previous Methods for Combining Uncertainties 81 LHR LCO B-1 82 DNB LC0 B-2 B3 References for Appendix B B-3 C.
Treatment of Uncertainties in Transient Analysis Cl Objective of Analysis C -1 C2 General Strategy C -1 C3 Analyses performed for Evaluation of R0PM for the
~
Limiting DBE's C -5 C 3.1 Loss of Coolant Flow Event C3.2 Single Full Length CEA Drop Event C4 Conclusions C -17 C5 References C -18 iii
LIST OF TABLES Chapter 1
~ Pace 1-1 Variables Affecting the LCO-Related Uncertainty _
l-5 1-2 NSSS Parameters Affecting the DNB and LHR LC0's 1-6 Chapter 3 3-1 Uncertainties Associated with the DNB and LHR LCO's 3-5 3-2 Impact of StatisticaLCombinaHen of Uncertainties on Margin to Limits 3-6 Appendix A A-1 Uncertainty [
] Components for the Evaluation of the LCO-Related Peripheral Shape Index A-4 Appendix C C -1 Design Bases Event and RPS Trip Protection C-19 C.2 Design Bases Event and Important Parameter Changes C-20 C -3 Uncertainties C-21 C -4 Key Input Parameters Used In the Loss of Coolant Flow Event C-22 C -5 Sequence of Events - Loss of Coolant Flow Event C-23 C -6 Comparison of Key Input Parameter Used in Safety Analysis and Best j
Estinate Cases for 4 Pump LOF Event C-24 C -7 Sequence of Events - Loss of Coolant Flow Event (Best Estimate)
C-25 i;
C -8 Key Input Parameters Assumed in the Single Full Length CEA Drop Event C-26 I
C -9 Sequence of Events - CEA Drop Event C-27 C-10 Comparison of Key Input Parameters Assumed in the Safety Analysis and Best Estimate Case for CEA Drop Event C-28 C-l l Sequence of Events - CEA Drop Event (Best Estimate)
C-29
I LIST OF FIGURES i
Chapter 2
' Page 2-1 Drib LCO Uncertainty Analysis 2-8 2-2 Excore Detector Locations 2-9
(
Appendix C C-1 Loss of Coolant Flow Event Core Flow Fraction vs. Time C-30 C-2 Procedures for Loss of Coolant Flow Event (STRIKIff-TORC Method)
-31 C-3 Procedures used to Determine Required Overpower fiargin During Loss of Coolant Flow Event (CESEC-TORC ilethod)
C-32 C-4 Loss of Coolant Flow Event Required Overpower fiargin (DilB) at 100% Power vs. Axial Shape Index C-33 C-5 Loss of Forced Coolant Flow Event Core Power vs.
Time C-34 C-6 Loss of Forced Coolant Flow Event Core Heat Flux vs. Time C-35 C-7 Loss of Forced Coolant Flow Event RCS Temperature vs. Time C-36 C-8 Loss of Forced Coolant Flow Event RCS Pressure vs. Time C-37 C-9 Loss of Coolant Flow Event (Best Estimate) 4 Punp Flow Coastdown vs. Time C-38 C40 Loss of Coolant Flow Event (Best Estimate)
Core Power vs. Time C-39 C41 Loss of Coolant Flow Event (Best Estimate) Core Average Heat Flux vs. Tine C-40 C42 Loss of Coolant Flow Event (Best Estimate) RCS Coolant Temperature vs. Time C-41 C4 3 Loss of Coolant Flow Event (Best Estimate) RCS Pressure vs. Time C-42 G.14 Procedures Used to Deternine Required Overpower liargin During Single Full Length CEA Drop Event C G15 CEA Drop Event Require,d Overpower fiargin (DflB) at 100% Power vr. Axial Shape Index C-44 616 Single Full Length CEA Drop Event Core Power vs.
Time y
C-45 J
LIST OF FIGURES (CONTINUED)
Figure Page C -17 Single Full Length CEA Drop Event Core Heat Flux vs. Time C-46 C -18 Single Full Length CEA Drop Event RCS Temperature vs. Time C-47 C -19 Single Full Length CEA Drop Event F.Cs reessure vs. Time C-48 C -20 Single Full Length CEA Drop Event Best Estinate Core Power vs. Time C-49 C -21 Single Full Length CEA Drop Event Best Estimate Core Heat Flux vs. Time C-50 C -22 Single Full Length CEA Drop Event Best Estimate RCS Temperature vs. Time C-51 C -23 Single Full Length CEA Drop Event Best Estimate RCS Pressure vs. Time C-52 G
vi
Definition s of Acronyms and Abbreviations ACU Axial shape index calibration uncertainty A00 Anticipated operational occurrence APU Axial shape index processing uncertainty AR0 All rods out ASI Axial shape index ASIDNB Axial shape index associated with P fdn l
ASI Axial shape index associated with P f ASIU Axial shape index units B
Unless specifically defined in context as representing AT power, B is used as core power 81 Rod average power at which fuel design limit or DNBR is reached for initial steady state B2 Power at which fuel design limit on DNBR is reached for transient conditions i
B Power level af ter inclusion of all DNB LCO uncertainties and allowances B
Power level after inclusion of the udR LC0 uncertainties and allowances BMU Power measurement uncertainty BMU Kth sampled value of BMU g
BOL Beginning of life BOPM Available overpower margin B pg Mean value of B distribution g
0PM B
The final B calculated from the.Kth simulation trial 0PM k 0PM B
COPM at lower 95% probability /95% confidence level of B g
i distribution L
vii
CEA Control element assembly CEAW CEA withdrawal CECOR Computer code used to monitor core power distributions CESEC Computer code used to simulate NSSS response to perturbations CETOP Computer code used to determine the overpower limits due to thermal hydraulic conditions CE-l C-E's critical heat flux correlation COAST Computer code used to solve conservation equations for mass flow and momentum CTM Centerline temperature melt DBE Design basis event DNB Departure from nucleate boiling DNBR Departure from nucleate boiling ratio f
Degrees of freedom Faug Fuel-densitication-dependent power peaking augmentation factor F
Primary coolant flow FDNB Coolant flow used to evaluate the ordered pairs (Pfdn. IP)
FLC0 Flow component of the DNB LC0 FMU Flow measurement uncertainty F
Total 30 nuclear power peaking factor including the effect q
of augmentation factors F
Total 3D nuclear power peaking factor including effects of 9
tilt and augmentation factors f
Fr Integrated radial pin peaking factor F
Core average axial power distribution peaking factor z
FTC Fuel temperature coefficient of reactivi.ty Fxy Planar radial peaking factor
~
~
I i Axial shape index of assembly i I
Core average axial shape index T(r)
Rod position dependent core average axial shape index for ROCS calculated power shape Ip Peripheral axial shape index I (r)
Rod position dependent peripheral shape index for ROCS p
calculated power share viii
IpRC(r),tpRControl(r)
Rod configuration dependent peripheral shape index based on control channel assembly weighting factors RS(r)
RSafety Ip
, gp Rod configuration dependent peripheral shape index based on safety channel assembly weighting factors (I-Ip) ROCS Difference between I and 1p for a ROCS calculated power distribution (I-Ip)CECOR Difference between T and Ip for a CECOR evaluated __-
power distribution (IpROCS_gpCECOR) Safety Difference between ROCS and CECOR calculated Ip using the safety channel assembly weighting factors (IpROCS_gpCECOR) Control Difference between ROCS and CECOR calculated Ip using the control channel assembly weighting factors k
Stochastic simulation trial number K
One-sided tolerance factor at the 95% probability /
95% confidence limit LC0 Limiting condition for operation LHR Linear heat rate LOF Loss of flow LPD Local power density LSSS Limiting safety system setting MTC Moderator temperature coefficient n
Normal distribution N
Total number of sampled cases in DNB i_C0 unceatainty analyses NA Not applicable NSSS Nuclear steam supply system P
Pressurizer pressure P
Axial integrated power of assembly i j
P1 Initial power level in CEA drop event analysis P
Final power level in CEA drop event analysis 2
DNB P
System pressure used in the calculation of the ordered pairs (Pfdne I)
P ix
Pfdl Power to fuel design limit on linear heat rate P
Pfdl for LHR LC0 including effects of azimuthal tilt PQ Pfdl for LHR LC0 not including the effects of azimuthal tilt P
Power to fuel design limit on CNB including fd n the effects of azimuthal til'.
PJdn Power to fuel desiqn limit rn 05B P fdnk Overpower from the kth simt.lation trial CETOP calculation LC0 P
Pressure component of the DNB LC0 PLCS Pressurization level control system PLHGR Peak linear heat generation rate PPCS Pressurization pressure control system PMU Pressure measurement uncertainty PU Uncertainty in predicting local power at the fuel design limit QUIX Computer code for solving the one-dimensional diffusion equations RCS Reactor coolant system ROCS Coarse mesh code for calculating power distributions R0PM Required overpower margin RPS Reactor protection system RSU Shape index separability uncertainty RTD Resistance temperature devices SAFDL Specified acceptable fuel design limit SAU Shape annealing factor uncertainty SC Approved credit in lieu of statistical combination of uncertainties SCO Statistical combination of uncertainties SIGMA Stochastic simulation code S100 Statistically combined uncertainty applicable ~
to the DNB LC0 SML0 Statistically combined uncertainty applicable to the LHR LC0 STRIKIN Computer code used to calculate fuel rod heat transfer i
X
T Azimuthal tilt allowance AZ T
Primary coolant inlet temperature, cold leg C
temperature T h Primary coolant hot leq temperature T DNB Inlet coolant temperatur<. used in calculating I"
the ordered pairs (Pfdn, Ip)
T LC0 Inlet temperature for the DNB LC0 in T 8.C0,0NB Inlet temperature for the DNB LC0 after accounting for the in temperature measurement uncertainty TM/LP Thermal marJia/ low pressure TMU Temperature measurement uncertainty TORC Code for calculating thermal hydraulic response to variations of system variables TORC /CE-1 Thermal hydraulic calculational model including CE-1 critical heat flux correlation W
Core average linear heat. ate gy g Wf Weighting factor of assembly i for excore detector set j
'g lC0-Maximum linear heat rate limit allowed by the rax LHR LC0 c><
Shape annealing factor 0PM'k kth sampled overpower uncertainty due to ASI AB uncertainties alp 3 Uncertainty in Ip due to uncertainty components other than electronic processing mI Uncertainty in Ip due to electronic processing P2 aP Pressure difference AT Temperature difference u
Axial shape index correction term uC r
l
]
l u2
[.
J i
UO(r)
[
-]
u SC
[
R j
uS E-3 Standard deviation o
xi L
l l
l
1.0 INTRODUCTION
1.1 PURPOSE Part 1 of the SCU report (I-l) describes the application of C-E's.nethod for statistically combining the uncertainties involved in the calculation of the limits for the local power density and thermal margin / low pressure limiting safety system settings (LSSS).
Part 2(1-2) describes the statistical basis for the revised departure from nucleate boiling ratio (DNBR) limit to be used in the evaluation of LSSS 's and limiting conditions for operation (LCO's).
The purpose of Part 3 of the report i-s to describe the method for statistically combining the uncertainties involved in the calculation of the limits for the DNB and LHR LCO's.
Uncertainties for the variables listed in Table 1-1 are considered.
1.2 BACKGROUND
1.2.1 Protection and Monitoring System The basic purposes and interactions of the LSSS and LC0's vere previously f
described in Section 1.2.1 of Part 1 of this report.
Part 1 describes the function of the protection system; Part 3 describes the function of the ONB and LHR LC0's.
Operation within the DNB and LHR LC0's provides the necessary initial DNB.and LHR margin to prevent exceeding acceptable limits during Design Basis Events (DBE's) where changes in DNBR and linear heat rate are important. A list of the Nuclear Steam Supply System (NSSS) parameters which affect the calculation of these LC0's is shown in Table 1-2.
A discussion of C-E setpoint methodology may be found in Reference 1-3.
1-1 1
1
1.2.2 Previous Uncertainty Evaluation Procedure The methods previously used to apply uncertainties to generate DNB and LHR' LC0's are presented in Reference 1-3 and are summarized in Appendix B.
As noted in Reference 1-3, these methods assume that all applicable uncer-tainties occur simultaneously in the most adverse direction.
This assumption is conservative.
Not all of the uncertainties are systematic; some are random and some contain both systematic and random characteristics.
As described in Reference 1-4, partial credit for statistical combination of uncertainties has been allowed for the DNB LC0 in view of the existence of this conservatism.
This report documents the methodology used to statistically combine the LCO-related uncertainties explicitly, in lieu of the credit previously used.
1.2.3 Design Basis Event Transient Analysis Evaluation The methods and procedures used in the report t) analyze DBE's were approved by NRC in References 1-5 and 1-6.
The purpose of the transient analysis evaluations is to determine the sensitivity of and variation in the required overpower margin due to the way the uncertainties are treated.
1.3 REPORT SCOPE
(
The scope of this part of the SCU report encompe,sses the following objectives:
1.
To define the methods used to statistically combine uncertainties applicable to the calculation of the DNB and LHR LCO's
]
l I
2.
To evaluate the aggregate uncertainties as they are applied in the determination of the DNB and LHR LC0's r
)
l t
3.
To evaluate how statistically combined uncertainties impact the l
selection of initial conditions for the transient analyses of DBE's, and to determine the magnitude of variations of R0PM within the range of the uncertainties of the key parameters.
t l-2
}
r One requirement for achieving the objectives is to define the probability distributions associated with the uncertainties being considered.
The develc,pment of those distributions which impact the LC0's differently than they impacted the LSSS's is discussed in Appendix A. To achieve the third objective, it is necessary to examine the sensitivity of R0PM to initial conditions for DNB and LHR-related DBE's. These evaluations are discussed in Appendix C.
The methods presented in this report are applicable specifically to the St.
Lucie Unit 1 (Florida Po~wer & Light Company) reactor.
1.4
SUMMARY
OF RESULTS The analytical methods presented in Section 2.0 are used to show that a stoch-astic simulation of uncertainties associated with the ex-core detector-monitored DNB and LHR LC0's results in aggregate uncertainties of [;
]
ressactively, at a 95/95 probability / confidence level.
The total uncertainties previously applied to the ex-core DNB and LHR LCO's are approximately [
]
respectively.
Therefore, the statistical combination of uncertainties program provides a reduction in the conservatism of the uncertainties applied in establishing the DNB and LHR LC0Js of approximately [
]
respectively.
The DBE sensitivity evaluations, described in Appendix C, show that the required overpower margin used in LC0 generation is relatively insensitive to the way un-certainties are combined.
1.5 REFERENCES
FOR SECTION 1 CEN-123(F)-P, Statistical Combination of Uncertainties, Part 1, 1-1 December 1979 1-2 CEN-123(F)-P, Statistical Combination of Uncertainties, Part 2, -
December 1979 l-3 CENPD-199-P, C-E Setpoint Methodology, April 1976 1-4 Docket No. 50-335, " Safety E' valuation by the Office of Nuclear Reactor Regulation", St. Lucie Unit 1, C.ycle 3, May 27,1979
/
l-3
REFERENCES FOR SECTION 1 (continued) 1-5 Letter, D. L. Zieman (NRC) to A. E. Lundvall, Jr. (BG&E) dated March 14,
~
1977, License Amendment 21 and SER for Cycle 2 Operation of Calvert Cliffs Unit 1.
Docket No. 50-317 s
1-6 Letter, R. W. Reid (NRC) to W._G.,_Counsil (NNEC) dated May 12, 1979, License Amendment 52 and SER for Cycle 3 Operation of Millstone Point Unit 2.
Docket No. 50-336 4
k 4
S s
i 1.
1 TABLE l-1 i
i VARIABLES AFFECTING THE LCC-RELATED UNCERTAINT.IES l
1 l.
Predicted integrated radial pin power at the fuel design limit 2.
Power measurement 3.
Shape annealing factor 4.
Shape index separability i
5.
Axial shape index calibration 6.
Equipraent processing of ex-core detector signals 7.
Flow measurement 8.
Pressure measurement i
l 9.
Temperature measurement l
O l-5
TABLE l-2 NSSS PARAMETERS AFFECTING THE DNB AND LHR LCO's DNB 1.
Core Power 2.
Axial Power Distribution l
3.
. Radial Power Distribution J
4.
Azimuthal Tilt Magnitude i
S.
Core Coolant Inlet Temperature 6.
Primary Coolant Pressure 7.
Primary Coolant Mass Flow LINEAR HEAT RATE l.
Core Power 2.
Axial Power Distribution D
3.
Radial Power Distribution 4.
Azimuthal Tilt Magnitude l
[
s l-6
2.0 ANALYSES 2.1 GENERAL The following sections provide a description of the analyses performed to statistically combine uncertainties associated with the DNB and LHR LC0's. The statistical combination technique involves use of the computer code SIGMA (Reference 2-1) to select data for the stochastic simulation of the DNB and LHR LC0 calculations.
The bases for the individual uncer-tainties not previously described in Part 1 (Reference 2-2) are presented in Appendix A.
The stochastic simulation techniques are described below.
2.2 OBJECTIVES OF ANALYSES B
The objectives of the analyses presented in this section are; 1.
To document the stochastic simulation techniques for the uncertainties associated with parameters that affect the LHR and the DNB LCO'.s, and 2.
To determine the 95/95 probability / confidence level uncer-tainty factors to be applied in calculating the LHR and DNB LCO'.s.
l 2.3 ANALYSIS TECHNIQUES
~
2.3.1 General Strategy The stochastic simulation code used for the statistical combination of the DNB and LHR LC0 related uncertainties is the computer code SIGMA.
It is l
described in Section 2.3.1 of Part 1 of this report (2-2).
2-1
2.3.2 DNB LC0 Stochastic Simulation For the Dh8 LCO, DNB overpower (Pfdn) divided by va required overpower margin (ROPM) is the dependent variable of interest.
The core coolant inlet temperature, reactor coolant system pressure, peripheral axial shape index and integrated radial peaking factor are the independent variables of interest.
As demonstrated in Appendix C, R0PM is relatively insensitive to these independent variables.
In addition, the maximum R0PM as a funcHnn nf shape index is used as input to generate the LC0's. This reduces the analytical evalus tion of the dependent variable to consideration of the Pfdn's response to the uncertainties of the independent variables. TORC /CE-1 (References 2-3, 2-4) is used to determine the functional relationship between Pfdn and the independent variables.
The probability distribution of uncertainties associated with some of the independent variables have been discussed in Appendix A of Part 1 of this report.
Those uncertainties specifically associated with only the LC0 are discussed in Appendix A of this part of the report.
The core coolant inlet temperature range of interest for the DNB LC0 stochastic simulation is defined by:
(1) the temperature at which the secondary safety valves open, and (2) the temperature at which the low secondary pressure trip would occur.
The reactor coolant system pressure range of interest for the DNB LC0 stochastic f
simulation is defined by:
(1) the value of the high pressurizer pressure tr:o setpoint, and (2) the lower pressure limit of the. thermal margin / low pressure trip.
It is noted that these ranges are the same as used in the LSSS stochastic
~
simulation (Ref. 2-2) and as such are bounding for the LCO.
l Figure 2-1 is a flow chart representing the stochastic simulation of 'the DNB limits.
This figure is similar to Figure 2-2 in Part 1.
The independent variables and their uncertainties are input to 11SE Each data set generated by SIGMA is evaluated with TORC /CE-1 to generate a Pfdn probability distribu-tion. The ratio of the mean value of Pfdn to the lower 95/95 value of Pfdn is the parameter of interest.
The details of the specific DNB LCO stochastic simulations performed are presented in Section 2.4.
2-2
2.3.3 LHR LCO Stochastic Simulation L
For the LHR LCO, the LHR LC0 overpower (P
) 1s the dependent variable fd of interest.
The three-dimensional (3D) pin power peak, the core average power level and the peripheral axial shape index are the independent variables.
The dependent variable is defined as LC0
~
p
~Wmax x 100 (2-1)
=
Wavg x F T q
where W is the peak linear heat rate allowed by the LHR LC0 and is m
determined by analvsis of DBE's.
W is the core average generated linear heat rate at rated power avg T
F is the synthesized core power peak including the effects of q
azimuthal tilting and augmented power peaking due to fuel densification.
In all other ways, the stochastic simulation procedure for the LHR LC0 is the same as the simulation procedure for the LPD LSSS described in Section 2.3.3 of Part 1.
2-3
2.4 ANALYSES PERFORMED
~
2.4.1 DNB LC0 Uncertainty Analysis e
Evaluation of the combination of uncertainties for the DNB LC0 is similar to the TM/LP LSSS analysis reported in Part 1.
The distributions of the uncer-talaties of the following parameten are input,to the analysis:
In order to combine the significant uncertainties in the same manner as shown in Figure 2-1 of Part 1, the LC0 stochastic simulation sequence shown in Figure 2-1 of Part 3 was used.
2.4.1.1 Simulation Module SIGMA The simulation process is carried out over all of the operating space, defined in Section 2.3.2, in the same manner as described in Section 2.4.1.1 of Part 1.
I 2.4.1.2 Axial Shape Index Uncertainty Simulation The basic relationships between the components of the axial shape index uncertainty were described in Appendix Al of Part 1.
However, only the set of ex-core detectors designated as " control channels" supply information for the calculation of the axial shape index used to monitor the LC0 on power versus shape index.
b 1
2-4
These detectors are used in the power ratio recorder monitoring system for St. Lucie Unit 1.
They do not have the same geometric placement as the safety channel ex-core detectors used in the evaluation of LSSS uncertainties described in Part 1.
The control clannel instruments are located at the same radial distance from the center of the core but at a different angular dis-olacement from the main diameter of the core compared to the safety channel instruments, and on the opposite side of the diameter (Figure 2-2).
Because of this angular difference, the shape index [
i] uncertainty components described in Appendix A of Part I are not appropriate for the stochastic simulation of LCO uncertainties.
The circuits used in monitoring the axial shape index with the control channel ex-core detector instruments for the LCO's also differ from those used for the LSSS.
The components of the LCO circuits therefore introduce a different unc.ertainty into the stochastic simulation process.
The shape index uncerta?nties specifically for the St. Lucie-1 control channels are developed in Appendix A of this report.
The procedure used to sample the shape index uncertainty distributions for the LC0 stochastic simulation are those described in Section 2.4.1.2 of Part 1.
The shape index uncertainties for the LC0 analyses differ from those for the LSSS analyses in the magnitude of [
] uncertainties based on control channel ex-core detectors rather than the safety channel ex-core detector uncertainties which were requ' red for the LSSS uncertainty evaluation.
2.4.1.3 Processing Uncertainty Sinulation As in the LSSS analysis described in Part 1, the signals generated by the ex-core detectors are processed into a power and an axial shape index (ASI) value. The electronic processing equipment introduces further uncertainty in these values.
Since the axial power distribution and the ASI value used in each simulation calculation are correl'ated, this uncertainty is incorporated in the stochastic evaluation of the LCO.
2-5
2.4.1.4 Overpower Calculation With Respect to DNB LC0
~
As in Part 1, the overpower limits due to reactor thermal-hydraulic condi-tions are determined by the code CETOP (Reference 2-5), which uses the CE-1 correlation, fil0P requires values of the pressure, inlet temperature, average coolant mass flow, and radial peaking factor, and calculates a limit on overpower.
2.4.1.5 Combination of Uncertainties As in Part 1, during each simulation trial (k),a calculation is performed to determine the ratio of the value of overpower at nominal (mean) conditions to the value at off-nominal conditions as the result of sampling values from the a1propriate uncertainty distributions.
These uncertainties are combined by using the following relations:
[
]
2-2 where
~"
2.4.2 LHR LC0 Uncertainty Analysis The stochastic simulation calculation used for the LPD LSSS uncertainty analysis 'in Section 2.4.2 of Part I was repeated for the LHR LC0 uncertainty analysis with only minor changes.
In the simulations, the overpower (E}j0) is derived from the technical specification value of the LHR LC0 limit.
The axial shape index uncertainty and the processing uncertainty simulations of Sections 2.4.1.2 and 2.4.1.3 were also applied to this analysis.
, 2-6
2.5 REFERENCES
FOR SECTION 2 2-1 F. J. Berte, "The Application of Monte Carlo and Bayesian Probability Techniques to Flow Prediction and Determination",
TIS-5122, February 1977
~
2-2 CEN-123(F)-P, " Statistical Combination of Uncertainties Part 1",
December 1979 2-3 CENPD-161-P, " TORC Code: A Canputer Code for Determining the Thermal Margin of a Reactor Core", July 1975 2-4 CENPD-206-P, " TORC Code:
Verification and Simplified Modeling Methods", January 1077 2-5 C. Chiu, J. F. Church, "Three-Dimensional Lunned Subchannel Model and Pr diction-Correction Numberical Method for Thermal Margin Analysis of PWR Cores," TIS-6191, June 1979 l
I 1
4 2-7
?
}
AX!AL POWER DISTRIBUTION eu 4
w C
p 3- ? ~
P3OBASILITY fd^
SIGIla E
F4 E DISTRIBUTICMS E E N' OT IMPUT
-D SAMPLING T
O E
l'ARAMETERS f.10DU LE P
g MAXIMUM VALUE OF S
Th ELECTRONIC f
T PROCESSING B
c o
CORE opm UNCERTAINTIES o
AVERAGE I
95/9-3 p
G c
ASI
!N g
opm A
- co rn
~3 h
d dl p2 l
V
---t ASI OVERPOWER
'P1 v5 I P"
E UNCERTAIflTY Pfdn + ABopm + BMU 0J 4
p SEf2SITI ITY w
STOCHASTIC (I )
-p p
R AION SIMULATOR O
W BUILDUP UNCERTAINTY DISTRIBUTION ON 2
OVERPOWER FOR N CASES
O EXCORE fBffR0fl roam a
O IFTECTOR LOCATI0flS 0
O 345 na 9
f t
O
?f 270 90 1
a5r@
~,
aa-8 O
n o:L
,g i
I8O _
-i; i
l EXCORE i+1ITR0tl DETECTOR ___.__
1 WF11 #
CHANNEL WELL #
CHANNEL l
5 SAFETY CHANNEL 1 (A) 8 c:acFTY CHANNFl 11 (T))
fi S&FETV CHANNEL 2 (a)f o
Cr+rmot cHANNF! l 7
SAFETY CHANNEL 3 (C 10 COffmOL CHANNEL 2 St. Lucie FIGURE Nuclear Pnwcr Station BCO E D E CTOR M TIO E Unit No. I 9-2 c
m
3.0 RESULTS AND CONCLUfl0NS 3.1 RESULTS OF ANALYSES The statistical ana'
- cal methods presented in Section 2 have been used to show that a stochastic simulation of uncertainti&s associated with the ex-core monitored DNB a nd LHR LC0's result in aggregate uncertainties of
[
]. respectively, at a 95/95 probability / confidence level.
Table 3-1 shows the values of the individual uncertainties which were statistically combined to yield the above aggregates. Appendix A contains a further discussion of the bases for these individual uncertainties.
The aggregate uncertainties are in units of percent overpower (Pfdn,P fdl) and are applied as such in the generation of the LC0 limits as discussed below.
3.1.1 DNB LCO The fuel design limit on DNBR for the DNB LC0 is represented by a combination of the ordered pairs (Pfdn' I ).
A lower bound is drawn under the " flyspeck" p
data such that all the core power distributions analyzed are bounded.
This lower bound is reduced by applicable uncertainties as follows:
(3-1)
(3-2) where:
lC0 B
- DNB power limit for LC0 aftel
' Jsion of uncertainties D
and allowances F
- Power to fuel design lim t on DNB including the effects fdn of azimuthal tilt SMD0 - Statistically combined uncertainties applicable to the DNB LC0 ASI
- Axial shape index associated with P DNE fdn-3-1
Temperature, pressure and flow components of the DNB LC0 are represented by equations as follows:
(3-3)
(3-4)
(3-5) where DNB DNB,TfB=Coolantconditionsusedinthecalculationsof F
,p (Pfdn' I ) ordered pairs of data.
p 3.1.2 LHR LC0 The LC0 on linear heat rate is represented by the ordered pairs (Pfdl' I )"
p A lower bound is drawn under the " flyspeck" data such that all the core power distributions analyzed are bounded.
This lower bound is reduced by the applicable uncertainties and allowances to generate the LC0 as follows:
l (3-6 )
l (3-7 )
i where:
Bh0 - Linear Heat Rate Power Limit for LCO after inclusion of-uncertainties.
PhC = the power to the LC0 linear heat rate limit. including the d
effects of azimuthal _ tilting.
SML0 - Statistically combined uncertainty applied to the LHR LCO.
3-2 4
3.2 IfiPACT OF STATISTICAL C0f1BIflATI0ft 0F UrtCERTAltlTIES 3.2.1 IllPACT O!1 f1ARGIf1 i
The motivation for using a statistical combination of uncertainties is to improve flSSS performance through a reduction in analytical conservatism in the uncertainties which must be taken into account. This section contains a discussion of the margin obtainable through a reduction in this conservatism.
Table 3-2 lists the uncertainty values previously used on St. Lucie I.
The approximate worth of each of these uncertainties in terms of percent overpower margin (P fdl) is shown.
fdn' The total uncertainties previously applied to the DilB and LHR LC0 are approxi-mately [
3,respectively.
The use of the statistical combination of uncertainties justifies a reduction in the conservatism in the uncertainty of approximately [
], respectively.
Although the conservatism in the uncertainty has been reduced, a high degree of assurance remains that acceptable limits will not be exceeded.
3.2.2 IltPACT ON CONSEQUEtlCES OF DBE'S The plant technical specifications restrict operation to within the DilB, LHR and equipment LC0's.'he statistical conbination of uncertainties only impacts the Ofl8 and LHR LC0's.For transient analyses of DBE's where changes in OflB and LHR are significant,the appropriate LC0 establishes the limits on initial
^
conditions assuned in the analyses.
Thus, the impact of uncertainties on these limits, and thus on the initial conditions fnr the transients, nust be evaluated.
As explained in previous sections the LC0's are generated based on the P (for DilB), PLC0 (for LHR) and the R0Pf1 for the limiting A00.
fdn fd l As explained in Appendix C,the maximum R0Pf1 which bounds the maximum variations in the R0Pf1 due to the range of uncertainties is used to generate these LC0's, Since the uncertainties will be combined statistically, the conservatism.
3-3
in the uncertainties used to generate the LCO's are reduced.However, the DNB and LHR LCO's based on the riethodology presented in this report will provide at least a 95% probability at a 95% confidence level that acceptable linits will not be exceeded during DBE's initiated from the extremes of the LCO 's.
i t
3-4
l TABLE 3-1 UNCERTAINTIES ASSOCIATED WITH THE DNB AND LHR LCO'S Uncertainty
- LHR LCO DNB LCO j
]
Core power (% of rated power)
+ 2%
+ 2%
Primary coolant mass flow (% Design flow)
NA
[
]*
Primary coolant pressure (psid)
NA i 22 Core coolant inlet temperature (*F)
NA 12 Power distribution (peaking factor) 7%
6%
l.
Separability (asiu)
See Table A-1 of Appendix Al 2.
Calibration (asiu)
[
][
+]
3.
Shape Annealing (asiu)
{
]
[
]
4.
Monitoring system processino (asiu)(2a)
[
]
~
5.
Monitoring system processing (psia)(2a)
- Note:
For complete description of these uncertainties, see Apoendix A.
3-5 1
TABLE 3-2 IMPACT OF STATISTICAL COMBINATION OF t!NCERTAINTIES ON MARGIN TO LIMITS Approximate Values of i
Equivalent Overpower Marqin Previous U"O Uncertainty Value LCO LCO Power 2% of rated Core coolant inlet temperature 2 *F Reactor coolant system pressure 22 psi Axial shape index:
Separatility
[
]
Shape Annealing
[
]
l Calibra tion
[
-]
[
]
Peaking factors 6% DNB, 7% LHR Equipment processing:
DNB LCO
[
]
LHR LCO
[
]
TOTAL Less Dreviously approved NRC credit for statistics Total Uncertainty Applied Previously Total Uncertainty Statistically Combined Net Margin Gain I
3-6
_.f I
APPENDIX A Basis for Uncertainties Used in Statistical Combination of Uncertainties l
l
-i e
A.1 Axial Shape _Index Uncertainties The discussion of Axial Shape Index Uncertainties given in Appendix A o'f Part 1 of this report (Al-1) is applicable to the stochastic simulation of DNB and LHR LC0' sand axial shape index uncertainties as well as to the generation of simplified uncertainty algorithms. However, some specific uncertainty values are not the same as in Reference Al-1 because the St. Lucie I ex-core detectors used for LCO monitoring are not l
at the same angular positions as the ex-core detectors used for the I
protection system. This angular displacement affects the assembly weighting factors used to calculate the peripheral shape index (I ) for power distribu-p tions generated by ROCS (Al-2) calculations or CECOR (Al-3, Al-4) evaluations of ex-core instrument signals.
The [
~
], however, is unaffected by the change in the mean values of -
- the factors because of the procedure used to calculate this uncertainty.
The effect of each ex-core channel's position has already been accounted for by the shape annealing factor averaging described in Section Al.3.3 of Part 1.
Thus, the shape annealing uncertainty component of the shape index uncertainty is also not affected by the difference between the angular position of the control channels and the safety channels.
A.l.1
[
.]
The f
] is calculated from the equation (Reference Al-1)
[
]
(Al-1 )
where
[
l Ip (r) is the rod position-dependent peripheral shape index calculated for the ROCS power distribution with the equation hPI R
gg Ip (r) = t
( Al-2)
)P 1
ig A-1
[' (r) is the rod position-dependent ROCS-calculated core average axial power distribution shape index 9 is the weighting factor of assembly i for excore detector set j P is the axially integrated power of assembly i' 9
l is the axial shape index of assembly i j
In order to use the existing [
] distribution, derived for the safety channels in Appendix A of Reference Al-1, for the LC0 simulations, the effect of the differences between the assembly weighting factors for the two sets of excore detectors must be considered.
Calculations of Ip (r) using the FP&L Safety Channel and Control Channel assembly weighting factors show that I R Control (r)-I R Safety (r) is rod configuration-independent.
p p
This difference has [
l
]
To incorporate this difference into the stochastic simulation of the LC0 uncertainties an [
] uncertainty distribution defined in Table Al-1.
A.l.2
[
]
Because the Safety Channel-Control Channel [
] difference is rod configuration independent and has been incorporated into the[ __
]
thet.
lterms are not modified for the uncertainty evalua-tions.
A.l.3
[
l Thel luncertaintyi
]was defined in Appendix A of Part 1 as:
[
]
(Al-3) where T
is the core average axial shape index I
is the peripheral axial shape index p
A-2
The difference ['
]
has been evaluated for the same cases as examined in Appendix A1, Part 1.
This difference hasi,
]
In order to incorporate this effect, an estimated uncertainty probability distributionwhichis[
] used in the stochastic simulations for the LCO uncertointies.
This distribution should be used with the stochastic simulation leading to the 95/95 probability / confidence level values of the DNB and LHR LC0 un-certainties.
1 A.2 Measurement Uncertainties The description of the measurement uncertainties given in Appendix A2 of Part 1 is also valid for the DNB LC0 uncertainty evaluation.
A.3 Monitoring System Processing Uncertainties The description of the Trip System Processing Uncertainties given in Appendix A3 of Part I is valid for the FP&L DNB LC0's because it is also based on ex-core detector input.
The specific uncertainty components which reflect the differences in circuitry between the Safety and Control Channel processors is explicitly accounted for in the evaluation of the instrument processing uncertainty and the stochastic simulation procedure.
The processing uncertainty on I p is given in Table Al-1.
A-3
TABLE Al-1 Uncertainty [
] Components for the Evaluation of the LC0 Related Peripheral Shape Index(I)
I Ko 95/95 X(f)(2)
{
j ASIU 1.
Separability Uncertainty-Calibration uncertainty (")
2.
Shape annealing uncertainty (")
3.
4.
Processing uncertainty (")
LHR (ASIU)
DNB (PSIA)
Notes on Table Al-1:
I (1) All components of the peripheral shape index have been tested'for normality [
]
(2) f - degrees of freedom (3) [
]
(4) [
]
(5) 2 Sigma values A-4'
A.4 References for Appendix A Al-1 CEN-123(F)-P Statistical Combination of U. certainties Part 1, December 1979 Al-2 R. E. Uhrig to V. Stello, Letter " Application for Cycle 3 License",
L-79-49, February 22, 1979 Al-3
" INCA, Method of Analyzing In-Core Detector Data in Power Reactors,"
CENPD-145-P, April 1975 Al-4
" Evaluation of Uncertainty in the Nuclear Fonn Factor Measured by Self-Powered Fixed In-Core Detector Systems," CENPD-153, August 1974.
1 I
=
Q A-5
f APPENDIX B Sumrrary of Previous fiethods for Combininq Uncertainties
/
l a
i
~
APPENDIX B The methods previously used for the application of uncertainties to the LC0's are presented in Reference B-1 and are summarized in this appendix.
~
B.1 LHR LC0 The LCO limit on linear heat rate is represented by the ordered pairs (Pfdl, I ).
A lower bound is drawn under this " flyspeck" data such that p
all the core power distributions analyzed are bounded.
Using the previous methodology this lower bound was reduced by the applicable uncertainties and allowances to generate the LHR LC0 as follows:
(B-1)
(B-2) where:
E 3
(B-3)
LC W
LCO limit on linear heat rate
=
,ax Wavg Core average linear heat rate
=
Planar radial peaking factor F,
=
F Core average axial power distribution peaking factor
=
7 F
g Fuel densification-dependent power peaking augmentation factor
=
T Azimuthal tilt allowance
=
AZ PU Uncertainty in predicting local power at the fuel design limit
=
BMU Power measurement uncertainty
=
SAU Shape annealing factor uncertainty
=
RSU Shape index separability uncertainty
=
ACU Axial shape index calibration uncertainty
=
APU Processing uncertainty
=
B-1
B.2 bHB LC0 The fuel design limit for the DNB LC0 is represented by the ordered pairs (Pfdn* I ).
A lower. bound-is-drawn under the " flyspeck" data such that all p
the analyzed core power distributions are bounded.
Using th'e prsvious methodology this lower bound was reduced by applicable uncertainties and allowances as follows:
(B-4)
(B-5) where:
/
Pfdn - Power to fuel design limit on DNB SC - approved partial credit for conservatism in uncertainty application i
BMU - Power measurement uncertainty bthercomponentsoftheDNBLC0werethenrepresentedbyequationsas follows:I UI-II (B-6)
(B-7. )
(B-8) i PMU - Pressure measurement uncertainty TMU - Tenperature measurement uncertainty FMV - Flow measurement uncertainty B-2
B.3 References for Appendix B B1-1 CENPD-199-P, "C-E Setpoint Methodology," April, 1976.
a 4
1 l
B-3 l
m
}
l l
g APPENDIX C TREATf4EtlT OF UtlCERTAIflTIES Ifl TRAflSIEllT AtlALYSES F
(
6 9
APPENDIX C TRANSIENT ANALYSIS C.1 Objective of Analysis As stated in Section 3.1, the DNB and LHR LC0's are based on the following:
(forDNB)andPfd(for LHR) for the reload core.
1.
The p
f 2.
Statistically combined process variable uncertainties.
3.
The DNB and LHR. Required Overpower Margin (ROPM) for the limiting DBE.
The methods used to combine uncertainties were discussed previously.
The objectives of this appendix are:
l 1.
To evaluate the impact of st&tistically combining uncertainties on the selection of initial conditions used in the transient analysis of DBE's.
2.
To determine the magnitude of the variation in R0PM att> ibutable to the uncertainties.
C.2 General Strategy This section of the appendix provides the basis for analyzing the Loss of Coolant l
Flow (4 Pump LOF) and Single Full Length CEA drop (CEA drop) events to determine the variation of the R0PM due to statistically combining uncertainties.
The Design Basis Events (DBEs) applicable to Florida Power and Light's St.Lucie Unit I are presented in Table C-1.
This table also lists the RPS trips which intervene to assure that acceptable limits
- are not exceeded. The table alse identifies which of these events has the potential of yielding the maximum R0 Pit used to generate DNB or LHR LC0's, or for setting the pressure bias input used to establish the TM/LP LSSS.
- The term " acceptable limits" is used in this appendix to include' limits on DNBR, kw/ft, and dose rates, etc.
C -1
i This table shows that nest of these events can be classified in the following manner:
1.
The events where action of the Thermal Margin / Low Pressure (TM/LP) trip and/or the Local Power Density (LPD) trip is necessary to prevent exceeding acceptable limits.
2.
The events where action of RPS trips and/or sufficient initial steady state margin is necessary to prevent exceeding acceptable limits.
These two categories of events are further discussed below.
C.2.1.
Events Where Action of TM/LP and LPD Trips is Necessary to Prevent Exceeding Acceptable Limits The TM/LP trip limits are calculated assuming a conservative pressure bias factor.
This bias factor accounts for the margin degradation due to processing, equipment I
and RTD response time delays. By accounting for these effects in a conservative manner, the TM/LP trip will be actuated when necessary to ensure that the DNBR limit is not exceeded.
As stated in Reference C-1, the maximum pressure bias factor is obtained either for the RCS Depressurization event or the CEA Withdrawal (CEAW) ovent.
However, the CEAW event r.ow has been classified (in Reference C-2) as not requiring the TM/LP trip.
Thus, this event is no longer analyzed to determine the pressure bias factor.
It is analyzed to determine R0PM as described in Reference C-2.
The pressure bias factor calculated for the RCS Depressurization event (which is the most rapid d epressurization event where mitigation by the TM/LP trip is necessary) is used to generate the Tit /LP trip limits.
The pressure bias term for the RCS Depressurization event, calculated using the methods arm procedures given in Reference C-1, is the maximum pressure bias term for the entire operating range of system parameters allowed by the Technical Speci-fication LCO.
Since the methods and the initial conditions used in this analysis are selected in the same manner as described in Reference C-1, there is no need to perform a C-2
sensitivity study on the calculated value of the pressure bias term.
That is, the method of combining uncertainties (either statistical or deterministic) does not affect the way in which the TM/LP trip is used for protection for DBE's-where actuation of the TM/LP trip is required.
The events listed in Table C-1 where action of the LPD trip is necessary to prevent
/
the kw/f t limit from being exceeded do not provide any bias term for input to the LPD trip limits.
These limits already include a three percent power ' bias to account for any transient variations in the measured power.
Since none of the DBE's requiring the LPD trip result in a three percent margin degradation from the time of LPD trip signal to the time of maximum kw/ft, there is no need to input an additional bias for the LPD trip based on transient analysis.
Therefore, the method of combining uncertainties has no impact on the method of analysis or the input data selected for transients requiring the actuation of tne LPD trip to ensure kw/f t SAFDL limit is not exceeded.
C.2.2 Events for Which Intervention of RPS T' ips and/or Sufficient Initial Steady State Thermal Marcin "aintained by LC0 is Necessary to Prevent Exceeding Acceptable Limits.
DBE's listed in this category are not solely protected by the TM/LP and LPD trips because some of the parameters (such as core mass flow rate or rad 41 peakina factors) that are important in some DBE's are not directly monitored by the TM/LP and LPD trips.
For these DBE's, the mitigating effects of RPS trips and/or sufficient iritial steady state margin maintained by operating within the LCO is necessary to ensure that acceptable limits are not exceeded.
The DBE's in this category can be further grouped according to a single key parameter change which has the greatest impact on the margin degradation. A grouping of DBE's in this manner is presented in Table C-2.
To detarmine the sensitivity of R0PM to the magnitude of uncertainties li.sted in Table C-3 during an event characterized mainly by a decrease in the core mass flow rate, the 4 Pump LOF event was analyzed.
This event was analyzed because it pro-vides limiting input (i.e., R0PM) to establish the DNB LCO. The variation of R0PM due to the magnitude of uncertainties observed for the 4 Pump LOF event bounds that for all events characterized by decreases in the core mass flow rate, since all of these events are characterized by the same physical effects.
C-3
The CI A drop event was analyzed to determine the variation of R0PM due to the method of calculating uncertainties for events characteriled by increases in integrated radial and planar peaking factors (F, F y).
This event was analyzed because a p
dropped CEA results in higher F and F increases than the Asymmetric Steam 7
xy Generator events.
Thus, the R0PM for the CEA drop is higher than for the Asymmetric Steam Generator event and is used to establish the LCO.
The CEA Withdrawal event was not analyzed because the R0PM for this event is approximately two percent lower than the CEA drop event.
In addition (based on the results showing insensitivity of R0PM to the magnitude of uncertainties for the 4 Pump LOF and CEA drop events) it can be stated that this event will not be-come limiting from the standpoint of establishing LCO's due to variations.in the R0PM attributable to the process variable uncertainties considered in the analyses.
_C.2.3 Impact of Statistically Combining Uncertainties on DBEs with Other Limits.
Tr.e statistical combination of uncertainties are only used to establish DNB and LHR LCO's and LSSS.
Therefore, it impacts only the DNB-and LHR-related LC0's and LSSS.
Statistically combining uncertainties does not impact events with other limits (such as deposited energy, time to lose Technical Specification allowed shutdown margin, etc.).
Therefore, events with other limits will be analyzed using the same methods and selecting the initial conditions in the same way as previously reported in the FSAR (Reference C-3) or as updated by approved reload license amendments.
O C -4
G.3 Analys<!s Performed for Evaluation of R0PM for the Limiting DBE's C.3.1 Loss of Coolant Flow Event (4 Pump LOF)
C. 3.1.1 Description of Transient The key input parameters for the 4 pump LOF event are determined from the description of the transient given below.
The 4 pump LOF event is assumed to be initiated by the simultaneous loss of AC power to all four reactor coolant pumps.
After the loss of power, the flow starts coasting down rapidly.
In a very short time (about 1.0 second) the low flow trip setpoint is reached.
After a delay for processing the trip signal (s0.5 second) and decay of the magnetic flux for the holding coils (s0.5 second), the CEAs start dropping into the core. After the scram rods reach about 20% insertion for an initially top peaked axial shape (or 55% insertion for a bottom peaked shape),
the CEAs have inserted sufficient negative reactivity to drop the core heat flux below that required to turn around the transient DNBR. The transient minimum DNBR occurs when the rate of heat flux decay (after scram) equals the rate of flow decrease.
The minimum DNBR occurs within 3.0 to 4.0 seconds of the initiation of this event.
Since the minimum DNBR is reached within the first 4.0 seconds, the power distri-butions and the peak linear heat generation rate have not had time to change.
The core inlet and fuel temperatures will not change appreciably, since the loop cycle time (s10.0 seconds) and the fuel time cons. tant (s6.0 seconds) are larger than the time required to terminate the transient DNBR.
Thus, the margin degra-dation for this event is determined by:
1.
The core flow coastdown 2.
The signal processing time delay 3.
The holding coil time delay 4.
The low flow trip setpoint 5.
The available scram worth 6.
The CEA reactivity versus insertioa characteristics.
C.3.1.2 Criteria of Analysis This event is classified as an A00 and hence is analyzed relative to the C_,
following criteria:
1 1.
Minimum Transient DNBR > DNBR limit based on CE-1 correlation.
5080 F 280 x 2.
Centerline Temperature Melt 5 0
50 0 0 t 0
)
Notes: 1) CE-1 DNBR shall have a minimum allowable limit corresponding l
to a 95% probability at a 95% confidence level that DNB will not occur on the limiting rod.
In this study, a DNBR limit of 1.23 was used (See Ref. C 4 forjustification.)
- 2) The CTM SAFDL is a criteria for this event, but this SAFDL is never exceeded since there is no change in PLHGR during this event.
C. 3.1. 3 Input Parameters and Initial Conditions The purpose of this study is to evaluate how much predicted margin degradations vary because of the way the uncertainties of initial conditions are combined.
An analysis parametric in 1.
the initial coolant temperature 2.
initial RCS pressure 3.
initial core mass flow rate 4.
initial axial shape index 5.
integrated radial peaking factor, and 6.
initial core power was performed to determine the sensitivity of R0PM to these parameters.
Other parameters were assumed to be at their limiting values to maximize the calculated margin degradation.
The input parameters used in the analysis of the 4 Pumo LOF event are presented in Table C -4.
A brief justification of values selected is
~
given below.
The key parameters for the loss of coolant flow event were identified earlier as the flow coastdown, the RPS delay times, the low flow analysis trip setpoint and the scram reactivity versus insertion characteristics.
The flow coastdown assumed in the analysis is presented in FigureC -1.
The coast-down was calculated assuming that the coastdown assist feature is inoperative. This produces the most rapid coastdown, and thus the maximum margin degradation due to the lower absolute flow a't time of minimum DNBR.
C4
The low flow analysis trip setpoint assumed is one corresponding to the minimum allowed Technical Specification limit of 93% of initial 4 pump flow.
The RPS trip processing r<iponse delay time and holding coil magnetic flux decay time e.sumed in the analysis are the maximum values allowed by the Technical Specifi-cations.
The use of maximum delay times results in the largest margin degradation since it takes longer for the CEAs to start dropping into the core and thus takes a longer time period to turn around the transient DNBR.
The scram reactivity versus insertion characteristics assumed in the analysis were calculated according to the methods given in Reference C-5.
Other important parameters are the available scram worth and the moderator and fuel temperature reactivity coefficients (MTC and FTC).
The available scram worths were conservatively calculated, including an allowance for the most reactive CEA being stuck in the fully withdrawn position af ter the trip.
A beginning-of-life (BOL) MTC was used in the analysis, since a positive MTC in combination with the slight increase in the coolant temperatures accelerates the rate of increase of both the coolant temperature and heat flux prior to trip.
Both these effects cause the transient DNBR to decrease at a faster rate.
A BOL FTC is assumed for the same reasons.
C_. 3.1. 4 Method of Analysis for the Four Pump LOF The Nuclear Steam Supply System (NSSS) response to a 4 pump LOF event was simulated using the digital computer code CESEC described in Reference C -6.
The code STRIKIN (Ref. C -7) was used to calculate the time variation in core average and hot channel heat flux during the 4 pump LOF event.
The thermal hydraulic code TORC (Ref.C 8) incorporating a CE-1 correlation and a 1.23 DNBR limit was used to calculate the thermal margin degradation during the event. The C0AST code, described in Reference C -9, was used to calculate the flow coastdown during this event. These codes and methods are L w.a as described in previously approved license submittals (Ref.
C -10) except for the use of a DNBR limit of 1.23 rather than the 1.19 value u, sed previously.
The calculational procedures used in the analysis to-determine the DNB R0PM deoend Jpon the initial axial power distribution.
The methods used to analyze 4 pump LOF are axial shape.index dependent because credit for the heat flux decay is taken only when the initial minimum CE-1 DNBR is located in an axial region of the core where the scram rods have passed the axial node of minimum DNBR before the time of mini-rum DNBR is reached.
C7
~
For axial power distributions characterized by negative shape indices, the STRIKIti-TORC metMd was used.
This method is schematically presented in Figure C -2.
For axial power distribution characterized by positive shape indices, the CESEC-TORC method, presented in Figure C -3.was used.
For a zero shape index both methods are used to calculate the R0PM and the maximum value obtained by these methods is then used to generate LCO's.
The two methods used to analyze this event are discussed below.
C. 3.1.4.1.
STRIKIN-TORC Method 1.
The time-dependent core flow, th.e individual loop flows and steam generator pressure drops are determined by using the code C0AST. C0AST solves the conservation equations for mass flow and momentum.
The general forcing functions for the fluid momentum equations consist of the pump torque values from the manufacturer's four quadrant curves, wherein the torque is related to the pump angular velocity and discharge rate.
2.
Limiting axial power distributions, characterized by shape index,are determined from a large sample ( 2,000) of possible distributions which are calculated as a function of axial shape index, core burnup, and CEA configuration, using the QUIX code (Reference C -11).
The limiting axial power shapes are those distributions that produce the lowest ini-tial steady state power to a DNBR limit of 1.23 at a given axial shape index.
The power at which the limit is reached is predicted by the TORC code.
It should be noted that steps 1 and 2 are independent of the LOF method,(i.e.,
STRIKIN-TORC or CESEC-TORC) used.
3.
The resultant core flows are used as input to CESEC to determine the hot channel mass flow rate and to demonstrate that Reactor Coolant System (RCS) prassure during the transient does not exceed the pressure limit of 2750 psia (110% of design).
4.
The RCS flow coastdown, the hot channel flow coastdown from CESEC, axial power distributions, and corresponding scram curves are input into STRIKIN-II to determine the time dependent hot channel and core average heat flux distributions during the transient.
The use of STRIKIN-II to calculate the absolute core average and hot channel heat flux distributions as a function of time is consistent with the methodology utilized and C-8
approved by the NRC on Calvert Cliffs Unit 1 Cycle 2 (Reference C-10) and Millstone Point Unit 2, Cycle 3 (Reference C.12).
5.
The TORC code is used to determine the time of minimum DNBR.
The inputs to the code are the core mass flow rate as well as the hot channel and core average heat fluxes predicted by STRIKIN-II at times cf interest.
Other parameters input to TORC are the initial RCS pressure, L ---
the-initial-inlet temperature, the initial integrated radial peaking factor and the net uncertainties combined statistically.
6.
The core mass flow rate and the hot channel and core average heat flux profile at the time of minimum DNBR are used in conjunction with the initial values of inlet temperature, integrated radial peaking factor, RCS pressure and the net uncertainties combined statistically to obtain a power at which the fuel design limit on DNBR is reached for the transient conditions.
The power at the time of minimum DNBR is denoted B,
2 7.
A TORC case is also run to determine the rod average power at which the fuel design limit on DNBR is reached for the initial steady state system parameters.
This value of power is designated B.
1 8.
The R0PM is then defined (Ref.C -1) to be:
)
(C-1) i C.3.1.4.2 CESEC-TORC Method 1.
Steps 1 and 2 outlined for the STRIKIN-TORC method are also used for this method to obtain the flow coastdown data and the limiting axial power distributions.
2.
The core coolant flow, as a function of time, along with axial power distribution, initial coolant inlet temperature, initial RCS pressure and the scram reactivity versus insertion associated with the axial power distribution of interest are input to CESEC to obtain the time dependent values of core average heat flux, RCS pressure, coolant inlet temperature and the core mass flow rate.
C-9
3.
A set of TORC cases are run with the time dependent values of core heat flux, temperature, RCS pressure and core mass velocity along with the initial values of integrated radial peaking factor, the axial power distribution and the net uncertainties combined statistically to determine the time of minimum DNBR.
4.
The core mass velocity at the time of minimum DNBR in combination with the initial values of RCS pressure, inlet temperature, axial power distribution, integrated radial. peaking factor, core average heat flux and the net uncertainties ' combined statistically are used to determine the power to the DNB limit.
This power is denoted B2 5.
A TORC case is also run with the initial steady state system parameters, including statistically combined uncertainties, to determine the power to DNB limit.
This power is denoted B1 6.
The R0PM is then,as before, defined to be
[
3 (C.2)
C. 3.1.5 Results The results of the sensitivity analysis perfonned over the range of uncertainties for the variables listed in Table C-3 about the nominal base conditions listed in Table C 4 are presented in Figure C-4.
This figure presents the R0PM obtained for the event inititated from the nominal base conditions and also presents the maximum variation in the R0PM due to the uncertainties as a function of initial axial shape index.
It should be noted that the absolute value of the R0PM is plant and cycle specific; however, the maximum margin variation is not plant and cycle specific.
The maximum variation in the R0PM, shown in FigureC -4 as a function of axial shape index, will be added to the cycle specific R0PM calculated for the nominal base conditions to obtain the maximum R0PM during the event.
This maximum R0PM will be used to establish the DNB LCO.
C 10
The sequence of events during a 4 pump LOF event is presented in Table C-5.
The NSSS response during this event is presented in Figures C -5 to C-8.
C. 3.1. 6 Conservatisms in the Analysis Methods.
I The purpose of this section is to identify the conservatisms that are included in the methods used to calculate the R0PM on DNBR during a 4 pump LOF event.
1.
The magnetic flux decay of the holding coil assumed in the analysis is 0.5 second. A more realistic value based on field test data is 0.4 second.
- 2. The low flow response time assumed in the analysis is 0.50 second, which is conservative by at least 0.1 second based on field measurements.
3.
The CEA drop time to 90". insertion value of 3.1 seconds assumed in the analysis is for slowest CEA.
A more realistic value for the slowest CEA drop time to 90% insertion is 2.90 seconds.
4.
The flow coastdown assumed in the analysis does not take credit for the coastdown assist feature. A more realistic flow coastdown. presented in Figure C -9, would be slower than assumed in the analysis.
5.
The R0PM is calculated without taking credit for the higher value of RCS pressure at the time of minimum DNBR. The higher RCS pressure at the time of minimum DNBR will lower the R0PM for this event.
To quantify the conservatisms outlined above a "best estimate" case was run.
A comparison of the input data used in the safety analysis case described in Section 2.3 with that used in the best estimate case is presented in TableC -6.
The R0PM for the best estimate case is ['
], which is lower by [
.] than that for the transient analysis case. The results of the best estimate case show that due to the slower flow coastdown and the higher low flow trip setpoint assumed in the-best estimate case, the low flow trip is initiated at the same time as in the safety analysis case. However, the faster RPS response time, the faster time to decay the magnetic flux of the holding coil and the faster insertion of the shutdown CEAs turns around the transient DNBR faster relative to ti.a safety analysis case. Due to the slower flow coastdown, the absolute flow at the time of C
11
minimum DNBR is higher than in the transient analysis case.
The sequence of events for the best estimate case is given in Table C -7.
The NSSS response for the best estimate is given in Figures C-10 to C -13.
C.3.2 Single Full Length CEA Drop Event (CEA Drop)
.C. 3. 2.1 Description of Transient The key input parameters for the CEA Drop event are determined from the description of the transient given below.
A CEA Drop event is assumed to occur as a result of:
1.
An inadvertant interruption of power to the CEA holding coil, or 2.
A failure in the latching mechanism when CEAs are being moved.
~
The drop of a CEA into the core reduces the fission power in the vicinity of the dropped CEA and adds negative reactivity on a core-wide basis.
The negative reactivity causes a prompt drop in power and thus the heat flux.
The magnitude of this prompt power decrease depends upon the worth of the dropped CEA.
Since no credit is taken for turbine runback in the analysis, a power mismatch exists between the primary and secondary system.
The power mismatch initially causes the primary I
side to cool down.
The decrease in the fuel and moderator temperatures in conjunction with an assumed highly negative fuel temperature and moderator temperature coefficients adds positive reactivity. The positive reactivity added by the feedbacks compersates for the negative reactivity added by the dropped CEA within approximately 100 seconds.
The initial decrease in the coolant temperatures also causes the pressurizer pressure to decreace (the analysis conservatively assumes that the pressurizer level and pressure control systems are inoperative).
In addition, the dropped CEA also causes an asymmetry in the radial power distribution and the radial power peaks. The radial peaks increase as a result of this distortion and achieve a new," tilted' asymptotic state. At approximately 100 seconds,the power and the core heat flux have returned to their initial values.
The coolant inlet temperature and RCS pressure achieve a new, lower, steady state value.
The DNBR also achieves a new, lower, steady state value.
l C 12
The margin degradation during this event is a result of the following changes in the key variables described above, which are:
1.
Increase in integrated radial peaking factor.
2.
Decrease in RCS inlet coolant temperature.
3.
Decrease in RCS pressure.
C.3.2.2 Criteria of Analysis The criteria of analysis for this event are the same as for the 4-pump LOF event.
C.3.2.3 Input Parameters and Initial Conditions This study evaluates how the uncertainties are applied to select initial con-ditions for the transient analyses in the CEA drop event.
An analysis parametric in l.
the initial inlet temperature, 2.
initial RCS pressure, 3.
initial RCS flow, 4.
initial axial power distribution, 5.
integrated radial peaking factor, and 6.
initial core power was performed to determine the sensitivity of R0PM to these parameters. Other parameters were assumed to be at their limiting values to maximize the calculated margin degradation. The method used for this analysis is schematically presented in Figure C-14.
The input parameters used in the analysis of the CEA drop event are presented in Table C-8.
For compidteness, a brief.iustification of each oarameter assumed in the analysis is given below.
The reactor state parameters of primary importance in determining the margin degradation are: (1) the integrated radial peaking factor for DNBR R0PM, (2) the planar radial peaking factor for LHR R0PM, and (3) the CEA drop worth.
The analysis conservatively assumed the maximum integrated and planar radial. peak changes and the minimum CEA drop worth.
The maximum radial peaking factor change results in the highest R0PM.
Assuming a minimum CEA drop worth is also conservative since it minimizes both the pressure and inlet temperature decreases. (It'should be noted that the analysis assumes an inconsistent set of radial peaking factor changes and CEA drop worth Realistically, a low reactivity worth dropped CEA will not produce the,,iximum radial peaking factor increases.)
C.13
End of life values of Moderator Temperature Coefficient (MTC) and the Fuel Temperature Coefficient (FTC) were assumed in the analysis.
These negative FTC and MTC in conjunction with the decreasing coolant and fuel temperatures insert positive reactivity.
The positive reactivity inserted offsets the negative reactivity inserted initially by the dropped CEA and thus enables the core power to return to its initial value.
The uncertainties on the FTC assumed are given in Table C-8.
All control systems-are-assumed-tc bc in the manual mode. The key control systems for this event are the Pressurizer Pressure Control System (PPCS) and Pressurizer Iovel Control System (PLCS).
The PPCS and PLCS are assumed to be in the manual mode because this allows the primary pressure to drop during the transient and thus minimizes the pressure at time of minimum DNBR.
This results in the largest DNBR margin degradation during the event.
C.3.2.4 Method of Analysis for the CEA Drop Event.
The Nuclear Steam Supply System (NSSS) response to a single full length CEA drop l
event was simulated using the digital computer code CESEC, described in Reference C-6.
The thermal hydraulic design code TORC, described in Reference C-8, used the CE-1 correlation and a DNBR limit of 1.23-to calculate the thermal margin degradation during the transient.
C.3.2.5 Required Overpower Margin for CEA Drop.
C. 3.2. 5.1 Required Overpower Margin on DNBR.
The calculation procedures used in the analysis to determine DNB R0PM are presented in Figure C-14. This procedure consists of the following steps:
l 1.
Determining the pseud'o hot channel powe-distribution both before CEA
_I drop and after CEA drop.
The radial peaking factors are synthesized from the core average axial power distribution and planar radial power distributions.
2.
Simulating the CEA drop event using CESEC to determine the final values of core average heat flux, RCS pressures and inlet temperature.
3.
Running the TORC code to determine the rod average power at which the final design limit on DNBR is reached for the initial steady state parameters, including uncertainties combined statistically.
This value of power is denoted B.
1 C - 14
4 The maximum heat flux, final inlet temperature and RCS pressure, the post drop integrated radial peaking factor, the post drop axial power distribution, the uncertainties combined statisticelly and the final value of the core average mass velocity are input to TORC to determine the power at which the fuel design limit on DNBR is reached for the transient conditions.
This power is der.oted B.
2 5.
The R0PM is then defined as:
(C-3) where P is the initial power level and P is the final power level of the 1
2 core.
C. 3.2. 5.2 Required Overpower Margin on PLHGR (KW/FT).
The R0PM on linear heat rate is calculated by the procedures given in Chapter 8 of Reference C-1 Since the methods used to analyze the PLHGR have not changed and since there is no sensitivity of the R0PM due to statistically combining uncertainties, no analysis is required.
C_. 3. 2. 6 Results.
The results of the sensitivity analyses performed for the CEA drop event is presented in FigureC -13.
This figure presents the R0PM obtained for the event initiated from the nominal base conditions and also presents the maximum variation in the R0PM due to the uncertainties as a function of initial axial shape index.
The maximum variation in the R0PM,shown in Figure C-15 as a function of axial shape index, will be added to the cycle specific R0PM calculated for the nominal base conditions to obtain the maximum R0PM during the event.
This maximum R0PM will be input to establish the DNB LCO.
C-1 5
The sequence of events during a CEA drop event is presented in Table C -9.
The NSSS response during this event is presented in Figures C-16 toc -19.
C.3.2.7 Conservatism in Analytical Methods.
The purpose of this section is to identify the conservatisms that are included in the methods used to calculate the R0PM on DNBR during a CEA drop event.
These conservatisms are qualitatively identified below.
An example case is pre-sented and compared with the safety analysis results of previous sections to quantify the conservatism.
1.
The analysis assumed a bounding value for the integrated radial peaking factor changes which is conservative by 2%.
The analysis also assumed a minimum CEA drop worth, which does not produce the maximum integrated radial peaking factor changes.
The use of con-sistent set of CEA drop worth and the integrated radial peaking factor change will lower the margin degradation.
2.
No credit for the actuation of the pressurizer pressure and level control system is taken in the analysis.
The actuation of the pressurizer pressure and level control system would maintain the RCS pressure at a higher value thereby lowering the margin requirement for this event.
l 3.
The moderator temperature coefficient assumed in the analysis is the most negative value of -2.5 x 10~4e/ F allowed by the Technical Specifications.
A more realistic end-of-life value, includino measurement uncertai nty, is -2.3 x 10-4ao/'F.
To quantify the conservatism outlined above a "best estimate" case ms run.
A comparison of the input data used in the transient analysis case described in Section C.3.2.6 with that used in the best estimate case is presented in Table L-10.
The R0PM for the best estimate case is [
] which is conservatiye by [
]
with respect to the transient analysis cese.
The sequence of events for the best estimate case is presented in Table C-ll and the NSSS response during this event is given in Figures C.20 to
-23.
C -15
t l
C.6 CONCLUSIONS Based on the results of the sensitivity studies, it can be concluded that:
1.
The R0P!1 is relatively insensitive to the range of uncertainties on the initial conditions.
The maximum R0PM established by the sensitivity study is'used to generate the LC0's.
2.
The use of a constant maximum R0Pf1 at each axial shape index to generate the LCO's eliminates the need to stochastically simulate the R0PM variations in calculating the net aggregrate uncertainty.
3.
The use of the maximum R0PM also ensures with a high degree of confidence
~
that acceptable limits for the DBE's will not be exceeded.
l
.C -17
_C.5 REFEREllCES FOR APPENDIX C C-1 CENPD-199-P, "C-E Setpoint liethodology, April,1976.
C-2 Cell-126 (F)-P, " Method of Analyzing Sequential Control Element Assembly Group Uithdrawal Event for Analog Prote~cted System", November,1979.
j C-3 St. Lucie Unit I FSAR Docket No. 50-335.
C-4 CEN-123(F)-P, " Statistical Combination of Uncertainties Part 2 - Combination of System Parameters Uncertainties in Thermal Margin Analyses for Florida Power & Light St. Lucie Unit I", January,1980.
C-5 Cell-122 (F), " FIESTA A One Dimensional, Two Group Space-Time Kinetics Code for Calculating PWR Scram Reactivities", November 1979.
C-6 CENPD-107, "CESEC TOPICAL REPORT, July,1974.
C-7 CENPD-135 (P), "STRIKIN-II, A Cylindrical Geometry Fuel Rod Heat Transfer Progran", August, 1974, i
C-8 CENPD-161-P, " TORC Code, A Computer Code for Determining the Thermal liargin of a Reactor Core", July,1975.
C-9 CEllPD-98, "C0AST Code Description", flay,1973.
C-10 Letter, D. L. Zieman (NRC) to A. E. Lundvall, Jr. (BG&E) dated March 14, 1977, License Amendment 21 and SER for Cycle 2 Operation of Calvert Cliffs Unit 1.
Docket flo. 50-317 C-l l System 80 PSAR, CESSAR, Vol. 1, Appendix 4A, Amendment No. 3, June, 1974.
C-12 Letter, R. W. Reid (NRC) to W. G. Counsil (flNEC) dated May 12, 1979, License Amendment 52 and SER for Cycle 3 Operation of Millstone Point Unit 2.
Docket No. 50-336
~
C -18
TABLE C -1 DESIGN BASIS EVENTS AND RPS TRIP PROTECTION l
LIMITING INPUT TO ESTABLISH SETP0INTS L
DBE RPS TRIP LC0 LSSS CEA Withdrawal High Power and No No Variable High Power Boron Dilution TM/LP and/or LPD No No Loss of Load TM/LP and/or LPD No No Excess Load TM/LP and/or LPD No No Loss of Feedwater TM/LP and/or LPD No No Excess Feedwater TM/LP and/or LPD No No RCS Depressurization TM/LP and/or LPD No Yes Loss of Coolant Flow Low Flow Yes No Loss of AC Power Low Flow No No CEA Drop None Yes No Asymmetric Steam Generator 6P Across Steam No No Transients Generator (i nput to No No TM/LP)
CEA Ejection High Power or Variable No No High Power Seized Puc.p Rotor Low Flow No No Steam Line Rupture Low Steam Generator Level No No or Low Steam Pressure Steam Generator Tube TM/LP and/or LPD No No Rupture 0
C-19
TABLE C-2 DESIGN BASIS EVENTS AND IMPORTANT PARAMETER CHANGES
~
~~
Parameter Changes Most Important to Design Basis Event, Margin Degradation Loss of Forced Primary Coolant Flow Decrease in Core Mass Flow Rate Loss of Non-Emergency AC Decrease in Core Mass Flow Rate Seized Rotor Decrease in Core Mass Flow Rate CEA Drop Increase in Integrated Radial and Planar Peaking Factors Asymmetric Steam Generator Transients Increase in Integrated Radial and Planar Peaking Factor 1
CEA Withdrawal Increases in Core Power and Core Coolant i
Inlet Temperature I
m 9
C -20
TABLE C-3 UNCERTAINTIES Uncertainties Values 1.
Uncertainty in integrated radial pin power (Fr) i.6%
2.
Uncertainty in local core power density (F )
1,7%
q 3.
Potser measurement uncertainty
+2%
4.
Shape Index uncertainty 5.
Flow measurement uncertainty R3%
6.
Pressure measurement uncertainty 1,22 psia 7.
Temperature measurement uncertainty 12 F i
- R0PM is an input value used to generate the LC0's. The ROPM values are l
generated parametrically in axial shape index.
The uncertainty associated with any given value of axial shape index is accounted for explicitly in 4
th'e analyses which determine the LCO's and does not have to be accounted for in the transient analysis.
C.21
TABLE C4 KEY INPUT PARAliETERS USED IN THE LOSS OF COOLANT FLOW EVENT Parameter Units Values Initial Power Level
% of 2710 MWt 100.0*
Initial In.let Temperature F
548*
6 2
Initial Core ikss Velocity X10 lbm/hr-ft 2.617*
Initial RCS Pre;sure psia 2225*
Integrated Radii 1 Peaking Factor -
-ARb 1.65*
- Lead Bank Inserted 1.70 Initial Axial Power Distributions Low Flow Analysis Trip Setpoint
% of initial flow 93.0 Flow Coastdown Fraction of Initial See Figure C -l flow vs. time Trip Delay Time sec 050 Holding Coil Delay Time sec 0.5 CEA Drnp Tine to 90% Insertion sec 3.1
~4 11oderator Tem 2rature Coefficient X10 Ao/ F
+0.5 Fuel Tenperature Coefficient
-15.0 Uncertainty CEA Scram Ilorth
%Ao
-5.3 Does not include uncertainty.
The uncertainties for these parameters are given in Table C-3.
The R0Pl1 is calculated as a function r,f axial shape index characterized by various axial power distributinns.
C -22
TABLE C -5 SEQUENCE OF EVENTS LOSS OF COOLANT FLOW EVENT Time Event Value 0.0 Loss of Power all Four Reactor Coolant Pumps 1.0 Low Flow Trip 93% of initial flow 1.5 Trip Breakers Open 2.0 CEAs Begin to drop into core 4.9 Maximun RCS Pressure, psia 2280 N I 4
4 C -23
TABLE C-6 COMPARIS0N OF KEY INPUT PARAffETER I!SFD Trl 9AFFTY AHal Y${$
_AND BEST ESTIMATE CASFS FOR 4 pilt4p LOF EVFflT Safety Analysis Best Estimate Paraneter Units Case Case Initial Power Level
% of 2710 MWt 100.0 100.0 Initial Inlet Temperature F
548.0 8
54,0 6
Initial Core flass Flow Rate X10 lbm/hr-ft 2.617 2.617 Initial RCS Pressure psia 2225 2225 Integrated Radial Peaking Factor (AR0) 1.65 1.55 Low Flow Analysis Trip Setpoint
% of initial flow 93.0 95.0 l
Flow Coastdown Fraction of initial Figure C-1 Figure C-9 flow vs. time RPS Tine Delay sec-0.50 0.40 Holding Coil Delay Tine sec 0.5 0.35 CEA Drop Time to 90% Insertion sec 3.10 2.90 CEA Scran Worth
%Ao
-5.3
-5.8 Moderator Tenperature Coefficient X10'4ao/ F
+0.5
-0.075 Fuel Temperature Coefficient
-15.0 0.0 Uncertainty Initial Axi.:. Shape Index asiu 0.0 0.0 t
l
--- C.gi)
I I
I TABLE C -7 SEQllEflCE OF' EVEllTS' ~~
LOSS OF C00LAtlT FL0ll EVErlT (BEST ESTIIIATE)
Time Event Value 0.0 Loss of Power all Four Reactor Coolant Pumps 1.0 Low Flow Trip 95% of initial flow 1.4 Trip Breakers Open
{
l.75 CEAs Begin to drop into core 1
l Maximur[RCS Pressure, psia 2257 4.0 e
l C-25 l
j t
TABLE C-8 KEY IflPUT PARAMETERS ASSUMED Ifl THE SIrlGLE FULL LErlGTH CEA DROP EVENT Range of Parameter Units Values Initial Core Power Level
% of 2710 Mut 100+
Initial Inlet Temperature F
548+
Initial RCS Pressure psia 2225+
Initial Integrated Radial Peaking Factor 1.693+
F, Lead Bank Inserted 25%
r 6
2 Initial Core tiass Flow Rate X10 ion /hr-ft 2.617+
Initial Axial Shape Index asiu
.3 to +.3+
CEA Drop Worth
%Ao
.08 Integrated Radial Peaking Factor Change 16.0 Moderator Temperature Coefficient X10'4Lp/ F
-2.5 Fuel Temperature Coefficient Multiplier 1.15 Values quoted are without uncertainties.
The uncertainties for these parameters
+
were given in Table C-3.
I l
h 8
C-26
TARLE C-9 SEQUENCE OF EVENTS CEA DROP EVENT Tine (sec)
Event Setpoint or Value 0.0 CEA Begin to Drop into Core 1.0 CEA Reaches Full Inserted Position 100% Inserted 1.2 Core Power Level Reaches itininun 90.4% of Initial and Begins a Return to Power due to Reactivity Feedbacks 1
100 Reactor Coolant Systen Pressure '
2184 Reaches a N w Steady State Value 100 Core Power and Heat Flux Returns 100% of Initial to its Maximum Value 9
\\
C -27
TABLE C -10 C0!! PARIS 0N OF KEY INPUT PARAMETERS ASSURED Ifl THE 5AFETY ANALYSIS AND BEST ESTIMATE CASE EDR CEA DROP EVEf[L l
l Safety Analysis Best Estimate Paraneters Units Values Values Initial Core Power Level
% of 2710 f1Wt 100.0 100.0 Initial Inlet Temperature F
548.0 8
54.0 Initial RCS Pressure psia 2225 2225 Initial Integrated Radial 1.693 1.59 Peaking Factor - Lead Bank Inserted 25%
6 2
Initial Core 11 ass Flow Rate X10 1bn/hr-ft 2.617 2.617 Initial Axial Shape Index asiu
.08
.08 CEA Drop Worth
.08
.13 Integrated Radial Peaking 16.0 14.0 Factor Change Moderator Tenperature Coefficient X10-4 ao / F
-2.5
-2.3 Fuel Tenperature Coefficient 1.15 1.0 f1ultiplier l
l e
9 C -28
TAllLE C-11 SEQUENCE OF EVENTS CEA DROP EVENT (BEST ESTIMATE)
T,ine (sec)
Event Sctpoint or Value O.0 CEA Begin to Drop into Core l
l 1.0 CEA Reaches Full Inserted Position 100% Inserted 1.2 Core Power Level Reaches itininun 83.5% of Initial and Begins a Return to Power due to Reactivity Feedbacks 40.6 Reactor Coolant Systen Pressure Zi/2 Reaches a liininum Value 103 Core Power Returns to its itaxinum Value 99.5% of Ini.tial I
103 Core Heat Returns to its itaximum Value 99.5% of Initial 9
C -29
i 1.0 4-PUMP COASTDOWN
- 0. 8.-
a S
b 0.6 CY.
t L.,
3:od u, 0. 4 m
O O
l l
0.2 i
0 I
I O
- 4. 0
- 8. 0 12.0 16.0 20.0 TIME, SECONDS si. ucie LOSS OF COOLANT FLOW EVENT figure
""'d,[i n,'slsu "
CORE FLOW FRACTION vs TIM C-1 C -30
MFC, F TC FLOW CUASTDOWN SCR AM WO RT H ---*I CESEC TO DETERMINE CEA MOTION CHARACTERISTICS, SCRAM REACTIVITY M (t)
INITIAL AXPD, TIN, PRESSURE, MASS FLOW R ATE (M (t))
g RESPONSE TIME -+
LOW FLOW ANALYSIS TRIP SETPOINT AWD HOLDING C0ll DELAY M (t) f40TE:
2 TIME
-~
STRIKIN TO.9IMULATE M2(t) = Hot Channel Mass Flow Rate 4 PUMP t_OF TO OBTAIN
+-
HF (t) AND HF (t)
HF1(t)= Core Average Heat Flux j
2 HF (t)= Hot Channel Heat Flux 2
HFg (t) HF2 III M (t) 3 1P t
INITIAL PRESSURE TORC TO DETERMINE INITIAL TIN TIME OF MINIMUM ONBR INITIAL AXPD
)
^
INITIAL F R
- UNCERTAINTY HF AND Mj 2
AT TIME OF MINIMUM DNBR TORC TO DETERMINE
+-
POWER TO DNB SAFDL FOR TORC TO DETERMINE TRANSIENT CONDITION.
+ -
DENOTE THIS POWER TRANSIENT CONDITIONS.
DENOTE THIS POWER 82 B j o
B 2 g
8
~
j INITIAL M AND HEAT Fluk PROCEDURES FOR LOSS OF COOLANT FLOW EVENT (STRIKIN TORC METHOD)
C-2 N
ea
l LOW FLOW ANALYSIS TRIP SETPOINT *
~MTC, FTC SIMULATE LOSS OF
-SCRAM WORTH FLOW COASTDOWN
- COOLANT FLOW l
eTIN, PRESSURE, WITH CESEC
~ MASS FLOW RATE I
SCRAM REACTIVITY %
~ INITIAL HEAT FLUX LOW FLOW TRIP RESPONSE TIME AND HOLDING col'L *
% INITIAL AXPD DELAY TIME TIME DEPENDENT VALUES OFTIN, PRESSUR E, M ASS FLOW RATE, AND HEAT FLUX u
TORC TO DETERMINE TIME OF MINIMUM DNBR e
UNCERTAINTY l
TIME OF MINIMUM DNBR AND MASS FLOW RATE AT TIME OF MINIMUM DNBR l
9 1
e INITI AL T TORC TO DETERMINE IN TORC TO DETERMINE POWER TO DNB LIM!T INITIAL PRESSURE POWER TO DNB LIMIT WITH MASS FLOW RATE
. WITH INITI AL AT TIME OF MINIMUM CONDITIONS. DENOTE DNDR. DENOTE THIS INITIAL F S
R THIS POWER AS B 3 R 2 UNCERTAINTY
~
INITIAL HEAT FLUX B2 n
l Bj e
St. Lucie PROCEDURES USED TO DETERMINE REQUIRED OVERPOWER f!OCl'8 Nuclear Pnwer Station MARGIN DURING LOSS OF COOLANT FLOW EVENT Unit No.1 (CESEC TORC METHOD)
C-3
______hW2- -
125 i
i I
I c:
w 3
124 O
c.
_J 5
123 t-E L'
122 O
at N
121 z
9 z
5 120 c-<2 119 g
w
$O c.
118 t
c-w s
>0 117 w
E 116 w
C
~
115 I
I i
i 0.3 0.2 0.1 0
+ 0.1
+0.2
+0.3 AXlAL SHAPE INDEX, ASIU l
l St. Lucie LOSS OF COOLANT FLOW EVENT Figure Nuclear Power Station REQUIRED OVERPOWER MARGIN (DNB) AT 100% POWER' Unit No.1 g _4 vs AXIAL SHAPE INDEX C-33
110 i
i i
i i
i i
100 /
90 I
g
- E 1
g 80 l
~
l E5 70 e
ci 60 J
w3 l
2 50 u
8 40 30 20 10 0
i i
i i
0 2
4 6
8
'10 12 14 16 18 20 TIME, SECONDS LOSS OF FORCED COOLANT FLOW EVENT figure si. i.ucie
"""' l,"d' s ia ti "
CORE POWER vs TIME C -5 C -34
( -
I l
l 110 100 I
E:s 90 8S 80 E5 70 x
Dd 60 s
50 u
8 40 30 20 10 t
0 0
2 4
6 8
10 12 14 16 18 20 TIME, SECONDS si. Lucie LOSS OF FORCED COOLANT FLOW EVENT Figure
""";,""[stati "
CORE HEAT FLUX vs TIME C-6 C-35
620 i
i i
i O
w 610
~
u3 h,600 S 590 T
OUTLET W
s 580 W
m
$,570 g
T
< 560 AVERAGE 8
u 550 xg T
o 540 6
INLET 530 i
i i
i 0
2 4
6 g
10 12 14 16 18 20 TIME, SECONDS
\\
\\
st.tucie LOSS OF FORCED COOLANT FLOW EVENT Figure
""*d,[;$"n"l.I "
RCS TEMPERATURES vs TIME r,-7 C 36
2300 i
2250 g 2200 E
s' O
$ 2150 u
a_
Wu
- 2100 2050 2000 0
2 4
6 8
10 12 14 16 18 20 TIME, SECONDS LOSS OF FC.\\CED COOLANT FLOW EVENT Fiqure si. Lucie nucica, power siation RCS PRESSURE vs TIME C-8 Unit No.1 C -37
1.0 I
i 0.9 3
3 0.8 u.
J t
J 2
0.7 u.
O o
0.G
-U<c:
0.5 u.
2' o
0.4 O
0.3 U
n 3
0.2 u.
0.1 0
i i
i i
0 4
8 12 16 20 TIME, SECONDS 1
St. Lucie Figure Nuclear Power Station LOSS OF COOLANT FLOW EVEN: FCST ESTIMATE)
Unit No.1 4 PUMP FLOW COASTD'.wM vs TIME C-O C 19
110 i
i e
i 100 90 80
$e e
70 R
o 60 at Cd W
50 0
c.
40 o
u 30 20 10 0
I I
0 4
8 12 16 20 TIME, SECONDS l
St. Lucie Figure Nucles ov er Station LOSS OF COOLANT FLOW EVENT lBEST ESTIMATE)
CORE POWER vs TIME f,-]O C-?"
110 i
i e
i 100 90 3:2 g
80 A
u.
C 70 et x'
3 GO u.
>-<w 50 I
wo j
40 w
w 30 c:
ou 20' l
l 10 l.
0 O
4 8
12 16 20 TIME, SECONDS St. Lucie Figure LOSS OF COOLANT FLOW EVENT (BEST ESTIMATE)
Nuclear Power Stat. ion Unit No.1 CORE AVERAGE HEAT FLUX vs TIME
(-))
cc4V51
630 i
i I
8 610 o'
ur C
l o
N 590
~
a.
T r
OUT H2<
d 570 8
TAVE o
t x
550 Tm
~
0 4
8 12 16 20 TIME, SECONDS St. Lucie Figure Nuclear Power Station LOSS OF COOLANT FLOW EVENT (BEST ESTIMATE)
Unit No.1 RCS COOLANT TEMPERATURE vs TIME C-12 L mal
2400 2300 m
S, 2200 w
e D
w f
d 2100 e
2000 t,
1900 1
0 4
8 12 16 20 TIME, SECONDS St. Lucie Figure Nuclear Power Station LOSS OF COOLANT FLOW EVENT (BEST ESTIMATE)
Unit No.1 RCS PRESSURE vs TIME C.6 C
'2
1 SYNTHESIZE 3 D AXIAL AND PLANAR RADIAL BEFORE DROP POWER DISTRIBUTIONS PSEUDO HOT CHANNEL 3.D PSEUDO POWER DISTRIBUTIONS HOT CHANNEL
~
POWER AFTER DROP 3 0 PSEUDO HOT CHANNEL POWER 1r
-lNITIAL TI N ---*
CEA DROP WORTH SIMULATE SINGLE
=
I FULL LENGTH CEA
__ INITIAL PRESS._*
I DROP EVENT INITIAL MASS MTC, FTC
=
WITH CESEC
- FLOW RATE
^
INITIAL HEAT __.
FLUX FINAL VALUES OF t
PR ESSURE, T IN MASS FLOW RATE AND HEAT FLUX l
y lf TORC TO DETERMINE POWER TO DNB LIMIT TORC TO DETERMINE FOR TRANSIENT POWER TO DNB LIMIT UNCERTAINTY FOR INITIAL CONDITIONS. DENOTE THIS POWE. 8 CONDITIONS. DENOTE 2
THIS POWER 8 3 B2 St. Lucie PROCEDURES USED TO DETERMINE REQUIRED OVERPOWER I UI'0 9
Nuclear' Power Station MARGIN DURING SINGLE FULL LENGTH CEA DROP EVENT unit No.1 b-1II
120 e
i I
i i
6 c
i
.J 5
119 I
_H 3
~
u.
O at 118 z9 5
0 1
C l
5 d
117 C
w B:
2 mw>0 116 Qw 5a O
we 3
115 I
i l
0.3 0.2 0.1 0
+0.1
+ 0.2
+0.3 AXIAL SilAPE INDEX, ASIU
\\
St. Lucie pI "
Nucicar Power Statioit CEA DROP EVENT 9
REQUIRED OVERPOWER MARGIN (DNB) AT 100% POWER Unit No.1 C.15 vs AXIAL SHAPE INDEX C
110
~
l 100 f 90 g
80
- E 8
70 g
8 60 e
p 50 s:o' 40 Eou 30
)
1 20 10 0
i i
i 0
40 80 120 160 200
~
TIME, SECONDS SINGLE FULL LENGTH CEA DROP EVENT r.gure si. Lucie
""'j,["*n",' sia ti "
C0RE P0WER vs TIME C -16 h MN
110 100 u-90 80 2:
- E g
/0 i
S g
60 x' 50 3
u_
s 40 5
x 30 w
x oo 20 1
10 0
t i
0 40 80 120 160 200 TIME, SECONDS SINGLE FULL LENGTH CEA DROP EVENT Figure si. tucie
""'"'0ll,'"d',8' "
CORE HEAT FLUX vs TIME C-17 i
^"
-_w
610 i
i i
TOUTLET O
vi E 590 R
\\
ts W 580 2
[TAVERAGE m
E 570 5
8 o
e R 560 UU I
550 INLET 540 1
I i
i 0
40 80 120 160 200 TIME, SECONDS SINGLE FULL LENGTH CEA DROP EVENT si. Lucie Figure
"'0,[,"7,'.siati "
RCS TEMPERATURES vs TIME C -18 d
2300 2280 2260 2240 5
m o-2220 N
o 2200 m
0
[ 2180 m
I E 2160 2140 2120 2l00 0
40 80 120 160 200 TIME, SECONDS SINGLE FULL LENGTH CEA DROP EVENT Figure si. Lucie
""'f,[, "d'siatio" RCS PRESSURE vs TIME C-19 M
/W\\
1 l
110 100 90 E 80
- E 8 70
~
Lt.
60 e
5 50 3:on-40 E
ou 30 l
20 10
{
l 0
0 40 80 120 160 200
~I TIME, SECONDS SINGLE FULL LENGTH CEA DROP EVENT Figure si.tucie Nuclear Power Station BEST ESTIMATE unit no.1 CORE POWER vs TIME C-20 t' _ A 6
120 110 j
100 p 90 s
80 2
h70 25 60 g
x'3 50 u.
&5 40 x
$30 u
20 10 0
0 40 80 120 160 200 TIME, SECONDS si. Lucie SINGLE FULL LENGTH CEA DROP EVENT hure Wuclear Power Station BEST ESTIMATE Umt No.1 CORE HEAT FLUX vs TIME C -21 C-5@
600 i
i i
i TOUTLET 590 O
viu E2580
- E T
570 AVERAGE m
}560 l
8 o
"o D 550 TINLET 540 530 i
0 40 80 120 160 200 TIME, SECONDS SINGLE FULL LENGTH CEA DROP EVENT si. tucie Figure Wuclear Power Station BEST ESTIMATE Unit No.1 RCS TEMPERATURES vs TIME C-22 C -51
2300 2280 2260 2240 t
G
'. 2220
~
l E
l 5 2200 lf)
E 2180 m
o" 2160 2140 2120 2100 i
i i
0 40 80 120 160 200 TIME, SECONDS
~
SINGLE FULL LENGTH CEA DROP EVENT si.tocie Figure helear Power Station BEST ESTIMATE RCS PRESSURE vs TIME C-23 UmtNo.1 C-52
.