ML20198B538
| ML20198B538 | |
| Person / Time | |
|---|---|
| Issue date: | 10/30/1985 |
| From: | NRC |
| To: | |
| Shared Package | |
| ML20198B535 | List: |
| References | |
| TAC-03912, TAC-3912, NUDOCS 8511060517 | |
| Download: ML20198B538 (13) | |
Text
,
ENCLOSURE SAFETY EVALUATION OF A COMBUSTION ENGINEERING LICENSING TOPICAL REPORT (TACS 03912)
Report No.: CENPD-199P, REV-IP (Proprietary) j Report
Title:
CE SETPOINT METHODOLOGY Report Date: March 1982 Originating Organization: Combustion Engineering, Inc.
Responsible Branch: Standardization and Special Projects Project Manager:' H. Bernard Reviewed By: Core Performance Branch and Brookhaven National Laboratory Introduction By letter dated April 13, 1982 (Ref. 1), the Contustion Engineering Co. (CE)
! . submitted for review a topical report entitled "CE Setpoint Methodology."
l The report describes the methodology and procedures for generating setpoints
. for CE plants equipped with analog Reactor Protection Systems (RPS). The report has been reviewed by the staff and staff consultants at the Brookhaven National Laboratory (BHL) under a Technical Assistance program contract (FIN
'A-3407). The result of the BNL review is provided in Reference 2. Additional
- information was requested from the applicant on July 19, 1984 (Ref. 3) and it was submitted by CE on August 13, 1985 (Ref. 4). The references on which this report is based have been approved previously. The staff reviewed all of the available information and our evaluation follows.
Evaluation The purpose of the RPS is to assure that the Specified Acceptable Fuel Design Limits (SAFDL) on the peak Linear Heat Rate (LHR) and the Departure from the bicleate Bolling Ratio (DNBR) are not exceeded during certain Design Basis
[Dk on
t Events (DBE). These limits translate into LHR requirements to avo_id centerlino i melt and DNS at a 95-95 probability-confidence level. The report deals with the methodology of implementation of Limiting Safety System Settings (LSSS) and "
corresponding Limiting Conditions for Operation (LCO) in order to satisfy the above requirements. The report is divided into 10 chapters and two appendices.
Chapter 1 provides an introduction and summary to the CE setpoint methodology.
a In particular it describes the DBEs which must be considered in establishing LSSSs and LCOs necessary for core protection. To this end a DBE list is
- j constructed and the available protection methods are determined. The DBEs and the available capabilities of the Reactor Protection System (RPS) are then used ,
to divide the DBEs into three groups depending on the level of the required l
protective action. Then the RPS, the technical specifications and the LCOs and LSSSs are integrated to assure that all design basis safety requirements are l l
satisfied.
Chapter 2 discusses the basic physics data required for the generation of monitoring and protection system limits and for the analysis of the limiting Anticipated Operational Occurrences (A00). The neutron cron sections are derived from the ENDF/B-IV file. Lattice constants are calculated using the CEPAK or DIT program (Refs. 5 and 6). The ROCS, QUlX, PDQ-X and FIESTA programs are then used for the calculation of multigroup multidimensional power distributions and reactor kinetics (Refs. 5, 6, 7 and 8). Core depletion calculations are
' performed using"the PDQ-X and the ROCS codes. The free oscillation mothei1 ology i
is used to generate axial power distributions as a, function of the axial shape" index that are used in the calculation of the kW/ft for the determination of 1 monitoring and protection system setpoints for steady state and transient .
. conditions. This chapter also presents several applications of the methods discussed.
- The ENOF/8-IV file for the neutron cross sections which is.uted for the cal-culation of the lattice constants is the one generally used in the industry.
Likewise,.the diffusion codes are either in general use or have been approved '
j by the NRC staff. The physics design models have been qualified in Appendix A ,' '
by comparison to calculational results of accepted codes or to meas.urements. I Therefore, the generation of the basic physics constants, power distribution and core depletion are acceptable.
Chapter 3 describes the generation of thermal-hydraulic limits which consist of curves of power to the fuel design limit on DNB vs the Axial Shape Index (ASI) and associated thermal margin information. The minimum DNBR is determined such that there is a 95% probability at the 95% confidence level that the limiting
{
fuel rod in the core will not experience DNB. In order to achieve this, the '
following thermal hydraulic limits are established:
(a) The MDN8R calculated using the CE-1 or W-3 correlations must be greater than or equal to the minimum established value, t
(b) At the HDNBR location the quality must be less than 20% for the CE-1 and I
15% for the W-3 correlations, and i i
^
(c) The average core exit coolant temperature must be less than the saturation temperature.
The thermal-hydraulic design methods are incorporated in the TORC and CETOP-D e
codes which are also used to express the available thermal margin in' terms of DNB (Refs. 9 and 10). The TORC code uses a two step calcul'ational process:
first a core wide analysis determines the lateral boundary conditions for the hot fuel assembly and then the hot subchannel analysis follows which allows'the' determination of the DONBR. Finally, this chapter deals with the determination of uncertainties using the statistical and the deterministic methods.
'The limits defined for the thermal-hydraulic design are appropriate for the' proposed, correlations and have been accepted by the staff.I The thermal-hydraulic
. design methods as incorporated in the TORC code have been approved by the NRC, hence, chapter 3 is acceptable.
Chapter 4 discusses the procedures used to synthesize the LSSS along with the i , way in which the LSSS is utilized by the RPS to shut down the reactor before '
the SAFDL on fuel centerline melt is exceeded. The parameters which are monitored are the axial power shape index and core power. Using the planar radial peaking factor, the LC0 is determined in combination with the variable high power t~ rip and the CEA block system. The peripheral axfal shape index is
! determined as a function of the external axial shape index and the shape annealing factor. The power level (auctioneered thermal and neutron power corrected for ASI) is then used to set a trip limit. The limits have been determined as a function of control element assembly configuration which have been selected over a range of configurations permitted by the transient insertion limits. Finally, the report deals with the application of uncertainties in the estimation of the LSSS limits.
In chapter.4 the determination of the limiting safety system settings for centerline melting is done by the proper comparison to the power level as determined by monitored parameters. This is similar to the general practice and is found to be acceptable.
Chapter 5 deals with the Thermal Margin / Low Pressure (TM/LP) LSSS for DNB core protection along with dynamic effects that must be accounted for in the trip
, system. In chapter 5 the specified acceptable fuel design limit DNBR is discussed in terms of the LSSS core power parameters and the LCO parameters of coolant flow and the CEA group position. In the core parameters the inlet
, temperature range is limited by the reactor power and the temperature corresponding to the pressure of the secondary safety valves and the secondary low pressure trip. The reactor pressure range is limited by the high and icw pressurizer trip settings. The DBEs and A00s are accommodated within these limits without violating DNBR.
The TM/LP trip"is available in combination with the following trip and monitoring systems: LCOs on CEA group positions, variable high power trip, 1ccal high power density trip, LCO on reactor coolant flow and the LCO on the total j integrated radial peaking factor. The calculational procedure for the TM/LP trip is described in some detail along with the statistical and the deterministic method for the quantification of uncertainties.
Finally this chapter deals with the dynamic effects which must be accounted for in the TM/LP trip system.
In chapter 5, the limits to avoid MDNBR are set by the core pressure and i temperature-range which are necessary to accon odate the A00s and DBEs. The spectrum of the A00s and the DBEs considered are those usually taken into consideration and, therefore, are acceptable. +
l l
~
C. . 2.1 i'T
"~ * " ~ " ~ ' ~ ~ ~ ~ ' ~ ~ ~ " ^
l Chapter 6 has been deleted from the revised version of the report as not being relevant.
Chapter 7 deals with the analyses of the loss of flow event and the asymmetric steam generator transients which are protected from exceeding the SAFDLs by the RPS and the initial steady state themal margin maintained by the technical specification LCO.
For the Loss-of-Flow (LOF) event the LCO and the LOF trip are set such that the
- DNBR SAFDL is not violated. The LCO is actually a set of operating limits on core inlet temperature, core flow, power, primary coolant pressure, axial shape and peaking factors. Total loss of flow is assumed to be a result of loss of power and is the DBE for the LOF. In the determination of the LCO conservative ass ~umptions are made for the required plant parameters. If the axial power distribution has a negative or positive shape index the STRIKIN-TORC or the CESEC-TORC methods are used, respectively (Refs. 11 and 12).
The asymmetric transient results from a secondary system malfunction which may change the core power distribution and which are not covered by the TM/LP nor the local power density trip. Therefore, this event must be analyzed separately to assure that neither the DNBR ror the power-to-melt (kW/ft) SAFDLs
. are violated. The events which can cause asyametric loading are loss or excess load or feedwater to a steam generator. The possible RPS trips to mitigate the
! . consequences of the asymmetric load are: low steam generator level, TM/LP, low steam generator pressure and the asymmetric steam generator transient protection j trip function. The trip which will be activated at a given time will depend on
~
the initiator and the particular transient conditions. Assuming initial.
conditions at full power, the report discusses four scenarios of asymmetric steam generator transients, i.e., loss or excess of load or feedwater to a steam generator. The required overpower margin on DNBR is calculated for each ,
of the four asymmetric events with conservative assumptions on core power increase, heat flux and peaking factors. The technical specification DNB LCO ,
establishes an operating margin greater than or equal to those calculated for the asymmetric events to assure that sufficient margin exists to prevent violation of the DNBR -SAFDL. Similarly, the required overpower margin for the power-to-melt (kW/ft) is determined for each of the four asymmetric events. The method ,
consists of computing the three dimensional local power changes in the hot
. - . . . - - . . . . . . . . . 7..-.._.-,.. .-- .- _. _. _ _ _
channel (at the hot spot) as a function of time and core power redistribution due to the asymmetric events. The results of the CESEC-TORC calculation are used. The statistical uncertainty determination methods used in the LOF analysis are also applicable here.
In chapter 7 a set of operating parameters are conservatively set to protect MDNBR from the A00 of total LOF. In addition, a separate analysis is described for the determination of the necessary RPS trip to protect against MDNBR and fuel melting due to power maldistribution from asymmetric steam generator loadings. All potential combinations of loss of load or excess . load are examined using approved methods and Appendix B examines asymmetries due to , ,
steam generator malfunctions; therefore, this methodology is acceptable.
Chapter 8 discusses the treatment of DBEs which do not result in (or do not require) RPS trips. The only such event is the single full length Control Element Assembly (CEA) drop transient. The power distribution resulting from a CEA drop has not been considered previously, hence, the required DNBR and power-to-melt (kW/ft) are considered here. To this end, ths CEA drop which produces the maximum DNBR margin degradation is selected on the
- basis of the maximum distortion produced in the hot channel power distribution.
The CESEC code is used to simulate the NSSS transient response. It is assumed that the reactor is at full power at the initiation of the transient. Other conservative as.sumptions are made regarding power distribution, reactivity feed-back, etc. In this manner the required DNB and power-to-melt kW/ft are determined for a CEA drop. The full length CEA drop at full power is the limiting A00. The statistical uncertainty determination methods applied in the LOF .
are also applicable here. . .
Chapter 8 examines the possibility of violating MDNBR or fuel temperature from power maldistribution which may not require a RPS trip. The;most severe such event is a CEA drop. Because the most severe event is used from full power conditions and the hot channel is examined, this analysis is acceptable.
. Chapter 9 describes the LCO for DNBR and fuel centerline melt required to ,
establish the steady state power operation such that the available margins are sufficient to accommodate the design basis A00 and other postulated accidents. The most limiting LCO calculated to accommodate the design basis ;
events in both' categories-then become the technical specification LCO. j i
- - ..... .~ .. . - -- .
O There are two methods for monitoring the peak LHR in order to maintain the initial required overpower margin, namely the incore and the excore neutron detectors. The incore detector monitoring system provides a direct measure of the three dimensional flux and, therefore, of the peak linear heat generation rate to ensure that the required overpower margin is maintained.
The excore neutron detectors are used for monitoring purposes when the incores .
l are not available. Monitoring is performed with the excore detector system by restricting the amount of rod insertion to limit the magnitude of the peaking factors and by maintaining the axisi shape index within the specified band.
The uncertainties for the excore monitoring of the linear heat generation rate can be estimated using either the deterministic or the statistical methods.
The uncertainties for the incore monitoring include, fuel densification, peaking factors, manufacturing tolerances and fuel expansion.
To assure that the DNB SAFDLs are not violated, appropriate LCOs on the axial shape index, the reactor inlet temperature, the reactor system pressure and the reactor flow are calculated. These LCOs assure that the available overpower equals or exceeds the required overpower when all the measurement errors, operational allowances and calculational uncertainties are taken into account.
In chapter 9, the thermal margins are examined to assure that A00s and DBEs can be accommodated. To this end, the incore detectors measure the power distribution and the excore detectors are used to restrict rod insertion and limit peaking. This direct measurement coupled with conservative uncertainty allowance, constitute an acceptable method.
In chapter 10, LCOs are established on the radial peaking factors to ensure that the LSSSs and LCOs on DNB and the peak LHR remain conservative and valid if the measured radial peaking factors exceed the assumptions incorporated into these limits. The enthalpy rise in the core for the hottest coolant' channel is related to the integrated radial peaking factor. Monitoring of the radial peaking factor is an indirect way of confirming that the radial peaks assumed are also adequate. When they exceed the Technical Specifications, the allowed power level is proportionately penalized. Such power reductions can be based on either the incore or the excore surveillance methods.
Chapter 10 provides a conservative and prudent means of safeguarding against erroneous settings of LSSS and LCOs by lowering the power level. This is an 4 acceptable technique.
Appendix A provides limited qualification of some of the physics design models used to generate input to the safety and setpoint analyses. However, extensive Comparisons to Calculational results are presented.
Reactivity predictions using fine mesh PDQ-X calculations for startup zero power conditions and from operating reactors were carried out. Comparisons with measured data indicate a 30-40 ppm soluble boron concentration difference.
j The flux shape differences were very small; therefore, the overall agreement is considered good.
CEA reactivity worths were evaluated using the fine mesh PDQ-X and were compared to startup test results. The comparisons were used to set conservative uncertainty bounds. No systematic bias was found in the results.
To estimate the CEA scram reactivity worth, the FIESTA (Ref. 8) and the QUIX (Ref. 6) codes were employed. Reference 8 contains the FIESTA description and verification. To demonstrate the conservative FIESTA results a comparison was made to the same calculation by TWIGL (Ref.13) and to experimental measurements.
4 The FIESTA results are everywhere conservative. -
The Doppler coefficient is represented in the design calculations of ROCS, PDQ or QUIX and accounts for cross section variation as a function of
', temperature and fuel temperature as a function of kW/ft. An empirical -
correlation has been established for the fuel temperature vs reactor power in terms of kW/ft. Use of this relationship in 2D and 3D calculations showed that the Doppler feedback is conservatively estimated.
The estimation of the moderator reactivity coefficient is part of the design analysis. Comparison of calculations to the measured values from Maine Yankee
-4 and Fort Calhoun indicated errors within *15 x 10 4/*F. .
The xenon generated reactivity effect is part of the design analysis. A xenon override test was conducted in which the reactor (Fort Calhoun) was shut
? ?~: 2 r r: r :2r7 7:7 ;T = 77.7=73_ _
L7__ __
. down from 69% equilibrium power and the resulting xenon transient reactivity worth was determined. The results of the measurement were compared to calculations and the reactivity worths were within the bounds of the experimental error.
Power distributions are calculated with the ROCS and DIT codes. Comparisons with measurements showed good agreement (Ref. 14 ). The Shape Annealing Factor (SAF) is defined as the ratio of the peripheral to the external shape index.
Although the SAF can be determined analytically, the preferred method is by i
measurement during startup tests. The standard procedure is to perform an 4
axial oscillation test (with all rods out) which is followed till convergence.
Measurement data are fitted to a linear equatica which allows the determination of the SAF. An extensive discussion of the values and its accuracy can be found in References 15 and 16.
- A peripheral shape index value is determined for each axial distribution. In
.the derivation of setpoints an allowance is made for uncertainties associated
- i with the value of the peripheral shape index. A discussion of these uncertainties is included in References 15 and 16. -
+
Asymmetric power distriLutions are calculated as part of the off-normal operating conditions. The standard design methods are employed for the asymmetric power distribution except that full core geometry is employed. Results derived using the ROCS and DIT computer codes are discussed in Reference 5.
The azimuthal power tilt (which refers to power maldistribution in the radial plane) can affect the margins to the specified acceptable fuel design limits.
An allowance has been made for tilt in the monitoring and the protection system-limits established for excore detector control. ,
Appe'ndix B discusses the asymmetric steam generator transient protection trip function which is designed to protect against exceeding the DNBR and the LHR SAFDLs due to primary loop asymmetries resulting from secondary system i malfunctions. The trip is, in effect, a steam generator differential pressure device. Its implementation is based on existing measuring and sensing devices.
-- 7 z z n _.i r ~ Z --- T ~~~' T U n -~ ' .i ___ Z -
Summary j We have reviewed the CENPD-199P, Rev-1P report and the material submitted in response to a staff request for additional information. In addition, the reviewer profited from discussions with staff members who were familiar with
- CENPD-199P-Rev. IP and the comments made by our consultants at BNL. The report consists of ten chapters and appendices A and B.
l Chapter one provides an overview of the report and a brief consideration of j what must be considered in the establishment of LSSSs and LCOs related l to the protection of kW/ft and DNB. Chapter two describes the physics data necessary for the analysis of A00s and the free oscillations method used to generate axial power distributions as a function of the axial shape index. The
~
third chapter describes the estimation of the thermal hydraulic limits i.e.,
power to the fuel limit on DNB vs the axial shape index while chapter 4 describes the local power density trip limit vs the axial power index. In both chapters, the associated thermal margins and calculational uncertainties are discussed.
l Chapter five deals with the Thermal Margin / Low Pressure LSSS which protects the core against exceeding the SAFDLs on DNB including the dynamic effects which must be accounted for in the trip system. The contents of chapter six have been omitted from this revision as being not relevant for the setpoint methodology.
Chapters seven and eight discuss Technical Specification LCOs for the DBEs when the Thermal Margin / Low Pressure trip set points are not adequate. The calculation of the required LCOs in such cases is covered in chapter 9. In the final chapter, LCOs are established on the peaking factors to ensure that the i
LSSSs and the LCOs on the DNB and the peak LHR remain conservative and valid in the event the measured radial peaking factors exceed the values assumed in the limits. Finally, Appendix A provides qualification of the physics design models and Appendix B discusses the asymmetric steam generator transient protection trip function.
i Conclusions The objective of the report is to assure safe operation of the nuclear steam supply syste's by specifying the methodology for:
10 -
4
~
- , . . , _ . - ~, .-._,.,e m , . - - - - - - - , - , -
(a) Limiting Safety System Settings in terms of parameters directly monitored by the reactor protection system, and (b) Limiting Conditions for Operation on reactor system parameters.
The report has been reviewed to assure that the methodology and procedures used, when applied to CE plants with analog reactor protection systems, will result in compliance with the provisions of 10 CFR Part 50 Appendix A criteria 10 and 13.
The methodology described has sufficient provisions to satisfy these requirements and is based on approved references. We conclude that the proposed setpoint methodology is acceptable and can therefore be referenced by CE in licensing submittals.
M E
e
REFERENCES
- 1. Letter from A. E. Scherer (CE) To J. R. Miller (NRC), dated April 13, 1982.
- 2. Letter from M. Todosow (8NL) to M. Dunenfeld (NRC), dated November 11, 1983.
- 3. Letter from C. O. Thomas (NRC) to A. Ec Scherer (CE), dated July 30, 1984.
- 5. CENPD-226-P., "The ROCS and DIT Computer Codes for' Nuclear Design," Combustion Engineering, December 1981.
i
- 7. WAPD-TM-678, "PDQ-7 Reference Manual", W. R. Cadwell, Bettis Atomic Power Laboratory Westinghouse Electric Corporation, January 1968.
i 8. CEN-122(F), " FIESTA, A One Dimensional, Two Group Space Time Kinetics Code for Calculating PWR Scram Reactivities" by V. Decher, Combustion Engineering, November 1979.
- 9. CENPD-161-P, " TORC Code, A Computer Code for Determining the Thermal i Margin," Combustion Engineering, July 1975. .
- 10. CEN-191(B)-P, "CETOP-D Code Structure and Modeling Methods for Calvert Cliffs 1 and 2," Combustion Engineering, December 1981.
- 11. CENPD-135-P "STRIKIN-II, A Cylindrical Geometry Fuel Rod Heat Transfer Program," Combustion Engineering, August 1974.
- 12'-
- 12. CENPD-107, "CESEC Topical Report," Combustion Engineering, July 1974.
- 13. WAPD-TM-743, "TWIGL - A Program to Solve the Two-Dimensional, Two-Group, Space-Time Neutron Diffusion Equations with Temperature Feedback" by J. B. Yasinsky, et. al. , Bettis Atomic Power Laboratory, Westinghouse Electric Corp., February 1968.
- 14. CENPD-153, Rev. 1-P-A, " Evaluation of Uncertainties in the Nuclear Power Peaking Measu' red by the Self-Powered In-Core Detector System," Combustion Engineering, May 1980.
- 15. CEN-123(F)-P, " Statistical Combination of Uncertainties" Parts 1 and 3, Combustion Engineering, 1980.
- 16. CEN-124(B)-P, " Statistical Combination of Uncertainties" Parts 1 and 3, ,
Combustion Engineering, December 1979.
4 s
N e
o e
~
13 -
, __, _