IR 05000269/1990010
| ML15224A662 | |
| Person / Time | |
|---|---|
| Site: | Oconee  |
| Issue date: | 04/09/1990 |
| From: | Belisle G, Burnett P NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION II) |
| To: | |
| Shared Package | |
| ML15224A661 | List: |
| References | |
| 50-269-90-10, 50-270-90-10, 50-287-90-10, NUDOCS 9004250183 | |
| Download: ML15224A662 (8) | |
Text
- PJR REG~,
UNITED STATES 0 oNUCLEAR REGULATORY COMMISSION
REGION 11 o
101 MARIETTA STREET, ATLANTA, GEORGIA 30323 Report Nos.:
50-269/90-10, 50-270/90-10, and 50-287/90-10 Licensee:
Duke Power Company 422 South Church Street Charlotte, NC 28242 Docket Nos.: 50-269, 50-270, and 50-287 License Nos.:
DPR-38, DPR-47, and DPR-55 Facility Name:
Oconee 1, 2, and 3 Inspection Conducted: March 12 -
16, 1990 Inspector :-a_1_ed Pl--T.'B ie t t Date Signed Approved by:
//
G. A. Belisle, Chief Date Signed Test Programs Section Engineering Branch Division of Reactor Safety SUMMARY Scope This routine, unannounced inspection addressed the areas of review of completed power escalation tests for Unit 3, cycle 12, review of procedures used in thermal power monitoring, and review of the bases and methodology used in flow venturi fouling coefficient determinatio Results As part of the power escalation tests, power distribution comparisons were made, on a fuel assembly-by-assembly basis, between measured and predicted values of relative fuel assembly power. Only comparisons for core locations with operating incore detector strings, using one-eighth core symmetry, were reported and considered in determining the acceptability of the measurement But, power must be measured at all locations to do the power distribution surveillance How well these alternate methods of measuring assembly power work, when there is no symmetric, operating detector string, is of concer The quality of the alternate methods of power determination affects the reli
. ability of the hotspot factor analyses required by Technical Specification Hence, comparisons should be made, with appropriate acceptance criteria, for measured and predicted relative assembly powers for all core location The licensee has agreed to review current practice and the bases for determining acceptable agreement between predicted and measured relative assembly power, regardless of the method of measurement. (Paragraph 2)
FDR ADOCK 0 00 02C FTDC
The current program for determining secondary side flow venturi fouling coeffi cients is based upon detailed and complex analyses of instrument uncertainties and their propagation into an uncertainty to be applied to the secondary side heat balance. That heat balance is used to calibrate the nuclear instruments and to confirm operation within the rated thermal power specified in the unit license The licensee engineers, who were instrumental in developing the program and are currently performing the required procedures and analyses, are fully familiar with the restrictions on data collection for and limitations in application of the fouling coefficient (If those restrictions are not satisfied, the coefficients must be reset to 1.0.).
However, those restrictions and limitations are not well enforced by acceptance criteria in procedures or by other administrative control The licensee has agreed to improve the administrative control (Paragraph 3.b)
Licensee engineers and operations personnel always provided a prompt and knowledgeable response to the technical issues raised by the inspecto No violations or deviations were identifie REPORT DETAILS 1. Persons Contacted Licensee Employees H. B. Barron, Station Manager
- M. Carter, Design Engineer, Site Office T. Curtis, Compliance Manager J. Davis, Technical Services Manager P. Gillespie, Engineer, Reactor Group
- M. Hone, Engineer, Performance
- D. Hubbard, Performance Manager M. Hutchinson, Engineer, General Offices (by telephone only)
- G. Lareau, Reactor Engineering Supervisor E. LeGette, Shift Supervisor, Compliance
- G. Rothenberger, Integrated Scheduling Superintendent Other licensee employees contacted included engineers, technicians, operators, and office personne Other Organizations NRC Resident Inspectors B. B. Desai, Resident Inspector P. H. Skinner, Senior Resident Inspector
- L. D. Wert, Resident Inspector
- Attended exit interview on March 15, 199 Acronyms and initialisms used throughout this report are listed in the last paragrap.
Unit 3, Cycle 12 Power Escalation Tests (72700, 61702, 61705)
Procedure TT/3/A/0811/12, Unit 3, Power Escalation Tests, controlled the activities reviewed her The procedure defined three power testing plateaus:
LPT (5-30% indicated power),
IMPT (40-75% RTP),
and FPT (90-100% RTP).
Tests performed at all plateaus included:
checkout of incore instrumentation, analysis of MDNBR and MLHR, and a heat balance and calibration of PRNI Core symmetry was checked at LP The PIDC test, scheduled for 73% RTP, was actually performed at 58% RTP, because power escalation was stalled by a delay in getting a second feedwater pump on lin At both IMPT and FPT a power distribution comparison was made, on a fuel assembly-by-assembly basis, between measured and predicted values of
relative fuel assembly powe Only comparisons for core locations with, operating incore detector strings, using one-eighth core symmetry, were reported and considered in determining the acceptability of the measure ment But, power must be measured at all locations to do the power distribution surveillances. How well these alternate methods of measuring assembly power work, when there is no symmetric, operating detector string, is of concer The quality of the alternate methods of power determination affects the reliability of the hotspot factor analyses required by T Hence, comparisons should be made, with appropriate acceptance criteria, for measured and predicted relative assembly powers for all core location The licensee has agreed to review current prac tice and the bases for determining acceptable agreement between predicted and measured relative assembly power, regardless of the method of measure men The inspector also reviewed the following tests procedures completed as part of the startup test program:
a. PT/O/A/0105/01, Checkout of Selected Computer Programs, performed during December 12-15, 1989, confirmed that offline computer programs used by the Performance Department in routine surveillance and test activities had been updated for cycle 12 parameters and dat Standard test cases were run with each program to confirm proper performance by the progra b. PT/O/B/0302/06, Review and Control of Incore Instrument Signals, was performed on December 19 and 24, 198 c. PT/O/A/0275/03, Calculation of Reactor Coolant Flow and Delta-T Power Constants, was performed on December 26, 198 d. PT/O/A/0302/04, Backup Incore Detector System Operability Verifica tion, performed January 31, 199 Following discussion of these tests with licensee personnel, the inspector had no further question The Startup Report for Unit 3, Cycle 12, was reviewed by the inspecto The report is a generally accurate description of the tests performed and the results obtained. The inspector identified an error in Enclosure 2.0 in the table of reactivity coefficient The licensee will issue a corrected pag During the observation of Unit 3 startup activities reported in NRC Inspection Report No. 50-287/89-39, it was noted that NI performance was less than optimum; since one IRNI and one PRNI were not operatin Observations in the control room during this inspection confirmed that both IRNIs are now operatin PRNI-9, however, can be calibrated to full power, but spikes too much to. use with the ICS. Currently PRNI-5 is used for input to ICS as well as for protectio No violations or deviations were identifie. Thermal Power Monitoring (61706)
a. Precision Heat Balance At the end of Unit 3 power escalation testing for cycle 12, a preci sion heat balance was performe Data were collected under PT/O/B/0325/11, Data Collection for Secondary to Primary Heat Balance Correction, during the period December 27-29, 198 Evaluation of the data and calculations of the actual heat balances were performed offsite at the corporate offices by the Oconee Systems Performance Section of the Technical Systems Divisio Those analyses were documented in an interoffice letter (and attachments), M. W. Smith to D. W. Hubbard, Subject: Oconee Nuclear Station Unit 3, BOC 12, Precision Calorimetric Test The thrust of the Smith report was the BOC volumetric flows for the RCS, which will be used in all PRIMARY2 thermal power calculations and in the determination of fouling coefficients for the feedwater flow venturi For little extra effort, far more product could be obtained. The secondary side precision calorimetric should have been compared with the results from CTPA on the OAC and POWER, the offline calculation, with the acceptance criterion that the latter two calculations should yield results at least as large as the precision calculation. The licensee is reviewing these consideration b. Feedwater Flow Venturi Fouling Coefficients The licensee has observed that the electrical output of each unit falls off over the first 30 to 60 days of operation in each cycl The falloff is about two percent and remains relatively constant throughout the balance of the cycl Since the feedwater flow venturis are cleaned during each refueling outage, the licensee has concluded that the buildup of a thin film on the nozzles is suffi cient to affect the measurement of dP across the venturi The result is an increased dP at constant flow. That dP is the primary variable in the calculation of feedwater flow, which, in turn, is the major contributor to the secondary side heat balance and power analysi The licensee has attempted to determine the extent of venturi fouling and to use an appropriate fouling coefficient to recover the lost power productio PT/0/A/0205/04, Determination of Feedwater Venturi Fouling Coeffi cient, is based upon simultaneous solutions of the primary and secondary side heat balances with primary side flow fixed, for calculational purposes, at the rate determined by the precision heat balance at the start of the cycl Since the primary flow was obtained at the start of the cycle from the precision heat balance and the flow coefficients for clean secondary side flow venturis, the
later determination of venturi fouling propagates the error or uncertainty in the measurement of thermal power by secondary side heat balanc The inspector asked the licensee to justify the propagation of error in recalibrating the secondary side flow ven turis and to demonstrate that thermal power measurement uncertainties did not exceed those assumed in the FSA The licensee had completed a propagation of error analysis before implementing the fouling coefficient procedur That analysis is contained in Calculation File OSC-3737 (Revision original), "Secon dary Power Uncertainty Analysis," dated September 8, 198 That analysis, which appears to be thorough in its identification of contributors to sensor and signal converter inaccuracies on an instrument-by-instrument basis, concluded that the secondary side precision heat balance on a loop had an uncertainty of 0.718% and that the corresponding primary loop flow uncertainty was 1.452%. The typical secondary side heat balance was calculated to have an uncer tainty of 0.898% without reference to a fouling coefficient. After application of the fouling coefficient derived from the primary side heat balance using BOC flow (the PRIMARY2 calculation), the secondary side heat balance uncertainty increased to 1.79% RT Extended discussions on the content of OSC-3737 were held with licensee engineers at the plant and, subsequent to the inspection, by telephone, culminating with a telephone conference on March 29, 199 Much of the discussions addressed the calculation of total instrument string uncertainties, their translation to measurement uncertainties, and, finally, the propagation of these individual measurement uncer tainties to heat balance uncertainty, as a function of the kind of heat balance performe At the conclusion of the discussions, the inspector accepted the licensee's methodology and calculation The effect of up to six failures of redundant instruments was calcu lated in OSC-3737 to increase the uncertainty, when using a fouling coefficient in the daily calorimetric calculation, to 1.97% RTP, which is very close to the uncertainty limit of 2% RTP assumed in the FSA Consequently, the report recommends resetting the fouling coefficients back to 1.0 if more than two instruments used in the secondary side heat balance are faile In addition to addressing the consequences of instrument failures on the validity of the fouling coefficients, OSC-3737 includes a list of eight other considerations required to calculate a valid fouling coefficient or other limitations, which would require resetting the coefficient to Reevaluation of OSC-3737 was required following the most recent precision heat balance on Unit 3 because of greater fluctuations than expected of some instrument That increased the uncertainty of the correlation. The effect on the number of instruments that may be out of service before resetting the coefficient to 1.0 was not evaluate Although OP/1, 2, and 3/A/1102/04, Operation at Power, have been revised to address two of the nine considerations in OSC-3737, most of these requirements are not well defended in procedures. Further more, there is no guidance extant on when the characteristics of a data set deviate sufficiently from the assumptions of OSC-3737 to require reevaluation using the methodology of OSC-3737. The licensee has agreed to improve the administrative control of these issues to assure that the fouling coefficients are properly maintained or reset to 1.0 when require The inspector witnessed the installation of new fouling coefficients in the OAC for Unit 3 and the checkout of the response of the OAC to the new value The administrative control of OAC parameters ap peared adequate, and the response checkout thorough. The completed procedure satisfied the data collection and analysis requirements of OSC-3737, but those requirements were not present as acceptance criteria in the procedur One good feature was observed in the procedures reviewed, that made use of computer points in surveillance test Those procedures contained enclo sures, which identified the computer points and for each point cross referenced both the IP procedure to calibrate the point and the standing work order to perform the calibratio The enclosures also required confirmation of current calibratio. Exit Interview The inspection scope and findings were summarized on March 15, 1980, with those persons indicated in paragraph 1 above. The inspector described the areas inspected and discussed in detail the inspection finding No dissenting comments were received from the licensee. Proprietary informa tion was reviewed in the course of this inspection, but is not included in this report. In a telephone conference with licensee personnel on March 29, 1990, the inspector completed his review of the venturi fouling coefficient determination and informed the licensee he had accepted their methodology given their commitment to improve administrative control of the proces. Acronyms and Initialisms Used in This Report BOG beginning of cycle CTPA computer program -
core thermal power analysis dP differential pressure FPT full power test plateau FSAR Final Safety Analysis Report ICS integrated control system IMPT intermediate power test plateau IP instrument procedure IRNI intermediate range nuclear instrument LPT low power test plateau
MDNBR minimum departure from nucleate boiling ratio MLHR maximum linear heat rate NI nuclear instrument OAC operator assist computer OP operating procedure OSC Oconee calculational file PIDC power imbalance detector correlation PRNI power range nuclear instrument PT periodic test RCS reactor coolant system RTP rated thermal power TS Technical Specification TT temporary test 0II