ML17290A481

LER 92-037-02:on 920815,reactor Manually Scrammed Due to Core Instability.Operating Strategies & Related Procedures Revised to Preclude Identified Precusor Conditions & Increase Precautions During Operation at Low Power Levels

ML17290A481
Person / Time
Site:	Columbia
Issue date:	06/18/1993
From:	Eades M WASHINGTON PUBLIC POWER SUPPLY SYSTEM
To:
Shared Package
ML17290A482	List: ML17290A481 ML17290A485
References
LER-92-037, LER-92-37, NUDOCS 9306290120
Download: ML17290A481 (18)

Text

LIGENsEE EVE/ REPoRT (LER)

AGILITY NAME (I) DOCKET NUMB R ( ) PAGE (3)

~

Washin ton Nuclear Plant - Unit 2 0 5 0 0 0 3 9 7 I OF 18 ITLE (4)

MANUALREACTOR SCRAM DUE TO CORE INSTABILITY EVENT DATE (5 LER NUMBER 6) REPORT DATE 7 OTHER FACILITIES INVOLVED 8 MONTH DAY YEAR YEAR 'EQUENTIAL E VISION HONTH DAY YEAR FACILITY NAHES CKET NUMBERS(S)

NUMBER UMBER 0 0 0 0 8 1 5 9 2 9 2 0 3 7 0 2 0 6 1 8 9 3 000 PERATING HIS REPORT IS SUBMITTED PURSUANT TO THE REQUIREHENTS OF 10 CFR 5: (Check one or more of the following) (11)

ODE (9)

OKER LEVEL 20.402(b) 0.405(C) 0.73(a)(2)(iv) 77.71(b)

(10) 0.405(a)(1)(i) 50.36(c)(1) 50.73(a)(2)(v) 73.73(c) 0.405(a)(1)(ii) 50.36(c)(2) 50.73(a)(2)(vii) THER (Specify in Abstract 0.405(a)(1)(iii) 50.73(a)(2)(i) 50.73(a)(2)(viii)(A) clew and in Text. NRC 0.405(a)(1)(iv) 50.73(a)(2)(ii) 50.73(a)(2)(viii)(8) orm 366A) 0.405(a)(1)(v) 50.73(a)(2)(iii) 50.73(a)(2)(x)

LICENSEE CONTACT FOR THIS LER (12)

TELEPNONE NUHBER M. G. Eades, Licensing Engineer REA CODE 0 9 7 7 - 4 2 7 7 COHPLETE ONE LINE FOR EACH COHPONENT FAILURE DESCRIBED IN THIS REPORT (13)

CAUSE SYSTEH COHPONENT HANUFACTURER EPORTABLE CAUSE SYSTEH COMPONENT MANUFACTURER EPORTABLE 0 NPRDS TO NPROS SUPPLEHENTAL REPORT EXPECTED (14) EXPECTED SUBHISSION MONTH DAY YEAR ATE (15)

YES (If yes, complete EXPECTED SUBHISSION DATE) NO TRACT (101 At 0301 hours0.00348 days <br />0.0836 hours <br />4.976852e-4 weeks <br />1.145305e-4 months <br /> on August 15, 1992, the reactor was manually scrammed due to indications of core instability. Plant control room operators noted Average Power Range Monitors (APRMs) swinging between 25 and 45% power with numerous Local Power Range Monitor (LPRM) downscale indications.

The Plant was scrammed within 80 seconds after the oscillations were detected.

The root cause for this oscillation event has been determined to be unanticipated interaction of operating conditions and components. The primary reason for the oscillation was the skewed radial'and axial power distribution in the reactor. The Supply System staff and vendor personnel failed to identify the extent of the tendency for this core design to become unstable under certain operating practices. The core consisted of about 26% 9x9 design fuel bundles and 74% 8x8s. The 9x9s have a higher pressure drop I compared with 8x8 fuel. In addition, the 9x9 fuel assemblies were operated at a higher power relative to the 8x8s. These conditions led the core to be susceptible to power oscillations. When the first Recirculation Flow Control Valve (FCV) was closed, in preparation for the Recirculation Pump shift to high speed, thermal-hydraulic instability was initiated and self-limiting power oscillations followed. The event was terminated by a manual reactor scram.

Short-term corrective action was taken to revise operating strategies and related procedures to preclude identified precursor conditions and increase precautions and controls during operation at low power levels.

Long-term design process, operating, and management corrective actions are also being pursued.

9306290i20 930618 PDR ADOCK 05000397 S PDR

LICENSEE EVENT REPOR ER)

TEXT CONTINUATION ACILITY NAHE (1) DOCKET NUHBER (2) LER NUMBER (8) AGE (3) ear umber ev. No.

Washington Nuclear Plant - Unit 2 0 5 0 0 0 3 9 7 37 02 2 F 18 ITLE (4)

MANUAL REACTOR SCRAM DUE TO CORE INSTABILITY Plan C ndi Power Level - 33%

Plant Mode - 1 vn D cri in At 0301 hours0.00348 days <br />0.0836 hours <br />4.976852e-4 weeks <br />1.145305e-4 months <br /> on August 15, 1992, the reactor was manually scrammed due to indications of core instability. Plant control room operators noted Average Power Range Monitors (APRMs) swinging between 25 and 45% power with numerous Local Power Range Monitor (LPRM) downscale indications in all regions of the full core display. Plant operators manually scrammed the reactor as required by Plant procedures and training. An Unusual Event was declared at 0320 hours0.0037 days <br />0.0889 hours <br />5.291005e-4 weeks <br />1.2176e-4 months <br /> in response to the power oscillations and Plant shutdown.

Prior to the event, the unit was reduced in power from 100% to about 5% power starting at 1855 hours0.0215 days <br />0.515 hours <br />0.00307 weeks <br />7.058275e-4 months <br /> on August 13, 1992,. due to high unidentified drywell leakage. A drywell entry was made to locate and isolate the source of the leak. The Reactor Bottom Head Drain Valve Bypass (RWCU-V-103) was found with packing and the valve was back seated to stop the leakage. 'ailed At 1710 hours0.0198 days <br />0.475 hours <br />0.00283 weeks <br />6.50655e-4 months <br /> on August 14, 1992, the operators commenced returning the unit to 100% power with rod pulls. Since this was approximately 22 hours2.546296e-4 days <br />0.00611 hours <br />3.637566e-5 weeks <br />8.371e-6 months <br /> after 100% power operations, Xenon levels were significantly higher than 100% equilibrium Xenon levels as well as equilibrium values for 5-7% power. This allowed the Station Nuclear Engineer (SNE) to establish a rod pattern highly beneficial to Plant maneuvering after the Recirculation Pump Shift to high speed (60 Hz) but which had high radial and axial peaking and. an axial profile peaked in the bottom of the core, The Advanced Nuclear Noise Analysis (ANNA) Monitoring System, WNP-2's stability monitoring system, was not turned on as the Plant was operating outside the region for which Technical Specifications required ANNA to be in operation. The unit attained 34%

power and 30% flow at about 2355 hours0.0273 days <br />0.654 hours <br />0.00389 weeks <br />8.960775e-4 months <br /> on August 14, 1992. The unit was held in this condition until 0258 hours0.00299 days <br />0.0717 hours <br />4.265873e-4 weeks <br />9.8169e-5 months <br /> on August 15, 1992, to support Turbine Bypass Valve testing and Control Rod Drive timing.

At 0258 hours0.00299 days <br />0.0717 hours <br />4.265873e-4 weeks <br />9.8169e-5 months <br />, power level was increased to about 36.5% as required by Plant procedures to provide adequate flow margin above the Recirculation Pump Cavitation Interlock. The control room operators then closed RRC-FCV-60A in preparation for recirculation pump 1A shift to 60 Hz speed. At 0259 hours0.003 days <br />0.0719 hours <br />4.282407e-4 weeks <br />9.85495e-5 months <br />, rod out block alarms and LPRM down scale alarms were received. The operators noted that the APRMs were oscillating between 25% and 45% power. The Shift Manager directed a Manual Scram in accordance with PPM 4.12.4.7 "Unintentional Entry into Region of Potential Core Instability," at 0301 hours0.00348 days <br />0.0836 hours <br />4.976852e-4 weeks <br />1.145305e-4 months <br />. Alt rods fully entered the core upon the scram. The shutdown of the unit to a cold shutdown condition went normally.

LICENSEE EVENT REPOR ER)

TEXT CONTINUATION AGILITY NAME (1) DOCKET NUMBER (2) LER NUMBER (8) AGE (3) ear umber ev. No.

Washington Nuclear Plant - Unit 2 0 5 0 0 0 3 9 7 2 37 02 3 F 18 ITLE (4)

MANUAL REACTOR SCRAM DUE TO CORE INSTABILITY Immedi rr iv A n At 0415 hours0.0048 days <br />0.115 hours <br />6.861772e-4 weeks <br />1.579075e-4 months <br /> on August 15, 1992, preparation began to place the reactor in cold shutdown and the Unusual Event was terminated at 0430 hours0.00498 days <br />0.119 hours <br />7.109788e-4 weeks <br />1.63615e-4 months <br />. Later on August 15, 1992, a root cause team was appointed and an analysis plan was formulated to investigate the cause of the event, assist the NRC Augmented Inspection Team (AIT) and recommend corrective actions. The Supply System root cause team was supplemented by experts from Siemens Nuclear Power, General Electric, and Institute for Nuclear Power Operations (INPO).

F herEv I i n R t ue nd rrectiveA i n A. ~~h~ i

1. This event is being reported per the requirements of 10CFR50.73(a)(2)(iv) as "....any event or condition that resulted in manual or automatic actuation of any Engineered Safety Feature (ESF), including the Reactor Protection System (RPS)." This event is also reportable under 10CFR50.73(a)(2)(ii)(B) as "....a condition that was outside the design basis of the plant"

'nd is reportable as a Technical Specification required shutdown under 10CFR50(a)(2)(i)(A) since the reactor was scrammed to counter the instability, as directed by the action statements 3.2.6 and 3:0.3. Finally, this event is reportable under 10CFR50.73(a)(2)(ii)(A) as an

"....unanalyzed condition that significantly compromised Plant safety." A 50.72 report was made to the NRC at 0327 hours0.00378 days <br />0.0908 hours <br />5.406746e-4 weeks <br />1.244235e-4 months <br /> on August 15, 1992.

2. Boiling Water Reactor (BWR).cores typically operate with the presence of global neutron flux noise. This noise can be characterized as oscillations having a dominant frequency of 0.3 to 0.5 Hz and a maximum peak to peak amplitude of 5 to 8% of rated flux. They are attributed to flow and boiling in the core, are stable due to the negative void reactivity feedback characteristics (i.e., power. increases, voids increase, reactivity decreases, power decreases) and are normally of no consequence to safety, fuel performance, or Plant operation. Power oscillations of concern are driven by thermal-hydraulic instabilities that occur under relative conditions of high power and low flow.. Under these conditions, oscillatory voiding and reflood conditions occur in the impacted fuel assemblies. When this thermal hydraulic instability is combined with the neutronic feedback mechanisms, the observed core oscillations occur.

3. Further evaluation found that radial and axial flux profiles established in the core and the core and fuel assembly design, combined to cause thermal-hydraulic induced power oscillations. The rod pattern selected caused the 9x9 bundles to be operating at higher power levels than the 8x8 fuel resulting in significantly reduced flow due to: 1) higher pressure drop across the 9x9 bundle, and 2) higher power level per bundle causing increased flow resistance. Specifically, the 9x9 bundles in the core have a greater hydraulic pressure drop

LICENSEE EVENT REPOR ER)

TEXT CONTINUATION AClLITY NAHE (1) DOCKET NUHBER (2) LER NUMBER (8) AGE (3) ear umber ev. No.

Washington Nuclear Plant - Unit 2 0 5 0 0 0 3 9 7 2 37 02 4 F 18 1TLE (4)

MANUAL REACTOR SCRAM DUE TO CORE INSTABILITY across them as compared to the 8x8 design. In Cycle 8, WNP-2 had about 26% of its core with 9x9 design and the remaining 74% of the core was 8x8 design including 12 LFAs. As can be seen from the attached Figure 1, a significant number of assemblies were operating at high power and relatively low flow. Thus, a number of individual assemblies were actually operating in the region where instabilities can occur. Enough individual assemblies reached the point of hydraulic instability and the overall core flux oscillations followed.

4, Additional investigations examined events which ied the SNE and operators to place the core in this condition. The high Xenon level allowed the SNE to proceed to the point at which additional shaping rods were withdrawn from the core resulting in a bottom peaked axial power distribution. Also, it allowed the SNE to establish a radial and axial core flux profile to optimize conditions after the pump shift. Specifically, the SNE set a radial flux profile ~

with a Critical Power Ratio (CPR) of 1.812 with a limit of 1.728 giving a Fraction of CPR of .953 for bundle 35;36. The Total Peaking Factor (TPF) was 3.782 for location 35-26-3.

The time for which these values were calculated for was 0257, three minutes before the power oscillations and just prior to closing the FCVs. The BWR Owners'roup (BWROG) guidelines indicate that these factors are contributors to power oscillations when in low flow and high power conditions.

5. At the time of the event one feedwater heater was out of service. This resulted in slightly lower feedwater temperature which did not contribute to the instability.

6. The time spent at relatively high power and low flow (i.e., 34% power, 30% flow) for CRD timing and bypass valve testing increased the chances of a power oscillation event occurring.

Since the Plant was operating in an area below the previously known region of potential instability there were no restrictions applied.

7. Further evaluation of the activities leading up to the event identified a Technical Specification violation. Technical Specification 3.2.2, Power Distribution Limits/APRM Setpoints, is applicable in Operational Condition i when thermal power is greater than or equal to 25% of rated. The purpose of this Technical Specification at less than full power conditions is to provide added protection against a highly peaked power distribution by temporarily adjusting the APRM setpoints to a more conservative value. This is done by evaluating Power Distribution data and determining the "T" factor used to adjust the setpoints. Technical Specification 4.0.4 states that entry into an Operational Condition shall not be made unless the Surveillance Requirements associated with the Limiting Condition for Operation have been performed within the applicable surveillance interval or as otherwise specified.

Surveillance Requirement 4.2.2.c states the surveillance is to be performed "...Initially and at least once per 12 hours1.388889e-4 days <br />0.00333 hours <br />1.984127e-5 weeks <br />4.566e-6 months <br /> when the reactor is operating with Maximum Fraction of Limiting Power Density (MFLPD).greater than or equal to the Fraction of Rated Thermal Power (FRTP)". During the power ascension associated with this event the Plant reached 25%

LICENSEE EVENT REPORTER)

TEXT CONTINUATION ACILITY NAHE (1) OOCKET NUHBER (2) LER NUHBER (8) AGE (3) ear umber ev. No.

Washington Nuclear Plant - Unit 2 0 5 0 0 0 3 9 7 2 3 7 0 2 5 F 18 ITLE (4)

MANUAL REACTOR SCRAM DUE TO CORE INSTABILITY power at 2320 hours0.0269 days <br />0.644 hours <br />0.00384 weeks <br />8.8276e-4 months <br /> on August 15, 1992. At that time, the "T" factor was approximately 0.85. A further decrease occurred between 0100 and 0300 hours0.00347 days <br />0.0833 hours <br />4.960317e-4 weeks <br />1.1415e-4 months <br /> with the "T" factor at the time of the event being 0.73. Plant Procedure PPM 7.4.2.1, "Power Distribution Limits",

implements the requirements of this Technical Specification..The procedure required the."T" factor to be evaluated and the APRM setpoints to.be adjusted every -12 hours when "T" is less than one. This was a violation of the Technical Specifications and is reportable under 50.73(a)(2)(i)(B).

8. Further investigation discovered a design problem with the ANNA Monitoring System which was installed in 1989. This system monitors the signal of six APRMs and 18 LPRMs. Its purpose is to provide advance warning of both global (in-phase) and regional (out-of-phase) oscillations. The ANNA Monitoring System is governed by Technical Specification 3/4.2.7 and 3/4.2.8, Stability Monitoring. The action statements of these Technical Specifications requires a decrease in thermal power or an increase in coie flow within 15 minutes if the decay ratio measured by ANNA is greater than 0.75. The decay ratio is the ratio between the amplitude of two consecutive peaks in the neutron signal. The decay ratio is less than one for a stable system. A decay ratio of 0.75 is selected as a decay ratio limit for operator response such that sufficient margin to an instability occurrence is maintained. The natural frequency of a BWR is approximately 0.3 to 0.5 Hz. The ANNA Monitoring System monitors between 0,2 and 0.7 Hz. The LPRM/APRM signals normally have 0.3 Hz low pass filters to remove noise from the signal used for normal power monitoring. However, this same signal was used by the ANNA monitoring system which, on subsequent design review, needed a 5 Hz filter to give accurate readings. The use of a 0.3 Hz filter on the input to ANNA resulted in nonconservative values of decay ratio in the region of interest.

For example, calculations by Siemens Power, Corporation for data associated with this event show that with a 0.3 Hz filter ANNA would calculate a decay ratio of 0.62 while the decay ratio calculated utilizing unfiltered signals would be 0.89. Thus, the design 'of the ANNA Monitoring System and its associated signal conditioning did not provide for accurate decay ratio determinations. An additional problem was discovered with Surveillance Procedure PPM 7.4.2.7.3, "Core Stability Monitoring". This procedure did not reflect the fact that the input to ANNA was filtered.,As a result, the peak to peak amplitude reflected in the procedure was nonconservative. This condition is reportable under 10CFR50.73(a)(2)(ii)(B).

LICENSEE EVENT REPORTER)

TEXT CONTINUATION AGILITY NAME (I) OOCKET NUMBER (2) LER NUMBER (8) AGE (3) ear umber ev. No.

Washington Nuclear Plant - Unit 2 3 7 6 F 18 ITLE (4)

MANUAL REACTOR SCRAM DUE TO CORE INSTABILITY B. MRQ~yM This event was analyzed by Plant staff with representatives from General Electric, INPO, and Siemens Power Corporation. The primary root cause for this oscillation event has been determined to be unanticipated interaction of operating conditions and components. The Supply System staff and vendor personnel failed to identify the extent of the tendency for this core design to become unstable under certain operating practices. In addition to the primary root cause, there were several contributing root causes.

1. The first root cause involves Plant/System operation since the effects of changing operating parameters were not fully evaluated in that:

ao The SNE selected a startup rod pattern with characteristics of aggressive critical power ratios and high radial peaking. This was standard operating practice to minimize stress on the Reactor Recirculation pumps and the fuel. Maximizing the amount of control rod movement during low speed recirculation pump operation minimizes the amount of rod movement required after the shift to high speed. With the control rods set to support power increases principally by increasing flow, the number of high power changes associated with rod movement is significantly reduced.

This results in less loc'al fuel stress at higher power levels. In addition, recirculation flow and power could then be increased quickly minimizing the time spent at high speed with the FCVs at the minimum position when the pump is subject to increased vibration.

b. Past startups and operating regimes during Cycle 8 had not disclosed problems with similar patterns.

C. The Plant stability monitor (ANNA) was not employed to provide early warning of the potential changes in core instability. This was, in part, due to past experience and a confirmation that the existing core exclusion region versus actual and planned Plant conditions for this startup were acceptable.

Thus, the SNE failed to consider the need to be conservative in his rod pattern selection and use'f monitoring tools for conditions that could promote and predict core stability.

LICENSEE EVENT REPOR ER)

TEXT CONTINUATION ACILTTY NAHE (1) OOCKET NUHBER (2) LER NUHBER (8) AGE (3) ear umber ev. No.

Washington Nuclear Plant - Unit 2 0 5 0 0 0 3 9 7 2 037 02 7 F 18 ITLE (4)

MANUAL REACTOR SCRAM DUE TO CORE INSTABILITY

2. The second root cause was design related. Specifically, there was inadequate independent review of design changes in that:

a. The Supply System design review process for the mixed core consisting of 9x9 and 8x8 fuel assemblies failed to discern the impact the differences in hydraulic resistance of the fuel assemblies would have on the core's susceptibility to instabilities outside existing instability regions defined by Technical Specification 3.2.7 and current BWROG guidance.

b. Design review included assurances of conformance to license requirements, but did not discern that core stability licensing analyses did not consider the effects of high peaking on core stability at operating conditions which existed at conditions other than the licensing basis.

Thus, the design review program responsible for setting limits on the Plant failed to adjust the. rod pattern and peaking conditions to assure core stability.

3. The third root cause'was analysis deficiencies in that:

a. The fuel designer did not perform sensitivity analyses for core instability at reactor conditions other than those required to perform reload licensing analyses.

b, Current licensing methodologies do not require these sensitivity analyses on startup power distributions.

C. Computer models to analyze cores to this level of sophistication are in development and not licensed for use.

d. The designer believed the hydraulic differences between 8x8 and advanced 9x9 would not contribute to the likelihood of instability under nominal startup conditions.

Thus, the fuel designers failed to identify that this core was less stable and did not provide recommendations for compensating through conservative operating conditions.

LlCENSEE EVENT REPQRER)

TEXT CONTINUATION ACILITY NAHE (1) OOCKET NUHBER (2) LER NUHBER (8) AGE (3) ear umber ev. No.

Washington Nuclear Plant - Unit 2-0 5 0 0 0 3 9 7 2 37 2 8 F 18 ITLE (4)

MANUAL REACTOR SCRAM DUE TO CORE INSTABILITY

4. A contributing causal factor was management methods. A Management Oversight and Risk Tree (MORT) analysis has been completed for this event. This analysis found significant weaknesses in the barriers management has established to prevent the event and in the

. management controls in place for the design and operation of the core. Management acknowledges it should have responded to prior industry information and critically questioned the design oversight and operating philosophy to minimize the potential for core power oscillations. Specific findings resulted from this review included the following:

ao Management's response to the Implementation Guidance for Stability Interim Corrective Actions issued on March 18, 1992, was weak. Training was provided to Shift Technical Advisors (STAs), SNEs and Plant Operators but procedures were not updated.

b. Management Policy allowed too much flexibilityfor the SNE/STA to determine the core flux profile, PI C. Management methods used to review the reload design did not ask penetrating questions in the area of core stability.

d. Management decisions and reasoning for reanalysis and acceptance of a lower feedwater interlock value were not well communicated. Consequently, procedures were inconsistently amended and SNE operating strategies were not appropriately influenced to take advantage of the lower feedwater value.

5. The root cause for the Technical Specification violation associated with the "T" factors for the APRM setpoints was less than adequate procedures. The Technical Specification requirements were not adequately written into the procedure requiring the adjustment prior to 25% power. A contributing root cause was management methods which failed to recognize and take corrective action for the noncompliant condition created by Technical Specification Surveillance 7.4.2.1.

6. The root cause for the problem with the filtered input to ANNA was a design configuration and analysis deficiency. The interface between the existing Plant hardware and the ANNA hardware received a less than adequate review and analysis. A contributing cause was an inadequate review and test of the design change to assure operability after installation.

LICENSEE EVENT REPORER)

TEXT CONTINUATION ACILITY NAME (1) DOCKET NUMBER (2) LER NUMBER (8) AGE (3) ear umber ev. No.

Washington Nuclear Plant - Unit 2 0 5 0 0 0 3 9 7 2 3 7 2 9 F 18 1TLE (4)

MANUAL REACTOR SCRAM DUE TO CORE INSTABILITY C.. h -T rm rr iveA i The following short-term corrective actions have been completed.

In order to maintain assembly power to flow ratio as low as possible, as well as maintaining radial peaking as low as possible, procedures were revised to require CPR greater than 2.2

.between 25% power and 50% core flow. Core total peaking factor will be maintained less than 3.4 prior to pump shift, These are initial parametric values and will be reevaluated for each cycle during the transition from a mixed 8X8/9X9 to a uniform 9X9 core.

Fifteen case studies were run by the Fuel Vendor to validate the stability of the Cycle 8 core.

This was accomplished utilizing one dimensional and three dimensional modeling codes. The calculations were performed under a variety of conditions including but not limited to:

August 15, 1992, restart conditions, August 2, 1992, restart conditions before and after FCV closure, and restart conditions with worst case under corrective action restraints both now (500 MWD/MTU), at 1000 MWD/MTU, and at 1500 MWD/MTU.

The results of these stability analysis show decay ratios for this core to be less than 1.00.

All cases showed decay ratios to be between 0.2 and 0.6 indicating all situations to be self dampening.

2. In an effort to minimize the inlet subcooling, which can contribute to power oscillations, a change was made to the Minimum Feedwater Temperature Curve in Plant Procedure, PPM 3.1.2, "Reactor Plant Cold Startup."

3. Procedures for monitoring power oscillations will require that the ANNA system be operable and in service from greater than 25% reactor power and less than 50% core flow (see attached, Figure 2).

4. In addition to general precautions recommended by the BWROG on stability, specific requirements were put in place to minimize Plant testing and time spent below 50% core rated flow and above 25% core rated power. This will require shifting recirculation pumps from 15 Hz to 60 Hz speed at power less than 33%.

5. An approved startup plan was written and approved that controlled and specified rod patterns for the Cycle 8 startup. This plan utilized a rod pattern that was analyzed for stability prior to closing the first flow control valve for recirculation pump shift to fast speed. Revisions, except for minor rod position deviations due to instrument or equipment malfunctions, will require a new stability evaluation of the alternate rod pattern and Plant Operating Committee approval prior to recirculation pump upshift.

LIGENSEE EVENT REPORTER)

TEXT CONTINUATION ACILITT NAHE (1) DOCKET NUHBER (2) LER NUHBER (8) AGE (3) ear umber ev. No.

Washington Nuclear Plant - Unit 2 0 5 0 0 0 3 9 7 3 7 10 F 18 ITLE (4)

MANUAL REACTOR SCRAM DUE TO CORE INSTABILITY

6. These actions were implemented by procedural changes and training of personnel on these changes. The Power to Flow Map was revised to reflect these changes by designating an "INCREASED AWARENESS" region (see attached Figure 2). BWROG Implementation Guidance for Stability Interim Corrective Actions have been implemented in Plant procedures and reinforced by training sessions and exam testing.

7. Plant Shutdown and Abnormal Condition Procedures were modified to provide increased monitoring, precautions, and direction regarding potential core instabilities.

8. The Supply System will continue its involvement in the,BWROG activities involving stability. A Supply System Principal Engineer has been spending approximately half time participating in these activities. This includes work as a primary representative on the Stability Committee ATWS/Stability Task Force and the committee involved with the long-term hardware proposal.

9. A memo was issued by the Plant Manager to all Plant Personnel informing them of the significance of this event. The memo provided information on the seriousness of the event, a summary of the causes, and an outline of corrective actions.

10. Plant Procedure PPM 7.4.2.1, "Power Distribution Limits", has been modified to require a "T" factor adjustment prior to exceeding 25% core power.

11. The six APRM and eighteen LPRM input signals to ANNA have been modified to eliminate the 0.3 Hz low pass filter. All ANNA signals now have a 5.0 Hz low pass filter.

12. A new Plant Procedure PPM 2.1.8, ANNA Stability Monitoring System, was written to describe the operation of ANNA outside of Region C on the power to flow map.

13. A peer review by the BWROG was performed of the current WNP-2 operating practices related to prevention, detection, and suppression of power oscillations.

The Supply System will continue to evaluate the appropriateness of the operating strategies implemented as short-term corrective actions. The actions taken will be considered during the evaluations that will be made during the implementation of the long-term actions. These results may identify changes needed in the operating strategies initially established.

D. n-Trm rr iv A in The following long-term design process corrective actions have been identified. These long-term corrective actions address the design process and the reviews by Supply System personnel to validate fuel vendor's calculations to ensure stability regions, are avoided throughout core operating cycles.

LICENSEE EVENT REPORER)

TEXT CONTINUATION AClLITY NAHE (1) OOCKET NUHBER (2) LER NUHBER (8) AGE (3) ear umber ev. No.

Washington Nuclear Plant - Unit 2 0 5 0 0 0 3 9 7 2 3 7 2 11 F 18 ITLE (4)

MANUAL REACTOR SCRAM DUE TO CORE INSTABILITY s'dditionally, verification of the fuel vendor's analysis to support design reviews for future cycles will be performed.

An important aspect of the development of the long-term corrective actions and the assignment of priority to each activity was in the determination of the implications on the design process and the operating strategies. Although the WNP-2 Cycle 8 core met the required reload design criteria, the power oscillation occurred outside of the currently identified region of instability. It has become apparent that a core reload design could satisfy all of the regulatory requirements and still allow undesirable operational situations. The Supply System has concluded that, as a long-term corrective action, it will be necessary to supplement our review of the design process. Because the root cause identified problems with operating strategies, the design process and the review of those designs must address those strategies. Additionally, the Supply System recognizes the importance of continued involvement with the industry in the resolution of the core stability issues. The following is a summary of evaluations performed to supplement the existing design process.

The scope of the Supply System design reviews was expanded to provide additional oversight of vendor reload design and analysis. This expanded review process, implemented for the Cycle 9 reload design, included increased scope and technical depth of the design. A plan for the design review for Cycle 9 was completed and the enhanced reviews included:

a. increased emphasis on the operating performance of the core in addition to meeting the licensing requirements;

b. increased awareness of the impact of core and fuel design changes on Plant operations;

c. increased attention to core stability and thermal hydraulic characteristics; and

d. the fuel vendor was required to perform additional stability analysis beyond the current licensing requirements.

2. The Supply System has evaluated the feasibility of changing the fuel design to one that is more stable than the current 9x9-9X design. The long-term objective is the use of fuel designs which create known and manageable stability characteristics during Plant operations and transients. The goal is that the next fuel fabrication contract, for fuel to be delivered in 1995, will meet the criteria necessary to satisfy this objective. Proposals are under evaluation, with a contract award scheduled for December 1993. The potential of earlier changes in fuel design was evaluated, but to assure understanding of the impacts of a new fuel design on the existing mixed core, the Supply System does not expect to be able to implement fuel design changes prior to the 1995 fuel delivery. The Supply System also evaluated the feasibility of changing the fuel design to one that is more stable than the cur-

LICENSEE EVENT REPOR ER)

TEXT CONTINUATION AGILITY NAHE (1) OOCKET NUHBER (2) LER NUHBER (8) AGE (3) umber ev. No.

Washington Nuclear Plant - Unit 2 0 5 0 0 0 3 9 7 2 037 2 12 F 18 ITLE (4)

MANUAL REACTOR SCRAM DUE TO CORE INSTABILITY rent 9x9-9X design. Design changes for Cycle 9 were evaluated and rejected because the Supply System did not believe enough operation experience" was available to justify those changes.

3. The Supply System encouraged the fuel vendor to accelerate the validation of the present stability code used for assessing selected rod patterns for the startup plan. Seimens has agreed to submit the STAIF code to the NRC for review and approval.

4, The stability of the existing core was evaluated as part of the startup plan discussed above under short-term corrective actions (paragraph C.5) to ensure that the corrective actions were valid during Plant startups during the remainder of Cycle 8 operation.

Beyond those issues directly related to core design and operating strategies, the Supply System identified some contributing factors that require further corrective action.

5. A short-term corrective action involved splitting signals for the LPRMs input into ANNA, the stability monitoring program. This implementation approach associated with the modification decreased the flexibilityof ANNA. The Supply System has evaluated the impact of this reduction in flexibility. A contract to implement modifications has been awarded.

6. The evaluation discussed above was also a reliability improvement evaluation of the core stability monitor (ANNA). This investigation included assessment of power supply, CPU redundancy, auto alarm features, and enhanced surveillance techniques to verify continued hardware and software operability. Recommendations are being implemented.

7. The Reactor Engineering Group within the Plant Technical Department provides on-shift, direction to the operating crews during power operation and maneuvering. The Fuels Engineering Group, within Engineering, is the primary interface with the fuel vendor. The Supply System has evaluated this division of responsibilities and the working relationship between the two groups in establishing strategies during startup and full power operation.

The recommendations have been made to Supply System management and are under review,

8. To regain flow margin for Cycle 9, the Supply System cleaned five jet pumps during the R8 outage.

9. An evaluation of long-term shutdown strategies ensured the correct procedures are in place for all conditions.

LICENSEE EVENT REPORTER)

TEXT CONTINUATION ACILITY NAHE (1) DOCKET NUHBER (2) LER NUHBER (8) AGE (3) ear umber ev. No.

Mashington Nuclear Plant - Unit 2 0 5 0 0 0 3 9 7 2 037 13 F 18 ITLE (4)

MANUAL REACTOR SCRAM DUE TO CORE INSTABILITY

10. The use of a startup plan with a specified rod pattern and peaking factors was revalidated for Cycle 9. This evaluation established that maintaining the same parameters used during Cycle 8 provided an adequate margin for stability. The Supply System will continue this approach, by requiring that a POC approved startup plan must be in place prior to exceeding 25% ..

thermal power.

ll. An action plan has been developed to improve the effectiveness of the Supply System's participation in industry activities.

12. Management issues associated with this event were addressed and corrective actions are ongoing. The MORT analysis has been completed. Management has reviewed this report and initiated the following actions:

a. Action was taken to strengthen programs and practices used to review and assimilate industry information. Specific changes to Plant practices have been incorporated to ensure BWROG and NUMARC information is critically screened for specific and for general relevance to WNP-2. A critical examination was performed to discuss other shortcomings in this area and mechanisms were put in place to strengthen the influence of industry initiatives on Supply System practices.

b. Action was taken to strengthen our reactivity management program. A Plant Procedure 1.3.59, "Reactivity Management Program", has been approved. This new program was developed using contact. with other utilities to adopt the best features of their programs. An assessment was done to assure the link betweeri design bases and operating constraints. The recommendations have been made to Supply System management and are under review.

C. A corporate level review, made of the overall relationship between Engineering Services, the fuel vendor, operators, STAs, and the SNEs assured that responsibilities and duties for all aspects of fuel design, "operation, fuel design related independent review, quality assurance, limit setting and recommendations on operating modes was well defined, active, effective and understood by all concerned. Included in this evaluation was a review to assure all barriers assumed to be in place (such as ANNA) to prevent reactivity related events are actually used in a fashion such that the barrier is effective in performing its intended function. This item is complimentary to item 12b. To strengthen the relationship, recommendations made to Supply System management are under review.

13. Disciplinary action has been defined for responsible individuals at all levels of management associated with this event (see Reference 6).

LlCENSEE EVENT REPOR ER)

TEXT CONTINUATION AGILITY NAME (1) DOCKET NUMBER (2) LER NUMBER (8) AGE (3) ear umber ev. No.

Washington Nuclear Plant - Unit 2 0 5 0 0 0 3 .9 7 2 37 2 14 F 18 ITLE (4)

MANUAL REACTOR SCRAM DUE TO CORE INSTABILITY The following is a summary of long-term corrective actions.

1. To support the enhanced reload design reviews and implementation of operating strategies, the Supply System will pursue obtaining core stability analysis codes. The Supply System will evaluate the existing codes and their availability in an attempt to implement their use in support of Cycle 10 core design review. This evaluation will be completed by January 1, 1994.

2. Replace FCVs and two speed pump operation with ASD pumps. This will eliminate the need to conduct operations under the restrictions on FCVs and 15 Hz speed pumps. The current two speed recirculation pumps will be powered from adjustable speed power supplies, allowing continuous flow adjustments from 15 to 60 Hz. With this modification, Recirculation FCVs are not required. Due to implementation restraints and concerns, this modification is now scheduled to be complete by June 30, 1996. The decision to install the ASD pumps was made as part of the Megawatt Improvement Project. The Supply System does not believe the decision to delay the installation of the ASD pumps will challenge the ability to operate the Plant with adequate margin for stability. The corrective actions taken and described in this report assure that WNP-2 is operated with adequate margin for stability.

fe i nifi n This event has safety significance since 10CFR50 Appendix A, General Design Criteria for Nuclear'ower Plants was challenged. Criteria 12, Suppression of Reactor Power Oscillations, states, "The reactor core and associated coolant, control, and protection systems shall be designed to assure that power oscillations which can result in conditions exceeding specified acceptable fuel design limits are not possible or can be reliable and readily detected and suppressed." Although exact wording of the criteria were met at all times the fact that the oscillations occurred unexpectedly results in some safety significance.

A water chemistry analysis shortly after the event showed no evidence of fuel failure. Analysis of the primary coolant gave no indication of iodine spiking which would be indicative of fuel failure following depressurization. A comparison of the coolant chemistry with previous shutdowns showed nothing abnormal.

During the initial assessment following the event, a bounding POWERPLEX analysis was performed to ensure MCPR limits had not been exceeded. The available APRM data indicated the maximum neutron flux magnitude was approximately plus to minus 9%; As a bounding case this steady state analysis assumed power was increased by 9% with no increase in flow. The results showed a delta CPR of 0.5.

With an initial CPR of 1.946, the minimum CPR would be below the Operating Limit Minimum CPR of 1.795 but well above the Safety Limit Minimum CPR (SLMCPR) of 1.07 including uncertainties in data and assumptions.

c LICENSEE EVENT REPOR ER)

TEXT CONTINUATION ACILITY NAME (1) OOCKET NURSER (2) LER NUNBER (8) AGE (3) ear umber ev. No.

Washington Nuclear Plant - Unit 2 0 5 0 0 0 3 9 7 2 3 7 2 15 F 18 ITLE (4)

MANUAL REACTOR SCRAM DUE TO'CORE INSTABILITY Transient analysis was also performed based on the peak to peak oscillations noted on the LPRMs. This transient analysis was performed with the Supply System VIPRE Transient code and was independently performed by Siemens using the XCOBRA-T code. A conservative hot channel analysis using a 30%

peak-to-peak input resulted in a delta CPR of approximately 0.20 from both codes.

It can, therefore, be concluded the oscillations did not result in any fuel failures or SLMCPR limits being exceeded.

The safety significance of the "T" factor adjustment not being performed prior to exceeding 25% power was evaluated. The purpose of this adjustment at nonrated (low power) conditions is to provide added protection against a highly peaked power distribution by temporarily adjusting the APRM sensitivity to a more conservative value. Had this been done at the 25% power point it would have resulted in a 15%

decrease in the margin to the trip setpoint. This is not safety significant, The safety significance of the 0.3 Hz filters associated with the ANNA input has been reviewed. If ANNA had been used with these filters it would have provided results that were nonconservative. The ANNA Monitoring System was one of the systems being relied upon for the "detect and suppress" strategy associated with core oscillations when operating in Region "C" on the power to flow map. The inoperability of this equipment because of filtered input was safety significant since it could have allowed if the Plant to go into the region of'instability unknowingly the system had been used. However, it is backed up by the Reactor Protection System (high neutron flux and flow referenced flux scram) which would have shutdown the reactor if the oscillations became too severe.

imilar Even There have been no similar events at WNP-2.

LICENSEE EVENT REPOR ER)

TEXT CONTINUATION AGILITY NAME (1) DOCKET NUNBER (2) LER NUHBER (8) AGE (3) ear umber ev. No.

Washington Nuclear Plant - Unit 2 0 5 0 0 0 3 ~ 9 7 2 3 7 2 16 F 18 ITLE (4)

MANUAL REACTOR SCRAM DUE TO CORE INSTABILITY EII Inf rm i n R f ~EII f

$ ggem Q~mgne~n Average Power Range Monitors IG MON (APRMs)

Local Power Range Monitors IG MON (LPRMs)

Recirculation Flow Control Valve AD FCV Reactor Bottom Head Drain Valve Bypass Valve (RWCU-V-103)

Advanced Nuclear Noise JC Analysis (ANNA) Monitoring System

n EA O m ~ v)

O f'nwt'.r l=lnw - lt>tlivitlttttlRssotnl)lies m

n O~

Figure 1 C)

HA C/l R 37 C) t= m

-I O

,7A ~

R I!)((st 1!i, 1'.)!)." at .'):AA AM, l(1 ".i(:c.  !) XQ C:

I'(Iw(.f -- .).).7 l'l(lw = 2'.7 C(A Kl m

C7

!In- CO 7C m

O X E((

it 0 m CI A

30 20 10 10 2!i  :)0 nn Ct(IW 1% t)I(tr(t)

S 0 0

~

a

~

I I

~ ~

l ~