05000413/LER-2004-001
| Docket Number | |
| Event date: | |
|---|---|
| Report date: | |
| Reporting criterion: | 10 CFR 50.73(a)(2)(i)(B), Prohibited by Technical Specifications 10 CFR 50.73(a)(2)(ii)(A), Seriously Degraded 10 CFR 50.73(a)(2)(v), Loss of Safety Function 10 CFR 50.73(a)(2)(v)(A), Loss of Safety Function - Shutdown the Reactor |
| 4132004001R01 - NRC Website | |
BACKGROUND
This event is being reported under the following criteria:
10CFR50.73(a)(2)(i)(B), any operation or condition which was prohibited by the plant's Technical Specifications, 10CFR50.73(a)(2)(ii)(A) and (B), any event or condition that resulted in the condition of the nuclear power plant, including its principal safety barriers, being seriously degraded, and the nuclear power plant being in an unanalyzed condition that significantly degraded plant safety; and 10CFR50.73(a)(2)(v)(A) and (D), any event or condition that could have prevented the fulfillment of the safety function of structures or systems.
Catawba Nuclear Station Unit 1 is a Westinghouse four-loop pressurized water reactor [EIIS: RCT]. The Chemical Volume and Control System [EIIS: CB]serves as part of the Emergency Core Cooling System (ECCS) to provide high pressure injection and recirculation of borated water to the Reactor Coolant System cold legs following a design basis accident.
The ECCS components are designed such that a minimum of one centrifugal charging pump (high head), one safety injection pump (intermediate head) [EIIS: BQ], one residual heat removal pump [EIIS: BP] and heat exchanger (low head), and three cold leg accumulators along with their associated valves and piping will ensure adequate core cooling in the event of a design basis accident.
Technical Specification (TS) 3.5.2 requires two trains of ECCS pumps to be operable during Modes 1 - 3. With one train inoperable, the train must be restored to operable status within 72 hours8.333333e-4 days <br />0.02 hours <br />1.190476e-4 weeks <br />2.7396e-5 months <br />. With two trains inoperable, TS 3.0.3 requires action to be taken within 1 hour1.157407e-5 days <br />2.777778e-4 hours <br />1.653439e-6 weeks <br />3.805e-7 months <br /> to place the unit in Mode 3 within 7 hours8.101852e-5 days <br />0.00194 hours <br />1.157407e-5 weeks <br />2.6635e-6 months <br />.
At the time of this event, Unit 1 was operating in Mode 1 at 100 percent power. During the event investigation, several safety systems were removed from service for routine surveillance testing and maintenance activities. No structures, systems, or components were removed from service that had any effect on the event or conflicted with Technical Specifications.
EVENT DESCRIPTION (Times are approximate) Date� Time Event Description 12/08/03 1557 ECCS venting procedure completed in preparation for unit startup. Unit 1 operating in Mode 6. No significant amount of gas identified during the venting procedure.
12/16/03 0145 Unit 1 entered Mode 3.
01/07/04 1622 ECCS venting procedure completed for the monthly surveillance. Approximately 1 minute of gas vented from the vent valve (1NV858) on the ECCS suction line from the refueling water storage tank. The vent valve 1NV858 was cracked open.
Because the valve was cracked open, there is no accurate means of measuring the amount of gas that was vented. Data from ultrasonic testing was not available.
Additional vent valves were opened and no additional gas identified.
Engineering conservatively approximated a maximum of 1.6 cubic feet of gas vented from 1NV858 based on the vent time and assuming the valve was fully open. It is possible that less gas may have been vented because the venting procedure is a qualitative assessment to determine if any gas is present in the system and not a quantitative measure of the amount of gas present.
Initial belief was that gas entered the piping during an outage alignment following the vent performed on 12/8/03.
ECCS venting frequency increased from monthly to weekly. Event entered into the corrective action program for further evaluation.
2213 1B charging pump placed in service. 01/7/04� 01/14/04 1200 ECCS venting procedure completed and approximately to 5 minutes of gas vented from 1NV858 and 1500 approximately 2 minutes and 10 seconds of gas vented from 1NV860 (charging pump suction piping from residual heat removal pump A). Engineering approximated 76 gallons (10.2 cubic feet) vented from 1NV858 and approximately 60 gallons (8.0 cubic feet) vented from the 1NV860.
A Failure Investigation Team assembled to identify the source of gas accumulation. A systematic investigation of all possible sources of gas intrusion initiated.
01/15/04 1850 In the absence of engineering analysis to confirm operability, Engineering recommended the centrifugal charging pumps suction in the containment sump recirculation flow path from A train was inoperable based on the rate and amount of gas accumulation at 1NV860 vented 1/15/04 1700 hours0.0197 days <br />0.472 hours <br />0.00281 weeks <br />6.4685e-4 months <br />.
01/15/04 1857 Periodic venting identified gas accumulation limited to the charging pump suction piping from residual heat removal pump A (1NV860) - "piggyback flowpath". Valve 1ND28A de-energized to prevent any gas at valve 1NV860 from being aligned to the charging pumps. The 72-hour action statement for TS 3.5.2 entered for one train of ECCS inoperable.
01/15/04 2339 8-hour phone notification to NRC completed based on evaluation at 1850 hours0.0214 days <br />0.514 hours <br />0.00306 weeks <br />7.03925e-4 months <br />.
01/15/04�Frequent ultrasonic testing (UT) of suction piping to was implemented to closely monitor gas 1/18/04 accumulation. Gas periodically vented to maintain the piping filled.
01/18/04 1725 Potential gas entry locations and generation mechanisms evaluated. Based on the decreasing gas volume, the gas origin believed to be related to makeup water pump 1B maintenance. Investigation of gas intrusion source continued.
Power restored to valve 1ND28A and TS 3.5.2 exited after completion of 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> without significant voids in suction piping. Ultrasonic testing was continued in order to closely monitor gas voids in the charging pump suction piping.
01/19/04 0350 Gas void discovered at 1NV860 using ultrasonic testing.
01/19/04 0430 TS 3.0.3 entered for both trains of ECCS inoperable. [Later analysis determined the ECCS was operable.] 01/19/04 0506 TS 3.0.3 exited after gas vented at 1NV860.
01/19/04 0715 Valve 1ND28A de-energized. The 72-hour action statement for TS 3.5.2 entered for one train of ECCS inoperable.
01/19/04 1123 8-hour phone notification to NRC completed.
01/19/04 Investigation of gas accumulation continued.
to 01/22/04 Frequent ultrasonic testing (UT) of suction piping continued to closely monitor gas accumulation.
Gas periodically vented to maintain the piping filled.
Source of gas accumulation not identified, however, the charging system was determined to be operable with actions taken to maintain the piping filled and vented. One of the actions included isolation of the primary sample purge line connected to the volume control tank relief valve header. [The isolated sample purge line was eventually determined to be the gas leakage pathway.] 01/22/04 0707 Power restored to valve 1ND28A and TS 3.5.2 exited with both trains of ECCS operable.
No significant amount of gas identified after TS 3.5.2 exited on 1/22/04.
01/22/04�Deliberate, systematic troubleshooting plan to developed and implemented to confirm the source of 02/7/04 the gas.
02/7/04�One source of gas identified as reverse leakage from the relief valve 1NV235 on the emergency boration line. The upstream primary sample purge valve, 1NM10, was also identified as leaking.
02/24/04�Relief valve 1NV235 and valve 1NM10 were replaced.
03/13/04 Troubleshooting activities involving 1NV235 indicated gas accumulation was still present when the primary sample purge line was aligned to the volume control tank relief valve header. The primary sample purge line was isolated before and after the troubleshooting activities to prevent gas accumulation.
Valve 1NM10 confirmed to be leaktight.
Valve 1ND28A de-energized, as necessary, during troubleshooting.
03/13/04 Primary sample purge isolation valve remained to closed to isolate the suspected gas pathway.
06/1/04 Ultrasonic testing continued to closely monitor the charging pump suction piping and ensure the piping remained within acceptable limits for pump operability.
Additional troubleshooting plans developed to determine the source of gas accumulation.
Plant conditions, manpower, and testing resources coordinated in preparation of troubleshooting activities.
06/1/04�Troubleshooting activities resumed.
Troubleshooting results indicated reverse leakage through relief valve 1NV235.
06/2/04 Relief valve 1NV235 replaced.
to 06/3/04 Troubleshooting activities indicated the relief valve leakage was terminated following valve replacement. However, as a precautionary measure, the primary sample purge line will be isolated until the long term corrective action to relocate the Unit 1 primary sample purge line connection to the relief valve header is completed. If the purge line is restored to normal alignment, then an alternate method will be implemented to maintain the gas accumulation in the charging pumps within an acceptable value for pump operability.
06/09/04 Analytical flow model completed. The results found that the charging pumps and safety injection pumps were operable prior to 01/07/04 and after 01/14/04.
Between 01/07/04 and 01/14/04, the 1A and 1B charging pumps were inoperable for the following conditions:
1.An accident condition that required emergency core cooling system flow, and 2.Reactor coolant system pressure below 2000 psig.
The 1A safety injection pump was inoperable for an accident that required emergency core cooling suction alignment to the containment sump.
During accidents that would not depressurize the reactor coolant system below 2000 psig, the 1B charging pump would operate within the vendor recommendation for void fraction.
01/22/04�No significant amount of gas has been identified to in the charging pump suction piping after TS 3.5.2 06/15/04 was exited on 1/22/04.
During this period, no gas has been identified in the charging pump suction except during troubleshooting evolutions. During the troubleshooting evolutions, the gas volumes were closely monitored.
CAUSAL FACTORS
The cause of the gas accumulation was reverse leakage of gas in the volume control tank relief header through valve 1NV235. The relief valve 1NV235 is connected to the emergency boration line to the charging pump suction. Contributing to this event is the piping configuration of the pressurizer sample purge line connection to the volume control tank relief valve header.
Maintenance was performed on relief valve 1NV235 during the past outage and the post-maintenance functional testing was satisfactory. The relief valve may have lifted during emergency core cooling system check valve testing when the charging pump suction was aligned to the residual heat removal pump discharge.
Another possibility is the relief valve reverse leakage may have occurred when the primary sample purge valves were manipulated.
Because the pressurizer sample line connection to the volume control tank relief valve header will be modified, no additional corrective actions are needed regarding 1NV235.
The review of the valve maintenance activities, check valve testing procedures, or plant equipment monitoring has not identified any performance deficiencies.
CORRECTIVE ACTIONS
Immediate:
1.Venting frequency increased upon initial discovery of gas accumulation.
2.Ultrasonic testing conducted at the charging pump suction piping to monitor for gas accumulation. Gas periodically vented to maintain system filled.
3.Failure investigation team assembled to determine the cause of gas accumulation.
4.Various actions established to maintain the system operable.
Actions included monitoring of the suction piping and temporary isolation of the primary sample purge line.
Subsequent:
1.Source of gas accumulation identified from leakage through valve 1NM10 and relief valve 1NV235. Valve 1NM10 was replaced on 02/24/04. The relief valve was replaced on 02/24/04 and 06/2/04.
2.Troubleshooting activities completed. With the exception of valves 1NM10 and 1NV235, no additional valves were identified as the source of gas accumulation.
3.As a precautionary measure, the primary sample purge line alignment to the relief valve header is isolated and under administrative control. The administrative control will be removed after completion of the modification to relocate the primary sample purge line connection. The administrative control may also be modified if an alternate method is implemented to maintain the gas accumulation in the charging pumps within an acceptable value for pump operability.
4. Unit 2 evaluated and determined to not be susceptible to similar leakage due to a different relief valve header pipe configuration.
Planned:
1. Relocate Unit 1 primary sample purge line connection to the relief valve header.
The planned corrective action is being addressed within the Catawba Corrective Action Program. The Subsequent Corrective Action Number three to isolate the primary sample purge line (until completion of the modification or implementation of an alternate method) is an NRC commitment. There are no additional NRC commitments contained in this LER.
SAFETY ANALYSIS
Throughout the gas accumulation, the charging pumps did not exhibit any degradation in pump operating parameters. The pumps flow, pressure, temperature, vibration, and noise were normal with no evidence of gas binding or gas intrusion during normal charging operation.
Ultrasonic testing of the suction piping was instituted early to monitor gas accumulation and maintain the system vented, thereby reducing the time that the pumps were inoperable. Testing frequency was as often as every two hours until an estimate of the gas accumulation rate was established. Testing frequency was adjusted based on plant conditions and venting results.
Prior to the event, surveillance testing was maintained within the Technical Specification requirements. The isolation of valve 1ND28A was within the Technical Specification Action time limits.
The leakage rates exhibited by the sample valves were within the containment leak rate acceptance criteria.
An independent, analytical flow model was conducted to determine the pump(s) operability for the amount of gas identified on 01/7/04, 01/14/04, and 01/19/04. The flow evaluations concluded that all charging pumps and safety injection pumps were operable on 01/7/04 and 01/19/04. The gas volume introduced to the charging pump suction on 03/12/04 during troubleshooting was bounded by the analysis conducted for the January gas analysis. Therefore, the emergency core cooling system would have operated as designed prior to 01/7/04 and after 01/14/04.
The amount of gas discovered on 1/14/04 resulted in both charging pumps and 1A safety injection pump being inoperable from 01/7/04 to 01/14/04. With the reactor coolant system pressure below 2000 psig, sufficient charging flow would be established to cause the gas to migrate into the charging pump suction and result in a void fraction above the vendor recommendation. The introduction of gas into the lA safety injection pump suction would occur during the cold leg recirculation alignment. For any event that does not depressurize the reactor coolant system below 2000 psig, the 1B charging pump would operate within the vendor recommendations. For any event that does not result in cold leg recirculation alignment, the 1A safety injection pump would be unaffected.
The core damage significance of this event has been evaluated quantitatively using actual maintenance configurations and consideration of the cumulative effect of each individual time period that gas entrainment impacted the injection and/or recirculation phases of ECCS. The conditional core damage probability (CCDP) for the event is approximately 4E-07.
The dominant core damage sequences associated with this event have the significant containment safeguards systems available. These include at least one train of the containment spray and hydrogen mitigation systems. Furthermore, most have low to moderate reactor coolant system pressures at reactor vessel failure. Sequences of this nature contribute insignificantly to the large early release frequency (LERF), which is dominated by the ISLOCA, station blackout, and seismic initiating events. Therefore, this event is judged to have no significance with respect to the LERF for Catawba.
ADDITIONAL INFORMATION
Within the last three years, no other LERs occurred at Catawba involving gas accumulation in the charging pump suction.
Therefore, this event was determined to be non-recurring in nature.
A review of industry operating experience indicates that gas intrusion is an industry concern and events have been identified at other sites. However, the Catawba event is a new, previously unidentified failure mechanism for gas introduction into the charging pump suction. Therefore, this event could not have been prevented from the review of operating experience.
Energy Industry Identification System (EIIS) codes are identified in the text as [EIIS: XX]. This event is considered reportable to the Equipment Performance and Information Exchange (EPIX) program.
This event met the reporting criteria of 10 CFR 50.73(a)(2)(v) and therefore will be recorded under the NRC Performance Indicators for Unit 1 as a Safety System Functional Failure.
There were no releases of radioactive materials, radiation exposures or personnel injuries associated with this event.