ML20113E931
| ML20113E931 | |
| Person / Time | |
|---|---|
| Site: | Limerick  |
| Issue date: | 05/12/2020 |
| From: | Christopher Hunter NRC/RES/DRA/PRB |
| To: | |
| Hunter C (301) 415-1394 | |
| References | |
| LER 353-01-001 | |
| Download: ML20113E931 (11) | |
Text
1 SENSITIVE - NOT FOR PUBLIC DISCLOSURE Final Precursor Analysis Accident Sequence Precursor Program --- Office of Nuclear Regulatory Research Limerick 2 Manual scram following inadvertent open/stuck open main steam safety relief valve Event Date: 02/23/200 LER: 353/01-00 CDP = 3x10-6 April 30, 2004 Condition Summary Description. On February 23, 2001, operators at Limerick 2 were performing a planned shutdown to repair the 2N and 2M main steam relief valves (MSRVs), which had been leaking for the past several months. During power reduction with the unit at 85% power, the 2N MSRV inadvertently lifted and remained open for two separate periods. Following the first opening of the 2N MSRV, a manual reactor scram was initiated approximately 2 minutes after the opening per the technical specifications. During the expected postscram level shrink, the (1) reactor protection system and (2) primary containment and reactor vessel isolation control system both received a valid actuation signal. Two loops of the residual heat removal (RHR) system were manually placed into service in the suppression pool cooling mode. The main steam isolation valves were closed to control the reactor depressurization and cooldown rate. However, the rapid depressurization resulting from the MSRV opening caused a violation of the technical specifications maximum cooldown rate of 100 F per hour. The first period of the 2N MSRV remaining open lasted 3 hours3.472222e-5 days <br />8.333333e-4 hours <br />4.960317e-6 weeks <br />1.1415e-6 months <br /> and 51 minutes with reactor pressure dropping from 1,040 psig to 122 psig.
Forty minutes after the 2N MSRV closed the first time, the valve again inadvertently opened for a 35-minute period, during which reactor pressure dropped from 206 psig to 122 psig (Reference 1).
Cause. The cause of inadvertent opening of the 2N MSRV was a loss of seating material from the first stage pilot valve due to long-term erosion and oxidation of the Stellite disc material.
This loss of material prevented the valve from reseating completely after inadvertent opening.
Also, due to pilot valve disc erosion, the 2N MSRV was found to repeatedly lift at pressures less than the rated reactor pressure.
LER 353/01-001 1 For the condition assessment, the parameter of interest is the measure of the incremental increase between the conditional probability for the period in which the condition existed and the nominal probability for the same period but with the condition nonexistent and plant equipment available. This incremental increase or importance is determined by subtracting the CDP from the CCDP. This measure is used to assess the risk significance of hardware unavailabilities especially for those cases where the nominal CDP is high with respect to the incremental increase of the conditional probability caused by the hardware unavailability. For the initiating event assessment, the parameter of interest is the absolute value of the CCDP.
2 Analysis Results
Importance or CCDP1 Condition assessment Conditional core damage probability (CCDP) 6.2 x 10-6 Nominal core damage probability (CDP) 3.0 x 10-6 Importance ( CDP = CCDP - CDP) 3.2 x 10-6 The Accident Sequence Precursor Program acceptance threshold for condition assessment is an importance ( CDP) of 1 x 10-6. The importance of the condition assessment for this event exceeds the precursor threshold.
The CCDP for the initiating event assessment (CCDP = 5 x 10-6) does not exceed the CCDP for transient with loss of all main feedwater (CCDP = 6 x 10-6) and would not be considered a precursor. Because the initiating event analysis did not indicate that the event was a precursor, documentation of the initiating event analysis is not provided in this report.
Dominant sequences Condition assessment: The sequence with the highest importance is a sequence that starts with the loss of a vital ac electrical bus followed by the failure of an MSRV to close, LOACB Sequence 51-06 (see Figure 1). The important top events for this sequence are shown in Table 2. These include:
LOACB Sequence 51-06:
- Reactor protection system (RPS) succeeds
- One MSRV fails to close
- Power conversion system succeeds
- Failure of suppression pool cooling mode of RHR system
- Failure of shutdown cooling mode of RHR
- Failure of containment venting
- Late injection to the reactor vessel is unavailable The importance of LOACB Sequence 51-06 is 3.1 x 10-6. No other sequence in the condition assessment had an importance higher than 4.5 x 10-7.
SENSITIVE - NOT FOR PUBLIC DISCLOSURE
Results tables The conditional probabilities of the dominant sequence are shown in Table 1.
The event tree sequence logic for the dominant sequence is provided in Tables 2a and 2b.
The conditional cut sets for the dominant sequence are provided in Table 3.
The definitions for the modified or dominant basic events are shown in Table 4.
Modeling Assumptions
Assessment summary This event was modeled as both a condition assessment and as an initiating event -
inadvertent opening of a relief valve (IE-IORV).
The condition assessment was run for 1,944 hours0.0109 days <br />0.262 hours <br />0.00156 weeks <br />3.59192e-4 months <br /> with one MSRV failed (TRUE house event) during the entire period. The condition assessment accounts for all initiating events using the nominal probability of occurrence for each.
Time period/condition duration. This event was analyzed as both a condition assessment and as an initiating event.
The degraded condition of the relief valve existed for a period of about 81 days. That is, the failure mechanism of interest for the 81 days is the opening of the relief valve and the failure to reclose following any plant transient (e.g., turbine trip, loss of offsite power
[LOOP]). The condition assessment model does not include an increased probability of the valve to inadvertently open on its own during this period.
The 81-day period for the condition assessment was selected based on a review of preevent and postevent data. The date when the 2N MSRV was degraded to the point of not being able to reclose occurred approximately the first week of December 2000. To monitor SRV degradation, the licensee used pilot valve temperature as a means to monitor first stage pilot valve leakage after pressure monitors on the valve failed. The licensee determined that once the temperature dropped to 497°F, the 2N SRV would need to be repaired or replaced. In August 2000, the licensee reduced the pilot valve temperature limit from 497°F to 475°F. However, the limit should not have been reduced below 492F, because at this temperature, licensee analysis showed that if the SRV opened, it may fail to re-close. On December 5, 2000, the pilot valve temperature dropped below 492°F. From that date until February 23, 2001, when the 2N SRV opened and did not immediately re-close, the 2N SRV was in a degraded condition.( References 3
& 4) Therefore, the time period from December 5, 2000, to February 23, 2001 (81 days or 1,944 hours0.0109 days <br />0.262 hours <br />0.00156 weeks <br />3.59192e-4 months <br />), was used for this condition assessment.
Recovery opportunities. Failure of the MSRV following a reactor trip would not affect (preclude) recovery actions following failure of other equipment.
SPAR model used in the analysis Limerick 1 & 2, Revision 3.02, October 2003 (Reference 2)
SENSITIVE - NOT FOR PUBLIC DISCLOSURE
Modifications to event tree and fault tree models The event tree for IE-IORV was changed to include a transfer to the anticipated transient without scram event tree for failure of the RPS. See Figure 2.
Event tree linking rule changes In the station blackout sequences with a stuck-open relief valve and successful high-pressure injection, the time to recover electric power was changed from 30 minutes to 2 hours2.314815e-5 days <br />5.555556e-4 hours <br />3.306878e-6 weeks <br />7.61e-7 months <br />. The following event tree linking rules were added for the LOOP event tree (between rules 13 and 14):
l13.5 lSBO1 if EPS*SRV[1]*/HPI1 then
/OEP = OEP-02H; OEP = OEP-02H;
/DGR = DGR-02H; DGR = DGR-02H; endif These linking rule changes were sent to ABS Consulting by personnel from the Idaho National Engineering and Environmental Laboratory (INEEL). The changes were the result of a number of conversations between personnel from these two organization. In these conversations, it became clear that assuming it was necessary to recover electric power within 30 minutes was too conservative with respect to IE-LOOP followed by loss of the emergency electric power system and successful high pressure injection. INEEL personnel changed the event tree linking rules to allow two hours to recover electric power.
Basic event probability changes Table 4 provides the basic events that were modified to analyze the event. The bases for some of the changes are as follows:
Probability of one safety relief valve fails to reclose (PPR-SRV-OO-1VLV). This value was changed from FALSE to 2.20 x 10-1 in the base case in order to conduct a condition assessment. The value of 2.20 x 10-1 was that used in Revision 3i. In the condition assessment, for the change case, this probability was set to a failure probability of TRUE (for the 81-day period). See sensitivity analysis section, below.
Probability of two safety relief valves fail to reclose (PPR-SRV-OO-2VLVS). This value was changed from FALSE to 1.30 x 10-3 in the base case in order to conduct a condition assessment. The value of 1.30 x 10-3 was that used in Revision 3i.
Probability of three or more safety relief valves fail to reclose (PPR-SRV-OO-3VLVS). This value was changed from FALSE to 2.20 x 10-4 in the base case in order to conduct a condition assessment. The value of 2.20 x 10-4 was that used in Revision 3i.
SENSITIVE - NOT FOR PUBLIC DISCLOSURE
Model update and modifications The SPAR model for Limerick 1 & 2 (Reference 2) was updated (Reference 5) to account for recent changes to the alpha factor for common-cause failures of essential service water, raw service water, and service water motor-driven pumps as follows:
Basic Event Description Value ESW-MDP-FR-04A01 Motor-driven pump alpha factor for 1 for 4 trains 9.69 X 10-1 ESW-MDP-FR-04A02 Motor-driven pump alpha factor for 2 for 4 trains 1.35 X 10-2 ESW-MDP-FR-04A03 Motor-driven pump alpha factor for 3 for 4 trains 4.28 X 10-3 ESW-MDP-FR-04A04 Motor-driven pump alpha factor for 4 for 4 trains 1.31 X 10-2 RSW-MDP-FR-04A01 Motor-driven pump alpha factor for 1 for 4 trains 9.69 X 10-1 RSW-MDP-FR-04A02 Motor-driven pump alpha factor for 2 for 4 trains 1.35 X 10-2 RSW-MDP-FR-04A03 Motor-driven pump alpha factor for 3 for 4 trains 4.28 X 10-3 RSW-MDP-FR-04A04 Motor-driven pump alpha factor for 4 for 4 trains 1.31 X 10-2 SWS-MDP-FR-03A01 Motor-driven pump alpha factor for 1 for 3 trains 9.68 X 10-1 SWS-MDP-FR-03A02 Motor-driven pump alpha factor for 2 for 3 trains 1.38 X 10-2 SWS-MDP-FR-03A03 Motor-driven pump alpha factor for 3 for 3 trains 1.86 X 10-2
Sensitivity analysis The analysis was done with a value of 3.1E-2 for PPR-SRV-OO-1VLV (Probability that one safety relief valve [SRV] fails to close) instead of 2.2E-1. This resulted in an importance = 4.0E-6, an increase of 25%.
It should be understood that what is measured in this sensitivity analysis is the importance or change in core damage probability from a base case. By reducing the probability that the safety relief valve fails to close, we are changing the base case core damage probability to a smaller value. Since the change case sets the probability of the safety relief valve failing to close to TRUE, the importance or change from the base case to the change case gets larger. In this case, the increase is 25%.
Related events A related event occurred in 1995 at Limerick 1 when a two-stage SRV inadvertently opened. This event was reviewed for unique modeling characteristics that could pertain to this analysis. None were found that would apply for the present analysis. No other related events were found for the 81-day period prior to the event on February 23, 2001.
SENSITIVE - NOT FOR PUBLIC DISCLOSURE References
- 1.
LER 353/01-001, Revision 0, Manual scram following MSRV failing open due to erosion and oxidation of the first stage pilot valve disc seating area, April 24, 2001 (ADAMS Accession Number: ML011170065).
2.
J. A. Schroeder, Standardized Plant Analysis Risk Model for Limerick 1 & 2, Revision 3.02, INEEL, October 2003. Computer model update October 2003.
3.
Limerick Generating Station - NRC Inspection Report 50-352/01-011, 50-353/01-011 December 7, 2001 (ADAMS Accession Number ML013410133) 4.
Final Significance Determination for a White Finding and Notice of Violation at the Limerick Unit 2 Generating Station (NRC Inspection Report 50-352/01-011; 50-353/01-011) January 11, 2002 (ADAMS Accession Number ML020110173) 5.
U.S. Nuclear Regulatory Commission Reactor Operating Experience Results and Database Common Cause Failures Database http://nrcoe.inel.gov/index.cfm/index.cfm?fuseaction=CCFDB.showMenu
SENSITIVE - NOT FOR PUBLIC DISCLOSURE Table 1. Conditional probabilities associated with the highest probability sequences for the condition assessment.
Event tree name Sequence no.
CCDP CDP Importance1 (CCDP - CDP)
LOACB 51-06 4.0 x 10-6 8.7 x 10-7 3.1 x 10-6 Total (all sequences)2 6.2 x 10-6 3.0 x 10-6 3.2 x 10-6 Notes:
1.
Importance is calculated using the CDP and the total CDP from all sequences. Sequence level importance measures are not additive.
2.
Total CCDP includes all sequences (including those not shown in this table). (File name: GEM 353-01-001 01-14-2004 092329 - Run for Record.wpd.)
Table 2a. Event tree sequence logic for the condition assessment.
Event tree name Sequence no.
Logic
(/ denotes success; see Table 2b for top event names)
LOACB 51-06
/RPS P1 /PCS SPC SDC CVS LI05CF Table 2b. Definitions of top events listed in Table 2a.
CVS CONTAINMENT VENTING FAILS LI05CF LATE INJECTION IS UNAVAILABLE PCS POWER CONVERSION SYSTEM IS UNAVAILABLE P1 ONE SRV FAILS TO CLOSE RPS REACTOR SHUTDOWN FAILS SDC SHUTDOWN COOLING MODE OF RHR FAILS SPC SUPPRESSION POOL COOLING MODE OF RHR FAILS Table 3b. Conditional cut sets for LOACB Sequence 51-06.
Importance per hour Percent contribution Minimal cut sets1 Event Tree: LOACB Sequence 51-06 5.3 x 10-10 25.9 RBENV RHR-HTX-TM-TRNB 5.3 x 10-10 25.9 RBENV RHR-XHE-XR-HTXB 2.6 x 10-10 13.0 RBENV RHR-XHE-XR-ERROR 2.0 x 10-9 Total2 Notes:
1.
See Table 4 for definitions and probabilities for the basic events.
2.
Total Importance includes all cut sets (including those not shown in this table).
SENSITIVE - NOT FOR PUBLIC DISCLOSURE Table 4. Definitions and probabilities for modified or dominant basic events.
Event name Description Probability/
Frequency Modified IE-LOACB Loss of vital ac electrical bus (condition assessment only) 2.1 x 10-6 per hour No PPR-SRV-OO-1VLV Probability of failure of one MSRV to reclose (condition assessment change case)
TRUE Yes1 PPR-SRV-OO-1VLV Probability of failure of one MSRV to reclose (condition assessment base case) 2.2 x 10-1 Yes1 PPR-SRV-OO-2VLVs Probability of failure of two MSRVs to reclose (condition assessment base case) 1.3 x 10-3 Yes1 PPR-SRV-OO-3VLVS Probability of failure of three or more MSRVs to reclose (condition assessment base case) 2.2 x 10-4 Yes1 RBENV Reactor building steam environment causes injection failure 2.5 x 10-1 No RHR-HTX-TM-TRNB RHR heat exchanger B is in test or maintenance 1.1 x 10-2 No RHR-XHE-XR-HTXB Operator fails to restore heat exchanger B after test or maintenance 1.0 x 10-3 No RHR-HTX-TM-TRNB RHR Heat Exchanger B is in teat and maintenance 1.0 x 10-3 Yes2 Notes:
1.
Basic event changed to reflect event being analyzed. See text.
2.
ASEP mean value, NUREG/CR 4550 Vol1, Rev. 1 Analysis of Core Damage Frequency: Internal Events Methodology.
LI LATE INJECTION CVS CONTAINMENT VENTING PCSR POWER CONVERSION SYSTEM RECOVERY SDC SHUTDOW N COOLING SPC SUPPRESSION POOL COOLING VA ALTERNATE LOW PRESS INJECTION LPI LOW PRESSURE INJECTION CDS CONDENSATE DEP MANUAL REACTOR DEPRESS CRD CRD INJECTION SPC SUPPRESSION POOL COOLING (EARLY)
HPI HIGH PRESSURE INJECTION PCS POWER CONVERSION SYSTEM SRV SRV S CLOSE RPS REACTOR SHUTDOWN IE-LOACB LOSS OF MEDIUM VOLTAGE AC BUS END-STATE 1
OK 2
OK 3
OK 4
OK 5
CD 6
OK 7
CD 8
OK 9
OK 10 OK 11 CD 12 OK 13 CD 14 OK 15 OK 16 OK 17 CD 18 OK 19 CD 20 OK 21 OK 22 OK 23 CD 24 OK 25 CD 26 CD 27 CD 28 OK 29 OK 30 OK 31 OK 32 CD 33 OK 34 CD 35 OK 36 OK 37 OK 38 OK 39 CD 40 OK 41 CD 42 OK 43 OK 44 OK 45 OK 46 CD 47 OK 48 CD 49 CD 50 CD 51 T 1SORV 52 T 2SORVS 53 CD P1 P2 SPC1 SDC1 LI05CF LI06 LI07 LI05CF LI05CV LI05CV LI01 LI02 LI05CF LI05CF LI05CV LI05CV LI06 LI07 SDC1 LOACB - LOSS OF AC BUS 2004/01/14 Figure 1. Limerick 1 & 2 LOACB event tree.
9 SENSITIVE - NOT FOR PUBLIC DISCLOSURE LER 353/01-001
LI LATE INJECTION CVS CONTAINMENT VENTING PCSR POWER CONVERSION SYSTEM RECOVERY SDC SHUTDOWN COOLING SPC SUPPRESSION POOL COOLING VA ALTERNATE LOW PRESS INJECTION LPI LOW PRESSURE INJECTION DEP MANUAL REACTOR DEPRESS CRD CONTROL ROD DRIVE INJECTION HPI HIGH PRESSURE INJECTION PCS POWER CONVERSION SYSTEM P1 ONE STUCK OPEN SRV END-STATE 1
OK 2
OK 3
OK 4
CD 5
OK 6
CD 7
OK 8
OK 9
OK 10 OK 11 CD 12 OK 13 CD 14 OK 15 OK 16 OK 17 OK 18 CD 19 OK 20 CD 21 OK 22 OK 23 OK 24 OK 25 CD 26 OK 27 CD 28 CD 29 OK 30 OK 31 OK 32 OK 33 CD 34 OK 35 CD 36 OK 37 OK 38 OK 39 OK 40 CD 41 OK 42 CD 43 CD 44 CD SPC1 SDC1 LI05CF LI06 LI07 LI05CF LI05CF LI08 LI09 LI05CF SPC1 SDC1 LI05CV LI05CV LI05CV LI05CV 1SORV - ONE STUCK OPEN SAFETY/RELIEF VALVE 2004/01/14 Figure 1. Limerick 1 & 2 LOACB event tree (continued).
10 SENSITIVE - NOT FOR PUBLIC DISCLOSURE LER 353/01-001
LI LATE INJECTION CVS CONTAINMENT VENTING PCSR POWER CONVERSION SYSTEM RECOVERY SDC SHUTDOWN COOLING SPC SUPPRESSION POOL COOLING VA ALTERNATE LOW PRESS INJECTION LPI LOW PRESSURE INJECTION DEP MANUAL REACTOR DEPRESS CRD CONTROL ROD DRIVE INJECTION HPI HIGH PRESSURE INJECTION PCS POWER CONVERSION SYSTEM RPS REACTOR PROTECTION SYSTEM IE-IORV INADVERTENT OPEN RELIEF VALVE END-STATE 1
OK 2
OK 3
OK 4
CD 5
OK 6
CD 7
OK 8
OK 9
OK 10 OK 11 CD 12 OK 13 CD 14 OK 15 OK 16 OK 17 OK 18 CD 19 OK 20 CD 21 OK 22 OK 23 OK 24 OK 25 CD 26 OK 27 CD 28 CD 29 OK 30 OK 31 OK 32 OK 33 CD 34 OK 35 CD 36 OK 37 OK 38 OK 39 OK 40 CD 41 OK 42 CD 43 CD 44 CD 45 T ATWS SPC1 SDC1 LI07 LI06 LI07 LI07 LI07 LI06 LI07 LI07 SPC1 SDC1 LI06 LI06 LI06 LI06 IORV - INADVERTENT OPEN RELIEF VALVE (IORV) 2004/01/14 Figure 2. Limerick 1 & 2 IORV event tree.
11 SENSITIVE - NOT FOR PUBLIC DISCLOSURE LER 353/01-001