IR 05000311/1999010
| ML20114E204 | |
| Person / Time | |
|---|---|
| Site: | Salem  |
| Issue date: | 05/12/2020 |
| From: | Christopher Hunter NRC/RES/DRA/PRB |
| To: | |
| Hunter C (301) 415-1394 | |
| References | |
| IR 1999010 | |
| Download: ML20114E204 (11) | |
Text
1 Final Precursor Analysis Accident Sequence Precursor Program --- Office of Nuclear Regulatory Research Salem, Unit 2 4160 Vac switchgear rooms did not reach or maintain the required CO2 concentration of 50% for Fire Suppression
'DWH 2/14/2000 2/14/2000
,5 50-311/99-10 CDP = 1x10-6 1.0 x 10-6 April 29, 2003 Condition Summary Description On December 7, 1999, the NRC completed an inspection of selected areas of the Salem fire protection syste The team identified that the carbon dioxide (CO2) concentration for the Unit 2 4160 Vac switchgear room did not reach or maintain the required CO2 concentration of 50 percent during testing (Ref. 1). The CO2 system also did not meet its design requirement, as stated in the final safety analysis report, which requires the CO2 tanks to contain a sufficient supply of CO2 for two full discharges into the largest protected area The installed tank was found to be only half ful This condition placed the CO2 system in a degraded condition such that the system may not have been fully effective in extinguishing fire Duration The CO2 system was in a degraded condition since the original construction in the 1970 For the purposes of the Accident Sequence Precursor (ASP) analysis, this condition was modeled for a maximum period of one year.
Recovery Opportunit Within the switchgear room, the 1B and 1C divisions of ac power are assumed failed with no recover IR 50-311/99-10 1 Since this condition did not involve an actual initiating event, 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 availabl This incremental increase or importance is determined by subtracting the CDP from the CCD 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.
SENSITIVE - NOT FOR PUBLIC DISCLOSURE Analysis Results
Importance1 The risk significance of the CO2 system being degraded and not effective, resulting in Division 1B and Division 1C 4160Vac power failing due to a postulated fire is determined by subtracting the nominal core damage probability from the conditional core damage probability for the point estimate CDP = 1.5 x 10-6 and a mean CDP = 1.0 x 10- This is an mean increase of 1.0 x 10-6 over the nominal CDP for the one year period where the Division 1B and Division 1C 4160Vac power was not available due to a postulated switchgear room fir The uncertainty about the mean is: 5% bound, 9.4 x 10-9 and the 95% bound, 4.1 x 10-6(see Figure 3).
NOTE:
The case where the fire does not propagate from Division 1B to Division 1C within the switchgear room was screened out, as the CDP was less than the Accident Sequence Precursor (ASP) program acceptance threshold importance ( CDP) of 1 x 10-6.
Dominant sequence The initiating event is a postulated fire in the 4160V ac switchgear roo The dominant core damage sequence for this condition is Fire-Induced Transient - Sequence 2 The events and important component failures in this sequence (see Figure 2) include:
-
Reactor trips successfully
-
Failure of AFW
-
Failure of Main Feedwater system during transient
-
Successful bleed portion of feed and bleed
-
Failure of HPI system flow, resulting in core damage.
Results tables
-
Table 1 provides the conditional probabilities for the dominant sequence.
-
Table 2a provides the event tree sequence logic for the dominant sequence.
-
Table 2b defines the event tree sequence logic elements listed in Table 2a.
-
Table 3 provides the conditional cut sets for the dominant sequence.
-
Table 4 provides the definitions and probabilities for modified and dominant basic event IR 50-311/99-10
SENSITIVE - NOT FOR PUBLIC DISCLOSURE Modeling Assumptions
Assessment Summary The risk significance of the Division 1B and 1C 4160Vac cabinets/equipment is determined by performing an initiating event assessment using the revision 3.02 model for Salem with the transient initiating frequency replaced with the product of the switchgear room initiating fire frequency and the probability of nonsuppressio The current probability of basic events that are assumed failed (TRUE) are used in the analysi This method is outlined in NUREG/CR-6544, Development of a Methodology for Analyzing Precursors for Earthquake-Induced or Fire-Induced Accident Sequences, Section 3.7 (Re ). For the case used in this analysis, the fire is assumed to propagate from Division 1B to Division 1C, The postulated fire in the 4160Vac switchgear room is assumed to fail both Division 1B and 1C electrical cabinets/equipment without recover However the fire is not assumed to propagate to Division 1A electrical cabinets/equipment because there are no cable/cable trays that traverse between the 1B/1C and 1A divisions (The case of a fire in Division 1A alone was determined to be of lower risk significance), adequate distance separation(approximately 20 feet between divisions), and no other basis for assuming propagation (see Fig. 1).
SPAR model used in the analysis Revision 3.02 Standardized Plant Analysis Risk (SPAR) model for Salem Units 1 and 2 (Ref. 3) was used for this assessmen The transient initiating event (IE-TRANS)
frequency is replaced (see below for details of fire-induced analysis considerations).
Fire induced analysis considerations The fire-induced analysis is based on NUREG/CR-6544 (Ref. 2). For this analysis all 4160Vac Division 1B and Division 1C equipment is assumed failed and the product of the initiating fire frequency for the plant location (switchgear room) and the probability of nonsuppression replaced the transient initiating event frequency in the fire-induced initiating event assessment.
-
Initiating Fire Frequency - The initiating fire frequency (Fi) was developed from NRC Report RES/OERAB/S02-01 power operation fire event data for severe fires (fires with duration greater than 5 minutes that were not self-extinguished) in the Switchgear Room during the 1986-1999 period (Ref. 4) and updated with 2000-2001 fire dat For the Salem 2 switchgear room, the fire frequency (Fi) used was 3.2 x 10-3 based on the following:
IR 50-311/99-10
SENSITIVE - NOT FOR PUBLIC DISCLOSURE Two switchgear room fire zones are assumed for this analysis: one for 4160 Vac divisions B and C and one for 4160 Vac division A.
Fi = (No. of Severe Fire Events + Jeffreys Prior of 0.5 Fire Events)
(No. of Switchgear Rooms x No. of Power Operation Reactor-Years)
Fi = (8 + 0.5) = 3.2 x 10-6 (2 x 1311)
-
Probability of nonsuppression - The assumption is that the fire will propagate between Division 1B and 1C, but not to Division 1A (see Assessment Summary, above).. Therefore, for this assumption, the probability of nonsuppression =
Unique system and operational considerations All three divisions of the 4160 Vac switchgear are located in the 4160 Vac switchgear room (Fig. 1). Design characteristics include:
-
The switchgear 4160Vac divisions are separated from each other by partial height, partial length marinite walls (Fig. 1). These walls form radiant heat shields and are not rated 1-hour fire barriers (Ref. 1.).
-
Fire protection in the area is provided by a manually actuated CO2 fire suppression system, smoke detectors, manual hose stations, and portable fire extinguisher The room is accessible at opposite end Overhead cable trays associated with 4160 Vac divisions 1B and 1C are located in close proximity of each othe There are no intervening overhead cables and there is adequate distance separation between 1B/1C and 1A (Fig. 1).
-
The 4160 Vac division 1A provides power to one train of safety systems, with the exception of intermediate pressure safety injection/CVCS pump These pumps are powered from 4160 Vac divisions 1B and1 Reactor coolant pump seal cooling is accomplished by either the chemical and volume control system (CVCS) or the component cooling water syste
Modifications to event tree and fault tree models None
Initiating event frequency changes
IR 50-311/99-10
SENSITIVE - NOT FOR PUBLIC DISCLOSURE The transient initiating event (IE-TRANS) frequency was replaced by the product of the initiating fire frequency for the Switchgear Room and the probability of nonsuppression
[(3.2 x 10-3) x 1.0 = 3.2 x 10-3]. Figure 2 shows the event tree for the fire-induced transient.
NOTE:
Since the postulated fire is in the switchgear room (applicable to a transient), no other initiating event is considered as applicable to this analysis (Ref. 2 provides this basis for replacing IE-TRANS only).
Basic event probability changes Table 4 provides the basic events that were modified to reflect the condition being analyze Division 1B AC power 4160 V bus fails (ACP-BAC-LP-1B) and Division 1C AC power 4160 V bus fails (ACP-BAC-LP-1C). These basic events were set to TRUE (i.e., 1.0 failure probability) to reflect the worst case fire damage state due to the propagation of fire between the two electrical divisions via the cable trays.
-
IE TRAN The transient initiating event was set to the product of the initiating fire frequency and the probability of nonsuppression (3.2 x 10-3) to reflect the worst case fire damage state due to the propagation of fire between the two electrical divisions (B and C) via the cable trays.
NOTE:
For this analysis, IE FIRE replaces IE TRANS (see Fig. 2).
Model update No updates were made to the revision 3.02 SPAR model for Salem.
NOTE:
The analysis assumes that high temperature seals were installed on all RCPs at the time of the even For this analysis, the RCP seal LOCA is not applicabl IR 50-311/99-10
SENSITIVE - NOT FOR PUBLIC DISCLOSURE References 1.
NRC Inspection Report, 272/1999-010, 311/1999-010, Salem Generating Station - Final Significance Determination and Notice of Violation, February 14, 2000 (ADAMS Accession Number: ML003683723).
2.
R.W.Budnitz, et al., Development of a Methodology for Analyzing Precursors to Earthquake-Induced and Fire-Induced Accident Precursors, NUREG/CR-6544, U.S.
Nuclear Regulatory Commission, Washington, DC, April 1998.
3.
J. K. Knudsen, et al., Simplified Plant Analysis Risk (SPAR) Model for Salem Unit 1 & 2, Revision 2QA, Idaho National Engineering and Environmental Laboratory, December 1997.
4.
J.R. Houghton and D. M. Rasmuson, NRC Report RES/OERAB/S02-01, Fire Events Update of U.S. Operating Experience, 1986-1999, U.S. Nuclear Regulatory Commission, Washington, DC, January 2002.
5.
J. P. Poloski, et al., Rates of Initiating Event at U.S. Nuclear Power Plants: 1987-1995, NUREG/CR-5750, U.S. Nuclear Regulatory Commission, Washington, DC, February 1999.
6.
C. L. Atwood, et al., Evaluation of Loss of Offsite Power Events at Nuclear Power Plants:
1980-1996, NUREG/CR-5496, U.S. Nuclear Regulatory Commission, Washington, DC November 1998.
7.
J. P. Poloski, et al., Reliability Study: Auxiliary Feedwater System,1987-1995, NUREG/CR-5500, Vol. 1, U.S. Nuclear Regulatory Commission, Washington, DC, August 1998.
8.
F. M. Marshall, et al., Common-Cause Failure Parameter Estimations, NUREG/CR-5497, U.S. Regulatory Commission, Washington, DC, October 1998.
9.
G. M. Grant, et al., Reliability Study: Emergency Diesel Generator Power System, 1987-1993, NUREG/CR-5500, Vol. 5, U.S. Nuclear Regulatory Commission, Washington, DC, September 1999.
9.
Public Service Electric and Gas company, Salem Generating Station Individual Plant Examination for External Events, January 199 IR 50-311/99-10
SENSITIVE - NOT FOR PUBLIC DISCLOSURE Table Conditional Probability of Dominating Sequences (Point Estimates)
Event Tree Sequence Conditional Core Damage probability (CCDP)
TRANS
6.5E-07 TOTAL All Sequences 1.5E-06 Table 2a Event tree sequence logic for dominant sequence Event Tree Name Sequence no.
Logic (/ denotes success; see Table 2b. for top event names TRANS
/RT, AFW, MFW-T, /BLEEDS, HPI Table 2 Definitions of event tree sequence logic elements listed in Table 2a.
AFW No or insufficient AFW flow BLEED Failure of Bleed portion of Bleed cooling HPI No or insufficient flow from HPI system MFW-T Failure of main feedwater system during transient RT Reactor fails to trip during transient Table Conditional cut sets for dominant sequence Event tree: TRANS, Sequence 20 CCDP Percent Contribution Minimum cut sets1 9.8E-008/hr 15.0 MFW-XHE-NOREC MFW-SYS-UNAVAIL AFW-TDP-FR-13 SWS-MDP-TM-1SWE5 9.8E-008/hr 15.0 MFW-XHE-NOREC MFW-SYS-UNAVAIL AFW-TDP-FR-13 SWS-MDP-TM-1SWE6 7.8E-006/hr 12.0 MFW-SYS-TRIP MFW-XHE-ERROR AFW-TDP-FR-13 SWS-MDP-TM-1SWE5 7.8E-006/hr 12.0 MFW-SYS-TRIP MFW-XHE-ERROR AFW-TDP-FR-13 SWS-MDP-TM-1SWE6 6.5-007/hr Total2 NOTES:
1.
See Table 4 for definitions and probabilities for the basoc events.
2.
Total CCDP includes all other cut sets for this sequence (including those not shown in this table).
IR 50-311/99-10
SENSITIVE - NOT FOR PUBLIC DISCLOSURE Table Definitions and probabilities for modified and dominant basic events Event name Description Probability/
Frequency Modifie d
IE-LOOP LOSS OF OFFSITE POWER INITIATING EVENT 0.00 YES1 IE-SGTR STEAM GENERATOR TUBE RAPTURE INIT. EVENT 0.00 YES1 IE-LDCA LOSS OF DC BUS A INITIATING EVENT 0.00 YES1 IE-LOA LOSS OF INSTRUMENT AIR INITIATING EVENT 0.00 YES1 IE-LLOSA LARGE LOSS OF COOLANT ACCIDENT INIT. EVENT 0.00 YES1 IE-MLOCA MEDIUM LOSS OF COOLANT ACCIDENT INIT. EVENT 0.00 YES1 IE-SLOCA SMALL BREAK LOSS OF COOLANT ACCIDENT INIT.
EVENT 0.00 YES1 IE-LOCCW LOSS OF COMPONENT COOLING WATER INIT. EVENT 0.00 YES1 IE-LOSWS LOSS OF SERVICE WATER INITIATING EVENT 0.00 YES1 IE-RHR-DIS-V RHR DISCHARGE ISLOCA OCCURS INITIATING EVENT 0.00 YES1 IE-RHR-HL-V RHR HOT LEG ISLOCA INITIATING EVENT 0.00 YES1 IE-RHR-SUC-V RHR SUCTION ISLOCA INITIATING EVENT 0.00 YES1 IE-SI-CLDIS-V SI COLD LEG ISLOCA INITIATING EVENT 0.00 YES1 IE-SI-HLDIS-V SI HOT LEG ISLOCA INITIATING EVENT 0.00 YES1 IE-TRAN INITIATING EVENT-TRANSIENT (FIRE-INDUCED)
3.2E-03 YES2 ACP-BAC-LP-1B DIVISION 1B AC POWER 4160 V BUS FAILS TRUE YES3 ACP-BAC-LP-1C DIVISION 1C AC POWER 4160 V BUS FAILS TRUE YES3 Notes:
1.
All initiating events, except IE-TRANS were set to 0.00.
2.
Transient initiating event frequency was revised to reflect the product of the initiating fire frequency and the probability of nonsuppression.
3.
Basic event was changed to reflect condition being analyze TRUE has a failure probability of Figure 1 Salem 2 4160 VAC switchgear Room Simplified Diagram
SENSITIVE - NOT FOR PUBLIC DISCLOSURE IR 50-311/99-10 Figure removed during SUNSI revie HPR HIGH PRESSURE RECIRCULATION RHR RESIDUAL HEAT REMOVAL COOLDOWN RCS COOLDOWN SGCOOL SECONDARY COOLING RECOVERED HPI HIGH PRESSURE INJECTION BLEED BLEED PORTION OF F & B COOLING PORV-RES PORVs CLOSE PORV NO PORVs OPEN MFW-T MAIN FEEDWATER AFW AUXILIARY FEEDWATER RT REACTOR TRIP IE-FIRE FIRE TRANSIENT
END-STATE F 1 OK
2 OK
3 OK
4 OK
5 CD
6 OK
7 CD
8 CD
9 OK
10 OK
11 OK
12 OK
13 CD
14 OK
15 CD
16 CD
17 OK
18 OK
19 CD
20 CD
21 CD
22 T ATWS
PORV-1 Figure 2 Fire-Induced Trans Sequence 20
SENSITIVE - NOT FOR PUBLIC DISCLOSURE IR 50-311/99-10
Figure 3 Uncertainty
SENSITIVE - NOT FOR PUBLIC DISCLOSURE IR 50-311/99-10