05000341/LER-1997-003, :on 970303,EECW Sys Ci Function Was Outside Design Basis Due to Design Basis Deficiency.Eecw Sys Primary Containment Penetrations Have Been Modified to Provide Diverse Power
| ML20141A910 | |
| Person / Time | |
|---|---|
| Site: | Fermi  |
| Issue date: | 05/08/1997 |
| From: | Antrassian A DETROIT EDISON CO. |
| To: | |
| Shared Package | |
| ML20141A877 | List:
|
| References | |
| LER-97-003, LER-97-3, NUDOCS 9705150051 | |
| Download: ML20141A910 (6) | |
| Event date: | |
|---|---|
| Report date: | |
| Reporting criterion: | 10 CFR 50.73(a)(2)(ii)(B), Unanalyzed Condition |
| 3411997003R00 - NRC Website | |
text
4
,n LICENSEE EVENT REPORT (LER)
F AcaJrv NAME (4 DOuE T NuMrsd4 c2, PAGE<h Fermi 2 0 l5 l0 l0 l0 l3 l411 1 l'l6 Emergency Equipment Cooling Water System Containment isolation Function Outside Design Basis And Subsequent identification Of Several New Bypass Leakage Paths N
va vR sEcut i Nuu RE sy 04v vn F Aciur < hAuf s r
- Engs, 0
5 0
0 0
l 03 03 97 97 0
0 3
0 1
05 08 97 0
5 0
0 0
o THis nEP<;nt is sveM rrED PVHsuANT
'O THE REQuihEMENTS OF 10 CFR OO 4
POwLR
'N X
Um oTHER.
0l0l0 (Specify in Abstract below and in text. NRC Form 366A)
LICEhGEE CONTACT FOR THis Lift DJ)
MEPeo*G MJMMR Andrew V. Antrassian - Compliance Engineer 313 586-1856 COMPLETE ONE LINE FOR EACH COMPONENT FA6 LURE DESCRIBED IN THl$ REPORT (13)
CAuSE
a rSTEM COMPONENT MANuf Actuf*ER AEP T ABLE CAvSE S V $ TEM COMPONENT MANuF AC TVHER NEPON I
l l
l l l
l l
l l
l l
l l
1 I
II I
II I
I I
I II suPPtf ueNT Ac nera r faEc rec o
EXPECTED "0N'"
0^'
'E^"
SUBMISSION
[ ] YES (if yes, Complete EXPECTED SUBMISSION DATE) l [X ] NO DATE (15) l On March 3,1997, it was determined that the Emergency Equipment Cooling Water System (EECWS) containment isolation function was outside the design basis. The EECWS is normally in standby and portions of the system piping distribute cooling water supplied by the Reactor Building Closed Cooling Water System (RBCCWS). The RBCCWS is isolated from the EECWS upon initiation of the EECWS.
The Updated Final Safety Analysis Report (UFSAR) Containment Isolation System (CIS) design basis identifies that no single failure will result in loss of the containment function. It was determined that a single failure of an electrical division could result in the loss of the EECWS containment isolation function. The EECWS containment penetrations have been modified to provide diverse power such that a single failure will not result in the loss of the containment isolation ftmetion. On April 8,1997, during review of this modification, a total of twelve previously unrecognized bypass leakage paths were identified in the EECWS, Post Accident Sampling System, and Compressed Air System. When leak rate data for these new bypass leakage paths was added to previous bypass leakage rate totals, the UFS AR bypass leakage limit of 4% La was exceeded on two occasions. Alarm Response Procedures (ARPs) have been revised to provide guidance to Operators to manually isolate the EECWS return header primary containment penetrations in the event that certain accident conditions exist. A UFSAR revision has been approved to properly identify the twelve new bypass leakage paths and these paths have been added to the bypass leakage program.
9705150051 970508 PDR ADOCK 05000341 S
- - i LICENSEE EVENT REPORT (LER) TEXT CONTINUATION mm
~~ a, u,
- - o, Fcrmi2 0 l5 l0 l0l0 l3l4 l1 97 0
3 0
2 l"l 6 TEA T (1?)
Initial Plant Condition:
Operational Condition:
4 (Cold Shutdown)
Reactor Power:
O Percent Reactor Pressure:
O psig Reactor Temperature:
122 degrees Fahrenheit Description of the Event:
On March 3,1997, with the plant in a maintenance outage, it was determined that the Emergency Equipment Cooling Water System (EECWS) [BI] containment isolation function was outside the plant design basis. At 1527 hours0.0177 days <br />0.424 hours <br />0.00252 weeks <br />5.810235e-4 months <br />, a four-hour non-emergency notification was made regarding this condition in accordance with 10CFR50.72(b)(2)(i).
The EECWS is an Engineered Safety Features (ESF) system consisting of two independent, redundant, full capacity cooling water divisions. During normal operation, the EECWS is in the standby mode and system piping is used to distribute cooling water provided through system supply and retum cross connect lines by the Reactor Building Closed Cooling Water System (RBCCWS) [CC). The EECWS is activated and system piping is isolated from the RBCCWS upon manual initiation, loss of offsite power, high drywell pressure, or low RBCCWS supply / return header differential pressure. Division 1 (11) EECWS containment and RBCCWS isolation valves are powered by Division I (II) electrical I
busses.
The Fermi 2 Updated Final Safety Analysis Report (UFSAR) identifies the EECWS primary containment penetrations as General Design Criteria (GDC) 56 penetrations (i.e., the associated piping connects directly to the containment atmosphere). While the EECWS piping inside containment is not physically open to the containment atmosphere, this piping does not meet the UFSAR definition of a closed containment barrier. The UFSAR Containment Isolation System (CIS) (JM] design bases identify that the electrical and mechanical designs ensure that no single failure will result in loss of the containment ftmetion. Further, CIS valves for a single division of a redundant ESF system may be powered from a single electrical division so that a single failure of an electrical division can not disable both divisions of an ESF system in these cases, a redundant mechanical barrier (i.e., a closed system beyond the isolation valves) exists so that the containment isolation function is not lost as a result of a single failure. The UFSAR CIS design requirements,in part, require that closed systems used as an isolation barrier inside and outside containment be designated Seismic Category I and at least Quality Group B.
l
I LICENSEE EVENT REPORT (LER) TEXT CONTINUATION wu - o, we - a, g
n ww, Fermi 2 0 l5 l0 l0l0 l3l4l1 97 0
3 0
3 l'l6 TLATOT)
Description of the Event (Continued):
The EECWS piping outside containment is designated Quality Group C and Seismic Category L The RBCCWS piping i:, &signated Quality Group D and Seismic Category ll/I. Consequently, even though this piping is physicaily closed, it does not meet the definition of a closed system and can not be considered a redundant mechanical barrier. A single failure of an electrical division could result in the EECWS containment and RBCCWS isolation valves of the same division failing in the open position which could result in the loss of the containment isolation function.
A modification to provide diverse power to the EECWS primary containment penetrations is discussed in the Corrective Actions section below. During review of this modification,it was identified that the supply and return primary containment penetrations for both divisions of the EECWS may be potential bypass leakage paths which were not previously recognized. UFSAR Table 6.2-2, Summary Of Primary Containment Penetrations And Associated Isolation Valves, indicates that these four penetrations were not considered bypass leakage paths and the penetrations were not included in the bypass leakage program. On April 8,1997,it was confirmed that the EECWS primary containment penetrations are bypass leakage paths and that this constitutes a condition outside the design basis of the plant. At 1104 hours0.0128 days <br />0.307 hours <br />0.00183 weeks <br />4.20072e-4 months <br />, a four-hour non-emergency notification was made regarding this condition in accordance with 10CFR50.72(b)(2)(i).
A review was conducted to identify any other bypass leakage paths that were not properly identified in UFSAR Table 6.2-2 and/or not included in the bypass leakage program. Seven additional penetrations in the Post Accident Sampling System (PASS) [lP] and one additional penetration in the Compressed Air System (CAS) [LF] were identified which were not previously recognized as bypass leakage paths.
llistorical Local Leak Rate Test (LLRT) data was reviewed for all twelve of the penetrations not previously recognized as bypass leakage paths. When the leak rate data for the new bypass leakage paths was added to previous bypass leakage rate totals, the as-found and as-left bypass leakage limit of 4% La identified in UFSAR Section 6.2.1.2.2.3, Bypass Leakage Paths, was exceeded during Refueling Outage 2 (RFO2) and Outage 88-01. The as-found and as-left bypass leakage rates calculated for the current maintenance outage (Outage 97-01) considering the new bypass leakage paths are below the 4% La limit.
The conditions discussed above involving (1) the EECWS containment isolation function, (2) the twelve previously unrecognized bypass leakage paths, and (3) the UFSAR limit of 4% La being unknowingly exceeded on two occasions, are conditions outside the design basis of the plant and are collectively being reported in accordance with 10CFR50.73(a)(2)(ii)(B).
I LICENSEE EVENT REPORT (LER) TEXT CONTINUATION
, - ~
mu -
e _..
m.o,,
VEAR SE%Eh hAL kf 4SK#e Fermi 2 0 l5 l0 l0l0 l3l4l1 97 0 0 3 0
1 4
l*l6 TDT 411)
Cause of the Event
The Fermi 2 Final Safety Analysis Report (FSAR) originally identified the EECWS primary containment penetrations as subject to GDC 57 (i.e., not directly connected to the containment atmosphere). In FSAR Amendment 29, dated April 1980, the EECWS penetrations were changed from GDC 57 to GDC 56. Based on a review of Fermi 2 correspondence from that time frame, the rationale for the change appears to be that because certain EECWS components and piping inside containment were designated Quality Group D, the EECWS could not be considered a closed system inside containment. The cause of this condition is a design basis deficiency in that the plant design was not adequately upgraded when the EECWS containment penetrations were changed from GDC 57 to GDC 56.
1 The cause of the four previously unrecognized EECWS bypass leakage paths is also due to the plant design not being adequately upgraded when the change in designation of these penetrations from GDC 57 to GDC 56 occurred. The cause of the seven previously unrecognized PASS bypass leakage paths is due to an oversight. At the time the PASS was installed, evaluation of the susceptibility of the penetrations through which the PASS was routed to bypass leakage was not performed. The cause of the previously unrecognized CAS bypass leakage path is due to an inappropriate assumption. The CAS was evaluated, and was not considered a bypass leakage path because system pressure was assumed to exceed post accident drywell pressures so that any leakage would be into containment.
This assumption did not consider a Loss Of Power (LOP) coincident with a Loss Of Coolant Accident (LOCA) which would result in loss of the Station Air Compressor. The CAS pressure may not exceed post accident drywell pressures and therefore a bypass leakage path could exist.
Analysis of the Event
The CIS is designed to provide protection against the release of radioactive materials to the l
environment by providing a minimum of one protective barrier between the reactor and the environs under all postulated conditions. This condition would result in faibre of the containment isolation function only in the unlikely event of a Design Basis Accident /DBA) coincident with a single failure and a failure of the EECWS/RBCCWS pressure boundary. In addition, Seismic Category II/I, Quality Group D, RBCCWS piping is designed to maintain structural integrity following a safe shutdown earthquake. Major differences between Quality Group B and Quality Group C EECWS components are in Non-Destructive Examination requirements at the time ofinstallation. From the standpoint of quality of material, manufacture, and installation, the differences are n gligible. Therefore, EECWS and RBCCWS piping would be reasonably expected to maintain structural integrity during and following a DBA. Based on this information, the condition related to the EECWS containment isolation function is considered of minor safety significance.
I
I LICENSEE EVENT REPORT (LER) TEXT CONTINUATION F AC '.RY hAME (1)
DOCAf f NUMUf H (2s LEA NUMBER &
PA0Ev3r
"^^
"f%"
M Fe rmi 2 0 l5 l0 l0l0 l3l4l1 97 0 0 3 0
1 5
l o' l 6 IE AT glia Analysis of the Event (Continued 1 The UFSAR bypass !eakage limit of 4% La assures that the major;q of primary containment leakage is processed by the Standby Gas Treatment System (SGTS) prior to release. With the twelve new bypass leakage paths considered, the UFSAR limit of 4% La was unknowingly exceeded during RFO2 (4.1% La as-found / 4.4% La as-left) and Outage 88-01 (12.5% La as-found /11.1% La as-left). Had these additional bypass leakage paths been recognized at the time, valve maintenance would have been performed as necessary to reduce the as-left leakage rates to within the 4% La limit prior to startup from these outages. The as-found bypass leakage rates for the outages following both RFO2 and Outage 88-01 were less than the 4% La limit. The dose consequences of the most limiting case (i.e.,
12.5% La) have been evaluated and have been determined to be within 10CFR100 guidelines. Based on this information, the conditions related to the twelve previously unrecognized bypass leakage paths and unknowingly exceeding the 4% La bypass leakage limit are considered of minor safety signi6cance.
Corrective Actions
A review of UFSAR Table 6.2-2, Summary Of Primary Containment Penetrations And Associated Isolation Valves, was conducted which veri 6ed that no other primary containment penetrations have design basis denciencies similar to that described in this report.
The EECWS primary containment penetrations have been modined to provide diverse power such that a single failure will not result in the loss of the containment isolation function.
Alami Response Procedures (ARPs) have been revised to provide guidance to Operators to manually isolate the EECWS return header primary containment penetrations in the event that certain accident conditions exist. This will limit the leakage of radioactive material which could pass through the EECWS primary containment penetrations and bypass secondary containment.
A revision to UFSAR Table 6.2-2 has been approved to properly identify the twelve new penetrations as bypass leakage paths. In addition, the twelve penetrations have been added to the bypass leakage program.
- - 4
- - 1 I
LICENSEE EVENT REPORT (LER) TEXT CONTINUATION F AciuT( teMdE gli DOCAE T MAMIER (2)
LER MAMBER se.
PAGE (3)
YEAR M
4 R
Farmi2 0 l5 l0 l0l0 l3l4l1 97 0
1 6
l'l6 0 0 3 TEA 1(11)
' Additional Information:
A.
Failed Components None r
B.
Previous LERs on Similar Problems LER 87-052 documented a design discrepancy involving the Primary Containment Radiation Monitoring System (PCRMS) primary containment penetrations. No previous LERs have documented unrecognized bypass leakage paths or bypass leakage rates in excess of specified limits.
1 i
l i
l 4
l l
1 d
J r
l 1