ML20198H418
| ML20198H418 | |
| Person / Time | |
|---|---|
| Site: | San Onofre  |
| Issue date: | 01/02/1998 |
| From: | Nunn D SOUTHERN CALIFORNIA EDISON CO. |
| To: | NRC OFFICE OF INFORMATION RESOURCES MANAGEMENT (IRM) |
| References | |
| 50-361-97-22, 50-362-97-22, NUDOCS 9801130285 | |
| Download: ML20198H418 (12) | |
Text
.
4 2 %
M)lllHI RN cal H OWM4
+
EDISON ma,-
An LicOW1471h4T'10A4L Company e
January 2,1998 U. S. Nuclear Regulatory Commission Document Control Desk Washington, D.C. 20555
- Gentlemen:
i
Subject:
Docket Nos. 50-361 and 50-362 Maintenance Rule inspection Supplemental Information San Onofre Nuclear Generating Station, Units 2 and 3
Reference:
NRC Maintenance Rule Inspection Nos. 50-361 and 50-362/97-22 During the period December 1 through December 5,1997, the NRC conducted the referenced specialinspection at the San Onofre Nuclear Generating Station, Units 2 and 3. The special inspection reviewed Southern California Edison (SCE) compliance with 10 CFR 50.65 (Maintenance Rule).
At the associated December 5,1997 Exit Interview, the inspectors identified a potential violation (s) for the failure to establish adequate performance criteria under 10 CFR 50.65(a)(1) for: (1) the instrument air system; (2) the containment isolation system; and (3) the po!ar crane. The inspectors also characterized the failure to scopc the turbine building HVAC system and the non-radiological sump and drain system into the Maintenance Rule program as an additionel potential violation.
At the Exit, it was indicated that SCE could provide additional information to the NRC after the inspection regarding the proposed violations and include any additional information regarding enhancements to the program. The purpose of this letter is to provide that supplemental information, which is included in the Enclosure to this letter, 9901130265 990102 PPft ADOCK 00000361 0
PDR 11
. I llill 101ll h1
. r m iu San Ocmente, CA 92074 0128
- ic ggM 714 41L1480 iat 714.%8-INO 1
Document Control Desk January 2.1998 With respect to the proposed violat;ons, as discussed in the Enclosure, SCE:
(1) believes adequate performance criteria was established for the instrument Air and Containment isolation Systems; (2) agrees appropriate polar crane performance criteria was not established until October 1997, however, the polar crane issue was licensee Identified as a result of corrective action from a previous vlotation (50-361/9614-03) where appropriate corrective actions have previously been implemented, and should therefore be categorized a non cited violation; and, (3) believes the turbine HVAC and non-radiological sump and drain systems were properly scoped as being outside the Maintenance Rule.
e SCE would like to assure the Commission that these issues do not represent examples where SCE failed to perform adequate maintenance on systems, structures, and i
components. SCE emphasizes that we have consistently maintained these systems at a level where they have remained capable of performing their intended safety function, if you have any further questions, please contact me.
Sincerely, b
b c(
4 L
v Enclosure cc:
E. W, Merschoff, Regional Administrator, NRC Region IV M. B. Fields, NRr 'roject Manager, San Onofre Units 2 & 3 J. A. Sloan, NRC benior Resident inspector, San Onofre Units 2 & 3
ENCLOSURE This Enclosure provides additional information regarding both the proposed violations identified in NRC Inspection Report Nos. 50-361 and 50 362/97-22 and enhancements to the San Onofre Maintenance Rule program which have been taken subsequent to tho inspection.
A.
PERFORMANCE CRITERIA The NRC Maintenance Rule Inspection team identified a potential violation for falling to establish adequate performance criteria under 10 CFR 50.65(a)(2) for: (1) the instrument air system (IA3); (2) the containment isolation system (CIS); and (3) the polar crane. Each individual example is discusced separately below:
1.
INSTRUMENT AIR FACTS San Onofre did establish speclHe performance criteria for the lAS: one SCRAM per 7000 hours0.081 days <br />1.944 hours <br />0.0116 weeks <br />0.00266 months <br /> critical, or 2 percent Unplanned Capability Loss Factor (UCLF).
ISSUE it was questioned whether the Mtsintenance Rule explicitly required additional performance criteria at the component lovel to monitor the effectiveness of maintenance on the IAS compressors.
RESPONSE
The IAS was placed into the Maintenance Rule program based on: (1) lAS is used to mitigate an accident or transient; (2) lAS is referenced in the EOl's; and, (3) lAS can cause a SCRAM or safety system actuation. While this puts IAS under the Rule, it is important to note the uniqueness of the IAS design as a four train system:
three compressors and a nitrogen backup system.
SCE selected the above plant level performance criteria to monitor the effectiveness of maintenance on the IAS in accordance with both the requirements of NUMARC 93 01, Section 9.3.2, and its safety significance as, defined in the Probabilistic Risk Assessment (PRA),
i NUMARC 93-01, Section 9.3.2 states ' Specific performance criteria are ostablished for all risk significant SSCs and for non-risk significant SSCs that are in l
1 ENCLOSURE standby (not normally operating) mode... Plant level performance criteria are established for all remaining non-risk significant, normally operating SSCs.
However there may be some non-risk significant SSCs whose performance cannot b9 praClically monitored by plant level critoria. Should this occur, otner performance criteria should be established as appropriate."
Because the IAS is a normally operating system with a high degree of redundancy in its source of supply, during the scoping effort, the Expert Panel concluded the IAS was a low safety significant (LSS) SSC. Th!s was based on the PRA, and Expert Panel judgement, because the IAS does not provide any Risk Significant functions.
As listed in PRA Calculation NSG 97-010, " San Onofro Nuclear Generating Sta' ion PRA Analysis Support Maintenance Rulo Sensitivity Study *, dated November 24, 1997, Section 11, the risk increase in Core Damage Frequency (CDF) for the event with all three instrument air compressors out of eurvice is on the order of 2E-9, and Large Early Release Frequency (LERF) shows no increase in risk associated with their loss. This is well within the allowed increase for CDF of 3.1E-6, and LERF of 1.44E-7, as determined in accordance with EPRI TR-105396 'PSA Applications Guide', dated August 1995.
The minimal consequence to the PRA from lAS is contrary to what many other plants experience. This is because most risk significant safety related valves at San Onofre are motor operated, and those safety related valves that are not motor operated either fail in their safety related positions, or utilize a qualified local back-up supply of high pressure nitrogen, independent of the IAS and back-up nitrogen systems. These local back-up nitrogen supplies are scoped with the system they support.
Individual instrument air compressor performance is monitored at San Onofre outside the Maintenance Rule in the State of the Sistem (SOS) Reports which are updated quarterly by the Station Technical Division. The compressor failures referenced above were evaluated as part of this effort, and corrective actions were developed, (including several cesign changes, parts replacements, and enhanced preventive mainte.)ance).
2
I i
ENCLOSURE SUMMe8Y As the IAS is an LSS system, the performance criteria were correctly set in compliance with 10 CFR 50.65 at the plant level in accordance with NEl guidance.
Notwithstanding SCE was in full compliance with the Maintenance Rule, as a minor i
program enhancement SCE has implemented an additional performance criteria of 10 percent unavailability over four quarters for each Instrument Air compressor, 2.
CONTAINMENT ISOLATION San Onofre established specific performance criteria for the CIS to nionitor the containment isolation function. The following is provided to clarify those criteria.
FACTS The peiformance criteria for dispositioning the (ClS) from 10 CFR 50.65(a)(2) to 10 CFR (a)(1)were stated as:
1.
Less than 94 percent cumulative availability of all containment penetrations over a refuelirig cycle, or 2.
Two, or more, repeat Functional Failures (FFs) among similar Critical Components, per refueling cycle, or 3.
Ten, or more, random and unrelated FFs during a refueling cycle.
Failure Criteria An FF of a Containment Prs,etration is determined by the followirg criteria:
1.
Any Appendix J component failure which results in a loss of containment isolation capability such that the total leakage rate exceeds 0.06 percent of containment air weight per day at the calculated peak pressure (Technical Specification limit is 0.10 percent), or; 2.
Failure of any containment isolation valve to close upon demand.
ISSUE 3-4
-yv
,-w.
..e.-,r.~,
,-,..me-:-
~y
.,-o.
ENCLOSURE The concerns raised by the inspection team were that the performance criteria was set too high at ten functional failures (FFs) per monitoring period and that the system should be monitored at the train level as opposed to the system level.
Specifically, the concern was that for a functional failure to occur, it appeared that both valves of the penetration would have to fail.
CLARIFICATION As stated at the exit meeting, either a single occurrence of a failure of any containment isolation valve to close upon demand (stroke) or an Appendix J component failure (for leakage) resulting in the total leakage exceeding 0.06 percent containment air weight would be ceunted as a Functional Failure. For the leakage criteria to be exceeded, both valves of a penetration would have to leak.
However, an individual valve failing to fully close on demand would also be a
. functional failure under the other criteria. It was believed, based on PRA analysis and past testing experience, that the failure of any c.,ontainment isolation valve to close would be a more frequent occurrence, which influenced the selection of the performance criteria as 10 random FFs, combining the two types of failures identified above. Thus we believe that our criteria was set commensurate with safety at 10, when both types of failures are considered.
Any combination of those events ec;ualing or exceeding a total of ten would exceed the criteria for random FFs. For example, a combined total of 9 events of individual valves failing to close on demand and one occurrence of an Appendix J component failure resulting in the total leakage exceeding 0.06 percent containment air weight would cause the CIS to be considered for goal setting. Thus, SCE is monitoring at the component level as the failure of any containment isolation valve to stroke is captured as a functional failure The performance criteria for the total leak rate is intended to capture any gross failures which may cause a loss of the containment isolation function. SCE's expert panel concluded that 0.06 weight percent dry containment air volume was an acceptable performance criterion t'ecause: (1) it was more conservative than the 0.10 weight percent dry containment air volume Technical Specification limit; and, (2) it was generally consistent with the range of values in use by the industry, it was expected that this type of failure would be an infrequent occurrence.
ENCLOSURE Therefore, SCE believes that the performance critoria established for the CIS were in compliance with 10 CFR 50.65 and were sufficient to adequately monibr the effectiveness of maintenance under the Maintenance Ruie.
ENHANGEMENT Notwithstanding that SCE believes it was in full compliance with the Maintenance Rule, as a program enhancement, SCE has revised its performance criteria. Upon further review it was determined that more emphasis could be placed on overall containment leakage. Accordingly, the criteria has been enhanced to include two additional criteria for containment leakage.
The erihanced San Onofre performance criteria for dispositioning the (CIS) from 10 CFR 50.65(a)(2) to (a)(1) are.
1.
Total containment "as found" leakage in excess of 0.03 percent of containment air weight per day at the calculated peak pressure, or 2.
More than 10 percent of Appendix J components tested, that exceed their "as found' administrative leakage limit during one fuel cycle, or (The administrative limit is the value for each Appendix J component that is determined by Station Technical dependent upon past performance, whose aggregate is less than the Technical Specification limit for containment leakage.)
3.
More than 6 percent cumulative unavailability of all containment isolation valves to close upon demand over a refueling cycle, or 4.
Two or more repeat FFs among similar critical components for failing to close upon demand, por refueling cycle, or 5.
Ten or more random ar d unrelated FFs where containment isolation components fail to close upon demand, during a refueling cycle.
Functional Failure Criteria An (FF) of a containment penetration is determined by the following criteria:
5-
.~.
-:,x
~
4 ENCLOSURE-
- A functional failure _is defin. < as any containment isolation valve or -
l
~ component that fails to respond or fully close (i.e., to properly seat) on -
1 demar d or the failure of a critical or suberitical manual valve to l
- completely close.
The performance criterion 1 above for leakage, will now result in the system being -
placed in (a)(1) for a ringle occurrence of exceeding tha 10 CFR 50, Appendix J t
overall "as found" leakage measurements. - In order to trigger this criteria, a -
complete penetration or train failure would need to occur.
Performance criterions 2 through 5 monitor CIS at the component level. During.
plant operat:on, most valves cannot be checked for leak rates, therefore these g
criteria for both availability and reliability of individual containment isolation valves
~
- wera estr.blished.-
~3.-
- POLAR CRANE 7
FFACTS 4
In response 1a Notice of Violation 50-361/96-14-03. 5. ; explicitly a w
-o re-examine our ut,e of plant !avel performance critarik 4 Iow safet*
3Cs.
This review [is] to confirm if non-risk significant functions provided within the Maintenance Rule scope could be practically monitored by appropriate plant level n
_ criteria."
p
{
ISSE
-The performance criteria for the polar crane prior to October 1997,. when the polar crr.no was monitmed as part of Struct.ures, was not sufficient to ensure that it could -
~
perform its mechanical function of lifting heavy loads over safety-related equipment and fuel.
~ RESPONSt: -
As a result of the re-examination program, SCE concluded that specific performance criteria should be implemented for the polar crane and certain other LSS systems. SCE developed and implemented the polar crane criteria in October
- 1997, in ret ponse to this previous citetion.
'i.
a 10-'Pb 4
M=- - - -e.2e m
ret' W -m
- T 7
a-4ew f
-7
-=p.m 9-'
y M
ys y'qy-g-
g
v ENCLOSURE S.UMMAELY b
SCE believes this issue (i.e., use _of plant level performance criteria for low safety significant SSCs) was the subject of previous enforcement action, and since it has been licensee identified and corrected, it should be categorized as a non-cited violation.
B.
TURBINE BUILDING HVAC SYSTEM AND NON-RADIOLOGICAL DRAINS i-ACTS SCE evaluated the turbine building HVAC system and the non-radiological drains, and in accordance with the Maintenance Rule guidance explicitly concluded they-
- should be excluded from the Maintenance Rule program.
ISSUE At the exit interview, it was stated that the term "significant" was not defined in the SCE Maintenance Rule program, and therefore, the inspection team could not determine if the systems called turbine building HVAC and non-radiological drains and sumps had been appropriately excluded.
RESPONSE
The Maintensnce Rule requires SSCs to be included in the A sintenance Rule program if they are referenced in the tacility Emergency Operating Instructions (EOls) protided they:
.. contribute SIGNIFICANTLY to the mitigation function of the system."
SO123-XIV-5.3.1, " Scoping for the Maintenance Rufe", using guidance provided in NUMARC 93-01, Rev. 2, defines the term "significant" as: "The SSC must add value to the mitigation function of the EOl by providing the total or substantial fraction of the total functional ability required to mitigate core damage or radioactive release."
The guidance provided in NUMARC 93-01, Rev. 2, Paragraph 8.2.1.3 states that, ".
.. for a non-safety related SSC to be considered important, it must add significant value to the mitigation function of an EOl by providing the total or a significant fraction of the total functional ability required to rnitigate core damage or radioactive release." _ _ _ _ _
l ENCLOSURE I
Proceedings of the NUMARC Industry Workshops on Implementation of the Maintenance Rule that were held in August 1993, provide guidance on evaluating an SSC's significant contribution in the EDP. On page C-1-11, in quest lon 30 it is stated, "... if no specific criteria has been established. PRA data and expert panel reviews could be considered in the determination of significant contribution."
The PRA determined that these systems were non-risk significant and in addition, the functions credited in the EOls have no impact on CDF or LERF. These systems and the documents justifying their exclusion from scope were reviewed by the Expert Panel at the October 4,1997 meeting. The October 4 meeting minutes specifically state the justification for exclusion of systems from the Maintenance Rule program as "... they are not significant contributors to the EOls, because in some cases these systems were listed in the EOls in trivia! or peripheral functions".
Accordingly, the nor:-radioactive aump system (SUMP) was scoped out of the Rule at the October 17,1907, Expert Panel meeting and the turbine building HVAC (HVAC-TB) was scoped out of the Rule at the October 31,1997, Expert Panel meeting.
The SUMP arid HVAC-TB are non-significant contributors to the EOls because:
HVAC-TB and SUMP are called out in step 22 of SO23-12-4, Steam Generator Tube Rupture (SGTR). This step calls for the turbine building HVAC to be stopped and for the turbine building sumps to be aligned to radwaste. There is no PRA significance f.o these functions.
These actions are also not includec'in the Combustion Engineering Emergency Procedure Guideline (EPG) for SGTR (reference CEN-152 and of SO23-14-4, SGTR Bases and Deviation Justification).
Since the EPG covers all actions necessary to citigate an SGTR and satisfy
) nine safety functions, the vendar also considers these actions as un-significant.
Step 22 related to the SUMP has no impact on offsite release. The discharge of radioactive liquid from the turbine building sumps to the outfall is monitored and can be terminated ay radiation monitor 2(3)RE7821.
2(3)RE7321 and the associated discharge valve to the outfall 2(3)HV5849 are inci'Jded within the scope of the rule in the LSS function MR-RAD-MON-05. Therefore, the action in step 22 is only to direct the sumps to raJwaste since their normal path would be isolated. The worst case consequence of this would be overflowing the turbine building sumps l
l l
ENCI.OSURE and spreading contamination on the floor of the 7' turbine building (the bottom floor).
Step 22 to stop all turbine building HVAC likewise will have negligible impact on offsite release. The design of the SONGS turbine building is important in that with the exception of a few small enclosed rooms (switchgear, lube oil, etc.) the turbine building is open to the atmosphere.
o Because of this design, only the enclosed rooms have exhaust fans end they are supplied solely from outside air. The remaining turbirie building HVAC circulates the air in the turbine building and provides no exhaust function.
f Therefore, the worst case consequence of not performing the action in step 22 would be to circulate contaminated air throughout the turbine building, which would have negligible impact on offsite release.
The purpose of HVAC-TB and SUMP in step 22 is to minimize the post-accident cleanup in the turbine building. This is supported by inclusion cf these identical actions in step 3.4.8.7 (actions to minimize contamination for a primary to secondary laak) of the Abnormal Operating Instruction SO23-13-14, Reactor Coolant Leak.
NEl has indicated that SSCs utilized in the EOls, whose only function is to minimize the spread of contamination within the secondary plant, are not within the scope of the Rule. This is supported by the examp'es of EOP-referenced non-eafety systems that do not provide significant mitigation functions identified in the NUMARC Industry Workshops on implementation of the NRC Maintenance Rule,10 CFR 50.65 (August 1993). These examples include:
Moisture Separator Reheater inlet Isolation Valves - Tube rupture procedure directs closure of these valves to minimize spread of contamination within the secondary plant. This has negligible effect on dose to the public.
Polisher Bypass Valve - This valve is also operated in the tube rupture EOP to minimize contamina* ion in secondary plant and dose rates near condensate polishers.
Tnese examples confirm that NE! does not consider minimizing secondary contamination as a significant mitigation function.
9
. ~ -
ENCLOSURE l
SUMMARY
SCE evaluated the turbine building HVAC syatem and the non-radiological sumps and drains, and in accordance with the Maintenance Rule a'nd NEl/NUMARC guidance concluded they should be excluded from the Maintenance Rule program.
a