ML20127H094
| ML20127H094 | |
| Person / Time | |
|---|---|
| Issue date: | 12/31/1992 |
| From: | Lobel R, Tjader T Office of Nuclear Reactor Regulation |
| To: | |
| References | |
| NUREG-1366, NUDOCS 9301220193 | |
| Download: ML20127H094 (93) | |
Text
_._
, s-WN&
~~~
InTarovements to Technica: Saeci~ica': ions
~
Survei :ance Recuirements U.S. Nuclear Regulatory Commission l
l Office of Nuclear Ituctor llegulation
- 11. label, T.11. Tjader
,,aatauq,
,fe, f
...../
3882 9 M 8 ' 22 2
1366 R PDR i
s AVAILABILITY NOTICE Availabibly of Raforence Materials Cited in NRC Pubhcations Most documents cited in NRC pubhcations will be availabio from one of the following sources:
1.
Tho NRC Pubhc Document Room, 2120 L Stroot. NW,, Lower Levol, Washington, QC 20555 2.
The Superintendent of Documents U.S. Government Printing Offico, P.O. Box 37082.
Washington, DC 20013 7082 3.
The National Technical information Service, Springfield VA 22161 Although the hsting that follows represents the majority of documents cited in NRC pubhca.
tions, it is not intended to bo exhaustivo.
Referenced documents available for laspection and copying for a fee from the NRC Public Document Room includo NRC correspondence and internal NRC memoranda; NRC bulletins, circulars, information noticos, inspection and investigation noticos; bconsoo event reports; vendor reports and correspondence; Commission papers; and applicant and ticonsoo docu-monts and correspondence.
The following documents in the NUREG series are availablo for purchase from the GPO Salos Program; formal NRC staff and contractor reports, NRC-sponsored conferenco proceed-ings, international agreomont reports, grant publications, and NRC booklets and brochures.
Also available are regulatory guidos. NRC regulations in the Code of Federal Regulaflons, and Nuclear Regulatory Commission Issuarcos.
Documents available from the National Technical information Service include NUREG serios reports and technical reports prepared by other Federal agencies and reports prepared by the Atomic Energy Commission, forerunnor agency to the Nuclear Regulatory Commission.
Documents availabio from pubhc and special technical librarios include all open literaturo items, such as books, journal articlos, and transactions. Federal Register notices, Federal and Stato legislation, and congressional reports can usually be obtained from thoso hbratios.
Documents such as theses, dissertations, foreign reports and translations, and non-NRC conferenco procoodings are available for purchase from the organization sponsoring the pubhcation cited.
Single copios of NRC draf t reports are available free, to the extent of supply, upon written request to the Offico of Administration Distribution and Mail Services Section U.S. Nuclear Rogutatory Commission Washington, DC 20555.
Copios of industry codes and standards used in a substantive manner in the NRC regulatory process are maintained at the NRC Library, 7920 Norfolk Avenue, Bothesda, Maryland, for use by the pubhc. Codes and standards are usually copyrighted and may be purchased from the originating organization or, if they are American National Standards, from the American National Standards instituto,1430 Broadway, New York, NY 10018.
l
NUREG-1366 Improvements to Technical Saecifications Surveillance Requirements Manuscript Completed: May 1992 Date Published: December 1992 H.1.nbel, T. R. Tjader Division of Operational Events Assessment Omcc of Nuclear Reactor llegulation U.S. Nucicar Itegulatory Commission Washington, DC 20555 ga a..y,,
.....e
AIISTRACT in August 1983 an NRC task group was formed to investi-(ISIP) was established in December 1984 to provide the gate problems with surveillance testing required by framework for rewriting and improving the Technical Technical Specifications, and to recommend approaches Specifications. As an element of the 'ISIP, all Technical to effect improvements. NURIIG-1024 ("rechnical Specifications surveillance requirements were compre-Specifications-linhancing Safety impact") resulted, and hensively examined as recommended in NURl!G-1024.
it contained recommendations to review the basis for test
'the results of that effort are presented in this reportJlhe frequencies; to ensure that the tests promote safety and study found that while some testing at power is enential do not degrade equiptnent; and to review surveillance to verify equipment and system operability, safety can be tests so that they do not unnecenarily burden personnel, irnproved, equipment degrad. tion decreased, and unnec-essary personnel burden relaxed by reducing the amount The Technict' Speelfications improvement Program of testing at power.
f-4 i
t
.i iii NURI:G-1366
5 i
CONTENTS d
Page Abstract...
iii Abb r e via t io n s............................................................................
ix -
IIxecu t ive S um ma ry.......................................................................
xi l
I n t rod uuon......................................
l.1 l i n c k g r o u n d................................................................... -....... --
1 i
1.2 Scope..........................
2-1.3 Quahtative Safety Assessment of Technical Specifications Surveillance llequirements......,......
4 --
2 Pros and Co ns of Test ing at Power............................................................
6 3 G e n e ra l Fi n o m gs...........................................................................
'8 e
3.1 Magnitude of Surveillance Testing Required by Technicd Specifications........................
8 3
3.2 Preventive Maintenance and Surveillance Testing.........................................
9-3.3 Japanese Approach to Surveillance Testing..............................................
9 3.4 Technical Specifications and the ASMI! Code............................................. -
10 3.5 Scheduling Surveillance Testing.................
12 3.6 Surveillance Testing and Radiation lirposure.............................................
12 3.7 Surveillance Testing and Plan t Transients...................................................
-13 3.8 Surveilla nce Testing and R eliabnity........................................................
-15 3.9 Surveillance Testing and Plant Design 15 3.10 Surveillance Testing and Power Reductions..............................................
16 3.11 Surveillance Testing and I!quipment Wear................................................
16 1
3.12 Survedlance Testing on a Staggered Test llasts............................................
16-4 R ea ct ivity Con tr ol Syst e m s....................................................................
17 4.1 Moderator Tem perature Coef ficient Measurement (PWR)........................... i........
17 4.2 Control Rod Movement Test................
17-4.2.1 Pressu rited Wa ter R eactors.....................................................
17 4.2.2 _ lloiling Wat er R eactors........................................................
18 4.3 Standby I.iquid Control System (llWR)...
...............s
................... ' 21 4.4 Closure Time Testing of Scram Discharge Volume Vent and Drain Valves (llWR)...............
21 4.5 Reactor Scram Testing To Demonstrate Operability of Scram D' charge Volume Vent and -
D rain Val ves (ll W R)..................................................................
22 5 Instru me n tation.................................
4.........................................
24 5.1 Nuclear Instrumentation Surveillance (PWR)..............................................
24_
5.2 Slave R elay Testing (PWR, IlWR)......................................................
25 5.3 Test Intervals for R PS and !!SFAS (PWR, llWR)...........................................
25 5,4 Ilydrogen Monitor Surveillance (PWR, IlWR)............................................
26 5.5-R eactor Trip 11 reaker Testing (PWR) ;.................................................... ' 26 5.6 Power Range Instrument Calibration (PWR)..................... _.......................
27 5.7 Control I!!cment Assembly Calculator Survedlance (Cl3 CPC PWR).........................,..
28 15.8 Incore Detector Surveillance (Cl! and Il&W PWRs)..............
28 5.9 Response Time Testing of isolation Actuation Instrumentation (PWR. IlWR)....................
29 Y
NURl!G-1366
.-.-_~_ - - ~ -.
i Contents f
i
.i i
Pate
\\
5.10 Source Range Monitor and Intermediate Range Monitor Surveillances (I!WR)...................
31 l
5.11 Calibration of Recirculation I' low Transmitters (llWR).......................................
31 5.12 Autoclosure Interlocks: Removal of the LCO From the Technical Specifications (PWR,11WR)....
31 l
5.13 Turbine Overspeed Protection System Testing (PWR 13WR)..................................
32 5.14 Radiat ion Monitors (PW R. IlWR).........................................................
34 l
5.15 Radioactive Oas liffluent Monitor Calibration Standard (PWR,13WR)..........................
35 5.16 Intermediate Range Monitor and Average Power Range Monitor Channel FunctionalTests (llWR)...............................................................................
35 1
6 R e acto r Coola n t Syst e m.....................................................................
36 6.1 R eactor Coolant System isolation Valves (PWR)............................................ -36 6.2 Power (or Pilot)-Operated Relief Valves and lilock Valves (PWR)............................,
.36-6.3 l iigh Point Vent Surveillance Testing (PWR)...............................................
37 6.4 Imw. Temperature Overpressure Protection (PWR).............
38 6.5 Specific Acthity of the Reactor Coolant-100T (PWR IlWR).................................
39:
6.6 Pre ssu rif er l icat e rs (PWR)............................................................
40 7
Em e rgency Core Cooling Syst em...............................................................
42 i
7.1 Surveillance of 110 ton Concentration in the Accumulator / Safety injection Tank / Core flood Tank (PWR)...............................................................................
42 7.2 Verification That !!CCS ljncs Are Full of Water (Contain No Air) (PWR)...................... - 42 7.3 Verification of Proper Valve Lineups of I!CCS and Containment isolation Valves (PWR. IlWR)....
43 7.4 Accumulator Water Level and Pressure Channel Surveillance Requirements (PWR).............. - 43 7.5 Visual Inspection of the Containment Sump (PWR)....................... o.................
44 7.6 Verification of Iloron Concentration in the lloron injection Tank (Westinghouse PWR)...........
44-8 Containment...............................................................................
45 8.1 Con tain me n t S pray Syst em (PW R).....................................................
45 8.2 Containment Purge Supply and lixhaust isolation Valves (PWR)',.............................
45 8.3 lee Cond e n se r Inlet Doors (PW R)....................................................... :46 8.4 Testing Suppression Chamber to Drywell Vacuum lireakers (HWR)...........................
47
(
8.5 liydrogen Recombiner (PWR, !!WR).........
47 l
8.6 Sodium Tetraborate Concentration in Ice Condenser Containment Ice..........................
47 9 Pla n t Sys t e m s..............................................................................
.49 9.1 Auxiliary Feedwater Pump and System Testing (PWR).........................,...........
49
- 9.2 Main Steam ljne Isolation Valve (MSIV) Surveillance Testing................................ 50 -
l 9.2.1 Piessu rized Wat er R eactors...................................................
50 1
9.2.2 lloiling Wa t e r R eactors..................................................,......
51 H
9.3_
Control Room timergency Ventilation System (PWR, !!WR)..................................
51 10 El ect ric Powe r..............................................................,,,.......
53 10.1 Emergency Diesel Generator Surveillance Requirements (PWR, llWR).......................,.
53 10.2 llattery Surveillance Requirements (PWR, HWR)........................................,..
57
vi l
l
.. ~. ~ ~.
._-.-,.._.~._m.m-
Contents l' age 1 1 R e f u e l in g..................................................................................
60 1
12 S pecial T e st IIxce ptio n s................................................................... _.
61 13 R ad ioact i ve liffl u e n t s........................................................................
62 14 Co n cl u si o n s............................................................................
63 71
! $ liibliography.....................................
Appendix Routine Inspections (Surveillance Testing) at Japaraese Power Plants.........................
77 Figure 3.1 PWR trips attributed to surveillance testing. January 1986-July 1988..............................
14 7
Tables 3.1 Number of surveillance tests required for RPS and liSFAS instrumentation currently, and number proposed in Westinghouse Owners Group topical report........................................
8 3.2 I n service t est quan tities for pu m ps..........................................................
11-3.3 ASMli Code.Section XI requireme nts for valves..............................................
11 3.4 Percentage of annual collective dose at LWRs by work function.................................
13~
4.1 M echanically stuck PWR control rods......................................................
19 1
8.1 Containment spray nonle problems at PWRs.................................................
45 9.1 Inservice testing required by Standard Technical Specifications and ASMll Code....................
49 10.1 Alt ernat e t esting requ ir ements..............................................................
55 10.2 Comparison of requirements of !!!!!!! Standard 450-1980 with requirements of Westinghouse STS.....
58 10.3 llat tery failure events r eported in LI!Rs.......................................................59
- 14.1 Summary of recommended ct.anges to surycillance requirements.................................. -64 I
I i
i ll' l
l l
l i,
vii NURIIG-1366 '
Y r k
W4"'
t gr g waM.,e eq qeq_.
4 q
g,.q,q.,,w.g_.,,,g,.,,
gwp.p.%-,4.y.p-w,- _. _.,gy..
p...
g,
,su__,
g ewg.g, y
y,
,>-y--%.v.g
.g gy
All11REVIATIONS ACI autoclosure interlock ADS automatic depressurization system AliOD Office for Analysis and Evaluation of Operational Data AFW auxiliary feedwater AIF Atomic Industrial Forum ALARA as low as reasonably achiesable AOT allowed outage time AOV air-operated valve ASilRAE American Society of Heating, Refrigeration and Air Conditioning Engineers ASI axial shape index ASMii American Society of Mechanical Engineers lilt boron injection tank ll&W liabcak & Wilcox IlWR boiling water reactor CE Combustion Engineering CEA control element assembly CliAC control element assembly calculator CFR Code of Federal Regulations CIV containment isolation valve COLSS core operating limit supervisory system CP construction permit CPC core protection calculator CRD control rod drive CRDM control rod drive mechanism CROR Committee to Review Generic Requirements DNilR departure from nucleate boiling ratio ECCS emergency core cooling system IIDG emergency diesel generator EFPD effective full power day EOC end of the operating cycle EPRI Electric Power Research Institute ESF engineered safety feature ESFAS eng;neered safety features actuation system FSAR final safety analysis report Fo ratio of maximum core power to average core power GE General Electric Co.
G1 generic issue OL generic letter GSER generic safety evaluation report j
HPCI high-pressure coolant injection llPSI high pressure safety injection l
IEEE Institute of Electrical and Electronics Engineers l
!!.RT integrated leak rate test INPO Institute of Nuclear Power Operations IRM intermediate range monitor IST inservice testing ix NURiiG-1366
Abtneviations I.CO limiting condition for operation 1JiR licensee event report 1.OCA loss-of coolant accident 1,00P loss of of fsite power 11C1 low-pressure coolant injection IFD local power density 1ll'OP low-temperature overpressure protection h10V motor-operated valve MSIV main steam line isolation valve hfl~C moderator temperature coefficient N!!S National Ilureau of Standards NPAR Nuclear Plant Aging Research (Program)
NPRDS Nuclear Plant Reliability Data System NRC Nuclear Regulatory Commission Nlth Office of Nuclear Reactor Regulation NSAC Nuclear Safety Analysis Center NSSS nuclear steam supply system PG&li Pacific Gas & lilectric PNI.
Pacific Northwest Iaboratory PORV power (or pilot)-operated relief valve PPS plant protection system PRA probabilistic risk assessment PWR pressurized water reactor RCC ral cluster control RCIC reactor core isolation cooling RCM reliability centered maintenance RCS reactor coolant system Ri!S Office of Nuclear Regulatory Research RG regulatory guide RiiR residual heat removal RPS reactor protection system RWCU reactor water cleanup RWST refueling water storage tank SAIC Science Applications International Corporation Sill.C standby liquid control system SCSS Sequence Coding and Search System SDV scram discharge volume SR surve; lance requirement SRM source range monitor Sil surveillance test interval STS Standard Technical Specifications SW service water 131P Technical Specifications Improvement Program UPS uninterruptible power supply USl unresolved safety issue VCr volume control tank Vl!PCO
. Virginia 111ectric & Power Company NURiiG-13t4 x
l
EXECUTIVE
SUMMARY
1he Technical Specifications Improvement Program (4) 1he surveillance places an unnecessary burden on (FSIP) was established by the Director of the Office of plant personnel because the time required is not Nuclear Reactor Regulation (NRR) on December 21 justified by the safety significance of the surveil.
1984 to cornpletely rewrite and streamline the Technical lance.
Specifications as well as to make line item improvements to existing'lechnicalSpecifications. Asa part of thiswork, in addition to applying these four criteria, the NRC staff many requirements are being relocated from the Techni.
performed a qualitative safety assessment on the Techni-cal Specifications to licensee-controlled documents, and cal Specifications surveillance requirements of the West-owners groups have submitted topical reports proposing ing house Standard Technical Specifications, Version 4 A, changes to suncillance test intervals and allowed outage and the llatch Unit 2 Technical Specifications.1hc im.
times on the reactor protection system and enginected plementation of the recommendations of this report as
',afety features inctuation system.1hc Nuclear Regulatory line-item improvements is consistent with the Commis.
Commission (NRC) staff has reviewed and approved sion policy on Technical Specifications improvements, these reports.
The policy endorses a program of short term improve.
ments to both the scope and substance of the existing To supplement this effort, the NRC staff performed a Technical Specifications (Commission's " Interim Policy comprehensive examination of all Tecimical Specifica.
Statement on Technical Specifications Improvements, tions surveillance requirements (SRs)in order to identify dated February 9,196h NRC,1987).
those that should be improved. The evaluation of the Five reactor sites were visited in 1988 in order to discuss safety benefit of changes to Technical Specifications SRs surveillance requirements with the people who plan, involves the consideration of the purpose of the SR (how manage, and perform these surveillances. 'the visits were a change affects safety, including the reduction of chal-productive and many of the comments received have been lenges to plant systems), the effect that the performance incorporated into this study.
of theSR hasonpersonnel(theexposureof pers(mnelto radiation sources and the burden on personnel re.
In addition, the NRC staff reviewed the dockets of several sources), and the effect that the performance of the SR reactors seeking plant rpecific Teclmicd Specifications has on plant equi. ment (equipment wear or degrada.
changes related to surveillance requirements that have t
tion).1hc sesults of that work are given in this report.
generic applicability.
The study recognized that testing is important to periodi.
Operational data from licensee event report (LER) cally verify that systems, structures, and components are searches, Nuclear Plant Reliability Data System avadable to perform their safety functions. Testing is
@PRDS) scarches, and other sources were relied on especially entical to reveal degradation and failures that heavily to assess the impact of Techni.al Specificatio~
occur while equipment is in standby mode. 'the study did find, however, that although testing at power is essential suncillance requirements on plant operation, i
1 to verify equipment and system operability, safety can be The Nuclear Plant Aging Research Program of the Office improved by reducing the amount of testing at power. Of of Nuc! car Regulatory Research (RES) proved a valuable the many existing surveillance requirements that were source of information on component reliability and types reviewed, only a small fraction, contained herein, are of degrada' ion.
recommended to be performed at longer surveillance intervals.
It was found that equipment failures and personnel errors during several types of surveillance tests cause a signific nt number of reactor trips. Reactet protection lhe staff used four criteria, derived from NUREG-1024 recommendations, to screen the surveillance require-syskm ms ng, tMme vake testmg, main skam mMatjon valve kss g,. M ear instrunwntadon - testing, ments. The criteria are:
engineered safety features (ESP) logic, and reactor trip breaker testing were all significant contributors to teactor (1) 'the surveillance could Icad to a plam transient, trips. Other TSIP work has decreased the frequency of reactor protection system testing and will rehicate the (2) The surveillance results in unnecessary wear to requirement'~ from the Technical Specifications for equipment.
turbm, e valve testing so tinat it can be changed to a rnore reasonable frequency on a design-specific basis without prior staff approval. Ilowever, turbine valve testing is l
(3) 1he surveillance results in radiation exposure to such a significant contnbutor to reactor trips that this l
plant personnel which is not justified by the safety report recommends changing the surveillance interva!
l significance of the surveillance.
from weekly and monthly (as presently rcquired) to xi NUREG-13(4 4
1 lixecutive Summary quarteily (if the tur bmc s endor agrees), rather than wait-dose. Tests that require containment entries while the ing for the cornpletion of toe new Standard Technical reactor is in operation (e.g., containment purge and ex-Specifications currently being developed.
haust isolation valve leak testing) cause significant doses.
Walkdowns of systems to check valve alignments and in addition to causing reactor trips, testing results in many snubber operabihty were also found to be significant con-spurious isolations of the control room, fuel handling tributors to radiation dose. IJcensees appear to have buildmg. autihary building, and containtnent ventilation.
taken the steps within their power to limit dose from j
inadverti nt emergency diesel generator starts are rela-testing.
tively common results of suncillance testing; actuations and isolations of standby safety equiprnent occasionally
'lhe report recommends several changes to testing to occur. It is recognited that difficulties in performing sur-further reduce exposmg personnel to radiation. Included veillance tests do not always represent a problem with the in these recommended changes are reducing the inspec-sunciliance requirement, and that some problems can tions necessary for assuring that there is no loose debris in arise from design deficiencies in the plant and its systems, the containment and reducing _ the frequency of Sometimes designs do not have a built.in capability for measuring the contents in the waste gas tanks (s). It ap-performing tests, or do not factor in test optimization and pears that, in general, it is not possible to significantly efficieng.
reduce radiation exposure that may result from surveil.
i lance testing since a balance must be maintained between Wear on equipment is also a significant concern; some the need to perform certain surveillances and the radia-instrument parts (such as connector pins and plugs) cxpe-tion dose incurred.
=-
rience wear from the amount of plugging and unplugging required for testing. Auxiliary fcedwater pumps were One utility exceutive has described the number of surveil.
found to be subjected to wear because of the small recir-lance tests required as " overwhelming." "Ihe number is culation lines used during testing.
large and the utility resources that trust be dedicaed to surveillance testirig are large. In an 18 month cycle, be-limergency diesel generators are subjected to an exces-tween 15,000 and 26,000 surveillances are typically re -
sive amount of testing, especially those diesels in plants quired (without counting simple channel checks). A com-with older Technical Specifications. A problem with one ment heard often durir.g plant visits conducted as part of diesti can result in testing the other(s)every eight hours.
this effort was that equipment was tested w'.dch never A problem with an emergency system other than a dicsci failed (except, perhaps, because of the testing). II:eause 3
generator can result in repeated testing of the dicsci of the large amount of testing and the fact that it may be generators. A problem with the support system to one greater than necessary on some systems, the application diesel generator can mandate that the diesel be declared of reliability methods to Technical Specifications surveil-inoperable and the other diesel or dicscis be tested, lance testing would result in a better allocation of utility resources to those systems and components which experi-
'lhe NRC staff recommends that the nutnber of diesel ence the most problems.
generator tests be greatly curtailed. Only when a valid concern (e.g., potential for common mode failure) is Although this effort concentrated on Technical Specifica-posed almut the availability of the other diesel or diesels tions surveillance testing, a major conclusion of this work when one dicsci generatot is inoperable, shouid the oper-is that preventive maintenance programs must be im-able diesel or dicscis be tested-and then, only one.
proved. A review of licensec event reports and other data shows that many of the failure? found from testing are due A significant contributor to the stress put on a dicsci to dirt or impurities in fluid systems, bent or broken parts, j
generator as a result of testing is the requirement that the loose parts, etc..which should have been corrected before -
diesel generator quickly come up to speed and full load.
they resulted in failure Surveillance testing can only lhe requirement for fast starting and loading of a diesel identify that a piece of equipment is in an inoperable generator comes from the assumptions of the analysis condition so that the time it is inoperable can be limited; of the.large-break loss-of. coolant accident ~(1.OCA)<
preventive maintenance, however, can limit the number Newer, mor e rc<distic assumptions in the I.OCA enalysis, of failures that occur. In this way, preventive maintanance supported by experimental data, show that slower diesel can make a greater contribution to reactor safety than is.
generator starting and loading times would be acceptable, being tr.de by surveillance testing.
since the LOCA criteria would still be met.
The combination of reliability concepts and preventive Radiation exposure to perumnel as a result of surveil.
maintenance in a reliability-centered maintenance pro-lance testing can cause up to approximately 20% of the gram, together with testing based on reliability character-total dose incurred at a site. The biggest contributor to istics of the system or component, would be an effective
. incurred dose is mainterumcc, not testing; however, some method to reduce system / equipment unavailabilities and surveillances do result in sigmficant incurred : stiation would represent better use of existing restmrces.
NUREO-1366 xii -
L
~ _-
_ ~ _ _ _ _ _ _. _. _ _ _ _. _ _
l 1 INTRODUCTION
-1.1 Ikickground ria to completely rewrite and streamline the existing Standard Technical Specifications.
In August 1983, the Deputy fixecutive Director for Re-gional Operations and Generic Requirements formed a (2) In addition to developing a new set of Standard Nuclear Regulatory Commission (NRC) task group to Technical Specifications as identified in recommen-investigate the scope and nature of problems with surveil-dation 1 above, a parallel program of line item im-lance testing required by Technical Specifications and to provements (in both scope and substance) to existing develop alternative approaches that would provide better Technical Specifications should be initiated.
assurance that surveillance testing does not adversely la accordance with the Commission.s
- Interim Poh.ey affect safety 't he staff issued NURI!G-1024. "I cchnical Statement on 'lechnical Specifications Improvement,"
Specifications-Enhancing the Safety impact," its re-dated February 9,1987 (NRC,1987), cach limiting conds-sponse to the Deputy lixecutive Director's request; that report contained five recommendations. 'the three acc-tyn for opetation (LCO)in each of the existing Standard 4
.Icchm, cal S,pecifications (FFS) was evaluated by the re-ommendations concerning surveillance requirements are Spective nuclear steam supply systern (NSSS) owners cited below as stated in NURiiG-1024.
group and by the NRC staff for retention m the new STS -
ot for relocation to a licensee-controlled document, As a Hecommendation 1 result of this review,35% to 45% of the LCOs ard their associated suiveillance requirements were identified for The testingfrequenctes in the l'echnical Specifications should tcmoval frotn the STS.
be rertcwed to assure that they are ad quately supported on a technical basis and that risk to the public is minimized.
l.icensees may m.ake changes to these sutveillance re.
quirements in accordance with plant-specific necds and Recommendation 2 cvaluations without prior NRC appnwal, provided the controls applicable to the document they are rclocated to
' The required suntillance tests should be reviewed to assure are followed (e.g.,10 CFR 50.59 for the requirements that important safety equipment is not degraded as a result of relocated to final saiety analysis teports (FSARs)). 'thus, testing and that such tests are conducted m a safe manner and for a significant numher ofIhe surveil'ance requirements -
in the appropriate plant operational mode to ensure that risk in current Technical Specifications, the licensec will be to the public is minimized.
able to climinate unnecessary tests or inspections. A let-ter from ' thomas E. Murley (NI C) to the chairperson of cach of the owners groups: May 9,1988 (c.g., a letter from Hecommendation 4 Thornas E, Murley, USNRC to Walter S. Wilgus, Chair-7he surwillance test requirements should be reviewed to as-man, The 11&W Owners Group) known as the " split rc-sure that they do not consumeplant personnel time unneces.
port," lists cach tcchnical spccification which is to bc sarily or result in undue radiation e.tposure to plantpersonnel rctained and cach that may be relocatcd for each typc oi without a commensurate safety benefit in terms ofminimi:ing rcactor, grouped by reactor vendor's design (i.e., Wcstin.
ghouse PWR, Combustion Engineering PWR, !!ahcock public risk.
& Wilcox PWR, and General litectric HWR).
On December 21,1984, the Director of the NRC Office After deciding which LCOs would remain in the new of Nuclear Reactor Regulation (NRR) established the Standard Technical Specifications, the owners groups hhnical Spectfications improvement Program (TSIP) commenced to rewrite each of these LCOs along with its
- to reconsidct Ihe entire subr. si techm, cal specifications associated surveillance requirements in an improved for. -
1 -
and to make recommendNons for improvement. Ihere mat. 'lhis work was completed in early 1990. Some pro-was close coordinatic.. octween this project and a similar posed changes to the new Standard Technical Specificac e
industry cffort sponsored by the Atomic Industrial Forum tions, including changes to surveillance requirements,
_ (AIF). A Commission paper was written (SECW86-310) were evaluated utilizing risk-based approaches (in
. and the Commission was briefed on the recommenda-i-
SAIC 90/1394) to ensure that the impact of each change -
tions of the project.There were two major recommenda' on plant core melt frequency or risk is acceptabic.-
tions.
(
In addition to restructuring the entire StandardTechnical L
(1) 'Ihc NRC should adopt the criteria for defining the Specifications as -discussed above, separate, parallel i
scope of the Technical Specifications proposed in efforts have been taking place on a specification-by-the AIF and TSIP reports.The NRC and each of the -
specification (line item) basis to focus on improvements industry owners groups should then use those crite-to specific technical requirements. Among other things, 1.
_.a,_..__._.____.._=_
,_ u._
i
] Introduchon such patattel efforts deal with changes to both surved.
through individual plant conversiom to the new Standard law e mitnals and allowed outage tunes.
Iechrucal Specifications or through individual licensee i
amendrnents. A gnetic letter rnay be prepared to address i
Asa rewit of probabihstic nsk an.dyses assmated with those surveillances winch are identihed as causing a sig-the effort to improve line iterns the NitC staff is review.
mficant number of reactor trips, to allow changes to be my, and m mme case', has aircady found acceptable, the made before individual plants convert to the new Stan-1 increaving of mme surveillance intervals for the reactor dard Technical Specificationt 'the actual implementa-j protectmn sptem and engmected saf ety features actua.
tion procew and f.chedule for these types of changes will tion systerns irmn monthly to quarterly and, in be based on the most cost-effective use of avadable NitC some cases, to stmtannually (Gli rep >rts Nii!W staf f icmurces recogniring that, although important, they 30X5). NI;l C-30851 l Supplements I and 2b rnay not have the same safety significance as other pro-NiiDC-30936, and NF.I)C-31677P-N; Cl! report posed changes.
- CI:N-327; Westinghouse report WCAP-10271 and Sup-plements 1 and 2; and ll&W report ilAW-10167 lall approved [ Sunilmly, inettases in allowed outage tunes 1.2 Scope have been found acceptable.
In this study, the recommendations of NUlti G-1024 in addition to these pumiams the Comminion tuned a were incorporated and expanded, the guidance in the stati requirements memhrandum on I ebruary 25., pigg "spht report" w. s ustd. and, with a few exceptions, only l
(SF.CY 87-314), requesting the NitC staff to:
those !.echons m ~he Standard Technical Specineations that are to be retamed in the new STS were evaluated.
. investigate the prm and cons of continuing to require the survedlance and testing of equip-
'lhis report only addresses Ihe subject of !.urveillance ment while the plant is at power. 'lhe staf f requirements as found in the Technical Specifteations.
should assure that NI(C does not requac un.
'lhe NI(C staff did not study, and this report does not necessary tests or inyections that result in addr ess such other aspects of the Technical Specifications equiprnent disassembly or unnecessary wear.
as allow ed outage times or r equired action w hen a limiting Stall should mform the Commissmn of the condition for operation is not met.
bases of present requirements and any pro-posed mahfications to present Technical
'this report is structured in a manner that parallels the Specihcation requirements.
structure of the StandardTecknicalSpecifications. A see-tinn has been included to specificdly address the Com-In response to the Commiuion's request, the staff imti.
rnission's question about the advantages and disadvan-ated an addihonal shott term cffort as part of the TSip to tages of testing at power. Next, a section follows oa gen-examine the Technical Specifications requirements w hich cral fmdmgs of this study, and then sections correspond-result in testing equipmer3t at power. 'lhis effmt is the ing to each of the major sections of the S'IS appear.
suhicct of this seport and is intended to respmd to the lleactivity Control Systems, Instrumentation, lleactor Comnustion and Oa the recornmendations made in Coolant System, limergency Core Cooling System, Con-NUldE1024 concerning surveillance requirementt tainment, Plant Systems, lilectric Power, Itefueling Op, erations, Special Test fixceptions, and Itadioactive lif Du-On June 20, 1958. the Commiuton was briefed on the ents, liach surveillance requirement in the Techniad simus of the ISIP; subsequently, the Commission re.
Specifications for which a change is recommended or quested that a paper be prepared on the staff's effort to considered is discuued in its appropriate section. Finally, l
reduce testing and survedlance requirements during overall conclusions and references at: presented.
p4wer operauon of a nuclear power plant (tec Clulk, 1968). 'lhe requested paper was to dtscurs the testing and
'lhe onginal goal was to study only surveillance testing at -
survedlances which would no longer be required during power that is required by the Technical Specifications.
l-p>wer operation. The paper, which was based on this
'lhis was later changW to encompass any surveillance l-report, w as inued on October 26,1988. Neither the Com-required by the Technical Specifications that rnet one or mission paper not this sepon provides the exact language more of four criteria selected by this study for screening for changes prop > sod to the wrvedlance requirements of surveillance requuemems to determine which surveil-j the Technical Speci6cationt it is expexied that others, lance requirements should be considered for change, using the recom mendations in this report, wdl rew rite the These four enteria are:
surveillance requirements for the Technical Specifica-tions (1) 'Ihe survedlance places an unnecessary burden on plant personnel because the time required is not
'the actual implementation of the approved changes will jusufied by the safety signifiamcc of the surveil.
be mtegratrd with the implementation of the TSIP lance.
j NUitIE1366 2
l
1 Introduction 1
(2) The survrillance could lead to a plant transient.
of these had generic application. 'lhe proposed changes to the Technical Spec fications of the following PWRs (3) 'Ihc surveillance results in unnecessary wear to were reviewed: Zion Nuclear Plant, Unitz 1 and 2; St.
equipment.
1.ucie Plant, Units 1 and 2: Prairic Island Nuclear Gener-ating Plartt. Units 1 and 2: North Anna Power Station, (4) 'the surveillance results in radiation exposure to Units 1 and 2; McGuire Nuclear Station, Units 1 and 2; plant personnel which is not justified by the safety and Millstone Nuclear Power Station, Unit 2/ihe follow-significance of the requirement.
ing ilWR dockets were also reviewed: laSalle County Shtion. Units 1 and 2: llatch Nuclear Plant, Units 1 and in addition to these criteria, qualitative risk was assessed 2; and Vermont Yankee Nuclear Power Station.
for the surveillance sequirements for the Westinghouse Standard Techracal Specifications, Version 4 A, as an ex-This study also made use of the work done as part of the ample of pressunited water reactor (PWR) Technical NRC Nuclear Plant Aging 1(esearch (NPAR) Program.
Specifications, and for the !!atch Unit 2 Technical Speci-(NURl!G-1]44). 'the reports on various systems and fications as an example of boiling water reactor (IlWR) components prepared under this program gave insight -
Technical Specifications. 'this part of the study is dis-into the ratc of failurc of specific systems and components cussed in more detail later in this section.
and also into the cause of the failures. This information was used to assess whether more testing is being done Within the resource and r,chedule limitations of this than could be justified on the basis of the failure rate of study, it was not possible to examine the identified items equipment.
in sufficient detail to make a recommendation in every case. As a result, sorne surveillance requirements are This effort emcentrated on tevis'ng surveillance require, identified as needing snore study before a conclusion can ments that may be detrimental to reactor safety, person.
be reached about whether a change is warranted, nel safety, or plant operation. It did not attempt to sys-tematically examine possible nonc<mservative surveil-In some cases, the NRC staff recommends that surveil-lance requirements.although a few such arcas were iden-lance intervals be extended. In some of these cases, ex-tified and are documented in this report It also recog-tending a surveillance interval will reduce risk. In other nizes that the majority of surveillance requirements are cases, the overall benefits of an extended surveillance essential to verify equipment and system operability and j
int erval offsets any negligible effect on safety. Since this is ensure safety. As a consequence, only a small fraction of a generic report, the recommendation for an extension those surveillance intervals reviewed were recommended '
could be justified on a generic basis. The basis for each to be performed at longer surveillance intervals.
recommendation is given. It may be possible that plant-specific or equipmenespecific surveillance test intervals Although every Technical Specifications surveillance re-could be even longerflhis must be decided separately for quirement for several sets of Technical Specifications was each plant, examined for this study, not all are discussed individually in this report. Only those identified from the sources As part of this study, the NRC staff visited five nuclear discussed above for which a change was considered are power plants to obtain information frorn reactor opera-discussed. A change was not recommended in even case, tions, maintenance, engineering, chemistry, planning, but where the recommendation is made to retain the and testing personnel on which Technical Specifications existing requirement or to study the issue further, the surveillance requirements meet one or more of the four reasons are given.
criteria used for the studyJlhe sites visited were Crystal River Nuclear Plant, Unit 3 (July 7,1988), San Onofre Technical Specifications are of utmost importance to the N uclear Generating Station, Units 1,2, and 3 (J aly 20 and operation os a nuclear power plant.'they ere the NRC's ~
l 21, 1988), Catawtu Nuclear Station, Units 1 and 2_
statement establishing the limits within which a nuclear (August 3,1988), North Anna Power Station, Units 1 and power plant must be operated in order to enhance safety 2 (August 10,1988), and laSalle County Station, Units 1.
and reduce risk to the public, and as such, they are taken and 2 (August 22 and 23,1988),
most seriously by leth the NRC and the licensees who' operate the nuclear pc cer plants. When a " limiting con-l Comments requested and received from the five NRC dition for operation" (LCO) is ecceded, the licensee is I
regions are incorporated in this report.
required to follow the remedial action stipulated in the Technical Specifications '(which may involve shutting in ano6er effort to determine which surveillance re-down the reactor) until the limiting conditions are met.
quirements in.the Technical Specifications might meet one or more of the criteria used in the study, the dockets it is, therefore, extremely important that they state their of several PWRs and llWRs were reviewed to study the
- requirements clearly (as well as the rewn for making the changes proposed by licensees to determine whether any requirement), that the requirement is really capable of 3
NURl!G-1366 i_
L
______m 1 Introduction being me t, and that the requirement really contnbutes in the compcment to be tested, level of redundancy provided a significant way to reactor safety, in the plant design for the component under considera-tion, and whether the function provided by the compo-Although this study, because of resource and schedule nent can be recovered given component failure. A Sener-limitations, could not be complete in the sense that it ally negative response implies that the component is im-addresses all of the potential weaknesses in Technical portant to plant risk and thus, from the point of view of Specifications surveillance requirements,it should con.
direct safety impact, should be considered for more fre-tribute significantly to the goal of making surveillance quent testing.
requirements in Technical Specifications more relevant and meaningful. Ilased on the overall assessment,includ-Typical direct safety impact questions
- ing consideration of the specific factors discussed in each section, it has been determined that the resulting recom-(a) is the component or system providing a relatively_
mendations pose no undue risk and provide significant unimportant function?
q benefits by achieving a more optimal overall system teli.
(b) is there a high level of redundancy?
ability and availabihty through a greater balance in the (c) Can the function affected by equipment be recov-factors involved (and thereby frequently improving cred?
" "O (2) - Indirect Safety impact 1.3 Qualitative Safety Assessment of ne objective of this criterion is to assess the negative Teclinical S iecifications imp a f twquent testing of compnents.ne types of I
questions asked in considering this criterion are listed Surve.llance Requirenlents below, nese questions are directed toward understand.
i ing whether testing performed on the specific compo.
Science Applications International Corporation (SAIC),
un nmn mynmawapmm bakPWon, under contract to the NI(C, developed a procedure for this study to evaluate surveillance requirements of Tech-u the, component testing could result in inad-an vertent plant trip or disabling of anothe component.
nical Specifications based on a qualitative assessment of A am, a Emuah mgak msynn imphs s ening STis (more frequent testing) and a generally po@s E
safety. SAIC roIused six review criteria for evaluatin8 P
the surveillance test interval (STI):
sponse implies lengthening Kils (less frequent testing).
Typical indlnct safety hnpact quesdons*
t c s ety i p ct (3) reliability (a) Does testing require aligning a portion of a safety or -
4 (4) occupational crposure support system away from the safe position?
(5) operator burden
. (6) NRC burden (b) Can testing cause a reactor trip?
Dese six criteria were chosen because of their direct (c) is a safety or support system also partially disabled impact on decisions associated with frequency of Fl'Is.
by the test and trip with no automatic return?
Each of these criteria represents a specific concem about the frequency of Sils, nese criteria address six major (d) Can testing result in Event V or other loss-of-cool-concerns, but consideration of these six criteria is not ant accident?
intended to constitute a complete analysis of any STL in particular, surveillance specific caigineering design con-(3) Reliability cerns are not addressed. For example, these criteria may b'& Midh im WWMe J
mdicate that a monthly chemical analysis mterval can be extended, but engineering considerations, such as cherni-the standby, normally operating, or cyclic components, cal depletion at a specific plant, may dictate frequent and to determme the dominant cause of the component testing. In any case, engineering considerations, as well as failure.,the types of questtons asked in considering this operating experience obtained from several data bases, criterion are listed below, in general, if it,is determined were used in making the fmal recommendations.
that a component is normally operatmg, is cycled, or is 1
contmucusly monitored, no additional testing is recom-mended for the component based on this criterion. Some (1) Direct Safety impact testing may be desirable to detect performance degrada-ne objective of this crbrion is to understand the impor, tion that otherwise would not be detected. nc remaining tance of the component to plant safety, ne types of questions on standby components are directed toward questions asked in considering this criterion are listed
- A generany negative response implies shonening test intervals; s gener-below. Rese include the importance to plant safety of any gaitne reysw miphes lengthening test miervals.
1 Introduction understanding what dominates the cornponent failure. If (b) Can there be an accidental dose to personnel from equipment wcarout and demand stress dominates a com-testing?
Finent's failure, less irequent testing is recommended, if standby stress is the dominating factor in a component's (5) Operator llurden failure, then more frequent testing would be recom-rnended.
'the objective of this criterion is to assess the Icvel of burden the test places on the operating staff.'lhe types of Typical reliability questions
- questions asked in considering operator burden are listed below. Similar to the occupational exposure critcrion, this (a)' is the component in a normal operational statc?
criterion can either lead to a recommendation of no -
change in Sil if there is no operator burden, or a recom.
(b) is the cmnponent c4mtinuously monitored and mendation to lengthening the STI if there is a burden on alarmed?
the operating staff.
(c) llesides demand testing, are there other frequent operability demands on the cornp(ment?
Typical operator hurden questions *
(d) is there likely to be test caused wearot.t?
(a) is the operator burden from testing characterized as high?
(c) is there likely to be test caused renewal (i.e., the flushing of a filter or geling of a battery)?
(b) 1)ocs testing frequently result in a repair action that threatens an allawed outage time (A(JI*)?
(f) What type of failure and repair causes dominates?
(6) NltC llurden (g) s e ratio of repair actions to catastrophic fahares
,Ihe objective of this criterion is to assess the level of burden on the N1(C staff.*lhe types of questions asked in c4msidering NI(C burden are listed below. Similar to the (h) If the test aligns a pmtion of the system away from mpational capome criterion, this criterion can either
' the safe position with no automatic return,is the test lead to a recommendau,on of no change to an Fil if there unavailability contribution large when compared to is no NitC burden, or a recommendation to lengthen the the contribution from standby stress catastrophic KIl if there is burden on the NRC staff.
faRutes?
(4) Occupational Esposure Typical NRC burden questions *
'the objective of this criterion is to understand the poten.
(a) is burden from current Technical Specifications tial for exposing plant personnel to radioactivity regularly high?
or accidentally as a result of conducting a specific test.
l
. Ihe types of questions asked in considering this criterion (b) DoTechnical Specification testing requirements re.
are listed below. If the surveillance test under considera, sult in frequent requests for extension?
tion does not result in regular or accidental exposure to plant personnel, no change to Kils would be recom-Application of the Proposed Criteria mended. If the surycillance test results in any exposure to plant personnel, longer Sils based on this criterion would it is important to note that the results of considering the be considered.
individual criterion must be considered together. -The final recommendation is based on the reviewer's judg-ment which considers the input from all of the criteria.
Typical occupational exposure questions *
(a) Is there regularly a dose to site personnel from test.
'this procedure was used for studying the surveillance ing the component?
requirements in the Westinghouse Vert. ion 4 A Standard Technical Specifications and the flatch Unit 2 Technical -
' A u ner any ne gauve rnpinw imphenhonening iest inienalo gener.
Specifications as exampics of PWR and IlWR Technical -
alfy swwve int inw implMensihening test uitervalt Specifications, respectively.
I NURilG-1366
- ~_
.i 2 PROS AND CONS OF TESTING AT POWER As is discuned in the miroduction (Section 1), the Com-a signif cant transient. A reactor is designed for a given mission specifically directed the staff to consider the ad-number of trips from full power. 'lhe number of trips is vantages and diudvantages (pros and cons) of continuing usually in the range of 200-250. Ilowever, a trip requires to require the surveillance and testing of equipment at systems to function so as to stop the nuclear reaction and power. Surveillance testing is essential to verify that remove heat to cool the core. A malfunction in one of equipment and systems are operab!c and that they per-these systems could turn a routine trip into a rnore serious forrn their safety functions. In some cases surveillance event. An error by an operator during a trip when plant testing is used to detect equipment and system degrada-conditions are changing rapidly could also turn a routine tion and thereby determine when maintenance (restora-trip into a more serious event. Even on an ordinary trip tion, repair, or replacernent)is to be conducted. Iliston-from full power, on a pressurized water reactor (PWR),
cal records of suncillance test data can be used to secondary safety valves may lift and there is then sorne determine reliability to evaluate system design, and concern that the valves may not rescat.1hc main feed.
thence to formulae system improvernents. The mode in water system may or may not still be available. If it is not which a particular surveillance is conducted should be availabic, the auxUiary feedwater system must function to selected to minimize risk and yet still be able to ade-provide a continued heat sink. The trip may put a power quately meet the surveillance requirements. Safety sys-transient into the electrical grid sufficient to cause a loss tems are designed with such redundancy that the failure of preferred power or the automatic transfer from of any single component will not prevent a system from reactor-generated power to an offsite (preferred) source performing its safety function. Ily choosing components may not be successful so that the emergency diesel gen-e that are high'y reliable, the probabihty is low that two crators must operate. lf the trip resulted from or causes a redundant components would be inoperable at the same safety injection signal, the containment will be isolated so time due to random fauures. Testing is performed to that important (but not essential)scryices are lost such as, identify failed components and thereby reduce the prots in a pWR (depending on the design), cooling to the reae.
abihty that a system could not perform its safety function for coolant pumps and control rod drive cooling. (The because two redundant compments have failed during control rods would have alteady performed their safety the interval between testing each component. In addition function.) I etdown is also isolated. 'these isolations can to system redundancy, safety is enhanced by diverse sys-be overridden or the containment isolation can be reset if tems that are capable of performing the same safety func-it is spurious, tion. Where reliable and diverse safety systems are pro-vided, the impact on safety of increased surseillance liesides reactor trips, testing can also cause engineered intervals is educed, safety feature actuations which unnecessarily start safety-related equipment. liven if such actuation does not trip l
Several safety concerns arise with testing at power. The the reactor, it diverts the operators' attention and can first is that a reactor at power should have all its safety tmpose wear on the equipment.
systems available in case circumstances require their use.
'thus, the concern about testing at power is valid and if a safety sptem is being tested, portions (trains, dm-sions)of the safety system may or may not be availabic for testing at power should be minimiicd on those sys' ns use, dependmg on the type of test and the logic built into required at power and those that might cause a plant transient.
the system's controls.17or example, a standby emergency pump in thq recirculating mode would not be available for it does not, however, appear to be possible or desirable dehvenng water to the vessel or steam generator until an to climinate all testing at power.The operability of some i
operator changed its flowpath. On the other hand, the safety components and systems must be assured during controls for an emergency diesel generator are designed reactor operation. The fundamental question is how can i
so that a loss of voltage to an emergency bus will cause the the testing be done so that the potential for causing a 1-diesel to pick up load. even if it is in the test mode at the plant transient is minimized? 'lhe way the test is con-i time. 'lhis feature is required to be tested, typically at ducted and the frequency of testing are both important every refueling outage. Safety-related instrumentation considerations.
channels are tested one at a time so that redundancy ensures that the operable channels will perform their A large fraction of the testing that must be donc does not function if required.'Thus, safety systems may or may not lead to a reactor trin For example, the emergency diesel be available immediately to respond to an emergency.
generators are required to be tested at intervals much shorter than a refueling interval to assure their proper Another concern with testing at power is that sometimes operation. Ilowever, this testing should not make the the reactor experiences a transient during testing that reactor more vulnerable to a trip. There are, however, may cause a reactor trip. A reactor trip from high poweris some systems or components, which are discussed in this NURiiG-1366 6
... - ~ - -
,...-_o.,_,.
%,,ws--#.%,_,_,ms4,m
._..,my-n,,ey_,,,._m,on.-3.,,.y.,-,,pu.&.
- -,,y,,-.--.--,%.r,..,yp,7,,-.f
2 prm and Cons of Testing at Power report, whose testing rnakes a reactor snore vulnerable to been developed and these provide insights. In a recent a trip.
study using Pl(A analysis done for the Arkansas Nuclear One, Umt i reactor using data from the NI(C Interirn In detidmg w hether it is really necessary to test a piece of 1(chatnhty livaluation Program (NUlti!G/ Cit-5200), an equipinent at power when that testing has the potential analysis of survedlance test intervals showed that per-for causmg a trip, one must wnsider the importante to forming approxirnately 70% of the survedlance tests in-safety of that system or wmponent, the failure rate of that signifKantly decreased risk. 'the analysis only considered syr, tem or component, the effectiveness of the test in the benefit from testing and not the tr,k due to the test. If dmovenng the fadure, and whether the test inight be the risk of testing (such as potential plant inps or equip.
better performed when thut down.
ment degradation) were included, thd percentage of sur-veillance tests wtach insignificantly decreased osk would As an exarnple, control rods are important to safety, in a be even larger, pWit they are required to be tested rnonthly to assure that they are riot stuck w that if a scram signal opens the
'lhus, even though a precise number has not been deter-trip breakers, the rods will drop into the wre. Ilowever, mined, the evidence from this study indicates that more operating data show that this test has the potential for testmg takes place than is necessary to reduce plant risk of causing a reactor scrarn. Operating data also show that cor e incit, the frequeng of 1, tuck wntrol rods is very low and that, when a stuck rW haunturred, the condition of the reac-
'lhe data, however, do show that mme testing is impor-tor has remained within the bounds of the acadent analy-tant to reducing risk and inust be donc during operation of sis assumption that the lungle highest worth control rod the reactor.
doch not top. (this assumption is made in every fina>
uifety analyras repnt and in deterrnining reactor shut-
'lhis teport, by the use of operating data, discussions with down snarym.) In addition, a review of the operating data plant personnel who perform the surveillance tests, engi-shows that most stuck mds are not found by performing neering judgment,and a qualitative assessment of the risk this test, but rather are found at the beginnmg of a cycle impact, has attempted to detcrmine which tests are im-winic withdrawing the control rods before rnaking the portant to safety and which are not, reactor cruical or during low. power physics testing.
l(eviewing data on equipment failure from licensee event Considenng these points, it does not seem necessary to reports leads to the conclusion that many of the fadures perforrn this test as frequently as is currently reqnited, could have been prevented by better maintenance prac-and the NI(C staff recommends that this test be per.
tices. Industry reports on equipment performance have formed less !requently.
also reached this conclusion. If the frequency of inoper-able equipment were reduced, testing could also be re-On the other hand, the testing frequeng of mam steam duced. As long as surveillance testmg at power finds lin: isolation valves (htSIVs) on PWits can not be as failed equipment, it rnust be continued. 'lhe recornmen-clearly determined. hiSIVs are important to reactor dations included in this report are based up n balancing safety. 'ihey are required to be tested quarterly Since the benefits for identifying potential component failures wmpletely closing an htSIV would trip the reactor, they by frequent testing and the benefits of reduced testing are only closed 10% of the way,llowever, esen this test because of the potential f or challenges to safety systems, snakes the : eactor vulnerable to a tnp. MSIVs are comph.
equipment wear, removal of equiprnent from service for cated devices and operating data show that they do be-testing, personnel radiation exposure, and diversion of come inoperable. In this case, the test does uncover the personnel resources frorn other safety-related activitics, inoperable A1SIVs. 'lherefore, the decision of how often l'or each recommendation for an increase in a suncil-to test htSIVs rnust take into account both that the test larme interval, the conclusion is that this action wdl pro-makes the reactor vulnerable to a trip and that the test vide a net benefit to plant safety, does detect inoperable htSIVs. In this case, the N1(C staf f recommends that a more careful study be done to deter-
~lhe answer to the Commission's question is, themfore, mine the correct surveillance test mtenal.
that some testing at power is necessary. Indications also exist that too much testing at power is currently being 1(chabihty and probabilistic risk awessment (PI( A) meth, done. 'Ihe Technical Speof cations Improsernent Pro-ods have not yet been developed to the point at which gram is reducing the amount of required testing. Ilow-they provide definitisc guidance on determining speafie ever, testing will remain necessary for f mdmp moperable surveillance test mienals. Approximate rnethods base equipment.
7 NUltI & 1366
3 GENEllAL FINDINGS
'llus section addresses certain general topics related to staf f counted the number of surveillance tests required in surveillance testing required in Technical Specifications the Technial Spectfications that would be relocated as a to pre the reader a broader perspective before the indi-r esult of rehicating Technical Spectfications identified in vidual suncillance requirements are discussed. 'lhe fol-the split report. A loiling water reactor (limerick) was lowing subjects are included in this section:
chosen. It was estimated that during an 18. month cycle, 36,620 individual survedlance tests would be relocated by e
Marnitude of Suncillance Testing Required by climinating approximately 42% of the limiting conditions Tec hmcal Specifications for operation (if0s) that did not rneet the split report e
Preventive Maintenance and Surveillance Tcsting criteria for retention in the Technical Specifications. Of Japanese Approath to Surveillance Testmg the surveillance tests that would be eliminated,20,280 e
Technical Specifications and the ASMl! Code surveillances are required to be done while the plant is Scheduhng Suncillance l'esting shut down.
Suncillance Testing and Radiation lixposure Surveillance Testing and Plant Trandents
'ihat count did not consider those Ifos that were to o
surveillante Testing and Reliabihty remain in the Technical Specifications. (It would not be e
Survedlance ~estmg and Plant 1: *ign correct to extrapolate the total number of surveillances o
Sunculance Tcsting and Power Reductiors from this estimate because most of the safety-related e
Surveillance Testing and I!quipment Wear mechanical, electrical, and instrumentation equipment e
Survedlancc Testing on a Staggered Test liasis was not included.) As part of this study (using thcTechni-e cal Specitications of a new Westinghouse reactor), the 3.1 Magriittule of Surve.llance,I,est.inM NRC staff counted the number of surveillances required i
lleytilre(I liy Tecli:11eal for just the reactor protection system (RPS) and engi-Specificalions neered safety features actuation system (liSFAS)instru-mentation assuming an IH-month cyc!c with three shut-In August 1981, the NRC staff published a report downs and startups during the cycle.
(NURiiG-OS39)on the results of a survey taken by r.cnior NRC managers to obtain the views of plant operators on The results are given in Table 3.1. Also shown in the table the safety impact of regulatory activities. Surveillance is the estimated number of surveillances required after testing was one of the areas examined. Some of the com-the approval and impleinentation of the owners group ments iccorded in that report are similar to comments topical reports which propose to extend the suncillance received during plant vi,its conducted for this study. The test interval for this instrumentation.
comments received during the August 1981 study were
~
summarized as follows: "lhe number of surveillances is lixcluding channel checks, just over 5000 functional tests overpowering according to licensees. A large percentage and calibrations of the RPS and liSFAS are presently of incidents occur as the direct result of suncillances. A required. This number will bc. educed to approximately frequent comment was that equipment is wearing out.
2300, or a reduction of approximately 54%, when the One group oflicensees observed that some of the require-recommendations in the owners group topical reports are ments for testing do not make sense when compared with implemented.
ot her s."
The magnitude of testing can also be measured in terms
'lhe number of surveillance requirements is indeed large.
of personnel time required for testing. Licensees gave As part of the preparation of the split report, the NRC several indications of this during this study. These are Table 3.1 Number of surveillance tests required for RPS and F.SFAS instrumentation rur-rently, and number proposed in Westinghouse Owners Group topical report RPS 1:SFAS Surveillances Curr ent Proposed Current Pr oposed Total surveillances e,000 000 3.000 1,500 excluding channel checks Percent reduction in 56 %
$1%
surveillances N URl!G-1366 8
2 Generall'indings given in the sections that follow on individual Techrocal ponent or system. This may be partuutarly true if the Spedfications. As one example, functional tests of instru-sun'elllance test stresses the equipment severely.
mentation account for a large amount of monthly testing.
Approximately 900 staff hours per unit are expcoded Safety systems spend most of their time in a standby each month for these tests on more than 300 instrument mode. A majority of the operational time of some safety channels at one multi unit site. This does not include systems comes from testmg. 'lhe combination of long administrative staff hours for supervision and paperwork times in standby and short test run times may be very hard processing.
on some equipment.'this is true, for example, of pumps, if the pump must operate on recirculation on a ddferent in terms of nurnbers of procedures, which is not necessar.
por1 ion of the head-capacity curve than the point at which ily an accurate measure of the number of tests performed, it was designed to operate, and of emergency diesel gen-one utihty provided the folhiwing statistics. 'the utility crators, which are subjected to thermal stresses during uses approximately 300 procedures for suneillance test.
fast startup and loading.
ing which contain approvimately 1100 requirements. Of these 300 procedures, only 65 do not contain Technical To illustrate the stress of testing standby equipment, one Specifications requirements. Some of these do, however, licensee compared the wear of a continuously operating contain surveillance requirements from such documents centnfugal circulating water pump with the wear of a as regulatory guides and bulletins.
low head saf ety injection pump which is in standby serv-ice. The circulating water pumps are typically 10% de-The hcensce for 1.imerick provided the following infor-graded after approximately 10 years of operation; the mation on the total number of surveillances done on an low head safety injection pumps may be 8% degraded in annual basis. l'or 1986 (with no refuchng) 14,888 surveil.
the same time even though the actual running time was lances were performed. l'or 1987 (with a refueling out-much less. If a surveillance test only discovers that a age) 17,540 surveillances were performed. Approxi.
component or system is inoperable, it performs the func-mately 98% of these were required by Technical Specifi.
tion of limiting the time after failure that the piece cations, the other 2% were required by other agreements of equipment is clied on. If, on the other hand, a test made between the licensee and NI(C. This includes in.
identifies degradation in the performance of a component strumentation (l(PS, l!SI AS, radiation monitors, secu.
or syst em w hich may lead to failure, and if the degradation rity, etc.), pump and valve testing, and testing of electrical is corrected, then the test can limit the number of failures eqmpment. It includes tests donc Imth during operation as well as the time between failures. *lhus, preventive and at refueling outages. At this particular reactor, all maintenance that identifies degradation combined with channel checks for a day are counted as one test, so operabdity testing is more useful in maximizing the avail-char,nct checks contribute only 365 tests per year to this ability of safety equipment than are operability tests total.'
alone.
These numbers, although not presenting a complete pic.
'lhere is a definite relationship between surveillance test-ture of the magnitude of the utilities' cifort necessary to ing and preventive maintenance.The Technical Specifi-perform surveillance requirements, do give an indication cations, as stated earlier, require testing only to deter-that a large effort is required.
mine operability. Ilowever, equipment that is frequently inoperable is a symptom of a preventive maintenance program that is too htx. Some of the most important safety 3.2 Preventive Maititeriantee an(I systems, e.g., emergency diesel generators, batteries, anu
""""P'*"'*'"4""
hi h'9""'I'Y P'"ti"'
8 Surveillance Testinb>
maintenance program to prevent inoperability.
Suncillance requirements in the Technical Specifica-3: rom another viewpoint, an effective presentive mainte-tions for a nuclear power plant are written. in general, to nance program will prevent the frequency of equipment ensure the level of operability of systems and components failures from increasing as the equipment ages. Thus,if a required to meet the assumptions of the plant's safety preventive maintenance program is effective, needs for analyses. '1 hey are not designed to track degradation of testing will not increase; instead, an overly conservative equipment by monitoring or trending performance and-testing requirement established previously will become a therefore, the surveillance requirements do not necessar*
candidate for test reduction.
ily predict the adequacy of future operabihty of the com-3,3 Japanese Approach to Surveillance
lhe statf count of sutwillances for timeiKk dncuv.ed on the prn uus Testing page counted each channel check as a separaic test. When the starf count is adjusted h (ountmg au channel chech done on one day as one test.as a done ty ihe t nuenck bcensee, the counts are more con-Studies comI3arinE the.laIianese and American aI3 stent.
proaches to managing and regulatmg commercial nuclear 9
NURiiG-1366
l 3 General I'mdings power (e p., NUlti!U/ Cit-3883) examine the thfferences their operabihty. Verification of operabdity appears to be in testmg between the utihtics of the two nations.
relegated to the shutdown inspection process.
'the performance of the Japanese utilities seems to be Another factor to be considered comes from an industry exceller t. In 1984, n ycar for whith data were avadable, observation: "Any test involving an itPS thannel trip Japan had 28 nuclear reactors in operation '!hese reac-receives close super isory attention which helps to raise tors had an average capacity factor of 733% and the the operator's awareness of the effects of an error."
frequency of unphnned shutdowns was 0.1 time / reactor /
year (4 inps for 28 units).
In summary, it appears that the Japanese succen with avoiding incidents while testing at power is due to:
Suth rood peifor mance is duc, m large put, to the ngor-ous, extensive presenlise maintenance program whith is (1) avoidance of testing that rnay cause a trip,if possible performed at all Japanese noticar power plants.
(2) meticulous care and vigilant supervision in conduct-An annual inspection n mandated, and the goverrung I"8 tChl5 regulation speedies the systems and components that (3) equipment designed for testing to avoid "jur).
must be inspected. A voluntary supplemcntalinspection rigging" to do tests at power is performed at the same time, as well as corrective ac-tions, backlits, and corrective maintenance. The effort W
eventive maintenance that keeps tht, equipment includes systems which,in the United States, ate consid.
operable (A significant number of tripsin tuts coun-cred both nuticar sicam supply and balance-of plant.
try appear to come from equipraent failures while ilus ngorous prograrn does af fect the capacity factor.
the equipment is being exercised during surveillance Note that even with a very small number of unplanned t e ng) shutdowns, the capacity fattor is good but not great, in addinon, in Japan routine inspections (surveillance 3A Tecliiiical Specifications niul the testing)are carned out durmg operation.The Appendix, ASME Code taken f rom NUlti!G/ Cit-38N3, provides a partial list of surveillance tests donc during nonnal operation for a 10 CFit 50.55a(g) requires that safety related pumps and Japanese PWit and ilWit. In general, the test intervals valves be tested according to the equirements of Section and the equipment tested are not sigmhcantly different XI of the American Society of Mechanical linginects from those in U.S. plants, in some cases, certain equip-( ASMii)lloiter and Pressure Vessel Code ( ASMll Code) ment is tested more often in Japan than here. Some of the and addenda. 'the requirements for the tests of pumps (csts identif ed during this study as causing a significant and valves are summarized in Tables 3.2 and 3.3 e this r
number of inps are performed even more of ten in Japan, report. Section 4.0.5 of the Standard Technical Specifica-accordmg to NUlltiG/ Cit-3883, for example, PWit con' tions refers to these requircinents.10 CFit 50.55a(g)(4) trol rmi movement tests and IlWit MSIV partial closure requires that the inservice inspection program for a facil-tests-ity must be revised at 120-month intervals and the pro.
gram for testing pumps and valves must also be revised hes cral factors seem to account for the thfference in luth every 120 months. Moreover, these programs must use, to (ountries m the rate of incidents while performmg these the extent practicable, the requirements in the latest edi-tests.
tions and addenda in the ASMii Code that are incorpo-rated by reference in 10 Cl 1150 55d(b) 12 months prior The Japanese, in pencrat, imnimlic the performance of to the start of the 120-month interval.
tests that have the potential for inpping the reactor.
Itclay actuahons are not done at power.11PS logic is
'the wording of Section 4.0.5 is general enough to accom-tested, but the equipment is designed so that it is not modate these changes to the program without requiring necessary to enter cabinets; test switches are proviJed.
changing the Technical Specifications with each program Other equipment, such as MSIW,is designed to be tested change.
casily, lessening the likehhood of a mishap.
'the testing requirements of the ASMl! Code only test in general, accordmg to NUltt!G/ Cit-3883, surveillance cettain aspects of the operation of a pump or a valve and testing at power appears to be targeted at discovering do not require the venfication of whether a whole system passive fattores in com[onents rather than at ensuring will perform its safety function. In this respect, the testing NUltliG-1366 10
3 Ocneral Findings Table 3.2 Inservice test quantities for pumps Quantity Measure Obscrie Speed, N (if variable speed) y inlet pressure, P y'
Differential pressure, AP V
l' low rate, O y
Vibration amplitude, V y
Proper lubricant level or pressure y
~
llearing temperature, T y
'Mmure befmc purop stattop and duting test.
Table 3.3 ASME Code,Section XI requirements for inives Exercise and leak rate test Fall safe Setpoint press-thertise stroke time nut to exceed test ure verification 2 ears quarterly 5 ) ears quarterly quarterly 3
Check valves h10Vs and AOVs Check valves htOVs and Safety / relief AOVs valves Certain manual Certain manual valves and locked-closed vahes MOVs and AOVs, e.g., containment isolation valves will perform its safety function. In this respect, the tes-Monthly testing of Class 2 and 3 pumps was required by tingrequired by the Technical Specifications goes beyond Section XI of the ASME Code before the winter 1979 that required by the ASME Code. For example,in testing version vcas issued; this requirement was iucorporated pumps, the ASMR Code requires the information in Ta-into the Technical SpecificationsJlherefore, some utili-ble 3.2 to be obtained. l'or the auxiliary feedwater pumps, ties are still testing their pumps monthly, rather than the Technical Specifications typically require that the quarterly as is required by the newer sersions of the flowpath to each ste.un generator be venfied following ASME Code and Standard Technical Specifications.
cach cold shutdown of more than 30 days, that each auto-Some utilitics have sought relief from the ASME Code matic valve in the flowpath actuates to its correct position,Section XI monthly testing requirements on the basis that and that each AITV pump automatically actuates on re-the newer versions of the ASME Code only require quar-ceipt of the proper signal. For emergency core coolin8 terly testing. These requests have routinely been ap-system (ECCS) pumps, the proper flow distribution to proved.
cach cold leg (assumed in the safety analysis)is verified as well as the automatic actuation of the pumps on the llWR standby liquid control system pumps and PWR appropriate signals.
auxiliary feedwater pumps are required by the Technical 11 NUREG-1366
3 General l'indings Speuf tations to be tested monthly, even for those plants uds was not to exceed 3,25 times the specific surveillance at which other pumps a.,. required by the Technical interval.
Speahcations to be tested quarterly. 'lhese requirements 1% inconsistency in the application of Section 4.0.2 was are discuwed in Sections 4.3 and 9.1, respectisely, of this remedied by Genene f.etter 8914. 'ihis allows utihties I C P"' l-more flexibdity in planning sunedlance tests so that test-mg required by the ASMii Code and testing required by Thn report recom nends that safety.relatcd pump testing k Technical Specifications can be better coordinated that is done inore often (e g, monthly) than required in (by deleting the 3.25 requirement).
the current sersions of the ASMi!( ode be performed quar terly.
Appendix J to 10 CFit Part 50 requires the containment to be tested for leaks.'lhree types of tests are discussed.
A study per formed ho the I!!cctnc Power llescarch Insti-
.lhe type A test is an integrated leak rate test, the type 11 tute (!! Pill) exammed the testing of emerpeney pumps W for leakage across penetrations, and the type that are on standby to dcter ruine ll the testmg contributed C test measures leakage across isolation valves Type A to the failure ratc of these pumps (NP-4264, Vol.1). lloth tests are required by Appendix J to be performed "at PWR and ilWit ernergency pumps were mcluded in dus approximately equal intervals during each 10-year senice study.'t he study found that the highest pump failure r ates period." llecause of this wording, which allows flexibility, were auociated with PWit auxihary Icedwater (Al%)
N provisions of 4.0.2 apply to type A testing.'lype ll and pumps and llWR residual hcat emoval pumpsimd redJ-C tests must be performed "in no case at intervals greater ual heat removal service water pumps. Ilowever, the re-than 2 years." llecause of this wording,4.0.2 is not appli-port concluded "lacept for the powibihty that some W to type 11 and C testing.
plants rnay be testing at too low of a flow, none of the Al W l.liRs l licensee event reports] appear to be caused
.this en present scheduling problems if the plant has by any aspect of the sun edlance test."(this is dneuued m been shut down during the gcle for a long time, then more detail in Section 4 of this report )
starts up, and must now per f orm type 11 or C leakage tests to comply with the time limit of Appendix J. Some of I.or the llWR systems, the liPRI report concluded.
these tests can only be done when the plant is shut down.
"None of the RilR liesidual heat remosalland SW berv-
.lh s is also true of other survedlance tests (e.g., liCCS ice water l 1.lills appear to be caused by any aspect of the Dow diMution tests in a PWit, or some emergency survedlance test.
diesel pencrator surveillances which a.c required at month intervals),
'lhus, cuept f or one problem with Al/W pump testing,.
putop testing does not appear to have any adverse effect it has been the position of the NRC staff that the inser-on unertency purnpt vice survedlance testing interval should not be extended by either storage under controlled conditions or by the 3.5 Sclieduling Surveillance Testing sistem remaining dormant or shut down for long periods p
of time The NRC staff reconsidered this guidance as part Since t hc amount of testing required by Technical Specifi' of this program and still considers it to be reasonable cations is large, the hcensee should be given as much since equipment could continue to degrade even if the freedom as possible to schedule testing so that tests on plant is in a long. term outage condition (e.g., lubricant the same system can be done at the same tune to minimi/c hardening or resdient seal degradation).
personnel resources and maximize the availability of the equipment.
'lhe Technical Specifications improvement Program has climinated the Section 4.0.2 tequir. ment that the total Standard Technical Speed.ications contain a survedlance maximum combined interval time for any three consecu-requirement (4.0.5)which clarifies the application of the tive surveillance intervals should not exceed 3.25 times insenice inspection and testing requirements of the the specified surveillance interval.This does not alter the ASMi! Code to surveillance testing requirements in custing 25% extension hmit on each surveillance test Technical Specifications. Some versions of 4.0.5 allow the nterval. This is presented by Generic Letter 89-14.This frequency of these tests to be extended by stating that wiu be helpful for those intervals controlled by Technical Section 4.0.2 is applicable; other versions state that Sec-Specifications and not by the regulations.
tion 4.0.2 is not applicable. Section 4.0.2 allows an exten-sion of the survedlance interval by '5% Previously, the combined time interval for any direc consecutive inter-3.6 Surveillance Testing and Radiation Exposure me pnucm u inung on rmrtutanon woh rmuutanon hna wtah One of the critena used for this study, as discussed in rou e tw to km than acun rk+ oiber safety retaica pune rnay aim hwe ihn probicm on a plant semik Nm Section 1,is that the radiation dose recened by personnel NURliG-1366 12
3 General i mdings performmg the survedlance should be commensurate not especiauy teliable. Also,is the radiation dose received with the importance of the rurveillance. liefore examin-by the workers w ho enter the containment justified? 'the ing mdividual survettlance tests, it is useful to put the Ni(C should study this issue.
radiation dose received dunng surveillance testmg into perspective with the iadiation dose received performing it appears that the occulutional radiation exposur e tends other plant activities.
to mercase as a plant ages. *lhis appears to be due to additional maintenance and inspections required as the Table 3.4, taken from NUltliG-0713, Vol. 7 shows that r eactors age (N U R l!O / Cit-5158). This, tr e nd, ho weve r. is survedlance and mstrvice testmg accounted for f rom probably independent of Technical Specifications re-13.8% to 21.4% of the annual collettive dose at 1. Wits quirements since they would not increase exposure as a over the time peruvj from 1975 through 1985. Itoutine function of plant age.
and special maintenance accounted for the largest pet.
centage of the incurred dose (from 67.1% to 76H1).
3.7 Surve,illance Test,ng and Plant i
I;ven though surveillante testing is not the tarrest con-Trailsienis tributor to dose, it is significant and should be reduced wnere powible by reducing the frequency of testing in A study was rnade to determine which surveillance tests those areas that have significant radiation. Some utthties resulted in the most reactor trips and other types of plant have acted to reduce radiation dose during surveillance transients.
testing (along with other routine activities) by pe rformmg walkthroughs of the tests in order to minimi/c time in
'this portion of the study was done by reviewing 10 Cl lt high radutton are'as as part of their Al.All A (as low as 50.72 reports of reactor tops over the period from Janu-reasonably achievable) programs.
ary 1956 through July 1988 and by reviewmg 1. lilts on reportable test-caused transients occurring between 1984 la some cases, a conflict emts between the concern for and 1988.'lhose 1J!!(s were retrieved using the Sequence incurred dose dunng testing and the need to test. An Codmg and Search System (SCSS).
example of this, discussed in Section 8, is the testing of contamment purge and exhaust isolation valves with r esd-
'lhe Ll!!( review provided msight into the frequency and ient seats.
severity of test caused events, but the analysis was bounded by (1) time and resource constraints, (2) detail Containment purge and exhaust isolation valve tests re-prosided in the I.l!R abstracts and in 10 CFit 50.72 re-quire pctsonnel to enter the mntainment. A general con-ports, and (3) the I.ER collection itself. The SCSS data cern exists about containment entries at pow cr, is the risk were neither reviewed nor cross checked, not was addi-from the problem requiring the containment entry at tional sampling puformed to ensure that all relevant power a greater risk than the risk from the potential loss 1 Jills were reviewed. In adthtion, not all tests that pro-of containment integrity due to the entry? Containment duced transients are tests that are iequired by theTechni-airkwks are troublesome and aithick door interkwks are cal Specifications.
Table 3.4 Percentages of annual collectise dose at 1. Wits by wor k function Percemage of collectise dose each year Work function 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 lleactor operations and surveillance 10.8 10.2 10.5 13.3 12.2 9.5 8.9 9.4 10.1 11.4 12.8 Routme maintenance 52.0 31.0 28.1 31.5 29.2 35.5 36.1 27.9 29.7 26.9 34.6 Inservice inspection 3.0 6.0 6.4 7.7 9.0 5.5
$.3 6.5 7.6 6.3 8.6 Special mamtenance 19.0 40.0 42.5 35.9 39.4 40.6 40.5 46.8 43.9 45.4 32.5 Waste processing tt9 5.0 5.8 5.0 3.6 3.0 4.2 5.0 4.6 3.6 5.1 Refueling 7.7 7.9 6.7 6.6 66 6.1 5.0 4.4 4.1 6.4 6.5 hr NtmlGnM WI L 13 N U RiiG-1366
3 General Findings For PWRs, reactor protection system (RPS) surveillances consistent with the results of the 10 CFR 50.72 report gave the most trips. Turbine valve testing caused the next review. RPS-caused trips occurred most frequently dur-highest number.
ing regularly scheduled tests during steady-state operr-tion. The most prevalent causes were human errors such The next categories had approximately the same number as mishandling of electrical leads, omitted procedural of trips: enginected safety feature (liSF) loga testing, sLps, and wrong train / wrong channel events.
reactor trip br eaker testing, and nuclear instrumentation testing. Monthly exercising of control rads also contrib-Examples of these types of human errors are given below uttd to the reactor trips during the time period examined for one 15WR.
for this study. Figure 3.1 shows the frequency of trips due
- o various types of testing for PWRs.
Poor communication and inattention caused the wrong equipment to be tested and resulted in a Tiw large number of reportable events (more than 2000 recirculating pump trip.
LliR abstracts were examined) reflects the number and complexity of tests done at nuclear power plants. The Grounding of a jumper during removal for test res-o number of events occurring during shutdown was similar toration caused an unnoticed blown fuse. A reactor to the number occurnng at power. The increased testing scram occurred during subsequent testing of an-and maintenance activity while the plant is shut down other channel' appears to offset the fact that plants spend more time critical than shut down. Overall, plants typicidly submit 5 Grounding of a jumper during installation car g
e to 101.liRs related to testing each year.These events fall into several general categories:
trip of an RPS bus.
Valving errors during functional test caused trips E
o ing of a reactor vessel level referenceleg and spm.ng partial emergency system actuations of other instruments on the same reference leg.This e
4 caused a major plant transie~ and an eventual trip.
conditions that could jeopardue the safe shutdown e
The most common cause of tops from power found in the a
'the review of the LER abstracts (for both PWRs and IFR review (which covered a longer period of time than llWRs) showed that the reactor trip events were domi-the 10 CFR 50.72 review) is turbine-related tests at nated by events caused dunng RPS mstrument and con-power. Examples include trips due to malfunctioning of trol testing and turbine and turbine control testing.This is control valves, bypass valves, governors, and stop valves.
20 18 N
N 7\\
N
,\\
h E R b wx sh.
Flametof Turbine Engineered Fleactor Nuclear Control Mlacolteneous protection vehre end sa' sty trip instrurnenta-rod and systern overspeed featurse breaker tion exercise unknown ioec Figure 3.1 PWR trips attributed to surveillance testing. January 1986-July 1988 NL REG-1366 14 i
3 General 17indings Many of the turbine trips occurred at less than full power 3.8 Surveillance Testing and because the tests are required during reactor stanup or ggjgjgg the initial test conditions required lowered reactor power.
Turbine tnps were caused by electrical and mechanical
'lhe comment was made several times during the site problems, and, like most events, were also caused by visits that there is a lot of testing of equipment that rarely human errors in writing procedures or following them-fails but that must be tested anyway; that is, a lot of The remaining trips found in the I.litt data were generally unnecessary testing takes place.
caused by steam 4ystem, feed-systern, and electrical-system testing. 'lhe steam-and feed-system trips wcre The best way to approach the problem of too frequent or caused by stroking and partial stroking of MSIVs, and unnecessary testing isin actmsof tcliability-basedTechni-other assortcd condensate and feed system valves.These cal Specifications which would give credit for good per-tests,like turbme tests, can c1use spurious trip signals, or formance. If a comp < ment failed infrequently, the surveil-cause stcarn flow, steam pressure, steam generator level, lance interval could be increased (if the failures were and power fluctuations that can lead to trips, The total predominantly due to standby stresses as opposed to de-test induced trip frequeng is estimated (roughly) to be rnand stresses);likewise,if a component faded of ten, the 0.6 per year. Note that this is about 10% of the total trip surveillance interval would be decreased to sorne mini-frequency.
mum value.
The largest number of events reviewed were attributed to Ilowever, degradation of the equipment caused by testing
,{
partial emergency system actuations. Most of these too frequently must be kept in mind. If a component is events have httle safety significance, but they burden and fatling too frequently, better preventive maintenance, a distract plant operators. Most of these events involve design change, or some other solution is preferable to ventilation ( ontrol room, turbine building, auxiliary more testing.
buildmg, eted shifting to the crnergency mode. Ileactor water cleanup (ItWCU) initiations and partial contain-An NitC/ industry working group has been established to ment isolations are also common. Inadvertent equipment mvestigate the feasibility of risk-based Technical Specifi-starts include some cmergency dicsci generator and auxil.
cations in conjunction with reliability centered mainte-iary feedwater actuations. 'lhese are not common but can nance (ItCM). A lead plant has been sclected to be used lead to excessive equipment wear.1.ike reactor trips, in a pilot study of this approach.
nearly all events can be attributed to human error; less than 10% can be attributed to component failures.
'the Canadians currently use reliability-based Technical Specifications which state a reliabihty goal to be met by Overall, reportable events caused by surveillance tests certain systems.
occur frequently, but are not generally severe. 'lhe test-caused trip frequency is not negligible, but relaxing sur-As discussed m. Section 10 of this report, personnel in-veillance test intervals cannot be expected to significantiv volved in Generic Issue (GI) H-56 have been stt. dying reduce the overallinp frequcney. Test-caused trips mos't how to implement these reliabihty goals 'lhus, work is frequently occur because of errors made when personnel begmnmg m the area of reliability-based Technical Speci-fications follow spectfically written procedures or when operators mishandle electronic test equipment and leads durmg regularly scheduled surseillances.
3.9 Surveillance Testing and Plant U
Iteduced testmg at power will reduce the number of hu-man actions that could cause trips, but less-frequen' tes'-
lhfficulties in performing surveillance tests or inspec.
ing may proportionately increase the number of pro ( >
tions are sometimes not inherent in the requirement but duraland other hurnan errors. A specific concern nov J in in the design of the plant anJ its various systems.
the review is that many surveillance test procedures re-quirc operators to lif t leads or attach jempers quroncs to Instrumentation systems may be designed so that testing complicated circuits and test panels. These tests are diffi.
requires hfting of leads, use of jumpers, and other expedi-cult to perform in an error free manner, and as operators ents to bypass channels for testmg.
and techmcians perforrn the tests less frequently nnd arc, therefore, less famihar with the tests, error rates may Access to valves or other equipment that must be oper-increase.
ated in order to perform a surveillance is sometimes difficult, putting an unnecessary burden on personnel or Thus there is a balance between reducmg the test fre-exposing them to rnore radiation than would be necessary quency to avoid human errors and keeping the tests fre-with a design that took better account of the need for quent enough that the human error does not increase testing. For exampic, at one PWit,in order to ymfy that because operators are unfamihar with the pnwedures.
the liCCS pipmg is filled and vented (to ensfre that a 4
15 NUlti!G-1366
3 General Findings water hammer could not occur), operators and health its best efhciency pomt which degrades the pump.This is physicists who must climb ladden to the top of a tank farm discussed in more detail later.
must be dressed in protective clothing because the tank farm is a radiation area.
I!!ectrical and electronic equipment wears or breaks from unplugging and removing equipment from cabinets for At another PWR, the surveillance on the containment testing or from lifting leads a ' using jumpers, area high radiation monitor requires that a heavy (be-
.The tue of valves for isolation or flowpath change causes cause of shielding) high-level source be lowered to the monitor.
leaks around the valve packing and other valve or valve actuator problems.
~lhe current industry effort on advanced reactor designs should include a study of how all required surveillance The testing of an emergency diesel generator in its emer-testing will be performed in order to (1) minimize the gency mode induces thermal stresses and causes other possibility of a transient caused by testing, (2) rninimize problems which are discussed later in this report.
the burden on plant personnel who will have to perform these tests, and (3) minimize the radiation exposure re-
'lhus, the importance of the test must be balanced against ceived by people in performing the required testing.
considerations of wear on equipment as well as on other considerations.
3.10 Surveillance Testing and Power Reductions 3.12 Surveillance Testing on a
- Staggered Test Basis Soine surveillance tests in both PWRs and ilWRs require power reductions in order to prevent a transient that can -
Staggered testing is the scheduling of tests for the subsys-tnp the reactor. In a PWR, a power reduction ir ncessary tems or trains of a system in which the surveillance test for stroking turbine valves. In a BWR, there are three interval is divided in to a subinterval for each subsystem or tests that typically require a reduction in power: MSIV
- train, testing, control rod movement testing, and turbine valve stroking. '!herefore, another incentive for eliminating un.
The advantage to testing on a staggered test basis is that -
necessary testing is the increase in capacity factor if such the chances of a common-mode failure and equipment testing were done at a reduced frequency.
unavailability are reduced. A staggered test basis can have disadvantages.
3.11 Surveillance Testing and One resident inspector stated that, at his plant; this type Equipinent Wear of testing requires additional licensed operators and over-time for operators, it also requires more individual -
l'!quipmer.t is sometimes operated in a different way for -
entries into protection cabinets which causes scheduling surveillance testing than the way it would be used per-problems for licensees and may increase the chance of a forming its design function. A simple example is an injec-reactor trip, it can also extend the time required to per-tion pump which, when tested, recirculates water back to form surveillance tests by requiring initial setup time for a tank through a line that is smaller in diameter than the test equipment to be repeated for each test rather than normal injection line, thus making the pump operate off setting up just once.
1 I -
l l'
l:
i l
i, l]
L l
NURiiG-1366 16 l-
~
l 4 ItEACTIVITY CONTItOL SYSTEMS 4.1 Moderator Temperattire The NRC staff approved this proposed chage to the Cocilicient Mcasurement (l' Wit)
"TC '" N'I * '" *'" "'" 9"i"" ' "' I" dated September 8,1986 (lingle,1986). The concept has All Standard Technical Specifications for PWRs contain a generic applicability but requires plant specific analyses.
requirement to momtor the value of the moderator tem-perature coefficient (M I C), twh the beginning of-cycle Findings value (which is typically positive but is always the ! cast Technietd Specifications require determination of e
negative) and the end of-cycle value (which is the most rnoderator temperature coefficient at 300 ppm bo-negative value). 'lhe least negative value is important for ron concentration.
predicting plant behasior in events in which the reactor coolant is heated and the most negatisc value isimportant If measured moderator temperature coefficient is e
for predicting plant behavior when the reactor coolant is more negative (less conservative than the technical cooled (the most limiting event is the postulated main specification v;due), the licensee must measure the steam line break accident) moderator tempcrature coefficient every 14 liFPDs until the end of the cycle.
The current surveillance requirement in the Standard Measuring the moderator temperature coefficient Technical Specifications for end of life MTC states that the MTC must be measured within 7 effective full power M low boron concentrations is difficult, days (EFPDs) after baron concentration in the reactor VEPCO proposed a method for eliminating this re-e coolant decreases to 300 ppm. If this measured MTC is more negativy than a specified value, then measurements quirement below 60 ppm.
of MTC must continue every 14 EITDs until the end of Method is plant specific.
cycle.
Itecommendation In a letter to NRC dated January 3,1986 (Stewart,1986),
Virginia lilectric & Power Company proposed for North Other licensees may wish to use the VEPCO approach, Anna Power Station, Units 1 and 2, addmg a footnote to the technical specification which states that once the equilibrium boron concentration is 60 ppm or less, fur-4.2 Control llod Movement Test ther measurement of the MTC may be suspended pro-vided that the measured MTC at an equilibrium boron 4.2.1 Pressurized Water Reactors
~
concentration of 60 ppm or less is less negative than the A test that seems to be of concern to reactor operators is predteted salue at 60 ppm' the determination, every 31 days, that each control rod, not fully inserted, is capable of movement of a specified
.Ihis w as supported by ca,n.u. tions which showed that the amount in either direction.
- a maximum possible change in MTC from 60 ppm to the end of the operating cycle (llOC)is less than the differ-This test is performed to determine if the control rods are ence in values of M iC from 60 ppra to EOC specified in immovable. The control rods may be immovable either this technical specification, because of an electrical problem in the control rod drive circuitry or because the control rod is mechanically stuck.
This change to MTC surveillance requirements is impor-Followmg discovery, the licensee thust determine tant from an operational viewpoint since once the equilib-whether the control rod is capable of being tripped, since rium boron concentration falls below 60 ppm, dilution the action requirements differ depending on that deter-operations take an extended amount of time be ause such mination. As long as a control rod c,m be tripped, shut-a large volume of dilution water is required-down margin is not as great a concern; however, require-ments for rod mis:dignment and rod insertion limits must The VEPCO letter of January 3,1986 (Stewart,1986) still be followed with an imuovable control rod.
. states that, as an example, dilutior, of the reactor coolant system (RCS) from 50 ppm to 40 ppm requin s ch@ng of The concern with this test is that it causes reactor trips or approximately 17,000 gallons of primary gra w.arand dropped rods. At first this may seem strange since control requires ove 2 hours2.314815e-5 days <br />5.555556e-4 hours <br />3.306878e-6 weeks <br />7.61e-7 months <br />. Extended dilution timt
. e reli-rods are designed to be moved and this test does not able MTC measurements difficult to obtain because of insolve moving the control rods in any way that differs fluctuations on system conditions that may take place from the way the rods would be moved when controlling over this time interval.
power or power d:stribution. However, rod drive 17 NURiiG-1366
=
' 4 Reactivity Control Systems mechanisms and control rod drive control systems are 4.2.2 Boiling Water Reactors complex and their mecharrical movements and timing requirernents are exacting. Most PWRs, except for these All control rods in a BWR are regt. ired to be tested by tests, operate with both the regulating tods and the shut-moving the control rod at least one notch every 7 days.
down nds withdrawn during normal operation.
He purpose of this test, as with PWR control rod move-ment tests, is to verify that the control rods are movable in Electrical problems with the control rod drive system, in response to a scram signal.
general, do not prevent insertion of a contral rod into the core when the reactor trip breakers are opened.
5 At any time during the operating cycle of a BWR, some Mechanical problems are much less common. Tabic 4.1 control rods may be fully out while others are in an inter-lists cases in which the control rods were mechanically mediate position, in the ccmtrol cell core operating strat-stuck ar, at least, behaved as though they were. For thes.
egy which is used on most BWRs, most rods are fully cases, a reactor trip signal would not have resulted in withdrawn. it is the testing of the control rods at interrac-these axis inserting fully into the core.
diate positions between fully in and fully out that presents -
operational problems for BWRs since flux peaks above Two observations can be made about the events listed in the rod tips require a power reduction in order to reduce Table 4.1 in which a control rod was found mechanically the stress on the fuel during rod movement. This power immovable. First, not all of these stuck control nx!s were reduction rnust be scheduled around other plant activi-found during thc control rod movement surveillance ties, espec ally fuel preconditionmg ramps. li controt rods tests. Most were discovered during control rod drop tim, are exernsed dunng a preconditioning ramp, power re-ing tests performed during startup physics testing or when ductions of 50 to 100 MWe may typically be requtred.
the rods were withdrawn from the core during plant startup.
If a control rod drive double. notches past'the original position, local power changes of 1.5 to 2 kW/ft can result.
Second, accident analyses assume that the single highest nis s significarit in terms of stress on the fuel. Testing worth control rod is stuck and will not insert, in only one fully withdrawn rods does not require a load reduction.
of these events was a second rod involved (Point lleach he use of barrier fuel (or other ful designs that elimi-i Nuclear Plant, May 1985)and in this case both of the rods nate high local cladding stress) should alleviate this were partially inserted.
problem.
. A review of plant trip data from 1986 through July 1988 found three reactor trips that occurred during fuel rod Also, to prevent a missed surveillance, there is a streng motion testing.
desire to perform inis surveillance or, the same day each week. His, coupled with the !oad reduction require.
- In view of the successful operationcI recor.! demonstrated ments, makes scheduling very difficult.
- by the control rod movement tests during power opera-tion, the NRC. sta "f recommends that the surveillance Problems detected during this surveillance include hy-interval for control rod testing he changed from every 31 draulic control enit directional control valves sticking in days to quarterly-the open position and failures of the rod manual control -
system electronics. However, these frilures are not re.
Findings lated to the scram capabil#y of the control rod.
o ne purpose of PWR control nxi movement tests is to detcet rods that cannot move.
A search using the Nuclear Plant Reliability Data System (NPRDS) found 212 cases of control rod drive failures e
Most stuck rods are discovered during plant starm between 1980 and the end of March 1988. Of these, only 8 i
were found dunng testing and of these 8 no failures were during initial pulling of the rods or during rod dro'p testing.
found that would have affected ability to scram.
.1
'llus test causes reactor trips, dropped nx!s, and he NRC staff has had discussions with General Electric e
unnecessary challenges to safety systems.
Co. (GE) concerning the extension of the surveillance test interval. GE is concerned that if this survettlance Recommendation interval were extended stagnant water in the seal area would not be sufficiently flushed. Without flushing of this Change frequency of the PWR control nx! movement stagnant water, the seals might degrade. GE statea that -
tests to quarterly.
more study and perhaps a test would be necessary.
NUREG-1366 18
, w.,
I 4
Table 4.1 Mechanically stuck PWR control rods Discovered by TS rod motion Plant Date Reference Power hvel -
surveillance Description No Radial vane of spider, to which ? rodlers were attached Haddam Neck 08/69 See Note 1 had broken. Found during startup.
Pt. Beach 1 1000 See Note 1 Before No During rod drop testing, several shutdown criticality criticality -
t;ank A rods stuck intermittently. More than 20 milling chips were found in upper ir:ternals. One chip was found lodged between the guide tube assembly and one control -
rod spider arm.
l Robinson
' 11n0 See Note 1 Trip No Piece of foreign material was found in guide tube.
l Indian Pt. 2 0402-See Note 1 Before power No Malfunction of movement of 4 CRDs.Two were caused 4
05/72 escalation by foreign matter.Two were caused by locally reduced testing section of guide tube in dash pot area.
Yankee Rowe 05n2-See Note 1 No Rod 18 could withdraw but would not drop properly.
0802 Failed to drop completely during an October scram.
Loose fasteners suspected of causfrg difficulties.
4' St. Lucie 1 Oln9-See Note 1 Trip
.No Two control element assemblics (CEAs) stuck at 8 inches withdrawn when CEAs were tripped. Manual lifting freed.
CEAs. Cause not determined.
Davis.Besse 06/81
. LER 81-038 No Control rod 5-8 was not withdrawing with the rest of its group. Leaf spring anti-rotational device of the Icad screw j
nut assembly had fractured into several pieces preventing g.
lead screw from rising..
g 3
2 San Onofre 1 -
12/81' LER 82-003 No
' Red 2.~ c 61 (RCC) assembly stuck. Freed after s'
.4 epeated manipulatton. Probable cause: failure of weld r
W;ching a vane supporting two rodlets to the RCC hub.
9
- s 1.
I' j
. Calvert Cliffs 2
'02/82 LER 82-10,
' Shutdown '
No
. CEA-19 stuck following trip at 8 inches withdrawn. Seven -[
Rev. I days later CEA-19 was " freed" and tested satisfactorily.
c[m Inspection at following refueling failed to find a cause.
}co
- ~ -
r!
l
=
g w
.g w
.n.
Table 4.1 (Continued) h n.
Discovered by
'3-TS rod motion C
Plant Date Reference Power level surveillance '
' Description ~
~
';li Turkey Pt. 3 12/82 LER 82-109 Yes While petforming periodic exercise, RCC 33 of shutdm E bank A failed to show proper movement. It was verified : 1$
that 33 stayed in fully within position.12/28/82 unrelated trip occurred. J3 remained withdrawn. Root cause not de-termined. On 12/31/82, RCC J3 began to move normally.
Surry 1 06/84 LER 84-017 29 %
Yes Control rod B-6 became stuck, One of two holddown spring clamps had separated from the top of a fuel assembly and had become lodged between twa RCC assembly rodlets.
hicGuire 2 12/84 LER 84-032-No Duke notified by Westinghouse that a Korean reactor control rod drive mechanism (CRDhi) guide screw E.
Catawba 1 12/84 LER 84-029 rotated out of position, fell, and became lodged on top of CRDht latch assembly. Imose guide screws could cause bending and prevent driveline motion. 5 out of 53 guide screws on hicGuire 2 dM not meet acceptance i
criteria and were replaced.14 dim. rods replaced on Catawba 1.-
i Davis-Besse -
.03/85 LER 85406 0%
No Set screw frequently jammed inside CRDht, preventing
' disengagement of the latching assembly during ctmtrol rod drop testing.-
' Pt. Beach 1
' 05/85 See Note l No Control nod stuck at mid-height during rod drop testing.
Second rod stuck at 90 inches out.
. AfcGuire 2
- 06/86
" LER 86-008 0%
No Control rod Ir3 stuck and unable to drop into core. On 6/21/86 while inserting all control rods into core, control rod L-3 inserted with other rods. Sticking attributed to small particle of debris.
Palo Verde 1 01/88-See Note 2
'No.
CEA-56 did not drop during rod drop testing. Bearings from multi. stud tensioner machine lodged in guide tube. '
1.-
Nuclear Peer Emperience, Volorne P%%2; Iht-1, Experiences *; IV ' Control Rods and Drives *; Section A-Tontrol Rods? pp.1-14 (January 197kJuly 1988); S. bl. Stoller Corp.,
Iknilder, Colorado. ~.
- 2.'
. letter from E E Van Brunt, Jr. (Arizona Nucicar Power Project), to J. B. htartin, Region V Administrator (NRC). February 25,1988. L
-.g.
4 Reactivity Control Systems lhe NRC staff therefore recommends that the BWR (ASM E Code) requires that at least 20% of the charges in Owners Group study the feasibility of extending the test explosively actuated valves shall be removed, fired, and interval for the BWR control rod movement test. If it is replaced every 2 years with charges from a fresh batch. A concluded that the extension of this interval is feasible sample charge from the fresh batch must be tested satis-from an engineering design viewpoint, the NRC staff factorily. Charges must not be older than 10 years. If a should analyze the study imdings and allow the extension, charge fails to fire, all charges with the same batch num-if found acceptable.
ber must be removed, destroyed, and replaced with charges from a fresh batch from which a sample charge Another 'echnical specification for llWR control rods has been tested satisfactorily. Because these valves must states that if one control rod is immovable because of function during a reactivity control emergency, the staff friction or mechanical interference, the other control considers the 18-month surveillance of the explosive rods most be exercised once every 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br />. This require-valves in the SBLC system to be reasonable. For those ment is vcq resource intensive, in view of the successful plants considering a 24-month cycle, testing the explosive experience noted during the control rod movement tests, charges once each refueling interval would be acceptable.
the NRC staff recommends that the requirement be However, for plants that operate on a 24-month fuel changed to "within 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> of discovery of an inoperable cycle, the bounding surveillance interval could be as long control nxl and every 7 days thereafter."
as 30 mon hs. In this case,25% of the charges should be tested each refueling interval so that the Code require-Findings ment that all charges must not be older than 10 years will BWR control rml movement tests take significar t e
time to do.
The Technical Specifications also require testing the SBLC system pumps every 31 days. 'lhe ASME Code A scan:h using NPRDS failed to find a single situ-requires that this test be performed quarterly.The NRC e
ation in which an immovable control nxl was discov-staff recomrnends that the interval in this surveillance cred as a result of this test.
requirement to test the pumps be changed to quarterly, i
e if a control rod is found immovable, all other control Finding rods are required to be tested every 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br />.'Ihis is Testing of the explosive valves and pump in the standby very resource intensive.
liquid control system is required more frequently by the Technical Specifications than by the ASME Code, Recommendations The BWR Owners Group should study the feasibil.
Recommendations e
ity from an engineering viewpoint of extending the Explosive valves should be tested once cach refuel-e surveillance test interval for the control rod move-ing interval for fuel cycles up to 24 months' dura-ment test from the present requirement of every 7 tion.
days. If the BWR Owners Group decides that it is feasible from an engineering viewpoint, the NRC The SBLC system pump ::st should be required by e
staff should examine the study findings and allow the Technical Specifications quarterly, in agreement extension if found acceptable.
with the ASME Code.
ne Technic i Spec &auonuhm!d buhanged t 4.4 Closure Time TeStinE of Scram e
require that,if a control nxlis immovable oecause of friction or mechanical interference, the et a con-Discharge Volume Vent and Drain trol rods should be tested within 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> and every Valves (InVR) 7 days thereafter.
The Standard Technical Specifications require that at least once crety 18 n.onths each scram discharge volume 4.3 Standby, L.igmd Control System (SDV) vent and drain valve be tested to assure that it will (InVR) close within 30 seconds nfter receipt of a signal for control The standby liquid control (SBLC) sy: um injects concen-trated boric acid into the reactor vessel of a BWR as an The 30-second criterion for vent and drain valve closure emergency measure to maintain shutdown margin.This is was proposed as Criteiion 1 of the generic safety evalu-done by opening an explosive valve and starting a pump.
ation report for the BWR scrara discharge system (GSER, Technical Specifications require testing the explosive 1980). However, many BWR closure tests pnxtuce clo-charge every 18 months.The Bailer and Pressure Vessel sure times that are very close to 30 seconds, leaving very Code of the American Society of Mechanical Engineers little margin.
21 NUREG-1366
+
4 Reactivity Control Systems Georgia Power Coi, the licensee for I f atch Nuclear Plant, Recommendation
.ts 1 and 2, proposed to increase the required closure Other !!WR licensees may wish to use the Georgia Power ColGli method on a plant specific basis to extend the SDV vent and drain valve closure time requir rnent.
General lilectric Co. provideo an analysis (MDE 103 1184) that took into account the following factors and made assumptions about the value of each factor:
4.5 Reactor Scram Testing To Demonstrate Operability of Scram (3) "" * ' i * "
Discharge Volume Vent and Drain Valves (llWR)
(2) displacement volume of water per individual control rod drive A Generic Safety Evaluation Report was issued Decem-ber 1,1980 (GSER,1980) for the BWR scram discharge -
(3) average expected post. scram leakage flow per indi-system.The report considered the Browns Ferry Nuclear vidual control rod drive Power Station, Unit 3 partial scram failure event of June 28,1980 and subsequent investigations, tests, and analy-(4) SDV drain flow prior to isolation ses, and failures of scram level instruments at other l
plants. 'lhe GSER provided an acceptable basis for con-(5) minimum scram discharge volume tmued llWR plant operation along with criteria for design and/or modtfication of the scram discharge system.'lhis
'lhe analysis considered (1) overfilling of the scram dis-also included surveillance critcria to be included in Tech-charge volume (whhh would produce a radiological con.
nical Specifications.
cern tecause of spillage of primary water outside the primary containment) and (2) t he temperature increase in One of the surveillance tests required by the GSER is a the SDV (if the temperature were greater than satura, verification that the scram discharge volume vent and tion, there could be flashing of the water to steam and drain valves are operable when the control rods are scram possible i.ydrodynamic loads),
tested from a normal control rod configuratiori oiS0% or less rod density. The vent and drain valves must close The analysis showed that the SDV would not overfill with within 30 seconds
- after the signal for control rod scram is a longer closure time. Hence there was no ndiological yeceived and must open after the scram signal is reset, concern. 'The analysts (based on Gli data) also showed this test is required at least once every 18 months.
- that there would be no flashing of water to steam and, While reviewing a proposed technical specification therefore, no hydrodynarnic loads' change for the SDV at Susquehanna Steam lilectric Sta-tion, Units 1 and 2, to eliminate this test requirement,
,this analysis is dependent on specific plant values for the Region I personnel noted that a large variation exists for five factors listed above, Ihus, m the case of flatch, the SDV surveillance requirements between plants (Kane, licensee was able tojustify a closure time of 60 seconds for ggg.g' Unit -2 but only 45 seconds for Unit 1 (Crocker,1987).
NRC Region I staff compiled information regarding SOV llence, tne NRC staff recommends that licensees who vent and drain valve surveillance requirements for all experience difficulty with the 30-second closure time of NRC Region i HWR facilities.~lhis compilation showed Criterion 1 of the Generic Safety Evaluation Report that:
dated December 1,19SO(GSER,1980)and the Standard I
Tecimical Specifications do a plant-specific analysis us'ng
...in [NRC] Region I, 5 facilities have surveil-the approved methods of MDE 103 !!84 to derive a r ew lance requirements for testing the SDV vent vent and drain valve closure time for their plant (s).
and drain valves from less than or equal to SO%
l control rod density; 4 facilities require the test l
Findings to be done from a shutdown condition; one facility requires the test to be done frorn both
=
Some llWR SDV vent and drain valves close in conditions; and 4 facilitics have no requirement times very close to the 30.second criterion in the for this type of testing. In addition, review of Technical Specifications.
Technical Specification surveillance require-ments for recently licensed facilities in other Georgia Power Co. and GE derived a method of e
assuring safe operation with longer vent and drain
- see sectx>n 4 4 tn iha rgort on ctosure ume testing of the SDV vent valve closing times. This method is plant specific, and drain utves.
NUREG-1366 22
r.
4 Reactivity Control Systems regions revealed that River llend 1, Perry 1 NRC Region I staff recommended that: "the Technical Clinton 1 and Ferm: 2 cach have the require.
Specifications be revised to include a requirement to ments for conducting the testing from a shut-evaluate SDV system response after each scram to verify down condition.
no abnormalities exist prior to plant restart," and further stated that: "this requirement along with system testing The GSER technical basis for including a scram is that:
during a scram from shutdown conditions and periodic valve operability checi:s fully mecis the intent of the Ge-
[ A] total integrated system test [IST) will dem-neric Safety Evaluation Report for BWR Scram Dis-onstrate that the system retains its capability to charge System."
monitor the accumulation of waterin the SDV and to scram the plant when required.The IST lt is concluded that the Region I recommendations are will allow operators to check for the proper reasonable and should be applied to all 13WRs.
operation of system components and instru-mentation under operating conditions normal to a scram operation.
The NRC staff therefore recommends that this require-ment be deleted from the Technical Specifications.
The Standard Technical Specifications require the inte-grated system test to be performed at least once every 18 Findings months.
e A scram test at 50% rod density orless to assure vent -
Pennsylvania Power and Light Company, the licensee for and drain valve operability is required.
the Susquehanna Steam Electric Station, justified the technical specification change deleting this test require-e Other tests are done to assure SDV vent and drain ment on the basis that: (1) deletion will eliminate the valve operability.
possibility of subjecting Susquehanna Units I and 2 to additional plant scrams soleiy for the purpose of meeting this requirement and (2) operability of the SDV vent and Recommendat,ons i
drain valves can be adequately demonstrated during a scram uutiated from shutdown conditionr. as occurs dur-Remove the requirement fora scram check of SDV e
ing the 18-month reactor mode switch shutdown position vent and drain valve operability at a 50% rod density functional check.
or less.
Require an evaluation of SDV system response after
'Ihus, there are iaconsistencies in the application of this e
requirement to different BWRs. There also does not ap, each scram to verify that no abnormalities exist prior.
pear to be a strong technical basis for this requirement to plant restart.
since testing of the SDV vent and drain valves can be etmducted at shutdown conditions without subjecting the Require vent and drain valve operability testing dur-e reactor to unnecessary scrams.
ing a scram from shutdown conditions.
l I-I l-s 23-NUREG-1366
~
L 5 INSTRUMENTATION 5.1 Nuclear Instrumentation automatic rod withdrawal.The set point for this rod stop Surveillance (PWR) is below the reactor trip set point.
Operating data on the nuclear instruments were reviewed 1
.The excore neutron monitoring system for pressurized to assess nuclear instrument failure and how to detect -
water reactors (PWRs)is typically designed to sense leak-such failure. Many of the problems with nuclear instru-age neutron levels from neutron source strength (0.001 mentation appear to be found by operators doing routine n/cm2/sec) to a level greater than the neutron flux corre' channel checks or observations of the instrument sponding to 100% power (typically 125% to 200% power).
readouts.
In order to accurately cover this region, the excore system
-is divided into three ranges: the source range, the inter-Relatively few failures of tw nuclear instrumentation mediate range, and the power range, the source range appear to have been found during regularly scheduled covers shutdown and subpower neutron levels. The inter-surveillance testing. Power supply failures occur fairly mediate range provides coverage of the subpower operat-often and they are found immediately by alarms or by ing levels as well as covenng the entire power range. The deviations from expected values. Other frequent causes power range yields information of core neutron level of failure of the nuclear instrumentation systems are am-when the ieactor is critical and is supplying heat to the plifier failures, or cabling failures (e.g., failures caused by reactor coolant system.The three ranges of neutron mdt' radiation or water). The detectors themselves also fail, cation are successively overlapped by at least one decade but not as often as the electronic instrumentation associ-to prevent a loss of indicatton when one range is operating ated with the detectors.
at the high end ofits scale and the next range is operatmg at the low end of its scale.
Some failures of the nuclear instrumentation system cause reactor trips, especially when the plant is atteady in These instruments typically perform safety, numitoring, shutdown ar.d being monitored by intermediate range and control functions.
instrumentation (which is a one-out-of.two to trip).Tur-bine runbacks are also caused by faulty nuclear instru-Jihere are typically four channels of power range nuclear gjr tation. (For a t urbine runback, only one channel need m
instrumentation. A channel functional test is presently required monthly. The testing involves simulat% a trip It appears from a review of operating experience that signal in one channelf this type of testing resul n reac-tor trips when a human error or an electrical disturbance nuclear instrumentation experiences significant set point causes a second channel to trip. As is discussed in the drift. Therefore, before the calibration requirements are introduction (Section N:this report, naclearin:4rumen-m dtfied, the set point drift of nuclear instrumentation tation testing is a Gquent cause of reactor trips from should be more completely e amined. However, there.
testing.
seems to be no reason, based on a review of the operating data, why the analog channel operational test could not be performed quarterly rather than monthly. (This has The safeguards built into the design of the neutron flux already been proposed by the vendor owners group and in measurmg mstrumentauon vary from plant to plant but some cases has already been approvud by NRC.)
typically, four power range nuclear instrumentation chan-nels are provided for overpower protection. An additional auctioneered high signal is derived by auctioneering the Findings four channels for automatic rod control. If any channel Nuclear instrumentation systems consist of source.
fails in such a way as to produce a low output, that channel range, intermediate range, and power range detec-ts mcapable of proper overpowe, protection but does not tors and supporting electronic instrumentation and cause control rod movement because of the auctioneer.
power supplies.
I'wo out of four overpower trip logic ensures an over-power trip if needed, even with an independent failure in Testing of nuclear instrumentation is a significant another channel, cause of trips.
l l-In addition, channel deviation signals in the control sys-Most failures of the nuclearinstrumentation system -
tem give an alarm if any significant power range channel are found from plant behavior (trips, turbine run-deviation occurs. Finally, an overpower signal from any.
backs, or from operator channel checks, or response nuclear power range channel bhicks manual and to alarms).
NUREG-1366 24 j
5 Instrumentation Recomine.sdation Findings Relay reliability is generally good.
Change surveillance interval of analog channel functional tests of nuclear instrumentation to quarterly.
Relay testing at power contributes to the frequency of inadvertent equipment starts and reactor trips.
5.2 Slave Relay Testing (PWR, BWR)
Recommendation Each instrumentation channel and interlock of the engi.
nected safety features aciuation system (ESFAS), as well Perform relay testing on a staggered test basis over a cycle as the actuation logic and relays, are required by the and leave the tests carrying highest risk to a refueling outage or other cold shutdown.
Technical Specifications to be demonstrated operable.
The actuation logic test and master relay test are to be performed monthly on a staggered test basis. Slave relay 5.3 Test Intervals for RPS and ESFAS testing is to be performed quarterly. 'lhe actuation logic (PWR, BWR) tests a'id the master relay tests do not appear to be a problem; the slave relay tests, however, involve the actua-During discussions with one of the utilities,Ihe staff noted tion of a large nurnber of components (valves and pumps).
that although there wasn't opposition to the requirement A great deal of coordination is therefore necessary be-for staggered testing, a strict application of staggered test i
tween the test technicians and the control room requirements caused problems because it results in a operators-different test schedule for three-and four-channel sys-tems.The NRC is reviewing or has approved the owners These tests lead to inadvertent actuations of safety equip-groups topical reports for extending the channel func-ment and reactor trips.
tional test of reactor trip system channels from monthly to quarterly. For testing on a quarterly basis (every 92 days),,
NUREG/CR-4715 examined the failure modes of relays the interval between tests of different channels for a c.f various types (undervoltage, control, timing, and pro.
staggered test basis for three-channel systems :s every 31 tective) and concluded that although the failure data days, but every 23 days for a four-channel system (92/3 vs.
showed age-related failure trends for relays, the data 92/4). The effect is that each cabinet, containing one available to date do not indicate a high failure rate. The channel, is involved in a test twice each quarter. The j
normalized licensee event report (1.ER) and Nuclear preferred solution would be a relaxation of the staggered Plant Reliability Data System (NPRDS) data indicate an test cycle for three-channel systems to coincide with the average faiiare rate of fewer than two reportable relay test cycle for four-channel systems. The net effect is that failures peryear per plant, which is small b; comparison instead of two different testing cycles with the concom;t.
to the number of relays in the plant, tant increase in the required access to mstrument cabi-nets, there would be one four-channel testing intervalin The reliability of sl ave relays is a reasonable basis for which a three-channel test would not be performed.
relaxing the testing requirements.
.Itc NRC staff recommends that this change be imple-mented with the approval of the owners groups topical The NRC staff recommends that the scope of each quar' reports and allow its application before then if a utility so terly test be reduced so that a different sample is tested chooses.
i I
each quarter. This approach has been applied to at least one plant that could not test some slave relays at power.
l Depending on the sample size, some slave relays could be Find "E' tested at power and others could be tested when the plant Some RPS and ESFAS functions have three chan--
is shut down.
nels; others have four.
With this alternative, one could be assured that those Staggered test basis testing would require more fre-e slave relays that offer the greatest potential for plant quent access to instrument cabinets when applied to upsets would be deferred to that group that is tested three-and four-channel functions.
during a plant shutdown. This would further the goal of reducing the potential of challenges to safety systems due Recommendation to testing during power operation. Ily testing a portion of the slave relays during each quarter, there is a reasonable Test three-channel systems on the four channel sched-time limit in which common-mode failure could be de-ule. Do not test cac of the three chanaels during a four -
tected.. Common-mode failure is generally a great:r channel test interval.Thus, the sequence of testing would -
threat to safety than are single-component failures.
be:
25 NUREG-1366
5 Instrumentation Three channel Four channel A
A the sensor inside the containment.The Tecimical Specifi-3 g3 cations allow these valves to be " opened on an intermit-tent basis under administrative control," There are four C
C contair.' nt isolation valves for two monitors. "Adminis-D trative control" means that measures are taken to assure A
A that the control room operators are aware of the position of these valves so that the appropriate action will be taken
'lhis could be implemented now for those three-and to close them, if required, four-chanael functions requiring staggered test basis testing.
There are other operability checks of the system. "Ihere are alarms to indicate electronic system and power fail-utes. In addition, operations personnel monitor the indi.
5.4 IIydrogen Monitor Surveillance cation daily and note any changes in indication. In view of (PWR, BWR) this, the NRC staff recomrnends that the surveillance test interval for hydrogen monitors be extended to a refueling.
Ilydrogen (or combustible gas) monitors are to monitor interval, and analog channel operational test to quarterly.
hydrogen concentration in containment following a loss-of-coolant accident (LOCA) and are designed to be con-sistent with Regulatory Guide 1.2," Control of Combusti-Fl.. dings
- ble Gas Concentrations in Containment Following a LOCA " These monitors are used only after a LOCA to Ilydrogen monitors serve onlyan indicating function e
tell the operator when to initiate the hydrogen recom-and are only required after an accident in which the biners. The hydrogen recombiners are not required for a core is damaged.
' period of hours to days after a large-break LOCA.
Calibration requires opening containment isolation v Ives for' a systeta with sensor inside the.
The Technica: Specifications require a channel check (at c ntainment.
leaet once every 12 hours1.388889e-4 days <br />0.00333 hours <br />1.984127e-5 weeks <br />4.566e-6 months <br />), an analog channel operational test (at least once every 31 days), and at least once every.
92 days, on a staggered test basis, a channel calibration Calibration is time consuming.
e using two gas samples, one containing 1% hydrogen, the other containing 4% hydrogen.
Recnmmendation Change frequency of calibration to once each refueling The monthly tests check only the electronics. The quar-interval and analog ch mnel operational test to quarterly.
terly test checks calibration.
Typical staff hours required for these tests are given in the 5.5 Reactor Trip Breaker Testing table that follows:
(PWR)
The Standard Technical Specifications require reactor -
Personnel
. trip breakers and reactm trip bypass breakers to be tested :
f Test required Time every 31 days.The St.
irdTechnical Specifications fur-ther allow a channel to be bypassed for surveillance test-Analog channel 2 technicians 4 hours4.62963e-5 days <br />0.00111 hours <br />6.613757e-6 weeks <br />1.522e-6 months <br /> ing for up to 2 hours2.314815e-5 days <br />5.555556e-4 hours <br />3.306878e-6 weeks <br />7.61e-7 months <br />.-
operational test A search of operating data was undertaken to attempt to Calibration with 2 technicians 32 hours3.703704e-4 days <br />0.00889 hours <br />5.291005e-5 weeks <br />1.2176e-5 months <br /> determine the cause of reactor trip breaker failures. In two es samples particular, the NRC staff attempted to determine if test-ing might be contributing to the failure of reactor trip a
breakers.
Typically, the hydrogen monitors have alarms to indicate electronic system or power failures. In addition, daily NUREG/CR-4715, prepared for the Nuclear Plant -
channel checks are done.
Aging Research (NPAR) Program, notes that most of the failures of reactor trip circuit breakers were age re- -
In order to perform the analog channel operational test '
lated, that is, related to service duration and the number and calibration, the containment isolation valves which
_ of operating cycles, and that, among the primary me-are nonnally closed must be opened for those plants with chanically induced stress mechanisms is routine opening NURI!G-1366 26 '
.l
.l
=- --
5 Instrumentation I
and closing operations. !! therefore seems that testing is a consistent with the calorimetric power. During the plant contributor to breaker failure. Ilowever, adjustment and visits conducted for this study, it was suggested that the
- I lubrication problems were also significant, as were bro-2% limit be retained for steady-state operation, but that a ken parts. 'lherefore, although testing may be too fre.
5% limit be permitted for suci transient operation as quent, it seems that preventive maintenance could be might occur during power changes a startup or kiad improved.
changes, because a less restrictive allowable deviation during transient operation would greatly lessen the prob-In order to assess, based on today's testing and preventive ability of reactor trip due to error while thb calibration is maintenance practices, whether less frequent testing performed. %ts was considered justified because of the might contribute to better reactor trip breaker reliabihty, temporary nature of the situation.
a more detailed study is necessary. For this present study, the answer was inconclusive. It is relevant to note that This proposal was made on the basis of f.he fact that many since diverse trip features were incorporated in reactor of the adjustments that are made at low power or follow-trip breakers (shunt trip feature) there h^.s been no re-ing transients may not ensure the same accuracy at full laxation of the test interval, although those changes re-power.Thus, many of these adjustments made at these suited in a significant impmvement in reliability.
conditions may not have a significant effect on the accu-racy at which high power trips would occur. Also, the flowever,it is clear that reactor trip breaker testing has plant computer provides almost continuous thermal caused reactor trips.ne reason for trips is usually human power calculaticms which are used to ensure that the error while conducting the testing. In order to attempt to 100% power rating is not exceeded and, thccefore, the reduce this error rate, the NRC staff recommends that errors in indicated nuclear power are not significant from the allowed outage time for one channel be increased to the.itandpoint of exceeding the licensed power level.
allow plant peoonnel to do the testing without being rushed.
However, reactor safety analysc;are done assuming a 2% -
crror in power and the safety analyses assume that the Findings accident bei.ig analyzed can start from any condition of normal operation. Therefore, to increase this uncertainty Testing reactor trip breakers has caused rewtor to 5% would be inconsistent with the accident analyses e
trips-assumptions on power variation. At least one reactor vendor, however, already takes into account a transient Most of these (nps are related to human error.
uncertainty as well as a steady-state uncertainty in calcu.
lating core power distribution limits.
Agmg of reactor trip breakers is a function of the number of cycles of operation. Reactor trip breakers De staff suggests that the owners groups consider this i
are most often operated for testing, issue, and if it is judged to be of sitfficier.t importance, the owners groups should present a proposal which will help Reactor top breakers tequire good preventive mam-prevent reactor trips and be consistent with the safety e
- tenance, analysis and Technical Specifications power distribution requirements.
Recommendation De vendor owner groups should consider whether more Findings recent operating experience would justify a change in the i
j test interval for reactor trip and bypass breakers. Licen-At greater than -15% thermal power, the nuclear sees sould pursue implementing an increase in the al-pwer must agree with the calonmetne power w ithtn If it does not, the gain of nuclear instruments lowable outage time for testing reactor trip and bypass breakers as addressed in the vendor topical reports for must be adjusted.
cxtending surveillance internis.
A recommerdation was made that the uncertainty.
should be increased to 5 % to help avoid reactor trips 5.6_ Power Range Instrument during the calibration when the plant was in a tran-Calibration (PWR) sient operating mode.
Technical Specifications require that above 15% power Safety analyses assume a 2% tmcertainty on power.
e the calorimetnc power be compared with the excore This must incit.de unsteady power operation unless power indication. If the absolute difference is greater a specific allowance is made separately for unsteady than 2%, the excore channel gains m ist be adjusted to be power.
27 NURifG-1366
i 5 Instrumentation Recommendation and may result in premature failures of these lhe owners groups should consider this issue, and if it is judged to be of sufficient importance, the owners groups The CEACs do not generate a penalty on inward single should presmt a proposal which will help prevent reactor CEA deviations and are not required to protect against trips and be consistent with the safety aaalyses and Tech
- multiple inward deviations The CEAC is not credited for nical Specifications power distribution requirernents.
protection against CEA ejection events. Its only purpose currently is to protect against single CEA withdrawals Imm deep Emup insertions at high power.1his combina.
5.7 Control Element Assernh.I}.
tien of conditions is unccmmon. The " withdrawal pro.
Calculator Surve.llance (CE CPC hfbit" signals generatcd by either CEAC protect against i
PWR) many withdrawal eventsJihe CEA drive mechanism con-trol system is designed to be single-failure-proof rio that Newer Combustion lingineering (CE) reactors utilize a the probabihty of core damage due to events normally computer system as part of the plant protection system covered by CEAC operation is low (PPS).This system is the core protection calculator (CPC) and is used to generate two reactor trips (there are several The CEACs perform self-checking routines during nor-other reactor ' rips included in the system design, for mal operation and the operatos check the CEAC opera-example, each of the San Onofre Nucimr Generating tion at least twice each day as required by the Technical Station Units 2 and 3 has 15 total trips), The two com.
Specifications. Sudden major failures would result in an-puter-generated trips are the low departure from nucle-nunciation. A slow drift would be discovered by the chan-ate boiling ratio (DN BR)and the high local power density nel checks.
(1.PD) tnps.
This surveillance requirement, therefore, appears to The control element assembly calculator (CEAC) pro-meet several of the stafI's crif cria. Its one-out-of-tvo trip vides the CPC with information about indi-idual CEA logic means that the plant is plated in a vulnerable posi-position devutionsJlhe CEAC *hioks" at the four CEAs tion. The people performing the surveillance are under in a subgroup and sends penalty factors to the CPCs if the some stress to work quickly on a sensitive task and there is positions of rods in a subgroup deviate from each other by some wear on the equipment. In addition, the equipment more than a deadband value, is self checking and reliable. The NRC staff, therefore, recommends extending the surveillance interval to quar-The CE AC also provides a CE A withdrawal protubit sig.
terly for the channel functional test. A further extension nal via the CPCs on a CEA misalignment, might be possible on a plant-specific basis.
The monthly channel functional test is required to in-Findings clude the injection of simulated process signals as close to lhe CEAC channel functional test places the plant j
e the sensors as practicable to verify the operability, includ, ing the alarm and trip functions. To perform this test in a vulnerable position.
requires two technicians to work about 4 hours4.62963e-5 days <br />0.00111 hours <br />6.613757e-6 weeks <br />1.522e-6 months <br /> per The test puts some stress on pers< nnel and the a
CEAC. lt involves an admimstrative burden in scheduling eqmpment.
sutveillance, since testing is not allowcJ by the licensee during PPS testing or during " hands-off" periods of high The CEAC is self-checking. In addnion,' channel e
system h)ad demand. Because of this, the technicians checks are required by Technical Specifications.
must work quickly, accordtng to the San Onofre licersee, increasing the risk of an inadvertent plant trip.
Recomn iendat.wn With one CEAC out of service for this testing,a failure of Extend the surveillance interval from monthly to the other CEAC wdl cause a plant trip. Hur an error quarterly.
a
- could also lead to a trip since the CEACs are one-of.two to trip.
5.8 -incore Detector Surveillance (CE The licensee for San Onofre stated that during the tests and B&W PWRs)-
the CEAC is powered off and powered on, which repre-sents a stress over time on the computers. The cabinet incore detectors in CE and B&W reactors are used to 4
doors remain open during this test. causing both the monitor core power (hstribution parameters such as CHAC and its associated CPC to overheat because the azimuthal (quadrant) tilt and axial shape index (CE) or normal air cooling Dow patterns are disrupted. 'lhe test-axial power imbalance (B&W). In CE reactors incere.
- ing invokes mechanical cycling of cables and connectors detectors are used to monitor the linear heat generation
- NUREG-1366 28
~.-
. - ~. -
5 Ir strumentation rate and in B&W tcactors, to measure FQ (ratio of maxi.
failure of a detector cannot result in core damage. The mum core power to average core power).
detector must fail in conjunction with some other initiat-ing event. Smcc there is no limitation on specific detec-In CE reactors in which the core protection calculator is tors that may be failed, operation with a newly failed part of the plant protection system, the incore detectors detector is not materially different from operation with a provide input to COI SS (core operating hmit supervisory detector whose failure has been recognized, system).
COLSS checks the incore de actor signals for validity and COLSS is a continuous online monitoring system whose monitors compli:mee with the Technical Specifications simits on failed detectors. COLSS also calculates axial program is located in the plant computer. Its inputs to the piant computer are control grade (not safety grade),
shape index (ASI) and azimuthal tilt rhich would show a.
"Ihus, COISS is not part of the plant protection system sudden change in value if a detector failed seriously.
4 and does not trip the plant, it monitors LCOs on thermal COLSS ASI is compared to CPC ASI on each shift and margin (DNilR), linear heat rate, azimuthal tilt, and raial any deviation would be identified at that time, shape index. It also assists the operator in rnaintaining
'Ihe NRC staff recomn7 ends that the surveillance re-j total core power equal to or below the reactor's licensed quirement for Clrl cenmcal Spectficatums be changtd to i
power. COtAS will alarm to warn the operator if any of be consistent with the B&W Technical Specifications; these LCOs are violated.
that is, the surveillance of the incore detectors would be i
required within 7 days before they are used fm power in CE reactors, the surveillance requirements for the dstribuhon meamcments.
incore detectors state that a channel check is required within 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> before its use for monitoring or czdibrat-It should be noted that this change does not change any ing and at least once every 7 days thereaf ter. lhos, a operabihty requirements foc incore detectors, just the channel check is required weekly, m, dependent of the surveilkmce requirements.
requirement to assure operability by performmg a chan-nel check before use in monitonng or calibrating. H&W Finding
)
reactors do not have this requirement. B&W Technical Specifications require a channel check only within 7 days
.Incore detectors are monitored within 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> before I
before use, use and weekly in CE reactors and 7 days before use for power distribution measurements in H&W reactors.
the channel check requires about I hour each for San Onofre 2 and San Onofre 3. This would be irc. eased to 3 Recommendation hours a week per unit if a combination of new and old
.Ihe H&W surveillance requirement for incore detectars detectors was used in the core.The Technical Specifica-
- should be used for CE plants.
tions require that Cae detectors be monitored before us-ing them with COISS on reactar startups.This results in critical path impact since the reactor must be above 15%
5.9 Ilesponse Tinie Testing of power to generate sufficient neutron flux and must re-
] solation ACttlation main below 20% power until the surveitlance is done.
Instrtinlentation (IMIt, IMit)
'lhere are nominally 280 incore det ectors in the core.The In order to limit the consequences of line breaks, BWRs Technical Specifications allow operation with 2$% of the are designed with the capability to isolate portions of detectors out of service. Thus, more than 70 detectors systems to retain vessel inventory. A typical list of these
,Q
- would have to be failed to fall outside the Technical isolations follows for a HWRF Specificatrons limit. On average, there is less th:m one detector failure per 100 effective full power days (EEPDs)
(1) PRIMARY CONTAINMENTISOLATION-of operation (according to the San Onofre licensee),
(a) Reactor Vessel Water Leve,t Therefore the probabihty of a single failure during the b I "' bC"I 1 weck interval is extrenicly small. A detector failure 2-
- I"' bCYCI 3 normally causes COLSS calculations to become more 3-
- I" U**k "II conservative, The incore detector signals a re verified on a (b) Drywell Pressure - Iligh -
monthly basis anyway as a part of performing the monthly U
"* U"C core parameters surveillance. COISS checks the incore L Raqation - liigh -
signals internally and will reject one that is obviousiv 2.
mc - Low failed. It will also alarm m the control room if more than plow -
1 i gh 25% of the detectors are failed, thus helping to ensure compliance with the Technical Specifications. Since the
_"tw im e for meam wy and h om cmpt*
incore detectors are a purely passive system, a simple 29 '
' NUREG-1366 -
5 Instrumentation (2) SFCOND ARY CONTAINMiiNT ISOLATION Another negative aspect of the response time testing besides the burden on personnel time is the increased risk (a) Drywell Pressure - liigh of plant trips and equipment damage. While these risks (b) lleactor Venel Water Level - 1 ow Low, are present during any testing, response time tests are g x. vel 2 particularly at risk for the following reasons:
First of all, a large number of lifted leads, jumpers, and (3) HI!ACIOR WKilitt CiliANUP SYSTliM 150-pull d fuscs are required to simulate t he necessary system iE110N conditions. 'lhis creates the potential for errors. Some (a) oflow - liigh examples from experience at 12Salle are given below:
(b) iIcat lixcitanger Area Temperature - Iligh
- 1. cad of temperatute compensatien resistor was bro-(c) Ilea' lixchanger Area Ventilation 6T thgh ken during post test restcration of lifted leads.
(d) Reactor Vessel Water Level - Im Low, 4g9g$)
Leve 2 e
Fuse olew when reinstalled, but condition was not (4) RiiACFOR Cold ISOLATION COOLING SYS-noticed. When another fuse was removed for testing TliM ISO! AllON in another channel, a reactor scram occurred. (1937)
(a) RCIC Steam 1.ine now - Iligh Jumper that was grounded during removal for test a
(b) RCIC S;eam Supply Pressure - 1.ow restoration caused unnoticed blown fuse. Reactor (c) Drywell Pressure - liigh scram occurred daring subsequent testing of an-(d) RCIC lic ipment Room A Temperature '
other charmel. (1987)
High Jumper that was grounded during installation (5) RilR SYTrliM SIRITDOWN COOLINO MODIi caused an RPS bus trip. (198S)
ISOLATION In addition, most response time tests are performed dur-(a.) Reactor Vessel Water I.evel - Low, Level 3 mg refuelingoutages when the configurations (i.e., jump-(b) Reactor Vessel (RllR Cut ~in Pernussive) ers, lift ed leads, etc.) of associated systems are morcelikely Pressure - liigh to be different from those covered by the test procedure.
Safety analyses assume that instrument channel actuation And fmally, the test equipment used to simulate hydraulic for non.MSIV channels occurs snaultaneously with diesel camp inputs to sensors ;s complex, difficult to use, and start. The Technical Specifications, therefore, give te, unfamtPar to technicians (due to its infrequent use),
quired response times which include both the instrument response time (typically measured in fractions of a sec, Failures and degradation which could affect response ond). the response tirnes of the actuation logic circuits time in.hese systems are normally detected through (typically less than a second), and the diesel generator other tests of the same or higher frequency (such as start time (10 to 13 seconds, r~pically).
monthly functional tests and logic functional tests). Two comptes of mch situations nottd during a review done by e 125alle licensee included the following.
Therefore, the chant.e is remote that a channel's respouse time would degrade to the po'mt where it exceeds the Torn diaphragm in RCIC exhaust diaphragm high 13-second diesel start time without a !ailure that would be pressure isolation switch. (Found during a monthly noticeable in other ways.The Technicai Specifications functional test.)
"flases" section stetes that the only purpose for checking these reponse times is "to enhance overaH system reli-Torn diaphrugra in RCIC high steam flow isolation e
ability and to monitor instrument channel response tiroe dP switch. (Found during a monthly functional test.)
tre nds."
2 PWR Technical Speelfications have 4 sinnlar table and the difficulties are similar to BWRs.
At LaSalle, approxtmately 230 staff-heurs/ unit / refueling interval are expended satisfying the non MSIV isolation The NRC stalf therefore recommends 0 Sninating re-systern response time testing regairements. hve days, sponse time testing of isolatiouctuation in.,irumentation l
around the ekxk, se required to perform these sarveil-where the requ'tred response time corresponds to the dtesel start time.
It appears this staff time could be berier spent on other As purt of f.hcTechnical SpeciDeations tmprovement Pro-work.
grare, the NRC staffis cut rently concidering extensions in NURiiG-1366 30
.~ --
~.
- -~
5 Instrumentation the surveillance test intervals and allowed outage times 5.11 Calibration of Recirculation Flow for isolation actuation instrumentation common to th Transmih s OlWR) reactor protection system and/or the f!CCS.
'the control rod withdrawal bhsck instrumentation sur-findings veillance requirements incluh a requirement to calibrate recirculation flow (pressure difference) transmitters Non MMV isolation response time testing is diffi-quarterly, in HWRs these are typically Rosemount units, e
cult, time consuming, and has a risk of tripping the
'!he normal frequency of calibration in other applications reactor.
using the same type of transmitter is refueling; for exam-plc, this is the case for low reactor water level isolation
'lhe test criteria are not meaningful compared to functions. Recent failures of Rosemount units, which i
measured isolation response time.
upon failures have displayed symptoms that can be de-tected d uring calibration (slow drift, lack of response over Herommendation transmitter's full range, increase in transmitter's time response, deviation from normal signal fluctuations, de-Delete requirement from both !!WR and PWR Technical crease in actectable noise level, deviation between sig-Specifications to perform response time test mg where the nals, "one-sided" signal noise, and poor response to tran-required response time corresponds to the diesel start sients) dictate that the calibration requirement remain at I""
- a quarterly interval.
5.10 Source Range Monitor and rinding Intermediate Range Monitor HWR pressure difference instrumeritation has different Surveillances (llWR) calibration requirements in different sections of the Technical Specifications. Calibrations are required quar-Present calibration interval for the source range monitors terly and at refueling. Rosemount units have been re-(SRMs) and intermediate range monitors (IRMs)is quar-cently failing and, in doing so, exhibiting symptoms that terly for surveillances required for the control rod with-are detectable by calibration, drawal bk3ck instrumentation. Another Technical Sccifications surveillance requirement for reactor protec-Recommendation tion system instrumentation specifies a refueling interval calibration for the IRMs.
Retain surveillance interval at its current quarterly frequency.
A review of set point trend data for the IRMs done by the laSalle licensee shows very little drift in equipment cMi' 5.12 Autoclosure Interlocks: Removal bration, lhese data, supported by the refueling interval frequency already incorporated in the technical specifica.
of the LCO From the Technical tion fm reactor protection system instrumentation sur-Specifications (PWR, llWR) veillances, provides a justification for extending the cali-bration interval for the SRMs and IRMs.
The normally closed residual heat removal (RHR) system isolation valves are opened only for RIIR after system l
'Ihe NRC staff recornmends that the interval be changed pressure and temperature have been reduced to the set to once cach refueling interval for calibration of the points for system operation which are below the design source range and intermediate range monitors in the temperature and pressure limits of the RHR system. An control rod withdiawat hkick instrumentation surveil-interkick prevents the valves from being opened when the -
lance requirements.
RCS pressure or the RilR pump pressure is above the
. RilR system design pressure A second interkick, the Findings autoclosure interlock (ACI), is provided to automatically close the valve if the RCS pressure increases above a set -
There are two different calibration requirements for pressuw.
'These instrumenu show little set point drift.
Gas & Illectric Company (PC&li) requested staff concurrence with its 10 C17R 50.59 cvaluation which de-Recommendation termined that removal of the RHR autoclosure interkick function does not cogstitute an unreviewed safety ques-The calibration interval for the HWR SRMs and IRMs tion or require modification of Technical Specifications.
should be changed to once each refueling interval.
The NRC staff focused its review of this issue on the 3l NURiiG-1366
5 Instrumentation effect that the propos6 change has on the livent V (in-plant-by-lilant basis because of numerous plant specific tersystem LOC A outside of the containment) sequence.
differences. The PG&li submittal (Shiffer,1987a) con-tains such a plant-specific analyses.
The staff reviewed the pmbabilistic risk assessment (PRA) submitted by the thablo Canyon licensee to sup.
1he Westinghouse Owners Group, however, sub.nitted, port this propos:d and also explored altcmatives to total and the NRC staff approved, a generie proposal for the removal of the ACI circuitry, deletion of the ACI(WCAP-11736-A).
Resolution of GI-99,"Lossof RilR Capability in pWRs,"
.lhe proposal, as agreed to by the NRC staff, will consist recommends (but does not require) that the ACI be re-of hardware and procedure mothfications' moved.
pG&li proposes to remove the autoclosure interhick Other licensees should review their safety bases for the function from the RilR system suction valves 'the open autoclosure interkick of RilR isolation valves to deter-permissive interhick will remam intact. An alarm will be mine if removal of the ACI results in a lower core melt udJed to cath valve which will actuate if the valve is open frequency and, if it does, should propose removing this and if RCS pressure is above a value set between 390 psig requirement from their Technical Specifications, and 450 psig,which is above the RiIR valve open permts-sive set point. In addition, the status lights on the opera-Findings tor's panel, which indicate that these valves are open or Analyses show that removal of the autoclosurc inter-closed, will remam functional after power has been re-moved from these valves.
lock decreases the core melt probability.
The licensee also developed improved operating proce-
'lhe Westinghouse Owners Group submitted a ge-e dures to assure closure of these valves.
neric proposal to t;emove the ACI and it was ap-proved by the NRC staff, The main reasons for removing the ACI are given in a Resolution of GI-99. "1.nss of RHR Capability in e
report from the Office for Analysis and Evaluation of PWRs," recommends (but does not require) that the Operational Data (AEOD) concerning decay heat re-ACI be removed.
moval problems at U.S. pressurized water reactors (A!!OD/C503). 'this report points out that, of the 130 Rec mmendation loss-of-RI1R events that were documented at U.S. PWRs between 1976 and 1983,37 were caused by the automatic Other owners groups besides Westinghouse should de-closure of the suction / isolation valves.The AEOD repet termine if the core melt risk is higher or lower with the also quotes a study prepared by Sandia Laboratory for the ACI deleted.
NRC (NURIIG/CR-4335) which evaluated the compet-ng risks associated with RI1R suction / solation valve clo-sures and Event V. Sandia concluded that:
5.13 Turbine OversIiced Protection-System Testing (PWR, BWR)
The lowest core melt frequency due to the
.h presem Technical Specifications for a pWR require a combination ofloss of RIIR suction durm.g cold shutdown and V-LOCAs is obtained when turbine overspeed protection system to be operabic in there are no autoclosure interlocks on the L1he system must also be operable in Modes 2 and a
RilR suction valves... removing the overpres-3 f all steam paths to the turbine are not isolated.
sure interlocks from the RilR suction valves
'the Technical Specifications for the turbine overspeed gives the best RilR suction arrangement for protection system in a BWR are equivalent; they require
. l'WRs based upon this analysis....when mter' the system to be operab!c in Opersing Condition 1 and. if -
hicks are present, loss of RilR suction is the the MISIVs are open, in Operating Condition 2.
i.
largest contributor to core melt frequency.
l:
llowever, when the interkicks are not presenti 1he Technical Specifications " split report" has desig-
- the core melt frequency due to loss of RilR nated this technical specification as one that may be relo-suction is comparable to or less than the V-cated from the Technical Specifications. However, it has 1.OCA core melt frequency for the best esti-become apparent in performing this study that, because mate cases.
of the problems caused by the turbine valve testing, faster action is called for. There are sescral reasons for this:
The AliOD report concluded that, even though it was most likely a good idea to remove the ACl, the effects of First, this surveillance has caused a significant number of ACI removal upon plant safety must be evaluated on a reactor trips.
l NURhG-1366 32-l.
i
-. ~
. -. - -. - - - _ _ - ~ - - - -. -. - -
l 5 Instrumentation 13 Second, the survedlance results in some wear to the The concern in this case, novmcr, is not burden on the valves and stress to the steam system.
licensee but the number of vertent trips and wear on the equipment.
i
' third, while the test is being donc, in order to avoid a Turbine overspeed protection is typically redundant and reactor trip, the steam flow to the turbine must be re-diverse. If a turbine accelerates from its normal speed duced. 'llus is done by either reducing reactor power' (c.., due to a change in load) the normal turbine control which results in a reduction of capacity factor, or by system positions the control (or governor) valves to rap-dumping steam to the condenser, which could be detn-idly return the speed to the proper set point.
mental to the condenser because of the damage done by vibration of the condenser tubes when large amounts of in addition to this, therc is typically both mechanical and steam bypass the turbmc and enter the condenser.
electrical overspeed protection.1hese systems are de-scribed in Section 10.2 of each plant's final safety analysis In addition, because power must be lowered to perform report.
this test, the test becomes very difficult to do at the end of gele when there may not be enough dilution capability (in Mechanical overspeed protection, which is independent a PWR) or rod worth (m a HWR) to os erride xenon.
of the electrohydraulic controller (EllC), is providc<J by the mechanical overspeed trip mechanism that is set to achate at Oypicah) N of rad spd The surveillance testing requires moving each of the tur-bine valves through one cycle (from the valve position Illectrictd overspeed protection, which is set m (typically) prior to the test, to fully closed, and returning to the 111.5% of rated speed, is provided as a backup to the origmal position). The test is typically performed by a mechanic;d overspeed trip device. The elect rical trip sole-control room operator with an observer at the valveflhe id valves are deenergized to trip the turbine upon re-test verifies freedom of movement of the valve compo' ce v ng an open contact from the IIHC. which represents nents; that is, it verifies that nothing inhibits the valve an overspeed condition.
from closing. Wesunghouse has stated (WCAP-11525) that "[t]his type of testing is beneficial for (1) detecting In addition, each steam line to the turbines typkally has a
non or slugptsh operation of the ndves, and (2)identdica-two valves in series so that a failure of a sin e valve tion of gross outward appearance of valve condition.
cannot cause the turbine to overspeed.
The surveillance requirements for valve operability en.
'the present requirements for the test frequency-arc surs that all the turbine steam inlet valves are capable of based on recommendations from the turhine vendor closing to protect the turbine from excessive overspeed.
(i~rammel,1987). This test interval was developed for Protection from excessive turbine overspeed is required fossil units and carried over to nuclear units due to the since excessive ove. speed of the turbine could generate similarity in design. Fossil units produced steam with potentially damaging missiles that could hit and damage much greater particulate (impurities) content than is per-safety-related components or structures.
mitted in nuclear units. These impurities required more frequent valve surveillance to ensure reliable operation.
A, fod uns anh dy M us utilized phosphate Thus, the test is beneficial in some ways and serves a c mistry in their condensate. The phosphate-based t
safety function.
chemistry control contributed to valve inoperability.1 or example, Turkey Point Plam, Unit 3 had stuck open tur-The issue here is not whether the test should be done, but bine stop valves which were found during shutdownJihe rather how often it should be done.
cause was phosphate' deposits between the shafts and bushings ci the valves (NPE-2).
As discussed earlier (see Section 3) turbine overspeed A review of the operating history of turbine valves protection system testmg ts a major source of reactor trips (NPE-2) shows that failures of these valves do occur and that occur durmg or as a result of surveillance testing. In that, uniike some other Technical Specifications tests i
addition, heensees lune stated that this test is hard on the which do not find failures, these failures are found during steam system, causing relief valves to hft and adding ther"
.ac turbine overspeed testing. Ilowever, the number of
. mal and mechanical stresses to the piping' trips attributable to this testing is significant.
Personnel at the San Onofre site estimate that perform-The NRC has granted increa,es in the testinynttrud for
-ing the weekly test takes a crew of technicians approxi-turbine overspeed protection testing. For example, in a mately 3 hours3.472222e-5 days <br />8.333333e-4 hours <br />4.960317e-6 weeks <br />1.1415e-6 months <br /> (per unit). 'Ihe monthly surveillance re-letter dated. April 16,1984 (Engle,1984), the staff ap-
. quires approximately 6 hours6.944444e-5 days <br />0.00167 hours <br />9.920635e-6 weeks <br />2.283e-6 months <br /> per unit, proved changing the requirement for cycling,cach of the 33 NURIGl366 a,.
5 Instrumentation
- turbine valves from every 7 days to every 31 days for North The capability to source check provides an integral verifi-Anna Power Station, Unit L cation of the response of the detector. Ihis is generally -
required monthly or before using a system that would Another factor in this case is the turbine manufacturers' release potentially radioactive fhud.
recommendations about the testing frequency of these valves. In some cases, these frequencies are comparable The testing of radiation monitors produces a significant to the existmg rechnical Specifications.
number of isolations of the control room, fuel handling building, auxiliary buildings, and various process lines. In The NRC staff recommends that, where the turbine addition, the testing requires significant licensee staff.
manufacturer agrees, the testmg m, terval for turbme Licensees also stated that the frequent testing tends to valves as part of the turbine overspeed protection system degrade the equipment, %e instrumentation must be suncillances be extended from weekly and monthly ter.s removed from cabinets and reinserted. A majority of the to one test done quarterly,in which a direct visual obser-instrumentation is scif-checking so that most failures will vation will be made of the movement of each of the be found in this way or by channel checks.
turbine valves currently required by Technical Specifica.
Lions to be tested.
The San Onofre licensee proposed that the surveillance A quarterly test corresponds to the most stringent valve test intervals for radiation monitors be extended. De-testing icquirement of the ASME Code.
tailed information on failure history was provided for some monitors. In addition, the San Onofre licensee made the following points in support of the extension of FindinP the surveillance test interval: The radiation monitors at San Onofre have had a minimal failure history. The fail-
. Turbine overspeed testing reWres a reduction in e
urcs are normally of remote meter indication which would power and is a mam cause c tor tnps during not impact the safety function, In addition, channel testmg.
checks and failure alarms would detect failures that re-e Testing of the turbine valves is necessary and the manufacturers' recommendations should be fol.
Iowed.
The extent to which these points are generic has not been determined as part of this study, it is, therefore, difficult
'ltecommendation to take the San Onofre experience and design and to-extrapolate it to other reactor sites.His appears to be a Where the turbine manufacturer agrees, the turbine situation in which reliability-based Technical Specifica-valve testing frequency should be changed to quarterly.
tions surveillance rcquirements could be utilized to de-crease the frequency of surveillance testing on reliable radiation monitoring systems while requiring more fre-5.14 Radiation Monitors (PWR,llWR) quent testing on radiation monitors that are not as reli--
able.
c He Technical Specifications contain three categories of rudiation monitors: those used for raseous and liquid effluent monitoring, those used for monitoring an area It does seem reasonable to give relief on the frequency of and indicating the radiation level, and those that are part channel functiona,1 tests since these tests do not involve L
of the reactor protection system and engineered safety the sensor (radiation momtor)itself Therefore, m order to decrease the licensee burden and increase the avail-r-
features actuation systems. The only radiation monitors with a reactor trip function are the main steamline radia-ability of the radiation monitoring systems, the NRC staff tion monitors on HWR main steam lines.The engineered recommends that channel functional tests on radiation safety features actuations are basically isolation functions m n@g ep)pment be performed quarterly. Forsome and air c!canup functions. Many radiation instruments r diation momtormg equipment, this surveillance is perform a monitoring function; these instruments moni-aheady donc quarterly, Channel checks, source checks, tor for reactor coolant leakage, accident conditions in and calibrations would be done at thett existmg surved-lance test intervals.
containment, and the release of gaseous and liquid
- effluents.
The NRCstaff also recommends that the vendor owners As with other instrumentation, radiation monitors are groups study the reliability, set point drift, failure modes, required to undergo three types of surveillances: a chan-and alarm capabilities of radiation monitors (with industry nel check, a channel functional test, and a calibration. In -
participation) to determine if further decreases in testing addition, a source check is performed, and calibrations are possible.
NURl!G-1366 34-
5 Instrumentation Findings dard may be excessive for one vendor's detectors, this Radiation monitor testing appears to require a large e
amount of resources.
Secondly, the NRC depends on the accuracy of the licen-secs' reporting of releases of effluents frem the site. In e
Most failures of radiation monitors can be found order to have confidence in these releases, a reliable from channel checks, source checks or alarms.
c;dibration standard is necessary.
1here is a large variation in the type and reliability of Individuallicensees may be able to justify a orogram for s
radiation monitoring equipment among utilitiet effluent monitoring instrumentation which does not in-clude this requirement but, on a generic basis, this re-Recommendations quirement appears to be necessary, e
in order to decrease hcensee burden and mercase the availabihty of radiation monitors, change the Finding monthly channel functional test to quarterly.
NHS calibration standards are necessary for effluent ng nementanon kcauw of tM vahy M in-nm The vendor owners groups should study whether struments used and the need for accurate measurements e
further reductions in radiation monitor surveillance c uen m activity.
testmg are possible.
Recommendation 5.15 Radioactive Gas Effluent Monitor Ret in th:s requ.irement.
Calibration Standard (PWR, InVR)
Sil6 Intermediate Range Monitor and HWRs and PWRs are required by theirTechnical Specifi-Average Power Range Monitor cations to calibrate noble ps activity monitors at refuel-ing. Some of these llWR Technicai Specifications have Channel F,unct,ional Tests (BWR) the following note attached to this requireme it.
IRM and APRM channel functional tests are performed 1hc initial channel calibration shall be per-every 7 day s, w hile all other RPS channel functional tests formed using one or more of the reference are performed once every 31 days, radntctive standards certified by the National llureau of Standards (N HS) or using standards in the time available, the NRC staff could not determine that have been obtained from suppliers that the reason for this dtfference. The NRC staff should participate in measurement assurance activi-d scuss this difference with the HWR Owners Group to ties with NHS. These standards shall permit determine whether there is a niid basis for this differ-calibrating the system over its miended range ente, if justified, the surveillance interval for the IRM of energy and measurement range. For subsc-and APRM channel functional tests should be changed to quent channel calibration, the initial reference radhiactive standards or radbactive sources every 31 days.
that have been related to the initial calibration shall be used.
Findings e
IRM and APRM channel functional tests are per-This requirement is newed by some in the industry as excessive since the equipment vendors supply informa.
formed every 7 days while all other RPS channel tion or kits for calibrating monitors.The requirement for functional tests are performed every 31 days.
an Nils standard makes instrument calibration more ex-The reason for this dif ference was not determined as pensive. A search of plant Technical Specifications shows e
that the NRC staff has not been requiring an NHS stan-part of this effort because of time restraints.
dard for cahbration consistently, but there appears to be a reasonable basis for the requirement' Recommendation First, there is a great variability between radiation detec.
The HWR Owners Group should determine tf the 7-day tors at different sites and the cahbration standards and requirement for channel functional tests on IRMs and procedures of these detectors. Thus, while an NHS stan-APRMs can be extended.
35 NUREG-1366
t 6 REACTOR COOLANT SYSTEM 6.1 Reactor Coolant System Isolation ciuding PORVs, block valves, and their control systems.
Mechanical failure or degradation accounted for 101 of ;
yggygg [pgg) the events; of these,91 were attributed to the PORV c ntr 1s. Of the 101 events,32 involved block valve fail-The StandardTechnical Specifications (STS) require that I which 4 mvolved block valve controls. Since urcs reactor coolant system isolation valves be tested " prior to PORVs and block valves were not considered safety re-entering Mode 2 whenever the plant has been in cold lated, their failures were not always reported (especially shutdown for 72 hours8.333333e-4 days <br />0.02 hours <br />1.190476e-4 weeks <br />2.7396e-5 months <br /> or more and if leakage testing has before 1979). Therefore, this may not be a complete ac-not been performed in tLe previous 9 months."
counting of failures.
The qualitative risk analysis and discussions wit h license es of pressurized water reactors (PWRs) indicate that the The breakdown of failures between types of PORVs is not surveillance requirements are burdensome and ctm result uniform. Habcock & Wilcox (B&W) plants with Dresser /
in occupational exposure; equipment reliability charac-Crosby PORVs accounted for 45% of the PORV me-teristics suggest longer surveillance test intervals (STis),
chanical failures. Also, more than 70% of the reported in addition, this STS requirement causes plant personnel fadures or degradation of PORV controls concerned to rush to recover from short forced outages.Thus, longer problems with the air / nitrogen control components re-surveillance test intervals are recommended.
quired to operate the air operated PORV.
In the judgment of the NRC staff, making the conduct of With this background, it appears that the current sttrveil-this survedhnee contingent upon being shut down 72 lance requirements in the Westinghouse Standard Tech-hou rs or more has a pote ntial for causing problems result-nical Specifications for the PORVs and their block valves ing from a hurried recovery. It is very likely that a plant are reasonable in that compliance with the Boiler and could have been operating satisfactorily for more than 9 Pressure Vessel Code of the American Society of Me-months without difficulty, and yet must delay restart after chanical Engineers (ASME Code) will check set point and a short shutdown in order to satisfy this surveillance re-leakage while the other Technical Specifications require-quirement. in addition, extending the STI associated with ments will verify operability of the control system. Note 4
I this surveillance does not signif;cantly alter the associated that the Standard Technical Specifications do not specifi-
- risk, cally require a surveillance of the air / nitrogen emergency backup system (such as accumulator pressure). The STS
(
The NRC staff recommends that the 72-hour time for do require that the emergency power supply for the i
remaining in cold shutdown withor 3csting the RCS iso-PORVs and block valves shall be demonstrated operable -
lation valves for leaks be increased to 7 days. This will every 18 months, help utilities perform repairs under less stress, but will make an insignific2mt difference in risk' He NRC staff has adopted the position that the pres-surizer power-operated relief valves (PORVs) should be l'inding
. included in the inservice testing (IST) program as Cate-gory B valves and tested to the requirements of Sec-RCS isolation valves are to be tested for leaks if the tion XI of the ASME Code. However, since the PORVs reactor is m cold shutdown for more than 72 hours8.333333e-4 days <br />0.02 hours <br />1.190476e-4 weeks <br />2.7396e-5 months <br /> and the hve shown a high probability of sticking open and are not valves have not been so tested m the last 9 months' needed for overpressure protection during power opera -
tion, the NRC staff has concluded that routine exercising Recommendation during power operation is "not practical" and therefore not required by IWV4412(a).
Increase the 72-hour time for remaming in cold shutdown without leak testing the RCS isolation valves to 7 days.
The PORV function during reactor startup and shutdown is to protect the reactor vessel and coolant system from
().2 Power (or Pilot)-Operated Relief low-temperature overpressurization conditions and the -
staff has concluded that these valves should be exercised Valves and Illock Valves (PwR) before untiation of system conditions for which vessel NRC report NUREG/CR-4692 discusses a survey of the low-temperature overpressure protection is needed. (See operating history of power (or pilot)-operated relief Section 6.4 of this report.)
valves (PORVs) and bkx k valves. The survey included
- 230 failures occurring between 1971 and mid-1986 in-The followirg test schedule is required:
I NUREG-1366 36
(1) Full. stroke exercise should be performed at each
(,.3 Iligli Point Vent Surveillance cold shutdown or, as a minimum, once cach refuel-rggj g
ing cycle.
'ihe Westinghouse Standard Technical Specifications re-(2) Stroke timing should be performed at cach cold shut-quire each RCS vent and bkick valve to be demonstrated down or, as a minimum, once each refueling cycle.
perable every 92 days by operating the valve through one complete cycle of full travel from the control room.
(3) Failsafe actuation testing shculd be performed at Westinghouse plants have a problem testing the bk)ck each cold shutdown.
valve upstream of the vent valve. When the bkick valve is opened at pressure, a pressure surge opens the vent valve rcsulting in a release of reactor coolant."Ihe discharge is (4) The PORV block valves should be included in the routed to either the pressurizer relief tank, a containment 150 program and tested quarterly to provide protec-ventilation system, or directly to the containment atmos-tion against a small-break loss-of-coolant accident phere.
(I OCA) should a PORV fail open.
A review of operating history (NPRDs and IJiRs) re-lance was being performed.g meMe 18 meL na nly one senous eve It should be noted that even this testing does not ensure this was at the Shearon llar-operability. During the steam generator tube rupture ris Nuclear Power Plant on October 9,1987; the RCS event that occurred at Ginna Nuclear Power Plant in leakage Technical Specifications limit was exceeded and January 1982. the PORV was used to reduce RCS pres' an Unusual livent was declared (see IJiR 87-058 and sure. ihe PORV opened and closed successfully three NPii-5). Ilowever, there have been numerous incidents times. On the fourth openmg/ closing cycle, the valve par-of smaller discharges.
tially closed, then opened completely, and stuck. The bkick valve was successfully closed to terminate the RCS blowdown.
S me Technical Specifications do not require testing these valves while at power. The Catawba Tecimical Specifications only require this surveillance during cold The most common reason for mechanical failure in shutdown or refueling, at least once every 18 months.
PORVs, according to NURl!G/CR-4692, appears to be the degradation of the scat / disc interface or other inter-The NRC staff recomraends that the licensees withTech-nal parts by high pressure steam and/or water. Internal nical Specifications requiring 92-day surveillances of RCS leaking was the most common failure mode. It therefore vent valves evaluate the requirements that appear in the -
appears that preventive maintenance is important for Catawba Tec.inical Specifications and revise their Tech.
these valves to assure that they do not degrade to the nical Specifications, if applicable, point where they fail to operate or seal.
~
This is an example of a design problem (the valve's ten-dency to open on a pressure surge) making a surveillance Findings test difficult.
Operational use of PORVs is covered under resolu-Findings tion of Generic issues (Glsb70 (PORV Reliabihty)
Testing of RCS vent valves at pressure has the po-and -94 (Additional 1. TOP for 1_WRs).
tential to cause a release of reactor coolant.
Not every plant with PORVs has PORV limiting Testing is required every 92 days by Westinghouse conditions for operation (LCOs) and surveillance Standard Technical Specifications and some plant-requirements in the Technical Specifications.
specific Technical Specifications. Other plant Tech-nical Specifications only require testing these valves at cold shutdown or refueling.
Recommendation Recommendation Direction concerning PORV and bkick valves surveil-lances will be provided a the resolution of GI-70 and I.icensees to evaluate applicability of Catawba Technical GI-94.
Specification Bases with respect to high point vent 37 N URIiG-1366
6 Reactor Coolant System surveillance testing and revise the frequency of testing of when used for overpressure protection is that they shall RCS vent vahes to cold shutdown or refueling if appro-be verified to be open at least once every 12 hours1.388889e-4 days <br />0.00333 hours <br />1.984127e-5 weeks <br />4.566e-6 months <br /> except priate.
when the vent pathway is provided with a valve that is locked, sealed, or otherwise secured in the open position.
'lhen these valves shall be verified open at least once 6.4 In,1,emperature Overpressure every 31 days.
Protection (PWR)
Also, in the emergency core cooling system (ECCS) sec-The Standard Technical Specifications recognize two tion of the Standard Technical Specifications, it is re-methods of protecting a reactor vessel against low-quired that all charging pumps and safety injection temperature overpressurization.' These are:
pumps, except the required operable pumps (centrifugal chcrging pump and RilR pump), shall be demonstrated (1) Two power operated relief valves (PORVs) with a inoperable by verifying that the motor circuit breakms are hft setting of less than or equal to [450]" psig, or secured in the open position at least once every 12 hours1.388889e-4 days <br />0.00333 hours <br />1.984127e-5 weeks <br />4.566e-6 months <br /> whenever the temperature of one or more of the RCS (2) The reactor coolant system (RCS) depressurized cold legs is less than or equal to [275)*F.*
with an RCS vent of greater than or equal to[ }"
square inches.
An event at Millstone Nuclear Power Station, Unit 3 on January 19,1988, illustrates the need for frequent surveil-l In addition, some Technical Specifications allow the lance of the low temperature overpressure protectior' R H R system relief valves to be used to protect the reactor system and the need for positive continuous indication vessel against a low-temperature overpressurization.The that the low-temperature overpressure protection system Standard Technical Specifications require:
is operable.
(1) Performance of an analog channel operational test The Millstonc 3 Technical Specifications required at least on the PORV actuation channet, but excluding valve one of the following for low-temperature overpressure operation, within 31 days prior to entering a condi-protection when in Mode 3 with the temperature of any tion in wnich the PORV is required operable and at RCS cold leg less than or equal to 350*F or when in least once per 31 days thereafter when the pORV is Modes 4,5, and 6 with the reactor vessel head on, required operabte; (1) Two residual heat removal suction relief valves each (2) Performance of a channel calibration on the PORV with a set point of 450 psig, or actuation channel at least once per 18 months; and (2) Two power-operated relief valves with lift settings (3) Verifying the PORV isolation valve is open at least which do not exceed the limit established in Techni-once per 72 hours8.333333e-4 days <br />0.02 hours <br />1.190476e-4 weeks <br />2.7396e-5 months <br /> when the PORV is being used for cal Specification figures of pressure set point as a overpressure protection, function of temperature, or
'the NRC staff recommends:
(3) The reactor coolant system depressurized with an RCS vent of 5.4 square inches or more.
(1) That licensees consider verifying that the PORV is On January 15,1988, Millstone 3 was in Mode 5 in water-not isolated at least once pn shift when the PO"Vs solid operation at 350 psia and 125'F.Two trains of RilR are used for low-temperature overpressure protec-were tn operation. The cold overpressure protection sys-tion.
tem (which uses the PORVs with a low-pressure set (2) That licensees consider the benefit of providing con.
point) had been erroneously deactivated in November
(
tinuous positive indication that the low-temperature when the solid-state protection system had been removed overpressure protection system is armed and oper, from senice at the beginning of the outage, but the licen-see had not detected this condition.
able.
On January 16,198&, the A train of RHR was removed Recommendation 1 is more consistent with the degree of from service for maintenance. Ihis put the licensee in an conservatism used in other sections of the StandardTech-8 hour action statement due to the non-operability of nical Specifications for overpressure protection. For in-stance, the surveillance requirement for the RCS vents, the cold overpressure protection system anC only one train of RilR (and consequently one relief valve)in op-erauon. However, the operator assumed that the PORVs
- tow tempruure overgenure bminiion n required in Mate 4 when the temperature M any h S coM leg is leu than or equal to were still operable as low-temperature overpressure tus}* r", and in MWe $ and Mode 6 nh the rextor vevel head protection.1here was no positive indication that the low-
"EiInugeme 61ues are uwd
- A plantapccific ulue wouki tie used.
NURiiG-1366 38 i
,33)A
,p %%
do v:a s
4
/*/f
$ 4+
% % *1 '
IMAGE EVALUATION
. N kfVj ? i %
c,/}f[& ~ (g TEST TARGET (MT-3) 49
,jjg!
N+.
%s, i
'w 49 sc v
($
th,
4 e
- W
$1 10
<.a e
,r au s.
Oh 1.1 e
1 M i.8 i!!!==8
._ _1.25l=L4 g=t6
=
n s
i 4
150mm w
6" e
4
<e* +
- s 4,, 5% ;f Q.g~@4.X
,:47,y,m 4' 6,qs s
>p '
p>
46'i O/
/
r en i
-1 4"
I d
r.
d v
..... m - t,4
4;. @
.q) %
IO e w
"#7" q.
8 Qf
%M IMAGE EVAL.UATION f
(+'#'vd9' TEST TARGET (MT-3)
). Y j <:f,,'f?,p,4(e d
\\
-,g, e
Y,4$$f' kk NV (1, r 28
,12.5 1,0
. A u-t r
L L3 2,
!!!22 L.
lute
- 3. y t._
E.,
I $ 1,.
l j,l h
{lll! l.8
~
Ili!!as a I.25 1.4 L 1.6 tm 4
150mm 6"
f!'
% bga g> */A eh h/,%o/
.,u--
E' p,t.n/fN a
.i +'
4.g e
/
ep,7js o%
e 2/
t g+
q
/
,1 t;
i bewivut i-
...at.
sa
6.
'b' p
ph 0
9 '49
.g%
g 4
51:
IMAGE EVAL.U ATION
/[ j
' di tf 9777 y' ~9, TEST TARGET (MT-3) qft,h
%of>t49' Alp /k4i*
y' ge;-
f 4
in 28
- 2,5 r
1.U i ' 2, 22 4.,sa u,,
a
,m
!Ui 2.0 11 hzu HIl 1.8 ilillE==
' l.25
[
I.4 91L.6
! re =
4 150mm 4--
6" b
+;b ///;*%
%y %;>gy N
6
,)f
_yg,[4 gy%,f9-y s
7.g
~
7;7
- x- - ; -- - - - -
Q%%
+
~<
m x
3gy b(
- M@y 6' Reactor Coolant System y
n,>
e rp,,
4 gy Ltemperature overpressure protection system was not op-overpressure protection system is armed 'and oper-g ~, :crable.
able. His indication should be available whenever r
~
the low. temperature overpressure protection sys-
& ~ Subsequently,:a maintenance activity on January 19 tem is required to be operable, 1 caused inadvertent closure of the suction valve for the p,
Implement resolution'of GI-70 (PORV Reliability) remaining operable RHR train, train D.ne reactor cool-s Lant system was in solid condition with both charging.
and GI-94 (Additional LTOP for LWRs).
. pumps and letdown in operation, ne letdown path was pC
. through the H train R11R pump.
6.5 Specific ActMt_,yof the Reactor l
(
LER 88-005 describes the event as follows:
]
'l At 10:55 the RilR low flow alarm was received The Technical Specifications stirveillance requirements 1
1 at the Main Control Board. He operators for specific activity of the reactor coolant vary significantly C
noted that the RIIR suction valve was going among different plant Technical Specifications.
6
' closed and stopped the operating RilR pump.
He reactor operator reviewed the condition of The Technical Specifications for both PWRs and BWRs
^
the plant and noted that pitssure was rising require that the specific activity (microcuries per gram)be 3
~3
. rapidly due to the charging system operatingon determined periodically by measuring two quantitles:
a solid plant. Within two minutes, the operator dose equivalent iodine-131 and the quantity 100/E,
manually isolated charging and increased let-F
. down through the chemical and volume control The design. basis accident which sets the limit for coolant -
system. The pressure' increase immediately activity in a'PWR is the steam generator tube rupture.
stopped and' turned. Within -15 minutes the The limit 'on dose equivalent iodine-131 assures that ra-
= transient stabilized with reactor coolant pres.
diation dose to the thyroid will be within the limit of.
- sure below 310 psia.
30 rem.
ne peak pressure was 535 psig, which was above the E is precisely defined in the " Definitions'f section of the Technical Specifications limit of 500 psig. lind a more Technical Specifications. It is basically a weighted sample
' effective and more frequent surveillance of the status ui (by concentration) of the' sum'of the be.ta and gamma-
< the cold overpressure protection system been required,-
energies per disintegration of each radionuclide in the j
.the Millstone 3 operators should have noticed that the reactor coolant.
P 2 system was not operable, ne Millstone 3 licensee has-In calculating the effects of offsite dose, both the dose to.
=
corrected procedures to assure better surveillance of the i ow. temperature overpressure protection system before the thyroid _and to the whole body are considered. The 1 l
it is required.-
dose to the thyroid is limited by thc doseTequivalent'
+
- iodine-131 limit.The whole-body dose is limited by limit-
. ing the concentration of noble gases since the assumption p,
, Fin' dings
-is made that an individual at the site boundary is im.
mersed in a cloud of released activity which is completely
- Low temperature overpressure protection is impor-3e i tant to safety.
made up of. noble gases. This limit on_ noble gases is o
100/E.
g U'
- I PORV isolation valves are to be verified open every De whole-body dose for the steani generator tube rup- -
72 hours8.333333e-4 days <br />0.02 hours <br />1.190476e-4 weeks <br />2.7396e-5 months <br />. His may not be often enough.
ture accident is th'e dose resulting from immersion in the Q
cloud of released activity. To insure that.the public is-4 Some PWRs have no positive contmuous indicat. ion 4
s e adequately protected, ' he specific activity of thc reactor t
that the low-temperature overpressure protecten coolant is limited ~to a value that will insure that the-1 i system is cperable, whole-body dose at the site boundary will not exceed 2.5 '
4
.z rem should a steam generator tube ruptui;c accident oc--
5 Recommendations -
-cur, e! Licensees should consider verifying that PORV iso-ne dose is calculated from the equation
- 1ation valves are open at least once per shift.
Dose = 0.25 EAV /O X
P t o j Licensees should consider the benefit of continuous -
, positive; -indication, that1 thei ow. temperature where l
m 39L NURiiG-1366
_m
=
E - average beta and gamma energies per disinte-cessive activity levels in the reactor coolant will be gration (Mev)* -
detected in sufficient time to take corrective action,"This A - reactor coolant activity ( Ci/ml) is true for gross beta and gamma and dose equivalent
. h is not tm for P-V - reactor coolant volume which leaks !nto secon-dary system (rn3) in addition to these problems, it appears that the industry x/O - atmospheric dispersion coefficient at site does not have a clear idea of the purpose of E. In some boundary (sec/m3) cases, chemists have stated that they have determined the gross activity in the primary coolant by boiling a sample to Using the 2.5-rem whole-body limit, and the plant-dryness and analyzing the residue, lloiling to dryness, of X O, A is given by course, drives off the noble gases and such volatile mate-specific values for V and
/
rials as radiciodines. He residue is dissolved solids, pri-A
- Constant marily corrosion products such as cobalt-60. In other cases, the coolant sample is degassed before measure-ment; there is then no noble gas left to me sure.
For 13WRs, the limiting accident in terms of offsite dose is The NRC staff recommends that industry (perhaps the steamline break outside containment. De same through the Electric Power Research Institute) derive a equation is applicable; ho;cever, V is the total volume of m te meanmgful measurement of noble gas activity. In, the meantime, the NRC should assure that the 100/E
" reactor coolant released to the atmosphere, limit is conservative for each plant and that it is being Values for the cor.stant vary considerably. A value of 100 determined m a meaningful way.
is now used as a standard and the Bases section ofTechni-cal Specifications states that this value is conservative Findings since it is based on a parametric evaluation of typical site conditions. It is not cicar that this is true since some older ne quantity 100/E is a limit on the accumulation of e
Techmcal Specifications use values lower than 100.
nob!c gas activity.
Practically, there are several difficulties with E. He It is time consuming to determine the quantity.
Technical Specifications require that it be determined every 6 months, but its valre vanes contmually. Gross He industry does not appear to understand the e
- beta and gamma samples are required by some Techmcal quantity well.
Specifications to be taken every 72 hours8.333333e-4 days <br />0.02 hours <br />1.190476e-4 weeks <br />2.7396e-5 months <br /> as an indication It is not clear that the quantity is conservative.
e of the total (pnmanly noble gas) radioactivity. In a PWR, the amount of noble gas in the system is, to some extent, a Recommendat, ns
. function of the letdown (cleanup) flow. In a llWR, noble m
gases are being continually released as offgas u p the stack.
He NRC should ask tha industry (perhaps through ~
Thus, in a 11WR, there is no significant accumulation of the Electric Power Research Institute) to derive a -
i noble gases, more meaningful measurement of noble gas activity,
-l possibly replacing the 100/Elimit.
He Techm. cal Specifications surveillance requirements for specific coolant activity require that the 1-131 value be The NRC should assure that the 100/Elimit is con-e obtained whenever there is a " problem" (> 15% power servative for each plant and that it is being detere
. change in one hour, offgas increase, etc.). his value can mined in a meaningful way.-
be obtained fairly quickly. Usually within 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> all of
- the radioactive iodines can be analyzed (usually involving i
an initial count and a 1-day decay count). To accurately 6.6 Presstirizer Heaters (PWR).
obtain an Evalue ordetermine if the 100/Elimit has been
)
exceeded, the sample would need to be held for further he Standard Technical Specifications fora PWR require
' decay counts It could be a week before the value is that the pressurizer heaters that are to be powered from known. This contradicts the statement in the Bases sec.
an emergency power supply should be energized to meas-tion of some Technical Specifications that "the surveil.
ure circuit current at least once every 92 days. He pur-lance requirements provide adequate assurance that ex.
pose of this surveillance is to demonstrate that the capac-ity of the heaters has not degraded.
E Most pressurizer heaters are in constant use, both the 1%'fy'("y))e whole-body dme, but is inckded here as proportional and to some extent the backup heaters.
l nerefore, operators should be aware of problems that i
1 NUREG-1366 40
Re operator should be aware of the status of both may arise with pressuriier heaters. In addition, pres-surizer heaters are fairly reliable.
proportional and backup pressurizer heaters during normal operation.
Therefore,92 days appears to be too frequent and the Some Technical Specifications require that the staff recommends that this test be done once cach refuel.
ing interval.
power supply be switched from normal to emergency power for testing every 18 months.
The Standard Technical Specifications also require that Some PWRs have some pressurizer heaters perma-e pressurizer heaters should be demonstrated operable least once every 18 months by manually transferring nently tied to a vital bus.
h power from the normal to the emergency power supply Some PWRs have dedicated safety related heaters.
and energizing the heaters. For some PWRs, the pres-surizer heaters are connected to the emergency bus so that no transfer to emergency power is required. For Recommendations those PWRs that have their emergency pressurizer heat-The capacity of pressurizer heaters should be tested crs powered by an emergency bus dunng normal opera-once each refueling interval for those plants without tion, this requirement should be deleted.
dedicated safety-related heaters.
Some PWRs have dedicated safety-related heaters not
'lle capacity of pressurizer heaters should be tested used during normal operation. In these instances, the every 92 days for plants with dedicated operator cannot be expected to have a grasp of the opera
- safety-related heaters.
tional status of the dedicated safety-related heaters. In these instances, the 92-day test is appropriate.
For those PWRs that have some pressurized heaters permanently tied to a vital bus, no testing of switch-Findings ing between power supplies should be required.
This surmillance requirement shall be changed on a Pressurizer heaters are required to be tested every 92 days to assure adequate heater capacity.
plant speclite basis.
\\
l 41 NURI!G-1366 k
7 EMERGENCY CORE COOLING SYSTEM 7.1 Surveillance of Iloron At mme plants, personnel who perform this check must-Concentration in the wear protectiv cl thing because of radiation from the ECCS.The proposed change would result in a decrease m Accumulator / Safety inj.ection the dose to plant worke s lhe amount of decrease was-Tank / Core Flood Tank (PWR) not determined in this study.
In a letter sent to Duke Power Co. on September 30,1986 In a recent event (July 20,1988) at San Onofre Nuclear -
(Ibod.1986), the staff approved a proposed Technical Generating Station, Unit 3, entrained air caused a water Specifications change to climinate the verification of the hammer m one tram of the high-pressure safety mjection boron concentration in an accumulator after a volume (IIPSI) piping. The plant was in Mode 5 undergomg an increase of 1% or more from normal makeup, integrated leak rate test (ILRT) at the time with the contamment pressurized. Several sump tsolation valves Normal makeup to an accumulator comes from the refu-were not ful;y closed, allowing air to enter the HPSI cling water storage tank (RWST) which is also borated.
suction piping. The pump became air bound; bowever, No dilution can be caused by adding water from this
""C or more slugs of water traveled down the discharge source as long as the minimum concentration of boron in pipmg before the pump was stopped. Normal practice at the RWSTis greater than or equal to the minimura boron San Onofre followmg an ILRTis to vent all the pipng and concentration in the accumulator, pumps so no pumps would be air bound and there would be no air entramed in the pipmg during normal operation.
Finding NRC Information Notice 88-23, " Potential for Gas Bind-ing of liigh Pressure Safety injection Pumps During a Technical Specifications require that the baron concen-Loss-of. Coolant Accident," discusses an incident at Far-tration in the accumulators be sampled to venfy that it is ley 1 in which hydrogen gas accumulated in the piping to within its limits after an increase in volume of 1% or more an HPSI pump. Hydrogen is normally used for the cover from makeup.
gas in the volume control tank (VCT); thus, water flowing through the VCT is saturated with hydrogen while it isin Recommendation the VCr. The dissolved hydrogen will come out of solu-tion downstream of the VCrif the pressure at some point it should not be necessary to verify boron concentration (i.e., elbows or tics) decreases to less than that at the of accumulator inventory after a volume increase of 1%
VCr. The hydrogen gas will not readily go back into or more if the makeup water is from the RWST and the solution. The incident at Farley indicates the need to minimum concentration of baron in the RWST is greater maintain this requirement to verify that the ECCS lines
. than or equal to the minimum boron concentration in the are free of air.
accumulator, the recent RWST sample was within specifi-cations, and the RWST has not been diluted.
Findings Air in ECCS piping could cause a water hammer e
that could damage the piping, rendering the system 7.2 Verification That ECCS Lines Are or portions of the system incapable of performing its Full of Water (Contain No Air) safety function.
(PWR) verifying that there is no air in the ECCS every 31 -
I l
The Standard Technical Specifications require that every days c:m result in a radiation dose to personnel.
l 31 days the emergency core cooling system (ECCS) be A review of operating data showed (as illustrated by e
verified to be fur of water by venting the ECCS pump the San Onofre event) that air entrainment water l
casings and accessible high points.
hammers in PWR ECCS piping occur during shut-l
< lown.
It appears from a review of operational data that air-he P2. ley m.cident has shown that air entrapment -
e becomes entrained in the ECCS during shutdown as a1 result of drain and fill operations but that once the system
' n occur during normal operation, due to H blan-2 has been placed in an operational condition there is no keting and saturation of the VCT.
mechanism for the intmduction of air,in fact, in a PWR there is an elevation head from the RWSTon the ECCS N """"" ' " d * ' I ""
L lines. Therefore, following testing or maintenarice, it is inspections should be conducted for entrapped air in essential to check the system before declaring it operable.
ECCS lines before startup from any condition in which l
l NUREG-1366 42 l
L Y
7 Emergency Core Cooling System -
testmg of the ECCS, performing an llRT, or draining of
'least hot standby within the next 6 hours6.944444e-5 days <br />0.00167 hours <br />9.920635e-6 weeks <br />2.283e-6 months <br /> and the any portion of the ECCS has occurred. The 31-day sur-pressurizer pressure must be reduced to less than veillances to detect entrained air in the ECCS should be 1000 psig within the following 6 hours6.944444e-5 days <br />0.00167 hours <br />9.920635e-6 weeks <br />2.283e-6 months <br />.
retained.
This action statement is extremely stringent since only one hour is. available to fix a problem. This 4
7.3 ' Verification of Proper Valve action statement affects the surveillances required -
LineuEs of ECCS and n the accumulator water level and pressure chan-net in the followmg way, It was noted during discus.
Containment isolation Valves sions at one site visit that accumulator level instru-(PWR, BWR) mentation has experienced downward drift.
Although it doesn't pose a problem to declare a Walkdown of the ECCS and containment isolation valves channelinoperable as a consequence of a channel are required by the Technical Specifications every 31 check (comparison of the readings from redundant 4
days. Although there was some discussion among utility level channels), there is a tendency to require a personnel about the radiation dose incurred in doing containment entry to recalibrate the transmitter these surveillances, the overall opinion was that such that indicates low just for the insurance it provides if walkdowns are desirable, and should be retained.1he the other channel becomes inoperable. This is a NRC staff concurs.
consequence of the time limits of the action require-ments that require a plant shutdown in one hour, Finding which is not enough time to rectify a problem with 1
one instrument. If the allowable outage time were i
Walkdowns of containment isolation and ECCS valves greater with both level instrument channels inoper-are good checks that containment integrity and ECCS able, the containment would not be entered just to operability are maintained.
recalibrate one level transmitter that has experi-R enced excessive drift.
Re ommendation A containment entry with only one instrument chan-
- Retain the requirements for containment isolation valve nelinoperabig would result in an unnecessary expo-and ECCS valve walkdowns to ensure correct valve pos.
sure to radiation.
tion.
The combination of redundant level and pressure instru-mentation may provide sufficient information so that it 7A Accumulator Water Level and m y n t be worthwhile to always attempt to correct drift associated with one mstrument if there were sufficient Pressure Channel Surve,illance time to repair one in the event that a second one became Requirements (PWR) inoperable, Because these instruments do not initiate a safety action, it is reasonable to extend the allowable (1) Each accumulator water leveland pressure channel outage time for them. The staff, therefore, recommends is to be demonstrated operable every 31 days by that an additional condition be established for the specific.
performing an analog channel operational test as case, where "One accumulator is inoperable due to the well as a channel calibration at least every 18 inoperability of water level and pressure channels " in -
months.
which the completion time to restore the accumulator to opcrable status will be 72 hours8.333333e-4 days <br />0.02 hours <br />1.190476e-4 weeks <br />2.7396e-5 months <br />. While technically inop-It is noted that licensees perform channel checks of etable, the accumulator would be available to fulfill its this instrumentation also, even though it is not safety function during this time and, thus, this change stated as a requirement in the Technical Specifica-would have a negligible increase on risk.
tions. Licensees should examine their channel checks surveillance and operational experience to Findings determine whether there is sufficie71 basis to justify Analog channel operational tests on accumulator the extension of the analog channel-operational test pressure and level channels are performed too fre-frequency-quently.
3 The action statement associated with an inoperable (2) One of the action statements for the accumulator-limiting condition for operation (LCO) states that accumulator is too stringent.
with one cold-leg injection accumulator inoperable, Recommendat, ions except as a result of a closed isolation valve, the inoperable accumulator must be restored to oper-Licensees to examine channel checks surveillance able status within I hour or the reactor must be in at and operational history to determine if there is a 43 NUREG-1366
7 l'mergency Core Cooling System basis for justifying the extension of frequency for Recommendation analog channel operational tests for pressure and level channels.
Inspect the containment at least once daily if the contain-rnent has been entered that day, and during the final entry e
Add a condition to the ECCS accumulator 1.C0 for to ensure that there is no loose debris that would clog the the case where "One accumulator is inoperable due sump.
to the inoperability of water level and pressure chan-nels," in which the completion time to restore the accumulator to operable status will be 72 hours8.333333e-4 days <br />0.02 hours <br />1.190476e-4 weeks <br />2.7396e-5 months <br />.
7.6 Verification of Boron Concentration in the Boron 7.5 Visual Inspection of the Injection Tank (Westinghouse Containment Sump (PWR)
PWR)
The Standard Technical Specifications require "a visual
.the boron injection tank (llff) is downstream of the mspection which verifies that no loose debris (rags, trash' charging pumps and is used to provide highly concen-clothing. etc.)is present in the containment which could trated (13,000-16,000 ppm) boric acid to the core follow-be transported to the containment sump and cause re-ing a main steam line break accident.He boron concen-striction of the pump suctions during LOCA conditions.
oMis M is m M vdid Wily m 7 5
,lhis visual, spection is required to be performed for all The llt f has been removed (or made inoperable) at many m
areas affected within containment at the completion of Westinghouse PWRs. For those reactors that still use the ca n inment entry when containment integrity is HIT, verifying the boron concentration can result in incur-ring a significant radiological dose.The bitis kept mixed The NRC staff recommends that this inspection should by circut tion of borated water between the HIT and the be done at least once daily if the containment has been bmc aad storage tanks tn order to keep a uniform tem-entered that day and when the final containment entry is perature to maintam boron solubility. One licensee rec-made (similar to the technical specification on air locks ommended taking the sample at the bonc acid storage for multiple entries). This would reduce exposure to peo.
tank rather than at the Urr to reduce the dose to person-ple doing work in this area by limiting the time spent in net. Licensees should be permitted to do this if they can the containment.
Justify that the mixing is adequate to give a boron concen-tration in the boric acid storage tank which is representa-tive of that in the Brr.
The recommendation would require that a record be kept of all areas visited during the multiple containment en-tries, but this should not be difficult.
Findings This requirement would apply mostly to containment en-
'Ihe Hfr contains highly borated water which is o
tries when at power, hot standby, or hot shutdown (bc.
cause the requirement states "when cont:.inment integ.
mixed constantly with the rest of the boric acid stor-rity is established") so that the entries would most likely age system to maintain a uniform temperature and be made for working on a specific problem in only a few boron concentration.
areas.
Verify!ag the baron concentration of the Urr can Findings result in a significant dose of radiation to plant per-Visual inspections of the containment assure that no e
loose debris will be transported to and clog the sump.
Recommendation Requiring inspections to be done at the completion Measure concentration of boron in the boric acid storage g
e of each containment entry results in a higher dose to tank rather than in the Bff if it can be justified that the
(
people doing the inspections than necessary.
concentrations are th6 same.
NUREG-1366 44
8 CONTAINMENT He test is qualitative. It does not measure the 8.1 Containment Spray System (PWR) flowrate.
Each containment spray system must be demonstrated ne only reported problems with this test have been o
operable at least once every 5 years by performing an air construction related.
or smoke flow test through each spray header and verify-ing that each spray nonic is unobstructed Recommendation He qualitative risk analysis indicates that the system is
%c surveillance interhl should be extended to 10 years, important to risk, especially for those plants in which the containment spray system performs the dual functions of removing iodine and cooling the containment, since the 8.2 Containment Purge Supply and heat removal function is redundant.
Exlutust Isolation V ilves (PWR)
The NRC staff searched for problems involving the con.
%c Technical Specifications s
. gre that each large-tainment spray system that had been uncovered by means diameter containment purge supply and exhaust isolation of this testing (NPE-1). Only three cases were found and valve shall be tested quarterly to demonstrate that the in all three cases the problem involved a construction leakage rate is less than 0.05 La when pressurized to error. These are listed in Table 8.1. (This is not true of pressure P., When the unit is operating, there can be boiling water reactors [BWRs] that have had blocked high radiation in the area where these valves are located containment spray lines, liowever, in a BWR, the con-and the testing then becomes an ALAR A (as low as rea-tainment spray system is not a safety-related system.)
sonably achievable) concern.
This testing gives no quantitative data on flowrates exiting He issue of leak tightness of large-containment purge the nonles. It only verifies that there is flow, which, from supply and exhaust isolation valves with resdient seals was raised by NRC Circular 77-11, "Contamment leakage the operating data, does not appear to be a problem.
Due to Seal Deterioration," dated September 6,1977.
lar d sse theunsaMactmypedormanceof Herefore, the staff recommends that this test interval be resdient seals on containment isolation valves (i.e., they extended to every 10 years, did not pass required leak rnte tests). nc valves had seat materials of neoprene and ethylene propylene.
Findings he causes of degradation (hardening and wear) of the-PWR containment spmy nozzles are checked every seat material were identified as environmental conditions 5 years to verify flow from the nozzles (humidity and temperature) and valve cycling.
Table 8.1 Containment spray nozzle problems at PWRs Plant Date Problem discovered Rancho Seco October 1973 Painters, painting the reactor building, covered up nonles of the reactor building spray system and did not subsequently remove tape from 16 of 199 nozzles. Only 4 had tape over spray openings.
Turkey Point 4 August 1978 While preparing for a spray nonle test, the licensee discovered that the re;tricung orifices were not installed in the branch con-nections from the containment spray headers to the emergency filter spray system. Unit 3's orifices were verified in place.
Farley 1 January 1982 Certain containment spray header nonles were found to be ori-ented incorrectly or positioned on the header incorrectly. Also, two nonles were not installed due to interference. Analysis showed that peak containment temperature would not have been affected 45 NUREG-1366
. w 8 Containment A generic issue (GI Il-20, " Containment leakage Due to Recommendation Seal Deterioration")was established by the staff to study the problem and propose an appropriate testing fre, This requirement should be reta, ed.
m quency. It was recognized during the study of this issue that plant personnel must enter the containment because 8.3 Ice Condenser Iniet Doors (PWR) that is where the leak test connection is located. It was considered necessary to test passive (closed) purge lines According to the Standard Tecimical Specifications for for leaks because of the potential adverse effect of sea-ice condenser containments, ice condenser inlet doors sonal weather conditions on the integrity of the isolation must be:
valves.
Demonstrated operable during shutdown at The resolution of GI 11-20 is reflected in the present least once per 3 months during the first year -
Technical Specifications surveillance requirement. The after the ice bed is initially f Ily loaded and at staff study appears to have been comprehensive.
least once per 6 months thereafter by testmg a E
sample of at least 25% of the doors and vertfy-The study recognized the high dose rate received by work-ing that the torque required to open each door ers doing this test.
is less than a specified amount.
The purpose of this surveillance is to provide assurance Duke Power Co., on the Catawba docket, proposed delet-that the lower inlet doors are capable of opening properly mg the requirement for survedlance on these valves and when required during a loss-of-coolant accident (LOCA) cited plant specific data that showed that these valveshad or other high energy line creak so that the steam released required neither maintenance nor repair (Tucker,1986).
in the lower containmer i compartment can enter the ice The proposal also cited the high radiation dose (3000 condenser and be condensed by the ice inside, person-mrem / test) and the high temperatures associated with this test as reasons for deleting this test.
Ilecause this testing cannot be performed at power, this specification requires a unit outage every 6 months to.
The NRC staff replied by letter dated July 31,1987 (Jab-perform this surveillance, bour,1987), that the test data presented were not suffi-cient and that a r. analysis performed by the staff indicated Duke Pawer Co. proposed a change for McGuire Nuclear that a quarterly test frequency could produce as much as Station, Units 1 and 2 (Tucker,1985) which would allow a an-order-of-magnitude reduction in the probability of maximum of 18 inonths between the tests for any one containment purge / vent system failure when compared door. He proposed surveillance interval is as conserva.
to the current 10 CFR Part 50 Appendix J requirement live as the current Standard Technical Specification sur-for leak testing at every refueling outage, veillance interval on an individual door basis.
ljcensees may be able to show, on a plant-specific basis or ge wasWe@ccam & McGuke he as part of a group effort, that their seals do not degrade design does not allow water condensatton to freeze, a c mmon cause of stuck doors.
over a sufficiently long period of time to permit an exten-sion of the surveillance interval.
The licensee also provided information concerning door reliability (Tucker,1985). Since 1981 to the date of the The NRC staff recommends that its position be retainea application in 1985, there had been 416 individual door until further data or analyses are presented to justify a inspections at McGuire Unit I and since 1983 there had change.
been 216 at McGuire Unit 2. In all of these tests, the doors met their acceptance criteria.
E' He NRC staff approved this proposed change in a letter l
ne surveillance on containment purge and exhaust (Hood,1988),
e isolation valves is in excess of 10 CFR Part 50 Ap-pendix J requirements.
Other utilities with an ice condenser contamment may wish to consider this change tf it can be justified on a This increase in surveillance testing results in an-
"" *E e
order-of. magnitude reduction in probability of con-tainment purge system failure.
Finding Duke Power Co. justified a surveillance interval for con-Workers performing this test receive a high dose of tainment inlet door testing that eliminated the need for a e
radiation.
shutdown.
NUREG-1366 46
8 Containment Recommendation Recommendations The monthly surveillance test should be retained.
ne Duke proposal may be used by other utilities if it can be justified on a plant specific basis.
The time cach vacuum breaker shall be tested fol-lowing any discharge of steam to the suppression 8.4 Testing Suppression Chamber to chamber shall be changed to 12 hours1.388889e-4 days <br />0.00333 hours <br />1.984127e-5 weeks <br />4.566e-6 months <br />.
Drywell Vacuum Ilreakers (llWII) 8.5 Ilydrogen Ilecombiner (PWII, BWR Technical Specifications require that at least once g g) every 31 days each vacuum breaker between the suppres-sion chamber and the drywell shall be tested by cycling De hydrogen recombiner system removes the hydrogen each vacuum breaker through at least one complete cyc!c and mygen gases that accumulate in the containment of full travel.ne American Society of Mechanicallingi-atmosphere following a design-basis loss-of-coolant acci-
~
neers Boiler and Pressure Vessel Code (ASMli Code) dent. It ts not capable of removing the highest hydrogen requires this test to be performed quarterly.
concentrations that could be present after a severe acci-dent.
His test is also required following any discharge of steam to the suppression chamber.The time for this test follow-The present Technical Specifications require testing the ing the Stcam discharge varies among Technical Specifica.
hydrogen recombiners at least once every six months by tions (e.g., for IIatch Unit 2 the time is 2 hours2.314815e-5 days <br />5.555556e-4 hours <br />3.306878e-6 weeks <br />7.61e-7 months <br />, for performing a functional test.
laSalle Units 1 and 2 the time is 12 hours1.388889e-4 days <br />0.00333 hours <br />1.984127e-5 weeks <br />4.566e-6 months <br />).nc NRC staff recommends that the time be established as 12 hours1.388889e-4 days <br />0.00333 hours <br />1.984127e-5 weeks <br />4.566e-6 months <br />.
A search of LliRs was performed to assess the reliability of hydrogen recombiners, Twelve failures of hydrogen Vacuum breaker operating problems were surveyed recombiners were found over the time period from 1980 briefly. From October 1978 through July 1988, a total of to June 1988. In most cases, only one train of the hydro-167 reported failures of suppression chamber to drywell gen recombiner system was inoperable. In one case in vacuum breakers were found. Some of these reported which both trains were inoperable, a backup hydrogen multiple vacuum breakers inoperable. In addition, a sig.
purge system was available. In two cases, the failure was nificant number were reported from only a few plants, fixed within two hours. Since the hydrogen recombiner is indicating either design or preventive maintenance prob.
manually started many hours after a LOCA occurs, the lems. In many cases, the vacuum breakers were inoper.
system would have been operable when called upon.
able only in the licensing sense because many of the The NRC staff recommends that, because of the redun-failures were problems with position indication and set points out of tolerance. In some of these events, however, dancy and apparent high reliability, the surveillance test the vacuum breakers were actually inoperable because of interval should be changed to once each refueling inter-V"I-mechanical binding. rust or foreign objects in air lines, or excessive leakage.
Findings The NRC staff recommends retaining the monthly sur-A search of LliRs found 12 failures of hydrogen veillance requirement because of the relatively high inci-recombiners between 1980 and the middle of 1988.
dence of problems and because these vacuum breakers are so unportant to safety.
in each case, either a redundant hydrogen recom-e biner or a diverse system (hydrogen purge system)
- /indings was available, Vacuum breakers have an important safety function.
Recommendation e
The Technical Specifications require a monthly sur-Change the surveillance test interval for hydrogen recom-biner functional tests to once each refueling interval.
veillance; the ASMli Code recommends quanctly surveillance.
8.6 Sodium Tetraborate Concentration There have been a significant average number of in lee Condenser Containment Ice failures pet year.
The Technical Specifications for the ice bed of an ice These failures seem to be concentrated at a few condensercontainment require achemicalanalysisof the BWRs.
stored ice to assure a boron concentration of at least 1800 47 NURI!G-1366
8 Containment ppm and a pil of 9.0 to 9.5 at 25'C. 'this analysis is Findings required at least every 9 months. 'Ihis surveillance re-quirement could mean that someone would have to enter Y,
g g
g the containment to take the sample while the unit is at condenser containment is requtred every 9 months.
power, which means that that person would receive a radiation dose. According to Duke Power Company
.lhis requires a containment entry while at power, chemists, meeting the boron concentration and plI re-quirements of this 1.CO has never been a problem.
Samples have always been within the required limits on boron concentration and pll.
The NRC staff therefore recommends that the frcquency for performing this chemical analysis be changed to once Change the analysis interval to once cach refueling inter-each refueling interval.
val.
NUREG-1366 48
=
9 PIANT SYSTEMS 9.1 Auxiliary Feedwater Pump and study that used data from the time period early 1979 to It 1982. NUREG/CR-4597 is a study donc for the System TcStinE (PWR)
NRC's Nuclear Plant Aging Research(NPAR) Program.
He Standard Technical Specifications and most plant Technical Specifications require the monthly testing of Iloth studies concluded that a significant cause of failures OI AN "*Ps is testing the pump by recirculating flow P
auxiliary feedwater (AFW) pumps. The 13 oiler and Pres-through a mm.tmum flow line which ts not adequately sure Vessel Code of the American Society of Mechanical sized. Both studies deduced this from a review oflicensee Engineers (ASME Code),Section XI, Paragraph IWP-3400, requires the testing of Class 1,2, and 3 cen-event report (LER) data and other data on the types of failures that occurred. This was reinforced by experience trifugal pumps "normally every three months." ARV pumps are the only pressurized water reactor (PWR) w th feed pumps at fossil power plants; these are also pumps required by Technical Specifications to be tested honzontal multistage centrifugal pumps similar to AFW more frequently than required by the ASME Code.
- pumps, lloth reports recommend, among other things, that the The Technical Specifications do not require the testing t size of the recirculation lines be increased. In many cases, be as thorough as required by the ASME Code. Tab!c 9.1 this can be achieved by modifying the ortfice in the recirc-compares inservice testing required by the ASME Code ulation line although, as the flow increases through the with that required by Technical Specifications' recirculation line, adequate flow to the steam generators must still be maintained.This could require a complicated Another difference in the testing is that, at some plants, interlock which would close valves on the recirculation one group may perform the monthly testing required by line when an actual demand signal is present.
the Technical Specifications, and a different group will perform the ASME Code testing every third month.
Although a change in recirculation flow from approxi-mately 10% to approximately 25% is the best solution to In both types of tests, the ARV pump takes suction from this problem, a reduction in the frequency of testing of the condensate storage tank and returns the water to the the ARV pumps from monthly to quarterly (the fre-condensate storage tank through a recirculation line. In quency specified in the ASME Code, Section.XI), could most plants, the recirculation line is sized between 5%
be a reasonable step to reducing the rate of wear, and 15% of the best estimate point flow (NP-4264, Vol.
1). This size was derived from considering temperature This problem should be put in perspective. Using the rise in the pump. Pump manufacturers are now recom-EPRI data,236 LERs reporting failures of turbine-driven mending that standby pumps be tested at a flow no less ARV pumps were found over the period from early 1979 than 25% of the best efficiency point flow (NUREG/
through late 1982. Of these,163 (69%) reported failures CR-4597). This is based on " hydraulic instability" of the related to the turbine (a figure which NP-4264 [Vol.1]
pump.This is a term for unsteady flow phenomena which states is roughly consistent with previous EPRI studies on can cause degradation of the pump and which become the high-pressure coolant injection [HPCI) and reactor more pronounced as the pump operates further away core isolation cooling [RCIC) systems in boiling water from its best efficiency point.
reactors [BWRs]) and 73 (31%) were pump related. Of those that are pump related,23% were failures of the Two studies have been done of the ARV system. Report
" rotating element" as opposed to instrumentation and NP-4264 is an Electric Power Research Institute (EPRI) controls or valves.nus, at the most,23% of the failures Table 9.1 Insenvice testing required by Standard Technical Specifications and ASME Code l
ASME Code Standard Technical Specifications l
Inlet pressure I-Differential pressure l
Flow rate Flow rate l
Vibration amplitude l
Lubricant level and pressure Bearing temperature
~
l Discharge pressure 49 NUREG-1366
1 9 Plant Systems -
could be reduced by decreasing the surveillance fre-AITV/EITV pump and valve surveillances have high quency for the turbine-driven AITV pumps.
risk impact per NUllEG/ Cit-520(L For the motor-driven AITV pump, the " rotating c!cmento The licensee burden is increased by monthly testing.
e accounted for 13 of 50 (26%) reported failures (instru-AITV pump availability is increased by quarterly
- mentation and controls, as well as valves, were the other testing on a staggered basis.
major contributors to failure of the motor-driven AITV pump). lhus, for the motor-driven AITV pump, at the Remmmendadon most,2695 of the failures could be reduced by less fre-quent surveillance or by decreasing the flow resistance of Change frequency of testing AITV pumps to quarterly on the recirculation hne.
a staggered test basis.
Another consideration is that, according to the NitC t e-port NUltEG/CR -4597,42% of the AITV pumpfailures M M a.in Steam L.ine Isolat. ion Valve were found during suiveillance testing and 29% were (MSIV) Surveillance Testing found during operation (6% were found during mainte-nance and, for 23%, the method of detection was not 9.2.1 Presstirized Water Reactors stated). 'lhus, surveillance testing is important in detect-ing failures in the AITV sys'em.110 wever, surveillance
'ihe ASMll Code,Section XI, Article IWV-3000 states testing also contributes to the problem.'the availability of that " valves shall be excretsed to the position required to the AITV pump, while related to the conduct of surveil-fulfill their function unless such operation is not practical lance testing, is not continuously linearly related to sur, during plant operation, if only limited operation is practo veilhnce testing. 'lhat is, at some point an increase in cal during plant operation the valve shall be part stroke surveillance testing (i.e., reducing the staveillance test exercised during plant operation and full stroke exercised interval) will not contribute to an increase in availability, during cold shutdown,"
and in fact could contribute to equipment unavailability.
Analysis of AITV pump failures mdicates that a monthly Since closin8 a main steam line isolation valve (MSIV) surveillance test interval (511) may be contributing to during operation would result in a plant trip, licensees do AITV pump unavailability through failures and equip,
(( ial-stroke testing quarterly as specified by the ASME C'
ment degradation. 'lhe changing of the AITV pump STI to quarterly, on a staggered basis, is consistent with this The purpose of this test is to demonstrate that an MSIV is analysis. Conducting the tests on a staggered basis will capable of movement.
permit system testing monthly, while r-ducing AITV pump testing to quarterly, thereby maintaining a consis-
'Ihe test is being done in several different ways.'lhe valve tent degree of reliability. The recommended change in may be closed rapidly in response to a close signal while testing frequency to quarterly is also consistent with the being tested. In this case, the valve closes until it is requirements of the ASMli Code-stopped by a limit switch at 10% closed (or less, depend-ing on the distance that the valve disc can be placed into Findings the steam line without fully closing because of the force of The ASME Code requires Class 2 pumps (such as a
AITV pumps) to be tested quarterly.
Another method is to drain hydraulic fluid to reduce Technical S ecifications re4uire testinE such pumps N* E' **" *
- e P
- monthly, lhe operator must respond to the valve closure and se-cure the bleeding at the 10% closed level. If the operator AITV pump wear is caused by recirculating water overshoots this level or if the test equipment fails, the -
e during tests through a line that has a smaller diame, MSIV will close and the plant will trip. likewise, if the ter than presently recommended by pump manufac-limit switch fails, the valve will completely close, causing a turers.
t rip, A review of 1. Ells shows that, at most,23% of tur.
'Ite MSIVs were not designed to be tested by the latter e
bine-dnven AITV pumps and 26% of the motor.
testing method.The hydraulic fluid outlets were designed driven AITV pumps had failures due to the " rotating to be used for maintenance and not far peritxlic testing.
- element" itself, as opposed to valves, controls, and so forth.
A review of operating experience with MSIVs (from in-formation in NPE-4) shows many cases in which MSIVs 42% of pump failures were found during surveil-could not be closed because of problems with the valves'-
e lance testing.
actuators or mechanical binding of the valves (as opposed NUlti!G-1366 50
~ _ _.. _ _ _ _..._
9 Plant Systems to control system problems). Some of these problems _
llWR MSIVs are required to be tested quarterly (ASME were found while the valves were being stroke iested.
Code) by verifying full closure within a specified time Most problems were found while the valves were being
. interval (typically between 3 and 5 seconds).
operated after maintenance or as part of other plant activities. Many of the problems seemed to arise from Some of the problems found during testing of IlWR inadequate maintenance.The buildup of foreign material MSIVs include:
and packing that was too tight were common problems.
Failed relays and limit switches so that no signal is Thus, it seems that surveillance testing is needed to find these failures, sent to (that division of) the RPS.
Air supply contaminated with fine particles or oil On the other hand, this test is considered, by PWR opera-which prevented the MSIV from closing.
tors, to have a high potential for reactor trip. Two recent Seat leakage in excess of technical specification al-trips as a result of MSIV testing occurr3d at Virgil C.
Summer Nuclear Station on May 12,192 92R 88-006) lowed values.
and at Waterford Generating Station.
c3 on Decem-ber 11,1937 (LER 87-028)/Ihus, a L
.4 study shows that
$*. hydraulic dashpot oil level resulted in fast clos-plants do trip during MSIV testing but that failures also ing times.
occur that affect the ability of an MSIV to close.
Reactor trips occur during IlWR MSIV tests; however, i
The NRC staff's qualitative safety assessment showed because significant problems were discovered during test-that the present quarterly survedlance test interval is ing of these valves, and because these valves are so tmpor-acceptable, llowever, a study of MSIV surveillance test-tant to safety, the NRC staff recommends that no change ing could be conducted to weigh the risks of testing versus be made m the test frequency of IlWR MSIVs.
the risks fram failures not being discovered.
Findings MSIVs are very imponant to reactor safety.
Findings e
Quarterly testing of MSIVsis required by the ASME Significant problems have been found during testing e
of IlWR MSIVs.
- Code, MSIV failures in which the valve would not close on Recommendation demand have occurred.
The surveillance frequency of IlWR MSIVs should not be MSIV testing has a high potential for trips, changed.
e A significant number of PWR MSIV failures appear to result from inadequate maintenance.
9.3 Control Room Emergency Ventilation System (PWR,.BWR)
Recommendation As part of the verification of the operability of the control room emergency ventilation system, the Technical Speci.
The vendor owners grou ps should consider the benefits of f cations require that the control room temperature be a study that would justify increasing the quarterly testing verified to be under a specified value every 12 hours1.388889e-4 days <br />0.00333 hours <br />1.984127e-5 weeks <br />4.566e-6 months <br />.
of MSIVs.
Some typical values are given below:
9.2.2 Boiling Water Reactors crystal River 120*F; Catawba 1/2 110*l-San Onofre 2/3 110*F The purpose of IlWR main steam isolation valves is to l
prevent the discharge of primary coolant outside the con-LSalle 1/2 None specified l
tainment following a pipe break in the main steam system.
-llatch 2 105*F l
.There are two MSIVs in series on each steam line. Since North Aana 1/2 120"U these valves must contain reactor coolant after an acci-Vermont Yankee None specified dent, minimizing leakage is an important consideration.
MSIVs also provide a signal to the reactor protection These values are derived, according to the Technical system (RPS) to trip the reactor upon 10% closure to Specifications "llases" section, to ensure that (1) the iun-anticipate the pressure and neutron flux transients that bient air temperature does not exceed the allowable tem-will follow a full closure.
perature for continuous duty rating for the equipment 51 NURiiG-1366
9 Plant Systems and instrumentation cooled by this system and (2) the ing the NRC staff's site visits, it is not clear that non-Class cont rol room will remain habitable for operations person-1E equipment would function at this temperature at all.
nel during and following all credible accident conditions.
He NRC staff recommends that the surveillance re-It is not clear that temperature limits this high accomplish quirement to record control room temperature every 12 -
thesc objectives. For one Pr "1, the following information hours be replaced with a more useful surveillance re-was available to judge the t, nservatism in the Technical quirement or possibly deleted if a more effective limit Specifications surveillarce requirement for control room cannot be established, temperature.
Findings The most conservative outdoor design temperature used The surveillance requirements for the control room by ASliRAE (American. Society of IIcating, Refrigera-tion and Air Conditioning Engineers) is the 0.1% level emergency ventilation system contain a requirement w hich is the highest temperature that occurs only 0.1% of that the control room temperature be verified every the year or 9 ho trs. AS11RAE suggests this value "only 12 hours1.388889e-4 days <br />0.00333 hours <br />1.984127e-5 weeks <br />4.566e-6 months <br /> to assure that it is less than a temperature for extremely conservative work." Using this temperature limit typically in excess of 100'F.
criterion at this site, the indoc ambient temperature would not exceed 104 *F. The tecnnical specification limit is 120*F for this plant.
This temperature limit is to ensure equipment oper-ability and human habitability. It does not appear to
- The control room temperature could exceed 111 *F at this be cffcctive for either purpose.
site for 1-1/2 hours per year but would always remain below 117'F.The FSAR for the reactor states that the Recommendation Class IE electrical equipment would be capable of oper-ating for only "short periods of time" at temperatures Replace this requirement with a more usciul surveillance above 117'F. From discussions with plant personnel dur-or delete it if a more effective limit can not be established.
.NUREG-1366 52
10- ELECTRIC POWER 10.1 Emergency Diesel Generator Research done by NRC and the industry has shown that f the assumptions in the analysis of the LOCA, sm Surveillance Ile9uirements (PWit' required by 10 CFR 50.46 and Appendix K to 10 CFR.
Im,R)
Part 50, are very conservative. In addition to the conserv-ative nature of the regulations, other conservatisms have Corresponding to their importance to safety, emergency been included in the vendors' LOCA models.
diesel generators (EDGs) have the most detailed Techni.
SECY-83-472 provides a method to eliminate those eon-cal Specifications surveillance requirements of any piece servatisms not specifically required by the regulations, of mechanical or electrical equipment in a nuclear power plant. Suneillance requirements for EDGs are currently Under the sponsorship of the Electric Power Research based on Regulatory Guides L108 and 1.9.
Institute (EPRI), calculations were performed, ubg the methods given in SECY-83-472, that show margin is available to the criteria of 10 CFR 50.46 that could be -
The safety function of the diesel generators is to supply ac used to extend the EDG start and load times. Studies electrical power to plant safety systems whenever the done for a typical four-loop Westinghouse PWR preferred ac power supply is unavailable. Through sur-
{NSAC-130) show that the diesel start and load time veillance requirements, the ability of the EDGs to meet could be increased to 45 seconds from 10 seconds. The their load and timing requirements is tested and the qual-45.second start time is limited by environmental qualifi-ity of the fuel and the availability of the fuel supply are cation considerations of equipment in containment.The monitored.
calculated peak cladding temperature was below 2200* F.
As part of the resolution of Unresolved Safety Issue (USI)
A similar calculation for a typical HWR 4 showed that the A-44, " Station Blackout," the NRC staff has prepared diesel generator start and load time could increase to 118 Regulatory Guide (RG) 1,155 to provide guidance on sec nds (NS AC-96), and still be withm acceptable limits.
EDO reliability levels RG L155 also specifies that the reliable operation of onsite emergency ac power sources Ilowever, for the purpose of evaluating the effects of should be ensured by a program designed to maintain and surveillance testing, start and load times should be ad-monitor the reliability level of each power source over dressed separately, time to ensure that the selected rehability levels are being achieved.
A fast start (i.e., start and acceleration to synchronous speed at full fuel rack position) has the potential to accel-crate the degradation of the diesel generator if conducted Generic Safety Issue (GSI) B-56, " Diesel Reliability,"
without the benefit of a prelube period. However, was established to develop guidelmes for an EDG reli-prelubricating diesel generators is now common practicp ability program. In addition to these efforts, the Office of and any remaining negative effects of fast starts are mmi-Nuclear Regulatory Research (RES) is conducting the rnal. Nevertheless, fast starts can be ehmmated on some Nuclear Plant Aging Research (NPAR) Program, which is diesel generators by changing the governor configuration,-
intended to resolve technical safety issues related to the but onh at the cost of reductng diesel generator reliabil-aging degradation of equipment important to reactor ity, by chmm, atmg a redundant overspeed protection fea-l safety. An important part of this program is the study of ture, that is, the backu p mechamcal governor. In this case, l
the aging of emergency diesel generators.
the gain associated with slow starts does not offset the loss I
of the backup overspeed protection.
The results of these programs were reviewed as part of
.this study to determine how these programs will affect Fast loading (i.e., zero to full load in 120 seconds or less) -
surveillance requirements for EDGs in the Technical during surveillance testing is, on the other hand, the mon Specifications.
sigmficant cause of accelerated degradation of diesel gen-l erators. It can cause rapid piston ring and cylinder liner 1-wear (up to 40 times greater than normal wear) and The current performance requirements are stringent.
should be climinated in favor of loading in accordance The EDG must start on any of several signals (e.g., man' with the manufactu rer's recommendations, except for the
'l L
ual actuation safety injection, or loss of normal power to 18-month loss of offsite power (LOOP) test. Manufactur-an em ergency bus), increase to rated speed in a short time ers' recommendations for diesel generator loading can be (e.g,10 seconds), and pick up its emergency load in blocks 30 minutes or more to reach full load, at programmed times (load sequencitg). These times are relatively short and are set by the requirements of the in an actual emergency, loads will be sequenced onto a large-break loss.of-coolant accident (LOCA).
diesel generator in approximately 60 seconds. This 53 NUREG-1366
10 Electric Power constitutes fast loading of the diesel generator regardless By a letter dated December 7,1987 (Murphy,1987), the of whether the sequencing started at 10 seconds or at licensee for Vermont Yankee submitted a request to either 45 or 118 seconds (as suggested by the studies revise this surveillance / alternate t ; ting requirement, referenced above) after the diesel generator starts.
'Ihe NRC staff reviewed this proposed change to the llence, design changes for slower diesel generator start-Technical Specifications and requested a more quantita-ing and acceleration would not significantly reduce the tive analysis than had been onginally supplied. in re-degradation of diesel generators which is inherent with sponse to this request, the Vermont Yankee Nuclear rapid loading that is necessary to meet safety analysis Power Corp, submitted an analysis dated July 15, 1988 requirements.
(Capstick,1988), using reliability methods. 'the NRC
'Ihe NRC staff recommends that all testing of the diesel generators, with the exception of the LOOP tests which The analysis quantified the unavailabilitics of the systems are performed with and without an ESF signal once cach when required to perform their intended function upon refueling, be performed by gradual loading in accordance demand, both with and without alternate testing. Two with the manufacturer's recommendations.
systems were chosen for detailed analysis: the core spray EDG tests were typically started with the EDG initially at ambient conditions with no prelubrication or warmup
'the pros and cons of testing were quantified, that is, time. Generic Letter (GL) 84-15 changed this, stating (1) the decreased potential for an undetected failure due that "[llicensees are encouraged to submit changes to to the alternate testing and (2) the increased unavaFabil-their Technical Specification [s] to accomplish a reduction ity due to (a) the alternate testing and (b) repair of in thc number of [ cold] fast starts " A typical technical demand-related and test-related failures. Other disad-specification was included in GL 84-15 which required a vantages to alternate testing which were not quantified in start from ambient conditions every 184 days rather than (nis study are:
cycry month, (1) reduced reliability due to equipment degradation Some nonstandard Technical Specifications require that, from excessive testing with an inoperable EDG, not only the remaining operable diesel generator (s) must be tested at a higher frequency (2) potential for unnecessary shutdowns that result in than normally required but, in addition, other emergency plant transients and challengas to safety systems equipment such as the emergency core cooling system (ECCS), safety-related ecoling water pumps (e.g., service (3) potential for plant transients initiated during sur-water), and other power supplies also must be demon.
veillance tests strated operable. This testing must commence "immedi-ately" up(m discovering that a diesel generator is inoper-(4) diversion of operating personnel time and attention able.
(5) increased radiation exposure to operating personnel Some nonstandard Technical Specifications also require that if a train or subsystem of certain safety systems other than the diesel generat, ors (for example, a low-head
'the analysis showed that, for the core spray system, alter-safety-injection pump of the ECCS) is declarnt inoper-nate testing (which is required daily by the Vermont Yan.
able, not only the other train of the particulai vstem but kee Technical Specifications) produced unavailabilities at -
also other equipment of the emergency core coeling sys-least a factor of 4 greater than monthly testing. For the tems and the diesel generators must be tested. 'lhus, a diesel generators, this factor was about 3.-
. failed train in one safety system can cause a great deal of testing of apparently unrelated systems. lhis type of test.
Considering this analysis and similar conclusions in ing is called " alternate testing "
NUREG-1024, the staff recommends that alternate test-ing requirements be deleted from the Technical Specift-An example of this in matrit 'ona is shown in Table 10.1 cations for all plants so that the failure of a train or (from a letter from Vermont Yankee Nuclear Power subsystem of a safety-related system other than an emer-Corp., J uly 15,1988 [Capstick,1988]) which is based on gency diesel generator would not require testing of the.
the Vermont Yankee Technical Specifications.
diesel generators or any other equipment, NUREG-1366 54 l
1
l-l l
Table 10.1 Alternate Iesting requirements Operation inoperable Standby RIIR Alternate Standby subsystem (days) '
. liquid Core 1101 Diesel Containment service Service cooling gas Inoperable components control sp ay s"bsystem generators cooling water water tower ADS RCIC HFCI treatment Standby '
liquid -
7 I!D control Core spray 7
1/D I
I I'
I ITC) pump 7
I IID*
ITCI subsystem 7-I/D
' I/D I!D
!!D t
. Diesel generators 7
I/D I/D UD I/D Containment cooling 30 1/D RIIRSW M
pump 30 1/D RilR senice
, uater 7
I/D I/D Service water 15 I<D I/D Alternate cooling tower 7
I/D I/D DD
' ADS 7
I RCIC 7
I/O,
IIPCI -
7
- I I
I!D I,D 5tsndby gas treatment 7
!!D 6
.t Note: I-Immediate; D-Daily
- 5 a-Redundant component only
- t> See ITC1 subsystem, core spray, and diesel generator alten: ate testing requirements.
2
.E
'c
.8
- c
- g. -
M-
.o-O
.c-L E
y-4
.t e
- O t
10 lilectric Power
'lhe NitC staff recommends that the requirements to test hours. During the first 2 hours2.314815e-5 days <br />5.555556e-4 hours <br />3.306878e-6 weeks <br />7.61e-7 months <br />, the diesel is to operate the remaming diesel generator (s) when one diesel gen-with its 2 hour2.314815e-5 days <br />5.555556e-4 hours <br />3.306878e-6 weeks <br />7.61e-7 months <br /> rated load and for the last 22 hours2.546296e-4 days <br />0.00611 hours <br />3.637566e-5 weeks <br />8.371e-6 months <br /> it is to crator n inoperable due to any cause other than operate at its continuous rated load.The Standard Tech-preplanned preventive maintenance or testing be limited nical Specifications require that, within 5 minutes after to those situations where the cause for inoperabdity has cornpleting this 24-hout test, the. mergency buses must not been conclusively demonstrated to preclude the po, be deenergized and loads shed
- ith a subsequent fast tential for a common male failure, llowever, when such start and full load acceptance, testing is required,it should be perfor med within 8 hours9.259259e-5 days <br />0.00222 hours <br />1.322751e-5 weeks <br />3.044e-6 months <br /> of havmg determined that the diesel generator is inoper-
- 1) uke Power Co., by letter dated ebruary 15,1988, on able.
the Catawba Units I and 2 dockets (lucker,1988) pro-posed to separate the 24 hour2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> test from the 5-minute test.
'lhe NPAlt Program founJ that regulatory surveillance The NitC staff approved Duke's proposal in a letter to requir ements are not the only contnbutor to !!DO degra.
Duke dated 5dy 2N 1988 (Jabbout,1988).
dation. NUlt!!G/ Cit-4590. Volurne 1, identified four categories of stresso's that contnbuted to emergeng die-The reason for requesting this change is that separating sel generator aging: vibration, infenor quahty of compo-these two required tests gives plant operators ndded flexi-nents, ads ene enuronment, and human error.
bility and prevents critical path complications during the outages.
The NPAll Propam did not specify the fraction of prob-lems found with emergency diesel generators which are Duke stated that it has been necessary to shut down the due to testmg. A study done for liPiti (NI'4264, Vol. 2) diesel generator faster than recoumended by the dicsci looked speciheally at failures of emergency diesel genera-generator shutdown procedure in order to perform the tors that r esult from sun eillance testing.'the data for this hot restart test within 5 minutes of the 24 hout test run, study consLt of 1. lilts from January 1979 through early Another problem with performing these tests in quick 1983, a penal of just over 4 ye.us. Note that this period succession is their potential for causing critical path com-preceded the issuance of Genene i.elter 84-15 so that, plications and delays during an outage. lingineered safety hopefully, the situation now would be somewhat better, features (l! Sit) actuation testing is performed at the be-A total of 555 failures of 136 diesel generators were pinning of refochng outages. Illock tagouts are delayed found. Of these 555 failures,70 (124) were determined until completion of I!SI testing. As a result of the testing to be related to suncillance testing. The components sequence currently dtetated by Tecimical Specifications, a that had the highest numbers of surveillance test-related minimum of 48 hours5.555556e-4 days <br />0.0133 hours <br />7.936508e-5 weeks <br />1.8264e-5 months <br /> of critical path time is spent each problems we r e: turbocharger, power assembly and bear-refueling outage running the two diesel generators. Ily mgs, startmg system, cooling system, lube oil system, gov-revtsing the survettlance equirements as requested, the ernor and exciter, and regulator. Ilowever, no specific two 24-hour runs could be completed later in the outage failures were widespread enough to be considered ge-or at some other convenient time.
nerie. Generic failures with diesel generators have oc-curred in the past, but solutions to these problems are Duke proposed to substitute a diesel generator run at avadab!c and, in most cases. have been implemented.
continuous 4ated hiad for 1 bour or until the operating temperature had stabilhed, followed within 5 mmutes by limergeng diesel generator testing appears to be an area a diesel engme start. To ensure that operating tempera.
that would benefit from a rehabihty-based testmg pro-tures have stabdued, the NI(C staff concludes that 2 gram (as discussed m Section 3.S of this report)Jihe NPf hours is a more appropriate time limit.
staff is evaluatmg rehabihty-centered concepts for the resolution of GI 15-56 that may further reduce unneces-
~lhe hot-restart test is performed to verify that the diesel smy testmg. NUltliG/ Cit-5078 describes an approach to generator does not have, m any way, impaired perform-a reliability based testing program for emergency diesel ance following operation at full load or equil7 Nm generators. As part of this rehability-based npproach, a te mperature.
y detailed root-cause analysis procedure and a good pre-ventive maintenance program (also reliability based) a lure to restart when hot, or extca Jed delay in restart-should be meluded. Detailed monitoring and trending are mg, is typically only experiet ced with small forced-air-important to assure good performance.
cooled diesel engines which, up(m being inpped undergo a temperature rise transient. The large diesel generators Diesel generator survedlance reqmromenis could also be are typically water cooled and do not experience any improved in another area.
sigmficant temperature rise transients dunng operation d
or after shutdown. In addition, diesel generators are The Standard Tect'nical Specifications contam a require-normally maintained at hot standby conditions (heated ment to operate cach emergency diesel generator for 24 cooling water and lubncatmg oil).
N UltlG 1366
$b
10 Electric Power
- Ihe NRC staff. therefore recommends that other utilities tems not associated with an inoperable train or sub-be permitted to char.ge their Technical Specifications to system (other than an inoperable EDG).
separate the 24-hour test and the hot startup test if they propose doing so.
10.2 Battery Surveillance Requirements Findings (IMR, BWR)
EDGs are very important to safety.
Industry guidance for testing large lead storage batteries e
of the kind used in nuclear power plants is found in EDGr are tested too often because:
Standard 450-1980 of the Institute of Electrical and Elec-tronics Engineers (IEEE). Regulatory Guide 1129, (1) Technical Spectfications at some plants require Revision 1 (Febrt.ary 1978) endorses an earlier version of testing if other safetyuelated equipment is this standard (IEEE 450-1975). The Standard Technical inoperable.
Specifications follow this standard to some extent but are
-~
more conservative in some requirements and less conser-(2) Technical Specifkations at some plants require vative in others Table 10.2 compares IEEE 450-1980 not just one start to verify operability but starts with the Westinghouse Standard Technic:d Specifica.
"immediately," or within 1 hour1.157407e-5 days <br />2.777778e-4 hours <br />1.653439e-6 weeks <br />3.805e-7 months <br />, and every 8 tions, Version 4 A.
hours thereafter Note that IEEE 450-1980 requires more visual inspec-Studies show that testing too frequently is counter-tions of the condition of the batteries (e.g., cleanliness, pnx! active to safety in terms of equipment evidence of corrosion, cracks and leakage of electrolyte) availability, than the Westinghouse Standard Technical Specifica-tions (S13). On the other hand, the Westinghouse STS Rapid loading is a major cause of diesel generator are more conservative with respect to the frequency of degradation.
measurements of battery charger output, pilot cell condi-tions, and total terminal battery voltage. ("the West-There is no safety reason for performing a startup of inghouse STS require these every 7 days while IEEE -
a dicsci within 5 minutes of the 24 hour2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> test run as is 450-1980 requires these surveillances only monthly.)
required by Technical Specifications.
It is apparent from this comparison that the Westing-house S13 are most concerned with measurements of the Recommendations operability of the batteries and not as concerned with When an EDG itself is inoperable (not including a mechanisms that degrade the batteiies.
support system or independently testable compo-nent), the other EDG(s) should be tested only once Perhaps the most significant surveillance not included in (not emy 9 hours1.041667e-4 days <br />0.0025 hours <br />1.488095e-5 weeks <br />3.4245e-6 months <br />) and within 8 hours9.259259e-5 days <br />0.00222 hours <br />1.322751e-5 weeks <br />3.044e-6 months <br /> unless the the Westinghouse STS is the surveillance for ambient absenes 4 any potential common-mode failure can mom temperature. IEEE 450-1980 requires a monthly be demonstrated.
surveillance. The Westinghouse STS do not. The West-inghouse KrS do require a quarterly surveillance of elec.
EDGs should be hiaded in accordance with the ven-trolyte temperature in a representative number of cells, dor recommendations for all test purposes other but the requirement is that the temperature t preater than the refueling outage LOOP tests.
than a minimum value, an operability r equirement/Ihere is no maximum temperature specified, The hot start test following the 24-hour EDG test a
should be a simple EDG start teat. If the hot-start A limit on maximum ambient temperature we'ald protect test is not performed within the required 5 minutes the batteries from degradation mechanisms, following the 24-hour EDO ter., it should not be necessary to repeat the 24-hour EDG test, The only NUREG/CR-4457, which studied the aging of Class IE requirement should be that the hot-start test is per-batteries for the NPAR Program, states that "thetmal formed within 5 minutes of operating the diesel gen-st.csses, whether caused by internal sources..or by th.
erator at its continuous rating for 2 hours2.314815e-5 days <br />5.555556e-4 hours <br />3.306878e-6 weeks <br />7.61e-7 months <br /> or until ruam temperature, are prolubly the most detrimental operating temperatures have stabilized.
with respect to accelerat% the aging of batteries." As an example, the report cites a major battery manufacturer as Delete the requirement for alternative testing that stating that an incicase in ambient temperature from requires testing of EDGs and other unrelated sys-77*F to 95'F reduces the life of the battery by 50%
57 NUREG-1366
- 10 Electric Power
- Table 10.2 Comparison of requirements of IEEE Standard 450-1980 -
with requirements of Westinghouse STS IEEE Westinghouse Requirement 450-1980-STS 1.
General appearance and cleanliness of battery M
and battery area 2.
Evidence of corrosion on terminals or connectors.
M Q
3.
Cracks in cells and leakage of electrolyte M'
4.
Individual cell condition Y
R
- 5. Tightness of bolted connections Y
R 6.
Integrity of battery rack Y
R 7.
Condition of ventilation equipment M
'I 9.
Electrolyte levels, each cell M
Q
- 10. Ambient temperature M
- 11. Vol! age, specific gravity, each cell Q
O
- 12. Electrolyte temperature, representative cells Q
Q
- 13. Total terminal battery voltage Q
Every 7 days
- 14. Pf m cell electrolyte level See item 9 Every 7 days
- 15. Float voltage M
Every 7 days
- 16. Specific gravity M
Every 7 days
- 17. Electrolyte temperature M
i Note: M = monthly O = quarterly, Y = yearly. R = not to exceed 18 months, 1
- The NRC staff therefore recommends a studyof the need IEEE Standard 535-1986 requires that batteries that for a maximum (and minimum) allowable ambient tem-have been aged to their end of life service be given a perature for batteries.
' pre seismic capacity test, a capacity test during a simu-lated seismic test, and a post-seismic test. IEEE $35-1986 '
There are other important-phenomena discussed in also requires seismic qualification of the ba'tery rack.-
NUREG/CR-4457 that are not covered by either the -
Therefore, batteries tested to IEEE 535-1986 should be -
Standard Technical Specifications or IEEE 450-1980.
acceptably qualified for seismic events, recognizing that.
These are the seismic vulnerabihty of the batteries and this is not a Technical Specifications issue and seismic -
excessive harmonic fluctuations in the battcry charger testing should not appear in the Technical Specifications, voltage, called ac " ripple."
The Standard Technical _ Specifications require several
' According to NUREG/CR-4457, excessive harmonic surveillances more often - than called for - by-IEEE-fluctuations in voltage from the battery citarger cause 450-1980. These are the electrolyte level, float voltage,.
stresses at the battery plate similar to overcharging, accel
- specific gravity of the pilot cell (Category A items of Table
. emte corrosion, and produce excessive internal tempera-4.8.2), and the total terminal battery voltage. Several of tures.The NRC Office of Nuclear Regulatory Research the PWR licensees identified these as burdensome sur-should continue to study these to determine if this situ-veillance intervals. However, as shown in Table.10.3
.ation is really a problem at nuclear power plants.
(taken from NUREG/CR-4457), some of the most com-mon causes of battery failure are associated with items The scismic event is the design-basis event for the me.
covered by these'surveillances. Note that the leading-chanical integrity of batteries. Seismic vulnerability is cause of battery inoperability is low specific gravity. Insuf-
_ caused by physical degradation of the structure of the ficient charge and low electrolvte levels are also signifi-battery.There are no good tests to detect this aging, but cant causes of ba'tery failure.
NUREG-1366 58 l
l
- 10 Electric Power Table 10.3 Itattery failure events reported in LERs
- Fa!!ure cause No.
Iow specific gravity 67 27 Personnel (operation, maintenance, testing) 52 21-Insufficient charge 27 11 Dcfective/ weak cells 22 9
l Low electrolyte solution level 14 6~
Faulty connections 13 5
Defective proced tres 11 4
Charger malfuni tion 9
4
]
Design, fabricatan, construction 8
3
.I liigh electrolyte solution level 8
3 Unknown causes 5
2 Corrosion 4
2 Short circuit 4
2 Normal wear / natural end of life 3
1 Extreme environment I
<1 Tote.1 248 100 Source: NUREG/CR-4457.
Note also that testing (grouped together with operation Findings and maintenance) is the second largest contributor to Operability surveillances of batteries required by e
battery failures. However, these 7 day surveillances should not be significant contributors to testing failures.
Techidcal Specifications are performed more often than the mdustry standard recommends.
' In addition, one utility representative told the NRC staff There is no Technical Specifications requirement e
during a site visit that in addition to the Technical Specifi-for' monitoring or controlling battery - room cations requirement, it was company policy to do these temperature.
checks every 7 days.
&ismk plification is an important consideration e
for Class IE batteriesand battery racks. All Class The NRC staff therefore recommends that the battery lE batteries and battery racks should be qualified to surveillance requirements remain as they are.
_ IEEE Standard 535-1986. This is not a Technical.-
Specifications issue.
As noted earlier, many factors specified in IEEE Alternating current (ac' ripple from battery charg '
e
~
450-.1980 are important for degradation of batteries that ers raay be a degradation etmcern.
are not covered by Technical Specifications. This is prob-ably appropriate,if the purpose of the Technical Specifi.
Recommendation cations is limited to operability concerna. Ilowever, the
' The NRC should consider the above findings and e
staff recommends that these factors be included in any determine whether any additional action is war.
preventive maintenance program.
ranted.
59 NUREG-1366 -
11 REFUELING The staff identified no problems with surveillance testing-regarding refueling in Technical Specifications.
i M
NUREG-1366 g)
12 SPECIAL TEST EXCEPTIONS '
Suspending Shutdown Margin First, Florida Power and Light Co. showed, using data from the St. Lucie plants (Units 1 and 2), that the prob-Requirements (PWR) ability of a stuck control rod was not significantly in-creased by this change. Second, there would be no Section 10 of the Standard Technical Specifications deals changes to core geometry between the time that the con _
trol rods are tripped for the drop time measurements and with special test exceptions to other Technical Specifica-the time that the shutdown margin limit would be sus-tions requirements needed to perform tests to verify that Pended (up to 7 days) because the vessel head and all the reactor will be operated within its approved limits, for vessel internals would be in their final position and example, low-power physics testing.
secured.
One of these test exceptions permits the shutdown mar-
^Ithough plant-specific data were used m. part to justify gin requirements in Mode 2 (startup) to be suspended for this change, the operation of the St. Lucie units with certain low-power physics tests.The surveillance require-respect to this testmg would be representative of the ments state that each control rod not fully inserted shall peration of other pressurized water reactors (PWRs).
~ be demonstrated capable of full insertion when tripped Therefore, if it is assured that rod drop timing tests will be
' from at least the $0% withdrawn position within 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> performed within 7 days, with no changes made that
- prior to reducing the shutdown margin to less than the w uld affect the capability of the control rods to trip limits required in Mode 2.
within this time, the verification of rod insertion prior to suspending shutdown margin should not be required for -
In a letter to NRC (August 17,1987), Florida Power and ther PWRs.
Light Co. proposed changing this time from 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> to 7 1
days (Woody,1987).The staff approved the request in a letter dated October 28,1987 (l'ourigny,1987).
Findings During low-power physics testing in PWRs, two rod e
Another Technical Specifications requirement also re-drop tests are required.
quires tripping the control rods but for a different pur-pose, in this case, the control rod; are tripped in order to Only one test is required as long as there is adequate e
verify that control rod drop times are less than the assurance that there will be no changes to the core or value(s) assumed in the safety analyses. These tests are control rods after the first test, usually performed more than 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> before the control rod worth measurements are needed. Thus, in prac' ice, Recommendation the control rods are tripped at least twice following a refueling outage: once for the drop time measurements All PWR licensees may select the Florida Power and and once prior to the time that the shutdown margin limit Light Co. proposal to eliminate one rod drop test if they
- is suspended.
satisfy the condition of performing a rod drop test no more than 7 days before reducing shutdown margin. If a This change would permit only one such test of the con-rod drop test has been performed within this ti. 4c, an-trol rods.The change is permissible for several reasons.
other test is not necessary, 1
61 NUREG-1366
13 RADIOACTIVE EFFLUENTS Waste Gas Storage Tanks (PWR) dose equivalent 1-131 of 1.0 gCi/ gram would result in a maximum waste gas storage tank activity of 75 pCi/cc.
Assuming Units 2 and 3 were operating with primary Waste gas storage tanks collect the radioactive noble coolant 1-131 dose equivalents of 1.0 Ci/ gram (the gases and airborne halogens to reduce the anticipated Technical Spectfications limit), a maximum storage tank annual releases and personnel exposure in restricted and activity of 150 Ci/cc can be calculated. A storage tank unrestricted areas in order to meet "as low as reasonably with 150 Ci/cc of noble gas at 300 psig would contain achievable" (ALAR A) guidelines.
60,000 Ci. 'Ihis is far below the Technical Specifications limit of 134,000 Ci for San Onofre.
'Ihe Technical Specifications contain a limit on the curies of noble gases that are allowed to be stored in the gas
'lhus, iodine concentration m the coolant will rarely rcach storage tanks and require that the quantity of materialin the Techmcal Specifications limit, if at all, at operating the tanks be determined to be less than this limit at least scactors, llowever, Xe-133 activity can accumulate and once every 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> when radioactive materials are beinr approach Technical Specifications limits during degas op-added to the tank. Measuring the contents of this ta' x crations prior to shutdown for refueling. Therefore, the exposes the chemists to high radioactivity, surveillance requirement should be changed to state that the quantity of radioactive material contained m each The " Bases" section of the Standard Technical Specifica.
waste gas decay tank shall be determined to be within the -
tions states that the Technical Specifications limit placed limit at least once every 7 days whenever radioactive ma-on the number of curies (Ci) permitted in the waste gas terials are added to the tank, and at least once every 24 tank ensures that the resulting total body exposure to an hours dunng pnmary coolant system degassing opera-II""8-individual at the nearest exclusion area boundary will not exceed 0.5 rem, which is consistent with the Standard Review Plan (NUREG-0800) guidance for calculating I'indings the consequences of the failure of a waste gas tank.
The total number of curies of radioactive gas in the waste gas tank will only approach the Technical It was pointed out by both staff and licensee personnel Specif cations limit during degas operations.
that the Technical Specifications limit placed on the al.
lowed curies in the waste gas tank is considerably above Personnel receive high doses of radioactivity in e
the vidue that would occur, even if the reactor were oper-measuring the number of curies in the waste gas ating at the Technical Specifications specific coolant ac-tank.
tivity limit,1 Ci!cc.'the San Onofte licensee did a calcu.
lation to demonstrate this (Katz,1988).
Itceommendation San Onofre Nuclear Generating Station, Unit 3 experi-The surveillance requirement for the limit on the number enced iuct failures during its first cycle. Data were tabu-of cuties in the waste gas tank should be changed to: "Ihe lated for the period when highest primary coolant activi-quality of radioactive material contained in each waste gas ties were experienced. When the Unit 3 primary coolant decay tank shall be determined to be within the limit at dose-equivalent i-131 averaged 0.48 Ci/ gram, the maxi-least once every 7 days whenever radioactive materials r e im activity measured in the inservice waste gas storage are added to the tank, and at lesst once every 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> tank was 36 Ci/cc. Using this relationship, a primary during primary coolant system degassing operations."
NUREG-1366 62-
14 CONCLUSIONS This study examined individual surveillance requirements it should also be kept in mind that surveillance testing in Technical Specifications and assessed their effective-required by Technical Specifications is not a major cause ness in ensuring reactor safety. In addition, as the study of reactor trips (such required testing contributes to ap-progressed, several more general conclusions were proximately 20% of the total for a mix of new and mature reached.These have been discussed individually in Sec-plants); nor is the testing a major contributor to radiation -
tion 3,
- General Findings," but they will be reiterated dose (approximately 20% of the total radiation dose can here.
be attributed to such testing).
First, there are a lot of surveillance requirements. Thou.
The amount of equipment wear or number of failures is more difficult to assess, but several studies of key safety-sands of tests are done in a cycle. It is necessary to test related equipmen: leve not identified testmg as a major equipment to ensure that it is operable or to uiscover that cause of fauure, although, as with the other categories it isn't functioning so that it can be restored to operable (trips and radiat,on dose), it is sigmficant.
i
- status. Ilowever, performing more tests than are neces-sary, or performing ineffective tests, obviously does not General Recommendations accomplish anything and can be harmful to people (radia.
tion dose) or equipment (wear).
Before presenting the detailed "line item" recommenda-
]
tions of this report (see Table 14.1), the more general The utilities should be given as much latitude as possible recommendations are given.
in scheduling surveillance tests. Few tests have timing requirements so critical that an extension would not be (1) A study should be done of the feasibility of using acceptable.
reliability-based Technict' Specifications on one or several systems. A lead plant should be selected and Scheduling the thousands of required surveillance tests the licensee of that plant should select, with the while maximizing availability of equipment, minimizing agreement of the staffs the systems that would be the operations that may cause scrams, ano ensuring that candidates for reliability. based Technical Specifica-tions.
at least one train of emergency equipment always remains operable is at times difficult. More thought should be (2) The current industry effort on advanced reactor de-given to which tests have surveillance schedules that are signs sould include a study of how all required critical and which tests do not, survallance testing will be performed in order to (a) minimize the possibility of a transient caused by Utilities should be given an opportunity to point out un' testing, (b) minimize the burden on plant personnel necessary types of tests at their facilities and to climinate wha will have to perform these tests, and (c) mini-them.
mize the radiation exposure received by people in performing the required testing.
Incorporating the concept of reliability into the Technical Specifications would be useful in climinatinf excessive (3) The NRC should study the issue of containment testing and in identifymg which components or systems entries at power to assess the risk to containment need the most surveillance, integrity from these entries and the effect on the people who enter containment with the reactor at Preventive maintenance is very important in reducing the PO*' f-number of failures. Operability testing does not reduce (4) Section 4.0.2 of the Technical Specifications, which the number of failures, it only discovers failures that have allows the extension of a surveillance test interval, occurred. Operability testing reduces the time period as-should be made applicable to Section 4.0.5 concern-sociated with inoperability, and increases availability. Pre-ing ASME Code testing in those Tecimical Specifi-ventive maintenance, however, should also be reliability cations which presently do not allow Section 4.0.2 to based to assure that resources are allocated where they be applied, are needed most.
The implementation of the recommendations of this Part of the reason that surveillance tests are difficult to study should reduce reactor transients, radiation dose to perform is the layout of the plant and the design of equip-personnel from testing, and wear on equipment. This ment in the plant. Although this is hard to fix in an operat-study, however, should be considered as only the begin-ing reactor, adequate planning should be required for the ning to a more enlightened approach by both NRC and advanced light water reactors (or other reactors) cur-the nuclear industry to assuring that equipment impor-rently being considered to facilitate surveillance testing.
tant to reactor safety is effectively and safely tested.
63 NURiiG-1366
114 Conclusions Table 14.1 Sumtnary of recommended changes to suncillance requirements Section -
t of this TS surveillance requirement _
report lapplicable criterla*]
Recommended change REACTIVITY CONTROL SYSTEMS 4.1 MTC testing at 300 ppm (PWR). [11]
VEPCO proposal rnay be used to assess whether testing is required at 60 ppm if
]'
crite: ion is not met at 300 ppm. Plant.
specific analysis is required.
4.2.1 Control rod movement testing every 31 days Change STI to quarterly.
(PWR) [B,T) 4.2.2 Control rod movement tects every 7 days (BWR).
BWR Owners Group should study the 18.'11 feasibility of extending the surveillance intervalIf it is feasible from an engineering --
viewpoint, the NRC staff should examine -
the study findings and allow the extension if found acceptable.
4.2.2 If a BWR control rod is immovable because of Change the frequency to "within 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> friction or mechanical interference, the othcr and every 7 days thereafter."
control rods should be tested every 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br />.
4.3 Standby liquid control system explosive valve Change STI to once each refueling interval -
- testing performed every 18 months (BWR). [B]
for fuel cycles up to 24 months' duration.
4.3 Standby liquid control system Change STI to quarterly, pump test monthly (BWR). [B]
4.4 Closure time testing of SDV vent and dram Closure time may be extended if justified by valves Verification that valves will close in 30 plant. specific analysis. (Approved for Hatch seconds (BWR). [B]
Nuclear Plant, Units 1 and 2.)
4.5 Reactor trip test to verify operability of SDV vent Delete requirement for a scram check of and drain valves. Required once every 18 months
_ SDV "".at and drain valve operability at n-(BWR). [B,T) 50% rod density or less.
Require an evaluation of SDV system
- response after each scram to verify that no.
abnormalities exist prior to plant restarti Require vent and drain valve operability -
testing during a scram from shutdown '
- conditions.
R INSTRUMENTATION 5.1 Nuclear instrumentation surveillance (PWR). [B,T)
- Change STI from monthly to quarterly..
Channel functional tests required monthly.
See footnote at cod of table.
NUREG-1366 64
14 Conclusions Table 14.1 (Continued)
Section of this TS suncillance requirement reporf lapplicable criteria'}
Recommended change INSTRUMENTATION (Continued) 5.2 Slave relay testing (PWR, llWR). [B,1 J Change quarterly requirements to a staggered test basis over a cycle with tests most likely to cause trip done at refueling outage.
5.3 Test intervals for RPS and ESFAS (PWR, BWR).
Test 3-channel systems on 4-channel
[B,T) schedule by not testing one of the 3 channels during a 4-channel surveillance interval.
5.4 Combustible gas control surveillance for systems Change analog channel functional testing with sensor inside containment (PWR, BWR), [B) to quarterly and calibration to once cach refueling interval. Presently done monthly and quarterly, respectively.
5.5 Reactor tnp breaker testing (PWR). [Tl The vendor owner groups should consider whether more recent operating experience would justify a change in the test interval for reactor trip and bypass breakers, Licensees should pursue implementing an increase in the allowable outage time for testing reactor trip and bypass breakers as addressed in the vendor topical reports for extending surveil-lance intenals.
5.6 Power range instrument calibration done to an The owners groups should consider allowing uncertainty of 2% between nuclear instrumentation 5% uncertainty and if appropriate make a and heat balance for steady-state and transient recommendation to NRC.
conditions (PWR). [B) 5.7 CEA calculator surveillance (CE CPC PWR). [B,T)
Change channel functional test from monthly to quarterly.
5.8 Incore detector surveillance done weekly and 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> Change CE surveillance requirement to prior to use on CE plants and 7 days prior to use B&W surveillance requirement.
for B&W plants (CE and B&W PWRs) [Bj 5.9 Response time testing of isolation actuation Delete for all signals that currently include d
instrumentation (PWR, llWR). [B,T)
EDG start time, 5.10 Present calibration intenal for SRMs and IRMs is Change ST! to once each refueling interval i
quarterly for rod withdrawal block instrumentation for calibration of rod withdrawal block Sut refueling in RPS instrumentation table (BWR).
[B) 5.11 Calibration of recirculation flow transmitters (BWR).
NOTS change recommended.
[B)
See footnote at end of table.
65 NUREG-1366
~_
14 Conclusions 1
Table 14.1 (Continued).
9 Section of this TS surveillance requirement report lapplicable criteria,j Recommended change INSTRUMENTATION (Continued) 5.12 Autoclosure interlock (ACI) removal (PWR, BWR).
The NRC staff approved a generic propos.al by the Westinghouse Owners Group (WCAP-11736-A) to delete the ACL Other -
owners groups should dett rmine if the risk of core melt is higher or lower if th'c ACI is deleted. GI-99 resolution supports' deletion i
of A CI.
5.13 Turbine overspeed protection (PWR, llWR). [B,T)
Change all turbine valve testing to quarterly if turbine vendor agrees.
5.14 Radiation monitoring instrumentation (PWR, BWR).
(1) Change analog channel operational test
[B,T.W]
to quarterly from every 31 days. (2) The vendor owners groups should study whether further reductions in radiation monitor surveillance testing are possible.
5.15 Radioactive gas effluent monitor caHbration standard No TS change recommended.
(PWR, BWR). [B) 5.16 1RM and APRM channel functional tests required BWR Owners Group should determine if -
it is feasible to increase surveillance test interval from 7 days for IRMs and APRMs.
REACTOR COOLANT SYSTEM 6.1 Leak test RCS isolation valves if in cold shutdown Change 72 hours8.333333e-4 days <br />0.02 hours <br />1.190476e-4 weeks <br />2.7396e-5 months <br /> to 7 days.
for more than 72 hours8.333333e-4 days <br />0.02 hours <br />1.190476e-4 weeks <br />2.7396e-5 months <br /> if not leak tested in last
~!
9 months (PWR). [B]
6.2 Test PORVs and block valves (PWR).
No TS change recommended.To be ad-
' dressed by resolution of GI-70 and GI-94.
6.3 High point vent valve tests required every 92 days Review applicability of Catawba TS and l
(PWR), [B,T) -
. revise test frequency to cold shutdown or refueling. if appropriate.
6.4 There is no requirement for positive indication that Licensee to consider the benefit of the low-temperature overpressure protection
. providing continuous positive indication system is armed (PWR) that the low-temperature overpressure -
protection system is armed and operable.
- Implement resolution of GI-70 and 01-94.
6.5 100/E as a measure of specific coolant activity (PWR, A more meaningful mcasure of noble gas BWR). [Bj _
accumulation should be derived by industry.
See footnote at end et table.
NUREG-1366 66
1 14 Conclusions Table 14.1 (Continued)
Section of this TS surveillance requirement report
[ applicable (rliesia')
llecommended change RI'ACIOIL CODIANT SYSTDI(Continued) 6.6 Test capacity of pressuriier heaters every 92 days Change frequency to once cach (PWR). lll]
refueling interval for those plants without dedicated safety-related heater.
h.6 Demonsttate emergthey power supply to pressuriier Retain for those plants in which heaterr are heaters is operable (done every 18 months)(PWit). [11]
not permanently connected to a vital bus.
Otherwise delete.
EMEltGENCY CORl; COOLING SYSTDI 7.1 Verify boron concentration in accumulator after Change to delete boron concentration check makeup and every 31 days (PWR). [Ill if makeup from normal source (RWST) that had latest test in specifications and the minimum itWST boron concentration is greater than the minimum accurnulator boron concentration.
7.2 At least every 31 days, check for air in liCCS (PWit).
No TS change recommended.
[11]
7.3 Verify proper valve lineup every 31 days for liCCS No TS change recommended.
and containment isolation valves (PWit, llWR). [ll,R]
7.4 (1) Do analog channel operational test on accumulator licensee to examine channel checks level and pressure instrumentation (PWR). [11]
surveillance and operational history to establish the basis for satisfying frequency of testing extension.
(2) Action statement on an inoperable accumulator Add condition to action statement to require r: quires shutdown of the reactor if not fixed in I shutdown of the reactor if the accumulator hour (PWR). [Il]
water level and pressure channels are not made operable withia 72 hours8.333333e-4 days <br />0.02 hours <br />1.190476e-4 weeks <br />2.7396e-5 months <br />.
7.5 Check areas entered in containment for hose debris Change to at least once daily when entries after each entry (PWR). [11]
are made and on last entry when successive entries are made.
7.6 Venfy boron concentration of IIIT (W PWR). [R]
Check illT boron concentration by measuring concentration in boric acid storage tank, CONTAINMENT f D'?it)
Dry, Ice Condensen :iubatmospheric 8.1 Test containment spray nonles for obstructions every lixtend to 10 years, 5 years (PWR). [11]
8.2 Test 42-inch containment purge supply and exhaust No TS change recoramended.
isolation valves for leaks (PWR). [it j See footnote at end of tabic.
67 NURl!O-1366
14 Conclusions Table 14.1 (Continued)
Section ofthis TS susseillante sequhement sepori lspplicable celleria'l Recoinmended thange CONTAINMl:NT (PWl()(Continued)
Dry, tre Condenser, Substmoyiheile (Continued) 83 Venfy operability of ice condenser doors every 3 Change facquently to once each refueling months (PWR). lllll')
intesval for all doors rather than 25% cach quarter.
8.4 (1) Vacuum breakers from suppression pool to No TS change recommended; retain drywell tested monthly. ASMl! Code sequirement is monthly requirements.
quatterly (llWR). lIl]
(2) Vacuum breaker testing (suppression chamber to Clumge to 12 hours1.388889e-4 days <br />0.00333 hours <br />1.984127e-5 weeks <br />4.566e-6 months <br /> for plants with less drywell) following any steam discharge (llWR). (11]
time than that.
8.5 liydrogen : combiner tested at least once every Change STI to once cach refueling 6 months (pWR, llWR). lll]
interval.
8.6 Chemical analysis of concentration of sodium Change the analysis interval to once each tetraborate and pit of ice every 9 months (PWR,
.refuelinE nterval, i
llWR). [lI,RJ -
PLANT SYSTl?MS 9.1 Al'W pump surveillance test (PWR). [ll,Wj Change frequency of testing AlW pumps to quarteily on a staggered test basis.
9.2.1 MSIV partial stroke testing quartetty per 4.03 (PWR).
More study required.
~
[T]
9.2.2 MSIV ustial stroke testing quarterly per 4.0.5 (llWR).
No TS change recommended.
l
[T]
9,3 Verify every 12 hours1.388889e-4 days <br />0.00333 hours <br />1.984127e-5 weeks <br />4.566e-6 months <br /> that control room temperature Delete or revise requirement.
is less than specif[ed value (typically greater than HK)*It)(PWR, llWR). [Il).
1:1.IXTRIC POWEll 10,1
!!mergeng Diesel Generator Survel; lance Requirements (PWR, llWR).
When one tram or subsystem (other than an liDG)
Delete alternativo testing requirements.-
is inoperabic, test the other train and other unrelated systems includmg the dicscl generators. (ll,Wl If an offsite power source becomes inoperabl:,
Delete this requirement.
test diesels that have not been tested within ptevious 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br />. [lI,W)
See fin $1ntitc at end of table, NURl!G-1366 '
1 14 Conclusions Table 14.1 (Continued)
Section of this TS suncillance requirement report l applicable criteila'l Ituommended change eel:CTitlC POWEll(Continued) 10.1 limergency Diesel Generator Surveillance llequirements (Continued)
Action statement calls for starting an !!DG. [ll,Wj Start only once within 8 hours9.259259e-5 days <br />0.00222 hours <br />1.322751e-5 weeks <br />3.044e-6 months <br /> unless the absence of any common mode failure has been demonstrated as the cause of the !!DG inoperability.
Verify that the I!DO is loaded to greater than Change the loading requirement to be in or equal to [ continuous rating) kW in less than accordance with the manufacturer's recom.
or equal to [60] seconds for monthly or mendation.
accelerated testing based upon prior failures.[il,W]
After performing 24 hour2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> !!DG test, start 11D0 on l'eplace the 11DG start on loss of offsite loss of offsite power /SI signal within 5 minutes. [11]
power / safety injection signal with the monthly 11D0 start surveillance requirement. Add or modify the existing footnote so that the 24-hour test need not be repeated if the hot-restart test is not completed within 5 minutes provided it is performed within 5 minutes of operating the 11D0 at its continuous h>ad rating for 2 hours2.314815e-5 days <br />5.555556e-4 hours <br />3.306878e-6 weeks <br />7.61e-7 months <br /> or until operating tempera-tures have stabilized.
10.2 llattery (PWit, llWit) llattery room temperature-no present requirement.
More study required.
AC ripple degrades batteries. There is no I.CO in the More study required.
Technical Specifications [W]
ItEl'UELING None identified.
SPECIAL TEST EXCEPTIONS (PWit) 12 liach full length control rod not fully inserted shall be Change 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> to 7 days. If a rod drop test demonstrated trippable from at least 50% withdrawn has been performcd within this time, another within 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> prior to reducing shutdown margin.
test is not necessary.
[Ill See footnote at end of tame.
69 NUltilG-1366
14 Conclusions
!alite 14.1 (Continued)
Stt tion of thh
'I S sun tillaine irtiuliement e r giur i lagi >litante tiiteria*l itetusinnended chante lt AlllO ACl XVI', l'.I i 1.1T.NTS (PWlO 13 Number of curies in waste gas tank inutt be monitored
'lhe quantity of radiivicine rnaterial every 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> when additions are being inade. [11,11l contamed in each waste gas decay tank shall be determined to be within the limit at least once every 7 days whenever radioactive materials are added to the ts:ik, and at least once every 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> during primary coolant system degr.ssing operations.
- 11,r f..us ar. pani.ic corim used in thiutody air. 11 - lur nwr t>mden m terrnoif imir, T - go.ni uanuent. W - unnnewuy
- car; and it -
t athaiion riluute to lwivinnel NUiti!G-13%
70
15 lillil.10GRAPilY Al!OD/C503 U.S. Nuclear Regulatory Commis-efficient limit hicasurements,"
sion, Office for the Analpis and September 8,1986.
litaluation of Operational Data,
- 11. O. Ornstein, " Decay IIcat Re-G1.84-5 U.S. Nuclear Regulatory Commis-moval Problerns at U.S. Pressurized non " Change to NURiiG-10211J.
Water Reactors," Al!OD/C503, censing Ihaminer Standards," Oc-December 1985.
neric letter 84-5, April 2,1974.
01,84-15 U.S. Nuclear Regulatory Commis-II AW-10167 liabcock & Wilcox Co., "Justifica-sion,* Proposed Staff ActionstoIm-tion for increasing the Reactor Tnp System On 1jne Text Intervals,"
prove and Maintain Diesel Genera-tor Reliability," Generic letter Volume 1, ll AW-10167, May 1986.
84-15, July 2,1984.
Capstick,1988 I ctter from R. W. Capstick, Ver-GS!!R,1980 U.S. Nuclear llegulatory Commis-mont Yankee Nuclear Power Corp.,
sion, "IlWR Scram Discharge Vol-to NRC, " Vermont Yankee Re-ume System," Generic Safety sponse to USNitC llequest for Ad-Evaluation Report, December 1, ditional Information-Surveillance g 939, Testing of l!CCS and St.C Equip-ment," J uly 15,1988.
Ilood,1986 lxtter from D.11000. NRC, to 11.11. Tucker, Duke Power Co.,
Cl!N-327 Combustion lingineermg Ownets
" Issuance of Amendment No. 64 le Group, "RPS/I!SFAS lixtended Facility Operating license NPF-9 Test interval Evaluation," CEN-and Amendment No. 45 to Facility 327, hiay 1986.
Operating 1.icense NPF McGuire Nuclear Station, Units 1 Chilk,1988 Memorandum from S. J. Chilk, Of-and 2," September 30,1986.
fice of the Secretary, to V. Stello, lido," Staff Requirements-llrief-llood,1988
- 1. citer from D. Ilood, NRC, to ing on Technical Specification R evi.
I1. II. Tucker, Duke Power Co.,
sions" (June 20, 1988 meeting),
" Issuance Notice of Amendment
~
July 6,1988.
No. 83 to l'acility Operating li-cense NPF-9 and Amendment No.
Circular 77-11 U.S. Nuc! car llegulatory Commis.
64 to Facility Operating ljcense NPF-17-McGuire Nuclear Sta-sion, " Containment Leakape Due to Seal Deterioration " Circular tien, Units 1 and 2 (FAC Nos.
77-11, September 6,1977.
59239/59240)," May 11,1988.
IN Ss-23 U.S. Nuclear Regulatory Commis-Crocker,1987 letter irom L P. Crocker, NRC, to s on, " Potential for Gas 11inding of J. P. O'Reilly, Georgia Power Co.,
High Pressure Safety injection
" Issuance of Amendment Nos.136 Pumps During a less-of-Coolant-and 75 to Facility Operating 1.t-Accident," Information Notice censes DPR-57 and NPF-5-Ed-88-23, May 12,1988.
win 1. llatch Nuclear Plant, Units 1 and 2," May 13,1987.
IN 89-42 U.S. Nuclear Regulatory Commis-sion, " Failure of Rosemount Mod-lingle,1984 lxtter from L 11. Engle, NRC, to els 1153 ind 1154 Transmitters,"
W. L Stewart, VEPCO, " Turbine informatic n Notice 89-42, April 21.
Overspeed Protection System Sur-
- 1989, veillance Intervals," Apnl16,1984.
Jabbour,1957 1 ettet from K. N. Jabbour, NRC, to Engle,1986 letter from L 11. Engle, NRC, to 11.11. Tucker, Duke Power Co.,
W. L Stewart, VEPCO, "End of
" Deletion of Technical Specifica-Cycle Moderator Temperature Co-tion Surveillance Requirement 71 NURiiG-1366
-... -. - _,.. - ~
O N;agurk j
4.6.1.9.4 for Containment Air ite.
N!!DC-30851 General !!!cctric Co., _"lechnical lease and Addition System Specification improvement Analy.
Valves-Catawba Nuclear Station, sis for ilWit Itcactor Protection Units 1 and 2 (l'ACS 63780/
System," N!!DC-30851. May 1985.
63781)," July 31,1987.
N!!DC-30851P General llicctric Co., " Technical Jabtuur,1988 letter from K. N.Jabbour, NitC, to (Suppl.1)
Specification improvement Analy-11.11, Tucker, Duke Power Co.,"Is.
sis for llWit Control Itod lilock t
suance of Arnendment No. 51 to Instrumentation," N!!DC-30851P -
Facility Operating IJcense NPP-35 (Supplement 1), June 1986.
and Amendment No. 44 to Pacility i
Operating 1.icensee NPP N!!DC-30851 General lilectric Co., "rechnical Catawba Nuclear Station, Units I (Suppl. 2)
Specification improvernent Analy-and 2 (TACS 67765/67766),"
s s for llWR Isolation Instrumenta-July 28,1988.
tion Common to itPS and !!CCS Instrumentation,"
NEDC-30851 Kane,1987 Memorandum frorn W. P. Kane, (Supplement 2), July 1986.
NitC llegion 1, to S. A. Varga, NitC,"1.icensing Action Iteview for NilDC-30936 Genernt lilectric Co.,
"l. LWR Susquehanna Steam I!!ectric Sta-Owners' Group Technical Specifi-tion Units 1 and 2, Technical Speci-cation Improvement Methodology fication Change Regarding Scram With Demonstration for llW R l
Discharge Volume System Surveil-liCCS Actuation Instmmentation,"
lance itequirements," August 31, N!!DC-30936, June 1987, 1987.
NiiDC 31677P.A General lilectric Co., "rechnical liatt,1988 letterirom 11. Katz, Southern Call-Specification improvement Analy.
fornia lidison, to !!. Ilutcher, NRC, sin for llWR isolation Actuation in.
" Waste Oas Storage Tank Activity strumentation," NHDC 31667P A, IJmit," August 26,1988.
July 1990.
IJ!R 874158 Shearon llatris, Docket No.
NP-4264, Vol. I lilectric Power itescarch Institute, i
50-400,111lt 87-058,1tev.1 Fultures Related to Surveillance Test-ingvfStandbyEquipment, Volume 1:
1 1J!R 884)05 Millstone 3, Docket No. 50-423,
_ llmergency -
Pumps,"
IJ'RI IJilt 884)05.
NP-4264, 1 Mollers !!ngineering IJilt88-006 Millstone 3, Docket No. 50-423,
! !!R 8M)06.
NP-4264, Vol. 2
!!!cctric Power Research Institute, failures Related to Surveillance MDl! 1031184 General lilectric Co., "lidwin I.
Terring of Standby Equipment, Vol-llatch Units I and 2 Relaxation of ume 2: " Diesel Generators "1.PRI Scrum Discharge Volume Vent and NP-4264, September 1985.
Drain Valve ClosureTimes," MDil 1031184, Rev.1, December 1984.
NP-5924 -
Electric Power Research Institute,.
Surveillance Monitoringand Diag-Murley,1988 letter from T.11. Murley, NitC, to nostic Techniques to improve W. S. Wilgus, Chairman, !!&W
~ Diesel Generator Itcliability," Pie Owners Group, May 9,1988, nal Report NP-5924,' July 1988.
l Murphy,1987 Letter frorn W. P. Murphy, Ver.
NPE-1 Nuc/ car Powr Orcrience; Volume mont Yankee Nuclear PowerCorp.,
PWR-2; !!ook-3, " Experiences";
to T.11. Murley, NRC, ' Surveil.
Vil-Safety
- Containment Pressure Sup-liquipment/' Supplement I to Pro-pression," pp.
1-51 (January psed Change No. 85, December 7, 1973; July 1988); S.. M. Stoller -
.1987.
Corp., lloulder, Colorado.
- 72
-~
a-
~.
13 Ilibliogmphy NPl!-2 Nuclear Pourr Everience; Volume Review of SJety Analysis Reports PWR-2; llook-2, " Experiences";
for Nuclear Power Plants," l WR VI-Turbine Cycle
- Systems, Edition, NUREG4)S00, July 1981.
Section A "Purbine," pp.1-38 (January 1973-July 1988): S. M.
NUREG-0839 U.S. Nuclear Regulatory Commis.
Stoller Corp., lloulder, Colorado.
sion, "A Survey by-Senior NRC Management To Obtain View-NPli-3 Nuclear Pourr Eyrrience; Volume points on the Safety impact of PWR-2; llook-1, " Experiences";
Regulatory Acthities From Repre-IV-Control Rods and Drives, Sec-sentative Utilities Operating and tion A " Control Rods," pp.1-14 Constructing Nuclear Power (January 1973-July 1988); S. M.
Plants," August 1981.
St911er Corp., lloulder, Colorado.
NUREG-1024 U.S. Nuclear Regulatory Commis-NPE-4 Nuclear Poner Everience; Volume sion, "rechnical Specifications-PWR-2; llook-2, " Experiences";
Enhancing the - Safety : Impact,"
VI-Turbine Cycle Systems Sec-NUREG-1024, November 1983.
tion D " Steam " pp.1-153 (Janu-ary 1973-July 1988); S. M. Stoller NUREG-1144 U.S.. Nuclear Regulatory Com-Corp., lloulder, Colorado, mission.
- Nuclear Plant Aging Research (NPAR)
Program,"
NPl!-5 Nuctrar Powcr Eyrrience; Volume NUREG-1144 (Revision 1), Sep-PWR-2; llook-1, " Experiences";
tember 1987.
V-Rcactor Coolant System, Sec-tion C " Relief and Safety Valves,"
NUREG/CR-1341 U.S. Nuclear Regulatory Commis-pp. 42-43 (January 1973-July sion," Regulatory Analysis for the 1988); S. M. Stoller Corp lloulder.
Resolution of Generie issue 115:
Enhancement of the Reliability of the Westinghouse Solid State Pro-NRC,1987 U.S. Nuclear Regulatory Commis-tection System," April 1989, sion," Interim Policy Statement on TS Improvements," $2 l'R 3788, NUREGICR-3883 U.S. Nuclear Regulatory Commis-February 6,1987, sion, " Analysis of Japanese-U.S.
Nuclear Power Plant Mainte-NSAC-96 Nuclear Safety Analysis Center, nance," NUREG/CR-3883, llat-
"Effect of Diesel Start Time on te!!c Pacific Northwest 12boratory, llWR/6 Peak Cladding Tempera-June 1985.
ture: licensing Basis Sensitivity Calculations," NSAC-96, General NUREG/CR-4335 U.S. Nuclear Regulatory Commis.
Electric Co., January 1986.
sion.
- Potential lienefits Obtained by Requiring Safety Grnde Cold -
NSAC-130 Nuclear Safety Analysis Center, Shutdown Systems," NURl!C/
"the Effect of Diesel Start Time CR-4335, Sandia National labora.
Delay on Westinghouse PWRs,"
tories, July 1985..
liPRI-NSAC-130, Electric Power Research Institute, Palo Alto. CA, NUREG/CR-4457 U.S. Nuclear Regulatory Com.
1988.
mission," Aging of Class 1E llatter-les in Safety Systems of Nuclear NUREG-0713 U.S. Nuclear Regulatory Commis.
Power Plants," NUREG/CR-4457, sion, " Occupational Radiation Ex-EG&G Idaho, Inc., July 1987.
posure at Commercial Nuclear Power Reactors and Other Facili-NUREG/CR-4590 U.S. Nuclear Regulatory Commis-ties,1985," Eighteenth Annual Re-sion, " Aging of Nuclear Station port, NUREG-0713, Volume 7, Diesel Generators: Evaluation of April 1988.
Operating and Expert Experience, Phase 1 Study" Vol.1 NUREG/
NUREG-0800 U.S. Nuclear Regulatory Commis-CR-4590, Pacific Northwest labo.
sion," Standard Review Plan for the ratory, August 1987, 73.
15 liibliography NUltliG/ Cit-4597 U.S. Nuclear llegulatory Com.
NUltliG/ Cit-5200 U.S. Nuclear llegulatory Commis-rnission, " Aging and Service Wear sion, "l! valuation of llisks Associ-of Auxiliary l'cedwater Pumps ated With AOT and STI Itequite-for PWit Nuclear Power Plants,"
ments at the ANO-1 Nuclear Vol.1. " Operating lixperience and Power Plant," NUlt!!G/ Cit-5200, l'ailure Identification," NUlti!G/
llrookhaven National laboratory, Cit-4 597, Oak Itidge National August 1988.
laboratory, July 1986.
SAIC 40/1394 Science Applications International NUltliG/ Cit-4692 U.S. Nuclear llegulatory Corntnis.
Corporation, "the liffects of Pro-sion, " Operating fixperience lle-posed Changesin the New Standard view of li ilures of Power Operated Technical Specifications on Nuclear a
Itelief Valves and lilock Valves in Power Plant itisk," SAIC-90/1394, Nuclear l'ower Plants," NUlti!G/
June 28,1991.
Cit-4692, Oak Itidge National 1atwatory, October 1987.
S!!CY-83-472 U.S. Nuclear llegalatory Commis-sion, "litnergency Core Cooling NUlti!G/ Cit-4715 U.S. Nuclear llegulatory Commis.
System Analysis Methods," SliCY-uon,"An Aging Assessment of ite-83-472, November 17,1983, lays and Circuit lireakers and Sys.
tems Interactions,"
NUltl!G/
SI!CY-86-310 U.S. Nuclear llegulatory Commis.
Cit-4715, I ranklin itesearch Cen.
sion, " Proposed Technical Specifi-ter, June 1987, cations Policy Statement," SliCY-86-310, Octolar 21,1986.
NUlt!!G / Cit-4999 U.S. Nuclear llegulatory Commis-sion, *listimation of Itisk Iteduction S!!CY-87-314 U.S. Nuclear llegulatory Commis-1 rom improved POltV lleliabihty sion,
- Staff itequirements Interirr.
in PWits " March 1988.
Policy Statement on Maintenance of Nuclear Power Plants and Ite-NUltt!G/ Cit-5078 U.S. Nuclear llegulatery Commis.
quirements I: rom Status liriefmg sion, "A lleliability Program for on January 7,
1988," S!!CY-limergency Diesel Generators at 87-314,1 cbruary 25,1988, Nuclear Power Plants Program Structure," Vol.
1, NUluiG/
Shiffer,1987a Ixtter from J Shiffer, PO&li, to Cit-5078, Science Applications in-NitC, *ltemoval of 111 111 System ternational, Inc., April 1988.
Autoclosure Interlock Function,"
August 4,1987.
NUlti!G/ Cit-5158 U.S. Nuclear llegulatory Commis-sion, " Worldwide Activities on the Shiffer,1987b 1.etter from J.D. Shiffer, PG&l!, to lleduction of Occupational lixpo-Document Control Desk, NitC, sure at Nuclear Power Plants,"
"limetgency Technical Specifica-firookhaven National 1.aboratory, tion Change to TechnicalSpecifica-June 1988.
tion 4.3.4.1.2a and b. ' Turbine Overspeed Protection'" Novem-NUlti!G/ Cit-5186 U.S. Nuclear llegulatory Commis-ber 8,1987.
sion, " low Ternperature Over-pressure Systems livaluation for Split report See Murley,1988.
Pressurized Water Itcactors," July 1988.
Stewart,1986 Letter fror, W,
I.
Stew 1trt, VliPCO, to 11,11. Denton, NitC, NUltliG/ Cit-5197 U.S. NuclGr Itegulatory Commis.
" Virginia Illectric and Power Com-sion, "I!valaation of Generic issue pany North Anna Power Station 115: linhancement of the lleliabil.
Units No. I and 2 Proposed Techni-ity of Westinghouse Solid State Pro-cal Specifications Changes," Janu-tection System," January 19S9.
ary 3,1986.
NU111iG-1366 74
15 liibliography Tourigny,1987 I etter from I!. G. Tourigny, NitC.
WCAP-10271 Westinghouse lilectric Corp.,
to C. O. Woody,11orida Power and (Suppl.1)
"livaluation of Surveillance 17 c-1Jght Co., *lssuance of Amend-quencies and Out-of Service Times ments (TAC Nos. 66014 and for the Itcactor Protection Instru-66015)," October 28,1987.
mentation System," WCAP-10271 Trammell,1987 letter from C. M.Trammell, NitC.
to j. D. Shiffer. PG& l!. " Issuance of g ggg,-10271 Westinghouse
!!!cctric Corp.,
Amendment (I AC No. 66541),
Buwl. 2)
livaluation of Surveillance 1 re.
November no,1987.
quencies and Out-of. Service Times for the !!ngineered Safetyl catures Tucker,1985 lxtter from 11.11. Tucker. Duke Actuation System, WCAI -10271 Power Co., tc !!. O. Adensam, (S.upplernent 2), Itevision 1, March NitC, '"I'echnical Specification I9" '
~
Amendment for Containment Ice Condenser," July 12,1985.
WCAP-ll525 Westinghouse
!!lectric Corp.,
Tucker,1986 Ietter from 11. II. Tucker, Duke "Probabilistic !! valuation of itedue-Power Co., to 11.11. Denton, NltC, tion in Turbine Valve. Test lire-
" Technical Specification Amcr.J.
quency, WCAP-11525, June 1987.
rnent, November 17,1986.
CAP-H ng ouse ecjnc
- Corp,
'lucker,1988 letter frorn 11.11. Tucker, Duke 1
System Auto Gome Inter-Power Co., to NitC, " Diesel Gen-erator Surveillance 4.8.1.1.2g.7..
lock itemoval lleport for the Wes, tinghouse Owners G roup, 12cbruary 15,1988.
WCAP-il736, l'ebruary 1988.
WCAP-10271 Westinghouse filectric Corp.,
"l! valuation of Surveillance 1 re-Woody,1987 lxtter from C. O. Woody, I'lorida quencies and Out of. Service Times Power and iJght Co., to Document for the lleactor Protection Instru-Control Desk, NitC, "Special Test mentation System," WCAP.10271, lixcepoons Shutdown Margin,'
January 1983.
August 17,1987.
75 NUl(1:0-1366
Al'I'ENDIX ROUTINE INSPECTIONS (SURVEllslANCE TESTING)
AT JAl'ANESE POWER PLANTS *
- $ource: NURI.GCR-3833.
77 NUl(EG-1366
Appendix Table 1 PWR sunelliance tests under normal operation items and test Frequency I)escription Diesel Generator o bianual stuting Once/ month Start and manually load the diesel generator from the central c(mtrol room to obtain operating data, including measurement of time to come up to speed.
Control flod Drive hicchanitm Twice/ month l$ 20 step insertion and pull-out for all banks (except control group banks) used in reactor control.
Safety injection System bianual starting of high pressure Onec/ month Checking of pump operation manually starting o
injection pump from control nom.
Operation of motor-operated Once/ month Checking of valve operation by manually valve opening / closing from the control mom.
Ilesidual lleat Removal System Once/ month Checking of pump operation by manually -
starting from the control mom.
Containment Spray System
- hianual starting of pump Once/ month Checking of pump operation by manually starting from the control mom.
Operation of motor-operated Once/ month Checking of valve operation by manually o
valve opening / closing from the control roorn.
licating and Ventilation System hianual starting of ventilation fan Once/ month Checking of fan operation by manually o
for containment vessel starting from the control room.
hianual starting of emergency Once/ month Checking of fan and dareper operation by e
filter fan for the control room manually starting from the control room.
'8 NUREG-1366 '
7
Appendix Table 2 IlWR surveillante conducted during normal operation items and test I'requency Description Diesel Generator o Manual starting Once/ month Start and manually load the diesel generator from the control room to obtain operating data including measurement of time to come up to speed.
Control Rod Drive System o One-notch test Once/ week Withdraw and insert all the control reds by one notch.
Reactor Protection System o Functional test of average power Once/ week Confirm the function of average power range range monitor with signals of malfunction and high neutron flux.
o Manual scram test Once/ month Push the rnanual sera n button on one side in the main control room and confirm the scram signal of the channel on the same side.
Stand-by Oas Treatment Systern o 17unctional test Once/ month Start the system manually and confirm the normal movement of auto valves and the normal function of the system.
Stand by Liquid Control System o Manual operation of pump Once/ month Confirm the nornial movement of the pump and the normal function of the system.
o ~ Density of poison Once/ month Sample and confirm density in the prescribed
- range, o Volume and temperature of Once/ week Confirm volume and temperature of poison poisen in the prescribed range.
l!cactor Core Spray System o - Manual starting of pump Once/ month Start the pump manually from the main control room and confirm normal function of.
the pump.
o Manual operation of motor valve Once/ month Open and close the motor valve manually from the main control room and confinn normal movement of the valve.
Residual 1-leat Removal System
.o Manual starting of pump Once/ month Start the pump manually from the main' '
contro' roorn and confinn normal function of J the pump.
i 79 NUREG-lM6
.?
~ ~. -
n r,
i Appendia Table 2 (Continued) ltrms and test l'requency Destilption 1(cactor Auxiliary Cooling Water Splem Manual starting of pump Oncc/ month Start the pump manually frorn the main o
control roorn and confirm normal function of i
the pump.
f Iligh.Picssure Coolant injection System Manual starting of pump Once/ month Start the pump manually from the main '
o contr01 room and confirm norinal function of the pump, Manual operation of motor valve Once/ month Open and close motor valve manually from o
the control room and confirm normal involvement of the valve.
Main Steam isola. ion Valve e
10% closed Once/wcek Close MSIV manually 10 90% opened posithm und confirm normal snovement of the valve.
17ully closed Once/3 rnonths Close MSIV *nanually at < 60'fc power and e
measure its f ull stroke tirne, 1
NUR!!O-1366 80
t#4C P OHM 3%
- 0. 8 NUCL E AR HE GUL AT 06W COMh05510N
- t. Ht POHT tMMM R q2493 (ASUg'ed t>y tJ54C. Add VOI.
tJALM 1t02, Supp. Hev., arid Anoenourvi Num-
"1. W BIDLIOGRAPHIC DATA SHEET
- " < d 8 (s
in.iruet.one on in. rev.r..'
NURliG-1366 2 IllLL Af JD 60tilllLt 3= DAIL HLPOH1 PVOL.lbtti D improvements to Technical Specifications Suncillance Requirements uun yg n December 1992 4, f lPJ 04 GHANT NUMo[H N/A f,
Av I t1J H 6 J
- 6. 1YPL OF HLPQHT R. lubel, T. R. Tjader 1 inal Topical L Pt HsOD COVt HLD UnclutWe Dates)
N/A 4 Pt HFGHMiJ;,4 OHO AN;/ alton - PvAME AND ADDHL 68 pf tJ64C, proviJe Division, Ottice w Heg*un, U 6, t Aspear Hegy: story Commission. and mathng a$1ress; it contractor. (ironos hame and ftietting address )
Division of s,perational Events Assessment Of fice of Nuclear Reactor Regulation U.S. Nuclear Regulatory Commission Washington, DC 20555 9 6PotADHtNO OHUAttl A fiON - NAML AND ADOHL S$ Of PetC. type '6ame as atme'; it contractor, provios NHC Division, Office or Hogm U B. Nuclear Hegulat.ity Commission, and ma:ltng address.)
Same as ah>ve 10 SUPPL t.ML N T AR Y NO I L S Report Completion Date: May 1992 n Ass:H Ac T troo ord. o"f I.. 3 In AuEust 1983 an NRC task group was formed to investigate problems with survedlance testing required by Technical Specifications, and to recommend approaches to effeci improvements. NUREO-1024 (" Technical Specifications-Enhancing Safety impact") resulted, and it contained ruommendations to review the basis for test frequencies; to ensure that the tests promote safety and do not degrade equipment; and to review surveillance tests so that they do not unnecessarily burden personnel.
'lle Technical Speelfications improvement Program (TSIP) was established in December 1984 to provide the framework for rewriting and improving the Technical Specifications. As an element of the TSIP, all Technical Specifications surveillance requirements were comprehensively examined as recommended in NUREG-1024.
The results of that effort are presented in this report. The study found that while some testing at power is essen.
tial, safety can be improved, equipment degradation decreased, and unnecessary personnel burden relaxed by re.
ducing the amount of testing at power.
- 42. xrv wcooszOesenioTOns iti.i ord. or pnra... inai iii a.gi.i,...aren.r. in tocatir.g in. report i is Av^iL^bLITY ST ATEMENT Unlimited 14, $EcultTY CLASSIFICATION Reactor Plant Technical Specif.ications Surveillance Tests Safety Enhancement Unclassified (This ReporO Unclassified 16 NUMOLH OF PA(.Ai. 8
- 16. PH6CE NHC FORM :Ds ( -69)
3
_m~4_sw4 J
mAes
.A.MA s.ddehMM-Ar. 4 M3""
'_," ' ""' ' ".Z4'
' I n Adha.4.a ABaa.m
&AJAh 4ss--c.d n+_
A
-AJJu-A%
_3-O--ss a_a MMLsh he d._m.6A_..
.6Li ami.wma-L.m.u h h lauwadAJ A A h. nen, 0
i F
G 5
- t v
i
-.l 4
h w
I' i
.t I
v 3
P 1
i 4
P f
e
(
i h
5 h
d 4
i t:
w ar m,-,mne,
.w,-,
s.
.--ae,
.,_m______
Printed on recycled paper r deral Recycling Program e
DECEMBER'1992
- NUREG-13M IMPROVBf ENTS TO TECHNICAL SPECIFICATIONS SURVEILLANCE REQUIREMENTS UNITED STATES sncat rounm class set NUCLEAR REGULATORY COMMISSION msuct ase nts emo uswc WASHINGTOff, D.C. 20555-0001 PERMIT NO, G 67 OFFICIAL BUSINESS PENALTY FOR PRNATE USE, $300
- - +.:
.--r
~
't^
^
n
-7
,, +.
7;
,;,;,+,.- - -
^
9
,e e ae 4
I
_ _ _