ML20206B151
| ML20206B151 | |
| Person / Time | |
|---|---|
| Issue date: | 06/15/1988 |
| From: | Advisory Committee on Reactor Safeguards |
| To: | Advisory Committee on Reactor Safeguards |
| References | |
| REF-GTECI-A-46, REF-GTECI-SC, TASK-A-46, TASK-OR ACRS-2585, NUDOCS 8811150352 | |
| Download: ML20206B151 (19) | |
Text
-
hbN ~$5 f ?, % ' %.
' T' f Q,.,
75.
CERTIFIED COPY
?
DATE ISSUED: June 15,
'd. $
. i 1988 ACRS SUBCOMMITTEE ON RELIABILITY ASSURANCE (EO-RISK SCOPING STUDY; A-46 UPDATE)
JUNE 14, 1988, WASHINGTON, D.C.
Purpose A resting was held by the ACRS Subcodinittee on Reliability Assurance on June 14, 19E0.
The purpose vf this freeting was to review the final cutcome of the Equiptrent Qualification (EQ) - Risk Scoping Study.
An update on the implementation of the USI A-46, "Seismic Qualification of Equipment in Operating fluclear Power Plants" was heard, Presentations were r.'ade by the NRC Staff and their contractors.
Notice of this meeting was published on June 1,1988, in the Federal Register. Attach-merit A is a schedule of presentations.
The meeting was entirely open to the public.
Richard Major was the cognizant Staff Engineer for the tre e t i ng.
8811150352 00061D PDR ACHS 2585 PNV Attendee 5
\\
ACRS NRC Staff C. Wylie, Chairnan M. Dey C. Siess, tieniber W. Fa rrer C. Michelson, Member R. Major, ACRS Staff i
Others L. Bustard, Sandia National Lab.
A. Kolaczkowski, SAIC p
1, NtWARC -
g
[
Dh ry.,
c w
l Minutes / Reliability Assurance 2
Mtg. June 14, 1988 C. Guild, Bishop Cook-Purcell & Reynolds P. Ecwell, Bechtel
- 5. Arndt, ACRS Fellow R. Schaffstall, KMC W. Schmidt, MPR Introducticn - Chairman C. Wylie Mr. Wylie began the meeting by reminding the Subcommittee members that the ACRS has been asked to com.ent on the E0-Risk S:oping study.
He thoucht the process used in this study was a logical way to determine prioritics in EQ research. How the St6ff intends to proceed with the information developed was not yet clear.
Mr. Michelion questioned how this effort complimented nther EQ efforts.
He wished to know what else is being done in the equipment qualification area. He cautioned that traditionally equiptrent is either assumed to faii or be operating normally, attention should be given to off-normal or aberrant operation. Analysis of the effects of strail pipe breaks outside containtrent for EQ effects was cited as an area that deserves more attention.
Dr. Siess questioned the usage of the term "accident" in the EQ-Risk Scoping study.
He explained that in PPA's accident often means a severe accident beyond the design basis event. However, EQ research has usually dealt with the enviror. rent created by the design basis accident, not severe accidents.
Mr. Michelson suggested a problem area in this study. Equiptrent re-liability estimates used in PRAs are based on normal operation l
l Pinutes/ Reliability Assurance 3
Mtg. June 14, 1988 conditions, however, given an accident environment, equipment will be challenged beyond normal operation circumstances. The question remains how valid are equipment reliability estimates given accident conditions?
Staff's Intrcduction en EQ-Risk Sec> ping Study - Moni Dey, NRC/RES Dr. Dey explained that hRC has traditionally funded EQ research to study the methods for qualifying safety-related electrical equipment to oemonstrate its ability to function during and following design basis accidents which produce harsh environments. The goal of this research wa; to confirm compliar.ce with 10CFR50.49 "Environmental Qualification of Electric Equipment important to Safety for Nuclear Power Plants," and the associated regulatory guidance. The EQ scoping study was initiated to determine the risk importance and priority for NRC funded EQ re-search.
Mr. Bohn of Sandia Labs. discussed a short coming of PRA methods.
Given a accident sequence where the power plant must remain at an intermediate state for several weeks before final actions are taken, EQ issues will have an inpact. Equipment could be required several weeks after the accident, not several hours later as assumed in pRA analyses. Equipment coulc be called upon to survive a harsh environment for long periods of tire.
The objective of the EQ scoping study was to determine the risk signifi-cance of electrical equipment that is essential to preventing core-melt accidents as well as ritigating the consequences of severe accidents.
P.inutes/ Reliability Assurance 4
Mtg. June 14. 1983 Also under study were EQ assumptions and issues pertaining to the performance of essential electrical equipment.
Results of the study will be used to:
Petermine the need for identifying new generic issues (none o
were found),
Support the implementation of the Comission's severe accident c
policy for existing and future plants (What EQ concerns should Le cddressed?),
o Determine if a need exists for further EQ research to charac-terize areas of large uncertainty ar.d.
o Frioritire any additional EQ research.
ACRS comments are being requested on this work in July. A final NUREG f
report is scheduled fer publication by July 29, 1988.
A research information letter is scheduled for issuance by September 30, 1958.
Sumnary of Other EQ Related Work - W. Farrer, NRC/RES Mr. Famer explained that the majority of EQ research related to electrical equipment has been conducted at Sandia Labs. The majority of the rechanical EQ work has been conducted at INEL. The electrical woi't has been basically unfunded since FY-86. Only small tasks such as
Minutes /Peliability Assurance 5
Mtg. June 14, 1988 report writing and editing was continuing. The mechanical work is still ongoing.
Valve seismic testing is being performed in Germany. Work is also underway at Wyle Labs, where motor-operated valves (MOVs) are being tested to see if they can isolate a high energy line break.
The mechanical work will continue through FY-89.
Containment isolation valves will be studied ur. der severe accident loads.
There is no EQ research scheduled beycnd FY-89, l'r. Michel3cn asked if the effects of harsh environments on solid state electrical equipment was covered in this study.
It was not. This was characterized as an important limitation.
Several makes of solenoid operated valves (S0Vs) were tested for the effect of aging. Numerous probler.s were encountered during the tests, where the valves failed to function. Mr. Michelson suggested 50V testing as a topic for the next Reliability Assurance Valve topical reeting.
EQ-Risk Scopina Study:
Project Objectives /EQ lssues - L. Bustard, Sandia Labs.
The scope of the EQ-Risk Scoping Study was electrical safety-related equipment that must function during a harsh environment.
Examples of this equipment are cables, valve actuators, and pump motors.
i
Minutes / Reliability Assurance 6
Mtg. June 14, 1988 Existing FRAs were used as a basis for the study. The objective was to assess the impact of electrical equipment environmental qualification or lack thereof on reactor risk and its uncertainties. The study was to identify ony analyses or testing that may be necessary to reduce the risk or its untertainties steming from lack of qualification of equip-cent important to safety.
The EQ-Rist Scoping Study is reviewing the appropriateness of design basis accident (DBA), EQ assumptions. Have accident conditioas been t.-deled correctly? The study is also attempting tc determine the wpurtance of ren-DBA accident envircrrents and sequences. Also under study was the lack ut equipment accident reliability information.
Examples of sinulating accident conditicns were given.
Is the sequential exposure of equipent to radiation followed by stean an adequate simulaticn of the accident or are synergestic effects important? Can a gamma radiction source be used to sir.ulate the mixed (accident) beta /gama radiation?
Other EQ issues coc.sidered included rethods for simultting aging con-ditions.
Special topics were also considered relating to equipment cualification.
For instance, are there test approaches (such as nulti-ple sample, fragility and overstress testing) which can compliment qualification type-testing to ar.hieve a broader safety perspective?
Other important severe accident environments and sequences should be considered such as steam venting from the BWR drywell to toe reactor building.
Another issue is PRA modelling of non-safety-related
l Minutes / Reliability Assurance 7
titg. June 14, 1988 4
ec,uipment such as that necessary for feed and bleed cooling.
A separate issue is wh:t PPA risk uncertainties occur from the fact that PRA typically empioys normal operation reliabilities to describe harsh l
environment equipment operation?
4 PRA and EQ Perspectives important to EQ-Risk Scoping Study Project Approach - L. Bustard, Sandia Labs.
Mr. Bustero discussed some PRA perspectives.
PRA risk significant time frarces are much shorter, than the accident time frames assuir.ed in EQ research.
PRA's are being used to model post-core melt eccident manage-l ment strattgies. As this work develops the staff can assess the risk i
significer.ce of what equipment is involved and the associated EQ issues.
PRAs typically c.ssume that normal operation reliabilities 6pply during accident conditior.s.
Finally, PRA does not have in its models details i
regarding plant status instrun.entation, and how the operator is going to respond to possibly erroneous plant status indicators.
Froni an EQ perspective equipent descriptions are in much greater detail l
i than represented in PRAs.
The EQ approach to equipment operability is l
to assure equipent is rr.uundant, and a successful type test on a single j
t j
component demonstrates an adequate safety n.argin, The EQ radiation source terms are about the same as the PRA assumptions.
The EQ-important accident sequences mo:leled in the PRA's have been accounted f
for in the EQ multi-environment testing.
l Project Approach - L. Bustard, Sandia Labs.
{
r
_ -. - ~,.,., _ _. _. _ _., _., - _ - _
Minutes / Reliability Assurance 8
'Mtg. June 14, 1988 In order to scope the risk significance of i.Q issues, several "candi-date equipment operations that must be accomplished in a harsh environ-ment were selected.
Then a list of candidate risk significant equipment whose operability is desired during harsh environments was developed.
Qualitative EQ oata sources were revieweo to identify those EQ compo-nents for which accidtnt reliabilities may d'..fer from normal operation relia bili tit.s.
This inethod focused the selection of equipment op-eration* for further study.
The selected equipment operations were evaluated in more detail. The evaluation determined hos FRAs model equipment operation.
The equipment operatior.s were assessed to deterrnine which secuences/ environments PRAs sucgested were ir..portant. Tre risk impact of equipment operation was evalu6ted. The applicability of historical equiptrent operability issues were assessed.
Finally, generalized conclusions were developed after evaluatior, of seve*al equipment operations for both PWR and EWR plants.
This project approac, provides a reasonable method for elintinating certain EQ issues frei further consideration.
It is also useful for identifying the direction cf further study.
I Identification of Harsh Environrrent Candidate Equipment Operations -
Alan Kolaczkowski SAIC The purpose of this task was to identify from a FRA perspective imoor-tant equiprent operations which could be impacted by a harsh environ.
ment. The process used was two fold.
The literature was reviewed to
Minutes / Reliability assurance 9
Mtg. June 14, 1988 e
find representative PRA importance studies to identify risk significant equipment.
Seccndly, a formalized process of rankirg equipment based on equiptent importance (from an accident mitigation or prevention sta.idpoint), equipment location, and potential envirnnrents for each ccident type reviewed was perforred.
The literature resiew identifico FORVs, SRVs, MSIVs, pressure and leve; senscrs, and ccntainment heat removal systems as pcte.ntial EQ signifi-cant equipment.
It Osn po','Ed out the need to develop a process that focuses on identif) g, significant equipment with EQ potential as past work has not been designed to accomplish this.
The formalized process wcn designuJ to give the task an EQ perspective.
Rather than specific :,ccident sequences, f our generic accident r,at-egories were coniidered fur the environrents asscciated with them.
Four logical plant areas were picked based on the severity of potential accident enviror. rents (insice the prircry pressure boundry. inside centainment, inside the reactor building, and irside the auxiliary buildir4).
Then equipment inportant to safety was listed, Hazardous envirunments to which the equipment is subjected is based on the acci-dent categcry and the plant area (environmental boundry). The hazardous environment is based on knowledge of thermohydraulic code predictions, previous studies, cperational experience, and expert knowledge.
The list of EQ-Risk significant equipment that resulted from the forcalized process included sets of PWR and BWR operatio,1al analysis
Minutes / Reliability Assurance 10 Mtg. June 14, 1986 issues.
(Operational analyses include consideration of applicable penetrations, splices, and cabling.)
The BWR issues included:
o Lcw pressure sensors o
Reference leg detector piping (pressures, levels),
o Safety relief valves, c
Inboard MS!Vs & bypass valves, o
Containtrent isolat:en valves, L
o High pressure injection purps, l
o Lcw pressure irjectica pur.;p notors, o
Low pressure injection MOVs, o
High range radiation fr.cnitors.
1 The FhR issues includi d:
o Containment fans, l
Cont't spray purps/notors (inside containment),
o e
Pressurine level / pressure instrumentation o
p0RVs & Foter-Operated Block valves, o
Cont't pressure instrumentation, o
Stean gen, level detectors, l
o Primary system RTDs, i
l o
Cerdainment isolation valves, o
H,vdregen detectcrs, I
[
o High range radiation monitors.
i
Minutes / Reliability Assurance 11 Mtg. Jur,e 14, 1988 This equiptrent could be subjected to a harsh environment when it needs to operate, so it becomes risk significant.
Additicnal Examination of Selected Equipn.ent Operations - Larry Bustard, SNL Having established a set of equipitent operaticra, the next step has to look at some of these in more detail to assess their risk significance within irportant FP.A sequences and environments to assess possibic eciu1p: rent cet.lificatico issues, PVR components selected for further analysis ircluded: PORV components, stean generator level detection trar.6mitters, ccritainmer.t fans, and high range radiation tronitors.
BWR components receivitig further aralysis included: SRV and MSIV solenoid operators, MSIV bypass valve POVs, high pressure injection pumps, and high rar.ce radiation monitors.
Examples of analysir en the above conponents were presented. The effects EQ issues could han on the frequency of certain fSA sequences was o plained.
Given the sensitivity of components such as terninal biccks, electrical penetrations, solenoid operated valves to harsh environrents the frequency of certain sequences could increase from that assurred in FRAS.
EQ. Risk Scopino Study: General Conclusions /Pecommendations - Larry Bustard, SNL i
i 7
Minutes / Reliability Assurance 12 Mtg. June 14, 1988
?
There were eight general conclusions that resulted from the EQ-Risk i
s Scoping Study.
j i
l 1.
'FRA accident timing perspectives provide a basis for modifying EQ l
practices.
In a pRA, core damage frequency and risk are substan-
- )
tially dependent on equipment operability during the first few hours to days of an accident sequence. The implications for EQ t
testing are thct less emphasis needs to be placed on demonstrating i
I
)
long curation accident eeuipment operability.
[
i I
r 2.
The importance of the accident radintion dose is overemphasized, 1
FRAs co not consider substantial in-conten. ment radiation con-ditiers to exist until core c.elt has occurred which is halt' an hour
[
to several hours af ter accident sequence initiation.
PRAs calcu-L late that risk significant containnsent failures will occur within a i
f i
few hcurs to a few days. EQ requires all equipeent to survive an l
t l
i instcntanecus release to th? containment of part of the core
{
inventory of radioactivity and radiation specifications are
)
l typically based on long term accident exposures.
I 3.
When PRA perspectives are combined with the existing EQ research j
perspectives, several historical EQ issues can be ceasidered not f
i i
(
risk significant.
Examples include: Straultaneous versus sequen-j f
tial application of accident radiation, steam, chemical spray, and pressure conditions; accident dose rate effects; and oxygen presence within the test chamber during LOCA simulations.
Minutes / Reliability Assurance 13 Mtg. June 14, 1988 4
When PkA perspectives are corbined with existing EQ perspectives, several historical EQ issues have possible risk significance.
Examples include: Adequate sealing / protection of safety-related circuits from moisture intrusion / condensation effects; choice of an,bient cnvironments as a basis for equipment qualification; and humidity aging effects.
5.
The lack of information regarding the reliability of certain corrrcr.ents in accident environments creates non-conservative biases in irrcrtant eccident sequences and base case core danage fre-quencies. The EQ process rarely surplies a.:cident equiptrent reliability information.
PRA analysis techniques possibly underes-tirate the importance of harsh environment equipment operations.
6.
There are risk irportant systein operations that rely on non-safety-related equipment or on equipn.ent thct must function during non-DBA environments.
Exarples include:
PORY solenoids and block valve motor operators during transient induced and small break LOCA enviro ments; and FF".S pumps when subjected to high temperature working fluids, such as a hot suppression pool.
7.
The risk irtortance of plant status equipment is difficult to quantify.
Plant status instrumentation is not well modelled by PRA fault trees.
FRAs assure plant status information is reliable and base actuation / failure response on human error.
i 1
l,-
c' Minutes / Reliability Assurance 14 Mtg. June 14, 1988 8.
The isk importatice of certain EQ issues could not be characterized by the Scoping Study using existing PRA rrodels and EQ data bases.
3 i
4 t
Theri are five gereral reconcendations that resulted from the EQ-Risk 7
Scoping Study.
1.
Develop harsh environment reliability information for selected equipment.
An example would be 50V and MOV reliability during a l
I FVR small break LOCA.
2.
There is a need for additioral definition of equipment with risk l
1 i
importance, such as plant status instrumentation, i
i 3.
A redefinition of the EQ regulatory basis should be considered for r
l exan.ple equipment opere.bility time requirements.
f i
4 There is a r4eo for improved guidance regarding operations, mainte-nance, and inspection activities.
l i
5.
There is a need for iriproved guidance regarding equipment aging
)
issues. An example would be humidity aging effects on electrical
[
1 penetratier.s with polyimide insulation.
l Chairman Wylie informed the Staff that one and a half hours had been l
I reserved to brief the full Comittee en the EQ-Risk Scoping Study during i
j; the July ACRS meeting (July 15, 1988: 8:30 a.m. - 10:00 a.m.).
He toirl l
the Staff thc Comittee would attempt to draft contents on this study.
L l
t Minutes / Reliability Assurance 15 i
Mtg. June 14, 1988
[
USl A-46. Seismic Qualificatien of Equipment in Operating Plants - T. Y.
Chang, NRC Staff Dr. Chang gave an update on the overall implementation of USI A-46.
He also surrarizee experience ceined from the Nine Mile Point Unit 1 (imF-1) trial plent walkdown.
l I
The first revision of the generic implementation procedures (GIP) has been prcduced by the Seismic Qualification Utility Group (SQUG).
f I
Plant-specific inplementation schedules are expected in early fall.
Staff review and SQUG updates to the generic implementation procedures will result in a final Generic Implementation Procedure to be ready by
[
i the end of February.
Staff appresal of the procedure and issuance of a safety evaluation report is scheduled fer April 1989.
Plant-specific implementation of USI A 46 reviews will start in 1989 and l
continues through 1994 The bulk of the reviews (49 of 59) will be l
conducted in 1990 and 1991.
i lhe NMP-1 pilot plant walkdown had three objectives.
It was a test of
{
the generic implementation procedures on a BWR.
It was a test to determine what improverents shculd be made to the training course and the generic implementation procedures themselves. The walkdown will also te used to verify the seismic adequacy of the NMP-1 equipment reviewed.
f i
i t
- - -. I
Minutes / Reliability Assurance 16 Mtg. June 14, 1988
, The total effort expended in the walkdown was about 2 man-years. Three i
seismic review tearts (SRT) with four engineers on each team were used.
There was a lead time of six months to begin preparations for the walkdown.
In general, the $RTs received adequate training and did an adequate and qualified job of evaluating ecuipment. Judgment used by the SRTs was t'asically cor.servative (i.e. equiprent which was later shown to be seismically acceptable was given a detailed evaluation, rather than being a proved during the walkdown).
Few seismic concerns and no serious seismic issues were observed during the walkoown, limp-1 is in good seisnic condition, fiRC observations and conclusions included the following points.
Utility ergineers can perform A 46 walkdowns if they have the proper qualifica-tions and experience. Taking the SQUG training course is not a substi-tute for seisnic engineering experience. The walkdown was smoothly executed.
Lessons learned from the Zion walkdown and pre-planning contributed to the sr'coth execution. There was considerable open space at tyP-1.
This fact contributed to the lack of seismic interaction hazards.
(The session on USI A-46 was for information only).
The neeting was adjeurned at 5:40 p.m.
m
Minutes / Reliability Assurance 17 o
Mtg. June 14, 1988 h0TE:
A transcript of the meeting is available at the NRC Public Document Room 1717 H Street, NW., Washington, D.C. or can be purchased from Heritage Reporting Corporation, 1220 L Street, NW., Washington, D.C. 20005, Telephone (202) 628-4888.
ATTACHMENT A JUNE 3, 1988 r
i )
L RELIABILITY ASSURANCE SUBC0m!TTEE MEETING JUNE 14, 1988 1717 H STREET, NW, WASHINGTON, DC - ROOM 1046 Equipment Qualification Scoping Study USI A-46 Implementation 8:30 A.M.
1.
Chairman's Introduction - C. Wylie I
8:45 A.M.
2.
Staff's Introduction on EQ/ Risk Scoping Study 9:00 A.M.
3.
Project Objectives /EQ Issues
.9:30 A.M.
4.
PRA.and EQ Perspectives Important to Project Approach 1
10:14 A.M.
BREAK 10:30 A.M.
5.
Project Approach 11:00 A.M.
6.
Identification of Harsh Environment Candidate Equipment Operations 12:00 NOON LUNCH l
1:00 P.M.
7.
Additional Examination of Selected Equipment Operations 2:15 P.M.
BREAK t
2:30 P.M.
8.
Conclusions fron EQ-Risk Scoping Study I
3:30 P.M.
9.
Update on the Implemer,sation of USI A-46 li l
i(
a) overall plans l!
b) summary of experience from NMP-1 trial plant walkdown
'l:
5:00 P.M.
Adjourn L
l L
1 3
l l
t
-. - -.. ~.
LIST OF HANDOUTS 1.
Presentation by Moni Dey, Project Mgr. NRC 2.
EQ-Risk Scoping Study: Project Objectives /EQ Issues - Larry Bustard, Sardia 3.
EQ-Risk Scoping Study - Alan K.
4.
Additional Examination of Selected Equipment Operations Larry Bustard 5.
EQ-Risk Scoping Study: General Conclusions / Recommendations Larry Bustard 6.
USI A-46 Seismic Qualification of Equipment in Operating Plants T. Y. Chang i