ML20195D866

Draft Rept, Evaluation of Air-Operated Valves at Us Light-Water Reactors

ML20195D866
Person / Time
Issue date:	12/31/1999
From:	Ornstein H NRC OFFICE OF NUCLEAR REGULATORY RESEARCH (RES)
To:
Shared Package
ML20195D865	List: ML20195D860 ML20195D866 ML20195D891 ML20195D921
References
NUDOCS 9906100060
Download: ML20195D866 (23)

Text

.

4 1

W r

e

,d.p-EVALUATION OF AIR-OPERATED V.n,,V..,ES E

w AT U.S. LIGHT-WATERjREACTORSF@4

/

w i ;#

1 ppp ~

g, N

g%

J

\\

9 wha.z e gf-1999 V B:#;y.

4 ss es a s : ;;[$Y$)$$$

u't) ;

ffD*tk pt"

% ide,.

I f($

'n'j$#

Prepared by:

cW Di.' Harold L. Ornstein y a.

.us.v;;

~

y:,,.

--.*/ ' 4

- g;;;,

~. d

(

i,[;.YF di %P afs &b

/q?v.

6 r 4; w v c.e,

gA h: +

.s w

gf

?lfC ;>

g,'

fyg? : '

4ip tot v%a.

1%

y;

w. :

y f fQ;%

j$

$2 a 1

v p. ;. y,.v g a;7t+'

Y.Y, 'f,hid,d[

.y Regulatpry Effectiveness Assessment and Human Factors Branch Division of Systems Analysis and Regulatory Effectiveness Office of Nuclear Regulatory Research 9906100060 990603 PDR ORG NRES PDR

o CONTEllTS

- AB B R EVIATION S........................................................... v EXEC UTIVE SU M M ARY..................................................... vi 1

IN TRODU CTIO N..............................................iW.......1 r

n

.h 2

' USE AND APPLICATION OF AIR-OPERATED VALVES...M.,Y,e0'f...........1 m

w >n w v

v a OPERATING EXPERIENCE............., gp4 3

g ph 3

d4 9

......f3

?

SITE VISITS...........................[bg}bfrj...

i

.7 4

INDUSTRY INITIATIVES................ 1@hv 5

. Q

..................... 10

,64 %

AIR-OPERATED VALVE FAILURES AND RISKr..j'.4 (

Overview........................w......y$p@y, y%....

6

..... 12 6.1

?.

p.

12 Accident Sequence Precursors.dh.l.y,........h;$.......

6.2 Risk Achievement Worth........g@h...... P.

13

+

. 6.3

.. Pl............... 13 ng;~;m.,

H FIN DING S.................., t.,.......i.. M.. ;w 7

Site Visits - AOV Progranis - Analysis and p,x.................... 13 Operating Experience d......,5........piti6g' of AOVs............. 13 7.1 -

7.2

)........................14 j

7.3 Risk Achievement Worths and Common-Cause Failures................ 15 7.4 -

Risk - Accident Seqdence Precursors,f........................... 15

.l 7.5 Dampers (" quarter,-turn AOVs")g..... '............................... 15 W3A pp%a n%

s n..

CONgUSIONSQ.MQ.y..... ?gJ.. f, 8

..................................15 83 A,,WP,.a.r.allelisms With~M.. otor-Operated Valves........................... 16 c -

w a

8.2 w A,pVggogramsq.gqq........................................ 16 Sup%pg[ ting Pneumaticpyslems,..................

8.3 fa Jystem kir-Operated Valves............................ 18 8.4 4@Other'P6 der-Operated Valves........................

8.59' sp

% Web gREFERENCES.W.,:93Wkp.................................................19 a

9 e

nc m

p g -

3

/

TABLES TQ -

Mdf Air-operated Valves in Plants Visited............................. 1 gift?

1' 2 mms ited..........................................................

3 Status of Air-Operated Valve Programs at Time of Site Visits

......................9 4 Elements of an Effective Alr-Operated Valve Program........................... 17 ill b

.j

[

i

?-

J:

ABBREVIATIONS I

-l i

~ADV' atmospheric dump valve eAEOD Analysis and Evaluation of Operational Data (Office for)

]

AOV;

air-operated valve AOV-JOG-Air-Operated Valve Joint Owners Group ASME

American Society of Mechanical Engineers N

iASP-

-Accident Sequence Precursor j,3

,,p 38:yr i

AUG Air-Operated Valve Users Group i

s QstAj NO BWR:

gb

./ ~ d"f

!L

,I boiling-water reactor 6/

f 4#

p g

.. u

.v CCDP conditional core damage probability -

?[#

Mt?{- 'W CCF common-cause failure CDF core damage frequency A

w EPRI

' Electric Power Research Institute

..y y

1 f' g%@.89:s N f.l$$P GL Generic Letter s

b,ki W

y.

~

IN

.information Notice l9,ggp

,~

g xjyW HOV hydraulic-operated valve [.

p#

NQF an v

v INEEL Idaho National Engirfeering and NnvironmeNtal Laboratory

-l lNPO

- Institute of Nuclear Power Opsrations -

IPE individual plant evaluation. ( (, g IPEEE_. Ng5 individual plant evaluation of extemal events sg8S.

4jh @

YW

"*9#116ensee evbn@ tis $oit r LER Alfidijlgerrea%g}fe:

LWR.

%#NA

^OW I

MOV.

gpatgrppwated valvey' MUG } pf Motor Ive Users Group aes.y <

NEl p Nuclear Energy-nstitute

,7 y -

.m s

PORV

_ power-ogerated relief valve

. IRVk pressure regulator valve :

ySAN.

probabilistic safety assessment a

gs

,,yr

?

S",

lenoid-operated valve L

.TEP

' top event prevention v

. ? !

\\

EXECUTIVE

SUMMARY

This study was initiated by NRC's Office for Analysis and Evaluation of Operational Data' to help the Nuclear Regulatory Commission determine if additional attention needed to be focused on air-operated valves (AOVs). This report and its companion document, Idaho National Engineering and Environmental Laboratory draft report (INEEL)/ EXT-98-00383, "A Study of Air-Operated Valves in Nuclear Power Plants," present the results of a comprehensive review of AOV operating experience, and visits to 7 U.S. light water reactor sites at which there are 11 operating reactors.

.><g Plant visits were conducted to obtain information about AOV operating, experience and AOV activities. Discussi6ns on operating experience focused on ths root causeiof'AOV failures and' on the plants' corrective actions. The licensees provided detailed infoini$ii6r) st360t the planti' AOV programs. Features of the AOV programs that were discussed included] debt'ification"of risk-important AOVs, desigr* margins, design verification [diagnositic testing, mili{tsiihnde' practices, ageing, participation in industry AOV activitiesfparallelisms between AOV'6nd MOV experience and activities.

$F%

myw%g, ?

K The study includes information on over 100 events whicb include; common-cause failures or

)

degradations of AOVs in important systems such as emergency core coo. ling systems, residual heat removal systems, auxiliary feedwater systems, emergency ac power systems, and boiling-i water reactor scram systems.

O n;,

Each plant visited had an AOV program. Those AOV projar'nsisssd risk-informed methodologies from the plant probabilistic risk assessments land the maintenance rule to categorize the plant AOV populations.' Many, but not all, of the licensees' AOV programs were l

or were planning to perform analysis and diagnostic testing to confirm that important AOVs had the capability of performing in accordance with their design. Similar to what was found with motor-operated valves, use _of n5wly developed diagnostic equipment has helped utilities discover deficiencies and weaknesses in the design, analysis, maintenance, and testing of AOVs. T6oseXOy prodrarnsVary significa'ntly from plant-to-plant. The plant AOV programs i

are voluntary'and the,re are'n'o. explicit regulatory requirements governing them.

l ynL

- Q. o.

The study concl6ds(thagmplementation ~of an effective AOV program can minimize the likelihood fop"commonicause AOV failures.

g

%Q(

f My y4

,p.

d. -Q p'

h

$M '/3 4rsg-Effective March 28,1999, The Office for Analysis and Evaluation of Operational Data was disbanded. The work described in this report which was initiated by AEOD is being completed by the Regulatory Effectiveness Assessment and Human Factors Branch of the NRC's Office of Nuclear Regulatory Research.

vi L

p I

1 INTRODUCTION To assess the status of air-operated valves (AOVs) at U.S. light-water reactors (LWRs), the Office for Analysis and Evaluation of Operational Data (AEOD) and the Idaho National Engineering and Environmental Laboratory (INEEL) engineers visited 7 reactor sites which house 11 operating U.S. LWRs representing about 10 percent of the currently operating U.S. LWRs. The site visits provided an important sampling of the AOV activities and programs at U.S. LWR plants. In addition, AEOD staff had discussions with engineers at many other U.S. LWR facilities and with members of nuclear industry groups such as_ the Air Operated Valve Users Group (AUG), Motor-Operated Valve Users Group (MUGX air-Operated Valve Joint Owners Group (AOV-JOG), Institute of Nuclear Power. Operations.,(INPO), Nuclear Energy Institute (NEI), American Society of Mechanical Eri' irieers (ASME) Operating and t

g Maintenance Working Groups on AOVs and hydraulic-operated valves'(HOVs)'[ASME O&M19],

and motor-operated valves (MOVs)[ASME O&M 8). &g p

w.4pgg4

%nm q v

To better understand plant AOV activities, AEOD visited $s$hhhlant sites to disc!$s$hbh licensee's AOV programs. The information gathered fro'th tiiiselvisits is an important part of 5

this study. The focus of this study is on important AOVsfVScQ8uid affect plant safety systems and as such are within the purview of NRC's rbgulatioks@N

%*Mm vi d's f} vy 2

USE AND APPLICATION OF AIR-OPERATED dLVES AOVs are used in all U.S. LWRs. They ars used in.5 wi dt 'of applications. Some AOVs perform important functions in safety and nonsafety-related[5ystems which could affect initiating event frequencies, accident mitigation, and radiological retcases.

An AOV is a complex system comprised of three majorfomponents: the actuator, the valve (body), and the controller. Each of the major comporients includes numerous " piece-parts" such as diaphragms, springs)lir$lt switches, solejiold operators, positioners, current / pressure (i/p) conyEte$ voltage /pressurs (e/p) converte'rs, accumulators, o-rings, lubricants, filters, regulators?pohss) bonnets, and seals. Electricity is required for control and air systems are required to provide,modve power. 4 3

,., m Table 1 containFa listing of,the AOV populations at the 7 sites (11 plants) visited during this i

study. The'licenseesYistied stated that their plants had between 400 and 2800 AOVs. Each of the planfivisited has c$tei(orised between 50 and 500 AOVs as " safet signififance,""importantho $fity," or a combination thereof. The remaining AOVs (the majopty of AOVs at eachl plant) were determined to have little or no safety-significance.

Sl@im'e AOV applicatior$ appear to be common to many plants. For example, a n

i dse"AOVs for contain~ ment isolation functions and for main steam systems. U.S. boiling-water i

r'eh6tMsi BWRs '6se AOVs in their scram systems. U.S. pressurized-water reactors use AOVs foV6onirb(lNdd su)xiliary and main feedwater and for condensate systems. T at U.S."LWRs are nonsafety-related and are generally associated with the non-nuclear balance l

of plant. Nonetheless many plants visited identified a number of "important" or " risk important" AOVs which had been classified as nonsafety-related.

l 1

i

Table 1 Populations of Air-operated Valves in Plants Visited Plant Safety-Related Category 1 Category 2 Category 3 GL 8910*

Name AOVs AOVs AOVs AOVs MOVs Palo Verde 41 + 131 = 172 AOVs 41 AOVs perplant 131 AOVs per plant are Approximately 2628 There are 831 1-2-3 per plant are classified are classified by the classified by the licensee AOVs per plant are MOVs on site by the licensee as licensee as as Category 2. The classified by the,.

which 336 are in (3 plants) of safety-related. See Category 1. The licensee refers to licensee as Category 3.

Category 1 and 2.

licensee refers to nonactive safety-related The licensee refers to the GL 89-10 active safety-related AOVs as Category 2 nonsafety related AOVs program.

1 AOVs as Category 1.

Q as CategoryJA,;

Fermi 2 22 AOVs in Category 1 403 AOVs are ft2 AOVs are clahs'IbedCategory]3h0Nare 147 MOVhare in and 34 AOVs in classified by the by the licenses ps those "having little or no the GL 89-10 Category 2 (56 total) licensee as Category 2 including

• safety-significance %,.progfah

^

are safety-related Category 1. The 24 safety-related AOVs. A economic accordbg to the licensee refers to The licensee desig, nates '

consequences."]QQ y

%d program plan draft.

AOVs having *high as Category 2,those less.

(Note: The original safety-significance" safety-significanf AOys p [,1995 rough outline for g

(There are also 2482 as Category 1.

that support safetyQh development of the solenoid-operated included ere 370 related functions or hayan Ferrni2 AOV program valves (SOVs) of which SCRAM inlet and relatively high economics ills!Ed t'otalof 2058 1442 are classified by outlet valves, consequences if they

$0Vhf which 598 the licensee as QA1.)

22 safety-related should fall;p$,

7 weyconsidered safety-valves, and 11 risk-

, 1. q ;

related vatves or significant valves 7

p:,

danipers, and 1460 M(p,

were considered g

nonsafety-related

' valves or dampers )

e ggg MJ Palisades The total number of 111 AOVs.Valvesin 40'3 AOVs are classified Approximately 586 There are 54

~

safety-related AOVs this category are by the licensee as AOVs which are not MOVs in the was not provided. See safety-related with

, Category 2cThese AOVs Category 1 or 2 are plant of which 30 Categcries 1 and 2.

active safety

'are safetyfrelated but of classified by the are covered by functions,importante low rislesignificance or licensee as Category 3 GL 89-10.

,m Akhg 9,p p$( M ksafety based on

'nonsafety-related but AOVs.

h

!! heir probabilistic used in " critical" sifetyhsipsament applications

1. y (PSA)lrtsi(Q significance, orj :,

ct dh ;ij, jncluded based on 4.h+w d 4,[

. Expert PanelF g

fiq determinations.

e. s.

LaSalle 12 84 for both units.

(AOys having high AOVs h6ving low safety AOVs having high There are 200 j.

3

' safety significance, significance. Number not economic significance.

MOVs in the I$[

4 In addition,370 controhp Number not provided.

Number not provided.

GL 89-10 rod drive hydraulic I provided.

hC valves in each unit ar$

(LaSalle categorizes program for both Mh classified by the ?

AOVs with no or limited units.

safety / economic

[h licensee as safety-significance as

@j yf 1575 nonsafety-related fetated; Category 4.) (There are i

!.s AOVs for both units.)

2 i

1

g

. Table 1 Populations of Air-operated Valves in Plants Visited (Cont.) -

l Plant Safety-Related Category 1 Category 2 Category 3 GL 8910*

.Name

. AOVs '

AOVs AOVs AOVs MOVs TMI1 98 AOVs are classified 98 AOVs are 328 AOVs are 484 AOVs are There are 81 as safety-related categorized as Class categorized as Class 2 categorized as Class 3 MOVs in the (designated "Q-class" i by the licensee.

by the licensee. These by the hcensee. These GL 89-10 or Class 1") by the These are AOVs with are AOVs with an EOP are AOVs not y

program for this licensee.

an active safety function or operational categorized 1 or 2.

plant.

function.

economic significance.

There%re a total of 910 g

AOVs ghTg41-1, The licensee d hd The 89 MOVs Ne classify AOVs,w'g".

Indian 263 AOVs are The licensee did not Point 3 classiried as safety-classify AOVs as s

cl _

within Ilkscope Category 1,2ar 3 related by the licensee.

Category 1,2, or 3.

Category 1

'There are 578let f,of GLf9-10.

1 7p%

(215 AOVs were K'

g a[78 03 = 315 0.

classified by the Q

hcensee as be,ng V M+ k i

re nonsafety-related.

within the scope of ghppygx QD 9 AL the Maintenance Rule,10 CFR 50.65 g

Sh 4pg j

(Ref.1))

g 9, e gg Turkey The licensee classified 174 AOVs (98 active, 53 (34 a' tiveh'$ passive, Yhere'[ri836 AOVs in 111 MOVs (total c

Point 3-4 191 AOVs(totalfor 78 passive, total for '

l total for both, units) areclassifiird tiy1he" licensee knosif thelicensee are within the both units. It is not for both units) both units) as safoty-both units) are related.

classified by the

'as Category 2p?jA

,apecifically designated scope of vn r p4 ;some AOVs as GL 89-10.

licensee as

[4

-t NC M Category 3.

Category 1.

V Generic Letter (GL) 89-10,

• Safety-Related Moto Operated Valve Testing and Surveillance - 10 CFR 50.54(f)," June 28,1989 (Ref. 2).

C p

The category deshtions in the tablEvary from pla'nt-N'M/Arit. The use of the cat NOTE:

a e i

s explained withy $g._

~Q$$m. m There may be SObm ;h5Ne plants thalMelclA$sified as part of the AOV. Figures for S w

u l.

separate data was prst 4 7 1

.e9_ p y? Aw%%%

h a

\\

l 3

, OPERATING EXPERIENCE 4

dW

^

INEEL draft report, "A Study of Air-Operated Valves at U.S. Nuclear Power Plants,"

lijgL/ EXT-98-00383,pril 26,1999 (Ref. 3), contains summaries of 109 AOV events that feported in LERsFand 24 additional events that were not reported in LERs. Selected i

' ents are liste"d below. Plant name, licensee event report (LER) number, and a short l

' of eachi% vent is presented. The conditional core damage probability (CCDP) l c'aled '

ENRC's Accident Sequence Precursor (ASP) Program is also shown when avallatilE "

\\

3 L

Palisades (LER 255/78-003 and 255/81-030)

On two occasions, water in the instrument air system prevented an AOV in the residual heat removal heat exchanger outlet line from opening (going to its " fall-safe" position). The two failures of the same AOV resulted in rapid core heat-up from cold shutdown to near boiling (215

'F and 197 'F in 45 minutes and about 1-3/4 hours respectively). Although these events occurred in 1978 and 1981, they are included because the poor quality air conditions that led to these events existed at the plant at the time of our site visit.

i Turkey Point 3 (LERs 250/85-019 and 250/85-021) a ASP CCDP = 9.0E-04 6'

yeyx While in hot standby recovering from a prior reactor trip, AOVs in the auxiliaiy feedwater and /

main feedwater systems failed on demand complicating lthe recovery 9bluitiplsidildres of AOVs and other pneumatic equipment were attributed to moisture and c6rrosion pr5duhs(lithe 7 instrument air system.

@gg '

Q[

+y Palo Verde 3 (LER 530/89-001)

M' y*

ASP CCDP = 4.9E-05 Palo Verde 1,2,3 (LER 528/89-005)

U,. /

'a1 i

Tgg

(

Failure of all four atmospheric dump valves (ADVs)[AOVs] to function'6h demand at Palo Verde Unit 3 subsequent to reactor trip contributed to a complicated recovery. The manufacturer found design deficiencies in allthree units'm12 ADVs - excessive internal leakage which could result in inability to operate the ADVs frohith'ejo'ntrot rooms or the remote shutdown panels at all three Palo Verde u' nits. The' licensee also found that incorrect valve set up and poor quality air contributed to the unit three ADV failures.

4 Hope Creek (LERs 354/93-006 and 354/94-017)

Repetitive AOV failures, including two sets of concijr' rent failures of AOVs in the Safety Auxiliary taking inf6acco0nt the effe' t of th;ted design cha' ge (modification of valve pack Cooling Systein. A licensee'initia n

c e new lower friction) compromised room cooling for d giant emergency diesggenerators'{8)Oys) and emergency core cooling systems (24 AOVs).

Mdh Oyster CreekJLER 219/85-012)

ASP CCDP = 2.3E-04 56 -

prymn sa y Two AOWs in series in a scrarn., discharge volume drain line failed to close on demand I

subseduent to a reactor scraiN resulting in an uncontrolled leak of hot pressurized reactor I

coolant outside primary containment. The failures were caused by inadequate AOV set-up and design (improper stroke adjustment and an improperly sized spring).

l%

gJ@$9eHW rmor t Yankee (LER 271/98-025, EN 35150) l

{

CoM6n%B6s0 failures (CCFs) of scram discharge drain lines (four AOVs). The AOVs had inad$dht6' design margin. Surveillance testing revealed multiple AOV malfunctions shortly after their installation.

J 1

4

E 1

)

l Millstone 2 (LER 336/97-011 EN 32070)

Nineteen of 23 AOVs serving in containment isolation functions failed to isolate under full system pressure. The failures were attributed to improper set up. Full pressure testing had never been done. Failure of the three AOVs which isolate letdown would result in offsite radiation doses higher than stated in the plant's final safety analysis report. Two of three AOVs in the letdown line had malfunctioned 4-years earlier but the problem was not corrected - see LER 336/93-023 below.

Millstone 2 (LER 336/93-023) qb/

While at ful1 reactor coolant system pressure with valve position indica %dRlstorts N

closed, two AOVs in the letdown line were leaking betwee'n 20 and 40jprii@8 licensee attributed the leakage to improper bench setting of the AOVs. The' licensee disid failure to test the AOVs at full reactor coolant system pfessurejas a contributin[EAusEThe licensee acknowledged the need to verify isolation of tiidse(valpes against full reacidr~ coolant i

system pressure however verification was not done untilpyears later (see LER 336/97-011 y

Ngg(g above).

LaSalle (LER 374/96-011), NRC Information Noticp (ItQ 96-68 ( f.'4) [

J.A LaSalle station's review of AOV diagnostic test data'and load calculations revealed errors in the q

AOV manufacturer's published data on effective diaphragm areasrUse of the manufacturer's erroneous effective diaphragm areas could result in incorrect' set-up values and consequent failure during design-basis events.

(

San Onofre 2 and 3 (LER 361/96-Ohi) o While pursuing an AOV testing program sirniffr t'o their MOV program, the licensee found i

several coritainment isolation Valves [AOVs] which~would not have been capable to closing under ddsign-ba$1s conditi' hs3 The licensee attributed the deficiencies to errors in the o

manufacturer'h;sh5 lysis andsetup;e'rrors that emanated from using the manufacturer's outdated and incorrectpetup instructions,

,w M: L.

&Y Haddam f9eck"(LER!213/9$005), NRC IN 95-34 (Ref. 5)

ASP CCDP = 1.4E-04 2

w&

a ens Bothp'f the pressurizer power-operated relief valves (PORVs)[AOVs] failed to open on demand during s test while the plant was in cold shutdown. The failures were attributed to air leaks eqdsed by improper AOV diaphragm installation by the licensee. Improper use of lubricant on I

the' diaphragms caused them to extrude enabling the air leakage.

Oh m

ilhdd$rn Neck (lek 93-007)

ASP CCDP = 6.5E-05 w ;n w +

th rgency air accumulator, used for PORV operation during feed and bleed At cooling, revealed excessive leakage through a PORV's diaphragm and pressure regulator. The diaphragm had been installed impnoperly. The pressure regulator leaked because of corrosion products which resulted from moisture in the air system which was caused by a malfunctioning air dryer SOV.

5

I Salem 1 (LER 272/91-030-01)

Recurring PORV [AOVJ failures at Units 1 and 2 and the failure of both Unit 1 PORVs to open on demand during testing. The apparent cause was air leakage around the diaphragms.

Contributing factors for many of the failures were: failure to use manufacturer's installation procedures, valve temperatures in excess of their design temperatures, and pipeline induced vibration. The LER noted similar experience with AOVs which are used as pressurizer spray valves at Salem units 1 and 2.

Fermi 2 (DER 97-1202-plant internal deficiency report) w[o:A.

.dib p.

Eighteen SOVs controlling safety related AOVs failed duringta'n 18-month cause was excessive use of thread locking compound ori' threaded joliit's* period, The

!c from the thread locking compound deposited on surface's),of the SOVs' causing [thyto,stic%I NRC IN 88-24 (Ref. 6) hh, b

Kewaunee (LER 305/87-12)

Vgg Calvert Cliffs (10 CFR 50.72 Reports 12013 and 12015 Akild4p198,8) f

'(pfgy NRC IN 88-24 notified all U.S. LWR licensees of conditions at Kewaupee1and Calvert Cliffs where common-cause AOV failures did/could result frorn,overpressurizing SOVs (which are piece-parts of AOVs). The IN indicated that failures of nonsafety-related pressure regulators could result in failure of safety-related AOVs.' Subsey06nt t(the,issijance of the IN, several licensees found similar situations at theirfplants. However, in recent years several licensees found similar vulnerabilities that their original review of IN 88-24 did not find, for example:

Clinton (LER 461/90-004)

)

j Indian Point 3 (LER 286/93-050)5 4i Millstone 3JLER 423/96-031)3, k

" ' f "i "

Cooper (Ins $ction Report"50!29'8/97-201 M'O D.C. Cook (LERi315/97-026-01;and 315/98-052/01)

Waterford (leg 382/,98-010) g44 w%m 8_elevant Non U.SIEvents..

, p[,

y m; g ly v

Darlingtos Unit 2 (Event,Ndtifibation Report D-1998-01497 and Detailed Event Report D-1998'-01497):

I

[

Op[hEptember 20,1998,6vhile restoring the instrument air system during an outage,18 of e

ig" pressure regulator valves (PRVs)" failed, exposing downstream AOVs to full-system pressure. The PRV failures were attributed to embrittled diaphragms coupled with the large loadthat was placed on the PRV diaphragms when the air system pressure was being restored.

Ttjeli6ensee noted that the occurrence of such an event could cause safety-related AOVs to be forhedid53$sition opposite from their " loss of air position," and that they could be damaged and remain in that "non-safe position." An analagous situation could occur at a U.S. plant during a recovery from a loss of offsite power or a loss of instrument air.

6

r, t '

Pickering Unit 2 (Canada] (SER A-94 plant root cause failure report):

On December 10,1994, a " thermally aged" diaphragm in a pressure relief valve [AOV] in the primary heat transport system cracked initiating a small break loss-of-coolant accident. As a result of that event, Canadian plants have implemented programs to replace existing diaphragms and implement programs for ensuring appropriate diaphragm replacement frequencies in safety-related systems.

,m.

rganiz'aiion for Bruce "A" [ Canada) (Technical Paper by Ferguson and Fitzgerald at thep Economic Cooperation and Development / Internal Atomic Energy Ager)cy, Meeting on Motor Operated Valve issues, April 1994):

)

In 1991 diagnostic testing was incorporated into an AOfpreventive ma n ra initial diagnostic testing (322 tests) found that: 65 perce' t of theJ/55 and poilSo d

n calibration,15 percent of the bench sets needed adjus(johkardonly 12 perce'n,

AOVs required no maintenance or adjustment. In subsequent diagnostic testing (88 tes,45 percent of the 1/ps and positioners required calibration,5 percent [

pc.h sets needed adjustment,

~

and 22 percent of the AOVs required no maintenancegria ent.h

' $7pOb sg ekpgy f hg Y

4 SITE VISITS l'

5%k

]

Seven site visits were conducted between October.1997, arid March 1998. Each visit lasted 2 days. Table 2 lists the plant name, the' dates of t$e visit,V hifesctor manufacturer, the

~

t architect engineer, and the year the_plarit began'commerciaT6peration.

The site visit team included one e$gineer froriiAEOD, td engineers from 1NEEL, and one engineer from either NRR orSES: During m6st of theVisits, the NRC resident inspectors attended thpentrance and/or, exit interviews %atw'ere held on site.

ffQ Q%Q

%W include 5 pl$nb$lk-throughs, discussions with plant management, plant The visits usually$l$nd engirie5ris;i snt operators and plant maintenance pers licensing peis8hi pi with AOV activitisikDiscussions were held regarding plant AOV operating experience and plant programs assocliteiifwithh0Vs. In addition comprehensive discussions were held with personnel asEociate'd Wt$lant PSAs (individual plant evaluations [lPE's] and individual plant evaluatpii of external events [IPEEEs)) and " maintenance rule" (10 CFR 50.65 [Ref.1])

i activities. Detailed trip rgorts from the site visits appear in Appendix C of Reference 3.

T. lants visited wereShosen in a manner to get a representative cross-section of the d RWR populationJn accordance with the following criteria:

I

% ' '4 q

andkEOD project schedule availability k an($$'rticipation in the Electric Power Research Institute (EP 1

hi R

3.

• Idiit participation in AOV users group activities 4.

plant type and age.

7

Participation by the licensees was voluntary and participants were assured that the visits were

" independent fact finding / lessons learned activities," not inspection or regulatory compliance activities.

Table 2 Plants Visited Plant Dates Plant Description /

Year Commercial Name of Visits Architect Engineer

/

L Operation Began s.

System 80 (no PORVs) IlWR/Bechtel g%.

@ $, 986 jr Palo Verde 10/28-29/97 Combustion Engineering, twagoop, 1-2-3 Fermi 2 11/03-04/97 General Electric BWR 4/ detroit Edi.s'onNNd88ef9 Palisades 11/18-19/97 Combustion EngineerinhEtklobp T1971' PWR/Bechtel Gg i.

g 4 LaSalle1-2 12/17-18/97 General Electric BWR 5/5arge,v w m

ntypg@,

1984 Lundy Q

s TMl1 02/12-13/98 Babcock and Wilco $ 16wered loop 1974 PWR/ Gilbert Ass 6ciatesdT.i

~

Indian 03/10-11/98 Westinghouse, four loop Phi 3/Ohited 1976 Point 3 Engineers and Constructorsy Turkey 03/24-25/98 Westinghouse, t$ree loop P'WR/Bechtel 1972 i

Point 3-4

[

.:/

{

s...

,o AOV Pro 72

$t Sites hi i k, -

v...

W h.a

~,g Allof the plantsWisited;h,ad AOVp,3 3rogra'ms in place. All of the programs were aimed at improving AOVpiforpance. However;there were many differences in the status and the depth of the, programs at,each station (see Table 3). Reference 3 provides details of the programsit the statiodsVdit$d; <

.e c

,e TheIbV programs at allhiibbstations visited had been or were in the process of surveying, 1

catedorizing, and rankingtheir AOV populations. Table 1 contains a summary of the cdtsgorizations and rariking efforts at each of the seven stations visited. The methodologies tised to categorize an'd rank the AOVs at the plants visited included: review of plant operating SNd{iisfiNnce,, consideiration of the results of plant PSAs, the use of expe 1

Frequ$rilly'these activities were part of licensee implementation of the maintenance rule. Many licensees' evaluations utilized IPE and IPEEE methodologies and results. Many licensees' categorizations considered risk achievement worths, Fussell-Vesley or other risk importance measures.

8 m_

Table 3 Status of Alr-Operated Valve Programs at Time of Site Visits Categorization Diagnostic Testing

• Plant Status Being Done Findings Palo Verde Complete Static and Dynamic Low margins-2 replaced / modified AOVs

~

Fermi Nearing Completion To be determined Calculations pl5nned f

a,.w g Palisades Complete Static. Dynamic beingyr, Low margins 9,dAOVs/

2.

replace 8/quidiffp planned.

of k(C,gp<gb margindkhh75[

LaSalle Complete Static replaced /modifiedfQVs!Found gg y, generic effective diaphragm c : area problem described in

&y

..] tlN 96-68.

TMI1 Complete Static planned..

1$w thargins-modified AOVs indian Point 3 Complete Static

!)Nk Low margins-

"W$D'h replaced / modified AOVs x

$(^ hocusing on maintenance /

Turkey Point Complete Static E

~ operations. Limited testing of i

problem AOVs.

i

/

Dynamic testing: testing conducted with system pressure and/or flow.

Static testing: testing conducted at ambient conditions without system pressure or flow, g{h:A f~

j h.

.n

k.

' The Palisades and Fermi pla'nts'drilead plants in a program funded by EPRI to develop /

of our visits to tii[osylshts, both plants had categorized and prioritized o A contractogh567 sr$.Mapalyses on the Palisades plant's AOVs. The Palisades p! ant staff p

had performed static testidof their AOVs, and was in the process of planning to perform dynamic" diagnostic tastingg in contrast, the Fermi plant had hired a contractor to conduct analyses of the most important (Category 1) AOVs, but had not established any specific plans fordiagnostic testing of AOVs.

j Q$

j Palo; rde's AOV program was initiated many years ago. Having experienced common-cause ilures as esfrly as 1989, the Palo Verde plant initiated an aggressive program to prevent 9erde staff performed static and dynamic test of AOVs which appeared to have., Apera ing margins. As a result of analyses which indicated less than desired design margins, coupled with the results of static and in some cases, dynamic testing, Palo Verde made modifications to certain AOVs to assure satisfactory operation during design basis events.

9 L

In order to analyze their AOVs, many of the utilities canvassed have had to purchase design 2

information and analyses from the AOV manufacturers since that information was not provided with the valves. It appears that some of the original AOV design information may have been provided to the architect-engineers but the utilities were not privy to many of the deta!!s of the AOVs' design analyses or available margins. Recently, there have been several cases in which utilities evaluated their AOVs and found errors in the AOV manufacturers' design calculations as well as errors in the valve designs (e.g., Crane-Aloyco, Fisher, Anchor-Darling /ACF/WKM/

BS&B [ described in Reference 3]). In addition, many AOV manufacturers have not provided sufficient guidance or instructions for AOV maintenance or changeout.<Similarly, regarding SOVs which are important piece-parts of AOVs, NUREG-1275, Vol. 6dOpersting Experience Feedback Report-Solenold-Operated Valve Problems," U.Sjpuclear Rygyldt'ory Commission,

where SOV manufacturers did not provide utilities with sufficient guid$,lpote Office for Analysis and Evaluation of Operational Data, February 1991 f

nce"fd@%4 pf, add ihtenance change-out of SOVs which control the AOVs.

f.1

/

y f

%V AOV programs at the stations visited either were using o%mf pwere planning to use AOV iagnostic testing equipment. In addition to the plants visited, feedbac(frorkir)dut'ry meetings indicates that plants have had favorable results using AOV diagnostic tesfingequipment to diagnose and fix specific AOV problems. In many cases, as a result 6f using" diagnostic, testing equipment, the utilities have made modifications to AOVs to improve their operatibMSome plants indicated that they use AOV diagnostic testing equipment rdutinyylto confirm'th'5[AOVs have been set up correctly, in a presentation at a recent AUG'md5tirigf aisalve engineer stated that at his plant, AOV diagnostic testing equipment "i.s the tool of ch61ce" to. assure that the AOVs have been set up correctly.

f f

^fd@

(

y l

Some plants have performed AOV diagnostic te' sting under prototypic dynamic conditions.

However, most plants' AOV diagnostic testing has been done statically, and not under prototypic design loading conditions. In some cases,, successful static diagnostic testing may i

not provide the assurance that a'n' AOV willbe abl5 to~ perform its safety function under design loading cotiditi'ons.

o69;,-3

ff 9tr Q Mfg Q % r; a p,g l

INITIATIV5dhh 5

INDUST

,p ;,

y, Operating e3pd,r'lencapa's shown that'many of the problems associated with MOVs such as valve sizing, packing, frictioq or actuator sizing, verification of valve capability, design loading, j

lack ojj$ndor informati6%norQirbtotypic surveillance testing, verification of design and j

operatfng capability alsoixist With AOVs. Industry organizations (INPO, EPRI, AUG) have j

endouraged licensees to;take the initiative to translate the lessons learned from the MOV o "Y5 ting experience and the diagnostic testing associated with MOVs to AOVs. As noted in p 4 of this repojt, licensees at the seven sites visited have initiated AOV programs to j

5 ress [ hose and"other similar problems. Those AOV programs vary in age, resources, and I

e@ie[ddthe maintenance rule, plant IPE and IPEEEs, plant operating exper j

8 Plants canvassed includes the seven stations visited plus others that had representatives at AOV industry meetings.

10

procedures, plant technical specifications, etc., to identify important AOVs, the design capability of which need to be verified. Some licensees have performed analyses and diagnostic tests to verify the capability of certain AOVs. Some utilities use diagnostic testing equipment to improve the set-up and maintenance of their AOVs. However, some licensees are not addressing the AOV issues discussed in this report. From a nuclear industry organization's presentation at a recent AOV users group meeting (Ref. 7), it appears that some plants' AOV programs are understaffed and underfunded.

In 1997, a nuclear industry organization notified all U.S. LWRs of concerns that operating experience was showing that many of the problems associated with MOVs were also present with AOVs, and that AOVs warranted further attention. At about the sam's time; EPRI implemented AOV pilot programs at the Palisades and FerJnl plants, andfedently expanded to important AOVs, the development of AOV calculational tschniques, aridTi(rific the Duane Arnold and Comanche Peak plants. EPRl's prografn support 5 t65' identification of.i e

g j%y MMS ~ g design capabilities.

Way?

4 %w 9

in 1998, similar to what was done with MOVs, U.S. LWR: licensees formed an Al pera ed Valve Joint Owners' Group (AOV-JOG). AOV-JOG's st$Nff! s'slon is *to develop a common i

and cost-effective U.S. nuclear utility AOV progran9hich*definss the minimum elements necessary to enhance safe and reliable AOV performafide an'd kil6w timely address of industry and regulatory AOV issues" (Ref. 6). The AOV-JOG initiatives are volurifary industry activities.

At the present time, details of what AOV-JOG is propos(og for resolutionpf AOV issues at their plants have not been presented to the NRC. 4MySg y

mln 3

in December 1998 at the AUG Number 16 meetingp~a nucleatJndustry organization representative stated that it was planning to increase its activitisd~in the area of AOVs including

" assistance visits" and holding AOV, workshops in 1999. In, addition, the nuclear industry organization indicated that it will make AOV in_ formation, piesumably operating experience, readily available on its WEB home page.

7 The nuclearin.dustry organibtion representative provided a summary of the AOV findings from 14 site visit'if0'ne strengtisjijd~tjas comnion to many plants was their use of diagnostic testing s

equipmedihggg

Jgg, 1

%gg M%

Weaknesses were observed in the following areas:

Q ?q Q

'l; Somil lantdakrjffdone d$ equate reviews of other plants' AOV events.

At some plants, testing;and sizing verifications of AOVs were not implemented.

AfAt some plants AOVs have not been set up properly in some systems.

- p For many AOVs,tthere were inadequate thrust assumptions.

-g Plants changed A'OV packing without considering the effect of the changed friction

_ pg values.

g

-ggPlants have not acted on information learned from industry groups on inaccurate thrust g% talculatiops j

@$!M Are'as o cbricern at some plants are:

AOVs have low station priority.

Some stations have a high tolerance for AOV problems.

Lessons leamed from MOVs are not applied to AOVs.

Lack of budget to support AOV programs - don't do much until something is required.

11

o e

Too Event Prevention Analvsis1 LReview of the technical paper,"Use of Top Event Presentation Analysis to Select a Safety-Significant Subset of Air-Operated. Valves for Testing" (Ref. 9), provided insight to the -

' Monticello plant's AOV program' In a discussion with the Monticello plant staff and their contractor, we leamed about the Monticello plant's AOV program. The Monticello plant's AOV program is currently under development and expected to be finalized later this year. The

Monticello AOV program will be similar to the AOV programs at other plants (noted' earlier in this report). One difference is that Monticello plant's AOV program will us Prevention" (TEP) methodology to identify " safety-significanC AOVs fofg~,

i periodic testing. The technical paper describes the results me of

~gn basis review and that the Monticello plant and its contractors have done to select ignifit.

').ing the TEPp g) gf methodology.

Y,, Q,. &,)f n

.V

The TEP methodology is commonly referred to as a "

' set method

.~

l

_ utilizes PSA techniques (the Monticello plant's IPE in t o determine whi uipment must work in order to prevent the undesired event (top

" ent) from occurring. The Monticello plant's TEP analysis identified 24 "important"J b technical paperindicated that if those 24 AOVs receive design-basis reviews andFperiod

. iige,they can be considered to be " reliable," and they would then be expected to have the reliabilit lan;lfailure rates that i

were used for AOVs in the Monticello IPE. The tech 61dal p, aper rep 6rtedth'at when the

. ' Monticello plant IPE's AOV failure rates were us6d.f6rithose 24 AOVs diid a failure rate of "one" was assigned to all other active AOVs, there Was a sjiilllj " ent) increase in the plant's

' base case core damage frequency (CDF) Pin contryst, the i paper reported that failures l

. of those AOVs would result in significant' increases'in CDF the base case.

U

[

l The technical paper also reported that when using the Fus' sell-Vesley importance and Risk L

Achievement Worth threshold or screening yalues of 0.5" percent and 2.0 respectively (per the l

- recommendations of NUMARC 93-01, "IndustiyGuideline for Monitoring the Effectiveness of Maintena

• t Nuclear ^ flants," Mayg%t the Monticello plant does not have any j

"potenti lj ificarit MJn contrast,ithe technical paper stated that "while no AOVs

. exceed l

r risk

, in combination with one another, they can have a significant e to

• in reliability.' On the other hand, TEP analysis identifies the combinati whic 'ereYpdrtant to safety."

i l

QS.

6' AIR FAILURES AND RISK

. f -OPERATED f l.

6.lfy Overview 4

' mary concemJ AOVs is that risk-significant AOVs satisfactorily perform their functions ed or ar gh? fed. ~ Nuclear plant safety in the U.S. is predicated on the single-failure FTthe failure of any single component should not result in unacceptable herefore, one would expect that single AOV failures would not result in una

_. consequences. However, common-cause'AOV failures can result in conditions more severe than those analyzed in the plant safety analyses.

l n

,2 r

i L>

l

-l 1

L

I e

6.2

. Risk Achievement Worth 8 From the plant visits, it became inherently obvious that there are major differences in use, type, and number of AOVs at each plant. Table i shows the results of each of the plants AOV classifications. The bottom line of which is that each plant has identified a number of AOVs that are "important" and which may also have high " risk significance." The quantity and the distribution of AOVs vary widely from plant to plant. Recognizing that plant IPEs have shown variations in which systems are most risk important, it is not surprising that the location of the 3

risk important AOVs vary from plant to plant. Several plants provided the study te'am with values of AOV risk achievement worth which indicated that failures of certain AOVs could significantly effect the risk from the plants. In some cases ttie plants {ouydhigh AOV risk j

n achievement worth values which indicated that failures of thos,e AOVs could~ledd,to CDFs many times greater than the CDFs which were estimated by thkplants' PSAs generic AOV failure rates.

M 4

s 6.3 Accident Sequence Precursors

\\

hk A review of NRC's ASP program results found that for the} ears /1984,to 1995, there were i

288 events that were classified as precursors (CCDP gFsater deGuallto,10E-06). Twenty-six of those events were AOV related (i.e., AOV malfunctions were invohled as.either initiators or contributors to the everts). Twelve of those AOV filated precursoFe066tshad CCDP of 10E-04 or greater. The highest CCDP was the 198'5]oss of all auxilia@^feedwater at Turkey Point in which water contamination of the instrument air system gesulted in common-cause AOV failures. The CCDP for that event was about 9E-00hihh hSd th'e fourth highest CCDP of the

~

40 events that were found to be precursdrs that year. No A,OV events after January 1,1995, were classified as precursors by NRC's ASP program.

i'

.f i

y

~

n

?

M@gY 7

FINDINGS

.f.

dh M

Y SithVisit's,- AOV Prog ~rar6s,- Analysis'@and Testing of AOVs 7.1

~%Qg W

.,a Visits to 7 U?$'.Ih{a'gnt ' ites provided an overview of the status of AOVs at abo s

the U.S. commercial LWR, population.9 Each plant visited was implementing an AOV program.

However, thge Wsrepidespread variations in the scope, focus, resources, status and future plans for each of the programs. All plants visited were integrating the knowledge gained from their rqairitenance rule activities /and their PSAs [lPE's] to categorize and prioritize AOV activities.

s 'Ty M

A,t/p% cases, AOVs that had been classified as "nonsafety-related" were found l

son slhnificant. Conversely, some AOVs that were classified as " safety-related" were found not to tie Msign,ificantf g% Q W-ww. -

3 Risk achievement worth is the ratio of the plant's core damage frequency calculated when the component of interest has a failure rate of one divided by the plant's base case overall core damage frequency.

13

p There have been instances where licensees' AOV programs found that risk significant safety-related AOVs were not capable of performing their intended safety function during design basis events due to inadequate valve design or inadequate valve set up. Previous AOV testing under nonprototypic conditions was incapable of detecting some of these deficiencies.

l Licensees found that AOV surveillance testing per technical specifications or ASME in-service-j testing requirements did not always provide assurance that AOVs would perform satisfactorily i

during design basis events.

I There was a wide variation in the status and plans for diagnostic testingfo,fAOVs. Some plants have done extensive static and dynamic testing of AOVs andpere lookin 'to" upgrade their diagnostic testing equipment, whereas some other plants have, as yet*

yerformed any AOV."

diagnostic testing.

g e

s g

i di f

Mg n

A significant problem facing the AOV engineers trying to deterrpine the AO i

M W'.v

@W@hy, margins is lack of AOV manufacturer design data.

Validation and verification of the accuracy and efficacy off ONdsgnostic testing equipment does not appear to be available at the present time. W My 4 s qWJ7, 7.2 Operating Experience x$'

i y;f A review of operating experience and visits to$~ M&Q f

seven U$ plant; sites found that many plants have experienced or discovered AOV malfunctions or vulnerabilities which affected plant operational safety or reduced plant operating margins. A riurnber of those malfunctions or vulnerabilities were due to a commori cause which affected several valves.

Licensees are more aware of the Ir'rkportance"of AOVs;tti n ever before through the AUG, Electric Powe,r Research Institute (EPRI), INPOr AOV-JOG, NE! and the AEOD AOV study.

Nonetheless(Tgcent operating,eyents through.1998 Indicate that plants are finding conditions in well as a%areyulnerable'to'CCRfrom design, installation and maintenance deficiencie which A ir system deficiendiss@ Q Qd Y

Almost all plants v] sited,had observed multiple CCF failures or degradations of AOVs and/or the deviation reborts, e@c':)}iut$ jose events ~were reported in plant internal repor SOVs contro30g th Tt t

sny were not reported to the NRC or to INPO's Nuclear Plant r

nonconservative PSA. Exam @ples which are described in Reference 2 in Fermi Unit 2 PSR580 thread locker, Ralis'ades pressure regulator failures; Turkey Point pilot-operated lockup valve failures / sticking; Three Mile Island Unit 1 Crane-Aloyco gate valve analysis errors.

&h 5

Almo'stpil of the pladts visited had experienced AOV failures that were caused by air-system chithmination ofjnoisture intrusion many years ago. Many of the licensees indicated that they andlWi5ii$ ant operation as a result of substantive improvements that we instrument air system and instrument air quality. The instrument air system and instrument air quality improvements were made as a result of NRC's GL 88-14, " Instrument Air Supply System Problems Affecting Safety-Related Equipment," August 8,1988.

14

.__.,yg 7.3 Risk Achievement Worths and Common-Cause Failures From the visits, it was learned that most of the plants calculated risk importance of their AOVs using their own probabilistic risk assessment models. Many of the licensees found AOVs having high risk achievement wonhs and failure of these AOVs would affect important systems.

Failure of the high-risk achievement worth AOVs to perform their functions would translate to corresponding increases in CDF estimates. Another major concern for AOVs is their susceptibility to CCFs, which may not have been analyzed.

These findings highlight the importance of verifying that important AOyp hill,be bapable of operating satisfactorily and for preventing common-cause AOV failures.h fp

f d %;Q Q l

g 7.4 Risk-Accident Sequence Precursors gf ig g g% b d47 Review of the results from NRC's ASP Program for the;% years 1984 to 1995 found tliat AOV

'M m'

malfunctions were involved as either initiators or contribdidrit6'about 10 percent'5ffh's' events that were designated as precursors. AOVs were involved id3bo'ut two-dozen events that had CCDPs of greater than 10E-06, with about half of those eV5Shivlng CCDPs of 10E-04 or e

greater. However, there were no AOV events after January 1,'1995,Lthat had CCDPs greater than 10E-06.

w g~p,

'v ecy%

7.5 Dampers (" quarter-turn AOVs")

, MQ 3

J Licensees at most of the plants visited arid licenseehth$t vierhc5nvassed at industry meetings have indicated that they are not including air-oper5ied dariphilh their AOV programs. This appears to be a carryover from their MOV programs (responses to GL 89-10 and its s

supplements). It is interesting to note that at one dual unit station visited (LaSalle), the containment purge valves which are air-operated darnpers have been evaluated by the plant (IPE results) to have a high risk importance. However, because the containment purge valves t

are categorized as " dampers,; they are not' covered by the licensee's AOV program. Also at the AUG meet!Egsin June 1998'ahd December 1998, spokesmen for AOV-JOG and EPRI both discountid thehe~od to consider alr-operated dampers in plant AOV programs. Further inquiry

~

s into this posill6hievealed that this position is simply an extension of the original position that the NRC had adopteyegarding motor; operated dampers under the GL 89-10 program. The omission of the: motor-operated dampers from the GL 89-10 programs was based solely on their fire p'dr6liiction f5ri6tl6MTheir performance during other design-basis events were not considere.

W *.J '

,a nu:

}

l C

8 iL CONCLUSIONS) b agts visited represented a reasonable cross-section of the U.S. LWR population. The

. lng conclu's'idhs are based on information obtained from those visits, discussions with

^

AT}$nh7AM3I of Sensitivity Analysis for AOV Study Job Code Numb 7prdOhjel'at industry meetings, operating data, INEEL's AOV sensitivity study, e

"Tr r r 15-00R-03-98," December 1998, plant PSAs and ASP analyses.

15 f

1 l

4 8.1 Parallelisms With Motor-Operated Valves Operating experience has shown that common-cause AOV design deficiencies have resulted in situations where multiple AOVs could not perform their intended functions under design-basis conditions as stipulated in plant safety analyses. The operating experience reported in LERs, inspection reports, and presentations made at AOV users group meetings indicates that there is a heightened awareness of the similarity between AOV and MOV issues. Similar to the industry's formation of a joint owners group to help address the MOV issues [MOV-JOG) after the issuance of GL 89-10 and its supplements, an industryjoint owners group fofAOVs has been formed [AOV-JOG). The AOV-JOG has not provided specific plans and' schedules or information regarding its members commitments to participat,lon and 5cyeEtsnce of the AOV-JOG's technical positions. Licensee participation in these programs is volun 5

W H %'P af AOV operability is subject to many of the same factors $5 MOVs such as: filg ~ h}btion [or' valve] factors than those used or assumed in manufactu'brs' design calculations,6 ' #

r underestimation of loads during design basis operationIextriphilation of valve op5ra51it' based y

on nonprototypic testing such as in-service testing, and ;kobignis with valve setup and maintenance practices.

[9yhi ms%

The analytical and testing methods for verifying AOV operating c5p'abilitibs are not yet proven and agreed upon. Continued discussions among NRC,, EPRI, nuclear plafit owners, AOV manufacturers, and diagnostic testing equipment manufacturers is needed to develop consensus.

t " y'(

,,q 8.2 AOV Programs e

>N During plant visits, we noted that work had been done to satisfy the requirements of GL 88-20,

" Individual Plant Examination for Severe Accident Vulnerabilities - 10 CFR 50.54(f),"

November 23,1988, (and its supplements) and the Maintena' nce Rule (10 CFR 50.65 [Ref.1])

regarding AOVs and their suMorting system's form'ed the foundations of many of the licansees' AOV progrhrhsMThe ob]'bbtives of those prograhis are to focus on AOVs and assure that they operate satisfd6torily to achieve safe reliable economical plant operation.

tW E..

.. (w All of the plantshisited had initiated'AO,V programs. Many, but not all of the programs focus on activities to co3firmgel capability an(operability of important AOVs. There are large plant-to-pjailfvariationsinithdtypes, numbers, applications, and risk importance of AOVs, and there arnignificant prdgramrnatic and schedular differences between the plants' AOV prograriis.

W"#

.n n

Tfe U.S. NRC is participating with industry groups to communicate the lessons of operating gxpegence.

gf Wm Eff$ctif6pOVgr'o,g grams would include the following elements:

y

1) *gggy1dentify risk-significant AOVs, including those which are considered pie alone systems such as the emergency diesel generators.

16 l

e 4

2)

Verify the operating capability of risk-significant AOVs under design-basis conditions, and periodically verify that the risk-significant AOVs' operating margins remain satisfactory.

I 3)

Follow manufacturers' recommendations for SOV installation and maintenance regarding the use of thread sealants and lubricants.

4)

Maintain instrument air quality in accordance with current industry standards and guidelines [ISA-S7.0.01-1996, and ASME OM-17].

, ey Utilizing risk-informed and maintenance rule methodologies, Affective OV programs can i

provide assurance that risk-significant AOVs are capable ofa6d will belable t6 operate satisfactorily in accordance with plant safety analyses. The elements of effectivs AOV g

programs are outlined in Table 4.

n ho :.ad 3,

g,r vQf Qffk eW Table 4 Elements of an Effective Air-OperatedpValve Program' AK a

.y:

1 A.

Catalogue the plant AOV population.

Q#

B.

Categorize the AOVs' importance based on:' [ ' !A N[E Safety analyses final safety analysis repbrt/ lice' rising requirements / technical specifications j gfg 4,

~'

Plant PSA (IPEs and IPEEEs) _ >

~. yn i

CDF, large early release frequency.

Fussell-Vesley importance, risk achievement worth, TEP/ minimum path set, etc.

Plant operating experience" Expert panet/ maintenance rule evaluations Transientinitiatorsi 2

,Ernegency oper@ingproceduresQh W

~

r

< bnormal operating procedures C. Verifyihat AOV installation, operating conditions, interface conditions and environmental

~

I conditions are within the AOV's design.

D. Confirp[$deEihhrNgins of "important" valves (and accumulators if applicable) for desihjn basis eventsfthrough:

[ Analytical techij @iisV y Diagnostic testi. g (st'atic and/or dynamic testing as appropriate) l EM. Verify pneumatic fluid quality periodically (on a frequency in accordance with industry

~,

/sMstandards/ guidelines [ISA-S7.0.001-1996/ASME OM-17]).

I 1

& ?(iw w.Y Lf g;W}f including " stand-alone systems'such as emergency diesel generators, BWR scram system, main steam relief and isolation systems, heat and ventilation systems, and air-operated dampers.

I 17 t'

\\

)

Table 4 Elements of an Effective Air-Operated Valve Program (Cont.)

{

l F.

Verify the adequacy and operability of safety-grade backup air accumulators Baseline: Verify accumulator sizing Periodic: Testing of check valves for leakage, monitoring / alarming accumulator j

pressure, verifying accumulator sizing, checking for the presence of water i

accumulation i

G; Perform preventive maintenance periodically on AOVs and their support systems i

AOV subcomponents (diaphragms, positioners, o-rings, springsfetc.T AOV supporting components (SOVs, pressure regulators, filtersjefc,.),

Supporting pneumatic systems (dryers, filter, moisture separators; compressors, u

pressure regulators, extemal compressed gas sy' stems, accifrnulato.rs' n

,g

)

a y..~ c H. Perform periodic surveillance testing to verify / confirm,that, operating mar $ns of'riske,

)

significant AOVs remain satisfactory for the duratior[6f plant life.

'% W j

l.

Train and sensitize plant operations and maintensoce'shf to the susceptibility of AOVs to CCFs emanating from their subcomponents, su6por~tind cSrhponents, and their

)

supporting pneumatic systems.

j Mcy W

j GQ$%

Tf 8.3 Supporting Pneumatic Systems ph c

W.

Many plants' AOV programs are addressing the design, opeisti8rYs, and testing aspects of the AOVs; however, it appears that many plants are not focusing on the AOVs' supporting pneumatic systems. At many plants, additional attention may be needed to focus more attention on AOV supporting pneumatic systems to minirnize the likelihood for unanalyzed common-cause AOV failures.. Of paramount importar)ce is the need to monitor the instrument air quality (moisture in particular)lin accordar)ce with' current industry standards and guidelines (ISA-S7.0;01i199.6, and'ASMEiOM-17).

<d

%& A.

' WM,

I e

.w w

8.4 StandwiSystem Air-Operat5d: Valves Alope y n m.

ve.

$ ' (' y %

u\\N At many plants,.important AOVs in stand alone systems (emergency diesel generators, main i

steam relief Aridisolatl66 s items, heat and ventilation systems) have not been included in the plants'AOV programs. Thess omissions were based on historical precedent, not on the risk-inforrned methodology that is being used for developing plant AOV programs. Inclusion of theseValves in plant AOV pro' grams may identify some as risk significant.

$'l N

8.5..Other Power-Ope, rated Valves

(

c't n

s O

Alt ough they are;not the subject of this report, considerations similar to those for AOVs may apply'toMOVs1 Consideration of these valves may identify some as risk significant.

gge 18

G 9

REFERENCES 1.

U.S. Nuclear Regulatory Commission,10 CFR 50.65,

• Requirements for Monitoring the Effectiveness of Maintenance at Nuclear Power Plants."

2.

U.S. Nuclear Regulatory Commission, Generic Letter 89-10, " Safety Related Motor-Operated Valve Testing and Surveillance - 10 CFR 50.54(f)," June 28,1989.

3.

Idaho National Engineering and Environmental Laboratory draft report,,"A' Study of Air-Operated Valves at U.S. Nuclear Power Plants," INEEUEXT-9,p:%

8 00383, April 26, 1999.

<;c.j %:h

,, g,

~M

,p /

x 4.

U.S. Nuclear Regulatory Commission, InformatiortNotice 96-68,'" Incorrect EffectiveJ Diaphragm Area Values in Vendor Manual Resultiin Potential Failure of Rne.urnaticf c

@ISgf MfP j

Diaphragm Actuators," December 19,1996.

p" W %#

5.

U.S. Nuclear Regulatory Commission, Informati$kNotics 95-34, " Air Actuator and j

Supply Air Regulator Problems in Copes-Vulcan Eiessurizer Power-Operated Relief Valves," August 25,1995.

W '.f(6, 3$\\

6.

U.S. Nuclear Regulatory Commission, Infornistion Notice 88224;;" Failures of Air-Operated Valves Affecting Safety-Related Systelms4May 13,1988.

g p.

e

)

..- Industry Perspective Air-Operated Valves,y,;eseilled at 2nd Joint Meeting of the

+

7.

pi Air-Operated Valve and Motor Operated Valve User',s'Grou' p Meeting (AUG Meeting Number 16), December 1998. ^

8.

Coleman, M., " JOG-AOV. Program," presented at 2nd Joint Meeting of the Air-Operated Valve and Motor Operated Valve User's Group Meeting (AUG Meeting Number 16),

December 1998.

r

[Qp 3y _

' ^...

9.

Pro @edings of the'4th intemational Conference on Probabilistic Safety Assessment and Top Ev6nflhe,v.Nierodei'C.F.",:Wellumson, T.P., Worrell, R.B., Blanchard, D.P Management)

'ention Anhiysis to Select a Safety-Significant Subset of Air-Operated

~

Valvesj,ofJesiijri " September 1998.

- ?, fQ;. <

.V m

r Yg' 1

y y

?

Ek.,

pi%

kh k <.&

19 L