ML20058B501

Summary of 900801 & 0907 Meetings W/Bwr Owners Group Re Potentially Degraded or Nonconforming Condition

ML20058B501
Person / Time
Issue date:	10/09/1990
From:	Gody A Office of Nuclear Reactor Regulation
To:	Zwolinski J Office of Nuclear Reactor Regulation
References
NUDOCS 9010300240
Download: ML20058B501 (71)

Text

t

'p 'cw [

I6 UNITE D sT AT[s I

[

j NUCLE AR REGULATORY COMMISSION gg[/j umiNotoN.o.c rom

%,]{,

October 9, 1990 MEMORANDUM FOR: John A. Zwo11nski, Assistant Director for Region III Reactors Division of Reactor Projects Ill, i

IV, V, and Special Projects

1. ) John N. Hannon, Director THRU:

Project Directorate 111-3 4^ Division of Reactor Projects !!!,

IV, V, and Special Projects T?.C'S Anthony T. Gody, Jr., Project Manager Project Directorate 111-3 Division of Reactor Projects !!!,

IV, Y and Special Projects

SUMMARY

OF NRC STAFF MEETING WITH THE BOILING WATER REACTOR (BWR) OWNERS-REGULATORY RESPONSE GROUP DN AUGUST 1, 1990, AND WITH THE BWR OWNERS GROUP ON SEPTEMBER 7, 1990 i

AUGUST 1, 1990 MEETING On August 1,1990 r,he NRC staff met with the BWR Owners Group (BWROG)

Regulatory Response Group (RRG) to discuss the actions necessary to address concerns about a potentially degraded or non-conforming condition existing incertainHighPressureCoolantJnjection(HPCI),ReactorCoreIsolation Cooling (RCIC) and Reactor Water jleanup (RWCU) containment isolation valves.

The NRC staff summarized its review of the data provided by the BWROG associated with the capability of Motor-Operated Valves (MOVs) used for containment isolation in the HPCI, RCIC and RWCU systems. The staff's

' review of the data provided by the BWROG indicated that the subject MOVs may i

require acceleration of the recommendations in Generic Letter (GL) 89-10, 's

" Safety-Related Motor-Operated Valve Testing and Surveillance." The BWROG l

presentation included a generic safety assessment of e postulated non-functional condition in one of the subject MOVs.

The BWROG's assessment confirmed the staff's determination that the safety significance of.this issue does not require imediate resolution.

The BWROG proposed action plan included short-term and.long-term recommen-dations to the holders of operating licenses for BWRs. The short-term recommendations would include operator briefings, frequent system walkdowns,

(

~ ] ('

G I.

jy /$

NRC RE :EIG COPY M'

2A 'W# PMc r

DFol C

PNU o" "

gh

Mr. John A. Zwolinski review and confirmation of plant-unique applicability of generic safety assessments. EPRI inspection of INEL test valves and assessment of the applicability of INEL test data of HPCI, RCIC and RWCU MOVs.

The proposed long-term recommendation was to address potential deficiencies in individual plant GL 89-10 programs.

Following the presentations and discussion, the NRC staff stated that the BWROG did not present adequate plans to resolve the ?otential deficiencies in the subject MOVs in advance of the recommended sciedule of 5 years in GL 89-10.

As a result, the staff indicated that it would consider the prepara-tion of a generic communication to licensees, addressing the need to evaluate the capability of the subject HPCI, RCIC and RWCU MOVs and correct any identified deficiencies within the GL 89-10 program but on a priority basis.

SEPTEMBER 7, 1990 MEETING On September 7, 1990, the NRC staff met with the BWROG to discuss the current activities and proposed future recomendations of the BWROG to address the potentially degraded or non-conforming condition which may exist in certain HpCl, RCIC and RWCU containment isolation valves as discussed

above, i

The BWROG presentation included a discussion of current activities, proposed future actions, and the BWROG position on mispositioning. Current BWROG activities include recommendations to all BWR licensees to begin a plant-specific safety assessment, review of Emergency Procedure Guidelines in e

response to leakage on an unisolable line break scenario, and participation in INEL valve inspection.

In addition, the BWROG stated that, during the next scheduled meeting with its MOV committee, the RWCU design basis will be reviewed with emphasis on the need to accelerate RWCU valves in the GL 89-10 program. The BWROG stated that the INEL data indicate that an indepth l

review of current safety-related MOV design methodology is apsropriate, however, the data is not appropriate for redefining MOV opera)ility bases at this time. The BWROG indicated that they would react responsibly to data presently being finalized and that a basis for reasonable assurance of operability should be made available by approximately June 1991.

The staff agreed that further review of MOV design methodology is appropriate. The INEL test data, however, will be considered by the staff as the best data available until more analysis is performed.

The-BWROG restated its views on mispositioning, which, for the most part, had' rejected the staff's recommendation in GL 89-10. The staff explained the mispositioning recommendation as discussed in Supplement I to GL 89-10.

Following the staff's explanation of the mispositioning issue, the BWROG agreed to reevaluate their position on mispositioning.

1 i

1

o Mr. John A. Zwolinski If you have any comunents or questions regarding these issues, please contact Anthony T. Gody, Jr. at (301) 492-1387 or Thomas G. Scarbrough at (301) 492-0794.

O!iginaltj nedb/

g Anthony T. Gody, J... Project Manager Project Directorate III-3 Division of Reactor Projects III, IV, Y and Special Projects

Enclosures:

1.

List of attendees, August 1, 1990 2.

Staff evaluation of M V data provided by the BWROG 3.

BWROG-RRG presentation slides 4.

BWR Owners Group Safety Assessment of MOV isolation Function 5.

List of attendees, September 7, 1990 6._

BWROG presentation slides cc: See next page Ni PD33 Reading NRC & Local PDRs Thiraglia JPartlow DCrutchfield JHannon EJordan MSlosson AGody. dr.

J ACRS (10)

OGC JClifford 4

i PD33:PM EMEB fPD3 D

AGody,J:k PD33iLA PKivitzer

./bj TMarsh JHannon l

j0/5/90 g)/r/90 10/ r1/90,O/]/90 DOCUMENT NAME: MEETING SUMMMARY 08/90 l

hf

• o' Mr. John A. Zwolinski

-3 1

. If you have any coments or questions regarding these issues, please contact Anthony T. Gody, Jr. at (301) 492-1387 or Thomas G. Scarbrough at (301) 492-0794.

H Anthony T. Gody, Jr.,

roject Manager Project Directorate 111-3 Division of Reactor Projects 111, IV, Y and Special Projects

Enclosures:

1.

List of attendees, August 1, 1990 2.

Staff evaluation of MOV data provided by the BWROG 3.

BWROG RRG presentation slides 4.

.BWR Owners Group Safety Assessment of MOV Isolation Function 5.

List of attendees, September 7, 1990 6.

BWROG presentation slides cc:.See next page l

F f

i

'.,3 om,.

1 I

l w

i J

cc:

]

s,

Steven D.

Floyd, Chairman BWR - Regulatory Response Group (CP&L) l 411 Fayetteville Ave.

l

- Raleigh, N.C.

27602 i

' Clive Callaway, NUMARC l

1776 I St. - N.W.

Washington, D.C.

20006 George-J. Beck,.Vice-Chairman BWR - Regulatory Response Group (PEco)

}

965-55 Chesterbrook Blvd.

t

Wayne, Pa.

19087 Steven J.

Stark, GE Nuclear Energy i

175 Curtner Ave.

San Jose,- Ca.

95125 t

I i

i 1

'f

.i.

i f

i i

s

't i-i

-l I

'I t

~

k

,. i s

l l, V:

"1 -

o 1

ii I.

'l l p

F

1 5

c.

6

'6

'o ENCLOSURE 1 LIST OT ATTENDEES FOR PUBLIC MEETING WITH THE BWR OWNERS GROUP - REGULATORY RESPONSE GROUP AND THE NRC AUGUST 1, 1990 NAME AFFILIATION Frank Orr NRC/NRR/DRP R.A. Hermann NRC/NRR/DET Alan Redpath CP&L Steve Gallogly General Physics G.H. Weidenhamer NRC/RES R. Steolo INEL W.

Farmer NRC/RES Richard J.

Kiossol NRC/NRR Owen Rothburg NRC/RES Robert S. Lewis PSE&G Robert D.

Binz IV PSE&G/BWROG Vice-Chairman Robert F. Janocok CECO / BWROG - RRG member Dave E. LaBarge NRC/NRR/DRP F.T. Grubolich NRC/NRR/EMEB Robert Stransky NRC/NRR/DRSP John B. Hickman NRC/NRR/DRSP Georgo Thomas NRC/NRR/SRXB David Allsopp NRC/NRR/PMAS Gerald Klinger NRC/NRR/PMAS John Hannon NRC/NRR/DRSP Larry Crockor NRC/NRR/DRP Warron J.

Hall NUMARC C.Y.

Chong NRC/NRR/EMCB Robert C. Jones NRC/NRR/SRXB A.

El-Bassioni NRC/NRR/RAB Anthony T.

Gody, Jr.

NRC/NRR/DRSP Tod Sullivan NRC/NRR/EMEB L.B. Marsh NRC/NRR/EMED T.G. Scarbrough NRC/NRR/EMEB J.E. Richardson NRC/NRR/DET J.G.

Partlow NRC/NRR/ADP W.T. Russell NRC/NRR/ADT J.D. Holdt GPCo/ BWROG - RRG member i

G.J.

Beck PECo/BWROG - RRG Vice Chairman l

S.D.

Floyd CP&L/BWROG - RRG Chairman Denver Aurcod GPCo/BWROG - MOV Committoo Chairman Ernie'Rossi NRC/NRR/DOEA Jim Linville NRC/OEDO-Frank J. Witt NRC/NRR/EMCB Stephen Kosciolny NRC/NRR/EMCB W.E.

Campbell, Jr.

NRC/NRR/EMEB R.B.

Borsum BWNS M.M. Slosson NRC/OEDO

I.

'e I

o t

ENCLOSURE 1 LIST OF ATTENDEES FOR PUBLIC MEETING WITH THE BWR OWNERS GROUP - REGULATORY RESPONSE GROUP AND THE NRC (CONTINUED)

NAME AFFILIATION i

Milton Vagins NRC/RES l

Dave Noonan Search Licensing - Bechtel Dave McGill IEL&P

=

John Hosler EPRI Michael J.

Bennett NPPD Paul Damerell MPR i

Ray Morris Entergy Operations l

^

Kenneth Dempsey NRC/NRR/EMEB George Wunder NRC/NRR/DRP Thomas Hicks Southern Technical Services Kevin Brown Liberty Technologies

~l t

1 i

i

+

4,iit '. P-p.

, e.,... ;,

(

u : - m.

" i A.

ua o e ',

i m

W,*

W1 V '., +.

ENCLOSURE ~2

, ny - 7,,

,m w.

_','?~

'l k. $

b 1^

4....

.e,4

.y s

2.,

[U'+

'I h,,.h n.p-y

.TI

4-1 g

s L

i l

1

.V 1

a e,

>w 1

4 s

'jf i

- i, w,

f I.,

- EVALUATION OF MOV DATA 1

f f

4 g

m. -

PROVIDED.BY THE BWR OWNERS' GROUP l,

ma c

.m j.

u.

n,

.x T. G. Scarbrough

j t

-a, 1.

1990 o

g August 1, l

i t i

f

- 4i.

~$

E-to sJ-h

,i a

c s

a

'l h,

I

,I

]

< z'. ; c ' ; g

x 7) e 3

"sf s

s' g

y:

1 lI

.s t

5 1

3 h

r 4

,+

.7. i t

t I

f

\\' l i

i '*

..f i

->I r

5 t

.\\

p.' u,

E

.h f

9 (b

..f

'..V-.

?

>,l nn 1

=- p, y ; t.

f I.,-

-f

\\>

b u'b (, t,

'.s.>

1,

'4 is?>,

4

'l i

m' 7

. y hj

. Y i *l T

1 sf_

t i <

w:41 e,

1 m

W ~ a.:'

{i',

j j

t ':,';

,\\ L s

_v.

e

_;z.,.

s.

.fs,t 2

'i=

n

- l-y J

e a 4

3

.' \\

'n.

e,.

i' ' i g

- 4

,!. r 3.

s

,._. r t,. fJJ-y# ' h 4' 5, f 4 s

i

+

2-

$. (.,.

g;

'4.,,

2.,;

,a x

3

~ l ^;

N

s-r i

mu A

s

'3 >

t

,1.5,'

z_

.," r/ 4..:

n i

k T.

+

,gj

-[f'

)

e i )y g

.,g-

.. gr.-

4 s

~

5 i

s u' '

.h,

.e

,*g.

(

• f.A. u s

'l g,,

. - i t

O.,5

.('

'3-S.,?-

m. -

",1 fy l 'f s t, '

Y.

0

_i 1

~,,i <

I a

6

-_.r t' 'ifli 5 ti 1

P

.e

" Ifj e

\\1 I

[

q 1

f m

g j

a*

y

.u

,. k 4.l.. -

~{

h

-( p..[...

r

'a

'!,- {[j g y 5,

b F

a i -~

q ! ':

{ h.}.-... - <.i '

_t s y-s<v t

r c

h

,.[ t

};.- -

I i

3.

o. nf.

u a i

o

d..fid%u.

Is>... '. 3. 2.j g J. ( c

, if s -

s I

,' u i "

t.r 1

_f se ry

_.+

r s

j

.~, '

+_

=v s

i j y i, _

v h * *., xpl'd c

" j s '

%%O kW u#

t i

di!)j { ',~' h j ^ s 7 'k l

c MOV DATA REQUESTED FROM THE BWR OWNERS' GROUP L i L FOR THE MOVs USED FOR CONTAINMENT ISOLATION IN THE STEAM SUPPLY LINES OF THE HIGH PRESSURE COOLANT INJECTION (HPCI) AND REACTOR J CORE ISOLATION COOLING (RCIC) SYSTEMS AND IN THE SUPPLY LINE TO THE REACTOR WATER' CLEANUP (RWCU) SYSTEM, THE FOLLOWING DATA WERE REQUESTED ^ 1. TYPE AND SIZE OF MOTOR, ACTUATOR, AND VALVE (INCLUDING DISK), l 2. MANUFACTURER OF MOTOR, ACTUATOR, AND VALVE, 3. DESIGN DIFFERENTI AL PRESSURE AND FLUID TEMPERATURE FOR OPENING AND CLOSING OF THE VALVE, AND g 4. THRUST DELIVERED AT THE CURRENT TORQUE SWITCH SETTING, DIFFERENTIAL PRESSURE AT WHICH TESTS CONDUCTED, AND BASIS FOR DELIVERED THRUST VALUE. l f \\ 'r ( r i t

t I ^* j -- i METHODOLOGY USED IN THE EVALUATION OF THE MOV DATA L INFORMATION NOTICE 90-40 PROVIDES ESTIMATED THRUST REQUIREMENTS FOR THE MOVs TESTED FOR THE NRC MOV TYPES NOT IN THE TEST PROGRAM (BUT OF THE SAME SIZE) WERE ASSUMED TO PERFORM IN A MANNER SIMILAR TO TESTED VALVES REQUIRING THE LEAST AMOUNT OF THRUST l THRUST REQUIREMENTS WERE ESTIMATED FOR MOV SIZES DIFFERENT THAN THE TESTED MOVs GATE VALVES WERE EVALUATED MOVs WITH IDENTIFIED GLOBE VALVES WERE ASSUMED TO BE ADEQUATE RATING OF ACTUATOR COMPARED TO REQUIRED THRUST ESTIMATES MOTOR SIZE COMPARED TO MOTOR SIZE USED IN TESTS AND ESTIMATES OF THRUST CAPABILITY I s i l ls A + w-,m,,

r l'. S.

3..

CATEGORIZATION OF MOVs + 's 1. NO CONCERNS IDENTIFIED DURING QUICK-LOOK PROCESS . 2.. MARGINAL MOTOR SIZE, ACTUATOR SIZE, OR TORQUE SWITCH SETTING 3. SMALL MOTOR OR ACTUATOR SIZE, OR LOW TORQUE SWITCH SETTING (AT THE PRESENT TIME, THE STAFF IS FOCUSING ONLY ON MOVs IN THIS CATEGORY.) 1: t \\ 6 i I 1

S i l 7/31/90 J BWROG MOV DATA OVERVIEW i i HPCI-TOTAL NUMBER OF VALVES = 46 ) l MOVs WITHOUT IDENTIFIED CONCERNS (INCLUDING 4 GLOBE VALVES) 18 MOVs WITH MARGINAL MOTOR, ACTUATOR, OR T. S. SETTING 16 MOVs WITH SMALL (OR LOW) MOTOR, ACTUATOR, OR T. S. SETTING 12 UNITS L, M, P, T, V, Z, HATCH 1, HATCH 2, MONTICELLO* (9 OUT OF 23 REACTOR. UNITS)

• JUSTIFICATION SUPPLIED

.RCIC TOTAL NUMBER OF VALVES = 62 MOVs WITHOUT IDENTIFIED CONCERNS (INCLUDING 7 GLOBE VALVES) 47 MOVs WITH MARGINAL MOTOR, ACTUATOR, OR T. S. SETTING 9 MOVs WITH SMALL (OR LOW) MOTOR, ACTUATOR, OR T. S. SETTING 6 UNITS E,G, N,Q,T (5 OUT OF 30 REACTOR UNITS) 5 RWCU TOTAL NUMBER OF VALVES = 71 MOVs WITHOUT IDENTIFIED CONCERNS (INCLUDING 8 GLOBE VALVES) 19 MOVs WITH MARGINAL MOTOR, ACTUATOR, OR T. S. SETTING 12^ MOVs WITH SMALL'(OR LOW) MOTOR, ACTUATOR, OR T. S. SETTING 40 L UNITS B,: D, H, I, K, L, N, P, Q,=R, S, T, U, V, W, Y, Z, AC, HATCH 2, QUAD CITIES 1, QUAD CITIES 2 (21 OUT OF 34 REACTOR UNITS) 8 UNITS WITH MOV PROBLEMS (SMALL/ LOW CATEGORY) IN MULTIPLE SYSTEMS j HPCI + RCIC + RWCU 1 (T) HPCI + RCIC= 0 HPCI'+ RWCU S (L, P, V, Z, HATCH 2) RCIC + RWCU 2 (N, Q). 1 i L 1 l l

7/31/90 NUMBER OF MOVs IN BWROG DATA VALVES SIZE TOTAL MOTOR ACTUATOR T.S. SETTING (in.) MARGINAL SMALL MARGINAL SMALL MARGINAL SMALL A/D 3 14 0 0 0 0 1 3 4 14 0 0 0 0 3 0 6 32 3 22 19 0 0 25 8 7 3 1 5 0 4 3 10 22 11 4 0 0 2 5 POWELL 3 0 0 0 0 0 0 0 4 8 0 0 0 0 2 0 6 6 0 0 1 0 3 0 10 6 3 1 0 0 0 0 8 0 0 0 0 0 1 0 CRANE 3 6 0 0 0 0 0 0 (CHAPMAN) 4 2 0 0 0 0 0 0 6 7 4 2 6 0 3 4 8 4 0 0 0 0 2 2 10 12 5 0 2 0 1 5 WALWORTH 3 6 0 0 0 0 0 0 4 0 0 0 0 0 0 0 6 3 0 3 3 0 0 1 8 4 0 0 0 0 1 0 10 2 2 0 1 0 2 0 VELAN 1 1 0 0 0 0 0 0 3 1 0 0 0 0 0 0 4 1 0 0 0 0 0 0 6 6 0 0 0 0 1 3 8 6 0 0 0 0 1 0 10 3 0 0 0 0 0 1 ROCKWELL 1 1 0 0 0 0 0 0 WEST. 3 1 0 0 0 0 0 0 BORG WAR. 10 2 0 0 0 0 0 0 6 2 0 0 0 0 0 0 (GATE VALVES, EXCEPT 4 A/D 10-inch, 6 A/D 6-inch, 4 A/D 3-inch, 2 VELAN 8-inch, 1 VELAN 3-inch, 1 ROCKWELL 1-inch, AND 1 WESTINGHOUSE 3-inch GLOBE VALVES.)

r. Tg l; il pil. 'J. ; i-/y /7/31/90 tE l v ii MOVs WITH_SMALL (OR LOW) MOTOR, ACTUATOR, OR T. S. SETTING 58 TOTAL 4 HPCI RCIC RWCU SMALL' MOTOR 4 0 2 l; i LOW T.S. SETTING 8' 4 13 i.., r .SMALL MOTOR +, LGW T.S. SETTING 0' 2 25 $f* L k '. .,j t i f ': I h i e y ? 5 r.-. 'YM [f, w ~ { -. A p l3 ' .'('. 4 . $ 'a l' i. g, ' :y s-s

.i, 1 7/31/90 MOST SIGNIFICANT FINDINGS FROM BWROG DATA UNIT SYSTEM VALVE SIZE D/P T.S. SETTING THRUST (in.) (psid) (lbs) ESTIMATE FROM TEST (Ibs) 1 M HPCI CRANE 10 1200 17460 29000 M' HPCI CRANE 10 1200 22540 29000 T-HPCI A/D 10 1250 26271 30000 T HPCI A/D 10 1250 20326 30000 V HPCI CRANE 10 1250 24017 29000 HATCH 1 HPCI CRANE 10 1080 23055 29000 i LQ RCIC A/D 10 1146 23478 30000 j D RWCU A/D 6 1020 12300 20000 -D RWCU A/D 6 1020 16100 20000 I RWCU A/D 6 1190 10039 20000 .K RWCU A/D 6 1040 12241 20000 K RWCU A/D 6 1040 14928 20000 L RWCU A/D 6 1150 13233 20000 L RWCU A/D 6 1150 13220 20000 N RWCU. A/D 6' 1250 13405 20000 1 N RWCU A/D 6 1250 13405 20000 l P RWCU A/D .6 1150 16069 20000 Q RWCU . A/D 6 1250 13786 20000 P RWCU A/D 6 1200 13405 20000 Q RWCU A/D 6 1250 13405 20000 R RWCU . A/D. 6 1173 13780 20000 S RWCU A/D 6' .1025 12800 20000 S RWCU A/D 6 1025 12800 20000 T. RWCU A/D^ 6 1020 9354 20000 T RWCU A/D 6 1020 11465 20000 4 W< RWCU-A/D 6 1135 15400 20000 'Y~ RWCU A/D 6 1025 12800 20000 Y- -RWCU - A/D 6 1025 12800' 20000 QC 1-RWCU CRANE 6 1250 6506 12000 QC 1 'RWCU A/D 6 1250-8333 20000 QC 2t RWCU CRANE 6 1250 4004 12000 QC 2 RWCU. A/D 6 1250 10190 20000 [ ?! l L

BWR Owners Group Regulatory Response Group-PRESENTATION TO NUCLEAR i REGULATORYCOMMISSION ON MOTOR-OPERATED VALVES i t l August 1,1990 l Rockville, MD i i I -l M : i 5i e2 i g. 1.ab l e i

BWR Owners Group Regulatory Response Group AGENDA Introduction S. D. Floyd/D. Atwood Leak-Before-Break S. Ranganath i Leak Detection Sys G. B. Stramback i 4 1 ECCS P. W. Marriott 1 Radiological P. W. Marriott l EnvironmentalQual H. P. Williams l I Valve Surveillance J. D Heidt l Testing & Actuation [ 1 Experience i I System Effects G.J. Beck i F Assessment & D. Atwood/S. D. Floyd i Recommendations i i i f t d

BWR Owners Group Regulatory Response Group INTRODUCTION o BWROG understands NRC concem based on present INEL data o BWROG believes safety assessment demonstrates that current configuration is adequate for interim period I BWROG and utility program GL 89-10 willprovide final resolution o s 1 l Page sDF 1 l l

c BWR Owners Group Regulatory Response Group. i MOV OVERVIEW l o INEL data implications I - Recognize significance - FinalINEL results to be published late 1990 - Lack of concrete conclusions o Chronology l l - IEB 85-03 design basis review (HPCI/HPCS/RCIC) \\ \\ - GL 89-10 design basis review (Oct. 89) l - INEL Data Public Meeting (Apr. 90) ( - NRC INEL Data Review with BWROG (May, 90) ( - BWROG Endorsed EPRI Review ofINEL Data (May, 90) - Safety assessment initiated (May, 90) - IN 90-40 (June,90) i - Industrysurveyby BWROG (June,90) l - Issued list of Concerned Valves to Utilities (July, 90) l - RRG Activated (July,90) / 1 l l Page sDF5 I

l !\\Ii !t Ii L {h!l! .it p u or G esnopseR y ro ta lu 6 g F e D R Se p g a u P o rG sren w sse O r gor R P s w n W i e i s v m e B W a R r E g s i I o s V r a g R P B n it E 0 n s 1 g e V T i 9 s O 8 e P V L D d G O M o .r I II1i 1 t ' i i a 1

BWR Owners Group Regulatory Response Group. SAFETYASSESSMENT

SUMMARY

o ' Individualplant licensing documents have established a basis for leak-before-break (LBB) scenario o Leak detection system exists to support LBB criteria l 0 Isolation occurs at flow /dp lower than design basis i r o Even with guillotine break, realistic assessment shows high probability ofline isolation i i L f l Page SDF2 4 ~

BWR Owners Group Regulatory Response Group. SAFETYASSESSMENT

SUMMARY

l (CONTINUED) l l o ECCS willprovide adequate core cooling following postulated design basisline break o Radiological release within 10 CFR 100 limits i o Typical ECCS equipment has been qualified for conditions l comparable to expected worst case conditions i l I i i Page SDF3

BWR Owners Group Regulatory Response Group-LEAK BEFORE BREAK CONSIDERATIONS o Piping Design Margin o Leak-before-break assessment I Leak rate vs Crack Length i o Conclusions relative to MOVisolation issue l i i. I i Page SR 1 ~

4 BWR Owners Group Regulatory Response Group _ PIPING DESIGN MARGINS o Piping designed to ASME/ ANSI Code requirements Safety factor of 3-4 forpressure loads [ Materialselection assures high toughness Fatigue margin built into design analysis I o Environmental etfects not significant i i Most of the piping at issue made of carbon steel (Not l susceptible to SCC) l' Low steady state temperatures on most stainless steel i piping Probability of Cracking Low I Page SR 2 t

BWR Owners Group Regulatory Response Group-LEAK BEFORE BREAK (LBB) ASSESSMENT l i o Through Wall Cracks in Pipes 'Announce' themselves in the form ofleaks wellbefore criticalsize O BWR plants have redundant leak detection systems Leaks are detected and system isolation occurs i o Leak rates are large enough for detection, but not comparable to that underpipe break condition I o Basis forLBB is alreadyin most SARs 1 o LBB concept accepted by the NRC and recognized in GDC-4 j,, y,.i '- [- i J ,,,4 n v // j, g, ' *l' LBB offers added margin above built in /f 3,I' design capability / . ?? S#- . ;,. <. w Page SR 3 4 L

BWR Owners Group Regulatory Response Group _ 1-LEAKRATE AND CRITICAL CRACK LENGTH l 0 Once through wall cracking occurs, leak rate increase with time L Steam cutting causes crack area to increase i Steam cutting effect not included in analysis l Sufficient ti,me for detection. Crack propagation rates are small i o LBB margin increases with pipe size Large pipes (> 6 in. Diameter) where breaks could be more significant have higher LBB margin. Smallpipes have lower LBB margins, but consequence r of break is less severe. Large data base exists on small l pipeleaks confidence o Leak rates are large enough for detection, yet flow is smallso that valve loadingis nearnormal I Page sR 4 1 i 1 = y

@ GENuckerEnergy Engineenng Calculation Sheet NUW8tR OATE SuSJECT - SY Swit?- _ 0 88 Mc DF, L IN C L C r)) N Q l Sre As cc TriNG AND PIPE-6 6 N DIN 6 j )N / PRE DIC T'O N f INCL U D I N6 STEAN / CC T T I NC1 / / Ld I b< Cc W C,6 R VAT I V E - M / LEAK A ATE McDEL / 2 \\ d l l _. /._. __ W zqp 1 / l i / / l / I I / i c l CRACK LENGTH l I e gg .: 9:: m -

~ < e-1

j':

j TABLE 1 VALUES OF PARAMETERS USED IN CRITICAL CRACK LENGTH M LEAK RATE CALCULATIONS Pipe Thickness .i, Schedule 80 U Pipe Internal Preuure Temperature 105g psi 528 F Normal Operation Bending Strenes 4ksi Material Stainless Steel or q Carbon Steel i TABLE 2 CRITICAL CRACK LENGTHS AND LEAK RATES FOR VARIOUS DIA PIPES Pipe Diameter Critical Crack Leak Rate at Critical (in.) Length (in.) Crack Length (gpm) Water Steam i 4 4

7. I 2.5 15 6

9.8 41 27 12 18 3 166 108 16 23.1 262 170 f GBS90A35 wp 1$- 'S R 6 '-

1 BWR Owners Group Regulatory Response Group - [ WATER HAMMER / SEISMIC LOADING l l 0 Testing and field experience show that under limiting water )^ hammerloads, piping integrity is maintained, and at worst only the pipe supports are likely to fail. Extensive water hammer tests under an EPRI/NRC sponsored program confirmed piping integrity o Seismic loads are not likely to cause pipe failure t EPRl/NRC funded tests on both crackcd and uncracked pipes could not fait pipes underseismic loads up to 10 times the ASME code limits Water hammer / seismic loads do not erode . LBB margins Page SR 7

BWR Owners Group Regulatory Response Group! EROSION-CORROSION o Erosion-corrosion is of concern in Carbon steelpiping High flow velocity. Low oxygen I t Continues system use t o Not an issue forstainless steel piping Carbon steelpiping affected by the MOVissue have o Low flow rates in steady state condition Infrequent operation Oxygenated water Erosion / corrosion not significant issue Page SR 8

~- = ~. BWR Owners Group Regulatory Response Group. CONCLUSIONS r I o Pipes have built-in design margins that protect against cracking i o Even if through cracks occur, the leak rates are high enough for detection, yet flow is smallso that valve loading is near normal o LBB offers added margin 1 i Page SR 9 l

BWR Owners Group Regulatory Response Group - Leak JDetection System. o RCIC/HPCI/RWCUpiping i o Primary method - air temperature monitoring s - Alarm -Isolation - Temperature equivalent to less than 25 gpm. -Smaller than criticalcrack size - Early detection Valve closure is completed long before line break and associated high valve DP occurs i t Page GBS 1 l t

BWR Owners Group Regulatory Response Group 1 ' Leak Detection System l o Additionalmeans ofleak detection l l Area / equipment room drain sumps Radiation monitors Physicalinspection Areas are accessible Manualoperatoraction { I l Immediate operatoractionis not requiredsince crack propagation occurs veryslowly Page GBS 2

!~ BWR Owners Group Regulatory Response Group 1 l. Leak Detection System l l Isolation o ' HPCI/RCIC and RWCU are provided with two MOVs in series o Valves mounted close as possible to containment wall (inboard valve)/(outboard valve) o Valve isolation - initiated in ditferent divisions o Valve closure power different sources (AC/DC) l Probability of both valves not closing is very remote even during high DP conditions Page GBS 3 t -~ s ~

BWR Owners Group Regulatory Response: Group ~ CORE COOLING SYSTEM IMPACT o Bounded byinside line break o Only one ECCS pump is required Provides adequate core cooling i i l Page PWM 1 j I i

BWR Owners Group Regulatory Response Group? RadiologicalRelease o Conservatively bounded by outside containment main steam line break analysis (SAR) o Large margin to 10CFR100 limits especially for smallleak detection system i l Within 10CFR100 requirements [ i Page PWM 2 l

BWR Owners Group Regulatory Response Group L l Equipment Qualification l o Smallleak - Leak Detection System range Enveloped by existing EO analysis i o Line break Inboard valve not affected by harsh environment Outboard typically enveloped by existing harsh EO conditions Typical ECCS equipment has been qualified for conditions comparable to expected worst case conditions Page PWM 3 ~-

1 BWR Owners Group Regulatory Response Group SYSTEM SURVEILLANCE TESTING & ACTUATION EXPERIENCE Technicalspecifications required testing - ASME Section XI required testing and Inspections Valvr. Operation during transient conditions l b i Page JDH 1

l. i BWR Owners Group Regulatory Response Group l' l TECHNICAL SPECIFICATIONS TESTING + Valve Stroking tests l System Operability tests System Walkdowns SECTIONXI TESTING Valve Functionality Inspections of system piping TRANSIENTS Plants have experienced isolation transients Many transients initiated via high flow Valves close i System testing and inspection plus transient i performance indicates valves operated underleak i before break conditions i Page JDH 2

[ BWR Owners Group Regulatory Response Group. SYSTEM SURVEILLANCE TESTING & ACTUATION EXPERIENCE Technicalspecifications required testing l l ASME Section XI required testing and inspections Valve Operation during transient conditions HVAC exhaust radiation level (SC/R) exceeds " Maximum Normal" l Area radiation level (SC/R) exceeds " Maximum Normal" Floor drain sump water level (SC/L) exceeds I " Maximum Normal" t 1 Area waterlevel(SC/L) exceeds " Maximum Normal" L Secondary containment delta-pressure reaches atmospheric Page JDH 3

.BWR Owners Group Regulatory Response Group u SYSTEM EFFECTS Expected dP under leak conditions similiar to normal system isolation Break flowlowerthan assumed Two valves in series help each other close - split pressure drop EPG guides operatoractions Page GJB 1

,I,,, y BWR Owners Group Regulatory Response Group. EPG CONTROL ACTIONS SECONDARY CONTAINMENT CONTROL - SC/T, SC/R, SC/L Double-Ended Break outside primary containment Scenario (i.e., into secondary containment) Motor-Operated Isolation valves fail to isolate due to excessive flow-induced differentialpressure Secondary containment parameters increase EPG Entry Conditions (Any One): Area temperature (SC/T) exceeds ' Maximum Normal" HVAC cooler delta-temperature exceeds (SC/T) " Maximum Normal" Page GJB 2

BWR Owners Group Regulatory Response Group l l - EPG CONTROL ACTIONS HVAC exhaust radiation level (SC/R) exceeds " Maximum Normal" Area radiation level (SC/R) exceeds " Maximum Normal" Floor drain sump water level (SC/L) exceeds " Maximum Normal" Area waterlevel(SC/L) exceeds ' Maximum Normal" Secondary containment delta-pressure reaches atmospheric Page GJB 3 i I i

i BWR Owners Group Regulatory Response Group l

SUMMARY

l l o Adequate safetymargin in design o Reactorisolation willbe achieved i o Core coolingprovided o Radiologicalrelease bounded o Typical ECCS equipment has been qualified for conditions comparable to expected worst case conditions t t Page sDF4

E BWR Owners Group Regulatory Response Group CONCERNS WITH APPLICABILITY OFINEL TESTDATA INEL Test results and present torque switch setting thrusts can not o conclusively be used to judge individualplant vake functionality Flex wedge versus double disc Unique valve intemal characteristics (design changes) Extrapolation of data to other manufacturers / sizes Criteria for evaluation of results (when was vake closed) i i l l \\ l Page SDF 11 1 l 1 .=

w BWR Owners Group Regulatory Response Group GENERIC' RECOMMENDATIONS SHORT TERM o Brief operators on MOVissue and remind their of RPV depressurization guidance in EPGs. o increase piping system walkdown frequency i o Review and confirm plant unique applicability of generic safety assessment O EPRIInspection ofINEL Test Valves o Assess applicability ofINEL test data and NRC evaluation to as-is capability of HPCI, RCIC and RWCU MOVs (subsequent to EPRI inspection) l LONG TERM i o Complete GL 89-10 Program j Page SDF 10

ENCLOSURE 4 i BUJR otuu as onoug --== BWROG 90103 clo Philodelphic (leCtric Compony + 955 65 Chestorbrook Blvd WC 638 5

• LUovne, PA 19067 5691 '

July 27, 1990 \\ Office of Nuclear Reactor Regulation United States Nuclear Regulatory Commission Vashington, DC 20555 Attention: James E. Richardson, Director Division of Engineering Technology Subj ect: BWR OWNERS' GROUP SAFETY ASSESSMENT OF MOV ISOLATION FUNCTION Inclosure: Safety Assessment: Isolation Function of MOVs for HPCI and RCIC Steam Supply Line and RVCU Water Supply Line, dated July 1990.

Reference:

BVR Survey, Data on Motor operated Valves, Ler G.J. Beck to L.B. Marsh, BVROG 9095, dated 7/6/90. The reference letter provided additional Boiling Water Reactor (BWR) plant l. Motor-Operated Valve (MOV) data requested by the NRC in response to an expressed Staff. concern about selected system MOV closure capability against design basis flow conditions. The Enclosure is a " safety assessment" of the l isolation function of these MOVs documenting the adequate safety margin of BVR plants. This assessment shows that a significant safety - concern does not exist, even if: the. system isolation MOVs any not have' adequately ~ sized actuators for saximum design basis conditions. Based on leak before-break considerations for the MPCI/RCIC/RWCU piping, it is not expected that system MOVs would ever be challenged at high flow design basis event. conditions. With the effective leak detection and isolation systems, leaks would be isolated early at low flow conditions. Additionally, realistic ' consideration of expected plant and system response to postulated accident conditions would lead to the conclusion that there is a significantly 'high probability of successful valve closure. Even without successful valve (- closures for a postulated rupture in these lines, there is adequate safety margin in the 1Energency Core Cooling Systems (ECCS) to handle the flow loss. The ECCS are designed for a much larger break. BVR utilities are already in the process of addressing this MOV thrust issue u as part of their response to GL 8910, which is a longer term versus intetim l-activity.' This letter has been endorsed by a substantial-number of the members of the BWROG. However, it should not be interpreted as a commitment of any e

James E. Richardson, Director .[ ' ' Oivi.sien of Engin2ering Technology BWROG.90103 July 27, 1990 Page 2 individual member to a specific course of action. Each member aust formally endorse the Bk'ROG position in order for that position to become the member's position. If you have questions regarding this issue, contact the undersigned, Denver Atwood, Committee chairman at (205) 877-7461, or Wendell Fiock, Committee Program Manager at (408) 925-1669. Very truly yours, _ h-George J. Beck, Chairman Bk'R Owners' Group Enclosure EXEC 1/GJB/rt i cc: Bk' ROC Primary Representatives BVROC MOV Committee RD Bin: IV, BWROG Vice Chairman BWROG Executive Oversight Committee SD Floyd, RRG Chairuan RR Galer, EPRI EP Shankle, INPO RL Simard,;NUMARC l' LB Marsh, NRC/NRR/MIB WT Russell, NRC (RI) TG Scarbrough, NRC (NRR/KEB) C Caloway, NUMARC W

I SAFETY ASSESSMENT ISOLATION FUNCTION OF MOVs FOR HPCI AND RCIC STEAM SUPPLY LINE AND RWCU WATER SUPPLY LINE GE NUCLEAR ENERGY FOR BWR OWNER'S GROUP l JULY 1990 L L CHI [ J. L LEONG l. G. L LEVY H. S. MEHTA L G. B. STRAMBACK APPROVED BY: b Y S' L S. J. S$ MANAGER l. BWR OWNERS GROUP PROGRAMS 1 l o l o l-s t

IMPORTANT NOTICE REGARDING CONTENTS OF THIS REPORT r PLEASE READ CAREFULLY The only undertakings of General Electric Company (GE) respecting information in this document are contained in the applicable contracts between GE and the BWR Owner's Group utilities as specified in GE Proposal 3551951, Rev. 3, accepted by the respective participating l utilities' Standing Purchase Order for the performance of the work L described herein, and nothing contained in this document shall be construed as changing those individual contracts. The use of this information except as defined by said contracts, or for any purpose other than that for which it is intended,is not authorized; and with respect to any such unauthorized use, neither GE or any of the contributors to this L document makes any representation or warranty, and assumes no liability as to the completeness, accuracy or usefulness of the information contained in this document. t i ,,.,..,4

1.0 Introduction On June 7,1990 the NRC, by letter to the BWR Owners' Group (BWROG), requested data concerning certain safety related BWR Motor Operated Valves (MOVs) capabilities. Data was requested for the primary containment isolation valves in the High Pressure Coolant Injection (HPCI) and Reactor Core Isolation Cooling (RCIC) steam supply lines, and the Reactor Water Clean Up (RWCU) suction lines. His request was the result of a BWROG and NRC May 24,1990 meeting. This meeting concerned the applicability of tlfe Idaho National Engineering Laboratory (INEL) test data performed to resolve Generic Issue 87. The NRC interpretation of this data is in Information Notice 90 40 'Results of NRC Sponsored Testing of Motor operated Valves' dated June 5,1990. The NRC interpretation of the test results appeared to indicate a 0.3 disk factor, normally used to calculate valve seating forces, is not conservative, ne calculated valve seating force is used to size the valve actuator and motor, and set the torque switch. Therefore, the actuator size or torque switch setting may be marginal or may not fully close the valve against postulated maximum design basis event Dow and differential pressure (dp). This safety significance assessment, requested by the BWROG, documents the adequate safety margin of BWR plants. It shows a si;;nificant safety concern does not exist, even if the HPCI, RCIC and RWCU isolation MOVs of concern may not have optimally sized or set - actuators for full closure under postulated maximum design basis event flow and dp - conditions. 2.0 Summarv e The isolation MOVs of concern were selected, sized, and set using good engineering judgement based on the state of the art at the time of purchase. On a plant specific basis, features were provided for early means of leak detection before a complete design basis pipe failure could occur. In addition, other systems which provide additional valve isolation capability are available. Materials were selected for low probability of pipe failure. In Service Testing in conformance with plant Technical Specifications is performed on the GBS90A35mp 1

jr piping and valves to confirm their suitability and readiness for service. Four of the six subject valves have been evaluated and tested based on IE Bulletin 85 03 [9]. Emergency Procedures Guidelines for other diverse plant systems provide means of rapidly redu:ing the MOV service conditions if a pipe break occurs. It is recognized that INEL testing has identiGed anomalous valve behavior in the test valves under their test conditions. The BWROG and utilities are following this testing and reviewing engineering data as it becomes available for plant application. Based on the data applicability to their plant and equipment capabilities, utility personnel are reviewing their MOVs to assure the valves will operate on demand under all possible conditions. This assessment employing a realistic integrated systems approach concludes existing BWR MOVs for HPCI, RCIC and RWCU systems supply line or suction line isolation have a very high probability of fullisolation under realistic conditions. In addition, HPCI and RCIC steam and the RWCU water supply line MOVs have demonstrated proper operation under l. conditions mimicing the likely demand event, a pipe leak. System isolation will occur before the postulated design basis event high flow dp condition. Based on this the presently installed and set equipment does not represent an undue risk to the health and safety of the I public. l In process utility actions responding to GL 8910 are proceeding with consideration of the INEL data to prioritize valves for review and testing. l Individual plant licensing documents (SARs) have established that pipe cracks produce leaks long before pipe failure would be expected. In addition, the NRC has accepted this l conclusion when approving the leak before break concept as a basis for pipe restraint l removalin Light Water Reactors. l 1 Leak detection equipment exists at all BWR plants to detect the small pipe leak condition and then to initiate system isolation. Smailleaks represent such a small quantity of fluid flow escaping from system piping that normal system flow parameters will not be noticeably changed. The system flow conditions during a smallleak will remain almost the same as the system normal standby and operational conditions. GBS90A35.wp 3 2-

These environmentally qualified MOVs, which perform the isolation function, have shown adequate operability for many years during normal, periodic, operational testing and inadvertent isolations. The most probable, realistic, safety (isolation) response required of these MOVs will be from a postulated pipe leak condition outside the containment. 'Ihe likelihood of a leak occurring in these lines is small. Even if a leak occurred it would be detected well before a high flow /dp condition develops. Substantial time exists for detection of such a pipe leak and completion of the isolation function by valve closure. The MOV isolation performance will be the same as already demonstrated by multiple isolations (both during periodic testing and inadvertent initiations) of these valves in most operating plants. L A realistic assessment of the consequences of a postulated design basis pipe break condition, or some intermediate pipe break condition, leads to the conclusion that there is adequate safety margin to protect the reactor core and isolate the system successfully. Any single ECCS pump is adequate to provide core cooling. Analysis has shown any single low pressure pump (i.e., RHR or core spray) has adequate capacity to overcome the inventory loss associated with the postulated failure in ene of the lines in question. 1-Additionally, the HPCI, RCIC and RWCU lines are equipped with two isolation valves. If either of these closes isolation is accomplished. Any action which reduces the differential L . pressure across either valve will allow system isolation. Some of these actions include i partial valve closure, depressurization through the postulated break and/or primary system depressurization as directed by the Emergency Procedure Guidelines (EPGs). L It is not expected HPCI/RCIC/RWCU system isolation MOVs will be challenged at high l flow design basis accident conditions because ofleak before break considerations. Leaks should be isolated early at low flow conditions due to the effective leak detection and isolation systems. There is a significant high probability of successful valve closure when realistic consideration of expected plant and system responses to postulated accident - conditions are used. Reactor coolant inventory losses can be made up even without GBS90A35.wp 3- ,y e = m

successful full valve closure for a postulated rupture in these lines, here is adequate safety margin in the ECCS to handle the losses. He ECCS are designed for a much larger break than these small line ruptures 10CFR100 off site dose limits are not expected to be exceeded even with a delayed isolation response for any of these three systems. 3.0 Safety Assessment - HPCI/RCIC/RWCU Pioe 12aks 3.1 Leakage Considerations It is industry experience that high energy pipes experience leaks long before a pipe break condition develops. Industry has referred to this phenomena as Leak Before Break (LBB). Most BWR plants have multiple channel, redundant leak detection monitoring of the high energy system lines external to the containment. This monitoring is sensitive to small leaks and causes both an alarm in the control room and at most plants automatic isolation signals to the leaking system's isolation MOVs. Isolation signals or operator action would initiate MOV closure long before the leakage could cause any significant Dow change, Duid loss or radiation release, and before a significant long term environmental challenge to the MOVs. The MOVs have been environmentally qualified to the more extreme Double Ended Guillotine Break (DEGB) environmental conditions. De MOVs are periodically inspected and tested to demonstrate operability during plant operation. In addition, these valves have occasionally been inadvertently closed during plant operation. This has demonstrated unscheduled dema~ d operability. n 3.2 - Lgak Before Break Justification Although the design basis for nuclear power plants, as discussed in the SAR, includes the - evaluation of a loss of coolant accident resulting from a postulated pipe break, considerable effort goes into designing piping and safe end systems to assure that such a break will not occur. Piping systems are analyzed using appropriate codes and standards, typically Section III of the ASME Code, to limit applied stresses, and materials are selected to provide adequate ductility and toughness. Piping design also provides implicit margins concerning fatigue initiation. Environmental effects are not considered significant. Piping materials 1 GBS90A35 wp..

(carbons steel in most cases) and steady state temperatures (less than 2500F in many cases) preclude environmentally assisted cracking. Thus, while cracking may be postulated, the probability is low.' Furthermore, leak detection systems are designed ts assure that, even if a pipe or safe end (nozzle pipe transition piece) should experience cracking, the crack would grow to a through wall leak and the leak would be detected well before it reaches critical crack size which could cause a pipe rupture in the long term. His concept is called the ' Leak Before Break' concept or approach. This critical crack basis already exists in most plant SARs as part of the plant design basis discussion. In more recent plants i h t typically covered in Chapter 5 of the SAR. In general terms, the LBB concept is based on the fact that reactor piping and safe ends are fabricated from tough ductile materials which can tolerate large through wall cracks without complete fracture under service loadings. By monitoring the leak rate from the through wall r teks and setting conservative limits on the leakage, cracks in piping can be l detected well before the margin to rupture is challenged. .In NUREG 1061, Volume 3 [1], the NRC Piping Review Committee outlined the limitations and general technical guidance on LBB analyses to justify mechanistically that breaks in high energy Guid system piping need not be postulated. In a recent modification to General Design Criterion 4 (2), the NRC has formalized the use of the LBB approach to ' justify the elimination of pipe whip restraints and jet impingement barriers as design requirements for a hypothetical DEGB in high energy reactor piping systems. nus there is NRC recognition the LBB concept provides added margin over and above the ASME Code piping design structural margins. A key parameter in the LBB evaluation is the critical crack length at which pipe rupture is predicted. The focus in the LBB evaluation is on the through wall circumferential cracks because such cracks could lead to a DEGB. A DEGB is one of the usual design basis event analysis assumptions. De LBB approach is not being applied in this assessment to eliminate pipe whip restraints or jet impingement barriers or reduce inspections. Derefore, explicit LBB margins are not b GBS90A35.wp 5

_ - - -. - -. ~ _ _ - - calculated nor are they necessary. Instead, the L3B concept is used in this muessment to demonstrate that the leakage from a through wall crack with a length up to but less than the critical crack length, would be large enough to be readily detected such that isolation actions can be taken well before the critical crar:4 length is achieved and long before max! mum design basis event flows and pressures are established. 3.3 Critient Crack Lenrth and Leak Rate Calculations Critical crack length ans leak rate calculations for typical BWR piping geometries have been documented in plant SARs. Reference 3 is an example of such calculations, ne calculations presented here use methods [4,5,6) more recent than used in the existing SAR calculations. Table 11ists the values of parameters used in the critical crack length and leak rate calculations. De results of the calculations for representative pipe sizes are summarized in Table 2. A limit load approach with a conservative value of flow stress equal to 2.4 Sm (where Sm is the value of material design stress intensity given in the ASME Code), was used in calculating the critical crack lengths. When based on test data, the flow stress for four inch diameter pipes was assumed to be 2.7 Sm. The leak rate calculation methods used for both the water and the steam lines are outlined in Reference 5. l An inspection of Table 2 shows that the calculated leak rate at critical crack length is, as cxpected, a strong function of pipe diarneter. Nevertheless, even for the 4 inch diameter water line, the predicted leak rate is 25 gpm ai close to the critical crack length. A 25 spm leak rate is larger than the leak detection rate sensitivity identified in the following section l on Leak Detection with the exception of the RWCU cold water lines. These calculations conservatively ignore leak rate increases due to steam cutting that can occur for a given crack length. Once leakage starts, due to steam cutting, it increases with time and the Table 2 leak rates can occur before reaching critical crack length. Full design basis MOV dp, corresponding to a DEGB, will not occur at these limits due to the down stream flow i restriction (crack). Thus complete MOV closure will occur under these conditions, ne RWCU cold lines have a much lower potential for cracking because of their constant cold condition and materials. GBS90A35.wp 6-

9 It is important to emphasize that the LBB margin increases with increasing pipe size. Dus, larger pipes where failure could be significant have inherent LBB advantages. Whue the LBB margin is somewhat lower for smauer pipes, there is stlU a large BWR experience database supporting the integrity of such piping. l Inspection programs (e.g., In Service Inspections (ISI) per ASME Section XI), other Generic Letter 88 01 [8] commitments and other periodic inspections on system piping outside the isolation valves provide additional assurance of continuing piping integrity and low probability of pipe leak and break conditions. Based on the results of th., and the following evaluation, it is concluded that the subject piping systems (HPCI, RCIC Steam Supply Line and RWCU Water Supply Line) are expected to develop a detectable leak long before reaching the point of incipient rupture. Thus, a DEGB in these lines is highly unlikely. 3.4 Leak Detection Monitoring and Isplah Most BWRs have been designed for compliance to General Design Criterion (GDC) 54 [7] . Piping.rystem pencreating containment. Piping systems penetrating primary reactor containment shau be provided with leak detection, isolation, and containment capabilities ..' This GDC was satisfied with a defense in depth combination of pipe break, high flow monitoring and isolation sensors for large leaks for each high energy piping system. These same high energy piping systems also have sensitive, smallleak, temperature monitoring and isolation sensors. At most plants the redundant, safety grade temperature monitoring equipment continuously monitors areas outside containment where high energy lines are routed. De temperature sensors for this monitoring are grouped with the piping of each system and will alarm and/or isolate that system when a leak condition is detected. At most plants the sensors and logic are applied in a redundant design configuration to be single failure tolerant. These temperature sensors can be configured in an ambient temperature and a differential temperature arrangerr Ont. De configuration is room dependent at each plant. GBS90A35.wp 7

) ne range of plant system area construction differences has resulted in alarm and isolation limits related to leaks typically from 5 spm to less than 25 spm. These isolation limits are converted to temperature values, and are expressed in terms of temperature in SAR Technical Specifications and other plant documentation. The temperature sensors sensitivity provides a fast response to a developing leak. Even though a temperature limit may relate to a specific leak rate, these same temperature limits can be attained with much lower leak rates. A smaller leak for a longer time period can reach the temperature limit too and allows recognition of smaller cracks. In addition to temperature monitoring in the RWCU system, most plants have cold water low Dow leakage monitoring capability. This cold water, small break, redundant, safety grade, differential Dow monitoring leak detection capability measures Dow in to and out of the system. It has an isolation limit ofless than 100 spm Dow mismatch between the system input and its outputs. It can quickly respond to a small break condition in the cold water portions of RWCU Typically this isolation limit would initiate MOV closure before any appreciable additional Dow could be developed. The RWCU heat exchangers dp drop will further limit any small break flow. This monitoring sensitivity has been inadvertently demonstrated numerous times during start up and realignment of the RWCU system. In addition to the temperature monitoring system and the differential flow monitoring (RWCU), the operator can detect smallleakage flow into the area or equipment room j drain Radwaste sumps. There are also area radiation monitoring system gamma detectors l that may alarm during smallleak conditions. Dese additionalleakage information sources i provide data to the operator which call for a visualinspection of the area. 1 Operating experience has shown relatively quick operator response to leaking conditions in safety systems and other monitored systems upon leak identification by routine inspection activities or by monitoring equipment isolations and alarms. The leak detection temperature monitoring capability installed in BWRs can detect the smallleakage condition and initiate isolation long before a pipe break condition would GBS90A35 wp 4-l e

develop. Derefore, the combination of the leak before break approach in conjunction with the leak detection capability provides early isolation at less than design basis conditions for a potential pipe break that might challenge the MOVs isolation capability at maximum flow induced dp. 3.5 Radiolorical Consequences of Leaksee Flow he radiological consequences of the leakage flow from the HPCI, RCIC or RWCU lines are bounded by the plant design basis radiological release. De BWR design basis event for offsite release is the DEGB of the main steam line, ne DEGB assumed in the evaluation of the offsite release results in a large amount of reactor inventory loss prior to break isolation, ne 11guld phase of the reactor inventory contains most of the radioactive material which is released into the secondary containment during the postulated break event. However, the resulting dose from the main steam line break is still only a fraction of the 10CFR100 limits. Furthermore, the total inventory loss for the smallleakage associated with the HPCI, RCIC or RWCU line is only a small fraction of that from a main steam line DEGB. For example, a 25 gpm hot water leak from RWCU typically can be detected within 10 seconds. nis means that the total inventory release before detection is less than 30 lbs. nis is a small fraction compared to the main steam line break 1: quid inventory loss which is l approximately 140,000 lbs total, of which 120,000 lbs is liquid. Derefore, even if the leak detection requires 4000 times longer to isolate the detected leak, the radiological release from the leakage flow will be a very small fraction of the 10CFR100 limit. l 3.6 Environmenta? Oualification Equipment Qualification (EO) of these MOVs has been performed to pipe break harsh environment envelope bounding conditions, which are much worse than smallleak environmental conditions. Satisfaction of EO requirements assures continued equipment safety function performance including MOVs up to these EQ bounding conditions. Therefore, no EQ concern exists for MOV isolation or the functioning of other safety systems equipment due to small pipe leaks. i GBS90A35.wp 9

= 3.7 Lgakane Flow and inadvertent Closure From leak before break considerations and with the capabuities of detection and isolation of a small leak, the leakage flow from a postulated leaking piping system would be small. Such smallleakage, when compared with normal or standby flow capabuities of the nstems, would not establish any appreciable dp across a closing isolation MOV until fully closed. Further, there have been some inadvertent isolations of these MOVs over the years at operating plants. Some of these isolations have occurred at or near 100% system Dow rates. 'Ihis demonstrates isolation capabuity well in excess of small pipe leak flow conditions. It should be further noted that as the HPCI/RCIC valves close they are subjected to the full reactor pressure, (dp of 1000 psi) across the valve seat. His dp v ul be equivalent to the isolation MOV end of stroke dp conditions for a DEGB. Derefore, in situ valve closure capability has been demonstrated. Successful RWCU isolations during normal full flow operation have occurred, which subjects the valves to full reactor pressure (dp of 1000 psi) across the valve seat. Therefore,in situ valve closure capabuity has been demonstrated. MOV isolation operability for small pipe leaks has been demonstrated for all three systems. 4.0 Safety Assessment. Desien Basis Pipe Break 4.1 Realistic Analysis Conditions i l An analytical assessment of a postulated design basis pipe break condition in one of the i L three BWR systems of concern can be looked at from a realistic perspective, just like the l postulated small leak condition. A realistic review, without all of the design ba:Is assumptions, was conducted because of the low probability (4 X 10-4/yr) of a high energy line break in cne of these systems. Any MOVs at BWRs which might be considered marginal or inadequate, when comparing their actuator size and deliverable stem force against expected required thrust, could stul be instrumental in achieving nstem isolation. l OBS90A35.wp 10-4

Some beneficial conclusions can be drawn from the system design, equipment design, and physical attributes of the systems and equipment. Dere are MOV design considerations which have been included during the design process which make MOV actuators more capable than their ratings state, ne actual Dow during a postulated leak would probably be closer to the 100% system flow rate rather than that attributable to the DEGB. This is because ductue pipe lines do not physicaUy guillotine rupture and there would be a Dow interference from the remaining l piping. Some plant valves have already demonstrated the ability to close under comparable, full flow conditions when inadvertent system initiation and isolations have occurred. There are two MOV isolation valves in series on each of these system supply lines. They are typicaDy mounted in the supply lines very close to one another, separated only by the containment wall. Upon receipt of isolation signals they will not close at exactly the same time, his is because of real world, small physical differences, as well as the fact that some are driven by AC motors whue others are driven by DC motors. Therefore, each valve may be subjected to different dp levels as they are closing. De possible alternate sharing of the break flow high pressure conditions and any cycling of this sharing between the two valves would probably allow at least one of the isolation va!ves to continue its closure motion until it becomes fully closed with the possibuity of the second valve fouowing thereafter. This possibility might better be described as a sharing or splitting of the high pressure condition between the valves. As the valves reach the end of stroke, they wiu be subjected to the fuli I dp condition. However, as discussed in Section 3.7, this is equivalent to the conditions that these valves would expe ience at the end of travel during inadvertent isolation. He control circuits for most MOVs contain limit switches for end of travel control, torque switches for valve seating (closing) control, and motor thermal overloads. Dese controls all l have the potential to stop actuator travel. In some plants the typical control arrangement has the limit switch bypassing the torque switch for 95% of the valve closure stroke. The torque switch controls only during the final 5% of the valve closure stroke. Thus full actuator torque capability is available until after valve orifice closure. In addition, many MOVs have the motor thermal overload bypassed except for testing. GBS90A35.wp 11-l l

s A full HPCI steam line break will reduce the reactor pressure. 'Iherefore, the resulting dp loads on the vah es will decrease with time during an outside containment line break event. Even if the isolation valves are not fully closed, the operator will be aware that the break has not been isolated due to the break detection system alarm in the control room. Control room operator response to the existing Emergency Procedure Guidelines will lead quickly to reactor scram and depressurization. Once initiated, reactor depressurization occurs in a few minutes. Reactor system depressurization through the break and through automatic or manual actions will reduce the dp on the valve, 'Ihis will allow time to isolate the line and ensure adequate core cooling. The combination of the above factors leads to the conclusion that Isolation MOVs will most likely respond to an intermediate pipe break condition or a design basis event with successfulisolation. 4.2 Nuclear System Imnact Assuming the high energy line break occurs, external to the containment,in one of the three systems, the impact on the nuclear system would be less severe than a Design Basis Accident (DBA). The high energy lines are smalllines (compared to the DBA) and would require less Emergency Core Cooling Systems (ECCS) flow for core cooling. Any one of i the low pressure injection pumps (Core Spray or Low Pressure Coolant Injection) would be sufficient to provide core cooling and handle the consequences of a postulated line break. Existing SAR analyses for the same line breaks inside the containment (which cannot be l isolated) show that there will not be any resulting core or fuel damage for the smaller line break events. ECCS components have spatial separation such that the impact of the postulated high energy line break should affect only one division of equipment. The remaining division will be more than sufficient to handle even the maximum line break considered in this analysis (as opposed to a more likely smallleak in the line). I i GBS90A35.wp

o . o e nerefore, BWR plants have adequate safety margin to protect the reactor core and provide adequate leak detection and isolation capability using the presently designed isolation MOVs and other mitigating measures. 4.3 Offsite Dose Release Imonet De radiological release from the DEGB of the HPCI and RCIC steam line is bounded by that of the main steam line break. These smaller lines do not depressurize the reactor vessel as fast as the main steam line. The reactor inventory release for these breaks is mostly steam. De dose from steam loss through an outside line break is small. Herefore, l the offsite release from the HPCI and RCIC steam line break will still meet requirements of 10CFR100. The reactor inventory loss from the DEGB of the RWCU line will be mostly Uguld. However, the radiologial consequences of the RWCU line is bounded by that of the main steam line, based on the assumed Valve closure times for the RWCU isolation valves. The radiological release from the main steam line la only a small fraction of that of 10CFR100. Therefore, any slightly longer valve stroke time for the RWCU isolation valves will not result in noncompliance with the requirements of 10CFR100, 5.0 conclusics Because of the leak before break considerations for the HPCI/RCIC/RWCU piping, it is not expected that system isolation MOVs would ever be challenged at high flow design basis accident conditions. With the effective isolation systems, leaks should be isolated early at low flow conditions. Additionally, realistic consideration of expected plant and system response to postulated accident conditions leads to the conclusion that there is a significantly high probability of successful valve closure. Even without successful full valve closure for a postulated rupture in these lines, there is adequate safety margin in the ECCS L to handle the reactor coolant inventory losses. The ECCS are designed for a much larger break than these smallline ruptures. Delayed isolation response for these three systems is expected to keep offsite dose releases within 10CFR100 requirements. l GBS90A35.wp 13 i I

o 6.0 References [1] Report of the U.S. Nuclear Regulatory Commission Piping Review Committee, NUREG 1061, Volumes 1 through 5,1984. [2] Federal kegister, Volume 52, p. 41288, final rule modifying General Design Criterion 4 in 10 CFR Part 50, Appendix A. [3] GESSAR II,238 Noclear Island, Section 5.2.5, GE Document No. 22A7007, Rev. O. [4] S. Ranganath and H. S. Mehta,' Engineering Methods for the Assessment of Ductile Fracture Margin in Nuclear Power Plant Piping.* ASTM STP 803,1983, pp. II.309 to II330. [5] A. Zahoor, R.M. Gamble, H.S. Mehta, S. Yukawa and S. Ranganath, ' Evaluation of Flaws in Carbon Steel Piping: Appendixes A and B,' EPRI Report No. NP-4824SP, i October 1986. l [6] Mehta, H.S.,' Determination of Crack Leakage Rates in BWRs,' Attachment 2 in Letter dated April 22,1985, from Jack Fox, Chairman, ANS 58.2 Working Group to K. Wichman of NRC. [7] 10 Code of Federal Regulations 50 Appendix A General Design Criteria l [8] NRC Generic Letter 88 01 NRC Position on IGSCC in BWR Austenitic Stainless Steel Piping, dated January 25,1988. 1 [9] NRC IE Bulletin 85 03 Motor Operated Valve Common Mode Failures During Plant Transients Due to improper Switch Settings, dated November 15,1985 GBS90A35.wp 14-

O

• O e

i TABLE 1 I VALUES OF PARAMETERS USED IN CRITICAL CRACK LENODi AND LEAK RATE CALCULATIONS i s Pipe 7hickness Schedule 80 Pipe Internal Pressure 105g psi Temperature 528 F Normal Operation Bending Stresses 4ksi Material Stainless Steel or Carbon Steel TABLE 2 P CRITICAL CRACK LENGTHS AND LEAK RATES FOR VARIOUS DIAMETER PIPES Pipe Diameter Critical Crack Leak Rate at Critical (in.) Length (in.) Crack Length (gpm) Water Steam l l I 4 7.1 25 15 l 6 9.8 41 27 12 183 166 108 16 23.1 262 170 t l l GBS90A35.wp 15-l l l

ENCLOSURE 5 i LIST OF ATTENDEES FOR PUBLIC MEETING WITH THE BWR OWNERS GROUP AND THE NRC SEPTEMBER 7, 1990 i T.G. Scarbrough NRC/NRR/EMEB J.E. Richardson NRC/NRR/DET L.B. Marsh NRC/NRR/EMEB C. McCracken NRC/NRR/SPLB E.J. Sullivan NRC/NRR/EMEB R.J. Kiessel NRC/NRR/OGCB O. Rothberg RES/DSIR/EIB G. Thomas NRC/NRR/SRXB S. Diab NRC/NRR/RAB F.T. Grubelich NRC/NRR/EMEB G.H. Weidenhamer NRC/RES/EMEB W.S. Farmer NRC/RES/EMEB A.T. Gody, Jr. NRC/NRR/DRSP Denver Atwood GPCo/MOV Committee Chairman Ralph Donges PSE&G Stephen D. Floyd CP&L/BWROG - RRG Chairman George Beck PECo/BWROG Chairman Bob Binz PSE&G/BWROG Vice-Chairman J.T. Beckham, Jr. GPC/Vice President - Hatch l r l 9 L l i l )

f BWROG MOV STATUS l 1 SEPTEMBER 7, 1990 l. l 1

BWROG/RRG M0V STATUS 0 BWROG PRESSING FORWARD WITH ACTIONS ASSIGNED UTILITY EXECUTIVE SPONSOR REQUESTED UTILITIES TO BEGIN PLANT-UNIQUE SAFETY ASSESSMENTS (REVISED TO ADDRESS EQ + FLOODING) NOTIFIED UTILITIES OF MOV ISSUE AND REQUESTED THEIR RE-REVIEW OF EPGs IN RESPONSE TO UNISOLATABLE LINE BREAK SCENERIO SCHEDULED MEETING WITH MOV COMMITTEE TO REVIEW RWCU DESIGN BASIS AND EMPHASIZE NEED TO ACCELERATE RWCU VALVES IN GL 89-10 PROGRAM PARTICIPATED IN INEL VALVE INSPECTION l --~,---------n,-- .v,----,------,

i BROG/RRG MOV STATUS (CONT'D) ) 0 INEL DATA INDICATES INDEPTH REVIEW OF CURRENT SAFETY-RELATED MOVs/ METHODOLOGY IS APPROPRIATE 0 INEL DATA NOT APPROPRIATE FOR REDEFINING MOV OPERABILITY BASES AT THIS TIME i 2 LIMITED DATA i SMALL VALVE SAMPLE MAY NOT BE REPRESENTATIVE OF INDUSTRY POPULATION EVALUATIONS NOT COMPLETE l 0 INEL TEST REPORT NOT FINALIZED l-o VALVE INSPECTION REPORT (EPRI) IN PROGRESS u o INEL REPORT LINKING INSPECTION DATA L TO TEST DATA IN PROGRESS l .)

BWROG/RRG MOV STATUS (CONT'D) 0 BWROG WILL ACT RESPONSIBLY ON DATA PRESENTLY BEING FINALIZED INEL NUREG CR-5558 REPORT DUE 12/90 INEL REPORT EXPECTED 3/91 ON MODELING OF VALVE INSPECTION DATA WITH INEL TEST i RESULTS EPRI ASSESSMENT OF INEL TEST RESULTS - 9/90 EPRI REPORT ON VALVE INSPECTION - 9/90 L O BWROG REVIEWING ADDITIONAL TESTING OPTIONS TO COLLECT MORE DATA ANCHOR DARLING BLOWDOWN TESTS EPRI MOV PERFORMANCE PREDICTION TEST PROGRAM BWROG-SPONSORED TESTING l l ADDITIONAL ANALYSIS OF t AVAILABLE DATA NECESSARY-MORE TESTING MAY BE NECESSARY

I % O4 O PROPOSED FUTURE ACTIONS 0 VERIFY APPLICABILITY OF TESTS USING EPRI AND VALVE SUPPLIER INPUT - 10/90 0 REVIEW EPRI DATA ASSESSMENT - 10/90 REVIEW INEL DATA ASSESSMENT - 1/91 0 DEVELOP BASIS FOR REASONABLE ASSURANCE OF OPERABILITY (USING INEL MODELING REPORT AND A/D TEST RESULTS) - 6/91 j}}