ML20127N562

Nonproprietary Sequoyah Units 1 & 2 W* Tube Plugging Criteria for SG Tubesheet Region of Wextex Expansions

ML20127N562
Person / Time
Site:	Sequoyah
Issue date:	11/30/1992
From:	Wootten M WESTINGHOUSE ELECTRIC COMPANY, DIV OF CBS CORP.
To:
Shared Package
ML19310D500	List: ML20127N536 ML20127N553 ML20127N562
References
WCAP-13533, WCAP-13533-R01, WCAP-13533-R1, NUDOCS 9212010311
Download: ML20127N562 (160)

Text

,

., . \ ,

.D "

v -

.. .. . . , . ., . . .M. . : - . . * .

. " _, '.. s.' .. .. ; ,. ,.

, . l- ., .l; .

. . :; . 1

.; <.. '.: . . ..r '. -,. .

• - . 5

..; . s

.. ~

.- .~ . . . . ..#

. , . , . . . ' .;  %.'+. ", . . ' . . , .. . .

' ' c. .. . . .

. . , i ,

,> .' . .' .f , -

,e , . s q: -

.  ; *. ..,. ~

• a -

.. y . . o;...~.o....... ,

a..

e *

...t '.,- - ~ -

..'.,p,y+s

'% -' . .$, y. .:-)m.

.-..*  : ; ' .....s. . .. - . ' * .

..;.x . .n..a., f.:*. . . . ' .'.9 ,*

' . i ~ . . . . .

' . . s.  :

s

.. .,..... .p- .._

.e..,4,.,.-.*

, , , e.1 .

.g # , . , ,, y, , , . ,..y

e .. .

- rW '. . .. ;, - l . .

... . L

- . .: .: . s; 3 ,

. . .', *: ., .6y - ,

g* h,. .i

s-

.. . g

~

.y

,F._ afn_ .. ._.

.'..\*.

',.s- ..

'. - .- - y y* *. .* -

,3;.:: . .', ' : , ,. ; k ', , .'.. .,

,: V . ; ; .M' .

.;;. .k ; :. .  :. . , .

. . a.

a - .; .

.'s.

...: = :: ~ . v . . .  :

' '.

. : ' .%.;. . s. :. . ... .' . , , - . .. .L

,W.:. . . .,e

- - - -; .:,j? M.;..ML%.. pp:ia'.'b.l: ,'.<rbc .!. . ; M . .-

r. -

• re. . :

t~ L . ::

- . . :. . i . ' a; . ,-

, s

. 3: we a$r .. . _ ' . .- y*.:._. ,; . .:....- s . .; .-' . . . sv '

' , ^ - .. .

.'..l :ll e r k % ; 3 .,, : . -; , , q. , , . ;. *l?:l':.. ._ , y. . .;[.J .:. :.

4 . :

c ' ."p.N -,.;; ; ,, _t  :

_. 2_. . . p . ::

q. v;. . n. , ..; .,;J . -- . ;y ' p. . -

,L p g., * .,; % s' , . , .

. ;s -- ..e. : - -..? .e. . . ...

.e. j

q. .

., ' * + 8 . +.f .l'...  ; ,* , . . . . .

. .:.-*- . . i....-. . : , ; ..' . s .

e. ;-).- .,:c

. ' ' . ^ ;*. g ..

<. q... t

.:.. : . *f *

. , y.g. ,, . = , ; .v a.

v . -

- l; . ,: ., is.*+

a,

. . ;4 . .

.. ,. *. . . - .,L.',.'.,.*-

........4.,_y

+ .. ,c  ;,,, ..

-e *s % .,. , y . :. - ..

. . ,. ?. ..a...

.,.g,,,. , . ,.. .; . . , . .. ...- - .

..,......e..,,7 *.,.t .

• ;v . p- \ l j; *s . . . , ,+_ .., ..-

w , ,

g ..., - . . ,. i. j; . . c .> ..

.g g. s. .. z.<

y , ' .**, . ....:,.%... . . r . h. . ..

'.'..,:. ~ y

?.  ? , p ;s  ; ' .y% , ... '

C '..2 . y : .rf. ,q - l j._. .

. f. *

h h k '.Thl hf '. . .

f. ; A . . *

. [ . ,h. ;s . ..Wf. .

'.;.{. . ' .

i

.. .S

-c ,

..-l~.

. r . i q.M .d

[A*, y.;-al}:s. f; '. v? , ni ::,  ! . ,,. .; .-': ..

.: 4

^

.s, u

• ,. & , g,' . '; . . .

., l -

. . . , .s ;3., ,- ..u ,..t*. .n. , ,. .,. 9 . .'. s' . .. , ,; '- . ( < g..,.,j.... . - . . , 9. r. . :. . . . . - .,

_f%..\g.4V.u- .. .. ,--f.* *y .e.._.. ....g . ., f,: ,

^ ,

. J,- .

..?;

• f ,4....

. ' ~ ',; y[ e* & , . . :;. ,: gM , .?.,1c % Cp q l' ? w' .: *-

n ]).)G b,.,.4..  :..s; .l:',. : Q yY? l.,,x..is Q .i ;: '.:s af.,.Q - e :s

,'tV.'.- ' : .

f.,. %

,..,8 a

J g, ., .,, g.

c..* . .?. :',;js ,3 Wfy.

-3 ,

-' .. . ;j;

• ,5 -

- ;-e ,d , t 9 L 8. ! . . , . . .

i

a. f: .;

; t .. s't g*', , .- G. h*.h Wg 3 ,4

'.( p'..'hA k jQ . @-N',t

1. 6 , y.'; j,a f .9.: t

l l.,iQg' , .._ .g,.*p, .,4 ^7 ,. t '..* 7'. d,,.j 1,..,h%

lf',7..'.

- *' s ? *,W . .;; . a

, p . ,

j g, S,t . 2l,.d. ,.....p'*

, ' .e 4 7 ' g:o.e . .:< . ** ,:; - ..vs

.r,c...

. ; e .3 . . ,.,l9...-

y , 9..n.~Aj.; . . 4 ,. ,[.

R . : . .;;};. - QR. *."f.,. f, . . ; . . . . . , -

c q:f.f. ai 4 ,.~;

. .; . .n  : . ,. v  :;y

.!,.s,- s a:. .f.:. .. ., 'n.f. m y,:' .w, e s -

t; J* J. e.h. , 4. . %~,) ; .q .3. . . . ; J . . . . . . .

s  ! .<. .y

. ..-.f.f.;O - ,py.;Q:. < ,e, M4;..o

' ' . . ;. s . . *- ; r- thy p:.

• ys  ?. . ..; ,y

..m .

e '.y ., < : ; . 3. . - ,

- 3~ *

. Q' y.U . * '.  ; .%

g .:. ; ,', , y,.J.>.:b. ,.

y. ..* . l %.. a, .T. .A 3...q/j y $ . M. : .c& :: .q. .* : :

,. : u f . ,e .

..t -

.w* c; C -

_s . ,.?.q,._s y . t .

.:j p : .,, y . - f. y ,.. _. j?%gp . :.. :y:h

,?_,, . ,._ ,_

e," i P.,' .} '

,1-.

d'* . ; '...; e, 6.... 'M'O ,% dI'

• ,..f-

.,- - - . _ , , t'. , *.r' ', .:;,s_.

.p' . , q:6,

._ ,'-  :. . .s . :: .* du Q 'r*_,js:e M < - (;' &g . , .

.g. ,q, , ,y, r ; . } : '; & s ,.;';q , ._f:,,c .:: ... .. ', . -.s . . , , -

.ft:h, ' '.w;f.- Q .w Y

.', . ;'_ _ v.f.'. . ); } .f:L . ,9f.;},&$j y.[y&,f..ony,

.L

,.:[ .. : %; o., ; ;: . v,Mq.3. m ,"*,*.  ;,. V .'q.,,y ,s .4. 'ig;;.g; ,F.s .c C v Veqpg. ..c

.h

....s e~ :,*r Q w . . s . . 9. ?. . y-

-b

~ : :p.,sv./ . .' . : * ;.Y. o; d;d. n.-

.1 ; . .. ;;- ,y ' , ,,. l-

. .". @. ,M 4.Q , '4#

.; .; . , ; .. . ,f c .. . . J. .fg . % .

g .

'4,..,.*

, :N.

-,?

f. :, , . . w l.;,-:l '.f Y [i f',[. *cf, ' Ll( {' . .;: '..M.,' . G ,y~.

, -: Q:  ;;."'f.y Rli',f ? [:[: j <.'. .+. ..; ,_\, ._C. - s i . ;. -

_ . ;_ i . .:. ; ' x

.q g. .. . .:..:,  ;

. {.j . . .. ..

. . f...( -

. . . . . . . . . . . . . . s s.

, . , , , ,,y.. . . . - - w- .,

., b. .c". ,, :,. ;* .< ,. ; * .

.- .; 4 , .

- p . . ' ..;.' .3,y . \, ' : . ,' . .

A n '.'.'

. , ,, :~, ..*

\ 1. ,,s. 'k e

.' . ' . . ~. ? . . ;. q',,-.,'.c , . . ._ . . . . . ,.N,'~."., _,,,...

• b, . \ . ' .

l.,

q,,.J ;<, . . . ,,1 , , . -

. . ; ., 7. ' ...e

..; 7,. y . ,,, .,.,, ..

. . ;'....,/ : . y '. . ... .,..,..i.'.7,y.,.....a ~ , ,' .../-...(

<?i : ,,.,, .g. .. -l i O,' -

' er .' .;..-. 4. . . ,. . ' J .{  ; , . . :.

. - .. g . , . .-

p ..4.,. -.',"....".41..- .,s. ,..g.

. . . ; 1 y ._, ,..z .. ,, -

~.' .:, . ;

..e

d w . .. . .,,-.; .

. * ~ .

> . ;. .-2.;. . .?a i,..,.

. . . .- ~ . , . - ., s . ...; ...

. , .;.v.

.; . ..;  : .. . ..* ....g-

..;r...,'..'- .' ' ,,:., ' . ( ^ . f . ( ... .

c . 4 . . . , ; .' - .. ) {. . . .(i.,;i. ;* r

, .. .g . . .. . :. , .) .; '. . 7' - - v.. '

. ~3f..~ , . , , ; :"*.-;c ,.

. c. . .

e. .. .- . . -

. ,< - . . , ..s

.g -.

a '- . . .g

' + ' f. -.,*. . .: ;' ' . . .

.' ~,$...'

~ - +. .. .. . . ';.

,s.: ;I ' ; i ' ..

l- ' .e

• e .. . . .

. .' i . * , -;; .. ;.  ? .. ..; . _ ' _ , p- .\ .'

. - : . , .'..,-l,.,  ?. .. \ (.,..

., "<- ., . . . , ,.,.._.tv , .i .i c.:. . , , - - ., ,. . ..

..ig._ *

.. s-

).

' . '. , . - w v.. ,'

f . s' 7 . ,[ . ' ', ,' ;,. b ' '.

's,

• 8' >. ***[,'~'

h. _.. * . ' . j%[ ' S ~. . -

, .-'l,- -* ' $ .,

-ll .'- **%,

..*,'.;i,3

.. ' . - ~

.s

L; . ., .. \ -

o'

• e

[ . % ' *(..*9 * .5 *'(' .[ # ,' - M.'. ,-. h *. l - . ' ,' .' ,-

...y.  :

' f

, , . . . .t . , . :

.a , . -

., .- .r . .. . . ',,.i. ,- k.. f, ,

..s.

. . .. . . .. - . :.r c.

. ......- l . , . , . . , -..p. ,,

. , .... v. -.1.'.. ....'.5.'c.~ ' .' .

.;'.....a -

.". ,Q ' ., *, .f.. . .J ., -* . _- . . ~ . '4 , .. ' ;; .}* ' , .* .. .-

7

. . ' . ' * . . . . . , ' . . ' . .'.s

.. . ., 4 . . ..

g : l j,;.p. J . -

' . t

. +-.

n, g . . .

. . , v .z. ~*.:_n'_-; -. f : :. .% . - . * '

.t, ',- (.,'.,  : ...  ; . .

.;. p , 6 ; ;. ,.a .' ,--l t. ., ' p . /.9$f ); . . .

.(..s-D,....'.." 1, ,. '. J. , ' 8 ..

, 7 -

. :- .6 * - . . '  ;  ; -

. , ' , + ;'; s . '.*e[.

..d-

$ . ~ , . . ! , ' ', ..c.!, ' . . /'. k. .g *.f',-

'.q $. , * / . .., .;~.4

. s,V ..',k

4

,.^ .. ..~; * ' ' : . ..,,..9';..,

3,.'

4

. -l 2.-, ,*

- . . .. . . ,!.s.- \ p? c: r ',s,. * ....,.:._.  ; : >

. :_ . , . ;i , ..

-f :> -

.. . * .-.'...,;,. p .. . . . . ... - 'e f ,.8s .-..-

a

$g; * . .,,4.

,....,.if,...*,.,...",f."g

., p: - ,. .. ' '. . ,

e... *..; '.,-)3 ', . ... .'.',7 ',5..

V #l ,

. g .'. ,/.- .. .'. .

.i.. ,.[. ,*. .. ; .,. *- - ' w5 t,. ,/.

.' .* Q;:

,,e

?

c .' ,5 ,. . . , . , . .

g

. .; . .9 . -, . $. l ..

3

. c.. ,-

_c . . , . , 4

. ' ,-. .... '*;' r. . / ';, y..

.*s-.',.. ., ,, ,

..s,. a'6 *...t,.

. '.*,.,,,. ..t.. .

i , - ' ..; ... ...

" N, - , !* '

, . , . . . . 4 ; - . .

• 3.f, . .

i,...,.- -

. . . , . '. 9 . 'g . -

' . 3.'.

- : ..i :v .'

' ' *

.?. !'.;: :. .;. .';. ' t '.;Q- .; 5 :' ,.; fs ,f. 'q;* _.[ l. ,:*., _' ,'. - -, . ' ., . ' . u-. ' . . ,' y. .: ',~ " r .,

.';,; c. . .. *.. . ...' . ...

..r,-

. . q. .

,.T',... ' . , .; . .....? 1 1 c.- c t ...l.t [ lf y. . . ;. .,; Q ' . %3' ..f : T R. y? ' . .. :...[%,. ).~.'..*, -_,.j,

b, .L > ,'rly %,:y;- . ' . 'l ~ ' '

- . , . .... .. m ; ' '.

._..- .._. . g c~ ,, f.',% {yl .* .', R,  : . ~.

D. _.w . _ .Q,, '.G. &.C > &., Y. g' .A R,&k ..*.* &.S., ' "'Wv; ] ~ .:l  ;~.., *,f W'. f,.: .g.':

' .~

d. N,l.: N'  ::% q: ( .' ' ' .2'(. l.T.' N.

'\

':W  ; <. '.. - l ,, . . r :

.Y. .- .s  :. -

.n.<g.j p,w?

, y s ;.-l,:% y v.c; Q  ;.

ri  ;  :

f.:1 '.g. ' u +:".,:- .. .. ..., u:, ': ._"' - . - : ..-

w.. : .:..".....v..-

- ,0.. s  : v.. .s u-

..?:

.. ., .A ~ ,;m... +,- < a,a. ~r

...?-. z -; ". ..-..: ,. ~ .: a . . . -:

u. s e :

v s

~.u. .*..:.,!.'..

r :;s ' -

.x.ce g u. %. % . ::.? &o r..

s ;: . t..  ; ...-

4. . . ;%.q% :.g: f:,9. .:,/4,.,,.

'. x y ~::y .....

.. . . :, -  ; :p A . < ' . . '; ~ g y ;* i.:* . . . L - - . a  %-  : ^

.r,y; ; ;g:. . . n.. . ;; : D. ., . . v, n.j y . . ... p'. f.viv.m, . '.-. x

- . v._ ;

. .;. m.~. 's." .: .

~

- . . . . J' .,::

. , s ;?- .*o3.g% .%. . . ;..F. ...,M.t-ig  ?* . , . . .i . 7 ".<*. v .-4**

- h. m J

-c- .l e. ;, w . W;,,i.. ,: . w ..

f a sl ..!....-

.. ** N .' . f. s - ; *. -

'% .  : .. .

• 9. l auh, ;. .'. ,c. ...

4 . . %' ..

) ,, A, 3,:,. )s w - .

?...,,...... - . . 't t - *6

.a . . . ' . . ' . . . .. ..

.; w,,.a.g; n a ; . , ... ,.:;.: 3. x. . s'.  !

s

... v..n. - r:; .; ra. . ..-.:. ;.;y: :.; .. 4

. w 3 b.r

,. 7 ,. . s : . . :.. .y.. . . .a-f bhh , p.:,.'

hh,5,Y [;.wg o.m ... .. .kh . , .h,jh. .

• y.,. . . . t

m

. . .. m. . . ,:.? ...  :

e[. .._h -..b . .. - -+ . . .

h. ,h_h.. g..

m' c,.h:._

.k_k h,vh 4 .. .~s ., - . * ' . .

9212010311 921118 M 2,. . ..

"?.

b PDR ADOCK 050 27 '%. . .i.; .%Y.7ar, :U .l-A.7.* ' +*t%n r.f. M.C;'. & % fr.  ?:, Wilv . : .1 :9..N . P a. 4 i . G' ', . .V 4:4- - B SJ ~L_ . W . f > :' " l ' D ':.  ::~: /'. 3 ?.' J. ' ' ' 'n e : ' M YT:: ' ' ':%. . ' 3'T*-." ..? -

.. : ~ ' -

WESTINGHOUSE CLASS 3 WCAP-13533 SG-92-10-023 Rev.1 SEQUOYAH UNITS 1 AND 2 W* TUBE PLUGGING CRITERIA FOR SG TUBESHEET REGION OF WEXTEX EXPANSIONS NOVEMBER,1992 Approved by: -

\

M. J. Wootten, Manager Steam Generah Technology & Engineering x

WESTINGHOUSE ELECTRIC LTPRPORATION NUCLEAR SERVICE DIVISION P. O. BOX 355 PITTSBURGH, PENNSYLVANIA 15230 C 1992 Westint - Electric Corporttie.

All userved 1

I SEQUOYAll UNITS 1 & 2 W* TUBE PLUGGING CRITERIA FOR SG TUDESilEET REGION OF WEXTEX EXPANSIONS ABSTRACT j l

Altemate steam generator tube plugging criteria wem developed to reduce the need to repair or plug Sequoyah Unit ! & 2 Series 51 steam generatcr tubes having indications in the WEXTEX explosive expansion region below the top of the tubesheet and below the bottom of the WEX" REX l transition which exceed the current Technical Specification depth based plugging limt t .1hese indleations meet RO 1.121 criteria for tube integrity, and indications left in service would have an 1 aggregate leakage below allowable limits for radiation exposure during a postulated SLB accident.

The alternate tube plugging criteria, or W* criteria, are patterned after L* alternate plugging criteria for roll expanded tubesheet joints and are based on maintaining structural and leakage integdty of tubes rtturned to service with indications in the WEXTEX rrgion. 'Ihe W* criteria for Sequoyah 1

& 2 are based on plant specific conditions. W* plugging criteria related to the structuralintegrity of the tube have been defined for two zones of the tubesheet. Tables have been provided to determine the leak rate contribution of cracks as a funedon of radial distance from the centerline of the tubesheet, position (depth) fmm the top of the tubnheet, and crack length, both for normal and faulted conditions.

t

, , - - a w , - , - - . - - . . - , , - , -..,v..

l i

TABLE OF CONTENTS PAGE 1.0 B A C K G R O U N D ........................................................................... 1.I 2.0 S U MMARY AND CON CLU S !ON S . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 1 2.1 OVERALL S UMMARY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 1 1 2.2 SUMM ARY AND CONCLUSIONS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 1 ,

3.0 T"CllNICAL APPRO A Ci l . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 1 3.1 R EQUIREM ENTS FOR W* CRITERIA....................................... 3 1 3.1.1 R e gulat ory Req ulte me n t s.. ...... . ... ... ... ... .... ... ... ....... ..... . . . 3- 1 ,

3.1.2 Requirements for W* Tube Plugging Criteria ...................... 3 3 3.2 LOADING CONDITIONS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 6 3.2.1 Normal Operatin g Loads . . . . . . . . . . . . . . . . . . . . . . . . . , . . . . . . . . . . . . . . . . . . . . 3 7 3.2.2 Other Translen Co n di ti on s . . . . .. . .. . . . . ... .. . . . . . ... . . . . . .. . . .. . . . .. . 3 7 3.2.3 Feedline B reak . . . . . . . . . . . . . . . . , . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 7 3 . 2.4 LOCA . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 8 3.2.5 Other Faulted Load Considerations . . ... ... . .. ... .... .. ... ..... .. .. 3 8 l 4.0 PULLOUT LOAD REACTION LENGTil . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . 4 1 4.1 EVALUATION METi lODS A ND LOADS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4- 1 4 . 2 TES TI N G . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 1 4.2.1 Test Samples.............................................................41 4.2.2 Te st De scription ' and Re suits........................................... 4 2 4.3 ANALYSIS OF TUBESliFET DEFLECTION EFFECTS ................... 4 7 4.3.1 Results for Normal and Faulted Conditions......................... 4 11 4.4 CALCULATI ON OF PULLOUT LE NGTil . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-27 i

4.4.1 Operating Contact Pre s sure s . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 27 -

4.4.2 WEX"I'EX Radial Contact Pressure and Friction Coefficient ..... 4 32 4.4.3 W

• Pullout Le ngth . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 34 4.5 ZONES FOR W

• LENGTI I . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 4 3 t

I i

- ~ _ . ~ . - _ . .

TABLE OF CONTENTS (CONT"D)

PAGE 5.0 DEG R ADED TU DE STRENGTII . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 1 5.1 EVALU ATION MED IODS AND LOADS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 1 5.1.1 Normal Ope ration Lnads . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 3 5.1.2 Feedli ne B re ak . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 3 5.1.3 Lo s 5 o f Coohn t A e cide n t. ... .. . .......... . ... .. .. . .. . .. .. ... . .. . .. . ... 5 3 5.2 AN ALYSIS OF DEGRADED TU BE STRENGTi!........................... 5 3 5.2.1 S ymme td c Case s . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 4 5.2.2 N on symme tric Ca se s . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 4 5.3 LIMITATIONS ON ALLOWABLE DEGRADATION ....................... 5 15 5.3.1 Circumfe m ntial Crack s . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 15 5.3.2 A x i al Cr ac k s . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 15 5.4 W* ADJUSTMENTS FOR TUDE DEGRADATION ... .. . . . . ... .... ....... 5 16 6.0 L E A K R A TE E V A LU ATI O N . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . 6 1 6.1 EVALU ATION METi lODS AND IDADS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6 1 6.1.1 Nonnal Ope ration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6 2 b.1.2. Faulted Condition - Feedline Break . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6 2 6.1. 3 LOC A . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6 2 62 WEXTEX EXPANSlON LEAK RATE TESTS............................... 6-2 6.2.1 S a mple and Te s t De s cri pt10n........................................... 6 2 6 2. 2 Te s t De s cri p tio n . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6- 3

6. 2. 3 Te s t R e s ult s . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-7 6.3 CONSTRAINED CRACK LEAK RATE TESTS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6 14 6.3.1 Te st De sign and De sed ption . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6- 14 6.3.2 S ample Pmparation and Te st A sse mbly . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6- 14 6.3.3 Te s t R e s ult s a nd Co nelu siens.................... ..................... 6- 17 6.4 LEAK RATE AN ALYSIS M ODEL . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-22 6.4.1 D ENTFLO M odel De scription . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6- 22 6.4.2 Leakage Rate Model for WEXTEX Tube s . . . . . . . . . . . . . . . . . . . . . . . . . . 6-23 6.5 AN ALYSIS MODEL FOR SLB LEAK RATE .. . . . . . . . . . . . . . .. . . . .. . . . .. . . . . . 6-3 8 6.5.1 S LB leaka ge B elow W * . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-39 6.5.2 Total S LB Leaka ge for WEXTEX Tube s . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-40 u

TA13LE OF CONTENTS (CONTD)

PAGE 7.0 S EQUOY A1i K 1 AND #2 1NSPECflON RESU LTS.................................. 7 1 7.1 S EQUOY Al l # 1 IN S PECrlON S . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7 1 7.2 S EQUOY Al i #2 I N S PECrlON S . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7 8 7.3

SUMMARY

OF INDICATIONS FROM 011IEF PLANTS ................ 7 8 7.4 ND E U N CERTA I NTI ES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7 8 7.5 CR A C K G R OWTl i RATES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7 13 8.0 W

• T U B E P LU G G I N O C R ITEl? I A .. . .. . . . .. . . . .... .. . . . . . .. . . . . . . . . . . . . . . . .. . . . . . . . . . . 8 1 8.1 G ENERAL APPRO AC11 'ID W

• CRITERI A . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8 1 8 . 2 W

• I E N G TI 1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8 2 8.3 APPLICATION OF GROWTil AND NDE UNCERTAINTY .............. 8 5 8.4 ALIDWAllLE tulle DEGRADATION IN W* LENGTH ................. 8-7 8.5 S LB LEA K RATE EV ALU ATIO N . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8 8 8.6 INS PECTION REQU1R EM ENTS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8 10 8.7

SUMMARY

OF W* TUl3E PLUGGING CRITERIA ....................... 811 9.0 REl'E R E N C ES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9 1 ill 1

- - - -- -____ __________________________U

T r

LIST OF FIGURES FIGURE TflLE PAGE Figure 1.0-1 Regions in the WEXT11X FuL Depth Tube to-Tubesheet Expansi0n.................................................................12 Fir,ure 4.2 1 As Fabricated WEXTEX Samp'e for Pull Tests ...................... 4 4 Figure 4.2 2 Pull Force Sample Configurations Tested .... . ................... .. ... 4 5 Figure 4.31 Tubesheet/ Channel flerd/ Lower Shell Finite Element Model ....... 4 8 Figure 4.3 2 Tube /rubesheet Contact Pressure for Faulted Conditions

- llot Leg...................................................................4-14 Figure 4.3 3 Through Thickness Tubefrubesheet Contact Pressure for Faulted Condition Ilot Le g . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 16 Figtue 4.3 4 Tubefrubesheet Contact Pressure for Faulted Condition

- Col d Le g . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 17 Figure 4.3 5 Through Thickness Tubefrubesheet Contact Pressure for Faulted Condition Cold Leg . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4- 19 Figure 4.3-6 Tube /rubesheet Contact Pressure for Normal Condition

- 110t Leg...................................................................421 Figure 4.3 7 Through Thickness Tubefrubesheet Contact Pressures for N ormal Condition - Hot 12 g . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 23 Figure 4.3 8 Tubefrubesheet Contact Pressures for Normal Condition

- Cold Le g . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 - 2 4 Figure 4.3-9 Through Thickness Tubefrubesheet Contact Pressures for Normal Condition - Cold Le g . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 2 6 Figure 4.41 Contact Pressure for Faulted Conditions on Hot Ieg Side as a Function of Depth from BWT and Radius........................ 4-28 Figure 4.4 2 Contact Pressure for Normal Conditions on Hot Leg Side as a Function of Depth from BWT and Radius........................ 4-29 Figure 4.4-3 Contact Pressure for Faulted Conditions on Cold Leg Side as a Function of Depth from BWT and Radius........................ 4 30 Figure 4.4-4 Contact Pressure for Normal Conditions on Cold leg Side as a Function of Depth from BWT and Radius........................ 4-31 Figure 4.4-5 Contact Pressure Used in PLRL Determination for Faulted Condition on llot12g Side. Includes Faulted Condition aad WEXTEX Contaet Pres sures........................ .............. 4-37 iv k

LIST OF FIGURES (CONT'D)

FIGURE TITLE EAGE Figure 4.4 6 Contact Pressure Used in PLRL Determination for Normal Operation on Hot leg Side. Includes Normal Operation and WEXTEX Contact Pressures....................................... 4 38 Figure 4.4 7 Contact Pressure Used in PLRL Determination for Faulted Condition on Cold leg Side. Includes Faulted Condition and WEXTEX Contact Press ure s.......................................,4 39 Figure 4.4 8 Contact Pressure Used in PLRL Determination for Nonnal Operation on Cold leg Side. Includes Normal Operation and WEXTEX Contac t Pressure s....................................... 4 40 Figure 4.4 9 Pullout lead Reaction Length for Normal Condition at Hot Leg and Cold leg. Length Shown is from B ottom of WEXTEX Tran sition. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 41 Figure 4.410 Pullout load Reaction length for Faulted Condition at Hot Leg and Cold Leg. Length Shown is fmm B ottom of WEXTEX Tran si tion. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-4 2 Figure 4.411 W* Length for Sequoyah 1 & 2 Measured frora Bottom of WEXTEX Transition (BWT) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-44 Figure 4.5-1 W* Zones A Corresponds to Expansion Transition Ins pection Zone s 1, 2, a nd 3 . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . 4 46 Figure 5.1-1 Ulustration of Degradation Bands Within Tubesheet Region........ 5 2 Figure 5.2-1 Sketch of Idealized Symmetric Crack Arrays.......................... 5 5 Figure 5.2 2 Crack Model for Allowable Circumferential Degradation............ 5 6 Figure 5.2 3 Sketch ofIdealized Non Symmetric Crack Arrays ................... 5-7 Figure 5.2-4 Model for Plastic Collapse of Slanted Straight Crack Array......... 5 8 Figure 5.2-5 Compadson of Load. Displacement Records, Computed Versus Measured for 30 Slots at & = 45*.... .. . . ... . . . .. . .. .... ... .. ... 5 9 Figure 5.2-6 Model for Plastic Collapse of Singly Hooked Slanted Cracks ...... 5 11 Figure 5.2 7 Model for Plastic Collapse of Doubly Hooked Slanted Cracks ..... 5 13 Figure 5.2 8 Yield load Versus Hook Length of Slanted Cracks.................. 5 14 Figure 6.21 As Fabricated _WEXTEX Sample for Leak Testing................... 6-4 Figure 6.2 2 Configuration of WEX'ITlleak Rate Samples...................... 6-6 Figure 6.2 3 Inside Diantter ofleak Rate Sample W4-018........................ 610 -

v

LIST OF FIGURES (CONT'D)  !

FIGURE TrIUI PAGE Figure 6.3 1 Constrained Crack Leak Rate Test Arrangement...................... 6-15 Figure 6.3 2 Tube Specimens Utilized in Constralned Crack leak Rate Tests ... 616 Figure 6.3 3 Alloy 600 Collar Utilized in Constrained Crack Leak Rate Tests ... 618 Figure 6.' 4 Restricted Crevice Measured vs. Predicted Leak Rate................ 6-21 i Figure 6.41 Flow Coefficient Versus Tube.to-Tubesheet Contact Pressure ..... 6 24 Figure 6.4 2 WEXTEX Faulted Condition leak Rate, Regiession Line  ;

Crevice Loss Coefficient, llot Leg,2.3" Tubesheet Radius......... 6-31 Figure 6.4 3 WEXTEX Faulted Condition leak Rate for Lower Bound 90% Confidence Limit Crevice Loss Coefficient,llot leg, 2.3 " Tube shee t R ad iu s . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6- 3 2 Figure 6.4-4 WEX" TEX Faulted Condition leak Rate for imwer Bound 90% Confider cc Limit Crevice less Coefficient, 1101 Le g, 12" Tu be sh e e t R adiu s... ......................... ............. 6-3 3 .

Figure 6.4 5 WEXTEX Faulted Condition leak Rate for Lower Bound 90% Confidence Limit Cn:vice less Coefficient,Ilot leg.

22" Tube shee t Radi u s . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6 34 Figure 6.4 6 WEXTEX Faulted Condition Leak Rate for lewer Bound 90% Confidence Limit Crevice Loss Coefficient, Ilo t Le g, 32" Tu be s he et Rndiu s ......................................... 6 35 Figure 6.4 7 WEXTEX Faulted Condition leak Rate for lewer Bound 90% Confidence Limit Crevice Loss Coefficient,llot 12g, 4 2.51 " Tubeshee R adiu s . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-3 6 Figure 6.4-8 Leak Rate Versus Circumferential Crack Angle....................... 6 38 Figure 7.1-1 Sequoyah 1 - Distribution of Circumferential Crack Angles......... 7 4 Figure 7.1-2 Sequoyah 1 Axial Position of WEXTEX Circumferential Cracks.....................................................................75 Figure 7.1-3 Sequoyah 1 Location of WEXTEX Axial Cracks vs.

Top . of Tu be shee t Ele v at10 n.... .... ... . ...... .... .. ..... ..... . . ..... ... .. 7 6 Figure 7.14 Sequoyah 1 - Distribution of Axial WEA" TEX PWSCC Crack 12ng th s . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7 7 Figure 7.31 Plant A 1 WEXTEX Axial Crack Locations with Respeet t o B W T. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7 - 10 vi

f i

LIST OF FIGURES (CONT'D)

FIGURE TITLE PAGE ,

Figure 7.41 Conclation Between Rotadng Pancake Coll. Detected Arc Length vs. Metallographic Exam Results ......................... 7 12 Figure 8.21 Tubesheet Map Showing Two knes for W* ......................... 8-4 Figure 8.3-1 Flexible W* Length and SLB Leakage Description................... 8 6

-i I

V 4

k V

l t

I i

LIST OF TAllLES NUMBER .TIII.E PAGE Table 3.1 1 Sequoyah Unit 1 & 2 Parameters For W* Criteria.................... 3 5 Table 4 2-1 WEXTEX Expansion Joint Pull Force Test Results.................. 4 6 Table 4.31 Faulted Condition Displacements and Contact Pressure Variations Through the Tubesheet Thickness (llot 12g)............. 4 15 Table 4.3 2 Faulted Condition Displacements and Contact Pressure Variations Through the Tubesheet Thickness (Cold Leg)............ 418 Table 4.3 3 Nomtal Condition Displacements and contact Pressure ,

Variations Thmugh the Tubesheet ' thickness (Ilot 1xg)............. 4 22 Table 4.3 4 Nonnal Condition Displacements and Contact Pressure Variations Through the Tubesheet Thickness (Cold Leg)............ 4 25 Table 4.41 Tube to-Tubesheet Fricdon Coefficient and WEXTEX Radial Ccatact Pressure Detennined from Pull Force Tests ................. 4 35 Table 4.4 3 W* Length for Sequoyah I & 2 Steam Generators................... 4 45 Table 6.21 Inside Diameter of WEXTrX leak Rate Samples........... ........ 6-5 Table 6.2 2 WEXTEX Expansion Joint leak Rate Test Results .................. 6 8N Table 6.2 3 WEXTEX Expansion Joint leak Rates, Crevice Lengths and A yerage Contae t Pre ssure........................................... 6-12/13 Table 6.31 Diametral Gap Sizes (mils) at Temperature and Pressure ............ 6-19 Table 6.3 2 Restricted Crevice leak Rate Test Results............................. 6-20 Table 6.4-1 Operating Contact Pressure at Cold leg for SLB Conditions....... 6 27 Table 6.4 2 Operating Contact Pressure at 110t Leg for SLB Conditions ........,6-27 Table 6.4-3 WEXTEX Leakage, Faulted Conditions, Hot leg (2.2 8 " to 12" R adli) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6- 2 9 Table 6.4-4 WEXTEX Leakage, Faulted Conditions. Hot Leg,

( 22 " t o 4 2.5 " R adii) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6 Table 6.51 Example of SLB leak Rate Determination for Sequoyah I & 2 W EX TE X Tu b e s . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-41 Table 7.1 1 Sequoyah Unit 1 IIL WEXTEX Transition RPC R e s ul ts ( I EOC5 ) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7 Table 7.1 2 Sequoyah 1 - Ultrusonic Testing of Largest Circumfere ntial Indication s . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-3 vill

LIST OF TABLES (CONTD)

NUMBER 'ITILE PAGE Table 7.21 Sequoyah Unit 211L WEX111X Transition RPC Re s ul ts (2EOC5 ) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9 Table 7.51 Estimate of Expansion Zone PWSCC- Avemge Crack Growth Rates (Based on Kiss Roll Data).............................. 14 IX

I SEQUOYAll UNITS 1 & 2 W* tulle Pl.UGGING CRrrERIA FOR SG TUllESilEET REGION OF WEXTIIX EXPANSIONS 1.0 llA C K G R O U N D Existing plant Technical Specification t.ibe repairing / plugging criteria apply throughout the tube length and do not tahe into account the trinforcing effect the tubesheet has on the external surface of an expanded tube. De presence of the tubesheet will constrain the tube and will complement tube integrity in that region by essentially precluding tube deformation beyond the expanded outside diameter. The resistance to both tube rupture and tube collapse is signincantly enhanced by the tubesheet. In addition, the proximity of the tubesheet in the expanded region signincantly affects the leak behavior of throughwall tube cracks. Ilused on these cor.siderations, the establishment of alternate plugging criteria specific to the region of tubes expanded by the Westinghouse explosive tube expansion (WLXTEX) process is justified. The W* criteria adapt elements of the L* criteria for hardroll expansions to explosive expansion conditions.

He Sequoyah 1 & 2 steam generators have a full depth WEXTEX expansion that can be genentily denned as follows. From the tube end and extending for a length of approximately 2.75 inches is a region expanded by the roll expansion process. From the top of the roll expansion to the vicinity of the top of the tubesheet the tube to tubesheet expansion is accomplished by the WEXTEX explosive expansion process. The resulting full depth tube-to-tubesheet expansion can be considered as four distinct regions. Rese are described below staning from the bottom of the tubesheet and are illustrated in Figure 1.0-1.

1. Roll Region That region of the tube which has been expanded by a rolling process.

This region extends from the bottom of the tube to approximately 2.75 inches above the bottom of the tube.

2. Roll Transition (RT) . Wat ponion of the tube which extends from the roll expanded region of the tube to the initially unexpanded region, and which is subsequently expanded by the WEXTEX process, 1-1 1

1

a,c,e Figure 1,0-1 Regions in the WEXTEX Full Depth Tube-to-Tubesheet Expansion 12

3. WEXTEX Region . That ponion of the tube which was expanded by the explosive ,

expansion process to be in contact with the tubesheet. This region stans at the roll I

transition and extends to the WEXTEX transition in the vicinity of the top of the tubesheet.

4. WEXTEX Transition - That portion of the tube which acts as a juncture between the WEXTEX region and the unexpanded region of the tube. This region starts at the top ,

of the explosively expanded region and extends for approximately 0.25 inches.

This report documents the development of criteria for reducing the need to repair or plug steam generator tubes which contain eddy current indications in the explosive expansion portion of the tubesheet joint below the expansion transition region. This repon summarizes the Sequoyah Units 1 & 2 tube integrity evaluation for tube eddy current indications occurring within the tubesheet relative to:

1. Maintenance of tube integrity for all power loadings associated with normal plant conditions, including stanup, operation in power range, hot standby and cooldown, as well as anticipated transients.

2. Maintenance of tube integrity under postulated limiting conditions of primary to- ,

secondary .md tecondary-to primary differential pressure.

3. Limitation of primary to secondary and secondary to primary leakage consistent with I

safety analysis assumptions and the desire to minimize leakage.

This report is an evaluation of alternate ph;gging criteria specific to Sequoyah Units 1 -& 2, and is based upon site specific operating parameters and allowable leakage requirements.

13 w_,.s,. .. ,- ,, n--rpm w w wr r -e. m---m-

2.0 SUhihiARY AND CONCLUSIONS 2.1 OVERALL SUMhiARY WEXTEX-expanded region altemate tube plugging criteria, or W* criteria, were developed for the tubesheet region of the Sequoyah 1 & 2 steam generators considering the most stringent loads associated with normal operation, including transients, and postulated accident conditions. The W* criteria air developed to help prevent tube burst and axial separation due to axial pullout forces acting on the tube, and help to provide that steam line break (SLB) leakage limits are not exceeded.

The W* criteria developer in this report permit tubes with indications in the tubesheet region to be retumed to service, and were performed utilizing plant specific operating parameters for Sequoyah Units 1 & 2.

The approach taken to develop the W* criteria is to utilize the general rnethodology of the L*

criteria for hardroll expansions and adapt the methods for WEXTEX expansions. The hardroll L*

criteria utilize an L* length of undegraded tubing to limit leakage and permit a flexible length (P) to resist pullout forces, with the length increased if degradation is present within the minimum P length. Since WEXTEX expansions have lower tube to tubesheet contact forces than hardroll expansions, limited leakage is possible under SLB conditions and the L* length is replaced in W*

by the requirement to calculate SLB leakage for indications left in service. The SLB leakage for a given crack size is dependent upon the length below the bottom of the WEXTEX transition (BWT). A flexible W* lengtn patterned after the flexible F* length of the L* criteria is applied in W*. In addition, the L* criteria are moditied to permit Ilmited length circumferentialindications within the pullout force distance. This option could be accommodated in L* as well as W*. 'lh general approach taken for W* can be applied to indications within the :ubesheet for attemate types of tubesheet expansions. The principal difference between the type of tubesheet expansion (hardroll, explosive, hydraulic, partial depth) is that the methods for calculating SLB leakage must be specialized for the method of expansion.

2.2 SUbiMARY AND CONCLUSIONS The development of W* criteria for Sequoyah Units 1 and 2 included testing of samples r:presentative of the WEXTEX expanded tube to tubesheet joints, analysis of contact pressures t etween the tube and tubesheet at relevant operating conditions, and analysis of axialloads acting en the tube Testing included pull force tests to determine the coefficient of friction between the tube and the tubesheet and the force irquired to cause movement of the tube within the tubesheet.

2-1

Leak rate tests were performed to develop a WEXTEX leak rate model relating leak rate to crack size, location of the crack within tubesheet, and operating conditions. The model is used to estimate potential leakage from tubes with indications left in service under SLB accident conditions.

The W* alternate plugging criteria developed for Sequoyah Units 1 & 2 can be summarized as follows:

W* Length The W* criteria include a W* length,i.e., the length of tube engaged in the tubesheet which is sufficient to react the applied axial loads to and help to avoid tube pullout.

For the most stringent loading conditions, i.e., nonnal operation, the applied axial load including a safety factor of three is about [ ]re e. The maximum W*

length necessary to resist this load under normal operation at the hot leg is [ ]6 C

• inches as measured from the bottom of the WEXTEX transition. W* has a radial dependence across the tubesheet with the nuutimum W* occurring near the tubesheet center. The [ ]a.e.e inch W* length applies throughout W* Zone B, whl:h corresponds exactly to transition inspection Zone 4. A [ p.c.e inch W* length applies throughout W* Zone A, which corresponds exactly to transition inspection Zones 1,2, and 3.

If indications are observed within the W* length, the W* 1ength must be adjusted to account for the potential loss of axial resistance force in the region of the cracks.

Allowable Decrndation Within W* Region The W* criteria limit degradation within the W* length to assure adequate strength to resist the axial pullout loads through the W* length. For axial cracks having the upper crack tip below the top of the tubesheet and the bottom of the WEXTEX transition, pull foire and leak rate requirements place no limit on crack length.

Evaluation shows that a tube could have an end of cycle thmughwall circumferential crack of [ las.e in circumferential extent in addition to a [ las,e throughwall OD-initiated crack over the balance of the circumference and not separate under an 22

e i applied axial load equal to three times the normal operating differential pressure (3AP). Decreasing this crack angle by the NDE uncertainty of 39* and potential growth of [ Jas.e permits an Beginning of Cycle (BOC) c:ack angle of up to

[ ]8 C e. Any tube degradation exceeding the above limits within the W* distance must be plugged or repaired. In addition, the circumferential crack must be located below the bottom of the WEXTEX transition and at least [ Ja44 inch below the top of the tubesheet.

Bands of parallel, axially oriented cracks are limited to five cracks. If the cracks are inclined relative to the tube axis by more than [ ]as.e. the total circumferentially oriented extent summed over all cracks must be less than [ ] Aceat the beginning of cycle. Similarly the circumferential extent of closely spaced axial cracks must le less than [ Ja4# between the " null points" on the RPC amplitude.

W* criteria are applicat,1c to those tubes which have typical WEXTEX expor.slons.

Characterization of candidate W* tubes may be necessary to determine if they are enveloped by the evaluated baseline conditions.

WEXTEX SLB Lenkact Tubes with indications in the tubesheet region may 12 retumed to service if the total potential SLB leakage from the tubes, as determined from the WEXTEX and potentially other leakage models does not exceed the allowable accident condition leakage guidelines determined by the offshe radiological dose evaluations detailed in Section 15 of the SQN FSAR.

i If the total leakage exceeds the limit, tubes must be plugged until the potential leakage fmm the non plugged tubes is within the allowable leakage limit.

l By maintaining expected SLB leak rates below the values assumed in the FSAR evaluations, offsite doses will be bounded by the existing licensing basis and inherently will remain within the guidelines of 10CFR100.

2-3

3.0 TECIINICAL APPROACil i 3.1 REQUIREMENTS FOR W* CIUTERIA 3.1.1 Regulatory Requirements Regulatory Guide (RO) 1.121, "Dases for Plugging Degraded PWR Steam Generator Tubes",

issued for comment in August of 1976, describes a method acceptable to the NRC staff for meeting General Design Criteria (GDC) numbers 14,15,31 and 32 by reducing the probability and consequences of steam generator tube rupture through determining the limiting safe conditions of degradation of steam generator tubing, beyond which tubes with unacceptable cracking, as established by inservice inspection, should be removed from service by plugging. The L recommended plugging criteria for the tubesheet region of WEXTEX expansions may result in tubes with both partial thmughwall and thmughwall (non leaking) cracks being returned to serviet.

, in the limiting case, the presence of a throughwall crack alone is not reason enough to remove a l tube from service. The regulatory basis for leaving throughwall cracks in service in the tubesheet region of WEXTEX expansions is pmvided below.

Steam Generator " tube failure" is defined by the NRC within RG 1.83 as the full penetration of the pdmary pressure boundary with subseouent leakace. Consistent with this definition. upon the imolementation of the tube pluccine criteria of this troort. known leakine tubes will te removed from service. 'Ite tube pluccine criteria of this reoort are established such that operationalleakace is not anticloated.

The NRC defines steam generator tube rupture within RG 1.121 as any perforation of the tube pressure boundary accompanied by a flow of fluid either from the primary to secondary side of the steam generator or vice versa, depending on the differential pressure condition. As stated within the regulatory guide, the rupture of a number of single tube wall barriers between pii. nary and secondary fluid has safety consequences only if the resulting fluid flow exceeds Ln acceptable amount and rate. This rate has been defined in NUREG-0844 as exceeding the make up capacity of the plant. Loss of steam generator tube integrity means loss of " leakage integrity". Loss of

" leakage integrity"is defined as the degree of degradation by a throughwall crack penetration of a tube wall membrane that can adversely affect the margin of safety leading to " tube failure", burst, or collapse during both normal operation and postulated accident conditions. Acceptable service, in terms of tube integrity, limits the allowable primary to secondary leakage rate during normal operating conditions and pmvides that the consequences of postulated accidents would be within 3-1

the guidelines of 10 CFR 100. In order to determine that steam generater tube integrity is not reduced below a level acceptable for adequate marg'a is of safety, the NRC staff position focused on specific criteria for limiting conditions of operation. These include: I i

1. Secondary Water Monitoring

2. Primary to-Secondary Tube Ixakage i

3. Steam Generator Tube Surveillance

4. Steam Generator Tube Plugging Criteria Tubes with throughwall cracks will maintain " leakage integrity" and are acceptable for continued operation if the extent of cracking can be shown to meet the following RO 1.121 criteria:

1. Tubes are demonstrated to maintain a factor of safety of three agains failure for bursting under normal operating pressure differential.

2. Tubes are demonstrated to maintain adequate margin against tube failure under postulated accident condition loadings (combined with the effects of SSE loadings) and the loadings required to initiate propagation of the Isrgest longitudinal crack resulting in tube rupture. All hydrodynamic and flow induced forces are to be considerrd in the analysis to detennine acceptable tube wall penetration of cracking.

3. A primary to secondary leakage limit under nonnal operating conditions is set 1.1 the plant technical specifications which is less than the leakage rate determined theoretically or experimentally from the largest single permissible longitudinal crack.

This action would ensure orderly plant shutdown and allow sufficient time for remedial action (s)if the crack site increases beyond the pennissible limit during serVICc.

In addition to the RO 1.121 criteria,it is necessary to satisfy FSAR accident condition limits for primary-to-secondary leak rates. Leak rate limits must be satisfied on a plant specific basis to the guidelines of 10 CFR 100. For Westinghouse plants, a steamline break event (SLB) is generally the limit event for radiological consequences and the SLB is applied in this repon as the reference event for limiting accident condition leakage.

32 i

7 3.1.2 Requirements for W* Tube Plugging Criteria Tube Burst Considerntions Tube burst is precluded for cracks within the tubesheet by the constraint provided by the tubesheet.

Thus the R.O.1.121 criteria are satisfied by the tubesheet constraint. Crack lengths do not need to be limited by burst considerations and operating leakage limits are not required to detect crack lengths associated with tube burst Conceivably, however, a 360 throughwall circumferential crack or many axially oriented cracks could permit severing of the tute and tube pullout from the tubesheet under the axial forces on the tube from primary to secondary pressure differentials. The W* criteria are required to prevent tute pullout from the tubesheet under axialloading conditions. A W* length is required such that the tube to tubesheet contact pressures integrated over the W* length are sufficient to compensate for the axial fortes on the tube and thus prevent tube pullout. Within the W* length, tube degradation must be limited to permit the axial loads to be transmitted over the t:ntire W* length without severing of the tube. The W* criterht limit the angular extent of circumferential cracks and/or the length or number of axial crackt left in service in the W* length to permit acceptable axial load capability. Below the W* length, any type of tube &andation is acceptable as tube integrity is not trquired to accomnxxiate axial kuds on the tube or to limit tule leakage.

Operntine Inkace Considerations WEXTEX region cracks could potentially occur as the result of prim: / water stress corrosion cracking (PWSCC). Extensive European operating experience has been obtained with axial PWSCC cracks left in service as summarized in Reference 1. This operating experience has demonstrated negligible nonnal operating leakage from PWSCC cracks even under free span conditions in roll transitions. PWSCC cracks (if they were to occur) in WEXTEX expansions in the tubesheet region would be even further leakage limited by the tight tube to-tubesheet crevice and the limited crack opening permitted by the tubesheet constraint.

Consequently, negligible operating leakage is expected from cracks in the tubesheet region of WEXTEX expansions and no W* requirements must be applied to limit operating leakage.

3-3 1

Accident Condition Inkage As noted in Section 3.1.1, the accident condition leakage must be limited to acceptable limits established by plani specific FSAR evaluations. The SLB event is limiting for leakage for the Sequoyah Unit 1 & 2 steam generators and is applied in this report. Since the higher pressure differentials associated with a SLB event can open cracks that had negligible operating leakage, the W* criteria must provide a method to calculate SLB leakage to deme ' strate compliance with FS AR requirements. Leakage is limited for cracks in the tubesheet by crack size and the small crevice l opening within the tube to tubesheet crevice. Based on Westing'i.ouse models for leakage through cracks, leak rate tests of WEXTEX expansions, a SLB leakage analysis model is developed in this report to calculate SLB leak rates for cracks left in senice.

Application of the W* tube plugging criteria requires the use of Sequoyah Unit 1 & 2 SLB leak rate.

limits. SLB leak rate analyses for cracks left in service must be performed using the methods of the repon to demonstrate leak rates less than the acceptable limits. If tubesheet region cracks, as j accounted for in the SLB leak rate analysis are the only deep (>40% depth) cracks left in senice,it is expected that SLB leakage will not exceed allowable limits. ,

Operating Conditions for the Sequoyah Unit 1 & 2 W* Criteria The key Sequoyah Unit 1 & 2 operating conditions utilized in this analysis and the basis of each are summarized ir Table 3.1-1. The values are from recent plant measurements reported by TVA.

The principal operating parameter influencing the W* criteria is the secondary steam pressure measured at the top of the tubesheet, which is [

4 ja.c.e Although a That reduction is currently being considered by TVA, an implementation decision has not been finalized, and it was deemed prudent to utilize current operating conditions for this analysis. If needed, an update to the W* analysis can be perfomied as a subcomponent of a That reduction pmgram.

l 3-4

a,c.e l ,

I l

[

i Table 3.1 1 Sequoyah Unit 1 & 2 Parameters For W* Criteda 1

3-5 l

e .. -v ,- _

The principal operating parameter influencing the W* criteria is the secondary steam pressure.- '

Since chemical cleaning of the secondary sid: in Unit 1 is planned, there may be an increase in steam pressure. However, increased steam pressure is typically more conservative relative to the W* criteria, and the new operating conditions would continue to be bounded by the values in this report.

The secondary side temperature of [

]* c.e ne saturation temperature differs from the actual temperature at the downcomer inlet by the subcooling due to mixing of fer Jwater with the reciseulation water.

3.2 LOADING CONDITIONS ,

The loading conditions for which tube integrity must be maintained include all power loadings associated with normal operation, including startup, operation in power range, and hot standby as well as anticipated transients and postulated accident conditions. Specific conditions evaluated were normal operation, feedline break (FLU) + SSE, and LOCA + SSE. The most stringent faulted condition is the FLB, The applied loads acting on the tube which could result in tube pullout from the tubesheet during normal and postulated accident conditions are predominantly axial and due to the internal to-external pressure differentials. For a tube that has not been degraded, the axial load is given by the '

product of the pressure and internal cross-sectional area of the tube. However, for a tube with degradation, e.g., cracks oriented at an angle to the axis of the tube, the internal pressure may also act on the flanks of the degradation. Therefore, for a tube which is conservatively postulated to be severed at some location within the tubesheet, the total force acting to remove the tube from the tubesheet is given by the product of the pressure and the cross sectional area of the tube outside ,

L diameter,i.e., the inside diameter of the tubesheet hole for an expanded tube.

The loads from normal and FLB conditions act from the inside of the tube and are tensile in nature,

, tending to pull the tube from the tubesheet. The LOCA loadings are due to pre:sure acting on the OD of the tube and are compressive and tend to push the tube into the tubesheet. The magnitude of these loads is addre. n ed below, l

l i

3-6

l 3.2.1, Normal Operating Loads For the analysis of the axial force loadings from normal operating conditions it is assumed that the tu" has degradation and it is also assumed that the pressure acts on the combined intemal and wall cross sectional area of the tube. For the Sequoyah Unit 1 & 2 tuben under normal operating conditions this area is detennined as the cross sectional area of the tubesheet hole.

'Ihe force acting on the tube can be expressed as i

I l 1

]se.e

The axial force acting on the tube under normal operating conditions calculated in this manner is j [ Jae.e lbs. A safety factor of three is applied to this load for design purposes. Therefore the axia' load bearing requirement for a Sequoyah Unit 1 & 2 steam generator tube is approximately I la.c.e Ibs. Other forces such as fluid drag forces in the U bends and vertical seismic forces 1 are negligible by comparison.

l 3.2.2 Other Transient Conditions

An evaluation was perfonned to consider operating transients which couid result in the condition l where the tube would be at a temperature lower than the tubesheet, in this situation some of the

} engagement preload would be lost as the tube would shrink relative to the tubesheet. The worst

case occurs for a Loss of Flow transient where the tube temperature becomes about 10 degrees lower than the tubesheet temperature. Ilowever, during this transient the decrease in primary side j pressure offsets the effects of the decrease in primary side temperature and the temperature j difference between the tube and the tubesheet, such that a net increase occurs in the contact 2

pressure between the tube and tubesheet hole.

i 3.2.3 Feedline Break The axial load on a tube under FLB conditions is approximately [

f 2

37 W

.-c.- - , - . m.-.. e -, , , - - . . - ,.u +

, - - ,. ~-,. , . , , ,,

As discussed in Reference 2 this load need not le funher augmented for dynamic loading effects during FLl!lecause of the sequence of events during an FLIL 3.2.4 LOCA l The axial load acting on the tube under LOCA conditions is approximately l ] ate Ibs, and is compressive Imm the tube OD. Therefore,it tends to force the tule into the tubesheet rather than to pull it out, and is not a factor in W* analyses.

3.2.5 Other Faulted Load Considerations Seismic analysis of the Sequoyah Unit 1 & 2 steam generators, Reference 3, has shown that axial loading of the tubes is negligible during a safe shutdown earthquake (SSI!).

3-8

4.0 PULLOUT LOAD REACI'lON LENGTil 4.1 EVALUATION ML'TilODS AND LOADS in the unlikely event that a steam generator tule is circumferentially degraded within the tubesheet to the extent that tube separation could occur, that portion of the tube from the separation to the top of the tubesheet must equilibrate the applied loads if tube pullout is to be prevented. The nomial

' la.c.e pounds, respectively, as and faulted operation applied axial loads are [

developed in Section 3.2. These are the most stdngent loads that may be applied to the tubes. To prevent pullout, these loads must be reacted by the axial restraint afforded by the contact pressure between the tube and tubesheet times the fdction coefficient of the tute to tutesheet interface acting over sorne interface length. For the purpose of this evaluation, the length of engagement of a tube in the tubesheet necessary to prevent pullout is defined as the pullout load reaction length, or PLRL.

Tmicennine PLRL it is necessary to know the magnitude of the contact pressure between the tube ano tubesheet, which is the contact pressure from normal or faulted operation plus that from the WEXTEX expansion process, and the value for the break away coefficie.a of friction for the tube /tubesheet material couple. Contact pressures resulting from nonnal or faulted operation have been calculated through finite element analysis as described in Section 4.3. The contact pressure from the WEXTEX process was detennined through test!ng of WEXTEX fabricated samples representative of the WEX" REX tube to tutesheet expansions. The fdction coefficient for the tube-to-tubesheet interface was also determined from these tests.

The engagement length,i.e., PLRL,is detennined as the depth, measured from the bottom of the WEXTEX transition, where the integrated irsistance load is equal to the applied load.

4.2 TESTING 4.2.1 Test Samples The samples used in the pull force tests were designed to simulate the tube-to-tubesheet interface characteristics. The samples consisted of a carbon steel collar approximately [

4-1

]bs.e The tube specimens were fabricated fmm Quality Assurance (QA) controlled stock and fabrication of the collars was done under QA st.tveillance. The WEX'ITW expansions were fabricated per approved procedures.

The collars used to simulate the tubesheet were [

)b.c.e The WEXTEX samples fabricated for pull force testing were of a [

]bt.e The samples were also noted to have [

)bs,e 4.2.2 Test Description and Results The samples were tested at various temperature and internal pressure conditions to assess the pull forces associated with the WEXTEX joint. The tests were performed in accordance with an approved, referenceable test specification under QA surveillance.

The actual test samples used for the pressurized and unpressurized tests were slight modifications of the as expanded double ended sample described in Section 4.2.1. The actual test sample configurations are illustrated in Figure 4.2-2. As noted in the figure, the samples were tested by applying an axial tensile load to the tube while restraining the collar.

4-2 i

. - _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ . - _ _ i

The testing was perfonned on a SATEC testing machine and a load /crosshead displacement curve was recorded as the test progressed. The pull force of the sarnple was taken to be the load where .)

'Tiret slip" of the tube in the collar occurred as reflected by a drop in the test load. In those tests at room temperature where there was no noticeable drop in the load, the pull force was taken as the load that is present when the tube was at [ ]b.c.e inch crosshead movement.

The pull force samples were tested at four nominal conditions to obtain pull forces at [- ]b.c.e tube to-collar interface contact pressures. The nominal test conditions included [

jbAe The results for the eight tests are given in Table 4.21.

The nominal interface length between the tule and the collar was approximately [

]b.c.e inches for the pressurized gests. As discussed in the analysis of the test data (Section 4.4.2), the effective interface lengtit is somewhat less than the nominallisted in Table 4.2 1 tecause a) the WEXTEX transition section was wi Mn the nominalinterface length, and b) there are end effects associated with the tube disconti3 ~ , .,i.e., WEXTEX transition and sectioned end of the tube.

The room temperature pull forces indicate some interference between the tube and the tubesheet simulant collar. The joint tightness [

]b.c.e These have been included in the table for information.

4-3

__ _ _ _ - _ . . _ _ . . - _ _ _ _ _ _. _ s

4 4

I i

i, a,c. e  !

i l

i i.

l i

I i

i 7

a 1'

4 4

b Figure 442-1 As Fabricated WEXTEX Sample for Pull Tests i 1

4-4 1 i

l i

. , _ - , . - ._. . . . . . , .. . . . . . _ . .;..-..,.#. s.-

. _ . . _ _ - _ . . . . ._- _ _ _ . . - _ . _ ~ . . - .. _ . . . .

t I

a,c.e k

Figure 4. 1 Pull Force Sample Configurations Tested o

i' I 4-5; L

i 3 l

'-m .- y

4 m

g F-t

. auf J

D sn W

M to W

W U

Cd O

~ W e

N .J

• J V D L

W J &

G3  %

aC =

l & O o

2 O

A L

X W

M W

M W

3 t

4 4.3 ANALYSIS OF TUBESHEET DEFIECTION EFFECTS A 2 D axisymmetric finite element analysis of the Series 51 tubesheet, channel head, and lower shell was performed in Reference 5. This model is shown in Figure 4.3-1. Displacements throughwt the tubesheet were obtained for the unit loads listed below:

Unit Load Macnitude Primary Side Pressure 1000 psi Secondary Side Pressure 1000 psi Tubesheet Thermal Expansion 500*F AT Shell Thermal Expansion 500 F AT Channel Head Thermal Expansion 500 F AT The axisymmetric analysis does not include the effect of the divider plate in restraining the tubcsheet displacements. The calculations performed with the 3-D finite element model of a Model D4 steam generator (Reference 6) showed that the displacements at the center of the tubesheet when the divider plate is included are [ Ja,e,e of the displacements without the effect of the divider plate. Since the Series 51 channc' Sead/tubesheet/ divider plate geometry is identical to that of the Model D4, the same factor of [ ]a,e.e was applied . the Series 51 tubesheet displacements.

This [ ]a.c.e factor is appUed to the tubesheet displacements produced by the pressure unit loads. The radial displacements produced by the thermal unit loads am unaffected by the divider plate.

The radial deflection at any point within the tubesheet is found by scaling and combining the unit load radial deflections at that location according to:

a,c.e (1) 1 I

4 a,c.e

h e

Figure 4.3 1 Tubesheevibnnel Head / Lower Shell Finite Element :131.,

4-8

I This expression is used to deterndne the radial deflections along a line of nodes at a constant axial elevation (e.g., top of the tubesheet, one inch below the top, and so on) within the perforated area of the tubesheet. A fourth order polynomial is fit to the UR vs R data so that UR may be expressed as a continuous function of the tubesheet radius as shown below, a,c.c

[ ] (2) 1 The expansion in the radial direction of a hole of diameter D in the tubesheet at a radius R is given by:

a,c.e (3a)

Substituting equation (2) into (3a),

a,c.e

[ ] (4)_

The expansion in the cimumfemntial direction of a hole of diameter D in the tubesheet at a radius R is given by:

[ ] ace (3b)

U R(R)is ohmined directly from equation (1). The expansion of a hole in the tubesheet is taken to be the average of the radial and circumferential expansions given in equations (4) and (3b) above.

During installation, the tubes are expanded within the tubesheet so that their diameter is equal to the diameter of the holes in se tubesheet. The thermal expansion of the outside surface of a tube at any location within the tubesheet is given by:

Dtube(thermal) = Datube(Trube - 70) (5) where atube = Coefficient of thermal expansbn of the tube T tube = Temperature of the tube = Tubesheet temperatum 4-9

The radial expansion of the outside surface of the tube when acted upon by intemal and external pressures is (Reference 7):

a,c.e (6)

Where

[

Ja,e.c The unrestrained expansion of the tube is obtained by combining equations (5) and (6), If this is greater than the expansion of the hole in the tubesheet, then the tube and the tubesheet are in contact. The thick cylinder equations of Reference 7 can be used to determine the contact pressure -

between the tube and tubesheet. The inward radial displacement cf the outside surface of the tube is given by:

a,c.e (7) <

~

The radial displacement of the surface of the hole in the tubesheet is given by:

a,c.c (8)

Thre r

[

ja.c.e The contact pressure is found from:

l 4-10 w- e y w. - 7-w93 -eee

a,c,e This results in :

a,c.e (9) 4.3.1 Results for Normal and Faulted Conditions l

The unit load displacements were taken from Reference 5 along lines of constant axial elevation in the perforated part of the tubesheet. These lines are atang the top and bottom of the tubesheet plus one and two inches from each of these surfaces, InbslTubesheet Dimensions and Material Procenics From Reference 5, the dimensions of the expanded tube and tubesheet unit cell are:

I Ja.c.e Material properties at 60(FF are:

I l )a.c.e -

Faulted Condition

! The faul'ted condition to be analyzed is :

l l 4-11

. .- . .- - - _ - . .- _. --..._~ .. . - .. . .

]ax,e The diametral thennal expansion of the outside surface of a tube on the hot leg side is givLn by:

ADt ube(hot) = (0.890) (7.82x10-6) (608.5 - 70)

= 0.003748 in.

The diametral thermal expansion of the outside surface of a tube on the cold leg side is giver. by:

ADt ube(cold) = (0.890) (7.82x10-6) (546.5 -70)

= 0.003316 in.

The radial expansion of the outside surface of the tube caused by pressum is:

a,c.e So that the diametral expansion of the tube on the hot leg side during the faulted condition is:

[ ]a.c.e and on the cold leg side,

[- 1a.c.e Evaluating the contact pmssum expmssion, a,c.e l

l l

l (10) 4-12 l

l

Results on the hot leg side were obtained for each of the tubesheet elevations considered. Equation (1) was used to determine the combined displacements [

]. Energraphics was used to fit founh order polynomials to the displacement vs radius data. The change in hole diameter was obtained using equations (4) and (3b), and the contact pressure between the tube and tubesheet was calculated using equation (10). 'Ihe contact pressure between the tube and tubesheet is plotted versus radius in Figure 4.3 2. Through thickness .,

variations at radii of 2.0,14.8, and 32.8 inches for the displacement, change in hole diameter, and contact pressure between the tube and tubesheet are listed in Table 4.31, The contact pressure is plotted in Figure 4.3-3. Note that the tube is in contact with the tubesheet for the entire length within the tubesheet.

Faulted condition results on the cold leg side for contact pressure between the tube and tubesheet is plotted versus radius in Figure 4.3-4. Through thickness variations at radii of 2.0,14.8, and 32.8 inches for the displacement, change in hole diameter, and contact pressure between the tube and tubesheet are listed in Table 4.3-2 and the contact pressure is plotted in Figure 4.3-5. Note that the -

tube is in contact with the tubesheet for all but the top two-thirds of an inch ofits length within the tubesheet.

Nonnal Condition The Normal condition to be analyzed is:

[

ja.c.e The diametral thermal expansion of the outside sudre of a tube on the hot leg side is given by:

4-13

a,c.e 4

J m

Figure 4.3-2 Tube /rubesheet Contact Pressure for Faulted Conditions ---

Het Leg 4-14 s- e _

a,c,e Table 4.3 1 Faulted Condition Displacements and Contact Pressure Variations Through the Tubesheet Thickness (Hot leg) 4-15

)

a,c.e -

e Figure 4.3-3 Through Thickness Tube /rubesheet Contact Pressure for Faulted Condition - Hot Leg 4-16 wka ,,.,y

a,c.e Figure 4.3-4 Tube /fubesheet Contact Pressure for Faulted Condition -

Cold Leg

.4-17

. ~ ~

- 8,C,e I

c 4

. Table 4.3-2 Faulted Condition Displacements and Contact Pressum Variations Thmugh the Tubesheet Thickness (Cold Leg) 4-18

h a,c.e r'

Figure 4.3-5 Through Thickness Tubefrubesheet Contact Pressure for Fauhed Condition - Cold Leg 4-19

ADrube(hot) = (0.890) (7.82 x 10-6) (608.5 -70)

= 0.003748 in.

The diametral thermal expansion of the outside surface of a tube on the cold leg si le is given by:

ADrube(cold) = (0.890) (7.82x10-6) (546.5 - 70)

= 0.003316 in.

The radial expanrion of the outside surface of a tube caused by pressure is:

a,c.e So that the diametral expansion of the tube on the hot leg side during the Normal condition is:

[ 3a.c.e and on the cold leg side,

[ ]a.c.c The contact pn:ssure between the tube and the tubesheet on the hot leg side is plotted versus radius in Figure 4.3-6. Through thickness variations at radii of 2.0,14.8, and 32.8 inches for the displacement, change in hole diameter, and the contact pressure are listed in Table 4.3-3 and the contact pressure is plotted in Figure 4.3-7. Note that the tube is in contact with the tubesheet for the entim length within the tubesheet.

The contact pressure between the tube and the tubesheet on the cold leg side is plotted versus radius in Figure 4.3-8. Through thickness variations at radii of 2.0,14.8, and 32.8 inches for the -

displacement, change in hole diameter, and the contact pressure are listed in Table 4.3-4 and the contact pressure is plotted in Figure 4.3-9. Note that the tube is in contact with the tubesheet for the entire length witnin the tubesheet except for a small region at the center.

4-20

a,c, e -

5 4

m Figure 4.3 Tube /rubesheet Contact Pressure for Normal Condition -

Hot Leg 4-21

a,c.e t

P Table 4.3-3 Normal Condition Displacements and Contact Pressure Variations Through' the Tubesheet Thickness (Hot Leg) 4-22 r*vr - _ _ _ _ _ - - _ _ _ - - . _ _ - _ _

a,c.e I

Figure 4.3-7 Thmugh Thickness Tubeffubesheet Contact Pmssures for Normal Condition - Hot leg 4-23

a,c.e t

}

Figure 4.3-8 Tube /Tubesheet Contact Pressures for Normal Condition -

Cold Leg 4-24

a,c,e b

^

Table 4.3-4 Normal Condition Displacements and Contact Pmssure Variations Through the Tubesheet Thickness (Cold Leg) 4-25

a,c.e

'l

)

Figure 4.3-9 Through Thickness Tubefrubesheet Contact Pressures for Normal Condition - Cold Leg  !

4-26

4.4 CALCULATION OF PULLOUT LENGTH I

The PLRL has been denned as the length of sound tube engagement in the tubesheet necessary to resist the analysis axialloads resulting from the pressure heads. The axial resistance to pullout is due to the force resulting from the radial contact pressure b: tween the tube and the tubesheet and the friction coef6cient between these two components.

The radial contact pressure between the expanded tube and the tubesheet during steam generator operation includes radial contact pressures related to the thermal expansion mismatch (0.40 E-6 in/in/ F) between the alloy steel tubesheet and the nickel base Alloy 600 tube at approx'imately 600 F, and to the differential pressure across the tube wall. Any radian interference contact pressure resulting fmm the WEXTEX expansion process also contributes to the total radial contact pressure. The WEXTEX related interference pressure may be due to a combination ofinterference pressure between the tube and tubesheet and/or lockup interference which may result from the tube OD surface flowing into grooves in the tubesheet ID during explosive expansion.

To determine the engagement length necessary to react the axial loads it is first necessary to define the levels of the operating radial contact (interference) pressure and the expansion process rdated radial contact pressure. The steam generator operating radial contact pressures are determined through calculation and the WEXTEX fabrication relatec interference pressures are deterrrdned from the al-e pull tests results. Both are discussed in the following sections.

4.4.1 Operating Contact Pressures Radial contact pressures resulting from typical normal and faulted operating conditions were determined in Section 4.3. From these data the contact pressme profiles at several radiallocations across the tubesheet were selected for use in the PLRL determinations. The contact pressure profiles used for the hot leg and cold leg at normal and faulted operation are shown in Figures 4.4 -

1 through 4.4-4. These profiles are for elevations referenced to the bottom of the WEXTEX transition,i.e., approximately 0.3 inches below the top of the tubesheet. The 0.3 inch reference position for the bottom of the WEXTEX transition (BWT)is based on the location of the BWT in the samples fabricated for this test program. The location is generally consistent with other WEXTEX samples and is close to the BWT location in the Sequoyah Unit 1 & 2 tubes.

4-27 4

i I

a,c.e Figure 4.4 Contact Pressure for Faulted Conditions on Hot Leg Side as a Function of Depth from BWT and Radius -

4-28

a,c.e I

M Figure 4.4-2 Contact Pressure for Normal Conditions on Hot Leg Side as a Function of Depth from BWT and Radius 4-29

1 e

a,c.e Figure 4.4-3 Contact Pressure for Faulted Conditions on Cold leg Side as a Function of Depth from BWT and Radius 4-30

a,c,e

,[

d-Figure 4.4 Contact Pressure for Nornn! Conditions on Cold Leg Side -

as a Function of Depth from BWT and Radius 4-31

i 4.4.2 WEXTEX Radial Contact Pressure and Friction Coefficient

)

I The magnitude of the WEXTEX radial contact pressure and the break-away friction coefficient are -

determined from tne test data shown in Table 4.2-1 and the following expression relating pull force to tube geometry and contact pressure. The pull force (PF) can generally be expressed in terms of friction coefficient and contact pressure as follows:

a,c.e (11)

The L ee and F ee terms are related to the fact that there are end effects in radially loaded tubes if there are material discontinuities such as those at the transition from expanded tube to unexpanded tube or at other locations. The stiffness and radialload of the tube is rtduced relative to that remote from the ends. The analysis of end effects in thin cylinders k based on the analysis of a beam on an clastic foundation. For a tube with a given radial deflection at the end, the deflection of points away from the ends relative to the end deflection is given by:

[ (12).

4-32

Jas,e For the radially Ivaded tube the distance for the end effects to become negligible is the location where the cosine tenn becomes zero. For the WEXTEX pull force test samples the distance corresponds to ' A" times "x" being equal to [

jax,e Ja.c.e Considering end effects and ad. lusted engagement length for the test samples, the friction coefficient and WEXTEX Sr were determined for Samples M f-007A and W4-006 in Table 4.21 by solving the above equation. The radial contact pressures use to thermal expanslut mismatch >

and AP across the tube wall used in the solution were determined by the sa ne method used to detennine operating contact pressures noted in Section 4.3.

The method of solving for p and SrW fmm equation 11 is shown in the following example for pull test sample W8-007A,

[

4-33

}a,C e b) At 600'F/1620 psi, Pull Foxec = [ ]*** pounds a,c.e Similar determinations m'p and SrW svere made for all tests and the results are summarized in Table 4.41. The differences in fricdon coefficients between samples W4-006 and W8 007A may be related to the collar ID surface iinish. While both were fabricated within the nominal surface finish range, Sample W4 006 had a rougher and less uniform surface finish than the other sample.

Since the tube OD surface flows into any grooves in the tubesheet ID during WFXTEX expansion, it is conceivable that a tubesheet or collar ID with more grooves would provide more anchor points and a higher apparent friction coefficient.

4.4.3 W* Pullout Length As a conservative appmach for determining the engagement length necessary to resist tube pullout, the p and SrW combination from the pullout force tests giving the larger PLRL, i.e., [

] axe was used in the calculations. Since the contact pressure between the tube and TS varies acmss and through the tubesheet as a result of Iabesheet deflection from the primary to secondary pressure differectlal, the engagement length wa determined by dividing the analysis load by the integrated axial resistance load (contact pressure ube OD circumference x coefficient of frictioti) at increasing depths in the tubesheet. All engagement lengths were determined fmm the 4-34

l l

8.C,e t

Table 4.4-1 Tube-to-Tubesheet Friction Coefficient and WEXTEX Radial Contact Pressure Detennined from Pull Force Tests 4-35

4 l

bottom of the WEXTEX transition which is approximately 0.3 inches below the top of the tubesheet.

The combined contact pressures across the tubesheet for the faulted and normal conditions, and hot leg and cold leg, are shown in Figures 4.4 5 through 4.4 8 for depths to eight inches below the bottom of the WEXTEX transition. The combined contact pressures include re.c trom operating conditions and those trlated to the WEXTEX process. ,

The pullout length for normal o}cration conditions was detennined at twelve radiallocations acmss the tubesheet. The PLRL was taken as the depth in the tubesheet whert

[ Jaee The value [

ja.c.e The > esulting pullout lengths, or PLRLs, across the tubesheet radius are shown in Figure 4.4 9 for the .:nt and cold leg for typical normal operating conditions and ir. Figure 4.4-10 for faulted conditions. The maximum hot leg PLRL occurs for normal operating conditions and is approximately [ las.e inches. The maximum PLRL occurs at the center of the tubesheet and decreases with radial location. These pullout lengths were d:Wned assuming that the tube had a uniform expanded diameter over its full length up to the WE) TEX transition and without consideration of end effects. The effect of the taper in tne tube diameter, which occurs fmm the 4

BWT to about 0.7 inches below the BWT (Section 4.2.1), is to ircrease the PLRL by about

[ la c.e inches across the tubesheet. inclusion of end effects for the tube discontinuity at the bottom of the WEXTEX transition and for an assumed discontinuity at the PLRL depth, increases the engagement length by less than [ las,e inches. Therefore, correcting the PLRL for end effects and taper eficcts adds about [ ] ate inches to the required engagement length.

4 36

a,c.e r '?

Figure 4. 5 Contact Pressure Used in PLRL Detemiination for Faulted Condition on llot leg Side. Includes Faulted Condition and WEXTEX Contact Picssures, I

4 37

a,c e t

i t

I l

P 7

4 t

L Figure 4.4-6 Contact Pressure Used in PLRL Detennination for Nonnal Operation on Hot Leg Side, Includes Normal Operation and '

WEXTEX Contact Pressures.

4-38

..__.:;~,___. . . - , .. . . , .

l l

R.C,e Figure 4,4-7 Contact Pn:ssure Used in PLRL Determination for Faulted Condition on Cold Leg Side. Includes Faulted Condition and WEXTEX Contact Pressures.

4-39

l

?

-i a,c.e .

1 t

i e

b

?

t f

l t

f Figure 4.4 8 ' Contact Pressure Used in PLRL Determination for Normal Operation on Cold Leg Side. Includes Normal Operation and WEXTEX Contact Pressures.

i i

l 4-40 l I

' ' * " * ~'y , = .. ,, . . . , , , , , . . . . , , , _

i n,c e l t

r 1

i t

.i r

1 I

fi 1

.. i Figure 4.4 9 Pullout Load Reaction I.ength for Normal Condition at Ilot Leg and Cold leg. Length Shown is from Bottom of j WEXTEX Transition.

4-41

, , . , , . . - , . , . - - , a ., - - . . , . -.

.. .. .- -. . -. . - . . - - - . . . - _ . . . _ ~ .

- - . . .. . . _ .-~. .. -- -..

a,c,e  ;

i I

T e

Figure 4.410 - Pullout Loaz ' eaction length for Faulted Condition at Hot Leg and Cold 12g. Length Shown is from Bottom of WEXTEX Transition.

4-42

. - .-. -.m y _. e --- , ., .. ,. ,r. .,e-

,.-- , ..m.. ,-

111storically, cracking in WEX" TEX tube to tubesheet joints has been observed only in the hot leg.

Therefore it is appropriate to use the above hot leg PLRL'to define W* for the hot leg rather than universally applying the more conservative cold leg PLRL. The hot leg PLRL, with the additional

[ ]8 c.e inch for tube taper and end effects, becomes the W* length for the Sequoyah Unit 1 & 2 steam generator hot leg. The hot leg W* distance is shown in Figure 4.411 as a function of radial position in the tubesheet and is also provided in Table 4.4 3. In the event cracking is c.bserved in the cold leg, the cold leg PLRL, with the [ la.c.e inch taper plus end effect factor, becomes the W* for the cold leg. Eddy cunent uncertainties for measuring the distance from the BWT to the W* distance r..ust be added to the W* length for a plant specific application.

4.5 ZONES FOR W* 1.ENG'Ill Eddy current inspection of tubes to the tubesheet radius-dependent W* depth shown in Figure 4.4-1I would result in the minimum amount of inspection but could be cumbersome from a data management standpoint. Therefore, a decision was made to define two W* length zones corresponding to the zones currently in use at Sequoyah Units 1 & 2 for inspection of the expansion transitions. Zones 1,2, and 3 corresponding to the expansion transition inspection zone are grouped into W* Zone A, and Zone 4 of the expansion transition inspection zone corresponds to W* Zone D. As seen in Table 4.4 3, the W* length for W* Zone B, nearer the center of the tube bundle,is [ ]s.c.e inches and the W* length for W* Zone A is [ ,

]as e corresponding to abe W* length of the tube the least distance from the centerline of the SO in the zone Although not intended to be implemented, ths. W* 1engths on the cold leg are

[ Ja.c.e inches in W* Zone B and [ las,e inches in W* Zone A. The expansion transition inspection zones are shown in Figure 4.51, Inspection can be limited to the maximum W* length calculated within these two zones, which will result in reduced inspection time relative to a single W* length criterion.

4-43

a,c.e ,

{

1 Figure 4.411 W* Iength for Sequoyah 1 & 2 Measured from Bottom of -

WEXTEX Transition (B\\T) 4-44

. _ . _ - _ , _ - . - - = _

- . - = . - _

a,c.e

)

4 i

i Table 4.4 3 W* Length for Sequoyah I & 2 Steam Generators 4-45 i

+.,4 i - rn.., .,--r,. -i,.-,, ,-~,,,,.w ,c- ,r, - - , - - - - ---v, ,-, ,,-c - , , , - + ~ - , . - - ---,r-,--~

4 i i 4

'l I

i t

4 1.

cn 4

t I-r 1

4 s

1 I

se w

Figure 4.5-1 W* Zone A Corresponds to Expansion Transition Inspection Zones 1,2, and 3; W* Zone B_ to Expansion Transition Inspection Zone 4

5.0 DEGRADED TUBE STRENGTil 5.1 EVALUATION MEDlODS AND LOADS Cracking related tube degradation in WEXTEX tubesheet regions has occurred prinutrily as single circumferential and single axial or near axial cracks. Circumferential bands of cracks, defined as multiple cracks occurring at the same axial elevation, have not been observed in the WEXTEX expansion region hut are judged to be a feasible degradation mode. At the time of this repon, few circumferential cracks have been identified at distances away from the WEXTEX transition.

Degradation below the WEXTEX transition,if present, could occur in single bands, referred to as single band degradation (SBD), or in multiple bands (MBD). An illustration of the degradation bands is given in Figure 5.1-1. Undegraded sections of tubing within the tubesheet region are referred to as a sound WEXTEX expansion region.

For single axial cracks there is no impact on the axial pull strength of the tube. For circumferential cracks, however, the portion enhe circumference that is uncracked must be sufficient to carry the axial load if tube failure is to be avoided within the W* region.

Should SBD or MBD occur within the W* region, the axial loads on the tube are assumed to be applied directly to the degraded bands. The degraded region must therefore have sufficient suength to transmit the loads to sound regions of the tube if tube rupture above the W* length is to be avoided, in the most conservative case, i.e., no axial restraint attributed to the degraded expansion, the applied axial loads are considered to be reacted by the aggregate length of sound tube ponions above and below SBD, and above, between and below MBDs. While the axialload on each successive degradation band decreases because of axial resistance afforded by the prior sound expansion regions, all degraded regions in this evaluation are assumed to experience the total axial load.

De evaluation of degraded tube strength for the WFXIT.X tubes included analysis of the response of degraded regions of tubing to axial loads based on data from previous tests performed te determine the axial load bearing capability of degraded tube-to-tubesheet rolled joints (Reference 8). De referenced tests were performed to determine the tensile strength for non-degraded and degraded 0.75 OD x 0.043 wall Alloy 600 MA tubes rolled into AISI 1018 carbon steel collars.

Some additional degraded samples were wsted in the unexpanded condition and others were tested in the decollare ~ condition. Throughwall degradation was simulated by electric discharge machined (EDM) slots in the tube specimens. The slots were [ la.c.e inches long and were 5-1 l

a,c e o

I b

i Figure 5,1 1 Illustration of Degradation Bands Within Tubesheet Region 52

machined at an angle of [ ]833 degrees from the axial centerline of the tube. The number of slots in the tubes was either l ]"e e.

The evaluation of degraded tube strength considered the most stringent axial load on the tube for the nonnal,ILB or LOCA condition.

5.1.1 Normal Operation loads The ultimate axial load to be borne by thc tube under normal operation conditions is approximately

[ lax.e. This considers the load to result from the pressure heads acting on the tube OD cmss section. A safety factor of three is applied to this load so the tube axial load bearing requirement for nonnal operation is [ Jax e pounds. Tids represents the most stringent axialload on the tube for all conditions considered.

5.1.2 Feedline Break The ultimate axial load to be borne under 11B conditions is approximately [

]a.c#. A safety factor of 1.43, corresponding to an ASME Code factor of 1.0/0.7 for allowable stress for faulted conditions,is applied to this load. Therefsre the axial loadbearing requirement for l~LB is [ ]*e.' pounds. As discussed in Reference 2, this load need not be further aug iented for dynamic loading effects during FLB because of the sequence of events during an FLB.

5.1.3 Loss of Coolant Accident An axial load from a LOCA event acts to move a tube downward. Thereff.e this condition does not applv a tensile load to the tube and is not considered further.

5.2 ANALYSIS OF DEGRADED TUBE STRENGTil The strength of a degraded tube is dependent on the crack geometry as well as the number of cracks and the angle of the cracks relative to the tube axis. In addition, the effect of the degradation on the tube strength is also dependent on whether the crack array is symmetric or nonsymmetric about axes parallel with the tube axis.

53

- - _ _ _ - - - - _ _ _ _ _ _ ~ _ _ _ _ _ _ - _ _ _ _ . _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ - _ _ _ _ _ _ _ _ _ _ - _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

5.2.1 Symmetric Cases Axial and circumferential cracks can combine to produce a variety of crack networks in the tubesheet region. Figure 5.21 illustrates a range of crack arrays symmetric about axes which are parallel with the rube axis. In these cases no ligament bending effect is present and a net cross section approach to axial pull strength is applic 21e. For a WEXTEX tube with an expanded ,

diameter of 0.890 inch, wall thickness of 0.049 inches, a lower tolerance limit of ultimate strength of[ Jaxe psi (Ref. 9), an axial load requirrment of [ ] ate po"nds (from Section 5.1.1),

ar.d a conserutive [ Jaxe throughwall degradation in the remaining portion of the tube wall (see Figure 5.2 2), the maximum allowable end of-cycle circumferential crack is [ Jaxe, 5.2.2 Nonsymmetric Cases Figurr 5.2 3 illustrates cases of slemed crack arrays which are not syminetric about axes which are parallel with the tube axis. In these cases, the transrission of axial force across the crack array creates a bending moment in the ligaments of material between cracks and lowers the load required for plastic collapse. The straight star'ted crack case, Figure 5.2 4, was presented in Reference S.

The load required to yield the slanted crack array is given by a,c.e (12)

The variables for the above are shown in Figure 5.2 4. The axial plastic displacement resulting fenn yielding and then rotation of the crack array is ,

a,c.c (13)

If the crack array plastic displacement, as a function of axial load,is added to the baseline load-displacement record of an otherwise identical tube without a crack array then the computed load- ,

displacement record closely approximates actual measurements as illustrated in Figure 5.2-5.

i

+

5-4

I a,c.e

.i.

8 Fi t;ure 5.2-1 Sketch ofIdealized Symmetric Crack Arrays 55

a,c,e Figure 5.2-2 Crack Model for Allowable Circumferential Degradation 5-6 N- - --- - . - - - - .

a,c,e Figure 5.2 Sketch ofIdealized Non Symmetric Crack Arrays 5-7

8 6 a,c.e  ;

i f

r i

Figure 5.2-4 Model for Plastic Collapse of Slanted Swaight Crack Array 5-8

.,--+....--.:-..-..,n.~,.-.L'---

- a ,C .C 1

l f

Y Figure 5.2-5 Comparison of Load-Displacement Records, Computed Versus Measured for 30 S! cts at Q = 45' 59

Laboratory measurements of the axial pull strength of the rolled 0.75 inch OD tubes showed that ,

the maximum pull strength is far above the yield load when substantial numbers of slanted cracks j are present (Reference 8). The presence of tbc slanted cracks leads to early yielding as the  !

ligaments between cracks deform in bending. As plastic deformation proceeds, rotation of the  !

slanted crack network occurs, and the tube rotates while becoming longer. As rotation occurs the '

effective moment arm for bending of the ligament between cracks decreases and plastic ,

deformation becomes more difficult. This is an example of geometric hardening. Maximum strength is reached when tearing at the crack tips occurs.

The onset of crack tearing depends on the cracked body toughness of the material, in the 3sts of the rolled tubes with slanted slots (Reference 8), the maximum pull strength correlated reasonably well with a J integral value of[ ] ate in lb/in. Applied J values can be obtained from computed load versus plastic displacement definition of J. The high toughness and small scale geometry of cracked Alloy 600 tubes makes neglecting the clastic loading a reasonable approach in the estimation of madmum pull strength.

If the cracks " hook over" in the circumferential direction as illustrated in Figure 5.2 6, the axial pull strength is decreased. This decrease is somewhat more than that obtained by an incarase in the effective slant angle. The crack hook decreases the net cross sectional area for tensile yielding and the size of the ligament subjected to bending. Using the same approach as the straight slant problem, the plastic strength of the hooked crack array is given by a,c.e (14) and the plastic displacement is given by a,c,e (15) where u is the ratation of the crack array as deformation proceeds. The above equations are for singly hooked cracks; the variables am shown in Figure 5.2-6. For doubly hooked cracks, top and bottom, the pull strength is given by 5-10

. . . . _ . _ _ _ _ . . _ _ _ _ _ _ . _ _ _ . _ _ _ _ _ = . _ . _ . . _ _ _ _ . . _ _ . _ _ _ _ _ _ _ _ . _ . _ - . . . . - . . ._

t I

b

- _ 8,C ,: 5 L

i L

P s

L d

Figtre 5.2-6 Model for Plestic Collapse of Singly Hooked Slanted Cracks -

S-11 ,

?

x,, . . . - . . , ,,_. .w,. 4... -, - - _ . . _ ,_,. . , . . ,,.m_.._,.. . . . - , , . . , , . - - , _ . . . . , - - . . . . . . , ~ . . , . . . . . . _ . . . . . ~ , . .

l a,c.e (16) and the plastic displacement is then a,c.c (17)

The parameters in these equations are defined in Figure 5.2 7.

Figure 5.2 8 is a plot of yield load versus the hook length of slanted cracks for 0.875 OD by 0.050 inch wall tubing. The results are applicable to the tubesheet expansion zone of WEXTEX tubes.

Since the double hook crack geometry is more limiting than the single hook case, Figure 5.2 8 bounds the slant crack cases.

As noted earlier, the yield loads of slanted crack arrays is substantially below measturd maximum loads. A very conservative approach to axial pull strength would be to accept crack geonttries on yielding eiteria. Thus all geometries above the [ ]a,c.e pound level (See Section 5.1.1) in Figure $. 8 would be acceptable.

A less conservative but still realistic approach to pull strength of degraded tube is to estimate the onset of crack tearing as this defines the maximum pull strength. On this basis, with a critical J integral value of about [ ]a.ce in lb/in2, all crack geometries above the 1000 pound yield load level would be acceptable as crack tearing o!!these geometries would not occur until the axial pull force increases beyond [ ]*.c.e pounds. Figure 5.2 8 thus illustrates acceptable numbers of cracks, crack angles and hook crack length which meet a minimum pull strength of [ la.c.e pounds.

i 5-12

..- . . . - _ _ - - - - - . - - .-- - _ -. . . . . - .-. . _ .. - - . .- -. - .. -. _ _ . .. . ~ _ _ _ .- .

i i

a,c e i

i p

Figure 5.2-7 Model for Plastic Collapse of Doubly Hooked Slanted '

.i Cracks J

5 13 x.N.,----,, sen,.-er m. N r,~ ..,--,,w..-- ~w ,,. ,.e -,,e - - . + - , s.n, r

(

a,c,e -

Figure 5.2-8 Yield Load Versus Hook Length of Slanted Cracks .

5-14 d

. , , - . - . .- , , _ . . . ..~

5.3 LIMITATIONS ON ALLOWABLE DEGRADATION 5.3.1 Circumferential Cracks The tube must have sufficient net cross section to bear the apphed axial loads if structural failure of the tube it to be avoided. The annlysis in Section 5.2.1 indicated that a tube opersdng under normal -:onditions and which, wie the W* region, has a throughwall circumferential crack

[ ] sac long and the balance of the tube wall [ ]as,e degraded, would not separate under the applied axialload of appmximately [ Jas.e pounds. The normal operating condition is limiting in that it has the higher applied load and lower radial contact pressure.

In the case where a circumferential crack regian may consist of a series of small cracks separated by uncracked ligaments, the crach length to which the [ ]as,e limit applies is the length of the cracked region less any uncracked ligaments within the region.

5.3.2 Axial Cracks Axial cracks in WEXTEX tubes are generally single indications as opposed to bands of axial cracks. Single axial or single near axial cracks in the W* region do not affect puilout load capability, since axial 1~. b m transmitted through the tube wall past the degraded portion of tube.

Therefore, for single axian m agle near axial cracks located in or bel a & W* region, there are no structural limitations on crack length.

Therefore, a conservat.ve model was developed consisting of up to five axial cracks in a band at a commor. elevation. Figure 5.2-8 indicates that an array of up to five cracks in which the cracks are oriented less than [ Jas.e from vertical would meet the axial load requirement of approximately

[ ]a,c.e pounds. The circumferential extent of five axial cracks oriented [ ]as,e to the tube -

axis is approximately [ ]as,e assuming the cracks are appropriately spaced. A bounding assumpdon is that a five crack limit can be interpreted as five cracks having less than [ ]ae c enveloping circumferential extent at the beginning of cycle (BOC).

In the event there are bands of multiple axial cracks containing more than five cracks, or cracks orierte i at angle; of greater than [ ]as,e to the tube axis, the allowable crack azTay which would not affece ube integrity can be determined from Figure 5.2-8. Multiple bands of cracks have no additional impact on tube strength as it is assumed each band sees the same axialload. They are evaluated in the same manner as single band degradation.

5-15

For simplicity in defindng allowable axial crack degradation within the W* length, the following limits are applied:

• For axial cracks having the upper crack tip below the BWT and below the top of the tubesheet no limits on crack length are applied.

• Cracks separated axially along the W* leagth are acceptW2. 'ne limit applies only to closely spaced cracks at an angle to the tube axis. Consequently, a simplified limit of five parallel axial cracks at any tube elevation within W* is applied for the W* criteria. If these cracks are inclined from the tube axis, the total circumferentially oriented angle summed over all five cracks at the beginning of cycle must be less than [ ]a.c.c. Similarly,if closely spaced axial cracks cannot be resolved separately, the circumferential involvernent between RPC amplitude " null points" must be less than [ ]a,c.e, 5.4 W* ADJUSTMENTS FOR TUBE DEGRADATION The fact that cracks have formed in the WEXTEX expanded region of the tubes suggests that at least locally there are regions of the tube ID which are under tensile stress rather than compressive stress. The pre.ence of a tensi!c stress also implies that there is no significant interference between the tube and tubesheet and it cannot be assumed that there would be significant axial resistance to tube pullout loads at these regions.

Therefore, if a crack occurs in the W* region it is assumed that the axial resistance to applied loads is effectively reduced. While the effect of the crack on axial resistance may be localized, it is conservatively assumed that a full circumferential region of the tube having a length equal to the axial length of the crack is not active in contributing to the axial restraint of the tube. As discussed in Section 5.0, the region will however transmit the axialloads to the sound portion of the tube below the crack. The total length necessary to react the postulated axial load is then the length of the undegraded region of the tube above the crack plus some length of sound, undegraded tube below the crack.

For the highest precision, the method of accounting for the non-active region of the tube (in terms i of providing axial restraint) would be to include *he length of the crack plus appropriate allowance for end effects (See Sections 4.4.2 and 4.4.4). Calculations of end effects for cracks at various -

locations below the WEX" REX transition, show that for most of the tubesheet,i.e., fmm the center 5-16

of the tubesheet to approximately the mid radies of the tubesheet, both the crack length and-associated end effects can be adequately accounted for by simply adding the crack length (or the sum of the enveloping lengths should groups of cracks be present) to the pullout load reaction length. To simplify data management for TVA, this requirement is applied to W* Zone B (which corresponds to expansion transition inspection Zone 4.) For W* Zone A (which corresponds to expansion transition inspection Zones 1,2, and 3), increasing the PLRL by 1.2 times the crack length (or the sum of the enveloping lengths should groups of cracks be present) accounts for the crack length and end effects. The fact that the increase in PLRL is less than the crack length plus end effects is related to the variation in contact pressure with depth in the tubesheet.

l l

l 5-17 l

6.0 LEAK RATE EVALUATION 6.1 EVALUATION ? TrliODS AND LOADS Extensive European operating experience with axial PWSCC cracks left in service has been summarized in Reference 1. This operating experience has demonstrated negligible normal operating leakage from PWSCC cracks even under free span conditions in the roll transitions.

PWSCC cracks in WEXTEX expansions in the tubesheet region would be funher leakage limited by both the tight tube-to-tubesheet crevice and the limited crack opening permitted by the tubesheet constraint. Consequently, negligible normal operating leakage is expected fmm cracks within the tubesheet region of WEXTEX expansions.

The steamline break (SLB) condsions provide the most stringent radiological hazards for postulated accidents involving a loss of pressure or fluid in the secondary system. To establish the leak rate criteria for faulted conditions, the SLB leakage rate is used with the feedline break p' essure differential. The evaluation of SLB leakage includes calculation of leak rate for all cracks left in service. The total leakage, i.e., the combined leakage for all such tubes must not .ceed the site specific radiologicallimits at the site boundary.

The evaluation method used for assessing leakage from cracks in the W* region considers the leak rate across the tube-to-tubesheet interface to be a function of the differential pressure across the tube wall, the radial comact pressure between the tube and the tubesheet, and crevice length. The crevice length is taken as the distance from the BWT to the tip of the crack. The radial contact pressure is calculated as described in Section 43 for the operating conditions being analyzed. The leak rates for the WEXTEX joints are determined through testing of prototypic WEXTEX tube-to-tubesheet joints and the test data are used to generate a leak rate model that will permit the calculation of leak rate for a givers crack length at any location within the tubesheet. From this '

model leak rates can be determined for cracks within the W* region, and possibly a distance within the W* can be defined below which leakage will be negligible. Below this distance leakage would not hav. to be considered in the SLB leakage limits.

The loads considered in the analysis are those resulting from nonnal operation, feedline break and LOCA as described below.

6-1 l

6.1.1 Normal Operation The differential pressure between the pdmary side and secondary side during normal operation is 1368 psi, i.e., 2250-882 psi. As described in Section 6.1 negligible leakage is expected from cracks under normal operating conditions and no W* requirements must be applied to limit operating leakage. Some leak rate tests were conducted at nonnal operating conditions to provide a more comprehensive leak rate data base from which to develop the WEXTEX leak rate model.

6.1.2. Faulted Condition Feedline Break The primary to-secondary differential for the FLB condition is 2650 psia. Dynamic loads in the tube joint, i.e., secondary side fluid drag on the tube during the postuls.ted accidents need not be added to this value. This is because the safety valve relieving event and the maximum fluid drag event are not concurrent. The axial load during the leak test .as prototypical because it is the end cap load caused by the prototypical differential pressure. The FLB pressure diffe ential is used for the evaluation of steam line break leakage.

6.1.3 LOCA The secondary-to-primary differential pressure for the LOCA event is about 964 psi. A temperature equal to the normal operating temperature is conservative for analyzed LOCA conditions. Leakage during a LOCA event will be negligible based on the European operating experience with free span cracks. In additior., the secondary-to-primary differential pressure will tend to close cracks and restrict leakage.

6.2 WFXTEX EXPANSION LEAK RATE TESTS 6.2.1 Sample and Test Description The samples for the leak rate tests were designed to simulate the WEXTEX tube-to-tubesheet interface characteristics. The samples consisted of a carbon steel collar approximately [

]b,c.e The tube specimens were fabricated from Quality 6-2

?

Assurance (QA) controlled stock and fabrication of the collars was done under QA surveillance.

WEXTEX expansions were fabricated per approved procedures.

The collars used to simulate the tubesheet were [

s

)b,c.e The WEXTEX samples fabricated for leak rate testing were of the configuration shown in Figure 6.21. The sample was rolled and welded at one end prior to WEXTEX expansion to create a seal at one end of the tube to-collar crevice. 'As was the case with the pull for test samples described in Section 4.2, the as-fabricated leak rate test samples had smaller diameters between the BWT and the fully expanded region which starts about [

]b,c.e locations am given in Table 6.2-1 along with the average diameter over the fully expanded region. These data are used in the determination of the average contact pressure along the tube-to-collar crevice leak path.

6.2.2 Test Description Samples were tested at various temperatures and pressures to assess WEXTEX joint leakage at nonnal and faulted conditions, which are more stringent than the LOCA condition, and to develop .

a leak rate model. The tests were performed in accordance with an approved, referenceable test specification under QA surveillance.

The expanded test sample had a series of [

]b,c.e 6-3

t 4

3,C, C 6

Figure 6.2-1 As-Fabricated WEXTEX Sample for Leak Testing .

t>4

- a,c,e .

Table 6.2-1 Inside Dianeter of WEXTEX Leak Rate Samples 65

a,c,e

+

J 1

,~

I d

I Figure 6.2-2 Configuration of WEXTEX Leak Rate Samples 6-6

The room temperature leak tests were performed in a multiple station . test stand with sample pressurization through a manifold system. The ID of the sample was pressurized and the OD was-at atmospheric pressure. The leakage from each sample was collected and measured by counting

_ drops of liquid for small leak rate semples, or by measurement of the collected volume in milliliters

^

in the case oflarger leak rates. The test time, i.e., the length of time me leakage was collected, was ten minutes for the room temperature tests.

The elevated temperrture tests were also performed in a multiple station test stand with sample pressurization through a manifold. The collars were enclosed in a fumace and the test temperature was maintained over the full lengtn of the collar. The pressurized water supply system was also heated. The sample was mounted in the position shown in Figure 6.2-2 and a water cooled skirt approximately [

]b.c.e was necessary to maximize condensation. The water was collected as it dripped from the bottom of the tube and/or condensing skirt. The collection time for the elevated temperature tests was ten to twenty minutes depending on the leak rate.

6.2.3 Test Results i

The results of the tests are summarized in Table 6.2-2. The nominal crevice length shown is the distance from the top of the collar to the centerline of the leak path hole, less the WEXTEX transition length. To put the leak rate results in a form which could be used to develop a WEXTEX joint leak rate model, the test data was reduced to determine an effective leak path length (tube-to-collar crevice) and an average contact pressure over this effective leak path length.

The staking operation noted above deformed the tube in the vicinity of the through-wall hole and effectively shortened the crevice length. As shown in the Figure 6.2-3 example of test sample W4-018, the staking, combined with the radius of the drilled hole, decreased the crevice length by

[

]b.c.e For the RT-1620 psi test condition with a nominal 6-7

TABLE 6.2-2 WEXTEX EXPANSION JOINT LEAK RATE TEST RESULTS

,g i

e lO 6-8

- - - e .4

^

- TABLE 6.2-2 (CONT.)

WEXTEX EXPANSION JOINT LEAK RATE TEST RESULTS b,c e

~

6-9

TUBE ID DIAMETER SAMPLE W4-01 J O.S O.7'35 -

<n O.7:3 -

u ai .

0.765 - -

[

(r I I

w '

g 15 0.75 - j 5 I a t

a 9 0.775 - .

W co S 9,77 _ Ann M O.7 65 -

0.76 - i e i i i

-1 1 3 5 POSITIOf J MOM TOP.Of COLLAR - ITCHES Figure 6.2-3. ID Diameter of Leak Rate Sample W4-018.

crevice length of 2 inches, this sample would be estimated to have zero contact pressure at  ;

approximately [

]b.c.e The average contact pressum f or the test is the integrated contact pressure over the effective crevice length after accounting for the reduced contact pressure resulting from the smaller tube diameter.

End effects, which were discussed in Section 4.4.2, were also included in the determination of integrated contact pressure.

The reduced data from the WFXIEX joint leak tests are summarized in Table 6.2-3 and include the leak rate, the effective crevice length, and the average contact pressure over this length for each test. The average contact pressure includes that due to differential thermal expansion and primary-to-secondary differential pressure. Contact pressure due to the WEXTEX expansion is inherent in both the test samples and SG tubes and is not included in the average contact pressme. These data are used in the development of the leak rate model for the WEXTEX tube to-tubesheet joints (Section 6.4).

t 6-11 l

l

TABLE 6.2-3 WEXTEX EXPANSION JOINT LEAK RATES, CREVICE LENGTHS AND

, AVERAGE CONTACT PRESSURE

_ _ b,c 6-12

TABLE 6.2-3 (CONT.)_

WEXTEX EXPANSION JOINT LEAX RATES, CREVICE LENGTHS AND AVERAGE CONTACT PRESSURE-b,c en 6 13

6.3 CONSTRAINED CRACK LEAK RATE TESTS 6.3.1 Test Design and Description Auditional testing was performed as a pan of the Sequoyah Unit 1 & 2 W* program to pmvide an enhanced basis for the prediction ofleak rates from WEXTEX-expanded tubes. Previous leak rate models utilized the crack opening area of free span lengths of tubes, and did not take credit for the reinforcing effect which the tubesheet will have in limiting crack opening. The test program described in this section was performed to provide an empirical basis for [

)b,c

[

)b,c Figure 6.3-1 illust ates the initial test program test arrangement. Since the correlation between crevice depth (or length) and loss coefficient are known from the testing reported in Section 6.2,

[

]b.c 6.3.2 Sample Preparation and Test Assembly

[

6-14

P f-E a,c. e -

1 Figure 6.3 Constrained Crack' Leak Rate Test Arrangement 6-15

a,c.e Figure 6.3-2 Tube Specimens Utilized in Constrained Crack Leak Rate Tests 6-16

throughwall, resulting in leakage of the pressure cycling fluid. A thin foil sheath was placed n the ID of the tube over the crack location and a Tygon bladder was inserted into the tube, With the Tygon bladder and foil in place, pr:ssure cycling was contmued at gradually reduced pressures, consistent with the crack length, until fatigue crack lengtl*; acre achieved close to the EDM notch length.]h.c Because of [

]b.c into the collar, A single collar, shown in Figure 6.3 3 with [ l b,e holes p was utilized in the test. [

]b.c The calculated gaps are shown in Table 6.31.

6.3.3 Test Results and Conclusions Leak rate data were first obtained for free span (i.e., without a collar) conditions at room temperature and at a nominal 600 F temperature. [

]he Leak rate data were also obtained [

]b c leak rate results are presented in Table 6.3-2.

Analysis of the data was performed to determine the correlation of the measured leak rates versus ,

the DENTELO predictions. [

]b,c Figure 6.3-4 provides the measured versus predicted leak rates. The numbers "1", "2", and "4" correspond to the three tube 6-17

a,c.e .

5 i.

Figure 63-3 Alloy 600 Collar Utilized in Constrained Crack Izak Rate Tests 6-18

--v I

b,c,e

~

E T

G .

Table 63-1 Diametral Gap Sizes (mils) at Temperature and Pressure (Adjusted for Pressure and Thermal Expansion)

c.

c, -

b s

t l

u s

e R

t s

e T

t e

a R

k a

I r

e c

i v

e r

C d

e t

i c

t r

s e

R 2-3 6

l e

b a _

T T I:$

l 11

a,c e i

~

Figure 6.3-4 Restricted Crevice Measured vs. Predicted Leak Rate 421 i

samples (WP-001, WP 002, and WP-004, respecively.) The initial test program demonstrated that the WEXFLO (or DENTFLO) leak rate model provides a reasonably good representation of -

actual ,s. measured leak rates for cracks in crevices with gaps; WEXFLO utilizes essentially the -

free span crack opening model with adjustments to the AP across the crack needed to obtain continuity of flow from tube to crevice and from crevice to TTS. [

]b,c To complete the modeling of the fully expanded region below the taper zone, [

ja.c.e 6.4 LEAK RATE ANALYSIS MODEL Calculation of leakage from an axial crack within the tubesheet and below the WEXTEX transition uses established technology in the form of the DENTFLO Code (Reference 10). This code calculates the primary to secondary leakage rate through an axial crack with a flow resistance between the exit of the crack and the secondary side pressure. In the case of the WEXTEX joint, the flow resistance is the crevice between the tube OD and the tubesheet hole surface.

Application of the DENTTLO Code requires the input of a crevice flow resistance, characterized by an average loss coefficient. The loss coefficient, in turn, must be determined in terms of parameters which are known, or which can be calculated for the field tube-to-tubesheet interface crevice. In the case of the WEXTEX generators, the parameter used is the contact pressure between the tube OD and the drilled hole in the tubesheet.

6.4.1 DENTFLO Model Description The DENTFLO Code was developed to assess leakage through tube cracks within the tubesheet.

. Leakage is determined by crack leakage characteristics as well as the flow resistance of the crevice.

The code calculates the flow-pressure drop relationship separately for the crack and the tube-to-tubesheet crevice. By continuity, the flow for the two elements must be the sene. The fixed primary and secondary side pressures establish the total pressure drop. DENTFLC iterates to 6-22

determine the crack exit pressure (crevice inlet pressure) such that these flows are equal. The flow rate for this condiuon is the leakage rate for the crack-crevice configt' ration under consideration.

In DENTFLO, the model for crack leakage uses a crack opening calculation based on the methods of linear clastie fracture mechanics adjusted by plastic zone corrections to account tor yielding near the crack tips. With the crack geometry defined, flow can be calculated. A one dimensional, two phase, isenthalpic form of the momentum equation is used to relate leakage to the pressure drop across the tube wall. The final crack flow model, incorporating a minimal number of empirical assumptions, shows good agreement with laboratory test data and field data as reported in Ref.10, 6.4.2 Leakage Rate Model for WEXTEX Tubes 6.4.2.1 Axial Cracks The crevice leakage model r:lates leakage flow through the crevice to the pressure drop between the crack outlet and the secondary side pressure. The governing equation is proviad by Darcy's Law for laminar flow in porous media or tight crevices. Tne law is adapted to two phase flow by incorporating a two phase multiplier.

Solution of the Darcy equation requires knowledge of the WEXTEX crevice flow resistance. The Ims coefficient needed to chancterize the crevice flow resistance was developed from the results of the tests cf simulated WEXTEX tube to-tubesheet joints reported in Section 6.2. The results were summarized in Tables 6.2 2 and d.2-3. The leakage model was used in conjunction with the test data for leakage rate and test conditions to determine the loss coefficient. Figure 6.4-1 shows the log of the resulting loss coefficient data (K) plotted versus the average contact pressure for each sample in the test series. Those samples with zem leakage in the tests e;c not included in the figure as the flow coefficient would be infinity.

To apply the DENTFLO Code to the WEXTEX field analysis requires input of the WEXTEX tube-to-tubesheet crevice loss coefficient. The contact pressure between the tube anf the tubesheet, determined at each tubesheet location (see Sectica 4.3), is used in conjunction with the lower 90%

confidence limit of the mean segression line (Figure 6.4-1) to develop a localloss coefficient value.

The resulting loss coefficient can be integrated from the crack location at a depth in the tubesheet to the WEXTEX transition. Then dividing by the crevice length above the crack yields an average value for loss coeflicient required for input to DENTFLO.

6-23 l

a,c. e l

I t

Figure 6.4-1 Flow Coefficient Versus Tube-to-Tubesheet Contact Pressure ,

6 24

The probability (1/RY , of core melt with a combination single steam generator tube nipture and main steam line break event is calculated to be 2.5 x 10 8 in NUREG 0844, "NRC Integrated Program for the Resohition of Unresolved Safety Issues A-3, A-4, and A-5 Regarding Steam Generator Tube Integraj". Although this core melt sequence has not been subjected to a detailed scenario specific analysis of mdionuclide releases to the atmosphere, the accident consequences were determined using "CRAC" (a code prmided by Sandia National Laboratories). The risk of this event in terms of early and latent fatalities was detennined to be 0 and 1.1 x 10-5/RY, respectively. Based on a review of the sequences included within the NUREG, the NRC staff was able to conclude that the increment in risk assoc 6ted with SGTR event is a small fraction of the accident and latent fatality risks to which the general public is routinely exposed. -

The radiological consequences of SLB event analysis utilizing average leak rates for steam generator tubes in which the W* criterion has been applied are bounded by the results included in NUREG-0844 for an equivalent (or lower) core-melt sequence frequency. Based on a comparative study with another plant, given a SLB/FLB break event, the probability of the break times the probability that the operator fails to terminate the event typically is approximately 10 5 fay, Although steam generator tube rupture cannot occur in the region of a tube m which the W*

criterion has ocen applied, for purposes of comparison, assuming a probabih,/ of consequential single tube rupture of 2.5 x 10-2 and the probability that the Refueling Water Storage tank empties prior to depressurizing the RCS to atmospheric pressure (as stated in NUREG 0844), the probability of core melt for this event sequence (1/RY) is 2.5 x 10-t0. This sequence leads to a maximum pressure differential across the steam generator tubes of approximately 2335 psi versus 2650 psi. Based on the low probabilities for ther events, it is judged that utilizing flow coefficients based on 90% confidence on the mean regression fit is adequate to determine expected steam line break leakage at Sequoyah Units 1 and 2. The primary to secondary leakage is expected to be maintained below 1.0 gpm and the radiological consequences are expected to remain within a small fraction of 10CFR100 limits. Most certainly, the resultant risk to the public is significantly lower than that identified for a SLB + SGTR event.

The methodology applied in this report utilizes leak rates at a lower 90% confidence limit on the mean regression fit of Figure 6.4-1 at 2650 psi pressure differential and measured crack lengths increased by average growth per cycle and NDE uncertainties on crack length at one standard deviation. The 2650 pressure differential can only be achieved following a SLB event if the PORV fails to open in addition to the operator failing to terminate SI. The probability of core-melt for this sequence is 2.5 x 10-II/RY.

T

With crevice loss coefficient known as a function of crack location for ctifferent tubesheet locations, crack leakage can be calculated. DENTFLO was used in a parametric study to calculate leakage under faulted conditions as a function of crack depths of [

]a,c.e In this example, ,

leakages for these crack ranges were determined for five radiallocations on the tubesheet ranging from 23 inches which is a location of least contae: pressure, and therefore maximum leakage, to a radius of 42.5 inches. The contact pressures from differential themial expansion and primary to  !

secondary side pressure differentials at these locations are shown in Tables 6.41 and 6.4 2 for the i hot leg and cold leg at SLB conditions.  :

i  ;

t 4

The results of this parametric study are given in Tables 6.4-3 and 6.4 4. Cases run include various assumptions with respect to the radial location on the tubesheet and value of the crevice loss l coefficient. Five tubesheet radli were censidered, At each radius, the crevice loss coefficient was conservatively taken at the lower 90% confidence limit on the regression line as shown in Figure 6.41. For comparison purposes, the smalle:t radius (2.3 inches) was also run with loss coefficient set at the linear regression value of this ligure. The leak rates are plotted in Figures 6.4 2 through 6.4 7. These results are used in conjunction with crack characteristics obtained in field .

eddy current inspections to determine potential SLB leakage rates for the cracks. Application of the leakage rates determined from the DENTFLO model to the analysis ofleakage !n steam generators is discussed in the following section.

Leakage in the slightly tapered 0,7" long zone below the BWT is calculated utilizing DENTFLO and a nominal gap. A linearly increasing crevice based upon measurements reported in Section .

4.2.1 is assumed ", m the contact location shown in Table 6.41 (e.g., BWT 0.69", BWT-0.67",

etc.) to the BW1. Leak rates in the to ; ion from the BWT to 0.7" depth have not been included Tables 6.4 3 and 6.4-4 however, the calculation basis has been developed and is available for the field leak rate analysis.

1 6-26

a,c.c I

Table 6.41 Op rating Contact Pressure at Cold Leg for SLB Conditions 6-27 1

. . . . . . , , . - . , - , - ,, , , , - . , , , , . ~ , . - , , , . , . . , . . . . , - , , , ,,,n.-- , ... ...-., - , . . . . . . .. . . , . . - . - - , _ . , - . . -.

a,c,e l

. J l

-j Table 6.4 2 Operating Contact Pressure at Ilot Leg for SLB Conditions 6-28

U Y.

,. P-t e

a b

9

'm S

o b

.5 8

a i

s 6

n 7

n d

.N

.o 6-29

A_ = o( 4 -._-1a 4 .5._ w e-._A -_+-_,Aa+.h--- -awAMEL- - - - -,-a- l hhe -4_a _;w.-.2 L.m_4ra*-s,1 m.-a.2m ..Mhsa_ amm. m , --S- - - -

i U.

U.

. D i

A 9

bo

.i i

u=.

k u,

G 1 1 6-30

,, , - -- , w r - +-. ,, , ~ .> . -,> -- ., ,--

l a,c.e l

1 J

i i

b i

Figure 6.4 2 WEXTEX Faulted Condition Leak Rate, Regression Line Crevice Loss Coefficient, Hot leg 2.3" Tubesheet Radius 6-31

~ '

, . + , , -. , . - . , - . , . . . . _ . , . -

..,-v4 - _ . . - ..,.' , ,. - , . _ , - , . , . , , ~ - - . _ , ,. ,

a,c.e Figure 6.4-3 WEXTEX Faulted Condition Leak Rate for Lower Bound 90% Confidence Limit Crevice less Coefficient, Hot Leg, 2.3" Tubesheet Radius 6-32

. - - . . . . . - - . - . . _ . . - - . . - .- - - .. - . .-.....~-.. - - - .-._-. .. -.

J 1

I j

I

}

a,e e  :

l t

i I

I

[

P

'h Figure 6.4 4 WEXTEX Faulted Condition 1x&, Pa!c for Lower Bound 90% Confidence Limit Crevice Loss Coefficient, llot Leg,

- 12" Tubesheet Radius t

-t

..--..r--,,_~--.-.-,--- - - , - , , - . - r.-- . . , , < - . - . . ,-v,- .-w.,-._,_.. .. _ . . . . - . - - - _ - ....-.- _ - - - - - .

i I

i a,c.e 9

b E

4 Figure 6.4-5 WEXTEX Faulted Condition Leak Rate for Lower Bound 90% Confidence Limit Crevice Loss Coefficient,llot Leg, 22" Tubesheet Radius 6-34

4 r

ACC  ;

i i

'I 1

P b

Figure 6A 6 -WEXT11X Faulted Condition Leak Rate for lower Bound 90% Conndence Limit Crevice less CicfDelent, Hot Leg, 32" Tubesheet Radius 6-35 I

<- --v-- + ,,o.w.,- ,-..m_, ,,,.w.-rr<,y ,r.--, ,-3.m7 y w .s_.E, , -om-.%,&- .me..,-.y.-,,,ge...e,. .__,.__m.-4.,~-. .--%-m.. ....M-. ..

P a,e e i

L t

l l

l i

Figure 6.4-7 WEXTEX Faulted Condition Leak Rate for Lower Bound 90% Confidence Limit Crevige Less Coefficient Hot 12g,-

42.51" Tubesheet Radius 6  :

i I

i 6.4.2.2 Circumferential Cracks The semi empirical relationship for leakage through cracks is given by:

[ ]bc (Reference 15)  ;

where Q is the leak rate in gallons per day (gpd), AP is the differential pressure in psi and the crack length is the total crack length in inches. 'Ihis relationship is benchmarked by selected data points on measured leak rates and the fact that for small cracks, [

]be Segmented PWSCC cracks have been found in pulled tubes (Reference 15) to have aspect atios of

(

)bteinch long inroughwall cracks. Figure 6.4 8 illustrates the leak rates for single and segmented-circumierential enttks.

The leak rate contribution of circumferential cracks,if any,is to be evaluated utilizing the DOC RPC crack angle (less than [ Jas,e) adjusted for NDE uncertainty and growth (39 and [ Jase) to calculate an EOC leak rate contribution utilizing the segmented crack leakage model of Figure 6.4 8. For total crack network angles which fall between the segmented crack angles, the leak rates are linearly interpolated.

F 6-37

~.,,-,9.-we..ww ---...r- s -w- ,- ,,-49,. m% 7, - , , , - - - , ,.,m, , - - -wrsn~---r,w,.. --v.--..-,e-

, , , , , w- - , ~r c.w ,4-r-e, . - - - - -.

l a,c,e Fig,ure 6.4-8 leek Rate Versus Circumferential Crack Angle 6-38 I

1

i For WEXTEX circumferential indications, the expected crack morphology is segmented cracks. ,

Segmented cracks are also present for ODSCC in the initial stages of crack font.4 tion, although the limited pulled tube data indicate that the ODSCC ligaments may be lost by corrosion for ,

throughwall cracks. For W* applications, the circumferential indications potentially left in service would be below the bottom of the WEXTEX transition (BWT). Due to the tight crevice below the BWT, both the axial and circumferential indications can be anticipated to be PWSCC. Thus the segmented rnodel is applicable for indications below the BWT. For W* applications, the segmented mcxiel is applied only for the circumferential indications. For axial indications, the leak rate modelis based on the test results for axial cracks within the expanded tubesheet crevice. The circumferential leakage model combines the segmemed model for leakage thmugh the crack with the leakage restriction provided by the expanded crevice above the circumferential indication.

Although plots and tables ofleak rates for circumferential indications are not given in this report as a function of depth in the crevice, the methods apphed are the same as for axial cracks.

6.5 ANALYSIS MODEL FOR SLB LEAK RATE The potential leak rate fmm cracks in tubes left in service during a SLB event can be determined from the WEXTEX leak rate model developed in Section 6.4. This model, which was based on leak rate tests of WEXTEX joints discussed in Section 6.2, was developed for the generic SLB conditions used in this evalation of WEXTEX Model 51 steam generators,i.e., SLB differential pressurt of 2650 psi, hot leg temperature of 611*F, or cold leg temperature of 543 F. ,

The discussion and example application of the SLB leak rate analysis model which follows is din:cted to the hot leg of the generator since, to date, cracks have only been observed in the hot leg.

The leak rate model can calculate cold leg leak rates should the need arise for a cold leg application in the future. The analysis model would be the same for the cold leg as for the hot leg. For the analysis of SLB leakage in WEXTEX tubes the conservative determination of the crevice loss i

coefficient uses the lower 90% confidence limit on the mean regression fit.

6.5.1 SLB leakage Below W* .

It is desirable to limit the inspection and characterization of cracks to the region between the top of the tubesheet and the W* distance. Inspection fc;r cracks below the W* region is not necessary ifit ,

can be shown that any leakage frorn cracks left in service is negligible, or if a reasonable l

accounting for potentialleakage t this region is made.

l t

l 6-39

1 he SLB leak rate for a long crack at 1 inch below the W* depth is about 7 x 10 4 gpm. This leak rate further decreases by about 40% per inch of additional depth in the tubesheet and as the tubesheet radius increases (almost a factor of two reduction between 2.3" and 22" radii). Hus only cracks within a few inches below W* and near the center of the tube bundle would contribute as much as 5% of the leava ge of a crack within the W* region. The number of cracks in the few inches below W* can be expected to be less than or no more than the number ofindications within W*. Therefore, indications below W* can be expected to contribute <5% to the total SLB leak rate and they can be neglected in the analysis.

6.5.2 Total SLB leakage for WEXTEX Tubes Determination of potential SLB leakage from WEXTEX tubes containing cracks within the tubesheet region, both from within the W* region and below W*, can be made from the leak rate model and a characterization of the cracks within the W* depth. As discussed above, the leakage analysis method includes summation of tF 'eakage from individual cracks within the W* region with the leakage from an assumed like number of cracks below W*, An example of the application of the SLB leakage analysis method to plant data is given below.

In this example,1t is assumed that a plant eddy current inspection identified thirteen tubes with crauxs within the W* region as summarized in Table 6.5-1. Excluding crack growth considerations for the moment (discussed in Section 7), the cracks are characterized as to length and axial elevation within the tubesheet. The leak rate for each crack was determined using the 90% confidence interval'ower bound leak rate fmm the WEXTEX leak rate model (Table 6.4 3) at the 2.3 inch tubesheet radius, which is the radial location of highest leak rate. The combined leakage for thc cracks within the W* region amounts to 0.036 gpm. This is significantly less than the Sequoyah 1.0 gpm leak rate limit, so none of the tubes with eddy current indications in this example would require plugging due to leakage considerations. If the tubes meet all other W*

criteria, they could be retmned to service.

1 i

6-40

l a,c,e Table 6.5-1 Example of St.B Leak Rate Detemnnation for Sequoyah 1 & 2 WEXTEX Tubes 6-41

i The analysis model for detemtining potential SLB leakage for Sequoysh 1 & 2 is summarized in the following steps. Note that the potential leakage must be based on crack location and crack size including allowances for measurement uncertainties and crack growth in the next cycle. i STEPS IN DETERMINATION OF POTENTIAL SLB LEAKAGE

1. Determine the leak rate for each tube identified by eddy current inspection within the W* region. Leak rate to be determined from Figures 6.4 2 through 6.4 7 or Tables 6.4 3 and 6.4 4, consistent with crack size and location within tubesheet. This includes indications found in prior and current inspections within the W* region.

2. Sum the individual tube leak rates from the above to obtain the total leak rate for the region within W*.

3. Cc.mpart potential leakage from Step 2 with SLB leakage limit. If potentialleakage is less than leakage limit, and all other W* criteria are met, tubes may be retumed to service. If the leakage limit is exceeded, an appropriate number of the tubes should be plugged to reduce the potential leak rate to a level within the SLB leakage limit.

l l

6-42 l

l

t 7.0 SEQUOYAll #1 AND #2 INSPECTION RESULTE ,

l 1

7.1 SEQUOYAll #1 INSPECTIONS .

)

l The most recent inservice inspecdon of Unit #1, IEOC5, took place in October 1991. At that time, allIIL WEXTEX transitions were subjected to RPC testing; this inspection was in addition to full length bobbin inspections ranging from 1069 to 1316 tubes in each SO, Sixty eight (68) tubes were identified as exhibiting RPC crack signals (see Table 7.1 1) including $2 circumferential indications and 16 axialindications, All of these tubes were plugged. Two tubes in SO-4, those with crack lengths exceeding 20C)',

were examined with ultrasonic testing (UT), revealing shorter cracks with large ligaments between the circumferential cracks as sumrnarized in Table 7.1 2.

Figure 7.1-1 presents the distribution of RPC arc lengths observed for the circumferentially-oriented indications (COI), as reported in the field. 'Ihe average crack are length observed is 109' 1 9' 4 with a range from 58' to 256*. All of the 52 COls were found within 0.3 inches below the top of the tubesheet, with the mean position at -0.10 0.06 inches. Figure 7.12 displays the axial distribution of the circumferential cracks.

Sixteen tubes with axial PWSCC in the WEXTEX region occurred within a one inch band extending from 0.13" above the top of the tubesheet to 0.86" below; two axial indications were found at approximately 2.10" below and one 1.33" below the top of the tubesheet. The axial distribution of the reported positions of the axial indications is given in Figure 7.13; the associated 1 crack lengths, as measured in the field, are given in Figurr 7.1-4. The average of the axial crack y leng0w observed is 0.32 0.017 inches, with a range from 0.07" to 0.86". The longest crack was one of two found 2.1" below the top of the tubesheet.

T 71 j

--,,-a

A.C,e Table 7.1 1 Sequoyah Unit 1 IIL WEXTEX Transition RPC Results (IEOCS) 7-2

a,c,e Table 7.1-2 Sequoyah 1 - Ultrason.c Testing of Largest Circumferential Indications 7-3

a,c.e s

'M 4

Figure 7.1-1 Sequoyah 1 Distribution of Circumferential Crack Angles 7-4

, , r, c w 94>.

I i

8 C.e Figure 7.1-2 Sequoyah 1 - Axial Position of WEXTEX Circumferential Cracks 75

8,C,C Figure 7.1-3 Sequoyah 1 - location of WEXTEX Axial Cracks vs. Top of Tubesheet Elevation 76

i I

1

. a,c e - i 1

i i

+

f

-t

?

t l

ll l

Figure 7.1-4 Sequoyah 1 - Distribution of Axial WFXGX PWSCC -

Crack. lengths 7-7

. . . _ , ~ . .

. . - . _...- ._ .. _ _ - _ _ . _ . _ . . .. __.:._._._._..__...._..u__...._.__.

7.2 SEQUOYAll #2 INSPECTIONS ]

During the 2EOCS inservice inspection of Unit 2 in April 1992, extensive RPC testing of the hot l

- leg WEXTEX transitions was conducted the results obtained are summarized in Table 7.21.

100% of the tubes in service in S/G's 2 and 3 were RPC tested but all four S/G's were subjected to the same sampling strategy; expansions were performed on the basis of findings in the first sample which comprised 100% of th: Zone 4 (central zone) tubes, predicated on inspection guidelines developed by the WEXTEX Subgroup of the Westinghouse Owners Group. All 19 of the tubes with WEXTEX PWSCC indications were plugged.

7.3

SUMMARY

OF INDICATIONS FROM OTHER PLANTS Figure 7.31 shows the distribution of WEXTEX cracks at Plant A-1. Of thirty five (35) indications, five (5) were in the full W* region, thineen (13) were in the 0.7" long section below ,

the BWT, one (1) was in the 0.03" long zone of uncenalnty in the BWT location, and sixteen (16) were above the BWT. Thus greater than half (18 of 35) of the indications in the tubesheet region may be expected to be suitable for W* criteria application.

7.4 NDE UNCERTAINTIES 7.4.1. Distance below the Bottom of the WEXTEX Transition In suppon of W* determination, measurements are provided from bobbin and RPC probes. The position of all cracks detected within the W* region must be reported relative to the bottom of the WEXTEX transition (BWT) which in turn is reported relative to the top of the ('ITS). Since the surface-riding feature of the RPC probes suppresses the signal from the WEXTEX expansion transition, RPC signals will be reported relative to the TTS. Absolute position of the B%T relative to the 'ITS is obtain:x! from bobbin probe profilometry, performed in prior inspections if available or from current inspection data if appropriate. The uncertainty for bobbin probes with 0.060 inch coils was established by Junker and Taszaryk (Ref.12) by comparing EC measurements with laboratory specimens representing true tube degradation morphologies and machined fiaws in and ,

near transition geometries. The uncertainty associated with bobbin probes measuring the displacement between two geometrically related locations was found to be [ 3a,c.e, When flaw indications att identified by the RPC probe, their positions relative to the 'ITS are 78

a,c,e Tabic 7.2-1 Sequoyah Unit 2 HL WEXTEX Transi: ion RPC Results (2EOC5) 7-9

i a,c.e l

I Figure 7.31 Plant A-1 WEXTEX Axial Crack Locations with Respect to B\VT 7-10

measured as the uppennost point at which application of the slope intercept method locates the  !

flaw. The uncertainty for RPC probes in locating this position is reported in Ref.12 as

[ ]84.8 inches.

7.4.2 Flaw Length Uncertainties RPC measurements are used to detect and size the extent of cracking in the WEXTEX expansions.

Circumferential crack length measurements for throughwallindication's have an uncenainty related to the diameter of the pancake coil used for detection; for a 0.100 inch diameter cod in a 0.775 inch ID tube, this uncenalnty is approximately one half the coil diameter [ Jae.e. This value was verified by Junker and Taszaryk (Ref.13) using laboratory specimens and mercury modeling. Field measurements of circumferential cracks were compared to metallographic pulled tube findings by S. D. Brown for the WEXTEX Owner's Group. This regression equation indicates that the IU'C probe tends to everestimate the crack angle up to about 120* and underestimate the angle above 120 . 'The data available are described by the linear regression equation:

Y(ECT) = 0.71

• X (Met) + 34.8 with a estrelation coefficient of r=0.85 and a standard error of139* Figt're 7.41 illustrates the correlation obtained from the industry.available tube pull results. For this analysis, the field-contlated NDE uncertainty of 39* will be utilized.

Axial crack length measurement error was found to va.y only slightly as crack length increased using the slope intercept method for determining the position of the crack ends. The length error determined in this fashion for pancake coils as well as unifomi field coils was [ la.c.e inches.

In Reference 14, an EPRI report perfonned by Westinghouse, an evaluation was performed of RPC crack length uncenainty based on pulled tube data. These de.ta are from many different organizations and include mixed methods (with and without coil end effect corrections) for calling l RPC crack lengths. Thus the EPRI repon leads to a conservative estimate of the measurement uncenainty. In the EPRI repon, the standard deviation of the measurement error is 0.0' inches (1.46 mm). For conservatism in the W* criteria, the uncertainty on RPC axial crack length

measurement is applied as [ - Jae.e inch.

t 7-11 l

_ . . _ . _ _ _ . _ , _ _ , - . _ ._ _ ~ . , _ .,, _. . .

a,c,e ,

t t

(

Figure 7.4-1 Correlation Between Rotating Pancake Coll-Detected Arc-Length vs. Metallographic Exam Results 7-12

~ _ .

7.5 CRACK GROWTH RA*IES 7.5.1 WEXTEX Circumferential Crack Grouth Rates Only a few of the COls found in the Unit 1 EOC5 RPC WEXTEX inspection had been inspected during EOC4. Of eleven (11) such tubes, eight (8) were judged to be NDD; the growth rates for the other three (3) were 13*,5' and 21*. The largest EOC5 are length for these 11 tubes was 128*;

thus with an effective threshold for detection of shallow cracks of the order of 60 , growths up to 68' could be inferred. Since this estimate is not based on actual measurements between sequential operating cycle examinations, it does not provide a sufficient statistical basis for growth rate estimates. While additional data analysis is underway to provide a growth rate estimate for Sequoyah Units 1 & 2, reference is made to Plant A-1 which has also experienced circumferential cracking in its WEXTEX transitions. Plant A-1 is operating with similar primary water chemistry and a higher hot leg temperature than the Sequoyah Units, and the growth rate is evaluated for the transition region, so the Plant A-1 growth rates are expected to envelope Sequoyah growth rates.

For two consecutive 100% inspections of the HL WEXTEX transition with RPC probes, an average growth of 30 was observed at Plant A-1; the 95% cumulative probability growth rate observed was [ Ja.c.c. The Plant A-1 data provide a suitable reference basis for growth rate in this analysis, however, they may be updated should additional Sequoyah Unit 1 & 2 crack circumferential gmwth rate data become available.

7.12 Axial Crack Growth Rate i Insuf ficient data on axial crack propagation for WEXTEX plants is available to predict growth in the WEXTEX zone. The growth of axial PWSCC has been studied for several years in European plants with mechanical roll expansions. It has been shown that growth rate is dependent on crack length, decreasing with increasing initial length, and leveling off at [ Ja.c.e per effective full power year [ ] Table 7.5-1 l' taken from Ref.1 provides a summary of the pertinent data. The data of Table 4.7-1 are utilized in the current analysis, however, they may be updated should additional Sequoyah Unit 1 & 2 crack axial growth rate data become available.

l l

7-13 1

1

a,c,e

.g-V9?.

H 1:#.;

8{

I I

l Table 7.5-1 Estimate of Expansion Zone PWSCC - Average Crack Growth Rates (Based on Kiss Roll Data) 7-14

8.0 W* TUBE PLUGGING CRITERIA This section integrates the results obtained in prior sections to summarize the W* tube plugging criteria for eddy current indications in the tubesheet region of the WEXTEX expansion. The W*

criteria provide the basis for disposition of indications found by bobbin coil or RPC inspections below the bottom of the WEXTEX transition.

S.1 GENERAL APPROACH TO W* CRITERIA The approach taken to devebp the W* criteria is to utilize the general methodology of the L*

criteria for hardroll expansions and adapt the methods for WEXTEX expansions. The hardroll L*

cnteria utilize an L* length of undegraded tubing to limit leakage and permit a fledble leng*h (F*)

to resist pullout forces, with the length increased if degradation is present within the minimum F*

length. Since WEXTEX expansions have lower tube-to tubesh:et contact forces than ...tdroll expansions, limited leakage is possible under SLB conditions and the L* length is replaced in W*

by the n:quirement to calculate SLB leakage for indications left in service. The SLB leakage for a given crack size is dependent upon the length below the bottom of the WEXTEX transition (BWT). A flexible W* length patterned after the flexible F* length of the L* criteria is applied in W*, In addition, the L* criteria are modified to permit limited length circumferential indications within the pullout force distance. This option could be accommodated in L* as well as W* The general approach taken for W* can be applied to indications within the tubesheet for altemate types of tubesheet expansions. The principal difference between the type of tubesheet expansion (hardroll, explosive, hydraulic, partial depth) is that the methods for calculating SLB leakage must be specialized for the method of expansion.

The general approach taken to develop the W* tube plugging criteria includes:

1. Davention of Tube Burst

. Tube burst is precluded for cracks within the tubesheet by the constraint provided by the tubesheet.

2. Prevention of Axial Separation Due to Pullout Forces

• Axial separation (tube pullout is precluded by requiring that undetected degradation be at least a distance W* below the bottom of the WEXTEX 8-1

l J

transition such that the pullout forces under normal operating 'or accident conditions are exceeded by the elastic preload between the tube and tubesheet. A l flexible W* length approach is applied whereby the minimum W* length is i increased by the length of any axial indications left in service to obtain an undegmded length equal to the minimum W* 1ength.

• Axial separation above the W* length is precluded by conservatively limiting the size of circumferentially oriented cracks left in service above W*.

3. Limit SLB Leakage Within Allowable Limits

. The FSAR requirements for allowable leakage under accident conditions are satisfied by demonstrating that the dose rate associated with potential leakage fmm tubes remaining in service is a small fraction of 10CFR100 limits.

. SLB leakage is limited by leakage flow restrictions reculting from the crack and tube to tubesheet contact pressores which provide a restricted leakage path above the indications and a'so limit the degree of crack face opening compared to free span indications.

4. Negligible Ixakage Expected Under Normal Opemting Conditions

• Extensive operating experience in Europe with PWSCC axial cracks left in service has demonstrated negligible operating leakage even for thousands of free span cracks remaining in service.

- Operating leakage is further limited for cracks within the tubesheet by the tube to tubesheet contact forces.

8.2 W* LENGTli

As developed in Section 4, the tube to tubesheet contact force resisting axial pullout forces is comprised of

radial contact presrure from the WEXTEX expansion process, radial contact pmssure from thermal expansion and pressure differentials at operating conditions, and the axially dependent radial contact pressures resulting from the tubesheet deflection due to primary to l secondary pressure differences across the tubesheet.. The radial contact pressures from tubesheet 82 l

l

deflection are a function of the radial distance from the tubesheet centerline. Consequently, the minimum W* distance for the tube :o tubesheet contact forces to exceed the pullout forces is also a function of the radial position of a tube in the tube bundle.

For the Sequoyah Unit 1 & 2 SGs, two zones have been defined, W* Zone A and W* Zone B, cormsponding to inspection zones cut._ ntly in use for inspection of the expansion transition. 'Ihe W* lengths for W* lone B and W* Zone A am 5.1 and 4.2 inches, respectively. The W* distance is defined as the distance below the bottom of the WEXTEX transition (BWT). Only hot leg values are given in this section as cold leg WEXTEX indications have not been found in operating S/Gs. Figure 8.2-1 shows a tubesheet map defining the two zones for W*.

Below the flexible W* distance, any degradation including a 360 circumferential crack is acceptable since the tube to tubesheet contact force integrated over the W* distance precludes tube pullout under normal operating of accident conditions. As developed in Section 4.4, normal .

operating conditions are limiting in determining the W* length. Allowable degradation within the W* length is developed in Section 8.4.

8-3 1

~

- 8,C m

U b

o ik H

e.o C

h O

c V)'

2 U.

v

.m.

j-h=

i N-oc w

C m

8-4

_ m.

8.3 APPLICATION OF GROWT11 AND NDE UNCERTAINTY 8.3.1 Flexible W* Length Since the required length of tubing to be inspected depends upon the presence or absence of indications and must include allowances for NDE uncertainty and growth of an indication, a variable or " Flexible W? Length" has been defined. The Flexible W* length is the distance fram the bottom of the WEXTEX transition (BWT) to the bottom of the requimd inspection zone, and includes the basic W* length (either 4.2 or 5.1 inches), increases in length if indications are present, conservative additional length for measurement uncertainty, and modifications to account for potential growth of indications.

The ficxible W* length below the BWTis defined as:

a,c.c

.w 8-5

1

-i

- a,c.e -j t

P 1

V 6

4

- Figure 8.31 Flexible W* Iength' and SLB 1.cakage Description ,

L- 86 M '

v& = mo %s 1 A m: a y,7 ,'g-->s-

a,c.e Probe speed variations are typically small, however, significant variations in probe speed could result in a difference in actual vs. intended measurement length. Gross disturbances in eddy currrnt signals associated with probe whip or cable snap should be carefully noted for potential effect on W* length. If necessary, a calibration versus known geometry (e.g., tube support plate spacing, tubesheet thickness, etc.) should be utilized to check probe delivery speed.

8.3 2 leakage Analysis NDE uncertainty and growth are included in the leak rate analysis by adding the NDE uncertainty

[ Ja,c.e and the estimated growth from Table 7.5-1 to each crack prior to the leak rate calculation for each tube. Although few circumferential cracks have been observed in W* region, NDE uncertainty and growth are added to beginning of cycle (BOC) RPC crack lengths (which must be less than 92') to calculate the leakage contribution at end of cylce (EOC) for any circumferential cracks. Figure 6.4-8 contains the leak rate correlation for circumferential cracks.

Note that the straight line ligament model is more appropriate for the W* analysis.

8.4 ALLOWABLE TUBE DEGRADATION IN W* LENGTH Tube degradation within the flexible W* length is limited to provide adequate tube strength to transmit axial pullout forces through the W* distance.

As described in Section 5.3.1, a crack model was developed demonstrating that a tube having a

[ ]8c+e circumferential throughwall crack provides sufficient strength against axial pullout forces to transmit axial loads through the W* length, even when the remaining ligament is conservatively assumed to have a [ ]a,c.e throughwall OD-initiated crack. Circumferential indications within the tubesheet can be expected to be PWSCC cracks due to the tight crevice.

PWSCC crack morphologies are segmented cracks with multiple initiation sites with limited corrosion of ligaments between the individual microcracks. PWSCC arc lengths tend to be continuous of with large ligaments between multiple indications and do not show extensive arc length involvement beyond the principal macrocrack or macrocracks. Thus the use of a circumferential crack model with a [ la.c.e deep crack extending the remaining circumference beyond the RPC-measured crack angle is very conservative for the expected PWSCC crack 8-7 l

1

morphology. This conservatism reduces the allowable circumferential indication by about 60*.

Because of detectability limits of the RPC probe, the reported RPC crack angle is based only on the portion of the tube which is greater than [ ]* C c thmughwall. The assumption that the reported RPC crack angle is throughwall is conservative, since the RPC crack angle is larger than the actual angle due to lead in and lead-out effects, and since no credit is taken for the perdon of the wall greater than [ ]a.c.e throughwall which may remain undegraded (other studi.':s have shown that up to [ ]as.e of the tube wall can be assumed to be undegraded for a given RPC length call.)

Because of the detection limits of the RPC probe, closely spaced circumferential cracks would be indicated as a continuous RPC crack angle. The allowable EOC circumferential crack length of

[ ]a.c.e represents the ar.gle that would be measured by an RPC probe. To obtain a BOC acceptable crack ongle, the allowable EOC angle must be reduced by allowances for crack growth as measured by an RPC probe and by the RPC measurement uncertainty. From Section 7.4.1, the RPC growth is [ ]a.c.e for an 18 month cycle. From Section 7.3.2, the RPC measurement error is 39'. Thus the allowances total [105 la.c.e and the acceptable BOC RPC crack angle that can be left in service is [ ]2Ae [ ]aAe, [ ja.c.e, Bands of paral'el, axially oriented cracks in the W* length must also be limited to permit load carrying capability through the W* region. Cracks separated axially along the W* length are acceptable. The limit applies only to closely spaced cracks at an angle to the tube axis. Detailed limits can be applied, as noted in Sections 5.2 and 5.3, that are dependent on the number and angle from the tube axis for the axially oriented cracks. However, bands of axial cracks, which are common in hard roll transitions, have not been identified to date in WEXTEX expansions.

Consequently, a simplified limit of five parallel axial cracks at any tube elevation within W* is applied for the W* criteria. if these ciacks are inclined from the tube axis, the total circumferentially oriented angle summed over all five cracks at the beginning of cycle must be less than [ la,c.e. Similarly, if closely spaced axial cracks cannot be resolved separately, the circumferential involvement between RPC amplitude " null points" must be less than [ ] sac.

8.5 SLB LEAK RATE EVALUATION The SLB leakage from the tubesheet region is limited by the contact pressure between the tube and tubesheet as developed in Section 6. The contact pressure results from thermal expansion mismatch between the tube and tubesheet, and imm the differential pressure between the primary and secondary side. The contact pressure is also influenced by the dilation or contraction of the tubesheet hole resulting from tubesheet deflection caused by the primary to secondary side AP acting on the tubesheet. Since the deflection effect is a function of radal distance from the 8-8

i tubesheet center, the tube-to-tubesheet contact pressure varies with tubesheet radiut ano. For the purpose of the SLB l-dage evaluation, the contact pressure from the WEXTEX expansion process is not included in the determination of the total contact pressure. The WEXTEX contact pressure was inherert in the leak rate test samples,is considered representative of that in steam generator tubes, and therefore is not accounted for separately in the SLB evaluation.

SLB !cak rate models have been developed for both the hot leg and cold leg from tests of prototypic WEXTEX joints, from tests for crack leak rates with crack opening constmined by the small tube to tubesheet gaps, and a flow model of the tube-to-tubesheet crevice (Section 6).

Knowing the length of a crack and its position (depth) in the tubesheet, the leakage rate can be predicted from the leak rate models.

The SLB leak rate model developed in Section 6 has been qualified using leak rate tests for the crevice restrictions of WEXTEX expansions. Ieak rates are developed in Section 6 as a function of the crack length and the distance of the top of the crack from the bottom of the WEXTEX transition. For cracks within 0.7 inch of the BWT for which the WEXTEX expansion diameter may be tapered (tube-to-tubesheet gap < 1 mil), the leakage modelincludes the crevice restriction of leakage due to small tube-to-tubesheet gaps .d the leak rates assume no credit for the potential limitation in crack face opening due to the small tube-to-tubesheet gaps.

The resulting SLB leakage for tubes with indications in the tubesheet region in the current inspection, is to be added to the SLB leakage from other tubes returned to service under additional alternate plugging criteria,if appropriate. The total must he 1:ss than the allowable SLB leakage for Sequoyah Units 1 & 2, a value af 1.0 spm for the faulted loop, as established by the FSAR. If the leakage exceeds the allowable limit, an appropriate number of the tubes with indications must be removed from service to satisfy the allowable leakage limit. The indications providing the largest contdbution to leak rate shall be prioritized for repair if the SLB leak rate limit is exceeded.

8-9

8.6 INSPECI'lON REQ'JIREMENTS The W* repair criteria provide for disposition of indications found by bobbin coil or RPC inspections. The extent of the bobbin and RPC inspections are determined by Technical Specification mquirements as supplemented by Sequoyah plant guidelines and RPC inspection guidelines for the WEXTEX transitions. For implementation of W*, indications within the flexible W* length that are found by bobbin coil inspection must also be RPC inspected to characteri e crack lengths and elevations. All indications left in service 'vithin the flexible W* length must be RPC inspected at each planned refueling outage. When RPC inspections are performed for the WEXTEX transition region, the inspection shallinclude the fulllength of the flexible W* region.

Recording of results for the bobbin coil inspection shall include, as a minimum: the tube location, the elevation of each indication relative to the top of the :ubesheet, the peak-to-peak voltage of each indication and the phase angle and/or depth of the indication. The recorded results shallinclude all indications in the tubesheet region.

Reconiing of results from the RPC inspection shallinclude:

- Tube location

- Length of RPC inspection relative to the top of the tubesheet on either an inspection basis if constant for all tubes inspected or on a tube basis if not consistent for all inspected tubes. The length to be RPC inspected below the bottom of the WEXTEX transition (BWT) is defined as the " Flexible W* length", as developed in Section 8.3.1.

- Elevation below the top of the tubesheet of each indication. The elevation shall be j reported as the distance from the top of the tubesheet to the top of the crack.

1

- Crack length of each axial indication. Axiallength shall be reported separately for l each indication. For axially oriented cracks inclined more than 30 to the tube axis, either the angle of the crack relative to the tube axis or the maximum cirrumferential length of the deviation from vertical shall be reported. For closely spaced, multiple axi-J inscations which cannot be individually resolved (retum to null level between inmcations), the total circ :mferential length and angle spanning the unresolved 8-10

indications shall be reported in addition to the axial lengths of each separate indication.

• For circumferential cracks, the circumferential are length or angular extent shall be reported.

• Maximum RPC voltage (peak to-peak) of each indication.

• RPC phase angle and/or depth of each indication.

8.7

SUMMARY

OF W* TUBE PLUGGING CRITERIA As develcped in the above sections, the W* plugging criteria can be summarized as follows:

W* Criteria The W* length in the hot les shall be 5.1 inches in W* Zone B and 4.2 inches in W* Zone . . The W* length is the length of tubing below the bottom of the WEXTEX transition (BWT) which must he demonstrated to be undegraded in accordance with the following criteria The flexible W*

length is the total RPC-inspected length as measured downward from the BWT, and includes NDE uncertainties and crack lengths within W* as adjusted for growth. Below the flexible W* length any type or combination of tube degradation is acceptable.

If cracks are found within the W* region, the flexible W* length must be applied to account for the assumed lack of axial restraint over the length of the crack and associated end effects. The adjustment is to be applied in two zones acmss the tubesheet as follows:

a) For W* Zone B, add the length of the crack to W*.

b) For W* Zone A, add 1.2 times the crack length to W*. (Reference Section 5.4)

Allowable Tube Decindation Within W* Lencth Tube degradation witt the W* length shall be limited as follows:

8-11

+ For axial cracks having the upper crack tip below the top of the tubesheet and the BWT, including [ ]a.c.e uncenainty on BWT, no length limit is applied.

• Bands of parallel, axially oriented cracks shall be limited to five cracks. If the cracks are inclined relative to the tube axis, the total circumferentially oriented inclination summed over all cracks must be less than 92* at the beginning of cycle. Similarly the circumferential extent of closely spaced axial cracks must be less than 92* at beginning of cycle between the " null point;" on the RPC amplitude.

- Circumferential crack angles shall be less than 92" at the beginning of cycle. Any tube degradation exceeding the above limits within the W* distance shall be plegged or repaited. The circumferential crack must be located below the bottom of the WEXTEX transition (including [ ]a c.c uncenainty on the BWT) and at least [ ]a.e.e inch below the top of the tubesheet.

. Any type or combination of tube degradation below the W* length is acceptable. W*

criteria are applicable to those tubes which have typical WEXTEX expansions.

Characterization of candidate W* tubes may be necessary to determine if they are enveloped by the evaluated baseline conditions.

SLB Leakare Evaluadon The SLB leakage evaluation shall be based upon:

. The total leak rate from RPC-inspected tubes in the W* region will be divided by the percent of tubes inspected including either bobbin or RPC probes to obtain the total leak rate from indications.

+ The crack lengths used in the SLB leakage analysis shall be increased by allowances for growth and NDE uncertainty. The adjusted length is then the RPC length

[ Jane + ALG from Table 7.5-1.

. The leak rate contribution of circumferential cracks, if any, shall be evaluated utilizing the BOC RPC crack angle (less than i ]a,c.e) adjusted for NDE uncertainty and growth (39' and [ ]a,c.e) to calculate an EOC leak rate contribution utilizing Figtue 6.48.

8-12

i

• Axial cracks located below the BWT and the W* distance shall be evaluated using the leakage analysis model of Figums 6.4 2 through 6.4 7 or Tables 6.4-3 and 6.4-4.

The combined predicted leakage from all tubes with indications, including those which have been mtumed to service pmviously, must be compamd to t' - Sequoyah Unit 1 &

2 allowable leakage limits. If the pmdicted leakage is less than the allowable limit, the tubes with indications may be retumed to service provided that all other W* criteria am met. If the allowable leakage limit is exceeded, tubes shall be plugged until the allowable leakage limit is met. Priority for repair to meet leakage limits shall be the indications having the largest leakage contribution.

8-13

9.0 REFERENCES

1. EPRI NP-6864-L, PWR Steam Generator Tube-Plugging Lirnits: Technical Support Document for Expansion Zone PWSCC in Roll Transitions, December,1990.

2. WCAP-11228, Rev.1, "Tubesheet Region Plugging Criterion for the South Carolina Electric and Gas Company V. C. Summer Units 1 and 2 Steam Generators", October 1986.

(Proprietary)

3. Smith, R. E., " Series 51 Generic Seistnic Analysis", Calculation No:e SM-91-97, NSD Structural Mechanics, Westinghouse, Pittsburgh, Pa., October 1991.

4. FWDC 9-61-5427, " Expansion of Alloy 600 Tubes Into Carbon Steel Collars Using The WEXTEX Process, Foster Wheeler Development Corporation, May 16,1991.

5. Plyat, S. N., " Generic Calculations of the Mechanical Plug-to-Tube Interface Stresses for s Unit Pressure and Thermal Loads (Series 51 Steam Generator)", Calculation Note SM 53, NSD Structural Mechanics, Westinghouse, Pittsburgh, Pa., April 1989.

6. WNET-142. Vcl. 8. "Model D4-2 Steam Generator Stress Report, Divider Plate Analysis, Westinghouse Tampa Plant, Florida, September,1977.

7. Timoshenko, S., Strencth of Materials, Part II, Third Edition, Vor. Nostrand Company,

~

Princeton, N. J.,1956.

8. WCAP-Il857, "Tubesheet Region Tube Alternate Plugging (L*) Criteria for Steam Generators in the V. C. Summer Nuclear Station", June 1988. (Proprietary).

9. WCAP-12522, "Inconel Alloy 600 Tubing Material Burst ad Strength Properties", J. A.

Begky and J. L. Houtman, January 1990.

10. NSD-RMW-91-026," An Analytical Model for Flow Through an Axial Crack in Series With a Denting Corrosion Medium", February 1991.

I 1. WCAP-13034, " North Anna Unit 1 Steam Generator Operating Cycle", August 1991 (Proprietary).

9-1

12. R&D Report 87-5D4 CYUDS-R2," Guidelines for Assessment of Discontinuity Geometry by Eddy Current Techniques", W. R. Junker and B. J. Taszarek, Dec. 30,1987.

13. R&D Report 89-8M4-GRETH R1,"An Assessment of Field Compatible Inspection Methods for Steam Generator Tubing", W. G. Clark, Jr., W. R. Junker, and M. J, Metala, May 1989.

14. EPRI-TR-101104, Project S404 28 Interim Report, " Mercury Modeling for PWSCC Length Sizing", W. R. Junker and R. E. Shannon, August 1992.

15. WCAP-13226, " North Anna Unit 1 1992 Steam Generator Operating Cycle Evaluation",

May 1992,(Proprietary).

i 9-2

.