ML20079P730
| ML20079P730 | |
| Person / Time | |
|---|---|
| Site: | Byron, Braidwood  |
| Issue date: | 10/31/1991 |
| From: | Connors H, Hall J, Hopkins G WESTINGHOUSE ELECTRIC COMPANY, DIV OF CBS CORP. |
| To: | |
| Shared Package | |
| ML19353B277 | List: |
| References | |
| WCAP-12609, NUDOCS 9111130273 | |
| Download: ML20079P730 (176) | |
Text
____ _ _ ._
'h ',, -
,.1,3 . .D , .y , ' - U ( '..J'I j (.- b
- . 3 /+A ?-
I j- . . . ' - ,P' il 'I . ' i "Q' - ' , ' ' .i,b
^
'\ !. 46, i .' r. ,i.t
- s 2,'! '((, 4 {[ '-. ['[$ 2'
( , , '/-' ;. N ' ',..; . , 'O Y , 9: 4I*. .. ;. %.*T - 's :. og i f.1 a }' ), y'f . .,. '; % ,
- _0 ,3 ",hg +'.y ;;l' (;*),
- . . i. :7 . '.;.. y .} , *. '..:
. .y . ,'f .4. ' ,- [. ;.'y' * } 'Q,p
. _, _,o. . . _ < :. j 2- * :q 6,6{ q, $_ j ,y ' ;
$ . a
.*.* . , . e4, n..
p 4 ._ , . . , .. 6 ' , s,' . .
. ... .. .- .,,.-i.y e.f' ... ' '. . . ,," 4...,.$'.'.-
/. L
' . . ' 'L .%C. '.,,.A .' .%,..,'
4
- q. -' ef - < . . .- ~ - * ;- ,'g. .f. 4., _ i s ..p 'an e.
'.e - '
,f,. ._. > . - '
4
'$.....',','.,.,f.4 , - ., ,
4,,
. AS . _ ,* , :. . . .s
.s ..{ s 4 9_ , : :. ' -.-. ; y . -
- - Q*?- .. .,-
.- - (. \ .Q 4
}y,
+.'
4 J. y _ _ l, i s _ .. , . . : .n . :.
- . .t,
. . , > '. . . . - (._;.-j'.,...;-_,.**'.s',I,
( ,.. . , .
.s.1 r< . .
,,q, y... g, -4. ~,%j.Q.7 .:
... .. . y;. . . , , 1.. , + :. . . Y: +. o
. . . . - . < , . p g, ,y*
, . . :. . .~, .
,\... ,. . ~ . .. . ' . . - . . . . . '"' ' . x .2 - gh . 4 . ,, ' * ~,. ,.~, W ~ " , ~"< . . ; e *C s ,.
'c 4 ' .e..
. '., ,. . . . . . .. s y,t. -' ' . . s E >
^
- w. [ / :) y.f _.:;_ . ,W ' ,, '.
_ , an -,*, ,,' .'. '; .
,, ,47, i * ~ ., ; f. ; '1
..,L ' ';. '
,', ' ,7,,,' i , ., -
f.L - _
, ' , . .; . j p f;/ ,+,
..; .'. -h'.- ,, '. .A.i
- ,3
,;; g*..= ^'. ,' ,g' . fi_ . : . , , . Q i;.: ) ?, . . q . ;f; g' ::;'tff .
- _ *\-
, i ,g
. . . . .,s Egy+ . . ..g . y '(. , A;3 ,. Fa. y ;<. . y'3 ',lljey *. W s i ,-4
,: c '., '. ; [; 4 *\.; V ,e,..,,'f. 4..o-s,,j
.5 ; . . , .. .. ..,g,..,. .
, 4 sl ! 9: . ' l. . , .
, ~ .., . ,h
,. O ll ' .h:*'..hkk '* '. f3f t{, .Y t%;. g.cs,uf . ,**. l
- o. -. ;- , e .
,.;l '. % g , . . . .\Ikfl p.ll'( h, ,$. c' {.'.*h.. c ' h. A;.) . ' h y
., . . . , < . . .- 4,.,. ,. am , - ' 7 < * .. .4 -4 -
- r. .' :'. ! ".: - $ v:. ye. ', "m .
'3g' :-
m.< : 6, .: ,. . .n W:c i.y,v4 .( *v
.~.,: '., 4p : b ',,, "~. . . h.;?;,, f A
.>,% , . . 't ; k +Q
, y,j
...v . . ? _ .. :;}5 y v. e o:< .t. ,- ,, . .3, ,.,-
' R.c . . .gy hy;,
us.y:s.s ..
- . . . . . ,,,, . .,. - , ,, i.71 . .. W - . .. ...t,, 4 Q
- L,T, 5 , , . . - Q . .,'. _ ,"_ .. ; ,;,g' , g 't7"i. . . ).+_'y%d y
, </..'_. . . . . Q:: Q
' ;'. t .K*
. . .; .e , -i ., ' -:- .
. . .$ j. ., j .[h[-[.: (..],s.h,.v h./[PN..;k,,,f,yjj w +y
,<'u r, ;. .. -i.;.. . . . ...
. x.-
'c~.... . 9' . . c .
a
. . ,.-a , , ~,.j.(
d-3 .
. , . , /. ,. /r ,
- - 4, 4.
.,,j. 3
'; . ....,m... .v+ .'.3 t
-;_ ..,[..
- 6 .
' ..: . , , ' . , cv ,
'.L l } ^.l..;. g.#. . , 'L ..
- , . , ..,.:b. .. .
- , ;, , .*1 ?. .
- k. .g, ' y _, 'py. . Ising_g.,.-9 .
_ .' > , .-..h
...- a
.M_...- * .. ' . . , _ ,-.,..'.',J,' . . , ,, : , - l . - , s.s . . o?
v m *i .. . . '. . ,
. .w.. .: ...,..n...OE ;
< t. - r,. . '+. ' .:: .
.. . ; s!
s
. .: . .: . .j. g v.g*s.n,l., :,v. .t- :.pnJ(. . ,, e ,s f t
7,,.
. W . - , : . -
,. : ,,1 . 44 t
- .- .:, z.. . ;
( < . . ~s. f, . - M' { .m - -- . 7%' . , ;; ., . .4 .'.,1 G tg. .p.J
- , . 4; ....(e^3 ',q c.
s,* '
y . .. . . .: . . f, b ,. ,"
.' - J- .;j 4 gM
., f~ .s.
- . ? . .['J< .%,' '4 dj y. .
.L., p . ' . h ;fg, ' *
< 3 .I.>.,p.....,..:
- . '. ....g.' ; *, - . ; . :, . - . 'y"y.. ' ^ :: ,. __y * '. -~ j",?% ,:* _.?;. : '._._.,',,,,..c..-.
s := su: .
,......". <;. \...: .y,>,;.g n. ,
/ -
,x o- _ ' . ' .
.y.*. .3 . , .- , , ' , .
. t , ,. .
,r e
.s,.. -& ,_
l... . . .;y, , .. -
.; 4. 'gV ;....' ..;(, ; ], . : ?
..\.. . ," .
,. ., . . , -* -. l,';. , ',, ._ . . .
. .; , . . ... :, t ,; ,, %, ?.q , , : . . : : Q..... . . .
s . . ;, y&. .. .c . x - . , . ~. .
...r..
' s .ew. . ,
- q +
'. : '-.c., ,- Ol . ' ... '-.* -. c. .:. -BsJ . . c,F . "., .P d., ;i,T.Jy
. .r
- ....,.s. * '
- . ,q . .s p
- 1. g- .r.h, z..-; ..- ,.,-.' I, '. 4. . ' . T.~ -,- . g . ,' .M. , .i:-lJ
- J2.". '. -T ..'".*V .. .; ' . 9 \ M. , s, h 3 Y; 'y i
', ; }~;l ,. :_ ls. .. - 0. , , ' ; y , , . f : ,
- , .
- , , : . . e
$.'....',"'l.....V,.,.'
. .. :, ,: .- ..s. .. s . . - - .u.. . *1. .. . . .
, . .' ~ ' ..
. . .,.'),.- '.. ';.;.. g- . %. .'.V"..r,.'..,9.',
' , ' , ' . . ' . * , , . . - -:.- . ' ' . . - V.
4 s:
.u.
s.
. ' i .. . . , , -.; ,. 4..., . , . ..
^
1 ,... . . , ,'. ; ;* - .7;<'
3
-. . ..c. ..
' . . 4 , . , . ". s - . -
...s
, ;., ,.. . - -.4::5
....7..
. - a . . . . .., _.,..- ,:. ; n,..;. .. .- ' . ' .; , . ..' . .
, . s. ..
s-
. '.:m.
. , . .. , .: __n., . v. .., c. . ,. .. < .?..
- . ~ .. c..- -s- r . .-
. , _, ,,. . . . . ,, ;y . t .n .' ...;.. . . . + . . - . j ..u ..a. a, - i '
g . , . :, . o
,. - . ; 7 ; .3 " af .r . . ;, ' ' . . , - N. ' . .'f.. . - ' ,....
' . E ..7. -. ,' . ';3. . - s
,; . . . , ' - : ....,........u. ..
...,....., e
. s '. . , . ...u,.
.. ......y
'. . , . ,'x
( .. .c.
'4. .. . . +.' . . ',
, ' . .L. ,... ' + ..'l n',- <:. = . ...o _m
, ; . . . . . e. -<. . . -,.. .':. . . , .L ; :.;'
,' ; .. .. .nv,
- i -.... .- . :, ' 4 Jp' - .f a ' >., '
- ). :' ; r )( ,:j ,- ': , , 's ; .
- t :. n.
p ' '; J :
s ., ,. . . . . . :;.y,, t
- y . _' 8 .~.s . . % 4, . . g. . .. ' ' . .
- i,.
.: .,:: ~.-
- , .t : .
t,'?.- .: . K . .:. ;,._ ' '..,
c..... .. ;
.- . .' - . ;. .* ,.. ~ . ,..
,.,.,,: n,*
+
. (., . - ' ' . }._'. .... , . . , . ._ :. ;y
,- - .' _., y,
'; - } ' [ G '_ _ . : . 's
- ; i. -
'. .. -;1c, . . ~ . , ,_ _ $l < ,
, _ ' . ',. ' .,;):. * ,
,j . . . ' _ .g' . ~ - , ; ;:, ;, .l[
. , ., ( . '. [.;'c [.\ ; - .. ,. '_. .j' .y
_ * ~ . . . .. ..
, . ' - .- . y.. ) .:
. ,. . , - d . .g - t ...:- ,
, g ; -
, ,_,... ,3.. . y ; ,. .
.- . .. S, ,.. ... - .. .
r .
. l . .- '
g,; .-
-,, -.g w
7g
. ::: '. . ' .- ' ' +..;...
,. .. : . .'n ,.. -
c
....,".....s.
.: .. L..l
.L - ,. - . . .
. .. .. . . .e
- p. .......;...,1 y..
g . .
v
....,...r .7 . ;..- . . , . , .. . ..,s.. ; ;. -
3 . ,. .,.- ,
s..
< .,A#. . - . ..
.. ;~- . ~. . .. ',
.-e,. - . -..,c.. . ; :: . .., i
. '.. . . . :r' . . .. .. . ;. , .
. . n , .. .
v.
.;. . .,i, .. .* .. .. ......;.y.#'V.-- ' ~ . . . -i .. .. ... <
- N
.F ...
!. . ,.M,..S.". N . ' ,s ,.
. .. .' . . ..; . .- ..'. .3 -
.,(
. , . + ., .,,,, . r
, ., y.., 3
..,-4...
. s ,..- . . - -
. . , .h .*,
., , , e ,, . .
a - . . , ,:. ,,.. . , .. . .m. 2.
- n. .. , .c c .. . . ,,...- . . -
9
......a, , ., .,% 4.,,4 .- .x... ,, .a, ,
, - , e. -
..o...
. . e.r . ..; .g -
,s.f,
. , , . . ; . . .e .,
- - . , . . -.. .,g,'..,. , . . ' .. .-
.. , . . . . - . ., .. ..., , ; , . - - t
=-
.y ;.. -
3 - ;. s .
.g .* 9 .. . . . , . . , , g,. .. . .
y , y j;,
'. Q , ..,
,y . .' .
l ':f ,Q .;;L. J.'. ['. ;H.s
. j .c .. , . . . : . ' -' O.. ; ,. ,,KlN.
'+ ..:,
f"' ' ' '.:n .
f :, ':..*.. .;J :.('?. ,.... .. f.- s , - > ;; ; [ 3.)' ..' ]' .' '-' . ...:, . .' glk '. ' )' ~ aQ/ .4
, ,. .. ..y. m. q ..c.
- y. . .
... . .: , . , = .,.t. . . ;. .c.. . .. ..
. . .:.s ... c.
..,.(' , I. s .
. . . . ,'\,....- ' -
. t
\ '. -.[
([#
t.. ,,f..'.'
8 ' .
..'.n
' . ' ' , .; i -
- q. .,j .yr - [ . ~ .g.' . ' '. . . ,,+ , .,, D . f e .. ;,2 ;=.7 ' i 1.
' . l . ' . . . I '. . ,. - :. c . i,- .;
e y - ..;..'...'_.. 3, .
, . . t, i,. ,
c7 .. t-4.+. .
?
s
- .. - ei.
. . ' .s.-
s
.,4.-.s.
4 -. .. , .. -.,3..-
r.-
.2 A.s .,,c. ..:.
.,s.,...
, .,.,3,..t.
. . a, 7 ;,.*...- .j. T ; ,- -
~o,... ;- -
a.. - ;- ....o, s
..,a.. t . ... . . .. \.- .... s. _,-'~,.. - -
. +, _ _
. , , . . . . 4 [, . : . .' .
-'K,. k b \.,2* f :i. j.l:.. wI s,, ,, < .' .:5,'i kt. ' .h- ... ? 's , '; , j' O .'I,',';,'... ,' [ q .', . ] ' ,' Q
, ;. : - .' : ?.5 t, J-
- c. .
. _h'.s ' . _ d' . . ... . .. . .r.. , .g' .<,. , . . . ' . :. .~.;-
' ^.
',-'.;......W 9_iy , .- '.r. . . .'
, . . :..n... .v
,. , ,,3 - ., ' ,g.;; .g a L,
g -.,,t ... -
.; .y . :-.,z. :.., ..c ,, ,;
,. .;{ .
,y . .',
3
- . > .- , ?. .,.. ;, ...- :.y Fr q
. _,e,~ . , ., . y: i..y;.,,,.s
., t :- . . ." 'r. 9 , ,. - , ;
..... .e.. ,% .: , . ~.y- 73. , .g-%. o. ,
. .'1 . '.. . .. ::,,' .' V..J u9~. g' y.
- , . . . . ... ,, yl..;. , .c,,, .
s, ,, , , . . ', . g ,, t . . c. ; . . ., f,, ., 3 ,,; . ..
...i i.' ,,)Q .> % .c y s ' _;, ,-.
.) . . .; .- n<..(_ ,,y
. . ,j, , ,',f, . h' g, - . ' ,' . . v, - ': .
i
., . ,.. ., . ~ l ' ...;) ' .:'.r;
- ..,. Y_
- $ n j. . .,,'- .a.,... s. . ~ . , . . . . * , 5' i, :.. .~kj l .,.. *J . , . ',:y .s ; J _ c._. :., . :q. ; . . . . . u
. , , - . , . i.
. ' ...c..x....,.,.a ,
r.. n, # . . . .*<9. -' a 49... ,.._f....':,,.
- . , . . _ . . t-J ::,...r .N. e, :, . -+; ;. '; 3 * :. . c' . ""; - *. a..:r; .r.,.._
~ , - .'
4'..
- ,V
- 9. ; .. . ...
- - -. - ~.
.' . > . .l . s -
.: ,e y. 1,
', -.'+,,.'.g.
v' !r . .; z . W.Q.% 't. v3.: .g'. _'~
,u,,- ..-'.:. : %. l*. ,
.' .=.--',,r'...t-, .
i,..
- ' s -. . e ' ' .' . . '.. . .. ,
'-' s
.. . ,{ . *. ag ' , ' ':
- ' ., ,. . . . . ..s ' '. '.6 3 . . . , '- ** .
'j, ,. Y .' ' ..;). s .'.b'..
- T s' ! **;q ' .; w;. -7
....g". . . .; . ; - .l l.'s 4 ka.:* '
.,(.._-
' :, . .} '.',_,..<*T :l . ' _, *l ;l . :. : ', :-'..: **:N*,,' MW'.f , ,
<,f..,..,','.',.';i'.;'.'
'g _
5 . . .; . ? .,z ., ,
G.. 2, . . .' 2
.' ; , / , f- ,. < A' ,. i . -'
- . -
- ,; ; , t, _ , , . 3 "%o. ' '
- . ,k y : - - ff ..'-
, . L&'f , .-'
.., ..' l[: K i , -.,.,_#, , ' , ',f ' ? 'a .',..4..l.,,;;'. . * : - ;. - , , 's'
- .,r. e c ; -4
- ,- ,9,. 7, .,,. . . . ' . . j ; . f,a[ .;; .p-% _
- '*.,4',.-
g- * ~ =,'7..'.**<,%, .,,
y..,: , ', . A. 6]* .,;. , ,. - ' - . . , ' , . 4..,*>.W.e ... :-
- -'..."..,. - 4:
- 4
.... t .. .
.4, * . . ; : '.
.' - - 4 : . .. . . . -. ,; .. , ,. a?. + -'..
,',g;'...,...c,'.,,.+f,-,:,,',,m'...,'.:.,J..- .,v.........e.r-.,,c<
f- .
y .;t e .. ,. -
- . p .c. ,-. J . . . ' . .,.).,
..3..'c,..'..:
..,,.,.g,.-(;,
c, e. '
. , c:,a ; .. *.' . .. c.Vw
- .... a
- g4....,-, , . ,,
-M ..'r %,...- /-- @lg-
,,- L. *,'l* . " ,;. . * . * ' '
.. . . . % y ,"- s' i- *
- .'. <.m v.,"
g . 3
.. -
- 1
- : ; '_e .- .e ... .'l..- '?
e l: 4 ::-, ; v ,: .~..q. n,. . ; . -,e - c->' a.,' ' '. n ' . . ,: - , . .v .:
- . ' , , I5 [ '.e
,1; - .' ;. ,I, .. : . . Ii 7, . ; A ',, , .' O '.. *(,/. ' ' Fly , }. N a ..([z .. . Q*
.; h. ; / [ f 7.3,[Q ( ,I' , ."". '.i . .
1.;
,..x'*..n . .. . .. s .u. h. ; q , .. y ,;. n . l j , 3 ,.c,' .y , .'[;.,'n[ .
.. . p s ., . ; s ,. ., . r s !"l$..i[*'...* .y.f. '...+.; .l. p.1, ' 4..c 1.e.a: ,j.',qL u i i g,[ ;.:..m '-'k'g.'.h@W; t . .Y
... l ..1.s. .,* H..e.. . ',.- ..*;. .- , ' . . . . .. . - .3,, .* . g,...
. ep g
A, ; 3'., ,,.,
.,.c,,
. , -.s.3,, .,2.0.,,4,- ,.
t..;..,3,,...,- S.,. I',..5,.,**-, ,
'd
.-.f
+
-g.
3^ .
- ' ,. . . ' h -l , j ' . ',%.. ' ,f ' . .. .*.. .jl r. -'y.-* ' ' ..,,'..,...'.,!;31. ; , ;..'\. a ,*i. '. 3 % Rg
'. ,*; .f.$.'.g. }l ' .'.', ,l'. . '. ' ? . p . 4,. .64 s 'j.y
' ~
_m . -
." .y-s .. .-
..- y. ';I* ? . . ;. .,.:' .. L' ..'. : . .' . o " sp ' y r.
.'Q .\,j ,, ?s h U o. ' _.. . :' *, s h '.* . . J' '- '. ') . , &p 3
. ;; .~ * . . , . '.: _ -
.,'r' 3- 'T s * : 9;,'. t .?l:::', t~l%' *
..g,'.'.
' '%. * ... F ,
+;. . . . . . .'.;.w*-
. .' , ~. ' 9' :"',.:*-
% , . ~ . ' , .'
' n , , ; ** ".r . ...., . ' ' .w:.*s ..
-et . , .:s,, s **
>.t. . . ' e
<, 5av..r n . ..- ':'f ;. 'T f - *;
i.-. . g i 73.4.'. '%.' i [v3 ;c g>. ?.e. 2,c',. . . ., .V *. ):1- ; .[;; ,' : ,? W. . . = . l . .,; . . .
.: .. _.', . t .:. ._ j>
ei . ' h 4': .-Qf,: .'.e 4. *.. ::,e
'. ,; ;.,; _. :;y ; f..f.[?k ' g _ ,. . : :q .s*K) ..'l _.g'.,~,:p.
- ), ..
, , ~ .' . ,3,W,
- e
.;. ,_.,n '. . , ;rj'.;: :',-
,' ..f.'.'.'p....,,'*"..::,.d-
.? ,ri .'.,. . ; ,.". ' , . '.s ~ 3 ' i s.', ;*, .
- . ' : _...s... . ' y _ ,. ., *.,*',% ; ,p
. *.. ,b.. *s*,..'. p s;g.o . n.; ' v,,
WESTINGHOUSE CLASS 3 WCAP-12609 SG.90-1 i 064 BYRON UNIT 1 AND BRAIDWOOD UNIT 1 EVALUATION FOR TUBE VIBRATION INDUCED FATIGUE OCTOBER 1990 AUTHORS: H. J. CONNORS M. H. HU J. M. HALL T. S. M AGG E G. W. HOPKINS R. M. WILSON J. L. HOUTM AN H.W.YANT T. A. PITTERLE R. A. TAYLOR M. J. SREDZIENSKI APPROVED: D 40 ---
dWOOTT , MANAGER S AM GENER TORTECHNOLOGY AND EN EERINC WORK PERFORMED UNDER SHOP ORDER XARD 89040 This document contains information propnetary to Westinghouse Electric Corporation. It is submitted in confidence and is to be used solely for the purpose for which it is furnished and is to be returned upon request. This document and such information is not to be reproduced, transmitted, dischsed or used otherwise in whole or in part without written authorization of Westinghouse Electric Corporation. Energy Systems Business Unit.
WESTINGHOUSE ELECTRIC CORPORATION NUCLEAR SERVICES DIVISION P.O. BOX 3377 PITTSBURGH, PENNSYLVANIA 15230 o 1990 Westinghouse Electric Corp.
00515:10 ot1/120490 _ _ _ _ _ . - _ . _ _ _ _ _ - - - - - - -
ABSTRACT On July 15,1987, a steam generator tube rupture event occurred > 'he North Anna Unit 1 plant. The cause of the tube rupture has been determinec m e high cycle fatigue. The source of the loads associated with the fatigue mech ..sm is a combination of a mean stress levelin the tube with a superimposed alternating stress. The mean stress is the result of manufacturing residual stress, applied stress and residual stress due to denting of the tube at the top tube support plate, while the alternating stress is due to out of-plane deflection of the tube U bend attributed to flow induced vibration. For tubes without AVB support, local flow peaking effects at unsupported tubes are a significant contribution to tube vibration amplitudes.
This report documents the evaluation of steam generator tubing at Byron Unit 1 and Braidwood Unit 1 for susceptibility to fatigue-induced cracking of the type experienced at North Anna Unit 1. The evaluation utilizes operating conditions specific to Byron Unit 1 and Braidwood Unit 1 to account for the plant specific nature of the tube loading and response. The evaluation also includes reviews of eddy current data for Byron Unit 1 and Braidwood Unit 1 to establish AVB locations. This report provides background of the event which occurred at North Anna, a criteria for fatigue assessment, a summary of test data which support the anaiyucai approach, field measurement results showing AVB positions, thermal hydraulic analysis results, and calculations to determine tube mean stress, stability ratio and tube stress distnbutions, and accumulated fatigue usage. This evaluation concludes L that one tube in steam generator C of Byron Unit 1 and one tube in steam generator B of Braidwood Unit 1 are potentially susceptible to fatigue and require i corrective action.
/
l l
SUMMARY
OF ABBREVIATIONS l i
ASME -
American Society of Mechanical Engineers ATHOS - Analysis of the Thermal Hydraulics of Steam Generators AVB - - Anti-Vibration Bar AVT -
All Volatile Treatment J ECT - Eddy Current Test EPRI -
Electric Power Research institute FFT -
Fast Fourier Transform FLOVIB - Flow Induced Vibrations MEVF - Modal Effective Void Fraction OD - Outside Diameter RMS - Root Mean Square SR -
Stability Ratio TSP - Tube Support Plate
'F -
degrees Fahrenheit hr -
hour ksi - measure of stress- 1000 pounds per square inch Ib - pound mils -
0.001 inch wrt - with respect to MW - mega wat?
psi -
measure of stress - pounds per square inch psia -- measure of pressure-absolute D0515:1D-pt1/112590 ii
TABLE OF CONTENTS SECTION 1.0 Introduction 20 Summary and Conclusions 2.1 Background 2.2 Evaluation Criteria 2.3 Denting Evaluation 2.4 AVB insertion Depths 2.5 Flow Peaking Factors 2,6 Tubc Vibration Evaluation 2.7 Overall Conclusion 3.0 Background 3.1 North Anna Unit 1 Tube Rupture Event 3.2 Tube Examination Results 3.3 Mechanism Assessment 4.0 Criteria for Fatigue Assessment 4.1 Stability Ratio Reduction Criteria 4.2 Local Flow Peaking Considerations 4.3 Stress Ratio Considerations 5.0 Supporting Test Data 5.1 Stability Ratio Parameters 5.2 Tube Damping Data 5.3 Tube Vibrat on Amplitudes with Single Sided AVB Support 5.4 Tests to Determine the Effects on Fuidelastic Instability of Columnwise Variations in AVB insertion Depths 5.5 References 6.0 Eddy Current Data and AVB Positions 6.1 AVB Assembly Design 6.2 Eddy Current Data for AVB Positions 6.3 Tube Denting at Top Tube Support Plate 6.4 AVB Map Interpretations 7.0 Thermal and Hydraulic Analysis 7.1 Byron Unit 1 Steam Generator Operating Conditions 7.2 ATHOS Analysis Model 7.3 ATHOS Results D0 sis:10 pt1/112590 iii
l TA8LE OF CONTENTS (CONTINUED)
SECTION 8.0 Peaking Factor Evaluation
- 8. North Anna 1 Configuration 8.2 Test Measurement Uncertainties 8.3 Test Repeatability 8.4 Cantilever vs U Tube 8.5 Air vs Steam Water Mixture
, 8.6 AV8 Insertion Depth Uncertainty 8.7 Overall Peaking Factor with Uncerteinty 8.8 reaking Factors for Specific Tubes 9.0 Structural and Tube Vibration Assessments 9.1 Tube Mean Stress 92 Stability Ratio Distribution Based Upon ATHOS 9.3 Stress Ratio Distribution with Peaking Factor 9.4 Cumulative Fatigue Usage 10.0 Operating Limit Zvaluation 10.1 Summary and Conclusions 10.2 Limiting SG Secondary Side Variables 10.3 Parametric One-Dimensional Relative Stability Ratio Analysis 10.4 Structural and Tube Vibration Assessment 10.5 Establishment of Operating Limits 10.6 Discussion and Application of Operating Limits APPENDIX A Evaluation of the[ la.b.c Tube Damper D0515:10 pt1/112590 iv
LIST OF FIGURES FIGURE 31 Approximate Mapping of Fracture Surface of Tube R9C51 S/G "C" Cold Leg, North Anna Unit 1 32 Schematic Representation of Features Observed During TEM Fractographic Examination of Fracture Surface of Tube R9CS1, S/G "C" Cold Leg, North Anna Unit 1 33 Calculated and Observed Leak Rates Versus Time ,
41 Vibration Displacement vs. Stability Ratio 42 Fatigue Strength of Inconel 600 in AVT Water at 600'F ,
43 Fatigue Curve for inconel 600 in AVT Water Comparison of Mean Stress Correction Models 44 Modified Fatigue with 10% Reduction in Stability Ratio for Maximum Stress Condition t
45- Modified Fatigue with 5% Reduction in Stability Ratio for Minimum Stress Condition 51 Fluidelastic instability Uncertainty Assessment 52 Instability Constant-p 53 Instability Constants, p, Obtained for Curved Tube from Wind Tunnel Tests on the 0.214 Scale U Bend Model 5-4 Damping vs. Slip Void Fraction 55 Overall View of Cantilever Tube Wind Tunnel Model 56 Top View of the Cantilever Tube Wind Tunnel Model 57 Fluidelastic Vibration Amplitude with Non Uniform Gaps ,
58 Typical Vibration Amplitude and Tube /AVB Impact Force Signals for Fluideiastic Vibration with Unequal Tube /AVB Gaps D051s:1D pt1/112590 V
LIST OF FIGURES (CONTINUED)
FIGURE S9 Conceptual Design of tlie Apparatus for Determining the Effects on Fluidelastic instability of Columnwise Variationsin AVB insertion Depths 5 10 OverallView of Wind TunnelTest Apparatus 5 11 Side View of Wind Tunnel Apparatus with Cover Plates Removed to Show Simulated AVB5 and Top Flow Screen 5 12 AVB Configurations Tested 5 13 Typical Variation of RMS_ Vibration Amplitude with Flow Velocity for Configuration la in Figure 512 61 AVB insertion Depth Confirmation 62 Byron Unit 13 team Generator 1 - AVB Positions 63 Byron Unit 1 Steam Generator 2 - AVB Positions 64 Byron Unit 1 Steam Generator 3 - AVB Positions 65 Byron Unit 1 Steam Generator 4 - AVB Positions 66 Braidwood Unit 1 Steam Generator 1 - AVB Positions t
67 Braidwood Unit 1 Steam Generator 2 - AVB Positions l
68 Braidwood Unit 1 Steam Generator 3 - AVB Positions 69 Braidwood Unit 1 Steam Generator 4- AVB Positions 71 Plan View of ATHOS Cartesian Mode! for Model D4 7 2 -- Elevation View of ATHOS Cartesian Model for Model D4 73 Plaa View of ATHOS Cartesian Model for Model D4 Indicating Tube Layout D0s1510 pt1/112590 VI
1 I
< W, OF FIGURES (CONTINUED)
FIGURE 74 Flow Pattern on Vertical Plane of Symmetry i 75 Vertical Velocity Contours on a Horizontal Plane at the Entrance to the l U Bend 76 Lateral Flow Pattern on Horizontal Plane at the Entrance to the U Bend 77 Vold Fraction Contours on Vertical Plane of Symmetry 78 Tube Gap Velocity and Density Distributions for Tube at R10/C5 79 Tube Gap Velocity and Density Distributions for Tube at R10/C20 7 10 Tube Gap Velocity and Density Distributions for Tube at R10/C40 7 11 Average Velocity and Density in the Plane of the U-Bends Normal to Row 10 81 Original North Anna AVB Configuration 82 Schematic of Staggered AVBs 83 AVB " Pair" in ECT Trace 84 North Anna 1, Steam Generator C: AVB Positions Critical Review "AVB Visible" Calls 85 North Anna 1, Steam Generator C, R9C51 Projection Matrix 86 North Anna R9C51 AVB Final Projected Positions 87 Final Peaking Factors for Byron Unit 1 88 Final Peaking Factors For Braidwood Unit 1 91 Axisymmetric Tube Finite Element Model 92 Dented Tube Stress Distributions- Pressure Load on Tube D0515;1D pt1/112590 Vii
UST OF FIGURES (CONTINUE D)
FIGURE 9-3 Dented Tube Stress Distributions-Interference Load on Tube 94 Dented Tube Stress Distributions - Combined Stress Results 9-S Relative Stability Ratios using MEVF Dependent Damping - Byron 96 Relative Stability Ratios using MEVF Dependent Damping - Braidwood Unit 1 97 Stress Ratio Vs. Column Number - Donted Condition -Byron Unit 1 98 Stress Ratio \o. Column Number- Dented Condition - Bra!dwood Unit 1 99 Stress Ratio Vs. Column Number- Undented Condition -Byron Unit ?
9 10 Stress Ratio Vs. Column Number - Undented Condition - Braidwood Unit 1 9 11 Byron Unit 1 and Braidwood Unit 1 - Maximum Allowable Relative Flow Peaking 10 1 Variation in RSR With Pressure and Flow, Byron 1/Braidwood 1:
Current (Reference) Thot (606.7T) 10 2 Byron Unit 1, Operating Limit on Pressure and Flow to Preclude Fatigue Rupture During 40 year Design Basis Operating Period 10 3 Braidwood Unit 1, Operating 1.imit on Pressure and Flow to Prcclude Fatigue Rupture During 40 year Design Basis Operating Period 10-4 Byron 1/Braidwood 1 RSR Multiplier vs. Flow, Thot, Plugging a.b.c A-1 A-2 A-3 A-4 D0515.10-pt1/112590 Viii
LIST OF TABLES TABLE 41 Byron Unit 1 Time History of Operation Used for Fatigue Analysis 42 Braidwood Unit 1 Time History of Operation Used for Fatiguc Analysis 43 Fatigue Usage per Year Resulting From 5tability Ratio Reduction E
f 51 Wind Tunnel Test on Cantilever Tube Model 52 Fluidelastic Instability Peaking Velocity Factors for Columnwise Variations in AVB Insertion Depths 61 Summary Listing of Unsupported Tubes-Byron Unit 1 62 Summary Listing of Unsupported Tubes Braidwood Unit 1 7-1 Byron /Braidwood Steam Generator Operating Conditions and Comparison with Model D4 Conditions on which the Reference 3D ATHOS Analysis was Based 8-1 Stability Peaking Factor Due to Local Velocity Ferturbation 8-2 Comparison of Air and Steam Water Peaking Factor Ratios 8-3 Effect of LocalVariation of AVB Insertion 84 Uncertainties in Test Data ana Extrapolation 8-5 Extrapolation of Test Results to Steam Generator Conditions 8-6 Final Peaking Factors 87 Stability Peaking Factors for Specific Tubes- Byron Unit 1 88 Stability Peaking Factors for Specific Tubes- Braidwood Unit 1 9-1 100% . 'wer Operating Parameters 7 D05151D 0t1/112590 iX
. _ _ _ . . _ ~ - . . . . . , _ - _ _ . , _ _ , . _ _. . - - . . _ _ . _ _ . ~ . . . _ . . - _ _ . - .
- - - _ ~ . - - . , - . ----
UST OF TABLES (CONTINUED)
- TABLE i
92 Byron Unit 1 - Tubes With Significant RSR's or Stress Ratios !
93 Braidwood Unit 1 -Tubes With Significant RSR's or Stress Ratios 10 1 Stress Analysis / Fatigue Results for Enveloping Tubes Based on a 40 year i Operating Period !
t e
i 4
h t
?
t t
D0515:1D-ret t/112$90 X ww-e- m -re sm-e vp-w*'e4 yp -.c r ww ree w o w- -upw-ee rw " N@ yewer Nm-m+ mem~>t'w,edr'-"'---'e*TN-'9rP*-'we'-'t're--W*m =7WW T"f 'w v - ww w TT P mT-9W-t m M T'M's"W's""P "tr T -Tyd'^WW g wed' of""M77-F"-PT-":*W"f' "'T*y*PTTw-Tr7
1
1.0 INTRODUCTION
This report documents the evaluation of steam generator tubing at Byron Unit 1 and Braidwood Unit 1 for susceptibility to fatigue-induced cracking of the type experienced at North Anna Unit 1 in July,1987. The evaluation includes three-dimensional flow analysis of the tube bundle, air tests performed to support the vibration analytical procedure, field measurements to este '86h AVB locations, structural and vibration analysis of seiected tubes, and fat e usage calculations to predict cumulative usage for critical tubes The evaluation utilizes operating conditions specific to Byron Unit 1 and Braidwood Unit l in order to account for plant specific features of the tube loading and re.ponse.
Section 2 of the report provides a summary of the Byron Unit 1 and Braidwood Unit 1 evaluation results and overall conclusions Section 3 provides background for the tube rupture event which occurred at North Anna Unit 1 including results of the examination of the ruptured tube and a discussion of the rupture mechanism. The criteria for predicting the fatigue usage for tubes having an environment conducive to this type of rupture are discussed in Section 4. Section 5 provides a summary of test data which supports the analytical vibtation evaluation of the candidate tubes.
A summary of field measurements used to determine AVB locations and to identify unsupported tubes is provided in Section 6 Section 7 provides the results of a thermal hydraulic analysis to estabbsh flow field characteristics at the top support plate which are subsequently used to assist in identifying tubes which may be dynamically unstable, Section 8 presents an update of the methodology criginally used to evaluate the tube rupture at North Anna Unit 1. Section 9 presents results of the structural and vibration assessment. This section describes tube n,ean stress, e stability ratio and tube stress distributions, and accumulated iatigue usage for the small radius U tubes in the Byron Unit 1 and Braidwood Unit 1 steam generators.
Section 10 addresses the effect of future modifications in the steam generator ~
operating conditions. Graphs of feedwater flow versus steam generator pressure are used to provide operating limit curves which can be used to validate that the steam generators remain within the analyzed limits of the tube fatigue analysis.
Finally, Appendix A summarizes the evaluation and qualification of multiple cable dampers for use in Row 12 tubes requiring corrective action.
\
costs <a amms90 1-1
l 2.0
SUMMARY
AND CONCLUSIONS The Byron Unit 1 and Braidwood Unit 1 steam generators have been evaluated for ,
the susceptibility of unsupported U bend tubing with denting at the top tube support plate to a fatigue rupture of the type experienced at Row 9 Column 51 (R9C51) of Steam Generator C at North Anna Unit 1. The evaluation of Byron Umt 1 and Braidwood Unit 1 was based on Eddy Current Test (ECT) data supplied by Commonwealth Edison Company and confirmed by Westinghouse.
2.1 Background
The initiation of the circumferential crack in the tube at the top of the top tube support plate at North Anna 1 has been attributed to limited displacement, fluid elastic instability. This condition is believed to have prevailed in the R9C51 tube since the tube experienced denting at the support plate. A combination of conditions were present that led to the rupture. The tube was not supported by an anti-vibration bar (AVB), had a higher flow field due to local flow peaking as a result of non-uniform insertion depths of AVBs, had reduced damping due to denting at the top support plate, and had reduced fatigue properties due to the environment of the all volatile treatment (AVT) chemistry of the secondary water and the additional mean stress from the denting.
2.2 Evaluation Criteria The criteria established to provide a fatigue usage less than 1.0 for a finite period of time (i.e.,40 years) is a 10% reduction in stability ratio that provides at least a 58 percent reduction in stress amplitude (to < 4.0 ksi) for a Row 9 tube in the North Anna 1 steam generators (SG's). A reduction of this magnitude is required to produce a fatigue usage of < 0.021 per year for a Row 9 tube in North Anna and therefore a fatigue usage factor objective of greater than 40 years. This same fatigue criteria is applied as the principal criteria in the evaluation of Byron Unit 1 and Braidwood Unit 1 tubing.
The fluidelastic stability ratio is the ratio of the effective velocity divided by the critical velocity. A value greater than unity (1.0) indicates instability. The stress ratio is the expected stress amplitude in a Byron /Braidwood tube divided by the stress amplitude for the North Anna 1, R9C51 tube.
Displacements are computed for the unsupported U bend tubes in Rows 11 and '
inward,(descending row number) using relative stability ratios to R9C51 of North Anna 1 and an appropriate power law relationship based on instability displacement versus flow velocity. Tubes having different U-bend radii will have different stiffness and frequency and, therefore, different stress and fatigue usage per year than the Row 9 North Anna tube. These effects are accounted for in a stress ratio technique.
The stress ratio is formulated so that a stress ratio of 1.0 or less produces acceptable
- stress amplitudes and fatigue usage for the Byron /Braidwood tubing for operating cesis to e m s9o 2-1
, . , - -,,g -. y w- "--
yr r--v- va--^-e e- M we* -w- T-*---7" - ^ ^ w g----:-- -ee*-*1'eg-- --^de g' + rT*W""-N' " ' ' 'Nr- g
conditions within the administrative limits identified in Section 10. Therefore, a stress ratio less than 1.0 provides the next level of acceptance criteria for unsupported tubes for which the relative stability ratios, including flow peaking,
. exceed 0.9.
The stability ratios for Byron Unit 1 and Braidwood Unit 1 tub:ng, the corresponding stress and amplitude, and the resulting cumulative fatigue usage must be evaluated relative to the ruptured tube at Row 9 Column 51, North Anna 1, Steam Generator C, for two reasons. The local effect on the flow field due to various AVB insertion depths is not within the capability of available analysis techniques and is determined by test as a ratio between two AVB configurations. In addition, an analysis and examination of the ruptured tube at North Anna 1 provided a range of initiating stress amplitudes, but could only bound the possible stability ratios that correspond to these stress amplitudes. Therefore, to minimize the influence of uncertainties, the evaluation of Byron Unit 1 and Braidwood Unit 1 tubing has been based on relative stability ratios, relative flow peaking factors, and stress ratios.
The criteria to establish that a tube has support from an AVB and therefore eliminate it from further considerations is that it must have at least one sided AVB support present at the tube centerline. The criteria is based on test results which show that one sided AVB support is sufficient to limit the vi'o ration amplitude for fluidelastic excitation. AVB support is established by analysis of eddy current (EC) measurements and is a key factorin determining the local flow peaking factors. The local flow peaking produces increased local velocities which cause an increase in stability ratio. A small percentage change in the stability ratio causes a significant change in stress amplitude. The relative flow peaking f actors for Byron Unit 1 and Braidwood Unit 1 tubing without direct AVB support have been determined by test.
These flow peaking factors normalized to the North Anna R9C51 peaking, are applied to relative stability ratios determined by 3-D tube bundle flow analysis, to obtain the combined relative stability ratio used in the stress ratio determination.
2.3 Denting Evaluation For conservatism in the evaluation, all of the tubes are evaluated for two possible conditions corroded, but not dented; and as being dented. The effect of denting on the fatigue usage of the tube has been conservatively maximized by assuming the maximum effect of mean stressin the tube fatigue usage evaluation and by incorporating reduced damping in the tube vibration evaluation.
2.4 AVB Insertion Depths t
The Byron Unit 1 and Braidwood Unit 1 SGs have two sets of Alloy 600 AVBs. The l
' inner' AVBs have a rectangular cross section and extend into the tube bundle approximately as far as Row 10. Discounting tube ovality, which tends to vary with bend radius, they provide a norrinal total clearance between a given tube and the surrounding AVBsof[ ja.c inch. Considering average tube ovality for a Row 10 oosis:io.ptirii2s90 22
tube, the nominal tota! tube to AVB clearance is approximately [ Ja.c inches. The outer AVBs have the same cross section as the inner AVBs, and extend into the tube bundle approximately as far as Row 21, providing a nominal tube to AVB clearance i comparable to the inner AVBs. Since the purpose of this analysis is to evaluate the potentially unsupported tubes at or near the point of maximum AVB insertion, only the dimensions and EC data pertaining to the inner AVBs are required.
For Byron Unit 1, eddy current data from the October 1988 inspections were supplied by CECO. Maps were provided by CECO showing the number of tube /AVB intersections in each U bend tube; these mapt were used to determine AVB insertion depths. For tubes in regions where more detailed information was required to determine the depth of AVB insertion, eddy current tapes supplied by CECO were reviewed by Westinghouse to identify the number of tube /AVB intersections and the location of these intersections relative to the apex of a given tube. This information was used in calculations by Westinghouse to determine the deepest penetration of a given AVB into the tube bundle.
For Braidwood unit 1, eddy current data from inspections performed by Westinghouse in October 1989 were used. Maps showing the number of tube /AVB intersections in each U bend tube were used to determine the AVB insertion depths.
For tubes in regions where more detailed information was required to determine the depth of AVB insertion, the October 1989 eddy current tapes were reviewed by Westinghouse to identify the number of tube /AVB intersections and the location of these intersections relative to the apex of the tube. This information was used in calculations by Westinghouse to determine the deepest penetration of a given AVB into the tube bundle.
For the area of interest in the Byron Unit 1 and Braidwood Unit 1 steam generators, the AVB support of the tube can normally be verified if EC data shows both legs of the lower AVB, one on each side (hot leg -cold leg) of the U-bend. This data, indicated by a listing of two or more AVBs in the insertion depth plots,is the method of choice for establishing tube support. If only the apex of an AVB assembly is near or touching the apex of a tube U bend, only one AVB signal may be seen. In this case, adequate tube support cannot be assumed without supplemental input.
Support can be determined if ' projection' calculations based on the AVB intercepts of higher row number tubes for the same column verify insertion depth to a point below the tube centerline. Maps of the AVB insertion depths for Byron Unit 1 and Braidwood Unit 1 are shown in Figures 6-2 thru 6 9.
2.5 Flow Peaking Factors Tests were performed modeling Byron Unit 1 and Braidwood Unit 1 Model D4 SG tube and AVB geometries to determine the flow peaking factors for various AVB configurations relative to the North Anna R9C51 peaking factor. The test results were used to bound the ratio relative to the R9C51 configuration.
oosi s:t omtm t 2s90 23
I 2.6 Tube Vibration Evaluation The calculation of relative stability ratios for Byron Unit 1 and Braidwood Unit 1
' makes use of detailed tube bundle flow field information computed by the ATHOS steam generator thermal / hydraulic analysis code. Code output includes three dimensional distributions of secondary side velocity, density, and void fraction, along with primary fluid and tube wall temperatures. Distributions of these parameters have been generated for every tube ofinterest in the Byron Unit 1 and Braidwood Unit 1 tube bundles based on currently defined full power operating conditions. This information was factored into the tube vibration analysis leading to the relative stability ratios.
Relative stability ratios of Byron Unit 1 and Braidwood Unit 1 (Row 8 through Row
- 12) tubing versus R9C51 of North Anna 1 are plotted in Figure 9 5 and 9 6. These relative stability ratios include relative flow peaking factors. The stress ratios for Byron Unit 1 and Braidwood Unit 1 are given in Figures 9 7 and 9 8. These also include the relative flow peaking effect, and are calculated based on clamped tube conditions with denting at the tube support plate.
For Byron Unit 1, examination of Table 9 2 and Figure 9-7 shows that R12C8 of steam generator C exceeds the limiting stress ratio criteria, and should be removed from servie.e, since the fatigue usage calculated for this tube is greater than 1.0. Removal may take the form of either' stabilizing and plugging' or' sentinel plugging' Of the remaining unsupported tubes in all four Byron Unit 1 steam generators, the highest stress ratio is 0.54 and occurs at location R11C4 of steam generator C. The maximum fatigue usage for this tube is calculated by combining the usage for Unit 1 operating history to date plus the projected usage for future operation. Assuming operation at 100% power with the currently defined parameters and plugging values for 100%
availability, the maximum calculated fatigue usage is approximately 0.1.
For Braidwood Unit 1, examination of Table 9-3 and Figure 9 8 shows that R12C5 of steam generator B exceeds the limiting stress ratio criteria, and should be removed from service, since the fatigue usage calculated for this tube is greater than 1.0.
Removal may take the form of either ' stabilizing and plugging' or ' sentinel plugging' Of the remaining unsupported tubesin all four Braidwood Unit 1 steam generators, the highest stress ratio is 0.48 and occurs at locat:on R12C4 of steam generator B. The maximum fatigue usage for this tube is calculated by combining the usage for Unit 1 operating history to date plus the projected usage for future operation. Assuming operation at 100% power with the currently defined parameters and plugging values for 100% availability, the maximum calculated fatigue usage is approximately 0.03.
These .;eding results are based on the assumption that future operation throughout the remeinder of the 40 year design basis period will be at 100% pcwer with the currently defined reference parameters (Wst = 3.78x106 lbm/hr and 908 psia). 5%ce some modifications in operating conditions are likely to occur at some l
00 sis:1 dst b112s 90 24
- -- ~ , , _ _ . - _ _ _ . . . _ _ . . ~ _ _ , . - . . . _ , _ _ _ _ _
l time in the future, a supplemental evaluation has also been completed to address I the effect of these changes on the fatigue assessment.
. The evaluation has defined the acceptable operating space for each unit with regard to satisfying the tube fatigue criteria. Figures 10 2 and 10 3 define the acceptable operating regions for the Byron 1 and Braidwood 1 steam generators, respectively.
The limits are defined in terms of two key thermal / hydraulic parameters, the feedwater flow rate (power level) and the steam pressure. Operation with steam pressures and flows in the " acceptable" region, above the limit line, will insure that ;
the U bend fatigue usage criteria (< 1.0) will ncit be violated for the remainder of l the design basis operating period. No tubes,in addition to the two tubes identified !
previously (one per each unit), will require preventive action to preclude a f atigue rupture so lorig as operation remains within the acceptable region. Note that the l recommended opr. rating limits include an allowance for steam pressure and feedwater flow measurement uncertainties.
Figures 10 2 and 10-3 also indicate that, based on the location of the current reference operating condition, margins of = 45 and 50 psi exist to the limits for Byron 1 and Braidwood 1, respectively, assuming the current 100% power /feedwater flow and primary inlet temperature are maintained. The steam pressure would have to decrease to about 865 and 855 psia, respectively, before any preventive action would have to be considered. Based on the flow and pressure measurement uncertainties described in Section 10, the level of tube plugging would have to increase to about 12% for Byron 1 and about 15% for Braidwood 1 to produce these reduced steam pressures at the reference Thot.
In the event that future operating conditions do reach the operating limits, only one additional tube in each unit would become a candidate for preventive action. These tubes are identified in Table 101. The adoption of even more adverse operating conditions below the limit lines in the " unacceptable" region would require action on larger numbers of tubes, depending upon the distance below the limits.
2.7 Overall Conclusion 1
The analysis described above indicates that, provided future operation remains in the acceptable region on the enclosed operating limit figures, the Byron Unit 1 and Braidwood Unit 1 tubes recommended to remain in service are not expected to be susceptible to fatigue rupture at the top support plate in a manner similar to the rupture which occurred at North Anna 1. Therefore, no modification, preventive tube plugging, or other measure to preclude such an event is believed necessary (for operation within the limits specified in Section 10), other than to address tube R12C8 of steam generator C of Byron Unit 1 and tube R12C5 of steam generator B of Braidwood Unit 1.
Commonwealth Edison Company has elected to install cable dampers and solid plugs in the tubes at R12C8 of SG C at Byron Unit 1 and R12C5 of SG B at Braidwood Unit 1.
oosts to-ptinus9o 25
The additional damping provided by the installed cable dampers has been shown to reduce the stress ratios in these tubes to a value significantly lower than 0.1. Thus, the fatigue usage per year for the aforementioned tubes is reduced to a negligible magnitude.
D0515:10 pt1/11259o 26
3.0 BACKGROUND
On July 15,1987, a steam generator tube rupture occurred at the North Anna Unit 1. l The ruptured tube was determined to be Row 9 Column 51 in steam generator "C."
The location of the opening was found to be at the top tube support plate on the cold leg side of +he tube and was circumferentialin orientation with a 360 degree extent.
3.1 North Anna Unit 1 Tube Rupture Event The cause of the tube rupture has been determined to be high cycle fatigue. The source of the stresses associated with the fatigue mechanism has been determined to be a combination of a mean stress levelin the tube and a superimposed alternating stress. The mean stress has been determined to have been increased to a maximum level as the result of denting of the tube at the top tube support plate and the alternating stress has been determined to be due to out of plane deflection of the tube U-bend above the top tube support caused by flow induced vibration.
These stresses are consistent with a lower bound fatigue curve for the tube material in an AVT water chemistry environment. The vibration mechanism has been determined to be fluid elastic, based on the magnitude of the alternating stress.
A significant contributor to the occurrence of excessive vibration is the reduction in damping at the tube to tube :upport plate interface caused by the denting. Also, the absence of antivibration bar (AVB) support has been concluded to be required for requisite vibration to occur. The presence of an AVB support testricts tube motion and thus precludes the deflection amplitude required for fatigue. Inspection data shows that an AVB is not present for the Row 9 Column 51 tube but that the actual AVB installation depth exceeded the minimum requirements in all cases with -
data for AVBs at many other Row 9 tubes. Also contributing significantly to the level of vibration, and thus loading,is the local flow field associated with the detailed geometry of the steam generator,i.e., AVB insertion depths. In addition, the fatigm properties of the tube reflect the lower range of properties expected for an AVT environment. In summary, the prerequisite conditions derived from the evaluations were concluded to be:
Fatique Requirements Prerequisite conditions Alternating stress Tube vibration Dented support Flow excitation
- - Absence of AVB Mean stress Denting in addition to applied stress Material fatigue properties AVT environment
- Lower range of properties DO51s:10 pti/112s90 3-1
l 3.2 Tube Examination Results l Fatigue was found to have initiated on the cold leg outside surface of tube R9C51
, immediately above the top tube support plate. No indications of significar.t accompanying intergranular corrosion were observed on the fracture face or on the immediately adjacent OD surfaces. Multiple fatigue initiation sites were found with major sites located at 110',120*,135'and 150*, Figure 3-1. The plane of the U bend is located at 45'with the orientation system used, or approximately 90' from the geometric center of the initiation zone at Section D D. High cycle fatigue striation :
spacings approached 1 micro inch near the origin sites, Figure 3 2.The early crack -
front is believed to have broken through wall from approximately 100* to 140'.
From this point on, crack growth is believed (as determmed by striation spacing, striation direction, and later observations of parabolic dimples followed by equiaxed dimples) to have accelerated and to have changed direction with the resulting crack front running perpend.cular to the circumferential direction.
3.3 Mechanism Assessment To address a fatigue mechanism and to identify the cause of the loading, any loading condition that would cause cyclic stress or steady mean stress he to be considered. The analysis of Normal, Upset and Yest conditions indicated e relatively low total number of cycles involved and a corresponding low fatigue usage, even when accounting for the dented tube condition at the plate. This analysis also
, showed an axial tensile stress contribution at the tube OD a short distance above the plate from operating pressure and temperature, thus providing a contribution to mean stress. Combining these effects with denting deflection on the tube demonstrated a high mean stress at the failure location, Vibration analysis for the tube developed the characteristics of first mode, cantilever response of the denttd tube to flow induced vibration for the uncracked tube and for the tube with an increasing crack sngle, beginning at 90* :0 the plane of the tube and progressing around on both sides to complete separation of the tube.
Crack propagation analysis matched cyclic deformation with the stmss intensities and striation spacings indicated by the fracture inspection and analysis. Leakage data and crack opening analysis provided the relationship between leak rate and circumferential crack length, Leakage versus time was then predicted from the crack growth analysis and the leakage analysis with initial stress amplitudes of 5,7, and 9 ksi. The comparison to the best estimate of plant leakage (performed after the event) showed good agreement, Figure -
a Dos 15 10-pt t /02590 32
l Based on these results,it followed that the predominant loading methanism responsible was a flow-induced, tube vibration loading mechanism. It was shown that of the two possible flow induced vibration mechanisms, turbulence and fluidelastic instability. that fluidelastic instability was the most probable cause. Due to the range of expected initiation stress amplitudes (4 to 10 ksi), the fluidelastic instability would be limited in displacement to a range of apprnximately [
Ja c. This isleu than the distance between tubes at the apex, { Ja.c.
It was further confirmed that displacement prior to the rupture was limited since no indication of tube U-bend (apex region) damage was evident in the eddy current signals for odjacent tubes.
Given 'he likelihood of limited displacement, fluidelastic instability, a means of establishing the change in displacement, and corresponding change m stress amplitude, was developed for a given reduction it, stability ratio (SR) Since the rueture was a fatigue mechanism, the change in stress amplitude resulting fro;n a reduction in stability ratio was converted to a fatigue u*, age benefit through the use of the fatigue curve developed. Mean stress effects were included due to the presence of denting and applied loadings. TLe resultsindicated that a 10%
reduction in stability ratic is needed (considsing the range of possible initiation stress amplitudes) to reduce the fatigue utage per year to less than Oh21 for a tube similar to Row 9 Column 51 at North Anna Unit 1.
msis so.pmms90 3-3 1
l 0-0 a
/
C4 C
q 180*
Region of herringbone Pattern I \ /
- 90' -
270' - ;
i k 3 .
I*I y E-E g,
A y A TAS q.
~{ \ F.T
- i^ Coarse Texture and Diaspled m Rupture
( Indicates Origins t
,, I, Figure 3-1. Approximate Mapping of acture Surface of Tube R9C51,5/G "C" Cold Leg, North Ann? nit 1 002:0 t o-cti c40990 3-4
8 = 1.5/1.6 w in.
t N Mesty Csica a %.
Attack 5 ar21 w in.
g 5 = 1.0/3.85 e in. gge.
3C i
80 h Parabolic 270* - i 14 01gles
) and V Internal Necking f
38 5 = 2.8/4.0 v in. I8d j 3,3 p,,,fd N M fB B Wearly tovi-Amed 5 = 4.1/6.9 e in. Otmples Note: Arrows Indicata Direction of Fracturt *ropagation l Figure 3 2. Schematic Representation of Features Observed During TEM <
Fractograhic Examination of Fracture Surface of Tube R9CS1, S/G "C" Cold Leg, North Anna Unit 1 l
[
- D0220- 10-ot11D40990 35 i
_ . . . . - . _ . . . _ . _ , . _ . _ _ . . . . _ _ .- _.---.._.-.._ __. --_ .- _ - _ _ . . _ . . _ _ - . _ - . . _ . . - . ~ - - - - _ - - _ - - -
- . _ . ._ . .- ... . . . . - . _ _ _ . . - . . - _ ~ . - - - . - . . - - . . -
i tenese l l l l l l l l l l l Calculated and observed leak rates versus time.
Observed values based on gaseous species condenser air ejector seeee -
SIGNA A = 5 KS!
SIGNA A = 7 KS! ,
SISMA A = 9 KS2 l }
^
O Ar-41 se.e-- D Xe-135 .
[ , g,,,7 W i U 1 I
b:
./ g 0 ..
l' o
sr.
o o
e . eene amme sees amo. == es TIME, MINUTES Figure 3 3. Calculated and Observed Leak Rates Versus Time cono io.ono4o990 3-6
4.0 CRITERIA FOR FATIGUE ASSESSMENT The evaluation rnethod and acceptance criteria are based on a relative compar ,
with the Row 9 Column 51 tube of Steam Generator C, North Anna Unit 1 This approach is necessary because (1) methods for direct analytical prediction of actual stability ratios incorporate greater uncertainties than a relative ratio method, and (2) the stress amphtude (or displacement) associated with a specific value of stability ratio can only be estimated by the analysis of North Anna Unit 1. For these reasons, the North Anna Unit 1 tubing evaluation was done on a relative basis to Row 9 Column 51 and a 10% reduction in stabihty ratio criteria was established to demonstrate that tubes left in service would be expected to have suffiaently low l vibration stress to preclude future fatigue rupture events l l
To accomplish the necessary relative assessment of Byron Unit 1 and Braidwood l Unit 1 tubing to Row 9 Column 51 of North Anna Unit 1, several cnteria are utihzed First, stability ratics are calculated for the Byron Unit 1 and Braidwood Unit 1 steam generators based on flow fields pred:cted by 3-D thermal hydraulic models and ratioed to the stabihty ratio for Row 9 Column 51 at North Anna Unit 1 based on a flow field obtained with a 3 D thermal hydraulic model with the same degree of refinement. These ratios of stability ratio (called relative stabihty ratios) for each potentially unsupported U-bend in the Byron Unit 1 and Braidwood Unit 1 steam generators shou!d be equivalent to 0.9 of R9C51, North Anna 1 (meeting the 10%
reduction in stabihty ratio criteria). This provides the first level of screening of susceptible tubes incorporating all tube geometry and flow field differences in the tube dynamic evaluation. It has the mherent assumption, however, that each tube has the same local, high flow condition present at Row 9 Column 51, North Anna Unit 1. To account for these differences, flow peaking factors can be incorporated in the relative stability ratios and the stress ratios.
The next step is to obtain stress ratios, the ratio pf stress in the Byron Unit 1 and Braidwood Unit 1 tube of interest to the stress ii, Row Column 51, North Anna Unit 1, and after incorporating the requirement that the,: ability ratio to Row 9 Column 51 (R9C51) for the tube of interest is equivalent ':o' Y.9, require the stress ratio to be
' 1.0. Thc stress ratio incorporates the tubeg eometry dif ferences with R9C51 in relation to the stress calculation and also inc .porates the rat:0 of flow peaking factor for the tube of interest to the flow pct ing factor for R9C51 (flow peaking factor is defined in Section 4.2). This should. rovide that all tubes meeting this criteria have stress amplitudes equivalent to 4.0 ksi.
Finally,the cumulative fatigue usage for plart operation to date and for continued operation with the same operating parameters is evaluated. A fatigue usage of 1.0 may not be satisfied by meetir.g the stress ratio criteria using the reference operating cyc'e evaluation since the reference cycle does not necessarily represent the exact duty cycle to date. Therefore, the time history of operation is evaluated on a normahzed basis and used together with the stress ratio to obtain a stress amphtude history. This permits the calculation of current and future fatigue usage oosts o as msw 4-1
l for comparison to 1.0. The time histories of operation used for fatigue usage calculations for Byron Unit 1 and Braidwood Unit 1 are summarized in Tables 41 and 4 2.
4.1 Stability Ratio Reduction Criteria For fluideiastic evaluation, stability ratios are determined for specific configurations of a tube. These stability ratios represent a measure of the potential for flow-induced tube vibration during service. Values greater than unity (1.0) indicate instability (see Section 5.1).
Motions developed by a tube in the fluidelastically unstable mode are quite large in comparison to the other known mechanisms. The maximum modal displacement (at the apex of the tube)is linearly related to the bending suess in the tube just above the cold leg top tube support plate. This relationchip applies to any vibration in that mode. Thus,it is possible for an unstable, fixe (! boundary condition tube to deflect an amount in the U bend which will produce f atigue inducing stresses.
The major features of the fluidelastic mechanism are illustrated in Figure 4-1. This figure shows the displacement response (LOG D) of a tube as a function of stability ratio (LOG SR). A straight line plot displayed on log log coordinates implies a relation of the form y = A(x)", whero A is a constant, x is the independent variable, n is the exponent (or power to which x is raised), and y is the dependent variable.
Taking logs of both side of this equation leads to the slope-intercept form of a str ght-line equation in log form, log y = c + n log x,where c = log A and represents the intercept and n is tha slope. In our case the independent variable x is the stability ratio SR, and the dependent variable y is tube (fluidelastic instability induced) displacement response D, and the slope n is renamed s.
From experimental results, it is known that the turbulence response curve (on log-log coordinates) has a slo;.e of approximately [ Ja.b.c. Test results also show that the slope for the fluidelastic response depends somewhat on the instability displacement (response amplitude). It has been shown by tests that a slope of
[ la,b,c is a range of values corresponding to displacement amplitudes in the range of [ ]Ac. whereas below, [ Ja.c are conservative values.
The reduction in response obtained from a stability ratio reduction can be expressed by the following equation:
3 S, C where Di and SR1 are the known values at the point corresponding to point 1 of Figure 4-1 and D2 and SR 2 are values corresponding to any point lower on this curve.
t 00515:10 pts /112590 4-2
Therefore, this equation can be used to determine the reduction in displacement response for any given reduction in stability ratio.
I This equation shows that there is benefit derived from even a very small percentage change in the stability ratio. It is this reduction in displacement for a quite small i reduction in stability ratio that formed the basis for demonstrating that a 10%
reduction in stability ratio would be sufficient to prevent Row 9 Column 51 from rupturing by fatigue.
The fatigue curve developed for the North Anna Unit 1 tube at R9C51 is from (
le.c. Thus, n.c where,6 is the equivalent stress aniplitude to o athat accounts for a maxinium stress of oy, the yield strength. The -3 sigma curve with mean stress effects is shown in Figure 4-2 and is compared to the ASME Code Design Fatigue Curve for Alloy 600 with the maximurn effect of mean stress. The curve utilized in this evaluation is clearly well below the code curve reflecting the effect of an AVT environment on fatigue and [the Smith-Watson-Topper technique]a.c for accounting for mean stress that applies to materials in a corrosive environment.
Two other mean stress models were investigated for the appropriateness of their use in providing a reasonable agreement with the expected range of init;ating stress amplitudes. These were the [ la.c shown in -
Figure 4-3, With a ( ]", the [
la.c.
The assessment of the benefit of a reduction in stability ratio begins with the relationship between stability ratio and deflection. For a specific tube geometry, the 00515-10 pt5/112590 43 m .-.- ,n _ me. . -,--r e- .--mm +---,m.- ,,a n.---v, . ,,---re .wm- ~ w- e v a e--,-- - s, -
-w -__
displacement change is directly proportional to change in stress so that stress has the same relationship with stability ratio, 4 A.f The slope in this equation can range from [ Ja.c on a log scale depending on the amplitude of displacement. Knowmg the stress resulting from a change in stability ratio from sri to SR2, the cycles to failure at the stress amplitude were obtained from the fatigue curve. A fatigue usage per year was then determined assuming continuous cycling at the natural frequency of the tube. The initial stress was determined to be in the range of 4.0 to 10.0 ksi by the fractography analr.is.
It was further developed that the maximum initiating stress amplitude was not more than 9.5 ksi. This was based on [
la.c. The corresponding stress levelis 5.6 ksi.
The maximum stress,9.5 ksi, would be reduced to [ la.( with a 10% reduction in stability ratio and would have a future fatigue usage of [ la,c per year at 75% availability, Figure 4 4. The minimum stress,5.6 ksi, would be reduced to
[ la.c ksi with a 5% reduction in stability ratio and would have future fatigue usage of [ la.( per year, Figure 4 5. In addition,if a tube were already cracked, the crack could be as large as [ la.c inch in length and thru-wall and would not propagate if the stress amplitudes are reduced to s 4.0 ksi.
Subsequent to the return to power evaluation for North Anna Unit 1, the time history of operation was evaluated on a normalized basis to the last cycle, confirming the conservatism of 9.5 ksi. [
la.(, the cumulative fatigue usage may then be computed to get a magnitude of alternating stress for the last cycle that results in a cumulative usage of 1.0 for the nine year duty cycle. The result of the iterative analysis is that the probable stress associated with this fatigue curve during the last cycle of operation was approximately [ la.( for R9C51, North Anna Unit 1, Steam Generator C, and that the major portion of the fatigue usage came in the second, third and fourth cycles. The first cycle was conservatively omitted, since denting is assumed, for purposes of this analysis, to costs io otsa12590 4-4
l i
have occurred during that first cycle. Based on this evaluation, the tube fatigue probably occurred over most of the operating history of North Anna Unit 1.
A similar calculation can be performed for the time history of operation assuming that[
la.c. On this basis,the effect of a 10%
reduction in stability ratio is to reduce the stress amplitude to 4.0 ksi and results in a future fatigue unge of ( la.c.
Other combinations of alternating stress and mean stress were evaluated with
-3 sigma and 2 sigma fatigue curves to demonstrate the conservatism of the 10%
reduction in stability ratio. Table 4 3 presents the results of the cases analyzed clearly demonstrating that the 10% reduction in stability ratio combined with a
-3 sigma fatigue curve and with maximum mean stress effects is cont ervative. Any higher fatigue curve whether through mean stress, mean stress model, or probability, results in greater benefit for the same reduction in stability ratio.
Further, for any of these higher curves, a smaller reduction in stability ratio than 10% would result in the same benefit. In addition,there is a large benefit in terms of fatigue usage for relatively small changes in the fatigue curve.
4.2 Local Flow Peaking Considerations I.ocal flow peaking is a factor on stubility ratio that incorporates the ef fect of local flow velocity, density and void fraction due to non unifarm AVB insertion depths.
The flow peaking factor is applied directly to tbc stability ratio obtained from thermal-hydraulic analysis that does not account for these local geometry effects.
Being a direct factor on stability ratio, a small percentage increase can result in a significant change in the prediction of tube retponse.
Since the evaluation of Byron Unit 1 and Braidwood Unit 1 tubing is relative to R9C51, North Anna Unit 1,the flow peaking factors are also applied as relative ratios, i.e., a ratio of Byron Unit 1 and Braidwood Unit 1 tubing to R9C51 at North Anna Unit 1. The flow peaking relative instability is obtained by testing in the air test rig described in Section 5.4, where the peaking f actor is defined as the critical velocity for R9C51 AVB pattern compared to critical velocity for a uniform AVB pattern. As explained in Section 8.0,the minimum value of [ ]a.b.c is appropnate for R9C51 of North Anna 1. The peaking factor for a tube in Byron Unit 1 and Braidwood Unit 1 tubing is therefore divided by [ ]a,b,c and the resulting relative flow peaking is multiplied times the relative stability ratio based on ATHOS results. If the peaking !
factor is 1.0,the relative flow peaking is [ Ja,b.c.
As a further demonstration of the conservatism of [ la b.c as the minimum flow peaking factor for R9C51,the stress amplitude of 7.0 ksi obtained from iterating on cumulative fatigue usage (and selected as the nominal value from fractography analysis) was used to back calculate the apparent stability ratio and then the aosis ius ms90 4-5
. - . _ _ . _ . _ _ _ . . . . _._ _ ~ _ _ - -_ __.
apparent flow peaking factor, e i swing for a range of slopes of the instability curve from 10 to 30,the stability ratio is in the range of 1.1 to 1,4 and the flow peaking factoris in the range of 1.8 to 2.2. This range of flow peaking agrees with the range
, of flow peaking factors measured in the air tests and is considered to be the best estimate of the range of the R9C51 flow peaking factor.
The range of stability ratios,1.1 to 1.4,is based on a value of 0.63 obtained with ATHOS results without flow peaking and with nominal damping that is a function of modal effective void fraction (MEVF) MEVF is calculated using the formula:
a,C
]
The nominal damping reflects the nominal reduction in damping that occurs with denting at the tube support plate. Therefore, a minimum damping scenario that is independent of void fraction is not considered to be credible and is not addressed in <
the evaluation that follows.
4.3 Stress Ratio Considerations in Section 4.1, a 10% reduction in stability ratio was established to reduce the stress amplitude on the Row 9 Column 51 tube of. North Anna Unit 1 to a level that would
- not have ruptured,4.0 ksi. Tu apply this same criteria to another tube in the same or another steam generator, the differences in (
Ja.c.
a,C 1
oosisa o.ns,n mo 4-6
a,c I
l I
I i
By establishing their equivalent effect on the stress amplitude that produced the tube rupture at North Anna 1, several other effects may be accounted for. These include a lower mean stress (such as for non-dented tubes), different frequency tubes from the [ Ja.(,e hertz frequency of R9C51, North Anna 1, and shorter design basis service.
oos t s;to-otsn us90- 47
In the case of lower mean stress, the stress amplitude hat would have caused the failure of R9C51, North Arina 1,wotdd have been nigher. !
Ja c.
A lower or higher frequency tube wCJld not reach a usage of 1.0 in the same length of time as the R9C51 tube due to tN different frequency of cycling. The usage accumulated is proportional toire frequency and,therefore, the allowable number of r.ycles to reach a usage of 1.0 is inve:sely proportional to frequency. The equiva!ent number of cycles to give the usage of 1.0 for a different f;equency tube
[
]a c.
For a different timt: basis for fctigue usage evaluation, [
]a,c.e.
Knowing the magnitude of the stress ratio allows: (1) the determination of tubes that do not meet a value of 51; and (2) the calculation of maximum stress in the acceptable tubes, u
Having this maximum stress permits the evaluation of the maximum fatigue usage for Byron Unit 1 and Braidwood Unit 1 based on the time history expressed by 9
normalized stability ratios for the duty cycle (see Section 7.4).
D05151D ot:1120490 4-8
Table 41 Byron Unit 1 6 Time History of Operation Used for Fatigue Analysis Effective Full Cycle
- Initial criticality (2/2/85) through cycle 3 (1/5/90) i 005 t 5:10- p15/112590 4-9
. . - . ~ . - - - . . _ _ . - . . . - . . .- - . .
1 Table 4 2
'dwood Unit 1 Time History of Or tion Used for Fatigue Analysis Effective Full Cycle Power Hours 1 10,182.5*
2 2,633.3**
Total 12,815.8 EFPH (534.0 EFP Days)
(At full temperature)
(With Thot reduction - cycle 2 EFPH as of 4/15/90) oos .e t o.ptsn uS90 4-10
Table 4-3 Fatigue Usage Per Year Resulting From 5tability Ratio Reduction SR,% Stress Fatigue Mean Stress Usage Reduction Basis (1) Curve l2) Modei Per Year
~
a,c 5, 9 yrs to ja.c l
fail [
5, 9 yrs to fail [ Ja.c
- 5. 9 yrs to fail [ la.c
- 10. max. stress amplitude (4)
[ ja.c
- 10. max. stress amplitude (4)
[ ja.c
- 10. max. stress amplitude (4:
[ ]a,c
- 10. max, stress amplitude (4)
[ Ja.c
- 10. max. stress based on duty cycle (5)
[ Ja.c _ _
(1) This give the br sis for selection of the initiating stress amplitude and its value in ksi.
(2) Sm is the ma .imum stress applied with Sm = Sm n + 5..
(3) [ ]* c .
(4) Cycles to failure implied by this combination of stress and fatigue properties is notably less than in plied by the operating history. Consequently this combination is a conservative. bounding estimate.
(5) Cycles to failuro implied by the operating history requires [ la c fatigue curve at the maximum strass of [ ]*A 0051 s:1o-otsn t 2590 4-11 m..
-i a,b,c L .
Figure 4-1. Vibration Displacement vs. Stability Ratio 00515:10-pts /112590 4-12
_ aiC
'-~
Figure 4-2. Fetigue Strength of Alloy 600 in AVT Water at 600 F .
1 00515:10-pt5/112590 4-13
. ,. .. m _ _ _ .
1 a,c E
i' i
. Figure 4-3. Fatigue Curve for Alloy 600 in.AVT Water Cornparison of Mean Stress Correction Models 00515:10-pt541290 4-14
i
,.-. B,C I
l i
)
l i
1 Figure 4-4. Modified Fatigue with 10% Reduction in Stability Ratio for Maximum Stress Condition 00515:1D-pts /112590 4-15
0,C e-Figure 4-5. Modified Fatigue with 5% Reduction in Stability Ratio for Minimum Stress Condition oos15.10-pts >112s90 4-16
S.0 SUPPORTING TEST DATA This section provides a mathematical description of the fluid-elastic mechanism, which was determined to be the most I;kely causative mechanism for the North Anna -
tube rupture, as discussed in Section 3.3, to highlight the physical conditions and corresponding parameters directly related to the event and associated preventative measures. The basis for establishing the appropriate values and implications associated with these parameters are provided. Where appropriate, test results are presented.
5.1 Stability Ratio Parameters Fluid-elastic stability ratios are obtained by evaluations for specific configurations,in terms of active tube supports, of a specific tube. These stability ratios represent a measure of the potential for tube vibration due to instability during service. Fluid-elastic stability evaluations are performed with a computer program which provides for the generation of a finite element model of the tube and tube support system.
The finite element model provides the vehicle to define the mass and stiffness matrices for the tube and its support system. This information is used to determine the modal frequencies (eigenvalues) and mode shapes (eigenvectors) for the lineaily supported tube being considered.
The methodologyis comprised of the evaluation of the following equations:
Fluid-elastic stability ratio = SR = Uen/Uc for mode n, where Uc (critical velocity) and Uen (effective velocity) are determined by:
U, = pf,D [(m,8)/ip,D )mj 2 [1]
and; N [2]
1 (p, / p,) U, 2 $,, 2 ,,
U ' = ' " 'N ern T (m, / m, ) p,,,
- z, jul where, D -= tube outside diameter, inches Uen = effective velocity for mode n, inches /sec 00515:1D pt6/112590 5-1
l l
N = number of nodal points of the finite element model
= number of degrees of freedom in the out-of-plane direction mj,Uj,pj = mass per unit length, crossflow velocity and fluid density at node j, respectively po,mo = reference density and reference mass per unit length, respectively (any representative values) sn = logarithmic decrement (damping) tjn = normalized displacement at node j in the nth mode of vibration zj = average of distances between node j to j 1, and j to j + 1 p = an experimentally correlated instability constant Substitution of Equations [1] and [2]into the expression which defines stability ratio, and cancellation of like tenns, leads to an expression in fundamental terms (without the arbitrary reference mass and density parameters). From this resulting expression, it is seen that the stability ratio is directly related to the flow field in terms of the seconaary fluid velocity times square-root-density distribution (over the tube mode shape), and inversely related to the square root of the mass distribution, square root of modal frequency, and the instability constant (beta).
The uncertainty in each of these parameters is addressed in a conceptual manner in Figure 5-1. The remainder of this section (Section 5.0) provides a discussion, and, where appropriate, the experimental bases to quantitatively establish the uncertainty associated with each of these parameters. In addition, Section 5.3 provides the experimental basis to demonstrate that tubes with (
ja.c. This implies that those tubes [
la.c would not have to be modified because their instability reponse amplitude (and stress) would be small. The very high degree of sensitivity of tube response (displacement and stresses) to changes in the v& city times square-root-density distribution is addressed in Section 4.0. This is iniportant in determining the degree of change that can be attained through modifications.
Frequency It has been demonstrated by investigators that analytically determined frequencies are quite close to their physical counterparts obtained from measurements on real structures. Thus, the uncertainty in frequencies has been shown to be auite small.
This is particularly appropriate in the case of dented (fixed boundary condition) tubes. Therefore, uncertainty levels introduced by the frequency parameter are expected to be insignificant (see also " Average Flow Field" subsection below).
00s15:10 pt6/112590 5-2
i instability Constant (Beta)
The beta (instability constant) values for stability ratio and critical velocity evaluations (see above equations) are based on an extensive data base comprised of both Westinghouse and other experimental results. In addition, previous field experiences are considered. Values have been measured for fulllength U-bend tubes in prototypical steam / water environments. In addition, measurements in U-bend air models have ben made with both no AVB and variable AVB supports (Figure 5 3).
To help establish the uncertainties associated with ATHOS flow velocity and density distribution predictions on stability analyses, the Model Boiler (MB-3) tests performed at Mitvabishi Heavy Industries (MHl) in Japan were modeled using ATHOS. A beta value consistent with ATHOS predicted flow conditions and the MB-3 measured critical velocity was determined. These analyses supported a beta value of
[ ]a.b.c.
A summary of the test bases and qualifications of the beta values used for these assessments is provided by Figure 5-2. The lowest measured beta for tubes without AVBs was a value of [ la.b,c. This value is used for the beta parameter in all stability ratio evaluations addressed in this Report (see also " Average Flow Field" subsection below).
Mass Distribution The mass distribution parameter is based on known information on the tube and primary and secondary fluid physical properties. The total mass per unit length is comprised of that due to the tube, the internal (primary) fluid, and the external (secondary) f!uid (hydrodynamic mass). Data in [ la c suggests that at operating void fractions [
ja,c.
Tube Dampina Test data are available to define damping for clamped (fixed) tube supports, appropriate to dented tube conditions,in steam / water flow conditions. Prototypic U bend testing has been performed under conditions leading to pinned supports.
The data of Axisa in Figure 5-4 provides the principal data for clamped tube conditions in steam / water. This data was obtained for cross flow over straight tubes.
Uncertainties are not defined for the data for these tests. Detailed tube damping data used in support of the stability ratio evaluations addressed in this report are provided in Section 5.2, below.
D0515:11pt6/120490 5-3
Flow Field - Velocity Times Square-Root-Density Distribution Average and U-bend local flow field uncertainties are addressed independently in 4 the following.
Average Flow Field Uncertair. ties in the average flow field parameters, obtained from ATHOS analyses, coupled with stability constant and frequency, are essentially the same for units with dented or non-dented top support plates. If the errors associated with these uncertainties were large, similar instabilities would be expected in the non-dented units with resulting wear at either the top support plate or inner row AVBs.
Significant tube wear has not been observed in inner row tubes in operating steam generators withnut denting. Thus, an uncertainty estimate of about [ ja.c for the combined effects of average flow field, stability constant and frequency appear, to be reasonable. To further minimize the impact of these uncertainties, the Byron Unit 1 and Braidwood Unit 1 tubes are evaluated on a relative basis, so that constant error factors are essentially eliminated. Thus, the uncertainties associat3d with the average velocity times square-root-density (combined) parameter are not expected to be significant.
U-Bend 1.ocal Flow Field Non-uniform AVB insertion depths have been shown to have effects on stability ratios. Flow peaking, brought about by the " channeling" effects of non-uniform AVBs, leads to a local perturbation in the velocity times square-root-density parameter at the apex of the tube where it will have the largest effect (because the apex is where the largest vibration displacement occur). Detailed local flow field data used in support of the stability ratio evaluations addressed in this report are provided in Section 5.2, below.
Overall Uncertainties Assessment Based on the above, discussions, and the data provided in the following sections, it is concluded that local flow peaking is likely to have contributed significantly to the instability and associated increased vibration amplitude for the failed North Anna tube. Ratios of stresses and stability ratios relative to the North Anna tube, R9C51, are utilized in this report to minimize uncertainties in the evaluations associated with instability constants, local flow field effects and tube damping.
5.2 Tube Damping Data The damping ratio depends on several aspects of the physical system. Two primary determinants of damping are the support conditions and the flow field. It has been shown that tube support conditions (pinned vs clamped) affect the damping ratio cos t snocen us9a 5-4 5
1 l
significantly. Further,it is affected by the flow conditions,i.e., single-phase or two-phase flow. These effects are discussed below in more detail.
[ la.c indicates that the damping ratio in two phase flow is a sum of contributions from structural, viscous, flow-dependent, and two phase damping.
The structural damping will oe equal to the measured damping in air. However,in two phase flow damping ratio increases significantly and is dependent on the void fraction or quality, it can Je shown that the damping contribution from viscous effects are very small.
Damping ratios for tubes in air and in air-water flows have been measured and reported by various authors, however, the results from air-water flow are poor representations of the actual conditions in a steam generator (steam-water flow at high pressure). Therefore,where available, results from prototypic steam water flow conditions should be used. Fortunately, within the past few years test data on tube vibration under steam-water flow has been developed for both pinned and clamped tube support conditions.
Two sources of data are particularly noteworthy and are used here. The first is a large body of recent, as yet unpublished data from high pressure steam-water tests conducted by Mitsubishi Heavy industries (MHI). These data were gathered under pinned tube support conditions. The second is comprised of the results from tests sponsored by the Electric Power Research Institute (EPRI) and reported in [
la,c.
The damping ratio results from the abeve tests are plotted in Figure 5-4 as a function of void fraction it is irnoortant to not<a that the void fraction is determined on the basis of [ la c. The upper curve in the figure is for pinned support conditions. This curve represents a fit to a large number of data points not shown in the figure. The points on the curve are only plotting aids, rather than specific test results.
The lower curve pertains to the clamped support condition, ootained from
[ ]a.c Void fraction has been recalculated on the basis of slip flow. It may be noted that there is a significant difference in the damping ratios under the pinned and the clamped support conditions. Damping is much larger fc : pinned supports at all void fractions. Denting of the tubes at the top support plate effectively clamps the tubes at that location. Therefore, the clamped tube support curve is used in the current evaluation to include the effect of denting at the top tube support plate.
The[ ja,c data as reported shown a damping value of 0.5% at 100% void fraction. The 100% void fraction condition has no two phase damping and is considered to be affected principally by mechanical or structural damping.
Westinghouse tests of clamped tube vibration in air has shown that the mechanical damping is only [ ]a.c rather than the 0.5% reported in [ la.c.
oos t s:t o. pts <i20490 5-5
1 Therefore the lower curve in Figure 5-4 is the [ ja,c data with all 1
damping values reduced by [ la.c.
5.3 Tube Vibration Amplitudes With Single-Sided AVB Support
}
"A series of wind tunnel tests were conducted to investigate the effects of tube /AVB eccentricity on the vibration amplitudes caused by fluidelastic vibration.
1 l
[
4 4
la.c. Prior test results obtained during the past year using this apparatus have demonstrated that the fluideiastic vibration charatteristics observed i in the tests performed with the cantilever tube apparatus are in good agreement
- with corresponding characteristics observed in wind tunnel and steam flow tests using U-bend tube arrays. A summary of these prior results is given in Table 5-1.
An overall view of the apparatus is shown in Figure 5-5. Figure 5-6 is a top view of j the apparatus. [ 1 i
1 l
4 i
i 4
1 la.c.
As shown in Figure 5-7, the tube vibration amplitude below a critical velocity is caused by [
]a,c.
Figure 5-7 shows the manner in which the zero-to-peak vibration arnplitude, expressed as a ratio normalized to [ ]a,c varies when one gap remains at [
]a.c. For increasing velocities, up to that D0515:10-pt6/112590 5-6
corresponding to a stability ratio of [
la.c. Figure 5-8 shows typical vibration amplitude and the tube /AVB impact force signals corresponding to those obtained from the tests which provided the results shown in Figure 5 7. As expected, impacting is only observed in the [ la.c.
It is concluded from the above test results that,[
Ja.c.
5.4 Tests Determined the Effects on Fluideiastic Instability of Columnwise Variations in AVB insertions Depths This section summarizes a series of wind tunnel tests that were conducted to investigate the effects of variations in AVB configurations on the initiation of fluidelastic vibration. Each configuration is defined as a specific set of insertion depths for the individual AVBs in the vicinity of an unsupported U bend tube.
The tests were conducted in the wind tunnel using a modified version of the cantilever tube apparatus described in Section 5.3. Figure 5-9 shows the conceptual design of the apparatus. The straight cantilever tube,[
Ja.c.
[
]**, Figured 5-11 shows the AVBs, when the side panel of the test section is removed. Also shown is the top flow screen which is [
la,' The AVB configurations tasted are shown in Figure 5-12. Configuration la corresponds to tube R9C51, the failed tube at North Anna. Configuration 2a cosis iocetusso 5-7
corresponds to one of the cases in which the AVBs are inserted to a uniform depth and no local velocity peaking effects are expected.
As shown in Figure 5 9,[
ja,c.
All the tubes except the instrumented tube (correspondmg to Row 10) are [
Ja c. As discussed in Section 5.3, prior testing indicates that this situatio, provides a valid model. The instrument tube [
ja.c as shown in Figure 510. Its [ la c direction vibrational motion is measured using a non-contacting transducer.
[
Ja.c. The instrumented tube corresponds to a Row 10 tube as shown in Figure 5 9. However, depending on the particular AVB configuration,it can reasonably iepresent a tube in Rc,ws 8 through 11. The AVB profile in the straight tube modelis the average of Rows 8 and 11. The difference in profile is quite small for these bounding rows.
[ la,c using a hot film anemometer located as shown in Figure 5-9.
Figure 5-13 shows the RMS vibration amplitude, as determined from PSD (power spectral density) measurements made using an FFT spectrum analyzer, versus flow velocity for Configuration la (which corresponds to tube R9C51 in North Anna).
Data for three repeat tests are shown and the critical velocity is identified.' The
- typical rapid increase in vibration amplitude when the critical velocity for fluidelastic vibration is exceeded is evident.
The main conclusions from the tests are:
1.- Tube vibration below the critical velocity is relatively small, typical of turbulence-induced vibration, and increases rapidly when the cntical velocity
. for the initiation of fluideiastic vibration is exceeded.
- 2. Configuration la (R9C51 in North Anna) has among the lowest critical velocity of all the configurations tested.
- 3. Configuration 1b,with a similar geometry, but slightly higher peaking factor than la, has been run periodically to verify the consistency of the test apparatus and its calibration.
oosis:io-pte<ii2s90 5-8
. . _ _ . . _ _ _ _ . . m ._._ _
The initial test results obtained in support of the Byron /Braidwood Unit 1 evaluation
'are summarized in Table 5-2. The test data is presented as a velocity peaking ratio; a ratio of critical velocity for North Anna tube R9C51 configuration la, to that for each Byr_on Unit 1 and Braidwood Unit 11 AVB configuration evaluated.
5.5 References O,C 9
00515:10 p 6/112590 5-9
Table 51 Wind Tunnel Tests on Cantilever Tube Model i
OBJ ECTIVE: Investigate the effects of tube /AVB fitup on floe.-induced tube
) vibration.
APPARATUS: Array of cantilevered tubes with end supports [
}
Ja.c.
MEASUREMENTS: Tube vibration amplitude and tube /AVB impact forces or preload forces.
RESULTS:
> a,b,c P
D0515 10-pt6/11259o 5-10
Table 5-2 Fluidelastic Instability Velocity Peaking Ratios for Columnwise Variation in AVB insertion Depths (Byron /Braidwood 1)
Type of Insertion Peaking Ratio Configuration U ic./Un
_ _a,b,c Note: Un is instability velocity at inlet for type n of AVB insertion configuration.
00515 ID-pt6 I12590 5-11
_ ___~
a,c Figure 5-1. Fluidelastic Instability Uncertainty Assessment 00515:10 pt6/112590 5 12
- - -- . . . .. -. - - -- - = . . .-
l l
U-Bend Test Data
- 1) MS-3 Tests p values of [ la,b,c
- 2) MB ? Tests yof[ ja,b,c
- 3) Air Model Tests pof[ Ja,b,cwithcut AVBs Tendency for p to increase in range of [ ]a.b.c with inactive AVBs(gaps at AVBs)
Verification of instability Conditions
- 1) Flow conditions at critical velocity from MB-3
- 2) Measured damping for the specific tube
- 3) Calculated velocities from ATHOS 3D analysis
- 4) p determined from calculated critical values Good agreement with reported p values
- 5) . ATHOS velocity data with p of [ ]c,b,c and known damping should not significantly underestimate instability for regions of uniform U-bend flow -
Figure 5-2. Instability Constant-oosis:io-stsci12590 5-13
a,b,c i
Figure 5-3 Instability Constant, p, Obtained for Curved Tubes from Wind Tunnel Tests on the 0.215 Scale U Bend Model cost sa omscumo 5-14
b 5
i I
a,b.c ,
l :.
I -
l>
l _ ._
1 Figure 5-4. Damping vs. Slip Void Fraction t
D0515:10 pt6/112590 5-15
_. . . . . .. . .-.. . . . _ . _ - . _ - - . . _ . .. ._ _ ._ -._. .. ~ . . _ - - - - -
a,b c
- ._ i I
L l
l l.
i:-
I' l
!- ~
I l
4 Figure 5-5. Over View of Cantilever Tube Wind Tunnel Model
- oasis:taptsn us90 5.16
8 b,C i
~
j F
Figure 5-6. Top View of the Cantilever Tube Wind Tunnel Model D0515 10-pt6/112 s,o 5-17
. . . . ._. . _ .. . . . . . , _ _ - _ . . - - . - - - _ . - . ~ _ - - . . - . . - .
a,b.c 1
l
.1 I
e j Figure 5-7.- Fluideiastic Vibration Amplitude with Non-Uniform Gaps i
i.
oosis:io-pteiusso 5-18
- -.-. .- . . . - . ~ _ _ . . = . _ . = . .
a,b,c
~ -
Figure 5-8. Typical Vibration Amplitude and Tube /AVB Impact Force Signals for Fluidelastic Vibration with Unequal Tube /AVB Gaps D0515:10-pt6/112590 5 19
l a b.c Figure 5 9. Cenceptual Design of the Ap 3aratus for Determining the Effects on Fluidelastic Instabi ity of Columnwise Variations in AVB insertion Depths D0515:10-pt6/112590 S.20 a.. . . _ . . . _ . .___~. , _ , . , _ _ . _ . , ., .. _ , _ _. _ , . _ , . _ . _ . _ _ _ . _ , . _ _ _
i f
1 i
1 4
a.b.c i
_ i
\.
4 I
C i
1.
f
- 7 4
Figure 510. Osetain View of Wind TunnelTest Apparatus ,
D051$.10 pt6/112590 - 5 21
~ ..-_ -_
i 4
f 1
i a,b;c 1 _
I
?
Figure 511. Side View of Wind Tunnel Apparatus with Cover Plates Removed to Show Simulated AVBs and Top Flow Screen oosis:to otsn us90 5 22
___ ~ _ _ _ _ _ _ _ . . . _ . - _ . _ _ _ _ _ _ _ . . . _ _ . _ . . . _ _ . _ . _ _ _ _ _ _ _ . . . . _ _ _ _ . . _ _ . _ _ _ _ . _ .
i 8iC ]
1 i
i h
-I, 5
4 i
h 5
E 1
?
?
i i
t n
t
-_ 1 Figure 512. AVB ConfigurationsTested for Byron /Braidwood 1
' D0515:1 D-pt6'11259o 5 23
.-. ,- ...p..7 y ~my+.,v--,,m,cy- ww.w, - , . , ,w w -.s .v , .-.,,-.,%,,,yv -,wr,.,yromsm-w,,,-= wiv -ww w .ry m- w w-r win.-*ww==ww---rs-- -we r*=---m<- * * * ' ~ + - 'vr*
a.b c r- -
Figure 5-13, TypicalVariation of RMS \/ibration Amplitude with Flow Velocity for Configura*. ion la in Figure 512 00 $15 10-pt 6'112590 5 24
6,0 EDDY CURRENT DATA AND AVB POSITIONS 6.1 AVB Assembly Design -
I I
L Ja.c.e.
Since the purpose of this analysis is to evaluate notentially unsupported tubes at, or near the point of maximum AVB insertion, only the dimensions and EC data pertaining to the ' lower' AVB's are used. Review of the EC data for Byron Unit 1 and Braidwood Unit 1 shows that AVB insertion depths follow the typical model D pattern, being r216tively uniform in the regions between Columns 31 and 84 which correspond to the ' flat' contour of the tube bundle in this region.
6.2 Eddy Current Data tar AVB Positions The AVB insertion detths were c%termined on the basis of interpretation of the eddy current test (ECD datc. To locate the AVB's, the ECT data traces were searched for the characteristic peaks seen in the signals, which indicate the intersection of the tube with an AVB (o a tube support plate). A typical ECT signal trace for an AVB is shown in Figure 61. Since arnb!guity can occur in the interpretation of the ECT data due to inability of most ECT sensors to differentiate on which side of a tube a ' visible' AVB is located, other information murt often be used to assist in establishing the locaticn of the AVB's. This information may include a review cf consistency with the design of the AVB assembiy. consistency of data from adjacent columns, and verification by means of ' projection' calculations to determine the AVB depth of insertion which will be plotted.
The flow peaking and tube support evaluation for Byron Unit 1 and Braidwood Unit 1 is based on the AVB insertloa plots shown in Figure 6 2 through 6 9. In these plots, the direct observation vsluet have been repeated from their original format, and supplemented with AVB position values based on ' projection' calculations where appropriate.
6.2.1 AVB Insertion Depths The direct observation data (the number of AVB intersections seen by the EC probe) are the principal basis for determining tube suppoit by the AVB's. Two or more AVB indications (on either >ide of the U bend apex) indicate that a tube is supported. The logic is that for an AVB to produce two indications,it must be inserted f ar enough oosis 10-o 3/11:590 61
i beyond the tube center line tc for its, ex to break contact with the tube, thus producing two separate peaks on the signal trace.
The guidelines used for interpretation of ' direct obervation' of ECT data for tubes provide a 'yes no' type of answer. Tubes having two or more signals can be assumed to be supported, while tubes with no signals can be assumed to be unsupported AVB plots based on ' direct observation'in this report show the AVB's as being i inserted to a point midway between the last row of supported tubes, and the first row of unsupported tubes ('zero' AVB siDnals). The accuracy of these AV4 p'ots then,is slightly less than 10.5 tube pitch. Where the combination of tube stability ratio for a given row of tubes and the plotted insertion distances of neighboring AVB's indicate that flow peaking limits will not be exceeded, AVG insertion plots based on ' direct observation' are sufficiently accurate for the tube fatigue evaluation. Where the combination of tube stability ratio and potential flow peaking effects requires more accurate placement, other techniques must be used to locate the AVB's. In the AVB insertion plots shown in Figures 6 2 through 6 9, columns in which no projection values are specified have AVB insertions based on
' direct observation' of the ECT data, as described above. For columns in which projection values are specified, the AVB insertions are based on the methods described in the following section; in these cases, the accuracy of the plotted AVB insertion is approximately ,1. 0.25 tube pitch.
Verification of tube support when only a single AVB indication is present requires additional analysis, The apex of the AVB could extend slightly beyond that of the ,
tube centerline providing support and producing a signal trace several inches long, .
but due to noise, surface deposits, or other anomalies, it may lack the two distinct peaks of a 'two' signal. It as more likely that only point contact exists between an AVB and the tube in question; or that an AVB may be close enough to a tube that its magnetic field is detected by the EC probe, but the AVB is withdrawn far enough ,
that it does not contact the tube 6.2.2 AVB Proiection The projection technique is useful where noisy or spurious ECT signals prevent positive location of an AVB or where data it unavailable due to a tube having been plugged. [
Ja.c.
In the case where the AVB characteristic signals can not be confidently determined due to a noisy signal or pre existing plugged tubes, data for locating the AVBs is provided from [ _
Ja.c.
005t5:10-pt3/112590 6*2
__ . _ - - _ _ _ _ . _ - _ _ _ _ _ _ . _ _ .- . _ _ _ _ ~ . . _ _ _ _ _ _ _ _ . _ _ _
i i
I l
t Jo.c.
i Ja.c ,
6.3 Tube Denting at Top Tube Support Plate Since the tube vibration analyses are based on the conservative assumption that all tubes in the area of interest are structurally ' fixed'in the TSP holes, as if by denting or corrosion, the current condition of the tube / TSP interf ace does not influence the disposition of the tubes found to be susceptible to f atigue. Subsequent to the identification of the susceptible tubes in Byron 1 (5G C R 12C8) and Braidwood 1 l (SG B R12C5), eddy current data for the susceptible tubes as well as adjacent tubes were examined to evaluate the incidence of corrosion and/or denting at the top tube supnort plate. The analysis showed no indications of corrosion with magnetite or denting in the regions of interest. ,
6.4 AVB Mapinterpretations The Byron Unit 1 and Braidwood Unit 1 AVB's have a nominalinsertion depth to Row
- 10. Evaluation of the EC data indicates thatin Byron Unit 1 in the area of interest, (Row 12 through Row 8) between columns 02 and 113, all but one of the Row 12 tubes, all but 15 row 11 tubes, and all but 58 row 10 tubes are supported. For Braidwood Unit 1 in the area of :nterest, all but four of the Row 12 tubes, all but 22 :
of the Ro v 11 tubes, and all but :W Row 10 tubes are supported.
Byron Unit 15G A The AVD map for Byron 1 SG A is shown in Figure 6.2. A listing of unsupported tubes is given in Table 61. Ten Row 10, ano thirty Row 9 tubes are unsupported. The highest significant flow peaking factors in the steam generator (see Sections 8 and 9) are found at tube locations R8C27, R8C75, R8C88, R9C19, R10C4, R10C110, and R10C111. The stress ratios for these and all remaining tubes are acceptable.
4 Byron Unit 1 SG B The AVB map for Byron 1 SG B is shown in Figure 6 3. A listing of unsupported tubes is given in Table G 1. Five Row 11, seventeen Row 10, and thirty seven Row 9 tubes are unsupported. For the tubes atlocations R10C104 and R11C6, AVB sensitivity analyses D0s15,1D eth112590 63 I
I have been performed to ensure that conservahva flow peaking factors are selected.
The highest significant flow peaking factors (see Sections 8 and 9) are found at tube locations R8C88, R10C104, R11C6 and R11C9. The stress ratios for these and all remaining tubes are acceptable.
Byron Unit 1 SG C The AVB rnap for Byron 15G C is shown in Figure 6-4. A listing of unsupported tubes is given in Table 61. One Row 12 tube, nine Row 11 tubes, twenty one Row 10 tubes, and thirty nine Row 9 tubes are unsupported. The highest significant flow peaking factors (see Sections 8 and 9) are found at tube locations R8C27, R8C31, R11C4, and R12C8. An AVB sens,tivity analysis has been performed for the tube at R12C8 to ensure that a conservative flow peaking factor is selected. In the dented condition, the tube at R12C8 exceeds the limiting stress ratio criterion and has a fatigue usage factor which is greater than 1.0 for 40 years of operation, leading to the recommendation that the R12C8 tube be removed from service. The stress ratios and fatigue usages for the remaining tubes are acceptable.
Byron Unit 1 SG D The AVB map for Byron 1 SG D is shownin Figure 6 5. A listing of unsupported tubes is given in Tabte 6-1. One Row 11 tube, seven Row 10 tubes, and forty Row 9 tubes are unsupponed. f or the tube at location R11C107, an AVB sensitivity analysis has been performed to ensure that a conte vative flow peaking factor is selected. The highest significant flow peaking factors (see Sections 8 and 9) are found at tube locetions R6C27, RSC88, R9C23, R9C92, and R11C107. The stress ratios for these and all remaining tubes are acceptable.
Braidwood Unit 15G A The AVB map for Braidwood 1 SG A.is shown in Figure 6 6. A listing of unsupported tubes is given in Table 6 2. Five Row 11 tubes, fifteen Row 10 tubes and forty-one Row 9 tubes are unsupported. For the tube at R11C104, an AVtl sensitivity analysis has been performed to ensure that a conservative flow peaking factor is selected.
' The highest significant flow peaking factors (see Sections 8 and 9) are found at tube locations R9C92 and R9C98. The stress ratios for these and all remaining tubes are acceptable.
Braidwood Unit 1 SG B The AVB map for Braidwood 1 SG B is shown in Figure 6 7. A listing of unsupported tubes is given in Table 6 2. Four Row 12 tubes, seven Row 11 tubes, seventeen Row 10 tubes, and forty Row 9 tubes are unsupported. Particular attention has been given to the AVB positions in the region of Row 12 Columns 2-5. In this region, the AVBs are not inserted beyond row 12. The potential for the AVB between Columns 2 3 to move in the flow and impact on the outside of the tube bundle has been D0s1510 pt3/112590 64
considered; this type o' behavior is not expectc-d, since this AVB is short and is welded to the retaining ring,which provides rigidity. Furthermore, testing performed by Westinghouse has shown that for all but the longest AVBs, an AVB which is not constrained by adjacent tubes does not experience fluideiastic excitation in the U bend flow field.
For the tubes at R11C8, R12C3, R12C4 and R12C5, AVB sensitivity analyses have been performed to ensure that conservative flow peding factors are selected. The highest significant flow peaking factors (see Sections 8 and 9) are found at tube locations R9C17, R9C18, R11C8, and R12C5. The AVB insertions at R12C2, R12C3 and R12CA do not produce significant flow peaking factors. In the dented condition, the tube at R12C5 exceeds the limiting stress ratio criterion and has a fatigue usage factor which is greater than 1.0 for 40 years of operation, leading to a recommendation that the R12C5 tube be removed from service. The stress ratios and fatigue usage factors for all remaining tubes are acceptable.
Braidwood Unit 1 SG C The AVB map for Braidwood 15G C is shown in Figure 6 8. A listing of unsupported tubes i< given in Table 6 2. Ten Row 11 tubes, twenty-five Row 10 tubes, and thirty-seven Row 9 tubes are unsupported. For the tubes at R10C92, R11C23 and R11C96, AVB sensitivity analyses have been performed to ensure that conservative flow peaking factors are selected. The highest significant flow peaking factors (see Sections 8 and 9) are found at tube locations R8C40, R8C43, R8C47, R9C9, R10C14, R10C92, and R11C23. The stress ratios for these and all remaining tubes are acceptable.
Braidwood Unit 1 SG D The AVB map for Braidwood 1 SG D is shown in Figure 6-9. A listing of unsupported tubes is given in Table 6-2. All Row 12, Row 11, and Row 10 tubes are supported.
The highest significant flow peaking factors (see Sections 8 and 9) are found at tube locations R8C23 and R9C9. The stress ratios for these and all remaining tubes are acceptable.
D051s 10-pt3"12590 6-5
_ _ . _ . _ . . . _ _, _ , _ _ . _ . _ _ .--.._ _ ~ ~ - - -
- - . _ _ - - . - . - _ . - - . - . . . - _ - . . _ - - - .- =.- . - . .
Table 61 Byron Unit 1 Summary of Unsupported Tubes SG A Row 12 No unsupported tubes Row 11 No unsupported tube Row 10 Columns 2 6 and 109113 )
Row 9 Columns 2-12,16 20,96 99, and 104113 Row 8 - Columns 2 27,31-85, and 87113 Row 7 All Tubes SG B Row 12 No unsupported tubes Row 11 Columns 610 Row 10 Columns 5-11 and 104-113 Row 9 Columns 219 and 95-113 Row 8 Columns 2 24,32 89, and 91-113 Row 7 All Tubes SG C Row 12 Column 8 Row 11 Columns 2-10 Row 10 Columns 212,1517 and 105111 Row 9 Columns 2 23 ano 95111 Row 8 Columns 2 85 and 88113 Row 7 All Tubes SG D Row 12 No unsupported tubes Row 11 Column 107 Row 10 Column 99 and 104-109 Row 9 Columns 2-11,15 23,27,92, and 95-113 Row 8 Columns 2 84 and 88113 Row 7 AllTubes D0515:10-ptF112590 6-6
Table 6-2 Braidwood Unit 1 Summary of Unsupported Tubes SG A Row 12 No unsupported tubes Row 11 Columns 104108 Row 10 Columns 96 99,103113 Row 9 Columns 2 19,22,92-113 Row 8 Columns 2 24,88113 Row 7 All Tubes SG B Row 12 Columns 2 5 Row 11 Columns 2-8 Row 10 Columns 211 and 107-113 ;
Row 9 Columns 2-22,95113 Row 8 Columns 2 25,68,81-83,90113 Row 7 All Tubes SG C Row 12 No unsuppoded tubes Row 11 Columns 16 24,96 Row 10 Columns 2,3,14 25,91 101 Row 9 Columns 210,14 27,89101,113 Rows Columns 2-34,40113 Row 7 All Tubes SG D Row 12 No unsupported tubes Row 11 No unsupported tubes Row 10 No unsupported tubes Row 9 Columns 9,95110,113 Row 8 Columns 2 24,27,31-79,87, and 92113 Row 7 All Tubes 00515.10-pt3/112590 67
, _ , . ~ . ~ _
-~. . . , . . , . ., . . . ,. .m.. m.m._ . . _ _ _ . _ - - .- = _ . . -
~
..._______v..~.,
l
!?_
- 4 *t*,-- o* e W = oes.e1 e - e as m st= it cra. ms
! i j %
e t y, p i eg y, j ;
o -e--
{ [ rua - i. . .
JMee ====. i M p
j i
I l l i 8Ff t,1CDe
- fit (SS i l / f Q'.'se O$ =to- ,e .eawe ,
N iM 4D.
g I mee au acts,1tpe e F73 (33 1,
f I
- e. ifp# t'_t t* Deert rtirt le toer 1FW6 1W
.Q*3 "[378,3 M #4 ' MyTarf1(pe . Up (no Q g. ..
d W W2.i _w y'2
.r. .
O ef Oe e C'
TWE 1If "h)k*, M .
as i:
.E an '
,n .s.
e ..a
_2 -
m
- o. _ . wet . . a. . ,- o.s.a . - e no = == is ca mi-4 a D.V -mim>re .-
se .
me.
- w. iw l '
er:2 ur mz i
e tTot* oeurt o - ..
I I
k e me i.I IM L
.M er m c.,. m can I i
-: ; essa vere wr, o - ,,
m i. .
94 tu
=,,De .rio . or arm ,
_ . g 4 a fi , 4_ = a a..
.... . . . . i
.r- . .ssa. o catsu .
il
\
i _ ,,,;'t'en , -m -
l l w a t i u
i \ Mr .-
~~
_ e. . -
L Figure 61 AVB insertion Depth Confirmation 1.
l 00220-ot).10072090 6-8
' i j ,
j Ilyron Unit 1 Steam Generator A '
1 i FH i n t @ @ NLA2
- T
! T @ @@@
j
@ @ @n;u@ @ @ @
- n. n3 i.2 in no i.
iu su ms me .o i.nr i. :osis.iupuiaii-, w w
- u- u u ta ju **
@ @ w@e @w@ @ @@
,s vi w =v .u oasu
@n @@ @ @ @ @
.v .c u w a ni j
i *H u :s to i i 3
y T@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@
c __ 1. n u n n n x. e s , , .s 3 i .2 ei
.o 3, s. n 4 33 9 33 32 si so e o u . .s ,
.:i .i n u. ., .
., n n n n ,
i 32 i i, li i , is i 4 2 7 @ l
- T @@@ @@@@@@@@@@
O @@@@@
- 2 @@
rm
, n @i. u n u n@ si v. n @u n @ u@
n n 3 u n@ n n@ @ @ n@ @ @ @ @ @ @ @ @@ @@
3, i, n >> is i. ii i: tii i,. .. . , , s . e_ i is is i se. e. e. n, o se se n er , e w e ,,, w . ,sif,,n1 i 3 w ,, ,. . ;
$ Naas Tw I
< t figure 6-2 Byron Unit 1 Steam Generator A - AVB Po9tions l i I !
i
ggE- g T- r-ee F- ei ese r?
ee l[ ei See 3[
ee l ST esezi- i ee s e ;
I sese i ee l- e i l Osee 7i l ee ;
- e i i ese@ ~ ;
es i e i
! @@ee !i s ii ei i Seee T S IT e i l*
OOOSII ; e e si ei essii j OO FT Se i Ge@ ET =
e ii e i- eee ii E
)
1 G TT O E- O@e II m i i @@ ii k
s il- see ?? B i es si e i- e@e TT 5 i l :: ee TT -
1EN se i -
see ET E S es i -'
WN ee ?T e@e i? $
.=5 ese i 1EN ee Ei GO TI j
. E st see II N -
ee?[ eeEI e5 eee , = $, em s eeni -
20 eee r? ee i -
ee st ;
&E
~ See i? i es i 09 51 3 5 es iT I Ge i-G 1
e il ! ,
es i? Ge ?- a S Ei :
ee fi GO T e fi m
@@e fT es i O BE T ess ii -
Se i -
Ei : 1 ese 11 es i
$1 c eee f i Ge T 11 Seee ll Oe s e fi esse ! i ee A~
~
el_
Gese ll ee ! ei ee n Seee i l el eee ll se 5_- OE _
OOe B i Oe s el _
4 ese 9 3 Ge c el _
ee ?
T- e i_g 1 7 --
e l : l e l .1 - 1,- T v
cl=lel.l-l- f Ol: 121*l J-I fj g
.u .<,, u -n - m 4
_ _ - _ _ _ - _ _ _ _ _ _ . _ _ _ _ _ _ -_ , . . -..--_.-,-..__-,_~......,%...m.,
. - - - ---.c___.w.,,-__--..,.,..-..~,,,,_,.,,,r, yc.
l r
e E-g 7r :-
eF er ~~eeOOe T' RGE_ O s e 5[
eOli e@ ii ee Ti -~OOOOe2
~~Oe@@e1T
-~tO@@@@ 9; G esii --Seese +;
e l_
es ii -~OOOOeEl5 es @@ ii ~~@OOOOe -
~
s il Ge ii -~OOOOe il;
-e 9 8 z
es ii ~~GeOOO 11 eO Ti ,~-OOOe = ;
j e [5 ee li --~@@@el'[ j es ;
ee n : --~OOe : : i e ei
- es ii ,~-Gee -i $
es ? se i
-~-Oee ?? O se s ee r ~~-@@@$
y se (( Oe ]
~
~~-OOsli B
3 u se a se x_: <~~-GOe : :
i
.53
=E @@e il . ee 'Ei ~
3 D$ ese r : ' --7ee E ,~-Gee -GOel[: n E ese TT '
es ii ~
OOeEi
!$ 698 [1 eeG L '-
( se ] <~OOO 55 3
ei b5 [ '
~~ Oehl
- g ese ! :
e 2
~~Geil 1
=>
G es ii '
ei
<~
Oe n : 8 ee Fi ~ u~e..,
ei
~~GeEl }
eee iii 'MN s i
~~GeSi N ~u l~
9 ees is " e i 931 : ;
ese ?J ~ RE e r - - -
es : ;
see f i MM e r '
ef1 I
eee II ,~
' j ee i eez 008 : 1' su se E ee T OE-see s- ' ~
G:
e S e t'.
e@ E - @L_
e @ e iig GE ei 6,e I Di
=_ p _ ei _
esp _' Eifw 9eG1. GE M{Ese e
~. L. . '
p=isi>i-i-
- )}
=i=iei-i i-
]- =i= lei-i-i- 1
-. c j
I l
~
~
ese i eF
~
ei @@77 eT
~
ee eT
~ I
- OF eT Ge T G7
~
ee c~1 ei T e (( ei OGe
~
i el _
el: ese _ l i l G il e2t @@ ell I
..- es : e see~s l esi .. ~ ~ ~gei see E
es i? t~ ~ ~
-esi eeE eF
~
ee ?E ei 6 eT
~
ee ei %
e TT e?
e@e '
es s i ei ees-i e c @@ [I e? Gee 51 s
~
L ee
- eli eee :: 4
.e$ ese (( -
eeli eeell 5
(
c: w i
see s : .-.. e : ese n : 3 9 ees si sii eeEi e E$ 9 ese i? 'ei ee Ei : ?
$E s' see 51
~
~B 95[ eee5 f ese L1 e2 ee4_ y es s e:
G : -
. Gehi o es i? ei : eeni 8 l
ese TT I eeee fi ,
ei ei l
Gefi eni :
)e J~ e s e e li ; sei '
- - ~
eti D~ e s e e 8 ; i 4 er : eli B esse [I ;
1~
' eil :- eli F ees s i er eg1 esee ?? '
ei
- eE _
i ese ll ; e5 ei _
ese "
- er ei _
ek e I
+ = m,.,_,-
g .
4,=,e,,_,-
y c
w=m.,_,-n.
i i
Braidwood Unit 1 Steam Generator A I u
f _b _m_m_
y uz 2x2x2m2m2m ,
j X2X2X27TiJ12)
. ea 1 @@@@@@@@@@@@@@@@@@@
7
@@@@@@@@@@@@@@@@@@@@@@@ I n.a n4 00000000000000000000000000 ii) i:2 in no io9 nos io7 ios nos ios io) io2 io: ion w 9e vs w w 9e 93 92 vi w ev sa sit sa 00000000 ss 34 33 s2 : si so r, 7s is r_. .= .u .o .u .u .= .o .= . .o ., ., . ., j
.. .. ,. . .. ..l,,, ,,
. - t
.- u u u u j
~
Ti liiiiiiiiiiiiii-m m m:m g g er l
zuza:m:m:m:m m l 2N2X2X2E2N2X2X ,
i i
l Cmounum w
2 00000000000000000000000000000000000000 76 13 x 73 72 71 70 69 68 67 66 65 64 63 62 6: f4 59 58 'r7 56 55 ,se 53 52 Si 50 del en ~ 47 4a es 44 41 er at ' eri ya I j '
i !
t Til t
\ Ti -
, io [
2 T l i OOOOOOOOOO i mini OOOOO@OOOOOOOOOOOOOOOO cs r
T0000000000000000000000000000000000000 is 37 i6 is 34 33 32 is in 29 2: 27 26 23 i 24 23 ' 221 2: 2o t9 in 17 i+1is is si 2 is so *. = 7 a s e t 2 t
~ ~ '
$ Plugged i obe @ rJoisy data, inconclusive data t
Figure 6-6 Braidwood Urut 15 team Generator A - AVB Poutions l _ _ _ _ - _ _ _ _ _ . _ _ - - _ _ _ - _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
j O T- - 07- -
!~
OF OT -
eOOeO !I 0:
OT - eseeO !Tt
- OT OGOOOO ; '
OF OF
- OT - e@eOTI!
OT OOOO !I OOeO 1; l
s :
O T- seO77 i
i O li O b- OGO T 3 0 5[ O I[ EGO!_-
Oi Of $
_ gg @ 01 g OL _ ;g >O y__ @O5_ .
0 5
- 0 ;
@O'
?
OO F '
I OT OOOTE 2
-= -
GO T l
'~
' Oi @@O Ei E j
jj GO [ ;g 0 3[ OGO l[
pg OO E _ 03 001i e
?u OO E _ O -
GO i g 55 50 eO S
- 07 GO n 3
GO T Oi OO E-a;' eO 8[
4 O2 e0 5_- 1, hu 80 <
03 O
~E <
eO F 03 O
-~
m 8
GO F
' 03 OI 4 eO F 03 OEi 5 l OO fi 03 Of[ i SO li 03 O Si ;
SO 8 : - 03 O IL 3 GeO E OT 05 f eeO 1
_ OE -
OE _
e@eO 1 _ OE O L_
eeeO L
eseO i _ O E- OE _
eeeO ? Oa Ot _
eeeO 1 _
i __
Oi _ OE _
=i=islei-i-
- a t
=l=lel.l-l-
- v 1 I
=l=lslal-dc c c l
P GO E[
OO Ri Ol[ -[
@O 00 3 _ 000: _
- Os GOO -
i OO I O T- OOOTI OO (( OO E OOOll SO L _
O1 OOO:L SO L _ O2 OO:L SO i _ SO i
_ 600 L - ;
@O li 00 3 00 11 : _ 0011 8 02 _ 0011 3
@O li O! _ 005 2
@O il 00 3 OOE:
@O : :
_ _ s 90 2 OO : 4
@@O [l 000011; i
, 00 y
@@O 5 i i n OO ]a[ GOOO '
e OGCO 51 ; OO ] GOOO ll ; i u 9990 1 i @O n OOOOOEI E
=k @@@O 3 3 0 7? OOOOO : : o
- $ ii @@eO fi '
60 37 OOOOOII 5 b @@@O TI "S O li ;; OOOOO II 3 35 ke 0000 [] 60 77 ; 9000051 :
OGOO
- i OO i OOOO n i E 3E OOOO EI GO 3 6000 51 ; ;
- 2 E 0 OOO8 -
3 00 3 OOOO : : 8 OOO fi i 60 3 OOOli ;
- GOOli i uq~ fi t u~ 0 3 I lg>@OO O0 O [I 1 ~ ~
% !GO @0 O]
SOO31 j OOE.:. 8 OO l N eO I -
O0 5 1 g
005 "
eO i _ _
eO i _ eOE _
@O 5 005 _
SO il
@O n O*
OO E : eO x 01_
00 41 GO E _ 05 _
@O E .
01_
ui= lei.l-i-j
. e.
u =l=le1.l-I- I
.n 3c e cl= le 1 1-1
.- .l
= c l
. . . _ . . , , . . . , . . . _ . . . _ . . . _ . _ . . . - - , , . , _ _ . _ . . - . _ . , , . , _ _ _ . . . . . _ _ . _ . - _ _ , , , ._,. ......, ~ _- . .. . . __
i GO E[
M~~~~~~~~ SOLGOi sol _[ I Oi
_ SARS60." _
\
@Oi @O31 Oi GOf _
Oli
' 01 _ GO ; 0:1 01 _
SO2 _ GOli OL _
SO5 _ @O:1 OL
,O L_
801 _ SO ; ; -
Oi SO2 _ SOLE; +
Oi SO3 _ @ Oil 8 Oi 901 _ 002_ 3 i Oi GOh _ SOE2 2 01
_ g @Of 031 $
g"@Of
_ ; O li 5 O El _
O li
- ~ g OOf _ 0011 e ,
-c O iE
- <~
eO3 _ OOiE S
=g OO E1 u~
OOI _ OOEI ?
$W OO EE gg 021 0021 E
'~
- gh b OGO * ; 03x ;
- 6 Oil y gy -> O 9031 35 m eeO E1 lggl Oli 05" ;'- ; 3 i: 0 900 8 <~ u~
eO s O -
1E OOO E[ l 00][ @O55; j C0 eO0 4 -
' 00 Oni
- 05 <~
eeO lii -
N GOT" OEE y
8
' ) Wgl@eO l ll OOi gg Oj~ _
- ggl@eO 1 _ ;g 903 gg O. @
ggleeO !
'~ >
GO s ;g~ '
OfE =
gg OGO ; 003-j! u~4 g O SE 3 lGGO8 i
-9 gg
+
'l God' gg ole: e
< ' ~
- GOO E T g
~ '
OOE 8, ll j!
l GOO i _
) d@eO(( O OOE _
C
- - SO li ; ; u~'
eOi gg @OE GO i l GO l i
- [ OE _ gg OOi _
OE _ ;g @OE _
90 11 ; OE OOa _
GO i _
- )Os_ SOE '
i._
TC?- OOE _
el=isi.i-i-j = l= le 1 1 I
- ca
_ e f = l = 12
- l + 1- =l ~ f -! .
l 7.0 THERMAL AND HYDRAULIC ANALYSIS This section presents the results of a thermal and hydraulic analysis of the flow field on the secondary side of the steam generator using the 3D ATHOS computer code, Reference (7-1). The major results of the analysis are the water / steam velocity components, density, void fraction, and the primary and secondary fluid and tube wall temperatures. The distributions of the tube gap velocity and density along a given tube were obtained by reducing the ATHOS results.
In the following subsection, the onerating condition data for Byron Unit 1 is presented. A description of the ATHOS modelis provided along with a sample of the output parameter distributions computed for Byron 1. Supplementary calculations were also completed to examine the effect of operation with increased tube plugging. For these calculations a one-dimensional stability ratio calculation technique is used to define adjustment factors which are applied to the reference 3D stability ratios.
7.1 Steam Generator Operating Conditions Recent full power steam generator operating condition data were supplied by Commonwealth Edison Company for Byron Unit 1 and Braidwood Unit 1. The d sta for Byron Unit 1 are applicable to operation with reduced primary fluid temperatures, whereas the data for Braidwood Unit 1 are for operation at " normal" primary fluid temperatures. From previous similar analyses,it has been clearly estabhshed that operation with lower pnmary temperatures and steam pressures results in more limitina operating conditions with regard to tube istigue compared to earlier operation with " normal" temperatures and pressures. Therefore, the Byron Unit 1 operating data is conservati vely used to represent Braidwood Unit 1.
The full power operating data supplied by CECO includes thermal power, steam flow rate, feedwater inlet temperatures, steam pressure, water level, and primary inlet / outlet temperatures for each steam generator. With the above data, calculations were completed using the Westinghouse SG performance computer code, GEND, to verify the plant data and to establish a complete list of operating conditions required for the ATHOS analysis. The GEND code determines the primary side temperatures and steam flow rate required to obtain the specified steam pressure at the given power rating. Besides confirming these parameters, the code calculates the circulation ratio which is of primary importance to the stability rati 3 analysis since it, together with the steam flow, establishes the total bundle flow : <te and average loading on the tubes. It also provides an overallindication of the vmds within the tube bundle since the bundle edst quality is inversely proportional to the circ ratio (X,xit = 1/ circ ratio).
With respect to their effect on tube stability ratios, three of the operating parameters are of primaryimportance: power level / steam flow, steam pressures, and the circulation ratio. (Primary side temperatures have only a very minor t
D051s 1Det3112s90 7-l 1
_ __ _ _ . . _ ._. m ___ _._._ _ ___ _.___ ._ _
influence on stability ratios). As mentioned previously, the steam flow rate and circulation ratio influence the total bundle flow rate and tube-to-tube gap velocity in the U bend. The steam pressure also influences the gap velocity via the void fraction and density; however,its major impact is on the tube damping. High U-bend flow along with low steam pressure results in higher loadings on the tubes with reduced damping. Both of these factors lead to higher, more limiting stability ratios, in order to obtain worst case stability ratios which bound the values for all of the Byron Unit 1 and Braidwood Unit 1 steam generators, the operating parameters which lead to the most limiting stability ratios are used. Based on analysis of the operating data for the individual Byron Unit 1 steam generators,it has been concluded that the operating data for SG C produces the most limiting stability ratios. Therefore, the operating conditions for Byron Unit 1 steam generator C are used for the ATHOS analysis. Table 71 summarizes the full power operating data supplied by CECO for SG C. The circulation ratio and bundle flow rate, which are parameters calculated by the GEND computer code based on the SG C operating data, are also summarized in Table 71.
It is noted that average values based on all SGs and parameters specific to other SGs (tuch as circulation ratio, water level and steam pressure) may vary from the values shown in Table 71 The values used for the ATHOS analysis are judged to be most limiting and are expected to produce worst case stability ratios and tube fatigue usages.
7.2 ATHOS Analysis Model The calculation of relatiu stability ratios involves comparing the stability ratio calculated for one or more tubes in a given plant to the ratio calculated for the ruptured Row 9 Column 51 tube in the North Anna Serls 51 steam generator. It makes use of ATHOS computed flow profiles for both tube bundles. Since the presence of AVBsin the U-bend region of a tube bundle could influence the overall flow field and/or the local flow parameters for a particular tube of interest, some discussion of the treatment of AVBs is necessary before presenting a description of the ATHOS model.
The ATHOS code does not include the capability to model the presence of the AVBs in the U-bend region. However, Westinghouse has mc,dified the code to include the capability to model the AVBs via flow cell boundary resistance factors. Practical lower limits of cell size in the ATHOS code, however, prevent a fine grid representation of the AVB V-bar shape which,in turn, limits the accuracy of the AVB representation. ATHOS calculations have been performed with and without AVB5 m the model. Calculations of stability ratios relative to North Anna R9C51 show that the relative stability ratios for tubes near the center of the steam generator are essentially the same for models with or without AVBs. The ATHOS AVB modeling ,
sensitivity studies with uniform insertion show some tendency for the AVB resistance effects to lower tube gap velocities near the central regions and to increase l velocities near the peripheral tubes. Howeser, the magnitude of this effect is
- l. 00st$ 10 pt3012s90 7-2
uncertain due to the limitations in ATHOS for modeling the AVBs. Further, the global flow resistance of staggered AVB insertion would be less than that from uniform insertion. Based on the sensitivity studies using ATHOS models with and without uniformly inserted AVBs, the most reliable relative stabihty ratios (for actual steam generators with non uniform AVB insertion depths) are expected using ATHOS models excluding AVB5 and effects of variable AVB insertion depths. Those AVB effects are accounted for by using flow test results of actual AVB geometries.
This approach has been utilized in the Byron Unit 1 ATHOS analysis which is described below. r The Model D4 analysis is based on a Cartesian coordinate system for the array of flow cells instead of the typical, and more widely used, cylindrical coordinate system.
With a Cartesian coordinate system, the tube array and any AVBs are arranged in a square pitched configuration which is in line with the coordinate axes. This ;
alignment provides an improved representation of the tube bundle.
The ATHOS Cartesian coordinate system model for the Model D4 steam generator consists of 25,536 flow cells having 32 divisions in the x-axis (perpendicular to the-tubelane) direction,19 divisions in the y axis (along the tubelane) direction and 42 divisions in the axial (z axis) direction. In the ATHOS analysis, the steam generator is considered to be symmetrical about the x axis of the tube bundle. The model therefore,cor'sists of one half of the hotleg and one half of the cold leg sides of the steam generator. Figures 71 and 7 2 show the plan and the elevation views of the model. These two figures show the layout of the flow cells and identify locations for some of the geometric features.
As shown in Figure 71, with the Cartesian coordinate system, the circular wrappet boundary is represented by a step-wise wall as indicated by the heavy lines. All of -
the flow cells outside the simulated wrapper boundary above the first axial slab were blocked off by specifying extremely high flow resistances on the faces of the appropriate cells. Tubelane flow slots in the tube support plates are also modeled.
Figure 7-2 shows the elevation view of the model on the vertical plane of symmetry of the steam generator. The feedwater nozzle is located at axialindices 12 = 11 and
- 12. Ten axiallayers of cells were iricluded in the U-bend near the top tube support plate (lZ = 30 to IZ = 39) to more closely model the flow conditions in the area of interest.
Figure 7-3 reproduces the plan view of the model but with the tube layout arrangement superimposed. This figure illustrates the locations of the tubes in the various flow ce; ,. The grid lines in the Cartesian model are in-line with the tube
- array, providing for all of the tubes to be within the boundary of the flow cells. The fineness of the cell mesh is evident; the largest cells contain only 36 tubes while some of the smallest cellsinclude only a single tube.
cosissowmmo 7-3 .
l l
I 7.3 ATHOS Results Thc results from the ATHOS analysis consist of the thermal-hydraulic flow parameters necessary to describe the 3 D flow field on the secondary side of the steam generator (velocity, density, and void fraction) plus the distributic,ns of the primary fluid and mean tube wall temperatores. The secondary side mixture velocity is composed of three components (Vx, Vy, and Vz) which ATHOS computes on the surfaces of the flow cell. Since the local gap velocity surrounding a tube is required in the vibration analysis, a post-processor is used which: a) interpolates among the velocity components for the cells located nearest to the tube of interest and, b) accounts for the minimum flow area between tubes to calculate the tube to tube gap velocity. The post processor performs the necessary interpolations to determine both in plane and out-of-plane gap velocities at specific intervals along the length of a tube. It also interpolates on the ATHOS cell-centered density and void fraction to determine the required local parameters along the tube length. The output of the post-processing is a data file which contains these parameter distnbutions for all the tubes in the generator and which provides a portion of the input data requimd for tube vibration analyses.
Figure 7-4 shows a vector plot of the flow pattern on the vertical plane of syrnmetry of the steam generator (the vectors are located at the center of the flow cells shown in Figure 7 2). The zig-zag flow pattern through the split flow preheater is clearly shown in the figure. On the hot ieg side, the vertical flow upward through the half-moon cut out at the center of flow distnbution Plate A is also clearly shown. The vertical velocity (V2) component entering the U-bend region is shown in Figure 7-5 (at 12 = 29). The figure shows the high 2V -component of the flowleaving the three flow s'ots on the top tube support plate (PLATE P) at the middle of the figure. The lateral velocity components, Vg = (V,2 + V y2) 5, on the same horizontal plane (12 = 29) are shown in Figure 7-6. Viawing Figu.es 7 5 and 7-6 it is seen that at the entrance to the U bend region the vertical velocity component can be about four times that of the lateral velocity resultant.
Figure 7 7 shows the plot of the void fraction contours on the vertical plane of symmetry of the steam generator on the hot leg side, void fraction develops rapidly from the lower bundle region. In the U bend region the void fraction is about 0.8-0.9 on the hot leg side, decreasing to about 0.60 at the bundle periphery on the cold leg side.
Figures 7 8,7 9, and 7-10 show a sample of the individual tube gap velocity and density distributions along three tubes at Row 10. In each figure the gap velocity and density along the length of the tube are plotted f rom the hot leg tubesheet end on the lef t of the figure to the cold leg end on the right. The mixture gap velocity and density distribution are required as part c,f the input for tube vibration analysis to determine the tube stability ratios. These data were generated by the ATHOS post-processor for each tube in the model and stored in a data file The data file was then utilized in the subsequent stability ratio calculations oosis m o w s90 7-4
)
l I
References:
+
4 71 L W. Keeton, A. K. Singhal, et al., "ATHO53; A Computer Program for '
Thermal Hydraulic Analysis of Steam Generators," Vol.1,2, and 3, EPRI NP-4604 CCM, July 1986.
1
+
e h
1
-I i
r I
t 3
s I
- l. D051510 pt3!112590 '7 3 1.
l
, , , . ., _ . _ . . . . , , , ~ - . , , - - , , , . . ~ . - . . . , . , . - - - , , - - - . . . , ~ , , , -. . , . - . . ,-. . . . -
l Table 71 Byron Unit 1 Steam Generator Operating Conditions Used For ATHOS Analysis Byron Unit 1
- l l
SG thermal power (MWT) 865 Steam flow rate 3.78 x 106 Feedwater inlet temperature (*F) 436 ;
Steam pressure (psia) _ 908 Water level (% of span) 61 Primary inlet / outlet 6C6.7/549.9 temperatures ('F)
Calculated Parameters a,c 4
m TaifsTT6 C i
oosis: to-pon 12590 7-6
. .. . . - - - . ~, - . .-. . - . . .__.- - . . _ - .
i a,c i
J Figure 7-1. Plan View of ATHOS Caretesian Model for Model D4 D0515:10 pt3/112590 7-7
a.c t
f
~ ~ ..
_]
Figure 7-2. Elevation View of ATHOS Caretesian Model for Model D4 ws15:io-pt3/112s90
a.c ;
l- i f
Figure 7-3. Plan View of ATHOS Caretesian Model for Model D4 Indicating Tube Layout
- D0515:10-pt3/112590 79
a,C
)
i l
i Figure 7-4. Flow Pattern on Vertical Plane of Symmetry oosi s.10-ptvi us90 7-10
k 3
a,c 1
L Figure 7-5.- Vertical Velocity Contours on a Horizontal Plane at the l Entrance to the U-Bend i i
l D0515:1D-pt3/112590 7-11
.. - . .. ..- .. . . ~ . . - . - . _ . - = - - _ . - . . . . - . . - . - . .. --. - _- . .
i l
a,c !
1 1
i i
l Figure 7-6. Lateral Flow Pattern on a Horizontal Plane at the Entrance to the U-Bend oosis;io.pon i2590 7-12
m !
d,C i
i 4
Figure 7-7. Void Fraction Contours on Vertical Plane of Symmetry
- 00515:10-pt3/112590 7-13
n.C
~
P '
I 1
1 l
l Figure 7-8. Tube Gap Velocity and Density Distributions for Tube Row 10/ Column 5 t
D0515:10-pti/112590 7,34
.e - . - . .
1 l
0,C
'.f .
- _J Figure 7-9. Tube Gap Velocity and Density Distributions for Tube Row 10/ Column 25 oosis.io ptariizs9o 7 15
a,c I
J 1
i Figure 7-10. Tube Gap Velocity and Density Distributions for Tube Rnw 10/ Column 45 00515:1 D-pt 3/112590 7-16
1 l
8,0 PEAKING FACTOR EVALUATION This section describes the overall peaking factor evaluation to define the test based I
peaking factors for use in the tube fatigue evaluation. The evaluation of the eddy
. current data to define the AVB configuration for North Anna-1 Tube R9C51 is described. This configuration is critical to the tube fatigue assessments as the peaking factors for all other tubes are utilized relative to the R9C51 peaking factor.
Uncertainties associated with applying the air model test results to the tube fatigue assessments are also included in this section. Included in the uncertainty evaluation are the following contributions:
e Extrapolation of air test results to two phase steam water
- Cantilever tube simulation of U-bend tubes e Test measurements and repeatability e AVB insertion depth uncertainty 8.1 North Anna-1 Configuration 8.1.1 Background The AVB configuration of the ruptured tube in North Anna, R9C51,is the reference case for the tube fatigue evaluations for other plants. In accordance with the NRC Bulletin 88-02, the acceptability of unsupported tubes in steam generators at other plants is based or tube specific analysis relative to the North Anna R9C51 tube, including the relative flow peaking factors. Thus, the support conditions of the R9C51 tube are fundamental to the analyses of other tubes. Because of the importance of the North Anna tube, the support conditions of this tube, which were originally based on "AVB Visible" interpretations of the eddy current test (ECT) data (Figure 8-1t were reevaluated using the projection technique developed since the North Anna event. The projection technique is particularly valuable for establishing AVB positions when deposits on the tubes tend to mask AVB signals such as found for the North Anna 1 tubes. The results of this evaluation are summarized below.
8.1.2 Description of the Method The basic method utilized was the projection technque in which the AVB position is determined based on measured AVB locations in larger row tubes in the same column. In this study, the projection technique was utilized in the " blind" mode, (AVBs called strictly based on the data) as well as the reverse mode (data examined on the basis of predicted AVB positions). The objective of this application was,with the greatest confidence possible, to establish the positions of the AVBs in an 8 column range arounct the R9C51 tube in North Anna 1, Steam Generator C.
oosts to cams 90 8-1
8.1.3 Data Interpretation The ECT traces for the U bends in Rows 8-12 (in one case,13) were examined for Columns 48-55. The original AVB visible calls are shown in Figure 8-1. The data were examined by an eddy current analyst experienced in reading these traces, and by a design engineer knowledgeable in the geometry of the Model 51 U-bend region.
The intent of this review was to determine if the presence or absence of AVBs as shown in Figure 8-1 could be confirmed using the AVB projection technique.
Preliminary projected AVB positions were based on geometric data provided for a few of the tubes near R9C51. The features which were sought were evidence of data
" spikes" where AVBs were predicted, offset indications (multiple spikes) where offset AVBs were predicted, single indications where single AVB intersections were predicted, etc. The data evaluation method used was a critical examination of the data, which was biased toward the presence of AVBs unless a confident call of "no AVB" could be made, and then checking the consistency of the data among the tubes in a column and against the theoretical data for the predicted AVB positions.
[
Ja,c.
. Figure 8 4 is the "AVB visible" map for columns 48 through 55, based on the critical review of the data. It should be noted that the original data interpretations and the review interpretations are consistent.
oosts.f o p:2n12590 8-2
8.1.4 Projections The[ ]a.c ECT traces were utilized for projecting the position of the AVBs according to the standard format of the projection method.
The re ults of the projections are presented in Figure 8-5, which shows a matrix of projections for tube rows 8 through 13 in columns 48 through 55. Fc. many of the tubes, more than one, and as many as three projection values are shown. Multiple projections are expected for a tube if the AVBs on either side of the tube are not at the same elevation, orif the upper and lower AVB support that tube. As many as four different projections are possible if it is assumed that the tube is supported by the upper and lower AVBs, and both upper and lower bars are staggered in elevation as shown in Figure 8-2.
The logic in arranging the projection data is based on the following two rules:
i Rule 1. The projections of the same AVB based on different tubes in the same column [ la.c.
[
]a c.
Rule 2. Two adjacent tubes in the same row [
]a,c. Consequently, the difference in the [
Ja.c.
The implication of this is that if the position (either left or right) of a projected AVB is assumed for a column, then the projections in the adjacent columns are also [
Ja c.
oosi s. io.pt2/i i2590 83
The arrangement of the AVBs as shown in Figure 8-5 satisfies the rules above and is consistent with the rupture of R9C51. The resulting AVB arrangements, based on the projection matrix of Figure 8 5 is shown in Figure 8-6.
8.1.5 Conclusions The general AVB arrangement surrounding the ruptured tube in North Anna-1, Steam Generator C, which was the basis for the analysis,is confirmed by a detailed critical review of the ECT data. Differences exist in the AVB pattern between tube columns 48 49,in which the AVBs appear to be less inserted than previously indicated.The pattern of Figure 8-6 is the best fit to the rules which were adopted for determining the position of the AVBs, as well as consistent with explanation of the tube failure.
The basis of the review was a projection technique which utilizes data from tubes one or more rows removed from the actualinserted position of the AVB to determine the position of the AVB. The intent of the review was to establish the positions of the AVBs by confirming or eliminating features of AVB alignments such as side to side offsets, etc. of the AV8s adjacent to the tubes. Overall, the conclusions regarding the positions of the AVBs around R9C51 in North Anna-1, Steam Generator C are based on consistency among all the available data.
8.2 Test Measurement Uncertainties The descriptions of the peaking factor tests and apparatus were provided in Section 5.4. All practical measures were taken to reduce uncertainties. Nevertheless, some still remain and should be properly accounted for. The important parameter measured during testing that has a significant impact on peaking factor is the air velocity. The air velocity at test section inlet was measured using a (
]a,c. Based on considerable experience with the use of such instruments,it is known that the magnitude of uncertainty is very small. A [ la,c measurement uncertainty is used in this analysis based on past experience.
8.3 Test Repeatability During the peaking factor testing of AVB configuration, each test was performed at least two times to confirm repeatability. It has been demonstrated that the tests are quite repeatable with the results often falling within 2 or 3% of one another for the repeat tests. An upper bound value of 5% was used in the current uncertainty analysis.
cos t sn o-ot2n 12s90 8-4
8.4 Cantilever ss U Tube A first order estimate can be made of the validity of modeling a U-bend tube by a cantilever tube in tests to determine the effects of AVB insertion depth on the initiation of fluidelastic vibration. The following assumptions are used:
a,c For the purposes of this estimate, the geometry of the cantilever measuring tube in the air test modelis compared with the geometry of a prototypical Row 10 tube. [
la.c, l The comparison between a U-bend tube and the model tube involve the consideration of an effective velocity associated with the flow perturbation caused by the AVBs, [
2 oosts to-pan us90 85 i
l 1
l l
(
Ja.c Using these values, the ratio of.the effective velocity for the cantilever measuring tube to that for the U-bend tube is about [ Ja.c for the case treated.
A similar evaluation can be made for a Row 10 tube that lies in the projection or shadow of an AVB thatis inserted to a depth required to support a Row 9 tube. [
Ja,c, The net result is that the ratio of the effective velocity for the cantilever tube to that for the U-bend tube is about [ Ja c.
These results indicate that, for the particular assumptions used, the cantilever tube model appears to be a reasonable representation of the U bend with respect to determining relative peaking factors for different AVB configurations. This evaluation also shows that, on the average, the magnitude of the systematic uncertainty associated with the use of cantilever tube to simulate the U-bend is about [ la.c.
8,5 Air vs Steam-Water Mixture The local peaking factors from the air tests can be applied to the steam generator steam / water conditions either as a direct factor on the mixture velocity and thus a direct factor on a stability ratio, or as a factor on the steam velocity only with associated impacts on density, void fraction and damping. This method leads to a reduction in tube damping which enhances the peaking factor compared to the direct air test value. For estimating an absolute stability ratio, this application of the peaking factoris a best estimate approach. However, for the evaluation of tubes relative to stability ratio criteria, it is more conservative to minimize the peaking factor for the North Anna Unit 1 tube R9C51 through direct application of the air D0515- 10-pt2/112590 8-6
test peaking factor. This conservative approach is therefore used for evaluating tube acceptability.
Under uniform AVB insertion (or aligned AVB insertion), tnere are no local open channels for flow to escape preferentially. Therefore, air flow is approximately the same as steam / water flow relative to velocity perturbations. Under non-uniform AVB insertion the steam / water flow may differ from air, as the steem and water may separate from each other when an obstruction, such as an AVB, appears downstream. The water would continue along the same channel while steam readily seeks a low resistance passage and thus turns into adjarent open channels.
Two phase tests indicate a tencency for steam to preferentialiy follow the low pressure drop path compared to the water phase.
Based on the above discussion, the F i are considered to more appropriately apply to the steam phase. Thus,it follows that mixture mass velocity for the tube subject to flow perturbation can be written as follows:
a,c where Dgis the vapor density, Df the water density, Fa the velocity peaking factor determined from air tests,jg* the nominal superficial vapor velocity, and jf
- the superficial water velocity. Steam quality can then be determined as follows:
a,c The Lellouche-Zotctar correlation (algebraic slip model), as used in the ATHOS code, is applied to determine void fraction. Subsequently, mixture density. velocity and damping coefficients for the tube which is not supported and subject to flow perturbation is evaluated. Therefore, similar to the air velocity peaking factor, local scaling factors of mixture density and velocity and damping coefficient can be readily determined. Finally, a local stabi!ity peaking factor for fluideldstic vibration can be calculated as follows:
aC where Fs is the stability peaking factor, Fd the density scaling factor, Fv the velocity scaling factor, and Fep the damping coefficient scaling factor. If we use the air velocity peaking factor without translating to steam / water conditions, then a,e oms mc2m2sw 8-7
As shown in Table 8-1 stability peaking f actors for the steam / water mixture are slightly higher than air vclocity peaking factors. Tha difference between the steam / water and air peaking factors increases as the air peaking factor increases For application to tube fatigue evaluations, the ratio cf the peaking factor for a specific tube to that for North Anna R9C51 is the quantity of interest. Larger values for this ratio are conservative for the tube fatigue assessment. The North Anna R9C51 peaking factor is one of the highest peaking factori As discussed in Section 8-7, a peaking factor of nearly [ ]"is determined for the R9C51 tube. The dif ferences between [
]" Typical values are shown in Table 8-2. These results show that the direct application of the air test data yields the higher relative peaking factor compared to R9C51. To obtain conservatism in the peaking factor evaluation,[
la c Comparing the values in the first and last columns of Table 8-1,it may be noted that the stability peaking fac+.or for steam water is [ ]" higher than the air velocity peaking factor. On the average, the uncertainty associated with the conservative use of air velocity peaking factor is [ }"
The conclusion that the peaking f actor for steam water flow would be higher due to the dependency of damping ratio on void fraction was supported by an alternate study. In this study, a section of steam generator tubes were simulated usmg the ATHOS code under prototypic flow conditions. The objective of this study was to examine the magnitude of the cnanges in void fraction and thus stability ratio as a consequence of non-uniform AVB insertion patterns. The current version of ATHOS has modeling limitations that prevent accurate modeling of local geometry effects, in addition,it is believed that an analysis using two-fluid modeling procedure is mandatory to a ca.lculation of the peaking factors for a steam generator to account for the preferential steam flow along the low resistance path. Consequently, the intent of this analysis is only to help bound the uncertainty on void fraction effects from extrapolating the air tests to steam-water.
- First the analysis was conducted with uniformly inserted AVBs in the ATHOS model.
The ATHOS results were processed by the FLOVIB code to determine stability ratios for the specific tubes of interest. The calculation was repeated using a non-uniform AVB insertion pattern in the model. The results show that the void fraction distribution changes as a result of flow perturbation. Further, the impact on stability ratio resulting from the changes in void fraction profiles was about [ ]" This alternate calculation providesindependent corroboration of the prior discussion regarding the stability peaking factors under steam-water conditions vs. in air.
costs scems* 8-8 '
___ - __- __ __ -a
i 8.6 AVB Insertion Depth Uncertainty The most significant uncertainty for the low peaking configurations is not in the test results, but in the determination of actual AVB insertion patterns adjacent to specific tubes. The methodology used for obtaining the AVB insertion patterns from eddy current data can ascertain the AVB location only to within approximately [
la.c. The effect on peaking factor resulting from this uncertainty is addressed using test results of AVB configurations that varied from one another by up to [ ]**
Based on maps of AVB insertion depth of various plants, several configurations have been tested for determining fluidelastic instability flow rate by an air cantilever model. Stability peaking factors were then determined from the ratio of critical flow rate for a uniform AVB insertion configuration to a specific configuration. Figure 8-7 summarizes the AVB configurations tested.
Position of AVB insertion depth is determined from Eddy Current Test (ECT) data.
Positioning of AVB from ECT data reading is subject to uncertainty; its accuracy is probably about [ la.c A change of an AVB insertion depth in a given configuration leads to a different configuration, and thus a different peaking factor.
A review of the tested AVB type has been made and results summarized in Table 8-3.
As can be seen, a decrease in depth of an appropriate AVB tends to decrease the peaking factor, for instance, a [
Ja c. Such a trend can be explained; a decrease in a specific AVB depth will open up more channels for incoming fluid to distribute and thus less flow perturbation. However, this applies only to those changes without inducing the reinforcement of flow perturbation from upstream to downstream.
On the average, the uncertainty in peaking factor resulting from small variations in AVB insertion (of the order of 1/2 tube pitch)is found to be [ ]**.
8.7 Overall Peaking Factor with Uncertainty As discussed in the previoe< subsections, there are several aspects to be considered in dpplying the laboratory test data to steam generator conditions. These considerations were reviewed one at a time in those subsections This section will-integrate the pieces into one set of stability peaking factors.
Looking forward to how these peaking factors are used in the analysis (Section 9),
the relative stability ratio calculated for a given tube without the consideration of flow peaking is corrected using the ratio of the peaking factor of the specific tube to that of the North Anna R9C51 tube (Configuration la).
It is to be noted that the test results would be applied as ratios of a spec fic tube peaking factor to the R9C51 peaking factor. This will reduce the influence of some uncertainties since the systematic uncertainties would affect both the numerator 00515:10-otbt 12590 8-9
..- . - - - -- - = , . -- -
and the denominator in the ratio of peaking factors. The major difference will be in those configurations whose peaking factors are significantly lower than that of R9C51- The approach employed here is intended to provide that conservative
- peaking factors are employed for such apparently low peaking configurations.
The uniform AVB configuration (2a)is selected as a reference configuration, and the peaking factors of all configurations tested are recomputed on the basis of this reference. As discussed below, some of the test uncertainties are applied to the rcierence case to account for its significantly low peaking relative to the R9C51 configuration.
The uncertainties in the test results and their extrapolation are those due to test measurements, test repeatability, cantilever tubes in the test vs U-tubes in the steam generator, and air tests vs steam-water mixture. These were discussed in more detail in the previous subsections. The magnitude of these uncertainties are listed in Table 8-4.
Of these uncertainties, those due to measurement and repeatability of tests are random errors and can occurin any test. Therefore, these are treated together. The total random uncertainties are calculated by [
Ja,c The RSS value of these is [ la.c. Since these can occur in any test, these are to be applied to all tests. One way of doing this is to apply it to the R9C51 value, that being in the denominator of the final peaking factor ratio. Thus the peaking factor for configuration la (R9C51)is reduced by this amount to yield a value of [ Ja,c instead of the [ ]a.c appearing in Table 5-2.
The next three uncertainties in Table 8-4 are systematic uncertainties. It could be argued that these appear in the peaking factors of both the specific tube under consideration and the R9C51 tube and are therefore counter balanced. However, the relative magnitude of these may be different, particularly for configurations with much lower peaking than R9C51. Therefore it was judged that the [
la.c. Similarly, as noted above, the effect on peaking factor due to the uncertainty in the field AVB configuration is also included in this reference case. Thus,[
]** The peaking factor of the reference configuration 2a (Table 8 5)is raised by this amount to a value of [ ]**
The change in peaking factors of configurations la and 2a resulting from the application of uncertainties as described above are shown in Column 3 of Table 8-5.
The peaking factors of all configurations are recomputed on the basis of this reference configuration (2a). These values are displayed in Column 4 of Table 8-5.
Some of the uncertainties were applied to the reference configuration (2a)in order to apply them to all low peaking configurations conservatively. Thus, no configuration should have a lower peaking factor than this reference configuration.
D051s tD pt2M t2590 8-10
Therefore, when a peaking factor value less than [ la.c is calculated for any configuration,(in Column 4 of Table 8-5),it should be altered to [ }". Further,
. for some of the configurations that are conceptually similar, the more limiting (higher) value is used. For example, a peaking factor of [ ]"is used for configurations 5a and Sb based on their similarity to configuration Sc.
The final stability ratio peaking factors calculated on this basis (with configuration 2a as the reference) are shown in Table 8-6.
The overall conclusions from the peaking factor assessment are:
- 1. As noted in Table 8 4, five elements have been included in the uncertainty evaluation for the peaking factors. The uncertainty estimates were developed from both test and analysis results as described in Sections 8.2 to 8.6. The largest single uncertainty of ( la,c is attributable to uncertamties of up to
[- Ja,c on determination of AVB insertion depths ' rom field eddy current data. Thb relatively large uncertainty is applicable only te low peaking conditions w,ure the AVB uncertainties can contribute to small peaking factors. The definition of "no flow peakmg" was increased to encompass the small peaking effects from AVB insertion uncertainties. For the AVB patterns leading to significant peaking factors, Aves were positioned within uncertainties to maximize the peaking factor. Fct these configurations, variations of AVB insertion within these uncertainties are expected to reduce the peaking factor compared to the final values of Table 8-6 and Figure 8-7.
- 2. Including uncertainties directed toward conservatively decreasina v ' king factor for the North Anna tube R9C51, the final R9C51 peaking i
[ la,c relative to a no flow peaking condition such as with unin insertion depths.
8.8 Peaking Factors for Specific Tubes Peaking factors for Byron Unit 1 and Braidwood Unit 1 were determined using the methodology described above. Table 8-7 summarizes the results of peaking factors.
The AVB positions on each insertion pattern of Figure 8 7 should be carefully noted.
Where the AVBs are shown at the top of the test tube,(configurations it,4n for example), the AVBs at least partially block the flow past the test tube and low flow peaking factors are typically obtained. Where the AVBs are shown at the centerline of the tube row above the test tube, the flow past the test tube is not restricted and significant flow peaking can be obtained.
In applying the nethodology to Byron Unit 1 and Braidwood Unit 1, maps of the AVB insertion depths shown in Figures 6-2 through 6-9 were first reviewed. The second step was to identify those unique and meaningful configuration of AVB insertion depths in locality. In doing so, maximum allowable flow peaking factors were also reviewed column by column for rows 8 through 12. Based on the Byron 1 DOS 15-10 pt2/112590 8-11
. -. . . - - ~ - - - - - - _- - __ - .-= - .- - - -- . _ - - - . -
and Braidwood Unit 1 tube vibration analysis, flow peaking factors greater than
[ ]a.c for row 8 and [ la.c for row 9 tubes would be required for tube fatigue to be a concern. l l
After conservative estimates of peaking factors were made for specific tubes, those having peaking factors near the maximum allowable value were identified and AVB insertion depth accuracy was reviewed for the tube involved and its neighboring tubes. If needed, stability velocity for the tube with identified configuration of the AV8 insertion depth was determined using the Westinghouse R&D Cantilever, Air Model. The peaking factor was then calculated using the stability velocity.
Determinations of peaking factors for identified tubes shown in Tables 8-7 and 8-8 are described in detail below. Tables 8-7 and 8-8 are broken into small tables for ease in following the description, 8.8.1 Byron Unit 1 Steam Generator A L
The following table gives the peaking factors for tubes with unique configurations of AVB insertion depths.
l Type of AVB Peaking l Steam Insertion Factor Row No. Column No.
Generator Depth l aC I
A 8 88 L 75 27 9 19 10 110,111 4
Type [ ja,c was a good selection for R8C75 and R8C27 tubes and a peaking factor of
[ Ja,c resulted for them. Tube R9C19 belonged to type [ la,c and thus a peaking factor of[ Ja.c was obtained. Type [ ]a.: was a good approximation for R10C110 and R10C111 tubes and a peaking factor of [ Ja,c was obtained. Type [ ]a c was a natural choice for tube R10C4 and a peaking factor of { ja,e was obtained.
l 0051s.10 pt2/112590 8-12
8.8.2 Byron Unit 1 Steam Generator B The following table gives the peaking factors for tubes with unique configurations of AVB insertion depths. For R8C88 tube, type [ la.c was a conservative selection and a peaking factor of [ la.c was thus obtained. Type [ la.c was a good approximation for R10C104, R11C9, and R11C6 tubes and a peaking factor of
[ la.c resulted.
Type of AVB Peaking Steam Row No. Column No. Insertion Factor Generator Depth a.C B 8 88 10 104 11 9 6
8.8.3 By. :n Unit 1 Steam Generator C T.1e folic, wing table I;sts tubes with unique AVB configurations together with the;r x peaking factors.
Type of AVB Peaking Steam Insertion Factor Row No. Column No.
Generator Depth B,C C 8 31 27 11 4 12 8 a
_ ~
For F. SC31 and R8C27 tubes, type [ la,c was a good v: conservative selection and a peaking factor of [ la.c was obtained. For R1'C4 tube, type j la.c was selected and a peaking factor of [ la.c resulted. R12C8 tube belceged tc; type [ la.c and a peaking factor of [ la.c resulted.
00515.10-pt2/112590 8-13
l l
l 8.8.4 Byron Unit 1 Steam Generator D The following table lists tubes having unique configurations of insertion depths of l- AV8s together with their peaking factors.
Type of AVB Peaking Steam Insertion Factor Generator Row Nc. Column No.
Depth l
a,c D 8 88 27 9 92 l
23 11 107 For R8C88 tube, type [ la,( was a good choice and a peaking factor of [ ~!a.c resulted. Type [ la,c was conservative for R8C27 tube and a peaking factor of
, I la.c was obtained Type [ la,c was good for R9C92 and R9C64 tubes, and a i peaking factor of [ la c resulted. For R9C23 tube, type [ Ja,c was conservative and a peaking factor of [ la.c was obtained. For R11C107 tube, type [ ja.c was a conservative selection and a peaking factor of [ ]a.c resulted.
8.8.5 Braidwood Unit 1 Steam Generator A The following table gives peaking factors for tubes having unique configurations of AVB insertion depths. .
Type of AVB Peaking Steam Insertion Factor venerator Row No. Column No. Depth j a,C ,
A 9 92 l
10 98 For tube R9C92, type [ la,c was a good choice and a peaking factor of [ la.c resulted . Tube R10C98 belonged to type [ Ja.c and a peaking factor of [ la,c was obtained.
oosisoo-oun usso 8-14
l l
8.8.6 Braidwood Unit 1 Steam Generator B The following table gives AVB types and peaking f actors.
Type of AVB Peaking Steam Row No. Column No. Insertion Factor Generator Depth a,c -
B 9 18,17 11 8 12 5 Type [ ']a.c was good for both R9C18 and R9C17 tubes and a peaking factor of
( Ja.c resulted. For R11C8 tube, type [ la.( was considered and a peaking factor of
{ la,c was obiained. Type [ la.c was a conservative selection for R12C5 tube and a peaking factor of ( . la.c resulted.
8.8.7 Braidwood Unit 1 Steam Generator C The following table lists unique AVB configurations and peaking factors.
, Type of AVB Peaking Steam Row Nc. Column No. Insertion Fa tor Generator Depth ,
a,C C 8 47,43,40 9 9 10 92 10 14 ,
11 23 For R8C47, R8C43 and R8C40_ tubes, type [ Ja,c was a good selection and a peaking factor of ( Ja,c resulted. Type [ Ja.c was appropriate for R9C9 tube and a peaking factor of { ]a.c was obtained. Type { Ja,c was conservative for R10C92 tube, and a peaking factor of ( ' la.c resulted. Tube R10C14 belonged to type { ]a,c and a peaking factor of [ la.c was obtained. For R11C23 tube, type ( ja.c was applicable and a peaking factor of [ la.c occurred.
oosts;io-o w i ns9a 8-15
8.8.8 Braidwood Unit 1 Steam Generato, a The following table shows tubes with unique AVB configurations together witc
> their peaking f actors.
Type of AVR Peaking Stearn Insortion Factor Genera tor Row No. Column No. Cepth a,e .
- p D 8 23 9 9
'Both R8C23 and RSC9 tubes beloc.ged to type [ la.c cnd a peaking factor of
{ }a.c resulted.
i D0515:10-pt2/112590 8-16
- - - _ _ _ _ _ - _ _ _ _ _ _ _ _ - - _ _ _ _ _ _ _ . _ _ - _ _ - - - - _ _ _ _ - _ _ _ _ _ _ _ _ _ _ _ _ _ _ - - _ _____-_-__-_-__-_____-____-_L
l l
I Table 8-1 i Stability Peaking Factor Due to L.ocal Velocity Perturbation Scaling Factors for SteamAVater ,
i l
Air :
Velocity Void Stability l Peaking Fraction Density Veiocity Damping Peaking Factor, Scaling, Scaling, Scaling, Scaling, Factor, 1 Fa Fv Fd Fv Fop Fs
_a t.
4
- NOTE: 1. Stability peaking f actor for steam / water mixture is calculated as follows:
- _ a,c
, 2. Damping scaling factor is calculated using modal effective void fraction of[ Ja.c for R9C51 tube, l'
l l'
l I
l
! oost s: so.o:zn i2sso 8-17
- . . . . - . . _ . _ _ , _ _ . . . . . .-. .._ -_. . _ . . . _ . . ~ . . _ _ ._
l 1
Table S 2 i Comparison of Air and Steam water Peaking Factor Ratios l Air- Air Steam Steam Peaking Peaking Peaking Peaking Factor Ratio Factor Ratio aC o
l 5
B4 GMb t
00515.to ot2n125so 3--18
Table 8 3 i
Effect of local Variation of AVB Insertion i,
l A to B AVS Peaking- Peaking Ratio
- Type A Type B . Variation Factor A. Factor 8 (B/A) i a,C
~~
a,C l
~~
l
[-
l l
I 00S iS;1D-pt2/112590 8-19 l
I i-
w immmu m I
Table 8 4 l Uncertainties in Test Data and Exitapolation l
Source of Uncertain *,y Type Mag nitude, "6 a,c 1.
2.
3.
4.
5.
- Thisin not at uncertainty associated with the test data.it results from the inaccuracy in s'etermining the true AVB position in the field using eddy current data.
oosts to pt2ni2s90 8-20
Table 8 5 Extrapolation of Test Results to Steam GeneWor Conditions Peaking Factor Test Data with Referenced to Configuration Data Uncertainties Configuration 2a a,C om v :o ytw.590 8-21
.. - - - . - - - . _ - - . - . . - - . ._ - - - - . - . _ ._ - - - - . . - . - ~ ... -
i
' Table 8 6 i
Final Peaking Factors i 1
Configuration Peaking Factor a,C !
i s
cosis:10-ptvt12s9a 8 22
l Table 8 7 ,
Stability Velocity Peaking Factors for Specific Tubes !
Byron Unit 1 Steam Type of AV8 Peaking Generator Row No. Column No. Insertion Depth Factor
~ ~
A 8 88 75 l 27 9 19 10 110,111 0 j B -8 88 10 104 11 9 6
C 8 31 27 11 4 12 8 D 8 88 27 9 92 23 11 107 l - ~
l-t t
f l
l oosis io.p e m s90 8-23 l
Table 8 8 Stability Velocity Peaking Factors for Specific Tubes Braidwood Unit 1 Steam Type of AVB Peaking Generator Row No. Column No insertion Depth Factor
~
a3
~'
A 9 92 98 f
B 9 18,17 ,
11 8 12 5 C 8 47,43,40 9 9 10 92 14 11 23 D 8 23 9 9 D051510 st2/111590 8-24
l I
2 O OO OO O dOOO
" O O OO OO C0 00
'o O OO OOL 6 00 0@ @
0 000000 0@@@@
- O, 54 55 54 O@@@@OOOOO 53 52 51 50 49 44 47 48 45 44
^V" V" 8 '"'
O VIStBLE @ ^ INVISIBLE
$ PLUGGED Figure 81. Original North Anna AVB Configuration (Configuration 1b)
D0220 pt2: 1D/071190 8 25
~
I l
i i
i i
i Figure 8 2. Schematic of Staggered AVB5 D0220-pt2:1D/071190 8 26
_ a,C l
Figure 8-3 AVB " Pair"in ECTTrace l
I cosis:io pt2n12s90 8 27
UOOOO O G0 00 000
" 00G00 Ge o OOO 00000 O O OO.OOO
.' '9 Ga0 0000000000 column OO@DO@OOOOOOO 54 55 54 53 52 51 50 49 48 47 46 l 45 44
$ Plugged Tube g FaaedTube based on the de this ans no Figure 8 4. North Anna 1, Steam Generator C, AVB Positions Critical Review
" AVB Visible" Calls D0220 pt2:1Dl071190 8 28
l 4
1 ... ..m _
t
.,..- M E l H, H Eb 5 E E H. ..!. n...
l mit - - I.uE l..,.E..,
.a G H N.....
=
.n . 2 3.. 4 B R ... ..
nio- - . .n .n ... ..n .,
- i. . .. .. ..s .. ,
M as -
k
- .= 'M ... ..., .. , ,it. it. .. u u it. 1*#
)
R8 = = AVS Avg ' y .) [. ' [v5 Avg VS AVS .A VE AVS AVS VS CES C54 C53 C82 C51 C30 C49 C44 1 "; O r %u U . Urwoopme.
d.C 8C
!a w sw.' H] ] w Sw.-[pr i an.]
l l
Figure 8 5. North Anna 1, Steam Generator C, R9C51 AVB [ Projection]a,c Matrix D0220-pt2:1D/071190 8 29
l 12 0000 0000O00 it OOO 000000 ta OOO 0 00 0 0.0.0.
000 00@O000'000
.0000000000000
$6 55 54 53 52 51 I 50 49 48 47 46 45 44 Figure 8-6. ort Anna R9C51 AVB Final [Projectedla,c Positions D0220 pt2:10/071190 8-30
_ _ _ _ _ _ _ _ _ _ _ - _ - - - - - - - - - - ~ ~ ~ ~
S.C Figure 8-7. Final Peaking Factors for Byron Unit 1 and Braidwood Unit 1 oosis io.pt2rti2590 8 31
i 9,0 STRUCTURAL. AND TUBE VIBRATION ASSESSMENTS 9.1 Tube Mean Stress This section summarizes the analysis to determine stresses in a dented but undeformed tube at 100% power. l.oads imposed on the tube correspond to steady-
. state pressure, differential thermal expansion between the tube and the support plate, and a thru wall thermal gradient. The analysis assumes the tube to be
( )" at cold shutdown.
A summary of the temperature and pressure parameters at 100% power in the vicinity of the top support plate are provided in Table 9-1. The tube temperature corresponds to the average of the primary side water temperature and the plate temperature. The resulting tube / plate radialinterference is [ ]"
Stresses due to differential pressure and interference loads are calculated using finite element analysis with the model shown in Figure 91. The model prescribes
[
ju Two reference cases were run using the finite element model, the first for a primary-to secondary side pressure gradient of 1000 psi, and the second for a [ }" inch radialinterference between the tube and plate. The pressure case incorporates the axialload on the tube by applying a pressure loading along the top face of the model. Plots showing the distribution of stress for the tube outer surface for the two reference cases are provided in Figures 9 2 and 9 3. Thermal bending stresses due to the thru wall thermal gradient are calculated to be 7.25 ksi using conventionc! t analysis techniques. The combined stress distribution along the tube length,in Figure 9 4,was obtained by combining the reference solutions, with appropriate multipliers based on 100% power operating parameters, and the thermal bending stresses. l The maximum axial tensile stress is 18.92 ksi and occurs approximately 0.133 inch above the top surface of the support plate. Adding, for conservatism, the surface stress due to pressure,0.91 ksi, gives an applied mean stress of 19.83 ksi. In addition to the applied stress, residual stresses exist in the tube as a result of the manufacturing process. For mill annealed tubes with subsequent straightening and polishing, residual stresses are com;iressive at the tune surface, but 5 10 mils belciw the surface, the stress levels change to 1015 ksi tensile. Combining the applied and residual stresses results in a cumulative mean stress of approximately 35 ksi, assuming tube denting without deformation.
l D0515 1Dmtat112590 91 '
If a tube is dented with deformation, the mean stress is limited by tube yielding. For the case of dented tubes with deformation, the maximum effect of mean stress was incorporated by using omax = oy in determining stability ratios and fatigue usage.
9.2 Stability Ratio Distribution Based Upon ATHOS An assessment of the potential for tubes to experience fluid clasticinstability in the U bend region has been performed for each of the tubes in rows eight through twelve. This analysis utilizes FASTVIB, a Westinghouse proprietary finite element based computer code, and PLOTVIB, a post processor to FASTVIB. These codes predict the individual responses of an entire row of steam generator tubing exposed to a location dependent fluid velocity and density profile. The program catalates tube natural frequencies and mode shapes using a linear finite element model of the tube. Tne fluid elastic stability ratio Ue/Uc (the ratio of the effective velocity to the critical velocity) and the vibration amplitudes caused by turbulence are calculated for a given velocity / density / void fraction profile and tube support condition. The velocity, density and void fraction distributions are determined using the ATHOS computer code as described in Section 7.3. The WECAN generated mass and stiffness matrices used to represent the tube are also input to the code. (WECAN is also a Westinghouse proprietary computer code.) Additionalinput to FASTVIB/PLOTVlB consists of tube support conditions, fluid elastic stability constant, turbulence constants, and location dependent flow peaking factors.
This process was performed for tubes in the Model D4 steam generator (used in the Byron /Braidwood evaluation) and also for the North Anna Row 9 Column 51 tube (R9C51) using similarly appropriate ATHOS Models. Ratios of the Model D4 results to those of North Anna R9C51 where then generated for each tube in each row of interest. These relative quantities are used to provide an initial assessment of the Byron /Braidwood Unit 1 tubes relative to the ruptured tube at North Anna Unit 1.
j Figure 9 5 and 9 6 contains the results of this process for each of the rows under investigation for Byron Unit 1 and Braidwood Unit 1. The relative ratios are obtained using the following conditions for Byron /Braidwood and North Anna Unit 1:
- 1) Tube is fixed at the top tube support plate,
- 2) Void fraction dependent damping, L
- 3) No AVB supports are active, l 4) location dependent flow peaking factors.
l-( It is to be noted that the stability ratios plotted on each figure are composites of all e four steam generators using mirror image plots to represent tubes on opposite sides of the SG centerline. That is, any peaking effect for a given tube location on the plot 0051510-pt4/112590 92
l represents the maximum value of the peaking factors in all steam generators at that location.
A horizontal kne is drawn at the relative stability ratio value of 0.90. This identifies the point where a ten percent reduction in stability ratio exists relative to North Anna R9C51. (See Section 4.1 for a discussion of the stability ratio reduction criteria.)
All the tubes with ratios above this line would be considered to have stability ratios larger than ninety percent of North Anna R9C51.
These figures indicate that most tubes in Rows 8 thru 11 of the Byron /Braidwood Unit 1 steam generators lie below the 90% line.
All unsupported tubes, with the exception of R12C8 in stearn generator C (Figure 9 S)of Byron 1 and R12C5 in steam generator B (Figure 9 6)of Braidwood 1 have RSR valuts (including flow peaking)less than 0.90.
9.3 Stress Ratio Distribution with Peaking Factor An evaluation was performed to determine the ratio of the Byron /Braidwood Unit 1 tube stress over the North Anna R9C51 tube stress. This ratio is determined using relative stability ratios discussed in the previous section, relative flow peaking f acters (Table 8-7 factors divided by [ ]") and bending moment factors. Sections 4.2 and 4.3 contain additionalinformation and describe the calculational procedure used to obtain the results presented in this section. The results presented below are based upon the following conditions:
- 1) Tube is fixed at the top tube support plate,
- 2) Damping is void fraction dependent,
- 3) Tubes have no AVB support,
- 4) 10% criteria with frequency effects,
- 5) Tubes are assumed to be dented or undented (fixed at top support plate but without tube deformation).
A tube can be considered acceptable if the stress ratio is less than 1.0 when calculated using the procedure described in Sections 4.2 and 4.3 and including the conditions listed above and subject to confirmation of fatigue usage acceptability.
Conformance to these requirements implies that the stress acting on a given tube is expected to be ir. sufficient to produce a fatigue event in a manner similar to the rupture that occurred in the R9C51 tube at North Anna Unit 1.
Figures 9 7 and 9-8 show the results of the stress ratio calculations for the dented (magnetite clamping with tube deformation) condition for each of the Byron Unit 1 D051513-pt4112590 93
i and Braidwood Unit 1 tubes in Rows 8 through 12. Figures 9 9 and 910 contain results of the stress ratio calculations for the undented (magnetite clamping without deformation condition for Byron Unit 1 and Braidwood Unit 1.
As in the case of stability ratios, the plotted stress ratios represent a composite set for all fout steam generators in this unit using mirror image tubes. The critical tube summary hst in Table 9-2 (Byron 1) and Table 9 3 (Braidwood 1)is prepared from these figures in conjunction with flow peaking and AVB support condition information for individual tubes in each steam generator.
The conditions at both Byron Unit 1 and Braidwood Unit 1 have been determined.
No denting or magnetite deposition has been observed at the top tube support plate in either of the units. However, the putential for magnetite deposition, or even denting, to occur during the next period of operation has not been eliminated just because magnetite has not been observed to date.
As can be observed in Figures 9-7 and 9 9, most tubes in Rows 8 through 12 of Byron Unit 1 and Braidwood Unit 1 fallin the acceptance region with respect to U bend fatigue, even when assumed to be unsupported. The only tubes that have stress ratios greater than 10 are R12C8 in steam generator C of Byron 1 and R12C5 in steam generator B of Braidwood 1. These tubes, R12C3 SG:C Byron 1, and R12C5 SG:B Braidwood 1, have stress ratios that exceed the allowable and it has been recommended that the tubes be removed from service using sentinel plugs or tube dampers to restrain motion. This recommendation has been made due to large stress ratios and fatigue usage factors calculated for these tubes in conjunction with the possibility of magnetite deposition occurring during a future operating period.
An evaluation has also been performed to determine the required relative flow peaking that will pre duce a stress ratio not greater than 1.0. Figure 9-11 contains -
the resuits of this prtu ss for all the tubes in Rows 8 through 12 and is applicable to both Byron 1 and Bre , wood 1. T he figure was generated using the conditions outlined previously with the additional constraint that the tubes are dented. Note that this figure reads opposite of the previous figures,i.e., the top curse in the figure corresponds to Row 8 and the bottom curve corresponds to Row 12. Maximum Allowable Relative Flow Peaking is the required relative flow peaking (0.68 corresponds to no flow peaking) that,if used on the given tube, will produce a stress ratio (with denting) not to exceed 1.0.
This curve can be used to identify the relative flow peaking required before preventive action would be recommended and, when used in conjunction with the actual flow peaking associated with each tube, to determine the margin present.
This has also been performed in Table 9-2 and Table 9-3. The column with heading
" Max Allow Rel Fpeak" identifies the relative flow peaking f actor that would be permitted, on a tube by tube basis, before the stress ratio criteria would be exceeded. As can be observed in the table and figure, the innt.r row tubes have larger values of allowable relative flow peaking when compared to the outer rows.
D051s 104tC11:sn 9-4
9.4 Cumulative Fatigue Usage All tubes that are unsupported and have a stress ratio i 1.0 have a maximum stress amplitude that is < 4.0 ksi(from 9.5 ksi) since a 10% reduction in the stability ratio for the North Anna Row 9 Column 51 tube was the criteria basis. The stability ratios i for Byron /Braidwood Unit 1 tubing are based on the currently defined operating I parameters and with future operation on the same basis, the tubes are not expected to rupture as a result of fatigue if: (1) they meet the stress ratio criteria of $1.0 and (2) their current and future fatigue usage will totalless than 1.0.
Based on the above analyses, most Byron Unit 1 and Braidwood Unit 1 tubes meet the relative stress ratio criteria. Table 9-2 and Table 9 3 provide summaries of the '
combined relative stability ratios and the stress ratios for the more salient
, unsupported tubes in Rows 8 through 12. Two tubes, R12C8 SG:C of Byron 1 and L R12C5 SG:8 of Braidwood 1, have stress ratios which exceed 1.0. These tubes also L
have fatigue usages which exceed the allowable value of 1.0, as shown in Table 9-4 Therefore, R12C8 SG:C of Byron 1 and R12C5 SG:B of Braidwood 1 have been recommended for preventive action. Based on the evaluation summarized in _
Appendix A, the installation of cable dampers and solid plugs in the aforementioned tubes reduces the stress ratios to a value significantly lower than 0.1. Thus, the fatigue usage per year after the installation of cable dampers will be reduced to a negligible magnitude.
The fatigue usages for limiting tubes at w o ng plugging levels are summarized in Table 9-4. Acceptability of the U-bend tuuing with respect to North Anna R9C51 type tube. fatigue is accomplished by demonstrating the worst case, but acceptable, tube would not have a total cumulative fatigue usage factor greater than 1.0 at the end of the 40 year operating period. This is accomplished for Byron Unit 1 by looking at tube R11C4 located in SG C. This tube has the largest stress ratio not greater than 1.0 with a value of 0.54. Assuming that this tube became dented shortly after the plant was first brought up to power, and operates at the currently defined operating conditions, the total fatigue usage at the end of the 40 year operating license would be less than 0.38.
l Looking at Braidwood Unit 1 the worst case but acceptable tube is located at R12C4 in steam generator B. It has been determined that this tube has a stress ratio (with denting) of 0.48. Assuming that the tube is dented and that Braidwood Unit 1 continues to operate at the current operating conditions, the calculated fatigue usage is less than 0.22, '
The effect of future modifications in operating cond.tions on the applicability of the reference tube fatigue analysis is evaluated in Section 10. The evaluation defines the acceptable operating space with regard to satisfying the tube fatigue criteria.
l oosts to-pt4>120490 9-5
Table 9-1 100% Power Operating Parameters- Byron /Braidwood Unit 1 Bounding Values for Mean Stress Calculation Primary Pressure = 2250 psia Secondary Pressure = '*U8 ps a Pressure Gradient = 1342 psi Primary Side Temperature * = 578'F Secondary Side Temperature = 533*F Tube Temperature = 556'F
- Average of Th ot = 606.7'F and Tcoid = 549.9'F.
oosis ioSt4n t2s90 9-6
Table 9 2 Byron Unit 1 -Tubes with Significant RSR's or Stress Ratios Flow Max Allow Stress Ratios S.G. Row Column Peak RelFpeak RSR*F.P. W Dent. W/O Dent.
a,c A 8 27 0.506 0.07 0.06 75 0.442 0.03 0.03 87 0.435 0.03 0.03 88 0.506 0.07 0.06 9 19 0.461 0.03 0.03 l 20 0.437 0.03 0.02 10 109 0.568 0.09 0.09 110 0.641 0.17 0.1 E 111 0.646 0.17 0.10 4 _ _ 0.578 0.09 0.08 11 No Unsupported Tubes 12 No Unsupported Tubes it , C
~ ~
B 8 88 0.500 0.07 0.06 89 G431 0.03 0.03 9 2 19 , 0.492 0.05 0.04 95 113 0.492 0 05 0.04 10 2-11 0.594 0.11 0.1 104 113 0.594 0.11 0.1 11 6 0.643 0.14 0.13 7-8 0.634 0.13 0.12 9 0.623 0.12 0.11 10 0.62 0.12 0.11
- a 12 No Unsupported Tubes 00$15;1D-pt4/112$90 97
Table 9-2 (continued)
Byron Unit 1 -Tubes with Significant RSR's or Stress Ratios Flow Max Allow /fp ratios S.G. Row Column Peak ReiFpeak RSR*F.P. ;V4 SD 4. O Dent.
aC s
C 8 27 0.500 0.07 0.06 31 0.492 0.06 0.05 9 2 23 0.492 0.05 0.04 95-111 0.492 0.05 0.04 10 2-12 0.594 0.11 0.1 15-17 0.533 0.06 0.06 105-111 0.568 0.09 0.08 11 5-10 0.653 0.16 0.14 4 0.817 0.54 0.49 2,3 0.7092 0.24 0.22 12 8 1.148 > 2.00 > 2.00 D 8 27 0.506 0.07 0.06 88 0.506 0.07 0.06 9 2 11 0.492 0.05 0.04 15-22 0.488 0.05 0.04 23 0.556 0.09 0.09 27 0.366 0.01 0.01 92 0.478 0.04 0.04 95-113 0.428 0.05 0.04 10 99 0.524 0.06 0.05 104 109 0.551 0.09 0.08 11 107 0.727 0.28 0.26 12 No Unsupported Tubes oosisno-m4/iirs9a 9-8
l Table 9 3 Braidwood Unit 1 - Tubes with Significant RSR's or Stress Ratics Flow Max Allow S.G. Row Stress Ratios Column Peak Rel Fpeak RS R
- F.P. W Dent. W/O Dent.
~ '
a.c A 8 88 O.435 0.03 0.03 9 2 19 0.505 0.05 0.05
!? 0.464 0.03 0.03 iG 0.704 0.33 0.30 03.$ 13 0.505 0.05 0 05 10 ut. 0.491 0.04 0.04 t/i 0.497 0.04 0.04 W 0.511 0.05 0.04 09 0.524 0.06 0.05 103 113 0.594 0.11 0.10 11 108 0.621 0.12 0.11 105 108 0.621 0.12 0.11 12 No Unsupported iv es a,C 8 8 68
~
[ 0.401 0.04 0.04 91 0.424 0.03 0.02 83 0.420 0.02 0.02 82 0.415 _
0.02 0.02 81 0.408 0.02 0.02 9 21 0.447 0.03 0.03 18 0.496 0.05 0.05 17 0.517 0.06 0.06 2 16 0.505 0.05 0.05 19,20,22 0.464 0.03 0.03 95 113 0.505 0.05 0.05 10 2 11 0.594 0.11 0.10 107 0.555 0.08 0.07 108 0.561 0.08 0.07 109 113 0.551 0.07 0.07 11 20 0,709 0.24 0.22 0
0.727 0.28 0.26 12 2-4 0.824 0.48 0.45 5 1.056 > 2.00 > 2.00 D051510 pt4'112590
Table 9 3 (continued)
Braidwood Unit 1 -Tubes with Significant RSR's or Stress Ratios Flow Max Allow Stress Ratios S.G Row Column Peak Ret Fpeak RSR*F.P. W Dent. W/O Dent.
a ,c
~ ~
C 8 34 0.408 0.02 0.02 40 0.442 0.03 0.03 43 0.486 0.05 0.05 47 0.466 0.04 0.04 9 28 0.505 0.05 0.05 9 0.528 0.07 0.06 10 0.500 0.05 0.05 14 27 0.494 0.05 0.04 89 101 0,494 0.05 0.04 113 0.356 0.01 0.01 10 2,3 0.594 0.11 0.10 14 0.540 0.07 0.06 15 25 0.533 0.06 0.06 91 0.526 0.06 0.05 F2 0.536 0.06 0 06 93 101 0.540 0.07 0.06 11 16 24 0.593 0.09 0.08 96 0.562 0.07 0.06 12 No Unsupported Tubes a,c
~ -
D 8 23 0.440 0.03 0.02 24 0.424 0.03 0.03 27 0.435 0.03 0.03 31 0.424 0.03 0.02 87 0.435 0.03 0.03 9 9 0.528 0.07 0.06 95 0.437 0.03 0.02 96-107 0.505 0.05 0.05 108 0.502 0.05 0.05 109 110 0.505 0.05 0.05 113 0.356 0.01 0.01 10 No Unsupported Tubes 11 No Unsupported Tubes I
12 No Unsupported Tubes 0051s 10 pt4/112590 9-10
Table 9-4 Fatigue Usage for Limiting Tubes at Varying Plugging Levels PoiCent Strela Aate F atigue SG Aom/ Cot Pivgging mth Ceniing usaDe Bron Unit 1 A ABC27 0? (0 07 0 000 5 < 010 0 000 to (0 to 0 000 20 0 10 0 000 A A10C111 02 0 17 (0 004 5 50 25 <0004 10 <0 25 c0 004 20 0 25 0 004 C R11C4 02 0 54 <0 384 5 (0 81 <0 384 10 (0 81 <0 384 20 0 81 0 384 C A12C8 02 >2 00 >1 00 5 >2 00 21 00 to >2 00 >1 00 20 >t 00 >1 00 0 A9C23 02 0 09 0 000 5 <013 0 000 10 <013 0 000 20 0 13 0 000 Bre4 wood Unit 1 A A9C92 02 0 33 0 057 5 <049 0 057 10 r.0 49 0 057 20 0 49 0 057 B A11CB 02 0 28 <0 02 4 5 (0 42 <0 024 10 <0 42 <0024 to 0 42 0 024 6 A12C4 02 0 48 <0 215 5 <0 72 <0 215 10 (0 72 <0 215
'O O 72 0 215 B A12C5 02 >2 00 >1 00 5 >2 00 >1.00 10 >2 00 >1 00 20 >2.00 > 1.00 C A6C'3 0.2 0 05 0.000 5 <0 07 0 000 10 <0 07 0 000 20 0.07 0 000 C A10C3 02 0 11 0 001 5 (0.16 0 001 10 <0.16 0 001 20 0.16 0 001 D0515.10 pt4/112590 9-11
a,c Figure 91. Axisymmetric Tube Finite Element Model cos t 5. io otan us90 9-12
i i
1000 PSI PRESSURE LOAD CASE 8
7 / % +- I l l
- ~ ~ /
f-4 h C-C 0 C 0- 0-- 0 -4 " *%%g_ ; ac
< 3 [
av wF 2 \, /if i
3 k k
- 8 0 '
84- _1 I ,l
I 5 L a -2 a N 3 f
?
m -a
-5 E
-6
-7 '
0 0.2 0.4 0.6 0.8 1 DISTANCE FROM TSP CENTERLNE N.
O AX1AL + HOOP Figure 9-2. Dented Tube Stress Distributions Pressure load on Tube l
D0220-c t4 3 0-041090 9 13
1 l
l l
l 1.0 uit itnERFERNCE CASE 30 20 ! -
l %
10 1, 0-o- M- 4--o-eO4 ]
, Pi - " ;_ ; ; ; ::
-10 D !
{ -20 - '
f d T- r/
u -
1 \ -a l / <
g -40 37 -50 !
$ -60 --
&f -10 T b5* -80 G I
-$0 .T If 5 --
-100 T I y )
E -110 b - 120 "I
~130 kk
,3 ,
~140
-150
-160 f 0 0.2 0.4 0.6 0.8 1 OfSTANCE FRou TSP CENTIRLNE, N.
O MtAL t HOOP Figure 9 3. Dented Tube Stress Distributions interference Load on Tube 00:20 ota o 04seno 9-14
, M t
tt]
Y '
g ,
o 9
, 4 e 1 6 F
Q 1{
O ~
e id A xat :.
[
4 s e a=s z
HW# 40 d 3 3 !3 .
F.
. 5 {U \s b. 4 s .a A O .
d L m t N n.
h b \ $
Q "h % \ , 2 g --y 'm
~J~~d gQ ,
l
- C%
g _ o__- A3 m
- (f) tt J h
W8 ~ J +
t;a CO 8 s Z 2 ea M d W
E-(f) ..
M Q O o
- .:> c o.-e o o o o o M 42 w l .e o. e cJ l I i (spuesnoq1)
ISd '33Y380S 301S100 NO SS2HIS Figure 9-4. Dented Tube Stress Distributions Combined Stress Results conoy4 in o4io9o 9 15
a,c l
I I
L l
l l
. Figure 9 5. Relative Stability Ratios Using MEVF Ocpendent Camping -
Byron Unit 1 (Composite of all Steam Generators with Umbrella Flow Peaking) oostsito.oienn9o 9 16
. __ . _ _ . _ _ . . . . _ . . . . _ , . . ,. _ . . . . . . .- . , . , _ . . . . , . . ~. . . . . _ - .
a,C Figure 9-6. Relative Stability Ratios Using MEVF Dependent Damping -
Braidwood Unit 1 (Composite of all Steam Generators with Umbrella Flow Peaking)
D0515:10-pt4d 12590 9-17
r J
}
d,C
~
b Figure 9-7. Stress Ratio vs. Column Number - Dented Condition - Byron Unit 1 (Composite of all SGs with Umbrella Flow Peaking) oosis: to pt4:112590 9 18
( -
a,C L --,
r k
E
$6
(;f Figure 4 8. Stress Ratio vs. Column Number - Dented Condition - Braidwood Unit 1 (Composite of all SGs with Umbrella Flow Peaking)
- cos15: m-pt41' 2590 9-19
a,C E
Figure 9-9. Stress Ratio vs. Column Number - Undented Condition - Byron Unit 1 (Composite of All SGs with Umbrella Flow Peaking) l D0515.10 pt4/112590 9-20 l
a,c Figure 9-10. Stress Ratio vs. Column Number - Uncented Condition -
Braidwood Unit 1 (Composite of All SGs with Umbrella Flow Peaking) oasis:to owii2s90 9-21
8.C Figure 9-11. Byron Unit 1 and Braidwood Unit 1 - Maximum Allowable Relative Flow Peaking oosts to-otaru2590 9-22
10.0 OPERATING LIMIT EVALU AllON The reference evaluation presented in the preceding sections of this report considers the operating conditions of Byron 1 and Braidwood 1 and projects performance based on continued operation at these reference operating conditions. These conditionsinclude: 10006 of full power,908 psia steam pressure,606.7"F primary inlet temperature, and 0.2% tube plugging. The evaluation concluded that only one tube in each of the units would require corrective action to preclude a North Anna type fatigue rupture.
A supplemental evaluati- is described in this section which addresses the effect of future modifications in ierating conditions. This evaluation defines the acceptable operati' terms of key thermal / hydraulic parameters, with regard to satisfying - _ . f atigue criteria. Operation with steam pressures and flows in the acceptaby region willinsure that the U-bend fatigue usage criteria will not be violated for the remainder of the design basis operating period. This information should enable CECO to develop the appropriate administrative controls to insure thct future operation is within the limits of the f atigue assessment.
10.1 Summary and Conclusions The parameters which affect the tube vibration analysis and which could change with time (or SG mechanical modifications) are the velocity, density, and void fraction of the steam / water mixture passing through the U-bend region of the tube bundle. The key SG operating conditions which influence these U-bend parameters and are to be monitored are the steam flow rate, steam pressure, and circulation ratio.
The analysis described in this section concludes that if no mechanical modifications are made which change the flow resistance of the SG secondary side recirculation loop, the provided graphs of steam generator feedwater flow versus steam generator pressure for Byron 1 and Braidwood 1 (Figures 10-2 and 10-3) can be used to validate that the steam generators remain within the analyzed limits of the reference tube fatigue analysis. No additional tubes will exceed the fatigue criteria and become candidates for preventive action as long as the operating conditions remain above the recommended operating limits shown in these figures. Based on the location of the current reference operating conditicn, pressure margins of about 45 and 50 psi exist for Byron 1 and Braidwood 1, respectively.
These conclusior.s assume that validation data will be obtained from the plant process computer with the instrumentation currently used and that these instruments will be maintained within their current stated limits of accuracy.
oosis to.mm us90 10-1
10.2 Limiting SG eacondary Side Variables J.11 Steam Flow "eam fit w rate is most easily deduced by subtracting the blowdown flow rate from he measured feedwater flow. Feedwater flow is monitored by instrumentation,
.th a stated, measurement accuracy of 12.9% over a maximum span of
, iO6'.m/hr, The resulting worst case flow uncertainty is, therefore, = 0.15 x 106 r " that the 2.9% value is based on a statistical combination of a number s . component uncertainties that influence the flow measurement sor and rack effects). Further, it is based on a flow indication obtained a it process computer.
un flow,if present,is typically a small fraction of the total feedwater flow typically 5 1%) Since blowdown flow subtracts from the volume of steam y ' w ed, a steam flow value based on the measured feedwater flow rate will be 1 e. sative.
Steam Pressure
+ wuming that the pressure signal to be used for comparison with the operating iimits derived herein is also obtained from the plant process computer, a worst case steam pressure measurement uncertainty of 3.3% of the total measurement span of 1300 psi is appropriate ( t 43 psi). Being measured in the steam line downstream
, of the SG's, this value is a conservative a'ssessment of SG pressure for use in tube vibration analyses due to line losses occurring between the SG and the measurement location.
10.2.3 Combined Allowance for Flow and Pressure Measurement Uncertainties l An algebraic summation of the preceding, individual measurement uncertainties for steam flow and pressure would result in an overly conservative uncertainty allowance. Such an approach could result in a final operating limit which is too conservative relative to the current operating condition, and lead to an unreasonably small operating margin.
Alternately, a more reasonable approach for applying the measurement allowance has been used. A combined uncertainty based on a " root mean square" of the two component uncertainties has been calculated. The resulting equivalent pressure a!!owance is about i50 psi, itis calculated from the 43 psi pressure measurement uncertainty and the pressure equivalent of the flow measurement uncertainty, = 30 psi:(FsMS = (432 + 302)1/2 = 50 psi). This 50 psi uncertainty allowance will be ddded to the best estimate operating limit determined from the tube stress / fatigue
, evaluation to produce the recommended upper bound operating limit.
l I
I D051510 pt7/11259o 10-2 l
l 10.2.4 Circulation Ratio Any mechanical modifications to SG secondary side hardware such as the AVB's or the primary separators could change the flow resistance of the SG recirculation loop, changing the amount of flow circulating through the tube bundle. This change would be relatively independent of thermal parameters. Any such change which reduces the SG secondary side flow resistance and increases the recirculating flow can be assumed to have a deleterious effect on tube vibration. Therefore, no modification should be made to AVB's or primry separators without evaluating the possible effect on secondary side circulation - s.
10.3 Parametric One-Dimensional Relative Stability Ratio Analysis 10.3.1 1D Relative Stability Ratio Methodology The assessment to determine the susceptibility of specific tubes to U-bend fatigue rnakes use of the reference three-dimensional (3D) flow, tube stability and stress ratio analyses described in Sections 7 and 9 along with appropriate adjustment factors. These adjustment f actors are used to scale the reference tube specific stability ratios to account for the effect of changes in operating conditions. They are derived using the one dimensional (1D) relative stability ratio (RSR) calculation technique described below. The 1D RSR adjustment provides a means of generating simulated 3D stability ratios for an alternate set of operating conditions without having to complete a specific, detailed 3D flow field calculation for each condition.
The 1D RSR ratio compares the fluid elastic stability ratio computed for a particular set of operating conditions to a reference stability ratio. For the Byron /Braidwood analysis, the relative stability ratio technique has been used in two ways: (1) to select the limiting reference operating conditions (SG C in Byron 1) as presented in Section 7.1, and (2) to define adjustment factors which are applied to the reference 3D stability ratios derived from the reference ATHOS analysis to account for operation with modified conditions (power, plugging, primary temperature, steam pressure).
The fluidelastic stability ratio is defined as the ratio of the effective fluid velocity acting on a given tube to the critical velocity at which large amphtude fluidelastic vibration initiates:
U (10 1)
FiuulelasticStability Ratio. SR = '#"'"
crstscal at ourt of mstalsisty in this ratio, the effective velocity depends on the distribution of flow velocity and fluid density, and on the mode shape of vibration. The critical velocity is based on experimental data and has been shown to be dependent upon the tube natural costs towmsw 10-3
l frequency, damping,the geometry of the tube, the tube pattern, and the fluid density, along with the appropriate correlation coefficients.
The detailed calculation of this ratio using velocity and density distributions, etc.,
requires three-dimensional thermal / hydraulic and tube vibration calculations which are time consuming. Alternately, a simplified, one-dimensional version of this ratio has been used to provide a relative assessment technique for determining the effect of changes in operating conditions on the stability ratio. The relative stability ratio is defined by the following equation:
GZ v
in this equation "ALTC" refers to an alternate condition (such as reduced steam pressure) and "REF" to the reference Byron 1 condition. While this simplified approach cannot account for three-dimensional tube bundle effects, it does consider the major operational parameters affecting the stability ratio. Four components make up this ratio: a loading term based on the dynamic pressure (pV2), a tube
- incremental mass (m) term, the natural frequency of the tube (fn), and a damping ratio (S) term. It should be noted that the ratio is relative, in that each component is ,
expressed as a ratio of the value for a particular operating condition to that of the reference operating point.
(The calculation of the relative stability ratio makes use of the overall steam generator operating conditions (steam flow, steam pressure, and circulation ratio),
along with the geometry, to calculate an average secondary-side U bend void -
fraction, density, and radial tube gap velocity. The velocity and density are then used in forming a ratio based on dynamic pressure (pV2)1/2, which is one component of the relative stability ratio. A second component is the tube damping which is calculated from an experimentally-derived correlation dependent upon the void fraction]a.c. The particular damping correlation which is used for all relative stability ratio calculations is based on a dented condition at the top tube support plate (a clamned condition, as discussed in Section 5.2). The clamped condition is also assumed in calculating the tube natural frequency.
Justification for use of a simplified, one-dimensional, relative stability ratio adjustment factor is provided by making comparisons with the results obtained from more detailed three-dimensional flow field / tube vibration calculations. This comparison has been made for over a dozen other steam generators which have been evaluated, to date, including both preheat generators like the Byron /Braidwood Model D4's and feedring units. The results indicate that the 1D oosts io#un us90 10-4
method provides a good or modestly conservative prediction of the 3D relative stability ratios. In particular, the plant to plant variation in the 1D ratios follows the 3D trend as the steam flow, pressure, and circulation ratio varies This indicates that the operating condition contribution to the relative stability ratio can be adequately accounted for by the 1D approach. The 1D to 3D adjustment is justifiable as long as it's applied within a group of steam generators which share a common or the same tube bundle configuration. In these situations, the overall tube bundle flow fields will be similar and the individual plant ratios will dif fer only as a result of the ef fects of variations in the basic thermal / hydraulic parameters.
10.3.2 1D RSR Calculations for Byron 1/Braidwood 1 The 1D relative stability ratio technique has been used to calculate adjustment factors for a matrix of operating conditions for the Byron 1/Braidwood 1 generators.
These conditions consist of 48 combinations of power level (90,95,100, and 105o6 of full power), tube plugging (0.2,10,20, and 30%), and primary inlet temperature (606.7,601.7, and 596.7#F). For each condition, a GEND code performance run was completed to determine the steam pressure, steam flow rate, and circulation ratio which correspond to the input power / plugging /Thot combination. Figure 10-1 identifies the matrix of operating conditions (0) for the reference Thot. Similar figures were also generated for the other two reduced That conditions.
Subsequently, a 1D RSR was computed for each condition. By interpolating among the matrix of resulting RSR's,it was possible to establish a number of constant RSR lines which are also shown in Figure 10-1. Note, that the RSR's are all relative to the reference thermal / hydraulic condition (RSR = 1.00) described previously in Section 7.
The reference condition is based on recent reduced temperature operation at Byron 1.
Plant Thermal Power = 856.25 MWt (100%)
SG(C) Thermal Power = 865.2 MWt (101.0%)
Steam Pressure = 908 psia Steam Flow Rate = 3.78 x 106 lbm/hr Thot = 606.7*F Tube Plugging = 0.2 %
Circulation Ratio = 2.30 This reference condition is also assumed to be applicable to Braidwood 1.
10.4 Structural and Tube Vibrati >n Assessment The reference evaluation reported in Section 9 examines the steam generator tubes which are at potential risk to deselop a North Anna R9C51 type tube rupture. The 1 analysis is performed using current and past operating histories and assumes that future operation would be at similar levels Based on an analysis of tube stability and stress ratios along with a fatigue usage calculation for the most highly stressed costs opmmsw 10-5 '
i tube,it is concluded that only one tube in Byron 1 (R12C8 in SG C) and one tube in Braidwood 1 (R12C5 in SG B) will require preventive action to preclude a fatigue rupture at the top tube support plate in a manner similar to the rupture which occurred at North Anna 1.
The methodology for calculating relative stability and stress ratios is the same as that described previously in Section 4 and will not be repeated in this section. However, a brief summary of the method and comparison with the evaluation criteria is as follows:
(1) Only dented tubes lacking AVB support with relative stability ratios above 0.90 (compared to N. Anna R9C51) are potentially at risk.
(2) Of the tubes with relative stability ratios above 0.90, only those with stress ratios exceeding 1.00 are potentially susceptible. The stress ratio screening provides a conservative assessment of tubes which could potentially accumulate fatigue usage greater than 1.00,i.e., actual computed fatigue usage for a tube with a stress ratio of 1.00 is less than 1.00. Note that stress ratio (and therefore fatigue usage) is a function of various parameters such as; relative stability ratio, tube stiffness, and tube frequency to name a few. Therefore, these parameters must also be considered in defining potentially susceptible tubes.
(3) For tubes with stress ratios above 1.00, only those with a computed fatigue usage above 1,00 for the design basis operating period are identified as requiring preventive action to preclude a potential rupture.
From the reference analysis described in Section 9, a listing of the critical tubcs having high stress ratios is presented (Tables 9-2 and 9-3). As indicated in the tables, only a few key tubes are evaluated in detail; however, these tubes envelope several other similar tubes having the same or slightly lower stress ratios.
For the operating limit analysis, the reference stability ratios are scaled by applying 10 RSR adjustment factors in the range 1.00 to 1.40. Tube stress ratios are then recoi. futed for these increased stability ratios to identify tubes having stress ratios above 1.00. For tubes with the highest stress ratios, individual fatigue usage factors are then calculated considering: 1) usage accumulated to-date at current and past operating conditions, and 2) future usage accumulated during assumed operation with higher stability ratios, i.e., with higher power and/or lower pressure.
By repeating this process with different RSR adjustment factors,it is possible to determine a particular factor at which the fatigue usage for the most limiting tube will be just slightly below 1.00 at the end of the 40 year operating period. These tubes are identified in Table 10-1. For Byron 1, the limiting RSR multiplier is 1.117.
This means that the steam generators would have to operate from the present time to the end of the 40 year design basis period with conditions that produce a 1.117 RSR multiplier, before any tubes are identified for preventive action to preclude a 1
D0515 10-pt7/112590 10-6
fatigue rupture. No tubes are identified for action if operation occurs with a
- multiplier below 1.117. If conditions are such that the 1.117 multiplier is reached, then only one additional tube will require action. Similarly,if operating conditions are adopted such that a multiplier of 1.257 is reached, one additional tube would be identified for action Were even more adverse operating conditions to be adopted such that the RSR multiplier is increased to 1.293,1.373, or 1.392, and additional two, two, and six tubes, respectively, would be identified for action.
Similarly for Braidwood 1, an increase in the RSR multiplier to the 1.10 Lmit results in one additional tube which will require preventive action to preclude a fatigue rupture during the 40 year design basis period. Further increases in the multiplier to 1.210,1.235, and 1.255 would lead to the identification of one tube,in each case, for preventive action. Were even more adverse operating conditions to be adopted such that the RSR multiplier is increased to 1.291 or 1.381, an additional six tubes and one tube, respectively,would be identified for action. No tubes are identified for action during the remainder of the 40 year design basis period if operation occurs with a multiplier below 1.141 As indicated by these results, even if the operating conditions were to change such that the R5R limits are reached or exceeded, only a few tubes would potentially be affected in each plant.
10.5 Establishment of Operating Limits The allowable engineering estimate operating limits determined from the preceding stress / fatigue analysis ao shown in Figures 10-2 and 10-3 for Byron 1 and Braidwood 1, respectively, in these figures, any combination of flow and pressure along these limit lines will have constant RSR's and identical results in terms of tube stress ratios and fatigue usage. Operation with any combination of flow and pressure on the upper side of the limit is " acceptable" (no preventive action required to preclude a potential fatigue rupture). Conversely, conditions having higher flows and/or lower steam pressures and located on the lower side of the line would require preventive action prior to the completion of the 40 year operating period and are labeled
" unacceptable".
The operating limits described above are based on a best estimate assessment. As discussed previously, the feedwater flow will provide a conservative representation of the steam flow rate. For this reason,it is listed as the cbscissa in these operating limit figures. Allowance must also be made for the potential for measurement uncertainties in the two key operating parameters, flow and pressure. Applying a statistically-combined allowance for the flow and steam pressure measurement uncertainties, a second operating limit is defined above the best estimate limit derived from the stress / fatigue analysis. Future,long-term, steady-state operation should be restricted to pressure and flow conditions lying above this limit.
D0515; 1 D-pt'//11259o 10-7
Figures 10-2 and 10-3 alsoidentify the location of the reference condition (Pst = 908 psia,Ws = 3.78 x 106lbm/hr). The distance between this point and the operating limits thus represents the allowable margin before preventive action needs to be considered, For the design basis 40 year operating period, the following approximate margins are indicated for Byron 1 and Braidwood 1:
Byron 1 Braidwood 1 !
Flow margin at constant pressure (Ibm /hr) 0.20 x 106 0.27 x 106 !
(5% power) (7% power)
Pressure margin at constant flow (psi) 45 50 The actual margin, however,is expected to be somewhat larger than indicated in these figures, This is a result of the tube stress ratio / fatigue analysis method which adds conservatism to the overall evaluation. In developing the limit lines, the more adverse operating conditions with higher RSR factors are assumed to initiate immediately and continue at this level throughout the remainder of the total operating period in actual operation, however, the change from the reference condition willlikely be a gradual process. Pressure reductions can be expected to occur gradually over many years as tube plugging rises. As a result, the actual RSR adjustment can be expected to also increase gradually and actual fatigue usage will be less than the usage calculated by the method used herein.
10.6 Discussion and Application of the Operating Limits Operating limits have been determined which define the acceptable operating conditions for which no additional
- preventive action,i.e., installing sentinel plugs, or stabilizers and plugs, will be required to preclude a U-bend fatigue rupture of the type which occurred at North Anna 1 in 1987. The limits are described in terms of two key thermal / hydraulic operating parameters which influence the calculation of fluidelastic U-bend vibration and fatigue - flow and pressure. In defining the limits, allowance has been included for the effects cf measurement uncertainties on the parameters.
The limits shown in Figures 10-2 and 10-3 are based on the results for the reference Thot operating condition. It was determined that this condition yielded a somewhat more conservative placement of the operating limit compared to the ' results
- obtained for operation with reduced primary temperatures. The decision was made to present only the limit corresponding to the reference T hot since: 1) it is more conservative and,2)it simplifies the implementation of the results since plant
- Excluding the single tube in each unit identified in the reference evaluation.
D0515 tD-pt7/11259o 10-8
l personnel need only be concerned with a sirgle lim;t line rather than a family of lines corresponding to various temperatures.
The fact that separate limit lines exist for each of the three primary temperatures considered in the evaluction implies that the results are somewhat sensitive to the manner in which a particular steam pressure is achieved. This sensitivity is related to the operating characteristics of the Model Od steam generator. The inclusion of a preheater in the Model D4 results in a dependency between steam pressure and the secondary side flow through the various regions of the generator. (This steam pressure / secondary flow dependency is not present in non-preheat steam generators.) Furthermore, the geometry and hydraulics of the D4 are such that the unit operates with a relatively low circulation ratio. At low values of circulation ratio the calculated damping and 10 RSR's are increasingly sensitive to incremental changcs in both circulation ratio and steam pressure. This characteristic, together with the pressure / flow dependency, account for the separate performance and limit lines derived in the evaluation (but not shown in Figures 10-1 to 10-3).
The results for Byron 1 and Rreadwood 1 indicate that the steam pressure would have to decrease to below about 865 and 860 psia, respectively, at the reference 100%
power level before any preventive action should be considered. This conclusion is appropriate to the 40 year design basis operating period.
A wide variety of operating conditions can be postulated which will yield the limiting 1D RSR multipliers at which additional preventive action would be required.
As an aid in determining these conditions, Figure 10-4 has been constructed. This figure presents the tradeoff between power level (or feedwater flow), tube plugging, primary inlet temperature, and the 1D RSR multiplier. By interpolating between the appropriate curves,it is possible to dc ' ermine various power (flow)/
plugging /That combinations which produce a given RSR multiplier. For example, assuming that full power is maintained, limiting RSR multipliers for each unit will occur at the tube plugging levels listed below.
Tube Plugging % at Limiting Best Estimate RSR Multipliers Byron 1 Braidwood 1 (1.117) (1.141)
Reference Thot = 24% = 27%
-5 F = 17 % = 21 %
- 10*F = 10% w 14 %
The corresponding plugging values for the situation in which the measurement uncer*ainty allowance is included (and the RSR multipliers are ef fectively reduced) will be reduced compared to these values. For example, with the measurement uncertainty included, the 1.117 limiting RSR for Byron 1 is reduced to = 1.04 and the reference That plugging is reduced from 24% to about 12%. Similarly, with the oasis to.c 7:s t:s90 10-9
measurement uncertaintyincluded, the 1.141 limiting RSR for Braidwood 1 is reduced to == 1.055 and the reference T h ot plugging is reduced from 27% to about 15 % .
costs to pon12590 10-10
l Table 101 Stress Analysis / Fatigue Results for Enveloping Tubes Based on a 40 Year Operating Period RSR Multiplier at which Evaluated and Number of Enveloped Tubes Evaluated leictemental -
Eag Reauire Action Tube Other Enveloped Tubes Affected Tubes Byron 1 1.0 (Ref) R12C8(C) None 1 1.117 R11C4(C) None 1 1.257 R11C107(D) None 1 1.293 R11C3(C) R11C2(C) 2 1.373 R10C111(A) R10C110(A) 2 1.392 R11C5(C) R11C610(C) 6 Braidwood 1 1.0 (Ref.) R12C5(B) None 1 1.141 R12C3(B) None 1 1.210 R9C92(A) None 1 1 235 R12C4(B) None 1 1.255 R11C8(B) None 1 1.291 R11C3(B) R11C2,4 7(B) 6 1.381 R12C2(B) None 1 l
l l-l oosis ic p:71112590 10-11
Figure 10-1 Variation in RSR with Pressure and Flow Byron 1/Braidwood 1: Current (Reference) T hot (606.7'F) a,c M enum D05151D pt7/112590 10-12
i Figure 10-2 Byron Unit 1 - Operating Limit on Pressure and Flow To Preclude Fatigue Rupture During 40 Year Design Basis Operating Period a.C t
M oosis:io pun us90 10 13
Figure 10-3 Braidwood Unit 1 - Operating Limit on Pressure an'd Flow To Preclude Fatigue Rupture During 40 Year Design Basis Operating Period B.C 4
l i
i i
i oosis:in-pi7esiis90 10-14
Figure 10-4 Byron 1/Braidwood 1..
RSR Multiplier vs Flow,Thot, Plugging a,c M
costs io-nt7/ii2sso 10-15 i
APPENDlX A EVALUATION OF THE [ Ja b,(TUBE DAMPER a,b,c m
m oos t s:10-ptan 20490 A-1
3,b,C mau M
A-2
,2J.__ A._44 6,W,,4 m Am. -Ah.a.144-m.44-.a4 e--ryu4m- .*T4 A 4- h.h A v4.**dhA.44%.4-2--4 a.254ir.44 E.J Ju.- A G -mea.eas mdb+h2a.i A- -A,,,
4 W
p a,b.C Wee i
l l
1 I
+
3 d
i o
A-3
) 4 4 - d. je 4 12ay ,aaE A-++.Sc*4.AW.- i$i._.A,J4Aa4,4-~E,*e.#--4 Fw+e'-_N-a9*A1. ara-4 4 ==E14.+5*~ - 4' e A + b n 944 a*-k de Mv3u>&- 4 4* )au eh P 2.*4's-. 4--h.rJ a a,b,C _. ,
M '
4
.J,-*
k
.l.
O m
Ae5
-- " - s . .e r ,e - ----~ w
d iar-- .A.AAS a,ML .a 4.a. # .4 h e., ,M J de._4 ,b-e A d af_a_w, u.sm_A E, ,s,4,,j..a 1 ,pLSa b J. g,,,aw -
,h_.a ,yp.h 4 :., , a gasAJ 4.m 4.aw,g__ aum , A 44w _ ,
afb.C eum
. . -h t
l I
a I
i I
l l __
A-4
a,b,C emans A-6
-___-____----_-__-_a
A a,b,C- !
egne I
E 1
I f
's i
t f
6 f
t : i i
b
~
f l
('. '
)
I F
f k
'h ,
e i
I
)
b L
I .. _ L l- I mean . >
mes-D A-7
5,b,C
~
7 A8
a,b C mitee A9
8,b,C
= -
essau A-10
O M
l cos15:10 pts /120490 A-10
_ _ _ _ _ __ _ _ _ _ _ _ ______ _