ML20079P179
| ML20079P179 | |
| Person / Time | |
|---|---|
| Site: | San Onofre  |
| Issue date: | 11/05/1991 |
| From: | Gartland G, Herschthal M, Strom W SOUTHERN CALIFORNIA EDISON CO. |
| To: | |
| Shared Package | |
| ML13302B812 | List: |
| References | |
| RCE-91-025, RCE-91-25, NUDOCS 9111120270 | |
| Download: ML20079P179 (76) | |
Text
{{#Wiki_filter:', . :
.. , . " ,. ,..- . , s ~. ..,
r g.. e
;p 1
g . g y lr g .. ,,
.,',.'l, y f, .l',. , ,', % p e; 4,,,t ' .* i[ o
.., } l , i. g ,, t
, e
.,i f ,g.^.,t,.
,a .,,i '
. '/ , , , ' . ,j
, M ' , , ' ri,. es,
- , 's ,g 9 , ..)r' , ]6,,., ] e* i.
r,'no ,, s . .4i
*. ,' ., 4.,5 ,e i .',,, ', .,m,.. -a(' jy(
.a . ,
6 ;. <
.m .. ...
i .. 1 . . ,, , ', . q, 7 . .
.m..
j .@4 ],
- Q ..' q,{
.A .c-' j ps ..
i, ' u i
. . . . , .- "), y, m4 30 i < r } }. ii :. c j.) ' ([ ',.v,.p. Sl, 4 . , ,
C . '.i, .
,t,.
[ [ h '. g '
' l ' ,. i -
. ' , , . ' ' - . i.' , ,
I
'. I o . '
s,, ' . , . *'
,I, s . i . , I ,'. ' J q, . . . .,,y,.-
" +.
~
.'y '
h.
..-+ .. .
-[ ' ' ' . j
. .. ' , ', h + . . ' ' ,
l., g.,d . p . a, .. - . . , .. , .'.,_?,2. . . , . .*I. .' - ,,j, 't N. . /- W,
,-l - .
' ' ,l, 9
- 1 (p. .T,. -
,.- y ,. . . m .,
, ' ,~ .'
s .. 3. . . 3,',
., . 3 , .
V 7._.
. p. .i . ,. , 7 j ;g y i.. . ,' .' .q, g. f g , , .' an j e '? . . . . ; ,; . . . ,- .=,f
, . . , ' ' . ;< ' . c .' . , . .. ' . .f.A
..e 7 " c'y' .: , . . . ' ' '. ,,'1 'Fi ., . . . f 4g ;';. 7, j ;,.
.- 3.;':-
,v ', 1 . ,
c ,- t, > m,.<'. 'i
,,, - ' '; -+<
, ',. ., c .; . ; ' 8' , . g* , . % ; ~ 1. ,w
...c, , , -','. :, . b .h ,,
',e',',",.-' 4,.. ' , , ' .,
- M :, .; .
;\ -
t
.-. - - . s - ? .
r ' t
%~- f .e 'q,.,, , ,. }' '.,- .,3
*' 4
', I 'j/ ' ( . [;' *~
1p1 #4hl* 49 ., i'[, .' $h *'
*. , p,
' * " '.*I. J.
\- ( }' *. '
f.*
,4, c' f 5, ,,'; [.,,' ,' / ' ir ' , , ' . >' ~
g ' [ yf.-f;. ,3 , p. - . ;, y ',. . [ _{., . ' &g', v$ ; %,.',c@,j . Gj ', y };;%, 't.. * ' ,
, , . .. 4 2;
.< .[ L , .
; ; g,: }. .'.. .. l, ,a. _ ,3v. i, . M' .: ,,
[ .,3. : ' ; ,y.. .
}e.
- ,. .. 4.
. - e e . . - or
~
, u cec 'j, .. 'y Q , ; ~, ', , . * ; 7 , , gw,. /
- $, .. , 9,', .'. .
- r: '.s.
,Q., ' .['. Y y : . ,.: , ,. ':..Jo.g
.. . *q
, , . .,a' '4., ;. q .. . h-'.c,:, , ~ ;., j: i, 'K 8t s= .. p ' ., , :..
o . .
%). ,. . *
;....f,
*+ M: y. . . . . , , , . , '. . .4 t- \ n )* 1 x..,, .
,,.;.'.;4'-
s ~ - '
\
' * ]* .'.u.;. < / 4. 4sis [ ! ,%,
..,i.'.,.'.g. *i'.. j .ab.'
'.,- i
', . , . t
.g. .8
. ', . '.' 8}'t} . ,a ' ' *.<
1 a 'q. 4
' 'y t - ) '4. J ' '
".'*f . .d
- Ti*' g . !. 2 ,,,...., . ; * . A
' - , ' ' [. c *.,.',i .. -'
.i., ,.i '* .
', [, .S . 4 , - * . . .,.. I $ . .. j, -g# . ,'.; . ,[ '.,P ,. ; s
.c 4 I.
,. . i
', . fj'\!.s-
,l y . i p. p ' y *4
- d. .f... *, ' I, '*g a.; \.'
*} ..i.-.'.',..,s g.. s
,. 4 ., e,',.J...i -f,.,*' y.< . ' . ' . - - ..
.\
i 4 '
.is , gi - .pi e, ...e . , . -
i.,c... J.*.4'"i.,,' g ., ;
- f. 1'-
3, g ,1 3 . c, *. .g
,s -
'.h ',.
i
,e *2.,.i.- .; . . . . ...ei,-
*t A / . .. . .-' ' .. >.....g ,
,e . ,.
.' .(
#t..
1
. y., ,4. .'- c .
. + ' ..,.t.i.,.- si . , .,.i**
. ,o, p
$. '.. , .n .,pg{.. ' A*.4.'- 'e . .i
'.,x e, .1
."....e.
. . .. . ,g ,. ,
s'; ' 3
'\J ..ll'. -. ;* .i."
, i' ' y . . ..
'3 4 , y. e , .r. [ ' .,*]e . \<,. 'e' dl, '? i ', j ., ' ), J' , c - '
.' : l' ,,w ',
i .,,., , cs :, ...' .. ' 4, . . c 5 ,
+ . . . J, n s ; ' . g ' .c ,, J J. . <
...'.! . s ' g, s
...f a
",. g Av , ' . . . . -:)*g, . , * ' *- . O .: /4,'
,,,t;* ..i " #. ,
. s . o, - .g., i.,4 s p '.e i .- h '. ' .. ' ' " .g ,,- -,
, f- .i W ". - v - * /. .
1 *
- p -
. r ) .
' i ,, .
4
,;' 3. . j o;. .. . .
..b., ,
,. , .at,.... -
. , . ... c o . -
4 . o ;, . g :. A . g. .- . . - ,h*. }. o r.*<- y,
, . .,# t& q,sc 4 ., .. , . 6 ;,
[;. n [ [4 '{,,,; . a .{ 5 , [,;.{.h .,]~c, j i g, j.[:hJ.
,$. Mf.h,, h: .hr4 ..%gg[,. L m.c . ; 9' n . ,.
.- h,. (f.-[ .i ,[ h...k.'.',k . ,f .. .f. i .,' . .
. ~. . , . + .
.. s. ,....- .r , , . .. y...
..s.
p, ,. . ; s t'..'+..
,., e ., .f * , t . . . . ;, * \. j - l.
l t ,,% '.i , q [;.y - ,,y , t ,[ . ' - c ,f..#. . [. 1.; t f-8g. . 3: '.',J,
. a
,4g . . ' . .. , ;n. 9 f;'.;..,: .L<,
y9,>. ',s.
; f. ; .s ' l4 ](,; + . . . , .. ..og ; .. ,j.
- a. . y . ,. 7..3,,,.....
. , . y t ia . c a . ,t,, . j;. ., w .
. ..i;
.a;....- ,
*y
>f >j- ; ; . .- 4, : l~ r... .4,. .. .s . : 3, + p . *...Jyi ~.S.
g-' , ,u , - ,
*; , . . . - o i ; 3.. .-l 3 3 g : .ic .
. . . . . c. , OP. ,
. # .-,. .J. c
. . . .. f. .. . . . . ..,,-,,,., . ....r. ,. 4 , . t~ 4-. .
..n ,e e
g. i f U ', '
' f' '- .. ,[ +S k .t.' ' : ,. ! . 3' . l -
'.'~,..
,j\,,*. I [ ' ..,.[ .,.. . ' '. J ' f. , s?' ' k , . . =.'.': [ , , vi
' .i .. ..'.. '... ',' 'aq- ",,
p . . y6' (' . . J{'
. f ,b . ,-
., ; ,,'"..'.},..'q,,',i,'l+sk
(-l . ' 1lc . [ I h.[. , ', ; .; f..,,2J l* . .;.,7. ;. .+ -
, v .1.'s . , .;, . ..',..
.-.,t, l) ; & 3 l
o".
- ... ' ' ' . a
- s. s - '\
.;..,.,,..'.i>*','.o.:.,
.',,f . i .. t. .' **
- ,'- .y f I s,. * ' . .
, f.
.r,*^f.. ,g ;. '. , ft f .',i ir t a
. . p' ' i . .A +,* '
t . - l {..,.., 3
'.c'
+
'. p ' .,
.- -i . 4; , l, . , . . e d, f. . .
.-a,.,. *.b.
g - . .
,g,'*'..,' ,r.'
,f,..
, .t*
. ,' , *j .- l j,$ e
., J , < ; . . . ; . ,*'g , y .
~...,' , , . . , .
*l*',e ,.f,' 2.- ,,*
' .;- '*. 'W s.. $w ,' e ' ,.. , . , . , .
. ' Ne r.? a .,.s . n .gl t,
-; 4,\,* y
.. ...p. ..'.t . . , * ,.
. . ..,,t a
+
s:,
*.\... ,.J.',,. ,
- t. ;" p'.
, I'i]',l-%.*...'. .j Q ] .. ' . : - ' f: . ., [.,J.' .', .' .. .. .,;.
.- .sq * . a * ^
- ll , *(.
l , ., - .. ';g .}q, . ' ,. . -l 't.', *f. y
*', .. . J . p. .*3* c. . ', i l . ; s. - ' ' "J' .
., o- .,,6,
-g
% ;, . . . gj, .f g$,
- r , ., ( t . p
.,. vg : .' (. 5 . v..,f H .-'
. . ,, ; ' .q. , t ,
. f .'**." y
.,,i g ., . . . . . . . 1 ,.7
.....',,,.. s.
],. ' '
.,4
,. 's i '
.t.
-# h
, o.' , .. .' ., '3' '., . ' /.. ' ..,'._.' [. ,' \ "'A. . z,.',["
, , a
', , i.' t ...;
9 c . ~: r a y. $'; \ . , . - ; - , ' m.,.$ .h. . ' . .j
,' *g n. .w, ., .m ,' ;. ' ;, s y. ... ;.,. ,. .. p.= I s . ,- ; . a A,
- e. , . m . . . .3. _ g_ , e . c. c. . . . , . . , - .,: : .:
a
...s . .
.. s
.. .,.. ,.. - 4,. ... . ..
. . .l
., < . . . . ~ . .g
. . ,...... . ., o .- .. 3
.' ' ,; ;. ' ..V. l v ' ."' . . , ' &,. , ..+ , :; L ' . , : .: r, .c.. ;. . . l? - ., ^
.% ', ': * " ,A < t . ' ' ' ;' ( n' l .h , , '
. :. ,. : ' .0.. 'f'.,.] - - .'lu. '
{l?;? i ' t ' t::' V '. ',. ; L .vl ;. p.'. , . ;.:, ., ~' ., R n-.'. , '. ', e, ,,., L l,l . : ,:.l.p.'. ;? l:9 : .'.. l?'.:
' 'M'
O.4 s ,99}.Ql f. ?
. . ' ' .['l ... .J
;' V 1 l.). 4 ! C [,I .p,h ;,
' ' 'k4 ( :.] .i., [,; Y ;3:'$ '.., f.' / ', :" . y k ..;5. A 7 2,'l, ~ .
m, $n '.#..A.."..,.v w %.. -" ..u" :.. ...s t.
'w':
n'f/lV;f,. h., '.:g
-. s ,'. ,, : ,n$.;
. ; . ..;i':::
r .*rgl ' .. ' . . $ ; -
. n . . m'Nu,.:;&' . l .M .
p':. ,.,b)m.j. . e. . v ...+-
.. .:. ~ ,a.' ~. s- .
- v. . \ .;",v- . ,e .
..$ .? , Y.;. t.o .. .l ; . *.' v..
x ,.' ; . -lg. .-
... \ )
l' .. .; _
/
- i.. . . ,. , '_ ; , ,,
' .n . i .. ,, e'. .
*s ,f 6
. . " , - - . v. : ., . . . , . .. ' ...'...;
... _ > , .. p, . . .; X : w.
Y.' .~ , c. . , . .. .. :.. . ? t ; .. , g: a . ;. ...: . . :.. ; ., J. y. ; , v.y :.y ., ., _ ; , . . y . . .,. v . ..c.: k. .. .. : :.,;.;;; '.. .; .. L.L. . _9, ,i. . f .
,'..f1s ' ' ,' - U i!
i' '*, j'
'/. L
'.**. *;', r/. 6 /* 'f'*
. ..< .'-, h . ' , .'. i : '. . ,, .[~ ..'e' v
' ..a. , [ . : n'. . o.-;4 , , . /. ? ' ': e t '.
*- ,s ,
;. ' 1
..a , .. ~.
.; N';.' . -+, '. .. .' :g....'s',7....n,.'-..
. : . <.t... n: . \. .. ....
.,. . g @. .' . . . .
. . , ....e r.,',.. T
..,.1.. . .. . . . , , , ,;
. . , . .'t,. J, g ,. ... ,, .'.
. .Mg s, , 6 p .. .
.O . . . .a.*, * - ' .i,,., -
.' . ., ', ,~;.,. . . - . , . ' ,- _, ,
w 8
., , y'. t- ..i, .' s .~ ,_ y , . , , - .,
. 1,.
., .. * . .. ,< . j 'f 4 .
..,,,,6, 3 ,g , , ..,g
. .' ' I
, , . ' .', +[. ,y'%,.k ' .. , **
s
, , 9,
' * *.' 4.- , , .
*I '
..6[.. L .J '. .. i , . Q .; ;, ' $( "' . . * !. . , , [. _ y* _, , *
.l.. .f.._,,.,. '. l,.. . L;:.'..., .. .. . ., ... .G.. .y_: _...t,;.,* .. 3.;,e.. .O~ . ,..' .. . ,,... .,. p. :[.[ ; " .
3 . . , ,,,, 3 J 't
; , c. ' - y . - l < . . ; ., , ; .s f";-.,..,y 14 .:i,. - -: .] r,. ....;..:< . . . . - .e.. c. , ' 4 ,"f
-(..,,'.i,.,..,,,.'4 , T. . . , . , , ,
$. , . , , , ' ,. . , ,] .,' ,'1 ..4, : - ) k.r, ).4. 1 .
.,.( ,s c . . . . g es. , .
<(
.,A.
j.. . ' ,
- Y' ' ,- . N ',[. $p M' 3.C " .[; ..D ) . . '.. j' .., b. g' ' ,,. . , . . . . . , l l ', '
,.l ,.
.....,h.. . . ' d., ,,
. : .,. '..,s , ' . V . . ,w ,?. * .:...
4..> . ~, : ct, . . .- . J, .
... \~..
. - . , q s. ..
<..e.'.%.,n.....*
.. ' ; ,r '*..+: .
. . 's - , . . ' ' *,.**.. .
i _ ,. " }s _ ,,
, p ..-
a '.L, , u. % i = , . '
'. ' : . .' [') fi . , ' ' . .
.' - ' , . ': . . , . , I * ,',
.: s .:. .s y & M / :? r m')' m:;
'. e* ' '*
, , , ,- j- . *. - lf * % .. .' .. , , !.,'.,. ,
.7
>..w...c w.w. %:., N.. .w y.% a,.; :. %...:.: .. .. v. ,. . . . m. . a . ., ,~.. . u. m . s .. r . . g. e ,, W, 3< , .3
.% ;.' .' j..' .4l ,1... , .. .,., .;
m-. I
. a. . .a. . . . .~w. -
'... a..r cc.' -
?w n, /e ;
',,'..;,. s : * . ! g, a
<,,.,y
-f
. l/ j :, 5 ' , +.* , j' kt
. ;.g. $ ;c; ,'i, .. _ , .' .l_,, ...\. ,_; :. ., ,3. ' a .. . q: ,: . .y : . _ .c & .'. ;
,6 ,
, .,v..r, .a.
( :,_j.v.) ?.3:; _ f,., ' , 'h, &. %,; ;;. s. ff_ ,, .. G -
.. j:,4;j. ;
- p.{.
' ' / '.;,. . l? .3 .- _.;%; '
_L :;;h. C.q N ];,ihh, l :!.ll i' Q% W.;t .JY WW; %;p:Q.G-C. / v) 6 L f ,W. .y ;\ @,. TQ: ;;< .. # N& . % ^ f . l j_, :".l. 'n 9; ; . 3 }Q:: y ,;>.G.,1: ;;,:a. ' ' . : l .j . :- _
) . ' ~:
1_.;P.:...[QJ 9:@y.i, j .;
+- %.% '.* ; y,s ,@ ':
& ;,U*,Q y;.Q%.Khg&2 4 M
... . .. ,. x. . q,p. .&, %,:..;,; M Q.. b.n.s.'.;:;;;V. w ,'. 4;..b:f,:q p,:W:: . %. n- .c2: -~ G.l.5 K,j.:$..;, y l .' .
(.
, .; .' f. .
I. ,1, . , , ,, MM: .w MW[,.lyy
- py b.,. .4MelpMM@z.a..;n.dMMN#@@
R..mn.
- v ;<.4q f; .hf &.M
_e . 4 s f .^,l: ;::n#; & ;) f e.;;.}l. m.&5ge-;; x, v:9443.w.9~.h.,
&. ,? .
A .L.a , );c, y.m: 3 . w; ;,;.w. w. . > .m.s.. . .. ,
.n.
. l. .
Y u ;.h' .c,7{.?.:.'; yl. ]'- s .;y. :a y ;:;;,.) .w . w .:V. .n.,;.:;;;:j
.r .: ,p.c ;.,;." .
. m. _f.
g sA m%.. ,y.::s .
? L .,.
f 7_.;:% :[m :-..:.q_
- _y;
.v j,l- xQ [...v.g n:,a J> ;:y;u n., , . ,z. _g.g._;, ~ . . ;..:y,
...?;;. . . ;; ,. .; y . _ ; <- y;,.:..~ _~ c;ty. ,.. q
. . . . ~ - . - - - - . . - - . - - . - - . - . . - . - - - ~ - .
4 f i 4 ( lI i a l i lI I 4 1 i 4 i I I iI i I I l MSIV 2HV8204 RAIL /GKIRT FAILURE !3 ROOT CAUSE EVALUATION !E RCE 91425 1 l NOVEMBER 5,1991 !I i i } 5 7 4 a l i ' ._..__.-__ . _ . .___..._ _ _ _ _. _ _ _ _ _ _ _ _ _ _ . . - _ _ . . . _ - ~ . _ _ __
l MSIV 2HV8204 RAIL / SKIRT FAILURE ( ROOT CAUSE EVALUATION 91025 NOVEMBER 5,1991 4
/
- G. W. Gartfind, Root Causo/ISEG
/N d REVIEWED BY- d! /da EW Strom t
.M. A. Herschthal, STEC
&rla 7//4/Ak M. A, Wharton, Manager ,
Nuclear Engineering Design Organization Approvod by: b - < s C. Chlv, Mc;ioger f 941 Safety Engineering i
l RCE.91025 November 5.1991 TABLE OF CONTENTS I EXECUTIVE
SUMMARY
. . . ........ .............. .... ............... 1 I BACKGROUND . . . . . . . . . . . . . . ... ....... . ...... .... ........ . . 3 DESCRIPTION OF VALVE OPERATION . . . . . . . . ... ..... .. ........ . .... 7 FAILURE SCENARIG INVESTIGATION . . . . . . . . . . . . . . . . . . . . .............. .. 8 ROOT CAUSE IDENTIFICATION . . .... ... .... .................... . 13 CO R R EC T IVE A CTI O N S . . . , . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15 IDENTIFICATION OF OTHER SUSCEPTIBLE ITEMS . . . . . . . . . . . . . . . . . . . . . . . . .. 15 10CFR21 REPORTABILITY . ..,.... . ... ........ ... ... .. ... 16 APPENDIX 1: Analysis of Gato/ Segment Back Angic Friction Forces Possiblo
. Failuro Scenario 3 .. ...... .. .. .... .... ...... . A1 1 APPENDIX 2: Shoo Motion as il1 ravels Across Rail Ramp . ., .. ........ . A21 APPENDIX 3: Kinematics and Contact Stresses . . ........,.............A31 APPENDIX 4: Valvo Drawing and Photographs . . . ..... ... ......... .. . A41 APPENDIX 5: KALSI Engineering Analysis Report . . . . . . . . . . . ....... ..... A5-1 R
I I
l l j RCE-91-025 November 5,1991 !l l EXECUTIVE
SUMMARY
EVENLGESCRIPTIQN l On August 25,1991 during the Unit 2, Cyclo 6 refueling outago, SCE performod a scheduled boroscopic inspection of both Unit 2 Main Steam Isolation valves (MSIV's). The inspection of ( [ MSIV 2HV8205 did not show any apparent damago, but MSIV 2HV8204 cxhibited damago to the valvo disk locking mechanism. As a result of the boroscopic inspection, a decision was mado to disassemble and further inspect the valvo galo ano segment locking mechanism of valvo 2HV8204. g Valvo 2HV8204 was disassemtled and the fo!!owing damago was noted: g
- 1) Three broken guMo rail to seat skirt capscrews on t"o north gate sido rail and one b oken capsciow on tho scuth gato sido rail.
- 2) The gato sido seat skirt was sovemd at the location of tho fourth capscrow from the top.
- 3) Both gato sido rails had sovero galling on the top chamfors, l
- 4) Both lever lock arm shoes had sovero galling on the flat bottom surfaco.
I lg BORLCAVE 3 Design Deficiency: 'I The materials used for the sliding surfaces of the lover lock arm shoo and the seat skirt rail I ramps are a poor design choice. The lever lock arm shoes have areas of stellito hardfacing, but the most high!y loaded arcas have 17 4 PH to 17-4PH stainicss stool sliding surfaces. 17 4 PH stainless stoolis a very poor choice of material for sliding surfaces that have appreciablo contact stresses. The threshold for galling of 17-4 PH on 17-4 PH surfaces is approximately 2000 pounds per square inch (psi) which is wcll below the contact stressos betwoon the lover lock arm shoes and tho guido rail ramps. 1 I
s - RCE 91025 November 5,1W1
~
L s The material problem was compounded by a modification of the rail ramp angles that was mado during the Unit 2, mid cycle outago prior to the Cycle 0 refueling outago. The iamp angles were hand machined from 45 degrees to approximately 30 doGrecs from the vertical with no chango to the companion angio on the lover lock arm shoe. This mismatch of , angles resulted in an increase in the loads and stresses between the shocs and the ramps. CORBE.01LYK.ACI19J15 Field Change Notices (FCN's) wore issued to modify the Unit 2 MSIV lover lock arm shoes and the guido rail ramps during the present refuciing outage. The ramp angles woro revised and the shoes woro reconfigured to match a 30 degree samp angle. The contact surfaces of both the snoes and the rail ramps woro overlaid with stellito 68. The Unit 3 MSIV's will be modified to the now r. hoc / ramp configuration at the next refueling outago, I I I I I t l i I 1 l 2 I l
RCE 01-025 November 5,1991 I BACKGROUND Unit 3 Cgfo 4 MSIV inspfstica In May of 1988, during the SONGS 3, Cycle 4 Refueling Outage, Southern Cahfornia Edison was informed by Louisiana Power and Light (LP&L) that the two Main Steam isolation Valvos (MSIV's) installed at the Waterford 3 Nuclear Steam Station had suffered internal damago. The MSIV's currently installed at San Onofro Nuclear Generating Station (SONGS) Un.ts 2 and 3 are of the same type as damaged at LP&L. A beroscopic inspection of the SONGS Unit 3 MSIV's and an ir.terim Root Cause Analysis (Referenco 1) of the LP&L falluto and its applicabilay to San Onofro MSIV's woro initiated to dotermino the falluto mechaniam for the I LP&L valves and to determino ,f the SONGS MSIV's are subject to similar failures. The results of the boroscopk t/ispection of the Unit 3 MSIV's are tabulated below: I (1) One broken capscrew was found in the body cavity of MSIV 3HV8205. The two guide rails (ono of the two segment skirt rails and eno of the two gato skirt raits) that could be inspected woro in place. The rails were found firmly attached to the skirt plates. They did not exhibit any separation that was observed on the LP&L M31V's. Only four capscrews could be inspected in place and they woro found to exhibit no signs of elongation, deformation or looseness. The results of this it'spection were documented in NCR 2 2343 Rev0. I (2) No broken capscrews were found in the body cavity of 3HV8204. The two guido rails that could be inspected (again, one of the two segment skirt rails and one of the two gato skirt ralh) were also in placo. (3) The chamfers of both the upstream and downstream guido ra:Is for both valves were inspected for impact marks. Only the top chamfors on tho downstroam, or gate guide rai!s of 3HV 8205 were thought to have galling markt,. The top I edge of the chamfers seemed to have been rounded by impact and some metal had been rolled over. This metal roll over and e relatively high contact stress during impact wat thought to be responsibio for the observed galling marks. 3-I
I RCE 91-025 _ , _ _ _ Ncvember 5,1991 (4) The shoes for the Icycr lock arms for both valves were inspected for impact or galling j rnarks. No vir,ible galling marks nero observed on the surfaces inspected. The conclusion of the Referenco t Interirn Root Causo Analysis was that the San Onotto MSIV's stroke more slowly than the LP&L valves and are not subject to the guido rail dislodgment failuro as experienced at Waterford 3. A dynamic analysis of the impact energies of both the Watntford 3 and SONGS, Unit 2 arid Unit 3 lover lock arm assemblics was perf;rmed. The Wateriord 3 valves had boon operated with a stroke timo much faster j than the SONGS valves. When the impact energies of the two valvos woro compared, the Waterford 3 impact crorgy was an order of magnitudo higher than the SONGS MSIV's. Further analysis concluded that the impact energy of the SONGS valves was capable of l bccaking only two capscrews while the Watoriord 3 impact energy was enough to shear all nina c?pscrews ai once. The conclusion of this study was that the SONGS rails could not fail from guido rail dislodgment due to shca'ing of all of the c:9 screws. No immediate action was required; however, as a long term measure an inspection plan for I all SONOS MSIV's was recommended. A study to evaluato revisions to the chamfor anglo on the rai!S from 45 to 30 degrcos and the addition of st3llito overlays on the contact curiacos of the lever lock arm sbocs and the guido rails was also recommendoi The addition of an
< r:fico piate in the commen dump piping of both valves wa$ also recommenced and imp'emented.
At the urging of the Nuclear Pogulatoly Commission, MSIV 3HV8205 was disassembled and inspected prior to the cr$d of the outage. The rest lts of the inspection were as follo.vs and are documented in NCR 3 2046 and NCR 3 2056:
'Thrco severed capscrews and one loose capscrew were found at the bottom of 1.
one of the segment side ' ails.
- 2. Too bottom surface of the segment sido rail chamfers woro found 9 bo galled. No ga: ling was observed on the gate sido rail chamfors, but some wear was noted. The boroscopic observation of galling was in error,
- 3. The lover lock arm shoes woro galled wh;ro they Contact ,the seament sido rails.
i 4
4 RCE.91025 November b.1991 All non conforming conditions were returned to acceptable configurations and the sheated capscrent were rep' aced. The upper ramps on the gato sido guide rails were a;50 reAorked at this timo per the disoosition of NCR 3 204G. The sharp cornera were remosed from both - ends of the ramps to a 1/4 inch nominal radius and the ramps were m thined/ stone ground to an angle of 45 dogicos 4 0 5 degrecs. The four damaged capscrews were replaced with l screws made of 17 4 PH material for increased stiongth. Reference 2, inspection Results of SONGS 3MSIV8205, determined that the root cause of the failuro of the gu',de rail capscrows was a self lim. ting galhng p!Ocess that builds up and removes the sharp comers of the guido rails. Once the sharp corners are removed, the I contact area between thG shoes and rails increar.cs. With the increase in area, the loading is reduced, reshiting in lower strelses and discontinuation of the gathng and adhes:on process. On August 9,1988, the NRC tt, sued information Notico 88 59 which discussed the main steam iso!ation valvo guido rail f a!!uro at Waterford, Unit 3. The correct;vo actions noted in the report included the SONGS corrective actions trom Referenco 1 along with adoitional items from Waterford 3. Theco recommendations included changing the shoo and f ait angles and adding stelhte hardfacing to the wear surfaces, ventying bolt abgnments and performing nondestructivo testing of all new bolts, increasing the valvo stroke timo and performMg periodic boroscopic inspections of the valvo interna!s. All corrective actions, with the exception of vonfication of bolt alignment and additional g I nondestructivo tetting of bolts, are being taken at SONGS. Sinco the valvo stroke timo at SONGS is much slower that at Waterford, the SONGS loadings are lower and tho requirements for the bolting inspections were not considered necessary. UML2JdSN_LNSffGION MSIV 2HV8204 was disassembled and inspected on January 19,1989 during a Unit 2, mid-cycle 4 outage. The bottom chamfors on both segment skirt guido rails woro found to be slightly ga"ed The top surfaces of the shoes were also slightly galled All capscrews were found to be in place. The noted cond;tions ware documented on NCR 2 2G12. All nonconforming conditions were repaired and all of the top guido rail chamfors were I machined /stono ground to 45 degrees 4 0/10 degrecs (approximate). The sharp corners I 5-
i 1 RCE 91025 November 5,1991
-woro also rounded to a t/4 inch riominal radius por the disposition 01.',"'R 2 2Gt2.
( URLT 2_01CI4SJAS1V INSPECIloti During tho SONGS 2, Cyclo G refueling outago an internal inspection of Main Steam Isolation Valvo 2HV8204 was portorrned with tho uso of a boroscopo. The inspection revoated damago
. to the ralla and a fracturo in the material of the gato sido skirt. Two of the countersunk machino scrows that hold the ralls to the skirts woro observed to bo protruding from the rail and appeared to either be backed out of the threadod holos or clongated duo to high loading. One of the rails was observed to havo a gap of approximately 1/8 inch betwoon what should be the contact surfaco of the rail and the shirt in the area of the top throo capscrows. The skirt plato materialin the vicinity of tho third capscrow appeared to be fractured,loaving a 1/2 inch gap in the skirt. Tho skirt and rail below this fracturo did not show any indication of loosonoss or damago. } MSIV 2HV8204 was disassembled and further inspected to dolormino the full extont of damage to the locking mechanism. A summary of the damago is as follows:
- 1. The gato sido rail had four broken capscrows. Throo of the capscrows woro from the north sido rail and one was from the south sido. The north sido capscrews woro the throo top scrows and the south sido capscrow was the top ,
scrow on its $1do.
- 2. The gato sido skirt plato was sovered at the location of the fourth from tho top cap screw on the north sido. Tho bottom portion of the plato had boon pushed down approximately 3/10 of an inch.
- 3. - Both top rail chamfors had metallic deposits and savoro galling.
4 Both lover lock arm shoes had galled areas on tho flat bottom surface at the transition of the radiused portion of the shoo. MSIV 2HV8205 was also inspected by boroscopo during the refueling outago. Tho guido rail ramps appeared to be discolored, but it could not bo dotorminod if galling had occurred. The valve was then disassembled and no galling of the skirt or guidoralls was found. I I RCE 91025 _ , November 5,1991 l DESCRIPTION OF VALVE OPERATION The Main Stcam isolation Valves are Cooper indus,tries, WKM flow Control Division, Power Seal gato valves. The valves are 40X30X40 900 Pound Class double disk gato valvos with pneumatic / hydraulic valve operators. The valvo disk is a two pioco assembly that consists of ;
'I a gato that is attached to the vatvo stem and a frco floating segment. The gato has upper and lower ramps that are followed by the segment during valvo operation to expand tho l assembly and provido positivo scaling of both the upsticam and downstream valvo soais at both ends of valvo travel. This action results in the valvo cavity being Isolated from the flow stream when tho valvo is in either the full open or full closed position. The free floa'ing
- I segment is articulated by a lover lock assembly that prevents premature wedging of the gato during mid travel. The lover lock assembly is guided during its travel by rails that ato Atached to skirts that fit over the va!vo seats. At both ends of the vaho stroke, the lever lock assembly disengages from the rails to allow the wedging action of the gato and segment to scal the valvo disk to the valvo seats.
The stem position of the MSIV's is controlled by an hydraulic system which excris lifting pressure on the bottom of a 24 inch diameter piston acttator. Nitrogen, in a closed space above the piston, provides the primary closing force along with the weight of the piston /gato/stom assembly to close the valvo. Dual pilot operated solenoid dump valvos
- poworcd from separato Class 1E DC poAct sourcos open and dump hydraulic oil from the r
bottom of the actuator which allows oil to flow throu0h the valvo timing onfico and tho ( nitrogen chargo moves the valves to the closed position. The design stroko timo for tho San f Onofro MSIV's was initially between 4 and 5 seconds. A design change to add flow rostricting orificos to slow the valvo closing stroke timo Oc ,vn to approximately 7 seconds was l initiated as a result of a failure of the MSIV's at Louisiana Powcr and I.ight, Waterford 3 Station. ? I 7 e l
s e RCE 91-025 November 5,1991 r - L FAILURE SCO!ARIO IN\n STIGATION [ o Possiblo Scenario 1 (Most Likely) Wrong Material Used for Sliding Contact Areas The baso material of both the lovo* lock arm shoos and the rails is 17 4 PH stainless stool. The shoes are overlaid with sicilito hardtacing on the radiusod portion of the shoo that contacts tho intido of the guido rails. Tho unhardfaced flat bottom of the shoe makes initial contact with the top of the rail as tho valvo begins closing. The shoo then rotates about the uppet outboard edgo of tho g ramp until the back surfaco of the shoo contacts the stop on the lover lock arm l (the stop forms a 45 degroo angle to tho arm contortino). Tho shoo then slidos across the 17 4 PH guido rail ramp untilit clears the ramp and ontors tho Spaco betwoon the rails. (Seoo gures 1 and 2.of Appendix 2). If the inglo of l the shoo matches the 45 degree angle of the guido rail ramp, the o is a largo contact area, as in figure 1. However, if the angles are Ceront, as in figuro 2, the contcet arca is a narrow lino. The 17 4 PH material, when sliding on 17-4 PH material, has a galling threshold stress of 2000 psi (Reference 4). This stress value will be excooded when the shoo is in lino contact with the ramp. SUDD211}D9_EvidCIlG Inspection of the lover lock arm shoos and the rails showed that galling had occurred on the 17 4 PH surfaces of both the shoes and the rails. The galling occurred at the top of the guido rail ramp. When the lover lock arm shoo is contacting this area, the shoo is in lino contact with the ramp and the contact I stressos aro extremely high. l The rails of valvo PHV8204 had been modified from a 45 degree angle to a 30 degroo from the vertical angle and the sharp transition had been smoothed to a 1/4 inch radius. Appendix 3, figuro 2, calculated the contact stressos under these conditions to be in excess of 123,600 psi. This stress levelis wellin
-C-l 1
s RCE 91025 November 5,1991 r L excess of the 2000 psi threshold stress for galling of 17 4 PH material, r [hWE09.fXidfa 2 ( Valvos 2HV8205 and 3HV8004 havo not shown any galling damago to dato. The loading between tho shoo and the ramp durinn vaWe closing with { matching ramp angles can vary from betwcon 514 pounds and 2961 pounds p depending on the value of the cuofficient of friction (500 figuro 1, Appendix 3). L The contact stress for those loadings while the shoo is in the unlocked position at the beginning of the stroke,is betwoon 987 and 3508 psi. Sinco tho threshold stress for galling is 2000 psi, there is some probability of galling. If ( the coefficlont of friction is lower than about .30 or the roatorial has somnwhat higher resistance to galling than the test threshold value, galling will not occur. o Possibio scenario 2 (Contr.buting Factor) inco.roct Ramp Angle ( The rail ramps on 2HV8204 woro reworked frcm a 45 degreo angte to a 30 degroo to the vottical engte prior to the Cycle 5 refueling outage. No modifications woro made to the shoes or lover lock arms to cause the shoe angio to match the ramp ar.glo at this t!mo. This mismatch resulted in lino contact betwoon the shoes and the ramp during unlocking of the lover lock ( assembly. The lino contact increased the contact stress at the transition betwoon tho _ I upper, flat rail surface and the rail ramp. Attor initiation of galling at the transition point, the friction forces increased which in turn increased the contact forces and stressos and the galled arca moved down the ramp. 9 m up
l il RCE 91-025 November 5,1991 l B e m i! a fxdenco Appendix 2, Figuro 2 shows the path of the shoo across the rail with the rails I cut back to a 30 degreo angle. Once the shoo has rotated to position 2, lino contact is maintained for the remainder of the travel. The radius at the transition point of the upper surface of the ramp to the railis the most highly stressed point in the travel. Onco tho shoo has locked in tho 45 degroo position in the lever lock arm, the shoo will slido across the radius and galling I- is extremoly liko'y. Appendix 3, Figuro 2 calculated the contact stressos for this condition as over 123,000 psi compared to a threshold galling stress of 2000 psi for the 17 4 PH material. l En!ulinnfaMtaca i Valvos 2HV8205 and 3HV8204 have not exporlonced similar damago. Valvo 3HVS205 had damage to the lower rails that was attributed to a combination of oiffering rail contant (the north arm w&; cather shorter than tho south or the valvo skirt was not squarely installed). This valvo has not shown I any damago to the upper rails. When the valvo was reworked to removo galling and wear on the lower rail ramps, tho (upper) gato sido rail ramps woro inspented and showed signs of wonr. The surfaces woro stono ground to remove the wear marks, but thero is no reason to believo that the ramp angle was appreciably changed from a 45 degroo angio. I o Possibio Failuto Scenario 3 (Less Likely) (Appendix 1) Gato/Sogment Back Angle Friction Excessivo if the friction coefficient of the tiiding surfaces betwoon the back angles of the gato and segment assembly are appreciably higher than the friction coefficients I between the seat surfaces, the gate /sogment assembly can ' stick
- together and travel down the ramps locked together. Under theso conditions, the forco roquired to unlock the assembly would be increased by the frictional forces built up during back seating of the valvo. This increased force would increaso 10-
I RCE 91025 November 5,1991 the contact stresses betwoon the lover lock arm shoes and the guido rail ramps. The increased stresses would, in turn, result in a higher probability of galling and subsequent valvo damage. , hDRQIllD91ddCD.CQ The calculation in Appendix 1 Indicates that tho segment can fail to unlock from the gato and maintain the clamping forco of backsoating for friction values on the back seat about 2 to 3 times the valvo seat friction. With the clamping forco holding the segment ar..f gato together, there would bo substantially more valvo seat frictional force than if tho wolght of the segment had allowed
- the assembly to unlock. The increased seat frictional forces would increase the shoe to rail contact stressos and the probability of galling.
BCh!1Lnglylgang ,e . f ' inspection of the valvo coat and the back engles of the gato and segment assembly did not indicato any scuffing, scratching or galling. A coeffielent of friction of betwoon .4 and .8 would requiro that somo damago had occurrod. o Possibio Failuto Scenario 4 (l.oss Ukoty) Binding in Lever Lock Arm Journal if galling or binding occurred on the bearing surface of the lover lock arm journal, the gato and segment can travel together until the lover lock arm shoes impact the guido rails. Under this condition, the forco required to unlock the assembly will be increased by the frictional forces in the bushing. This incroated force would increase the contact stressos betwoon the shoos and guido rail ramps, This,in tum, would result in a higher probability of galling. hpportina Evidence Nono. 11
M ~ RCE 91025 November 5,1991
~
Ec!gting Evidenra Manual actuation of the lover lock arms of MSIV 2HV8204 did not indicato any unusual drag or binding that would be associated with journal galling or other 3 material distress. The kinomatic study in appendix 3 shows that galling of the guido rail ramps can result from stresses found during normal valvo opc*ation without addalonal loadings. I I I I - 1 I I I I l 12-
RCE 91025 November 5,1991 ROOT CAUSE IDENTIFICATION ( The root cause of the 2HV8204 failuro was the use of 17-4 PH material that is tubject to galling for the sli6ng surfaces betacon the lover lock arm shocs and the guido rail ramps. This inadequacy was compounded by a revision to the guido rail angle from 45 degrees to 30 degrees without a companion chango to the lever lock arm snoe. The change in the ramp angle resuHed in an increase in the contact stress betwcon the sutiacos by causing lino contact between the lover lock arm shoo and ramp at the beginning of the slido down the ramp. Tho 45 degroo tamp anglo prior to the chango allowed surfaco contact with a locked shoo while the segment was being lifted. Damage to the lover lock mechanism can occur when the valvo is in either the full opun or full closed position and tho valvo is actuated. When the valvo moves from the scaled position toward midstroke, the icvor arm shoos contact the guido rails and move the floating f segment up and away from the valvo scat, thus collapsing the gato assembly. The contact forces betwcon the lever arm shoo and 'ho guido rails,if excessivo, can result in impact damago or galling of the shoo and rail surfaces and possible failuto of the rails, mounting screws, or skirts. Appendix 2, Figure 1 illustrates the normal travel path of the lover lock arm shoe across the rail ramps as the valvo closos. The ramp angle is 45 degrees to vertical and the icycr lock arm has a travel stop that limits rotation of the shoo to 45 degroos from vertical. When the shoo first contacts the ramp with its pivot z' punt 1, the shoo is nrobably in a horizontal position since the shoc pivot is above the centor of gravity of the shoc and gravity will pull the shoo to this position. As the valvo closos, the shoo will pivot about the outboard, upper, odge of the ramp and the shoo pivot centerline will move toward point 4. Points 2 and 3 illustrate the pat'i assuming that the shoc contacts tha ramp h a 45 oegree from horizontal position which could occur due to fluid dynamic forces or if the pin frictbn is high. Tha taco of the shoc will slido across the ramp, with full contact, unt:1 point 5 is reached. At point 5 the back surface of the shoo WW contact ino stop in the lover lock arm and relative rotation between the shoo and arm will cease. As tho valve travels further, contact betwcon the shoo and ramp decrease toward lino contact, f.rst on the outboard upper edge of the ramp and then as the outboard edge of the shoo travcis past this point, on the lamp itscif, D 13-C u__ _ -___ _ _ _ _ _ _ _ . _ . _ _ _ _ _ _ _ _ _ _ _ _ . _ _ _ _ _ . _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
I i
" RCE 91-025 November 5,1991 Appendix 2, Figure 2 illustrates the lover lock arm shoo travel path during valve closing with the rails reworked to a 30 degrecs from "crtical angle. Noto that nothing was dono to the lever lock arm cr the shoo to change the locking point at a 45 degree angle. The shoe will contact the top surface of the rail, point 1, as before and will rotato about the autocard upper edge of iho rail ramp until pc.nt 2 is reached and the shoe locks in the lever lock arm, At this point, the shoe will start to slido on a narrow line of contact with the cdge of the rail ramp.
The shoo will continuo sliding on this line of contact, with increasing load until the edge of the 3 shoe reaches the ramp. At this point the shoe material changes from 17 4 PH to a stellito ) overlay. The shoc continues down the ramp until it cicars the rail at point 3. ,
- The material of the rail ramp and the bottom of the shoo are both 17 4 PH stainless stool. ;
This material combination is particularity susceptibio to galling with a threshold galling stress as low as 2000 psi (Reference 4). The force betwoon the shoe and the ramp is a function of the geometry, the weight of the segment and friction. As tho friction coctficient between tho shoo and the ramp increases, the resultant fo'ce vector mcycs closer to the conter of rotation g E of the lever lock arm and Inquires more and more force to slide the shoe down the ramp. Onco initial gal!ing has occurred, the roughened surface increases the friction and the resultant force, which causes an increase in ccitact stress and increased galling. After sevo al galling cycles, the forces can get high enough to sever the set screws attaching tho l' rail to the skirt plate and skirt plate fracturo can occur. I I + I I L RCE 91025 November 5,1991 i L
, CORRECTIVE ACTIONG
[ The Unit 2 MSIV's have been modified to eliminato the possibility of galling of the guido rail ramp to lover lock arm shoo interfaces. Tho followirp changes have been made:
- 1. The guido rail ramps have been reconfigured to a 30 degroo from the vortical angle,
- 2. The guido rail ramps have been hardfaced with Stellito 6B.
{
- 3. The Lever lock arm shoes have been reconfigured to match the 30
[ degrco to vertical anglo of the guido rail ramps.
- 4. The lever lock arm shoes have boon hardfaced with Stellite 6B on all guide rail contact surfaces.
The Unit 3 MSIV's will be modified to the now shoc/ rail configuration during the next refueling oatage. IDENTIFICATION OF OTHER SUSCEPTIBLE ITEMS All WKM Pow R-Scal gate valves use a similar mechanism for expanding the valvo disks into the valve seats. As such, all WKM Pow-I Scal valvos should be examined. Engir.cering will g review the designs of all WKM valves ust 1in safety related applications and determine if they l are subject to similar fal!ure modes. Appr;priate action will be taken it design changes are required. I I E _._ _ . _ _ . _ _ _ . _ __ _ _ . _ _ _ _ . . _ _ _ _ _ _ _ _ _ _ _ _ . __ _
4 - RCE 91-025 Novt.mber 5,1991 10CFR21 REPORTABILITY L The generic problem of lever lock arm / guide rail ramp galling was prev _ reported to the NRC by Southern Cahfornia Edison in July of 1988. The NRC issued Information Notico 88-s 59 in August of 1988 to inform the industry of the potential probicm. REFERENCES
~
- 1. INTERIM ROOT CAUSE ANALYSIS Por The Guido Rail /Lov R Lock Interaction Problem For MSIVs SONGS 61988
- 2. Inspection Results Of SONGS 3MSIV 8205 8-24-88
- 3. Advanced Mechanics of Materials, Boresi and Sidebottom
- 4. Armco product Data Bulletin No S-45 I
I l l l l l l
.,8 l
l
W RCE 91025 November 5,1991 I I I I I T'. W I APPENDIX 1 Analysis of Gate / Segment Back Angle Friction Forces Possible Failure Scenario 3 I I I I I I x, ., I
-A
RCE 91025 November 5,1991 ANALYSIS OF GATE / SEGMENT FORCES r b Th's analysis examiras the gate and segment forces with respect to geometry and friction. When the valvo is backscated in the open position, the disk segment contacts a stop on the valvo connet and the gate continues its upward travel. As the gato continues to move, the back angles of the gate segment assembly cause the disk to expand into the seats. The clamping forco betwoon the seats is a function of the applied stem force, geometry and the friction coefficients. l When the motion of the gato is reversed and the disk drives down, there is a possibility that the clamping I forces will not relieve. In this case, the clamping force will increase the frictional dra; on the segment and will add to the forces required to unlock the mechanism. This increased force can add to the contact stresses betwoon the lever lock arm shoc and the ramp on the skirt rails. FORCES WITH STEM DRIVING UP (IGNORING WElGHT OF GATE AND SEGMENT) l P huI ,
\
f = M;N 1 l u/ P e j-t =hP 1
\r ~
l l En " O Ev = 0 P = N COS 0 - p, N S/N 0 F = p, P + 2 N COS 0 l P N ( COS 0 - p2 SIN 0) COMBINING EOUATIONS YlELDS I N - P/(COS 0 -p2S/N 0 ) F - p, P 4 (4 2P COS 0 )/(COS 0 - p, S/N 0) F = fl( 3 + H2 COS 0)/(COS 0 - p2 SIN 0)) l l A t .2
~ RCE 91-025 November 5,1991 FORCES WITH STEM DRIVING DOWN (IGNORING WElGHT OF SEGMENT)
\ \
t = .O 'N r t = As P L _ f-
- P I f E
I H I P - N COS 0 + p, N SIN 0 N - P/(COS 0 + pg SIN 0) Ev
- O .
p, P + N S/N 0 - p2 N COS 0 COMBINING EOUATIONS YIELDS 9: " (4 COS 0 + S/N 0)/(COS 0 - p, S/N 0) 1 BASIC PROGRAM FOR FRICTION VALUES 10 CLS 30 LPRINT 40 LPRINT 50 LPRINT
' 60 LPRINT
- SEAT FRICTION VS. SEGMENT FRICTION FOR SEGMENT NOT UNLOADING DURING TRAVEL' 70 LPRINT 80 LPRINT
- SEAT FRICTION U(1)",' SEGMENT FRICTION U(2)'
90 LPRINT I 100 A=15*3.1417/180 110 FOR U1 =.01 TO 9.000001E-02 STEP .01 112 GOSUB 120 I 115 NEXT 116 GOTO 150U1 120 U2 = (U1 *COS(A) + Sid(A))/(COS(A) U1
- SIN (A))
.# # .# # *;(U1),(U2)
I 130 LPRINT USING ' 140 RETURN 150 FOR U1 = .1 TO .5 STEP .05 g 151 GOSUB 120 3 155 NEXT U1 160 LPRINT::- I A13
I I RCE 91-025 November 5,1991 BASIC PROGRAM CONTINUED 170 LPRINT
- BASED ON EOUATION U1*COS (A) 4 SIN (A)
U2= - -- - - - COS A - (U1)(SIN A) 180 LPRINT
- WHERE: A = BACK ANGLE I 181 LPRINT
- 182 LPRINT
- 190 END U1 = COEFFICIENT OF FR!CTION BETWEEN SEATS U2 = COEFFICIENT OF FRICTION BEMEEN G ATE AND SEGMENT RESULTS OF BASIC PROGRAM SEAT FRICTION VS, SEGMENT FRICTION I FOR SEGMENT NOT UNLOADING DURING TRAVEL SEGMENT FRICTION U(2)
I SEAT FRICTION U(1)
.01
.02
.28
.29
.03 .30
.04 .31 .
.05 .32
.06 .33
.07 .34
.08 .36
.09 .37 I .10
.15
.38
.44 49
.20
.25 .56 I_ .30 .62
.35 .68 I .40
.45
.75
.82 With a seat coefficient of friction of .20 (which could be reasonably expected) the back angle friction would have to bc .49 in order for the gato and segment to fail to unlock due to seat I friction while the stem is driving down. Visualinspection of the back angles of the gate /sogment assembly did not show any signs of galling, roughness or damago. Self locking of the gate / segment is, thereforo, highly unlikely.
I I A14 I
rq' \ RCE 91025 November 5,1991 l-l l l l APPENDIX 2 Shoo Motion as it Travels Across Rail Ramp l I I I I I I l .2. , I l M
I RCE-91025 November 5,1991 I I a I 'N
/ 'N 6 g
I O y' N/ - I _ ,O
- O
'-twar-
.7 N -tecceo m ;
I \ s 4%, '~
/ -(
?
\s I im..
l l $
% ,,/ /' ..-_l I :
FIGURE 1 (ROTATED 90 DEGREES)
"AS DESIGNE9" (45 DEGREE) RAll RAMP CONFIGURATION
- 1) INITIAL CONTACT (1) Assumes that shoe is in a horizontal position when it contacts the top of the rail.
- 2) INITIAL CONTACT (2) Assumes that choe is rotated to a 45 de. gree angle when contact occurs. Note that contact is a line between the rail corner and the flat surface of the shoe.
3 LOCKED SHOE (5) Shce back anglo contacts the stop in lever lock arm and shoe locks in I) position. Sliding starts. I I ^'
RCE-91025 November 5,1991 { E N r N l - \ / e , i.4r r Z 0 C K ff M *d E s j3 **; + sp _- s < .
~
t % ['. 5 z ,: h ' n,, f nit.
, )
.i
\
ll
-[ - . ]
i FIGURE 2 j REVISED (30 DEGREE) RAll RAMP CONFIGURATION 5 (ROTATED 90 DEGREES)
- 1) INITIAL CONTACT (1) Assumes that shoe is in a horizontal position when it contacts the top of the rail.
- 2) INITIAL CONTACT (2) Assumes that shoe is rotated to a 45 degrco angle when contact occurs. Note that contact is a line between the rail corner and the flat swface of the shoe.
- 3) END OF RAMP (3) Once the shoo passes this point it can rotate to a horizontal position and
'I will remain between the rails.
.I A2-3
RCE 91-025 November 5,1991 6 I APPENDIX 3 Kinematics and Contact Stresses A31
j RCE 91-025 __ November 5,1991 l KINEMATICS AND CONTACT STRESSES f This appendix contains scale drawing layouts of the lever lock arms and rails. These layouts define the motion of the arm and shoo as the valvo moves from the full open position and the lever lock arm shoes contact and move along the rail ramps. Once the motion is determined, it is possible to defino scale vectors for forces and friction. Theso vectors can then be used to calculate the forces and moments that are imparted to the shoe and from theso forcos, the cuntact stresses. Figure 1 is the kinematic diagram for the valve *as designed' condition. This diagram has a 45 degree rail ramp angle from the vertical and assumes that the transition from the flat top of the guide rail to the rail ramp is a sharp edge. _ Figuro 2 is the kinematic diagram for the *as found' condition of valvo 2HV8204. This diagram has a 30 degree rail ramp angle from the vertical and has a 1/4 inch radius betwoon the top of the rail and the ramp. Both figures contain force vectors for a coefficient of friction of both .20 and .40. These friction coefficients were chosen as fair cstimates of the upper and lower limits for normal friction of 17 4 PH stainless steel sliding on 17 4 PH stainless steel. The force vectors were drawn and used to scale the moment arms between the vectors and the lever lock arm pin. The momeat arms were then used in the calculation of the forces on the ramps. All dimensions used vsere obtained from measurements taken from valvo 2HV8204. The segment weight of 2300 pounds was measured using a dynamometer. I CALCULATION OF CONTACT FORCES The contact forces of the lever lock arm shoc on the guido rail ramp will be estimated by determining the force at the shoe that is required to lift the segment weight of 2300 pounds. The valve stem was assumed to move at a constant velocity of 7 inches por second end all j acceleration of the mass was assumed to occur while one lever lock arm shoe was in contact l with the guide rail ramn. It was further cssumed that the motion during the acceleration was simplo, undamped, I harmonic motion that could be modeled as a spring / mass system. The natural frequency of the systen was calculated usirr o spring constant of the lever lock arm in bending and the mass of the segment. The r* s of the arm, shoe and pins woro ignored as being too small to affect the results of the analysis. I I A,2 l 1
4 RCE 91025 November 5,1991 r . The spring constant of the lover lock arm was determined by modeling the arm in the ANSYS finito clomont program. The program determined that the deflection of the arm due to the l 2300 pound weight of the segment was .00061 inch. This value was used to calculato the natural frequency of the spring mass system as follows:
/ 1/2n/ Afm
{ / - 1/2tv' afm6
/ 1/2n/dl
/ = 3.13//5
[ / 3.13/[056BT
/ = 126 cps,P .008 soc The natural period of the arm is .008 seconds. This can be used to calculate the average accoloration assuming that the deflection occurs in 1/4 period and the initial velocity of the arm is at .823 inches per second, constant acceleration ace - 2(6 -(-)Vo t)/t2 acc = 2(.00061 + 823(.002))/(.002)2 a aco - 1128 inches /sec g - 2.9 l simplo harmonic motion i acc = Vo u I acc .823 (/554//.0005T) g - 1.7 , The acceleration due to the impact of the arm with the guido rail ramp was calculated to bo j 2.9 g'1 This can be considorod an upper bound value for the accoloration loading.
j Assuming simplo harmonic motion, the ma.wimum accoloration is 1.7 g's. 1 i
)
)
) .3 f
s RCE 91-025 November 5,1991
. . . . . - . . . ,t~w .. . ..a. t. t . . , _. :
;ip'~~
i Fi i t- s ,,, m / l n h%
!!ht nI!I u.. .. m t 111
!! l' \ '
I ,
!. '?" 'I -
ri 3
\
I \ I \l \
/
f{ j o, oce. -. e, ... ca>
;, ; m.c.woo .i.i c n.mu .= .c. e + = w. . .c
!,. 1, i..
,, Se# A TN1 69 8 GW87 cr 23W DOJte A80 TAG LE** 9 8+ 08 9 El 1 **M S gi
.. ......., w . .. .
. . s w . n , . . .. . soi , mee.
i ;3 , ; sn>,s 's . . re o, a ji i 'r ... cu iss. e
- s, c .=,o,a um. .c eu, ecua j;
- /,;N . . , . vo x s u , . , n m,<.
flg k*4 * *en psict eo crer os to I ll i-
) ,
e . i n a e . n i . . $ . 314 me, asseen e.s o LoAsium f {/ i pmu.$gN $94 e 94 40 $UJ45 %
,1 .
ass.w.a ,. x a to. e im F*mM e s.10 M tia . 01) NJ.CS
"* ..y I.
i
. 93 i
; - u _it wicew .
see c, m . c, ,, ..s..
. ~,,,.,. . ..
i e k (C#d ACT AIEA . $H f /s i 4/g! i meict sr ru . nv . ... . in ,= i 3 f4 } (\ comace ar.eu . was,, . e +4 . nsas psi
/<
-, 'N I .e.
. . .. _v
_ . ,, w _ m,cic, p a.ssa_ .=
., . =_ _. ., A. c.
I
%, E 'i
- w. ., . s -n i
l.. - s.- F1GURE 1
-15 Degreo 4rv
.34 l
l i _______--_-_-_____- _-
1
'y RCE 91-025 November 5,1991 I I
1 8 1 i t i . .............,..n..., /
......._.7* .2:: .
j 3 s\x 8
;- N e pt id 5.* men
, l+ *
); g PC54 & 5.r
/\
e dWf 9 ff LV 491. .hP4?9 Aa0 & 18Jt Lt.C?ot LP 3 SS s u et'$ lE tin n;)
/ mwt, <.
. A...m. . .-. t n m,. . . x o B . . . . .. . . . . . . . . . . .
a
; ( e, e n a ct io cc .c . os .o t . . .mx...,..,,.. :
lq aj ji .sm.fa.e,u.o= , ii i e mi in . a e s ei. . etav meos jl l a ht M3 1.10 0 LCmDl%)
= if ii 4 , c. e, . $ == s,. . ms m ece hI I I. / o<se em ict ioee crue rp so i
fl ! t !,
=.,m.. u , .c in . 3, .u.a.
~, s.shMisa; $ 9 g LLapeada f
, mu.
l; e i, - ic. . . v , a e . 4
,l 93 . .m s . n
.,i; e uno . s?,e umo. a i.,o .ast an>i
- I ll '
} uw t, o w s: s (s. , o...re o w e c. . t . mso
; , , k A * . ,1.- d. },',
, , f. wl\ / . . . ,
, l ;l . ; 7 Me
- 2 ( L..i.' 4.,) . . m . d '
l ,t ,a 4 1 sussise c. sesta49 t s eeseen
' . < e .* re .n . , e
,! j 9 .o., .u.,. .
'/,,/
'I , ; a *hl . MA
- j i; i E
9
>jI . .p.ms s o.3.,4d 17 m
. . p3. 94,7m,c
( , g' v y, ,<y ;3 ... . . .>,. m . . 4.
.. . , , , , , , . 4. ' - --- -yv -t -~ ! .y . , , , , , , , , .v . . m ...a .
Y 4 4, . 3 ! wn v ,asmo vs esi r nwmass.ao emract sem os is aci. w..<a T*. N. 4 e nras as con wti e ce is. p.4
. j
.w. s s2 I :,
; s, t s i ,t i
s[
% a. 5 t.+e M meep
,g
.e_ ,
i
'*\, .
s i y i i
\
\ s\F
- g. V t C#l 3 30 DEGM f AAup s
I I "s
- I
/7
\ AA 4 s f*
^
Nj$*** 4 4 s* p+ / p 9/ pcf o+ / woo
- p o-s
,* ./' - -- - - - '-/-'//
RCE 91-025 November 5,1991 {E ANSYS FINITE FLEMENT PROGRAM OUTPUT The following table is a corg of the highest values of the icvor lock arm deflection output that was used for calculation of the natural frequency of the arm. The maximum value at point 843 was used in the calculation. The completo results of the analys:s are on fi o with G. -I Gartland. THE FOLLOWING X,Y,Z DISPLACEMENTS ARE IN GLOBAL COORDINATES lI
"*" POST 1 NODAL DISPLACEMENT LISilNG "*"
LOAD STEP 1 ITERATION = 1 SECTION = 1 TIME = 0.00000E + 00 LOAD CASE = 1 THE FOLLOWING X,Y,Z DISPLACEMENTS ARE IN GLOBAL COORDINATES
- NODE UX UY UZ 843 0.70484475E 05 -0.61042662E 03 0.49095384E-06 848 0.70921406E-05 -0.37221059E-03 0.49736207E-06 j 838 0.72524740E-05 -0.37205007E-03 0.48804520E-06
-B 738 0.21954611E-03 0.29013213E-03 0.15824420E-03 736 0.21971333E-03 0.28993812E-03 0.49808936E-05 ; I 729 0.21632150E-03 -0.27437074E 03 0.21143691E-05 687 0.21681921E-03 -0.27378469E-03 0.12844332E-05 739 0.21749919E-03 -0.27278684E-03 0.20477262E 05 g 802 -0.19765024E-03 -0.28744497E-03 0.28863970E-05 3 735 0.21783288E-03 -0.27240034E-03 013507744E-05 804 -0.19741483E-03 0.28729217E 03 -0.3F256814E-05
- en 845 0.15357781E-03 -0.30719584E-03 0,12627243E'05 s
g MAXIMUMS NODE 762 843 695 VALUE 0.22038163E 03 0.61042662E-03 0.20153209E 04
" " RuN COueteTeD " " ce= 5942200 tim E = 10.0685
{ 41
^3 7 si
-I l
,I RCE 91025 November 5,1991 I I p I I I I I I APPENDIX 4 Valve Drawing and Photographs I I I I I I 1 l A4-1 I
s .
- 2. -
~_ : .,-- -
3 :,,1 I 7 qq
=~
L td - :::.,- :, . O
- .v
- t
- I S*, b N* 4 m
f;- ,D d, % ...'n.
* ~
[ '
. - J: , n' n
l w: /,e -
~ I~_
,-g. -
li >;l 33 m y l Ih h[ 3 g l j - r .s1tv l SEAL ' t.o. -@ *
/:b-1 i
3 d I5 O} Jg g E --<- I 30s.ET ~ . s' i vm e s
,D -
"8 , p 0 -fi a :1 1 fl fil I ...-
Q .Q,, s E ; v E s i s ' 0 -)
.- p l Ii
,cq7 erg
/.,,---- c/ ' J#
y S E ", v E N T _ a. ,,
$AI[ --- -4 - /~% ,t . i / a 8 i /<,
I ;E <-4-tc u l l l 'i, g l-H staytst gg Av i
]L+i
,. v :,u v E s E A N . \ <, v ,,',\
,/ l! l'
,l. ! V, . ?
A'.L ,,,, ' LEss - L O L, < n u% o H. .m _..
- A v S-;E f If j.
i l . . .
.i , i ti
'i o
,l
\l l} * .'-~" LOW
,\ '
.z- ,
~~.
2 C ATE SE AT SCPp "y !' GwCE R AIL fl 'y$%,)t A ' d st; vest SEAT 3 cat G ATI SE AT SCRT \s 1 N, '/
~., .. .
- u' s v r ' qi s E :.T
" Main PL'-'O ' N' 5 c ,'a 7 -g;t ::rt
;d@E 1 C onfip a tion o f $A ain S t e am
;satation Val < e 1
ll il
\
t: f
/
s
%{
,% m.f.s n
o -
${. { ,-
y t. j [7l. y ar l b
;' ,W:/yf .g%m x I
.(f- \
n j . ,l'llj. y
\ ,
s- Ii l\ L 7.g O - i .W'>'
, /
n ) ar n
,9 Il . J, ' u u
g c
, ag.
1 1q l l u t r n c o i n [- ; f{
- g lA T
E D
- " g Y
!;f; p!l, y.
l't v i ' {a -"y%. U L B g ,g g! .. M t > E t (, k t t S
- c S I u A 2 A
O. t x T S c E N u A R E
' c U M L/
I S
\ ,Jj, n t
G I F G E 3
?' g _
'Ng\$FQ ?J ~. , w o
t S D
+
j' d.- ' ::n%
- w s -\ N
%. A g? .d-E T
A s. f' , i [ l 1 f. A - h G t 1
?s - . o
. %.t ,%3 1{.k \.
G N r A n K
" K c C N
\,
- C u C
L A . t c A D W M U o c G N R c n s R OS E A " E A C H K , 'L P w S E A U P U
~ o t
- G TE R
a c e /v. ,, 44
.a s
. ~'s ND c?%4' E 0 r 7 ~
u .
. . M0
^ ~ .M;s G1 U~- . _
E
- /s ?sf,, S D
# ~
4 - E
- 6 ~
t*
~ y' '
ET
\
1
; SA tg (g\ \ \
\
RT
'c:' ,
. s. ,, e,,vy EO VH E
H l l I:l ll,j ,l1
} i e ', w1 l ep i
;4= . 3 _ u ,3 - -, s a .- t.
I e. ( 4< e
. . ;. ;fi " PIf, h
4 . k '[ -'s 1
'L
( .
- 't.,,:-(.}<** } / '
.4.., -
~
- l l ,~ 2 Jsm ,. - ' -
%; [ niq
~/ ' ' 'E ,
..) ,,
. . / ,. ', Ag" ,h .h j
, .., - lt
.. -q.
+ y -
w
.. . , _. .,9y 3 n.. ::3 "
, % ;..na e.; A Q
7._.~
* ' . D, . fy q[,E , . .
E . i . , ; . , ,, 4 ,, , . - ~ . I , , . , . > 3, .; . . . ,' .
'.p. '. :'
~
' 'p. *_ ,, '
.. '. F-l-]j I;;;' _ '
'a u
- .% e m b,i.. L a
c'm. .- . PilOTO #1 flORTil S 1 DE 1 l'VER 1OCK ARM SilOE DJ. MAGE I . - - R - i @ I - ..
]ft , . (; %g I- .
- rj --5 f}*py'teg 4'$h. '
,; \ . '. ,
3 {;lr p:l [ . *- ' -
~ -
w.
, _\ -
3. A 1 y s.
}'lj
' dplf, Y u. f .'.cw; 4 .
@L d.h.l.. i . ' P N' .
PilOTO #2 SOUTil SIDE I.EVER LOCK ARM SilOE DAMAGE
s L l L,
;y .L,%
s g f},, 6 o c E .. l (, I( - s
. % ,.a - ~ p' s
i s i
- t. .
.oj i py a QP;;my[s: ;
b; ;[:..l:%.
;; 8 .
>zm,~W e,,
f: y y ,
.. p n
*;n{ ~, . ' , , ,I ..
j f . . N cfllY l o.sAussbi.d!.dM4 PHOTO #3 i NOR') H SIDE SKIRT RAIL DAMAGE l I I %e,-e. .
%[j*jki j ,- F
%>y m
4 .. hf
- abnscyv a \fy l ~
- n j$, fhhMi; 5 P_ ,l ,
p.s
+
.w w
,- 0 ,, ,e
~'Wupt ohk ;
*f;;jgg -
l
.,)h;!j$_)t$t3, l' A PHOTO #4 SOUTH SIDE SKIRT RAIL DAMAGE
y { Qm ' L &y ij - 4 1 -c ,
}{
, ]
/ r
^
~J ,
r , . ,[ t i 1
! 1 1
?
P110TO #5 I. CRACK IN DISK SKIRT (END VIEW)
~
y_ ,j ' '\ 'r . h I 9
\
/ %
wg 1
\ ' # -'
;,'y
.[%
I - I L I'llOTO i/ 6 CRACK Ill DISK SKIRT ( l'LAll VIEW) I a - - - --
w_ - - - _ _ - - - - - - - _ - - _ - - - - _ - - - - - - - - - , - - - , . - - - - - - - , . _ - _ - - - - - - - - - - - . y s BROKFt1 AP SCREWS I
. . . . - . . - . - . ~ . - . .- . - .- . . - . . .- . - - - .-. .-
l l I RCE-91025 , November 5,1991 l l i t 5 I 'I i + I a APPENDlX 5 l KALSI Engineering Analysis Report I e
- I
; I
I il Pagn1 !- ANALYSIS FOR DESIGN MODIFICATIONS g OF MSIV GATE GUIDE I DOCUMENT NO.1719, Rev.1 OCTOBER Si,1991 g 'I Prepared for l SolJnIERN CALIFORNIA ENISON CCMt' ANT SGSGS UNIT 2 SAN ONOFRE, CALIFORNIA I I I Precared by Re:;iewed by
,\d
- N Q lO*24 9l J. }C. Wang, PhD, P.E. M. S. Kalsi, PhD, P.E.
I KE! Ttle Nu. 335
'I , K ALsl ENGINEE AING. INC. __, ,
, qc-e.s c.n :esas . a ant *v s s
- , . - , s.
g ., . m. :.. . . .~.
L Document No 1719, Rev,1 October M,1991 Page 2 I TABLE OF CONTENTS Page I 1. INTRODUCTION 3
- 2. DESCRIPTION OF W.K M GATE AND SEGMENT ASSEMilLY E PARALLEL PXPANDING OPERATION 6 I
SUMMARY
OF RESULTS AND CONCLUSIONS 8 3, 3.1, Valve Closing Action 8 3.2. Valve Opening Action 10 3.3, Overell Conclusions n
- 4. RETERENCES 17 I APPENDIX A: Transient Dynamle Impact Ane,1ysis During Gate Clc5ing Acticn APPENCIX B: Analysis of Forces nr.d Strenes During Cate Opening Action APPENDIX C: Verifiestion of Computer Program for MSIV Gate Guide Impact Lead Analysis I
I I I I y I mm e~m~e em~m.
= .. . _ _
,~c.
I _____________.___-__________-____..____-_____________-___________-_.-_________--__.____-__.m._m____-__-..s_____-__ _ __- ___ .s
t Document No,1719, Rev,1
- Octolar 34,1991 Page3 s
t I L ANALYSIS FOR DESIGN MODIFICATIONS OF MSIV GATE GUIDE
- 1. IN'IRODUCTION This repcrt summarizes the results of the gate guide analysis performed for the Main l Steam holation Valve (MSIV) to support Southern California Edison's design rnodification effort for SONGS Unit 2.. These main steam isolation valves are designed and manufactured by W.K M Product Dhision of Cocper industries. The basic design consists of a through. conduit, parallel expanding gate valve of 40" x 30" x 40' site which can provide the wedging action in t i closed as util as the open position. Figure 1 shows the details of the internal components and the associated nomenclature. In addition, a desedptica cf W K.M gate and segment assembly. operation and its wedging! expanding action in the open and cloeed position is given in Section 2 immediately following the
- ntroduction.
The failures cf the guide rail capserews and skirt were dis. covered during the August 1991 refueling outage inspection of 2HV 8204 M81V using an optical boroscope. The MSIV was subsecurntly disassembled and inapected to determine the cause of failu.e and the design mod.ifiertius necessary to correct it. l Inspection of the disassembled gate and segment assembly showed what the failures occurred at these gate skirt locations:
- 1. The skirt fractured at the fourth capscre" from the top location. This is the narrower.t skirt width under tansile load in the r.en axis direction. The fracture appears to be tensile everload with seme neeking at the fractured cross section. Only the left-hand side of the skirt was fractured Oooking from downstream).
- 2. Three capscrews above the fractured skin section were broken.
- 3. The top capscrew on the right hr.nd side of the skirt was t.lso broken Goeking from d - st,ea a I ~ ~ m ~ mm , ~ e, , ~ c.
.. _ . _ _ l
Oricument No.1719, E4 v.1 Octeder W, tW1 Page 4 in add .. a '., the cbviously fractured cc mponents listed abose, the following damage or abnorm litio were observed during inspection of the gute and segment assernbly:
- 1. The gate rail chamfers, where the !cver Wk arms and guide rails interact during I valve elesing, show severe signs of ( iling. Onlling resulte d in metal renioval from the bottom of the shoe surfaces and d6 posited it onto t.ho guide rail chamfers. The Sigment side guide rails and the top shee Surfaces are in good eendition, indicating that the vt.lve did not trperience problema during valve opening 2 Sliding snarks esist on both gate and sagtutut stating surfaces. The back angle contact surfaces appear to be in good eendition.
I 3. Sliding marks also exist on all four guide tsil sliding surfaces.
- 4. Indentation marks exist en the matching shoes and lever lock arms at the int 4raction peints where the shoe articulation is stopped by a slot in the arm.
Ka}si Engineering, Inc. wer engaged by SCE to per)crm a detalkd review of the current MSIV problems ar.d provide recommendations fer corrective .,etione. The current efTort, as documente! in this repcrt,is focuestd on reviewing the current design deSchney and tnt,urug that the proposed dulcn modi 6 cation will eliminate the problems and provide the required safe operation of the MSIVs. More dettdled analysis on the ofTects of optrating pressure, cperating temperature, local centact stressa s, and cate r,ssembly dimensional tc! cran:e stack.up will be pe;rformed and doetunented in a later report. The prepcsed design modi 0 cations (cr the lever lock arm shoes and i aide railt, as shcwn in Figures 4 and 5, are basically the same design modincations used at leuisiana Power
& L:ght, Waterford SES Unit 3, for their MSIVs with the identical valve design. The modjS ations were successfully implemental in 1988 with no reported probkms to dat+.
The main difference between SCE and LP&L's MSlVs is in the operating nquiremento I the LP&L valvts are required to close in a thcrttr time (as low as 1.6 seconds) while SCE rerguires the MSIVs to clos,e witMn 6.9 seconds (normally oprated within 3 to 6 orconds). A m:.thtmatical medel that simulates impact ferces and motlens of the cate and segment anembly during closing .setion was used to conduct the rakr fcr the current design e.nd g proposed modificaticns. The model takes inio secount the velocity of the shte at impact, the 58 coefileient of friction betwttn the shoe and guide inil, and the lccal etntact gametry, I R ALGl ENGINEG AING, INC. t.w -AN:^ a w ,c.As.s m s
Document No,1719, itsv.1 { October 24,1!41 Pace $ including the contact angle that the thee can acquire witt. i the elevis of the lever. lock g arm. The math m.idel provides a w ay of anaiyring key centributing facters to the guide L rail impact force during a valve closing. It quantines the carrent design defi.lency as well as determines the rnagnitude of design imprevement that can be achieved under the proposed modi 6 cation A comparison of the two design 616 presented in this repcrt which shows the advantages of the new des!gn in reducing impact forec and in the ability to operate under a high cramelent of friction. The detailed dynntnie impact iorce analysis is der.cril+d in Appendix A. Problems with the valvt crening stroke on the stgtntnt guido rail 6 are annlyzed in Appendir B. The available stem thrust, the effect of frictional co6Melent. and the load on the segment rail are also dibcuned. Inspecticna cf gate and segment anwmbly t.re sugge6ted to asse that the MSlV will eterate within t.he design range. semion e , _ a,1e.the ana,,si. , emit a m e, _..i.niom . I I I _K ALSI G NGING E AING, INC.
..w -. c &. m . t. s. ~,
) Dccun.ent No l'11% Kev.1
- O&te r l'4. t!"91 Part6 7
F L 2. DESCitlFnON OF W.K M GATE AND SEGMT.NT ASSE.hulLY PAlui.1.F.1, EXPANDLNG OPDMTION [ Pigure 1 shews the dttails of the internal compenents cf the W K-M parallel expanding gate vahe and the nemenclature used in this repcrt Bulcally this salve is a wedge ghte
~
valve design in which two wtdge piacca are emplotd which expnd in a lateral dircclicn whenever relative asial motion (in the direction cf the stem)is allcwed to occur between I them This design relies upon the two pieces to travelin, ether as an assembly without any rehtivt s.xial motien (or lateral txpnsicr.) during the entire streke except at the very ends whet. the assembly approachts the fully cpen er fully cit.ted position. To keep the two wedge pieces called *gst4" and *stgment" from moving relative P each other, a special mechemism is em;1cyed which provides the nenssary kinematic restraint to the entire assernbly. This mechanism is comprised of a ' lever lock" assembly, which consitts of a lever lock arm, a IcVer lock shoe, a cam assembly between the gate and segmer.t, and guide redis which are fast +ned to a stat skirt on each ste.t.. During the mid travel position, the levt r lock shoe is guided b(tween the guide raib which pnvents any relative alal metjen I between Eate and segment thenby preventing any tendency (cr the two pieces to expand laterally which can result in an inerense in the force required to move the assembly. The guide rails are terminated at tithcr end cf the streke cf this valve to permit the lever + lock shoe to mcse outside the parallel restraint provided between the guide ralle. When going toward the My cpen pce; tion, this in turn p(rmits the gate to continue to move upwards abr the ugmcnt hits the Mitem of the bonnet which steps the upw crd movement of I the segment. This relative meticn b(tween the gate and segment is transtr"ted through a
~
special cam mechanism (nwt vinble in this sketeh) through th+ Intr Inek arm whleh kicks the shoe to the left as shown in Figure 1. This allcws a wedging / climbing action between the gate and segment to take place which is acecmpanied by 80 increate in the dimension between the parallel faces cf the gate and r+gment.
-l During normal valve cperaticn, whtn the vah e is given a signal to go clcted, the gate and I f4gmtnt atumbly starts to mtve down with the shoe stillin the kft petitien until it hits the tcp of the guide rail (et shown in Figure 1). In the basie desten, there is a 45. degree chamfer prov;ded at this location cf the guide rail which tends to rulde, or fcree, the thc e from this extreme left petition to a central position htween the guide railt. It should be noted thtt the *dgts c. he 45 dtgrte chamfer on the guide infle are relatively sharp with no eignificant redit, Th(se tre pctentiel titt s for initieting galling. In the curant SCE QALBt E NGIN E ERING, INC:.
..m.m s m c. . m l
Document No.1719. R4 v. i Odober 24,1991 l' age 7 design for 21{VM04, the chamfer angle was modified to a smalle r atr.le (from stain uis) 'g M and the corr.tr indius at the chamfer ws.s entstgtd. The gate s.nd segment assembly travels tegethtt through the entire stroke until at the ytry and w hen the segment stop hits the etop pad previded in the !cdy upstream condu!t. Afler the tegmtnt tr.otion is stepped and the gtte continut-s to be pushed down by the stein, the telt,tive motion letween the gate and segment now engages the cther set cfinclined planes and st.ntts the wedging hetion. This is secompanied by a motion of the lever lock arm and the shoc asu mbly which is kicked to the right in the clobed position. In the fully closed posit 3cn, both the gate and the Ettm wedge pieces are laterally txpanded to fittnly centact their respective tents and provide the desired Stating force. I I I I I I I I I I I K ALGI ENGINEG AING, INC.
~
.tc~ w,ca -eaz,.,r. u,a w.n
.., ,,. m . . . - - - - - - r Documen* No.1719. llev.1 Octobe r 24,1991 hge S I .
- 3. ltr$ULW ANT) CONCLUSIONS I 3.1. Valve Closing Action
- 1. Inspectjen cf the failed MS!Y ekirt and gaide rails (2HV4204) and raiew cf the detailed valve intemal compontnt designs indicate that the current lever lock ann assembly and guide rail designs have two majer defleitncies;
- a. The high contact angle (45 degreen mt:nsured from the stcm axJs) at the shc.e>
guide rail int 4 taction peint is ineffective in gulding the lever lock shoe toward the spare between the guide rails during n velve eloeing stroke. It results in a high impact force en the pida rail chamfer, b, The absence of overley at t.he centact surfaces hnd lack of radius at the guide rail chamfer enuse accelerated surface deterioratien, high c emeient of friction, reductd torque to relate the lever lock arm into :ts ccrrect positicn, and excessively high impact force.
- 2. A mathematical model capable of simulating the dynamic impact actien of the assembly was used to analyze the key design facters tEceting the gate closing (detailed in Appendix A). The model was used to investigste the efTect of cotme!cnt of frictien en the guide railimpact fcrte, the effect cf shouguide rail c:ntact angle fcr the existing and prepcsed designs, and the effect of the vahe clcaing stud en the guide rail impact force within the valve cperatinE rance. The results of this analysis I are tummarized in Figures 2 and 3.
- 3. E/Tect of ec4/Scient of friction. As shown in Figure 2, the current design (45. degree contact angle) has low impact force for the ncrmal range cf emmdent of friction (0.2 to 0.4 range). As the contact surfaces deteriorate N - 0.6 or higher), there is an I asymptotic rise in the magnitude of impact feree delivered to the guide rt.ils.
The:retically, as the coemeient cf friction reaches e. meignitude of 0.72 (for the 45 I depee contact t,ngle desten), no torque is generated at the shoe to guide rail contact to caus.c rotaticn of the lever lock arm and force it between the guide rails. Under this conditien, e.ll the kinetic energy cf the moving gate assembly is absorb (d by the lond. resisting components and converted into strain (nergy. Whcn this occurs, the mrgnitude cf the fcrecs transmitted to the guide rail becc7r.es very high, and can easily cause shearing cf the guide rail.to-tLitt heltir.g er the wakut n: tion cf the I KALSI ENGINEE AING, INC. ve c.u , ca.. c e sos r;. w s. s n _ _____ _______ _
9q g; - ; i,G - [;] T K r v -- W. .s i LVJi xu i W - ;* i;' ic4U:' i'J1 I Document No.1719, Ha e 1 Octdwr 24,1991 Pact 9 skirt itself, as cbeened in the fa:ltd suhee. The art trent ca fuch r.t of 6ietion c,f 0.72 is peeuble wit.h the galling centact surfaces e nd sharp corner indo.taticn.
- 4. Effect of thce to. guide rail contact anglt, Yer the W%1 propond duica ined,fice.
tions (30 degrees of contact angle WITil RESPECT TO STEM AXIS), the impact fr.ree is much lower than the current design under the ncrmal range of coefncient of friction. 13ut inore importantly the new design can tolcrate a high ectfncient of I friction, up to 1.0, without significantly increasing the cont. net forte. The advantage of a shallcw contact angle t. hat can (ficctively convert contact force to leWr. lock arm torque to rotate the arm towt.rd the center of the guide rails is elet.rly illustrated in Figure 2. This shallow angle kups the impact force low and avcids surfuee degradation. It thould be noted that the normal centnet force at the shoc/ guide rail interface is pcrpendicular to the bottom shoe surface, at shown in Appendit A fcne disgram, I while the (Hetional force is parallel to the shee surface. Thuefore, the bott:m thee surface angle during the closing contact deterrninu the magnitude cf the rotating terque to the lever lock arm instead cf the guide rail chemfer t.ngle. Figure 3 shows the time histcry plct of the impact forec6 for the old t.nd riew du!gns, I As the coefUcient cf friction changen frem 0.6 to 0.07, the old 45 degree aqle design increases the impact force signifierntly from 7,050 to 11,190 pounds while the new design keeps the impact fcree r.t less than 2,160 pounds cfimpact ferce (for clesing speed of 10 irib.e).
- 5. Effect e/ rolte clo* Int IPecd. The SCE valve closing
- peed is citimated in the ramge cf 5 to 15 in'r4c, Figure 2 plots three cloring speeds of 5,10, and 15 in/ree fur the MSIV I closing speed range. As shown in the figure, the impact force increate s rcpidly as the closing tpeed increaus. Cornparing the current SCE design of 10 insec versus the old LP&L design cf 25 in'see shows that the irr. pact forces r.t = 0.4 ere 4,252 pounds for SCE .nd 10,630 poands fcr LP&L, Thenfere, the SCE dukn thculd be e,ble to perfcrm better if cther factors rernain the same Fct the prcposed design, simila r relatienthips (xist with scnoclly lower irnpact leads ove r a wider I coefficient cf friction range, as shcwn in Figare 2.
friction increases (contact surface decradation), both SCE and LP&L duirns liowever, at the esemeient of K ALSI ENGINEE AING, INC. W ::>- e.s A, MM C. 4.N 4 6 b B_ . _ - - - - - - - - - - --
ram Lv;1 Tu.K 1 VJ - -- ,;e i:t itiv : o .c m3 3, - a, s -ai Document No.1719, Rev.1 0;tolor 24,1M1 Page 10 l approach an asyrnptotic rise in the magnitade cf irrtatt fcree delisera d to the guide rails at p. = 0.72 for the current de dens and u = L2 for the troposed doign modificatien reprdless cf the gate closing sped.
- 6. In addition to the improvernent in contact angle to a mere favorab'e orkntatien, the new thc4 and pide rail designt provide the following dt'61gn impresemtnis:
- a. Stellite overlay on toth shoe and guide rail chamfor surfaces;
- b. Nitto.httdening of the shoe surface;
- c. Generous corner radilin both lever. lock thee 01 = r) and guide rail (R = W).
'Ihcrefere, based on the above cornbined design improvements, the new deeign should have a sufficient safety margin to keep the contatt autfares in geod working condition after repen.ed valve eleting.
I 3.2. Valve Opening Acthm As shown in Appendh 11 calculaticns, the valve actuator has sufr.ciant hydraulie force (649 kips) to pull the gate assembly upwcrd if the reihtance exists due to the shoo and rail sticking. If the gate and seginent back angle turfacts nie in good ccndition, the gate and ulcment 6hould be in a c0llapsed con 0guration with very little lead applkd to the segment rails. if the back .ngle surfaces deterierate much faster than the seat surfaces to reach a I signiacent difference in cocflicirr.t of friction as tabulated in Appendh 11, then an upward wedging movercent may occur, resulting in high centact load on the railn. Another possible shoe and segment rail chamfer contact rr.ay occur during the Snal stage ef a valve closing 6trcke if the dimensione.1 Stack up is outside of the design lim.t6 (discutud in Appendix 11). Inspecticn cf the 21IV 8204 MSIV thewed this valve to be free from both of thue pctential pr oblem s. To ensure freedom from thtte potential problem areas related to the e;ening etrcke cperation in other htSIV5, the fcilowing inspecticn should be urfortned at disanen.bly: I K ALSI ENGINEE AING, INC. W Om e :.s cEra:.N c h. n v m n _. _ _ _ _ _ _ _ _ _ _ _ - _ ______- _ _ _ _ _ _ _ ____ _ .-
my ; i v ,;).19 ; v' f,. r; 4 N i vJ " ~. iT P.T R WIT. e Dt.cument No.1719, flev.1 Cktoler N,1991 l't.re 11 I .8
- 1. The back eagles (.f t.he pte and acment theuld be inspcted to enture that they are in good werking condition.
- 2. The gate and tegment asumbly, including the urment skirt, should be inspected to ensure that the top lever. lock shoe surface is not contacting the segment rail chamfer during the fint! wedging action of the ghte and segment in n (lcting stroke.
The force equilibrium analysle in the opening directicn shows that,in ceder in prevent the lever. lock arm from relating to the center cf the guido rolle, the coemelent cf friction would have to be as high as 1.29. Thle magnitude of friction le not encountered under normal contact surface conditions even with some degree of suriate damage. The new improved shoe and guide rail design with Stellite overlay, nitro hardening, and ample I radil should prodde sumelent margin to aveld contact surface galling and high segment rail lead. I 3.2. Chrrull Concittsicma The analysis has identined the current design dencieneks. The prepond design modifications should be t.ble to eliminate current design poblems. A review of the modificatkna shows no new problems are introduced. Therefore, it is recornmended that the preposed design should he implemented for the SONGS Unit.s 2 nnd 3 MSIVs, I I I I K ALB' ENGINEE A!NG, INC. ' ~ wcwa cA r t t oy i. uat ytw; n _ _ . _ - _ _ _ - _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
y
; y. ;4 ry1 cn, a cn r m rscch m m r a ----- --
5 Document No.1719, Rev.1 (ktdwr 24.1941 l'oce 12 il .
** K ALGl ENGINEERING, INC.
y y eru,* a moss casaws I 54 s ,1 ~)k'k O \
$ e L $
g I t \
?'i h I h s p- SEGMENT STOP
{+{ g $ YJ - SEGMENT 0^ -
-7}) / , -- LEVER LOCK ARM { .,
gg - LEVER LOCK SHCE ) I LEVER LOCK ASSEMBLY' g ', ,, [ sw , . -- GulCE RAll i N 'i GUIDE RA!L SKIRT } [ [ k l QUICE RAIL ASSEMBLY I SHOE HIT HERE h NORMAL FLOW WHILE CLOSING d _q d - h y. I SHOE HlT HERE WHILE OPENINGd S I .. - Figure 1 Nomenclature of Compononts Used in WKM MSIV Cato Galve I K ALSI ENGINEE AlNG, INC. rac-w<uc cenas c. A.Nu vss I b _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ . -
m m W W M M M Mi f s 'X ou
> O$
r r - T a.
$ S >? U g - -
~h =*
3 ~ v. m - (' Z 50000 8 r 8?- i
!" 9 1: O=45*. 5*/sec - SG N 9 Z 3. p=4y, lu /sec - SG ! f'A ?
- , g a
~'
- 3. p--45*,1Y/sec - SCh ::f '
e s T d y E 40000 e o=30 . 3 /sec - SG /& - 5- 0-30',10~/sec - SG :; ; T f --E II F v :
'f 5 6: Q=30*.15~/sec - SG (o\
, 31 2 c o
7: O=49. 25'!* - 7I-y,_ o- 30*. 2Ylst-c - I3,I-I' h4 '-J l l i s 8 33300 ~ - I
! ,h* .
g } c,,,fod j oogk
-)
n 4 .$4 w t n- 2c000 , lV/I $f G i
,/ /1 j
1 2 .- :
~~
, , ' j' Y} f /
' l/.-- ,
399g3 '
^' !
h ,
',l 5'
,. s s ,,,/~', ,,__'~_,~/,,.. '
- . - , ,,,.~.
A
-~__ -_,..#../,,,,. ) h4 U
0 -1 . 1 l 0.2 0.4 0.G G.8 1.0 1.2 1.4 l C.0 i l Coc!Ilcient of Friction rs Figure 2 z* Comparing Impact Forres on GateItait for t! c Old and New Guide Designs _
- a l
Document No.17;9, Rev.1 Octaher 2.4.11491 Page 14 iLO'O
-010 0 S IB G IB 600'O =
j y 9a y - 2 2 d l' lA *A R 1 1 ,o 800 0 .h8 G G G .a e A l I p po# ~ L 400'O g a Y p A I 1 # , th
* ' p 5&
# -900'0 e .o 0 aa e '
as
,, 0 $ g to, c'
~500'O ,
j]
- L e
r.& d .cE 0 ,E
$ .. t 0 0 ' 0 f ,}
E l~b T 4,
-000'O ,o T e
d v . tiLe o u
/ - ?,. 0 0 ' O gg
,; n s' ~6 ___
v e b-N N I C
-j v ! -100'O .-
.b u$
ii -000'O .B O b
, , ,- . . . . 000'0 o o o q a o 8 8 g k4 9 N
=
3 u !. SQl '118W uo *DJod lordWI f K ALEI ENGINEE AING, INC. W.cn A ct.u ce vas a As % v e, s
Document No 1719. IdV 1 Octdat 24,1PB1 . Page 15 i .,
- -.., ..' ]
111111Q ! % di:i1 a tt E O I O l'i:s< s . 34C
% u h
.60*,1)
* ,, F ;-, ,% ,*,. g I n'. '
'O g * ! f in 9, [ e :t ,
I h;.kd, k 3 k I .. I .Q.9 i B i p a { I;jU'? I W 1 Ql JJ 1 d.!!" " .L 5 gL'i;gra ';g g 3
\
2 . 5 i : L 05 : l . hh#'[ D
+;
I fV \ ~
'I NE 3*j h
n
~
- s l l l :!
a *t ' I, 2 [ gi . .. f
! s it e d g I ' l i l%ll s P ! , hhi t; , s> - 'g J j y I k 4 -
w i j v' n i g F- 1, d t! h _ , , _ ek 1 .,) -
' ,*f4
*g
~
=
*j g F
$ l C' E; f eA.
g p..; .= m l e h #
- k. ,
i l ( q 3. WP e -
* .I. 2 I
z p( ';) d s 8 , I
, , h Ill '
MS d ]- . h ;';'; ; ;'; ,' 9 i t !, y t 'Di!!.!.L.
;j > i
{
+, ~ --h. . $ ~
l. gb'!!. ,y '.
-n -
r i iu,ii[d.,jjjs., ji 1 j ~; p, ij . g,,i . . ... . It I qiis If K ALSI ENGINEE AING, INC.
~'
ve ce:Au CEr.GN c Av tva o
Document No 1719. R4v.1 I. Oct4,lm r 24. l'.61 Fset IC I i ! 1 g
'e$"
4
' i' \ g;e.,'
bg* ' I?g]If ! . O( , I. 1.- 8 __ -- _ s fil ![.1
,1 u rf\y. A .
m i et .. pt ot_ t.i 3 ._ i j j @, g[3 j: E I l s et , y a ,CR 't
;g tv- g
,, C 4 :
J
...., . ~~ f "* * ~ t r DI ~
i l 2. N h
, C k' l I l
l a al< i, ,
# . %a . .
. [~.g -c q .. g.. + .. +.. ]
-.p. ..e.-.9.-.g..
.p.
3, p
, Ili f
{ i l ithi i [f . U Lpk l >. l f, e,-
. l p
Ur te
!'b i.o
,,t, I S I st2 djj d: i.g 3
e I i. i'J CJI i l E 3 W IT ; 3 l ,. . A - l _ _J D,: e'$
)l fI 4
ull;E o'j l3c- g ---- - - _ _ _ _ i a _1 , ,,,, Jig ' ,1
.5
+ I, I \,f-- -- - - - _-
$' e FBjiHilpl i a.
ijlij iM,::. 3 o 3
;4 I :J;, <g g { ,l i l *7.g ;;p i ,
1[ :
'i , 6. ,
i I, I
, E l ,
.g.. .... .
i.
" [I ' !
- 3i ul { - i
( - - _ . _ - _T i ( 5;a _T~- - __J1. gj = i
-Q '
i {'dby
= - -.
- ~_ ~.- - n
.S , e,1 ',
\ , f,
,t s i 4
I EI
.' g) r
{ - 3
.--. v 0
)
)
i _,.__.._ ._, ,_J q' K Al.Sl E NGIN E E RING, INC. MI! C.e A* J : 4 L ,~~l E A f. 4 *.A L v 6 3 I m - _ _
Docurnent No.1719, lkv.1 ~ Octthr :4,1991 l' age 17 1 H p-I im1TIENLTA
- 1. l'riction and Wcur of htaterials by E. Habinowicz, John Wiley & Sons,1965.
- 2. Rcet Cause Analpis of MSiv Gatt Guide Failure, kisi Engineering report to 14uhlana Power i Ught Company, KE! Document No.15600, hv.1, June 1958.
'I I
I I I I I I I I I I I I K ALEli ENGINE!EMING, INC.
..w 0+ c% cuw; c. <.s,.i., a n
m -
,nrn Tw'r~- m e r u ,w r ,an -si ----- - - - - - - 7 i v v . v - w v v - , v . s-Documetet No.1719. Rev.1 4ptndia A Octder 24,1991 hge 1 A
AppendirA*. Tnmstent Dyruunic Impact Annlynis During Gate Ciming Action KALSI ENGINCE AlNG, INC.
*k h"1SEJ C d*b (s h ND M b Ng&(Th 3 a - _ _ _ _ _ _ _ _ _ _ . - - - _ _ _ _ _ _ _ _ _ _ _ . - _ _ _ _ _ _ _ _ _ - _ _ - _ - - _ . _ _ _ . _ - _ _ _ - _ . -
,- m,n, . . . . . m. n + m w m - - -ae m u m a; s Documer.t No.1719. Eev, 3 Appendix A Oetd= t 24. tM1 Page 2
!I . Ap;rndit A Transie nt 1)yn arnte impact A.nalysts During Gate Closing Action la order to take into account the algnideant factors that afTect closing i:npost lond, a simplified meth model was developed based on the cloting mechan!sm, relative stiffness, and some assumptions. The signincant factors that tJtect closing impact lends are the z . contact anEl e. coefncient of friction during impact, valve closing speed, segment welght, end lever. lock arrn geometry and stiffness. The math model was developed to enable un to simulate the closing impact from inillal centeet cf the shoo and rail till their sept.ratien. It provided un a way to analyse the contributlen cf these signi0 cant factors and their perfonnence rangt's. The results of this analysis enabled in to identify the current design deflekney and assure the performante of the prcposed new design. A.1. CLOSING IMPACT ANALYSLS During early val,e clesing, the bottom of the lever.1:ck shoe contacts the tall chamfer, The ter.ultant reaction force from the te!) normally pievides a necese.ry torque to push the lever lock assembly to the center cf the guide tr.ils. This rnerhanism in desigried to prevent gate tnd segmerit wedging during closing, The closing impact analystr was develcad tv ct.lculate the transknt impact lead using Inc.1 mental time steps to simula;e the imptet action from initial contact to finish. Velocity. force, tareue, and accelerration relstionn were develcred based on some almplified assumptions to provide a reasonable site of the inath model. The (cIlowing ecntervative rssumptiens were used in devcleping the math inodel:
- 1. The lever lock arm is the only Dexible member under impact lead : compared with the othtr gr.te and segment components), If additional fleubility taists in the sisembly,it will mt.ke the syst+m sefier and the impact lead lower. i
.I 2. The cate and segment assembly trcsels at a constant speed htfore and after the impact lead (unless a very high impact lead cttutt, the clesir g daw n tpced will not be algnifict.ntly changed frcm the tisumptkn).
I K A LG 1 U N_Gl_N. E E 0 TIN G, f N C. , vannas cat na as n 4.v,p e n --
y y 7--,,w - m - m i m n i y - 4e np , o m , r, . c Document Ne 1719, Hrv.1 Appendit A Ocider 04,1991 pg,3 a
- 3. Segment weicht is the only weicht used in the lever lock arm acceleration ,
cokulations. The lent leck t.rin Might, which 16 relatively 6rnall ecrnrared with the tscment weight. is neglected. Impact simulation is perftrmed by the following stept: o
- a. Select a small time step.
I b. Unleulate the dcwnward (1 Ait) and ILitral di0etticn.
~
A
- c. Calculate the feree components, resultr.nt force, and tcique due to thlh added l travtl and de0cetkn.
l
- d. Calculate the nettletation velocity, and displseemtnt of t.he segment.
s
- c. Modify the lateral de0ection due to segment moument and go to the next time I step.
A.2. EQUATIONS FOR T11E IMPACT ANAINSIS
'ihe follcwing mathema.tical equatiens were derived fcr the impact simulatiens:
- 1. Lever. lock ari.: Cexibility The lever 1ctk arrn Onibility is calculated based en the following simplifkd meith model: ,
Ft 3 DeGeetien a I y, 2 I75 " d where i 3EI n 2 2. 5 g '
--J.
E t-3E7 3 p, , ,,4.5 "og evr. ;, L75 x 4.5 3 12 l " g , 3 .- -
? .-
. . - .. DtDection 9.524 x 304 F t-Firure A.1 I.crer.Inc.k Arm Math Model
.I h,
R ALGI ENGINEER 21NG, INC. UE6%Ncat. ceu ca c e.Acvs s --
3 y. n-- pyy-, 7, w u s u , , u.v. , , , - ,,,..y ,v.,g._ , ,, Document No.1719 Rev.1 Appendix A October 24,1991 p,.g, 4
- 2. Vertist.1 ta.d it. ural diglstemerits at the cernact themfer
- AY = V x at ce a ent
, ,,pg where Ve elesing niecity, in/ tee I 3y at = time inenment, see n \'
AX n t.an 0 AY . AX' M where 0 = centact angle Figure A.2 4X' = AX movemer.t due to segment jpward Shre Displannment meme as derNed Mowy
- 3. Force and terque relations Lateral force increment fer a riven lateral t, gg de0ecticn increment-
'k '
'.524 9 10 4
J h l
/l d ~ Normal force increment:
Ah
/,,s Jr.e AFg cos 0 f fe' ai
! 6. Resulte.nt force increment'
? gy , bin 6
ees (t.an~I p) 6
) u where u = coemeient of friction t.t the centset surfaces ff, F4 mment arm distance, d:
Figure A.3 d = Li sin 02
- L isin I tan'lp- 0) 3 Lever loc.lt Arm Forve and Torque '2 /
I Relationships Verticat displacement . lever.1cek arm's car, or, due to lattral denectien, aX: 9 aY' = L aX 03 AX K ALSI EENGINEE AING, INC.
*,ec.,wce.t cE rr.N s AN AL vs's
_ ,_ _ __ . _ _ __ --- - ~ ~ - -- --
~
Document .No.1719, }kv.1 Appendis A October 24,1H1 Tate 5 i . Torque increment, a Torque: A Torque w AFg x d Force increment at ear, AFy ': AFyrem L 2 Acceleratien increment for the tegment: A Ace = m where m a segment mass = 2,50&3SG.4 Segment vertical displacement for At increment: A Y ' = Vo at + 1/2 ao At2 where Vo a Prnicus iteration's velocity, Irt see / no a Previout iteration's seceleration, irt /ste2 Segment vertical selocity change fer At incremont-V u Vo + ao At Undemeted AX movement due to AY' movement An AX' . bl AY ' = 2,2,,5,,, 37, L2 2.63 Based on the above derived relationships, the parametiic study of these f.ignificant factors can easily be performed using a FORTRAN computer program. W re6ults cf this study were plotted as shown in Figures 2 and 3. Figure 2 shows the overrill trend of the lever lock air.embly perkrmance as functions of the gate elesing velocity, the chamfer angle, and the coemcitnt cf friction. The 45 degno angle represents the old design while the 30-degree arii;1e is the new (hamfer derign. Fcr the cid design, as the coemejent of friction increases, the impact force increates eslowly; but the impact force picks up quickly aft (r the coemeient of friction exceeds 0 6, it reaches a maximum cf 0.72 where no torque can be generated under this high coefficient of I friction. At this peint the actual impact force is bcyond the capability of this math model to K A LGl E NGIN EI E S::llN G, IN C. Me se cu. ::u s aN c. e, A;vs 3
c., u . _ , . . -. .._ - Docarr.ent No.1719, N
- 1 A;gendix A 0:uh r 24 IW1 Pere 6 calculate. (It rne.y be estirr.ett d from the 1. uncinr, for t o ) llows-vt t, the tre nd ch erly shows that hy;h irnpart forcu occur afta the coeflicient of friction exceeds 0 6 due to the I
fast disappearing torque as the resu' tant ferre a: Trenches the leu r hirk rais (d - O For Fmocth stal cn4 teel surfacts, the cctf%cnt cf friction is estimated to be in the C 2 t.e 0 4 range {ll For a piled surface, the toen!chnt of fricticn can quickly go be> cnd 0 6, resalting in high im;nct ferets that caute enere guide rail and gett skirt damsges f The cate elciing speed c.t the lever bk shce and guide rail centact is critical to the ptrfermance of the lever. lock mechanism. It nct only affects the magnitude of the impact l load (ni shown in Figure 2 plett), but e.lt o e a a n s the initiation of surface degradatier/ galling, as cbserved on the guide rail chamftri and the lettom of the leser-lock shoes, The gate closing speed at the thoe and guide rail contAet is entirrat4 d to l# in the range c f 5 to 15 f r,'ree for SCE's MSIV design. The impact forces, as o.leulated frcm the math model, are in the range of 2,126 to 6,375 paunds bcfere et ntact surface degrtdttien (a = 0.4 use d). The impatt feree for the SCE doign is sigt.incurtly lower than t.be estimsted ,I impet force cf 10,630 pounds fer the LPL design with the maximum closing spred cf 25 l in/se e, as shown in Figure 2 compriscn (2}. Howeve r, as the coefGeient cf friction inerenses tecnts,tt surface decradatien), heth SCE and LPL doigns approach 5, f asymrtetic rise in the mr.;-nitude cf the impct force delivuod to the guide rails at = 0.7 2 rect.rdieu cf'he gate closing speed. q The new thee ar.d g. tide rail designs, as shcun in Figures 4 and 5, use n shallow er centact h Engle of 50 degrees (meuvred from the stem c.xis.' This new design angle enn rncre l ! (fkethcly generate hi,,' 1mr leck ttrm torque to push the shoe toward the center of the 5,.:ide ra:Is. F:gure- 2 shews that the impact ft,rce (c'r the new desten is in the range of 525 to l 2,471 pcunds (same at LP&L's medied shoe and guide rail ar.gle doirnh which is enly 38.S pc tnnt cf criginal doi;n i impet fcrea with 45 degreer, contact ang!e. I l It should be not(d that the normtl ccntact force at the shoc' guide ml intoface is pergndicular to the b0ttom shoe surfr.ce, at shc wn in che force dir. cram, while the frinion force is parallcl to the : hoe surface. Thertfere, the shee bcttmi turface Engle insteed cf the guide rail chamfer angle wdl determine the effect:vuess of the sha te cuule rad design in dh erting the ccrtact fcrce to the rchtiry: k rc;ue fcr th e les er h4k arm. !I g K ALSI ENGINEE AING, INC. r ,m. - w 2. ~w. c. .:t. :. ~ - I
- g. , m a- rw..-= , w . , u w r m.ra ,. - ..,...v.>.. u 4,
- Decutt.et.t No.1719, ltev.1 Appenda A N Octrikt 24,1991 Pore 7 h
L The new shoe and guide rail desigt.s cito provide sign.0 car.t impreve me nt in centntt surface galling n61 stance with Stellite nerlay cn both thee end chan,fer surfens end with nitro. hardened sutis.cc on the thort, t.s shown m Figures 4 and 5. The rail shamfe,r angle is al6o 30 degress (from stem aus) to match the shc.e angle to ptcaide a le.rge Oat surface ecntact area. Ovnercue cerntr radli art t rovided in both la ver. lock thc.e 6 (lt r) rnd guide rails (It = 5/S') tn keep the contact surfaces from degrading. Beard on the math m9 del calculations, the new det.lgn etntact force will retnnin low tmtil the oppt. rent coefUcitnt of friction teache6 the extremely high vahte of 1.2. Therefore, the new detien hat Sufhcient safety margin to racid damtre to the centnet surfaces nihr ren nt4'd closing. Figure 3 shows the time history plot cf tbc impact foret s for the old etnd niw desicus. As the coefBeitnt er friction changes from 0.6 to 0.67, tk cid 45-degree str.gle design incre ases the imphet force significantly from 7,050 to 11,192 pounds while the new design keeps the impact fcree at lei, then 2,100 pounds cfinmaet fcree (for ci ' ; speed of :0 irt'See).
.I .I I
I -I LI
- I I
I K ALGl ENGINElIFilNO, INC. w =-w,zu. uvv..~,c. w.e n,.
g _ . 3yyyr - e = r,7v -- m m pu i m , y - o,,w w. a . . . e tiocument No.1719. Ibv.1 Appendix B Oetder R 1991 reg,1 I I I I I I I Appendix B Analysla of Forces and Stmssaw During Gato Opening Actiot I I I f I -1l. I g K A LG) ENQlNEES:4tNG, INC, vct> o,cta. ce e as c. e.ut ysh - I
________,..-.--_m,,, . Document Nc.1719. He v.1 Appendix 13 I Octoter 24,1W1 Page2 o l I Appendis 11 AnaJysis of Forces and Strwea During Cate Opening Action .I 11.1, VALVE OPENING FORCE EST15tATION During gate opening, the gate speed is slow. Therefcre, the load on the stgment rajls can be treated as a static load. The availt.ble force from the hydraulic pressure can be 66timated as: Fnet = net upward force e hydraulic force - nitrogen force
'I - actuater piston drag
- actuatcr packing drag
-- valve packing drag
- moving component weight
= 2,760 x 0.7654 (24 ,42 ) -- 1,150 x 0.7854 x 242 2
- 15FJ2
- 12,983
, - 5,369 I n 648,f69lbs
- 6,124
'I %crefcre, the actuator can previde sumcknt forte to overecmo the seg:nent rail resistance (thearing the bolts)if the shce t.nd tall are stuck together. D.2. EPFECTr 0F 17,ICTIONAL COE171CIEN'IB I During valve opening, the stem slowly pulls the gate upward. If the gate and segment are already in a allapsed position, then *.he lever. lock arn will stay betwetn the gate and segment rails;i.e., no sigtdficant centiset force in applied to the 5tg~nent rails. For the gate and segmtnt to rot collapse, and remain wedged du;ing open!ng, the cor:mcients of friction Lt sett surface and itcment back angle have to be significantly d:fferent. (Nete This is an unde:sirable rnode cf cperataen.) The (cree equilibrium relatienships for opening direction can be exprtised at: I K ALGl ENGINEIEFtlNG, INC. e,tcr,uc, A. rscsss s aves s -
37.;;q1 773n r mu nymtuaw+-^-- 714 i!c it i'b G0 '.? , Document No.1719. }tev.1 Arpendix B CM.ober 24,1991 i' age 3 y E Fx e 0 i Nccs15'+p2 N sin 15' t D3 u -s., M ne E IN a 0 4 * [ , D, P2 N ecs 15' . N sht !5' . W . pt Ds = 0 o
- y. 4 D3 f ,, Da .
- J, , _( c o.-s 15' + p2 in 15')
I 4#S '"8d"
/. D (p2 ces 15'- Sin W) = W +3p g D S (cos 15' + p2 sin IS')
I Figurv 11.1 Segment rare Ndy DL,griur. lp If DS n W, then . I d"" p3 ces 15' + sin 15' ces 15'- p g sin 15' l Therefore, we have the following relations for causing tha undesirable mode cf Este/ segment wedgirig while opening: p1 92 trir-I 0.20 0,494 0.25 0.555 0.30 0.618 0 35 0682 0A0 0.743 i OAS 0 816 010 0 887 The eniculated H2 in the minimum ceemc4nt of fricticn required to move gate nnd segment upwcrd te.gether in a wedged (expanded) pcsiticn with segment weight assumed to be small ccmpared with the 5(ct centatt fctce, Ps. The alcVe calculated pg itnd n; indicate that the difnrences between pg and p2 are higher than the usual valuts for sent and back K ALSI ENGINEE AING, INC.
~mn a un a. a .a.
Document No.17 3 9, het 1 Appendia D Octota r 34, lys t Page 4 I - angle surfaces that are in good condition (example: D1 = 0.25 and p2 = 0,4 are ecmmenly ustd in design eilculatiens). Therefore, in an ideal dt. sign condition where bcth centact eurfaces are in goed condition, the gate e.nd regment should be in a cellapsed tendition with sery little lead applied to the segment rails. If the back angle surfacu deteriorste faster than the sect surfacts to reach the above tabalat+d ranges, then the wedged upwbrd movement can occur, which may result in high contact load at the rails (if a high coefDeient of friction also exista at the segrnent rail chamfera). Another possible shoe and segment rail the.rnfer contact may occur during the final stare of a valve clcriing stroke when the gate and segment are wedging to make upstream and downstream stat contacts and the Itver. lock arm is swinging toward the eegment rail. I For a normrd gate end stgTnent assembly, the lever. lock shoe should go under the segment rail chamfer withcut contacting it. Should dimtnsionel tolerances sta.k.up outside the design limits (a shtrt lever.4ck arm and'cr a long segment rail), then shes and segment rail chamfer contrct may occur while clor.ing. The contact load under this condition may be high due to the intuMelent lever. lock arm swing that prevents complete wedging and expar4rg cf the ge.ta and regment. Shoc and rail chemfer surface damage mey result in high apparent coemeient of friction in the subsequent opening streke. The inspection cf 211V.M04 MSIV show this valve to be frte from this potential prehlem. Whenever disesserr.bly c' any MSlY is donc, this area should be examined to ensure proper eperation. 11.3. LOAD ON SEGMENT RAIL SectJen 3.1 discusses the condition that can stJck the leur. lock arm shoe at the segment rail chamfer. In this sectien, we astume that the shoe sticks at the chamfer anJ estimate the mtaimuan leed on the rail.
, Assume as the werst case that PR paists through the lent.leck rais. If all the Icad-carrying comperents are streng enov;;h, then the lead applied tv both raib can reach 649
' tips, as chleplated in Section El. From the design lnyout, we have the requirt:d cocMejent of friction as:
gn te 0 "I 0=lo0.W'iF=5? a = 128 K ALSI F.NGINI!rd AING, INC.
. w tw. a c e s y, c <#.t.t vro .l
,,-_,, w, - - , . . . - - - - - --- -- -- - - --- - - - - - - - - -
Document 14.1719, kv.1 4pendis 11 Octd.trr 24,1991 Ps re 5 I - lt is not possible fer this high cctf%citnt (f frictirn to 3"7 c.ccur on usual contact surfaces esen with seme gsliing. flow ever, the actual centact, based c.n the cld h- - MSIV layout, cccurs et the lower cerntr (f the rail I M chamft r as show n in Figure 11.2. The indentntien of the lower rt.il (hamfer (ctner en the "Y tcp shoe surface may tauto the initiatten cf meterial galling, especially for the old 17 4 Pil snaterial without 5 surface hardening cn both sides. I N i
=
p
; The arparent coefficient cf friction under galling and corrier
[ge indentation may be incr(ased to a levtl that is clete to l ' the above talculated value.
~ Segme4f The new shce and segment rail designs, nr. thown in g, Pigures 4 and 5. provide sigmificent imprevemint to (cntact surface gallir.c resistnnee with Steillte overlay 8
Dx # en bcth shoe and rail centact turfaces. The rail
\ etamfer cerrar radius chances from an oticinally 6 harp i;ctner to a gentrcus b'8. inch radius to avoid surface degradt.ticn. The shce surface in cho nitro.
Firm D.2 hard(ned to provide cdditicnal galling resistance. I.rsd on Segment Itail To assure the MSIV cpening stroke eperated within the detign vance, two areas are y recommende:1 fcr insp<ction after the valve f a disnuembLd: 1 The back angles of the gate and segment c(ntact turfacts should be inspected to e.ssure that they are in goed cend.ticn. Therefcre, the coefkient cf friction on the bath angles veill not be exceedingly higher tht.n the stat surfaces, as discutred in Section IL2.
- 2. After th;e r;nd guide rail modifications, the CEte and ttgment attembly, includit:g the r>egment ghlit, sheuld be intpectGI to t. pure that the top lever.ltch thee surface is I net ccntr.cting the s<cment rail themfer during the final wedginc acticn cf the get tnd let;Tntnt in a clesirg stroke, I K ALal GNGINGE AING, INC.
a c-i.,=ir a rm sw s . -
,7 , , . .._m - _ . __ _
K ALBt ENGINEERING, INC. 000 'd o - -
"EV ho.
I
- tsesti u.n sc wsa n u -_
r Act .lf_ or SC c o ne : ._ _ EE._. _., _ . . . _ . -. _ . . _ Pno.'EC1. A.V.ALULL.f2.'1 n 6D&}L.Mt EWil!J/}.1J t W ML!L' 64 75 .$48. .. _ I rearAnte ev. d.. > . isMVA P t vif wi D e v . f_'%_.MAf L . , .__ _ _ _ ,
--. D Att: to
- 4:2L nsos so __3 r r C At t .tL'A.t'lL C A- _m_ _ _. D A t t __. . .. __ _
ArtsNotx C 1.-- __ VEi!)RCAT/04 Cf CCMNfEtt PatGUM l for' M.51V dATC 00/02 /MMcTLMD ANALYS!$ I I I I iI t 4 I !I i
- l. E d
i_ 8
- g ,
f E M 'O $MF%#f 8 8108*$ M Qgl4a44$#g H f I l$ hg
, .m r _- , , , , , .-,m.,_, , m , m ., , m ,,, ,v,,..., ,,
I K A LSI GNGIN EE A h'4G,1NC. D00 'd o -- o ast u.a.u tc:.,s..ts3s Sh ' 0- . E Ig r Fra LE_ cr ic_ CutNT ~'E -- - _ FROJECT lx4J 6,.L ! Ri llL.Ll! '??/ l~1Gh*/ILN.}- S!L!.LV ' O * 'l.4 *.1C # mtr AAto ay _.c._.CLWF I l f.L..FROJ NQ 33f _ DATE- , , , , ,
*4EVIE WE D 8 B)
/'I l.:._ISl '. I . " ATE d *l CA _ _._ ___.._.._._..__ ._ CATE _
dased v> thc </ >ath w del .v;d cf u t ?* 5 der.wd m ,4ppe.. d i a A c .? tai.s ref>tst, c .: ,, pl e /*Ce rtM c cmp:dce pecf m n, was cleve!cpo rJ h 'I perhnn f4e r>americc./ calculatina 6 -the NS/V pte guide ><nyact g hercf a n.n ;y.sls . 7Ais appe ,dee prw: des tA: et w p . J e. p m g.8a ,,, vahdd:m /vret110cert)m Jy em /m e,,,9 14 e a rial. n .5 tw att,.J;.,,,, hof4 0
'the tiend .<ha /dirra and ersnpdre ev' par ca y a fyp;cc./ case ca p h-f e ,( :n E~yure 2 cf de na-lv text l .
-C. l TG.S T CA$E DATA -
5 Kc /e ve- -Ink orn, ikxiLillly (ene . side) = - I a it.;,993 t y,,
.I 7 mm '
.6 t a ! ins e int te m Gs 0 a p.ccio2 3ec
$ = Shoe a ul rail confed ausg e i= J o* ( nea cle.s ty r, a,se..)
l 9, = cmisd p;n+ + /enr- /s ek a rm axa a,,<yle (q.s., ) - 9. s ' , l., u l.t?vi t- fo r.k a f m len q lb : 22.E io n . 4 5 . n = si,s a,, /~ . . i. a -,., - u, ~ ca.,,w n ,,, h L 3 .= % r/iu.I le , ee -h e 4 arr,, /ec9 Re22 s' i n u - c. a.m i., cu,,sim . .c g
<- .r e w a n, - ic +
to = Sczj m ent n e ;q /, r v- 2;7co Ib I I . . . . _ , . . _ -
. g-c s ; erne, wsw,merwxrw~ oa e 4. 4a I K ALOI ENGINEE AING, INC, D00' NO -
Ott G% & ANAns:5 COW.l ANT! *EU ND' - - - j F AGE _.i.C. OF SC l CUCNT 3C E FROJECT: f'14L131-.=M V ' ' A r' *
- r ? ' t !' 4:!D *J 6? 11 V -
~
C2. .d;tu DF
)
pan F AA ED B Y; lJ ., T., h M M I REVTWED BY; /% 1. @ ?I O AT E ; _.//, .' 3 d/ DATE /8-2 M/ O A. _ pnOJ No,; 2 ?e CATE: C 2 8/ M Pl t fled GQUA7/C AI S Ft R IME Th"S T CA J G
&nd on tAe equzth as cQrived d
, .A ppged,'y A , tjey w je i s-
- 1~
sj,n p lifa d & +/ e- s);ec , ! c f e st cea e S mie Cenvente nc e gf Aa nd alcdc hrrs. Y = S/oe p:ihen t, Y-axis (sfes n als) s % f V sc
= % rocc2. (in )
talterc Yu = prev sus Nire sfepi p;sihm A X = net le ver-lock chrm cleflec+im = fa,, $ a Y s y'
= D. Colts 4 7 - sk' (in) wbere su' t. sX in.ees,,ev+ dat k sqiaeir upwd mweme tt s Fa = rewis,2 impas acc (uide.s) = 2 v %
- i. '*cs ( t%-% )
E =Z^ #* 5 en 0 a(L5 ) i = 2x sX
- 9. 5:4 x op" ccs 9 (n (M'p )
= 26275 0 sX cib )
a hju e = 'Ing::e c>, le ve t'- lcc le a ern = a FR *d
= are t si ,,,( } - o, o )
-I a C. 63 s t CCsFg (it !L) l '
,-r w f*
= I,=N ( f h ~l j,oy g g f 4 0, c f., Qq rt ) CAY :
, s T<r q u e 4 :_
2
-! G 2 7# .d Tew u e ( /4 )
.. .. . . . . . ----:~ '"" "'
K ALSI ENGINEERING,INC. 000 No t t s sN & AN A0 5 S CONT.i'A s't REV. h0. . PAGE A.f OF *>C CutNT: -(E - - c: -+- - F RQJECT. J *l*u L ! !S ' C O' (! S '] h*l)' N ? N *b * ** fl b * ?!A** $ L*! W . , . _ dY !va fM I PREPAAED BY' REVIEWED BY: /1 I- # '** O ' p AT E _LL.2) - 4 /
- - -_ D A7 E O ' a' '"! .'_ C A-PROJ NO .2,,,2f DATE<
4 I' b bcc = $t.;1 mL*1bt C:ltltll<fi^1)
^'
t = f-n1
/
=. D. I$4 5 6 3 Ey I
(ih et')
.h 'f ' : Seh eneet vof,' cal dis):!cccm ea r & si n>cretocar i 1
=,, K .s t + r d e at --
I :. p . 00 0 2. Vo + <!.klt Or ( it1 ) where t), = pe ric.a o tenho.$ wh ciQ , i~ lm Ce 1 : ec .iin J ise a ii. n c.t: tic <&ien , L' /td '
. v. .
t sy = ex e n ve n >:. ~ du e h 5 9 m e. h ..,; uc e d m v m e, . ' L' ~
, _L a y! = g, g5 gt.rh sy' (A V = Ssp e , t a l. e ik, I = %4 0,st -
= V, + 0. C 0 0 2 Q u ( ^ lSe c .)
6 t 1 C2 &moiFe Cx77Pv7 VE2irl C:-71;I) Nhh he A //s y,riv y fa ble Ce napa re a f)e !rn pi:4 cvlpat wi r.In + f.e ha ad CalcalUIm S b vec:Aj N c a ? ( u e a Ct,' ! t!c ;!CiG IcFe:I !=h" TU U p . -) !a . n o I I - I Z c, . u~, u
.... .. ~.. . , . . .
..si "' '
r K ALSI ENGINEERING, INC. W: 'O - - " OE5 ON A ANavs ! COW.'av! m NO _ _ FAGE.f C 05 & CUENT. . __SEE f pnOJtCT. AWALHI!/Y< OLS.f;iSR12]ElLL1Lth!
- H!i / k"1* _f2:M L pq pay c sy; ._' G NAN 6 "A?f 3'2'*?/ F AOJ NC -. .Mt"
#4EVIEWC0 Br M 5 ' @)I m- C ATF _/li E 91. 0 A _ _ C AT E-
[
- j ' y,,g 08 m/'w Y j' df t. Q 4 is p 4 4 /cc g47 l a X' /
g/ N h a n sl l Al. (,n ) , lin) (ib ) ((i. ib ) (it 2 (om/pf) (, rt ) , (in) ' ( e ffe t) (;'CitZ Ct m p . 00 : lCoatISC) Ji7 20) 9.u 7 14 Cl D.; n [ herg. (s ** F .
- 0. C01 t,Ctu
- 0. c t ieff pre;}s?
it7 dd 2!> 7 i W, 7 ,_ 100 '
!4 l 0_ 0 C O' CH 4 ha d t.Ct4 o c C 25 t') ) ($3 1 4tf 415 I t '.1 215 Lt3 k11:t W4:Ost -m,jfl'pcm s re?) t tn a d b(n1 Q
[ Cen t C.f t i C CCI43f! 0 73 \ I A 0' ' fil Z!!? I td ' !!I1csts tteradR(L&_7]
/a.d C.C'S llCCl4)f 473 fl E ' Z ilf 4hG ll.11cn to f f get, riltatnf
~
0.0000 ~ Cs w fr 0 00$ l0.00444 C /ttl PC 7 ! 3ff!J Sff kl.!L7]I5 2 4froff 1%f Aos o en koon?4 ft7/ #7 : 3tv/ ] r(1 3 7 16 7 5i t s,& o.17+1
- e. ,1 c c1C k00:w) in </ 91r .Pt I
~
,' m *, e 7 % ev i' y p sam c. u te hand 6.t t 0 !0.00 SMM is34 977 as4f I (t-7 l. ,3yxiii3y(n,i4 /,at??
t%,n . l t cis \ roll!L 17c8 /iri S:!tJ 7t) 7 t rtriof t.>it tii? O.ta t t O' " I'- hand wc it >CCit!L 1748 ini Oli! 7?3 >ist nch !!7x$ r. inn 0.cel4 evn . c m lo ome 1902' aa sen e n :ia,e , w o n & ,9 'r g haed \ e.cla it itins. /10 2. ' tre t sslA i 952 'ica,4 46.m aiiY ut29 l ^"'g Cmr \ . 016 9.ro7cti I /?to izH l C7 t1 CE 'i ';.w ii"t ws ai;3lt. 7 c;3 h a n.1 i s.c/I lo oc7eig / m _ /c63 57to eH lin t o c4J,c3,,R s.7 en cu~;. \ c.ciP lo.coln 7 E10 \pfy d o n 1x & c. w l o, g ,3 . haad \ U lR la co7c2t 19b1lim s it i 1967 / tit ! C7tl ' ?10 M&stN :11 t soil o. % ll A cc20 l-- Co-9 \ c:0\tCCl741
- JACC i la ' f is:( tr4 [Mca to Ar ialth J. It 1 l A a nd I c o w lace (74/ /fe t I / 2. i/ . Ir:4 tr4 ::< r m9 /.74(rdt /./c 9 i c m f. \ c.cy s l .c:lito _tn1 i n q_r ! Sect c .cc2 .,- v u la maii"2.ccas e /.ao
; hand ) C D L' 9 C:EIT) 1731 ! ll c r ' rt 4 L 779 tst f a ri); tiax,i f. it s f 1 e n \cen/\o.cositt ins - 9ac 4i} 6v ',z.:itn, % nv;h i_cu hand l e c:d \c.te s'2 61 14H : WJS .:513 CC7 i . n ts , Qn g l,,T i. cy g c, cc a cen. c,.ou to can /ir7 7sr i 1355 57 1 13 era *1: s 7/ccM f. et 7 h a .,d . o.ctb b.tc4c1L tis 7 73 r l As ss Sy9 V ,,gs ,Mr s 7tm% i.rg, 0.C02)
- e. n I o n t kre u:s 7s ? , s ei \ zi u ess h u sia ,i11 n < asi ,. ;, ;
had \ c.c n-).e cun 7s7 \ vri l :314 ! $39 buv,5+!z nu,i! i. ni' cm. f. I a.0 3 0 ';cle s 7 251 /10 i T6 7 '
/fes 9 44%- c]; ;2 ,T,-{f,7t<f ~
C. c C.3 # ha 4 \ o,can Leetor7 !?o zn \ in' is & oci,; % g w ,,1, t.7ui s awe % cAcws Wt the rewin +k Wh e pw .ap.t a A Av:t cJe<JN::a are Mes ccJ scend a 6.> da+ p,,,is a i ,e t f;u14!\ c!,'jif f:a n s - C$ e r rW 0 cvm lem d co] w l.1ti.m s. 7/ s t ?!rs c , ile aceu r9 e f fJ e y eq rs.1 2 c duf . u .-
.. u . . . . . - . - . . .
_}}