ML20012B735

Metallurgical Evaluation of Four-Inch Pipe-to-Elbow Weld from Plant Hatch,Unit 2.

ML20012B735
Person / Time
Site:	Hatch
Issue date:	11/30/1989
From:	Delwiche D, Toro Diana Diaz, Gordon G GENERAL ELECTRIC CO.
To:
Shared Package
ML20012B730	List: ML20012B729 ML20012B733 ML20012B735
References
89-178-021, 89-178-21, NUDOCS 9003160193
Download: ML20012B735 (18)

Text

ri .

1, .

"3 ENCLOSURE 3'

' FUEL &'PIANT MATERIALS TECHNOIDGY F&PMT Transmittal- -

l PRODUCTS &. SERVICES. -No. 89-178-021 DRF A00-03514.  ;

i r

i i

' METALLURGICAL EVALUATION OF-A-FOUR-INCH PIPE-To-ELBOW WELD FROM PLANT HATCH UNIT 2 (WELD 2G31-3RWCU-4-D-11)

' f November 1989 Prepared-by:

T. P. D16z-Plant Materials Technology i

Approved by: 11

.G. M. Gordon,-Manager Fuel &' Plant Materials Technology Verified by: A l'_^ J D. E. Delwiche Materials Technology Programs n

.h pf, w

a.

-- + . , , - ,

w .

.+

i o; 1,1 IMPORTANT NOTICE REGARDING CONTENTS OF THIS REPORT PLEASE READ CAREFULLY The only undertakings.of' General Electric Company respecting information in this document are contained in the contract between Georgia Power company and General Electric Company, Purchase Order

Number 89-6001259, and nothing contained in this document shall be construed as changing the contract. The use of this information by anyone other than Georgia Power Company, or for any purpose other.than that for which it is intended, is not authorized;.and with respect to any unauthorized use, General Electric Company makes no representation or warranty, and assumes no liability as to the. completeness, accuracy, or usefulness of the information contained in this document.

.[

2

t i

p. .

l 1.- INTRODUCTION Ultrasonic _and penetrant examinations conducted at Plant Hatch ,

Unit 2 revealed the presence of a 3/8 inch axial indication in a reactor water cleanup four inch stainless _ steel-pipe-to-elbow weld (2G31-3RWCU-4-D-11).. Upon excavation to a depth of approximately 170 mils, the length of the axial indication was reduced to 1/8 inch; but at this depth it intersected two parallel circumferential indications. The circumferential indications were

-5/16 and 7/16 inches in length, respectively.-At this point the a elbow was removed for replacement and a specimen containing the

• excavated weld was shipped to GE's Vallecitos Nuclear Center for metallurgical. evaluation.

2. RESULT 8 AND CONCLUSIONS ,

o The parallel circumferential indications detected by field ,

P.T. examination _were produced by the intersection of the excavated surface =with a crescent shaped weld defect. The defect was caused by lack of fusion between the root pass and ,

subsequent weld bends, o Extensive lack of fusion was observed well beyond the '

excavated area at a depth of approximately 180 mils from the I.D. surface, and it is likely that similar weld defects are present along the entire weld circumference. -

o The cause of the original axial linear indication could.not be ascertained because insufficient fracture surface material was'available-for examination after' excavation. However, >

considering'the poor quality of the weld it is very likely that the axial defect was produced by hot cracking during:

fabrication.

o Meta 11ographic crossections of the weld indicate that up to one half of the weld thickness (180 mils) was deposited in a single pass. The unusually thick root bead had a> ferrite

- content of 2.2 to.4.3 percent vs. 5.0 to 7.5 percent.for the balance.of the' weld, indicating heavy dilution with base metal at the root.

o- Metallographic evaluations have shown no evidence of base metal cracking.

o SEM evaluation of the weld defect surface confirms that the defect was produced by lack of fusion. Evaluation of the ligament between the excavat~ed surface and the weld defect yielded no conclusive results.

3

e o P.T. of the I.D. surface revealed a 1/16 inch rounded .

indication approximately 1/4_ inch from the excavated surface.

No physical,co.nnection or leak path between the root porosity and the crescent shaped weld defect was found, o since the weld defects observed are analogous to laminations in structural terms, their presence should not compromise the ,

structural integrity of the system.

• o' It would be prudent to examine a weld centerline transverse section of an additional joint to verify that the poor welding practice observed was an isolated instance, and/or continue performing ultrasonic examination to NUREG 0313 requirements.

3. ' DISCUSSION 3.1 Visual and Licuid Penetrant Examination.

The as received specimen (Figure 1) was approximately 7 inches long and consisted of the excavated weld, and equal lengths of-straight piping and elbow base metal on either side of the weld. '

The O.D. surface was clean and shiny and the I.D. surface had a matte finish typical of the red-brown oxide film found on stainless steel exposed to the BWR environment. The inscriptions l

i "BS6KM" and " flow 41nSNBSA"_were vibroetched on the O.D surface of the specimen, and three' linear defects resembling the letter "N" could be seen in the excavated area with the unaided eye. Since no azimuthal markers were present, the excavated area was arbitrarily designated as-the 90 degree azimuth, with 0 degrees being defined by the elbow extrados.

h Following the visual examination, the excavated region was liquid penetrant examined and photographed -(Figure 2) . The specimen was '

then sectioned longitudinally as shown in Figure 3-to gain acess L

to the I.D. surface . The weld root was irregular in places, and near the 70 degree ar.imuth, the root topography changed from.

concave to convex, giving the appearance that a repair had been o performed. P.T examination of the I.D. surface revealed a 1/16 inch rounded indication at this location (see Figure'4) which is approximately 1/4 inch from the begining of the excavation. The indication was caused by weld root porosity at this transitional L , . region and no other relevant indications were found on the I.D.

surface..

3.2 Metallocrachic Evaluation The 180 degree portion of the specimen containng the excavated area was sectioned normal to the pipe axis 1/2 inch on either 4

t 3:

e,-

side of the weld and four specimens were removed for 1 . metallographic and SEM. evaluations as shown in Figure 5._ A crossection of,the. weld approximately 1/4 inch from the excavated-region-(specine'n A) is shown in Figure 6. A crescent shaped weld defect produced by lack of fusion is visible at midwall in the boundary between the unusually' thick root pass and subsequent weld beads. The defect is confined to the weld metal and no evidence of  !

cracking in.the base metal was found.

Figure 7a'is a 2X magnification view of specimen D prior to mounting, showing the excavated surface with the circumferential i indiations and the thin ligament remaining between-the excavated surface and the weld-defect. Figure 7b is a crossectional view at the excavated region where only the base material, the weld root  ;

Lpass'and the thin ligament remain below the excavated surface.

~

j p In order to check how extensive the lack of fusion-was beyond the-

" excavated area, a macro specimen was prepared from a 3 inch long segment of the weld (105 to 180 ) which was sectioned normal to [

[ the pipe axis at the weld centerline. The etched specimen is

~

shown in Figure 8 at 4X magnification. Large voids and extensive lack of fusion is again visible between the root pass and  !

subsequent weld beads, which suggests that the entire weld contains similar flaws, p '

o The above results show'that the circumferential indications were li . produced by the intersection of the excavated surface with the crescent shaped weld defect resulting_from-lack-of fusion.

Unfortunately,-insufficient ligament material remained from the excavation to permit a metallographic evaluation of theJaxial indication, but considering the poor quality of the weld it is most likely that the. axial indication was produced by hot cracking f

of the weld metal =above the large weld' defect present in this-  :

i L region.

l-3.3 Scannina Electron Micoroscony (SEM) .l r

SEM evaluations of the weld defect surface were conducted on specimen C,(shown at:10X magnification in Figure'9) following separation of the remaining ligament-from the bulk of the specimen. The exposed defect and excavated surfaces are shown in

Figure 10. The-dendritic texture of the defect surface is typical '

L of cast structures where the solidification process has been L halted as is the case when lack of fusion between weld beads ,

occurs. No evidence of mechanical fracture or tearing was observed.

li 5

i c

L;<

I Figure 11 presents underside views of the removed ligament-

_ showing the oxidized dendritic = texture on-the surface of the weld defect. Figure-12-a is an oblique view-of the excavation. side of the ligament showing the sharp edged boundaries which gave rise to the' linear-indications. The arrow points to the location of Figure 12-b- which shows a 40 micron wide region between the excavated

> surface and the underside of the ligament. Unfortunately, since th4 excavated surface and the underside meet at such a sharp angle, there is insufficient surface area associated with the axial indication to make a conclusive determination of the fracture mode.-However, the flatness of the surface suggests that hot cracking of the weld metal might have occurred above the weld defect'and considering the poor quality of the weld in general the occurence of hot cracking would not be surprising.

3.4 Ferrite Measurements Bulk ferrite measurements were performed at several locations using a Ferritesecpe. In general the ferrite content of the root pass was lower than that of subsequently deposited beads

' indicating that dilution with the base material occurred. A differential in ferrita content between beads is often associated with weld defects, and the results tabulated below are consistent with the findings of this report.

DEGREE -FERRITE % (3 readinas)

AZIMUTH SPECIMEN ,

RQQT BAL. WELD 170 E 2.2 5.0 145 E 3.4 7.2 110 E 3.5 7.5 90 D 2.8 N/A 85 B 2.6 N/A 75 A 4.3 6.3 C

1 l

l I

l l

l l l l

l a 5 -

l

. r

. y. ' .' , . . . .% zn

.y .. +

. 1

. d ,, . i

. 'y$g ,. . . . - a ~

p

_. 3gi' .

..w .- ,- _

l ,

j l .

T

/ ,

~.

l ' Nx lN _. y

.- .. . . . % ;- . .: - r,

-~~.',,,.;..,w.'.W e t -

'$'l 'y, ' e g ,

! 'ff' h' - -

.l -

o

' t.?2 2),Y.L'b.4. ' - ' *l 'i* % 1L' S L h W W ' ; l d .

1

[htg

- ~ '

%. g.7 1?

\ . , .

..7 A,14 .. Br

.p . n . n.. -- ~~ + , fy;-

.q ;p ,~ & hfi% , ,l _-

e7 %f f r~ ~* '[;

4 '

'E h.;:.N i e#dd (f A g.

m. ,.; g .~e . .? + 4*. '9,4.:.'ser ,...%g.~ .'

1 A. EXCAVATED SIDE 0-180 l

l

.g. . . .. .. .

. . .a ~

1 i *

? '

u~

1

.# , r-s s' . l ~,agn. % %x% ;

e

+ ,

3.m -

........ j

-gx -:.,-c .

m... g ..n. s. , .

s  : '5c7 ~ V b@,

\ L ., .. .

l- _

3 Th.6,u3!'-

9, y 6 71 BI' k '.'05.*N. . I0tl,

>.- r c r i.? , s. . n .n m. . t E.,i...s.1,.u...Lis. e . ,, is a ..c . .

. . gw y... - J. -

.- ,.. .., g s

n.

e D. -

xa.-g

+. k.~,,~f

.. $ ,, , y,

, +

f:b

. _ h,g: (

$/*...^3 14 Q ds N wJS' .

m.

. (. . .

l l

B. UNEXCAVATED SIDE 180-360 1

FIGURE 1. AS RECEIVED SPECIMEN 7 '

l

\

l l

J j

[ ".

l' I

f 1

l 1

1 1

l i

9 k

• .y

'A g

e. - me '.- 4. . '

J

. . g.

v <+. ;-

.4Y1 ..

.s.

.h' S '

4 L h t

~

g

.y #

Y

k. , l . , , 'AT e -. ,

may n . . . . _,

-'l ._

r

^ '

~ r . ,, j _

}$

! S' " '

FIGURE 2. P.T. INDICATION ON EXCAVATED REGION 1

8

,4 p 4.th e4.wA.s ** nM esip (- 4 J " k-l I

l l

l t

6 l

l

. .j . . . . . . . . . . . .

. ' - 9 ,y _

. . - e J. g' e .

4 3 .. , q- ,, ., . .. .

' ~

e 4c. -, . ' ' *, ; ,

. g ,- . .

( 'g '.i fk# A.-  :

1

< .. x.

4 y',

-~

t t ?, -

I .,** ' .'

• w -

. . ' ., i ,

g

'* e t #*', 'j - g s ye l .. .. ..s.... . . . . . .

I FIGURE 3. LONGITUDINAL SECTIONING TO EXPOSE I.D. SURFACE 9

4

.g '}

-~~r

_g L g-4 ,

)NE' ~l4 5 :

~

r~ ~

.__y____

• ~ + Q, n t p i p .;.*(r,T i

%' A W *M , . .

?

W . p;. ,

%.- y;dk i': ' ' ; .h.1w -h*

.u,_  ;

.. $l; ' y , .. ..

-w,+

1. p.D g_'3,=.

t; .

y s.. h. 4 4

. t

. % b4 ~

.- .% 71 J

~* ' '

~

-A

- } gy _

?? ~~

,w ..

, m

-M

.q m . \

-Y

.. *%%;th:W. - .- . . - ..-S

[".

i

,~l *

[.y ' ( x ,-lk'- ?i 51

.y:f'.'f'

' - 9. y g *i E . ggt I Q , ,.y ,7 y - - ^

g,,, .

...c

. .~ *4: '4

-.. i ,

, f , -

r -

x e .u a - .c s ' . sg.

g . ..

1 f

I ,, , e .gm4,,

,;. c, ,. .--.

, g om, .. . . .

. w

• f.- 4;

p

• l, ' '" s l w.

• s-';.*.W,.

hi?7' 1 -

,q

, .,.s

+ 4 4., e .

[( 4  ;.q e  % 4 .

e,.,.h 'I .

uyy .,

yg,,,~ . : _ . -

-c am, y... ,_s

,7 ,, ,, y 1 xw \ + N, ,

,  ? + , cc %,, ,). _g y .f,"N. , -J .

4 , ,.

,- s. . . _ . .- .? <*.

^. ~ -- ; .m , ,

l' l 4 c

fQ. [%. . s .. ...;% ~'O;m ,, e. . .N - e7~..

.2 e. .:+ .s .? :

yfp

( .-

i 1

[ .~

• } =,, . Mi (a) I.D. SURFACE (b) P.T. INDICATION AT I.D. (WELD ROOT)

RWCU C--129 "

• WELD 11 V'25.;%

* ? .5 g a-_ g- _L. .

• -;._.- g.g e.9 -. .

1*%' - . . .Q:.

M;. ,:-v. ..

(c) < 0,iN,' Ms M -9[edT N YQ' ,',LM*(

LOCATION OF P.T. INDICATION

^+ -#n, " i.g ' " 'r L -

(WELD ROOT POROSITY) ,* - . #3_ ;. . . ^ ., ..

um ,4

n. a > m,

- +- ..

' Q. ,.4 9 99 % ~ , Q '? 4 's

.f.

}' &lAm {& " '

si' '

4* WWW N.,. hTW. 7 .

. .y.~ [, .

- . . s 4X

. . -. . .. .m . _ . .

FIGURE 4 I.D. SURFACE AND LOCATION OF P.T. INDICATION DUE TO POROSITY 10

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

1 90' 0* 180'

'f D  ;

-l

-l A B CD E l

r  % ~f  %  ;

I l

'l l$

1 A. B C- D

\" -

s. i;~ ~

11 *uosa 1 l ,.an n et

. 1.. .. 3. 2 v. .iraaa

. .. ..n

. t a 3. i. ..

lAIAJJAI4l4lJJAl4lAlAllilAIAIAldal lAlidJ),d,ljAliljalalalalli iljdelljall),dil),

I i. y , ;  ; ' v.rKui,4:fa':'<r 'c ! j_: .: s FIGURE .- 5. SECTIONING PLAN: A,B,D, METALLOGRAPHY; C, SEM; E, MACRO 4

r 11

.. , . . .. e 4,., :' ,,p

.-p .. . . < . ,.

't. ,',. .

< L. , . r .'

.gf *

.4 ou .

OXALIC ACID 8X FIGURE 6. CROSSECTION OF WELD AT 75 AZIMUTH, SPECIMEN A 12

i' i

l l

l l

1 \

\ l l

r l . -:

_g

• f' .

k, '

>G , t % .

1

, .y ' .

+; a j c

.m, +4 . :w

- : vg .,o ,

l

.y,

~

• jn

• 4 4, ,

.

• 4 7 ,, .-

• - .,_ i - .

".. ; 3,, -,

f . .- i y , .; ',,-

l

;y. < .'

t ,, ' . .

..- . 5. . +

QW ),l.... A:'

,g ' . g) * '

, ,j' . ..,r/, - ..$ h . s- Cf '~

l ii g y,n f J.

,1,*1--- .- -

~

'W'M'i r . , 7 q.3., g 7

!f

'[h-

.} . [ I .. ll@

FIGURE 7A. SPECIMEN D, PRIOR TO MOUNTING 2X J4 r.

n

-.# 4 . s ..

)y.,. . . . .

-.. g. :, :..

t~=:

y '

j.3 j.ys..,4<.

s.

. w.n*.m h,w a, e

.. , a 1.Q -

' c.

• W . 4;% ,4 '**",.;

g .. .

Y OXALIC 8X FIGURE 7B. WELD DEFECT UNDERNEATH EXCAVATED AREA SPECIMEN D 13

1

,. e - --

+.. . . . .

A ,4 -

?

. .v_ ~ ~

'i

.g %,

,e m ,. . .

. i' ,

.- t

. , .. g. .

. . .s ..

~~ .

2 . 9 .

. ~ . '

M

~

M~

. .l_,

.,A-

. .? . A 3 ._ . .

4. 'r . n g.

y ... - -:. , ,"!: J J.f

. ,,.s ( j

. p. .

• s- .

.  % e,. ' o, . R'. . *

%t*f,;< . . .= r$ L'-

e, .

b OXALIC 4X FIGURE 8 WELD CROSSECTION NORMAL TO PIPE AXIS, SPECIMEN

• E, 1050 TO 1800 14

I I

l l

l t

. _ ~ . . _ _ .

1; , , .-

Y '

s .

.a . ^^' ' y -

ef ' * .

~

V.,:;n. k R"~ W:[\'yh -

%' , . .# l#5 i  !

l a

V l

l l

. 10X i

A. EXCAVATED SURFACE

{ " '

s l 4' ~ .'

l 4 ,.

+. - av  ;$ , . ..

W p

. ' ' * ' ;', ,,'p*'g

&~ .

E?fny, ;. .,. c _ ,__ WELD DEFECT LD .4.... -): 1. w (LACK Oi' FUSION)

..i1E k-

k hf I,Y.b$

IEE B. WELD CROSS SECTION 10X FIGURE 9. EXCAVATED SURFACE AND WELD CROSS SECTION, SPECIMEN C USED FOR SEM 15

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

l i .

ik

} I W .,.,f* _p </ '

f,

• l [ . )  % {

y. :

. ,, ' )y

'N , , .

~.

5 t

l A. EXCAVATED SURFACE - DEFECT SURFACE BOUNDARY l

• 1. $, , k'(+ - p  ;

',W* -h i - $ ' *k.) j

.f

>T sky

} ~ fj}

1 C

, ]

'.48, a - -a .\ i N., 2 i.f g

%, =.:

yv ,;t 4 p; l1, 4'- tI x

a. .t* $t5% ;c

~ '

. ff' [Qj 3' &

. h* ~~

,m Ar

**

B. OXIDIZED DENDRITIC TEXTURE OF DEFECT SURFACE 4

FIGURE 10. SPECIMEN C, DEFECT SURFACE ON ROOT PASS 16

5 I' l .

, < l( -.f. .

h, 4[ g h{ O.sg}

js,fft ~ <,: y 4'% .f n 9

^?-

- r x '. _

q ' fy ": .

4- V y;dph.g R f,,f. % g , ,}

c -

,y ^

g .. ; .

gl .> k .;$8-,.L. ',+. ... -%:' ) _ Q .,.c,;y3 .'if-}t ynf .y .u a ,- w y

3 ,

G .-

, . > y- 3

+

.. r; t <

• 4

3. #7 . . . , ..l. .

> _5

.y, l

x. ... ,

p.

r

u l W ,

.s '{f\} t 2

> . e z. s . j y

, . " ', -  ; . . n >. %:, l !},

, s . , f v ; . - ..,

9. ss 8e . <

- , s.

. . ' ' '- .e ] .

~

,  % {.

• Q, ,

h 1.. .. . .

..f*~*f .

i l A. LIGAMENT UNDERSIDE l

' '~ ~

we

'l i 7 #Q w.

9, *+

• W -Q 4T

#i '

3i ag -

e3 ( !** '*

, 22:

l [ $ .( .

,k. . _ .

41

..Jty%p.y)a~

-:g w+5 3 4 gd i  : T

• lfl Qf(a. *. .

4

?

l fI

~

1 p*[- :4

..p

. * .~ .

j n y#g Y-s j Q', -- ' d y .g W 44duc$w.

e'a a'2. ,J

.A l' b. e274 1&n. 5.

B. OXIDIZED DENDRITIC TEXTURE OF DEFECT SURFACE FIGURE 11. SPECIMEN C DEFECT SURFACE, LIGAMENT UNDERSIDE 17

.. . . ~ -

r, T

-.g x .

f r

l-t' .

i s.J .. . . '

4 J

4 y . . . . . . . . .

._-.. B Y .,

s

~

c ..

f'

+

2147 2H U ge5 i ng gD37 :

A. OBLIQUE VIEW OF LIGAMENT EXCAVATED SURFACE l

l -gggg/ ,

%-  %* t J

( s1

!  ! pf 5+ \ ,

fj I

'?

,: j }' j .

l. -

A;. A . pf j _ ,, 3

% +. b?a;';

1. .k fl.:.g( g)n .%

y  :; ":n [ :_ a f. :tf'pd ,;

g '- o, ',!?g - . -

.,\

i.

f/ .

I

^' ^

h i $442- $ p , th @ , [ gaar l

B. REMAINING FRACTURE SURFACE OF AXIAL DEFECT t

9 FIGURE 12. EDGE VIEW OF LIGAMENT 18 l

.a