1995:70 Crack Characterisation for In-service

SKI Report 95:70

Research

Crack Characterisation for In-service

Inspection Planning

Peter Ekström

Jan Wåle

November 1995

ISSN 1104–1374 ISRN SKI-R-95/70-SE S T A T E N S K Ä R N K R A F T I N S P E K T I O N

Swedish Nuclear Power Inspectorate

POST/POSTAL ADDRESSSE-106 58 Stockholm BESÖK/OFFICEKlarabergsviadukten 90 TELEFON/TELEPHONE+46 (0)8 698 84 00 TELEFAX+46 (0)8 661 90 86

E-POST/E-MAILski@ski.se WEBBPLATS/WEB SITEwww.ski.se

In-service Inspection lanning

Jan Wale

November 1995

".= .... r' ... concerns a the ~W~Ofllsr

Power Inspectorate (SKI). The conclusions and viewpoints presented in report

Abstract

During in-service inspection by non destructive testing the reliability is highly dependent

on how the equipment is adjusted to the specific object and to the anticipated crack

fea-tures. The crack feature and morphology vary widely between different cracking mecha-nisms and between material types, in which the cracks appear. The major objective of this study was to characterise a number of morphology parameters for common crack mecha-nism/structure material combinations. Critical morphology parameters are crack orien-tation, shape, width, surface roughness and branching. The crack parameters were evaluated from failure analyses reported from the nuclear and non-nuclear industry. In addition, a literature review was carried out on crack parameter reports and on failure analysis reports, which were further evaluated.

The evaluated crack parameters were plotted and statistically processed in data groups with respect to crack mechanism and material type. The fatigue crack mechanisms were classified as mechanical, thermal or corrosion fatigue and the stress corrosion crack mechanisms as intergranular, transgranular or interdendritic stress corrosion cracking. Furthermore, some common weld defects were characterised for comparison. The materials were divided into three broad groups, namely, ferritic low alloy steels, stainless steels and nickel base alloys.

The results indicate significant differences between crack parameters when comparing data from different crack mechanism/material type combinations. Typical parameter values and scatter were derived for several crack mechanism/material data groups, where the amount of compiled data was sufficient for statistical significance.

Sammanfattning

Yid ofOrstorande provning ar provningens tillfOrlitlighet beroende av hur provnings-systemet anpassas tin det objekt som skall provas och till de fOrvantade defektemas morfologiska egenskaper. Dessa egenskaper varierar avsevart mellan olika sprick-mekanismer och meHan de materialtyper dar sprickoma upptrader. MaIet med denna smdie har varit at[ kartlagga et! antal parametrar som beskriver de morfologiska egen-skapema hos de vanHgaste sprickmekanisrnfmaterialkombinationema. Kritiska parametrar ar sprickorientering, form, bredd, ytfinhet och fOrgreningsgrad. Dessa parametrar har hamtats direkt fran skaderapporter eller uppmatts fran foton over trvarsnitt pa sprickor

som redo visas i rapportema. U tvarderade sprickor ar skador som intraffat inom

karn-teknisk och inom konventionell industri. Dessutom har en Htteratursmdie genomforts med syfte art sammansililla data fran liknande parametersmmer och darutOver samla in resultat fran skadeutredningar fOr egen utvardering.

De uppmatta parametrama har behandlats statistiskt och grafiskt i grupper indelade efter

sprickffiekanism och materialtyp. For utmattnmgssprickor sarskiljs mekanisk, termisk och korrosionsutmattning. For spanningskorrosionssprickor sarskiljs interkristallin, transkristalli:n och interdendritisk spanningskorrosion. Dessutom har nagra av de

van-ligaste svetsfelen utvarderats fOr jamfOrelse med sprickoma. Materialen har delats in i tre

stora grupper, namligen ferritiska laglegerade staI, rostfria sill och nickelbaslegeringar.

Resultaten indikerar art signifIkanta skillnader fOr utvarderade sprickparametrar fOreligger nar olika sprickmekanism/materialgrupp-kombinationer jamfOrs. Typiska varden och spridning fOr de utvarderade parametrama har bestamts fOr alIa de datagrupper dar ett tillrackligt undedag fOrelegat.

of contents

Abstract. ... 2 Sammanfattning ... , . " ... 3 Table of contents ... 4 1 Introduction ... 7 2 Objective ... 8

3 Crack evaluation methods ... H • • • • • • • • • • • • • • • • • • • 0 0 • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • 8 3.1 Recorded morphology parameters ... 8

3.1.1 General data ... 9

3.2.1 Crack morphology data ... 9

3.2.3 Limitations ... 14

4 Results ... 15

4.1 Survey of compiled data from evaluated cracks ... 15

4.2 General description of crack morphology ... 16

4.2.1 Mechanical fatigue ... 16

4.2.2 Thermal fatigue ... 17

4.2.3 Corrosion fatigue ... 18

4.2.4 Intergranular stress corrosion cracking, IGSCC ... 18

4.2.5 Transgranular stress corrosion cracking, TGSCC ... 19

4.2.6 Inter dendritic stress corrosion cracking, IDSCC ... 20

4.2.7 Weld flaws ... 20

4.3 Orientation in surface and through thickness direction ... 22

4.3.1 Mechanical fatigue ... 22

4.3.1.1 Ferritic low alloy steeL ... 22

4.3.1.2 Stainless steel ... 22

4.3.2 Thermal fatigue ... 23

4.3.2.1 Ferritic low alloy steel ... 23

4.3.2.2 Austenitic stainless steel ... 23

4.3.3 Corrosion fatigue ... 23

4.3.3.1 Ferritic low aHoy steeL ... 23

4.3.3.2 Austenitic stainless steel ... 23

4.3.4 IGSCC ... 24

4.3.4.1 Ferritic low alloy steel ... 24

4.3.4.2 Stainless steel ... 24

4.3.4.3 Nickel base alloys ... 24

4.3.5 TGSCC ... 25

4.3.5.1 Ferritic low alloy steel ... 25

4.3.5.2 Austenitic stainless steel.. ... 25

4.3.5.3 Nickel base alloys ... 25

4.3.6 IDSCC ... 25

4.3.6.1 Nickel base alloys ... 25

4.3.7 Weld flaws ... 26

4.3.8 Uteraturedata ... 26

4.4 Crack shape and branching ... '" ... 27

4.4.1 Mechanicalfatigue ... 27

4.4.1.1 Ferritic low alloy steel ... 27

4.4.1.2 Austenitic stainless steel ... 27

4.4.2 Thermal fatigue ... 27

4.4.2.1 Ferritic low alloy steel ... 27

4.4.2.2 Austenitic stainless steel ... 28

4.4.3 Corrosion fatigue ... " ... " ... 28

4.4.3.1 Ferritic low alloy steeL ... 28

4.4.3.2 Austenitic stainless steel ... 28

... 34 ... 35 u..:.; ... "., ... . alloys ... . u . " .... , ... ,. or .. " Il' ... -'" . . . . " ~ .. '" . . . ~ " .. III " ., .. ,. ... ~ ~ .. " .. "' .. " ,. ,~ . . . ~ . . . " .. ~ .. " " . . . .. 1 Nickel alloys ... . 4.5.7 'Veld flaws ... . Concluding ... . width ... 37 1 Mechanical fatigue ... .. 1.1 low alloy ... 31 ... ,. ... 39 4.6.2 '-',JI"-'"'-' . . . 52 low alloy ... . Aul~telllitic stainless ... .. alloys ... .. alloys ... " ... 56 ... , ... , ... 58

5 6

"' ... "' ... rougmlteSS ...•... 80

1

The "''''IJ'''J'''''''~

as the nature structure under eXilWU.ml.!;lOIl

mance

the subsequent ... ,"""""'v .... and qualification stage.

~"DT equipment and procedure compo··

nent to type

straightforward

is known, optimisation with respect to defects can he problematic.

The main reason for this is that data not always are available as to which crack chara!:reristics depend on underlying degradation mecha.nisrns.

extreme """'., .. " orientation,

to ri"",","'n-n

surface roughness.

as as more

project are presented this report, which been the form of a Ilel()pmlent and

quali-data handbook can be used by NDT enE:RI-e~erS working with

is a """,,-,,,... ... n

want to em.ol:llilSHie not to atexmO'OK.

is ana

are rl{.~,eal:;a

wa.r.; to most common

scatter were deternrunt;;{1,

... ", ... ".. from a large u .... , ... V' ... FUfitller:moc-e, a conrpre:beIllsi for 1 0-15 years. ~,",",,",,""'AL a ... "' ... u .. , .... dlesCfl[l!tlOU. l1ffilH,aUOllS of

...,u.,'I-' ... was designed

... 'lJu.u as 1-'''''''''' .... '''.'''.

limited number of eX1JreSS11JllS

of n~I''l'!n,p.tp.r~

~Ol[fipleIl[leIU to the recorded

reports for this pru"OOlie

n-'.::I'U~.:l.U'LJU3 over

recordmg of the

morpho-and crack ... n ... h

mate-nmrlel)ers a

""'U'll-'U·'Y both the of

can

alternating thermal etc.

or cOlmnOn!~nt

on

thickness was also rec:orclea.

to a

centre

some eXl'Jre,SSlIJflS are

was

etc.

etc.

cold worke,d.

crack

is is to a

to is L

are

are

sha~ in surface same are as

Number damaged area.

was was was

Crack

Crack

Crec

direction

1, ru-e to a

stone is

amount

was measur~,a as a mean

Straight Winding Bend BiUnear

Branched Heavily branched

Zigzag

Fig. 2, Scbematic illustration of different types of crack sbapes.

measure. mea..;;urements are an accurate a mea-surements. SnlJill..G uate anmrnel~lC,u mean

of

ugtme1SS on a macro scale not

ona

was,

measurement length is range

length range of 1-2 mm.

:---mm. see

measurements. anl;;:.mipm. see

I

Where:

was recorcled at Su.rface, at hrut' can a mme bY=~'~L"'§ t\vo um~m~ss,

not

to 11 to a

radius: was it wa. ..

IH"""""..!.U;;'-to represent

a amount

3.2.3 Limitations

with such

investiga-" ... 'investiga-" u .... investiga-"investiga-" ... 'u ... ~.u. for each case. A importance.

information

none one

means

most registered cra,:;kS. The lack data shown in the presentation

results in 4.

A crack a three-dimensional UI.'.Ii.,;..;. ... ~. this do not cover

a ofa course

of mtc)rmaU()fl

over must H""" .. ""-Ih", .. ,,,

example, crack width at the surface is to be varying along crack length

on surface. The crack width measurements are made on micro graphs showing the

through direction. These photos are from a cross of

are

aelJen,aeEU on were

both

two UULl'i\U. .... me~CnaniSmS.

"'" .... "'''''' .... z;; was CH1SSlIlea groups fatigue 0/6 s 1 1120 two or more 1112 1/6 39/4 7 1 :I 1

Ferritic low aUov Nickel base anows

Low high temperature Ferriticlaustenitic .:nruJlu ... ».:'!I Alloy 182 steels

Alloy 82

4.2 General description of crack morphology

evaluation. Hv'''' ... ·n!

UltV .. U<lI,UJ.i)JlU is

If''-P'Cl'''' derived

typical crack leaturt~S are shown by micro graphs in

Appendix

4.2.1 Meclumical fatigue

by

"-"" .. ,. .... .,., .. stresses or .11.'"'''', ...

and corrosion surrounding emn.rc.nm.ent are considered insignificant

or red!isuibl!ti

see

the stress mtf:ns:i:tv factor

stress diI'eC1tlOJrl. an onentaticlD to con-is perpendicular to growth for fatigue

soot"ace rou~h11!ess mc]us,es with an increa~iin~ stress intt~nsttv

same as

deviates

ber addition, the

... 1[4 .... 8'"- c:racmig com.ull1red to me~h~Wcal .lQ,iJ~Jr.I,"'''',

thermal occur a

is IJi:I.'l,ll,;J:llll

some ~.oocts

pattern on the sur-or

see no 8 - 9

or tnermial

most cnlUill;;WJristlc features of C01:ros~lOn

Int,er~U"rulwa~ stress cmros;lOn ... g,.., ... , . . ,I;, A ... ' ... , combined en~~cts

stresses and COl!1'O;~101n. stresses.

of aus,ten.1t!c stainless steels.

stresses are ",,,,,,,'1" .. '\(11""" CrillCK:Ulg to occur. KeSU1Uw. stresses

important

causes faijltm~s

case to cracks. scale. are parameters as thickness UU'\.i~i.lVU morphology to

2.

more more common not it or in a coarse base not

of

mate~ .not

4.2.6

crficking, IDSCC occurs

structure.

are

logy some cases it

The reason structure nf'T\lTPf'.n

a difference 10Iligit1udinal dirloction cornDa:.red to transverse omectlOl1

in the transverse dendritic

same diIlection as the tle1J!dntes. an 1..U1oC11"""·'-'-~,Ul!;\"~ sm()om T:ypical examples

a'l"~'''''I'>''' no -41 Aor~endix2

Weld are included in the study comparison with service induced

to

joint surfaces.

Hot cmClaIlg

are fusion

,""u"',,,,"',"'''' induced cold cracks the heat affected zone next

are to

fusion between weld pMses also occur. Typically. they are

ex-and

nOl1-$t:ralJ~!lt. The most common

tation ·c:faiClO.!lg parallel to the direction, and

occur

cracks are located in the

centre. However, to

4.3

Thenun:mer

in surface

through thickness

1.

suo:ac.e orientation is relllu:eu to welds or ge()mt~tnc:u ,U.,"'\'I.U""·~,

on the

are one or

trolled geometrical features.

",n"~'O;t"p, on~:;:ntilti(Jln is

con-in the through tb14::kIJ:eSS di~ion are for only exc:eotlon

crack tends to follow the geometry a angle. No informatiQn were available on

were seven.

n . . .""""" are

cmmel;::tU:fn to tm~::K.ll,ess an~~les are 65'>

craCKS from the non-nuclear maustJt'V only one

Cr'dlCg are lm1:mk~

are is to steel were are at are no

due to the However. of the stone

pattern on seems to

pipes.

4.3.3 Corrosion f~tigue

4.3.3.1 Fenitie low alloy stsel

r<:i'l:I:,rolr~ were the nm'HliuclJ~r industry. The nmnrn~r

Ten are ,L'V'\<'U,,",U at were On~;!nUX1

parallel to the weld. Out the six cracks running parallel to the weld,

three are l00ttea coarse zone at a 0-1 mm

fusion the weld metal and one 5 mm away from the weld. The

OJien-but

O'ne

was

through tbtc~kness orientation but

are lOcate:o at

are are

across

were me:astlreil,

4.3.4.2 Stainless steel

Out evaluated cracks four were non-nuclear industry.

'Ha.!.U.l""!'!i'l> steel one each in terl'l11C ",~"uU''''''''' steel and preCIPItatIOn llaI'oo:nlD,g

cmme,cUcrn to

located worked austenitic snurue;ss pipes and six cra<:ks are from laIJ10nttoty

test spe:CH'llen.S.

.,,,1"l'!!'i .... '" orientation,

CnltCKS, even cracks which are

located is parallel to the The 'Orientati'On

m'Ore unclear due to lack of mtlilmultion.

4.3.4.3

600. Ten those are labom-is a

basft evaluated 0-is common four cases as laboratory tests. were

onentattcln in t h .. ", ... h thickness ri."'.,...~" • ..,. ...

two cases mont:. £Ui.VU;t1I;'U.UJA. is available on

cracks were are

mens. tnl'!OUlilm, ttucii;:ness angles are

a nmnug,'h .~u",..,.u .... '>" ,.., .. ""'~ .... """.

not meam.:ngtUl to

6

non-emfirc~rHr&enl:s at

growth

were seven

onem:amm was , ... h'~"""'!7",r1

are that of common

at cirl::amaferential

onl~nti!.t1cin C!\:!;pe:nas on

stresses and serls.Uisa:llou P"'~""''''.u t''"''''' ... ,''''

starts at surface then it turn~

cracking me;Chjlm~;mS

weld IGSCC was not

of

line.

1

cracks occur seven .... "w, .... "',

... w"'~ from the nu~:::le(U' i!lldul~trv and

two are

area. In two cases more

iJ.I..a,UvJUJ.UlJ;:; was rec:or{ied

ten

of

nOl:1-nUClea1' industry were

through thil~kness The shape cracks. one available. to the surface dinectl!Cm one two the

nOlll-naC.l,eaI maustlrv there no infoD'lllation

partly

measureaas

111m.

common

nents

stone p,uu~r:n.

'"'''''''VA''' stone pai:tern.

"",""",>"1.>-,,, bent or "'~"E>'U"~J

one pattern.

~vv'u_sronep~emA~'~'~

"' .... 'U,b"U "'.,a"' ... have, in mostca..qes, more five the damaged area.

rusltan,~e between cobble stone pattern was in cases

nnn.

4.4.3.1 Ferl'iticlow alloy steel

ten are ID

the through thickness direction. Four the cmcKS are more or branched. The number branches each are typically 1 - 3.

4.4.3.2 Austenitic stainlers steel

One nuclear industry was evaluated. The the through thickness no LAJU ... A . . " ' ... .,'vu .... 4.0"'".." ... on

4.4.4 IGSCC

more than ten cracks erist. Macroscopic branching was recorded five The number ofbmnches those cracks were measured. as 0.2 ~ 2.O/mm.

is

were recorded

011

cases, all

mlJo:l~il IGSCC-TOSCC

were three cases of pure Oral!lCnmg is not common for

evaluated crack

a

pre1Cipltail0n hardeIlmg stainless acm.ck

branching 10 brru:lCh1eSltnm.

evaJua:red.

!:':tt".'lIl:':n labor'dl:Orv

cracks

and ..."" .... "roe. cracks

The 'Vat'llatllQIl show

mat

straight, or

one {OSee crack that .... JlLu.<un

Alloy 800 and the renuulilmg in 600. no du]:erence in shape between laboratory cm!cKS and

pure ... ""-... ll .... eJ[Cel,t b:mnctresJ'm.tJrl. The num.ber cracks the damaged a:re,a

are two cracked areas Qh.""'t'lf1'n

"' ... .." .. "''!-,''' •• " test and one <!;lPf"""1/'"'P. UIUU\A;;;U

alloy .l.uu ... u.,!';ID mente;(! in one is more two pnrvp" or WUIUIIlg one 15 cases more

from the nU!::;le;ar industry, and 11 are

was recorded one crack L-;

Fig.

steel.

4.4.7 'Veld flaws

cold cracks and lack fusion aeI€;::cts are controlled by the weld joint

geo-the cold cracks follow an mtt~r~I:amtlar crack path and show a ia~jgeG ... ,'"

shape. The of flaws are located fusion line, which produce straight

occur as

are "'V~j'''''''''''''. each case is

occur as

Concluding

Crack typ~ Material

grOM}} Stainless steel Nickel base alloy TGSCC IDSCC 80-90 60-90 3) 2 1) <4 1) straight 1 - 3 1 >5 4)

Table 3, Typical crack orientation, shape and branching data.

1) Conirollied

"'V"""""."I or cold forming.

Lack of fusion flaws and cold cmcks are controlled by

4) Lack of fmdoo flaws and cold Cl."aCks are ... _____ to the weld

5) by

Commoo on a micro scale.

Number of macro c:racks ID ilie damaged anYd, dose to the evaluated. Clack.

no

occasionally

have a

tip are

a tip

cracks in austenitic SUiUUI;:;/SS

crack was .u","""''"' ... ,~

crack radii is shovm

a crack

radii ~ 5 other

are .---'-J transgranular. The crack

22 recorded crack

to crack occur

out 18 evaluated cracks.

One was

steel

""~ ... , ... , ... The crack tip

Crack

is

<1J.Un.

less 1

all cases

. . . 'AL . . . ,'''"' wt>oe:ars at the tip

stainless is S; 2 ten of tip 2 < 1 IlID, show radii 2 was two radius

one exc:C'PttOn. all

70 60 50 40 30 .Ill: ₂₀ (]I III

...

<.> 10 0 Cl} Cl} _Cl} (.) 'iJ) 9 'iJ) 'iJ) Cl}

gs

z

E i

_E (.) Cl)

z «

0:,; 0:,;

z

u

0:,; 0:,; u 0:,; ..c ..c E E {) u { ) <.:> (.) <.:> (.) (.) (.) _{li li}

5

'iJ) 'iJ) Cl) Cl) Cl} (J) _(J) (!) (!) f3. ..c S2 S2 ~ ~ G

rE

9 ::it ::it I - <.:> I

-measurements

at sm·ni(~e niolli',fI versus

at point I.JJ.V!LLvy versus rus!taD.(;e

versus dls'!:am:;e from

connc~ted. to

me~cb:amcal A"""4 .... ~ • .., cracks in +''''' .. · ... ti,,.

t11:l.(;KDleSS and crack ,,~ ... ,'n,,,, ,

and 2- respectively.

J.I;';.:lUUIl are nTje;sente:d

midway are LI,nnu.;.u versus crack J"i"", .... n',' .. 'nl

m(},[[f:a versus distance from

l l u .... ...,.,""' .... 1lO crack

are

in

mm

1000 1 I) 0 1-··· .. ···0··· .... · .. ···•··· .. ··· .. · .... f;}··· .. ··· .. ····,.·· .. · ... ., o o .loi1 10 0 rI:

...

_Cl (.) 0 1 0 20 40 60 80 100 versus ",r"",'OT 1000 Cl 100 n .n. 0 0 0 Cl Cl Cl Cl D o 2 8 10 [mm] at

HlOO Non-nuclear Cl Surface cracks ₀ 0 100 Nuclear crack 0 Cl .3:\ _iO (,) !!J

...

0 1

o

20 40 60 80 100 [%] venus

1

4.6.2

1000 Cl f Cl Cl D j~···a I Cl iJ Cl Cl I 0.1

o

low aUoy U .... ,UU'iU fatigue and 7-50%, resoec Cl 5 10 alloy were

are more homogenous,

Cl .".' 15 range mm to the measured crack HI"""-",,,, are larger and more sca,ttel~ed. is too

for accurate ,...ro. ... r-,,. _{versus distance to}

Cl o a o Cl 10 ~ ___ . ____________ .~ __ ",_. ______________________ J

o

20 40 60 80 100

}l'ig. 12, CriiCk width at surface and midway for 5 thermal fatigue cracks

'Versus crack depth/wan thickness ratio.

1000 r---.---~ Cl 00 Cl Cl .l!!1 1 0 fr··· .. · ... · .. · .. ··· .. · .. · .... · .. ···· .... · .. · ... · .... · .. · .... · ... ·· ... --... --... --... --j t)

=

...

U Fig. 13, o 2 4 """', .. ,..." at from crack tip.

130 g;:

...

"IC:$

-

--

-i 40 -. .l!!: (,) I!J

...

_'" 0 ₂₀

o

o

0.2 OA 0.6 0.8

Fig. 14, Crack width at tbn:!e locations for each crack versus distance

mm

depth/wall

thickness ratio 12·24%.

4.6.2.2 Austenitic stainless steel

The crack width was evaluated thermal fatigue cracks austenitic stainless steel.

The range mm,

5 - 40 mm and 3.25 - 100 %, respectively. results are shown in The

1000 100 ~ Hi u II:j

..

(.) 0

IIII! Weld materia! o Base material w ~~~u .. ~"~" ... ~ .. ~_~~~~" .. ~~_~~~~.~"~~~,.~~~~~~.~~.~"~".~~.~~ •• ~ .•• ~ .. ~,.~~a~_.~." ••••• ~ •• w*~~~".".~.~ •••• ~g."~_~~¥~_'~~." •• _~ •• ~ ••• n~.~.'~~.~'~~~~~.' 0

og

0 0 0 IIII!DD IIII! 0 0 5 10 15 20

Distance from crack

100 80 '0000 X oe . . o o o o o o ,

-..- -e

--

, -~" , _~~

-/ " ,

-

~

--/" ,~ ~'-'" 60 / ' # / ' ~ , A -=.

...

"

i 40 ..tI: / ' ... x / ; "

-(.;I (IJ ./ . ' "'

/~ ~ ~-

---

4)- -

-

- -

--

----~-

---

- -0

...

0 ₂₀

o

o

0.5 1.5 2.5 3

Fig. 11, Crack width at three locations for four thermal fatigue crack.s in mm.

4.6.3 Corrosion fatigue

4.6.3.1 Ferritie low aUoy steel

In 20 corrosion fatigue cracks :in ferritic low alloy steel were evaluated. The range

u,",~",.u. wall ratio was mm, 4

-1 - in

number of evaluated cracks is larger compared to the ",,..,.,'un,,, two fatigue mechanisms,

but scatter width data considerably The reason exc:eSS·lve

1000 0 Cl 0 ₀ 0 0 00 0 0 00 0 0 ID 0 0 0 0 0 0 0 .:.:: 10 u (ij "-(.;I o 20 40 60 80 100

Fig. 18, Crack width at surface and midway for 17 corrosion fatigue

I''W'~U''~:'" versus 1000 0 Cl 0 Cl 0 0 0 Cl b~ Cl Cl "' ~ ~ 0 p 0 DD 0 DO 0

~

Cl 0

_B

Cl rIl ID Cl ~ ~ Cl ~ 1 o 2 4 6 10

one

mm 12 10 4-, ... _'", .... /' /'

o

, / , / / ' / / 0.5

is

-":;;;.' '"Q .-'

..

-/ ' / ' / " /" ... , ... ... 1.5 2.5 3

Fig. 26, Crack width at three locations for one corrosion fatigue crack in

versus tip.

4.6.4 IGSCC

4.6.4.1 Ferritie low alloy steel

13 were

wan thickness and crack depth/wall thickness ratio is 1.0 ~ 23 mm, 5

-are In Fig. 21

surface and midway are plotted versus crack mcme,(] versus (11s1tarM;e

each crack is included graph. The crack width lll¥.lVQ.'~ inCjrea~;in2 depth/wall

Cl Surrac \') Midwa 00 1-... , ... ,' ... , ... , ... , .... ,., ... ,', ... ,'"" .. , ... {.l ... ," ... " ... , Cl Cl o o o o \') o 10 ~".,.-... -"" .. ~ ... ~'""~ ... -.~ ... ' ... ' ... - ... , ... ,''''~ Cl El o o 0 1 o 20 40 so 80 100

Fig. 21, Crack width at surface and midway for 13 IGSCC versus: crack

1000 0 100 0 p Cl [] .=

...

Cl 00 Cl Cl Cl Cl 1'3 i .lII: 10 ~ "~ ~

-

~ Cl P Cl Cl [] u ~

...

0 ~ Cl [] 1

o

4, 8 10 12

is mm, 8 are 80 60 J: ... "C1 i 40 .¥ ~ €C'

...

-,- --(.) ₂₀

-(j

o

2 3 4 5

Distance from crack tip, {mm]

Fig. 23, Crack width at three locations for each crack versus distance

were

4.6.4.2

Staifl.lefS

IGSCC cracks in stainless depth, 100%, steel The are

cracks are from the nOll1-11UClear

iOOO 100 0 Cl 0 ..ii: 10 u «Il

..

Cl (,) 0 0 Cl 0 tJ 0 0 ODD [jJ 0

D°

Cl Cl 0 0 0 0 0 20 40 60 SO

Crack depth/w£'Ili thiekl'Hi!1!38 ratio, [%J

Cl Cl

100

Fig. 24, <:rack width at surface and midway for 37 IGSCC versus crack depth/wan thickness ratio..

70 60 _, , 50 , ,

.

40 .~, J£ .... "0 30 i ",,-

.

.,..- , .,..- . / ' ..ii: U ₂₀ «Il

...

(,) 10

.

18' _.0' ... + / ' ",.. / ' " , '

-

-

-. -:;:-:--~ ,,~.- ~ .,,;::'

-~/. in

_.

-'

-?;

;,"'-0 2 3 4 5 7

D!st8lm::e from crack tip, [mm)

25, width at

'1000 Cl 100 o Cl Cl Cl DD Cl Cl Cl 16 20

Fig. 26, Crack width at three locations for each crack versus distance

4.6.4.3 Nickel base alloys

Twelve were wall and

depth/wall thickness ratio is 0.3 - 8.56, 1 - 8.56 and 50 - %, respectively. results are shown in 27 through 29.

1000 0 Surface 0 100 D ,£

-'j 0 ~ ₁₀ C.lI till

...

4'.> 0 0 1 0 20 40 SO SO 100 Crtill©~ t~lc:tne$6

Fig. 27, Crack width at surf'ace and midway for nin.e IGSCC cracks in Nickel base alloys versu"r.; crack depth/wall thickness ratio.

1000 100 0 Cl Cl D D 0 .9.., 0 I~ B 8 D DD i

o

at from crack tip.

D .'" Cl Cl 0 0 0 :2 :3 4 5

~=---

...

-40 -

-30 '" ."."" '" " /' .-20 -x· .• . x 10 . . CUI 1

Fig. 29, Crack width at three locations for each crack versus distance from crack tip. Four typical IGSCC cracks in Nickel base aHoys.

4.6.5.1 Ferritic low alloy sted

In aU, four TGSCC in ferritic low alloy were evaluated. in crack depth,

thickness and depth/wall •. u .. " ... ...,"'" is

are in Fig.

' ; c 0 (I _Cl

...

o .2 0 g 00 .'Iii: 0 eJ III

...

0 0 o 0 10 0 10 20 30 40 50 60 4 •.• -,,, .... ,-"" versus 1000 Cl 100 0 J:

...

"Cl

i

J Cl Cl Cl .'Iii: 10 t.) lII:I

..

0 J J

o

0.5 1.5 2 2.5 3

Fig. Crack width at three locations for each crack versus distance froOm

:Jti. u

•

...

0 350 300 250 200 150 100 50 0 0 :2

IJillltSlnCG from crack tip, [mm]

... ,...""iIJ,,, with ri""· .... t-I-.'"

3

thickness 5~16 mm and crack deptblwall thickness ratio 7~50%.

stainless

number of evaluated TGSCC cracks in stainless was range

crack depth/wall thickness ratio is 0.5-20 mm, mm

000 w 0 100 CJ 0 0 0 .I!i1 iO (,) «IJ ₀ 00

...

0 0 (',) i

o

20 40 60 80 100 ['Ye] versus 1000 0 100 0 0 0 0 0 0 0 0 0 10 .:a:: iJ 0 0 I:!I

...

(',) ₀ 0

o

2 .4

"

El 10 at versus

were a scatter

1

... ""'"""'"", one a

five are "'",.'n"'",,,,,, liiOUI:.:e.cl cracks. is no dHferen{3e _data

the two 'Tr;~"",(1 _{COl1Sequently all} are <UiUCQM..U as one group range

crack depth/wall 'WtUI~2>..UvO" is 0.5 - 3 mm, 3 - 7.62 mm

i"",')!1,.U1..:! are 37. HIO 0 Surface 0 Midway o 0 0 ~ ~ .= 10 0 .... 'g i 0 1 o 20 40 60 80 100

Fig. 35, Crack width at surl'ace and midway fer three IDSCC cracks versus crack depth/wan thickness ratio.

o o i

o

0.5 1.5 2 2.5 o o 3 3.5 4

~'ig, Crack widtb at three locations for each crack versus distance

from crack tip.

iOOO 100 s::. .... "D i .laC: !J ₁₀ IJf$

...

u Midway

Fig. 37, Crack width at sudace, midway and at crack tip for six IDSCC in

at were too

upper dashed scatter, the

are

'vVLU"/i:!>.iLVU U"~U:;lU'''''' data is considerable compared to the other two mechani..,ms .

... "" ... "'UJl.., ... and thermal stainless ,n,.,'.,''''_ see

depth/wall I:h1c.kness

In 41

along versus distance to the tip is shown. In both plots the mechanical and

reasonably data

Table 4, Cracli width statistics of fatigue cracks alloy

Number 6 I

3 5

deviation

800 100 €lOO 500 ;-... .;:: ₄₀₀

-

!_ ... 'I':lI "fi 300 ... ..:=: c I'll .... 200 0 t .. · - -... - . ... 100 0 -- _ ... _{f···· f····}

r

1

!. -- -)

-_._--

_... i- ...

Fig. 38, Range of crack widths at surface (s) and midway (m) locations of fatigue cracks in ferritic low aUoy steels.

400 350 f··· 300 250 .::;.

-'I':lI 200 'i t···

,

... .¥ c ₁₅₀ I'll i···

,

... ...

...

U 100 ! ... !-~- ... ---50 .. I""" f··· .. · ... _.. ---

---

---0

Mech fat • s Therm fat - s Mach fat - m Therm fat - m

Fig. 39, Range of crack widths at surface (19) and midway (m) locations of fatigue cracks in stainless steel.

Cl! tl <i!II ....

_...

::'I !Ill .... «> s:. .... "§ 'j JII: <C

"

..

(,.) Fig. s:.

...

"

i JII: <C I'll

...

(,.) 000 100 0 1 steel 1000 <> 0 <> - <? - ... ~ .. ~ ~ ~ .. ~ : .. "''-::''-~ ~ :: .• <>. -- - ... ~ " ~, 0 0 0 20 40 60 80 100

Cl'l'lIck depth/wall thlci:.nees ratio, [%]

~idtb at surface for fatigue crack..~ in fenitic low alloy

versus crack depth/wall thickness ratio.

o <> <> <> <>

..-

.1> -,,0 0 _ . _ .. _ . .. • _ .. • .0-100~"·~·.··""··~-~··~~":·~··r-'~"·'.~--~"C"o"",,·-···· .. D .... ···· .. ·· .. ·· .... ···· .. · .... -· .. ·· .... ··· .. -· .. ·~ 0 <> 10 0<> - 0 -Meehan!celi 0 ---e ... Therm~i -.0- -Corrosion fatigue 1 0 2 :3 4 5 S 7 8

@ u !Il:I

-

'= ~ lIIII .... _I;\J &.

....

1Z "it ~ e !Il:I

...

U .c _... :si! 11 1000 0 0 100 0 0 10 1 0 20 40 60 80 100 120

Crack depth/wall thickness I'l!Itio, [%]

at in ve:rsus HlOO o o ~ ... ~ 100 ~'·"·'·'·""""··"··"'···7~;···"···"·'"·=--··'·~·;·" ... . o o o 00 o ~ 10 mn~Cr···8c ... '" ••..•....•... ····~7···'"···"'··· ...•.•... ~ e !Il:I

...

U 1

5---J

o

4- 12 Hi 20

4.6.8,2

6-300

~I

, -250 .... , 200 150 , ... r -..lo!l ₁₀₀ (.j ~ i .'., ...

...

- -0 50 _- -r - - -

_-

_- r - - -

-0 r - - _boo::-=- ' - -

-

-

- _' -~- ' - - -

-

-~ ~ -

=

;::::;;:;:

Table Crack width statistics of IGSCC for low aHoy steel and steet at surface [J..IID.] IGSCC 4 100

inside the crack [

IGSCC

nickel base

2 1

260 20

1

Table 7, Crack width statistics of IDSCC and IGSCC for nickel base alloy.

4.6.9 Literamre data on crack width

measurements

pipe means a low was

mea-sured by ultrasonic or by testing, examination.

Statistics the crack width measurements and plots on crack width versus

depth/wall are shown Table 9 and 45 and 46, respectively.

measurements are

were

LfaI.J.A .... '-'i!I in two 14I"n,1",,'I"C' Ine!alIlog:rap,mc examinations

was measured both at

in at

the crack. there no COIrellittlclfi between crack

width. crack and pipe dimensions. a given crack depth, width is assumed to

be controlled by bending moments. applied and he suggest the

to presen.rea as a

water reactor

not are

cornp~trin:~ data

.... V'.i"'M .... JI.~'U of this comparison is that crack measuremems on me~uu!ogralJ!mc

sam-[5] and an good accuracy as measurements made

metal sun:ace

cracks were turn

on

to-gether with some data points from thermal fatigue in cast austenitic stainless steels and corrosion fatigue Type stainless steel. cracks were produced laboratory and the specimens used were of equal or almost equal thickness. The crack width

mea-surements were all made at the crack/surface The width are

re-plotted versus depth/wall values are

more

narrow thermal crack

corrosion fatigue cracks. may partly be explained by the fact that all the width

data DoctorlStenefjlill were obtained in laboratory where both loading and

environmental conditions were well controlled. The major part of the cracks evaluated in

are attack

IIi (.) Q

-

...

:::I 1111

...

Q .l:

...

:E ~ .Ill! I1J Q

...

0

stainless steels. Compar.ison between results of this work and literature data, [2], [3] and [5].

200 150 100 -... -...

--- --- --- --- , ... " ... 50 -

"--

.... - --- -

----_

... -

-0 {2J [5]

Fig. 45, Range of crack widths at su.rface of IGSCC cracks in austenitic

0:;"'''' .... ''' of [2],

1000 HiD 0 tJ

-

0 II:i .t:

_....

₁₀ o:JD 0 "a i 0 ~ ~ I'll

..

U 1 0 20 40 60 80 100

Crack thickness ratio,

at versus 1000 ,

,

0 0 0 ri 100 u III .'"' 0 0 0 0 u 't: ;::, 0 0 I@ ....

"

0 ~ 0 0 0 0 J:: ₁₀

...

~ "a i .iiII: (,l III

..

0 J. o 20 40 60 80 iOO [2].

.... Cl! .:::;

...

"'a

'i

.lliIl I\lI Cl!

...

0

....

_!::\I .:::;

_....

1:1 i .lliIl Q I!%

...

0 o o -+ -+ 0 0 10 -+ 0 0 20 40 60 80 100 Crack: thickness iOOO,---~ 100 <> 10 1 0 20 40 Crack: at versus

of

o 0 SO o o TtU!lIrm#i1 COl"rolllon Mechanicai 80 th!cknlillli [%] 100

HIO _., f·-~~--- ... --.- ....

ao

_

... c·-_.' >.-60 r' '., ... ' .... f-. ..

-.... !\'I 40 r' ~ ... - - - .. -.c

_....

"

'[it f-. ... , ... f·-.ill: ₂₀ 0 f-' r··· f·'···

,

... «flY

...

--,

- --

.. -,",,-() _r' --- ---,---0 J J stainless steels [6].

As mentioned previously the crack width strongly dependent on the stresses acting

pe!'pel:l(:H~cm;ar to When cutting out a

often is a:n]~u~a due to

or

crack width, and NDT rl"""'~n+...,.

CaIlrlea. out Brickstad in 1991 crack

closure was divided two categories. namely, I1P"r1Irp induced and crack gro\vth induced

"'J.'-,"' ... ,.., the by

hVldrCIStiLtic loads, or "'U'''''U~'''''''

as, weld residual stresses or residual stresses forming. Crack closure during

crack growth occurs to local plastic residual strain in the of the crack surfaces

"'."'..,""" ~ .... "".ul""~ ID a cirl;;Wlrnfe]rentiru

u"", ... ' .r,r",.,"",..., iV'''''' ... ID :a cmoollnterential

in service __ ~ .... ,.., shut

bent

CQJ1W1U0I1S at

stress

occur

a nuclear power plant.

positions along pipe

... "",.",.fh un •. ",,,,,,", ... v.:), ... was c!a:lme~\1 to be well iIr .... ,., .. ",...,

more difficult to pre~ru.c:t •... ~,J ... !F'.u. ~u'"'.uu, ... fatigue at LUUlkU.l§

where water different temperature mix. was found possible.

Considering estimations, the of crack closure must be accounted

where the

SlU.nlectto severe ti"lP'lrm;:!

Crack width According to

mechanical

are presented ';:!v"'I'1.VU 4.6.9. A reasonably

is 'I.',JI ~iCUJ.tJi.i3tt",U

....

_III J::.

_....

"

i :;,t.

"

fl'J

...

(.) 400 350 G@it!l. 300 250 2:00

L _

!.l

i' 150

r

I I-I ~~~~~~~.~ ~ ••• n~~~"'~U'~.~ 11 1 100

r

50 0

s:

_{~ E} ~ w ~ ~ _"? "'" ~ _J5 ~ "" m Z Z u "? "'-' "." ~ _{,~ ~} (,.) "'" u i$ "'?

.,.,

~ .;<! _Jij

B

u

8

_~

u Jl!

8

i

1

J

~

""

..§

&; (,.) V.I (,.) 0

J

~ <I!l "" _{~ ~} Ill) Cl _''"' Cl ₉

..

" ;:::; ,... on the surface of fatigue and stress corrosion cracks.

a:

~ "'"

_""

"., _"?

B

K

...

""

v.> 52 £a

4.7 Crack surface roughness

merumrements

on

4. '1.1 If'atigue

evaluation seven mechanical fatigue, therm.al fatigue 20

... ~ .. "., ... ' , ... ",,,.,-., are Table

.lau~"",",· ""u ... A.a show lower and less i:>'\.',",''''"",lVU compared to 'v, ... "'''' ...

fatigue Cl'acl.(S highest 4-values were measured mechanical fatigue cracks, while

evaluation fatigue thermal fatigue are

shown in Table 11 and Fig. 52 and 53. Mechanical fatigue cracking in stainless

a comparable roUg.!l1'1lesS to ferotic steel. One data

212 !lm, U"e.~'''''A than the the scatrer this data group

more sca1~r~~a to "'''''u''''',",

Cracks

7

260 200 1·· ... • ... · .. ···· ... ••··· •• •••· .. ·•• .... • .. ···.. •· .. •••• ...

-...

I!@ i:: ₁₅₀ ,-" ... 01

..

(» "E ... N 00 .."." .. I··· ... tt 50 ... ~---

₁

~ \) 1000 800 ... I· .. ··· .. ';' 600 c: ... 0

..

(3 'E

...

~ 400 .. .. _ ...

_

.... - .... -0 ...: 200 .. -.. ! .... I " ' ' ' ' -- -1 ~---.--l

Fig. 53, Comparison of correlation length, of fatigne cracks in

cases is .."h,",p''Mu·£1 below deviation evamiilClon of less 13 and

rou.g.!ll1!ess and

1

250 200 _ _ _ " ' . 0 " _ _ 150 f .. • .... • .. - .. 50 , ... " .... _{i .. ·"··} i ... _ • • • 0 . . . ..

---

..

--.--IGSCC-SS I(.;iSCC·Ni iDSCC-Ni

base alloys in Fig.

same data illUluns are compared 57.

in

300 250 ...., :;:00 ~ 1:: (I

..

!J

'e

HiO

....

IS m: tOO 50 0 ""

...

"" ~

...

q

i

~

_~

_~ 12 _~ :! _{""' tl} u u

1 1

J

""

_g

_'"

u W> ~

~

::2 ::2 Sf! A.4h

two areas rest were the lnItuenCe

11lttuelrlCe on se~re~!H IIletJtlO(lS to measure

are many .... ..,.UUU ... 'JA .... which depend on

measurement is .uu" .... " .. ""' .... '.5 methods are 1-''''''''''''''''''''''' measurement on a

a surface roughness or measurements on a

over a cross section The most common measW'es are and

surface measurements on ratl,gue test spt::cu:neIlS of

auste-stainless and ferritic low

median

derived

ness

25 20 -... , ,-... 'i' IC _H> 0

..

Q

!

'Ii 10 -a: I !i [1]-CS

Fig. All data are converted to

Ra

for

with dam established

exc~'Dtion is

roughness,

stainless the local data reported in [9] are mLlch

work. a median value 1 Jlm to

I

r

-

_I

~---f

are "i'\f1,'~"'"

agreement

lIi:UI~" MtWAI>:n :; and

HiO 7: I: C

..

u

!

100 Ii' a; 50 0 ] .s:;

!

I

] _] ]

~

₁

_""

~ it3 5: ~ u

j

"'t ~ ~

i

m tl ~

!

sa

i

"'"

sa

60~---T---·---4 0 -! ... o •••• o • • o • • ooooo • • o _ _ _ ._o.oo . . . . o_o_~

30

1

~M.~'~ ___ .~~ ... ~~_

I

1- -20 i 10

~IU

0 00 en <;I) """

t1

<n ~ ~ U <h U "'i' v" ~ ...

g

u

_B

u _Ji ,i Ji 'lll (1) ""'

₈

~ u

_~

_j

_{J J}

~ ~

_'"

... ~ y:; ~ ~ ~

-

sa

pe:rfo:rm:ing a ... <Lv accurate ""'1'"<""'" conditions. component !llDllelltSlGlns. length, along crack • amooot

1 2 3 4 5 6 7 8 9 cast ompOIlents in Piping aeu~cn.~:>D of plemelnt to 1991.

hhmUfic"H!«)gl Crack mecnaoi$m

Ml!tef~a! gr:ade Materia! cOl"ldmll!1

Crack dim@flsicl'I [Iltml

to .• tnleknes$ thickne-s Numool1' of cr~ci(s M aU::f(!scof»h: brallchh;t{l 10 Amount of (utidli';s Surhlc(;!! 1

I

Sketci'l R",fliilUHIC.e Crack iOClltio!1 M;mteriOlI group form Surface C·S Ilattem dilt~jlnce Micl'o-struchlr£, L {mm] Midway ₃

o

Crnek rndius Midway 1

o .... er crack location <ll"ld shape

\

\

\

\

\

Tip Tip

\

\

\

Appendix 1 A, G ImmJ 2 0 1

\

\

\

\

\ 30

\

\

\ \

\

I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I

SKI Report 95:70

Research

Crack Characterisation for In-service

Inspection Planning

Peter Ekström

Jan Wåle

November 1995

ISSN 1104–1374 ISRN SKI-R-95/70-SE S T A T E N S K Ä R N K R A F T I N S P E K T I O N

Swedish Nuclear Power Inspectorate

POST/POSTAL ADDRESSSE-106 58 Stockholm BESÖK/OFFICEKlarabergsviadukten 90 TELEFON/TELEPHONE+46 (0)8 698 84 00 TELEFAX+46 (0)8 661 90 86

E-POST/E-MAILski@ski.se WEBBPLATS/WEB SITEwww.ski.se