ML20058K149

Paper Entitled, Fatique Behavior of Nozzles of LWR Pressure Vessel Model

ML20058K149
Person / Time
Site:	Davis Besse
Issue date:	06/25/1978
From:	Kodaira T, Miyazono S, Ueda S JAPAN
To:
Shared Package
ML20058K143	List: ML20058K140 ML20058K146 ML20058K149
References
NUDOCS 9003020268
Download: ML20058K149 (8)

Text

1 Mi,,

~

~

~~

~

4

't,4' L Ms -

E(

h Dockst Numb 3r 50-3 v

!.icense Number NPF-

~.-4 Serial Number 1768 c+

g#

)

Fatigue Behavior of Nozzles of Ught Water Reactor Pressure Vessel Model i

l S. MlYAZONO, S. UED A, T. KODAIRA, K. SHIBATA T. ISOZAKJ, and N. NAKAJIMA j

Japan Atomic Energy Researen institute

.j Tokai Researen Establisnment Tokai mura, Ibaraki ken. Japan 1

.i ASSTRACT ttu j

In this paper, the expenmental results of a cyclic in-

,p ternal pressure test are descnbed on the pressure vessel

~

model.m which three kinds of nozzle models were

,,_. gg g a{

welded, in the inner corner surface of each nozzle two kinds of art:Reial cracks were machined perpendicular to

'[-

the direcuon of maximum circumferential stresses, which h,,,,4,,,,,,,,_,,,s_,,

occur at both tne head and bottom sides.

5 Cyclic intemal pressure was given from zero to 10.8 MN/m by otl after;the stram distribution of the inner i

g,

i and outer surtace ot ear.., aozzle was examined by static N

e=s internal pressure of 10.3 MNimi. In the cyclic internal pressure test. the fatigue crack length from the artincial cracks at any pressure cycles imposed was measured by Fio,1 Cross SECTroN OF NO.5 PRESSURE VESSEL MODEL an electrical resistance method and crack gages.

INTRODUCTION model was designed based on the criteria of Sec,!!!, AS-In the assessment oiintegnty for light water reactor ME Boiler and Pressure Vessel Code (4) and nearly 2,000-

! pressure vessels,it is very important to study experimen, mm length.1,000-mm diameter and 23 mm thickness, as tally and theoretically the trutistion life and propagation shown in Fig.1 while three nozzle models were respes behavior at the inner corner surface of the nozzles where tively selected from the nozzles of the primary circuit of the maximum circurnierential stress is induced by inter.

boilin(water reactor pressure vessels which are being nal pressure, n'tmmreastructed or operated in Jap:n. In the inner cor-In the Japan Atomic Energy Research Institute, cy.

ner surface of each nozzle two kinds of artificial cracks -

clic intemal pressure tests have been so far performed to (notenes) were machined perpendicular to the direction expenmentally investigate the iniuation life and propaga-of maximum circumferenual streras,which occur at tion rate of crack of some artincial cracks at the inner both the head and bottom sides.The artificial cracks corner usmg nye steel models oilight water reactor pres.

were machined by a thin gnnder of 0.3 mm thickness.

sure vessels [1], (2), [3]. In this paper, the experimental and 30 mm diameter as follows: one was of nearly 20 results of a cyche intemal pressure test are desenbed mm surface length and 3 mm depth with a straight crack

' mamly on the fifth pressure vessel model,in which three front ( A type), while another wu ornearly 8 mmsurface kmds of nozzle models were welded.The prtssure vessel length and 3 mm depth with a crack front of a circular 9003020260 900220 PDR ADOCK 05000346 N

P PDR

y 1

}^

6

^

i arc (B. type). Cyclic internal pressure was given from zero signed based on the enteria of Sec. ll!, ASME Bouer and to 10.8 MN/m by oil after the stress disttsbution of the Pressure Vessel Code and.three nozzle models were re.

1 a

inner and outer surface of each nozzle was obtamed by -

spectsvely designed being selected from the nozzles of -

1 static internal pressure of 10.8 MN/m and then each ar.

the primary circuit of the pressure vessels of boding 8

tificial crack was machined. In the cyclic internal pressure water reactors of 800 ~ I,100 Mwe grade,which are be.

p test, the faugue crack length was measured by an elec.

ing now constructed or operated m Japan and they were -

K *trical resistance method and crack gages, nearly 1/6 scale nozzle models of their prototype. As After 09200 eveles of internal oressure, tnis test was shown in Table I the shell was made of a low carbon

'l C

-'J[,st3pped unce one ot'six artincial cracks oroparated to steel. ASTM Type A30? Oisde C and the nozzles wers t

/

_the outer surface and od leaked from it. From the experb made of a forgmg steel ASTM Type A336 Modified.The j

mental results,any remarksole differences of crack initia-cross secuens of three nozzle models of N1, N2 and N3.

tion were not recognized between the A.and B. type are shown in Figs. 2. 3 and 4 in detail and they were aruSetal cracks and it was determined that the propaga-welded at 45*,135' as ?25' of the A.A section of the tion rate of both cracks was the order of 10~# mm/

pressure vessel model. 2esenbed in Fig.1, respecuvely.

cycle, i:'

STATIC INTERNAL PRESSURE TEST:

PRESSURE VESSEL MODEL-The stram distribution of the inner and outer surface The pressure vessel model used ir, tms expenment is

.of each nozzle was exammed by stram gages when a J

shown m Fig.1.The chemical and nechanical properties static intemal pressure was given to the pressure vessel -

of the materials of the shell and nozzle are desenbed in model before two kinds of arttiicial cracks were ma.

]

Tables 1 (1) and C). This pressure vessel model was de.

chined at the inner corner surface of each nozzle.The

-1 Table 1. Performance Tesung Results on ShellMatertais of.Vo. 3 Pressure VesselModelil)

Specuication l Extractea Yield t.lltimate Elonsauon Reduction Charpy %

Position Strength Tensde of Area

' Notch Impact (MN/m')

Strength (f.)

(f.)

Test t

(MN/m )

(N.m) 8

^

130' f

343

$49/686

17 hoduction R = 1.23 t (at -12' C)

No.

longitudinal Top Aversee 110 4K 1623 t/13 i-

!!$.106.110 587 683 20 Good Transverse Aversee 60.8 60.8 J 7.9.63.7

. Bottom I

ChemicalComposttion if5)

Shell Elements

'C Si Mn P

S Ni Cr Mo ASTM Specifdicauon f 0.00 0.15/0.30 1.15/1.30 0.033 0.00 040i0.70 0.45/0.60 A302CradeC Ladle Analym 0.19 0.29 1.28 0.014 0.015 0.57 0.31 742

1:

W 1

i, 3

a f

s 1

~*

R i

4p.

?

+

. 1

-~'f

. Table 1. Performance Testant Results of Nottle MateHals ofNo. 3 hesmre YesselModel(2)--

Specifiesuon Yield Ultimate.

Elongation Redumuon Bend Charpy V..

]

Strength Tensile

..' Arca Test - Noten impact o,

(MN/m')

Strength

(%)'

(%)

Test 8

(MN/m )

(N.m)

L l

Average 33.3 Matenal Minimum 26.5 l

]343 549 20 (at -12' C) number 45 E 253 Average 155.9 2/325 453 608 27.9 69.4 160.8.143.2.163.8 -

O *mscal Composition (fo)

Nostle Elements C

Si Mn P

S Ni Cr Mo-N1,N2,N3 ASTM A336 modified Specincanon 0.27 0.15/0.35 0.50/0.80 0.040 0.050 0.50/0.90 0.25/0.45 0.55/0.70 -

ASME Case 1236+

Ladle Analysis l 0.15 0.32 0.69 0.008 0.013 0.88

.0.32 0.62 ye

)

i P.

v sy,,

si 7.-y.e i_-

25 g

J h.

p E

f,

5

,a m

Q l

'* J,

";OM g

y n

1

- -- -F j

~

.es.

(

_L p

y,

)l

1. , - i 3

i q

- +

w Nv;9 f

p,., yfA ^**.., e f, 5.a gl?

A*

1

.i i

i.f

}-

. i.j nr.

a m.

res.

FIG.2 CROSS SECTION OF NI NOZZLE MODEL FIG. 3 CROSS SECTION OF N2 NOZZLE MODEL s

2 static internal pressure test was carried out step by step and Poisson's ratio were used at 2.06 x lo MN/m and to a maximum internal pressure of J0.8 MN/m which 0.3 respectively. The stress distnbutions of inner surface.

2 was selected as nearly 1.5 times of operating pressure of of each nozzle are shown in f:gs. 5,6 and 7,in which the boiling water reactors, in this internal pressure test a experimental results of N3 nozzle are compared with the j

kind of machine oil was used as the pressunzing matenal.

calculated values and they are consistent comparatively All the stresses of the inner and outer surfaces of each as shown in Fig.7.

nozzle were calculated from the strain values ootained The circumferential stresses of the inner corner sur.

expenmentally by strain gages with an assugnption of face of each nozzle are maximum and their results are elasti stress concentration, where as Young's modulus denoted in Table 2.The stress concentration factors of i

743

'h

,)

gt 4 s l,

m.

58

t l

f

,y'!

w

,a'

.h 4

,(

j i

g.

il

,/

1 1

i

?

i o-izee q

[

c:

e tt 7

-'06e.m

,,, o g y

\\

).,

22ee \\

~ g.

tile

~

)'

. i c-f

+ i N

/ oven f

h I

, g Soo i

y" w.*

• 800 P e t0.8 MNMI

......M...

_ i

.l.\\

apge s,

{300 t

• Moos strees FIG 4 CROSS SECTION OF N3 NOZZLE MODEL g,,,,,,,,,,

1 i 2xe,,

O i

100^

~._

_m

~

e V

0 C SO 100,00 200 220mm

=

FIG.6 STRESS OtSTRIBUTION OF INNER SURFACE OF N2 NOZZLE

- l b~

~ ! ll

/

,,,e_.

e 4

e-622 9 gl

-- zue-s e

\\

4 x

i-a4 e i

i 172e i c i

--iod e.,

3 i

[ \\

,0 e l i

i J

i (

k' s 2

N l

Fl 1 I

$00 l l.

P e tO.8 MN Ant 500 Iace

• Moos strees g

/ ;l

- a s a.,.

ex

$ {,

~

')

ace s

/ y,

p. e.e uNm a

a, \\?

n p#

300

',).

w 5, s >

,b, n..

v, Aai..,e.

~

,2,

,M.g

~ Cateuwted Wiue

.,e,dl I

,,, i m ay.-l[ gi f b t

~

0 0 So 100 iso 200 250==

g g,, g,

FIGS 5 STRESS DISTRIBUTION OF INNER SURFACE OF N1 FIG.7 STRESS OISTRIBUTION OF INNER SURF ACE OF N3 NOZZLE NOZZLE 9

a e

2-

s t

l Table 2. Sires COncentranon factor of

!/

Inner Corner Surface of Nonle ll

//

l w.Ml i....

, u; j

NI N2 N3

,e g

4

.:' /

{

3

f Cucurrierental $ tress of

/

Innet Surface of Shett 20:

02 to:

~

(MN/m')

$co l

Cacumferential Streu of

)

v 4

Inner Corner Surface of Notste 493 856

$17 rMN/m')

'o -

. Stren Concentration Factor of inner Corner Surface of Nossa

.3

.3

,.6 o

woo

'como

+>oco cocco esooo wooo Nwneer er emas.N FIG.8 CR ACK LENGTH $ ROM NOTCH TIP OF N1 NCZZLE I,

A TYPE NOTCH

. 'the inner coiner suriaCe of each nozzle are 23 0.3 and 6 of NI.N: and N3 nozzie respectively as shown.n so Table 2. It has been determined from thest expenmental results that three are not any remarkable differences of g*

stress concentration factors oi circumferential stress as among the three nonles, g"

CYCL.lC INTERNAL PRESSURE TEST After the static internal pressure test, m order to in.

f ec vestigate the effect of crack geometry on the initiation life and propagation rate two kinas of artificial cracks, j

A type and B type,were machined. These artincial cricks are depicted in Figs. 2. 3 and 4 respectively. They were machined at the symmetrical places of the inner corner surface of each nouie,which were the Gange and bottom

' *** y" ',, g '"*' *****

ude et the pressure vessel model.

,, g,,. CRACK LENGTH FROM JOTCH TIP OF A DiMECTION Cyclic internal pressure was given to the pressure (sHELL 5:04) OF N1, N2 AND N3 NOZZLE vessel uodel with a cycling rate of 5 epm from 0 to 10.8 MN/m.The fatigue crack length from the attincial 2

cracks at any pressure imposed was measured to the nozzle side respectively. it is seen that the crack propaga.

direction of the depth (0 = 45')by an electncal potential tion rate of a. and b. directions are higher than one of c.

method and at the inner surface of the shell(0 = 0*)

oirection except for the eart:et stage. in Fig. 9, as an ex.

and nozzie side (0 = 90* or 9 = 10$*) by crack gages, ample,is denoted the relationship between the crack whrte d is denned as the circumferential angular co.

length from the notch tip of a. direction (shell side) of ordmate, as shown m Fig. S. The cyclic internal pres.

NI N2 and N3 nonle and numoer of cycles.obtamed sure had been applied to sne leak +;c of oil from one about B type notch by crack gages. As shown in this Og.

of six cracks.

ure, there are some cifferences of the crack propagation The oilleaked out from the outer surtace of A type rate among three noules of N1,N: and N3. Figure 10 notch of N1 nozzle after cyclic internal pressure of shows the expertmental results of the crack length from 29.000 cycles. In Fig. 8. for example,is shown the rela, each notch tip of the a, b. and c. directions, which we;e tionship eetween number of cycles and crack length obtained by measunng the cracklength of the fracture from the tip of A type notch of NI nozzle.The crack surface of NI.N: and N3 nouie respectively after each initta' ion life of c. direction was smaller than gne of a.

cross section was cut from ese ;oule part.Tht c ack and b. direction,whiuere the mner surface of shell and length of NI noule is the largest for both A and B type 745.

m

.. (

1 I

4 Mork l Noter i5re7c 70 type i tien a

i A

o I

A I

8 i

O A

y A

g

?

g

\\

z O

A j

j i

e i

L gm e

o e

2 A

= C' 2 %C p

-Noten b

j 5

j d

a o

g j

e i

30 3

J l

e J

J to N1 N2 N3 l

Nozzle number l

FIG.10 CRACK LENGTH FROM NOTCH TIF OF A.,5 AND C.

C DIRECTION OF N1, N2 AND N3 NQZZLE c

l.

2 g

I notch,while one of N2 nozz!e is the smallest for both fractographs,which were taken at four locations of the notches and one of N3 nozz!e lies between Ni and N2 surface of fatigue failure. As denoted in this figure, the '

neules. it can be presumed from Figs.5,6,7 and Table stnations caused each cycle by cyclic internal pressure 2 that the crack propaganon eehavior of the nozzle is can be observed from the fractographs of S2, S7, S8 and dependent on the stress distnbution and stress concentra.

S10 respecuvely. In Fig.12 are compared the expen.

L

- tion factor of the circumferenttal stress in its inner sur-mental results of crack propagatior rates which have l'

face. As one reason,it seems that the stress concentration been obtained by an electrical potential method, crack l

- factors of the N1 and N3 Aoules are larger than one of gages and measurement of striation spacing for a. b and the N2 nozzle. As shown in Fig.10, the crack length of c direction. The crack propagation rate obtained by A type notch is larger than one of B type notch for each measunng the stnation spacing of the fractography is nozzle of N1,N2 and N3 It is supposed that this differ-consistent with the rate obtained by the electrical poten-ence depends on the geometry of both notches.

t:al method and crack gages, as shown in Fig.12. The

!n Fig,11 are shown the macroscopic appearance of crack propagauon rate was the order of 10-3 mmicycle fracture surface of N1 nozzle (A type notch) and its and increased with the crack length, 746

u a

a

-m.-e A

2-A

..,u, 4

)

we M e

5 10 e

.3 mm

\\

"~

tossa 52 37 58 w-i..

0.01 mm w-

}I

-c 1==

.,aEsLAFACE08N1 N O 2 ' E A T Y P E N OT C - ' ', ', ~ 3 = a ; ;-, c m a p = 5

i C

)

g 1

4) The crack prop:3 tion rate of nozzle obtamed by e frec'ien,,)

the electncal potentialinethod and track pages is con-i 9 ',,,

0 8 sis'ent with the measutt ment of strianon spac.ng and 1

the order of the crack p.'opagatio. rate is 10~3 mmicycle kj under the ermronment of otl and room temperature. The l

D eL expenmental result is ;onsistent with the results which g

were obtained by the cyclic internal pressure tests of the j

3rd and 4th pressure vessel models [:1.[3].

l 7

3*

-w

{

{

ACKNOWLEDGMENTS t

t*

The authors wish to acknowledge Prof. T. Udoguchi.

i

} l,7 Prof.Y. Ando and Prof. X. !ida of the Uruversity of j '* "f

- Tokyo and Dr. T. Fujimura. Technical Research Associa.

tion for Integrity of Structures at Elevated Semce Tem-l peratures (ISES) for his helpiul suggestions and discus, j

'o ao so ao so ao sions to this subject and also thank Mr. T. Ohba. Mr. R.

' Croca ientra tem noecn vio **

Xawamura. Mr. M. Matsumoto.ind Mr. D. Song. the sio.t: :mAC< emoracATioN RATE N1 NOZZLE CROSS DMa Momtc bugy ban.5 Insutm ad W. K.

IE: Tion (A TYPE NOTCH)

Hashimoto for their collaboration and asustance.

j CONCLUSIONS REFERENCES j

1 The conclusions obtained from the experimental re.

(1) Funmura. T., Myarono. 5.. and Ueda. $. et al..

sults of the static and cyclic internal pressure tests of the anctutsnesy raca pagayson in Ovalaid Norslu of a j

Nucleat Rsactor.hocernant: or the ist Internanonal Con.

fifth pressure vessel model are as follows:

/erence on hetsure ressel Technotoey, Delft. Netherlands.

!) There are not any remarkable differences of stress september :9.oct. :.1969. Part !!. pp.1:131::0.

i distnbutions and stress concentration factors of circum.

[2] Funmura. T., Miyarono. 5.. Kodairs. T. and Shibata. K.,

ictentlai stress at the inner corner surfaces among three

" Integrity Aucument of StructuralModels of a Reactor Venel wh NotchH Nottin undu Cyche heuun LoMang".hocud.

1 nozz!es of N1. N and N3.but it is,if anything.seen ints of the.'nd Internettonal Conference on Messwe Vessel Tech.

that the stress concentration factors of N1 and N3 nozzle notory. San Antoruo.Tesas !!.S.A. 0ctober 1-8.1974,2 58, pp.

are a little larger than one of N: nozzle.

801 819.

) The crack propagation rate of the nozzle of the (31 Miyatono.S: and Shibata. K. " Crack Propasation Charactnistics in the Nottle Corners of a Prenure Veuel Stut j

pressure vesselis dependent on the stress concentration Nel of Wht Watn Ructor.Nudinn of the Maurne.

factor et the inner nozzle cornet.

tional Conference on Strucrwel.Vechames ut Reettor Technol.

3) The crack propagation rate of the nozzle is in.

ory. London, United Kingdom Septemoer 1 3.1975. C4/8.

Quenced by the geometry of notches at the inner nozzle (4l Rules for Construction of Nuclear Vessels". ASME corner.

Boiler and Pressure Vesal Code.Sec. !!!.1968 Ed.

Aeorintec from Third international Conference on Preuwre Vesses Technoloey Part 11 - Meteriais eno Peorication Puohsnea o/

The American Society of Mecnanical Eneineers United Eneineerine Center 345 East 47th Street. New Yort. N. Y.10o17 b

- -,