ML20080J761

Forwards Analysis Performed by Structural Integrity Assoc to Determine Impact of Induction Heating Stress Improvement on Welds.Intergranular Stress Corrosion Cracking Indications Will Be Evaluated for Impact on Piping Structural Integrity

ML20080J761
Person / Time
Site:	FitzPatrick
Issue date:	02/09/1984
From:	Bayne J POWER AUTHORITY OF THE STATE OF NEW YORK (NEW YORK
To:	Vassallo D Office of Nuclear Reactor Regulation
References
JPN-84-10, NUDOCS 8402150184
Download: ML20080J761 (21)

Text

,

, 123 Main Sim4 Whte Ptsm, fbm York 10A01 Old 6816200

1. > NewYorkPower 1# Authority

=i_

February 9, 1984 J PN-8 4- 10 Director of Nuclear Reactor Regulation U.S. Nuclear Regulatory Commission Washington, D.C. 20555 Attention: Mr. Domenic B. Vassallo, Chief Operating Reactors Branch No. 2 Division of Licensing

Subject:

James A. FitzPatrick Nuclear Power Plant Docket No. 50-333 Induction Heating Stress Improvement (IHSI)

References:

1) NYPA letter, C.A. McNeill to T.E. Murley, dated July 22, 1983 (JAFP-83-0769).

Dear Sir:

During a short outage scheduled for early March of this year, a-demonstration of Induction Heating Stress Improvement (IHSI) will be performed at the James A. FitzPatrick Nuclear Power Plant. This outage is tehtatively planned to start on March 2 and to end on March 12, 1984.

As presently planned, the IHSI process will be applied to eleven (11) welds in the reactor water recirculation system. Table 1 lists the number, type of joint, pipe diameter, stress rule index, carboa content and EPRI Damage Index for the affected welds. These welds are identified on Figures 1 and 2 which are simplified iscmetrics of the FitzPatrick recirculation system. The stress rule indices and carbon contents have been factored into the EPRI Damage Index calculations along with variables such as post-weld grinding and plant thermal cycles to give a quantitative measure of possible future IGSCC. The welds were selected on the basis of equipa+nt accessibility and their future susceptibility to IGSCC.

The object of this proposed demonstration is two-fold: to obtain plant specific information on the effect of IHSI on possible pre-existing IGSCC; and to evaluate the usefulness of IHSI as a plant-wide IGSCC countermeasure. It is also expected to reduce the susceptibility to IGSCC of some of the most vulnerable welds. The first goal is a direct recult of the experiences other operating plants.have had with IHSI. That is, post-IHSI ultrasonic testing revealed suspected IGSCC indications that were previously not detected or were evaluated as weld / joint geometry. DDf 8402150184 840209 PDR ADGCK 05000333

/ff f

P PDR

r A pre-IHSI and post-IHSI ultrasonic examination will be performed on the sample welds. It is planned that both of these inspections will 1

be executed by the same technicians using the same equipment and procedures. In this way any changes in ultrasonic indications will be more clearly defined and interpreted.

Because the sample is heavily weighted towards welds most susceptible to IGSCC, and in light of the extensive inspections performed during last summer's refueling outage, it is the Authority's view that in the event of discovery of IGSCC indications, additional inspections would not need to be performed during this outage. Any indications of IGSCC will be evaluated for impact on piping structural integrity and, if required, will be .

repaired. Discovery of IGSCC during the examinations performed during this March's outage will be considered in selecting the sample size for inspection during the next refueling outage, scheduled for early spring of 1985.

One of the welds selected for treatment by IHSI is the distribution manifold end cap in which IGSCC was discovered as reported in Reference 1. Attached is an analysis performed by Structural Integrity Associates to determine the impact of IHSI on this weld.

This analysis indicates that no crack propagation is expected to occur during the induction heating of the pipe, and that the tip of such a crack would experience a negative or compressive stress for a depth of up to 70% of the wall. Note that the depth of this crack was determined to be no more than 20% of the pipe wall, which, even if doubled for conservatism, is well below that depth at which residual stresses become tensile. Performing IHSI on this weld will help clarify the effects of IHSI on the ultrasonic detectability of pipe cracks while enhancing the resistance of this particular welded joint to future crack growth. Finally, if this demonstration of IHSI confirms the feasibility for full scale application to the large diameter ctainless steel piping, implementation would be tentatively scheduled for September, 1984.

If you have any questions, please contact Mr. J. A. Gray, Jr. of my staff.

Very truly yours, h /b vfA J.P. Bayne [

rt Executive Vice President Nuclear Generation cc: Office of the Resident Inspector U.S Nuclear Regulatory Commission i P.O. Box 136 Lycoming, New York 13093

• ,*

• l

, 3 . . . . . ,

1 FIGURC 1 #2- 2 N* 2 A* SE N 28*SE O t2 5

~... 5,

__ 3 , N 3C+$f p ....l

"-zc "

~.2.. , ,2 n ,24 9n '

p,ai,.,,} us n ,,,,, n21s n5 20*90 82-l3 12-19 f2 64 923 50 A II*IO U 22 11 82 26Q 2* * ' W 22 2, 22 28 02858 .

28 30 i,,,,, i 9

91120 0

N Awa 24-130 22 16 23 3r 22 28 28 32 ,,47 4 44 28 52 28 53 4 45 28 54 28 53 28 34 25 57 4 44  !:

N 29*56 4 46 28 54 4 42 28 37 "'

4 39 4 40 I5Ct'E RIC OT RECIRC 'I.ATICr; TYS TM - LCCP 18 NO"T : ABBPT.'IA*!O IE C:.0 ??O'*BERS

4' - ",. l PIGURE 2 12 70 12 88 /I2-69 P '-'

/

Na26.SE 12 64 it.st

/ , g ,9, @ ' '~]

jt 75 yh 32 65 12 84

\ N 2H 58 N.27-SE N.27 3g

,,. N 11-84

'u 22-73

• I* Ekd---

22 7s N .so 22 a1 54 *M-12 59 g,gg A22 a7 M O18.e4 it.

,( aaa A st-M f)28*49 A.ro? l 22-64 r  :

3 i28 404 s'

% 28 ics NelA 1 a 12*68 22-62

& .94 4 4956 22*4 4, 28 90 s.co 4 94 ** *o8 4 s32 28.+os 4-97 28-to3 4M Z4 - 6 4 2, 98 gng g y g,,,,

28 91 g g.cos 2s-92 0 28.elo 20-lEC tc-s4\

2 C- t 4 4 ,

l C- s \ ? C s-ne .

l

-- ?O - t * *!25-115 l t t

C:9 'v o '

G9q l

i 26*i13 0 :s. . .

a

s-siz 20C.':ITTIC CT PIC;pc.LA;;cs gys;p,3  ;,777 g NME: ASEpI y;,*, rc ;g . wg. ; gg,gg.,

= _ , _ - _ _ . - __. . . ___ .- _. - _ - - . . . .-

.O TABLE 1 Stress Rule EPRI 2 M21d No. Joint Type Ncminal Pipe Diameter Q Index Damage Index j 12-4 Shucpolet-pipe 12" .070 1.80 2.13 12-9 Pipe-olbx 12" .070 ' l.53 1.13 12-15 Shucpolet-pipe 12" .050 2.02 3.16 I

i 12-23 Pipe-safe-end 12" .044 1.59 1.22 12-75 Pipe-safe-end 12" .057 1.53 1.14 12-76 Pipe-eJ1xu 12" .079 1.84 2.03 1

12-77 Pipe-elixv 12" .070 1.67 1.41 11-73 Sweepolet-pipe 12" 044 2.28 1.04 22-22 Pipe-end cap 22" .079 0.98 0.95 l

l 28-48 Pipe-safe-end 28" .058 1.06 0.87

! 28-49 PiI u-ellxu 28" .053 1.36 1.15 l 28-106 Pipe-cafe-end 28" .051 1.v6 0.69 i

i 28-107 Pipe-ellui 28" 0.52 1.31 0.80 I

I J

! 1) Altemate hulds to be substituted in case of accessibility problems.

2) Values after seven years of operation.

1 1

1 l

1 t <

/

' ,)

STRUCTURAL INTEGRITY ASSOCIATES

"

• 3150 ALMADEN EXPWY. SUITE 226 . SAN JOSE. CA 95118 (408) 978 8200 ASSOCIATES: December 22, 1983 T.L G ERSER PCR-83-054 A.J. GIANNUZZ1 A.JAmo P.C. R e.cAmotu.A Mr. David Sancic New York State Power Authority 123 Main Street White Plains, New York 10601 .

Subjects IMPACT OF IHSI ON POTENTIALLY CRACKED PIPE WELDS AT FITZPATRICK Dear Daves This letter report addresses several questions you raised regarding the upccming IHSI sampling program planned at Fitzpatrick. The questions arise due to the fact that IHSI will be performed on one weld with a known, IGSCC-like UT indication, and several welds which, although not called out to be crs ked in the recent UT examination program, did contain some " geometric" indications, and were identified as potentially susceptable to IGSCC by the Structural Integrity (SI) Associates damage index evaluation (Reference 1).

This report summarizes a substantial body of data (References 2 and 3) which exists on the impact of IHSI on cracked or potentially cracked welds, and addresses it to three potential concerns at Fitzpatrick:

1. To what extent will the IHSI residual stress improvement be realized in welds which are already cracked?

2. What will be the effect of IHSI on predicted IGSCC crack growth, (such as that reported for end cap weld 22-02-2-22 in Reference 4)?

! 3. Is there any potential for propagation of existing cracks during the IHSI process?

Details of the SI review in each of these areas are presented below. In summary, the preponderance of analytical and experimental data, both in the U.S. and Japan suggest that IHSI is effective in producin6 a deep, compressive residual stress field in the vicinity of existing IGSCC in pipe welds.

Incorporating this residual stress field into fracture mechanics based IGSCC crack propagation analyses (such as Refer'ence 4) results in highly compressive stress intensity factors, even for crack depths of up to 60 or 70 percent of the wall thickness, from which it can be concluded that v

M M3 y ..

.'..3...,

23 G lIT M ./ "

blIdMS *

.' Page 2 PCR-83-054 IHSI treatment of existing IGSCC should promote crack arrest under operating condition loads. Finally, analytical as well as destructive metallographic evaluations have confirmed that no crack extension occurs during an IHSI treatment. Some data have shown, however, that changes in UT amplitude can occur either as a result of operational variables or changes in UT reflectivity as a result of the IHSI process. It is noteworthy, however, that where such UT reflectivity changes have been observed in the laboratory, they were not due to physical growth of the defects, but rather were due to a change in crack tightness or acuity, such that it produced i UT signal change.

RESIDUAL STRESS IMPROVEMENT The effect of pre-existing defects on IHSI resid.ual stress Unprovement was studied by both analytical modelling and experimental measurements (X-ray, strain gage and magnesium chloride) as reported in section 5.1 of Reference

2. Results of the analytical modelling are summarized in Figure 1. This plot shows through-wall axial residual stress distributions in a 26-inch diameter pipe as a function of crack depth (a/t) for a pre-existing 360' crack. The plot shows that the ragnitude of compressive stress and the depth or the compressive zone actually increase with crach depth for depths up to 40 percent of the wall thickness. Experimental confirmation of this analytical prediction is, illustrated by Figure 2, which is typical of many such strain gage residual stress measurements reported in Reference 2.

This plot represents data on a 4-inch pipe weld which was precrackeG by the pipe test technique and then given an IHSI treatment. "he strain gage residual stress measurements were also borne out by surface residual stress measurements on pre-cracked pipe using X-ray defraction and boiling magnesium chloride techniques. The later tests were particularly noteworthy in that .

they produced transgranular cracking TGSCC on the cutside surfsce of the pipe, in which tension is predicted after IHSI, but none on the inside surface (see Figure 3).

Analytical modelling performedin Japan (Reference 3) for 12 inen diameter pipe, precracked up to 25 percent of the wall thickness yielded essentially identical results as the GE 26 inch pipe results discussed above (see Figure 4). IHI also performed magnesium chloride testing of a pipe with an EDM notch in the weld HAZ, both with and without IHSI treatment. As illus-l trated in Figure 5, cracking at the notch was experienced with no IHSI, but not at the IHSI treated notch.

Tne above analytical and experimental results confirm that IHSI is effective l in producing compressive residual Otresses even in the presence of pre-existing IGSCC in a pipe weld.

EFFECT ON PREDICTED IGSCC CRACK GROWTH GE used the above analytical residual stress data as input to a fracture ,

mechanics formulation to determine crack tip stress intensity factor as a function of crack depth. The results, for a 26 inch diameter pipe, are illustrated in Figure 6. The predicted stress intensity factor is increasingly negative for crack depths up to 60% of the wall thickness. This result should be sufficient to counteract any positive stress intensity factor f STRUCTURAL g INTEGRITY oem

. *. Pagel!3

. . PCR.83-054 data due to applied loading, and effectively negate any further propragation of the crack.

SI has incorporated an expected pest IHSI residual stress pattern into our crack growth analysis of the Fi:: patrick end cap weld 22-02-2-22 (Reference 4).

The results, illustrated in Figures 7 and 8, are similar to the GE results described above. The crack tip stress intensity factor for residual and applied stress is negative for crack depths up to 70% of the wall indicating no crack growth. Thus a substantial improvement is predicted over the pre-IHSI crack growth results of Reference 4.

~

Finally, experimental crack growth data en pre-cracked, IHSI treated pipe obtained using pipe test methodology are reported in Reference 2. These data indicate a slight improvement in time to failure versus non-IHSI treated pipe (See Figure 9), but the degree of improvement is not nearly as dramatic as that predicted by :ne above analyses. The reason for this apparent discrepancy is that the pipe test methodology is not a good one for testing residual stress improvements such as IHSI. Pipe tests are loaded significantly above expected service loads to accelerate the cracking phenomenon, and the subsequent yielding overwhelms any apparent benefit of IHSI. In order for pipe tests to adequately represent IHSI, they must be conducted at lower loads, which results in much longer failure times in non-IHSI treated piping, and extremely long (bordering on impractical) test ti=es to demonstrate factors of knprovement associated with IHSI.

It is thus concluded, based on analytical results, that IHSI of pipes con-taining pre-existing IGSCC, should promote crack arrest, and that the GE pipe test data, although not substantiating the magnitude of this claim, does not contradict it.

. CRACK PROPAGATION DURING IHS,I_

Extensive analysis and teeting, both in the U2S. and Japan have been condusted which demonstrates that the IHSI process in itself does not propagate pre-existing defects. The elastic plastic residual stress analysis reported in Reference 2 was further examined to determine the maximum strain excursion in the innermost vicinity of the crack-tip. The resulting stress-strain history is illustrated in Figure.10. This figure does not exhibit nearly enough strain to produce ductile tearing or other forms of crack extension in a material as inherently tough as 304 stainless steel. Similar analytical work conducted by IHI in Japan [ Reference 3] demonstrates that the maximum crack-tip opening displacement [CTOD: during the IHSI process is well below the critical CTOD value for crack extension.

These analytical results are borne out by extensive destructive metallurgical examinations of pipe test specimens reported in Reference 2. In these tests, specimens were pre-cracked, IHSI treated, then destructively examined.

Metallurgical examination of several such specimens showed no evidence of crack propagation in any mode other than intergranular stress corrosion,

(} STRUCTURAL y.7 INTEGRITY ocni

~ -

+. -

. . .. ~ ~.

Pago 4 1- y " ~ ' ^

, PCR;83-05'4 , -

~'

which,is positive evidetice that IHSI did not propagate the pre-existing ~

cracks in these specimens.- An interesting by product of these experiments, however, is .ths fact that IHSI did appear to change the UT reflectivity of the cracks in some cases: 'This phenomenon is illustrated by Figure 11 in which there was a substantial increase in UT, amplitude (> 40% of full screen height) in a precracked weldment befor6~and after IHSI. GE did ex- -

tensive investigations of UT operational.varisoles, crack morphology, and actual cracx size of these specimens frontwhich they concluded that the IHSI process, while not causing,any physical crack growth, may affect the crack tightness and acuity,suc hthat,,it will? p,coduci a UT signal change.

Ihopethattheabove[informationsatisfiesyourlicensin6/internalreview ^

needs for the upcoming:1HSI' outage. L?leasolcall if you have any questions on it or reed our e.ssistance on any%tner ' aspect of the job.

~

' ~

t . .

e V y r y yo s,, '

^*

t 4 7

^

, .t ,. l ~' ' ~~ -

~

Peter C. IMddedell'a -

- s

~

/dh ,

s.w, .

encl.

4_

M  % *

~. -

\ , , - .,

~

~

l 's;, ,

i ,'

l

-? .

~~ ~

. s , i~. .,

s.

^

g# -

. -+ /

, , , . n I'

N' 4

,./

l N

( -c ._

\

l .-

l . . . . -

P I STRUCTURAL l

i e ,

j INTEGRITY .ac t.

t_

4 List of References

1. Structural Integrity Associates Report No. SIR-83-005, Revision' 0 "A Damage Model Based Cost / Benefit Alternatives at the Evaluation of IGSCC Remedy / Repair alternatives at the James A. Fitzpatrick Nuclear-Power Plant," September, 1983.

2. EPRI Report NP-3375, " Induction heating Stress Improvement",- Prepared by General Electric Co , EPRI Project Mariager A.T. Giannuzzi, November, '1983.

3. EPRI Report NP-81-4-LD, " Residual Stress Imporvement by Means of Induction Heating," Prepared by Ishikawajima- Harima Heavy Industries Co., Japan, EPRI Project Manager A. J. Giannuzzi, March,1981.

4. Structural Integrity Associates Report No. SIR-83-002, Revision 2,

" Fracture Mechanics Evaluation of observed flaw indication in 22-inch pipe to end cap weld 22-02-2-22, JAMES A. Fitzpatrick Nuclear Power Plant",

August, 1983.

_ _ ~

i l

1

[

STRUCTURAL g INTEGRlW .oc1.,.

. 70 ao -

so -

ale = 0.4 ah = 0.064 40 -

ah = o.2 l

,,===

a

,30 20 -

to -

1 e

N 0 ff l E

E

/

t -io

/

/

e ~~#

_a _

/

I

-30 - / -

% UNCRACKEO l s an _  %~~m.s i i st=0.o64 l -so -

/

-60 -

a/t = o.2

/

/

-70 - # /

/

/ ah = o.4

-m - l

/

/

I

/

e I l s .d .

1/4 THICKNESS 1/2 THICKN ESS 3/4 THICKNESS od-R ADI AL POSITION Figure 1 IHSI Througn-Wall Axial Stress Distribution at Weld Center Line for Preexisting Cracks (From GE Analysis of Reference 2)

STRUCTURAL p INTEGRITY noc u

STRESE(MPs)

,o .g= -300 -2= -im 0 iOO 2= 3= 4= sw I I I l l l l COMPRESSION TENSION 0.M - '

O.71 -

N E

t 9 0.57 -

5 d

E o 0.43 -

2 o

5 a.

0.29 -

0.14 -

f f O f f t f 10 -56 -42 - 28 - 14 0 14 28 42 56 70 STRESS (kes)

Figure 2.

l Through Wall Axial Residual Stress 'DJstribution for Specimen BS-06 Joint E. Pre-Cracked Plus IHSI-279.5 Azimuth, 1.52 mr.

(0.05 Inches) From Weld Fusion Line (Strain Gage Measurements from Reference 2) l

' S TRUCTUR AL

[j INTEGRITY a x es

r - - _. ..

l

. . l

' O.o.

. }

q ;w.5<4G.:  %;.4. ,

s w

t. '.' y ' ' . ,- .m

..-Q- t .A 0

-  : , ne, 4 .,

. i, ': +

. .;:GQ .. C ./:o.V .-* . .t --.

, ,- a * .-?!

.*y.

~

c- g-

'%.;,.- . , - ..t

+ ~4 p n A . .

l

s. c .F~% ; Qt .2t. j

~

~

Tascc ZONE - >

'f ' ..

f- " r4 t . . - .

l

' .: ?z,.- vt

9 . .-

=-

, .ti-

• e g

~ q,a"?

Ai kW. ,=

~ -.-

. gg..q3, 4 l

.A c4 '

o -

,p: . . .- j,

.b.3,..c..

sx u .

< [-).'..:... -~ ,. : .. ..)., * <

.v y ,- '

l

.,,s._. \, .

~

.\.\.]

z .

f'

f' _ l k ;.T'n 5.. *: ' .%..~-i&+E. %..%) ."~ -

. r* -  :' ' . .<^

..,. l

.. c

a r:u .v ..

,.r:-.i -* _fg ... ~Q.

-~ .

&.w%. m. _ ?.~: 3 ,i :v ..g?. r%s.7 i..Q. - D. . .=i.

. &,..r

.\ .-

.- . . . c .. .

.: .W ... . n  : g... ,.e .9,-

.g ' ' ~

- _v .

u. .* c~ g*a 3 -

w.

.j,.,. n

..2- , %-.#yg-X . J.

-~

.? . :.

,e. , '<* '" e.. .' *- , .- - !?",,y-@7~ .93 .

e.: =*._2: -V

. '%p .2..:.t.'.&R.-%. eg..

-': " . ' . 'y',~. . <,;B- Q,u.

s *1

. . ..q.-

M. . . 'w L  ?.- .gr -.  :. .  : ;.4 i

h. . , / 3.. Q_ ' ,, ' eg,.

3 42- .. - - M. _ -

s. Y .-

l m ._ _

1. 2*L -i byLi^.f L,,

n.- c~

t 9 ' -y .7 h. ".'.".Ld.:45 a  ;. ,.. ,

. g/"- y.! .s -

d i &; . ?. '-  % . ' .;'5 p.n 4; ., n.; G. .l. ., = .; .;, .<. .

;*. .?h._T;.Y,.? .

[,ss, f u.. ' '- '

-h,j. ':# - -

.-r

. r.8;a 3 . .3,,.y .s+ l

.e ' . ,

• • 4

.w % # m3s $

Of $ Y.M g l

.s,. ?gi;.. M .

~

.,' ' ? . Figure 3 ID IGSCC from pipe test and OD TGSCC Produced by Localized

. ?.,i 'p7 ", ~A P ]' .. ., ' . , -

' ^f ~ f ; f3 7 ,-  ;

MgC1 2 Test on 0.D. TGSCC stopped l

'4 7f,y.4, h P in compressive zone ahead of V- '*

' ,;;1'1,& 3'n .'I IGSCC crack, and shows the ability l

-- rosec go,k3 '

of IHSI to produce deep compre-

-t .. ZONE 3 siVe stress on pipe ID surface.

~ '

vl . 63' %W'I ^ .

TGSCC on OD results from tensile (Frem Tief. 2)

~

stress.

l I.D.

I f STRUCTURAL gb INTEGRITY w.i.

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

1

. . {

CRACK DEPTH l'. 4 m:n

--- - 2.4mm

~

4.2mm 4.

50

+

/- ,

J I

l l 1 76 l

l

'/

I //

1 ', '

I

. I el .

I l '}

-50 -

g,l- 7 1

II f

I

-100 0 5 10 15 19

' WALL THICKNESS (mm)

Figure 4 Calculated residual stress through the thickr.ess of 12 in pipe with,a flaw (Reference 3)

STRUCTURAL

[g INTEGRITY asso

(a) EDM - _. .

~

[EDM: ELECTRIC h .I.

DISCHARGE ~

( MACHINING / -

y -

l i- >

(b) EDM s

r' 5 G' lI a .

Mgci2  :' -

I llrogI ~

?+..IIriof,Ist

'5-.

.- O'y g g,-

,v, .

\

Propagation

_. . m. 7 n >

\ .

l l  !

> (c) EDM

,v. >

l HSl / !! s oI!j o '

r,,I'%:o,,

, ,  : IIII Mgt12

__ ~

~:

\

No Propagation ogg O

i i Figure 5. Japanese Experiment Showing IHSI Residual Stress l Improve" lent in Pre-Notched Pipe (Reference 3' i

j

$TRUCTURAL i-g, INTEGRITY ux.. .

I i

l

[

l I

3 .

I l

10 -

0

),-10 =

l 8 I

G m

-20 -

2 Y

iE 3

w m

t; -m _. _ _ _

40 -

- 50 -

I M =

l l /

I /

/

-70  !  !  !  ! l l iA 0.1 0.2 OJ OA 0.5 0.8 0.7 sh A ATIO Figure 6 Stress Intensity Factor vs. Crack Depth (a/t ratio) for Pipe Subjected to IHSI (Fron GE Analysis of ,

26 Inch Diameter Pipe - Reference 2)

STRUCTURAL

[p' g INTEGRITY .oc. e.

s - _ - a . .

.e

• e

.e .

.t., .

30 .

w.. .

. . 4

. . . n .

.go. .. . . .. . . . .

e e . . s

..e . . .

e c

e-7.

m 0.2 0.4 0.6 0.8 1.0 ae . . .

o.,..... .

H . ..

O CRACK DEPTH (a/t) r- - - - -

s.

E-m ..

n -- . .

z - - -

w s .

. g.z .

w - ...

m. .

V- -20 ...

-30 _

.-40.

Figure 7. Crack-Tip Stress Intensity Factor Vs. Crack Depth for Fitzpatrick 22 Inch Pipe to End Cap Weld Subject to IHSI (SI Analysis) m i STRUCTURAL l

d lNTEGRITY .is

r.

a .

- 1.2 9 6

~1.0 _

s a**

~ .

m a' a

,,,, e==

a0 = 0.78 in. _

0.8 - .

_ __ _ _ _ _ .ao 2.p_:.012 L in. _ _ _ _ _ __ _ _ _. _ _ _. __ __ _ ,

4 ao

[

0.6 -

= 0.585 in.

l l-a

_ _ _ _ --0_= _0. 487_5 in_ - - -_-- _-- -.__ _ _ ___--

i, .

+-4..

0.4 __ aO._= 0. 3 9 in .-

l ..

a3 = 0.293 in.

J l

0.2 _. _ _ ag_= 0.195 in.

i

! I I  ! t i ,

5000 1000 15000 TIME OF PLANT OPERATION (Hours) i Figure 8. Predicted Crack Growth Rates With IHSI For Fitzpat. rick  :

22 Inch Pipe-To-End Cap Weld (SI Analysis)

~

STRUCTURAL k lNTEGRITY oc.no

__ - , , - . _ ~ .-. ._ _ . _ . _ . . _ . . . . .- . . - _ . _ _ . - - . . . _ .

,e ,

4 33 -

228 4

32 - -

228 E

31 - -

214 A -

VV 30 - -

207

- *;G REFERENCE E 29 - -

200 m g V RSP 03 (ALSO 1084 CYCLES AY 201 mPa (29.1 koil g 4

g A RSP 031 (JOINT E NOT IHSil , g U Q RS082 (F ALLURES PEFORE IHSil 0 O O O "

28 -

O E4 08 (F AeLURES BE FORE IHSil _ ,93 IHSI (ALSO 771 CYCLES AT 152 mPa (27 ksil 27 -

h MPO 3 74 CYCLES AT 172 mPa 125 ksil.

186 417 CYCLES AT 210 mPa l30 5 ksel A RSP02 ,

T RSP O4 (ALSO 1084 CYCLES AT 201 mPa 129.1 ksil 26 -

9 RS 012 (ADDITION AL CYCLES TO F All AFTER IHSil

~

II9 E E408 I ADDITIONAL CYCLES TO F All AFTER IHSil O O E

-l p_ - 25 - -

172 a

\r s i i i e i i eil i e i i e ieil i e i e i i ei 30 iOO i .O* i O.*O Z

M$ TEST CYCLES oc mo j Figure 9. Results of PreCracked Pipe Testing of Four Inch gp{ Diameter Pipe Welds Before and After IHSI (From Reference 2)

376 (dos f

/

207 __ /

(301

/

./

/

lE -

./

/

/

/

es j

~

tiol .

/

2 /

1 /

I /

$ o / /

g to) -e m /

s /

• /  %

j .

49 /

7

(-ioi

/

/  :'=

/

-iss /

t-201 y

/

/ e

/ 10 00 LOs . TION OF ELEMENT

- 207

(-30)

/

/

/

- 276

(-40) o 075

[4 o g

0.1 TOTAL AXI AL STR AIN 1%)

Figure 10. Axial Stress-Strain History in Vicinity Of Crack-Tip During IHSI (From. Finite Element Analysis of Reference 2)

~

STRUCTURAL

[h INTEGRITY.sm

o -

l .- ..

3

.' l BEFORE IHSI i

l l

80 -

( 60 -

g a

$ 40 -

4 20 -

A 2 '~ e

^

- ^ - ' '

0 "f" * =^ L

• I I I O 180* 360 l

AFTERIH$l l

80 -

6j 60 -

a b 40 -

20 -

2 0

l I I C 180* 360*

AZIMUTH Figure 11 UT Amplitude Versus Azimuth for Specimen RS-01-2 Before and After IHSI as Recorded by Same Operator. Small changes in amplitude do not represent crack grawth. ( From Reference 2)

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