ML20064A818

Rev 1 to Feedwater Sparger Crack Growth Assessment

ML20064A818
Person / Time
Site:	Brunswick
Issue date:	02/28/1990
From:	Deaver G, Legate R, Pyron J CAROLINA POWER & LIGHT CO.
To:
Shared Package
ML20064A817	List: ML20064A816 ML20064A818
References
B13-10487, RDE-42-1289, RDE-42-1289-R01, RDE-42-1289-R1, NUDOCS 9009280130
Download: ML20064A818 (16)

Text

..

m'

, l'

(/ , ,

y{ ,, p . ;

V ,

RDE 42-12s9- Rev.1 DRF,# BI3-104s7- ,;

y i

l -l >

BRUNSWICK STRAM ELECTRIC PLhMT UNIT II" j FEEDWATER SPARGER CRACK GROWTE AssEssxEur l I

I

! - FEBRUARY 1990 I

d

J PREPARED BYt AM SlGl40 L4. A. DEAVER,-MAN #GER j REACTOR COMPONENT DESIGN ..

1

'1 C 0 PREPARED-BYt-

7. W.-PYR)fH, PRINCIPAL ENGINEER j REACTOR COMPONENT DESIGN 1

-1 VERIFIED'BY '\ L L T'O R. E. LEGATEkJI'ECHNICAL LEADER 1

REACTOR COMPONENT DESIGN  ;

REVIEWED BY: *

• 9O A. R. SMITH, SOUTHERN REGION ,

LICENSING SERVICES MANAGER .

i- .

APPROVED BY _ A 4 [iO N. W. BIdLIERI, MANAGER l j_ REACTOR DESIGN ENGINEERING

/, *~ Yl; i i M I'5'27h M M'

l APPROVED BY:  ;

G. L. SOZZI, MANAGER PLANT PERFORMANCE ENGINEERING ,

1 9009280130 900914 PDR ADOCK 05000324 P PNV

.e

y . _- _. . __ . ._ _ .

([, A **. $

% / l

V '?

)

IMPORTANT-WOTICS REGARDING {

00NTENTS OFcTEIS REPORT' l 1

PLEASE READ CAREFULLY {

t l

F  !

b This report was prepared by GE Nuclear Energy sclely for the Carolina Power & Light Company. The information contained in this report is believed by GE Nuclear Energy to be an accurate  ;

and true representation of the facts known, obtained or provided j to GE Nuclear Energy at the time this report was prepared. }

The only undertakings of GE Nuclear Energy respecting information

~

., in this document are contained in the Work Authorization a 2570020034 (Task 2) of Master Agreement ZM70020000 between CP&L and-GE Company. The use of this information except as defined by 'i said proposal, or for any purpose other than that for which it is intended is not authorized; and with respect to any such  !

unauthorised use, neither GE Nuclear Energy nor any of the contributors to'this document makes any representation or >

warranty (express or implied) as to the completeness, accuracy, ,

or-usefulness of the information contained in this document or that use of such information may not infringe privately owned i rights; nor do they assume any responsibility for liability of .

damage of any kind which may result from such use of such information.

}

s

" 4 ,

l 4., - .- l INTRODUCTION &

SUMMARY

l.  !

The currently installed feedwater spargers at the Brunswick Unit 2 plant have a single row of side drilled holes in the six' inch {

diameter sparger header pipes. There are 36 side drilled holes  ;

in each of the four feedwater spargers which have various sizes  ;

for balancing the flow distribution around the circumference of l the reactor vessel. This flow distribution is important in order to maintain a uniform power distribution within the reactor core..

one phenomenon which has been observed in this type of sparger is radial cracking at 1.he side drilled flow holes. This cracking l has been previously evaluated to be caused by thermal fatigue which results from the turbulent flow condition as the flow exits l the sparger at a lower temperature than the reactor vessel.

During startup, the most extreme temperature differentials are l experienced when the reactor vessel fluid is at 5500F and the '

f'" feedwater sparger flow can be as low as 1000F. After startup, the feedwater heaters are in operation and the temperature of the i I

feedwater flow increases to approximately 4200F. An initial assessment previously concluded that the cracking should arrest l as the cracks propagate away from the edge of the hole. Actual experience has shown that propagatic7 of the longest cracks does reduce to a very slow rate, but that it is'possible to have crack lengths which are larger than originally predicted. The most plausible explanation for this behavior is that the cracks are initiated by the unstable discharge transients at the flow holes and are propagated beyond the self limiting length by. "

intergranular stress corrosion cracking (IGSCC) due to the ,

creviced environment created by the crack surfaces.

4

The feedwater sparger is not a safety related component and the cracking'will not affect the safety of the plant. While the potential for loose parts is not of immediate concer,n, a likely concern in the future would be for small sognents of pipe becoming loose around the flow holes. This and other scenarios are discussed in the safety analysis portion of the report.

A calculation was performed to ascertain the potential change in 0

enthalpy balance due to the loss of a segnent of pipe at the 315 y sparger tee exit holes. The results of this calculation show, as expected, that a lost segment causes the enthalpy balance to degrade slightly but remain within General Electric's specified design limits for rated flow conditions.

Based on analysis results and the observed cracking behavior of the BWA plants with side exit flow hole feodwater spargers, it is scceptable to operate Brunswick Unit 2 at least another cycle

.with the existing feedwater spargers without any repairs.

4 I

i

-'e i

I

2. CRACKING CAUSES Thermal stress, fatigue, and fracture mechanics analyses.of the l feedwater sparger flow hole region have been performed in l Reference 1. Two thermal transients were studied and were found f to be capable of initiating and propagating flow hole fatigue i cracks. A high cycle thermal transient caused by " unstable i discharge" flow at the flow holes is the most likely cause for crack initiation and was found to be self limiting after a growth  ;

of approximately 0.2 inches. An " infrequent startup" transient l vas also found capable of initiating cracks and, while not self i 1

limiting in growth behavior, is not expected to be a primary contributor due to the limited number of cycles likely to occur. 1 once cracks are initiated, another potential cause for continued-

~' cracking would be from intergranular stress corrosion cracking ,

(IGSCC) driven by oxide wedging. This behavior is further l enhanced near weld joints such as found at the " Tee" fitting {

vhere weld residual stresses may-still exist. l t

k I

1 r

)

i 4

W W d +_'a ss.,.m __s- __---w - a-m -- 1-*T+ me-rr-- - -

I. .

~

3. PREVIOUS BWR EXPERIENCE Feedwater flow hole cracking in Boiling Water Reactors (BWR's) has been monitored by General Electric since 1979. Prior to 1979, . several instances of flow hole cracking were observed but the spargers were subsequently replaced with an improved top mounted nozzle design. Table 1 summarizes the crack history at the feedwater sparger flow hole region for all applicable ~BWRs which currently have side flow holes. The most significant cracking data is provided by the BWR-4 type plants listed.  !

Brunswick Units 1&2 (Plants H & G).have experienced flow hole cracking since 1979 and 1982 respectively. Small cracks were ,

discovered in the outermost flow hole of the 3150 sparger at i Plant I in a recent 1989 outage inspection. However, Plants D and F have not detected flow hole cracks after 14-15. years of j operation. The other listed plancs are of limited use to the i data base due to either the uniqueness of the design or limited l operation time.

l The Brunswick flow hole cracking has been observed to be cf two 1 types. All cracks emanate radially from the flow holes. The  ;

f

( cracks observed in the sparger arms, away from the " Tee" section '

holes, appear as a " sunburst" pattern as seen in Figure 1. These are consistent with the cracks predicted to occur in the l Reference 1 theoretical analysis. The cracks observed at the flow holes in the welded " Tee" fitting are observed to be of two different types. In addition to the " sunburst" pattern, cracks are also present which follow along the edge of the horizontal velded seam as shown in Figure 1. These cracks are likely the result of weld residual stresses not fully relieved by solution heat treatment.

'i>

g _' ' . -'

a  ;

7l ,

L'.'  !

i l'

m 4.- REVIEW OF BRUNSWICK II INSPECTION DATA {

As previously mentioned, flow hole cracking of the feedwater i spargers at Brunswick II was first observed in 1982. Since then, 'l the spargers have been visually inspected at each refueling outage. In 1938, a liquid penetrant examination was performed to j establish a baseline measurement of the most evident feedwater i sparger flow hole cracks. This examination was repeated in the i

current refueling outage to ascertain the extent of crack propogation occuring. A comparison of these inspections verifies I that slow crack growth is occurring. Generally, the longest 6 length cracks show little or no evidence of continuing growth. i Detectable crack growth is more readily indicated on the cracks ,

which had shorter lengths in 1988. Also, at this time there have  ;

not been any loose segments of the sparger header pipe as a -

L result of the cracking. However, cracking at the 315 0 sparger -

tee section has propagated to link cracks from two adjacent holes 7 I

as shown in Figure 2. These cracks follow the heat a:fected zone of the weld which runs across the four holes in the tee section.

i Since cracking is occurring on both sides of the weld, it is conceivable that a segment of material which is approximately 2" X 3/4" X 3/8" could become a loose r3rt. If this happens, the flow distribution in this local area will change (Section 5.2 addresses this issue) . Also, at hole f20 on the 3150 sparger, there is a small triangular shaped area at the 4 o' clock position which has cracking on all sides. This piece may also become a i loose part. In that event, the effect on the flow distribution would be insignificant.

I

5. SAFETY EVALUATION The topics which require discussion for the feedwater flow hole cracks are the sparger structural integrity, flow distribution and loose parts.

5.1 Structural Integrity The feedwater sparger is not a safety related component and there will be no effect on the reactor vessel pressure boundary integrity. For feedwater sparger flow hole cracking, the two potential scenarios are that small fragments of pipe material will become loose between the flow Mies or that a full circumferential crack of the sparger pipe will occur. The most likely near term event would be to lose small fragments of pipe-material. In this case, the structural integrity of the foedwater sparger header pipe will not be adversely affected.

The stresses in the feedwater sparger are primarily produced by hydraulic loads, pressure differential loads, and thermal gradients. Since the sparger pressure differential is less than '

25 psi, both the hydraulic and pressure stresses in the sparger header pipe are low. Losing a segment of material between the flow holes weakens the cross section; however, this alone will not affect the structural integrity of the feedwater sparger header pipt. The thermal gradients in the feedwater sparger cause secondary stresses which will not adversely affect the structural integrity of the header pipe. Locally, in the area where the segment has become dislodged, the flow will be turbulent and may initiate new thermal fatigue cracks on these

. surfaces. An additional consideration is the presence of the

^

flow hole cracks remaining at locations after losing a segment.

Some of these cracks may propegate faster when the segment is missing due to the hydraulic forces which are greatest at the tee section and could eventually causc additional loose parts. In I

1 (

> e p-.1, B~

Tconclusion, for the case where small= segments' separate from the w

L ,

. header pipe,,the' structural' integrity of the feedwater header i pipe will not be' adversely affected.  !

. )

l The other scenario involving full- circumferential cracking of a >

L

'I header pipe .s not a probable near term event. None of the  !

['  ! cracks oriented in the circumferential direction are long enough.

0 . to expect-full cracking'of a' header-cross-section within the next iseveral~ operating cycles. Even in this scenario, no looso parts are likely to result within the reactor vessel because of'the ,

p*1 pinned end' connections to the. reactor vessel. However,_ flow.

P.

' dist'ribution would be significantly affected and would be -!

L detected by the core instrumentation. There is a potential for L flow impinging on the reactor pressure vessel wall which could (

h result in vessel cracking.

- E, 15.2 Flow Distribution s The-basic functional requirement for the.feedwater spargers is to ,

1 distribute the feedwater uniformly within the reactor so that it j l

will form a. homogeneous mixture with the reactor recirculating 7 by coolant _ water. ~The feedwater must be distributed and mixed with L the recirculating saturated water discharged from the steam L /separatorsLand dryers to. provide adequate net positive suction 1 L head by subcooling at-the inlet to the jet pumps to prevent cavitation, 'The most demanding requirement for uniformity of

, mixing results from the need to have a uniform temperature

= mixture entering'the reactor core'to prevent asymmetrical core power distribution. The design requirement is to maintain the

~

. core inlet enthalpy uniform within plus or minus 0.2 per cent ,

. from average'enthalpy..

For the flow hole cracking problem, the near term ccadition which S' may occur is the' loss,of a segment of material between holes at f

___~_______.__a_ _ _ _ _ _ _ _ _ _ _ . _ _ _ _ _ _ _ _ _ _ _ - . _ _ _ _ _ _ _ _ _ . _ . _ _ . . _ _ _ _ _ _ _ +

j- . ,

h - , , ,, , i

= ;;'

-the 3150 sparger tee sect i on. The additional; flow area created <

- by: the lost segment will~ cause more flow to-exit'at the tee  ;

region. To evaluate-this: condition, a calculation was performed

[3

[ to ascertain the change'in enthalpy balance due to the loss of a'- f l' flow. hole segment. piece using the HEATR02 computer program.  ;

.(Reference 2). This~ calculation simulated the' loss of a 2" X 3/4" area: segment piece at the 315 0 sparger-tee exit holes.and  ;

l.

compared the results with those for the existing, intact-I spargers. The results of these calculations show, as expected, that a lost segment causes the enthalpy balance to degrade U

~

L

slightly but remain ~within General Electric's specified design i limits for both 100% and 105% rated flow conditions; the maximut l; pressure drop across the sparger remained below 25 psi, while tne ,

L core inlet enthalpy remained uniform within plus or minus 0.2 per o , cent.-

h t 6

5.3 Loose Parts

.Two potential lost parts have been evaluated to determine whether there is any effect on safety. One part is a segment of. material

[

between the flow holes. An assumed size of approximately 2" X l

X .3"'was postulated based on the results of the liquid penetrant _

examination. 'The second assumed piece is triangular shaped 3/8" X 3/8" X 1/2" and .3"-thick. These pieces, which could separate

.from the sparger, ere likely to.end up resting on the shroud 4

support shelf. It would also be possible, but unlikely, for small pieces to ente: into the jet pump or the recirculation suction nozzle. A segment pasa!.ng through the pump would enter t the vessel bottom head area and remain there since there is'not 4 1 ,

enough' lift to carry it upward to the orificed fuel support .:

casting. Fragments entering the recirculation-piping from the suction nozzle would. pass through the recirculation pump and, depending on size, would either get caught in the jet pump inlet '

mixer nozzle or reach the vessel bottom head. Pieces which.

i g

g- . . - .-.

h. t

, . . - 1

". L ' ' ' ]

l reach the. shroud. support shelf or the vessel bottom head willinot .,

l c

cause'a safety problem,.and'no significant' damage to other-components should' occur. Although unlikely, any pieces which are' >

L caught in: the jet pump inlet mixer would cause a change 11n the j l

pressure reading on the jet pump instrumentation. The blocked "

l.3 t

inlet mixer will indicate a lower than normal flow reading and the adjacent' inlet mixer on the same riser will have a higher than normal reading. Other BWR plants which have had flow.

. blockages in an inlet mixer have been able to remove the object: j at the next outage and no damage to.the inlet mixer.has been observed. There-is, however, some potential.for erosion which is 1

I

(

dependent-on the size and shape of the object and the flow characteristics of the-jet pump, t

I '5.4 Confirmation With 10CFR50.59 i i

The feodwater sparger ' flow hole cracks observed in Brunswick Unit' 2 do not constitute an unreviewed. safety question as defined in 10CFR50 59. A review of the structural integrity, loose parts-l probability and materials' confirm that: (1) the probability of the occurrence of, or the consequence:of an accident are not .

increased; (2) the possibility of .an accident or malfunction of a dif ferent type than previously, evaluated is' not created; and (3).

L the margin of safety.as defincd in the basis for any technical  !

.sp'ecification is not reduced. ->

4 5.5 Continued Operation Based on the analyses performed in this report and the results of:

the 1989 inspections of the Brunswick Unit 2 feedwater sparger

. flow hole crack region, it is considered acceptable to operate i H with the existing feedwater sparger for at least another fuel m

cycle.

l

.so k? '

i li$j, -

.  ?

'i' t

6.- CONCLUSIONS & RECOMMENDATIONS Based on previous analysis and' observed cracking _ behavior.in the 9 -flow' hole region of-the-feedwater-sparger, it is con *cluded'that the cracks will grow at a very slow rate'and'do not-present a i safety concern. The feedwater sparger11s not a safety related component and the cracking will not affect ~the safety of the-  ;

. plant. In the event that a small segment ~of the pipe-becomes l

loose around the flow holes, the change in flowidistribution does:

!! inot resul't in an unacceptable core enthalpy distribution.

Based on the above, it is acceptable to. operatet Brunswick' Unit = 2, at least another cycle with the existing feedwater spargers without any repairs. To reduce the' potential'of segments:of pipe

?

-becoming loose,-repairs to increase the structural integrity <of

'the segment should be considered. Liquid penetrant sparger examinations-should be continued at future outages to insure that-iflow hole cracking has not progressed to a' stage requiring complete sparger replac < "s.

y L

u

\'

4 .

L l

u f

? 'e

?

k

, .- 4 -- - - -

w y s

t.  ;)

1 L91 , 'n

.*e- 6

i '.f'; l -t

+

.e i .- - 7.: #

REFERENCE 8' S. R. Sharma and P. C. Riccardella, "Feedwater Sparger Hole  :

~

1.

Thermal. Stress Analysis", RA (RSA-76-04,-General-Electric Internal Report, 3/32/76. ,

-g  ;

2. . A. B. Burgess', "HEATRO2 Computer Program,. Technical Description, Qualification, and Users Manual", NEDE-25158, ,

December, 1979. .

.i 4

m o

• l '

li ,

t L'  :

'4

'In I

L si l

l i a

f Y $1 .

j r l

'I i

-' l

^

L c

'\

'j

- ui-~

- - 4 <

.~ ..:

m; = ' -

p. _

' # ~'

n  : TABIZ 1 - '

= .. .- -

BUR FLANT FEEUUATER SPARGER FLOW HOLE CRACK HISTORY _

}

! -.I. D. ~ Plant-- Commereial Flow Hole Cracking ,

# Iyp.e - _Startuo___ First Reported __ Crack Growth History /Coauments t

Not Reported Square cross section sparger; donable row flow holes.

~

~ A&B BUR-2 12/69 Carbon steel material.

i C BUR-3* 4/71 Not Applicable Has had periodic sparger changeouts which limits data base information.

D BWR-4 11/72 No cracks as of No cracks found in 1986.(extensive visual exam).

5/86 Inspection E BWR-4* 11/72 None Reported Spargers changed out in 1974; no cracks reported' since.

F BUR-4 2/75 No cracks as of No cracks in 1983' penetrant exam; also none noted 1988 Inspection @ visual exams through 1988.

i G BUR-4 1/76 Cracks noted in Cracks growing slowly; 1989 last inspection.

l 1982 Outage 2" maximum crack length (hole-to-hole) @ Tee' veld:

• Inspection seam.

l H BWR-4 3/77 Cracks noted in Cracks growing slowly; 1988 last inspection.

l 1979 Outage Inspection

, I BUR-4 8/78 Cracks noted in Three small cracks initiated in outermost flow hole l 1989 Outage- of 315' sparger.

Inspection-(

J BUR-5 6/84 None Reported Limited reactor operation time. l ODenotes Foreign Reactor -

I- _- v - - - _ _ _ - - _ _ - - , . - _ . _ _ - - _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ - - . - _ _ _ _ _ _ _ _ _ - - - _ - - - - -____-_=_=____-=~_-___~_-.__2__.:_=_.-

._ - ,- _ - - - - - - _ a n. _ _. - - , - _ _ .

,. . _. .;- - . 2g.:. - ' - -

~

u';.,

- $NGj_

-- . . - -

• ca n.

, .~ ? - l:

._ Afl BWR FEEDWATER SPARGER FLOW HOLE CRACK . PATTERNS-TEE FLOW HOLES " SUNBURST" CRACKS hob

(

Qf W HEADER ARM HEADER ARM

-LINEAR WELD JOINT WELD JOINTSO INDICATIONS

=

VIEWHA-A A(f -l 1 A-

/ A l i N a' 3 r . -

TEE HEADERJARM

- t l FIGURE 1 O v i

4

. 4 w-

-. i p _

-~_.u. -

_.. = -.:u .. . - - _ . ~ . . . . . . _ 2_... - _ _ . . . . . . . . . . ~ . - ~ - _ , . _ . . _ .;

g,_ . . . . .- - -

o e e e p BRUNSWICK STEAM ELECTRIC PLANT

' "" ^ '

GE NUCLEAR ENERGY INVESSEL LIQUID PENETRANT PHOTOGRAPHS .

FIGURE 2

s mmy , ~m . ._ _

v c ,fe. y  ; 7' .. . . . .. a 4- i -

7 . s .

3 .

.'  ? ' .- ~ Eyi'i

%d{.. .'hi ..

.f ,

, "[ '[

t W:"Vf.{? . . .

g ,

s

, 9 s y '.(R y

• W ,

.,, , n . ;

m

~ . p., .y m .v ...

. ,;. ." . ~: , . - . ., : \

, ra dD A.

e-.E M

..g q ,  ; P* ' .

3 ','  ;

.g -h ,. ..

y . :; .

. ' ;: : . 44&% M" w. -17 ' .

A, .. e .

. , ye

.~

j*)F

~

.;~. ~:

n - +

- . . W.

. wj '  :

?  % ..

.. a + - .

] *

, s

- M --

315' HE E 4 17, 18, 19, & 20

' I A . . .q  : . ,, .

4, _ . .~ '

3.s , y e- , s. 6 J.,

m4...+w . . . s t, :7:. ; e c.-, .

. . ..n  ;

~

c . : . .e .: i [,- '.

. ;.y V ; ; -l ; g ~ +

4

)- )g .: , s .-~., ;, i . - :; c <

'T T.c

• jy * - !t .

4 g

, ~ ;.. +

J.,*.  ;

jf ,

i r,4.y .y e . .

g#. .' . , ;+

, s.~ ..~,e.  ;:. p g gg.;r , ; .,

q, ;

_. +

5 e,.. ...g.++. , ,

[ D :lr,[$$i [fh d, ;;d , h .

' Y .T@h. ", e * '

y .

4, .

- ~ ,

ge .

315' HOIE i 17,18,19, & 20 FEEDWATER SPARGER FLOW HOLES

.