ML20012A172

Core Spray Line Crack Growth Analysis Update for Brunswick Steam Electric Plant Unit 2.

ML20012A172
Person / Time
Site:	Brunswick
Issue date:	02/23/1990
From:	Cornwell K, Gordon G, Ranganath S GENERAL ELECTRIC CO.
To:
Shared Package
ML20012A170	List: ML20012A169 ML20012A172
References
EAS-03-0190, EAS-3-190, NUDOCS 9003080440
Download: ML20012A172 (16)

Text

_ . , . . , , , , ..

o m, -

~

o-O . - - - - - - - - .

o:

L,

_( "=_.._- .--..a_. .. .-~a<<-- -x---*I--~+-+-- +-+~~#

. o:

.O -

( )

a GE Nuclear Energy

.9003080440 900223 2'S PDR O -ADOCK 05000324 PNU ,o O.~ _

. ~!

} ,3 .

E EAS-63-0190 (Supplement 1 of

i. EAS 14 03BB) ,

DRF E21-00094-1 January 1990 t ji  :

l l

} CORE SPRAY-LINE CRACK GROWTH  :

.. ANALYSIS UPDATE FOR BRUNSWICK j . STEAM ELECTRIC PLANT UNIT 2 f l

i k

i I

Prepared by:

} K.F. Cornwell i

Reviewed by: . .

A.R. Smith, Southern Region Licensing Services Manager .,

Approved by: ,b 4 _

D.: G.M.'Gordon, Manager -

!: Fuel and Plant L Materials Technology 4

^4.~-

Approved by: M" -

O~ S. Ranganath," Manager Materials Monitoring and Stru ural Analysis Services l .

Approved by: h 3 G.C.~aozfi,Fa/iger Plant Perforfnince Engineering l

Q > ,

l

3; < :* .

7, EAS-03-1090  :

GE NUCLEAR ENERGY

} . ,.

t t

j IMPORTANT NOTICE REGARDING  !

CONTENTS OF THIS REPORT  !

is h Please Read Carefully 4

The only. undertakings of General Electric Company respecting information. in this ~ document are contained in the contract between Carolina Power & Light Company and General Electric Company, and nothing contained in this document. sha.11 be construed as changing the contract.  ;

g The use of this information by anyone other than Carolina-Power & Light P~ Company, or for any purpose other than that for which it is -intended under such contract is not authorized;' and with rt Ject to. any unauthorized use, General Electric Company makes no representation or warranty, and assumes no liability as to the completeness, accuracy, or usefulness of the information contained in this document.

I.

Q L .

L o >.

I 1

l C

-i-

..i

+1 I

[ ,2

t. 8 EAS-03-1090  !

SE NUCLEAR ENERGY l 1

1 TABLE OF CONTENTS

]

EASA I f .

l 1

1. lWTRODUCTION AND BACKGROUND l-1 1.1 Crack History 1-1 1.2' Recent Inspection Results 1-2 ,.

I

2. . REVIEW OF. PREVIOUS ANALYSIS 2-1 l 2 '.1 Applicability of Previous' Analysis 2-1 2.1.1 Crack Leakage Estimate Review 2-1

-2.1.2 Core Spray Pipe Structural Integrity Review 2-2 2.1.3 Lost Parts Analysis Review 2-2 2.1.4 - Loss-of-Coolant Accident Analysis Review 2-2 2 .' 2 Sumary of Review 2-3

3. SAFETY EVALUATION CONCLUSIONS 3-1 I

l h 4 ~. CRACK GROWTH ESTIMATE UPDATE 4-1 i

p i - .. 5. LOCA ANALYSIS UPDATE 5-1 1

i. _6. REFERENCES 6-1 3 -

> -15

o j y, , , ' : . '

e.< i EAS-03-1090 GE NUCLEAR ENERGY l  ;

I y

l.0 INTRODUCTION AND BACKGROUND .

1.1 CRACK F STORY Q~-

A crack indication was found during the previous refueling outage for Brunswick Unit 2 (February 1988) on the north core spray line  :'

(header) where the piping and junction box meet. Following the identification of the crack indication, both an ultrasonic test (UT) and  :

_.)

a liquid dy: penetrant test (PT) were performed to confirm the indication was e crack and to determine its dimensions. The examinations revealed the following information:

I 1) The crack was approximately 3.50 inches in length- along the inside of the pipe (80 degrees of the inner circumference),

2) the crack was approximately 1.75 inches in length along the l outside of the pipe (36 degrees of the outer circumference),

and

3) the crack was through wall.

4 At the time the crack was found, it was postulated that the crack.

could have been caused by either intergranular stress corrosion cracking (IGSCC) or thermal fatigue, with the most likely cause being IGSCC.

Analyses, documented in Reference 1, were performed to evaluate the

g. effect on safety, and support operation for an additional cycle. The l, previous evaluations examined:

l

1) Crack leakage estimate (Section 2 of Reference 1),

k. 2) core spray pipe structural integrity (Section 3 of Reference 1),

3) lost part analysis (Section 4 of Reference 1), and

'4) loss-of-coolant accident analysis (Section 5 of Reference 1). *

^)

The results of these evaluations provided the basis to support continued safe operation without repairs to the core spray line crack for one additional cycle, g

1-1

f- :

v ,.

EAS-03-1090 GE NUCLEAR ENERGY l.2 RECENT INSPECTION RESULTS Brunswick Unit 2 has now completed the additional cycle of

operation (Cycle 8) that was supported by the Reference 1 evaluation.

~

Because of the previous crack identification, CP&L planned additional inspections for the refueling outage following the Cycle 8 fuel cycle.

These inspections have now been completed and the results are discussed i below.

CP&L performed UT examinations on both the previous crack and on the three other geometrically similar locations of the core spray line (on the piping adjacent to both sides of the T-box on both the north and south core spray line headers). These examinations were performed in addition to those required by the Inspection and Enforcement (IE)

=

Bulletin 80-13, and those suggested by GE Nuclear Energy Service

Information Letter (SIL) Number 289. These inspections revealed no new

[' crack indications on the core spray line.

The PT examination- of the previous crack site resulted in the following updates to the previous outage findings:

i i 1) The crack is approximately 4.90 inches in length along the inside of the pipe (111 degrees of the inner circumference),

and 2 2) the crack is approximately 1.90 inches in length along the outside of the pipe (39 degrees of the outer circumference).

Thus, during the operation of Cycle 8, a crack growth of 1.4 inches occurred along the inside of the pipe (an additional 31 degrees of the inner circumference) and a growth of 0.15 inches occurred along the outside of the pipe (an additional 3 degrees of the outer circumference).

1-2

[ . ., [ * -

I h

EAS-03 1090 GE NUCLEAR ENERGY.

' 2.0 PREVIOUS ANALYSIS APPLICABILITY REVIEW

~

2.1 APPLICABILITY OF PREVIOUS ANALYSES D. .

The evaluations performed to determine the acceptability of continued operation for the previous cycle (Cycle 8) were reviewed to determine which areas might require updating due to the crack growth-experienced, or to extend the evaluations to support o'peration in :Se upcoming cycle. The results of this review, including the primary conclusions from each of the evaluations areas, are documented below; and those areas requiring updates are identified.

2.1.1 Crack Leakaae Estimate Review The core spray line crack leakage rates utilized in the previous evaluation are discussed in Section 2 of Reference 1. Based upon assessments performed at other BWRs, including both visual inspections and air-bubble tests, the upper bound leakage .for the current crack at Brunswick was previously and is still estimated to be less than 7 gpm during the core spray injection phase of a- Loss-of-Coolant Accident (LOCA).

The previous structural evaluations indicated that crack growth would be small leading to a virtual arrest condition prior to reaching about -180 of . the piping circumference (this maximum crack size expectation is still valid, see Section 2.1.2). The leakage for a 180 0 crack in the Brunswick core spray line was and is still conservatively estimated to be 20 gpm.

A crack growth rate of 0.5 inches for the previous operating cycle was estimated in the Reference 1 evaluation. This growth rate estimate has been reviewed in context with the actual crack growth rate experienced and has been updated as discussed in Section 4 of this report.

2-1

q 1

I*

)l g l

. . . 1 EAS-03-1090 I GE NUCLEAR ENERGY 1

[ 2.1.2 Core ~Sorav Pine Structural Inteority Review ~

The core spray pipe structural integrity evaluations documented in  :

Section 3 of the Reference 1 report were based on the Brunswick core

[ spray line pipe configuration and are not a function of the existing crack. ,

The basic- conclusions from the structural integrity evaluation are k summarized below:

i

1) .The driving force for crack extension is expected to-be small when the crack length reaches 1800 ~ of the piping circumference. Therefore, slowdown or virtual crack arrest is -

expected at this point.

2) A through wall crack of up to 2350 of the piping circumference can be tolerated without gross failure of the core spray line.

2.1.3 Lost Part Analysis Review l .

The lost part analysis documented in Section 4 of Reference.1 is k independent of the size of the crack and thus does not require updating.

The basic conclusion froa this evaluation is that no safety concerns are posed i,y any postulate 6 loose parts.

2.1.4 Loss-of' Coolant Accident Analysis Review The Brunswick LOCA licensing basis was epdated subsequent to the analysis documented in the Reference 1 report. Section 5 of this report replaces the Section 5 evaluation of Reference 1. The updated LOCA

-licensing basis more than adequately accounts for the effects of the core spray line crack on the LOCA response. Therefore, no restrictions on plant operation are required.

e i

1 2-2

3'.-

t s EAS-03-1090 GE NUCLEAR ENERGY

-2.2

SUMMARY

OF REVIEW One area of ~ the Reference 1 report, the crack growth rate estimate, was identified as requiring an update as a result of the new inspection'information from the outage. Also since the time the Reference 1 evaluations were performed, the LOCA licensing basis for Brunswick has been completely revised and updated. Although the Reference I conclusions were based on the previous LOCA licensing basis, T they- remain valid. The LOCA evaluation of the core spray line _ crack has, however, been updated to be consistent with the new LOCA licensing basis.

The necessary updates are documented in subsequent sections of this report. The updated evaluations are consistent with the previous conclusions and support continued operation of Brunswick Unit 2 for an additional cycle -(Cycle 9) without repairs to the existing core spray line crack.

M I 2-3

y ..

EAS-03-1090 GE NUCLEAR ENERGY

~

i 3.0 SAFETY EVALUATION CONCLUSIONS Based on the evaluations presented in Reference 1 as updated herein, it is concluded that continued operation with the current core spray line crack for one additional cycle - does not constitute an unreviewed safety question or a significant safety hazard for the reasons stated below:

1) The core spray line crack will not increase the probability of occurrence of any accident previously analyzed in the Updated i Safety Analysis Report (USAR) because the crack is located on the core spray piping inside the reactor pressure vessel and thus cannot cause a loss of reactor vessel inventory (i.e., a LOCA).

2) The consequences of any accident previously analyzed _ in the USAR (as amended by Reference 2) will not be increased. The crack may

-result in a bounding loss of core spray flow of 20 gpm and the

! Reference 2 analysis was performed with an assumed reduction in the delivered flow rate of 725 gpm from the designed pump flow rate.

Thus, the current analysis more than adequately accounts- for the potential flow loss through the crack.

'3) The core spray line crack does not increase the probability of occurrence of a malfunction of equipment important to safety because this could only occur if the core spray line function, to 1

deliver coolant to the core, could be defeated. The core spray-line function will not be impaired because:

a. The crack growth is expected to slow down significantly N prior to reaching about 180 of the pipe circumference, and j l

7 3-1 I

[s.*" ..

EAS 03 1090-GE NUCLEAR ENERGY 1 b. The structural integrity of the core spray line will be maintained for cracks up to 235 0 of the piping ,

circumference. 5

4) The consequences of a malfunction of equipment important to safety will not be increased as previously evaluated in the USAR (as amended by Reference 2) because the Reference 2 evaluation more

.. than adequately accounts for the potential core spray flow loss

[ through the crack.

_ 5) The probability of an accident or possibility for a malfunction of equipment important to safety of a different type than already -

evaluated in the USAR (as amended by Reference 2) will not - be increased because:

a. the crack is located on the core spray piping inside the 4

reactor pressure vessel and thus cannot cause a loss of reactor vessel inventory,

b. the core spray line structural integrity will be maintained during core spray injection and thus will be able to fulfil its function of supplying inventory to the core during a LOCA.

c. The existence of potential loose parts has been investigated and it was determined that if they occurred plant safety would not be affected.

6) The margin of safety as defined in the basis to any technical specifications will not be reduced. Currently, the Brunswick

) Unit 2 Technical Specifications require surveillance testing to demonstrate that the core spray pumps develop 4625 gpm of flow h 3-2

q

. .t EAS-03 1090 lJ, GE NUCLEAR ENERGY against a vessel head of 1113 psig. This flow requires.ent stems from the previous LOCA evaluation which conservatively assumed a L core spray flow rate of 100 gpm less than the actual design flow of-E 4725 gpm. Since a bounding value of the potential crack leakage is O much less than 100 gpm, the margin of safety is not reduced.

Additionally, the current LOCA licensing basis for Brunswick j~ assumes a core spray flow rate of 4000 gpm, allowing an actual  !

analytical margin of 725 gpm. (Technical Specification changes to M. reflect the new minimum core spray flow rate supported by the l Reference 2 LOCA licensing evaluation have not yet been submitted.)

i O

h0:

y

'O

b -

O-i OJ 3-3

f EAS-03-1090 GE NUCLEAR ENERGY l 4.0 CRACK GROWTH ESTIMATE UPDATE _

in the previous evaluation, it was estimated that the crack would '

grow approximately 0.5 inches during the previous cycle of operation. >

[

)

This estimate was based on an 18 month fuel cycle and assumed an IGSCC mechanism was responsible for crack growth (crack growth rate contributions due to thermal fatigue are negligible). It did not explicitly consider new crack initiation sites. The estimate of growth

/ rate was also based on the use of crack growth rates corresponding to water conductivity levels of approximately 1.0 yS/cm and assumed growth from both ends of the crack.

l j The actual crack growth experienced during the previous cycle L of operation (Cycle 8) was 0.15 inches on the outside of the pipe. This

[ is much less than the predicted value. However, on the inner surface, the increase in crack length (1.4 inches) is greater than the predicted.

[ value. The difference is probably due to additional crack initiation l sites circumferentially close to the previous indication where cold work I may exist. Additional crack initiation sites can lead to a longer crack ,

than that previously predicted. IGSCC initiation in cold worked low l

[ carbon stainless steel material has been observed in other BWR- plants L

and the NRC is aware of this. Differences in the actual water chemistry

! in the region of the crack compared to that assumed in estimating the I growth rate may also be a contributor to the differences in the actual b versus estimated crack growth rate.

Because of potential new initiation sites on the inner surface, it is difficult to predict crack growth for future operation. On the 7 outside surface, the crack growth estimate was conservative. This is

! expected since no new initiation is likely on the outside due to the l absence of cold work and since upper bound weld sensitized crack growth rates were used. The difficulty of predicting crack growth in regions J where new initiations are occurring is recognized in NUREG-0313. In NUREG-0313 it is recommended that a factor to accommodate the additional C 4-1

y , .:.

EAS-03-1090 GE NUCLEAR ENERGY

f. change in length compared to the depth change be applied. ,, Based on the previous inspection results, a factor of approximately 10 is realized (0.15inchesversus1.4 inches),

f These results can be used for the purposes of estimating crack growth during the next operating cycle. For the outer surface of the ,

pipe, the previous estimate of 0.5 inches is retained since it was ,

greater than the 0.15 inches of growth based on the inspection results.

For the inner surface, applying the factor of ten to the 0.15 inches of

) growth experienced on the outer surface, the estimated change in the f

, length . for the inside surface is 1.5 inches. These estimates are considered conservative since the change in compliance of the pipe (i.e., the stresses relax as the crack grows) should tend to slowdown the crack growth.

While new initiation sites cannot be ruled out, the driving force becomes small when the crack extends to approximately 180 of the piping circumference. .Thus, slowdown in crack growth and eventual arrest is

.likely. This estimate is based on through-wall cracking; since the

' actual crack profile is longer ~ on the inside than on the outside, some extension beyond 180 on the inside may occur. This would especially be h true if the region _of; cold work extends all the way around the inside l: ~

L circumference of the pipe. Despite this, crack arrest is still expected 0

I when the average crack length reaches 180 of the piping circumference.

1 1

1 1

L p- 4-2 L

I . _

U.

EAS-03-1090 GE NUCLEAR ENERGY

! 5.0 LOCA ANALYSIS UPDATE _

The Brunswick LOCA licensing evaluation has recently been updated utilizing the SAFER /GESTR-LOCA evaluation methodology (Reference 2). As a part of applying this . new methodology to Brunswick, CP&L revised j- several Emergency Core Cooling System (ECCS) parameters. These revisions were a part of a program which is designed _to establish new plant safety parameters and Technical Specifications. The minimum 7 allowable rated core spray flow rate (into the core region) in the ,

new LOCA licensing basis was assumed to be 4000 gpm. This corresponds-to a pumped core spray flow rate of 4100 gpm. (The 100 gpm difference conservatively accounts for the vent hole leakage which is actually j expected to be less than 13 gpm during core spray operation.) The core i spray design flow rate is 4725 gpm. Thus, a margin of 725 gpm exists between the expected design flow rate and that assumed in the LOCA licensing evaluation. This margin of 725 gpm is more than adequate to compensate for the calculated crack leakage of 7 gpm for the existing crack, or the 20 gpm maximum leakage expected for a- 180 0 crack during

core spray operation,.and.the vent hole leakage. Thus, the current LOCA licensing basis conservatively bounds any LOCA effects associated with the cracked core spray line and no additional evaluations are required.

)

l-D 5-1

-m___ _ _ _ -

_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ . _ _ _ _ _ _ _ _ _ _ _ _ _ _ . _ _ _ _ _ . _ _ _ _ _ _ _ _ . _ _ _ _ _ . _ _ . _ _ _ _ _ _ . _ _ . _ _ _ _ _ _ _ _ . _ _ _ _ ___ __ _ mm

b 1.

j(: , .. l EAS-03-1090 GE NUCLEAR ENERGY s

I

)b _

6.0 REFERENCES

1. " Core Spray Line Crack Analysis for Brunswick Steam Electric Plant 1: Unit 2", . General Electric Company, EAS-14 0388, March 1988. ,

2. " Brunswick Steam Electric Plant Units 1 and 2 SAFER /GESTR-LOCA i

Loss-of-Coolant Accident Analysis", GE Nuclear Energy, NEDC-31624P, Class III, Septenber 1988.

9 i;

. r 1

L3' 6-1 i.

, . .