ML20078A740

Provides Results of Analytical Evaluations of Cracking Identified at Quad Cities Nuclear Power Station Unit 1

ML20078A740
Person / Time
Site:	Quad Cities
Issue date:	06/13/1994
From:	Walsh R COMMONWEALTH EDISON CO.
To:	Russell W NRC OFFICE OF INFORMATION RESOURCES MANAGEMENT (IRM), Office of Nuclear Reactor Regulation
Shared Package
ML20078A742	List: ML20078A740 ML20078A745
References
NUDOCS 9407010114
Download: ML20078A740 (7)

Text

I ,

1

, a '

1 June 13,1994 I Mr. William T. Russell, Director Office of Nuclear Reactor Regulation U.S. Nuclear Regulatory Commission Washington, D. C. 20555 Attn: Document Control Desk

Subject:

Analytical Evaluation of Cracking Identified at Quad Cities Nuclear Power Station Unit 1 NRC Docket No. 50-254 l l

References:

(a) M. Lyster letter to T. Murley, dated June 6,1994, Providing Response to Request for Additional Information Conceming Core Shroud Cracking at Dresden, Units 2 and 3, and Quad Cities, Units 1 and 2.

(b) General Electric Company Report GENE-523-02-0194, dated March l

1994, Evaluatic,n and Screening Criteria for the Quad Cities 1 and 2 l Shrouds (Attachment 1).

(c) General Electric Company Report GENE-523-30-0294, Revision 1, dated June 1994, Recommended Inspection Criteria for the Quad Cities 1 and 2 Shrouds (Attachment 2).  ;

(d) General Electric Company Report GENE-523-A79-0594, dated June 1994, Evaluation of the Indications Found at the H5 Weld Location in the Quad Cities Unit 1 Shroud (Attachment 3).

Structural Integrity Report RAM-94-159, dated June 11,1994, j (e)

Evaluation of Flaws in Circumferential Core Shroud Welds at Quad

! Cities Unit 1 (Attachment 4).

l (f) General Electric Company Letter GLS-94-ll, dated June 8,1994, Response to Commonwealth Edison Technical Audit Questions Regarding the H5 Weld Flaw Evaluations for Dresden Unit 3 and Quad l l

Cities Unit 1.

l

Dear Mr. Russell:

In reference (a) Commonwealth Edison (Comed) submitted the results of the core shroud visual examinations and supplemental ultrasonic examinations performed at Quad Cities, Unit

1. The analytical evaluation of the core shroud cracking consisted of structural margin  ;

assessments utilizing limit load and, where appropriate, linear elastic fracture mechanics at g

n<nn -

9407010114 940613 PDR P

ADOCK 05000254 PDR h( / '

J

each horizontal weld location, H1 through H7 in accordance with ASME Section XI,IWB-3142.4. The purpose of this letter is to provide the results of the evaluations to the NRC staff for review in accordance with ASME Section XI, IWB-3144(b). The following is a synopsis of the evaluation results at each weld location. The detailed evaluations are contained in the j reference (d) and (e) reports. l During the current refuel outage (QlR13), visual inspections (VT) were performed on the core shroud welds H1 through H7. Subsequently, additional examinations using ultrasonics (UT) were performed to corroborate the results of the visual examinations.

I Using ASME Section XIIWB-3142.4 Acceptance by Analytical Evaluation, the evaluations l were performed to support operation, without repair of the shroud welds, for an 18 month operating cycle. The Screening Criteria, reference (b) forms the basis of the evaluation methodology in accordance with ASME Section XL The Inspection Criteria, reference (c), is also based on the Screening Criteria and was developed prior to the performance of the visual ,

inspections to provide the minimum acceptance standard for each weld, in the form of equally l distributed unflawed material. Due to accessibility limitations and inspection results, l additional analyses were performed to determine the actual structural margin based on ASME l Code factor-of-safety. The analysis for welds H1, H2, H3, H4, H6 and H7 is documented in reference (e). For weld H5, the inspection and screening criteria was not met. Therefore, additional analysis was performed and is documented in reference (d).

The inspection results, previously transmitted per reference (a), were utilized as input to the evaluations and are summarized below.

Additional evaluations on the H1 and H2 welds were necessary because the spacing of unflawed material varied slightly from the Inspection Plan. No indications were observed visually. The H2 visual inspection results were confirmed by UT. Approximately 78% (539 inches) of H2 was UT examined. Indications were noted at 20% (139 inches) of the weld with a maximum depth of 0.35 inches, with a majority of the flaw depths at 0.20 inches. No l 1

flaws were detected in the areas visually examined.

The H3 weld meets the applicable Inspection Criteria. The visual inspection was performed on 100% of the weld ID and OD surfaces.

The H4 weld visual inspections covered 66 (116 inches) of the ID and 90 (162 inches) of the OD with a total of 15 inches ofID/OD overlap. Two 1/2 inch long flaws were observed on the ID surface by VT. These flaws are separated by ~30 inches. No flaws were observed during the OD visual inspection. The H4 weld ID inspection results were bounding when l

compared to the OD inspection and , hence were used in the evaluation.

The HS visual inspection covered ~150 of the circumference and noted numerous, random linear indications in the ring material located below the weld at all locations inspected. 'Ihe observed cracking is not connected but has distinct starts and stops. Subsequently, UT examination was performed at four (4) locations for a total of 112 (207 inches). The UT recorded three (3) flaws, with a maximum depth of 0.57 inches. An assumed conservative  !

l

)

flaw depth of 1.24 inches for 360 was used in the evaluation coupled with a conservative crack growth rate of 5x10 4inches / hour. The results of the evaluation show an end of 18 month cycle flaw depth of 1.84 inches. This depth results in a structural margin factor of 9.7 when compared to an allowable crack depth of 2.88 inches using limit load analysis.

To aid in determining the root cause of the cracking, two (2) " boat" samples were removed from the H5 weld. The preliminary boat sample results are:

1. The cracking was caused by IGSCC.

2. The material composition is 304 stainless steel.

3. Irradiation effects were not observed.

4. Confirmed cold working depth up to 0.050 inches.

The H6 weld meets the applicable Inspection Criteria. The visual inspection was performed l

on the OD and covered 36% (231 inches) of the weld with one 7 inch flaw detected. The l

weld was also UT examined at four locations for a total of 6.5% (42 inches). One 2 inch flaw was detected on the ID. The UT examination was not performed in the area where the 7 i

inch flaw was detected. Therefore, the VT results were reduced by the UT results and used in the evaluation. 1 The H7 weld was evaluated because the areas inspected differed from the inspection plan due l to accessibility. The visual inspection covered 23% (146 inches) of the weld. This weld was l also UT examined at four locations for a total of 6.5% (42 inches). The UT areas coincided l i

with the areas visually inspected. There were no recordable indications at H7.

The analytical evaluation of the core shroud cracking consisted of structuial margm

)

assessments utilizing limit load and, where appropriate, linear elastic fracture mechanics (LEFM) at each horizontal weld location, H1 through H7. The structural margin assessments determined the minimum factor-of-safety available in terms of required unflawed areas for a 18-month cycle of operation at each weld location. The operating margin consists of any margin above the Code required minimum factor-of-safety of 1.4 under faulted conditions.

The following is a synopsis of the evaluation results at each weld location. The detailed evaluations are contained in the reference (d) and (e) reports. Also, the reference (f) letter provides the rationale for the structural analysis criteria and methods used in the reference (d) report.

J

A summary of the Code margins available at each location is as follows:

Weld Limit Load Factor of Safety g, ;

HI 7.1 H2 5.8 H3 56.3 i

H4 5.6 (1.7 LEFM)

H5 > 9.7 i H6 4.5 H7 1.5 l In closing, the above evaluation results, coupled with the substantial conservatisms that were built into the flaw evaluations, demonstrate that the flaws observed in the core shroud welds represent no immediate safety concern, and that all applicable ASME Code safety margins will be maintained well beyond the end of the next operating cycle for Quad Cities Unit 1.

If there are any questions concerning this matter, please contact this office.

Respectfully, 1 fNW Robert J. Walsh Core Shroud Project Manager l

Quad Cities Station cc: J. B. Martin, ilegional Administrator - RIII l

. C. Patel, Project Manager - NRR C. Miller, Senior Resident Inspector - Quad Cities l.

ATTACIIMENT 1 Genem! Electric Company Report GENE-523-02-0194, dated March 1994, Evaluation and Screening Criteria for the Quad Oties I and 2 Shmuds.

e l

h

O GE Nuclear Energy TECHNICAL SERVICES BUSINESS GENE-523-02-0194 GE Nuclear Energy DRF 137-0010-7  ;

175 Curtner Avenue, San Jose, CA 95125 Class II March 1994 I

i r

Evaluation and Screening Criteria for the Quad Cities 1 and 2 Shrouds i

i Prepared by: J W, F. Weitze, Seni/ Engineer Structural Mechanics Projects Verified by: @b '

H. S. Mehta, Principal Engineer Structural Mechanics Projects Approved By:

Dr. S. Ranganath, Manager Structural Mechanics Projects

.a=4"'*""***"*=+m%

j Gl@ yo6037Ds 44 - ,

GENuctear Energy GENE-52MMIN i

IMPORTANTNOTICE REGARDING CONTENTS OF THIS REPORT l

Please Read Carefully l l

The only undertakings of the General Electric Company (GE) respecting information in this document are contained in the contract between Commonwealth ,

I Edison Company and GE, and nothing contained in this document shall be construed as changing the contract. The use of this information by anyone other than Commonwealth Edison Company, orfor anypurpose other than thatfor which it is intended under such contract is not authori:ed; and with respect to any unauthori:ed use, GE makes no representation or warranty, and assumes no liability as to the completeness, accuracy, or usefulness of the information contained in this document, or that its use may not infringe privately owned rights.

GE Nostear Energy GENE-523 03-0194 l

Table of Contents l

PAGE l

\

1.0 INTRODUCTION

, . .. . . . . . . I 2.0 DETERMINATION OF THE EFFECTIVE FLAW LENGTH., .4 1 l

4 2.1 Proximity Rules.. . . . . l 2.1.1 Case A: Circumferential Flaw -- No Axial Flaw.. . .. .5 2.1.2 Case B: Circumferential Flaw - Axial Flaw . .6 2.1.3 Case C: No Circumferential Flaw -- Axial Flaw... .

.6 7

2.2 Application of Effective Flaw Length C.iteria.. .. ..

. . . . . . . . . 12 4 3.0 STRUCTURAL ANALYSIS.. . .. .. .

12 3.1 Applied Loads and Calculated Stresses.. ... . . . . .

l 15 3.2 LEFM Analysis... .. .. .. . . .

3.3 Limit Load Analysis.. . . . . . . . . . . . 16 16 3.4 Shroud Thickness Considerations. . . . . . .

3.5 References. . . . . . . . . . . . . . . .. .. . . . . . .. . . . . I7 4.0 ALLOWABLE THROUGH-WALL FLAWS .... . . . . .. . . . . . .. 21 4.1 Allowable Through-Wall Circumferential Flaw Size .. . . . . . . . . . .21 4.1.1 LEFM Analysis . ... ........ . .. . .. . . . . . .. 21 4.1.2 Limit Load Analysis........ . .. . . . . . . . . .22 4.2 Allowable Axial Flaw Size... .. . . . . . . . . . . . . . ... 22 4.2.1 LEFM Analysis... .. . . . . .. .. . . . . . . 22 4.2.2 Limit Load.. . . . .. .. . .. ... .. . ... . 23 4.3 References . .... .. . . . . . . . .... . . . . . . . . . . .. . 24 -

.25 5.0 SCREENING CRITERI A .... . . . . . .. . . . ... . . . . . .. . . . . . . . . . . . .

l j

i

- i l

GE Nuclear Energy GENE-523-03-0194 List of Tables PAGE Table 1-1: Conservative Assumptions included in Screening Evaluation. .3 Table 2-1: Flaw Combinations Considered in Proximity Criteria . . .5 Table 3-1: Dynamic Bending Stresses at Shroud Welds . . 13 Table 3-2: Pressure Differences . 13 Table 3-3: Shroud Weight and Seismic Shear Loads. 13 Table 4-1: Stresses and Allowable Flaw Lengths at Shroud Welds. .22 iii

2 GE Nuclear Energ GENE-523 02-0194 List of Figures PAGE Figure 2-1: ASME Code Proximity Criteria. .8 Figure 2-2: Application of Proximity Procedure to Neighboring Circumferential Flaws.. . .9 Figure 2-3: Application of Proximity Procedure to Neighboring Axial and Circumferential Flaws. 10 Figure 2-4: Process for Determining Effective Flaw Length. . I1 Figure 3-1: Sketch Showing Circumferential Welds in the Core Shroud. 18 Figure 3-2: Comparison of J-R Curves Developed for Two Irradiated Stainless Steel Specimens.. . . . .. . . .19 Figure 3-3: Schematic Illustrating Flaw Interaction. . . .20 1

1 l

l i

i IV

.s

GENaclear Energy GENE-53-02-0194

1.0 INTRODUCTION

In preparation for the Quad Cities 1 and 2 shroud inspections, Commonwealth Edison Company has requested GE to develop a screening criterion for indications that may be found at the shroud welds. Recently, indications have been discovered in some BWR shrouds as a result ofin-vessel visualinspection (IVVI). When indications are found by IVVI, only the lengths of the indications are known. Given that non-destmetive examination (NDE) of every visually detected indication would be difficult and time consuming, a method of screening indications for subsequent evaluation is required. This report presents such a screening criterion.

The guiding parameter used for the selection of the indications for further evaluation is the allowable through-wall flaw size, which already includes safety factors. If all of the visually detected indications are assumed to be through-wall, then the longest flaws, or combination of flaws, would have the limiting margin against the allowable through-wall flaw size. In reality, the indications are likely not through-wall, and therefore the criteria and methods presented in this report are conservative.

The result of this procedure will be the determination of the effective flaw lengths which will be used to compare against the allowable flaw size and selection ofindications for more detailed evaluation. The determination of effective flaw length is based on ASME Code,Section XI, Subarticle IWA-3300 (1989 Edition) proximity criteria. These criteria provide the basis for the combination of neighboring indications depending on various geometric dimensions. Crack growth over a subsequent cycle is factored into the criteria.

The proximity rules described here also conservatively assume that there is interaction between two perpendicular flaws. It is assumed that circumferential and axial indications could increase the effective flaw length depending on the unflawed distance between them.

This effective circumferential flaw length must be compared against the allowable circumferential flaw length. The effective axial flaw length would be compared against the allowable axial flaw length.

I

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

i GENuclear Energ GENE-523-02-0194 Flaws are considered in the same plane if the perpendicular distance between the planes is 4 inches or less. Any flaws which lie at an angle to the horizontal plane should be i

separated into a circumferential and axial component. These components can then be used l

separately in the determination of effective flaw lengths.

The selection ofindications for further investigation can be performed by evaluating the resulting effective flaw lengths. Indications with effective flaw lengths greater than l the allowable flaw sizes would require further characterization by NDE or more detailed analysis. The procedure described here is conservative, since all of the l indications are assumed through-wall and are being compared against the allowable I through-wall flaw size. l This report describes the following steps:

. Determination of effective flaw length including proximity criteria for adjacent l 1

flaws.

i

. Determination of allowable flaw sizes based on both linear elastic fracture l mechanics (LEFM) and limit load criteria.

. Screening criteria. l The report covers the limiting stresses for all the shroud welds (HI through H7 welds).

Therefore, the screening criteria developed here cover all shroud weld indications. A list of conservative assumptions used in this evaluation is summarized in Table 1-1.

2

i GE Naclear EnerO GENE-523-03-0194 l l

l Table 1-1: Conservative Assumptions included In Screening Evaluation

1. Postulated surface indications were assumed to be through-wall for analysis.

2. The bounding crack growth estimated for the next fuel cycle was included in postulated flaw lengths used for evaluation.

3. ASME Code primary pressure boundary safety margins were applied even though the shroud is not a primary pressure boundary.

4. ASME Code,Section XI proximity rules were applied.

5. A proximity rule to account for perpendicular flaws was applied, although not required by Section XI.

6. An additional proximity rule which accounts for fracture mechanics interaction  ;

between adjacent flaws was used.

7. Fracture toughness measured for similar materials having a higher fluence was  !

used (fluence comparable to end-of-life prediction), i

8. For welds H4 and H5, both LEFM and limit load analyses were applied, even though LEFM underestimates allowable flaw size, and is not required for austenitic materials.

l 1

3 i

GENaclear Eberzy GENE-523-02-Ol94 2.0 DETERMINATION OF THE EFFECTIVE FLAW LENGTH The effective flaw lengths are based on ASME Code,Section XI proximity criteria as presented in Subarticle IWA-3300. The procedure addresses both circumferential and axial flaws. Indications are considered to be in the same plane if the perpendicular distance between the planes is less than 4 inches. All flaws are considered to be through-wall. Therefore, indications on the inside and outside surface should be treated as if they are on the same surface. When two indications are close to each other, mies are established to combine them based on proximity. These rules are described here.

2.1 Proximity Rules l

The flaw combination methodology used here is similar to the ASME Code,Section XI l proximity rules concerning neighboring indications. Under the rules, if two surface indications are in the same plane (perpendicular distance between flaw planes < 4 inches) and are within two times the depth of the deepest indication, then the two indications must be considered as one indication.

In Figure 2-1, two adjacent flaws L1 and L2 are separated by a ligament S. Crack growth would cause the tips to be closer. Assuraing a conservative crack growth rate of 5x10-5 n/hr, i the crack extension, Aa, at each tip is 0.625 inches for an 18 month fuel cycle (12,492 hours0.00569 days <br />0.137 hours <br />8.134921e-4 weeks <br />1.87206e-4 months <br /> using a 95% capacity factor), and 0.833 inches for a 24 month fuel cycle (16,655 hours0.00758 days <br />0.182 hours <br />0.00108 weeks <br />2.492275e-4 months <br /> using a 95% capacity factor). Therefore, combining the crack growth and proximity criteria, the flaws are assumed to be close enough to be considered as one continuous flaw if the ligament is less than (2 x Aa + 2 x shroud thickness). For a shroud thickness of 2.0 inches, this bounding ligament is 5.25 inches for an 18 month fuel cycle and 5.67 inches for a 24 month fuel cycle. Thus, if the ligament is less than 2Aa + 2t, the effective length is (L1 + L2 + S + 2Aa). Note that the addition of 2Aa inches is to include crack growth at the other (non-adjacent) end of each flaw (See Figure 2-2).

If the ligament is greater than 2Aa + 2t, then the effective flaw length is determined by adding the projected tip growth to each end of the flaw. For this example, L1eff = L1 + 2Aa, and L2efr= L2 + 2Aa.

A similar approach is used to combine flaws when a circumferential flaw is close to an exial flaw (See Figure 2-3). If the ligament between the flaws is less than Aa + 2t, then the 4

GEhelear Energy GENE 523-03-0194 efTective flaw length for the circumferential flaw is Leff = L1 + S + Aa (the bounding ligament for these cases). If the ligament is greater than Aa + 2t, then the flaws are treated separately.

After the circumferential and axial flaws have been combined per the above c iteria, a map of the effective flaws in the shroud can be made, and the effective flaw length can be used for subsequent fracture mechanics analysis.

To demonstrate the proximity criteria, three examples are shown in Table 2-1 and described below.

Table 2-1: Flaw Combinations Considered in Proximity Criteria Case Circumferential Flaw Axial Flaw A Yes No B Yes ,

Yes C No Yes 2.1.1 Case A: Circumferential Flaw -- No Axial Flaw This case applies when two circumferential indications are considered. Figure 2-2a shows this condition. If the distance between the two surface flaw tips is less than 2Aa + 2t, the indications must be combined such that the effective length is (See Figure 2-2b):

L efr= L1 + S + L2 + 2Aa where: L1 = length of first circumferential indication L2 = length of second circumferential indication S = distance between two indications 5

switar Energy GEVE-523-02-0194 u the distance between the two tips is greater than 2Aa + 2t, the effective flaw lengths are (See Figure 2-2c):

Llefr= L1 + 2Aa L2eg= L2 + 2Aa 2.1.2 Case B: Circumferential Flaw -- Axial Flaw This case applies when both a circumferential and an axial flaw are being considered.

Figure 2-3a demonstrates this condition. For this case, only growth of the circumferential flaw is considered. If the distance between the circumferential indication tip and the axial indication is less than Aa + 2t, then the effective circumferential flaw length is ,

(See Figurc 2-3b):

L efr= L1 + S + Aa where: L1 = length of circumferential indication S= distance between the circumferential tip and i axial flaw.

and the effective axial length is (Figure 2-3b):

L ef r= L2 >S 2Aa where: L2 = length of axialindication If the distance between the circumferential indication tip and the axial indication is greater than Aa + 2t, then the flaws are not combined (See Figure 2-3c) and the effective lengths are:

l Llefr= L1 + 2Aa (for circumferential flaw)

L2eg= L2 + 2Aa (for axial flaw) 2.1.3 Case C: No Circumferential Flaw - Axial Flaw This case applies when only axial flaws are being considered. The effective length is determined in a manner similar to that used for Case A for circumferential flaws.

6 ,

GE Naslear Energy GENE-523-02-0194 2.2 Application of Effective Flaw Length Criteria The application of the effective length criteria is applied to two adjacent indications at a time. Figure 2-4 is a schematic which illustrates the process. For example, using the 0 azimuth as the starting location for a circumferential weld or plane, the general procedure would be as follows:

. Moving in the positive azimuthal direction, the first indication encountered is indication 1.

. The next indication is indicatico 2.

. Apply proximity rules to the pair ofindications (indications I and 2). Combine the flaws if necessary (L1 + L2 F S). If the flaws are combined, the resulting flaw becomes indication 2.

. Continue along positive azimuthal direction until the next indication is encountered. This becomes indication 3.

. Apply proximity rules to indications 2 and 3.

. Continue proximity rule evaluation until all indications along the subject weld or plane have been considered.

7 s

i l

l GE Naclear Energy GENE-523-02-bl94 \

i l

t )

w >- 1 1

I j j( '

s. j x., j

\  !

\ l l

l Combined L1 rwi  ;  !

f l Flaw l

/ l ,

/ l -

/  : /

f

! ,' /

/

V '/ D1 l

1 Jk  :

4 > l \

L  :

i s

D2 V 4 > j JL 'N i

O Fine 2 Nd Figure 2-1: ASME Code Proximity Criteria 8

GENuclear Energy GENE-52.1-02-0194

)

/ ~

As-Found

<,>  % j  !

v%

\ l

\*~-

1 " ~Y ~ *

- u __

+

1 f t

r

- - r- )

[ S < 2t+21 a  ;

, L,, = L1 + S+L2+ 23a j 2 L, > ,

3 l

1 I

S > 2t+21 a Lt.e =L1+2sa (c)  % \

L2,,,=L2+2se 4- u ,

- ~ -* Ml-a l

+

f Figure 2-2: Application of Proximity Procedure to Neighboring Circumferential Flaws 9

GE% clear Energy GENE-323 03-0194 j

w-w u

/

< . As Fevnd (a) m / \

\ ^ws

)Y l l

l

\ l 4- u -M-s->

l 1

reae : w (b)

)

{k.

[

S < 2t+ aa Li g = L1 + S ei a +

I

^^ # L2 =L2+2 aa en

) Y 4- u ->4 s ->

M L1 ., >

J

• • = w w i

A

{ ,,

S > 2t+ Aa (N " [ \ L1 en

.L1+2 as y L2,,=L2+2 as 4- of p c >

Figure 2-3: Application of Proximity Procedure to Neighboring Axial and Circumferential Flaws 10

GENaclear Entro GENE-523-03-0194 Start at Theta = 0 Move in + Theta Direction 1

I T

r t=1 l

lf First Flaw is Flaw i i

lf Next Flaw hl is Flawi+1 I Y

Apply Proximity Rutes to Flaws i and i+1 Y

Combine Flaws if Necessary and Determine Effective Length l

Y ini+1 if Flawis Combined Combined Flaw = Flaw i+1 f

I No /

Last Flaw?

Yes v

-> Done Figure 2-4: Process fur Determining Effective Flaw Length 11

=

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

GENcwlear Entry GENE-323-03-0194 3.0 STRUCTURAL ANALYSIS r The preceding section of this report described the determination of effective flaw lengths from the IVVI results. These efTective flaw lengths have to be compared to the allowable l flaw lengths to assess the structural integrity of the shroud. This section describes the j details and the results of the stmetural analysis performed to determine the allowable flaw lengths. The stmetural analysis consists of two steps: (1) the determination of axial and circumferential stress magnituies in the shroud, and (2) the calculation of the allowable I flaw lengths. Both the fracture mechanics and limit load methods are used in the calculation of allowable flaw lengths. ,

3.1 Applied Loads and Calculated Stresses The applied loads on tbc shroud consist ofinternal differential pressure, weight, and dynamic. The dynamic loads consist of a horizontal shear force and an overturning bending moment. The shear force acts in a direction which does not influence crack growth significantly, so it is not considered. The bending moment stress at a shroud cross-section varies as a function ofits vertical distance from the top of the shroud.

Because of the inherent ductility of the material (which will be discussed in Section 3.2 of ]

this report), residual stresses and other secondary stresses do not affect stmetural margin. ,

Thus, they need not be considered in the analysis.

The magnitudes of the applied loads were obtained from the dynamic stress analysis (Reference 3-1) and Updated Final Safety Analysis Report (UFSAR, Reference 3-2). The nominal shroud radius and thickness (Reference 3-3) were used to calculate the stresses .

from the applied loads. Stresses are calculated based on strength of materials formulas.

Figure 3-1 shows the weld designation and relative locations in the shroud. Table 3-1 shows the calculated dynamic bending stress magnitudes for both the upset and faulted conditions. The appropriate pressure differences for the normal / upset and faulted conditions are shown in Table 3-2. Axial membrane stresses are calculated based on these pressure differences, as well as cumulative weight (Table 3-3), vertical seismic (0.08 g's OBE,0.16 g's DBE), and buoyancy. Shear forces are given in Table 3-3, but, as

~

mentioned above, are not used in the analysis.

l s

12

.- l

GENuclear Energy GENE-12.102-0194 Table 3-1: Dynamic Bending Stresses at Shroud Welds Weld Moment, (in-kip) Stress. (ksi)

Designation Upset Faulted Upset Faulted HI 5.19x 103 1.04x104 0 07 0.14 112 116x104 2.31x 104 0.15 0.31 H3 1.24x104 2 47x104 0.19 0.37 114 4 31x104 8.61 x104 0 65 1.30 l I

H5 7.72x104 1.54x105 1.17 2.34 H6 7.96x 104 1.59x105 1.28 2.56 H7 1 13x105 2.26x105 1.82 3.63 Table 3-2: Pressure DifTerences Pressure DifTerences (psi)

Component Normal / Upset Condition Faulted Condition Shroud Head and 8 20 Upper Shroud Core Plate 17 30 Lower Shroud 25 43 ,,

Table 3-3: Shroud Weight and Seismic Shear Loads Effective Effective Shear Weld Wt.* (kips) Wt.* (kips) (kips)

Designation OBE DBE OBE H1 174.91 157.32 43 H2 197.87 177.97 338 H3 198.99 178.98 338 H4 254.89 229.26 415 HS 329.10 296.00 604 116 330.62 297.37 604 H7 345.71 310.94 592 These are cumulative weights, not lumped masses. Buoyancy and vertical seismic efTects are included.

The structural analysis for the indications uses two methods; linear elastic fracture mechanics (LEFM) and limit load analysis. Both the limit load and the LEFM methods were used in determining the allowable flaw sizes in the shroud. Since the limit load is conectned with the gross failure of the section, the allowable flaw length based on this 13 s

GENaclect Energy GENE-d3-03-0194 approach may be used for comparison with the sum of the efTective flaw lengths, determined in Section 2.2, of all the flaws at a cross-section. On the other hand, the LEFM approach considers the flaw tip fracture toughness and thus, the allowable flaw length based on this approach may be used for comparison with the largest effective flaw length, determined in Section 2.2, at a cross-section. The fluence levels at welds H1, H2, 113, H6, and 117 are such that no significant embrittlement effects are expected.

Therefore, only the limit load approach was used at these welds. The technical approach for the two methods is described next.

i 14

GEbelear Ervizy GENE-523-01-01N 3.2 LEFM Analysis The shroud material (austenitic stainless steel) is inherently ductile and it can be argued that the structuralintegrity analysis can be performed entirely on the basis oflimit load. In fact, J-R curve measurements (Figure 3-2) made on a core shroud sample taken from an overseas plant having higher fluence (8x102o n/cm2) showed stable crack extension and ductile failure. The ASME Code recognizes this fact in using only limit load techniques in Section XI, Subsubarticle IWB-3640 analysis. Nevenheless, a conservative fracture mechanics evaluation was performed using an equivalent Kjc corresponding to the materialI J c. The Kjc for the overseas plant shroud was approximately 150ksid. Use of this equivalence is extremely conservative since:

i) The actual fluences for Quad Cities 1 and 2 are lower than that for the overseas plant from which J-R curves were obtained.

l ii) The J-R curves show Jmax values well above the Ji c, confirming that there is load capability well beyond crack initiation (See Figure 3-2).

Also, Kjc is divided by ASME Code safety factor: 3.16 for normal and upset condition stresses, and 1.4 for faulted condition stresses. For the analysis presented here, the LEFM analysis is confined to welds H4 and H5. The fluence corresponding to welds at and below the core plate and above the top guide is an order of magnitude lower and the associated fracture toughness is comparable to that of the unirradiated material. For those locations, only the limit load analysis is used.

An additional consideration that applies only to the fracture mechanics analysis is the question, "When is a flaw independent of an adjacent flaw?" The ASME Code proximity rule described in Section 2 considers how flaws can link up and become a single flaw as a result of proximity. However, even when two flaws are separated by a ligament that exceeds 2Aa + 2t, they may not be considered totally independent of each other. That is, the flaw tip stress intensity factor may be affected by the presence of the adjacent flaw.

This can be accounted for by using the finite width correction factor for a flaw in a finite plate. For a through-wall flaw in an " infinite" plate, the stress intensity factor is:

K = cd(na)

For a finite plate, the K value is higher as determined by the finite width correction factor, F. In this screening evaluation it is assumed that the plate is " infinite" if the 15

t clear Energy GENE-523-02-Ol94 correction factor F is less than 1.1. As seen in Figure 3-3, if the width of the plate exceeds 2.5(L1 + 2Aa) (or a/b less than 0.4), then there would be no interaction due to plate end edge effects. If this same condition is applied to two neighboring flaws, then there will be no interaction between the two indications if the tips are at least 0.75(L1 + L2 + 4Aa) apart. Thus, if the distance between indications is greater than 0.75(L1 + L2 + 4Aa), then they may be considered as two separate flaws. If however, they are closer, for the purpose of fracture analysis, the equivalent flaw length is the sum of the two individual flaws including crack growth. Alternately, the precise equations using specific assumed flaw lengths and actual applied stresses may be compared to the appropriate allowables to account for interaction.

3.3 Limit Load Analysis A through-wall circumferential flaw was assumed in this calculation. Limit load calculations were conducted using the approach outlined in Subsubarticle IWB-3640 and Appendix C of Section XI of the ASME Code. The flow stress was taken as 3Sm. The Sm value for the shroud material (Type 304 stainless steel) is 16.9 ksi at the approximate normal operating temperature of 550 F.

Safety factors from the ASME Code (for circumferential flaws - 2.8 for normal and upset r.nd 1.4 for emergency and faulted, and for axial flaws - 3.0 for normal and upset and 1.5 for emergency and faulted) were used in the analysis. Separate criteria are prepared for each weld, based on location-specific stresses.

3.4 Shroud Thickness Considerations A shroud thickness of 2.0 inches was used in developing the screening criteria. However, there are locations in the shroud with wall thickness greater than 2.0 inches. Therefore, it must be detennined if the use of 2.0 inches is applicable to all other shroud locations.

The screening criteria based on the 2.0 inches thickness is considered applicable to locations of greater thickness since stresses were determined based on the 2.0 inch thickness. This results in conservative stress values when applied to locations with thickness greater than 2.0 inches, such as the weld between the 2.0 inch shroud cylinder -

and 2.5 inch top guide support ring.

l l

16 I

GE Nucitt.r Energy GENE.523-02-Ol94 3.5 References 3-1. Letter PBS9401 from P. B Shah (GE) to Hien Do (CECO) dated March 16,1994,

" Shroud Seismic Loads for Quad Cities."

3-2. Quad Cities 1 and 2 Updated Final Safety Analysis Report (UFSAR).

3-3. GE Drawings:

a. 718E861, Rev. 6, " Shroud - Spec, Control," Part 2, GE-NED, San Jose, CA.

b. 886D485, Rev. 4, "Re2etor Vessel - Spec. Control," Part 1, GE-APED, San Jose, CA.

l

]

I 17

GENE-5 3 02-0194 GENculcar Ezergy 0*

0' ,

Shroud Head

,) Hi H2 Top Guide _

Support Ring H4 H5 cor. PWe Support Ring O H6 Shroud Support H7 Ring

~- ,

0-0- '

l NOT TO SCALE 4

i l

l l

Note: Vertical weld locations are not shown for clarity. 1 l

l Figure 3-1: Sketch Showing Circumferential Welds in the Core Shroud 18

GENucleae Enero GENE-523-02-Ol94  ;

i

~~ '

0.3T - CT Specimens 2500 -- ,

'l 2000 --

o i I *

• i r g*

1 E o

^

1, 6

^ a

@ 1500 --

1 o, l 5

t I I o^

• I l

i ^ l

• ^

I l 1000 --

g o, '  ;

It I a 1ASB k I af I

• 1A6B

• l 1 500 --

6g g

?I I

I i 1

I I O -II l l l' l l l l 0 0.02 0.04- 0.06 0.08 0.1 0.12 0.14 Crack Extension, Delta a (inches)

Figure 3-2: Comparison of J-R Curves Developed for Two Irradiated Stainless Steel Specimens 19 x,

. gggg,33,_gg,,,,,

GE Nuke km .

A A I l

- m m-7 at F 00 1.0 2* 2. o1 1.006 8

M-> 4 _p 0.2 1.0246 l 03 1.0577 0.4 1.1094 0.5 1.1867 I

2n A > ) 1

\ ,. , .

I O

K=0 *F K=0

'K / )~ /

0.75(L1'+t2)

A >

-> 0.75L1' 4

-> 0.75L2 4-L1' L7 A W 4 >

-> 0.5L1' 4-

~W 4~

4 1.2512 >

4-1.25L1' +

' N 7 X Li' = L1 + 2 Aa L2 = L2 + 2 Aa Figure 3-3: Schematic Illustrating Flaw Interaction 20

l GE Nuclear Energy GENE-5D-02-0M 4.0 ALLOWABLE THROUGH-WALL FLAWS Allowable through-wall flaw sizes were determined using both fracture mechanics and limit load techniques for both circumferential and axial flaws. It should be emphasized that the allowable through-wall flaws are based on many conservative assumptions and are intended for use only in the screening criteria. More detailed analysis can be perfonned to justify larger flaws (both through-wall or part-through when measured flaw depths are available). However, since the intent of the screening criteria is to determine when additional evaluation or NDE characterization is needed, a conservative bounding approach is utilized.

4.1 Allowable Through-Wall Circumferential Flaw Size Both the LEFM and limit load methods were used to evaluate the allowable through-wall flaws. At welds H4 and H5, LEFM and limit load analysis methods were used, and the limiting locations for through-wall cracking occurred at the H5 weld. For the limit load analysis, the governing case is the H7 weld location where the pressure and dynamic stresses are high.

4.1.1 LEFM Analysis The total axial weight, pressure, and dynamic stresses are 0.66 ksi (weld H4) and 1.17 ksi (weld H5) for the upset condition and 1.63 ksi (weld H4) and 2.62 ksi (weld H5) for the faulted condition. Using the ASME Code safety factors for fracture analysis (3.16 for normal and upset and 1.4 for faulted), the faulted condition is limiting for H4 and upset is l

limiting for H5. 1 To determine the allowable flaw size based on LEFM methods, the conservatively

~

estimated inadiated material fracture toughness K cI value of150ksidin was used.

Applying a safety factor of 1.4 for the faulted condition, the allowable I K of ~ 107ksiE was obtained. The allowable flaw size was calculated using the following equation:

Ki = G.cJ(xa) where Gm is a curvature correction factor as defined in (Reference 4-1), o is the axial membrane stress, and 'a'is the half flaw length. The bending correction factor Gb. which varies through the wall from a positive to a negative value, and has an average of zero, 21

GE Nuclear Ewry GENE-523-02-0194 was not used since the objective is to obtain the average Kg through the thickness. The l allowable through-wall circumferential flaw length (2a) was determined as E 281 inches for H4 and 183 inches for H5.

4.1.2 Limit Load Analysis A through-wall circumferential flaw was assumed in this calculation. The limit load '

calculations were conducted using the approach outlined in Subsubarticle IWB-3640 and Appendix C of Section XI of the ASME Code. The flow stress was taken as 3Sm. The S value for the shroud material is 16.9 ksi at the approximate normal operating m

temperature of 550 F.

The stresses and allowable flaw length for the limit load analysis are shown in the table -i below. The allowable flaw length is based on the faulted condition, which was found to be limiting for each weld, and includes the ASNE Code,Section XI safety fr.ctors.

Table 41: Stresses and Allowable Flaw Lengths at Shroud Welds Weld Axial Force Stress (ksi) Bending Moment Stress (ksi) Allowable Flaw ,

Upset Faulted Upset Faulted Length (in) ,

H1 0.09 0.43 0.07 0.14 541 H2 0.07 0.41 0.15 0.31 532  ;

H3 0.05 0.37 0.19 0.37 501 0.01 0.33 0.65 1.30 466 H4 H5 0.00* 0.28 1.17 2.34 438 0.13 0.61 1.28 2.56 407 H6 H7 0.12 0.60 1.82 3.63 388

• The calculated value is negative and, therefore, conservatively assumed to be zero for allowable flaw calculations.

i 4.2 Allowable Axial Flaw Size 4.2.1 LEFM Analysis l

The allowable axial flaw size is governed entirely by the pressure hoop stress. As with the l circumferential flaw case, the allowable axial flaw size was determined assuming a ,

22 1

GENaslear Energy GENE 5:3 03-0194 through-wall flaw. For a through-wall flaw oflength 2a in the shroud, the applied stress intensity factor is given by:

K = M

• g,* M where M is the curvature correction factor given by:

2 M = [1 + 1.61a /(Rt)]o.5 (from Reference 4-2)

In the above expression, the allowable flaw length 2a can be determined by equaring the calculated K to the fracture toughness of 150ksid. The hoop stress for the faulted condition is 1.04 ksi, the ASME st.fety t' actor of 1.4 is applied and the result is used in the previous equation.

The allowable flaw length was conservatively determined to be 2a = 150 inches above the core plate.

4.2.2 Limit Load I

An alternate approach to determining the allowable flaw size is to use limit load techniques. The allowable flaw length is given by the equation:

j ch " Of /(M

• SF) i where M is a curvature correction factor as defined above, cr= 3Sm is the flow stress, SF is the safety factor (3.0 for upset conditions,1.5 for faulted), and ch = the hoop stress corresponding to the AP of 20 psi (faulted) above the core plate and 25 psi (upset) below the core plate. The allowable flaw length based on the limit load analysis is 706 inches above the core plate (using the limiting shroud diameter at welds H1 and H2) and 294 inches below the core plate. Since the value above the core plate exceeds the LEFM value, the allowable axial through-wall flaw length is 150 inches between H3 and H5.

l 1

I i

23 1

GENmelear Energy GENE-523-02-0194 4.3 References 4-1. Rooke, D.P. and Cartwright, D.J., " Compendium of Stress Intensity Factors," The Hillingdon Press (1976).

1 4-2. Ranganath, S., Mehta, H.S. and Norris, D.M., " Structural Evaluation of Flaws in Power Plant Piping," ASME PVP Volume No. 94 (1984).

l l

l I

l 24

GE Naclear Energy GENE-523-02-0194 l

5.0 SCREENING CRITERIA l

The determination of the allowable through-wall flaws has been described in Section 4. I The objective was to use the allowable flaw size as the basis for the screening criteria.

Since the screening rules represent the first step in the evaluation, they are by definition ,

conservative. If the criteria are exceeded, the option of doing further detailed evaluation or performing additional NDE remains. The allowable through-wall flaws were:

. Circumferential Flaws

- Hl: 541 inches (limit load only)

- H2: 532 inches (limit load only)

- H3: 501 inches (limit load only)

- H4: 466 inches (limit load), 281 inches (LEFM)

- HS: 438 inches (limit load),183 inches (LEFM)

- H6: 407 inches (limit load only)

- H7: 388 inches (limit load only) l

. Axial Flaws

- Above Core Plate: 706 inches (limit load),150 inches (LEFM)

- Below Core Plate: 294 inches (limit load)

A conservative approach in developing the screening rule is to include both the LEFM and limit load analysis. For circumferential flaws, LEFM provides the limit on an effective single flaw length for H4 and H5, while the limit load analysis provides the limit on effective cumulative flaw length. For axial flaws, the allowable flaw length is 706 inches between H1 and H3,150 inches between H3 and H5 (LEFM), and 294 inches below the core plate (limit load).

For circumferential flaws at welds H4 and H5, tne limits are applied as follows. At weld H5, for example, the fracture mechanics based limit for a single effective flaw length, as determined in Section 2.2, is 183 inches. This in itselfis not sufficient, since there could be several flaws (each less than 183 inches) in a circumferential plane that cumulatively add up to greater'than 438 inches (the allowable circumferential flaw size based on limit load analysis). Thus, the sum of the effective flaw lengths, as determined in Section 2.2, should be less than 438 inches.

25

GE Nuclear Energy GENE-5:3 02-0194 When considering LEFM based evaluations, the crack interaction criteria described in Section 3.2 must be applied in comparing against the allowable lengths. For example, for adjacent flaws where the spacing, S, is less than 0.75 (L1 + L2 + 4Aa), the length L = Ll' + L2' is used for comparison with the LEFM based allowable flaw length. The lengths Ll' and L2' are as determined in Figure 3-3.

The criteria presented in this report are conservative in that continuous flaws (for limit load) were assumed. Additional analysis assuming the flaws are non-continuous (that is, distributed around the circumference of the shroud) or part-through wall will yield larger cumulative flaw lengths.

l l

l I

l l

l I

i f l 26 i

)

.- ~ .- - -. . . ..

ATTACIIMENT 2 General Electric Company Report GENE-523-30-0294, Revision 1, dated Anne 1994, Recommended Inspeedsc Criteria for the Quad Cities 1 and 2 Shmud;.

I i

i

O GE NuclearEnergy TECHNICAL SERVICES BUSINESS GENE 523-30-0294 GE NuclearEnergy Revision 1 175 Curtner Avenue, San Jose, C A, 95125 DRF 137-0010 7 Class D June 1994 Recommended Inspection Criteria forthe Quad Citles 1 and 2 Shrouds 1

l l

Prepared by: b I, 6 khl/

O. L.3tevens, Senior Engineer ' '

Structural Mechanics Projects l

'- C A}4y Verified by:

H. S. Mehta, Principal Engineer Structural Mechanics Projects Approved By: I b Apr ' kY '

Dr. S.%nganath, Manager l Structural Mechanics Projects

l 1

GENudearEmrgy GEhE SDJMD4, Rev.1 i l

l IMPORTANTNOTICE REGARDING CONTENTS OF THIS REPORT Please Read Carefully l The only undertokings of the General Electric Company (GE) respecting information in -

this document are contained in the contract between Commonweahh Edison Company and GE. l and nothing contained in this document shall be construed as changing the contract. The use of l this information by anyone other than Commonwealth Edison Company, orfor any purpose other than thatfor which it is intended under such contract is not authorized; and with respect to  ;

any unauthorized use, GE makes no representation or warranty, and assumes no liability as to the completeness, accuracy, or usefulness of the information contamed in this document, or that its use may not infrhtge prhutely owned rights.

l l

l i

i l

~s

,;9. ' eg'k :.

~

0 4 IMAGE EVALUATION ,

/ 4  %

- TEST TARGET (MT-3) .

pp ///g .

" 4?l 9,,,,/ .gs ,

4 j+

1.0 h22 l,l l323 l@

f.25 1.4 l

.-__ l limme 1.6 4 - - - -

- --- 15 0 m m -- - - >

6" - - - - - - - -- -

i Ox

, a b

l4~~//fA y,J>p,&y,,,

Q;f/h o

yg;

- ..n

x R ^O '

s%

' C' 1

IMAGE EVALUATION /

z..

A%

s

~ hf* TEST TARGET (MT-3)

/ f)[f /g,

////

'V$f pf k"; th'

+4 s 1.0 m J '4 122 d a _t l 1.8 imm 1.25 1.4 i 1.6 l

_=

- - -- 15 0 m m >

4 -

_..--------6" >

4_ _ _ - _

a%

p

/+

$r,f! ?ty lsp% 4 4$~$ O egn, p sg&

N I  ;.

alb

.kh?es',,

% ?> '

P 1

IMAGE EVALUATION

[O s

, '~

+ 4%

%. -: .- j/

TEST TARGET (MT-3) ,

//o/

l.0 #?S Jlll05 c

4.

(

2 1.25 1.4 I

\ _1.6_

4 . _...-- - - 150mm >

4 _ _ _ _ . _ _

_. 6 " >

ff

. n ,s *h// 4 fkzzzz

,o,h, u. c= z b p s

• g ,3 Of f(h!b 'f4<'

f(64 s>

~

---

_-__:_-t___-_.-: ____._s D'^ ?lY -

N

>> .e ,

[j[/7

' gp*'

IMAGE EVALUATION  %

Q/ - O 4),

N\ /o/

TEST TARGET (MT-3)

%$9 If/2 V

l.0 '

Il 2 2 l,l  !$li li, S z2 0 lM4 l.25 1.4 41

==

\1 m.6 m 4 . . ---- - 15 0 m m - - - - - - - - - - - - - >

6" - - - - - - - - - - - - - - - - -

h:Y N l~%.y,y, , .

f+ // p ' \

////p y;;&,,,,,,,,  ?;3

. n

ass.a ra.e _ _ cz s c.s s.s u zoa,aa. 2  !

J l

Table of Contents .

PAGE ,

1

1.0 BACKGROUND

.. .... ..... . . . . . . . . . . . . . .... ...................................]

1 2.0 TECHNICAL APPROACH.... . .... . .. . . . . .. . .. .. . . . ... .. . .. ..... . . ... ... . . . . . ... . . .. . . . .2 3.0 INSPECTION CRITERIA. .. .... . . . . .. .. .......................................6 .

4.0 REFERENCES

........ . . . .. . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . .......7 List of Tables PAGE Table 1: Limit Lond/LEFM Results for the Next Fuel Cycle... . .. . . ... .4 Table 2: Limit Lond/LEFM Results for the Nex: Two Fuel Cycles. . . . . . .5 List of Figures PAGE Figure 1: Shroud Inspection Flow Diagram............. ......... . ...... .. .. .. ......8 Figure 2: Limit load Approach .. . ......... .. . . . . . .. ... . .. ..........................9 11

GENandaarFanvar GENEJ8 MOM 94 Rev. I

1.0 BACKGROUND

The purpose of this report is to develop inspection criteria for the Quad Cities 1 and 2 shrouds in accordance with recommendations presented in GE Senices Information Letter (SIL)

Number 572, Revision 1 (SIL 572).

SIL 572 recommends examinations of accessible areas on both the inside diameter (ID) and outside diameter (OD) surfaces of the core shroud at the next scheduled refueling outages for all plants with Type 304 stain'ess meel shrouds with six or more years of power operation, and for all plants with L-Grade stainless steel shrouds with eight or more years of power operation.

Power operation is defined as operation where the reactor is above 200'F. The Quad Cities I and 2 shrouds are fabricated from 304 stainless steel, and each plant has more than 6 years of power operation; therefore, inspection of the shrouds is warranted during the next scheduled refueling outa8es.. SIL 572 recommends that any visual examinations be performed with an enhanced VT-1 system that can resolve a standard one (1) mil wire on the inspection surface. If no cracks are observed, it is recommended that the shroud be re-examined at every second refueling outatre. If cracking is observed, the shmud should be re-examined during each refueling outage, and a structural margin analysis should be performed to assess operability. The inspection sample should be based upon a statistically significant sampling of the accessib e areas.

Figure I shows, in a flow diagr.sm format, the steps involved in the evaluation of the inspection results. The Reference 1 report documents screening criteria which may be used as a basis for dispositioning flaws discovered in the shroud during inspection, as suggested by the

" COMPARE AGAINST VISUAL SCREENING CRITERIA" box in Figure 1. Reference I uses limit load and linear elastic fracture mechanics (LEFM) techniques to determine allowable flaw sizes musumkg continuous length flaws. Altemate screening criteria are presented in this report which use similar methodology to that used in Reference 1, but take account for non-continuous flaws (i.e., separate uncracked regions distributed around the shroud circumference). This shamate approach is based on the premise that inspection times may be significantly shorter if the objective is to find the minimum amount of material needed to maintain structural margins rather than determining the full extent of any cracking which might be present.

The altemate approach described here, along with SIL 572, provide a basis for the plant-specific recommendations made for Quad Cities I and 2. The recommended inspection plan that follows can be performed by using enhanced visur) examination, ultrasonic examination or some 1

GENedear Emra GENE 5B.30 04, Rev )

1 l

combination thereof There are distinct advantages and disadvantages associated with each of these inspection techniques which should be considered before selecting the inspection method, such as the cost to perform each type of exam and the associated impact on critical path time.

The relative merits of each inspection technique are not discussed in this report. j i

l 2.0 TECHNICAL APPROACH l

In this section, inspection criteria are developed which are used as a basis for the inspection recommendations which follow. The methodology used here is consistent with the limit load and LEFM techniques presented in Reference ? However, the criteria developed here i take into account distribution of uncracked material around the circumference of the shroud This approach is less restrictive than that used in Reference 1, where it was assumed in the limit load i approach that cracks were condnuous. Therefore, the technical approach used here is to find the minimum amount of uncracked material at each weld location to meet the necessary structural  ;

margins (including safety facters). The condition of the uninspected locations of each weld remains unknown with this approach; in fact, much ofit may be uncracked. However, for this conservative approach, it is assumed that they could be cracked. This assumption has no consequence to the structural adequacy of the shroud ifit can be shown that the inspected regions are adequate from a structural standpoint, taking into account the necessary safety factors and  !

future crack growth.

J The ilmit load approach used here is depicted in Figure 2. In this figure, four equally distributed uncracked regions have been assumed. The length of each of these regions is to be determined for each shroud horizontal weld such that structural margins applicable to limit load methodology are realized and, where applicable, margins resulting from a LEFM approach are realized as well. From Reference 1, crack growth, Aa, is estimated to be 0.625 inch for an 18 month fuel cycle and 0.833 inch for a 24-month cycle. Therefore, the analyzed length of each uncracked section usei in this evaluation was assumed to differ from the minimum required 1 inspection length by twice the crack growth (i e., 2 x Aa, assuming crack growth from both ends f of the uncracked region). The same methodology applies for any other number of equally-spaced uncracked regions.

The neutral axis shown in Figure 2 is first determined by equilibrating the force resuhing from the applied membrane stress, P , in the uncracked cross section with the force resulting 1 from a stress equal to the flow stress in each of the uncracked regions. The flow stress of the 1

2 l l

l

GENacieer Enery GENT.-523-30-0294, Rev.1 materialis taken to be 3S m . Once the stress distribution in the uncracked regions is determined, the resulting moment about the centroidal axis is calculated, :;nd this is used to determine an equivalcat bending stress, Pg, in the cracked section. Finally, consistent with ASME Code,Section XI, IWB-3640 procedures, PJ + Pmis compared to the following:

(Pb + P )

• S.F. l where: Pb = Applied bending stress in uncracked section.

Pm = Applied membrane stress in uncracked section. -

S.F. = Safety factor consistent with Appendix C of Section j XI of the ASME Code.

= 1.4 for fsulted conditions

= 2.8 for normal / upset conditions i

The lengths of equally spaced, uncracked locations is stocturally adequate if )

(Pb+ P )

• S.F. is less than P ' + P m . As a final step, verification that the length obtained from j this limit load approach exceeds the proximity criteria for adjacem flaws (from Reference 1) is l perfortned. l I

Consistent with Reference 1, LEFM proximity limits must also be satisfied for welds H4 and H5. Calculations were performed for welds H4 and H5 to account for the LEFM cffect adjacent flaws have on each other to ensure that flaws separated by the uncracked region length are acceptable (assuming the emire distance between uncracked regions was cracked). These l calculations resulted in a minimum distance between flaws such that the stress intensity, K,  !

(including the appropriate safety factor) was within the allowable material toughness value of 150 ksidinch used in Reference 1. The minimum spacing between flaws at these locations is the limiting G.e., greater) result ofboth the limit load and LEFM methodologies.

Based on the loading described in Reference 1, the results of the limit load approach (as  !

well as the LEFM approach for H4 and HS) described above are given in Table 1 for each weld for the next fuel cycle (18-month cycle). Based on Commonwealth Edison Company's intention of switching to a 24-month fuel cycle after the next cycle, results are also given in Table 2 for the next two fhel cycles (one 18 month cycle and one 24-month cycle). The weld locations are as identified in Reference 1. These results demonstrate that equally spaced uncracked regions, with quantities and lengths corresponding to that shown in Tables 1 and 2, provide adequa:e material to maintain au structural margins.

l 3

GENuclear Energy GEVD.523J6.MN, Men 1 Table 1: Limit Load /LEFM Results for the Next Fuel Cyde Number of Minimum Required Weld Lhniting Equally-Spaced, Inspection Length per  :

Condition Uncracked Uncracked Region (l) j Regions (inches)

Hi Fauhed 4 5.25 H2 Fauhed 4 5.25 H3 Fauhed 4 5.25 H4 Fauhed 4 18.75(2.3)

H4 (alternate) Fauhed 8 5.25(2)

H5 Upset 8 7.75(2,3)

H6 Fauhed 4 13.25 i I

H7 i Fauhed see note 4 see note 4 NOTE (1) From Reference 1, crack growth is estimated to bc 0.625 inch for an 18-month fbcl cycle. Therefore, the length of each nneracked section used in the evaluation was assumed to be the minimum inspection length reported here less crack growth from both sides of the uncracked region.

Additionally, this length must be greater than the proximity criteria spacing requirement of Reference 1. Thus, the following was used in the evaluation for weld H1:

Proximity criteria spacing requirement from Reference 1 = 5.25" Lagth used in evaluation = 3.00" Inspection length = Length + crack growth = 3.00" + 2(0.625")

= 4.25" Thus, the minimum required inspection length is 5.25" (2) For these locations, LEFM techniques ut applicable due to fluence considerations.

(3) The LEFM proximity criteria produced more limiting results than the limit load criteria, so the LEFM limits are tabulated at this location.

(4) Due to access restrictions at the H7 weld, the following distribution of uncracked regions was evaluated (refer to Figure 2): 34.25' at 90" and 270', and 7.25" at 15',45',135',195", 225' and 315". These lengths are inspected lengths, so the actual leogths analyzed tooit into account crack growth, as pernote 1.

4

l GEN:eJeer Enegy GENC-313.J6 N94, Rev.1 Table 2: Limit I4ad/LEFM Results for the Next Two Fuel Cycles Number of Minimum Required Wekt Limitingr.ondition Equally-Spaced, Inspection Length per Uncracked Uncracked RegionO)  ;

Reasons (inches) l H1 Faulted 4 6.92 H2 Faulted 4 6.92 H3 Faulted 4 6.92 H4 Faulted 4 20.42(2,3)

H4 (alternate) Faulted 8 6.92(2) i HS Upset 8 9.42(2,3)

H6 Faulted 4 14.92 H7 Faulted see note 4 see note 4 ._

NOTER (1) From Reference 1, crack growth is estimated to be 0.625 inch for an 18-month fhel cycle and 0.833 inch for a 24-month fuel cycle. Therefore, the <

length of each uncracked section used in the waluation was assumed to be the minimum iWon length reported here less crack growth from both i sides of the uncracked region. Additionally, this length must be greater than the proximity criteria spacing requirement of Reference 1. Thus, the '

following was used in the evaluation for weld H1:

Proximity criteria spacing requirement = 4.0+2(0.625"+0.833") = 6.92" Length used in evaluation = 3.00" Inspection length - Length + crack growth = 3.00" + 2(0.625"+0.833")

=5.92" Thus, the minirmrm required inspection length is 6.92" -

t (2) For these locations, LEFM techniques are .yylkJ,le due to fluence considerations.

(3) The LEFM pua4 criteria produced more limiting results than the limit load criteria, so the LEFM lirnits are tabulated at this location.

r (4) Due to access restrictions at the H7 weld, the following distribution of uncracked regions was evaluated (refer to Figure 2): 35.92" at 90' and 270*, and 8.92" at 15*,45*,135",195", 225* and 315*. These lengths are inspected lengths, so the actual lengths analyzed took into account crack growth, as per note 1. l k

5 t

l l

GENudoet Eny GENE-323-30-D294, Rev.1 3.0 INSPECTION CRITERIA Based on the limit load /LEFM approach described in Section 2.0, the recommended inspection criteria for Quad Cities I and 2 are summarized below- ,

i l

1. Based on the results of Tables 1 and 2, inspect equally spaced regions around the j circumference of the shroud for each of the horizontal welds HI - H7 (except as I noted for H7), in quantities as noted in Tables 1 or 2. The length of each inspected l region should be at least equs] to that shown in Tables 1 or 2 in order for all .

structural margins to be met. Since the approach used here is to identify regions which have uncracked material, inspection of accessible areas that meet the criteria presented here should be performed for both the ID and OD surfaces, if accessible, to verify the absence of cracking. However, considering the various factors (end grain effects, fluence, environment / water chemistry and potential inaccessibility), it is recognized that the examination of the HI OD, H2 OD, ID ID, H4 ID, H5 OD,  !

H6 OD and H7 OD surfaces will provide the most critical data for use with this t

examination approach (e.g., the areas noted are judged to be more susceptible to  !

crack initiation).

2. If the extent of uncracked rnaterial identified in step (1) is insufficient to meet the limit load /LEFM results presented in Tables 1 or 2 for any given weld, additional areas of that we!d must be examined in order to demonstrate structural margin through analysis #) The extent of the additional examination is dependent upon the amount of uncracked material identified in the initial regions. Using an approach similar to that described in Section 2.0, additional regions distributed around the circumference can be identified for further inspection with the intent of l locating an amount of distributed, uncracked material which satisfies structural margins (limit load for all welds as well as LEFM for welds H4 and H5). This iterative approach will be pursued by Commonwealth Edison Company during the shroud inspection based on the results achieved.

3. From Reference 1, the allowable crack length for axial welds is longer than the width of the plate material used in the shroud fabrication. Based on this, it is not possible that axial flaws would exceed allowable flaw lengths. Therefore, examination of vertical welds is not necessary for demonstrating structural margins.

6

GENuclearFaergy CENE.$2MM294, Rev. I

4. In addition to the areas identified above, inspection of additional areas of the shroud should be considered based on a review of the fabrication records contained in Reference 2. The information contained in Reference 2 suggests 4 actions may have been taken which would cause local areas of cold work on the shroud. Because of the known susceptibility of such areas to IGSCC, inspection of these areas is appropriate. However, the amas of cold work were not identified in Reference 2. Based on this, a location-specific inspection recommendation cannot be provided. However, if cold worked locations are identified during the.

inspection, it is recommended that visual inspection of these local areas also be performed.

NOTE: (a) It is recognized that access limitations at the H7 weld may preclude the possibility ofidentifying sufficient uncracked material to positively demonstrate structural margin through limit load analysis. Should this prove to be the case, a conclusion as to the integrity of the weld may be based upon the results of the examinations performed.

4.0 REFERENCES

[1] GENE-523-02-0194, Revision 0, " Evaluation and Screening Criteria for the Quad Cities I and 2 Shrouds," W.F. Weitze, GE Nuclear Energy, San Jose, CA, March 1994.

[2] GENE-771-04-0194, Revision 0, " Shroud Fabrication and Operational History Data, Quad Cities 1," G. L. Hodson, GE Nuclear Energy, San Jose, CA, March 1994 7

[ ._. .- _. .

GDE 523JM194, Ren i GENadnerEnergy VLSUALLY INSPECT HORIZONTALWELDS lI COMPARE AGUNST O.K.

VISUAL SCREENING CR;TERIA NOT OX lI EXAAND VTINSPECTION TO WTOK '

---

LOCATE ADDITIONALREGIONS QF UNCRACKED MATERIAL OK lf

} CHARACTERtIE FLAWS PERFORM UWT LOAD ANALYSIS NOTOK BY UT OF UNCRACKEDREClONS IDENTIFIED BY E7 AMINATION OL

! U O.K.

EVALUATE UT RESULTS BY ANALYS S NOT0X 1f 1f l 1I aHRoUo ggpyg STRUCTURAL iMTEGRRY Z AssupEo Figure 1: Shroud Inspection Flow Diagram S

i l

GENU &aarJOsexy GDE.523 30 0294, Reu i Stress Distributions At Net Section Collapse Cross Sections o.

Pm P,+P ,

i ml m7 9 '

270* 7 r- 90' /

Neutral Axis

, m i

I '

180' ,

0 Uncracked Section

- 4 Equally Spaced Uncracked Regions l0' h I 27o=

fa-f" 90' Neutral Axis

.y r m

[--

C r

,p. ,-

A 180' Cracked Section Figure 2: Limit Load Approach 9

ATTACllMENT J f

+

General Electric Company Report GENE-523-A79-0594, dated hat 1994, Evaluation of the Indications Found at the IIS Weld Location in ths, Quad Cities Unit 1 Shmud.

1 i

4 t

I i

- -