ML111810286

Handbook on Flaw Evaluation Kewaunee,Unit 1 Steam Generators Upper Shell to Cone Weld.

ML111810286
Person / Time
Site:	Kewaunee
Issue date:	05/31/1991
From:	Schmertz J, Adamonis D Westinghouse, Div of CBS Corp
To:
Shared Package
ML111661099	List: ML111661097 ML111810286
References
NUDOCS 9109050080, WCAP-11476, WCAP-11476-R03, WCAP-11476-R3
Download: ML111810286 (213)

Text

I Docket #& 4, AccessiOfl # C L of LUr

--NOTICE THE ATTACHED FILES ARE OFFICIAL RECORDS OF THE INFORMATION &

REPORTS MANAGEMENT BRANCH.

THEY HAVE BEEN CHARGED TO YOU FOR A LIMITED TIME PERIOD AND MUST BE RETURNED TO THE RE CORDS & ARCHIVES SERVICES SEC TION P1-22 WHITE FLINT. PLEASE DO NOT SEND DOCUMENTS CHARGED OUT THROUGH THE MAIL. REMOVAL OF ANY PAGE(S) FROM DOCUMENT FOR REPRODUCTION MUST BE RE FERRED TO FILE PERSONNEL.

-NOTICE-PDR ATXIOC 0C500012-0E:5 F . IPDR..

WESTINGHOUSE PROPRIETARY CLASS 3 WCAP-11476 Revision 3 HANDBOOK ON FLAW EVALUATION KEWAUNEE UNIT 1 STEAM GENERATORS UPPER SHELL TO CONE WELD May 1991 W. H. Bamford D. S. Drinon D. Kurek Y. S. Lee R. D. Rishel Verified by: CJ

/J. C. Schmertz, PE Structural Mechanics Technology Approve D. C. Adamonis, Manager Materials, Mechanics & Diagnostics Technology Although information contained in this report is Westinghouse Proprietary Class 3, no distribution shall be made outside Westinghouse or its licensees without the prior written approval of Westinghouse Electric Corporation, Energy Systems Business Unit, Nuclear and Advanced Technology Division.

Westinghouse Electric Corporation Nuclear and Advanced Technology Division P. 0. Box 2728 Pittsburgh, Pennsylvania 15230 1991 Westinghouse Electric Corporation

WESTINGHOUSE PROPRIETARY CLASS 3 TABLE OF CONTENTS SECTION TITLE PAGE 1 INTRODUCTION 1-1 1.1 Code Acceptance Criteria 1-2 1.1.1 Criteria Based on Flaw Size 1-2 1.1.2 Criteria Based on Stress 1-3 Intensity Factor 1.1.3 Primary Stress Limits 1-4 1.2 Geometry 1-4 2 LOAD CONDITIONS, FRACTURE ANALYSIS METHODS AND MATERIAL PROPERTIES 2-1 2.1 Transients for the Steam Generator 2-1 2.2 Stress Intensity Factor Calculations 2-1 2.3 Fracture Toughness 2-3 2.4 Critical Flaw Size Determination 2-5 3 FATIGUE CRACK GROWTH 3-1 3.1 Analysis Methodology 3-1 3.2 Stress Intensity Factor Expressions 3-2 3.3 Crack Growth Rate Reference Curves 3-3 3.4 Fatigue Crack Growth Results 3-5 4 SURFACE FLAW EVALUATION 4-1 4.1 Scope of Evaluation 4-1 4.2 Code Criteria 4-1 4.3 Basic Data 4-2 4.4 Typical Surface Flaw Evaluation Chart 4-4 4.5 Procedure for the Construction of a 4-5 Surface Flaw Evaluation Chart 5 EMBEDDED FLAW EVALUATION 5-1 5.1 Scope of Evaluation 5-1 i

WESTINGHOUSE PROPRIETARY CLASS 3 TABLE OF CONTENTS SECTION TITLE PAGE 5.2 Embedded vs. Surface Flaws 5-1 5.3 Code Criteria 5-2 5.4 Basic Data 5-3 5.5 Fatigue Crack Growth for Embedded Flaws 5-4 5.6 Typical Embedded Flaws Evaluation 5-5 Chart 5.7 Procedure for the Construction of 5-7 Embedded Flaw Evaluation Charts 5.8 Comparison of Embedded Flaw Charts 5-8 with Acceptance Standards of IWB-3500 6 FLAW EVALUATION CHARTS-UPPER SHELL TO 6-1 CONE WELD 6.1 Evaluation Procedure 6-1 6.2 Modification of Hydrotest and 6-4 Leak Test Temperatures 7 REFERENCES 7-1 APPENDIX A RESULTS OF THE INSPECTION OF SPRING 1987 A-1 APPENDIX B RESULTS OF THE INSPECTION OF MARCH 1988 B-1 ON STEAM GENERATOR "A" APPENDIX C RESULTS OF THE INSPECTION OF APRIL 1991, C-1 KEWAUNEE STEAM GENERATORS "A" AND "B" ii

SECTION 1 INTRODUCTION This flaw* evaluation handbook has been designed for the evaluation of indications which may be discovered during inservice inspection of the Kewaunee Unit 1 steam generators. The tables and charts provided herein allow the evaluation of any indication discovered in the upper shell to cone weld region without further fracture mechanics calculations. The fracture analysis work is documented in this report. Use of the handbook will allow the acceptability (by analysis) of larger indications than would be allowable by also only using the standards tables of the ASME Code Section XI. This report provides the background and technical basis for the handbook charts. This handbook was prepared as a result of the discovery of indications in the upper shell to cone weld of the "B" steam generator in spring of 1987. Details of these indications and their evaluations are contained in Appendix A.

The geometry of this region is shown in Figure 1-1.

The highlight of the handbook is the design of a series of flaw evaluation charts for both surface flaws and the embedded flaws. Since the fracture mechanics characteristics of the two types of flaws are different, the evaluation charts are distinctively different in style. One section of this handbook deals with surface flaws, and another section concentrates on the evaluation of embedded flaws.

The flaw evaluation charts were designed based on the Section XI code criteria of acceptance for continued service without repair. Through use of the charts, a flaw can be evaluated by code criteria instantaneously, and no follow-up hand calculation is required. Most important of all, no fracture mechanics knowledge is needed by the user of the handbook charts.

The use of the term "flaw" in this document should be taken to be synonymous with the term "indication" as used in Section XI of the ASME Code.

2310s/0359s/071987:10 1-1

It is important to note that indications which are large enough that they exceed the standards limits, and must be evaluated by fracture mechanics, will also require additional inservice inspection in the future, as discussed in Section XI, paragraph IWC-2420[1]. Note that subsection IWC applies speci fically to the upper shell to cone weld, but it is not yet complete, and the user is often referred to subsection IWB. This is presently the case for subsection IWC-3600, which refers the user to IWB-3600.

1.1 CODE ACCEPTANCE CRITERIA There are two alternative sets of flaw acceptance criteria for continued ser vice without repair in paragraph IWB-3600 of ASME Code Section XI [1. Namely,

1. Acceptance Criteria Based on Flaw Size (IWB-3611)

2. Acceptance Criteria Based on Stress Intensity Factor (IWB-3612)

The choice of criteria is at the convenience of the user, per IWB-3610. Both criteria are comparable in accuracy for thick sections, and the acceptance criteria (2) have been assessed by past experience to be generally less restrictive for thin sections, and for outside surface flaws in many cases.

In all cases, the most beneficial criteria has been used, generally criteria (2). Although the steam generator wall thickness in the region of concern is slightly less than 4 inches, both sets of criteria from IWB 3600 may be applied.

1.1.1 CRITERIA BASED ON FLAW SIZE The code acceptance criteria stated in IWB-3611 of Section XI are:

a < .1 a For normal conditions (upset & test conditions inclusive) and af < .5 a. For faulted conditions (emergency condition inclusive) where a - The maximum size to which the detected flaw is calculated to grow in a specified time period, which can be the next scheduled inspection of the component, or until the end of vessel design lifetime.

2330s/0359s/1 10487: 10 1-2

a = The minimum critical flaw size under normal operating conditions (upset and test conditions inclusive)

a. The minimum critical flaw size for initiation of nonarresting growth under postulated faulted conditions. (emergency conditions inclusive)

To determine whether a flaw is acceptable for continued service without repair, both criteria must be met simultaneously. However, both criteria have been considered in advance before the charts were constructed. Only the most restrictive results were used in the charts.

1.1.2 CRITERIA BASED ON STRESS INTENSITY FACTOR As mentioned in the preceeding paragraphs, the criteria used for the construction of the charts in this handbook are from the least restrictive of IWB-3611 or IWB-3612 of Section XI. The criteria in IWB-3612 are based on safety margins between the applied stress intensity factor and the fracture toughness of the material.

The term stress intensity factor (KI) is defined as the driving force on a crack. It is a function of the size of the crack and the applied stresses, as well as the overall geometry of the structure. In contrast, the fracture toughness (KIa, KIC) is a measure of the resistance of the material to propagation of a crack. It is a material property, and varies as a function of temperature.

The criteria are stated in IWB-3612:

KI K1 < For normal conditions (upset & test conditions inclusive)

K K1 < - For faulted conditions (emergency conditions inclusive) 4 7 2330s/0359s/080 8 :10 1-3

where K = The maximum applied stress intensity factor for the flaw size a to which a detected flaw will grow, for a specified time period, which must equal or exceed the time until the next inspection.

Kia = Fracture toughness based on crack arrest for the corresponding crack tip temperature.

KIc Fracture toughness based on fracture initiation for the corresponding crack tip temperature.

To determine whether a flaw is acceptable for continued service without repair, both criteria for normal and faulted conditions must be met simultaneously. However, both criteria have been considered in advance before the charts were constructed. Only the most restrictive results (for either normal or faulted conditions) were used in the charts.

1.1.3 PRIMARY STRESS LIMITS In addition to satisfying the fracture criteria, it is required that the primary stress limits of Section III, paragraph NB 3000 be satisfied. A local area reduction of the pressure retaining membrane must be used, equal to the area of the indication, and the stresses increased to reflect the smaller cross section. All the flaw acceptance tables provided in this handbook have included this consideration, as demonstrated herein. The allowable flaw depth "a" determined using this criterion is 1.20 in. for a surface flaw in the upper shell to cone weld region, and for an embedded flaw the allowable depth "2a" is 2.6 inches. Thus the fracture mechanics criteria are governing.

1.2 GEOMETRY The geometry of the upper shell to cone weld region of the Kewaunee Unit 1 steam generators is shown in Figure 1-1. The dimensions shown are the minimum values from the design drawings. For purposes of heat transfer, the outside surfaces have been assumed to be insulated. The notation used for both surface and embedded flaws in this work is illustrated in Figure 1-2.

7 2330s/0359s/07198 :10' 1-4

FIGURE 1-1 Geometry of Upper Shell to Cone Intersection for Kewaunee Unit 1 31.20 Note: All dimensions in inches.

4 87 2330s/0359s/0 09 :10 1- 5

Figure 1-2 Typical Notations of Surface and Embedded Flaw Indications Wall Thickness t Wall Thickness I.

S~

TYPICAL SURFACE FLAW INDICATION TYPICAL EMBEDED FLAW INDICATION 2330s/0359s/071987:10

SECTION 2 LOAD CONDITIONS, FRACTURE ANALYSIS METHODS AND MATERIAL PROPERTIES 2.1 TRANSIENTS FOR THE STEAM GENERATOR The design transients for the Kewaunee Unit 1 steam generators are listed in Table 2-1. Both the minimum critical flaw sizes, such as a under normal operating conditions, or ai under faulted conditions for criteria (1) of IWB-3611, and the stress intensity factors, K1 , for criteria (2) of IWB-3612, are a function of the stresses at the cross-section where the flaw of interest is located, and the material properties. Therefore, the first appropriate step for the evaluation of a flaw indication is to determine the limiting load conditions for the location of interest.

For the region of interest, the upper shell to cone weld, the full range of design transients was considered. Transients such as pressure tests, including both hydrostatic and leakage tests, can be controlled by setting the test temperature. Therefore, in determining the governing normal condition only the operational transients were considered, and a separate determination was made as to any required changes in the pressure test temperatures, to ensure that they would not be limiting. A discussion of this subject is is the provided in Section 6.2. On this basis, the governing normal condition heatup condition, while the governing emergency and faulted condition is the feedwater-line break. All the transients were considered in calculation of fatigue crack growth, as discussed in Section 3.

2.2 STRESS INTENSITY FACTOR CALCULATIONS One of the key elements of the critical flaw size calculations is the determination of the driving force or stress intensity factor (KI). This was done using expressions available from the literature. In all cases the stress intensity factor for the critical flaw size calculations utilized a representation of the actual stress profile rather than a linearization. This was necessary to provide the most accurate determination possible of the critical flaw size, and is particularly important for consideration of emergency and faulted conditions, where the stress profile is generally 7

2330s/0359s/0 1987:10 2-1

nonlinear and often very steep. The stress profile was represented by a cubic polynomial:

2 3 a(x) = AO + A1 + 2t + 3 where x is the coordinate distance into the wall t = wall thickness a = stress perpendicular to the plane of the crack In construction of the surface flaw charts (Section 4) three flaw shapes were used, continuous (a/k = 0.0) semielliptical, with length six times the depth in Section (a/z = 0.167) and semi circular (a/z = 0.5). As will be seen 4, the charts cover the full range of shapes between these values.

For the surface flaw with length six times its depth (a/z = 0.167), the stress intensity factor expression of McGowan and Raymund [2] was used.

The stress intensity factor KI (o) can be calculated anywhere along the crack front, where o is the angular position, as defined in Figure 1-2. The point of maximum crack depth is represented by 0 = 0. The following expression is used for calculating K1 (0):

0.5 2 1/4 2a K 4)=(co 4 +ac si~ 4) (A0HO 2 at A1 H1 c

2 3

+ a 2A H+4 a3AH 7 at A2 H2 3- aA 3 H3 )

t The magnification factors HO(o), H1 (o), H2 (o) and H3 (o) were obtained by the procedure outlined in Reference [2].

The stress intensity factor calculation for a semi-circular surface flaw, and (a/e = 0.5) was carried out using the expressions developed by Raju Newman [3]. Their expression utilizes the same cubic representation of the stress profile and gives precisely the same result as the expression of

• McGowan and Raymund for the flaw with a/ = 0.167, and the form of the equation is similar to that of McGowan and Raymund above.

7 2330s/0359s/0 1987:10 2-2

The stress intensity factor expression used for a continuous surface flaw was that developed by Buchalet and Bamford [4]. Again the stress profile is represented as a cubic polynomial, as shown above, and these coefficients as well as the magnification factors are combined in the expression for KI below:

0 .5 2 4

=[a 2a 2 KK1 = [Ta]0. [ 0 F1

[AO F1 I a Al1 F22 + 2- A2 F3 ++

A2 F3 A3 a3 A F4 3 F4]

where Fl, F2 , F3 , F4 are magnification factors, available in [4].

The embedded flaw charts were constructed for a wide range of flaw sizes and shapes. The stress intensity factor calculation for embedded flaws was taken from work by Shah and Kobayashi [5] which is applicable to an embedded flaw in an infinite medium, subjected to an arbitrary stress profile. This expression has been shown to be applicable to embedded flaws in a pressure vessel in a recent paper by Lee and Bamford [6].

2.3 FRACTURE TOUGHNESS The other key element in the determination of critical flaw sizes is the fracture toughness of the material. The fracture toughness has been taken directly from the reference curves of Appendix A,Section XI. In the transition temperature region, these curves can be represented by the following equations:

0 KIc = 33.2 + 2.806 exp. [0.02 (T-RTNDT + 100 F)]

0 KIa = 26.8 + 1.233 exp. [0.0145 (T-RTNDT + 160 F)]

where KIc and KIa are in ksiV in.

The upper shelf temperature regime requires utilization of a shelf toughness which is not specified in the ASME Code. A value of 200 ksi Vin has been used here. This value is consistent with general practice in such evaluations, as shown for example in reference [7], which provides the background and technical basis of Appendix A of Section XI.

7 2330s/0359s/0 1987:10 2 -3

The fracture toughness of steam generator malerials has been examined in recent years relative to the reference toughness curves of the ASME code.

Dynamic fracture toughness tests were conducted on base metal, weldments, and heat-affected zones, and were all found to be bounded by the ASME Kia curve. Behavior was found to be very similar to that of the reactor vessel steels and weldments for which the Kia curve was developed. Thus, even of though the minimum specified yield strength of these materials is in excess the 50 ksi value specified for the ASME reference Kla curve, these results show that these materials should also be covered. Further discussion and details are found in References 8-11.

The other key element in the determination of the fracture toughness is the value of RTNDT, which is a parameter determined from Charpy V-notch and drop-weight tests.

To allow determination of RTNDT for the upper shell and cone materials, a compilation was made of the properties listed on the original material test certificates. The materials used in the steam generators were tested after a post-weld heat treatment cycle of 1050-1150'F for approximately 7 hours8.101852e-5 days <br />0.00194 hours <br />1.157407e-5 weeks <br />2.6635e-6 months <br />, as shown in Table 2-2. The Charpy impact properties of these materials are listed in Table 2-3.

The U.S. Nuclear Regulatory Commission has established guidelines for estimating the value of RTNDT from Charpy properties in their Standard Review Plan [12]. Review of Table 2-3 shows that in general the materials in the shell and cone region have excellent Charpy properties, and therefore the 0

value of RTNDT is equal to the test temperature, which is 10 F for all the materials. This value has been used in the development of the flaw evaluation charts.

Once the value of RTNDT is established, the reference toughness curves of the ASME Code discussed above may be used directly, since the materials are SA533 grade A class 1 which has a minimum specified yield strength of 65 ksi.

7 2330r/0359s/07198 :10 2- 4

2.4 CRITICAL FLAW SIZE DETERMINATION The applied stress intensity factor (KI) and the material fracture toughness values (KIa and KIc) were used to determine the allowable flaw size values used to construct the handbook charts. For normal, upset and test conditions, the critical flaw size ac is determined as the depth at which the applied stress intensity factor K exceeds the arrest fracture toughness KIa*

For emergency and faulted conditions the minimum flaw size for crack initiation is obtained from the first intersection of the applied stress intensity factor (KI) curve with the static fracture toughness (Kic) curve.

2330s;/0359s/071987:0. 2 2-5

TABLE 2-1 TRANSIENT GROUPING FOR FATIGUE CRACK GROWTH ANALYSIS Transient Description Cycles Total Cycles Group In Group 1 Heatup and Cooldown* 200 210 Turbine Roll Test 10 2 Plant Loading 15% to 100% 18300 18300 and Plant Unloading 100% to 15%

3 Large Step Load Decrease* 200 4200 Small Step Load Increase 2000 Small Step Load Decrease 2000 4 Hot Standby Operation* 18300 18300 (includes feedwater cycling) 5 Loss of Load 80 80 6 Loss of Power 40 40 7 Loss of Flow 80 80 8 Reactor Trip 400 400 9 Secondary Side Pipe Break 1 1 10 Secondary Hydrostatic Test 5 5 11 OBE 50 50 12 RCS Pipe Break 1 13 Primary Hydrostatic 5 Notes

• Umbrella Transient

• • These transients do not affect this region 7320s/0359s/071987:10 2- 6

TABLE 2-2 POSTWELD HEAT TREATMENT OF UPPER SHELL - CONE WELDS Steam Generator A Heatup to 1120*F 11.5 hours5.787037e-5 days <br />0.00139 hours <br />8.267196e-6 weeks <br />1.9025e-6 months <br /> Soak at 1050-1150aF 7 hours8.101852e-5 days <br />0.00194 hours <br />1.157407e-5 weeks <br />2.6635e-6 months <br /> Cooldown in air 6 hours6.944444e-5 days <br />0.00167 hours <br />9.920635e-6 weeks <br />2.283e-6 months <br /> Steam Generator B Soak at 1050 - 1150 7.5 hours5.787037e-5 days <br />0.00139 hours <br />8.267196e-6 weeks <br />1.9025e-6 months <br /> Cooldown in air 6 hours6.944444e-5 days <br />0.00167 hours <br />9.920635e-6 weeks <br />2.283e-6 months <br /> 23.3C/0359s/071987:10 2-7

TABLE 2-3 MATERIAL PROPERTIES OF UPPER SHELL-CONE REGION KEWAUNNE NUCLEAR PLANT Charpy Values (10-F) Lateral Expansion Location Material Type (ft-lb) (inches) RTNDT Cone materials, SG/A heat C5798-1 SA 533-65 Gr A Cl 67,63,72 0.067, 0.060, 0.068 10aF heat C5798-5 SA 533-65 Gr A C1 110,132,111 0.088, 0.092, 0.076 10aF heat 5816-4 SA 533-65 Gr A C1 110,82,110 0.082, 0.084, 0.087 10OF Upper shell materials, SG/A heat 75E553 SA 533-68 Gr A C1 75,65,51 0.050, 0.044, 0.040 100 F heat 6589-4 SA 533-67 Gr A C1 44,95,71 0.036, 0.056, 0.040 10aF Cone materials, SG/B heat 5798-1 SA 533-65 Gr A C1 67,63,72 0.067, 0.060, 0.068 10aF heat 5798-3 SA 533-65 Gr A C1 96,80,114 0.063, 0.078, 0.087 10aF heat 5816-4 SA 533-65 Gr A C1 110,82,110 0.082, 0.084, 0.087 10aF Upper shell materials SG/B heat 216881 SA 533-68 Gr A C1 79,70,95 0.053, 0.048, 0.067 10aF heat 796419 SA 533-66 Gr A C1 95,70,86 0.065, 0.057, 0.064 10aF Weld - Shielded Metal Arc top location 104,105,102 Not 10aF 1/4 T location 95,45,103 Available 10aF 2330s/0359s/1 10487:10 2-8

SECTION 3 FATIGUE CRACK GROWTH In applying code acceptance criteria as introduced in Section 1 of this report, the final flaw size af used in criteria (1) is defined as the flaw size to which the detected flaw is calculated to grow at the end of the specified service period. In this handbook, ten-, twenty-, and thirty-year service periods are assumed.

These crack growth calculations have been carried out for the upper shell to cone weld of the Kewaunee Unit 1 steam generators for which evaluation charts have been constructed. This section will examine the calculations, and provide the methodology used as well as the assumptions.

The crack growth calculations reported here are rather extensive, because a range of flaw shapes have been considered, to encompass the range of flaw shapes which could be encountered in service.

3.1 ANALYSIS METHODOLOGY The fatigue crack growth analysis procedure involves postulating an initial flaw at a specific region and predicting the growth of that flaw due to an imposed series of loading transients. The input required for a fatigue crack growth analysis is basically the information necessary to calculate the parameter AK1 which depends on crack and structure geometry and the range of applied stresses in the area where the crack exists. Once AKI is calculated, the growth due to that particular stress cycle can be calculated by equations given in Section 3.3 and Figure 3-1. This increment of growth is then added to the original crack size, and the analysis proceeds to the next transient. The procedure is continued in this manner until all the transients known to occur in the period of evaluation have been analyzed.

The transients considered in the analysis are all the design transients con tained in the Final Safety Analysis Report and the steam generator equipment specification, as shown in Section 2, Table 2-1. These transients are spread equally over the design lifetime of the vessel, with the exception that the preoperational tests are considered first. Faulted conditions are not 2330s/0359s/080487: 10 3-1

considered in the crack growth analysis because their frequency of occurrence is too low to affect fatigue crack growth.

Crack growth calculations were carried out for a range of flaw depths, and three basic types. The first type was a surface flaw with length equal to six times its depth (a/ = 0.1667), and whose analysis was previously reported.

The second was a continuous surface flaw (a/R = 0.0), which represents a worst case for surface flaws, and the third was an embedded flaw, with length maintain equal to five times its width. For all cases the flaw was assumed to a constant shape as it grew. Calculations for other flaw shapes were unnecessary because the selected types conservatively model the crack growth of the other flaws of interest for construction of the charts.

3.2 STRESS INTENSITY FACTOR EXPRESSIONS Stress intensity factors were calculated from methods available in the literature for each of the flaw types analyzed. The surface flaw with aspect ratio 6:1 was analyzed using an expression developed by McGowan and Raymund

[2] where the stress intensity factor KI is calculated from the actual stress profile through the wall at the location of interest.

The maximum and minimum stress profiles corresponding to each transient are represented by a third order polynomial, such that:

2 3 0(X) = A + Al + A x + A x t t The stress intensity factor KI (o) can be calculated anywhere along the crack front. The point of maximum crack depth is represented by o = 0. The following expression is used for calculating K (W), where o is the angular location defined in Figure 1-1.

K (O) = [ a 0.5 (Cos2 a 2sin2 1/4 (A OH 0+ 2tA 1H 1

+ 1 a 2 A H + 4 a3 A H3 )

t7 2 2 T 3 3 3 2330ls/0359s/071987:10 3- 2

The magnification factors H0 (0), H1 (o), H2 (0) and H3 ( ) are obtained by the procedure outlined in reference [2].

The stress intensity factor for a continuous surface flaw was calculated using an expression for an edge cracked plate [13]. The stress distribution is linearized through the wall thickness to determine membrane and bending stress and the applied K is calculated from:

K =I am Ym /a + oB YBB a The magnification factors Y and YB are taken from [13] and a is the crack depth.

For embedded flaws, the stress intensity factor expression of Shah and Kobayashi [5] was used, as discussed earlier in Section 2.2. The flaw shape was set with length equal to five times the width (a/z = 0.10), and the eccentricity was varied, as shown in the Table 3-2. This flaw shape was chosen to provide a worst case calculation of stress intensity factor for embedded flaws. The calculated crack growth was very small for this case, so no other shapes were considered necessary to analyze.

3.3 CRACK GROWTH RATE REFERENCE CURVES The crack growth rate curves -used in the analyses .were taken directly from Figure A4300-1 of Appendix A of Section XI of the ASME Code. Water environment curves were used for all inside surface flaws, and the air environment curve was used for embedded flaws and outside surface flaws.

The materials used for the pressure boundary of steam generators are basically higher strength versions of the reactor vessel steels, SA508 Class 2 and 3 and SA533 Gr B CI, and early designs had exactly the same materials as the reactor vessel.

A large number of specimens of steam generator materials, SA508 C 2a, SA553 Gr. A Cl. 2, SA508 Cl. 3a, and SA533 Gr. B C2 materials and two associated submerged arc weldments were tested at Westinghouse. The environments used 233;s/3 9s/o71987:13 3-33

were low and high temperature air, PWR primary water, and secondary side steam. These environments cover all the possible environments for both the primary and secondary side of the steam generator, but do not include any contaminants which could be present in the secondary side environment. Load ratios of 0.2 and 0.7 were employed for the air environment, and values of 0.2 and 0.5 were used in the PWR and steam environments.

Results showed that the reference crack growth rate curves for ferritic steel contained in Section XI were also applicable to these steels. The PWR environment was found to produce the highest growth rates, but the data were well below the ASME reference curves. The data obtained in the steam environment showed crack growth rates equal to or below the rates obtained in the PWR environment under the same conditions. These results are discussed in references 14-16. Therefore the ASME Code reference curves are applicable.

For water environments the reference crack growth curves are shown in Fig.

3-1, and growth rate is a function of both the applied stress intensity factor range, and the R ratio (Kmin/Kmax) for the transient.

For R<0.25 (AK <19 ksiv/in) = (1.02 x 10- 6) AK 5 (AK >19 ksiv/in) = (1.01 x 10-1) AK 1.95 where = Crack Growth rate, micro-inches/cycle.

For R>0.65 (AK <12 ksiv/in) = (1.20 x 10-5) AK 5.95 (AK >12 ksiV in) = (2.52 x 10-1) AK 1.95 For R ratio between these two extremes, interpolation is recommended.

233us/0359s/071987:10 3-4

The crack growth rate reference curve for air environments is a single curve, with growth rate being only a function of applied AK. This reference curve is also shown in Figure 3-1.

da = (0.0267 x 10 ) AK3.726 where, da Crack growth rate, micro-inches/cycle AK = stress intensity factor range, ksil in

= (KImax - K Imin) 3.4 FATIGUE CRACK GROWTH RESULTS The fatigue crack growth results upon which handbook charts were developed are summarized in Tables 3-1 and 3-2, and shown graphically in Figure 3-2.

7 2330s/0359s/0 1987:10 3-5

Table 3-1 Fatigue Crack Growth Results - Kewaunee Unit 1 Steam Generator Upper Shell to Cone Weld Region - Surface Flaws Continuous Flaw (a/z = 0)

INITIAL CRACK DEPTH AFTER YEAR DEPTH 10 20 30 40 0.500 0.71154 0.99030 1.35720 1.88252 0.600 0.84452 1.16395 1.59752 2.26478 0.700 0.97438 1.33694 1.84992 2.70801 0.800 1.10453 1.51543 2.12762 3.23917 a/. = 0.1667 INITIAL CRACK DEPTH AFTER YEAR DEPTH 10 20 30 40 0.700 0.79232 0.90274 1.03399 1.19217 0.800 0.91140 1.04488 1.20470 1.39380 0.900 1.03072 1.18819 1.37449 1.57976 1.000 1.19063 1.33190 1.53259 1.75105 1.500 1.71524 1.94732 2.19663 2.46658 2330s/0359s/071987:10 3-6

Table 3-2 Fatigue Crack Growth Results - Kewaunee Unit 1 Steam Generator Upper Shell to Cone Weld Region - Embedded Flaws 6 = T/16 INITIAL CRACK DEPTH AFTER YEAR DEPTH 10 20 30 40 0.120 0.12015 0.12031 0.12046 0.12061 0.150 0.15024 0.15049 0.15073 0.15097 0.160 0.16027 0.16056 0.16083 0.16112 0.165 0.16529 0.16559 0.16589 0.16619 6 = 3T/32 INITIAL CRACK DEPTH AFTER YEAR DEPTH 10 20 30 40 0.150 0.15017 0.15035 0.15052 0.15070 0.200 0.20032 0.20064 0.20096 0.20129 0.240- 0.24047 0.24096 0.24144 0.24194 6 = T/8 INITIAL CRACK DEPTH AFTER YEAR DEPTH 10 20 30 40 0.280 0.28048 0.28098 0.28147 0.28198 0.300 0.30057 0.30116 0.30173 0.30232 0.320 0.32066 0.32135 0.32202 0.32271 0.330 0.33071 0.33145 0.33217 0.33292 6 = 3T/16 INITIAL CRACK DEPTH AFTER YEAR DEPTH 10 20 30 40 0.400 0.40059 0.40121 0.40180 0.40243 0.440 0.44076 0.44154 0.44231 0.44310 0.450 0.45080 0.45164 0.45245 0.45329 0.460 0.46085 0.46174 0.46259 0.46349 6 = T/4 INITIAL CRACK DEPTH AFTER YEAR DEPTH 10 20 30 40 0.400 0.40031 0.40063 0.40095 0.40127 0.440 0.44040 0.44081 0.44121 0.44162 0.450 0.45042 0.45086 0.45128 0.45172 0.460 0.46044 0.46091 0.46136 0.46182 7

2330s/0359s/07198 :10 3-7

Fig. A-4300-1 SECION XI - DIVISION 1 1980 EDITION 1000 700 500 200 S

U 100 U

S U 70 C

0 U

50 b

S b

S 0 20 U

S 0

10 7

• Linear interpolation is recommended 5

to account for R ratio dependence of water environment curves, for 0.25 R < 0.65 for steep slope:

5 95 da = (1.02 X 106) 01 6K .

dN 1 2 5 7 10 20 50 70 100 Stress Intensity Factor Range (AK 1 ksi .Ti.)

Figure 3-1 Reference Fatigue Crack Growt.h Curves for Carbon and Low Alloy Ferritic Steels 2330s/040787: 10 3-8

-- 7 Continuous Flaw, a/l = 0.0 a/1 = 1.167 Embedded Flaw

=T/8 a/1 =.0.10 (Initial Flaw, in.)

AO Figure 3-2 Fatigue Crack Growth Results - Upper Shell Cone Weld Section Circumferential Flaws for 10 Year Period 2:330s/040787: 10 3-9

SECTION 4 SURFACE FLAW EVALUATION 4.1 SCOPE OF EVALUATION The surface flaw evaluation covers the upper shell to cone weld region. This section describes the development of the inside surface flaw charts for that region.

4.2 CODE CRITERIA The acceptance criteria for flaws have been readily presented in Section 1.

For convenience they are repeated as follows:

af < 0.1 a c For normal conditions (upset & test conditions inclusive) and af < 0.5 a. For faulted conditions (emergency condition inclusive) where af - The maximum size to which the detected flaw is calculated to grow for a specified period, which must can be the next scheduled inspection of the component or until the end of vessel design lifetime.

a = The minimum critical flaw size under normal operating conditions (upset and test conditions inclusive)

= The minimum critical flaw size for initiation of nonarresting growth under postulated faulted conditions. (emergency conditions inclusive) 2330s/0359s/080487:10 4-1

Alternatively, criteria based on applied stress intensity factors may be used:

K KI < 7 10 For normal conditions (upset & test conditions inclusive)

K K < For faulted conditions (emergency conditions inclusive) where K = The maximum applied stress intensity factor for the flaw size a to which a detected flaw will grow, for a specified period, which must be at least until the next inspection.

KIa = Fracture toughness based on crack arrest for the corresponding crack tip temperature.

KIc Fracture toughness based on fracture initiation for the corresponding crack tip temperature.

The larger flaw size determined by these two criteria is used to develop the flaw charts.

4.3 BASIC DATA In view of the criteria, it is noticed that three groups of basic data are required for the construction of charts for surface flaw evaluation. Namely, af, driving force (KI), and fracture toughness (Kia and KIC).

The preparation of these three groups of basic data will be discussed in the following paragraphs. They are the key elements of the allowable flaw size, and fatigue crack growth calculations upon which the evaluation charts are based. A schematic diagram of the evaluation procedure is shown in Figure 4-1. KIc and Ka are the initiation and arrest fracture toughnesses (respectively) of the vessel material at which the flaw is located. They can be calculated by formulas:

2330s/0359s/071987:10 4-2

K = 33.2 + 2.806 exp. [0.0 2 (T-RTNDT + 1000 F)] (1) and 14 5 0 (T-RTNDT + 160 F)] (2)

KIa = 26.8 + 1.233 exp. [0.0 Notice that both KIc and KIa are a function of crack tip temperature T, and the material property of RTNDT at the tip of the flaw as discussed earlier, in Section 2.3. The upper shelf fracture toughness of the vessel steel is assumed to be 200 ksilin, as discussed in Section 2.

The driving force, K, used in the determination of the flaw evaluation charts is the maximum stress intensity factor of the surface flaw under evaluation. The methods used for determining the stress intensity factors for surface flaws have been discussed in Section 2. It is important to note that the flaw size used for the calculation of K is not the flaw size detected by inservice inspection. Instead, it is the calculated flaw size which is projected to grow from the flaw size detected by inservice inspection. That means that the surface flaw size used for the calculation of KI had to be determined by using fatigue crack growth results. This is equivalent to working backward in the chart of Figure 4-1 to determine the largest allowable flaw size.

As defined in IWB-3611 of Section XI, af is the maximum size resulting from growth during a specific time period, which can be the next scheduled inspection of the component, or until the end of vessel design lifetime.

Therefore, the final depth, af after a specific service period of time must be used as the basis for evaluation. The charts have been constructed to allow the initial (measured) indication size to be used directly. Charts have been constructed for operational periods of 10, 20, and 30 years from the time of detection.

The final flaw size af has been calculated by fatigue crack growth analysis, which has been performed covering the range of postulated flaw sizes, and flaw shapes and locations within the wall needed for the construction of surface flaw evaluation charts in this handbook. All crack growth results have been summarized in Table 3-1, and a sample plotted in Figure 3-2.

7 2330s/0359s/08048 : 10 4-3

Notice that all the finite surface flaws and embedded flaws analyzed are semi-elliptical in shape. Crack growth analyses for finite surface flaws with aspect ratio (a/e) greater than 0.167 have utilized the results of 0.167, and for any flaw with aspect ratio less than 0.167, the results of the continuous flaw are used. This is conservative in both cases. It is noted that only the crack growth analysis for circumferential flaws was performed, because of the orientation of the indications found in the spring inspection of 1987, and the orientation of the upper shell-to-cone weld. Charts were not prepared for longitudinal flaws.

4.4 TYPICAL SURFACE FLAW EVALUATION CHART The two basic dimensionless parameters, which can fully address the characteristics of a surface flaw are used for the evaluation chart construction. Namely, O Flaw Shape Parameter a/z o Flaw Depth Parameter a/t

where, t - wall thickness, in.

a - flaw depth, in.

- flaw length, in.

Now, consider the chart for the governing transient. Section 2.1 indicated that the most limiting normal condition expected to occur during the remaining plant life is the heatup transient. In addition, the governing emergency and faulted condition is the feedwater line break. The fracture and fatigue analyses showed that the heatup is the most governing of these transients.

Figure 4-2 shows the results for the heatup transient, and it is constructed as follows:

o The flaw shape parameter a/. was plotted as the abscissa from 0 (continuous flaw) to 0.5 (semi-circular flaw) 0 The flaw depth parameter a/t in % was plotted as the ordinate.

2330s/0359s/080487:1 0 4-4

o The lower curves are the code acceptable flaw depths tabulated in Section XI. These curves indicate the acceptance standards below which analytical evaluation is not required. Three curves are provided for the code acceptance standards, covering the versions of the ASME Code from 1980 until the present. The lowest curve is from Table IWB-3511, which was revised with the 1983 Winter Addendum resulting in the middle curve. Beginning with the 1986 edition of the ASME Code, acceptance standards for this region are provided in Table IWC 3510-1 and these have also been plotted, and are slightly more liberal.

o The upper boundary curves show the maximum acceptable flaw depth by code criteria beyond which no surface flaw is acceptable for continued service without repair. These upper bound curves have been determined by the fracture and fatigue evaluations described herein, and they are applicable for 10 years, 20 years, or 30 years as indicated.

o Any surface indication which falls between the two sets of boundary curves will be acceptable by the code, with the analytical justification provided herein. However, IWC-2420 of ASME Section XI requires future monitoring of such indications.

The inside surface flaw evaluation charts constructed for the upper shell to cone weld region of the Kewaunee Unit 1 steam generators are presented in Figure 4-2, and repeated in Section 6, where instructions are given for their use.

4.5 PROCEDURE FOR THE CONSTRUCTION OF A SURFACE FLAW EVALUATION CHART This section describes how the inside surface flaw evaluation charts were constructed for the upper shell to cone weld region.

7 2330s/0359s/08048 :10 4 -5

Step 1 Determine the critical flaw sizes from Table 4-1. These flaw sizes are used to determine allowable flaw sizes per IWB-3611.

Load Fl aw Critical Flaw Depth (in.)

Condition Orientation a/e = 0.0 a/z = 0.167 a/2 = 0.5 N/U/T* Circumferential ac = 3.70 a = 3.70 ac = 3.70 c

Circumferential a. = 2.246 a. = 3.70 ai = 3.70 E/F*

Note that in some cases here the critical flaw depth is set equal to the wall thickness. This is for the case where the stress intensity factor for of flaw postulated flaws never exceeds the fracture toughness, regardless depth.

The maximum code allowable flaw depths using the criteria of IWB-3611 are then determined, using a factor of 10 for normal upset and test conditions and a factor of 2 for emergency and faulted conditions. The results are presented below:

Load Allowable Flaw Depth (in)

Condition a/k = 0.0 a/Z = 0.167 a/k = 0.5 N/U/T 0.37 0.37 0.37 E/F 1.123 1.85 1.85 487 2330s/0359s/080 :10 4 -6

Step 2 Determine the maximum code allowable flaw depth (ac or ai), per IWB-3612:

Load Flaw Code Allowable Flaw Depth (in)

Criteria a/Z = 0.0 a/z = 0.167 a/Z = 0.5 Condition Orientation N/U/T Circumferential 1.20 2.17 2.55 Kla//102 K //2 1 E/F Circumferential 1.51 2.74 3,70 Step 3 The allowable flaw depth is then determined from Step 1 and Step 2 allowable flaw depths. The most liberal results are taken for each type of. load condi tion. Then the load condition which produces the smallest allowable is chosen, and this becomes the final allowable. For normal, upset and test conditions the allowable depths of step 2 are larger, and this is also the case for emergency and faulted conditions. Thus, from the results of step 2 we find:

a/Z = 0.0 allowable a = 1.20 in.

a/z = 0.167 a = 2.17 in.

a/z = 0.5 a = 2.55 in.

Step 4 Determine the corresponding initial flaw sizes which will grow to the above critical flaw sizes after 10, 20, and 30 years of service.

We define the above limiting critical flaw depth as a . The initial flaw size a can be found from the fatigue crack growth results of Table 3-1 and have been plotted in Figure 3-2.

• N/U/T normal, upset, and test conditions E/F emergency and faulted conditions 2330s/0359s/1 10487:10 4-7

The values of a which are applicable to 10 years of service, for example, are listed as follows:

Continuous Flaw a/. = 0.167 a/z = 0.5 af 1.20 2.17 2.55 a 0.87 1.90 2.24 This shows that the effect of fatigue crack growth in this region is very small.

Step 5 Determine a/z vs. a/t% in the upper shell to cone weld region where t =

3.7", and a = a o . For 10 years of service, the values are:

Continuous Finite Surface Finite Semicircular Flaws Flaws, a/z = 0.167 Surface Flaws a/ 0 .167 .5 a/t 0.235 0.514 0.605 Note that the allowable flaw depths here exceed 20 percent of the wall thickness, which has been set as an arbitrary limit, based on engineering judgement. The charts therefore reflect this value as an upper limit.

Step 6 The upper bound curves result from the plots of a/k vs. a/t for 10, 20, 30 years of service as shown by Figure 4-2.

7 2330s/0359s/08048 :i0 8 4-8

Step 7 Plot a/z vs. a/t data from the standards tables of Section XI as the lower curve of Figure 4-2.

The values of the acceptance standards for this region from the various editions of the ASME Code are:

Aspect IWB-3511-1 IWB-3510-1 IWC-3510-1 Ratio, (1980) (1983, W83 Add.) (1986) a/. a/t, % a/t, % a/t, %

0.00 2.0 1.9 1.9 0.05 2.1 2.0 2.0 0.10 2.3 2.2 2.2 0.15 2.6 2.5 2.5 0.20 2.9 2.8 2.8 0.25 3.2 3.3 3.3 0.30 3.7 3.8 3.8 0.35 3.7 4.4 4.4 0.40 3.7 5.0 5.0 0.45 3.7 5.1 5.1 0.50 3.7 5.2 5.2 The above six steps would complete the procedure for the construction of the surface flaw evaluation charts for 10 years, 20 years, or 30 years of operating life.

In the interest of prudence, Figure 4-2 only shows the allowable flaw depths for these inside surface flaws up to 20 percent of the section thickness.

7 2330s/0359s/08048 :10 4-9

TABLE 4-1 BASIC DATA FOR SURFACE FLAW EVALUATION AT UPPER SHELL TO CONE WELD SECTION FLAW MINIMUM CRITICAL FLAW SIZE REGION CONDITION ORIENTA

& TION CONTINUOUS FLAW ASPECT RATIO = 6:1 ASPECT RATIO = 2:1 LOCATION INCHES ( ) INCHES () INCHES (a)

C).

N/U/T LONG. ar- = -- a. x a.

Do CIRCUM. ac = 3.70 1.0 a, 3.70 1.0 ac" 3.70 1.0 CL E/F LONG. a. = 25 a. =30 at = 1.

4- 0- CIRCUM. a~=2. 25 0.61 a. = 3.70 1.0 aZ 3.70 1.0 V) ZD U LEGEND :

ac Minimum critical flaw size under normal conditions

a. Minimum critical flaw size under faulted conditions 2330s/040387: 10

0 Component System Must be Transients Repaired, Expected Replaced, or I After Flaw Discovery Retired Reject Fatigue Acceptable Flaw to Crack End-of-Life Yes for Continued Be Evaluated - Growth Flaw Size Operation Until (ad) Analysis (all Next inspection (LEFM)

I-.

I-.

Initiation and I initiation KIc, Kia Arrest Smallest Kla Curve - Smallest Curve Critical Flaw Fracture Critical Flaw Fracture Size for Mechanics - Size for (ac) Mechanics --. w Accident (ai)

Severest Analysis Normal. Upset Severest Analysis Conditions Normal- (LEFM) Test Conditions Accidenti (LEFM) Initiation Transient Condition I Without Arrest Must Arrest Within 75% of Wall Thickness Enhanced Nondestructive Examination Figure 4-1 Schematic representation of Appendix A flaw evaluation process' 2330s/040387: 10

01 10 years 20 years 30 years 20

'II . .

18 TINII I ii i l 16

? 14 1N ItI ITHIN THIS ZONE THE SURFACE LAW IS ACCEPTABLE BY CODE z1 2 NALYTICAL CRITERIA (IWB 3600)

I-.

U,10 r\)

6 CODE ALLOWABLE LIMIT 1983, Winter 83 Add. through 1986 4 Prior to 1983, Winter 83 Addenda I~~

Hl lHl.

2 . Il.li ll 1 41 il iiiIn

11. 'n NO ANALYTICAL JUSTIFICATION IS REQUIRED 0

0 0.1 0.2 0.3 0.4 0.5, FLAW SHAPE (a/)

Figure 4-2 Flaw Evaluation Chart for Circumferential Inside Surface Flaws in the Upper 2330s/040387:10 She)) to Cone Region

SECTION 5 EMBEDDED FLAW EVALUATION 5.1 SCOPE OF EVALUATION Embedded flaw evaluations were performed for the upper shell to cone weld flaw charts region. This section describes the development of the embedded for that region.

5.2 EMBEDDED VS. SURFACE FLAWS According to IWA-3300 of the ASME Code Section XI, a flaw is defined as embedded, as shown in Figure 5-1, whenever, S > 0.4 a (5-1) where S - the minimum distance from the flaw edge to the nearest vessel wall surface a - the embedded flaw depth, (defined as the semi-minor axis of the elliptical flaw.)

use of The parameter 6 has been defined in.this document to facilitate the the charts. 6 is defined as the distance from the centerline of the flaw to the surface of the vessel. Therefore, 6 = S + a. Substituting into the as a proximity limit in equation 5-1 gives a limiting definition of 6 function of a, for the proximity limit.

a 6 - S (5-2) 6 > 1.4 a (5-3) 2330s/0359s/0 7

1987:0 5 5-1

Therefore, the limit for a flaw to be considered embedded is a = 0.714 6.

A flaw lying within the embedded flaw domain is to be evaluated by the embedded flaw evaluation charts generated in this section of the handbook. On the other hand, a flaw lying beyond this domain should be evaluated as a surface flaw using the charts developed in Section 4 of the handbook instead.

The demarcation lines between the two domains are shown graphically in Figure 5-2.

In other words, for any flaw indication detected by inservice inspection, the first step of evaluation is to define to which category the flaw actually belongs, and then to choose the appropriate charts for evaluation.

5.3 CODE CRITERIA As mentioned in Section 1, the criteria used in most of the cases for embedded flaws are of IWB-3612 of Code Section XI. Namely, KI KI < ;7 0For normal conditions (upset & test conditions inclusive) (5-4)

K KI < For faulted conditions (emergency conditions inclusive) (5-5) where K= The maximum applied stress intensity factor for the flaw size af to which a detected flaw will grow, during the period of evaluation, which must be at least until the next inspection.

Kla = Fracture toughness based on crack arrest for the corresponding crack tip temperature.

KIC Fracture toughness based on fracture initiation for the corresponding crack tip temperature.

7 2330s/0359s/07198 :10 5-2

The above two criteria must both be met. In this handbook only the most limiting results have been used as the basis of the flaw evaluation charts.

5.4 BASIC DATA In view of the criteria based on stress intensity factor, three basic groups of data are needed for construction of embedded flaw evaluation charts. They are: af, driving force (KI), and fracture toughness (KIa and KIc).

Kic and Kla are the initiation and arrest fracture toughness (respectively) of the vessel material at which the flaw is located. They can be calculated by formulas:

K = 33.2 + 2.806 exp. [0.0 2 (T-RTNDT + 100 0 F)] (5-6) and 0 (5-7)

KIa 26.8 + 1.233 exp. [0.014 5 (T-RTNDT + 160 F)]

K is the maximum stress intensity factor for the embedded flaw of interest. The methods used for determining the stress intensity factors for embedded flaws have been referenced in Section 2.

Notice that both KIc and K Ia are a function of crack tip temperature T, and the material property of RTNDT at the tip of the flaw as discussed in Section 2. The upper shelf fracture toughness of the vessel steel is assumed to be 200 ksilin.

K used in the determination of the flaw evaluation charts is the maximum stress intensity factor of the embedded flaw under evaluation. It is important to note that the flaw size used for the calculation of KI is not the flaw size detected by inservice inspection. Instead, it is the calculated flaw size which is projected to grow from the flaw size detected by inservice 1

2330s/0359s/07 987:1 0 5-3

inspection. That means that the embedded flaw size used for the calculation of K had to be determined by using fatigue crack growth results, similar to the approach used for surface flaw evaluation, as illustrated in the previous section.

However, unlike the surface flaw case, the fatigue crack growth for an embedded flaw (even after 30 years of additional service life) is very small in comparison with that of a surface flaw with the same initial depth.

Consequently, in the handbook evaluations, the measured flaw size has been used for evaluation by the charts independent of the service period* because fatigue has little or no influence for embedded flaws as discussed below.

This simplifies the evaluation procedure without sacrificing the accuracy of the results. A detailed justification of this conclusion is provided in the next section.

5.5 FATIGUE CRACK GROWTH FOR EMBEDDED FLAWS The environment of an embedded flaw is considered to be inert, or air. The crack growth rate for air environment is far smaller than that of the water environment, to which the surface flaw is conservatively considered to be exposed. Consequently, the fatigue crack growth for an embedded flaw is far smaller than that of an inside surface flaw (of the same size and under the same transient conditions). Numerically, the fatigue crack growth of an embedded flaw is so low that the difference between the initial flaw depth and its final crack depth is negligible, as demonstrated in Table 3-2 for the upper shell to cone weld.

Therefore, in the construction of the evaluation charts for embedded flaws, the accuracy of the charts would not be impaired using the flaw size found by inservice inspection directly.

• This conclusion holds for the range of flaw sizes acceptable by the rules of Section XI, IWB-3600. It would not necessarily hold for very large flaws of the order of 50 percent of the vessel wall thickness.

7 2330s/0359s/07198 :10 5-4

5.6 TYPICAL EMBEDDED FLAW EVALUATION CHART The details of the procedures for the construction of an embedded flaw evaluation chart are provided in the next section.

In this section, instructions for developing a chart are provided by going through a typical chart, step by step. This would help the users to become familiar with the characteristics of each part of the chart, and make it easier to apply. This example utilizes the surface/embedded flaw demarcation criteria of the code, as discussed earlier.

Following are the highlights of auxiliary charts used to construct the embedded flaw evaluation chart for the upper shell to cone weld region.

1. The abscissa of the chart in Figures 5-3, 5-4, and 5-5 represents the flaw depth a, of the embedded flaw.

2. As defined by code, embedded flaws with a depth less than a = 0.714 6 should be considered as embedded flaws. Any embedded flaws beyond the domain of a = 0.714 6, should be evaluated by means of surface flaw charts instead.

3. A key parameter for evaluating an embedded flaw is 6, the distance between the centerline of the embedded flaw and the nearest surface of the steam generator wall.

A range of 6 between 16 t and 1t has been considered in constructing Figures 5-3, 5-4, and 5-5.

4. For each specific value of 6, such as t,8 16 t, 4t, etc., a family of curves were plotted for a range of a/. values ranging from .333 to .100.

For any specific flaw depth a at the abscissa, a corresponding value KI at the ordinate can be found in Figures 5-3 through 5-5, for any distance to the surface, 6.

7 2330s/0359s/0 1987:10 5-5

5. The range of a/z values from 0.333 to 0.10 was chosen to encompass the range of flaws which might be detected. For the upper shell to cone region, fracture results are independent of the aspect ratio, as will be discussed further below.

6. In developing this specific chart, the code acceptance limit line of KIa//10 as a function of flaw depth is shown in Figures 5-3 through 5-5.

7. The intersection of the K1 curve with the code acceptance limit line is the maximum flaw size acceptable by code for the specific curve, in accordance with the K1 < K //10 from IWB-3612.

8. In view of Figures 5-3 through 5-5, it is seen that none of the curves intersect with the code acceptance limit line. That means that, up to a distance of 6 = t (= 0.925"), all embedded flaws are acceptable by the code criteria so long as their depth is within the domain of a = 0.714 6.

9. The maximum acceptable flaw size can be found from the chart by determining the abscissa of the intersection points. Namely, for 6 = 0.25 t, Maximum Acceptable a/ Flaw Depth a*(in.)

.100 0.4625

.167 0.4625 (= a = 0.4625)

.333 0.4625

• Maximum Acceptable Flaw Depth a is set at 1t, based on engineering judgement, to limit the allowable through-wall penetration to 25 percent of the wall thickness.

8 4 87 2330s/0359s/0 0 :10 5-6

1

10. The maximum acceptable embedded flaw size for 6 = t has been depicted in Figure 5-2. This simple flaw evaluation chart, described in the following paragraph, is the type to be used for evaluation, as may be seen in Section 6.

These embedded flaw evaluation charts, constructed for the upper shell to cone weld region of the steam generators, are presented in Figure 5-2 and are repeated along with instructions in Section 6.

5.7 PROCEDURE FOR THE CONSTRUCTION OF EMBEDDED FLAW EVALUATION CHARTS This section shows how an embedded flaw evaluation chart was constructed for the upper shell to cone weld region during the governing transient which is the heatup (including feedwater cycling). The example here is for the case of RTNDT = 100F.

Step 1 Calculate KI values for embedded flaws of various size, various aspect ratios, and at various distances underneath the surface. In total, 129 cases were analyzed by closed form stress intensity factor expressions. These 129 cases are listed in Table 5-1.

Step 2 The KI results of the 129 cases were plotted in Figures 5-3 through 5-5.

Step 3 10 or KI < KIa/ 1 10 criteria as Determine the allowable flaw size, from ac /

determined by Figures 5-3 through 5-5. Similar results could be obtained for the emergency/faulted conditions, but it can be seen from the surface flaw evaluation that they will not be governing so they have not been included here.

2330s/0359s/071987:1 0 5-7

5.8 COMPARISON OF EMBEDDED FLAW CHARTS WITH ACCEPTANCE STANDARDS OF IWB-3500 The handbook charts for embedded flaws do not show the acceptance standards of Section XI, as the surface flaw charts do. Therefore, it is not clear from the charts themselves how much is gained from the analysis process over the standards tables contained in IWB-3500. Such a comparison cannot be made directly on the embedded flaw handbook charts, because the charts are purpose of applicable for a full range of sizes, shapes and locations. The this section is to provide such comparisons, and to discuss the results of those comparisons.

The handbook chart values have been compared with the acceptance standards tables in Figure 5-6. In this figure the values from Table IWB-3511-1 have been plotted as the base curve, and the limit curve for embedded flaws that the justified by analysis is shown as the other line. It can be seen to range of embedded flaw shapes and depths justifiable by analysis is related the flaw location within the wall. The deeper the indication, the more benefit is obtained from the analysis.

2330s/0359s/071987:0 5-8

SURFACE N

the maximum embedded flaw size o (in depth direction) allowable per ASME XI*

EMBEDDED So

• the corresponding minimum depth FLAW -y of an embedded flaw (less than DOMAIN which it must be considered a a a surface flaw) 0 FOR ALL EMBEDDED FLAWS:

a S a0 NOTE: If a > a , the flaw must be characte~ized as a surface I flaw, with depth

• a + a.

[a = 0.7146 for the 1980 Edition of the ASME Code and later editions]

Figure 5-1 EMBEDDED VS. SURFACE FLAW 4

2330s/0359s/0 0987: 10 5-9

SURFACE/EMBEDDED FLAW DEMARCATION LINE, BEGINNING WITH 1980 CODE 0.13 a idBEDED LW t 8  :~~9ON!itURATION~

0.12 ABOVE THIS LINE ARE NOT ALLOWABLE 0.11 0.10 0.09 0.08 U. SURFACE .....

U 0.07 0

0.06 REGION MUST BEatr CO NS IDERED SURFACEC& 44

-FA S NTI 0.05

-J T4 0.04 0.03 0.02 0.01 0

0 0.05 0.10 0.15 0.20 0.2 5 DISTANCE FROM SURFACE (A) @ We stinghouse 1987 FIGURE 5-2 EMBEDDED FLAW EVALUATION CHART FOR CIRCUMFERENTIAL INDICATIONS IN THE UPPER SHELL TO CONE 7

2330s/0359s/04098 : 10 5-10

IH [Vi~

I., I ji. 'V LI 4L4444L4L141U4L14.LLJIJWIJJIAIJJ I Lull I I I IJ11I1L~iI!1III!IlII1!I; [1 I

0 re wTJ Q)

H 3i I~i~ii~r Half Crack Length (in., ~ ~ ji11LI~

FIGURE 5-3 STRESS INTENSITY FACTOR PLOTS FOR a/ = 0.333 USED IN CONSTRUCTION OF EMBEDDED 2330s/040387:10 FLAW CHARTS

II 4 11 7111 I ~ 1 I..

.I 01,

.. I -:;T. ... ...

OT

• r-v

.c LA IT nn 47g

... Half Crack Length (in.)

FIGURE 5-4 STRESS INTENSITY FACTOR PLOTS FOR a/. 0.1667 USED IN CONSTRUCTION OF 2330s/040387: 10 FLAW EVALUATION CHARTS

0 IA FIGURE 5-5 STRESS INTENSITY FACTOR PLOTS FOR a/t = 0.10 USED IN CONSTRUCTION OF 4 7 2330s/0 038 .10 FLAW EVALUATION CHARTS

=

I C

Y= S/a = 1.0 I

6 Y= S/a = 0.8 Y= S/a = G.6 Y= S/a = 0.4 0.I 0.2. 0.3 0. 0s.

FLAW SHAPE (a/t)

FIGURE 5-6 ACCEPTANCE STANDARDS FOR EMBEDDED FLAWS, FROM TABLE IWB-3511-1 (Note that for Y < 0.4 the flaw must be assumed to be a surface flaw) 2330s/040387:10 5-14

0 TABLE 5-1 EMBEDDED FLAW CASES ANALYZED FOR THE UPPER SHELL TO CONE WELD REGION DISTANCE OF FLAW EMBEDDED FLAW DEPTH (IN.)

TO SURFACE (6 in.) a/1 = 0.1 a/1 = 0.167 a/1 = 0.333

_______________ 4 T 0.03 0.06 0.03 0.06 0.03 0.06 T/16 0.09 0.12 0.09 0.12 0.09 0.12 0.15 0.1652 0.15 0.1652 0.15 0.1652 6 = 0.2313

__________ I F 1~

0.03 0.06 0.03 0.06 0.03 0.06 3T/32 0.120 0.09 0.12 0.09 0.12 0.09 Ln 0.15 0.18 0.15 0.18 0.150 0.180 6 = 0.3469 0.2478 0.21 0.2478 0.21 0.2478 0.210 I I 0.04 0.08 0.04 0.08 0.04 0.08 T/8 0.12 0.16 0.12 0.16 0.12 0.16 0.20 0.24 0.20 0.24 0.20 0.24 6 = 0.4625 0.28 0.3304 0.28 0.3304 0.28 0.3304 I tI_

0.05 0.10 0.05 0.10 0.05 0.10 3T/1 6 0.15 0.20 0.15 0.20 0.15 0.20 0.25 0.30 0.25 0.30 0.25 0.30 6 = 0.6938 0.35 0.40 0.35 0.40 0.35 0.40 0.45 0.4955 0.45 0.4955 0.45 0.4955 0.06 0.12 0.06 0.12 0.06 0.12 0.18 0.24 0.o8 0.24 0.08 0.24 T/4 0.36 0.30 0.36 0.30 0.36 0.30 6 = 0.925 0.42 0.48 0.42 0.48 0.42 0.48 0.54 0.60 0.54 0.60 0.54 0.60 0.6607 0.6607 0.6607 2330s/040387: 10

SECTION 6 FLAW EVALUATION CHARTS-UPPER SHELL TO CONE WELD 6.1 EVALUATION PROCEDURE The evaluation procedures contained in ASME Section XI are clearly specified these in paragraph IWB-3600. Use of the evaluation charts herein follows procedures directly, but the steps are greatly simplified.

Once the indication is discovered, it must be characterized as to its location, length (e) and depth dimension (a for surface flaws, 2a for embedded flaws), including its distance from the inside surface (S) for embedded indications. This characterization is discussed in further detail in paragraph IWA-3000 of Section XI.

The following parameters must be calculated from the above dimensions to use the charts (see Figure 1-2):

a o Flaw Shape parameter, -Z a

o Flaw depth parameter, T o Surface proximity parameter (for embedded flaws only), t where t = wall thickness of region where indication is located

= length of indication a = depth of surface flaw; or half depth of embedded flaw in the width direction 6 = distance from flaw centerline to surface (for embedded flaws only) (6 = s + a) s = smallest distance from edge of embedded flaw to surface 2330s/o35s/080487:10 6-1

~

as Once the above parameters have been determined and the determination made to whether the indication is embedded or surface, then the two parameters may Its location on the be plotted directly on the appropriate evaluation chart.

chart determines its acceptability immediately.

Important Observations on the Handbook Charts Although the use of the handbook charts is conceptually straight forward, of observations experience in their development and use has led to a number which will be helpful.

Surface Flaws The handbook chart for inside surface flaws is shown in Figure 6-1. For outside surface flaws the chart is shown in Figure 6-2. The flaw indication parameters (whose calculation is described above) may be plotted directly on the chart to determine acceptability. The lower curve shown (labelled "code allowable limit") are simply the acceptance standards from IWB-3500 (or IWC-3500, for the newer code edition), which is tabulated in Section XI. If is the plotted point falls below the appropriate line, the indication If the acceptable without analytical justification having been required.

lines plotted point falls between the code allowable limit line and the labelled "upper limits of acceptance by analysis" it is acceptable by virtue of its meeting the requirements of IWC 3600, which allow acceptance by fracture analysis. (Flaws between these lines would, however, require future monitoring per IWC-2420 of Section XI.) The analysis used to develop these lines is documented in this report. There are three of these lines shown in the charts, labelled 10, 20, and 30 years. The years indicate for how long the acceptance limit applies from the date that a flaw indication is discovered, based on fatigue crack growth calculations.

As may be seen for example in Figure 6-1, the chart gives results for surface flaw shapes up to a semi-circular flaw (a/z = 0.5). For the unlikely occurrence of flaws which the value of a/z exceeds 0.5, the limits on acceptance for a/z = 0.5 should be used as required by article IWA-3300 of Section XI. The upper limits of acceptance have been set at (a maximum of) twenty percent of the wall thickness in all cases, as discussed in Section 4.

2330s/0359s/080487:1 0 6-2

Embedded flaws The evaluation chart for embedded flaws is shown in Figure 6-3. The heavy whether the diagonal line in the figure can be used directly to determine indication should be characterized as an embedded flaw or whether it is sufficiently close to the surface that it must be considered as a surface flaw (by the rules of Section XI). If the flaw parameters produce a plotted point below the heavy diagonal line, it is acceptable by analysis. If it is above the line, it must be considered a surface flaw and evaluated using the surface flaw chart in Figure 6-1 or Figure 6-2.

The standards for flaw acceptance without analysis cannot be shown in the embedded flaw charts because of their generality. Therefore, they have been plotted separately in Figure 6-4.

Detailed examples of the use of the charts for both surface and embedded flaws are presented in the following sections.

Surface Flaw Example flaw and Suppose an indication has been discovered which is an inside surface has the following characterized dimensions:

a = 0.12"

= 1.2" t = 3.7" The flaw parameters for the use of the charts are a - 0.0324 (3.24%)

t a - 0.10 Plotting these parameters on Figure 6-1 it is quickly seen that the indication is acceptable by analysis. To support operation without repair it is 2330s/0359s/08048 7

10 6-3

necessary to submit this plot along with this document to the regulatory authorities.

Embedded Flaw Example Assume that a circumferential embedded flaw of 0.24 x 5.00", located within 0.2817" from the surface, was detected. Determine whether this flaw should be considered as an embedded flaw.

2a = 0.24" S = 0.2817" 6 S + a = 0.2817 + 1/2 (0.24) = 0.4017" t = 3.7"

£ = 5.0" and, a = 1/2 x 0.24"

= 0.12"

.Using Figure 6-3:

a 0.12 = 0.0324 0.4017 0.109 t3.1 Since the plotted.point (X) is below the diagonal demarcation line, the flaw must be considered embedded. Since it is below the a/t = .125 limit line, the indication is acceptable.

6.2 Modification of Hydrostatic and Leakage Test Temperatures If an indication is discovered in the Kewaunee Unit 1 steam generators which is justified for further service without repair by the flaw evaluation charts of this report, an increase in the minimum temperature at which the hydrotest and leak tests must be conducted may be necessary to ensure the required margins of Section XI are maintained. In this section, charts are provided 2330s/0359s/080487:1 0 6-4

for determination of this temperature, which is a function of the size and location of the indications discovered. Separate treatments have been developed for embedded and surface indications.

6.2.1 Embedded Flaw Hydrostatic and Leakage Test Temperature Requirements The charts herein provide a simple method for determining the required minimum temperature for any subsequent hydrostatic or leakage tests. Once an indication has been characterized, its size and location within the wall of the vessel (6/t) determine the allowable hydrostatic or leakage test temperature. This may be done by simply plotting the indication on the appropriate chart.

This determination has been made using 'the same methodology described earlier in Section 5. As discussed in Section 2 of this report, the value of RTNDT 10'F is conservatively applicable to all the steam generators. Figure 6-5 therefore covers the steam generator vessels for the hydrostatic test tempera ture, and Figures 6-6 through 6-8 cover test temperatures for a range of leakage test pressures. These figures cover the entire range of embedded flaw sizes and shapes.

6.2.2 Surface Flaw Hydro and Leak Test Temperature Figures 6-9 through 6-12 provide charts for the determination of hydrostatic and leakage test temperature requirements in the event that surface flaws are detected and shown to be acceptable by the surface flaw evaluation charts of Section 6.

These figures provide test temperatures for a range of pressures, and it can be seen from these charts that in some cases the test temperature must be increased above the presently specified value, for flaws in a small range of sizes. The figures show that slightly more restrictive temperatures are required as the test pressure increases.

7 2330s/0359s/08048 :10 6-5

10 years 20 years 30 years UPPER LIMITS.0F 20 ACCEPTANCE BY ANALYSIS 16

.14 .. ......I pil i I WITHIN THIS ZONE THE SURFACE II FLAW IS ACCEPTABLE BY CODE LU ANALYTICAL CRITERIA (IWB 3600)

0) ~10 lilli 6 CODE ALLOWABLE LIMIT 1111111111111 l!!I 11111111111111111111111 IHIHIRL'. t auag *1983, I I l~ Iii lillliibblill!lb  !!ll!!!!lilIll Iiji

.11lii; Winter 83 Add. through 1986 I1 ir,11111.1.1 I~I i .1i iii tillI lllU !W 11r.

4 Ifi I fiW1,I till___lfiL~~1 HII . .

ill ifH

",Prior to 1983, Winter 83 Addenda 2 NO ANALYTICAL JUSTIFICATION IS REQUIRED 0,

0 0.1 0.2 0.3 0.4 0.5, FLAW SHAPE (a/f)

FIGURE 6-1 FLAW EVALUATION CHART FOR CIRCUMFERENTIAL INSIDE SURFACE FLAWS INTHE UPPER SHELL TO CONE REGION 2330s/040387: 10

0 10 years 20 years 30 years 20 --UPPER LIMITS OF P IW!~ P PM I~ PMP9 ~

litjil w~ ~ I~

HISIi.11J!

I 1Htiii'14~

Ill! !Ill 1111 11Iill ACCEPTANCE BY

!II]  !;I tI II I j; II I ANALYSIS 18 II 111i ih: I I 16 11Hil 1. ii ill0t IP 14 Oil Ul..-

m PH "iTi. III 1111711 IT]i ilWi lit HIT.!; 1!q! iiiiii 111 il ' H! WITHIN THIS ZONE THE SURFACE 12 - TT

tj j., 4 it 11.1 FLAW IS ACCEPTABLE BY CODE Ii ANALYTICAL CRITERIA (IWB 3600) 310 TrT qi Iilit i iii f 1 I.Iiiiiiii I.. IIk TI fill - :-i t

T7 ;iip '- 1! t It I U.

4.- 1ih R 6 CODE ALLOWABLE LIMIT m r,

lj il jH I rl 9

i if

...... IM 1983, Winter 83 Add. through 1986 4 :i. mo-I oI ... ... 7rr Prior to 1983, Winter 83 Addenda

!!!T AT4, 2 1! 1111!!!1 H11 NO ANALYTICAL JUSTIFICATION lill !Ih I Ll iii: Ilk I 7:1:

0.

0 im 0.1 I Of III 0.2 hi I I i, 0.3 I till 511i 0.4 Tih IN i 0.5 IS REQUIRED FLAW SHAPE (a/1)

FIGURE 6-2 FLAW EVALUATION CHART FOR CIRCUMFERENTIAL OUTSIDE SURFACE FLAW IN THE UPPER SHELL TO CONE WELD 2330s/040387: 10

SURFACE/EMBEDDED FLAW DEMARCATION LINE, BEGINNING WITH 1980 CODE 0.13 a t s MBEDDED FLAW

--*FLAWS

-n CONFitlURATION WITH 0.12 ABOVE THIS-LINE ARE NOT ALLOWABLE 0.11 0.10 FAv 0.09 0.08 U SURFACE-------- -

U 0.07 0

FLAWS IN THIS.

0.06 REGION MUST BE CONSIDERED

-SURFACE .

0.05 -FLAWS 0.04 0.03 0.02 0.01 0

0 0.05 0.10 0.15 0.20 0.25 DISTANCE FROM SURFACE (1) @ Westinghouse 1987 FIGURE 6-3 EMBEDDED FLAW EVALUATION CHART FOR CIRCUMFERENTIAL INDICATIONS IN

• THE UPPER SHELL TO CONE 4

2330s/0 0387:10 6-8

=

I 0~

LU Y= S/a = 1.0 LJ~

Y= S/a = 0.8 Y= S/a = 0.6 Y= S/a = 0.4 0.I 0.2. 0.3 0. o.S FLAW SHAPE (a/t)

FIGURE 6-4 ACCEPTANCE STANDARDS FOR EMBEDDED FLAWS, FROM TABLE IWB-3511-1 (Note that for Y < 0.4 the flaw must be assumed to be a surface flaw) 2330s/040387:10 6-9

SURFACE/EMBEDDED FLAW DEMARCATION LINE 0.13

.:1M - MEDDDFLAW- 140

-:CONFIGURATION 0.11 0.10 120 S0.09 S0.08 5 SURAC z

. 0.07 0

FLAWS IN THIS 0.06 REGION MUST BE CONSIDERED U. 0.05 FLAWS SURFACE

= 0.04 .. 70 0.03 50 0.02 ------MI 11 0.01 0 0 Westinghouse 1987 0 0.05 0.10 0.15 0.20 0.25 DISTANCE FROM SURFACE )

FIGURE 6-5 DETERMINATION OF HYDROSTATIC TEST TEMPERATURES FOR CIRCUMFERENTIAL EMBEDDED FLAWS (P = 1356 psi) 2330s/04038 7

10 6-10

SURFACE/EMBEDDED FLAW DEMARCATION UNE 140 0.13 120

-E- - R: EMBEDDEO FLAW::

-.,CONFIGURATION 0.12 17 :11:

0.11 0.10 0.09 0.08 S0.07  ::URFACE 70 FLAWS IN THIS 0.06 REGION MUST BE::

CONSIDERED 50 0.o SURFACE U. 0.05

-j 4

X 0.04 -

0.03 0.02 0.01 .....

-d .. ...

0 Westinghouse 1987 0

0 0.05 0.10 0.15 0.20 0.25 DISTANCE FROM SURFACE )

FIGURE 6-6 DETERMINATION OF LEAKAGE TEST TEMPERATURES FOR CIRCUMFERENTIAL EMBEDDED FLAWS (P = 1085 psi) 2330s/040387 10 6-11

SURFACE/EMBEDDED FLAW DEMARCATION UNE 0.13 T.:12 EMBEDDED FLAW:

- CONFIGURATION 0.12 0.11 70oF 0.10 50 0.09 S0.08 U.

. 0.07 0

FLAWS IN THIS I 0.06 REGION MUST BE 0 -CONSIDEREDT ISURFACE U. 0.05

.1 Z 0.04 FLAWL :4:. .........

0.03

..77 - -- - .....

1:7 0.02 -  ::-~ 1 0.01 0.15

. . ..... 0..20.

. ...0..2.

C Westinghouse 1987 0

0 0.05 0.10 0.15 0.20 0.25 DISTANCE FROM SURFACE (-a)

FOR CIRCUMFERENTIAL FIGURE 6-7 DETERMINATION OF LEAKAGE TEST TEMPERATURES EMBEDDED FLAWS (p = 750 psi) 6-12

SURFACE/EMBEDDED FLAW DEMARCATION UNE 0

0.13 70, 50 F

- - . : :EMBEDDED FLAW CONFIGURATION 0.12 0.11 0.10 0.09 S0.08 U. 0.07 0

-FLAWS IN THIS 2 0.06 ~REGION MUST3BE~

a CONSIDERED U. 0.05

. di&B FLAWS1 i~ll  :"lii~=T ~i .1.i y) :it s aniB s

= 0.04 0.03 0.02 0.01

) Westinghouse 1987 0

0 0.05 0.10 0.15 0.20 0.25 DISTANCE FROM SURFACE )

FIGURE 6-8 DETERMINATION OF LEAKAGE TEST TEMPERATURES FOR CIRCUMFERENTIAL EMBEDDED FLAWS (p = 500 psi) 6-13

0 U-)

0 0

"3 U)

'to 200 180 20 18 16 160 14 12 O~)

w 10 70, 120, 140 8

LL 6 50 4

2 0

0.5 FLAW SHAPE (alP)

FOR FIGURE 6-9 DETERMINATION OF HYDROSTATIC TEST TEMPERATURES CIRCUMFERENTIAL SURFACE FLAWS (p =1356 psi)

0 to 0.

20 18 16 140 14 12 70, 120 w 10 I-.

8

-j U-6 4

2 0 0.5 0.1 0.2 0.3 0.4 0

FLAW SHAPE (a/2)

FOR FIGURE 6-10 DETERMINATION OF LEAKAGE TEST TEMPERATURES CIRCUMFERENTIAL SURFACE FLAWS (p = 1085 psi)

O to 120 20 70 18 16 500 F 14 0 1o 6

4 2

00 0.2 0.3 0.4 0.5 0.1 FLAW SHAPE (a/2)

FIGURE 6-12 DETERMINATION OF LEAKAGE TEST TEMPERATURES FOR CIRCUMFERENTIAL SURFACE FLAWS (p = 500 psi)

0 Lo 0

0 (0

w 140 120 20 18 16 14 12 I-.

10 70, 50 4 8 U.

6 4

2 0 001020304 0.5 FLAW SHAPE (a/2)

FOR CIRCUMFERENTIAL FIGURE 6-11 DETERMINATION OF LEAKAGE TEST TEMPERATURES SURFACE FLAWS (p = 750 psi)

SECTION 7 REFERENCES

1. ASME Code Section XI, "Rules for Inservice Inspection of Nuclear Power Plant Components", 1980 Edition; 1983 and 1986 editions (used for updated standards tables, Section 4.5), and 1980 edition [Winter 1981 Addendum]

(for revised reference crack growth curves).

2. McGowan, J. J. and Raymund, M., "Stress Intensity Factor Solutions for Internal Longitudinal Semi-elliptic Surface Flaw in a Cylinder Under Arbitrary Loading", ASTM STP 677, 1979, pp. 365-380.

3. Newman, J. C. Jr. and Raju, I. S., "Stress Intensity Factors for Internal Surface Cracks in Cylindrical Pressure Vessels", ASME Trans., Journal of Pressure Vessel Technology, Vol. 102, 1980, pp. 342-346.

4. Buchalet, C. B. and Bamford, W. H., "Stress Intensity Factor Solutions for Continuous Surface Flaws in Reactor Pressure Vessels", in Mechanics of Crack Growth, ASTM, STP 590, 1976, pp. 385-402.

5. Shah, R. C. and Kobayashi, A. S., "Stress Intensity Factor for an Elliptical Crack Under Arbitrary Loading", Engineering Fracture Mechanics, Vol. 3, 1981, pp. 71-96.

6. Lee, Y. S. and Bamford, W. H., "Stress Intensity Factor Solutions for a Longitudinal Buried Elliptical Flaw in a Cylinder Under Arbitrary Loads",

presented at ASME Pressure Vessel and Piping Conference, Portland Oregon, June 1983. Paper 83-PVP-92.

7. Marston, T. U. et. al. "Flaw Evaluation Procedures: ASME Section XI" Electric Power Research Institute Report EPRI-NP-719-SR, August 1978.

8. Logsdon, W. A., "Dynamic Fracture Toughness of ASME SA508 C2a Base and Heat-Affected Zone Material, in Elastic-Plastic Fracture, ASTM STP 668, 1979.

2330s/0359s/080487: 10 7-1

9. Logsdon, W. A., "Dynamic Fracture Toughness of Heavy Section, Narrow Gap Gas Tungsten Arc Weldments" Engineering Fracture Mechanics, Vol. 16, No.

6, 1982.

10. Logsdon, W. A., "Dynamic Fracture Toughness and Fatigue Crack Growth Rate Properties of ASME SA508 C1.3 and SA508 C1.3a Base and Heat Affected Zone Materials" in ASTM Journal of Testing and Evaluation, Vol. 10, July 1981.

11. Logsdon, W. A., and Begley, J. A., "Dynamic Fracture Toughness of SA533 Grade A Class 2 Base Plate and Weldments" in Flaw Growth and Fracture, ASTM STP 631, 1977.

12. U.S. N.R.C Standard Review Plan, (Rev. 1), Section 5.3.2, Report NUREG 0800, July 1981.

13. Plane Strain Crack Toughness Testing of High Strength Metallic Materials, ASTM STP 410, March 1969.

14. Logsdon, W. A., Liaw, P. K., and Begley, J. A., "Fatigue Crack Growth Rate Properties of SA508 and SA533 Pressure Vessel Steels and Submerged Arc Weldments in Room and Elevated Temperature Air Environments" Engr.

Fracture Mechanics, Vol. 2, No. 3, 1985.

2330s/0359s/071987:10 7 7 -2

APPENDIX A RESULTS OF THE INSPECTION OF SPRING 1987 A-1

SUMMARY

During the Spring 1987 ultrasonic examination of the Kewaunee Unit 1 Steam Generator "B" upper shell to cone weld, nine recordable indications were noted. Two of these were detected with the 45 degree, 2.25 MHz shear wave examinations, and the remaining seven were detected with the 60 degree, 2.25 MHz shear wave examinations. The location of these indications in the weld and past experience with the same weld in other steam generators at other plants indicates that all these indications are volumetric in nature, i.e.,

small slag inclusions and/or voids. An evaluation of these indications (using 50% DAC sizing criteria) to the acceptance standards in Table IWB-3511-1 of the ASME Code Section XI, 1980 Edition results in seven indications which are unacceptable. In a similar evaluation using the acceptance standards in Table IWC-3510-1 of the ASME Code Section XI, 1986 Edition results in six unacceptable indications.

Using the fracture analysis rules of IWB-3600 and the guidelines of Appendix A, both from the ASME Code Section XI, 1980 Edition, all the indications are acceptable using 50% DAC sizing levels (2.25 and 5.0 MHz transducer data), and 20% DAC sizing levels without beam spread correction factors (5.0 MHz transducer data).

A-2 ULTRASONIC EVALUATION AND DISCUSSION Nine recordable indications were noted during the recent examinations of the Kewaunee Unit 1 Steam Generator "B" upper shell to cone weld. Summary tables of the indications are presented in Tables A-1 and A-2. Table A-1 provides the measured "2a" value, the measured "S" value, and the measured length all with respect to the normal to the inside pressure retaining surface of the component and determined using a 5.0 MHz transducer and 50% DAC sizing criteria. Table A-2 shows the same parameters using a 2.25 MHz transducer and 50% DAC sizing criteria. These values are measured using indication plots 2330s/0359s/080487: 10 A-1

rather than calculated from the raw data due to the geometry of the weld. The majority of the indications were detected from the outer diameter surface of the transition cone but are physically located in the upper shell portion of the weld. The indication parameters ("2a", "Z", and "S") therefore have been taken from the surfaces of the upper shell. The 45 degree sizing data was taken using a 2.25 MHz transducer and a 50% DAC sizing criteria. The 60 degree sizing data, with the exception of Indication C, was taken using a 5.0 MHz transducer and the same 50% DAC sizing criteria. Sizing data using a 60 degree, 2.25 MHz transducer and 50% DAC sizing criteria were also taken.

Although both 2.25 MHz and 5.0 MHz sizing data were taken, the primary sizing data used for the fracture mechanics analysis was based on that taken with the 5.0 MHz transducer. Experience has shown that 2.25 MHz testing is excellent for detection in this application, but tends to oversize when used in conjunction with the Section XI criteria.

The 2.25 MHz transducer produces a beam spread which is wider than that of a similar size 5.0 MHz transducer. This factor typically results in an unavoidable overestimate of the true size of volumetric reflectors such as slag, which is believed to be present in this case. An example will illustrate this fact. Consider an indication which is being sized with a 2.25 MHz, 450 shear wave transducer, as shown in Figure A-1.

As the transducer is moved along the examination surface it picks up an indication (shown by the dot), and the first step is to locate the peak response of the indication, as shown in illustration (b). For illustration purposes, assume the amplitude is 100% of the distance amplitude correction curve (DAC). The peak response of the indication is then plotted in illustration (e), at an angle of 45 degrees from the transducer location. The distance along the 45 degree line is determined from the time base of the ultrasonic test instrument, which is a function of the speed of sound in the material.

The extent of the indication is then determined by moving the transducer along the surface until the amplitude drops to 50% DAC. This point is shown in illustration (c) for one direction, and corresponds to a reduction in the 2330s/0359s/071987: 10 A-2

signal amplitude of the indication of 6 dB or one half.Section XI requires this point to also be plotted at an angle of 45 degrees [see (f)] even though it is clear from (c) that the angle is less than 45 degrees. A similar procedure is then followed to get the extent of the indication in the other direction (d) and the location is again plotted at 45 degrees [see (g)] even though in this case the angle is clearly greater than 45 degrees. The through wall dimension of the indication, "2a", is then determined from projection of a line through the peak point perpendicular to the vessel inside surface, as shown in (g). The through wall dimension then follows from projection of the end points onto the perpendicular.

An illustration of how the flaw sizing and location changes with a narrower beam is shown in Figure A-2. Here the example is exactly the same, but a 5.0 MHz transducer of similar size is used. The peak location or center of the indication is found to be identical to the previous example, as shown in (e) but the outer extent of the indication is considerably different, because the beam is narrower, and the projection of the outer 50% DAC limits of the indication is less, as shown in (f) and (g). The through wall depth is much smaller, and also the distance from the inside surface is also much greater.

This is exactly the situation which occurred with the indications in steam generator B, although the actual details were more complex.

Therefore in the case of volumetric flaws a reduction in beam spread is desired. There are a number of ways to minimize the beam speed, including use of a higher frequency transducer, a focused transducer, a larger transducer size or a combination of these. The beam spread, 8, can be shown by simple physics [All to be related to the diameter (D) of the transducer and its frequency (f) as follows:

sin 8= kX kC where K = a constant C = speed of sound in the material X = wave length 8 = beam spread angle, defined in Figures A-1 and A-2 2330s/0359s/080487: 10 A-3

Beam spread effects can also be minimized by use of beam spread correction, which is essentially a correction on the plotted extremities of the indications, but data to support the accuracy of these calculations is limited. The use of other transducers is permitted by Paragraph T-451.1 of the ASME Code Section V, Article 4 which states that "other ultrasonic techniques and nondestructive examination methods may be helpful in determining a reflector's true position, size, and orientation".

A-2.2 Experience With Other Plants The indications in steam generator B appear to be quite characteristic of experience with various welds in steam generators and pressurizers at other plants where preservice ultrasonic examination results based on 2.25 MHz, 50%

DAC sizing methods-predicted reflectors detected in weld backchip regions had dimensions in excess of those allowable values provided in Section XI of the ASME Code. Attempts were made to confirm the size, location, and orientation of these indications by complementary nondestructive examination methods, i.e.

O degree longitudinal wave examinations, and both fabrication and field radiography. No reliable responses could be observed from the shear wave indications using the straight beam examinations. In terms of the radiography, the fabrication radiographs of the areas in question were reviewed with no conclusive results. Additionally, field radiography was performed in selected areas but again no confirmation of the shear wave examination indications could-be obtained.

These inconclusive results led to physical removal of some of the suspect indications by mechanical means for complete metallurgical characterization.

The indications were found to have been caused by small slag inclusions and voids between weld passes in the weld backchip area near the inside surface.

Measurements made during the destructive analysis showed that the ultrasonic sizing using 2.25 MHz, 50% DAC sizing methods exaggerated the true size of the discontinuities in terms of length and/or through-wall dimensions. These results are presented in Table A.3, and plotted in Figure A-3. These results agree closely with the illustrations previously presented.

2330s/0359s/1 10487:10 A-4

Furthermore, this experience correlates well with investigations to date which have shown that when sizing volumetric-type reflectors by amplitude drop methods, i.e. 2.25 MHz, 50% DAC, the typical result is that the beam size rather than the reflector size is measured. For example, the lower the test frequency, the larger the beam width resulting in a larger than actual apparent flaw size (References A2-A7).

A-2.3 1987 Inspection Conclusions Since the indications found in these examinations are ultrasonically similar to those detected at other plants it was appropriate to use higher frequency transducers to obtain more realistic data concerning the through-wall dimensions of the indications. Since the 45 degree indications sized with 2.25 MHz, .50% .DAC .methods.were within the acceptance standards in Table IWC-3510-1 (ASME Section XI, 1986 Edition), no high frequency data were taken. This is shown in Table A-5.

Using the data in Tables A-1 and A-2, two sets of evaluation calculations were performed. The first evaluation compared the characteristics of the indications to the acceptance standards described in Table IWB-3511-1 of the ASME Code Section XI, 1980 Edition. This evaluation resulted in seven indications which were unacceptable (Table A-4). The second evaluation used the acceptance standards of Table IWC-3510-1 of the ASME Code Section XI, 1986 Edition as the acceptance criteria. This evaluation resulted in six unacceptable indications (Table A-5). The latter ASME Code was considered for information only because it contained acceptance standards strictly for Class 2 component welds such as the upper shell to cone weld.

To be more conservative, additional data were taken using a 20% DAC sizing criteria but without the use of beam spread correction factors. The use of this sizing criteria is specified in Nuclear Regulatory Guide 1.150 but with the use of beam spread correction factors. Of course, the size of the indication as delineated in this same regulatory guide is determined, though, by using the greater of the values obtained by the 50% DAC sizing criteria and the 20% DAC sizing criteria with beam spread correction. The 5.0 MHz, 20% DAC sizing data are summarized on Table A-6.

2330s/0359s/1 10487:10 A-5

A.3 FRACTURE ANALYSIS There are two alternative sets of acceptance criteria for continued service without repair in paragraph IWB-3600 of the ASME Code Section XI:

1. Acceptance criteria based on flaw size (IWB-3611)

2. Acceptance criteria based on stress intensity factor (IWB-3612)

The more beneficial criteria of IWB-3612 have been used for evaluating the nine indications.

To determine the allowable flaw sizes in a weld, finite element analysis methods were used.

All applicable plant transients were analysed to select the most severe stress were profiles through the thickness of the weld. The actual stress profiles then approximated by third order polynomials and used for calculating the stress intensity factor (KI) for various crack sizes and aspect ratios.

The resulting K values were compared to fracture toughness values (Kia and KIC ). Critical flaw sizes were then obtained, and allowable flaw sizes determined using the acceptance criteria discussed above.

The final step involves calculation of crack growth due to fatigue loading.

All anticipated plant transients were utilized in determining the resulting flaw size for a specified period of time. This was done for 10, 20, and 30 year intervals.

In addition to satisfying the fracture criteria, it is required that the A local primary stress limits of Section III paragraph NC-3000 be satisfied.

area reduction of pressure retaining membrane must be used, equal to the area of indication; and the stresses increased to reflect the smaller cross section.

2330s/0359s/1 10487:10 A-6

The nine indications found are all subsurface flaws as defined by IWB-3500.

As shown in Figures A-1 and A-2, all nine indications are acceptable per the fracture analysis criteria of IWB-3600. The fracture evaluation methods used for these analyses have been documented in the main body of this report.

It should be mentioned that some elevation of the hydrotest and leak test temperatures over the specified temperature will be required to ensure the margins of IWB-3600 are maintained, and these temperatures have been provided along with the complete technical details of the analysis in the main body of this report. The revised hydrotest and leak test temperatures from this inspection are provided in Figures A-6 and A-7.

A-4 REFERENCES Al. Krautkramer, J., and H. Krautkramer, Ultrasonic Testing of Materials, Springer-Verlag New York Inc., New York, 1969, page 83.

A2. Gruber, G. J., Hendrix, G. J. and Schick, W. R., "Characterization of Flaws in Piping Welds Using Satellite Pulses", MATERIALS EVALUATION, April 1984.

A3. Cook, R. V., Latimer, P. J. and McClung, R. W., "Flaw Measurement Using Ultrasonics in Thick Pressure Vessel Steel, " final report on Contract No. W-7405-eng-26, prepared by Oak Ridge National Laboratory for the U.S.

Nuclear Regulatory Commission, Aug. 1982, Oak Ridge, TN.

A4. Doctor, S.R., Becker, F. L., Heasler, P. G. and Selby, G. P.,

"Effectiveness of U.S. Inservice Inspection Techniques - A Round Robin Test," Proceedings of Specialist Meeting on Defect Detection and Sizing, Ispra, Itlay, May 3-6, 1983. Joint Research Center, Ispra (Va), Italy.

A5. Jessop, T. J., Mudge, P. J. and Harrison, J. D., "Ultrasonic Measurement of Weld Flaw Size," National Cooperative Highway Research Program Report 242, prepared for the Transportation Research Board by The Welding Institute, Dec. 1981. The Welding Institute, Cambridge, England.

4 7 2330s/0359s/080 8 :10 A-7

A6. Mudge, P. J. and Jessop, T. J., " Size Measurement and Characterization of Weld Defects by Ultrasonic Testing: Findings of a Collaborative Programme," Proceedings of NDE in Relation to Structural Integrity, Paris, France, Aug. 24-25, 1981. Applied Science Publishers, Ltd.,

London, England.

A7. Rishel, R.D., "Summary Report: Volumetric Flaw Depth Sizing,"

MT-SMART-807, September 12, 1985 (submitted to Seabrock Power Station).

2330s/0359s/071987:10 A-8

TABLE A-1

SUMMARY

OF ULTRASONIC TEST INDICATIONS FOUND IN THE KEWAUNEE UNIT 1 STEAM GENERATOR "B" WELD 2-5 (5.0 MHZ TRANSDUCER, 50% DAC SIZING) 1S11 DATA INDICATION MEASURED LENGTH "2a" (inside surface)

1. 45 degree A

2. 45 degree B

3. 60 degree A 0.37" 1.02" 0.75"

4. 60 degree B 0.35" 0.75" 0.50"

5. 60 degree ** C

6. 60 degree D 0.28" 0.69" 1.10"

7. 60 degree E 0.65" 0.75" 0.63"

8. 60 degree F 0.26" 1.93" 1.00"

9. 60 degree G 0.35" 1.79" 2.75" NOTE:

• • Using the 5.0 MHz transducer this indication only had a ultrasonic signal response of 50% DAC. According to the examination procedure no further sizing data needed to be taken. As a result the 2.25 MHz transducer data is given.

2330s/0359s/071987:10 A-9

TABLE A-2

SUMMARY

OF ULTRASONIC TEST INDICATIONS FOUND IN THE.

KEWAUNEE UNIT 1 STEAM GENERATOR "B" WELD 2-5 (2.25 MHZ TRANSDUCER, 50% DAC SIZING)

DATA INDICATION MEASURED "S" LENGTH "2a" (inside surface) 1.-4---------- ---------- --- --------------- -----

1. 45 degree A 0.61" 0.37" 0.60"

2. 45 degree B 0.43" 0.45" 0.10",

3. 60 degree A 0.69" 0.67" 1.20"

4. 60 degree B 0.81" 0.24" 1.50"

5. 60 degree C 0.65" 0.63" 0.90"

6. 60 degree D 0.55" 0.53" 1.10"

7. 60 degree E 0.46" 0.37" 1.50"

8. 60 degree F 0.61" 1.67" 3.10"

9. 60 degree G 0.41" 1.63" 1.80" 2330s/0359s/071987:10 A-10

TABLE A-3 NONDESTRUCTIVE VERSUS DESTRUCTIVE TESTING RESULTS USING 2.25 MHZ, 50% DAC SIZING PHYSICAL DISTANCE FROM ID THROUGH-WALL LENGTH SAMPLE SURFACE DEPTH UT ACTUAL UT ACTUAL UT ACTUAL CORE #1 ** ** .37" 0.09" 1.18" 1.15" (Plant 1) to 1.03 to 3.18" CORE #2 ** **

.16" 0.02" .63" 0.45" (Plant 1) to .58" to .75" CORE #1 0.00" 0.08" to 0.24" 0.01" 0.88" 0.25" to (Plant 2) 0.33"

• to 0.33" 0.28" CORE #2 0.16" 0.82" 0.53" 0.18" 0.88" 0.27" (Plant 2)

GRINDING 0.05" **

0.37" ** 1.00" (Plant 2)

GRINDING 0.00" 0.375" 0.45" 0.094" 3.5" (Plant 2)

GRINDING 0.00" 0.125" 0.51" 0.156" 3.25" (Plant 2)

GRINDING 0.02" 0.156" 0.43" 0.219" 0.75" 0.375" (Plant 2)

GRINDING 0.00" 0.24" 0.75" (Plant 2)

GRINDING 0.00" 0.219" 0.33" 0.343" 1.0" 0.438" (Plant 2)

• One UT indication was found to be four indications upon metallurgi cal evaluation. The values show the range of sizes for these four defects.

• • Dimensions not reported.

2330s/0359s/071987:0 A A-11

TABLE A-4 RESULTS OF THE ASME SECTION XI, 1980 EDITION CALCULATIONS USING THE ACCEPTANCE STANDARDS OF TABLE IWB-3511-1 DATA IND. MEASURED TYPE OF "a" "1S"1 a/t a/t

• n2a" [1] IND. ALLOW. ACT.

1. 45 deg. A 0.61" subsurf. 0.31" 0.37" 0.60" 7.2% 8.4%

2. 45 deg. B 0.43" subsurf. 0.22" 0.45" 0.10" 7.2% 5.9%

3. 60 deg. A 0.37" subsurf. 0.19" 1.02" 0.75" 4.1% 5.1%

4. 60 deg. B 0.35" subsurf. 0.18" 0.75" 0.50" 5.3% 4.9%

5. 60 deg. C 0.65" subsurf. 0.33" 0.63" 0.90" 5.4% 8.9%

6. 60 deg. D 0.28" subsurf. 0.14" 0.69" 1.10" 3.1% 3.8%

7. 60 deg. E 0.65" subsurf. 0.33" 0.75" 0.63" 7.2% 8.9%

8. 60 deg. F 0.26" subsurf. 0.13" 1.93" 1.00" 3.1% 3.5%

9. 60 deg. G 0.35" subsurf. 0.18" 1. 79" 2.75" 2.8% 4.9%

NOTES:

• The measured base metal thickness of 3.7" was used rather than the measured weld thickness of 3.9" due to the irregular nature of the weld crown.

[1] From Table A-1, except for indications 1.2 and 5 which are from Table A-2.

2330s/0359s/071987:10 A- 12

TABLE A-5 RESULTS OF THE ASME SECTION XI, 1986 EDITION CALCULATIONS USING THE ACCEPTANCE STANDARDS OF TABLE IWC-3510-1 DATA IND. MEASURED TYPE OF "a" "1S"1 "t"I a/t a/t *

"2a" [1] IND. ALLOW. ACT.

1. 45 deg. A 0.61" subsurf. 0.31" 0.37" 0.60" 8.9% 8.4%

2. 45 deg. B 0.43" subsurf. 0.22" 0.45" 0.10" 8.9% 5.9%

3. 60 deg. A 0.37" subsurf. 0.19" 1.02" 0.75" 4.4% 5.1%

4. 60 deg. B 0.35" subsurf. 0.18" 0.75" 0.50" 6.0% 4.9%

5. 60 deg. C 0.65" subsurf. 0.33" 0.63" 0.90" 6.2% 8.9%

6. 60 deg. D 0.28" subsurf. 0.14" 0.69" 1.10" 3.1% 3.8%

7. 60 deg. E 0.65" subsurf. 0.33" 0.75" 0.63" 8.9% 8.9%

8. 60 deg. F 0.26" subsurf. 0.13" 1.93" 1.00" 3.1% 3.5%

9. 60 deg. G 0.35" subsurf. 0.18" 1.79" 2.75" 2.6% 4.9%

• The measured base metal thickness of 3.7" was used rather than the measured weld thickness of 3.9" due to the irregular nature of the weld crown.

[1] From Table A-1, except for indications, 1, 2, and 5 which are from Table A-2.

2330s/0359s/071987:10 A-13

TABLE A-6

SUMMARY

OF ULTRASONIC TEST INDICATIONS FOUND IN THE KEWAUNEE UNIT 1 STEAM GENERATOR "B" WELD 2-5 (5 MHZ TRANSDUCER, 20% DAC SIZING)

DATA IND. MEASURED "S" LENGTH I.D. "2a" (inside surface)

1. 45 deg. A *

2. 45 deg. B *

3. 60 deg. A 0.52" 0.96" 1.05"

4. 60 deg. B 0.47" 0.79" 1.05"

5. 60 deg. C 0.57" 0.65" 0.90"

6. 60 deg. D 0.73" 0.51" 1.50"

7. 60 deg. E 0.65" 0.83" 0.95"

8. 60 deg. F 0.39" 1.89" ** 1.20"

9. 60 deg. G 0.35" 1.67" 3.13"

• Data not taken.

• • The reflector is nearer to the outside surface than the inside surface. "S" to the outside surface is 1.77".

2'330s/0359s/071987:10 A- 14

45 SHEAR, 50% DAC SIZING EXAMPLE OF 2.25 MBZ, EXAMPLE ASSUMPTIONS  :

100% DAC

- MAXIMUM AMPLITUDE OF RESPONSE,

- DIAMETER OF TRANSDUCER = D FIGURE A-1 SCHEMATIC EXAMPLE OF FLAW SIZING WITH 2.25 MHZ TRANSDUCER, USING 50% DAC.

2330s/0359s/071587:10 A-15

EXAMPLE OF 5.0 MBZ, 45 SHEAR, 50% DAC SIZING EXAMPLE ASSUMPTIONS

- MAXIMUM AMPLITUDE OF RESPONSE, 100% DAC

- DIAMETER OF TRANSDUCER =D FIGURE A-2 SCHEMATIC EXAMPLE OF FLAW SIZING WITH 5.0 MHZ TRANSDUCER, USING 50% DAC.

2330s/0359s/071 587:10 A-16

0 NONDESTRUCTIVE VERSUS DESTRUCTIVE TESTING RESULTS (2.25 MHz TRANSDUCER. 50% DAC SIZING) 1.1 f.'

I 1.0 u

0.9 CL z

0.8 0cc 0 0.7 0.6 F-.

0.5 O

0 Ir 0.4 I

0 0.3 I

I-_

0D UJ 0.2 CE)

Li YK0.1 0.0 l 0.0 0.1 0.2 0.3 0.4 0.5 ACTUAL THROUGH-WALL OEPTH (INCH)

FIGURE A-3 NONDESTRUCTIVE VS DESTRUCTIVE TESTING RESULTS, 2.25 MHZ TRANSDUCER 2330s/071587: 10 WITH 50% DAC SIZING

SURFACE/EMBEDDED FLAW DEMARCATION LINE .

0.13 - - -- ~ -- 7 l 8E EMBED E CONFIGURATION 4-FLAWS WITH 0.12 ABOVE THIS LINE ARE NOT ACCEPTABLE 0.11 0.10

- 0.09

-T-aws ..... 3 A-S0.08 LL 0.07 0

x 2 0.06 o

U. 0.05 ALL EMBEDDED FLAWS X 0.04 (ON THIS SIDE OF DEMARKATION LINE) 0.03 ARE ACCEPTABLE PER CRITERIA OF IWB 3600 FLAW5 IN THI5S 0.02 AS LONG AS -0.25

. REGION ARE t i ACCEPTABLE -- --- ----

0.01 0

0 0.05 0.10 0.15 0.20 0.25 DISTANCE FROM SURFACE (a)

MEASURED TYPE OF as" "la a/t

• S/t Accept DATA IND. able?

"2a" IND. ACT.

0.61" subsurf. 0.37" 0.60" 8.4% 0.182 Yes

1. 45 deg. A subsurf. 0.45" 0.10" 5.9% 0.180 Yes

2. 45 deg. B 0.43" subsurf. 1.02" 0.75" 5.1% 0.326 Yes

3. 60 deg. A 0.37" subsurf. 0.75" 0.50" 4.9$ 0.250 Yes

4. 60 deg. B 0.35" subsurf. 0.63" 0.90" 8.9$ 0.258 Yes

5. 60 deg. C 0.65" 0.28" subsurf. 0.69" 1.10" 3.8% 0.224 Yes

6. 60 deg. D 0.65a subsurf. 0.75" 0.63" 8.95 0.291 Yes

7. 60 deg. E subsurf. 1.93" 1.00" 3.55 0.557 Yes

8. 60 deg. F 0.260 0.35" subsurt. 1.79" 2.75" 4.9% 0.531 Yes

9. 60 deg. G FIGURE A-4 FRACTURE ANALYSIS RESULTS FOR INDICATIONS FOUND IN THE KEWAUNEE UNIT 1 STEAM GENERATOR "B" WELD 2-5 (50% DAC SIZING, 5.0 MHz) 7 2330s/0359s/08048 :10 A-18

SURFACE/EMBEDDED FLAW DEMARCATION LINE 0.13 EBEDDE M

- -:.E-CONFIGURATION D FLAU

-*-FLAWS WITH -!

t 0.12 ABOVE THIS LINE ARE NOT ACCEPTABLE 0.11

• LEI 0.10 6 0.09 7 1-7 0.08 5 U. 0.07 =-"SURFACE 3

0 4

0.06 FLAWS REGION IN THIS MUST BE o CONSIDERED SURFACE 8 I.- 0.05 FLAWS 9

ALL EMBEDDED FLAWS 0.04 THIS SIDE OF A-------4--i(ON A--- DEMARKATION LINE) 0.03 ARE ACCEPTABLE PER CRITERIA OF IWB 3600 0.02 FLAWS IN THIS AS LONG AS 1-a0.25 REGION ARE t ACCEPTABLE 0.01 0

0 0.05 0.10 0.15 0.20 0.25 DISTANCE FROM SURFACE (S)

SUMMARY

OF ULTRASONIC TEST INDICATIONS FOUND IN THE KEWAUNEE UNIT 1 STEAM GENERATOR "B" WELD 2-5

( 20% DAC SIZING )

DATA IND. MEASURED "St" I.D. "12a" a/t (inside surface) /t

1. 45 deg. A *

2. 45 deg. B I

3. 60 deg. A 0.52" 0.0703 0.96" 0.330 1.05"

4. 60 deg. 8 0.47" 0.0635 0.79" 0.278 1.05"

5. 60 deg. C 0.57" 0.0770 0.65" 0.253 0.90"

6. 60 deg. D 0.73" 0.0986 0.51" 0.236 1.50"

7. 60 deg. E 0.65" 0.0878 0.312 0.95"

8. 60 deg. F 0.39" 0.0530 1.89" *' 0.560 1.20"

9. 60 deg. G 0.35" 0.0470 1.67" 0.498 3.13"

• Data not taken.

• • The reflector iz nearer to the outside surface than tr.e inside surface. "S" to the outside surface is 1.77".

FIGURE A-5

SUMMARY

OF ULTRASONIC TEST INDICATIONS FOUND IN THE KEWAUNEE UNIT 1 STEAM GENERATOR "B" WELD 2-5 (20% DAC SIZING, 5.0 MHz) 2330s/0359s/080487.10 A-19

SURFACE/EMBEDDED FLAW DEMARCATION LINE 160 0.13 140

" FLAW 8EMBEDDE

-CONFIGURATION 0.12 0.11 0.10 120 0.09 0.08 16L 0.07 ~SURFAC 0

0.06 0

4SURFACE FLW U.

-A 0.05 4HE 70 X 0.04 0.03 0.02 0.01 D Westinghouse 1987 0

0 0.05 0.10 0.15 0.20 0.25 DISTANCE FROM SURFACE (4)

FIGURE A-6 DETERMINATION OF HYDROSTATIC TEST TEMPERATURES FROM RESULTS OF THE SPRING 1987 INSPECTION 2330s,0359s/080487:10 A-20

SURFACE/EMBEDDED FLAW DEMARCATION LINE 0.13 . ........

120

-EMBEDDED FLAW-----

-:CONFIGURATION :.J 0.12 0.11 7- 7; 0.10

-At: 04::

0.09 g 0.08 a S

IL. 0.07 0 7=: F_-

70

3 7- FLAWS IN THIS :t7:

0.06 REGION MUST BE CONSIDERED SURFACE LL 0.05 FLAWS

7. fi;.:

H.

= 0.04 0.03 H = 17--f-71 L.0 0.02

7' 0.01 j7 C Westinghouse 1987 0

0 0.05 0.10 0.15 0.20 0.25 DISTANCE FROM SURFACE ()

FIGURE A-7 DETERMINATION OF LEAKAGE TEST TEMPERATURES FROM RESULTS OF THE SPRING 1987 INSPECTIONS 2330s/0359s/08047:1 0 A-21

APPENDIX B RESULTS OF THE INSPECTION OF MARCH 1988 ON STEAM GENERATOR "A" B-1

SUMMARY

During the March 1988 ultrasonic examination of the Kewaunee Unit 1 steam generator "A" upper shell to cone weld [SG-W2 (Weld 1-5)], nineteen recordable indications were noted. Seven of these were detected with the 45 degree, 2.25 MHz shear wave examinations, and the remaining twelve were detected with the 60 degree, 2.25 MHz shear wave examinations. The location of these indications in the weld, past experience with the same weld in other steam generators at other plants, and supplemental examinations performed on this steam generator indicate that all these indications are volumetric in nature, i.e., small slag inclusions and/or voids. An evaluation of these indications (using -6 dB drop or half maximum amplitude sizing criteria) to the acceptance standards in table IWB-3511-1 of the ASME Code Section XI, 1980 Edition with the Winter 1981 Addenda results in fourteen indications which are unacceptable.

Using the fracture analysis rules of IWB-3600 and the guidelines of appendix A, both from the ASME Code Section XI, 1980 Edition with the Winter 1981 Addenda, all the indications are acceptable using 50% DAC sizing levels (2.25 MHz transducer data), and using -6 dB drop or half maximum amplitude sizing levels (5.0 MHz transducer data).

These examinations were performed with the same personnel and procedures utilized on numerous other plants. These other plants exhibited both inner diameter cracking conditions, subsurface fabrication flaws, or a combination of both at the recording levels established in the test procedures. The evaluation of examination data and the performance of supplemental investigations were conducted by engineering personnel directly involved with the evaluation of data from the same plants as specified above.

2949s-061788:10 B-1

B-2 ULTRASONIC EVALUATION AND DISCUSSION Nineteen recordable indications were noted during the recent examinations of the Kewaunee Unit 1 steam generator "A" upper shell to cone weld. Summary tables of the indications are presented in tables B-1 and B-2. Table B-1 provides the measured "2a" value, the measured "S" value, and the measured surface length all with respect to the normal to the inside pressure retaining of the component and determined using a 5.0 MHz transducer and -6 dB drop or half maximum amplitude sizing criteria. Table B-2 shows the same parameters using a 2.25 MHz transducer and 50% DAC sizing criteria. These values are measured using indication plots rather than calculated from the raw data due to the geometry of the weld. This evaluation scheme is to maintain adherence to the flaw indication characterization criteria provided in IWA-3300 and table IWB-3511-1 of Section XI. The majority of the indications were detected from the outer diameter surface of the transition cone but are physically located in the upper shell portion of the weld. The indication parameters

("2a", "z", and "S") therefore have been taken primarily from the surfaces of the upper shell. The 45 degree sizing data, with the exception of indications 4, 5, and 6, was taken using a 5.0 MHz, 45 degree shear wave transducer and a -6 dB drop or half maximum amplitude sizing criteria. The 45 degree shear wave indications 4, 5, and 6 were sized only using the detection data (2.25 MHz, 45 degree shear wave transducer, and 50% DAC sizing criteria.) The 60 degree sizing data was taken using a 5.0 MHz, 60 degree shear wave transducer and -6 dB drop or half maximum amplitude sizing criteria. Sizing data using a 60 degree, 2.25 MHz transducer and 50% DAC sizing criteria were also taken. Although both 2.25 MHz and 5.0 MHz sizing data were taken, the primary sizing data used for the fracture mechanics analysis was based on that taken with the 5.0 MHz transducer. Experience has shown that 2.25 MHz testing is excellent for detection in this application, but tends to oversize when used in conjunction with the Section XI criteria, and volumetric-type reflectors.

The 2.25 MHz transducer produces a beam spread which is wider than that of a similar size 5.0 MHz transducer. This factor typically results in an unavoidable overestimate of the true size of volumetric reflectors such as 2949s-061788:10 B-2

slag, which is believed to be present in this case. An example will illustrate this fact. Consider an indication which is being sized with a 2.25 MHz, 450 shear wave transducer and 50% DAC sizing criteria, as shown in figure B-1.

As the transducer is moved along the examination surface it picks up an indication (shown by the dot), and the first step is to locate the peak response of the indication, as shown in illustration (b). For illustration purposes, assume the amplitude is 100% of the distance amplitude correction curve (DAC). The peak response of the indication is then plotted in illustration (e), at an angle of 45 degrees from the transducer location. The distance along the 45 degree line is determined from the time base of the ultrasonic test instrument, which is a function of the speed of sound in the material.

The extent of the indication is then determined by moving the transducer along the surface until the amplitude drops to 50% DAC. This point is shown in illustration (c) for one direction, and corresponds to a reduction in the signal amplitude of the indication of 6 dB or one half in this case.Section XI requires this point to also be plotted at an angle of 45 degrees [see (f)]

even though it is clear from (c) that the angle is less than 45 degrees. A similar procedure is then followed to get the extent of the indication in the other direction (d) and the location is again plotted at 45 degrees [see (g)]

even though in this case the angle is clearly greater than 45 degrees. The through wall dimension of the indication, "2a", is then determined from projection of a line through the peak point perpendicular to the vessel inside surface, as shown in (g). The through wall dimension then follows from projection of the end points onto the perpendicular.

An illustration of how the flaw sizing and location changes with a narrower beam is shown in figure B-2. Here the example is exactly the same, but a 5.0 MHz transducer of similar size is used. The peak location or center of the indication is found to be identical to the previous example, as shown in (e) but the outer extent of the indication is considerably different, because the beam is narrower, and the projection of the outer 50% DAC limits (or in this 2949s-061 788:10 B-3

case -6 dB drop or half maximum amplitude limits) of the indication is less, as shown in (f) and (g). The through wall depth is much smaller, and also the distance from the inside surface is also much greater. This is exactly the situation which occurred with the indications in steam generator A, although the actual details were more complex.

Therefore, in the case of volumetric flaws a reduction in beam spread is desired to obtain a more realistic size. There are a number of ways to minimize the beam spread, including use of a higher frequency transducer, a focused transducer, a larger transducer size or a combination of these. The beam spread, 6, can be shown by simple physics [B1] to be related to the diameter (D) of the transducer and its frequency (f) as follows:

sin ~ kX kC TU where K = a constant C = speed of sound in the material X = wave length 6 = beam spread angle, defined in figures B-1 and B-2 f = frequency Beam spread effects can also be minimized by use of beam spread correction, which is essentially a correction on the plotted extremities of the indications, but data to support the accuracy of these calculations is limited. The use of other transducers is permitted by Paragraph T-451.1 of the ASME Code Section V, Article 4 which states that "other ultrasonic techniques and nondestructive examination methods may be helpful in determining a reflector's true position, size, and orientation".

The raw indication data from the detection examinations in steam generator A clearly indicate that the detected reflectors are embedded rather than surface. This is seen in the location of the peak responses. No peak response is observed at or near the inner diameter surface which would be expected for a surface breaking flaw. In addition the test operators did not observe any low level amplitude signals below the recording level located at 2949s-061 788:10 8-4

the inner diameter surface indicative of those found in plants having an inner diameter surface cracking condition. Supplemental examinations on three of the 60 degree shear wave indications originally determined to be surface by the rules established in Section XI resulted in the fact that these indications could be observed from both sides of the weld in a normal half-vee technique fashion as well as a 5/8-node technique with the peak locations embedded within the weld. The longest indication (approximately 12 inches long) was scanned with a 0 degree, 5 MHz longitudinal wave probe resulting in a confirmation of a cluster of reflectors at positions approximately 3.2 to 3.4 inches below the outer diameter surface for the entire length of the indication. At the same transducer position that this cluster was detected, a backwall response at 3.9 inches below the outer diameter surface was noted.

This indicates a thickness of 3.9 inches and a difference in position between the volumetric reflectors and the inner diameter surface of 0.5 to 0.7 inch.

All examination data, therefore, clearly suggest embedded flaws.

B-2.2 Experience With Other Plants The indications in steam generator A at KNPP appear to be quite characteristic of experience with various welds in steam generators and pressurizers at other plants where preservice ultrasonic examination results based on 2.25 MHz, 50% DAC sizing methods predicted reflectors detected in weld backchip regions had dimensions in excess of those allowable values provided in Section XI of the ASME Code. Attempts were made at other plants to confirm the size, location, and orientation of these indications by complementary nondestructive examination methods, i.e. 0 degree longitudinal wave examinations, and both fabrication and field radiography. No reliable responses could be observed from the shear wave indications using the straight beam examinations. In terms of the radiography, the fabrication radiographs of the areas in question were reviewed with no conclusive results.

Additionally, field radiography was performed in selected areas at these plants but again no confirmation of the shear wave examination indications could be obtained.

These inconclusive results led to physical removal of some of the suspect indications by mechanical means for complete metallurgical characterization.

294es-061788:1o B-5

The indications were found to have been caused by small slag inclusions and voids between weld passes in the weld backchip area near the inside surface.

Measurements made during the destructive analysis showed that the ultrasonic sizing using 2.25 MHz, 50% DAC sizing methods exaggerated the true size of the discontinuities in terms of length and/or through-wall dimensions. These results are presented in table B-3, and plotted in figure B-3. These results agree closely with the illustrations previously presented.

Furthermore, this experience correlates well with investigations to date which have shown that when sizing volumetric-type reflectors by amplitude drop methods, i.e. 2.25 MHz, 50% DAC, the typical result is that the beam size rather than the reflector size is measured. For example, the lower the test frequency, the larger the beam width resulting in a larger than actual apparent flaw size (references B2-B7).

B-2.3 1988 Inspection Conclusions Since the data clearly suggested volumetric-type reflectors at KNPP the use of a more realistic volumetric flaw sizing approach was implemented. This sizing approach consisted of using a 5.0 MHz transducer and a -6 dB or half maximum amplitude sizing criteria. The angle used in sizing was dependent on the angle which detected the indication. The 5.0 MHz transducer resulted in a smaller beam spread in comparison with the true size of the suspect reflectors. The -6 dB or half maximum sizing criteria was selected because it has provided the better accuracies when compared with 50% DAC or 20% DAC sizing levels (reference B8).

Using the data in tables B-3 and B-4, two sets of evaluation calculations were performed. The first evaluation compared the characteristics of the 2.25 MHz detection data to the acceptance standards described in table IWB-3511-1 of the ASME Code Section XI, 1980 Edition with the Winter 1981 Addenda. This evaluation resulted in sixteen indications which were unacceptable (table B-5). The second evaluation compared the characteristics of the data composite sizing (5.0 MHz and 2.25 MHz data) to the acceptance standards described in table IWB-3511-1 of the ASME Code Section XI, 1980 Edition with 2949s-061 788:10 B-6

the Winter 1981 Addenda. This resulted in fourteen indications which were unacceptable (table B-4). All indications sized with the 5.0 MHz transducer are classified as subsurface indications.

Since the indications found in these examinations are ultrasonically similar to those detected at other plants it was appropriate to use higher frequency transducers to obtain more realistic data concerning the through-wall dimensions of the indications. Since 45 degree indications numbers 4, 5, and 6 sized with 2.25 MHz, 50% DAC methods were within the acceptance standards in table IWB-3511-1 (ASME Section XI, 1980 Edition with the Winter 1981 Addenda),

no high frequency data were taken.

B.3 FRACTURE ANALYSIS There are two alternative sets of acceptance criteria for continued service without repair in paragraph IWB-3600 of the ASME Code Section XI:

1. Acceptance criteria based on flaw size (IWB-3611)

2. Acceptance criteria based on stress intensity factor (IWB-3612)

The choice of criteria is at the convenience of the user per IWB-3610. The more beneficial criteria of IWB-3612 have been used for evaluating the nineteen indications.

To determine the allowable flaw sizes in a weld, finite element analysis methods were used.

All applicable plant transients were analyzed to select the most severe stress profiles through the thickness of the weld. The actual stress profiles were then approximated by third order polynomials and used for calculating the stress intensity factor (KI) for various crack sizes and aspect ratios.

The resulting K1 values were compared to fracture toughness values (KIa and K ). Critical flaw sizes were then obtained, and allowable flaw sizes determined using the acceptance criteria discussed above.

2949s-061788:10 B-7

The final step involves calculation of crack growth due to fatigue loading.

All anticipated plant transients were utilized in determining the resulting flaw size for a specified period of time. This was done for 10, 20, and 30 year intervals.

In addition to satisfying the fracture criteria, it is required that the primary stress limits of Section III paragraph NC-3000 be satisfied. A local area reduction of pressure retaining membrane must be used, equal to the area of indication; and the stresses increased to reflect the smaller cross section.

The nineteen indications found are all subsurface flaws as defined by IWB-3500. As shown in figure B-4, all nineteen indications are acceptable per the fracture analysis criteria of IWB-3600. The fracture evaluation methods used for these analyses have been documented in the main body of this report.

It should be mentioned that some elevation of the hydrotest and leak test temperatures over the specified temperature will be required to ensure the margins of IWB-3600 are maintained, and these temperatures have been provided along with the complete technical details of the analysis in the main body of this report. The revised hydrotest and leak test temperatures from this inspection are provided in figures B-5, B-6 and B-7.

B-4 SECONDARY WATER CHEMISTRY AND TUBE MATERIAL Between 1979 and 1988 copper tubing in all the major secondary side heat exchangers and the condenser were replaced with stainless steel. The Plant Chemistry Group has interfaced with the Plant Operations Group to ensure favorable secondary side water chemistry. (Table B-6) KNPP S/G Chemistry limits are based on both EPRI and Westinghouse chemistry guideline philosophy. Efforts towards the optimization of water chemistry inside the steam generators has resulted in maintaining a suitable environment in terms of preventing the corrosion of the steam generator girth weld.

2949s-061 788:10 B-8

All of the indications noted during the 1988 inspections of Steam Generator "A" are volumetric in nature, subsurface, and are prior existing slag inclusions and/or voids. Further, the steam generator water chemistry and sludge data would support the conclusion the indications are not corrosion induced or crack-like.

B-5 REFERENCES B1. Krautkramer, J., and H. Krautkramer. Ultrasonic Testing of Materials, Springer-Verlag New York Inc., New York, 1969, page 83.

B2. Gruber, G. J., Hendrix, G. J. and Schick, W. R. "Characterization of Flaws in Piping Welds Using Satellite Pulses", Materials Evaluation, April 1984.

B3. Cook, R. V., Latimer, P. J. and McClung, R. W. Flaw Measurement Using Ultrasonics in Thick Pressure Vessel Steel, final report on Contract No.

W-7405-eng-26, prepared by Oak Ridge National Laboratory for the U.S.

Nuclear Regulatory Commission, Aug. 1982, Oak Ridge, TN.

B4. Doctor, S.R., Becker, F. L., Heasler, P. G. and Selby, G. P.

"Effectiveness of U.S. Inservice Inspection Techniques - A Round Robin Test," Proceedings of Specialist Meeting on Defect Detection and Sizing, Ispra, Italy, May 3-6, 1983. Joint Research Center, Ispra (Va), Italy.

B5. Jessop, T. J., Mudge, P. J. and Harrison, J. D. Ultrasonic Measurement of Weld Flaw Size, National Cooperative Highway Research Program Report 242, prepared for the Transportation Research Board by The Welding Institute, Dec. 1981. The Welding Institute, Cambridge, England.

B6. Mudge, P. J. and Jessop, T. J. "Size Measurement and Characterization of Weld Defects by Ultrasonic Testing: Findings of a Collaborative Programme," Proceedings of NDE in Relation to Structural Integrity, Paris, France, Aug. 24-25, 1981. Applied Science Publishers, Ltd.,

London, England.

2949s-061 788:10 B-9

B7. Rishel, R.D. "Summary Report: Volumetric Flaw Depth Sizing,"

MT-SMART-807, September 12, 1985 (submitted to Seabrook Power Station).

B8. Willetts, A. J., Ammirato, F. V., and Kietzman, E. K., Jones, J. A.

Applied Research Company. Accuracy of Ultrasonic Flaw Sizing Techniques for Reactor Pressure Vessels, EPRI RP1570-2 Draft Interim Report, March 1988.

B9. Letter, Tomes WPS to Kurek W PSD. June 10, 1988,

Subject:

"Secondary Side Water Chemistry."

2949s-061788:10 B-10

TABLE B-1

SUMMARY

OF ULTRASONIC TEST INDICATIONS FOUND IN THE KEWAUNEE UNIT 1 STEAM GENERATOR "A" WELD 1-5 (5.0 MHZ TRANSDUCER, -6 dB DROP SIZING, SIZING DATA)

MEASURED 1"S DATA INDICATION "2a" (inside surface) LENGTH

1. 45 degree 1 0.14" 0.08" 0.35"

2. 45 degree 2 0.30" 1.04" 0.50"

3. 45 degree 3 0.37" 0.28" 0.95" 4 -- -- -

4. 45 degree

5. 45 degree 5 -- -- -

6 -- -- -

6. 45 degree

7. 45 degree 7 0.31" 0.59" 1.0"

8. 60 degree 1 0.23" 0.35" 0.75"

9. 60 degree 2 0.52" 0.26" 1.2"

10. 60 degree 3 0.35" 0.12" 1.55"

11. 60 degree 4 0.52" 0.26" 1.5"

12. 60 degree 5 0.30" 0.65" 2.1"

13. 60 degree 6 0.47" 0.41" 1.4"

14. 60 degree 7 0.30" 0.69" 2.9"

15. 60 degree 8 0.47" 0.20" 12.1"

16. 60 degree 9 0.35" 0.69" 13.25"

17. 60 degree 10 0.41" 0.71" 1.0"

18. 60 degree 11 0.47" 0.37" 1.8"

19. 60 degree 12 0.37" 0.30" 2.8" 2949s-061788:10 B-1

TABLE B-2

SUMMARY

OF ULTRASONIC TEST INDICATIONS FOUND IN THE KEWAUNEE UNIT 1 STEAM GENERATOR "A" WELD 1-5 (2.25 MHZ TRANSDUCER, 50% DAC SIZING, DETECTION DATA)

"S"'

MEASURED "2a" (inside surface) LENGTH DATA INDICATION

1. 45 degree 0.35" 0.12" 1.25"

2. 45 degree 0.43" 0.51" 0.85"

3. 45 degree 0.39" 0.39" 0.85"

4. 45 degree 0.12" 0.59" 0.60"

5. 45 degree 0.23" 0.67"

6. 45 degree 0.87"

7. 45 degree 0.23" 0.53" 1.4"

8. 60 degree 0.76" 0.0" 0.75"

9. 60 degree 0.46" 0.35" 1.0"

10. 60 degree 0.76" 0.12" 1.55"

11. 60 degree. 0.52" 0.41". 0.9"

12. 60 degree 0.41" 0.34" 0.75"

13. 60 degree 0.64" 0.41" 1.0"

14. 60 degree 0.47" 0.64" 0.7"

15. 60 degree 0.42" 0.07" 12.1"

16. 60 degree 0.29" 0.82" 13.25"

17. 60 degree 0.46" 0.65" 1.0"

18. 60 degree 0.46" 0.47" 1.8"

19. 60 degree 0.58" 0.23" 1.8"

• To small to measure 2949s-061 788:10 B-12

TABLE B-3 NONDESTRUCTIVE VERSUS DESTRUCTIVE TESTING RESULTS USING 2.25 MHZ, 50% DAC SIZING PHYSICAL DISTANCE FROM ID THROUGH-WALL LENGTH SAMPLE SURFACE DEPTH UT ACTUAL UT ACTUAL UT ACTUAL CORE #1 ** .37" 0.09" 1.18" 1.15" (Plant 1) to 1.03 to 3.18" CORE #2 ** **

.16" 0.02" .63" 0.45" (Plant 1) to .58" to .75" CORE #1 0.00" 0.08" to 0.24" 0.01" 0.88" 0.25" to (Plant 2) 0.33"

• to 0.33" 0.28" CORE #2 0.16" 0.82" 0.53" 0.18" 0.88" 0.27" (Plant 2)

GRINDING 0.05" ** 0.37" 1.00" (Plant 2)

GRINDING 0.00" 0.375" 0.45" 0.094" 3.5" (Plant 2)

GRINDING 0.00" 0.125" 0.51" 0.156" 3.25" (Plant 2)

GRINDING 0.02" 0.156" 0.43" 0.219" 0.75" 0.375" (Plant 2)

GRINDING 0.00" 0.24" 0.75" **

(Plant 2)

GRINDING 0.00" 0.219" 0.33" 0.343" 1.0" 0.438" (Plant 2)

• One UT indication was found to be four indications upon metallurgi cal evaluation. The values show the range of sizes for these four defects.

• • Dimensions not reported.

294gs-061788:10 B-1

TABLE B-4 RESULTS OF THE ASME SECTION XI, 1980 EDITION WITH THE WINTER 1981 ADDENDA CALCULATIONS USING THE ACCEPTANCE STANDARDS OF TABLE IWB-3511-1 (COMPOSITE SIZING DATA)

INDICATION MEASURED TYPE OF a/t a/t NO. "2a"[1] IND. "a" "S" "." ALLOW. ACTUAL DATA

1. 45 deg. 1 0.14" subsurf. 0.07" 0.08" 0.35" 3.6% 1.9%

2. 45 deg. 2 0.30" subsurf. 0.15" 1.04" 0.50" 4.6% 4.0%

3. 45 deg. 3 0.37" subsurf. 0.19" 0.28" 0.95" 3.6% 5.1%

4. 45 deg. 4 0.12" subsurf. 0.06" 0.59" 0.60" 2.9% 1.6%

5. 45 deg. 5 0.23" subsurf. 0.12" 0.67" 7.2% 3.2%

6. 45 deg. 6 subsurf. 0.87"

7. 45 deg. 7 0.31" subsurf. 0.16" 0.59" 1.0" 3.3% 4.2%

8. 60 deg. 1 0.23" subsurf. 0.12" 0.35" 0.75" 3.2% 3.2%

9. 60 deg. 2 0.52" subsurf. 0.26" 0.26" 1.2" 3.8% 6.8%

10. 60 deg. 3 0.35" subsurf. 0.35" 0.12" 1.55" 2.0% 4.8%

11. 60 deg. 4 0.52" subsurf. 0.26" 0.26" 1.5" 3.4% 7.0%

12. 60 deg. 5 0.30" subsurf. 0.15" 0.65" 2.1" 2.8% 4.0%

13. 60 deg. 6 0.47" subsurf. 0.24" 0.41" 1.4" 3.4% 6.4%

14. 60 deg. 7 0.30" subsurf. 0.15" 0.69" 2.9" 2.8% 4.0%

15. 60 deg. 8 0.47" subsurf. 0.24" 0.20" 12.1" 2.2% 6.1%

16. 60 deg. 9 0.35" subsurf. 0.18" 0.69" 13.25" 2.6% 4.9%

17. 60 deg. 10 0.41" subsurf. 0.21" 0.71" 1.0" 3.7% 5.7%

18. 60 deg. 11 0.47" subsurf. 0.24" 0.37" 1.8" 3.1% 6.4%

19. 60 deg. 12 0.37" subsurf. 0.19" 0.30" 2.8" 2.8% 5.1%

• To small to measure.

[1] From table B-1 except for 45 degree indications 4, 5, and 6 which are from table B-2.

2949s-061 788:10 B-14

TABLE B-5 RESULTS OF THE ASME SECTION XI, 1980 EDITION WITH THE 1981 WINTER ADDENDA IWB CALCULATIONS USING THE ACCEPTANCE STANDARDS OF TABLE 3511-1 2.25 MHZ TRANSDUCER, 50% DAC SIZING DETECTION DATA INDICATION MEASURED TYPE OF a/t a/t NO. "2a" IND. "a"l "" ALLOW. ACTUAL DATA

1. 45 deg. 1 0.35" subsurf. 0.18" 0.12" 1.25" 2.1% 4.8%

2. 45 deg. 2 0.43" subsurf. 0.22" 0.51" 0.85" 4.2% 5.9%

3. 45 deg. 3 0.39" subsurf. 0.20" 0.39" 0.85" 4.0% 5.4%

4. 45 deg. 4 0.12" subsurf. 0. 06" 0.59" 0.60" 2.9% 1.6%

5. 45 deg. 5 0.23" subsurf. 0.12" 0.67" 7.2% 3.2%

• 0.87"

6. 45 deg. 6 subsurf.

7. 45 deg. 7 0.23" subsurf. 0.12" 0.53" 1.4" 2.9% 3.1%

8. 60 deg. 1 0.76" surface 0.76" 0.0" 0.75" 3.7% 20.4%

9. 60 deg. 2 0.46" subsurf. 0.23" 0.35" 1.0" 3.9% 6.2%

10. 60 deg. 3 0.76" surface 0.76" 0.12" 1.55" 3.7% 20.4%

11. 60 deg. 4 0.52" subsurf. 0.26" 0.41" 0.9" 4.5% 7.0%

12. 60 deg. 5 0.41" subsurf. 0.21" 0.34" 0.75" 4.3% 5.5%

13. 60 deg. 6 0.64" subsurf. 0.32" 0.41" 1.0" 4.8% 8.6%

14. 60 deg. 7 0.47" subsurf. 0.23" 0.64" 0.7" 5.1% 6.17%

15. 60 deg. 8 0.42" surface 0.42" 0.07" 12.1" 1.7% 11.4%

16. 60 deg. 9 0.29" subsurf. 0.15" 0.82" 13.25" 2.64% 4.1%

17. 60 deg. 10 0.46" subsurf. 0.23" 0.65" 1.0" 3.9% 6.3%

18. 60 deg. 11 0.46" subsurf. 0.23" 0.47" 1.8" 3.08% 6.17%

19. 60 deg. 12 0.58" subsurf. 0.29" 0.23" 1.8" 2.59% 7.77%

• To small to measure.

2949s-061788:10 B-15

TABLE B-6 SECONDARY SIDE HEAT EXCHANGER TUBING MATERIALS COMPONENT MATERIAL Condenser 439 Stainless Steel Feedwater Heaters 439 Stainless Steel 11A, 11B, 12A, 12B, 13A and 13B Feedwater Heaters 304 Stainless Steel 14A, 14B, 15A and 15B Moisture Separator Reheaters 439 Stainless Steel 1A1, 1A2, 1B1 and 1B2 Gland Steam Condenser 439 Stainless Steel SECONDARY SIDE CHEMISTRY:

- Phosphates were eliminated from the secondary water during the middle of the first cycle.

- The in-line chemistry monitors, their ranges and the alarm setpoints are based on vendor recommendations, EPRI guidelines and plant experience.

- KNPP cleans-up the secondary water systems during start-ups.

- Sludge lancing, tube bundle washdowns, and wet lay-up are routine steam generator activities during refueling outages.

- Feedwater heater tube cleaning is routinely performed during refueling outages.

- There are chemistry hold points at low power operations at KNPP.

2949s-061788:10 B-16

EXAMPLE OF 2.25 MHZ. 45 SHEAR- SO% DAC SIZING (ALSO - 6 DB DROP OR HALF MAXIMUM AMPLITUDE SIZING IN THIS CASE ONLY)

EXAMPLE ASSUMPTIONS

- MAXIMUM AMPLITUDE OF RESPONSE, 100% DAC

- DIAMETER OF TRANSDUCER = D Figurle B-1. Schematic Example of Flaw Sizing with 2.25 MHz Transducer Using 50% DAC Sizing Levels. (This particular example also shows -6 dB drop or half maximum amplitude sizing.)

2949s/032588: 10 B-17

50% DAC SIZING XAMPLE OF 5.0 MHZ. 45 SHEAR, AMPIFLITUDE SIZING (ALSO - 6 DB DROP OR HALF MAXIMUM IN THIS CASE ONLY)

EXAMPLE ASSUMPTIONS 100 DAC

- MAXIMUM AMPLITUDE OF RESPONSE,

= D

- DIAMETER OF TRANSDUCER Figure B-2. Schematic Example of Flaw Sizing With 5.0 MHz Transducer Using 50% DAC Sizing Levels. (This particular example also shows -6dB drop or half maximum amplitude sizing.)

2949s/032588:10 B-18

NONDESTRUCTIVE VERSUS DESTRUCTIVE TESTING RESULTS (2.25 MHz TRANSDUCER, 50% DAC SIZING) 1.1 I 1.0 U

0.9 L

z 0.8 0.7 I

0 0.6 w d I-.

0.5 W_

X 0

0.4 0

a:

0.3 0.2 0.1 0.0L 0.0 0.1 0.2 0.3 0.4 0.5 ACTUAL THROUGH-WALL DEPTH (INCH)

Figure B-3. Nondestructive vs. Destructive Testing Results, 2.25 MHz 2949s/032588:10 Transducer with 50% DAC Sizing

SURFACE/EMBEDDED FLAW DEMARCATION LINE 0.13 EMBEDDED FLAWE .  :: a CONFIGURATION FLAWS WITH 0.12 ABOVE THIS LINE ARE NOT ALLOWABLE 0.11 0.10 0.09 0.08 U 0.07 SURFACE- ------- 1 0 = FLAWS IN THIS 0.06 REGION MUST BE CONSIDERED SURFACE ..

L 0.05

_j 0.04 FLAS . ALL EMBEDDED FLAWS

~ t -+EMARKATI4N LINE) 0.03 DARE ACCEPTABLE PER CRITERIA OF IWB 3600 LFLAWS AS t ASONG 0.02 0.01 Westinghouse 1987 0

0 0.05 0.10 0.15 0.20 0.25 DISTANCE FROM SURFACE ()

DATA IND. TYPE OF MEASURED MEASURED MEASURED a/t C/t ACCEPT IND. "2a" "S" Mtn ABLE

1. 45 deg. 1 subsurf. 0.14" 0.08" 3.73" 0.019 0.040 yes

2. 45 deg. 2 subsurf~. 0.30" 1.04" 3.73" 0.040 0.320 yes

3. 45 deg. 3 subsurf. 0.37" 0.28" 3.73" 0.051 0.126 yes

4. 45 deg. 4 subsurf. 0.12" 0.59" 3.73" 0.016 0.174 yes

5. 45 deg. 5 subsurf. 0.23" 0.67" 3.73" 0.032 0.212 yes

6. 45 deg. 6 subsurf. 0.00" 0.87" 3.73" 0.000 0.233 yes

7. 45 deg. 7 subsurf. 0.31" 0.59" 3.85" 0.042 0.195 yes

8. 60 deg. 1 subsurf. 0.23" 0.35" 3.73" 0.032 0.126 yes

9. 60 deg. 2 subsurf. 0.52" 0.26" 3.73" 0.068 0.139 yes

10. 60 deg. 3 subsurf. 0.35" 0.12" 3.73" 0.048 0.080 yes

11. 60 deg. 4 subsurf. 0.52" 0.26" 3.73" 0.070 0.139 yes

12. 60 deg. 5 subsurf. 0.30" 0.65" 3.79" 0.040 0.211 yes

13. 60 deg. 6 subsurf. 0.47" 0.41" 3.73" 0.064 0.174 yes

14. 60 deg. 7 subsurf. 0.30" 0.69" 3.37" 0.040 0.225 yes

15. 60 deg. 8 subsurf. 0.47" 0.20" 3.93" 0.061 0.112 yes

16. 60 deg. 9 subsurf. 0.35" 0.69" 3.68" 0.049 0.236 yes

17. 60 deg. 10 subsurf. 0.41" 0.71" 3.68" 0.057 0.250 yes

18. 60 deg. 11 subsurf. 0.47" 0.37" 3.73" 0.064 0.164 yes

19. 60 deg. 12 subsurf. 0.37" 0.30" 3.73" 0.051 0.131 yes Figure B-4. Fracture Analysis Results for Indications Found in the Kewaunee Unit 1 Steam Generator "A" Weld 1-5 (Composite Sizing Data) 2949s/032588:10 B-20

SURFACE/EMBEDDED FLAW DEMARCATION 160 0.13 140 FLAW.:

I17:

-:- -EMBEDDED

- CONFIGURATION 0.12 0.11 0.10 120

-0.09 U.. ,

S0.08 U. 0.07 0.0 u 0.05 REGION MUST BE 0.07 SURFACE3 2 0.06

-J 70 z 0.04 FLAS

.;EGOINTH MUS BE:: 50 0.03 0.02 0.01 0 Westinghouse 1987 0

0 0.05 0.10 0.15 0.20 0.25 DISTANCE FROM SURFACE )

Figure B-5. Determination of Hydrostatic Test Temperatures from Results of the March 1988 Inspection (Composite Sizing Data) (p = 1356 psi) 294gs/060288: 10 B-21

SURFACE/EMBEDDED FLAW DEMARCATION UNE 140 0.13 120

- EMBEDDED FLAW:

-" CONFIGURATION 0.12 0.11 0.10 0.09 0.08

-- 2 0.07 --

70 0.06 AWS INT A EGION MUSTIBE=

CONSIDERED 50 SURFACE 0.05 X 0.04 0.03 0.02

. -::... = : ,  :.1 0.01 l A_. 0 Westinghouse 1987 0

0 0.05 0.10 0.15 0.20 0.25 DISTANCE FROM SURFACE a Figure B-6. Determination of Leakage Test Temperatures from Results of the March 1988 Inspections (Composite Sizing Data) (p = 1085 psi) 2949s/060288:1 0 B-22 SURFACE-*

SURFACE/EMBEDDED FLAW DEMARCATION LINE 0.13 g .CONFIGURATION ....

0.12 0.11 70 0 F 0.10 50 0.09

• 1 S0.08 s a 0.07 -SURFACE 0

FLAWS IN THIS 0.06 REGION MUST BE A7

=CONSIDERED SURFACE U. 0.05 FLAWS--4 X 0.04 T" 1t i,:

0.03 0.02 0.01 j!~L4~ I iw?

O Westinghouse 1987 0

0 0.05 0.10 0.15 0.20 0.25 DISTANCE FROM SURFACE (a)

Figure B-7. Determination of Leakage Test Temperatures from Results of the March 1988 Inspections (Composite Sizing Data) (p = 750 psi) 2949s/060288: 10 B-23

WESTINGHOUSE PROPRIETARY CLASS 3 FW APPENDIX C RESULTS OF THE INSPECTION OF APRIL 1991, KEWAUNEE STEAM GENERATORS A AND B C-1

SUMMARY

During the Spring 1987 manual ultrasonic examinations of the Kewaunee Unit 1 Steam Generator "B" upper shell to cone weld, a total of nine indications were detected and recorded with 45 and 60 degree, 2.25 MHz shear wave examinations. At that time, all available evidence including inspection experience with the same weld joint at other plants resulted in a conclusion that the indications were volumetric in nature, i.e., slag and small voids.

In the Spring 1988 examinations of the Kewaunee Unit 1 Steam Generator "A" upper shell to cone weld, a total of 19 indications were detected and recorded using the same examination methodology.

The indications were quite similar to those found on Steam Generator "B" in terms of through-thickness location. Again, the evidence suggested that all of the recorded indications were volumetric in nature, occurring primarily in the weld "back chip" area within an inch of the I.D. surface.

C-1

WESTINGHOUSE PROPRIETARY CLASS 3 In both investigations (detailed in Appendix A and B of this report) the indications were sized again with 5.0 MHz search units which resulted in a more realistic assessment of the through-wall dimensions. Investigations were also conducted with straight beam search units and several indication locations were confirmed, supporting the conclusion that the flaws were embedded.

The March 1991 examinations of the Kewaunee Unit 1 Steam Generator "A" and "B" upper shell to cone welds (weld numbers 1-5 and 2-5),

were conducted in two phases which will be identified herein as manual and automated. In the manual examinations, which were conducted primarily as a detection and surveillance activity, both upper shell cone welds were completely re-examined with the previously used code compliant techniques requiring 2.25 MHz transducers and calibrations performed on the standard steam generator calibration block using side drilled holes. This examination served the purpose of relocating and measuring previously disclosed indications in both welds for comparison.

All recordable indications identified in the 2.25 MHz examinations on both upper shell to cone welds were sized again with 5 MHz transducers using -6dB drop or half maximum amplitude sizing criteria. In terms of the amplitude drop sizing methods, the 5 MHz data provides a more realistic estimate of flaw sizes due to the C-2

WESTINGHOUSE PROPRIETARY CLASS 3 smaller beam size and the nature of embedded discontinuities. Since the majority of the indications in both girth welds are essentially point reflectors, where beam size rather than flaw size is measured, the 5 MHz data proved to be conservative and, therefore, appropriate for the current fracture mechanics evaluation.

In the automated examinations, all indication areas in both upper shell to cone welds were examined with techniques and equipment specifically designed to resolve flaw features in the greatest possible detail. Scanning routines were designed to allow for complementary angle/scan direction studies at high sensitivity to establish a) the through-thickness location of the indications, b) perform amplitude independent sizing measurements on all indications revealed in the manual examinations and c) examine the inner diameter surface of the steam generator for evidence of cracking.

The Dynacon Systems Inc. Ultrasonic Data Recording and Processing (UDRPS) system was utilized for all of the automated data acquisition and served as the primary analysis tool for indication through-wall sizing, for the assessment of indication through-thickness location, and for the measurement of the ligament of metal between the indication and the inside diameter surface. Nearly 100 data sets comprising approximately 1000 scans were performed and stored on optical disk for off line analysis.

C-3

WESTINGHOUSE PROPRIETARY CLASS 3 F Using the fracture analysis rules of IWB-3600 and the guidelines of Appendix A, both from the ASME Code Section XI, 1980 Edition with the Winter 1981 Addenda, all the indications are acceptable using -6dB or half maximum amplitude sizing levels with 5.0 MHz manual examination data (Ref. Figures C-1.1, C-1.2 and C-2).

Indication analysis conducted on the processed UDRPS data using amplitude independent sizing techniques resulted in the best possible estimate of through-wall extent. When combined with the 50% DAC length measurements from the manual examinations, the dimensions of all recorded flaws in both upper shell to cone welds are within the allowable limits of the ASME Code,Section XI, Table IWB 3511-1.

Figures C-3 and C-4 show the UDRPS sizing data for Steam Generator A and B plotted on the fracture analysis flaw chart.

C-2 ULTRASONIC EVALUATION AND DISCUSSION C-2.1 MANUAL EXAMINATIONS In the 1991 manual examinations, both Steam Generator upper cone welds were completely re-examined with 2.25 MHz, 45 and 60 degree shear wave transducers. Calibrations for the examinations were conducted on block #WPS-36, a 3.5 inch thick carbon steel reference block used for all previous inspection work on the upper cone circumferential welds. The examinations were performed as the C-4

WESTINGHOUSE PROPRIETARY CLASS 3 primary detection technique, for the purpose of identifying and measuring all previously disclosed ultrasonic indications, and to identify additional areas requiring further investigation. The results of these detection examinations are as follows:

In Steam Generator "A" (weld 1-5), a total of 32 indications were detected. This total included 19 recordings which were directly related by position to the 19 indications recorded in 1988 (12-60 degree and 7-45 degree). In 5 instances, the same indication was recorded by two beam angles or by the same beam angle in opposing directions. Comparison of the 1991 data and 1988 data was possible in all cases, and the recorded lengths, amplitudes and through-wall sizes were essentially unchanged. In addition, 13 previously unidentified volumetric type indications were recorded for further investigation (Figures C-5.1 through C-5.3).

In Steam Generator "B" (weld 2-5), a total of 17 indications were detected. This total included 9 recordings which were directly related by position to the 9 indications recorded in 1987. In 2 instances, the same indication was recorded by two beam angles.

Comparison of the 1991 and 1987 data was possible in all cases, and the recorded lengths, amplitudes and through-wall sizes were essentially unchanged. In addition, 8 previously unidentified volumetric type indications were recorded for further investigation (Figures C-6.1 and C-6.2).

C-5

WESTINGHOUSE PROPRIETARY CLASS 3 All 32 detections in Steam Generator "A" and the 17 detections from Steam Generator "B" were sized with a 5.0 MHz, 0.5 inch square transducer using the -6dB or half-maximum amplitude technique, regardless of their size or acceptability status. In the -6dB technique, the indication peak response was normalized at a level of 80% full screen height. The examiner then recorded the peak, and minimum and maximum sweep position and transducer location data as the amplitude dropped to 40% full screen height.

The purpose of the 5 MHz half maximum amplitude sizing data was to apply conservative amplitude based flaw measurements for the fracture mechanics analysis and to provide for a reliable baseline of recorded data if future manual examinations are required.

Supplemental 0 degree examinations using a 5.0 MHz, 0.5 inch dia.

single and dual element transducers were performed on all the indication areas in both upper shell to cone welds. All indications were identified with these examinations.

C-2.2 AUTOMATED EXAMINATIONS Automated examinations using the UDRPS system were conducted on all the indications areas identified in the detection examinations.

Prior to conducting the examination, a system and sizing technique qualification demonstration was performed in the presence of the C-6

WESTINGHOUSE PROPRIETARY CLASS 3 Authorized Nuclear Inservice Inspector, and representatives from Wisconsin Public Service Plant Engineering, Quality Control, and Quality Assurance departments. The qualification process involved automated data acquisition with 45 and 60 degree shear wave transducers on the 3.5 inch Code calibration block, other reference blocks, and the detection and sizing of embedded and surface connected mechanical fatigue cracks in carbon steel qualification blocks.

.C.2.2.1 .TEST.PARAMETERS Transducer selection was based primarily on the potential for high resolution of specific flaw features. The KB Aerotech Alpha, 2.25 MHz 0.75 inch square element was chosen as the primary acquisition transducer in both the 45 degree and 60 degree examinations. The KB Aerotech Gamma, 2.25 MHz, 0.5 x 1.0 inch 45 degree search unit was also used effectively in both the demonstration and in a limited number of scans performed on Generator "A". Transducer frequency was limited to 2.25 MHz for the purposes of maximizing penetration with the least practical amount of system gain, enabling the analyst to increase gain in the range of 5 to 10 times ASME Code sensitivity (+14 to +20dB).

C-7

WESTINGHOUSE PROPRIETARY CLASS 3 UDRPS acquisition parameters were adjusted for the highest possible effective digitization rate (EDR), which is expressed in range resolution cell size (RRC). The spacing or distance between consecutively recorded A-scans is referred to as the inter pulse period (IPP). On the UDRPS display, these values represent one pixel movement in the scan direction and one in the time domain as follows:

• I- I I I / N I V i _

C 1 I I 4 -

I I

SYSTEM. DELAY (S) SUFFER LENGTH (MC)

SUFF DEL (PZC)

AMPLITUDE Ipp sacreen he;ght InT voltage counLs (0-55) color Scale u5 C-8

WESTINGHOUSE PROPRIETARY CLASS 3 These values are a reflection of measurement capability in the processed data. In the case of the 45 degree transducer, the RRC size was 0.25 microseconds and in the 60 degree data, the RRC size was 0.5 microseconds, or metal path resolution values of 0.0313 and 0.0626 inches respectively. In both cases, the IPP was set at 0.02 inches.

System set-up parameters were established on the 3.5 inch Code calibration block. It is here that refracted angle, velocity, and pulses per beam width can be set, and system performance was verified at scanning speed.

The motorized scanner selected for this application was a standard Dynacon Systems "Dynascanner" equipped with encoder feedback. Scanning speed was maintained smoothly at 1.5 inches per second in both the qualification trials and during data acquisition. The distances between scan lines was maintained at 0.25 inches.

C-2.2.2 RESULTS OF THE SIZING QUALIFICATIONS A total of 4 blocks were used in the sizing qualification demonstration. The 3.5 inch thick Code calibration block, identified as #WPS-36, contains 0.187 inch diameter side drilled holes and a 2% deep buttress notch. Scanning was performed over both reflector types. The side drilled holes (Fig. C-7) provided C-9

WESTINGHOUSE PROPRIETARY CLASS 3 a reference source for verifying the calculation in the Satellite Pulse Observation Technique (SPOT)(1) as applied to an ideal cylindrical void. The buttress notch provided a good representation of a surface intersecting planar flaw. The notch reflector was scanned at a gain value of calibration sensitivity

+6dB. From figure C-8, the notch target image is saturated, (highly reflective) showing an image on both sides of the block O.D. surface. Even though the notch is an ideal surface reflector, the response characteristics are similar to actual surface breaking cracks in ferritic specimens.

A second reference block, identified as the DSI block, provided a side drilled hole reference source for verifying the SPOT technique on ideal cylindrical void sizes significantly smaller than the code block (0.046" vs. 0.187"), (Fig. C-9). This range of hole sizes enabled the analysts to derive a simplified formula which could be effectively applied to the rounded inclusion sizes which might be expected in the examination.

do= AZt x .7145 where: do = void diameter AZt = metal path (synchronous satellite) - metal path (specular)

C-10

WESTINGHOUSE PROPRIETARY CLASS 3 F The satellite pulse observation technique has been proven effective in the initial classification of flaws and as the primary sizing method when synchronous satellite pulse responses are observed.(1)

In an effort to prove the capability for detection and accurate sizing of surface and near surface cracks, two 3 inch thick specimens containing cracks were included in the demonstration.

The blocks, identified as NATD #3 and NATD #15 (Figures C-10 and C-11), were scanned at various sensitivity levels. Generally, for both the 45 degree and 60 degree search units, a sufficient number of target secondary responses can be observed to permit sizing by Backward Scattering Tip Diffraction (BSTD) at calibration sensitivity +12dB (Fig. C-12, C-13 and C-14).

Adequate detection of both crack types is shown with the 45 degree transducer at calibration sensitivity +6dB (Fig. C-12 and C-13). Actual examination sensitivities in the automated scans ranged from +12 to +20dB above the calibration sensitivity.

C.2.2.3 AUTOMATED EXAMINATION TECHNIQUE The automated scanner was applied to each indication area disclosed in the manual examinations. The indications were bounded by the Level II operators manually first, and then a scan routine was programmed to sweep across the indication at C-11

WESTINGHOUSE PROPRIETARY CLASS 3 increments of 0.25 inch. All indication areas were scanned with at least 2 routines. The first routine was conducted with the same beam angle and scan direction as in the manual recordings.

System sensitivity was generally increased during the first scan to identify the indication and obtain as much sizing information as possible. Since the majority of the indications were detected with the 60 degree beam angle, the.target images were seen as single specular reflections embedded within the examination volume with a limited amount of sizing information available to the analyst. A second scan routine was performed over the target area with a complimentary beam angle, generally 45 degrees. Care was taken during the scanning operations to insure that both transducers started at the same point, insuring that the spatial coordinates for a given target could be verified.

The second scanning routines proved to be the most beneficial in terms of revealing discrete flaw secondary responses, which were useful for sizing, and for confirming the through-thickness locations of the embedded flaw indications by visualizing the flaws in the 5/8 node (skip response). In most cases, the girth weld inside diameter surface provided few direct geometric-type responses, even at +20dB, indicating that the surface is smoothly finished. For indications revealed originally with a 45 degree transducer, the necessary sizing and location confirmation information was obtained with complimentary 45 degree scans.

C-12

WESTINGHOUSE PROPRIETARY CLASS 3 As previously stated, the gain control was adjusted frequently during the scanning routines to reveal flaw secondary responses and perform through-wall sizing by amplitude independent methodologies. Unfortunately, this did not permit assessment of indication length, which can only be done in automated scanning when the gain control knob is left alone. Therefore, the length measurements from the 2.25 MHz manual detection data was used in the calculations for code acceptability. From a limited number of length estimates obtained in the automated data, our conclusion is-that the manual length data is realistic.

C-3 DISCUSSION OF RESULTS The results of the analysis of the UDRPS processed data indicate that all of the ultrasonic detections in Steam Generator "A" and "B" are from small embedded welding type discontinuities having relatively smooth, simple geometric shapes. The distribution of the discontinuities in and about the weld I.D. back chip area (Fig. C-15) is consistent with results obtained in identical weld configurations at other plants where these types of indications were confirmed by non-destructive and destructive testing. All indications were determined to be within the allowable limits of the ASME Code,Section XI, Table IWB-3511-1, 1980 Edition with Addenda through Winter 1981.

C-13

WESTINGHOUSE PROPRIETARY CLASS 3 This conclusion is supported primarily by the automated scan data.

The results of investigations on a total of 49 indications (32 in Generator "A" and 17 in Generator "B") show clearly that the flaw indications have small through wall dimensions (ranging from 0.1 to 0.27 inch), and are not connected with the inside surface. Further, in all the processed data, there is no evidence of cracking on the inside diameter surface. Scanning sensitivities were conducted in a range of 5 to 10 times Code sensitivity with the 45 degree beam, leading to a high probability of detection if cracking were present.

Summary tables.for the through thickness location methodologies are included as Figures C-16.1, C-16.2 and C-17.

The UDRPS processed color hard copy data for each indication is presented in Supplement 1. The indication assessment summaries preceed the UDRPS color hardcopy for each weld. Note that several indications thought originally to be unique and separate were confirmed in the processed data as being multiple recordings of the same indication. Where this is the case, only one evaluation was conducted, as referenced in the comments section of the indication assessment sheets.

For a discussion of the evaluation technique supporting the aforementioned conclusions, three examples from the supplement will be described in further detail. These include an example of a small rounded volumetric indication judged to be typical for the majority C-14

WESTINGHOUSE PROPRIETARY CLASS 3 of the indications (Indication #60-B Fig. B-5), a small embedded flaw close to the I.D. surface which proved to be the most difficult interpretation (Indication 45-1, Fig. A-2), and finally, an embedded flaw judged to have planar characteristics in the processed data (Indication #60-10 Fig. A-16).

For indication #60-B (Fig. B-5), a single embedded target is identified on the 60 degree transducer sweep. The lines which are transcribed on the plot for weld centerline and I.D. surface are points calculated.and interpolated directly from the display. Peak reflectivity along the target line is represented by a black dot, drawn in later to identify the target through thickness location. As is the case in most 60 degree scans, very little information is revealed in the scan sweep other than the specular or direct response from the target. As noted previously, the I.D. surface did not provide a great deal of geometric shadowing, even at higher gain settings, as evidenced by the fact that no geometry is noted on the 60 degree scan sweep, conducted at approximately 2x calibration sensitivity. The 45 degree hi-resolution scan (figure insert) of the same indication reveals more features of the indication for sizing (SPOT), and provides proof that the indication is embedded and isolated from the backwall by the strong 5/8 node response with no flaw targets identified along the I.D. surface. If indication 60-B was connected to the I.D. surface, both the half-vee and 5/8 node C-15

WESTINGHOUSE PROPRIETARY CLASS 3 targets would probably be connected, as one continuous target image intersecting the I.D. surface as shown in the 45 degree scan sweep over the cracked specimen (Fig. C-12).

Indication 45-1 (Fig. A-2 and A-3) is closer to the I.D. surface than any of the other 48 indications. The indication was originally recorded manually in 1988 from scan direction 5, and was confirmed as having identical dimensions in 1991 (ref. sketch, Fig. A-3). In the automated scans, more quantitative information was available from scan direction.2. (Fig. A-2), where a strong indication of satellite pulse is observed on the UDRPS processed data sweeps. The I.D.

surface shape in the area of interest is drawn on the screen by connecting measurement points from direct I.D. reflections and by estimating the surface contour by interpolating between half-vee and 5/8 node responses from other low amplitude indications in the area of interest. The location of indication 45-1 was also confirmed by using a 5.0 MHz, dual element, 0 degree beam applied directly beneath the weld crown on scan surface 2 (Fig. A-3), where the indication is seen approximately 0.2 inches from the backwall surface. The straight beam indication has a smooth echodynamic response in both the X and Y scan planes, supporting the indication classification as volumetric.

C-16

WESTINGHOUSE PROPRIETARY CLASS 3 Indication 60-10 (Fig. A-16) is an embedded flaw providing a single specular response in the 60 degree scan. The weld centerline and I.D. surface are plotted directly from position data on the display.

In the 45 degree high resolution scan (figure insert), the indication is highly reflective in the 5/8 node, indicating a preferred orientation. A secondary response, asynchronous in nature, was judged by position data to be related to the primary response, and was seen for a duration of at least two scan sweeps. All the indication points were combined for the through-wall estimate. It should..be noted.that a 0 degree beam reflection was noted in the area of interest of indication 60-10, leading to the possibility that the indication could be volumetric with a complex shape rather than purely planar.

Using advanced sizing techniques, the through-wall estimates for the majority of indications is around 0.1 inch. Since all the indications in both generators are judged to be embedded, and are therefore, prior existing, the radiographs were again carefully reviewed by qualified personnel in an attempt to confirm the presence of the ultrasonic indications. From available information, it was determined that a panoramic technique was used, with the film wrapped around the outside diameter surface of the girth weld and the Cobalt source placed in the center of the generator. Considering the radiographic quality measure of 2-2T, with a 4.1 inch weld section C-17

WESTINGHOUSE PROPRIETARY CLASS 3 thickness, the maximum defect detectability would be around 0.080 inches. Since the majority of indications lie in a band separated from the film by a distance greater than 3 inches, it is not unreasonable to assume that detection would be borderline.

Another possibility considered is whether or not the U.T. indication locations correlate with fabrication repair areas. The results of this investigation, conducted by the Kewaunee Plant technical staff, is presented as Supplement 2. Results indicate that there is no general.correlation to be made between the two.

C-4 CONCLUSIONS From the results of the automated and manual test data, the following conclusions are offered:

1. All of the indications detected in the Kewaunee Steam Generator "A" and "B" upper shell to cone welds are the result of prior existing welding type discontinuities.

2. Indication bounding measurements are within the allowable limits of the ASME Code,Section XI, Table IWB-3511-1.

3. The indications are embedded, and generally located, within an inch of the inside diameter surface.

C-18

WESTINGHOUSE PROPRIETARY CLASS 3

4. There is no evidence that the indications originate from or extend to the weld inside diameter surface.

5. Discontinuities remaining in the weld after fabrication were below the detection limits of the radiographic examination.

C-19

WESTINGHOUSE PROPRIETARY CLASS 3 REFERENCES (1) Kurek, D., "Alternative Sizing Techniques for Embedded Flaws",

Westinghouse Electric Corporation, 1990, WCAP-12691.

C-20

SURFACE/EMBEDDED FLAW DEMARCATION LINE, BEGINNING WITH 1980 CODE 0.13 ,r~fl? -"IEMBEDDED FLAW I a

rI.Et -CONFIGURATION

.. L t 0.12 ABOVE THIS LINE ARE NOT ALLOWABLE 0.11 + 2V&~. L+ LT+3

.31 H1 C4. +/-L 'ri w ir 0.10 i-4 ztMF 1

it 44-

1. dv' trir.*

,ILI TrJ-

-.. 1 0.09 0.08

-J 23 U 0.07 32 0S.. M44 2.9

<aI-.

C..L 0.06 oJ aC ifiw w 11 21 0.05 22 IG E 0.04 Is 2.0

,24 0.03

3 0.02 (0 3

0.01 9 Westinghouse 1987 0

0 0.05 0.10 0.15 0.20 0.25 DISTANCE FROM SURFACE (4)

NDIATION IAT a& Ant s At 1 45-1 0.21 .027 0.01 .030 2 45-2 0.3. .038 0.36 .131 3 45-3 0.18 .023 0.98 .274 4 45-4 0.12 .015 0.8 .221 45-5 0.19 .024 1.47 1.53

.396

.409 6 45-6 0.17 022 7 45-7 0.23 .029 1.16 .323 8 45-8 0.24 .031 0.99 .284 9 45-9 0.12 .015 0.97 .264 10 4 5-10 0.24 .032 0.95 .285 11 45-11 0.43 .054 0.95 .295 12 60-1 0.39 .050 0.73 .237 13 60-2 0.52 .066 0.46 .182 14 60-3 0.46 .058 0.69 .232 15 60-4 0.59 .074 0.52 .206 16 60-5 0.35 .044 0.86 .261 17 60-6 0.29 .037 0.69 .211 18 60-7 0.31 .039 1.26 .357 19 60-8 0.40 .051 0.44 .162 20 60-9 0.29 .037 0.92 .270 21 60-10 0.41 .052 0.97 .297 22 60-11 0.29 .049 0.94 .287 23 60-12 0.55 .070 0.83 .280 24 60-13 0.25 .032 0.98 .283 25 60-14 0.38 .049 0.68 .223 26 60-15 0.42 054 0.7 .233 27 60-16 0.42 *053 0.73 .237 28 60-17 0.48 .061 0.67 .230 29 60-18 0.53 .067 0.76 .259 30 60-19 0.29 .037 0.81 .241 31 60-20 0.59 .079 0.42 .192 32 60-21 0.54 .068 0.76 .261 FIGURE C-1.1 FRACTURE ANALYSIS RESULTS FOR INDICATIONS FOUND IN THE KEWAUNEE UNIT 1 STEAM GENERATOR A WELD 1-5 (5.0 MHz DATA)

10, YEARS LEGEND 20,YEARS A - The 10, 20, 30 year 30,YEARS acceptable flaw limits.

B - Within this zone, the surface flaw is-acceptable 18 by ASME Code analytical criteria in IWB-3600.

16 C - ASME Code allowable since 1983 Winter Addendum.

14 D - ASME Code allowable prior 12 to 1983 Winter Addendum.

I CL W 10 0

8 U.

6 4

D 2

0 0 0.1 0.2 0.3 0.4 0.5 DWestinghouse 1987 FLAW SHAPE (a/t)

INDICATION 45-1 ( REF. FIG. C-1.1 FIGURE C-1.2 FRACTURE ANALYSIS RESULTS FOR INDICATIONS CLASSIFIABLE AS INSIDE SURFACE FLAWS IN THE UPPER SHELL TO CONE REGION, KEWAUNEE UNIT 1 STEAM GENERATOR A

SURFACE/EMBEDDED FLAW DEMARCATION LINE. BEGINNING WITH 1980 CODE 0.13

: MEBEDDE D FLAW

- z &CONFIGURATION  :

0.12 0.11 REGION MUSTNOTALLOABE ti~-p 44-r4 0.10 d~: Tilt i t"t FLW WI T tt 0.09 FI.A VVE sunn ri.i~l -- -' 4'a TISLIEHAR I' FLAWSEINHTHISINE ARE REGIONALLOWABLE 0.08 0.07 0.06 SURFACEACE .

LL 0j FLW I HI .- -,

0.05 0.04

+/-4.

0.03 ItI -tt 0.02 0.01 0 © Westinghouse 1987 0 0.05 0.10 0.15 .0.20 0.25 DISTANCE FROM SURFACE (a)

INDICATION # ZA 1 45-A 0.27 .035 0.98 .286 2 45-B 0.14 .018 1.10 .300 3 45-C 0.16 .021 0.95 .264 4 45-D 0.12 .015 0.57 .162 5 45-E 0.17 .022 0.69 .199 6 45-F 0.31 .040 0.22 .096 7 60-A 0.47 .060 0.75 .253 8 60-B 0.40 .051 0.55 .192 9 60-C 0.55 .071 0.57 .217 10 60-D 0.58 .074 0.49 .200 11 60-E 0.39 .050 0.78 .250 12 60-F 0.35 .045 1.80 .306 13 60-G 0.28 .036 2.3 .636 14 60-H 0.35 .045 0.4 .147 15 60-I 0.46 .058 0.55 .197 16 60-J 0.31 .039 0.86 .257 17 60-K 0.56 .072 0.81 .279 FIGURE C-2 .- FRACTURE ANALYSIS RESULTS FOR INDICATIONS FOUND IN THE KEWAUNEE UNIT 1 STEAM GENERATOR "B" WELD 2-5 ( 5.0 MHz DATA )

48/45/10930-1 SURFACE/EMBEDDED FLAW DEMARCATION LINE, BEGINNING WITH 1980 CODE 0.13 trrEMBEEDDED FLAW:"' a id::-CONFIG3URATION -FLAWS WITH t 0.12 ABOVE THIS LINE ARE

++

NOT ALLOWABLE 0.11 0.10 0.09 0.08 LL 0.07 Note: Multiple recordings judged to be from the same indication 0 were reduced to one data plot.

(ref. Attachment 1) 0.06 -15 0

W_

U. 0.05

_J 0.04 ~

-~~ -- -- -'

FLAWS~ INTI 2.

0.03 0.02 3 %7- 19-21 --2 0.01 I2 0 O Westinghouse 1987 0 0.05 0.10 0.15 0.20 0.25 DISTANCE FROM SURFACE (a)

STEAM GENERATOR A - WELD 1-5 PLOT# INDICATION # PLOT# INDICATION # PLOT INDICATION #

1 45-1 9 60-4 17 60-12 2 45-5 10 60-5 18 60-13 3 45-8 11 60-6 19 60-14 4 45-9 12 60-7 20 60-15 5 45-10 13 60-8 21 60-16 6 60-1 14 60-9 22 60-17 7 60-2 15 60-16 23 60-19 8 60-3 16 60-11 24 60-20 25 60-21 FIGURE C-3 FRACTURE ANALYSIS RESULTS FOR INDICATIONS FOUND IN THE KEWAUNEE UNIT 1 STEAM GENERATOR A, WELD 1-5 ( UDRPS SIZING DATA)

48/45/10930-1 SURFACE/EMBEDDED FLAW DEMARCATION LINE. BEGINNING WITH 1980 CODE 0.13 2V EMBEDDED FLAWL

-FLAWS WITH t T12 CONFIGURATION 0.12 ABOVE THIS LINE ARE NOT ALLOWABLE 0.11 0.10 0.09 7:::

0.08 oLL 0.07 Note: Multiple recordings judged U

to be from the same indication 0 were reduced to one data plot.

0.06 REGION MUST BE:

(ref. Attachment 1)

~CONSIDERD . ..

L 0.05

-J 4 0.04 SURFACE1 E R~ -EL FLAWS -----

p-r

7 T7 .....

0.03 0.02

~~j~j::4:_ Al

.~ ti~'~ . .- 12~

4 -S- 9 -1ci~3 K~iL~i~IV2iI f T I 0 LI.LI 1 T7 0 ~LLLLW.LA1 Il-7:7 Q Westinghouse 1987 0 0.05 0.10 0.15 0.20 0.25 DISTANCE FROM SURFACE ( )

STEAM GENERATOR B - WELD 2-5 PLOT INDICATION # PLOT INDICATION #

1 45-E 8 60-F 2 45-F 9 60-G 3 60-A 10 60-H 4 60-B 11 60-I 5 60-C 12 60-J 6 60-D 13 60-K 7 60-E FIGURE C-4 FRACTURE ANALYSIS RESULTS FOR INDICATIONS FOUND IN THE KEWAUNEE UNIT 1 STEAM GENERATOR B, WELD 2-5 ( UDRPS SIZING DATA)

0

SUMMARY

OF SIZING RESULTS - CONVENTIONAL AND ADVANCED TECHNIQUES Found in the KNPP Unit 1 S/G A Weld 1-5 1988 1991 1991 2.25 MHz (1) 5.0 MHz (2) 2.25 MHz (1) 5.0 MHz (2) UDRPS Ir I I I I "l "

BEAM ANGLE Peak Peak IND. NO. (DEGREES) 1"2aII I s Sweep 112an s I s Sweep "12al I S 12a"(3) 1 (4) s (3) 1 45 0.35 1.25 0.12 6.8 0.14 0.35 0.08 0.29 0.80 0.06 6.8 0.21 1.13 0.01 0.10 1 0.80 0.17 2 45 0.43 0.85 0.51 8.3 0.30 0.50 1.04 0.24 0.60 0.61 8.3 0.30 0.31 0.36 Ref. 60-4 3 45 0.39 0.85 0.39 5.6 0.37 0.95 0.28 0.21 0.50 0.82 5.7 0.18 0.56 0.98 Ref. 60-3 4 45 0.12 0.60 0.59 5.3 --- - (6) 0.81 5.3 0.12 0.44 0.80 Spot 5 45 0.23 --- 0.67 5.0 --- --- --- 0.27 0.60 1.31 5.0 0.19 0.63 1.47 0.18 1i0.60 0.75 6 45 0.87 5.0 --- (6) 1.47 5.0 0.17 0.31 1.53 Ref. 60-7 7 45 0.23 1.40 0.53 5.4 0.31 1.00 0.59 0.34 0.50 1.12 5.4 0.23 0.31 1.16 Ref. 60-11 8 45 (7) (7) 0.30 0.80 0.30 5.6 0.24 0.56 0.99 0.10 0.80 0.95 9 45 (7) (7) 0.12 0.55 0.80 5.0 0.12 0.50 0.97 0.11 0.55 0.24 10 45 (7) (7) 0.18 0.30 0.35 5.7 0.24 0.38 0.95 0.204 0.30 0.46 11 45 (7) (7) 0.36 0.70 1.40 8.5 0.43 0.63 0.95 Ref. 60-11 1 60 0.76 0.75 0.00 5.9 0.23 10.75 i0.35 0.92 0.75 0.55 5.9 0.39 0.75 0.73 ] 0.75 0.45 (1) 2.25 MHz data recorded at 50% DAC. (6) Judged to be a "SPOT" reflector having no measurable length or "2a" dimension.

(2) 5.0 MHz data recorded at -6dB (half maximum amplitude). (7) Indications 45-8,9,10, and 11 and 60-13 through 21 (1991 data)

(3) Measurements using advanced sizing techniques. are previously undisclosed volumetric type indications.

(4) Lengths taken from 1991 2.25 MHz manual data.

(5) Depth varies along Length. E: AlL 1988 indications were located and measured in 1991 and were judged to be similar in 1SGA1A l amplitude and size. Rev. 0 FIGURE C-5.1 Summary of Recorded Indication Sizes Conventional and Advanced Sizing Techniques Steam Generator A - Weld 1-5

0

SUMMARY

OF SIZING RESULTS - CONVENTIONAL AND ADVANCED TECHNIQUES Found in the KNPP Unit 1 S/G A Weld 1-5 1988 1991 1991 2.25 MHz (1) 5.0 MHz (2) 2.25 MHz (1) 5.0 MHz (2) UDRPS BEAM ANGLE Peak Peak IND. NO. (DEGREES) " a t s Sweep "2a" I s "2a" t s Sweep "2a" ( s 12a"(3) 1 (4) s (3) 2 60 0.46 1.00 0.35 5.2 0.52 1.20 0.26 0.46 1.00 0.58 5.2 0.52 0.60 0.46 0.11 1.00 0.75 3 60 0.76 1.55 0.12 5.9 0.35 1.55 0.12 0.72 1.55 0.84 5.9 0.46 0.50 0.69 0.10 1.55 0.49 4 60 0.52 0.90 0.41 5.3 0.52 1.50 0.26 0.53 0.90 0.42 5.3 0.59 0.75 0.52 0.10 0.90 0.75 5 60 0.41 0.75 0.34 4.9 0.30 2.10 0.65 0.41 0.75 0.66 4.9 0.35 0.65 0.86 0.19 0.75 0.72 6 60 0.64 1.00 0.41 5.0 0.47 1.40 0.41 0.65 1.00 0.63 5.0 0.29 0.35 0.69 0.10 1.00 0.73 7 60 0.47 0.70 0.64 5.0 0.30 2.90 0.69 0.55 0.70 1.30 5.0 0.31 0.50 1.26 0.10 0.70 1.68 8 60 0.42 12.10 0.07 6.0 0.47 12.10 0.20 1.19 12.10 0.21 6.0 0.40 7.95 0.44 0.14 12.10 0.45/

0.85 (5) 9 60 0.29 13.25 0.82 4.6 0.35 13.25 0.69 0.29 13.25 0.99 4.6 0.29 12.65 0.92 0.10 13.25 0.35/

1.10 (5) 10 60 0.46 1.00 0.65 4.7 0.41 1.00 0.71 0.46 1.00 0.92 4.7 0.41 1.80 0.97 0.17 1.00 0.93 11 60 0.46 1.80 0.47 5.4 0.47 1.80 0.37 0.53 1.80 1.05 5.4 0.39 1.20 0.94 0.10 1.80 1.25 12 60 0.58 1.80 0.23 5.5 0.37 2.80 0.30 0.62 1.80 0.79 5.5 0.55 1.80 0.83 0.10 1.80 1.15 (1) 2.25 MHz data recorded at 50% DAC. (6) Judged to be a "SPOT" reflector having no measurable length or "2a" dimension.

(2) 5.0 MHz data recorded at -6dB (half maximum amplitude). (7) Indications 45-8,9,10, and 11 and 60-13 through 21 (1991 data)

(3) Measurements using advanced sizing techniques. are previously undisclosed volumetric type indications.

(4) Lengths taken from 1991 2.25 MHz manual data.

Depth varies along length. similar in were located and measured in 1991 and were judged to be (5) 1SGA1B E All 1988 indications

lamplitude and size. Rev. 0 FIGURE C-5.2 Summary of Recorded Indication Sizes Conventional and Advanced Sizing Techniques Steam Generator A - Weld 1-5 (Continued)

SUMMARY

OF SIZING RESULTS - CONVENTIONAL AND ADVANCED TECHNIQUES Found in the KNPP Unit 1 S/G A Weld 1-5 1988 1991 1991 2.25 MHz (1) 5.0 MHz (2) 2.25 MHz (1) 5.0 MHz (2) UDRPS BEAM ANGLE Peak Peak IND. NO. (DEGREES) "2a I I s Sweep I s (

s Sweep s 12a"(3) 1 (4) s (3) 13 60 (7) (7) 0.54 1.10 [0.45 5.8 0.25 11.30 0.98 0.10 1.10 1.00 14 60 (7) (7) 0.36 0.80 0.50 5.6 0.38 0.50 0.68 0.10 0.80 1.00 15 60 (7) (7) 0.54 0.90 0.50 5.4 0.42 0.70 0.70 0.10 0.90 0.89 16 60 (7) (7) 0.30 0.80 0.70 5.7 0.42 0.50 0.73 0.10 0.80 1.05 17 60 (7) (7) 0.60 0.70 0.60 5.6 0.48 0.60 0.67 0.27 0.70 0.83 18 60 (7) (7) 0.54 3.80 0.60 5.4 0.53 2.70 0.76 Ref. 60-19 19 60 (7) (7) 0.48 1.60 0.60 5.4 0.29 0.35 0.81 0.17 1.60 0.61 20 60 (7) (7) 0.54 1.20 0.50 5.3 0.59 1.15 0.42 0.10 1.20 1.10 21 60 (7) (7). 0.30 1.00 0.60 5.3 0.54 0.80 0.76 0.14 1.00 0.75 (1) 2.25 MHz data recorded at 50% DAC. (6) Judged to be a "SPOT" reflector having no measurable length or "2a" dimension.

(2) 5.0 MHz data recorded at -6cB (half maximum amplitude). (7) Indications 45-8,9,10, and 11 and 60-13 through 21 (1991 data)

(3) Measurements using advanced sizing techniques. are previously undisclosed volumetric type indications.

(4) Lengths taken from 1991 2.25 MHz manual data.

(5) Depth varies along length.

[E All 1988 indications

lamplitude and size.

were Located and measured in1991 and were judged to be similar in 1SGA1C Rev. 0 FIGURE C-5.3 Summary of Recorded Indication Sizes Conventional and Advanced Sizing Techniques Steam Generator A - Weld 1-5 (Continued)

SUMMARY

OF SIZING RESULTS - CONVENTIONAL AND ADVANCED TECHNIQUES

. Found in the KNPP Unit 1 S/G B Weld 2-5 1987 1991 1991 2.25 MHz (1) 5.0 MHz (5) 2.25 MHz (1) 5.0 MHz (2)

Peak Peak IND. NO. BEAM ANGLE "2a" I s Sweep u2a" I S "2a" I s Sweep 1"2a" I s A 45 0.61 0.60 0.37 5.6 --- --- --- 0.35 1.00 0.69 5.6 0.27 0.38 0.98 B 45 0.43 0.10 0.45 5.9 --- --- --- 0.44 0.60 0.61 5.9 0.14 0.25 1.10 C 45 (7) (7) 0.35 (8) 0.50 5.5 0.16 0.50 0.95 D 45 (7) (7) 0.18 0.80 0.50 5.6 0.12 0.75 0.57 E 45 (7) (7) 0.12 0.25 0.60 5.4 0.17 0.38 0.69 F 45 (7) (7) 0.29 0.80 1.20 8.1 0.31 0.88 0.22 A 60 0.69 1.20 0.67 5.5 0.37 0.75 1.02 0.66 1.20 0.76 5.5 0.47 0.60 0.75 B 60 0.81 1.50 0.24 6.0 0.35 0.50 0.75 0.27 1.50 0.46 6.2 0.40 1.40 0.55 C 60 0.65 0.90 0.63 5.4 --- -- --- (8) 0.95 5.4 0.55 0.70 0.57 D 60 0.55 1.10 0.53 5.3 0.28 1.10 0.69 0.36 0.80 0.68 5.4 0.58 0.90 0.49 E 60 0.46 1.50 0.37 5.5 0.65 0.63 0.75 0.78 1.10 0.69 5.5 0.39 0.50 0.78 F 60 0.61 3.10 1.67 3.9 0.26 1.00 1.93 0.53 2.40 1.81 3.9 0.35 1.30 1.80 (1) 2.25 MHz data recorded at 50% DAC. (6) Value taken from inside surface, indication is 1.53" from outside surface.

(2) 5.0 MHz data recorded at -6dB (half maximum amplitude). (7) Indications 45-C,D,E, and F; and 60-H,I,J, and K are previously undisclosed (3) Measurements using advanced sizing techniques. volumetric indications.

(4) Lengths taken from 1991 2.25 MHz manual data. (8) No measurable dimensions.

(5) 5.0 MHz data recorded at 50% DAC. 1SGB1A Rev. 0 FIGURE C-6.1 Summary of Recorded Indication Sizes Conventional and Advanced Sizing Techniques Steam Generator B - Weld 2-5

0

SUMMARY

OF SIZING RESULTS - CONVENTIONAL AND ADVANCED TECHNIQUES Found in the KNPP Unit 1 S/G B Weld 2-5 1987 1991 1991 2.25 MHz (1) 5.0 MHz (5) 2.25 MHz (1) 5.0 MHz (2) UDRPS Peak Peak IND. NO. BEAM ANGLE " I2a" s Sweep "2a" t s u2au t s Sweep "2a" I s 12a"(3) 1 (4) s (3)

G 60 0.41 1.80 1.63 3.1 0.35 2.75 1.79 (8) 2.03 3.2 0.28 0.80 2.34 (6) 0.10 2.30 2.00 I.-

H 60 (7) (7) 0.36 0.90 0.50 5.7 0.35 2.00 0.40 0.10 0.90 0.30 I 60 (7) (7) 0.35 0.60 0.50 6.0 0.46 0.40 0.55 0.15 0.60 0.28 J 60 (7) (7) 0.23 0.80 0.70 5.4 0.31 0.80 0.86 0.10 0.80 0.80 K 60 (7) (7) 0.17 1.20 0.65 5.4 0.56 0.30 0.81 0.10 1.20 1.10 (1) 2.25 MHz data recorded at 50% DAC. (6) Value taken from inside surface, indication is 1.53" from outside surface.

(2) 5.0 MHz data recorded at -6dB (half maximum amplitude). (7) Indications 45-C,D,E, and F; and 60-H,I,J, and K are previously undisclosed (3) Measurements using advanced sizing techniques. volumetric indications.

(4) Lengths taken from 1991 2.25 MHz manual data. (8) No measurable dimensions.

(5) 5.0 MHz data recorded at 50% DAC. 1SGB1B Rev. 0 FIGURE C-6.2 Summary of Recorded Indication Sizes Conventional and Advanced Sizing Techniques Steam Generator B - Weld 2-5 (Continued)

FIGURE C-7 UDRPS 450 Automated Scan Data 3.5" Thick Code Calibration Block (WPS-36) Side Drilled Hole Reflectors

FIGURE C-8 UDRPS 450 Automated Scan Data 3.5" Thick Code Calibration Block (WPS-36) Buttress Notch Reflection

FIGURE C-9 UDRPS 450 Automated Scan Data DSI Block - 3/64" Dia. Side Drilled Holes in Line

NOTES:

1. CRACK MECHANICALLY INDUCED IN LABORATORY.

2. BLOCK USED TO DEMONSTRATE UDRPS EOUIPMENT FOR CHARACTERIZATION/SIZING OF INDICATIONS IN SC GIRTH WELDS.

3. BLOCK PROVIDED. OWNED, AND RETAINED BY 2.880 DEMONSTRATION BLOCK #3 17"L X 2.88"H X 4"W REV. 1 APRIL 25. 1991 FIGURE C-10 NATD Block #3 Technique Qualification Specimen (Surface Connected Crack)

FIGURE C-11 NATD Block #15 Technique Qualification Specimen (Embedded Crack)

FIGURE C-12 NATD Block #3, 450 Scan at Sensitivity Levels +6dB and +12dB FIGURE C-13 NATD Block #15, 450 Scan at Sensitivity Levels +6dB and

+12dB

FIGURE C-14 NATD Block #15, -600 Scan at Sensitivity Level +12dB REF DIRECTION 5 15.50' TOP Steam Generator A Weld 1-5 Flaw Distribution REF DIRECTION 5

15. 50' TOP Steam Generator B Weld 2-5 Flaw Distribution FIGURE C-15 Flaw Distribution - Steam Generator A and B Upper Shell to Cone Welds Plots from UDRPS Data

Methods for Verifying the Through-Thickness Locations of Embedded Flaws Found in the KNPP Unit 1 S/G A Weld 1-5 I.D. SURFACE MANUAL 5.0 MHz "SHADOWING" FROM UDRPS COMPLIMENTARY INDICATION NO. STRAIGHT BEAM UDRPS HI-RES-SCANS ANGLE/5/8 NODE RESPONSE COMMENTS 45-1 x (1) X X 45-5 X X X 45-8 X X X 45-9 X --- X 45-10 X X X 60-1 X X X 60-2 X X X 60-3 X X X 60-4 X --- X 60-5 X --- X 60-6 X X X 60-7 X X X 60-8 X --- X 60-9 X X X (1) Indication located with 5.0 MHz dual element 0 degree transducer. Indication is 0.2 inches from I.D. surface having smooth echodynamic response in X and Y plane of transducer manipulation.

2SGA1A FIGURE C-16.1 Summary of Through-Thickness Location Techniques - Steam Generator A - Weld 1-5

Methods for Verifying the Through-Thickness Locations of Embedded FLaws Found in the KNPP Unit 1 S/G A WeLd 1-5 I.D. SURFACE MANUAL 5.0 MHz "SHADOWING" FROM UDRPS COMPLIMENTARY INDICATION NO. STRAIGHT BEAM UDRPS HI-RES-SCANS ANGLE/5/8 NODE RESPONSE COMMENTS 60-10 X X X 60-11 X X X 60-12 X X X 60-13 X X .x 60-14 X X X 60-15 X X X 60-16 X x x 60-17 X --- X 60-19 X X X 60-20 X --- x 60-21 X --- X FIGURE C-16.2 Summary of Through-Thickness Location Techniques - Steam Generator A - Weld 1-5 (Continued) 2SGA1B

Methods for Verifying the Through-Thickness Locations of Embedded Flaws Found in the KNPP Unit 1 S/G B Weld 2-5 I.D. SURFACE MANUAL 5.0 MHz "SHADOWING" FROM UDRPS COMPLIMENTARY INDICATION NO. STRAIGHT BEAM UDRPS HI-RES-SCANS ANGLE/5/8 NODE RESPONSE COMMENTS 45-E X X X 45-F X --- X 60-A X --- X 60-B X --- X 60-C X --- X 60-D X X X 60-E X --- X 60-F X --- --- Shallow - mid-wall 60-G X --- --- ShatLow - mid-walt 60-H X X X 60-1 x --- X 60-J X --- X 60-K X --- ---

FIGURE C-17 Summary of Through-Thickness Location Techniques - Steam Generator B - Weld 2-5 2SGB1A

PROPRIETARY INFORMATION NOTICE THE ATTACHED DOCUMENT CONTAINS OR IS CLAIMED TO CONTAIN PROPRIETARY INFORMATION AND SHOULD BE HANDLED AS NRC SENSITIVE UNCLASSIFIED INFORMATION.

IT SHOULD NOT BE DISCUSSED OR MADE AVAILABLE TO ANY PERSON NOT REQUIRING SUCH INFORMATION IN THE CONDUCT OF OFFICIAL BUSINESS AND SHOULD BE STORED, TRANSFERRED, AND DISPOSED OF BY EACH RECIPIENT IN A MANNER WHICH WILL ASSURE THAT ITS CONTENTS ARE NOT MADE AVAILABLE TO UNAUTHORIZED PERSONS.

COPY NO.

DOCKET NO.

CONTROL NO.

REPORT NO.

REC'D W/LTR DTD.

NRC Form 190 (10-89)

NRCM 2101 PROPRIETARY INFORMATION

WESTINGHOUSE PROPRIETARY CLASS 2 SUPPLEMENT #1 KEWAUNEE UNIT 1 STEAM GENERATOR UPPER SHELL TO CONE WELDS STEAM GENERATOR "A" WELD 1-5 STEAM GENERATOR "B" WELD 2-5 WITH INDICATION ASSESSMENT TABLES FIGURES A.1 THROUGH A.33 - STEAM GENERATOR A WELD 1-5 FIGURES B.1 THROUGH B-13 - STEAM GENERATOR B WELD 2-5

INDI ION ASSESSMENT PLANT 1AWAC6 page of WELD WELD5 -W sa &\kc) SSl-cc WPS- DESCRIPTION £teAI'V\C. SteAM C, e-lzcCt A-L ___p, cRorRIS%-loCGIRrN 0 0 0 o0 x 0 COMMENTS z390 L 4-1 t IWB - 3000 reference UDRPS File No.

kAble. / (ul3-3 S -I Sil 1 e80 V /3. C t 3.78 0.8 0.10 395 0.1N I S b .0di ?I- 2.I. ts 45

-2 - -

-- - Fo-- S,/ z//

%- --- REF. 60-3 fz SJZ.JiAs

' H 'a V Z. 19 3.ro S_ _a_ _ _ __e_ __,

,5 / V 29Z 3-10 0 s. \1 3 S.b iS e-31 32 1, Spo LCN&TH, IEF 60-7 Foi. .S/Z/tvA 9 - - 60 8 - 2-S 2-( 0.8 0.1 3-5 0ci \ Sb ,ot (.3 29 cc L__ _ _ar 4 5/5 3 3S 3 3 OS0.11 50z1\ Sub 1 13 211 eF bo -S FOg Si2Nn Aee 41V 31Z6 3.39 o. c.20 3Z0 S5 \ Stb .51 i 5.\ e .7 NOTE: "2a" dimensions calculated at less than 0.1 inch were assessed at 0.1 inch.

Analyst .

Level -a- c/

FIGURE A-1.1 INDICATION ASSESSMENT TABLE STEAM GENERATOR A WELD 1-5

INDI ION ASSESSMENT PLANT page of 3 WELD i DESCRIPTION . A (Ierr lrAN jnoe T-r Celc o 0c 5 0 2 Lf 0..0a a . . O w x5 u "COMMENTS

- 00 S .*.0

.4 . 4 4 a 3 IWB - 3000 reference (n UDRPS File No.

- 3 7 3 q 015 S 3 85 oqs \

/2 ~ /c 2 Mj 3 \ . G0 1. 3 85 e-7 \_ _

\/ 3, 5O 3-3 I s \/ 3. 3 1-O\ 3 -,5 S 10 __3 2_7 5_ _75 V 2c1 31 075 0 3 85 07 -2 11 2 V Z.L 2\lC 6.7 01 31 z I s

.S 1 IS 13.Z.S 01. 3S 8 d jsO -I2 ,. Tej( ~~'i ~~i f J \V 31o 3 . H 35 &73 _

- - -7 I- - 7 3 3 9_i5 j 11-3 Soc, T Analys I Le 33091 cr-. / ;F 2 v.Analyst~-i 27 3-S 13 9 I 3 8 \s ilk \ez 2ee

-. 3:

T 7 __ __ __ _

05 00 28 c 11sto

__ \ 27 S 3;Cs 2 F5 S (-c, T 12 75 1 .5 l0.11 3-8 5 _0 2 _3__ __ _

6 ~~ ~- -138~

S/V 75 25 0 0 I1S

\. ~

0

~ 0L% 3 . 87ZtTD2 2 7 V 2t -,e S03F FIGURE A-1.2 INDICATION ASSESSMENT TABLE STEAM GENERATOR A WELD 1-5 (CONTINUED)

INDI*ION ASSESSMENT COMMENTS IWB - 3000 reference UDRPS File No.

ZIis C..'I 3i~s 2A IL.1-7 27S LE i I 3 ,IS j-u "'-1 FIGURE A-1.3 INDICATION ASSESSMENT TABLE STEAM GENERATOR A WELD 1-5 (CONTINUED)

FIGURE A-2 S.G. A INDICATION 45-1 45 DEGREE HI-RES SCANS REF DIRECTION 5 S.0 ON Manual Detection direction

. I-, I, 11//1 Geometric Reflections (UDRPS)

TOP FIGURE A-3 S.G. A INDICATION 45-1 DATA POINT PLOTS

kNPP SG A INDICATION 45-5 NE 45 DEGREE SCAN DIRECTION 2 12 110 1 7 YT 4 956 ZT 5 412 FILE: DA0542 63 45.00 DEGREE DUCER

-IN S EEP f6: YE 1I -,N Z8B 3 827 MPL 362 ASCAN 141 AMP 0 SCAN 6 FIGURE A-4 S.G. A INDICATION 45-5, 45 DEGREE HI-RES SCAN

FIGURE A-5.1 S.G. A, INDICATION 45-8, 45 DEGREE HI-RES SCAN FIGURE A-5.2 S.G. A, INDICATION 45-8, 45 DEGREE HI-RES SCAN FIGURE A-6 S.G. A, INDICATION 45-10, 45 DEGREE HI-RES SCAN 01 FIGURE A-7 S.G. A. INDICATION 60-1, 60 DEGREE SCAN WITH 45 DEGREE HI-RES SCAN (INSERT)

FIGURE A-8 S.G. A INDICATION 60-2, 60 DEGREE SCAN WITH 45 DEGREE HI-RES SCAN (INSERT)

FIGURE A-9 S.G. A INDICATION 60-3, 60 DEGREE SCAN WITH 45 DEGREE HI-RES SCAN (INSERT)

FIGURE A-10.1 S.G. A INDICATION 60-4, 60 DEGREE SCAN FIGURE A-10.2 S.G. A INDICATION 60-4, 45 DEGREE HI-RES SCAN FIGURE A-11.1 S.G. A INDICATION 60-5, 60 DEGREE SCAN FIGURE A-11.2 S.G. A INDICATION 60-5, 45 DEGREE HI-RES SCAN FIGURE A-12 S.G. A INDICATION 60-6, 60 DEGREE SCAN WITH 45 DEGREE HI-RES SCAN (INSERT)

-IN X P = 008 ZT = 7.903 FILE: DA0762 83 60.00 DEGREE OUCER

-IN, SWEEP 4 = 1.165 Z8 = 3.953 MPL 267 ASCAN 7 AMP 8 SCAN 4 FIGURE A-13 S.G. A INDICATION 60-7, 60 DEGREE SCAN WITH 45 DEGREE HI-RES SCAN (INSERT)

I Ii~

4 FIGURE A-14 S.G. A INDICATION 60-8, 60 DEGREE SCAN WITH 60 DEGREE HI-RES SCAN (INSERT)

FIGURE A-15 S.G. A INDICATION 60-9, 60 DEGREE SCAN WITH 45 DEGREE HI-RES SCAN (INSERT)

FIGURE A-16 S.G. A INDICATION 60-10, 60 DEGREE SCAN WITH 45 DEGREE HI-RES SCAN (INSERT)

FIGURE A-17 S.G. A INDICATION 60-11, 60 DEGREE SCAN WITH 45 DEGREE HI-RES SCAN (INSERT)

FIGURE A-18.1 S.G. A INDICATION 60-12, 60 DEGREE SCAN 0A FIGURE A-18.2 S.G. A INDICATION 60-12, 45 pEGREE HI-RES SCAN

FIGURE A-19 S.G. A INDICATION 60-13, 60 DEGREE SCAN WITH 45 DEGREE HI-RES SCAN (INSERT)

DA1401 SWEEP FIGURE A-20 S.G. A INDICATION 60-14, 45 DEGREE SCAN WITH 45 DEGREE HI-RES SCAN (INSERT)

FIGURE A-21.1 S.G. A INDICATION 60-15, 60 DEGREE SCAN FIGURE A-21.2 S.G. A INDICATION 60-15, 45 DEGREE HI-RES SCAN FIGURE A-22 S.G. A INDICATION 60-16, 60 DEGREE SCAN WITH 45 DEGREE HI-RES SCAN (INSERT)

FIGURE A-23 S.G. A INDICATION 60-17, 60 DEGREE SCAN WITH 45 DEGREE HI-RES SCAN (INSERT)

1 0 VT = 6 611 2T = 7.688 FILE 0A1965 83 60 00 DEGREE DUCER

-IN SWEEP 5 B, = -8046 Z = 3.845 MPL 260 ASCAN 298 AMP 12 SCAN 5 FIGURE A-24 S.G. A INDICATION 60-19, 60 DEGREE SCAN WITH 45 DEGREE HI-RES SCAN (INSERT)

FIGURE A-25 S.G. A INDICATION 60-20, 60 DEGREE SCAN WITH 45 DEGREE HI-RES SCAN (INSERT)

ISS FIGURE A-26 S.G. A INDICATION 60-21, 60 DEGREE SCAN WITH 45 DEGREE HI-RES SCAN (INSERT)

INDI ON ASSESSMENT 0 PLANT. eve u. N page of WELD ( SG -Lu (o ) DESCRIPTION c o0 0 Z '- COMMENTS

..ou

.. c Al 3 3IWB - 3000 reference 6 1 > 3UDRPS File No'.

a/ - - - - I- - -K (-3-25 h. X 1 L5 NOTE: "2a" dimensions calculated at less than 0.1 inch were assessed --- O - o at 0.1 inch. SPI - /V0

___ __ _ _ _ _ - - -(-R & - f1-6et ur cLa ". Nr 95 0 6/L 'S 21 (

L5 V 27S 3 5 CI LEo; \ ub .20 . 3 46vb 3V15 3 55 0.9 I- -5 I -0 L 2 ro 1/ 2l 1-.. 0,13 2 S .Z LA-L arc)

_h275 625 0,1 73 _ 2_ _ __a_

Ic 2 i 3. 3.\ G.j OdI G04 .or blA 23 (8 L ,t ,io 3,1 77 317 6. 3 23 3.0 .srD

2. V I 2 l 1 12 5 oS 52- 2 E Star L , V I 5 Z 2.1 6t2 2 I - 1 2 2.-7 55I 5 br__ _Z 2-. _k 7V 3S 3(. 0 1 O 3 .0 2- 24 /1_ _ a

-1 v 17 3v 2. o -- ig - . 7 1.i 3.0 IA 2__

Analyst Level Y Date' FIGURE B-1.1 INDICATION ASSESSMENT TABLE, STEAM GENERATOR "B", WELD 2-5

INDI ION ASSESSMENT PLANT CA 't page of Q.

WELD - Ltd ) DESCRIPTION C (. 0 ( 1 TInco r I-s U c Z aL COMMENTS 5 a - 8 08 8 0

- - - *U* - = .0 o

U .12

• C350 > -1 Ili 3

iUu IWB - 3000 reference UDRPS File 4I )

.No. 1920 c Analyst Level -*7- .te Da FIGURE B.1.2 INDICATION ASSESSMENT TABLE, STEAM GENERATOR "B", WELD 2-5 (continued)

FIGURE B-2 S.G. B INDICATION 45-E, 45 DEGREE HI-RES SCAN FIGURE B-3 S.G. B INDICATION 45-F, 45 DEGREE HI-RES SCAN FIGURE B-4.1 S.G. B INDICATION 60-A, 60 DEGREE SCAN FIGURE B-4.2 S.G. B INDICATION 60-A 45 DEGREE HI-RES SCAN F 58 N 0DE GC-B kD 4

WELD C L 125

/ KN P SG 6 INSICATION 60-B 60 DEGREE SCN DIRECTION 2 I.-'

1 5 %T 7 964 ZT 7.933 FILE: DBO62 B3 60 00 DEGREE DUCER

~fl S~iEEP ?YB I 094 Z8 3.968 M L 268 ASCAN 359 AMP I SCAN FIGURE B-5 S.G. B INDICATION 60-8, 60 SCAN WITH 45 DEGREE HI-RES SCAN (INSERT)

FIGURE B-6 S.G. B INDICATION 60-C, 60 DECREE SCAN WITH 45 DEGREE HI-RES SCAN (INSERT)

01 FIGURE B-7 S.G. B INDICATION 60-D, 60 DEG E SCAN WITH 45 DEGREE HI-RES SCAN (INSERT)

FIGURE B-8 S.G. B INDICATION 60-E, 45 DEGREE HI-RES SCAN FIGURE B-9 S.G. B INDICATION 60-F,' 45 DEGREE HI-RES SCAN FIGURE B-10 S.G. B INDICATION 60-G, 60 DEGREE iCAN WITH 60 DEGREE HI-RES SCAN (INSERT)

FIGURE B-11 S.G. B INDICATION 60-H, 60 DECREE SCAN WITH 45 DEGREE HI-RES SCAN (INSERT)

FIGURE B-12 S.G. B INDICATION 60-1, 45 D;GREE HIGH-RES SCAN i

FIGURE B-13 S.G. B INDICATION 60-J, 15 DEGREE HIGH-RES SCAN FIGURE B-14 S.G. B INDICATION 69-K, 45 DEGREE HI-RES SCAN 1.

WESTINGHOUSE PROPRIETARY CLASS 2 SUPPLEMENT 2 KEWAUNEE UNIT 1 STEAM GENERATOR UPPER SHELL TO CONE WELDS STEAM GENERATOR "A" WELD 1-5 STEAM GENERATOR "B" WELD 2-5 COMPARISON OF THE LOCATION OF FABRICATION REPAIRS WITH ULTRASONIC TEST DETECTION DATA ( 1991 )

FIGURE 1 STEAM GENERATOR "A" WELD 1-5 FIGURE 2 STEAM GENERATOR "B" WELD 2-5

'A

COMPARISON OF FABRICATION RADIOGRAPHS AND 1991 XI INSPECTION RESULTS FOUND IN KNPP UNIT 1 SG "A" WELD 1-5 36 37 38 39 40 35 34 42 33 32 CENTERLINE OF 35 /

28 36 FW INLET NOZZLE 31 262 30 25 29 '24 23 SG // // 1 OUTER NUMBERS CORRESPOND TO UT REFERENCE POINTS.

28 2

-22 27- 3 21 LONGITUDINAL WELD #2

.20 (LONG. SEAM "H")

INNER NUMBERS CORRESPOND 25 -19 TO RT REFERENCE POINTS. 5

.18 6 3

17 INDICATES THAT A FABRICATION REPAIR 23 7 RADIOGRAPH EXITS FOR THIS LOCATION 116 22 5 15 6 <3 INDICATES LOCATION OF UT INDICATIONS 21 14 7 DETECTED WITH 2.25 MHZ TRANSDUCER 13 20 i 12 CHART4 FIGURE 1 COMPARISON OF REPAIR RADIOGRAPHS AND STEAM GENERATOR A WELD 1-5

0 COMPARISON OF FABRICATION RADIOGRAPHS AND 1991 SECTION XI INSPECTION RESULTS FOUND IN KNPP UNIT 1 SC "B" WELD 2-5 36 37 38 39 40 35 1- I 1 40 41 34 33 43 32 33 34 32 CENTERLINE OF

\28 35 36 FW INLET NOZZLE 31 27 30 25 38 OUTER NUMBERS CORRESPOND 29 24 SC " / 1 TO UT REFERENCE POINTS.

28 23 41 2 22 27 42 -3 21 LONGITUDINAL WELD #2 4 (LONG. SEAM "H")

-5 2

24 18 6 3

23 17A INDICATES THAT A FABRICATION REPAIR RADIOGRAPH EXISTS FOR THIS LOCATION 2 2 5 211651TO RT RE E E C POINTS. 6 0 222 1610 INDICATES LOCATION OF UT INDICATIONS DETECTED WITH 2.25 MHZ TRANSDUCER CHART4 FIGURE 2 COMPARISON OF REPAIR RADIOGRAPHS AN:) UT INDICATION LOCATIONS STEAM GENERATOR B WELD 2-5