ML20196L203

Non-proprietary Rev 0 to BAW-2346NP, Alternate Repair Criteria for Tube End Cracking in Tube-to-Tubesheet Roll Joint of Once-Through Sgs

ML20196L203
Person / Time
Site:	Crystal River
Issue date:	05/19/1999
From:	Jonathan Brown, Burgess J, Costa D FRAMATOME
To:
Shared Package
ML20196L198	List: 3F0599-22, Forwards non-proprietary Version of B&Wog Topical Rept BAW-2346NP, Alternate Repair Criteria for Tube End Cracking in Tube-to-Tubesheet Roll Joint of Once-Through Sgs ML20196L203
References
BAW-2346NP, BAW-2346NP-R, BAW-2346NP-R00, NUDOCS 9905270224
Download: ML20196L203 (141)

Text

)

BAW-2346NP May 1999 G$h

=OW

/if STEAM,.

TOR? COMMITTEE s,

4 g) 7 xy v

, -~~mmq ALTERNATE REPAIR CRITERIA FOR TUBE END CRACKING IN THE TUBE-TO-TUBESHEET ROLL JOINT OF ONCE-THROUGH STEAM GENERATORS r=, -

nm,

1 BAW-2346NP f

May 1999 ALTERNATE REPAIR CRITERIA I

FOR TUBE END CRACKING IN TIIE TUBE-TO-TUBESHEET ROLL JOINT OF ONCE-THROUGH STEAM GENERATORS Prepared for B&W Owne.rs Group Steam Generator Committee l

B&WOG Non-Proprietary Version Prepared by:

Framatome Technologies, Inc.

P.O. Box 10935 Lynchburg, Virginia 24506-0935 mieir

l 1

BAW-2346NP APPROVALS PREPARED BY:

DATE: 5

/

f D.E. Costa This report was reviewed and was found to be an accurate description of the work reported.

REVIEWED BY:

b 3Il DATE:

5 l/9 /H 2 nm kA. Burges0 Verification of Independent Review.

5//9/?9 APPROVED BY:

DATE:

Jf. Brown j

Manager. Steam Generator Engineering This report has been approved for release.

s8 RELEASED BY:

DATE:

I7 N D.J. Binh' Program Manager B&W Owner's Group Thk document is the non proprietary version of the proprietary document BAW-2346P Revision 0. In order to qualify as a non-proprietary document, certain blocks of proprietary information have been withheld.

The criteria used for withholding information is provided below.

b) The information reveals data or material concerning B&WOG research or development plans or programs of potential economic advantage to the B&WOG.

c) The use of the information by a non member would decrease his expenditures, in time or resources,in designing, producing or making a similar product.

d) The information consists of test data or other similar data concerning a process, method or component, the application of which results in an economic advantage to B&WOG.

e) The information reveals special aspects of a process, method, component or the like, the exclusive use of which results in an eonomic advantage to B&WOG.

ii

BAW-2346NP RECORD OF REVISION REVISION DATE SECTION DESCRIPTION 0

May 99 ALL Original Release iii

BAW-2346NP EXECUTIVE

SUMMARY

This report documents the technical justi6 cation to implement alternate repair criteria (ARC) for axial cracks oflimited extent in the tube within the originally fabricated tube-to-tubesheet rolled joints of Babcock and Wilcox (B&W) Once-Through Steam Generators (OTSG).

This assessment demonstrates that a number of tubes with axial cracks in the rolled joints may remain in service, without repair, while still conservatively satisfying allowable primary-to-secondary leakage limits set forth in plant technical speciGeations and licensing submittals.

Numerous tasks were performed to develop the technical basis for the ARC. A detailed finite element analysis of the OTSG was performed to determine the general structural behavior of the OTSG (i.e. tube loads and tube-to-tubesheet joint interface behavior) under normal operating and accident loading conditions. A second finite element analysis was performed to assess the structural integrity of the tube-to-tubesheet weld for both normal and accident loads. Shop fabrication records and plant repair history were reviewed to establish the most representative (and bounding) geometry for the tube-to-tubesheet joints. A detailed mockup was designed to simulate actual plant operating geometry and loads. In order to insure the mockup was truly representative, a comprehensive parameter study was performed. Flaw (crack) definition and eddy current characteristics were deGned based on plant inspection results and mockup testing.

An analysis was also performed to assess the potential of crack growth between inspection periods. Leak testing was performed under simulated field conditions to establish leak rates for normal 100% steady state power operation and limiting accident conditions. The leak rates were then coupled with the finite element results of OTSG behavior to create curves of the " design" steady state and accident leak rates as a function of tubesheet radius. These curves provide the utility with a means for determining leak rates for cracks in specific tube locations within the tube bundle.

In addition, the qualification process outlined and used for this ARC provides each utility with the capability to perform plant specific assessments using plant specific data. This data may include, but is not limited to, plant specific transient loads and resulting tube-to-tubesheet joint interface dilations, plant specific qualified inspection techniques, and plant specific crack growth rates. Plant specinc changes to the leak rates presented in this ARC will be provided in 10 CFR 50-59 licensing submittal.

In summary, the details contained within this Topical Report provide evidence that leaving a quantified number of tubes in service with tube end axial cracks is acceptable and will not compromise the integrity of the steam generator.

iv l

l

BAW-2346NP Requirements for ARCImplementation The following items provide a summary of the requirements necessary for application of this ARC.

The ARC will be applied to any axial cracks detected in the original fabrication tube roll of the upper or lower tubesheet, and only to those indications that do not extend into the portion of the tube adjacent to the carbon steel portion of the tubesheet. " Adjacent to I

carbon steel" refers to the portion of the tube that was rolled into the carbon steel portion of the tubesheet, as opposed to the portion of the tube that was rolled into the tubesheet cladding.

This ARC is not applicable to tubes with circumferential, mixed mode, or volumetric indications.

Site specific measurements to determine the average thickness of the tubesheet cladding are required. The cladding thickness is used to define the maximum allowable crack I

length. The measured clad thickness must be less than ( c ] inches in order for this ARC to be applied. A procedure for determining clad thickness is provided in Appendix E.

Site specific inspections of tube rolled joints are required to identify locations and orientation of tube end cracks. Location shall include tube location within the bundle as well as the crack location in relation to the clad-to-carbon steel interface.

A plant specific growth rate analysis for detected tube end cracks must be performed to I

establish a plant specific growth rate, or to verify the applicability of the growth rate determined within.

l The total number of identified (detected) TEC indications must be increased to account for the probability of detection (POD). This increase is required to assess the population of undetected flaws.

I Total calculated leakage from all identified axial cracks addressed by this ARC, e

combined with leakage from other sources, will be limited to the allowable leak rates defined in plant specific licensing commitments. Note: For tubes with multiple indications, a separate leak rate for each Indication must be used.

Axial indications addressed by this ARC with a combined calculated leakage less than the allowable limit may remain in service. If the combined calculated leakage exceeds i

the allowable limits, selected tubes may be repaired until the combined calculated leakage of indications remaining is service is less than the allowable limit.

J V

J

BAW-2346NP TABLE OF CONTENTS APPROVALS................................................................................................................................li R E C O R D O F R E V I S I O N........................................................................................................ 1i 1 E X E C U TI V E S UM M A R Y.......................................................................................................... i v T A B L E O F C O NT E N TS........................................................................................................... v i L I S T O F FI G U RES............................................................................................................ 1 x LISTOFTABLES.................................................................................................................x A C RO NYMS A N D AB B REVI ATIONS.................................................................................. xi 1.0 I NT RO D U CTI O N.......................................................................................................... 1 1.1 Purpose.................

... I 1.2 B ac k g rou nd..................................

.I 1.3 Qualification Requirements.................

....I 2.0 D ES C R I PTIO N O F TH E OTS G................................................................................ 3 2.1 Functional Description...........

.3 2.2 General Design Information.

.3 2.3 Tube Material Properties....................

...4 2.4 Tube-to-Tubesheet Joint Geometry..

.5 2.5 Tube.to-Tubesheet Joint Repair History..

........5 2.6 Conclusions.

..........................................................5 3.0 DES CRIP' TION OF TUBE END CRA CKING........................................................ I 1 3.1 B ac k gro u n d...................................

... I 1 3.2 Characteristics of Tube End Cracks......................................................................... I 1 3.3 Potential Effect on Tube-to.Tubesheet Weld.

.12 3.4 Potential Effect on Structural Integrity of Steam Generator.......

.12 3.5 Summary / Conclusion.......................

.... 13 4.0 TEC H N I CA L B A S IS FO R A R C.............................................................................. 14 4.I G e neral A ppro ac h......................................................................... 14 4.2 Conservative Assumptions / Parameters.........

.16 5.0 ANALYSIS OF ROLLED J OINT INTE RFACE................................................... 17 5.1 Development of Finite Element Model.......

. 17 5.2 Model Parameter Assessment.......

. 21 5.3 Transients Considered.........................

...................22 5.4 Thermal Boundary Conditions......

.. 22 5.5 Thermal Results.............................

.23 5.6 Structural Boundary Conditions.........

.25 5.7 S truc t u ral Re s ul t s....................................................

. 25 5.8 Co n c l u s i o n s...............................................................

.26 6.0 S UMMA R Y O F TEST PR O G RA M........................................................................... 29 6.I Determination ofInstallation Torque............

....... 29 6.1.1 B ac k gro u n d................................................

...... 29 6.1.2 Summary of Tests.................

.29 6.1.3 Discussion of Test Variables...

. 29 I

I I

V!

l

L BAW-2346NP 6.1.4 Evaluation of Data.......................

. 31 6.1.5 Summary of Results/ Conclusion -Installation Torque.........

. 33 6.2 Assessment for Bounding Joint Mockup...

.. 33 6.3 Description of Production Mockup.........

.39 6.4 M oc ku p Condi tio nin g......................................................................... 41 6.4.1 Dilation Cycling..................................

.41 6.4.2 Axial Load Conditioning..

... 41 6.5 Description of Leak Tests.....

.. 43 6.5.1 Initial Leak Tests................................

... 44 6.5.2 -

Fi nal Leak Testin g........................................

...... 45 6.6 Results of Mockup Leak Tests.....

.....46 7.0 APPLICATION O F LEAK RATE RESULTS.......................................................... 49 7.1 Method of Analysis................

.49 7.2 Steam Generator Tube Loads and Delta Dilations...

.49 7.3 Leak Rate versus Tubesheet Dilation.............

.52 7.4 Leak Rate versus Tubesheet Radius..................

.... 54 8.0 E DD Y C U RR ENT T E C H NI Q U E................................................................................ 60 8.1 EC Background...............

.......... 60 8.2 EC Te s t................................................

........ 61 8.2.1 EC Test Samples...........................

............61 8.2.2 EC Te s t Fix t ure..............................................................

.. 67 8.2.3 EC Test Acqu isition............................................

. 68 8.3 EC Te s t Re su lt s.................................................................................. 69 8.3.1 EC Interface Location........................

.70 3

............74 8.3.2 Coil Look-Ahead................

8.3.3 Flaw Location s.............................................................

... 75 8.3.4 Data Acceptance............

..........77 8.4 ARC EC Application......................

.. 78 9.0 TE C G RO WTH A SS ESS M ENT.............................................................................. 7 9 9.1 In t rod u ctio n.....................................................

.. 79 9.2 Methodology.......................

... 79 9.3 Evaluation of Indications in the Cladding Region.

. 79 9.4 Results From Growth Rate Study.......................................

.80 9.4.1 Distribution of Indications Relative to Clad-to-carbon Steel Interface.......... 80 9.4.2 Change in Distance Relative to Clad-to-carbon Steel Interface.......

......... 80 9.4.3 Statistical Evaluation.............

........ 80 9.5 Growth Rate Results/ Conclusions..............................................

...... 81 10.0 PR O B A B I LITY O F D ETE CTI O N.............................................................................. 84 11.0 STRUCTURAL EVALUATION OF TUBE-TO-TUBESHEET WELDS................ 85 11.1 Introduction............................

..... 85 11.2 Finite Element Model...................................

.85 11.3 Transients Considered......................................

.......... 8 6 11.4 Boundary Conditions...........................................

. 86 11.5 Analysis Results...............................

.87 11.6 Conclusions....................

.88 12.0 PLANT S PECI FI C IMPLEMENTATION................................................................. 89 Vil

BAW-2346NP 13.0 R I S K A S S ESS M E NT................................................................................................... 9 1 13.1 Principle 1: Satisfaction of Current Regulation.

........... 91 13.2 Principle 2: De fense in Depth........................

.92 13.3 Principle 3: Safety Margins......................

.. 92 13.4 Principle 4: Effects on Core Damage Frequency...............

.92 13.5 Principle 5: Performance Monitoring.................

.. 93 14.0

SUMMARY

................................................................................................94 A PPENDIX A : RES ULTS O F L EA K TESTS.................................................................... 95 APPENDIX B: STATISTICAL EVALUATION OF LEAK RATES................................102 A P PEN DI X C: TY PI CA L ETS S........................................................................................... 1 13 A P PE N D I X D: G RO WTI I A N A LYS IS................................................................................ 118 APPENDIX E: PROCEDURE FOR MEASURING CLAD TIIICKNESS.......................126 I

I I

I I

I I

I I

I I

h viii

BAW-2346NP LIST OF FIGURES

(

Figure 2-1: OTSG Lon gitudinal Section....................................................................... 7 Figure 2-2: OTSG Tube-to-Tubesheet Joint.......................................................................... 8 Figure 5-1: Lower Portion of OTSG Finite Element Model............................................. 20 Figure 5-2: MS LB Thermal Results................................................

......................24 Figure 5-3: Delta Dilation versus Tubesheet Radius..................................................... 28 Figure 5-4: Axial Tube Loads versus Tubesheet Radius.............

...........................28 Figure 6-1: Summary of Torque Test Results................................................. 32

[

Figure 6-2: Mockup Assembly...........................

...... 40 Figure 6-3: 95/50 Room Temperature Leak Rate vs. Delta Dilation............

............47 Figure 7-1: 95/50 Leak Rates for Upper Tubesheet......

. 58

[

Figure 7-2: 95/50 Leak Rates for Lower Tubesheet.................................................... 5 8 Figure 8-1: Look-A head Tu be S ampl e....................................................................... 62 Figure 8-2: [ d ] Tube Sample....................................

...................................................63 Fi g u re 8-3 : Tube shee t B l oc k.................................................................................. 64 Figure 8-4: CCI, View from ID of Tube Hole 3.........................................

....... 65 Figure 8-5: Assembly Map of Tubesheet Block................................................... 66 Figure 8-6: X Y-Positioning Fix t u re.................................................................................. 67 Figure 8-7: Center of [ d ] Shown in Lissajous................................................................ 70 Figure 8-8: Pancake Coil Response from the CCI........................................................... 71

[

Figure 8-9: C-Scan of CCI On [ d ] Pancake Channel.................................................... 72 Figure 8-10: Plus Point Coil Response from the CCI....................

.......................73 Figure 8-11: C-Scan of CCI on [ d ] Plus Point Channel.....................

.............................. 74 I

Figure 8-12: A Flaw 0.050 Inch Away From The CCI [ d ]................................................. 76 Figure 3-13: A Flaw At The CCI [ d ] Carbon Steel Region................................ 77 Figure 9-1: Distance of FTL from CCI, SG-A......................................................... 82

(.

Figure 9-2: Distance of FFL from CCI, SG-B...............................

...........................82 Figure 9-3: Differences in Flaw Tip Location, SG-A....................................................... 83 p

Figure 9-4: Differences in Flaw Tip Location, SG-B.

..............83 t

Figure E-1: Typical Map of OTSG Tubesheet Face................................................ I 29

(

(

L lx

BAW-2346NP LIST OF TABLES Table 2-1: OTSG Design Information.......................

.9 Table 2-2: Geometry of the Tube-to-Tubesheet Joint............

.... 10 Table 2-3: Tube-to-Tubesheet Joint Materials................

.10 Table 5-1: Summary of Selected Results..........

..... 27 Table 6-l: Torque Test Results Regression Summaries..............................

.32 Table 6-2: Summary of Parameter Study.............

39 Table 6-3: Load Conditioning Test Loads and Cycles......

. 43 Table 6-4: Matrix of Test Parameters...

.46 Table 6-5: Pertinent Leak Test Results.

.48 l

Table 7-1: A Dilation versus Tubesheet Radius..........

.. 50 l

Table 7-2: Results of 95/50 Leak Rate.

.. 53 Table 7-3: Summary of 95/50 Leak Rate Results....

. 54 Table 7-4: Steady State Leak Rate versus Tubesheet Radius..........................

... 56 Table 7-5: Accident Leak Rate versus Tubesheet Radius......

.56 Table 7-6: Summary of 95/50 leak Rate vs. Tubesheet Radius...

.59 Table 8-1: Look-Ahead Tube Samples......................

.. 62 Table 8-2: Assembly of Tube Samples into Tubesheet Block.......

........ 66 Table A-1: Initial Leak Rate Results..................

.96 Table A-2: Final Leak Rate Test Results...........

.... 97 Table B-1: Accident Condition leakage Data From Testing......

.105 Table B-2: Accident Condition Statistical Results.............

..I10 Table B-3: Steady State Condition leakage Data from Testing.

.....I11 Table B-4: Steady State Condition Statistical Results.

.I12 Table D-1: Re-Analysis of SG-A Data.............

..... 121 Table D-2: Re-Analysis of SG-B Data................................

... 123 I

[

X

r.

BAW-2346NP

ACRONYMS AND ABBREVIATIONS ANO-1 Arkansas Nuclear One - unit 1

ARC-Alternate Repair Criteria

.ASME:

. American Society of Mechanical Engineers ASTM

- American Society of Testing and Materials B&W =

Babcock and Wilcox B&WOG.

B&W Owners Group.

CCI Clad-to-carbon steelInterface CD-Cooldown CR-3 Crystal River-unit 3 DB-1 Davis Besse - unit 1 EC-

. Eddy Current

'ECT Eddy-Current Technique -

EDM Electrical Discharge Machine EFPY Effective Full Power Years EPRI Electric Power Research Institute ETSS

. Examination Technique Summary Sheet FA Fuel Assembly FE-Finite Element FTI-Framatome Technologies Incorporated FTL~

Flaw Tip length -

HU Heatup ID Inside Diameter MSLB Main Steam Line Break MR Mid Range -

NDE Non-Destructive Examination 1

NRC.

Nuclear Regulatory Cunmission OD Outer Diameter

. ONS Oconee Nuclear Station - unit 1 -

ONS Oconee Nuclear Station - unit 2 ONS-3 Oconee Nuclear Station - unit 3

-OSUTL One Sided Upper Tolerance Limit OTSG Once-Through Steam Generator POD Probability Of Detection PWSCC Primary Water Stress Corrosion Cracking

=RC Rotating Coil technology, such as RPC or Plus Point RHR Roughness Height Rating SI,nu Maximum Stress Intensity SG' Steam Generator T,

. average of reactor inlet and outlet temperature Teow reactor inlet temperature, Cold leg

Tho, reactor outlet temperature, Hot leg

-TEA Tube End Anomaly TEC Tube End Crack

- TMI-l

- Three Mile Island - unit 1 TSP.

Tube Support Plate TW Through-Wall UTS Upper Tubesheet xi

BAW-2346NP j

1.0 INTRODUCTION

i l

1.1 Purpose

]

The purpose of this document is to provide a technicaljustification to implement an ARC for tubes with axial indications of limited extent near the tube ends. This ARC will be applied to indications in the original fabrication tube-to-tubesheet rolled joint located in the upper and lower tubesheets of once-through steam generators.

1.2 Background

Inspections at plants with B&W designed steam generators have revealed crack-like indications near the ends of the expansion roll of the tubes' upper tube-to-tubesheet j

joints. These indications, originally defined as " tube end anomalies" (TEA), were initially believed to be in the non-pressure boundary portion of the tube. Subsequent inspections have shown that some of the crack-like indications actually extend below the tube-to-tubesheet weld into the pressure boundary region of the tube. These crack-like indications are now referred to as " tube end cracks" (TEC). These types of indications have since been identified in NRC Information Notice 98-27, " Steam Generator Tube End Cracking"(dated July 24,1998). It is noted that no laboratory examination data on tube-to-tubesheet rolled joints is available which would verify the indications are actually cracks. The indications are believed to be cracks based on the eddy current response.

A program has been implemented that would allow for the continued operation of steam generators without repair of the tubes with TECs. Allowing tubes with TECs to remain in service is justified based on a cornbination of structural analyses, mockup testing and in-service inspections.

1.3 Qualification Requirements l

l This ARC will allow tubes with indications of axial orientation located within the clad l

region of the tube-to-tubesheet rolled joint, or in the portion of the tube protruding from j

the cladding, to remain in service without repair. The limitations of this ARC are:

l This ARC does not apply to tubes with circumferrntial, mixed mode, or l

volumetric indications.

The ARC applies only to tubes with axial indications adjacent to the tubesheet cladding or in the tube end protruding from the cladding. Tubes with any portion of an axial indication adjacent to the carbon steel tubesheet must be repaired. The clad to-carbon steel interface (CCI) must therefore be located by inspection.

The tube-to-tubesheet weld in the degraded tube must be capable of carrying all normal operating and accident tube loads.

I

BAW-2346NP The combined postulated leakage from all indications for which the ARC will be e

applied, as well as other primary leakage from throughout the plant, must be less than the leakage limits defined in the plant's Technical Specification or licensing submittals.

The analysis and testing documented in this report address the effects of axial indications only. The effects of circumferential, mixed mode, or volumetric indications on the structural integrity and leakage performance of the tube-to-tubesheet joint have not been evaluated. In order to limit the ARC to axial indications, inspection techniques must therefore be able to characterize the flaw morphology (i.e. circumferential, axial, mixed mode, or volumetric).

Axial cracks of limited extent in the tube-to-tubesheet rolled joint do not diminish the axial load carrying capability of the tube. Although the load carrying capability of the rolled joint may be diminished, the tube will not slip if the weld is strong enough to carry the loads. Therefore, a detailed analysis of the tube-to-tubesheet weld is required to show it is capable of resisting the tube axial loads associated with both normal operating and accident conditions. In addition, the support provided by the tubesheet prevents the affected tubes from rupturing at the axial crack location due to internal pmssure.

The presence of the tube-to-tubesheet weld and the support of the tubesheet provide adequate structural strength for tubes with TECs. Therefore, the most important technical issue for the subject indications becomes the evaluation of their postulated leakage. The total leakage from axial tube end cracks left in service, when combined with leakage from other sources throughout the steam generator, must be less than the leakage limits identified in the plant licensing commitments. Therefore, quantifying the leak rate associated with TECs is the main focus of this Topical Report.

l 2

l BAW-2346NP

2.0 DESCRIPTION

OF THE OTSG There are currently seven operating nuclear power plants in the United States that use model 177FA OTSGs. They are Arkansas Nuclear One Unit 1 (ANO-1), Crystal River Unit 3 (CR-3), Davis Besse Unit 1 (DB-1), Oconee Nuclear Station 1 (ONS-1), Oconee Nuclear Station 2 (ONS-2), Oconee Nuclear Station 3 (ONS-3), and Three Mile Island Unit 1 (TMI-1). The OTSGs at all these plants are nearly identical in design and function, and have similar tube material properties.

2.1 Functional Description The OTSG is a straight-tube, straight-shell, vertical, counter-flow, once-through heat exchanger with shell-side boiling. By nature of its design, the OTSG eliminates the need for steam separating equipment.

In the OTSG, shown in Figure 2-1, primary fluid from the reactor enters through an inlet nozzle in the top head, flows down through the tubes, enters the bottom head and exits through two primary outlet nozzles. The feedwater enters through a series of spray nozzles near the top of the annular feedwater-heating chamber. Here, the feedwater is heated to saturation temperature by direct contact with high-quality or slightly superheated " bleed" steam. The resulting saturated feedwater enters the tube bundle through ports near the bottom of the tube bundle. Nucleate boiling starts immediately upon contact with the hot tubes. Steam quality incmases as the secondary fluid flows upward between the tubes in counterflow to the primary fluid inside the tubes. The departure from nucleate boiling occurs at about the 25 foot level at design conditions.

The mode of heat transfer then changes from nucleate to film boiling. Steam quality continues to increase but at a slower rate. After 100% quality is reached, the steam becomes superheated, leaves the tube bundle at the upper tubesheet, flows down the steam annulus, and exits through two steam outlet nozzles.

All B&W designed plants operate at similar conditions. Because the plants have similar reactor coolant system flow rates, the core AT's are similar and thus the values for L and Tcoia are similar. For the plants that operate with a 579 F Tm (all but DB-1), b and Tcoia are approximately 601 F and 556 F, respectively. For DB-1, which operates with a 582 F Tm, T o, and Teoio are approximately 605 F and 559 F, respectively. These values h

can vary due to Tm control and tube plugging. (Tube plugging reduces reactor coolant system flow rates and increases the AT across the core.)

2.2 General Design Information j

All of the operating OTSGs were manufactured by the B&W Company in their Nuclear

[

Equipment Division in Barberton Ohio, using essentially the same manufacturing process. Each steam generator weighs approximately 570 tons, has an outer diameter of 12-1/2 feet, and an overall height of 73 feet. Table 2-1 contains a summary of OTSG design data.

I 3

BAW-2346NP Each steam generator has more than 15,000 alloy 600 tubes that are spaced on a [c] inch.

triangular pitch. These tubes are [c] inch OD x [ c ] inch nominal wall x [c] foot long.

They were expanded into the upper and lower tubesheets (at the primary face) with a

[c] inch minimum roll and then welded to the primary face cladding of each tubesheet.

Figure 2-2 provides details of the rolled joint and weld. The use of straight tubes results I

in almost pure counterflow, which results in improved secondary flow distribution and primary-to-secondary temperature differentials. This design also has the benefit of compressive loading on the tubes during normal operating conditions. This is mainly due to the average tube temperature being greater than the average shell temperature.

Additionally, the alloy 600 tubes have a thermal coefficient of expansion slightly greater than that of the carbon steel shell. This compressive load tends to inhibit the initiation and propagation of some stress related damage mechanisms.

Proper lateral spacing of the tubes is maintained by [c] tube support plates. They were fabricated from [ c ] inch thick carbon steel plate, drilled and broached to provide surface contact and support along three axes for each tube at each tube support plate. An exception is the periphery holes of the uppermost ([c]*) support plate which were not l

broached. The support plates were non-uniformly spaced vertically to prevent resonant vibrations along the tube length, thus providing the highest possible damping factor.

l The OTSG upper and lower tubesheets were clad on the primary side face with inconel alloy 600 and examined by liquid dye penetrant and ultrasonic inspection. The design thickness of the cladding was [ c ] inch minimum for all plants except DB-1 which has a l

minimum design thickness of [c] inch. It is noted that the cladding thickness may vary however over the tubesheet.

A thermal cycle of [ c ] F to [ c ] F was applied to the completed OTSG to perform stress relief to all welds in accordance with the ASME Code. The thermal cycle was monitomd using data from thermocouples attached to the vessel. Following stress relief, I

the OTSGs were hydrostatically tested.

2.3 Tube Material Properties The OTSG tube material in all seven operating plants is alloy 600 (ASTM SB163). The OTSG tubing raw materials were melted into the alloy 600 ingots and fabricated into hollow rounds by B&W Tubular Products Division (TPD). TPD and two other outside vendors performed the tube finishing processes (tube drawing, etc.). The tube material

[

was later thermally treated between [ c ]*F and [ c ] F for a minimum of [c] hours l

during the full furnace stress relief of the completed steam generator. As a result, the installed tubes are both sensitized and stress relieved, which provides improved resistance to caustic stress corrosion cracking but makes them more susceptible to intergranular attack caused by reduced sulfur species.

4 s

1 BAW-2346NP 2.4 Tube-to-Tubesheet Joint Geometry The tube-to-tubesheetjoint is comprised of a [c] inch minimum roll at the primary face of both tubesheets and a fillet weld attaching the tube to the clad face. The cladding is an alloy 600 weld overlay with a design minimum thickness of [c] inch to [c] inch. The joint roll was installed using a torque-controlled rolling process with a tool that indexed the depth of the roll off of the cladding face. The roll length and position were therefore controlled by the geometry of the tool. Approximately.[ c ] inch of excess tubing extends past the top of the cladding. This tube " stick out" is required to accommodate the fillet weld. A summary of the joint geometry is provided in Figure 2-2. Pertinent dimensional and material information for the joint are shown in Table 2-2 and Table 2-3 respectively.

2.5 Tube-to-Tubesheet Joint Repair History In order to assure that the ARC properly accounts for the current conditions at all the B&W operating plants, a review of the design drawings, manufacturing records, and repair history of the tube-to-tubesheet joint was performed. With the exception of the tubesheet bore diameter, the design geometry and materials comprising the joint were the same in all of the B&W 177FA plants. The diameter of the tubesheet holes is nominally

[ c ]-inch for ANO-1, DB-1 and ONS-3, compared to a nominal [ c ]-inch for ONS-1, ONS-2, CR-3, and TMI-1.

Damage and repairs to the tube-to-tubesheet joint of the OTSGs were also reviewed, including records of tube end damage due to loose parts at ONS-1, CR-3, and ONS-2, and repair rolls in the field. Repairs that were assessed for their possible impact on the application of the ARC include the flush weld repairs at ONS-1 and the recessed flush welds at CR-3.

i in addition, the history of the OTSGs was reviewed to assess the impact of the application of chemical cleaning and water slap, as well as abnormal transient conditions l

on the roll joint integrity. The abnormal transient conditions included boil-dry events, loss of offsite power, turbine trip, and loss of reactor coolant pumps. These conditions were reviewed for axial tube loads associated with the transients.

2.6 Conclusions All currently operating U.S. plants with OTSGs have B&W model 177FA steam generators. These OTSGs were all fabricated in the same time period, utilizing the same general design and materials. Variability in tube material properties is limited due to the sole source melting of the alloy 600 material, consistent fabrication specifications, and consistent thermal treatments. These plants also operate at similar conditions. With regard to application of this ARC,it was particularly important that the tube-to-tubesheet joint design be similar for all plants. This has been verified by review of design information and repair records.

Thus, all once-through steam generators can be 5

I BAW-2346NP considered generically with regard to design, function, and materials relative to application of this ARC.

I I

I I

I I

I I

I I

I M

[

9

[

6

BAW-2346NP Figure 2-1: OTSG Longitudinal Section Primaryinlet Nonie -

k.

n\\

/

Manway inspection Opening

~ ~

Upper Hemispherical ~

\\

'g Upper Tubesheet Head Vent & Level Sensing '

N k

Connection pgg g.gg k

bh3

- - Manway AFW Header c(

- ggess Upper Cylindrical Bame

' ' f' f'fjfg[][*j Tube Support Plate Steam Outlet Nonle (2) x

}b.

d m.pgM NF

^

Handhole (2 to 8)

';[hg i

Upper High Level bl l

6 g/ y Drain (2)

Sensing Connection (2) %

wW Main Feedwater Risers

$jjg$!gd a

ee ater Memal Main ion (2) ~._- >--

f ff N

e in onn Feedwater Header

.=.

ffh Downcomer

~

Temperature Sensing Connection (2) yj

$g%j)g{y:

Mjj;g;)}jfjj Tubes pmg g

l hk YIO Lower Cylindrical Bame Upper Low Level Sensing Connection (2) N kf f

/

Handholes(7) N rhd -

I dm&"*

o --

Drain (2)

NJ

l

%g j

=

' *[i eNion (2) j c Temperature 3

o I

Drain (4) #~ r,

Sensing Connection Lower Hemispherical Head '

\\

Lower Tubesheet Primary Outlet Nozzle (2) -

' s.

_ __ __ __ Manway SupportSkirt

% /

p -\\

t kp inspection Opening 96C129G Primary Drain I

7

J 1

BAW-2346NP Figure 2 2: OTSG Tube-to Tubesheet Joint a

Tube-to-Tubesheet Weld.

[c]"

[ c ]" minimum leg Clad Face a

n g g jae 4

Cladding I

cc

'][tG.);W-e%a:E TQ ' %?Th '

intedace r.,

.r w.

v o-xP,3 %.3.~y#v..,'.%..

gt l

g ,.h D i. C, #

• r';

'L~ h [.

Yhd?MUN "$11 Jb.T

,a,;;c.::.;a.. c Mm?Mrit, Z ry. m

. y q$gvg.e::,

s L "wa.6%

m n

v$vW gr.-

't ti

. W.9R4 %( ;

Tubesheet ma.Y.M$

[ g ]"

wQsu i

wx;s

.,' W - 1

+. n: o,..; f GL,

t. YW k

W' 9#!!!yg[W MYfMQ k+DO knew a

..' j$$Q mru e

.ormer A ? dr.;

I

,+ 7f4df

,. Q iQ:x. L:Kph."(4-.

oj % f??A Q K V Q.

a

<.e.. a % pc

W:';Tr%f MFJ-.( A*J Sr i. l'-- ^ ^

.h i

[',4, l T bh 4

a @W.6QAMQIWfi q#9:n(!$,r1.3 NE kM& :2U1%

i

} N b N k MI[h b Mhk[h'iN M3N!h

$555YN$

wmcr&n@:o

> +.e.rcem

[j$2i$$

kM$dNIkM l

, Adh&%d%iM AG$EA?f:a:lKW

, n ata.Ml1 t%h:44p::

G.,s

.D w;1:6 4;f*. ?W. uQ^

i :

2. +'e&

.y or-.

T-v.3,.Qy iW '.

. s: Ag-nA

% < vg

1. o,o#6 a A--y

• ~ *W my-w y

v+w1 &,:m*

s N AM d.Qla 4?dp47Ajd7X'6fd <

l u tyjid'd p q

+

c J.."t

.p *T*. '

'r E32% %]Q"S WTCN" ST Q l

S

.M.didjN N$N::::idf;$sb m.:n n;h.w;

.,aadv '"d sqwww g#y},-. 4 m Tube

. C..

. ~..

,d%hz..,@ c'CA$w(

I

• * : &.HO?.Sf6::f n ? a.Y.;jip,;NcQ

.fC"l$5%gD.,

w%n:.r:.:nx. 5 n..,unn::

dktd'4CCW L

h M7,4.Zr.h

• Design minimum Values. Actual clad thickness varies.

l 1

l l

l' 8

jl l

llj l'

ljl j!l1l PN 6432 W

3-S A

N B

O 2

-SN O

1

-SN n

O o

ita mr

)

o fn C

I n

I

[

g I

M ise T

D l

A 9

G ST O

1 -

1 2

-B e

D lbaT 3-RC I

O NA T E D

l t

t l

a r

s e

O y

W W

)

o s

e W) r l

h p

a e

)

h it e

l M M p

m h

c p

ir k c

e h

d d s n

h s

)

l ec n

u e

e n b c

l lCn a

a s a

s a

u b n e

D..ic l

l l

e i

m ui i( S p

a e

e r

a ic i(

t u

n C t

v v

s b

e i

T (s h f

y h

T i(

k i

ic n e

e l

G o

t n

c o

T M T e s t

t t s e

o M M) lae i

r t

t t

e e l

r e

eic r

r o L L d

S T a

i s

I C

ei t

n te e

e a

hR) t nn P

e r

r r

h n t

k l mt h

e r

r f

f k

b s o

o o) e a

a r

e e M o o e

eh imc e

s s

s mte p

p ph ek imem l

o i

me w

w

. oG u

u n oh u

ual S S

S(

TT NBD TT TMi ee e

.h n

c c

a op o

o G o

b b c i

b p

p pc bi a

b p b n c P D it r

CO P P S N NS T Ti NT T NP i

uh ooi uy uin u

u u n n

i

(

(

BAW-2346NP Table 2-2: Geometry of the Tube-to-Tubesheet Joint Item Dimensions Tube Geometry

[cl Tubesheet Hole Geometry (c)

Length of Roll Expansion Icl Length of Excess Tube (stick out)

Icl Cladding Thickness (c)

I'I Tube-to-Tubesheet Weld Table 2-3: Tube-to Tubesheet Joint Materials Component Material Tubes

,c; Tubesheet (Upper & Lower)

[c]

Cladding

[c]

Tube to Tubesheet Weld

[c]

l 10

~

l BAW-2346NP

3.0 DESCRIPTION

OF TUBE END CRACKING

3.1 Background

[

During recent inspections, some B&W plants have reported indications from eddy current (EC) data in the tubes near the upper tube end. An indication above the primary face of the upper tubesheet was called a tube end anomaly (TEA) and initially left in service at f

some plants.

These indications included single end anomalies and multiple end anomalies. All anomaly indications were originally considered to be outside the pressure boundary.

During the [

c

), a reverse pressure bubble test identified leakage from two tubes, one in each generator. One of the leaking tubes had been identified as having a tube end anomaly during the previous inspection. Subsequent inspection of the leaking tubes using updated EC analysis guidelines (described in Section 8 within) indicated that some of these indications were in the portion of the tube adjacent to the tubesheet cladding and therefore are within the pressure boundary. The indications were identified as mixed mode cracking (axial and circumferential) in the OTSG-A tube and an axial crack in the OTSG-B tube. Both of these indications were in the upper roll region just below the upper tubesheet seal weld.

[ c 1 performed additional eddy current data analysis of all tubes previously defined to

(

have TEA. The inspection showed that the majority of the indications were identified to be in the rolled joint adjacent to the cladding. Re-rolling [

c

]

repaired approximately three thousand of these indications.

During the [

c

), approximately 250 tubes with tube end y

indications were re-rolled. Based on re-analysis of EC data, tubes with indications were

[

confirmed to be in service at [

c

). In the spring and summer of

[

c

] petitioned the NRC to allow the affected plants to operate the remainder of their current fuel cycles with TEC indications in service. The NRC granted these petitions with the commitment that the indications would be repaired or plugged during the next outage of sufficient duration.

All known TEC indications at ONS 1 and ONS-3 have since been repaired.

3.2 Characteristics of Tube End Cracks Based on a review of the EC data for tubes with TEC indications, the indications are typically characterized as crack-like and axially oriented. However, circumferential

(

indications have also been identified as well as a small number of volumetric indications.

l Multiple axial indications and combinations of axial and circumferential indications have been identified in some tube ends.

Based on the EC data, the TECs are believed to initiate on the inside surface of the tube.

They are typically short, axially oriented, and located in the rolled portion of the tube near the heat affected zone created by the tube-to-tubesheet weld. The rolling process and weld create residual stress that may make the material mom susceptible to Primary 11

~

I BAW-2346NP Water Stress Corrosion Cracking (PWSCC). For this reason, it is believed that the TECs are PWSCC initiated.

3,3 Potential Effect on Tube-to-Tubesheet Weld Based on industry experience of alloy 600 weld metal, it has been concluded that the tube-to-tubesheet weld (wire type [c]) is not likely to crack and that the weld would not be affected by the TECs. liistory of PWSCC failures of alloy 600 material (control rod drive nozzles and pressurizer nozzles) in pressurized water reactors shows that the cracks typically initiate on the ID of the nozzle in the heat affected zone (created by the application of the weld) and propagates through the nozzle wall. The crack does not initiate at the weld. The crack initiates at the highest stress location and follows the stress profile. Based on industry experience, stress corrosion cracking (SCC) susceptibility of Alloy 600 materials can be ranked as follows:

1. [

c

]

2. [

c

]

3. [

c

]

The rankings provide additional evidence that the [

c

]

[

c

}.

[

The only observed cracking of alloy 600 weld material in pressurized water reactors has occurred in repair welds. The repair welds were installed as part of the replacement of an alloy 600 nozzle that had failed due to PWSCC. For the repairs, portions of the original weld material were left in place. Although the cause of these weld failures is not known, it is believed the source may have been poor welds. There is also evidence the cracked welds were isolated to those repairs that used an alloy 690 nozzle.

This conclusion that the weld is not cracked is also supported by bubble tests of steam generator tubes performed at various plants. The tests clearly showed that the leakage was from a crack in the nozzle wall in or near the heat affected zone caused by the weld.

There was no evidence ofleakage from the weld and no evidence of cracks in the weld.

3.4 Potential Effect on Structural Integrity of Steam Generator Cracks that are within the pressure boundary of the OTSG tubes have the potential to contribute to primary-to-secondary leakage and may impact the structural integrity of the tube.

f Primary-to-secondary leakage during normal operation is monitored according to the plant's Technical Specifications to ensure that any leakage remains less than the f

acceptable limit. Total operational leakage at the plants with repairable TECs left in service has remained well within the acceptable limits of the Technical Specifications for these plants. The primary safety concern for primary-to-secondary leakage is postulated leakage during an accident condition. The axial loads and increased pressure differential 12 r

BAW-2346NP during a postulated accident condition have the potential to increase the primary-to-secondary leakage, compared to the normal operating conditions. Therefore.

TECs that are left in service must be assessed for postulated leakage during the worst case accident conditions.

The presence of the tubesheet precludes the possibility of burst for tubes with TECs. The tubesheet also prohibits the tube from bending in the vicinity of the TECs. Therefore, only tube axial loads (tension) can cause catastrophic failure of the tube within the rolled joint. Thus, the structural safety concern associated with TECs is the axial load carrying capability of the rolled joint and/or weld. Since the effect TECs have on the load carrying strength of the rolled joint has not been quantified, the axial load strength of the joint is relinquished to the weld. Provided the weld can resist the axial tube loads, the structural strength of the joint is maintained and the leakage remains the only concern.

Since the effects of circumferential, mixed mode, or volumetric flaws have not been evaluated for their effects on the structural integrity and leakage performance of the tube-to-tubesheet joint, this ARC is not applicable to these types of indications.

3.5 Summary /Conclusien Tube end cracks are indications near the end of the tube within the tube-to-tubesheet roll joint of the OTSG. These indications are typically axial in nature, but may include multiple axial indications at the same location, circumferential indications, or a combination of axial and circumferential indications. Some indications are volumetric.

The most likely cause of the tube end indications is PWSCC. TECs have been identified in the OTSGs since 1996 during upper roll transition inspections using rotating coil probes. The indications would have been difficult to identify using eddy current data from a bobbin coil probe because of the signal distortion caused by the tube end, tube weld interface, and the clad-to-carbon interface.

Many of these indications were originally considered to be outside the pressure boundary. However, updated EC analysis guidelines (described in Section 8 within) indicate that many of these indications are in the cladding region of the tubesheet and therefore are within the pressure boundary.

TECs that are left in service must be evaluated for contribution to primary-to-secondary leakage during a postulated worse case accident condition. Burst and failure by bending are precluded by the presence of the tubesheet. The structural integrity of the joint is confirmed based on the axial load carrying capability of the tube-to-tubesheet weld.

13

l BAW-2346NP 4.0 TECIINICAL HASIS FOR ARC There were numerous tasks required to develop the technical basis for the ARC. A brief description of each step is provided below with more detailed discussions contained in other sections of the document. A list of conservative assumptions / parameters used in developing the technical basis for the ARC is also provided.

4.1 General Approach A detailed finite element analysis of the OTSG was performed to determine the e

general structural behavior of the OTSG (i.e. tube loads and tube-to-tubesheet joint interface behavior) under normal operating and accident loading conditions. The I

critical parameter forjoint interface is the relative difference in tube OD dilation and tube bore dilation. This " delta dilation" represents the extent to which the tube-to-tubesheet joint is tightened or relaxed during normal operational transients and I

accident loadings. It is directly related to the joint interface pressure (stress) and therefore the load carrying ability and leak tightness of the joint. It is noted that due I

to the oval displacement of the tubesheet holes created by the bowing of the tubesheets under normal and accident loadings, the delta dilations are not symmetrical around the tube. The axial tube loads and joint interface delta dilations from the I

analysis were used as direct input to mockup testing for leakage. The delta dilations help to define the joint interface pressure or residual stress in the joint. The analysis also included a comprehensive parameter study to ensure that the model used was bounding for all the B&W steam generators.

In order to ensure the analysis and test mockup represented actual plant geometry, a e

I detailed review of shop fabrication records and plant repair history was performed.

The review included available data from the time of original roll expansion of tubes to the repair of damaged tube ends and welds in recent years. The results were used as input to both the finite element analyses and mockup design.

In order to assess potential leakage from axial cracks, a detailed mockup was designed. Before the production mockup design and construction were completed, a comprehensive parameter study was performed to ensure the mockup was truly representative and bounding of the tube-to-tubesheet joint in operating plants. The l

parameter study looked at variations in tube thickness, tubesheet hole bore diameter, roll installation torque, tube and tubesheet material yield strengths, as well as many other parameters. The results of the study were incorporated into the final mockup l

design.

l The production mockup consisted of two general components. The first was the e

I actualjoint mockup that included a [c] inch thick segment of tubesheet with a number j

of attached tube segments. The test tube, with an EDM (electrical discharge i

machine) notch to represent a crack, was rolled in place using actual procedures from the original shop installation. The second component of the mockup was a test frame l

used to dilate the tubesheet bore. The dilations were needed to simulate tubesheet bore ovalization under plant operating conditions as defined by the finite element 14

l BAW-2346NP analysis. The mockup also had the capability to apply axial tube loads and tube internal pressure.

Using the mockup to provide simulated plant (tube joint) conditions, pressure testing was performed to establish leak rates for a variety of operating and accident I

conditions. A number of different loading cases were tested. The cases included variation of pressure, tube hole dilations, and tube axial loads. When all the data was compiled, a set of curves ofleak rate versus tubesheet bore dilation was created.

1 The leak rates from the mockup testing were then coupled with the finite element (FE) results (tube axial loads and joint interface dilations) from the SG behavior l

analysis to define the normal operating and accident leak rates as a function of tubesheet radius. These relationships provide the utility with a means for determining potential leakage from cracks at specific locations within the tube bundle identified l

during outage inspections.

To ensure indications could accurately be located and characterized during outage I

inspections, mockup testing of actual inspection techniques was performed. Various EC inspection techniques were tested in order to find the best method of defining the location of the indication relative to the clad-to-carbon steel interface. The clad-to-carbon steel interface is used to establish an upper bound length for the axial indication. Tubes with indications of circumferential extent or indications of axial extent that extend beyond the interface, adjacent to the carbon steel portion of the I

tubesheet, are excluded from the ARC. To ensure the tube end axial indications will not propagate longitudinally to a length greater than [ c ] inch from the face of the cladding, an evaluation of plant inspection data was performed to assess the potential I

of indication growth between inspection periods.

Since the ARC will allow tube end axial cracks to be left in service, and axial cracks I

e could reduce the rolled joint interface pressure and axial load strength of the rolled joint, an analysis was performed to show that the tube-to-tubesheet weld is capable of I

resisting all normal and accident loads. This finite element analysis considered both original fabrication fillet welds and flush welds used for repairing tube end damage from loose parts. The detailed ASME Code analysis consicered normal operating stn sses and resulting fatigue usage as well as maximum stresses due te faulted loads.

A detailed summary of the tasks performed to justify this ARC is provided in the following sections.

I l

l l

15

_--a

l

?

BAW-2346NP 4.2 Conservative Assumptions / Parameters l

The following is a list of some of the conservative assumptions, design parameters, and geometry data used in the analysis and/or testing supporting the technical basis for the ARC.

I The Gnal production mockup used for leak testing incorporated many conservative e

features defined in a comprehensive parameter (geometry, material strength) assessment. See Section 6.2 for details.

A 100% through-wall EDM notch was used in the leak testing mockup to simulate the l

tight TECs observed in actual SG tubes. This results in a conservative leak path (now area) relative to the actual TECs.

For each tube leak tested, the axial EDM notch was [

d

].

This alignment conservatively maximizes the effect of dilation on leak rate.

The leak test was performed at [

d

], which eliminated the potential for

[

d

] as would be seen in the actual SG. Thus, the observed [

d

] leak rate is higher than would be expected I

[

d

).

l The accident leak testing was conservatively performed with the transient maximum delta dilation (and axial tube load) applied with the transient maximum primary-to-secondary pressure difference. This is conservative because the transient pressure at I

the time of maximum dilation is [

d J.

Although the actual dilation at the time of maximum pressure has not been determined, it should be [

d

].

The ARC assumes that all axial indications are 100% through-wall for the entire length of the tubesheet cladding. Therefore, all indications are assumed to contribute I

to the leakage. No reduction in leak rate has been taken to account for indications that are less than 100% through-wall.

I The ASME Code analysis of the tube-to-tubesheet weld conservatively assumes that the weld is exposed to the entire range of axial tube loads (i.e. the roll joint resists none of loads).

I Based on the conservative approach demonstrated in the preceding list,it is concluded the actual leakage that could result from TEC indications left in service by application of this l

ARC will be much lower than predicted in this report.

I I

16

l BAW-2346NP 5.0 ANALYSIS OF ROLLED JOINT INTERFACE g

The effon to qualify the TEC ARC required the determination of the general structural l

behavior of the OTSG for the various applicable loadings. This behavior was quantified by the development of a finite element model of the overall OTSG (tube bundle, tubesheets, shell, heads and support skirt).

Due to the straight tube design of the OTSG, the axial stiffness of the tubes interacts with the axial stiffness of the adjacent secondary shell, thus generating axial tube loads for I

various combinations of temperature and pressure. Additionally, the differential pressure across the circular (flat plate) tubesheets causes a " diaphragm effect" that results in tubesheet bowing. This bowing effect can produce a dilation of the tubesheet tube holes I

in the region of the tube-to-tubesheet joint. Such dilation may reduce the load carrying capability of the rolled joint and/or generate stresses within the weld. Because of the l

need for evaluation of the tube-to-tubesheet joint, modeling of the " general structural I

behavior" of the OTSG (i.e., deflections, axial tube loads, etc.) and the local structural behavior (i.e., hole dilations) was required.

Two key outputs from the analysis of this model were axial tube load and, by subsequent calculations, tube and tubesheet hole dilations. The tube dilations were calculated based I

on the applicable tube temperature and pressure values. The tubesheet hole dilations, when coupled with the tube dilations, provide the required joint interface data (i.e., delta-dilations).

5.1 Development of Finite Element Model I

As an input to the tube-to-tubesheet joint qualification effort, the general structural behavior of the OTSG was determined. The " general structural behavior" is considered to be the structural response of the OTSG to applied thermal (such as temperature I

transients) loads as well as mechanical loads (preload, primary pressure and secondary pressure). This behavior has been quantified by the development and execution of a finite element model of the overall OTSG. As an example of the model grid, Figure 5-1 depicts the lower portion of the OTSG model.

To appropriately capture the structural behavior affecting the axial tube loads and associated tube hole dilation, the OTSG FE model included the tubes, tubesheets, secondary shell, upper & lower heads and support skirt. The tube. sheet components have been sub-divided into two regions - 1) the region perforated with tube holes (over 15,000 l

per tubesheet) and 2) the solid region (outboard of the perforated region). This division was required to facilitate the application of the " equivalent plate" material properties to the perforated region.

I Because the key results are a function of the overall OTSG stiffness (as reflected by the shell and tubes), it was important to represent the [

c

] of these subcomponents. Therefore, the model used the [

c

] for the shell and tubes.

The [

c

] of the tubes was taken as [ c ] inch based on measured data (vs.

[

c

]). Since the shcIl courses were fabricated by rolling flat plates into 17

BAW-2346NP cylinders, the [

c

] of the shell courses was taken as the average between the [

c

] and the [

c

}. For the thick shell courses, the [

c

] inches (vs. [

c

])

and for the thin shell courses, the [

c

] inches (vs. [

c

]

[

c

]).

The tube-to-tubesheet joint included both a rolled joint (typically [

c

] at the primary face of the tubesheet) and a fillet weld. The rolled joint resulted in an interface pressure between the tube OD and the tubesheet bore ID. This residual pressure generated a localized (relative to overall OTSG model) residual stress in the tubesheet around each tube hole. The collective effect of these residual stress fields (acting in the

[

c

] at the primary tubesheet faces) on the effective stiffness of the [

c

]

tubesheet was [

c

]. This [

c

] was made because any effect on stiffness is a function of the tube-to-hole interface pressure and this pressure I

varies with both temperature and tubesheet bowing magnitudes. Thus, any effect on tubesheet stiffness would vary with transient conditions (for temperature) and with tubesheet radiallocation (for bowing). h was therefore concluded that the effect of this l

[

d

] was [

d

] influence on the general structural behavior B

of the OTSG.

I The OTSG internal structures are [

d

). An investigation determined that the lateral suppon/ restraint provided by the Tube Support Plates (TSPs) had a

[

d

] on the general structural behavior of the OTSG. The investigation I

included a) [

d

], b) [

d 1

face of the tubesheets (due to small tube deflection [

d

]) and c) [

d

] of the tube under displacement. Even with [

d

],the I

collective [

d

] of the tubes remained [

d

] that [

d

]

(i.e.,[

d

] was registered at their connections to the tubesheets and, thus, the associated tube hole dilations [

d

].

The vertical support provided to the TSPs by the tie-rods and spacers was [

d

]

the model. During all but the most severe conditions, the [

d

] of these components would [

d J. Therefore, these components generally produced a

[

d

] to the Lower Tubesheet. This [

d

] would be applied as a [

d

] to the Tubesheet at [ d ] support spacer locations. For the case of secondary side blowdown (as would occur during the initial phase of a Main Steam Line Break transient), the collective differential pressure across the support plates would potentially result in [

d

] support rods. This load would be reacted as l

[

d

] on the [

d ] tie-rod locations of the Lower Tubesheet. This load was considered as [

d

] when compared to the accompanying [

d

]

differentials and axial tube loads. Note also that this [ d ] would occur only during the

" blowdown" which lasts only a shon time and [

d

] before other

[

d

] are developed (i.e., tube-to-shell temperature difference and maximum primary-to-secondary pressure differential).

u 18

I DAW-2346NP

[

d

] OTSG internals was specifically included for its potential effect. This was the mounting plate of the Lower Shroud. This plate mounts directly on the top of the Lower Tubesheet (in the recessed area). Due to its presence, it [

d

] from (or into) the secondary water into (or out of) the interfacing tubesheet rim. To account for this [

d

], effective [

d

] were determined and applied to the interfacing tubesheet area. Therefore, the physical mounting plate was [d ]

[

d

] its effect on the [

d

]of the Lower Tubesheet was included.

The [

d

] structure feature that was represented in the model is the Upper Shroud Shelf plate. This was included [

d

] of the steam and feedwater downcomer regions of the model. It had [

d

] to the structural behavior of the overall model.

Based on the nature and magnitude of the loadings, a linear-elastic, axisymmetric model was chosen as the appropriate model type. The FE model was constructed to te representative of all B&W 177FA OTSGs. The major sub-components for the various plants have essentially the same dimensions and material designations. However, there l

are some minor differences that required assessment and disposition (see Section 5.2).

I I

I I

I I

I I

l l

l l

19

BAW-2346NP Figure 51: Lower Portion of OTSG Finite Element Model

-[

[d)

[

f

-(

(

{

(

(

20

[.

{

BAW-2346NP

\\

5.2 Model Parameter Assessment As mentioned above, there are some minor differences between the OTSGs. To assure that the final analytical model conservatively includes any significant effects that these

[.

' features might have, parameter assessments were performed.

Parameter assessments were made for several features, including (but not limited to):

[c]

. [-

[

[

[c

]

[

[c]

{

[c]

[

(

c 21

- BAW-2346NP

[d]

a-The parameter assessment consisted of comparing the results from representative test loading on the model "with the feature included" to the results for the same test loading "without the feature"(base case). The key results used in the comparison were the axial tube loads and the associated maximum tube hole dilations. The conclusions from the assessments resulted in the incorporation of model features as noted above (either due to conservative effects or as a more realistic representation of the physical OTSG).

5.3 Transients Considered The OTSG' structural model described above has been used to analyze transient conditions for' use in assessment of the tube-to-tubesheet joints. The analyzed transients / conditions included:

- ) Heatup from 70F10 8% power a

b). Cooldown from 8% steady-state power c) Power Changes from 0% to 15% power and 15% to 0% power d) Plant Loading / Unloading from 8% - 100% power and from 100% - 8% power e) Steady-state conditions at 0%,8%,15% and 100% power levels f) Bounding Main Steam Line Break (MSLB) g) Small Break Loss-of-Coolant Accident (SBLOCA)

The selected transients included those representing the controlling transients for consideration of cyclic loadings (i.e., items "a" through "e", above) that have their parameters described in the Functional Specifications.

Also included were the

. controlling transients for consideration of single accident load application (i.e., items "f" and "g", above).

5.4 -. Thermal Boundary Conditions For the analyzed transie ":, die [

c

] load (relative to axial tube loads and dilations) is the thermal load. Thi thermal loads were applied to the FE model by 1) specifying a bulk fluid temperature a'. a surface and a corresponding heat transfer film coefficient or

2) by specifying an assigned temperature. For example, for surfaces such as the upper &

lower head inside surfaces, solid tubesheet rim surfaces (both primary & secondary sides), shell steam annulus, shell feedwater downcomer (above water level) and shell i

feedwater downcomer below water level, the appropriate bulk fluid temperature and heat 22 L

BAW-2346NP transfer coefficients were applied versus time. Whereas, the [

c

] of the

[

c

] and the [

c

] each had temperatures versus time assigned (based on the parameters of the specific transient).

For the normal operating transients, the heat transfer film coefficients were calculated for input based on standard correlations using the transient specific parameters (including temperature ramps and flow rates). However, for the analysis of the accident transients, the heat transfer coefficient values used came directly from the thermal-hydraulic transient analyses. Use of the heat transfer coefficients generated by the thermal-hydraulic analysis assured a continuity and consistency with the structural analysis for the rapidly varying anJ " change-of-phase" heat transfer regimes.

The outside surface of the OTSG was assumed to be perfectly insulated. This assumption has no significant effect on the " range" of loads / dilations for consideration of cyclic effects. The assumption yielded conservative results for the accident transients since these transients result in the tubes becoming much cooler than the shell. Thus, the assumption tended to maintain the shell temperature at a higher temperature thereby maximizing the tube-to-shell temperature difference.

The lower portion of the support skirt was assigned a constant temperature of [ c ] to approximate [

c

). This has no significant effect on the loads or dilations since this region is well removed from the tube /tubesheet/shell interaction.

The thermal boundary conditions described above were applied to the FE model and the

{

temperature distribution was calculated versus transient time (as nodal temperatures).

5.5 Thermal Results For each subject transient, the results of the FE model thermal analysis were reviewed to determine the transient time points that contain the most severe temperature distribution.

For considerations of maximizing axial tube loads and associated tube hole dilations, the crucial time point is the time of maximum tube-to-shell temperature difference.

Experience with analyses of the OTSGs shows that [

c

]

[

c

]

[

c

]. Of these [ c ],

the dominant contribution is the "I c

]". Therefore, the results of the thermal analysis were post-processed to provide the time-history of the tube-to-shell temperature difference.

An example of such a history of the tube temperature, average shell temperature and the tube-to-shell temperature difference is shown in Figure 5-2 for the MSLB transient.

The time point of maximum tube-to-shell temperature difference (delta-T) was analyzed as well as other time points such as the initial condition, time point before maximum delta T, time point after maximum delta-T and near the end of the transient. For each time point, the nodal temperatures throughout the FE model were used as input to the static structural analysis.

23

i BAW-2346NP Figure 5 2: MSLB Thermal Results l

[c]

I I

I l

l l

1

[

[

[

[

[

r L

24

BAW-2346NP 5.6 Structural Boundary Conditions As described above, a static structural analysis was performed for critical transient time points determined in the thermal analysis. For each time point, the applied thermal load for the static analysis was the nodal temperatures from the thermal analysis. Also applied were the mechanical loads that occur simultaneously with the nodal temperatures --

primary pressure, secondary pressure and installed tube preload.

Consistent with the actual OTSG geometry, the pressure loads on the perforated portion of the upper & lower tubesheets were applied such that pressure only acted on the metal surface (i.e., accounted for tube holes). Also, [

c j

[

c

}. Because the tubes were modeled as beams, [

c 1

(

c

].

I

(

c

]the axial tube load [

c

], the tube preload was applied to the model as an axial strain equivalent to the force associated with the tube elongation applied at the time of OTSG fabrication.

The static solution of the FE model was performed for each selected time point. The results were post-processed to glean and tabulate the key results relative to axial tube loads and associated tube hole dilations.

5.7 Structural Results Tube loads and dilations were determined for the case of a steam generator with no plugged tubes (0 %) and also for the case of a steam generator with 25% plugged tubes.

Two plugged tube configurations were considered. One configuration incorporated the 25% plugged tubes uniformly distributed over the tube bundle. The other configuration used 10% of the plugged tubes concentrated at the periphery of the bundle and the remaining 15% uniformly distributed'over the remainder of the tube bundle. The distributions chosen for analysis were determined to be representative of the existing I

plugging patterns of the OTSGs. The 0% and 25% plugged tube cases were concluded to be bounding for the OTSGs. It is noted that no OTSG is approaching having 25% of its tubes plugged.

The difference in the analyses of the plugged and non-plugged tube cases was the temperature imposed on plugged tube and the absence of primary pressure in the plugged I

tube. The plugged tube temperature was more representative of [

c

]

temperature of the SG while the non-plugged tube temperature was more representative of the [ c

].

Il h

25

BAW-2346NP j

I As mentioned above, the FE model was solved for each critical transient time point as determined from the thermal analysis. The key results are:

a) axial tube loads as a function of tubesheet radius b)

" tube-to-tubesheet hole" differential dilations as a function of tubesheet radius and c)

" tube-to-tubesheet hole" differential dilations as a function of depth into the tubesheet Using post-processing routines, these results were summarized for both the upper and lower tubesheets. Table 5-1 contains a summary of maximum axial tube loads and accompanying tube hole dilations at the primary face of the upper tubesheet for the subject transient conditions. The table includes the maximum axial tube load at any radial location and the corresponding tube hole radial and circumferential (tubesheet directions) delta dilations. Typically, the location of maximum axial tube load is at the bundle periphery.

Example values are plotted in Figures 5-3 and 5-4 for the 100% power steady state condition and the limiting time point for the MSLB transient. Similar type results for each transient are used as input parameters to the physical testing of tube-to-tubesheet joint mockups.

5.8 Conclusions To evaluate the adequacy of the tube-to-tubesheet joints, it was necessary to determme the structural response of the OTSGs when subjected to various transient loadings. For the tube-to-tubesheet joints, the primary measures of structural response were the axial tube loads and the accompanying tube hole dilations. These values were a function of the overall stiffness interaction of the tube bundle, tubesheets, secondary shell and upper / lower heads. A linear-elastic, axisymmetric finite element model was developed to represent these subcomponents of the OTSGs. This model was applied in the analysis of the controlling transients for both cyclic conditions (normal operating transients) and conditions of single load application (accident transients). Results indicate that the model was appropriate for the analysis of the general structural behavior of the OTSGs. It was concluded that the resulting axial tube loads and associated " tube-to-tubesheet hole" differential dilations were appropriate for subsequent use in the qualification of tube-to-tubesheet joints (i.e. TEC ARC).

i l

l i

26 1

l

BAW-2346NP

{

Table 51: Summary of Selected Results

(

Transient Maximum Maximum Accompanying

.t Axial Radial Circumferential Load Delta Delta

[

(Note 1)

Dilation Dilation L

(Notes 2,3 & 5) (Notes 2,3 & 5)

(lbs)

(mils)

(mils)

(

IIcatup, flooded Heatup, minimum level

(

Cooldown without hold Cooldown with hold Power change 0%-15%

(

Power change 15%-0%

Power loading 8%-100%

(c]

Power unloading 100%-8%

0% steady state power 8% steady state power

[

15% steady state power L

100% steady state power Bounding MSLB

~ Small Break LOCA limiting 25% plugged tube cases Heatup, minimum level Bounding MSLB

[c]

Small Break LOCA Notes:

1. Axial loads are the maximum for any location in the tube bundle.

2. Dilation values presented are for the primary face of the upper tubesheet.

[

3. The differential dilations reported are independent of any residual"springback" associated with the rolled-in joint.

4. The 25% plugged tube values are based on 10% plugging at the periphery of the SG with the remaining 15% equally distributed over the remained of the bundle. The dilations for the plugged tube case are [

c

] than those for non-plugged case because the effect of the plugged tubes [ c ] the " bowing"of the tubesheet and therefore the tubesheet hole dilations.

5. Negative delta dilations represent a tightening of the joint (i.e. the dilation of the tube OD is greater than that of the tubesheet bore).

27

i BAW-2346NP

[

Figure 5-3: Delta Dilation versus Tubesheet Radius

[

[

[

[C]

[

[

Figure 5-4: Axial Tube Loads versus Tubesheet Radius

[

[

[

[

[C]

[

[

[

(

{

r 28

I BAW-2346NP

- 6.0

SUMMARY

OFTEST PROGRAM 6.I' Determination ofInstallation Torque 6.1.1

Background

As mentioned in Section 2, the tube to-tubesheet rolled joints were installed using a torque-controlled process. The procedure used to determine the torque was common to all the currently operating plants. It required that samples of OTSG tubing be roll expanded into a test block, adjusting the torque to achieve an acceptable range of ID expansion.- The acceptable expanded tube ID was defined as being equal to the

[

c

]

['c J.

6.1.2 Summary of Tests In order to determine a representative roll torque, FTI installed 108 rolls in a test block that was fabricated with the same hole geometry'as the original fabrication test block.

These tests utilized three different tube heats of material, which represent the range of yield strengths installed in the operating plants. The installation torque ranged from [c]

to [c] in-lbs. The roll expander used was the same model as that used during the original shop fabrication. After the tubes were expanded into the block, the tuba ID was measured and the change in ID was compared to the acceptance criteria given above to determine if that combination of tube yield strength and installation torque satisfied the acceptance criteria.

6.1.3 Discussion of Test Variables A number of geometry and material variables were tested to insure that the torque used in the tubesheet joint mockup conservatively bounded those'of the original fabrication process. These variables are discussed below.

MaterialProperties of the Tubet A sampling of the Certificate of Material Test Repods representing over 167,000 tubea used in the fabrication of the B&W steam generators was performed. The results are as foGows:

Average Yield Strength = [ c ] ksi(room temperature)

Yield Strength Standard Deviation = [ c ] ksi The requirements for performing the mechanical property tests on alloy 600 tubing (SB-163) stipulated that the mechanical properties be tested at a minimum frequency of once every 5,000 lbs for each lot of material. Based on the density and geometry of the OTSG tubes, at least one test was required for every 400 tubes.

Based on this requirement,' the population of 167,000 tubes required at least 418 mechanical tests be performed. - A statistical two-sided 95/95-tolerance limit "k factor" for 400 samples is 2.1 Based on this information, there is a 95% confidence that 95% of the tubes had a tested yield strength of [

c

] ksi. Therefore, the mom temperature yield i

E-29 L

BAW-2346NP l

strengths of the tubes used in the fabrication of the steam generators ranged from a 957c l

lower bound of [ d ] ksi to a 957c upper bound of [ d ] ksi l

l For the subject mockup testing of the tube-to-tubesheet joint, three heats of tube material were used. The room temperature yield strengths of the materials used are:

heat # 93542-5001, Yield Strength = [ d ] ksi i

heat # 93381-5011, Yield Strength = [ d ] ksi heat # 93452-5060, Yield Strength = { d ] ksi The range of room temperature yield strengths used in the mockup testing is therefore

[ d ] ksi.

Based on the data provided above it was concluded that the mockup tube yield strengths adequately represent the range of tube yield strengths used in the operating SGs.

Material Properties of the Split Block: Based on available information, it was concluded that the original B&W split block was made from [

c

].

Although documentation of the actual material strengths of the split block were not found, the strengths [

-d

}

[

d

]. The ASME Code minimum room temperature yield strengths for the material used [

d

]

[

d

] ksi.

The room temperature yield strength of the [

c

] used for the subject mockup split block testing [ d ] ksi. It was concluded that the material used in the mockup testing was representative of the original split block used during fabrication.

The split block material yield strength utilized, along with the range of tubing yield strengths are considered to adequately represent the range of possible material properties used in fabrication of the SG.

Roll Expander and Expansion System: To minimize variability associated with tooling differences, the roll expander geometry and lubricant used in the subject mockup testing was identical to that used in the original fabrication of the SGs. The original roll system (tool) was not specified in the expansion procedure, so an [

c

] system was utilized for the mockup testing. Use of the [ c ] tool allowed for the monitoring and recording of i

the actual torque applied to each roll. Any potential differences in the way that the torque l

is measured or applied to the roll expander during the ARC testing compared to the original fabrication were considered negligible.

Tube and Test Block Geometry: The wall thickness of the tubing, the gap between the tube and the split block hole, and the ID of the split block hole all affect the amount of ID expansion for a given torque. The equation used to determine the amount of ID expansion included [

c

], [

c

] and the t

l 30

{

BAW-2346NP

[

c

]. The use of this equation in the calibration process was assumed to

(

address geometry variability.

The original expansion procedure did not specify how to measure the geometry of the tube and the split block. Different measurement instruments, techniques and locations could affect the dimensions used to determine the change in tube ID. It was assumed, however, that the combination of using accurate measurement equipment and utilizing regression techniques to trend the data (versus strictly using the raw data) would provide results representative of those obtained during the fabrication process.

b Number of Rolls: The original expansion procedure did not state if it was permissible to re-use the split block portion of the test block assembly. Multiple rolls in the same block p

could result in a slight increase in the ID of the split block hole (from test to test) and L

potentially increase the hardness / yield strength of the split block. This, in turn, could effect the amount of ID expansion for a given torque.

The subject mockup testing performed for this ARC re-used the same split blocks. Any change in ID of the holes was accounted for by measuring the ID after removing each tube sample and using that dimension in the calibration equation. Any change to the hardness / yield strength appeared in the data scatter and was therefore accounted for.

6.1.4 Evaluation of Data A total of 108 rolls were installed in the test blocks, with each of the three heats of tube material accounting for 36 rolls. The installation torque ranged from [

c

] in-lbs. A summary of all the torque test results in provided in Figure 6-1.

The data was then separated into three categories based on tube material strength and plotted for each of the three yield strengths tested. Linear regressions were then performed for each data set and the results were superimposed on the figures. The results showed [

d

], the torque required to achieve an acceptable roll [ d J. Based on the regression line, the range of acceptable torque was determined for each data set. The criteria for determining the acceptable torque values were the same as the criteria outlined in the original roll torque procedure. The criterion was based on the measured change of the inside diameter of the tube. This change in tube ID, referred to as " delta tube ID", was defined as:

delta tube ID = [ c ]

As previously stated, the criterion for the establishing an acceptable torque was that the calculated delta tube ID be [

c

]. Table 6-1 provides a summary of the acceptable torque ranges for the three tube material strengths tested.

L 31

BAW-2346NP Table 61: Torque Test Results Regression Summaries Material Yield Lower Torque Upper Torque Heat Strength Limit Limit Number (ksi)

(in-lbs)

(in-lbs) 93542-5001 93381-5011

[d]

93452-5060 Based on the data in Table 6-1, torques in the range of [

d ] in-lbs were acceptable for I

all three heats of material tested. This acceptable torque range is shown in Figure 61 along with the raw torque results. A review of the figure shows thirteen of the rolls installed in this torque range met the [

d

] inch A ID acceptance criteria.

l Four rolls in this torque range did not pass the criteria. Three failed low by [ d ] inch B

and one failed low by [ d ] inches. Based on measurement variability, these 4 rolls could be considered acceptable. In addition, [ d ] torque values are conservative for purposes ofleak and load testing. This validates the range chosen.

Figure 6-1: Summary of Torque Test Results I

I

[d]

I I

i 1

1 I

I 1

l 32 c

~

BAW-2346NP 6.1.5 Summary of Results/ Conclusion -Installation Torque The geometry and material strengths used in the mockup testing adequately bounded those of the original SG fabrication. Based on the testing described above, the minimum torque range that satisfied the acceptable expansion criteria for the range of tube yield strengths tested was [

d

] in-lbs. It was concluded an installation torque of [ d ] to

[ c ]in-lbs should be targeted when creating an " original" OTSG rolljoint for ARC leak testing.

6.2 Assessment for Bounding Joint Mockup The critical parameters for the OTSG roll expanded joint were evaluated to determine the effects that each could have on the finaljoint strength and leakage. Some of the factors were evaluated through testing while others were evaluated through analysis and engineering judgement.

The testing was conducted using multi-hole mockup blocks with various hole sizes, ligament widths, and material yield strengths.

Tubes of various diameters, wall thicknesses and yield strengths were rolled into the blocks and then load and leak tested to assess joint integrity. The range of geometries and material strengths used were selected to bound the range of those in the actual steam generator. The test results were

[

tabulated and the geometry that provided the weakest rolled joint (lowest pull out load and/or greatest leak rate) was incorporated into the production mockup.

A list of the parameters evaluated is provided below. Detailed discussions of each parameter are provided in the paragraphs following the list.

Variation in Rolling Torque Material Properties of Tubesheet and Tubing e

Variations in Tubing Dimensions (Wall Thickness) e Variations in Tubesheet Bore Diameter e

Variations in Tubesheet Ligament Thickness e

Variations in Tubesheet Bore Surface Finish e

(

Variations in Tubesheet Cladding (Presence or Thickness) e Effects of OTSG Stress Relief Effects of Axial Loading Effects of Flow Induced Vibration (FIV) Cycling Effects of OTSG Tubesheet Dilation Comparison of Operating Temperature vs. Room Temperature Testing Results Variation in Rolling Torque: Leak and load testing was performed on rolled joints that

[

had been installed at various torque values in a 14 hole mockup block. The targeted L

installation torque range chosen in Section 6.1.5 represents the lowest range of torque that pmduced an expansion that meets the original shop acceptance criteria. A summary of the test results is provided in the following table.

33

p i-BAW-2346NP l

j Torque Value Number of Leakage @

Pull Out Load l

Samples AP of[ d ] psi 3

(in /hr)

(lbs)

I d ]in-lb average results l

[ d )in-lb average results

[d)

[ d ]in-lb average results Results are the average of all samples with similar installation parameters.

[d]

[d]

Variations in Tubing Yield Strength: Leak and load testing was performed on rolled joints in tubes of two different yield strengths. The low ([ d ] ksi) and high ([ d ] ksi) yield tubing were evaluated as part of the testing of the OTSG tube-to-tubesheet joint.

j These tests, in both 14 and 38 hole mockup blocks, were performed to determine the i

effect that tubing yield strength had on the leakage and strength of the roll expanded joint. Both high and low yield tubing was installed using an equivalent torque. The results of this testing are summarized in the following table.

i l

Test Block Setup Number of Leakage @

Pull Out Load Samples AP of[ d ] psi l

(in /hr)

(lbs) 3 14 hole, low yield tube 14 hole, high yield tube

[d]

38 hole, low yield tube 38 hole, high yield tube Results are the average of all samples with similar installation parameters.

The results of this testing showed that joints fabricated with high yield tubing had

[

d

]. This was expected, since [

d

]

[

d

], into the tubesheet I

34 m

l l

BAW-2346NP I

bore. From these results,it was concluded that the mockups would be fabricated using

[

d J.

Variations in Tubesheet Material Yield Strength: Leak and load testing was pesformed on rolledjoints in tubesheets of two different yield strengths. The yield strengths of both I

blocks used in the parameter testing were [ d ] and [ d ] ksi. The minimum yield strength of the OTSG tubesheet was [ c

] ksi and is within the range of mockup tubesheet yield strengths.

Joints installed in the [

d

] of the two blocks (the 38 hole block) had significantly higher leak rates and [ d ] pullout loads than those installed in the other block (the 14 I

hole block). The average leak rate and joint strength data is provided in the following table.

Mockup Configuration Number of Leakage @

Pull Out Load Samples AP of[ d ] psi 3

(in /hr)

(lbs) 14 hole unciad average

[d]

(block oy = [ d ] ksi)

I 38 hole unclad average (block oy = [ d ] ksi)

Results are the average of all samples with similar installation parameters.

Based on the results of this test, it was concluded that the material used for the mockup blocks would have a yield [

d

] ksi as possible.

Variations in Tubing Dimensions (Wall Thickness): A thicker walled tube will [ d ]

the [

d

] of the tube which, as seen when comparing low and high yield tubing, will [

d

].

The high yield tubing that was used for testing (Heat 93452) had an OD that was

[d]

I

[

d

]

[

d

]. The wall thickness was also [

d

] than the [

d

] value

([

d

]). Therefore, the use of the [ d ] wall satisfied the conservative criteria. The use of an OD that was near the

[ d ], rather than the [

d

], was considered representative of the

[

d

]. Its use was justified by the inclusion of the remaining conservative l

properties inherent in the testing.

It was therefore concluded that [ d

] OD tubing with a wall thickness [

d

)

[ d ] would be used for leak and axial load testing.

L 35

BAW-2346NP Variations in Tubesheet Bore Diameter:

[d]

It was therefore concluded that [

d

] tubesheet bore diameter ([ d ] inch) woeld be used for testing.

Variations in Tubesheet Ligament Thickness:

l

[d]

It was therefore concluded that [ d ] tubesheet ligament (hole spacing of [ d ] inch I

compared to [ d ] inch) would be used for testing.

Variations in Tubesheet Bore Surface Finish: From testing performed on the 14 and 38 l

hole mockups, [

d

] conclusions about which type of bore finish produces the best results. Surface finishes from [ d ] (roughness height rating, I

smooth) to [ d ] (rough) were used for test purposes. While smooth tubesheet bore surface fmishes typically result in [

d

], their [

d

].

The differences between smooth and rough TS bores [

d

]. Therefore, I

the surface finish called out [

d

] was used in the fabrication of the test blocks with [

d

] on surface finish. The average leak and joint strength data from the parameter testing is provided in the following table.

Mockup Configuration Measure Leakage @

Pull Out Load I

of Surface AP of[ d ] psi 3

Finish (in /hr)

(Ibs) 14 hole average smooth bore 14 hole average rough bore

[d]

38 hole average smooth bore 38 hole average rough bore Results are the average of all samples with similar installation parameters.

Based on the results of this test it was concluded that the tubesheet bore surface finish would be [

c

).

Variations in Tubesheet Cladding (Presence of and/or Thickness): The purpose of this test was to determine the effect cladding has on joint leak rate and pull out strength.

Testing was performed on a 38 hole mockup block that was clad on one side and unclad 36

BAW-2346NP on the other. The results of this testing, listed la the table below, show that the performance [

d 1

This was attributed to the [

d

], relative to the

[

d

] on the mockup block. In the 38 hole mockup block, approximately

[ d ]of the effective roll expansion was in the [

d

] cladding.

Mockup Configuration Number of Leakage @

Pull Out Load Samples AP of[ d ] psi 3

[

(in /hr)

(lbs) 38 hole unciad

[d]

38 hole clad Results are the average of all samples with similar installation parameters.

I

!! was therefore concluded that [

d

] would be used on the g

mockup blocks. [

d J

B

[

d J.

Effects of OTSG Stress Relief:

I

[d]

It was therefore concluded that the mockups [

d j

[

d

],

[

d J.

Effects ofAxialLoading:

l

[d]

Effects of Flow Induced Vibration (FIV) Cycling:

[d]

37

BAW-2346NP It was therefore concluded that [

d

] to account for FIV loading.

Effects of OTSG Tubesheet Dilation: During OTSG normal operation the tubesheet holes dilate, while at main steam line break conditions the holes are [

d

]. As a result of this dilation, the tubesheet bore may pull away from the tube, which could result in an increase in leakage and a decrease in holding force. Therefore, as part ofjoint testing, the mockup must be loaded under various conditions to simulate the opening of the tubesheet bore.

It was concluded that [

d

] applied to the I

mockup. In addition,[

d

] the mockups would be [

d

]a

[

d

}

l Comparison of Operating Temperature vs. Room Temperature Testing Results: At operating temperature, the modulus of elasticity and the yield strength of the tube and tubesheet materials are reduced from the room temperature values. Also, there is a I

difference in the thermal expansion between the tube and tubesheet. To account for these differences, [

d

]

differences between room and operating temperature values.

It was therefore concluded that the effects of temperature on the joint interface pressure would be accounted for by [

d

] would be l

used in the mockup testing.

Summary of Bounding Parameters: Table 6-2 provides a summary of the conservative parameters that were determined from the parameter study to be used for the final qualification testing. Each of these items has been discussed in detail above. These parameters were chosen because they collectively result in a conservative estimation of leakage through the rolled joint.

I I

I I

I h

3

BAW-2346NP l

Table 6 2: Summary of Parameter Study l

Parameter FinalTest Configuration Rolling Torque

[d]

Tube Yield Strength

[d]

Tubesheet Yield Strength

[d]

Tube Dimensions

[d}

Tubesheet Bore

[d]

Tubesheet Ligament

[d]

Tubesheet Bore Surface Finish

[d]

Tubesheet Cladding

[d]

Mockup Stress Relief

[d}

Axial Fatigue

[d]

FIV Fatigue

[d]

Tubesheet Dilation (d]

l

~

Room Temperature Testing vs.

[d]

Operating Temperature 1

6.3 Description of Production Mockup I

I l

[c]

39

I i

BAW-2346NP The tubesheet block was mounted in a frame that was used to apply load to the mockup block. The frame had hydraulic jacks mounted on both the designated x-and y-axes to place dilation loads on the mockup block. The mockup block was attached to the frame and the hydraulic jacks using 1-1/8 inch diameter bolts. As pressure was applied to the jacks, a reaction load on the mockup block caused the block to " stretch" thereby dilating the holes. The hole dilations were controlled by the pressures applied to the jacks. The tubesheet block and mounting frame assembly are shown in Figure 6-2.

Figure 6-2: Mockup Assembly

[d]

Since the dilation of the test hole (hole #5) could not be measured after the test tube was rolled in place, it was necessary to develop an indirect measure. This was accomplished by [

d J. A " window" tube (measurement hole) was created by [

c

] such that a gap existed between the two sections near the mid thickness of the mockup block. As the 40

BAW-2346NP mockup block was loaded, the dilation [

d

] was measured by means of an

[

d

].

[d]

After this [

c

] was complete, an OTSG tube with an axial [ d ] wide 100% through-wall EDM notch was rolled into the test hole using a torque of

[

d

) in-lbs. The EDM notch was positioned so that one end was even with the cladding face. The other end of the notch extended into the tubesheet bore. Fifteen (15) total mockups were constructed, including five (5) each containing a [ d ] EDM notch of

[

d

] inches long. After installing the tubes, a fillet weld was placed on the test tube to simulate the tube-to-tubesheet weld on the actual OTSGs. For each test tube, the axial EDM notch was aligned [

d

].

This alignment conservatively maximizes the effect of dilation on leak rate. The entire mockup was then stress relieved at [

d

] hours to simulate the effect of the shop stress relief on the rolledjoint.

6.4 Mockup Conditioning 6.4.1 Dilation Cycling During operation of the OTSGs, the tubesheet holes undergo dilation cycling during various transients. These cyclic dilations could possibly affect the contact stress between the tube and the tubesheet in the rolled joint. In order to account for this possible effect, the tubesheet mockup was dilation cycled to simulate OTSG conditions. Each block was dilated [ d

] to the maximurn normal operating transient dilation determined in the analysis of Section 5.0. The limiting dilations occurred during the [

d

] transient and included a dilation of [ d ] mils in the radial direction (relative to the center of the SG) and a dilation of [ d ] mils in the circumferential direction (normal to the radial).

These [ d ] cycles of the normal operating ranges allowed the mockup to shakedown to an clastic action prior to leak testing.

L 6.4.2 Axial Load Conditioning in addition to tubesheet hole dilation, nonnal operating transients create axial loads in the OTSG tubes. These loads are imposed on the roll joint and could affect the integrity of the joint. In order to account for the effect of these loads, the mockups were subjected to axial load cycling to simulate the cumulative effect of up to 40 years of plant operation. The design transient load ranges were determined fmm the finite element analysis discussed in Section 5.0. The load ranges used for load conditioning were based on the design load 41 1

BAW-2346NP ranges with an adjustment (increased) to account for the number of mockup test samples.

The adjustment factor was determined in accordance with the criteria of Article I-10 of Appendix I of Section III of the ASME Code (1965 and 1968 Editions). Article I-10 allows the option to increase either the number of design cycles or the load range for application during mockup testing. A summary of the design load and cycles, adjustment factor used, and the dilation conditioning loads is provided in Table 6-3.

The design transients axial tube loads and design cycles were combined into three transient groups. These groups were created to bound all the axial tube loads and transient design cycles for the design life of the steam generator.

Transient group I was selected to include the range of maximum compressive and maximum tensile tube loads. The tube loads were bounded by the

[d]

Transient group 2 was selected to include the next largest load range. The load range was bounded by the loads associated with the

[d]

Transient group 3 was selected to bound all remaining transients. The load range was bounded by axial load associated with the

[d)

According to Article I-1080, the minimum number of cycles (or loads) which must be applied to the test samples shall be determined by multiplying the design service cycles (or loads) by a specified factor Kru (Krs). The procedure was followed and the resulting loads and cycles are summarized in Table 6-3.

1 42 s

Table 6-3: Load Conditioning Test Loads and Cycles l

Transient Design Load Design Factor Test Load Test Group Range (1)

Cycles Adjusted Range (1)

Cycles (lbs)

(lbs) 1 2

[d]

3 Notes:

1. The design loads shown are based on 0% tube plugging case. The test loads included an additional factor [

d

] as l

we!! as additional margin to account for potential instrumentation and measurement error.

2. Increase cycles by Appendix 1 factors.

I

3. Increase loads by Appendix I factors.

All axial fatigue cycling was conducted after imposing a [

d

] on the mockup. This corresponds to the worst case [

d

] transient I

determined in the analysis discussed in Section 5.0. This [

d

] was conservatively imposed because it was not practical to [

d

].

6.5 Description of Leak Tests The purpose of the leak testing was to determine leak rates for rolled joints with 100%

l through-wall axial cracks of various lengths for both 100% steady state power and I

bounding accident transient conditions. Leak rates were required at steady state power conditions so that the primary-to-secondary leakage expected during steady state I

operation could be determined.

In addition, leakage at accident conditions was determined to compare with allowable off-site dose leak limits.

[

The testing accounted for the effect of the difference in dilations between the tute OD I

and tubesheet bore resulting from the tested transient condition. All leak testing was performed at room temperature. The leak test parameters are summarized in Table 6-3 and discussed in the paragraphs that follow.

Pressure The primary-to-secondary pressure differences (AP) associated with steady state 100% power operation and the limiting accident condition (MSLB) were used for the mockup testing. Pressure was applied to the mockup tube ID using a hydrostatic pump. Pressure was applied such that the only leak path was through the EDM notch in the rolled joint. The steady state 100% power design AP was [ d ] psi. The required accident test pressure was [ d ] psi and was derived from the safety relief valve setpoint with a 3% allowance for the setpoint tolerance. The actual pressures tested included an 43

BAW-2346NP additional factor to account for potential inaccuracies in equipment or measurements.

The actual target differential pressures tested were [ d ] psi for steady state and [ d ] psi for accidents.

Arialleadt Steady state leak testing was performed without the direct application of an axial force. However, there was an axial tensile load created by the tube end cap being exposed to the test pressure. This net axial tensile stress was conservative when compared to the normal steady state tube loads. The normal steady state tube axial loads and resulting tubesheet bore dilations are [ d ] and in the [ d ] direction, and thus testing without taking credit for these [ d ] loads is conservative.

The bounding accident condition for leakage integrity is the [ d ]. [ d ] is limiting because it produces the greatest pressure differential across the tube-to-tubesheet joint and the greatest potential for primary-to-secondary leakage. During this transient, the tubes experience an axial tensile load due primarily to the temperature difference between the tubes and the OTSG shell. The load is greatest [

d J. The analysis of Section 5 determined the maximum axial load was [ d ] lbs at the [

d

] of the bundle for the [ d ] transient. This design load was increased for testing to account for potential inaccuracies in equipment or measurements. Therefore, the bulk of the testing was performed with an applied axial load of [ d ] Ibs. This maximum load bounds all other accident conditions loads currently defined for all the OTSGs. A limited number of tests were also conducted at a lower load of[ d ] Ibs to determine the effect of axial load on leakage. All accident testing was performed at the same delta-pressure.

Dilations: Biaxial loads were applied to the mockup block in order to dilate the tubesheet hole, simulating the effect of tubesheet bow during the transient conditions. As discussed in Section 6.3, measurements were made [

d

] the tube being leak tested based on [

d

] of each test block. For 100% steady state power, [ d ] dilation was applied. This is conservative because the actual delta dilation during steady state operation is [

d

] which would

[ d ] the joint interface pressure and therefore [

d

] the leak rate. For [ d ]

conditions, various dilations and aspect ratios were used so that the leak rate could be quantified as a function of dilation. This leak rate as a function of dilation was needed due the variation of dilations associated with the actual SG tubesheet.

i The dilations used in testing were not the absolute dilations of the tubesheet bore, but rather the delta dilation between the tubesheet bore and tube outside diameter. The applied tubesheet block loads and resulting dilations did take into account the tube dilations resulting from the test pressure.

6.5.1 Initial 12ak Tests Prior to performing axial cyclic testing, a small number of initial leak tests were performed. The leak tests were performed with 100% steady state power and [ d ]

pressure differentials across the tube. These tests provided a baseline leak rate for use in determining the effects of cyclic conditioning on the leak rate. The hole dilation, 44

BAW-2346NP hydrostatic test pressure, and tube axial load values applied during these tests are listed in l

Table 6-4.

6.5.2 Final Leak Testing i

The final set of leak testing was conducted by applying a range of axial load and tubesheet bore dilation values. Leak tests were performed with both 100% steady state power operation and [ d ] pressure differentials across the tube. The hole dilation, I

hydrostatic test pressure, and tube axial load values used for this testing are also listed in Table 6-4.

I I

I I

I I

I I

I I

J

[

45 I.-

BAW-2346NP Table 6-4: Matrix of Test Parameters Initial / Final Test Pressure Total Axial Load x-dilation y-dilation Leak Test I.D.

(i 20 psig)

(i50 lbs)

(i 0.1 mils)

(10.1 mils)

INITIAL A

l B

[d]

c I

FINAL D

E F

I o

H

[d]

I I

J K

I L

M N

I o

P I

NOTE:

1. These test parameters bound the actual design parameters determined by the analysis of Section 5.

I

2. The [ d ] lb axial load is the end cap load associated with a pressure [ d ] psi.

3. The [ d ]Ib axialload is the end cap load associated with a pressure of [ d ] psi.

4. The [ d ] and [ d ] Ib loads are the total of end cap pressure load and an applied l

axial force.

B

5. The dilations reported are the mockup bore dilations.

6. For relative comparison, the x-dilation corresponds to the "circumferential" dilation from the FE analysis of Section 5. The y-dilations correspond to the " radial" dilations.

6.6 Results of Mockup Leak Tests l

As discussed previously, testing involved the application of constant pressure while applying biaxial loads / dilations with an accompanying tube axial load. Test assemblies with axial notches of three different lengths were used. Measurements of hole dilations (biaxial) and leak rate were made for all leak test sequences. The results from each individual test are provided in Appendix A.

46

BAW-2346NP

[

Table 6-5 provides a summary of the pertinent results used in for the ARC. The pertinent

[

results include the average leak rate of the five mockups tested for each test condition (Test # as shown in Table 6-5). The average leak rate computed at a 95% confidence level is also provided. Application of the confidence level adds assurance that the f

average leak rate reponed herein would not increase if more samples were tested. See Appendix B for results and the method used to calculate the 95/50 limit.

Figure 6-3 contains a summary the average 95/50 leak rates for the three notch lengths tested as a function of delta dilation. The values shown are for the [ d ] psi pressure case defined in Table 6-5.

Figure 6-3: 95/50 Room Temperature Leak Rate vs. Delta Dilation (for bounding accident test pressure of[ d ] psi)

(

(

[d]

t

{

r 47

BAW-2346NP Table 6-5: Pertinent Leak Test Results I

Test EDM Pressure Axial Bore Dilation Leak Rate Length Load (mils)

(gpm)

(inch)

(psi)

(lbs)

X axis l Y axis average l

95/50 D

H I

I J

K L

I

~

O P

I D

H I

(d]

J K

I t

M I

O P

D I

I g

J B

K L

I O

P Notes:

1. Dilations are the mockup bore dilations, for tube-to-tubesheet bore delta dilations subtract the tube pressure dilation ([ d ]).

2. The average leak rate is the average of the five mockups tested at the specified conditions. Individual test results are provided in Appendix A.

3. 95/50 values are used for calculations ofleak rate vs. tubesheet radius.

4. Leak rate values come from Appendix B.

5. Leak rates are from room temperature testing.

48 I

BAW-2346NP 7.0 APPLICATION OF LEAK RATE RESULTS As previously stated, the key to this ARC was being able to postulate leakage for defined TECs. The desired leak rate output for the ARC was the leak rate (per indication) as a function of tubesheet radius. This relationship will allow each utility to determine the postulated leakage for an indication in any given tube under normal and accident type I

loads.

Leak rates were determined for steady state 100% power operation and also for the I

limiting accident condition transient, [

d

] plugged tubes. The [ d ] was limiting because it produced the greatest tube axial loads and tubesheet bowing, and also the greatest potential pressure differer.:e across the tube. The tubesheet bow reduces the I

rolled joint interface pressure, which in turn increases leakage. The combination of analysis results discussed in Section 5 and leak rate test results from Section 6 provided the necessary information for determining the required leak rates.

7.1 Method of Analysis Section 5 contains a detailed analysis of the behavior of the stcam generator under normal operating and accident loadings. The results of the analysis include tube axial loads and tube-to-tubesheet joint delta (A) dilations as a function of tubesheet radius.

The I

A dilations represent the net difference in dilation of the tube OD relative to the tubesheet bore dilation. This A dilation defines the interface pressure of the joint, which relates directly to the axial load strength and leak tightness of the rolled joint.

Section 6 contains the results of leak tests performed on mockup tubes with tube end I

cracks (simulated by 100% through-wall EDM notches). Leak rates were determined for three different crack sizes and various combinations of tubesheet bore dilation. The mockup tests were designed to bound the geometry and operating conditions of the operating 177FA plants.

The A dilation versus tubesheet radius was used with the leak rate versus A dilation to create a table and plots ofleak rates as a function of tubesheet radius (relative location of tube within the tube bundle).

l 7.2 Steam Generator Tube Loads and Delta Dilations The analysis of Section 5 provided the tube loads and A dilations associated with both l

100% power and bounding accident condition loads. The A dilation, or " contact interface" between the tube OD and tubesheet bore "ID", has a direct impact on the amount ofleakage across the joint. The greater the A dilation, the lower the joint contact pressure, and the greater the leakage. Table 7-1 contains a summary of the a dilations at the primary face of both the upper and lower tubesheets.

Values of radial and circumferential A dilation, relative to the center of the steam generator, are provided for both the 100% power and bounding accident condition ([

d

]

[ d ]). For reference, Figure 5-3 shows the variation of A dilation as a function of 49

___A

BAW-2346NP tubesheet radius for the upper tubesheet. Figure 5-4 provides a plot of tube load as a function of tubesheet radius.

Table 7-1: A Dilation versus Tubesheet Radius Limiting Accident ([ d ])

100% Steady State T/S A Dilation (mils)

Tube A Dilation (mils)

Tube Radius Upper T/S Lower T/S Load Upper T/S Lower T/S Load (inch)

Rad l Cire Rad l Cire (lbs)

Rad l Cire Rad l Circ (Ibs) 1 I

I I

[d]

I I

I I

I I

50

BAW-2346NP

{

Table 7-1: A Dilation versus Tubesheet Radius

[

Limiting Accident ([ d 1) 100% Steady State t

T/S A Dilation (mils)

Tube A Dilation (mils)

Tube Radius Upper T/S Lower T/S Load Upper T/S Lower T/S Load (inch)

Rad l Cire Rad l Cire (Ibs)

Rad l Cire Rad l Cire (lbs)

[

{.

[

{

[d]

[

{

[

L i

51

BAW-2346NP 7.3 Leak Rate versus Tubesheet Dilation The mockup testing of Section 6 provided the leak rate per crack (100% through-wall j

notch) as a function of A dilation. Leak rates for both normal 100% power operating pressure and maximum accident pressure were determined for various A dilation values and three different crack sizes.

Because of the conservatisms applied during development of the ARC (Section 4.2), the

" average" leak rate was used rather than maximum leak rate. The average leak rate is the average leak rate of the five mockup samples tested for each " test #"

The average rate was selected to get a representative leakage of the total population of indications rather than a maximum leak rate for a single indication. As discussed in Section 6 and Appendix B, a statistical 95/50 upper tolerance limit was then applied to the average leak rates. It was concluded that this statistical relationship would be used for assessing TEC leak rates. A summary of these statistical leak rates is provided in Figure 6-3 and Table 7-2.

It is noted that the dilations listed in Table 7-2 are the mockup tubesheet bore dilations.

They represent the dilation of the hole (tubesheet bore) and not the A dilation between the tube and tubesheet bore. The A dilation is obtained by subtracting the tube OD dilation associated with the tube's internal pressure from the bore dilation. The tube pressure j

dilation for the pressures tested were.

tube dilation = 2(P)(ID)(OD')/(E(OD -ID ))

2 2

OD = mockup bore diameter = [ d ]

ID = expanded mockup tube ID

=[ d j

[d]

P = internal pressure = { d ]

E = tube Modulus = [ d ]

Substituting the appropriate values into the equation yields:

tube OD dilation [ d ]

tube OD dilation [ d ]

These values of tube displacement are used with the bore displacements given in Table 7-2 to determine the final dilation to leak rate relationship provided in Table 7-3.

52

BAW-2346NP Table 7-2: Results of 95/50 Leak Rate Test Crack AP Axial Bore Dilation 95/50 Length Load (mils)

Leak Rate (inches)

(psi)

(Ibs)

X l Y (gpm)

D H

I J

K L

M O

P D

H I

J

[d}

K L

M O

P D

H I

J K

L M

O 6

P Note: Dilations are " bore" dilations and not tube-to-tubesheet bore A dilations.

Refer to discussion on previous page for explanation of the differences.

53

BAW-2346NP The leak testing resulted in [

d

] relationship between crack length and leak rate. It was therefore concluded that the [

d

] three crack sizes for each tested dilation would be used. The final results are presented in Table 7-3.

I Table 7-3: Summary of 95/50 Leak Rate Results (as a function of A Dilation and notch length)

I A Dilation Leak Rate (gpm)

(mils)

[ d ] inch

[ d ] inch

[ d ) inch Bounding X l Y long notch long notch long notch Leak Rate LIMITING ACCIDENT CONDITION I

I

[d]

I STEADY STATE 100% POWER I

As previously discussed, the leak testing performed and summarized in Table 7-3 was based on a tube with an EDM notch to represent a tube end crack. The notch was aligned

[

d

]. These maximum [

d

] orleast

[

d

] dilations (compared to the [

d

] dilations) resulted in the smallest contract pressure between the tube and mockup bore, and therefore the greatest leakage.

Therefore, the bounding leak rates as a function of the [ d ] dilation were used in Section 7-4 for determining leak rates as a function of tubesheet radius.

1

[d]

7.4 Leak Rate versus Tubesheet Radius To best support the implementation of the ARC, it was desired to represent the leak rates as a function of tubesheet radius. This will allow the utility to easily define the leak rate for identified indications based on their relative location within the tube bundle. The A dilations versus tubesheet radius described in Section 7-2 and the leak rate versus 54

_-.a

BAW-2346NP A dilation from Section 7-3 were used to calculate the desired leak rates as a function of tubesheet radius.

For the 100% steady state operating condition, Table 7-1 showed that the radial and I

circumferential A dilations calculated for the steam generator tube-to-tubesheet joint [ d )

[ d ] with relation to tubesheet radius. In addition, there was [

d

]

between the radial and circumferential dilations. A comparison of the steady state A dilations to the A dilations from the physical leak testing (Table 7-3) revealed that the

[

d

] than the calculated SG dilations. Therefore, the leak rates from the mockup testing were [

d

] for the steady state condition. Table 7-4 provides a summary of the steady state leak rate values for all l

tubesheet radii. Since [

d

],

the steady state leak rate [

d

] as a function of tubesheet radius. The leak rate values in Table 7-4 are representative of a single indication. For tubes with multiple I

indications, a separate leak rate for each indication must be used.

For the limiting accident condition, Table 7-1 did reveal that [

d

]

[

d

] tubesheet radius. As a result, the leak rates for the accident condition loading [

d

]. Table 7-1 also showed that the A dilations [

d J. Therefore, the l

[ d ] that resulted in the [

d

] was used to determine the leak rates as a function of tubesheet radius. The [

d

], and therefore largest leakage, occurs at [

d

]. For the cases where both the I

circumferential and radial dilations [

d

], the [

d

]

A dilation was used. The limiting A dilations from Table 7-1 were used to select the I

appropriate leak rate from Table 7-3. A summary of the limiting A dilations and resulting accident leak rates is contained in Table 7-5 and also in Figures 7-1 and 7-2.

A condensed numerical summary of the leak rates as a function of tubesheet radius is I

provided in Table 7-6.

The leak rate values provided are representative of a single indication. For tubes with multiple indications, a separate leak rate for each indication must be used.

NOTE: Plant specific assessments may be performed based on actual plant specific transient loads and resulting tube-to-tubesheet joint interface dilations.

To ensure I

compatibility with this ARC, the procedure outlined in the section must be used to determine plant specific leak rates. These plant specific leak rates may be used in place of the generic bounding values provided within. Plant specific changes to the leak rates presented in this ARC will be provided in a 10 CFR 50-59 licensing submittal.

55

I BAW-2346NP Table 7-4: Steady State Leak Rate versus Tubesheet Radius (per indication for steady state 100% pcwer) tubesheet upper tubesheet lower tubesheet radius limiting leak rate limiting leak rate A dilation A dilation (inch)

(mils)

(gpm)

(mils)

(gpm)

[d]

Table 7-5: Accident Leak Rate versus Tubesheet Radius (per indication for bounding accident condition) tubesheet upper tubesheet lower tubesheet radius limiting leak rate limiting leak rate A dilation A dilation (inch)

(mils)

(gpm)

(mils)

(gpm)

[d]

56

BAW-2346NP i

Table 7-5: Accident Leak Rate versus Tubesheet Radius (per indication for bounding accident condition) tubesheet upper tubesheet lower tubesheet radius limiting leak rate limiting leak rate A dilation A dilation (inch)

(mils)

(gpm)

(mils)

(gpm)

I I

I

[d) 1 I

I I

I I

See NOTE in Section 7.4 for potential use of plant specific data.

I I

57

BAW-2346NP Figure 7-1: 95/50 Leak Rates for Upper Tubesheet

[d]

I I

I Figure 7-2: 95/50 Leak Rates for Lower Tubesheet i

I I

l

[d]

I 58

]

1

BAW-2346NP Table 7-6: Summary of 95/50 Leak Rate vs. Tubesheet Radius Bounding Accident Condition (MSLB)

Upper Tubesheet Lower Tubesheet Radius Leak Rate Radius I2ak Rate (inch)

(gpm)

(inch)

(gpm) 1

[d]

I I

i 100% Steady State Power Condition I

[d]

Notes:

1. Radius = location of tube center relative to the center of SG.

2. See NOTE in Section 7.4 for potential use of plant specific data.

I I

I I

L)

J 59

BAW-2346NP I

8.0 EDDY CURRENT TECIINIQUE The Eddy Current (EC) technique used to suppon this ARC must be able to determine if I

an axial indication (in the tubing, near the tube end) protrudes into the portion of the tube adjacent to the carbon steel region of the tubesheet. A series of EC tests were designed g

and completed to establish a method for locating the clad-to-carbon steel interface and to l

determine the location of an axial indication relative to this interface. The thickness of the cladding will be known. Thus, bounding an axial indication at the clad-to-carbon g

steel interface will limit the length of the portion of that indication which is within the l

pressure boundary to the cladding thickness. This method of bounding the length of an axial indication is not dependent upon establishing an accurate axial scale for the EC data. This EC Section documents the EC tests performed in support of this ARC and 1

presents a method for applying the EC technique to an ARC inspection.

I The type of EC support for this ARC is different than typical NDE ARC support, in that neither the detection nor the sizing of indications is a part of the NDE development for this ARC application. To support this ARC, the developed NDE technique needs only to I

locate axial indications relative to the interface location. Existing qualified techniques will be relied upon for detection. Existing methods will be relied upon to determine indication orientation. Sizing of indications is not mquired. For the purposes of this I

ARC, the conservative assumption that any detected axial tubing degradation is 100 percent through wall will be made. Since Appendix H of the EPRI Steam Generator Examination Guidelines applies only to qualification of techniques for either detection or I

sizing, Appendix H does not directly apply to this locating application. However, the EC test was developed to be consistent with the intent and protocol of EPRI Appendix H.

8.1 EC Background Recent in-service inspections of once through steam generators (OTSGs) have found I

axial indications in the OTSG tubing near the upper tube end. These axial indications can be located in the tube end portion that protrudes from the tubesheet face, in the cladding region of the tubesheet, and in the carbon steel region of the tubesheet. The location of I

the axial indication determines the significance of the indication for application of this ARC.

I For either the pancake coil or the Plus Point coil, the rotating coil EC response from a tubesheet location such as the clad-to-carbon steel interface is not an abrupt and immediate response, but is rather a gradual change in response that occurs over a distance of probe travel.

As a rotating EC coil approaches a detectable flaw, the EC response from the flaw begins l

to form before the center of the EC coil reaches and coincides with the edge of the flaw.

l This effect is expected, since the field of the EC coil extends beyond the coil center, and "looks ahead" of the coil. This look-ahead distance is that distance from the first part of the flaw response to the pan of the response which corresponds to the physical location of the beginning of the flaw. Similarly, there will be a look-behind distance, which is the distance from the part of the response which corresponds to the other end of the flaw to 60 J

BAW-2346NP the last part of the flaw response. For a symmetric coil, such as the Plus Point coil or the f-pancake coil, the look-ahead distance and look-behind distance are expected to be equal for machined flaws of uniform depth. Thus, for large-amplitude through-wall flaws, the physical flaw location will lie within the EC flaw response. Flaw length-sizing methods typically use the entire EC signal to locate and length-size the flaw.

This is a conservative method for length sizing and locating flaws.

A detailed study of the EC response from the clad-to-carbon steel interface was required to support this ARC. A tubesheet block and a set of tube samples was built specifically for this ARC development, and the following test plan was developed to establish a reliable means for EC to identify interface locations and relative flaw positions.

8.2 EC Test A rotating pancake coil and a rotating Plus Point coil were the coil types used in the EC test.

The EC test plan was designed to locate axial indications relative to the clad-to-carbon steel interface in the tubesheet. To accomplish this, the EC test had three main objectives for each coil type. First, the area of the EC response which corresponds

(

to the clad-to-carbon steel interface was identified. Second, the look-ahead of the coils I

was verified. Third, a method was developed to determine whether or not an axial indication penetrates past the interface and enters the carbon steel region.

8.2.1 EC Test Samples

[-

Several tube samples and a tubesheet block were fabricated for the EC test plan. Some tube samples were rolled into the tubesheet block, and some were placed in plastic holders.

Three types of holes were placed in the tube samples and tubcsheet block. These holes were denoted as locator holes, tube holes, and interface holes. The three hole types were

{_

defined as follows.

[

[d]

{-

61

BAW-2346NP There were two main types of tube samples: the look-ahead tube samples and the locator l

tube samples. Figure 8-1 shows a diagram of the look-ahead tube sample with 20 %TW axial notches.

Figure 8-1: Look Ahead Tube Sample l

I

[d]

l l

l Each look-ahead tube sample had two axial EDM notches of equal design depth and I

length. [

d

] each EDM notch.

The [

d

] of the EDM notches. Thus, each look-ahead tube sample had two axial EDM notches and [

d

]. The

[

d

] of the notch ends in an EC response. Eight look-ahead tube samples were made and are listed in Table 8-1.

I Table 8-1: Look-Ahead Tube Samples As Built Number Type Of Axial Notches as built 1231223B 20 %TW OD notches I

as built 1231224B 40 %TW OD notches as built 1231225B 60 %TW OD notches as built 1231226B 80 %TW OD notches

]

as built 1231227B 100 %TW notches J

as built 12312288 100 %TW notches as built 1231229B 20 %TW ID notches as built 1231230B 60 %TW ID notches b

)

BAW-2346NP Each [ d ] tube sample consisted of a set of [

d

] holes (see Figure 8-2). The

[ d ] holes in the [. d ] tube samples were used to [

d

] the tubesheet

[.

d-

]. These tubes were positioned in the mockup block tube holes [

__ d

]

[

d

] ofinterest. There were seven [ d ] tube samples made, denoted by as built numbers 1231231B through 1231237B.

Figure 8-2:[ d ]TubeSample

[d]

63

BAW-2316NP l

The tubesheet block was made of carbon steel, with an alloy 600 cladding on two sides.

Both the top clad and the bottom clad were used in the EC test. Twenty-four tube holes

[

d

] were placed in the tubesheet block. There were [

d

]

holes for each of the [

d

]. Figure 8-3 shows a diagram of the tubesheet block.

Figure 8 3: Tubesheet Block g

[d]

I I

I I

I I

I i

BAW-2346NP Prior to rolling tube samples into the tubesheet block, the [

d

] locations were

[

d

] of the tubesheet block. [

d

] were made using a

[ d )in the tube holes of the tubesheet block.

I Figure 8-4 shows the view of the clad-to-carbon steel interface from inside tube hole I

number 3 of the tubesheet block. The lighter colored material at the top is the carbon steel. The darker colored material at the bottom is the cladding. The actual interface is

[

d

] throughout the tubesheet block. These are [

d

]

[

d

] which occur from the welding process. Note the initial interface [

d

]

the block is about [

d

]. Compared to this, the distance from the [

d

]

the [

d

] of the tubesheet block (not shown in Figure) appears to [

d

] in the region shown.

I Figure 8-4: CCI, View from ID of Tube IIole 3 I

i 1

[d]

I I

I I

The tests in this study use both tube samples which are rolled into the block and tube l

samples which are moveable in the block. Figure 8-5 shows the assembly map of the tubesheet block. Table 8-2 lists the tube samples which are rolled into tubesheet block.

The tube samples which end in a -0## are clean tube samples. In Figure 8-5 and Table 8-2, the tube samples which end in a 23#B are [

d

] samples.

I L

65

BAW-2346NP Figure 8-5: Assembly Map of Tubesheet Block i

I

[d]

I I

I I

I Table 8-2: Assembly of Tube Samples Into Tubesheet Block Tube Hole Tube Sample l Top Side Protrusion l

[d]

1 1

2 3

4 g

l 5

6 8

[d]

I 9

10 l

11 B

12 13 J

14 15 16 t

19

[

r L

66

l 8.2.2 EC Test Fixture I

Some of the acquisitions in the EC test required aligning or offsetting a [ d ] in a moveable tube sample with an [

d

]in the tubesheet block. The moveable tube samples were held in place relative to the tubesheet block using the XY-positioning fixture, shown in Figure 8-6.

This fixture accurately moved the tube relative to the tubesheet block in 0.001 inch increments. [

d

]

{

d

]

[

d

]

Figure 8 6: XY-Positioning Fixture (Places Samples At Precise Locations within Tubesheet Block) 4.pyyg3 g

9.

wt I

,i E

a y

4 I

q

~,

7 f

n Q

w ny e s

yM o;

g

[

y t

&\\

6 E

ma wh~

a 4

p

~;

y 3 9,;;,

s-1 Depending on which tube hole was selected, a free floating tube sample had a clearance

~

space between 0.004 inch and 0.009 inch of radial distance between the tube OD and the tube hole ID. This small space between the tube and the tubesheet had no measurable effect on determining tubesheet locations and relative flaw locations. The EC test used both rolled and non-rolled tubing at a clad-to-carbon steel interface, and confirmed that the absence of the roll had no effect on determining tubesheet locations or relative flaw

~

positions. The EC test also had two types of non-rolled tubing at a clad-to-carbon steel interface. The first test type was a tube which was permanently held in place by a roll expansion in another location. The second test type was a moveable tube which was temporarily held in place by an XY-positioning fixture. The EC tests showed that the

{

XY-positioning fixture accurately positioned and held the tube samples relative to the

[

d

] the tubesheet block.

67 i -

BAW-2346NP:

8.2.3 EC Test Acquisition The EC test plan implemented to support this ARC had four parts: an Interface Location Test, a Coil Look-Ahead Test, a Relative Flaw Location Test, and an Orientation And Extension Test. All of the data was acquired in accordance with an examination technique summary sheet (ETSS) such as the ETSS in Appendix A. Data was acquired in both the push and pull directions, at both 0.1 ips and 0.2 ips speeds. [

d

]

[

d-

]. Data was stored on FTI in-house optics.

The interface location test identified the areas of the EC responses which correspond to the location of the clad-to-carbon steel interface. Data was acquired for six tube holes in the tubesheet block. Each tube hole had two clad-to-carbon steel interfaces, one on the top side of the block and one on the bottom side of the block. Four clean (unflawed) tubes were roll expanded into the tubesheet block (tube holes 1, 2, 3, and 4) with a 0.187 inch protrusion from the top side of the block face. [

d

]

[

d

]

[

d

]

[

d

]

[

d-

]

[

d

]

[

d

}

[

d

].

The coil look-ahead test measured the axial distance from where the coil first responded to an axial notch to where the notch actually began. Tube samples were made with a -

variety of notch depths; then: were OD notches-(20 %TW, 40 %TW, 60 %TW, and 80 %TW), through wall notches, and ID notches (20 %TW and 60 %TW). The tube samples [

d

]

[

d

]

[

d

]

[

d

] The look-ahead distance was plotted against both flaw depth and EC response magnitude. The eight tube samples which were used for the coil look-ahead test were as built numbers 1231223B through 1231230B.

The relative flaw location test placed axial flaws at different locations relative to the clad-to-carbon steel interface. This demonstrated the ability of the rotating coils to

. conservatively disposition a flaw as either being only within the cladding or into the carbon steel. [

d

]

[

d

}

[

d l

[

d

]

[

d

]

[

d

)

[

d j

[-

d

]

68

BAW 2346NP b

f d

j

{

f.

'd j-f d

]

I d

I

[-

The orientation and extension test verim y 7 d.-

d

)

f.

d

_]

(

d f.

d J,

8.3

' EC Test Results The EC test data was analyzed, and a method for locating the clad-to-carbon steel interface was determined. The EC responses from the tubesheet mockup were like those from in-service tubesheets. The look-ahead for each of the coils was measured. The analysis technique was used to determine whether or not a flaw was in the carbon steel region of the tubesheet. A data acceptance criterion was also established.

(-

(

(

(:

[.

L 69

I l

BAW-2346NP 8.3.1 EC Interface Location Data from the interface location test was analyzed to identify the areas of the EC responses which correspond to the clad-to-carbon steel interface. The EC response from each of the four frequencies (300 kllz,200 klfz,100 kHz, and 20 klfz) for each of the three coils (0.115 inch Pancake, Plus Point,0.080 inch Pancake) was evaluated. The goal of this test was to find a clearly identifiable pattern in the EC response from the interface which corresponded to the interface location. The location of the clad-to-carbon steel interface [

d

]. Figure 8-7 shows the

[

d

] shown in the lissajous. For each coil, the responses from the four frequencies were compared to determine the most reliable frequency for that coil. The best frequencies for each coil were then compared to determine the best channel for locating the clad-to-carbon steel interface.

Figure 8-7: Center of[ d ] Shown in Lissajous i

I

[d]

I I

I

.I

-M 70

BAW-2346NP The pancake coil responses to the clad-to-carbon steel interface were shaped like [ d ].

On the higher frequencies, the location of the interface was near the [

d

].

As the frequency decreased, the interface location [

d

]

[

d

]. Figure 8-8 shows the 0.115 inch pancake coil responses from the four frequencies.

Figure 8-8: Pancake Coll Response from the CCI I

I

[d]

I I

I I

I I

l l

J 71

BAW-2346NP For both pancake coils, the best frequency for locating the clad-to-carbon steel interface was [

d 1

[

d

]

Based upon EC data analysis, it was estimated that locations could [

d

] of data when using a pancake coil to locate the interface.

Figure 8-9 shows the location of the interface on the [

d j pancake channel.

The arrow near [

d

] corresponds to the interface location, which was

[

d J.

Figure 8 9: C. Scan of CCI On [ d ] Pancake Channel 1

1 I

I

[d]

I I

I I

I

]

[

L 72

BAW-2346NP The Plus Point coil has [

d

] response in the interface region. The interface occurs [

d

] response on each of the four Plus Point frequencies. Figure 8-10 shows the four frequencies of the Plus Point coil at the interface location.

Figure 8-10: Plus Point Coil Response from the CCI I

I l

[d]

1 1

I I

I 73

BAW-2346NP

\\

For the Plus Point coil, the best frequency for locating the clad-to-carbon steel interface l

was [

d

]. This [

d

] frequency had the interface located [

d

]

l

[

d

] response. Picking a location by finding the [

d

]

response was concluded to result in consistent locations. Based upon EC data analysis, it was estimated that locations would [

d

] when using a Plus Point I

coil to locate the interface. Figure 8-1I shows the location of the interface on the [ d )

[

d

] channel. The arrow [

d

] corresponds to the interface location. Note that [

d

] is located at the

[

d

].

I Figure 8-11: C-Scan of CCI on [ d ] Plus Point Channel i

I I

[d]

I I

I I

Thus, any of the [

d

] of the [

d

] can be used to locate the l

clad-to-carbon steel interface. The most accurate method of locating the interface uses the [

d

]. The interface is located in the [

d

]

response on this channel. This [

d

] method is repeatable and objective, unlike the

[ d ] method which requires a [

d

] to select a location near the

[

d

].

8.3.2 Coil Look-Ahead Coil look-abead was measured for the pancake coils and the Plus Point coils for a variety of flaw depths. For each flaw and for each coil, there was always a measurable look-ahead distance. Look-ahead varied depending [

d

]. For the chosen method of measuring look-ahead, [

d

]

74

l BAW-2346NP distance than the [

d

]. Coil look-ahead will be depicted in the next section, which shows the method for locating flaws relative to the clad-to-carbon steel interface.

8.3.3 Flaw L.ocations l

The data acquisitions made for the relatire flaw location test contained axial flaws at different locations relative to the clad-to-carbon steel interface. The [

d

]

[

d

] was used to locate the interface in the EC response. The [

d

]

I

[

d

] were used to locate the edge of the flaws. A repeatable method for identifying the end of a flaw was established for both the Plus Point and the pancake coils.

I In all of the data acquisitions for flaws which were positioned at the interface or beyond the interface into the carbon steel region, both pancake coils and the Plus Point coil

[

d

]

l

[

d

]

There were cases in which flaws that were located within the cladding region and 0.050 inch or 0.100 inch away from interface were [

d

}

[

d

].

I I

I I

I I

I

[

c L

75

BAW-2346NP Figure 8-12 shows the EC response from a flaw that was positioned entirely in the cladding region, with the edge of the flaw 0.050 inch away from the clad-to-carbon steel interface. Because of the [

d

]

[

d

).

I Figure 812: A Flaw 0.050 Inch Away From The CCI [ d ]

I I

I I

l

[d]

l I

i 1

E

[

[

76

i BAW-2346NP Figure 8-13 shows the EC response of a flaw which was positioned [

d

]

[

d

). The cursor is located at the edge of the flaw response, which EC indicates [

d

].

Figure 8-13: A Flaw At The CCI[ d ] Carbon Steel Region

[d]

i The test's method of locating flaws relative to the clad-to-carbon steel interface was successful. Using the [

d

] to locate the interface and [

d

]

[

d

] to locate the flaws identified any flaw which was [

d

]

[

d

)

[

d

] In some cases, flaws which were [

d

] were

[

d

]. The amount of

[

d

] will depend upon the [

d

].

l 8.3,4 Data Acceptance The method for locating flaws relative to the clad-to-carbon steel interface does not require an [

d

]. With the currently used methods of data acquisition, some stalls orjerks in the axial probe motion can occur. Some stalls or i

jerks in probe motion [

d

)

indications are [

d

] in the EC responses. The use of the probe

~

extension [

d

]

[

d

)

77

BAW-2346NP Any data record which contains stalls or jerks [

d

' ] must be rejected. To apply these test results during field applications, the EC analyst must

.[

d

] was acquired in the regions ofinterest

. for each data record.: A good data record can be verified by [

d

]

.[

d J. Good Plus Point data will have the characteristic [

d

]

.[

d-

) in the interface region as shown in the previous EC Figures.

Two speeds,0.1 ips and 0.2 ips, were used in the test. Each speed had a sampling rate of 400 samples 'per second. Thus, the data acquisition resolution for the 0.1 ips data was twice that of the 0.2 ips data. [

d

]'

conservatively locate flaws. The [

d

] will increase data resolution and may be used if[

d

] the interface location.

8,4 ARC EC Application An ETSS such as the ETSS in Appendix A will be used to acquire and analyze EC data for this ARC application. A [

d

] will be used to locate the clad-to-carbon

[

d

}

[

d

)

[

d

] These techniques will be used as described in the ETSS and in the EC Test Results section above. Channels not used in the location of the clad-to-carbon steel interface or the location of flaws may be eliminated. Alternate channels numbers may be used to support multiple probes or other setups.

There are two conservative EC aspects of this ARC application: the depth assumption and the method of flaw location. The depths of the flaws are not sized by EC, and any axial tube degradation is assumed to be 100 percent through the wall of the tubing. This location technique will [

d

]

[

d

)

[

d

]

[

d

]

l

[-

d

]

NOTE: Other inspection techniques may be qualified in the future for use with this ARC.

78

BAW-2346NP 9.0 TEC GROWTH ASSESSMENT A growth rate analysis is necessary to determine if an indication is likely to grow between inspection periods. For this ARC, the growth of the indication is measured relative to the clad-to-carbon steel interface.

The ARC may be applied to axial indications that are determined to remain within [ d ) inches of the face of the cladding I

during the subsequent examination cycle.

9.1 Introduction I

The ARC developed in this report utilizes location of the indication in relation to the clad-to-carbon steel interface as a measure ofleakage and structural integrity. As a part I

of any tube integrity assessment, the growth of tube degradation over time must be addressed to ensure that indications that remain in service over an inspection interval will not exceed structural or leakage limits. Therefore, rotating coil (RC) inspections from I

consecutive operating cycles were evaluated to assess growth rates with respect to the CCI for TECs. Specifically, the data from tubes inspected during the [ d ] and [ d ]

outages at [ d ] were reviewed in order to quantify the magnitude of change in the I

position of the indication relative to the clad-to-carbon steel interface. Plant specific data will be evaluated during future outages to ensure that the overall apparent growth at a given plant is consistent with that determined in this report.

9.2 Methodology The objective of the historical review was to assess the progression of the indications in the cladding region of the tubesheet rolled joint. The 0.115 diameter mid range pancake I

coil data of the upper tubesheet for the last two outages for [ d ] (1996 and 1998) for these indications were re-evaluated. The primary goal of the analysis was to determine if any indications grew from above the interface to below the interface over the duration of I

the cycle. The cycle length was [ d ] effective full power years (EFPY). The secondary goal was to quantify the change in proximity of the indication to the interface.

9.3 Evaluation ofIndications in the Cladding Region For this re-evaluation,156 indications were reviewed from the last two outages at I

[ d ]. A selection of upper tube end indications (single end anomalies and multiple end anomalies) reported during both outages were compared side-by-side using the 0.115 MR pancake coil to determine the signal characteristics of amplitude, length, position relative I

to the clad-to-carbon steel interface and number of indications. The 0.115 pancake coil was used since it was the common inspection coil for both outages.

I The axial scale was set based on the clad thickness, which had been determined to

[ d ] inch at [ d ]. This scale was used to measure the axial distance between the lowest extent of the axial indication and the clad-to-carbon steel interface signal.

h 79

1 1

BAW-2346NP 9.4 Results From Growth Gate Study 9.4.1 Distribution of Indications Relative to Clad-to-carbon Steel Interface The growth rate is based on the relative distance from the tip of the indication closest to the interface (designated as the Flaw Tip Location, FTL) to the clad-to-carbon steel interface. The distribution oflocation of the FTL relative to the CCIis shown graphically

(

in Figures 9-1 and 9-2 for SG-A and SG-B respectively. A positive number means that the tip of the indication was above the CCI. As seen in the graphs, all indications re-evaluated were located above the CCI for both outages.

9.4.2 Change in Distance Relative to Clad-to-carbon Steel Interface From the re-analysis, it was noted that none of the indications extended below the cladding interface into the carbon steel in either outage. However, the measured distance between the FTL and the CCI was noted to change between the two outages. The

[

distribution of this change is presented in Figures 9-3 and 9-4 for SG-A and SG-B respectively. Note that " growth" in this respect would be represented as a decrease in the g

distance between the FTL and the CCI. For this reason, the distribution of the change in i

measurement is presented as the 1996 measurement minus the 1998 measurement. Thus, a positive number represents apparent growth (meaning the FTL was observed to be closer to the CCIin 1998 than in 1996).

9.4.3 Statistical Evaluation The data for the distance from the tip of the indication to the CCI for steam generators A and B, and for 1996 and 1998 were acquired as discussed in Section 9.3. It was assumed that a difference in the mean value of the SG-A and SG-B populations from 1996 to 1998 could be inferred to be the growth of the indication toward the CCI.

[

Each of the four populations of data was verified to approximate a normal distribution by using a Chi-squared goodness-of-fit test. With the assumption of normality, a null hypothesis (Ho) was proposed for the mean of the difference (pn) between the 1996 and 1998 measurements for each steam generator.

Ho: po = do (do = 0)

The analysis was performed under the condition that the mean difference in each SG from 1996 to 1998 (i.e., inferred growth) was zero (do = 0).

b-For steam generator B, the null hypothesis of a = 0 was not rejected at a 57c significance level. For steam generator A, the null hypothesis of po = 0 was rejected at a

{

5% significance level, but the null hypothesis of o = 0.01 was not rejected at a 57c significance level.

b 80

BAW-2346NP 9.5 Growth Rate Results/ Conclusions j

From the perspective of this ARC, the length of the indication in the cladding is not of significance, as long at the identified indication does not extend past the CCI. The leak rate data for this ARC has assumed that the crack may be as long as [ d ] inches.

Therefore, growth of the indication is allowed as long as the propagation of the indication does not extend past the maximum length of [ d ] inches from the face of the cladding.

From the re-analysis of ANO-1 data from two consecutive outages, it was shown that the j

FTL relative to the clad-to-carbon steel interface did not show any significant growth towards the interface. The change in location from the FTL to the CCI is normally distributed around zero change in location. As shown in Appendix D, for SG-A, the implied growth is [ d ) inches per EFPY ([ d ] inches per [ d ] EFPY). For SG-B, the implied growth is [ d ] inches per year. Considering that the testing allowed for a growth of [ d ] inches ([ d ] inch tested EDM length minus [ d ] inch bounding clad

)

thickness), the implied growth from the statistical evaluation is well within an acceptable margin. Based on the results of this study, it is concluded that the growth rate of the indications toward the CCIis [ d ], and therefore [ d ] leak rate needs to be made to account for growth over an inspection cycle.

In order to assure that this conclusion is applicable to all plants that apply the ARC, the plant specific growth rate of the population of TECs remaining in service must be evaluated and monitored each outage. If this new data indicates at any time that growth towards the clad-to-carbon steel interface has occurred, the plant specific growth rate must be compared to that calculated within. If the plant specific growth rate exceeds the value calculated within,its effect on the ARC must be evaluated.

l l

81

BAW-2346NP Figure 91: Distance ofITL from CCI,SG A f

{

[d]

i Figure 9 2: Distance of FTL from CCI

-B f

}

[d]

[

[

{

82

{

BAW-2346NP Figure 9-3: Differences in Flaw Tip Location, SG-A

[

{

[d]

Figure 9-4: Differences in Flaw Tip Location, SG-B

[

[d]

{

i

{

rt 83

l 10.0 PROBABILITY OF DETECTION For this ARC, the frequency distribution of TEC indications (based on radius location) found during EC inspection will be scaled upward by a factor of 1/ POD to account for non-detected cracks which could potentially leak during accident conditions during the I

next cycle of operation. The probability of detection reflects the ability of the inspection method to detect all of the tube end cracks that exist in the steam generator tubing. The g

adjusted population, minus detected indications for tubes that have been plugged or l

repaired, should constitute for pumoses of the tube integrity analyses the total number of indications remaining in service. The assumed distribution by radius can be expressed as:

Niraaius = [(l/ POD)(N s Found) radius]-[(Nacpain.a)raaiud1)]

A where:

I NIraaius = estimated number ofindications at given radius bin (NA.s Found ) radius = number of indications actually detected at given radius bin I

(Nacpairca) radius = number of repaired indications at given radius bin POD = probability of detection of TECs The recommended method for inspecting the roll expansion zone is the rotating coil. An RC inspection is accomplished using a surface riding coil that is rotated around the tube l

axis. The coil is spring-loaded to maintain contact with the tube inner surface as it moves I

through the expansion region. Gaps between the coil and the tube surface caused by a change in tube diameter are significantly reduced compared to bobbin coil.

As discussed in Section 3.2, the damage mechanism responsible for the TECs is believed to be PWSCC. The MR Plus Point (ETSS 96508) technique has been qualified per I

Appendix H of the EPRI Steam Generator Examination Guidelines for detection and sizing of axial PWSCC. The POD for this technique is [ d ] based on a 90% confidence for cracks greater than 50% through-wall. The 0.115 diameter pancake coil (ETSS l

96502) technique has also been qualified per Appendix H. The POD for this technique is I

[ d ] based on a 90% confidence for cracks greater than 50% through-wall.

I NOTE: Other inspection techniques may be qualified in the future for use with this ARC. The qualification process will determine the appropriate POD for application with TEC inspections.

Applying POD As defined in Section 7, leakage is a function of tubesheet radius. Therefore, the number ofindications found during the EC inspection, for each radial zone (bin) defined in Table 7-6, must be increased by the inverse of the POD. This resulting number of indication must be multiplied by the associated leak rate to obtain the total leak rate for the zone.

The total leak rate from all zones for each SG must be compared to plant specific allowable leak rates.

84 s

BAW-2346NP 11.0 STRUCTURAL EVALUATION OF TUBE-TO-TUBESIIEET WELDS In order to evaluate the structural integrity of the tube-to-tubesheet weld under normal

(

operating and accident loads, a detailed finite element analysis was performed. The analysis considered both the original fabrication fillet weld and the flush weld used to repair tube end damage at [ d ] and [ d

]. The analysis evaluated the welds for maximum stress and fatigue resulting from normal operating transient conditions and also for maximum stress associated with accident (faulted) transient conditions. The analysis considered all loading mechanisms including primary and secondary pressures, tube axial loads, and weld (tubesheet bore) dilations resulting from bowing of the tubesheets.

The results and conclusions of the analysis showed that both the original fabrication fillet weld and the flush repair weld met the requirements of ASME Code Section Ill for Class

! components for all loads associated with both normal operation and accident loads. A brief summary of the analysis is provided in the following sections.

I 1.1 Introduction Axial cracks in the rolled joint region of the SG tubes could potentially reduce the load carrying ability of the rolled joint. The diminished load strength of the joint may require the tube-to-tubesheet weld to resist higher loads. In order to allow axial indications to remain in service, the weld must be shown capable of withstanding the higher loads.

Since the reduction in load strength due to the presence of axial cracks has not been quantified, a detailed analysis of the weld was performed conservatively assuming the weld experiences the full range of tube loads (i.e. the roll has no load strength). The detailed analysis was performed using finite element techniques as implemented by the ANSYS software program.

I 1.2 Finite Element Model An axisymmetric finite element model with " harmonic loading" capability was constmeted to determine operating and accident condition stresses in the weld. The harmonic element option allows the user to impose nonsymmetrical boundary conditions on an axisymmetric model. The option was required to accurately represent the nonsymmetrical tubesheet bore (and weld) dilations. The model consisted of the weld, a portion of the tube, and the weld-to-tubesheet interface.

85

BAW-2346NP 11.3 Transients Considered l

The transients considered in the analysis of the weld were identical to those used in the analysis of the general structural behavior of the SG, Section 5.0. The transients included the significant normal operating transients as well as the faulted accident transients. The transients analyzed included:

a) Heatup (HU) from 70F to 8% power b) Cooldown (CD) from 8% steady-state power c) Power Changes from 0% to 15% power and 15% to 0% power d) Plant Loading / Unloading from 8% - 100% power and from 100% - 8% power e) Steady state conditions at 0%,8%,15% and 100% power levels f) Main Steam Line Break g) Small-Break Loss-of-Coolant Accident The selected transients included those representing the controlling design basis transients for consideration of cyclic loads (fatigue) as described in the plant Functional Specifications. Also included were the controlling transients for consideration of single accident load application (faulted loading).

l 11.4 Boundary Conditions The structural boundary conditions imposed on the finite element model included dilations due to tubesheet bowing, tube axial loads due to interaction of the tubes and shell, and primary and secondary pressures. Load cases were executed for each of the transients defined in Section 11.3. The loads for each transient came directly from the detailed finite element analysis of the general structural behavior of the OTSG, Section 5.0.

Dilations: The nonsymmetrical dilations of the tubesheet bore and weld were imposed on the model by use of nodal displacements. Displacements were imposed on the tubesheet face-to-weld junction using a Fourier series. The procedure allowed the I

application of the oval displacement of the tubesheet bore predicted by the analysis of Section 5.0.

Axial Tube Load The tube axial loads were imposed directly on the modeled tube end by means of an axial force. The applied force represents the effects of axial loads associated with differential thermal growth between the tubes and shell, the original fabrication installed preload, and the axial load resulting from the Poisson's effect from pressure.

Pressure: Primary and secondary side system pressures were also applied to the model.

The pressures were imposed by using an effective primary-to-secondary pressure differential applied to the tube ID. The applied delta pressure was equivalent to applying separate primary and secondary pressures to the tube ID and OD.

L 86

BAW-2346NP 11.5 Analysis Results Orieinal Fabrication Fillet Weld:

The calculated maximum stress intensity range for the normal operating transients and the resulting fatigue usage factor, which incorporated a fatigue strength reduction factor I

of four, were compared to ASME Code Section III allowables. A summary of the results is provided below.

I Normal and upset. or erating conditions:

maximum primary + secondary stress intensity:

SImaximum Range = [ d ] ksi < allowable of[ d ]

Fatigue usage factor:

Ummi = [ d ] < allowable of[ d ]

The calculated maximum primary stress intensities for the accident condition loads were compared to the allowable stress intensities defined in Section III Appendix F of the ASME Code. A summary of the results is also provided below.

Maximum accident (faulted) condition:

I maximum primary stress intensity:

SImaximum = [ d ] ksi < allowable of[ d ]

l Repair Flush Weld:

The geometry for the repair consisted of a semicircular weld with a minimum throat dimension of approximately [ d ] inch. This geometry differs from the production fillet weld geometry previously discussed in that the repair weld joins the top of the tube to the adjacent " flush" clad face of the tubesheet. The analysis method used for the flush weld was identical to that employed in the evaluation of the original fabrication fillet weld described in the preceding paragraphs.

The most severe transient case, as defined by the highest calculated stress intensity for l

the production fillet weld, was utilized for comparison purposes. The repair weld was evaluated using the bounding [ d ] transient temperatures, differential pressures, and tubesheet hole dilations used in the assessment of the fillet weld previously discussed.

The results of the comparison are provided below.

Original fillet weld maximum SI = [ d ] ksi l

Repair flush weld maximum SI = [ d ] ksi The comparative assessment of the maximum stresses showed that the calculated stress intensities for the repair flush weld [

d

] those calculated for the production fillet weld. It was therefore concluded that the analysis results for [

d

]

tube-to-tubesheet [

d

].

I I

87

BAW-2346NP 11.6 Conclusions Based on the results of the finite element analysis, both the original fabrication fillet weld and the repair flush weld meet the requirements of Section III of the ASME Boiler and Pressure Vessel Code for the normal operating and accident load conditior.s. Both the tube-to-tubesheet original fabrication fillet welds and repair flush welds are structurally capable of resisting the loads associated with both normal operational transients and accident conditions. Furthermore, this conclusion can be achieved without assistance from any frictional forces that may be available from the mechanically roll expanded tube to-tubesheetjoint.

The analysis and conclusions provide a strong basis for not requiring the roll expansion to carry any portion of the tube loads. Therefore, leaving axial indications in the rolled region of the tube will not adversely affect the structural integrity (axial load strength) of the tube-to-tubesheet joint.

I I

I I

I I

I I

I m

88 a

BAW-2346NP 1

12.0 PLANT SPECIFIC IMPLEMENTATION The following general steps will be followed when applying the ARC:

Site specific measurements of the tubesheet cladding thickness, as outlined in Appendix E, are required in order to verify that the tested geometry bounds the i

actual plant condition.

The measured clad thickness must be less than

[ d ] inches in order for this ARC to be applied.

The tube-to-tubesheet joints will be inspected with a rotating coil probe in accordance with the requirements of Section 8. All indications detected will be characterized as axial, circumferential, volumetric, or mixed mode. Tubes with I

axial indications located adjacent to the tubesheet cladding or in the tube end that protrudes from the cladding will be candidates for application of the ARC. Tubes I

with circumferential, mixed mode, or volumetric indications are not covered by application of this ARC.

l A growth rate assessment ofindications identified in a previous outage must be e

B performed using the same methodology defined in Section 9. The growth rate of the population will be monitored each outage to ensure that the conclusions of Section 9 remain valid.

The position of each axial indication with respect to the CCI will be established according to the analysis technique specified in Section 8. Indications that extend into the portion of the roll joint adjacent to the carbon steel tubesheet are not covered by this ARC.

I The number of indications found during the EC inspection, for each radial zone e

(bin) defined in Table 7-6, must be increased by the inverse of the POD. This resulting number of indications must multiplied by the associated leak rate to obtain the total leak rate for the zone. For tubes with multiple indications, a separate leak rate for each indication must be used.

I The postulated leakage for all zones in each OTSG will be summed to determine e

the total postulated accident leakage from the TECs to be left in service. This total leakage will be compared to the plant specific allowable leak rate. If the total leakage from each OTSG is less than the allowable, then all the indications may remain in service through the next operating cycle. If the postulated leakage from one or both OTSGs exceeds the allowable, selected tubes will be repaired until the postulated leakage from the remaining indications is acceptable.

All indications remaining in service as a result of this ARC will be inspected at each subsequent outage to ensure that indications have not grown past the CCI and that no circumferential or mixed mode indications are present.

'P 89

BAW-2346NP

. Plant specific assessments may be performed based on actual plant specific data.

This data may include, but is not limited to, plant specific transient loads and resulting tube-to-tubesheet joint interface dilations, plant specific qualified inspection techniques, and plant specific growth rates.

In order to ensure compatibility with this ARC, the process outlined in this ARC must be used to evaluate this plant specific data. Plant specific changes to the leak rates presented in this ARC will be provided in a 10 CFR 50-59 licensing submittal.

I I

I I

I I

I I

I I

I I

L 90

BAW-2346NP 13.0 RISK ASSESSMENT The NRC has established five key principals that should be met in order to implement B

risk-informed decision making relative to license basis changes. These principais are listed in Regulatory Guide 1.174,"An Approach for Using Probabilistic Risk Assessment I

in Risk-Informed Decisions On Plant-Specific Changes to the Licensing Basis" (July 1998). These principles are as follows:

1

1. The proposed change meets the current regulations unless it is explicitly related to a requested exemption or rule change, i.e., a " specific exemption" under 10 CFR 50.12 or a " petition for rulemaking" under 10 CFR 2.802.

2. The proposed change is consistent with the defense-in-depth philosophy.

3. The proposed change maintains sufficient safety margins.

4. When proposed changes result in an increase in core damage frequency or risk, the increase should be small, and consistent with the intent of the commission's Safety Goal Policy Statement.

5. The impact of the proposed change should be monitored using performance measurement strategies.

l The implementation of the ARC relative to each of these principals is discussed in the following paragraphs.

13.1 Principle 1: Satisfaction of Current Regulation The analyses and testing performed in support of this ARC have demonstrated that,if the ARC is applied consistent with the requirements set forth in this report, the deterministic structural integrity criteria of the plant's current licensing basis as defined in the plant Technical Specifications is satisfied. This includes appropriate margins for failure under normal operating conditions and postulated accidents established in Regulatory Guide 1.121. In addition, the impact on postulated leakage during a design basis accident has been assessed, and it has been shown that the leakage will be less than the limit I

established to satisfy 10 CFR 100 limits for off-site dose defined in the plant licensing documents. The approach taken to conservatively quantify the leakage is consistent with the methodology outlined in NRC Generic Letter 95-05, which established an acceptable I

framework for submittal of alternate repair criteria which could result in leaving tubes with known through-wall degradation in service. Therefore, it is concluded that the proposed change to the plant's licensing basis meets all current regulations set forth for I

implementation of this ARC.

I h

a 91 a

BAW-2346NP 13.2 Principle 2: Defense in Depth The proposed ARC will allow potentially through-wall cracks to remain in service, which potentially compromises the leakage integrity of one barrier between the public and the fission products in the reactor core, namely the primary-to-secondary pressure boundary.

However, as discussed earlier, the structural integrity of the tubes is maintained under application of this ARC, therefore rupture of the boundary is prevented. Leakage that could be attributed to these indications during a design basis accident has been conservatively estimated and shown to be less than the plant acceptance criteria when the ARC is applied in accordance with the requirements in this document. Furthermore, the ARC has no effect on the mmaining containment structures, or on any plant process or procedure that would increase the likelihood or consequences of any accident. Therefore it is concluded that the defense-in-depth design attributes are satisfactorily maintained under application of this ARC.

13.3 Principle 3: Safety Margins it has been shown that the indications to which this ARC can be applied cannot burst due to the support provided by the tubesheet. In addition, analysis has shown that the structural integrity of the tube-to-tubesheet joint, including the fillet weld, is maintained with safety margins defined by the ASME Code of Construction. Therefore it is concluded that safety margins consistent with the design basis of the plants have been maintained.

13.4 Principle 4: Effects on Core Damage Frequency The ARC for tube end cracks will be applied only to axial indications that are located in the portion of tubing that is roll expanded into the tubesheet. Burst of these indications is not possible due to the constraint provided by the tubesheet. The thermal challenge conditions associated with a severe accident cannot increase the probability of tube burst at this location. Therefore, application of this ARC will not increase the probability of tube rupture during a postulated severe accident.

In addition to the above, preliminary results from an EPRI study indicate that B&W plants are not susceptible to the hot leg counter-current natural circulation phenomenon I

that presents a challenge to the steam generator tubes in other PWR designs. The B&W OTSG design is inherently protected against this natural circulation process because of the vertical orientation of the hot legs. This is in general agreement with the conclusions I

made by the NRC in NUREG-1570, " Risk Assessment of Severe Accident-Induced Steam Generator Tube Rupture", which states " severe accident thermal challenges to steam generator tubes are not a concern for the B&W design."

Another potential challenge to the steam generator tubes during the severe accident is from re-starting (or bumping) of the reactor coolant purms (RCPs). By clearing the cold I

leg loop seals, the hot gases could enter the steam gennator tubes via full-loop natural circulation. Use of the RCPs in this situation is not presently prohibited by plant I

92

BAW-2346NP operating procedures. The B&W plants will review the results of the EPRI study as it relates to the B&W plants, and will address any actions warranted based on this review to mitigate the consequences of the postulated severe accident.

As discussed above, the probability of tube burst in the region of interest is not increased and a thermal challenge is unlikely for the B&W designed steam generators. It can therefore be concluded that no impact on Core Damage Frequency or Large Early Release Frequency will occur as a result of the proposed ARC.

13.5 Principle 5: Performance Monitoring All indications remaining in service as a result of this ARC will be inspected in each planned future inspection outage in order to ensure that the requirernents of the ARC are continually satisfied. The inspection will be conducted in accordance with the EPRI Steam Generator Examination Guidelines, as supplemented by the requirements in this report. In addition, primary-to-secondary leakage monitoring during normal operation serves to ensure that the tubes are not degrading at a rate significantly higher than that which is assumed in this application. Therefore it is concluded that sufficient monitoring measures are in place to ensure the continued satisfactory performance of any tubes with indications left in service as a result of this ARC.

93

BAW-2346NP 14.0

SUMMARY

The testing and analyses presented in this report support leaving tubes with axial indications in service without repair,if the indications do not extend into the portion of the tube adjacent to the carbon steel portion of the tubesheet. Application of this alternative repair criteria within the guidelines presented in this topical report will ensure that adequate margin is maintained against challenges to the structural integrity and leakage of the affected tubes.

In addition, the' qualification process outlined and used for this ARC provides each utility with the capability to perform plant specific assessments using plant specific data. This data may include, but is not limited to, plant specific transient loads and resulting tube-to-tubesheet joint interface dilations, plant specific qualified inspection techniques, and plant specific growth rates. Plant specific changes to the leak rates presented in this ARC will be provided in a 10 CFR 50-59 licensing submittal.

?

94

APPENDIX A: RESULTS OF LEAK TESTS I

This appendix contains the results of the individual leak tests described in Section 6.5.

I I

I i

l l

I I

I 95 APPENDIX A

Table A-1: Initial Leak Rate Results Test Test EDM Test Axial Dilation (mils)

Measured Leakage l

Block Length Pressure Load (inch)

(psi)

(Ibs) x y

gpd ppm i

I I

I I

[d]

I I

I l

l

[

96 APPENDIX A

BAW-2346NP Table A-2: Final Leak Rate Test Results Test Test EDM Test Axial Dilation (mils)

Measured Leakage Block Length Pressure Load (inch)

(psi)

(lbs) x y

gpd gpm i

I I

I I

I

[d]

I I

I I

97 APPENDIX A i

BAW-2346NP Table A-2: Final Leak Rate Test Results Test Test EDM Test Axial Dilation (mils)

Measured Leakage Block Length Pressure Load (inch)

(psi)

(Ibs) x y

ppd gpm l

I I

I I

[d]

I I

I I

I c

98 APPENDIX A

l BAW-2346NP Table A-2: Final Leak Rate Test Results Test Test EDM Test Axial Dilation (mils)

Measured Leakage Block Length Pressure Load (inch)

(psi)

(Ibs) x y

gpd gpm I

I I

I I

I rei I

I I

i I

[

[

99 APPENDIX A

BAW-2346NP Table A 2: Final Leak Rate Test Results Test Test EDM Test Axial Dilation (mils)

Measured Leakage Block Length Pressure Load (inch)

(psi)

(Ibs) x y

gpd gpm i

I I

I I

[d]

I I

I I

I I

M 100 APPENDIX A i

F l

BAW-2346NP Table A-2: Final Leak Rate Test Results Test Test EDM Test Axial Dilation (mils)

Measured Leakage Block Length Pressure Load (inch)

(psi)

(Ibs) x y

gpd gpm i

[d]

1 l

l 101 APNNDIX A

BAW-2346NP 4

-APPENDIX B: STATISTICAL EVALUATION OF LEAK RATES This appendix contains the statistical evaluation, defined in Section 6.6, for the individual leak

. test results presented in Appendix A.

102 APPENDIX B

~

BAW-2346NP B.I Introduction This appendix addresses leak rate characterizations for various flaw sizes (EDM notch lengths) for accident conditions ([

d

]) with [

d

]

dilation values, and for steady state conditions ([

d

] ) with h

[ d ] dilation. These are treated as two cases, and will be addressed separately in this 3

appendix.

I Leak rate characterization will be by statistical one-sided upper tolerance limits at various confidence / population coverage values. For each EDM length and dilation value, these tolerance limits on leak rate will be computed.

The data provide only five (5) observations for each EDM length and dilation value.

B.2 Analysis for Accident Conditions The data used for accident conditions are shown in Table B-1. The variable of interest is leakage in units of gallons per minute (gpm).

This data was imported to the NCSS 2000 statistical analysis software package for basic analysis, and tolerance limits were calculated using necessary multipliers. The formula i

for computing a one-sided upper tolerance limit (OSUTL), assuming the variable of interest is normally distributed, is:

xbar + k(C, P, n, df)

• s where:

xbar is defined as the average of the n observations of the variable s is defined as the sample standard deviation of the n observations k(C, P, n, df) is the multiplier, with C = confidence coefficient P = population coverage n = degrees of freedom for the average f = degrees of freedom for the variance estimate (equal to n -1) for these analyses NCSS 2000 software consists of several analysis modules. Of particular interest for this work is the Descriptive Statistics module, which includes an option that computes summary statistics for selected variables in the database. In this option, the module I

allows for tests of normality using several techniques. For this work, the Shapiro-Wilk W ten was used for assessing the hypothesis that the data is from a normal or Gaussian distribution ihioughout this section.

The use of statistical one-sided upper tolerance limits for samples as small as in these data categories (5) requires normality of the parent population to produce reliable results.

I Generally, for the accident condition data sets by EDM length and dilation combinations, the raw leak rate data did not pass the Shapiro-Wilk W test for normality. The small 1

103 APPENDIX B

BAW-2346NP l

sample size may be a factor in this. Transforming the leak rates by taking logarithms (base 10) proved useful in establishing the assumption of normality. When using a transformation such as this, the tolerance limits are calculated using the average and standard deviations of the transformed data, and the result is then exponentiated to provide the OSUTL in gpm.

The results for the logarithmic transformation of the raw leak rate data in gpm are presented in Table B-2 for 507e/95% OSUTL,95%/957c OSUTL, and 957c/50% OSUTL.

The 95%/50% OSUTL is also shown in Table 6-4 and in Figure 6-3.

The k(C, P, n, df) values were computed using a Fortran program. For the confidence (C) and population coverage (P) values of interest in this document, the values are summarized in the following table:

C P

n k(C,P, n, n-1) 50 %

957c 5

1.78 95 %

507c 5

0.953 95 %

957c 5

4.203 B.3 Analysis for Steady State Conditions The data used for steady state conditions are shown in Table B-3. The variable of interest is leakage in units of gallons per minute.

The methodology used for the statistical analysis of the steady state condition is identical to that used for the evaluation of the accident conditions presented in Section B.2 (above). The results for the logarithmic transformation of the raw leak rate data in gpm are presented in Table B-4 for 507e/95% OSUTL and 95%/50% OSUTL. The 95%/507c OSUTL is also shown n Table 6-4.

The k(C, P, n, df) values were computed using a Fortran program. For the confidence (C) and population coverage (P) values of interest in this document, the values are summarized in the following table:

C P

n k(C,P, n, n-1) 50%

95 %

5 1.78 957c 50 %

5

.953

(

104 APPENDIX B

!I l

1l,ll l\\ll l

l 1

P B

N X

6 I

4 D

32 N

E W

P 0

P A

1 B

g A

)

o m L pg

(

e gla ka d e

e L

rusaeM gn i

tseT

)s mg Y

l i

n oi m

rn

(

Fo n

l i

o t

ai i

t d t

an la Do i

X eC D

gc al i

k c ay eC Lr d

nte a

5 of o

0 iA L )s

]

1 t

b d

i l

d aI ng i(

[

on x

Ct A

i s

t e nT ed ka ice cL er Al u) a i

s s

n s p 1 i e

F r(

-B P

e lb ht a

g)

T ns ee Lh c Min D(

E tse#

T k

ts c eo TlB ll

lll ljjl j

. P B

N X

.. 6 I

4 D

32 N

E W

0 PP A

1 A

B g

)

m o

L pg

(

e gla ka d e

e L

rusaeM gn itseT

)s Y g

mn l

i oi m

r n

(

Fo n

l it o

ai i

t d t

an a

l Do X

i eC D

gc al i

k c ay eC Lr d

e nt a

6 of o

0

]

iA L )s 1

t b

d i-l d

aI ng

(

i

[

on x

Ct A

i s

t e nT e k da ice cL e

Al u) r a

s s i

n s

e (p 1

iF r

B P

e lb h

t a

g T

n )s ee Lh c Min D(

E tse#

T k

ts c eo TlB Il l

l Il{ljl!l ll l1 P

B N

X 6

I 4

D 32 N

E W

P 0

P A

1 A

B g

)

o m L pg

(

e gl a

ka d x e r

I usaeM gn i

tseT

)s Y g

l mn i

oi m

r n

(

Fo n

l i

o t

ai i

t d t

an a

l X

Do i

eC D

gc ail k c ay eC Lr d

nte a

7 of e

0 iA L )s 1

t b

i-l d

al ng i(

on x

Ct A

i s

t e nT ed ka ice cL e

r Al u) a i

s s

n se (p 1 i F r

B P

e lb ht a

g)

T nse eh I

c Min D(

E tse#

T k

ts c eo TlB

l!!

l llli1i lil P

B N

X 6

I 4

D 32 N

E W

P 0

P A

1 A

B g

)

o m L pg

(

e gl a

ka d e

e L

rusaeM gn i

tseT

)s Y g

l mn i

oi m

rn

(

Fo n

l i

o t

ai i

t d t

an a

l X

Do i

eC D

gc al i

k c ay eC Lr d

e nt a

8 of o

0

]

A L )s i

1 t

b d

i l

d aI ng i(

[

on x

Ct A

i s

t e nT ed ka ice cL er Al u) a i

s s

n se (p 1 i F r

-B P

e lb h

t a

g)

T ns ee Lh c Min D(

E tse#

T k

ts c eo TlB

~

PN 6

432 W

0 A

1 B

g

)

o m L pg

(

e gla ka d e e L

rusaeM gn its eT

)s Y

g mn l

i oi m

rn

(

Fo n

l i

ai o

t i

t d t

an a

l X

Do i

eC D

gc ai l kc ay eC Lr d

e nt a

of o

]

A L )s it b

d i

l d

aI ng i(

[

re on x

m Ct A

i m

s u

t e nT ed ka w

ic e s

cL e

m r

u Al u) a i

s s

n s p 1 i e

- F r(

==

B P

=

e

=

lb h

a tg) 1 T

ns ee Lh c Min D(

-r E

tse#

T k

ts c eo c

TlB r

1

i

,lj!i i

iIl1{!

ll1i l

P B

N X

6 Ls I

g 4

To D

3 Ul N

)

2 Sn m

E W

Oo p

P d

g 0 e

(

P A

5 s

/ a A

B 59B Lsg IJol t

)

S n m

Oo p

5 d g

e

(

9 s

/ a 59 B Ls g Ho l

t

)

S n m

Oo p

5 d g

e

(

s 9 s t

/ a 0

lu 5 B se 0

?

al R

1ga a t

od m la Lf r s o o c

n i

I ts it g

a o0

)

am t

1 S

Lga t

p d oDg n

L t

(

o S

it i

a d

ga 0

t n

oD

)

]

1 o

L m

0 1

C e1 p

d v

g g

(

Ao

[

tn L

e d

i a

c vt c

e a

)

A DD m

p d w g

t a

(

2 S R

-B e

a l

t b

a

)

eD m

a r

p T

Aw g

a

(

R n

l m

)

o i

ls t

i a

i

(

D lad

)

ia s

x o b

AL I(

e r

s i) u s

s p

e

(

rP

-c h

)

Mt h

g c

Dn n

Ee i(

l 1

i 1

L

l I

IIjj!!

P B

N X

6 I

4 D

32 N

E W

P 0

P A

1 B

g A

)

o m

f pg (lega d k

e a

r x

us I

aeM gn i

tse

)s T

l Y

i m

m g

(

o n r

n i

f n o

l ao i

t t

i a

t ai l

Dd D X i

n eo gC a c k i d

al ec a

Ly o

C L )s

]

n b

d r

l o e aI i

t i(

1 t f

[

i x

i dA A

I no-g Cn i

et t s e

ae r

T u) t i

S s s k

s ya m (p d e P

aL e

l t

Sa n

i

F h

3 t

g)

B ns ee e

Lh l

c b

Min aT D(

E tse#

T k

ts c eo TlB ll.

II!

I l

{!lllil 1Illlll l

PN 6

432 W

n A

Lo

)

0 T 5

d sm B

/Ue g p

5 Sslo 9

g Oa

(

B n

Lo 5 T dsm

)

9 Uegop

/0Ssl g

5 Oa

(

B 0

?

al 1

a t

s ga odm t

lu L

r f

o se so n

i R

g la o0 a m

)

c L1 t

i ga p

t doD g

s i

t L

(

ta S

tS n

g o

o0

)

1 a L

m i

t t

ga p

i d

e n

v oD g

M o

A L

(

C e

v a

ew t

)

a Dat m

t m

S Ra p

un y

d D

g f

t u

d So

(

ae t

s S

f

)

m owta m

u 4

eaa p

-B vRD g

A

(

e s

l u

b m

a n

u

)

o T

it ils u

la m

i

(

e D

mm d

l l

)

a u

ix o I

a s

b AL

(

se e

m r

m u

i) u s

s s

p e

(

u rP h

)

Mt h

g Dn cn Ee i(

l t-

-c 1

-3 i

ji

BAW-2346NP i

i APPENDIX C: TYPICAL ETSS I

I This appendix contains examples of typical examination technique specification sheets used for I

the qualification of the eddy current technique defined in Section 8. These sheets will also be used for inspection of steam generators for tube end cracking.

I I

I I

I I

I I

{

113 APPENDIX C s

6,wa--

BAW-2346NP Examination Technique Specification Sheet ETSS TEA ARC 3-Coil RPC (.115/+PT/.080) l Revision 0 l Page: 1of4 Examination Scope Applicability:

B&WOG TEA ARC Project Instrument Tubing Manufacturer /Model: Zetec MIZ-30 Material Type: Inconel600 Data Recording Equipment OD/ Wall (inch): 0.625 OD X 0.037 Wall Media: HP Hard Drive 1.3GbOptical or Equiv.

Calibration Standard Software Type: ASME with TSP Wear and EDM Manufacturer: Zetec Analog Signal Path I

Version / Revision: EN95, Patch 95_4_26 Acquisition or Probe Extensio.n - Optional: Manufacturer: Zetec later, Version 4.30 Analysis or later.

D#3414-11-A,48" 5-2 pin Examination Procedure Extension Type & Length: Universal 945-1760. 50 ft.

Number / Revision: 54-ISI-400-08 Slip Ring Model Number: 508-2052 Scan Parameters Scan Direction: Pull or Push Digitization Rate, Samples / Inch (Min.)

l Axial Direction l 25/30 l Cire. Direction l

30 Probe Speed Sample Rate RPM Set RPM Min RPM Max

.IIPS 400 300 180 407

=

.2 IPS 400 300 300 407 l

Probe / Motor Unit 5

Description (Model/Dia./ Coil Dimensions)

Manufacturer /Part Number Length 520 (555) 3-C Delta i15/+FT/.080(Unshielded)

Zetec D# 3610-7-A (.010".013" coil shoe)

N/A I

.520 (555) 3-C Delta l 15/+PT/.080 (Unshielded)

Zetec D#4259 A (.014". 017" coil shoe)

N/A

.500 (5-2) M/U - 36 pin Zetec 810-4077-011 83 ft.

Data Acquisition Calibration Coil I (.115" Pancake) Channels Channel &

Ch I Ch4 Ch7 Ch l I Frequency 300 kHz 200 kHz 100 kHz 20 kHz Phase Rotation 40% ID Axial 40% ID Axial 40% ID Axial Broach TSP l

15*

15*

15 90*

B Span Setting 40% ID Axial 40% ID Axial 40%ID Axial Broach TSP 3 divisions 3 divisions 3 divisions 3 divisions g

Calibration Coil 4 (Trigger) Channel l

l Ch 8 100 kHz l

Encoder Pulse l

90 l

4 disisions l

Calibration Coil 5 (+IFT) Channels I

Channel &

Ch2 l

Ch5 Ch9 Ch 12 I

Frequency 300 kHz 200 kHz 100 kHz 20 kHz Phase Rotation 40% ID Axial 40% ID Axial 40% ID Axial Broach TSP 15*

15*

15' 90 l

Span Setting 40% ID Axial 40% ID Axial 40% ID Axial Broach TSP l

3 divisions 3 divisions 3 divisions 3 divisions Calibration Coil 7 (.080" Pancake) Channels Channel &

Ch3 Ch6 Ch 10 Ch 13 Frequency 300 kHz 200 kHz 100 kHz 20 kHz Phase Rotation 40% ID Axial 40% ID Axial 40% ID Axial Broach TSP 15*

15*

15*

90 Span Setting 40% ID Axial 40% ID Axial 40% ID Axial Broach TSP 3 divisions 3 divisions 3 divisions 3 divisions 114 APPENDIX C

BAW-2346NP Examination Technique Specification Sheet ETSS TEA ARC 3-Coil RPC (.115/+PT/.080) l Revision 0 l Page: 2 of 4 Configuration Board Settings tree off I ocwn coritiouration s O name t Isamples I sec : 400 rec. media = Opoc tester.

board # 1 boarde 2 board # 3 board e 4 board e 5 board # 6 coard e 7 board a 6 e of channels =

13 probe # 1 probe # 1 probe # 2 probe 2 probe # 1 probe # 1 probe # 1 probe # 1 DRIVE DRIVE DRIVE DRIVE DRIVE DRIVE DRIVE DRIVE A D B C A D B C A D R C A D B C A D B C A D B C A D B C A D B C Drive Polanty N

N N

N N

N N

N N

N Group Number 1

1 1

1 1

2 2

2 2

2 I

Coll Number 1

4 5

7 8

1 4

5 7

8 FREQ#1 Time Slot s 1 300 kHz G: u 2 l 12 0 V D

D D

FREQ#2 Time Slot a 2 200 kHz G-x 2 l12 0 V D

D D

FREQs3 Time Slot s 3 100 kHz G 2 l 12.0 V O

D D

D FREQ#4 Time Slot a 4 20 kHz G: x 2 l 12 0 V D

D D

FREQ #5 Time Slot # 4 FREQ so Time Slot s 6 I

FREO s7 Time Slot a 7 I

FREQs8 Time Slot s 8 l

Special Instructions

1. Verify smooth data acquisition in the areas of interest, without probe stalling orjerking.

I I

I I

I I

I k

y 115 APPENDIX C w

BAW-2346NP Examination Technique Specification Sheet ETSS TEA ARC 3 Coll RPC (.115/+PT/.080) l Revision 0 l Page: 3 of 4 Data Analysis Calibration Coil I (.115" Pancake) Channels Channel &

chi Ch4 Ch7 Ch 11 Frequency 300 kHz 115MR 200 kHz 115MR 100 kHz 115MR 20 kHz i15MR Phase 40% ID Axial 40% ID Axial 40% ID Axial Broach TSP Rotation 15 t3*

15 i3*

15 13*

90 13*

Span Setting 40% ID Axial 40% ID Axial 40% ID Axial Broach TSP Minimum 3 divisions 3 divisions 3 divisions 3 divisions Calibration Coil 5 (+PT) Channels Channel &

Ch2 Ch 5 Ch9 Ch 12 Frequency 300 kHz + AXIAL 200 kHz + AXIAL 100 kHz + AXIAL 20 kHz + AXIAL Phase 40% ID Axial 40% ID Axial 40% ID Axial Broach TSP Rotation 15 3

15 i3*

15 13*

90 *3*

Span Setting 40% ID Axial 40% ID Axial 40% ID Axial Broach TSP Minimum 3 divisions 3 divisions 3 divisions 3 divisions Calibration Coil 7 (.080" Pancake) Channels Channel &

Ch3 Ch 6 Ch 10 Ch 13 Frequency 300 kHz 080MR 200 kHz 080MR 100 kHz 080MR 20 kHz 080MR Phase 40% ID Axial 40% ID Axial 40% ID Axial Broach TSP Rotation 15 13*

15 23" 15 i3*

90 3*

Span Setting 40% ID Axial 40% ID Axial 40% ID Axial Broach TSP Minimum 3 divisions 3 divisions 3 divisions 3 divisions Calibration Process Channels Channel &

Ch P1 (Ch 2)

Ch P2 (Ch3/10)

Ch P3 (Ch2/9)

Ch P4 (Ch1/7)

Frequency 300 kHz + CIRC 300/100 kHz 80MR 300/100 kHz + AXIAL 300/100 kHz i 15MR Adjust N/A Supp. Broach TSP Supp. Broach TSP Supp. Broach TSP Parameters Phase Ch 2 Rotation 40% ID Axial 40% ID Axial 40% ID Axial Rotation

+180 15 i3*

15* i3" 15 13" Span Setting 40% ID Axial 40% ID Axial 40% ID Axial 40% ID Axial Minimum 3 divisions 3 divisions 3 divisions 3 divisions Voltage Normalization Calibration Curves (optional)

CH Signal Set Normalize Type CH Set Points 1

100% Axial 20 Vpp Coil 1 Chnis Volts (Vmx)

P2 0% & 2 Wears 2

EDM (Note 4)

Coil 5 Chnis Phase (Vpp)irnq d 2

Axial OD/ID 100,60,40 3

(Note 4)

Coil 7 Chnis Phase (Vpp)itwq d Pl Cire OD/ID 100,60,20 Data Screening Left Strip Chart Right Strip Chart Lissajous Ch 2 Vertical Ch 12 Vertical or Analyst Discretion Ch2 116 APPENDIX C

1 BAW-2346NP Examination Technique Specification Sheet ETSS TEA ARC 3 Coll RPC (.115/+PT/.080) l Revision 0 l Page: 4 of 4 Special Instructions 1.

Span, Phase and Volts are to be set using the center of the notch or hole signal.

2. Channels P2, P3, and P4 are optional for tube roll expansion examinations.

3. Rotate data using " Data Stew Menu" so coils 5 and 7 are aligned with Coil 1. Label the coils using the acronyms shown in the " channel & frequency" column of the data analysis calibration section.

4.

When the 100% Axial EDM notch saturates, substitute the 60% ID Axial EDM notch for voltage normalization and set it to a value of seven (7.00) volts (Vpp).

5.

Use the tube outside diameter (0.625") in user selects for tube diameter.

I

6. Set the clad-to-carbon steel interface location as shown in the Figure below. The interface occurs in the middle of the V-shaped signal on the 20 kHz Plus Point response. The interface occurs near the end of the ramp towards the carbon steel region on the 20 kHz pancake responses.

12 20

' AXIAL 11 20 115t1R 13 20 000 tit

- C1 C5 Vert C1 Vert C1 C7 Vert v.w I

I E

I

.il i

_E J

ee,_

I r

I I

0.9f I

If any axialindication is present, determine if any portion of the indication is beyond the interface location 7.

into the carbon steel region of the tubesheet.

l 117 APPENDIX C

.s.

BAW-2346NP APPENDIX D: GROWTH ANALYSIS The appendix contains the details of the TEC growih assessment defined in Section 9.

I18 APPENDIX D

BAW-2346NP D. !

Introduction This appendix summarizes the analysis techniques used in the comparison of upper Gibe end indications (SEA and MEA) reported during the [

d

]

outages at [

d

]. The statistics analysis used to determine the growth for the re-analyzed indications is also included in this appendix.

D.2 Re-Analysis Information D.2.1 Scope The scope included analysis of the upper tube ends as follows:

SG-A: A total of 141 tubes with the SEA or MEA codes were analyzed. Of this total.

I 98 tubes were compared between the outages.

The remaining tubes were determined to be unusable based on the quality of the IR14 pancake coil data.

SG-B: A total of 80 tubes with the SEA or MEA codes were analyzed. Of this total, 50 tubes were compared between the outages.

The remaining tubes were determined to be unusable based on the quality of the IRl4 pancake coil data.

Five (5) tubes were not retrievable because of optical disk errors in the IR14 data.

D.2.2 Methodology The data analysts assigned to this project were certified in accordance with FTI Procedure 54-ISI-24 (Quality Assurance Program) and in addition held QDA qualifications in accordance with the PWR Steam Generator Examination Guidelines.

The data analysts were trained on ETSS # 1, [ d ], Revision 0, dated 2/23M9 prior to the task.

I Analysis was performed in accordance with FTI procedure 54-ISI-400-08 and ETSS #1, using Zetec Eddynet 98, Version 1.15, eddy current analysis software.

ETSS #1 l

describes the essential variables and procedure used for this analysis. The basic outputs from the analysis were the signal characteristics of amplitude, length, position relative to the clad-to-carbon interface and number of indications. These signal parameters were l

recorded for each indication in both outages in a standard eddynet report format for comparative evaluation. In addition, graphics of each of the indications eddy current response were produced I

D.2.3 Summary of Re-Analysis The distance from the FTL to the CCI for some tubes could not be defined due to the poor quality of some of the IR14 pancake coil data. The primary inspection coil for this outage was the Plus Point coil and consequently the pancake coil was not required to produce acceptable data for all examinations. The poor data quality is best explained as signal saturation and excessive coil lift-off. These parameters can affect the signal 119 APPENDlX D

BAW-2346NP amplitude and length measurements and may degrade detectability, thereby affecting the aemacy of the flaw tip location. In some instances,it was evident from the data that the probe did not achieve a smooth scan across the area of interest making it dif6 cult to set an accurate axial scale for determining the distance from the FTL to the CCI. In cases where the analyst could not make a proper measurement, the tube was not analyzed. In marginal situations the analyst made the measurements on a best effort basis.

D.2.4 Results of Re-Analysis The results of the EC re-analysis are shown in Tables D-1 and D-2 below. At least three EC calls were made for each TEC reviewed. The three calls were 1) "SAI" or " mal",

I which notes if it is a single or multiple tube end crack, 2) "LEN", which provided a length to the flaw, and 3) "FTL", which determined the distance from the tip of the TEC to the CCI. Since only the FTL measurement was needed to determine the change in the I

location of the Haw relative to the CCI, it is the value provided. The data noted in the tables were used for determining the growth of the TECs for the ARC.

I I

I I

I I

I

[

F L

~

120 APPENDIX D

BAW-2346NP Table D-1: Re Analysis of SG A Data Re-Analysis of CCI Re-Analysis of CCI SG-A SG-A Nw I tube l call l1996 Location l1998 Locationrow l tube l call l1996 Location l1998 Location ;

I I

I I

[d]

I I

I I

I

]

[

121 APPENDIX D

BAW-2346NP Table D-1: Re Analysis of SG-A Data Re-Analysis of CCI Re-Analysis of CCI SG-A SG-A row l tube l call l1996 Location l1998 Location row l tube l call l1996 Location l1998 Location

[d]

I I

CCI - clad-to-carbon steel interface FTL-flaw tip location i

I I

I I

I L

122 APPENDIX D E

t

BAW-2346NP Table D-2: Re-Analysis of SG B Data Re-Analysis of CCI Re-Analysis of CCI SG-B -

SG-B rowl tube l call l19% Location l1998 Location row l tube l call l1996 Lxation l1998 Imation 1

-)

[d]

t CCI-clad-: carbon steel interface FTL-flaw tip location 1

l l

123 APPENDIX D

BAW-2346NP D.3 Statistical Approach for Determining Growth Rate The results for determining growth for TECs are established for each steam generator.

The analysis process is the same for each steam generator. Because interest is in average growth, the statistical method used is that of Student's t test for paired differences. The average growth is the average change in distance between the tip of the indication and the clad-to-carbon steel interface. This is a test of the hypothesis that there is no growth.

The null hypothesis here is that the average growth is 0.0 inches. The alternative hypothesis ofinterest is that the average growth is greater than 0.0 inches.

The software package, NCSS 2000, contains a procedure that computes the necessary statistics to analyze the paired differences. This tests and assumes that the data (defined as the growth of indications identified in 1996 and also examined in 1998) is a random sample of such indications, with the growth measurement as well as the individual length measurements following the normal probability distribution.

D.3.1 Growth Analysis for Steam Generator-A The data shown in Table D-1 for steam generator A was imported into NCSS 2000. The results from the Student's t test computer model show that the average growth is greater than 0.0.

A reasonable quantification of the average growth is a one-sided upper confidence limit, say, at the 95% level. This is calculated as xbar + t(0.95, d0

• s Nn Where xbar is the average growth, s is the sample standard deviation of the growth, n is I

the number of indications for which the growth is determined, and ((0.95, d0 is the Student's t distribution for a confidence level of 95% with df=n 1 as the degrees of freedom. The value for t(0.95, dO can be obtained from NCSS 2000, using the I

Probability Calculator module of th:s software.

For SG-A, the calculated growth is:

[ d ) inches The hypothesis that the average growth is [ d ) inches was also investigated for SG-A.

The test of the hypothesis that the average growth is [ d ] inches is supported by the 5% level of significance.

l D.3.2 Growth Analysis for Steam Generator-B The data shown in Table D-2 for SG-B was imported into NCSS 2000.1hc results show that the average growth is [ d ] inches at the 5% level of significance.

!n

,L 124 APPENDIX D I

L

=

BAW-2346NP D.3.3 Conclusion From the statistical analyses it was concluded that the hypothesis that the average growth for SG-A is greater than [ d ) inches at the chosen level of significance. Since SG-B supported the hypothesis that growth is essentially [ d ] inches, SG-A average growth of

[ d ] inches will be bounding.

I I

I I

I I

I

~

l i

I L

E 125 APPENDIX D I

BAW-2346NP 1

I I

APPENDIX E: PROCEDURE FOR MEASURING CLAD I

THICKNESS I

I This appendix contains the procedure for determining the thickness of the cladding on the primary faces of the steam generator tubesheets.

I I

I

.4 L

[

126 APPENDIX E E.

BAW-2346NP

(.

. E.0 Introduction An important element required for implementation of this ARC is the measurement of the

(

thickness of the Alloy 600 cladding on the primary face of the tubesheet. The clad thickness is required because it is used to limit the maximum extent of axial indications remaining in service, and to ensure that the length ofindications is bounded by the testing

(

that has been performed.

E.1 Equipment and Setup

(

[

[d]

[

[

E.2 Measurement Locations

[

[d]

{

f-L-

127 APPENDIX E

BAW-2346NP l,

[d]

I I

I I

E.3 Interpretation of Data As stated previously, the cladding thickness is expected to vary slightly across the tubesheet. Since the cladding thickness is being used in this ARC as an upper limit on I

the length of axial TEC indications, an over-estimation of the cladding thickness is conservative. For this reason, the maximum thickness measured at any of the [ d ]

locations identified above will be used to bound the thickness of the cladding at any I

location on the tubesheet. If it becomes apparent that the thickness varies significantly from one location to the next, the utility may elect to define two or more zones with different bounding values, and apply the ARC differently to those zones.

As stated in Section 9, the maximum length of the indications to which this ARC can be applied is [ d ] inch minus the growth rate. For example, if the cladding thickness is 1

[ d ] inch, then 0.125 inch of growth could be allowed over the subsequent fuel cycle.

Thicker cladding reduces the margin for acceptable growth between inspections.

[

F L

128 APPENDIX E F

BAW.2346NP Figure E-1: Typical Map of OTSG Tubesheet Face I

Y ma

...*m...e WM.. * * *=...m....m....=... m....

1

. M..m...,.m. wo m...m...,.a..n..,W...m.. s..m...a..u..W..m...M..m...W..u...W..m...M..

.................................u..............

.wM A. W...,.

M...M.. w...w....=...m...,.m...M...w.e.....n.

A..W. m..

1 S

.A. w. e..m........w..m...M...u......m...m...A..m...W..,........w..w..M..m....=...=........... m...M.........

I ww.

.Mw.

--Wm---mmt....................................................A. Aumw-m m.... yw....t.w.................

Mw-A.

.......................................................u........................

uWA.W..w..e.rWdWt................................................................................

s wW YA.

WWW

.... n u.M..n.W..A..MM.o u n.He..t.w.. A.

... n............. u.

~. ~. n....... u.......n.... u'M.MM..w..wM..W WM. u.4.M. u.W.W.=.M..%# Wa.twNM.. MMWA u.

u.

n u

u u

u...

a u.

u.u I

%We.N..M..W. 64.M...w.f.n..%..t.WAWAv M..%..W.M. wtWee.M. o.W.u. = =.u.

n u = =.u. o.......a.... n. u.....a u...........

.....................................................u..............

.. u s.a.~ u...

u...u.........................u........

u.. u...~. n a. u. u

~.o u..

.....o.u

.......u

.u....

.u

.u

...u....

-.n

..a..==u u.u..

u.u..n..a...=.u

.u...u.u.u.u. u..'

.a.n.m

... u..n. bw n u.. n

...........................~....u....u........o....a..s.u..n

..n.

u..n.u..... u..

u u

u.

u.

a. u. u..n..u.. u.

.u

~

u.

u.

.u.. n.a..u....u

.n n.

u neu u

..M.u..AM.P..A.no..n.n u..o...

.a u..u..

~......n......u.... u..

4.A.W..A.W.ev..

4WWMmWWWM4M4.AW/

W

u.... u. eu. n. t u

..... un u u..

.o.....~

u u

.u..u.u. u.u. n u.. o. l.... u t.u. u.....

u.=

u n.

....u.n. u u.u..n u.. u... u u...u. o.. u......u.a.

. u. u.u.

.. u n......u.u n.

.o.

u.

u

..... o

.u..u..u...u u....

1

.u un u

....n. u.un.un...u.n n u...

n..u. u.

u n..

.u.

...u.

.u.

.u u

...u....u.o

u. l...... u.

u.......

u.

.... u u

...... u.... u... u u.... u. uu... u. n..u.u. a u.. u.n. n u.. ~..u. u. n.

. n.... n e..o.u u

.o.u.u u.

n u.n.

n un.u

..u.a

. e..u.u.o.n..e

..n.o. u.. ulu un..e n u... u.n n u..u.a...u..u.u n.

n.

.n

.. u. u n

...*lo u u..

en un u.u u.u.u. u.n. u.u nu u u..n..n. uu.....u.u.n. u u.n..

uu ne.u n ene u..

..u.

.~.. ~

n u WAW A.

WWM%WW v

WW A

W u

u.

A40..M. Ah. WM. nMM...W..W.n t.u n.o. n. 4.M.u..M.4.A.u.e.n.4.r.N..W.A.W.n.nMhMMWA. 4.f.W'MM..%.t.WAMfwM..e..%. AwfW mu.u u u

%o u no.u u...&...u. u..u...t.....u.a u. u o

n au== no u

nu u.n. u.. u..n.n. n... u..n. u. a.u.

u.n....n o..n..u.u.n......u u a. u...u.a u.....u.

..n..u. u. u. u. n

. up.o.n.n u u un a u..o....

u

.n u.

un.

u..

n u un u. u.

u

.u.u....

i. u u

n..

..u ooun u

...u.

...u..

a uu.

u a..

u i

...a ule... un.=..n...... u.

....... A..W.A.W.A..M.M.%tWAMMMMWWWA.a. u.M..M..ewWu...M..t. 4.Ao...MM. n..'

W W

W f.t... u a = =..

e n. u.... u..

e

u.. u.n u u n.

.n...

.=

n

... u... u...a..u n u. u n. =

a..u...n

....==.u.un.u....=....uu u n.u

.. u...u...u. u.n.o

.u a==.u n.u.n u.n.

.u..u.

u..n.u...o..u..

.n u...

..u n u

u o.

n

.a u..

u

.. n u.....u..u o

uu.unu u.=....

u. un...u...n n.

.n u...u n..

n

.....u.

u............u..u.u..nu..u.n...u......u...u.

nu

...u.==.=..

u. u.o..o n.~ u u.. u u.

.. u.n.

...u u.... un..u. n.

.. u. ! S.o.ot....n u.... u.o u..n... un..u u tn....

. u. u.u. uu n.

. u..n.u. u u..

n.

u..nu

.a u u....... u u

~ un un no en u.u. u.

.u..

.... n

....u..=..u. n. u n u..n u.u..a

.u uu

.nu u..u..u..u u

. u.. u n

.u A

t

.u.=.Z....u.u..w..u. u.

.u. u.u.m u..u.

..n....u.....

u.

.u. u u.e...... u. u.

u MM..W. u.t4AWA.M.u.AWA.

u o

. u e. u

.u.o.n..

un o.

u n.

.. u

..u.

l

. nu. %n.

u....

. u. u.

...n.. u..u. u_u..u nu..=..n. u.u n..u.u.u..u.n.u..n.

.e.u.

f 1n..Leun.....aneu.....u......u.u....u.o.u...

...u.

n

.v o

unu u.n u..n.....o... u... u o u o u.

u a

...n_..o..e.u......u......u.

u.u....

. eneu.

u

..u........

..o..s.u....

........=..u...

..u... u

.u..n u..e. o. un n..... u........'E.Yli '.'i.'AW.'.'i.YE.Y.'li.YE.'."Ei YE Y.l.i.A"Z.E Y.i Yri.YT.Y.i.i Y.i.Y.i Y.Z

.u.2

....ueu........

.%..o.?!.\\a.1%o..a.u.1.Z

..u.p.......

.....o..u.u.n.

..u.

..o u..n.

n

....o u.

u o n u..o..u.

.a.

. u.

.S

u......... 52..

u

..u. u u u au.....o.

=. 2.a u.. uu o. o..n.... e.

.u. u. u..n.*C

...n. u. u.

u

. uu en u.

u u.....

a.

u

.u.

u.

'A%.A.u.M..M.Z. u.

9wWe F

MnM u

e.. s u.. u.

...........n. e... u...u. u.u. *..u u.

.."::e un. "." "** "": " :::T ".": "":: ". M..WAt%..M..M..w.

u n

.. n u e...

.u. u u.

un......a. n e u.....n.n un un u..u n..

n..e u u n..

u a un u.

...u....u.

u..u.

.u....n..e.u.....u._n u

u

. u u. un.

u....

...... u.u._..

....f_..u..a..=.=..a...u.....=..nu.....u...

I

....o o..u.o......n u. u.u.......... uo.o.........u.u.= a.ma. a.nna. amam ana aama. anna an.na an a...

..n. anna man. aa.n.s anna ma.n mana anna man

=u...

.u..u..--..u..4....=..=..n....u..n...u.....ouu......uu..u.

-- u

.u n. u..

. u.

u.

u.

.....u n

...e...n...e.o..u...u......

=..u..

....u

u....n u.. u

.a.n...u. ~e. u..u. e....a..u. u.. u. u....

u

.u u

u. u.u.....n..

%A.A.M..A.UMN...a.n

.o......... u

u..n

.... l u.

u.

..u.u..n.n u.u...u.u

..u......

..u. u.u.

o.

u m

.e..o u o..e.n.

....u

..n.u u.

..m

.e. u...o n ue

..a..

.e. u eu

= =..

.. an eu.

u u

.o.

n.p u.

n..u...

un...n.u.su........u...a....u......u......u.....u.u.u.

..c

...n...........

.=u.......u...........o.o..u....

n..o u. a e.u.

.... u n..

u

-unun ei u

......o.

o. u.. n.a. u..n

u..n.o. o.

u.

u..

.o u. n u.

.........n.

.u..n u

u.. u..n. i

.. u. e n....u.

....u....

u

.... u u

.u......u

.u u

n..

...u.

V....n..........

I

• H. A ""*: ".i..."L." : ".*.,A**L. "".47 **:". *"."A."T"Y:: ". *A." *.".*.* *:

.....u.

n....u. u...n....u.

......n.u..

. u.

u.

.u.

.u..

.n.u...........

......................n......n.u....

.....u........ u..u...u.u....u..u......u............u...u....o.

'IA.%.M..M..A..W....................................................................................................

..u...n..

n...

.o..a.

.u o e u u. u.

... u

u. ".... ::::: *:.' ".L...:.".:. "."."..:.:.".:.:.:.".A.*.:.:.:.:.".;.;..;.:.:..:.:"*t

• n...a.u...e.u.............

......u.u..

.n...u. n. u. u... e.n..... u...n.a. e.u. u. u... u........ u..n.

un..

.....o..u.........u..n..

.a...u......

%M..h. W...M. ".f.lM. 9.M.. %. A................................ A..'M..M..M..4.h. M..M..M. M..M..M.. M.M.. T. A..LM. A.R. A. M.M..M..M...M.. W i

...n..u..u. u.

..n......a......n....u...........

WA.........................................,.....u.....u.......o..u..n...

u....................o..n..u..

......n........

..n..........

..u........un==

u............u.u..........o..nu.

u..u.u

.oeu..

.. A.14,'c.:.:..:.:.A.*A..X..:. ".".L..".^.. ":.L. :: :

n........................... "...a."...A. N.A. A.M. AWMMe" e.......................u...........u....s...............................................

g y

MA*N?::.4 :::"A" :*:." ":: " L. ".*,*.A."o."A."".l.A.":.n.*

A

....u..=.=........o....u...o

...........n....e.n.u..u...

e n..n...

u. on. o u

.......s............e.

..e........

O

.....................................................u...............

.......u............u......................................

W

.u.......unu...n..........

u.........

• WMvdWA..T.M..M.M4A.W.. A.W AN.A.M. M.N.E. MM..M..M. 9.M..Y.M. 4 HV

• ee

. u..e.n.

.....u..u..u..n..n.nu.....

. u

.ea.......................

.u.a9a.ym..w..a.

ygg w,w p

.......a.......n...u..m....

W MO

..e.........

1 129 APPENDIX E u

~

.._