ML20059D485

Steam Generator Sleeving Rept Laser Welded Sleeves Jm Farley Units 1& 2

ML20059D485
Person / Time
Site:	Farley
Issue date:	08/16/1990
From:	WESTINGHOUSE ELECTRIC COMPANY, DIV OF CBS CORP.
To:
Shared Package
ML19307A549	List: ML20059D461 ML20059D468 ML20059D481 ML20059D485
References
WCAP-12673, NUDOCS 9009070059
Download: ML20059D485 (137)

Text

- -,

WESTINGHOUSE PR",PRIETARY CIASS 3 ,

WCAP-12673 SG-90-06-052 l

SG 90 06-052 I

• STEAM GENERATOR SLEEVING REPORT Laser Welded Sleeves J. M. Farley Units 1 and 2 f

t

. t l .'

t' i

This document contains information proprietary to l

Westinghouse Electric Corporation; it is submitted in j confidence and is to bs used solely for the purpose for t

, which it is furnished and returned upon request. 'This document and such information is not to be reproduced, transmitted, disclosed or used otherwise in whole or in l . part without authorization of Westinghouse Electric Corporation. .

e 11to Westinghouse Electric Corp.

I -

WESTINGHOUSE ELECTRIC CORPORATION NUCLEAR SERVICE DIVISION P.O. BOX 355 PITTSBURGH, PA 15230 9009070059 900816 PDR ADOCK 05000340 P PDC

TABLE OF CONTENTS Section litig hgg l.0 INTRODUCTION 1-1 l.1 Objectives 1-1 1.2 Report Applicability 1-1

- 1.3 Experience 1-2 2.0 SLEEVE CHOICE AND SLEEVING BOUNDARIES 2-1 2.1 Sleeve Choice 21 2.2 Sleeving Boundary 2-1 3.0 DESIGN 3-1 3.1 Sleeve Design Documentation 3-1 I

3.2 Sleeve Design Description 3-1 3.2.1 Weld Qualification Program 3-7 I 3.2.2 Sleeve Weld Joint Qualification 38 3.2.3 Acceptance Criteria 38 3.3 Design Verification: Test Programs 3-9 3.3.1 Design Verification Test Program Summary 39 3.3.2 Corrosion and Metallurgical Evaluation 3 10 3.3.2.1 Materials 3-10 3.3.3 Laser Welded Joints 3-21 3.3.3.1 Expansion Processes 3-21 3.3.3.2 Outside Diameter (00) Surface Condition 3-22 3.3.3.3 Stress Corrosion Testing of 3 24 i

- Laser Welded Sleeve Joints 3.3.3.3.1 Corrosion Test Description 3-24 3.3.3.3.2 Corrosion Resistance of Laser Weided Joints 3-28 ,

)

i 1

I

TABLE OF CONTENTS (Continued)

Section 11.1.11 hat

l. 3.3.4 Mechanical Test Program for the Laser

'. Welded Joints 3-35 3.3.4.1 Description of the Laser Welded Joint Test Specimens 3 35

- 3.3.4.1.1 Description of Out of Tubesheet Test Specimens 3 35 3.3.4.1.2 Description of the Elevated Tubesheet Sleeve Lower Joint Specimens 3 35

• 3.4.1.3 Description of Full Length Tubesheet Sleete Lower Joint Specimens 3 38

' 3.4.2 Description of Tests for the Laser Welded Joints 3-38 ,

3.3.4.3 Leak Test Acceptance Criteria 3 40 f 3.3.4.4 Results of Mechanical Tests of the i Laser Welded Joint 3-40 3.3.4.5 Rewelding 3 40 l .'

3.4 Post Weld Heat Treat 3 40 3.5 Analytical Verification 3 48 3.5.1 Introduction 3 48 3.5.2 Component Description 3,/8

- 3.5.3 Material Properties 3-48 3.5.4 Code Criteria 3-48 3.5.5 Loading Conditions Evaluated 3 53 I

. 3.5.6 Methods of Analysis 3 54 3.5.6.1 Model Development 3 54 i! 3.5.6.2 Thermal Analysis 3-55 3.5.6.3 Stress Analysis 3-56 l-ii l

TABLE OF CONTENTS (Continued)

Seetion 11111 Elat

. 3.5.7 Results of Anaik 3 58 3.5.7.1 Primary stress Intensity 3 58 3.5.7.2 Range of Primary Plus Secondary Stress Intensities and Fatigue

- Evaluations 3-60 3.5.8 References 3 68 3.6 Special Considerations 3 69 3.6.1 Flow Slot Hourglassing 3 69 3.6.1.1 Effect on Burst Strength 3-69

. 3.6.1.2 Effect on Stress Corrosion Cracking (SCC) Margin 3 69  ;

3.6.1.3 Effect on Fatigue Usage factor 3 69 3

3.6.2 Tube Vibration Analysis 3-70 3.6.3 Sludge Height Thermal Effects 3 70 3.6.4 Allowable Sleeve Degradatic'.2 3 70 3.6.4.1 Minimum Required Sleeve Thickness 3-70 l

3.6.4.2 Determination of Plugging Limits 3-73 3.6.4.3 Application of Plugging Limits 3 75 3.6.5 Effect of Tubesheet/ Support Plate Interaction 3-75 3.6.6 St/t etural Analysis of the Laser Welded Sieeves with Interfacial Weld Widths of 3-75

[ ja,c,e 1

3.6.7 Minimum Sleeve Wall Thickna**. 3-79 3.6.8 Effect of Potential Gap Openings at the Welds 3-79 3.7 Evaluation of Flow Effects' Subsequent to Sleeving 3 79 3.7.1 Safety Analysis and Design Transients 3-79 3.7.2 Equivalent Plugging Level 3 81 iii

l TABLE OF CONTENTS (Continued)

Seetion Ij,tJg Eggg

.o {

3.7.3 Fluid Velocity 3 83 i i

[ 3.7.4 Flow Effects Summary 3 83 l 4.0 PROCESS DESCRIPf!ON 4-1 4.1 Tube Preparation 41 4.1.1 Tube End Rolling (Contingency) 4-1  ;

4.1.2 Tube Cleaning 4-3 4

4.2 Sleeve Insertion and Expansion 4-3 4.3 Lower Joint Hard Roll ( Full Length Tubesheet Sleeves ) 45 4.4 General Description of Laser Weld Operation 45 4.5 Post Weld Heat Treatment 46 4.6 Inspection Plan 4-7

'O

. NDE INSPECTABILITY 5-1 5 5.1 Ultrasonic Weld Inspection 5-1 5.1.1 General Process Overview 51 5.1.2 Principle of Operation and Data Processing 5-2 5.1.3 Equipment and Tooling 57 5.1.4 Laser Weld Test Sample Results 59 .

5.1.5 Summary 59 5.2 Eddy Current Inspection 5 13 5.2.1 General Process Overvie,w 5 13 5.2.2 Principle of Operation 5-13 5.2.3 Transitioa Region Eddy Current Inspection 5 14

$ 5.2.4 Laser Weld Region Eddy Current Inspection 5 24 g 5.2.5 Sumary 5 29 6.0 INSERVICE INSPECTION PLAN FOR SLEEVED TUBES 6-1 iv

a LIST OF TABLES M 11111 EAEt 3.1-1 ASME Code and Regulatory Requirements 32 >

- 3.3.2 1 Summary of Corrosion Comparison Data for Mill 3 14 i Annealed Alloy 600 and Thermally Treated Alloys 600 and 690 3.3.2 2 Effect of Oxidizing Species on the SCC Suscepti- 3 18

- bility of Thermally Treated Alloy 600 and 690 C rings in Deaerated Caustic 3.5.4 1 Criteria for Primary Stress Intensity Evaluation 3 49 (Sleeve) i 3.5.4 2 Criteria for Primary Stress Intensity Evaluation 3 50 (Tube) 3.5.4-3 Criteria for Primary Plus Secondary and Total 3 51 Stress Intensity Evaluation (Sleeve) 3.5.4 4 Criteria for Primary Plus Secondary and Total 3 52 Stress Intensity Evaluation (Tube) ,

3.5.7.1-1 ilmbrella Pressure loads for Design, 3 59

'aulted, and Test Conditions 3.5.7.2-1 Pressure and Temperature Loadings for Maximum 3-61 Range of Stress intensity and fatigue Evaluations 3.5.7.2-2 Maximum Range of Stress Intensity and Fatigue 3 64 Full Length Tubesheet Laser Welded Sleeve (Without Hardroll) 3.5.7.2-3 Maximum Range of Stress Intensity and Fatigue 3 65 Full Length Tubesheet Laser Welded Sleeve l

(With Hard Roll) 3.5.7.2 4 Meximum Range of Stress Intensity and Fatigue 3 66 Elevated Tubesheet Laser Welded Sleeve 3.5.7.2-5 Maximum Range of Stress Intensity and Fatigue 3 67 Support Plate Laser Welded Sleeve 3.6.6-1 Maximum Range of Stress Intensity and Fatigue

. [ la.c.e 3 77 3.6.6-2 Maximum Range of Stress Intensity and Fatigue

[ ]a,c.e 3 78 3.7 1 Summary Hydraulic Equivalency Numbers 3 85 Normal Operating Conditions y

LIST OF TABLES ( Cont.)

.' Fiaure ling gg ,

3.7 2 Detail Hydraulic Equivalency Numbers 3 86 4.0 1 Sleeve Process Sequen u Summary 42 4

k 4

e I

vi 1

^

LIST OF FIGURES

, Fiaure 11111 EAER

- 2.2 1 Sleeving Boundary 22 3.2 1 Full Length Tubesheet Laser Welded Sleeve 3-3 Installed Configuration 3.2 2 Elevated Tubesheet Laser Welded Sleeve 35 Installed Configuration 3.2-3 Support Plate Laser Welded Sleeve 36 Installed Configuration 3.3.2-1 Influence of Process and Microstructure on

• 3-12 SCC Resistance of Alloy 690 3.3.2 2 SCC Growth Rate for C rings (150 percent YS and 3 15 TLT) in 10 percent NaOH 3.3.2 3 Light Photo Micrographs 111ustrating IGA After 3 16 5000 Hours Exposure of Al'oy 600 and 690 C Rings to 10% NaOH at 332'C (630*F) 3.3.2-4 SCC Depth for C-Rings (150 percent YS) in 3 19

- 8 percent Sodium Suifate 3.3.2-5 Reverse U-bend Tests at 360'C (680'F) 3 20 3.3.3 1 Accelerated Corrosion Test Specimen Welded Joint 3-26 Configuration 3.3.3-2 Accelerated Corrosion Test Specimen Roll Transition 3 27 Configuration ,

3.3.3-3 Cumulative Percent Cracking For C02 Laser Welded 3-29 Sleeves in 7500F Accelerated Steam Corrosion Test {

3.3.3-4 Cumulative Percent Cracking For C02 Laser Welded 3-30 Sleeves in 7500F Accelerated Steam Corrosion Tests 3.3.3-5 Cumulative Percent Cracking For YAG Laser Welded 3-31 Sleeves in 7500F Accelerated Steam Corrosion Tests 3.3.3-6 IGSCC in Alloy 600 Tube of YAG Laser Welded Sleeve 3-33 Joint After 109 Hours in 7500F Acclerated Steam Corrosion Test 3.3.3-7 Minor IGSCC in Alloy 600 Tube of Stress Relieved .3-34 YAG Laser Welded Joint After 1000 Hours in 7500F Accelerated Steam Corrosion Test vii

l LIST OF FIGURES (Contitiued) l 1

Fioure 3 g

. 3.3.4.1.1-1 Out of Tubesheet Laser Welded Joint Test Specimen 3 36 Elevated Tubesheet Sleeve Lower Joint Test Specimen 3 37 3.3.4.1.2 1 3.3.4.1.3 1 Full Length Tubesheet Sleeve Lower Joint Test Specimen 3 39 3 41 3.3.4.5 1 Laser Welded Sleeve With Reweld Vertical Test Stand Hock up 3 44 3.4 1 Initial Test Samples Detailed 3-45 3.4 2 Field Prototypic Test St.mples Detailed 3 46 3.4 3 Typical Stress Relief Power Profile 3 47 3.4 4 Ultrasonic Inspection of Welded Sleeve Joint 5-3 5.1.2 1 5.1.2 2 UT Signal in Good Weld Region 54 5.1.2-3 UT Signal in Lack of Weld Region 55

. 5.1.2 4 UT Signal in Small Lack of Bond Region 56 5.1.3-1 C-Scan From UT Examination of a Typical 58 Laser Weld

~

UT Setup Standard 5 10 5.1.4-1 ,

1 C-Scan from UT Examination of Equipment Setup 5 11 1

E 1.4-2 j

Standard C-Scan from UT Examination of Workmanship 5 12 5.1.4-3 .

of a laser Welded Sleeve. J Ja,c.e Calibration curve 5 16

, 5.2.3-1 [

Eddy Current Signals frc the ASTM Standard, 5-17 5.2.3 2 Machined on the Sleeve 0.D. of the Sleeve / Tube Assembly Without Expansion ( Cross Wound Coil Probe)

Eddy Current Signals from the ASTM Standard, 5-18

- 5.2.3 3 Machined on the Tube 0.D. of the Sleeve / Tube Assembly Without Expansion ( Cross Wound Coil 1 ,* Probe )

viii

LIST OF FIGURES (Continued)

Figure Title Page Eddy Current Signals from the Expansion Transition 5-19 5.2.3 4 Region of the Sleeve / Tube Assembly (Cross Wound Coil Probe )

Eddy Current Calibration Curve for ASTM Tube 5 20 4 5.2.3 5 i Standard at ( yn,c.e kHz and a Mix Using the Cross Wound Coil Probe Eddy Current Signal from a 20 Percent Deep Hole, 5 21 5.2.3 6 Half the Volume of ASTM Standard, Machined on '

the Sleeve 0.D. in the Expansion Transition Region of the Sleeve / Tube Assembly (Cross Wound Coil Probe)

Eddy Current Signal from a 40 Percent ASTM 5-22 5.2.3 7 Standard, Machined on the Tube 0.0 in the i l

Expansion Transition Region of the Sleeve / Tube Assembly (Cross Wound Coil Probe)

Eddy Current Response of the ASTM Tube Standard 5 23 .

5.2.3 8 at the End of the Sleeve Using the Oross Wound Coil Probe and Multifrequency Combination ja,c.e kHz Eddy Current Baseline of 5 25

- 5.2.4-1 Crosswound (

Laser Weld Crosswound Mix Eddy Current Response Baseline of 5 26 5.2.4 2

~

Laser Weld la.c.e kHz Eddy Current Response After 5 27 5.2.4 3 Crosswound (

40% Flat Bottomed Hole was Placed in 00 of Tube at Center of Weld 5.2.4 4 Crosswound Mix Eddy Current Response After 40% 5 28

- Flat Bottomed Hole was placed in 0.0. of Tube '

at Center of Weld i

l iX

l.0 INTRODUCTION

- This document contains technical information in support of the licensing of the I tube sleeving maintenance process as applied to the J. M. Farley Units 1 and 2

',- Series 51 steam generators using a laser welding installation technique. This repair technology is based on extensive development and field implementation programs in steam generator maintenance and repair, including prior laser weld programs.

1.1 Objectives i

The sleeving concept and design are based on observations to date that the tube degradation due to operating environmental conditions has occurred near the~

tubesheet and tube support plate areas of the tube bundle. The sleeve has been designed to span the degraded region in order to retain these tubes in service.

The sleeving program has two primary objectives:

l

a. To sleeve tubes in the region of known or potential tube degradation.

1 .'

b. To minimize the radiation exposure to all personnel (ALARA) l 1.2 Report Applicability This report is applicable to Westinghouse Series 51 steam generators. These steam generators are U-tube heat exchangers with mill annealed Alloy 600 heat l

transfer tubes which have a 0.875 inch nominal outside diameter (0D) and 0.050 inch nominal wall thickness.

The sleeves described herein have been designed, analyzed, and tested to meet the service requirements of the Series 51 steam generator through the use of conservative and enveloping thermal boundary conditions and structural loadings. The structural analysis and mechanical performance of the sleeves are based on installation in the hot leg of the steam generator. Installation in the hot leg is the bounding case and therefore demonstrates applicability of the sleeve design for cold leg installations as well.

1-1

I 1.3 Experience

' i

)

- Sleeving technology was originally developed te sleeve degraded tubes (including leaking tubes) in Westinghouse Series 27 steam generators. A process and remote sleeve delivery system was subsequently developed and used in Westinghouse Series 44 and 51 steam generators. The sleeving technology has also been applied to the installation of sleeves in a plant with non Westinghouse steam generators.

Westinghouse has maintained in service over 25,000 steam generator tubes at operating nuclear power plants world wide through application of the sleeving technology. Mechanical-joint, brazed-joint, and laser welded joint sleeves of Alloys 600, 690, and bimetallic 625/690 have been installed by a variety of techniques hands on (manual) installation, Coordinate Transport (CT) system installation, and Remotely Operated Service Arm (ROSA) robotic installation.

Westinghouse sleeving programs have been successfully implemented after approval by licensing authorities in the U.S. (NRC - Nuclear Regulatory

- Commission), Sweden (SKI - Swedish Nuclear Power Inspectorate), Japan (MITI -

Japanese Ministry of International Trade and Industry) and Belgium (Vincotte).

Laser welding sleeving technology has been previously appit;d at Doel 3, which has Series 51 steam generators, in Belgium in July 1988. Fifty five sleeves were installed and inspected in a demonstration program. Two tubes with laser welded sleeves were removed after one year of operation and metallographically i examined. The results of this inspection verified that both process parameters and inprocess inspection were accurate barometers of the as produced laser weld.

l

- The current laser weld system produces a weld equivalent to the system applied l at Doel 3. The major differences are that the previous system used a gaseous l

C02 laser while the current system uses a solid state Nd:YAG laser and the CO2 system was mounted to the steam generator manway and directed by mirrors

!. while the Nd:YAG system is delivered from outside the containment building by fiberoptic cable.

1-2

m _ si 2.0 SLEEVE CHOICE and SLEEVING BOUNDARIES

.' 2.1 Sleeve Choice

." This report addresses three distinct types of sleeves a full length tubesheet sleeve, an elevated tubesheet sleeve, and a support plate sleeve. The full length tubesheet sleeve is appropriate for all plants which have degradation at the top of the tubesheet, since the lower joint is formed at the bottom of the tubesheet. The elevated tubesheet sleeve is appropriate at Farley Unit 2

^

because Farley Unit 2 has full depth rolled tubes and an alternate plugging criterion ( F* ) accepted by the NRC. The alternate plugging criterion alows i.

tube degradation below a predetermined distance from the tubesheet secondary '

face to remain in service due to the support the tubesheet provides in the area of the defect. This condition is necessary because the sleeve lower joint is pf formed at an elevation above the primary face of the tubesheet, which means the tube is the pressure boundary for some portion within the tubesheet. The support plate sleeve may be installed to bridge degradation located at tube support plate locations or in the free tube span.

~

- 2.2 Sleeving Boundary

- Tubes to be sleeved will be selected by radial location, tooling access (due to channel head geometric constraints), sleeve length and eddy current indication elevations and size. Figure 2.2-1 shows theoretical radial sleeving boundaries for tubesheet and tube support plate sleeves as determined by a geometric radies computed from the channelhead surface-to-tubesheet primary face clearance distance minus the tooling clearance distance. (The actual "as is" bowl geometry will be reduced in certain areas.)

The specific tubes to be sleeved in each steam generator will be determined based on the eddy current analysis of the extent and location of the degradation. A tube plugging / sleeving map for each steam generator will be

. provided as part of the contract specified deliverables at the conclusion of the sleeving effort.

2-1

o e o o n c n m v v M M n) OJ - - n oo88 88 o o8888o .:o.

o oo8h$$288!$:888!$$8!$$8om d co88!o000.00000o000$0080$o8*88888 oe ::::8888 3

80 8 e

88838>8!gm88g[2888$8888888!

• 888$880 8882.

000 0000.800000o00080. 000000 0000000og00 000o0000000.000o.o... -

0888 8 000o.88888888888.!!8 000[88888888888888 8 ..oooooooooooo O ooog.0o00o000o.oooooo..ooooooooooooog g o

o8888088loooooo.0.o00 88888888888228888888888888888o 0u838888888888888888888888 888888 088ja8888888288888:38888888!!!!888888!$$$o n; gjjo888888888838888888 888888888888 m 88888888888888 a ols80888888!.:ooooooooooooooo$!8!888888888888o o .ooooooo ooooococoooooooo M -

o[8:8888888j88888888888[88888((88jj!!g888o888!0

<d 2 -

808:88888828088 38888888oo888880088co 888:[88888888j8888g88888888g8888888888g88888 M

E @[ 888:o88888 8808 888o8888888808888888888088888 0 x u a 088j'880888888888880888888088co8888800088888[0888j8888888888 s

0 z 888 e T i .

~

Z 5  !!!g!$$8!!!!!!!!!!!$$$$!!!!$!!$$$!!!$$$$!!!!!8 8

00 $$"

$o 888888888888888 888888888888888888888888888888 , .

- 0o0 0oo00o0 000 000000000000000o00000000000000 L @

o>u U o00 000000.o ooo.cocooo 00.

888 88888888888 888 o0000o o 888:88888888$88880008000000o00o080008$8888880 oooo.oooooooooooooooococoooooooooo e

2 4

_a 8888j888888888888888888j8888j88888[888j88j888e u

Q d 888.888888828888888888088880888880888088o888 e i i

88888888888j888888888888888888j8888j8888j8

~

888888888888.888888888888888888o88880888808 0 8j8888888j88288888888888j8888888888jj88888*

888088828888828888088888888880088888 0 088:!888888888288888288888888888888888 888 888828888888888888888288888888888888880 8888288888888 28888882 88888888888888 0 ' -

888888888888888888888888228888888888p 8888288888888888888888888888822:2 0 4

882888888888888882288888688888 m 888888888888888888888 22888888 i

. 888882288888888888888888888. C -

8 288888888888888888

• 8888*!888882 Sjj88j8888888 m$ (

88888008:02%28:3 + z i I

888888888888 ooooooo 2~2

3.0 DESIGN 3.1 Sleeve Design Documentation The sleeves have been designed and analyzed to the 1986 edition of Section 111 of the American Society of Mechanical Engineers (ASME) Boiler and Pressure Vessel Code, as well as applicable United States Nuclear Regulatory Commission (USNRC) Regulatory Guides. (The use of guidelines and recommendations of the relevant regulatory guides is only with respect to the laser welded sleeve design. The &cceptance and exceptions to regulatory guides previously provided by Alabama Power Company in the FSAR and other submittals is unchanged.) The associated materials and processes also meet the rules of the ASMC Boiler and Pressure Vessel Code. Specific doctments apolicable to this program are listed in Table 3.1 1, 3.2 Sleeve Design Description ,

The sleeve is sized to address design requirements and to be compatible with

- the dimensional constraints imposed by the tube inside diameter. These constraints include variations in tube wall thickness, tube ovality, tube

inside diameter, tube to tube sheet joint variations and runout / concentricity variations created during tubesheet drilling or misalignment of tubesheet and support plate holes. The remaining design parameters such as wall thickness and material are selected to enhance design margins and corrosion resistance ana/or to meet ASME Code requirements.

The reference design of the full length tubesheet sleeve, as installed, is

, illustrated on Figure 3.2-1. At the upper end, the sleeve configuration (see Figure 3.2 1) consists of a section which is hydraulically expanded. The

. hydraulic expansion of the upper joint is sufficient to bring the sleeve into contact with the parent tube and achieve proper fitup geometry for welding.

Following the hydraulic expansion, a fillerless weld is made between the sleeve and the tube using the laser welding process. This joint configuration is known as a laser welded joint (LWJ).

The s' ave extends from the tubesheet primary face to above the tube degradation. In the process of sleeve length optimization and allowing for axial tolerance in locating degradation by eddy current inspection, the guideline is that the welds are to be positioned a (

]a,c.e, 3-1 j i

l

~

l l

Table 3.1 1

~

ASME CODE AND REGULATORY REQUIREMENTS l

Aeolicable crittril Reauirement 111m Sleeve design Section 111 NB 3200, Analysis

• NB 3300, Wall Thickness Operating Requirements Analysis Conditions Reg. Guide 1.83 SG Tubing Inspectability 1

Reg. Guide 1.121 Plugging Margin Sleeve Material Section 11 Material Composition

' action 111 NB 2000, Identification.

Tests and Examinations Code Case N-20 Mechanical Properties Sleeve Joint 100FR100 Plant Total Primary-Secondary leak Rate Technical Specifications Plant Leak Rate ,

Section IX - QW Weld Qualification Code Case N-395 Laser Welding i

i i

3 1

3-2

~

l

-- a,c.e <

1 H

E t

t  :

}

b I

1

\.

i l

l l

l t  !

l l

l

\

- 1 l

i- -

Figure 3.2-1 1

1

Full Length Tubesheet Laser Welded Sleeve Installed Configuration 1

3-3 j

The out of tubesheet, i.e. free span,' joint is located so as to provide [

3 a , c ', e ,

At the lower end, the sleeve joint configuration (Figure 3.21) consists of a section which is hydraulically expanded into the original tube and a smaller section within the hydranlically expanded zone which is roll expanded. The sleeve is rolled to a preset torque. This torque has been established through laboratory testing as the range which provides leak-tightness, mechanical strength and resistance to stress corrosion cracking. A single pass weld is then made between the sleeve and tube approximately 1 inch above the tubesheet cladding. The weld forms a hermetic seal at the lower joint. Tb lower end of the sleeve is preformed to facilitate the seal formation and reduce residual stresses in the sleeve.

The elevated tubesheet sleeve is shown in Figure 3.2-2. The design of the

- out of tubesheet LWJ is the same as that for the full length tubesheet sleeve.

The configuration of the lower joint is different in that the lower joint is located approximately 16 incnes above the primary face of'the tubosheet and the roll expansion is omitted. The main reason for this type of sleeve'is that it allows much greater radial tubesheet coverage than the full length tubesheet sleeve whila wrintaining the same axial degradation coverage above the.to of the tubesheet. This sleeve is operationally equivalent to a full length tubesheet sleeve for plants which have full depth expanded tubesheets. Again the welds will be loctted (

ja,c,e, The support plate sleeve is shown in Figure 3.2-3. Each and of the sleeve has l

[

a hydraulic expansion region withir. which the weld is placed. The weld configuration is the same for both upper and lower joints and is the same as t

the out of tubesheet weld in the tubesheet s?eeves. Tl.e inside diameter of both ends of _the sleeve are t'apered to reduce the end ef fects of the sleeve on eddy current inspection. The welds will be located-at [

a,c,e, 3-4 L_

i a,c.e l.

0 O

9 1

i l

i

\

\

i l

1 Figure 3.2-2 i 4

Elevated Tubesheet Laser Welded Sleeve Installed Configuration i

3-5

1 a,c.e l l

l 1

5 o

Figure 3.2-3 m

e Support Plate Laser Welded Sleeve Installed Configuration 3-6

The completed out of tubesheet LWJ will be given a post weld heat treatment (PWHT) that relieves the residual stresses in the tube. Residual stresses in areas adjacent to welds develop due to weld metal shrinkage upon solidification l l

of the weld metal. Such residual stresses are usually benign in most welded i structures for most servi:e. However, in mill annealed Alloy 600 tubing that l is susceptible to PWSCC, the post weld residual stresses could contribute to the induction of PWSCC. Tests of as-welded LWJs in susceptible mill annealed l Alloy 600 tubes indicate that the residual stresses adjacent to the laser weld do not lead to the degree of PWSCC as that experienced by roll transitions.

[

(These tests are described later in this report.) Nevertheless, the reference LWJ process includes a PWHT that tests demonstrate provides further enhancement in PWSCC resistance of the tubing. (These test observations for PWHT LWJ specimens are also described later in this document.)

The sleeve material, thermally treated Alloy 690, is selected to provide additional resistance to stress corrosion cracking. (See Section 3.3.2 for further details on the selection of thermally treated Alloy 690).

3.2.1 Weld Qualification Program 9

~

A weld qualification program will be conducted to qualify the laser welding process for the use in installation of 0.740 nominal OD sleeves into 0.875 nominal 00 tubes. The qualifications performed in accordance with the rules of the ASME Boiler and Pressure Vessel Code,Section IX, Subsection QW will remit in the generation of information necessary to prepare a procedure qualification record as well as the welding procedure specification.

In the course of qualifying the welding process, welding processes will be generated for:

- Sleeve weld joints made outside of the tubesheet i

- Sleeve weld joints made within the tubesheet These two basic processes cover all of the weld joints necessary for f l

Weld installation of any of the three types of sleeves discussed earlier.

process qualification will be completed before field implementation.

I 1

3-7 1

j

WESTINGHOUSF PROPRIETARY CLASS 2 The qualifications for each of the weld joints will be performed separately.

The essential welding variables for all the weld qualifications are in accordance with ASME Code Case N-395. While the 1986 ASME code (the latest NRC approved edition) does not address the installation of sleeves, and hence does not have any specific rules relative to the design and testing of the weld, the

[ rules of the 1986 ASME code are foslowed to the maximum extent possible.

Sleeving was added to the 1989 ASME code. Rules delineated in the 1980 edition are used where they provide more guidance than, and do not conflict with, the

}

1986 edition.

To provide similitude between the specimens and the actual installed welds, l representative field prot. esses are used to assemble the specimens. The out of tubesheet joints are representative in length and diametral expansion of the hydraulic expansion zone. The in tubesheet joint specimens are assembled in collars which are representative of the unit cell of the Series 51 steam generators. The sleeve and tube materials are consistent with the materials and dimensional conditions representative of the field application.

3.2.2 Sleeve Weld Joint Qualification

[

Five specimens of each weld process to be qualified will be assembled and welded in accordance with the prescribed welding procedure. The specimens will be sectioned longitudinally and liquid penetrant examined. Two cross sections of each specimen will be polished, etched, and examined at 10X.

An additional five specimens of the upper joint will be welded and tested for information. Two will be tension tested in acccrdance with ASME Code Section IX,SubsectionQW-462.l(e). Two will be cut 3xially to perform two face and

~

root bend tests and two side bend tests. One specimen will be archived for future reference.

3.2.3 Acceptance Criteria The qualification welds shall be free of cracks aan ;ack of fusion and meet

. design requirements for weld throat and minimum .eakage path. The welds shall meet the liquid penetrant requirements of NB-3530.

The guidelines for the tensile tests are:

3-8

~

WESTINGHOUSE PROPRIETARY CLASS 2 The specimen fails in the weld metal or fusion line at a tensile strength  ;

of 80,000 psi r, ;reater, or '

('

The specimen fails in the tube outside the weld or fusion line at a tensile  !

. strength of 76,000 psi (95% of the tube minimum tensile strength) or greater, or The specimen fails in the sleeve outside the weld or fusion line at a tensile strength of 95,000 psi (95% of.the sleeve minimum tensile strength) '

or greater.

3.3 Design Verification: Test Programs 3.3.1 Design Verification Test Program Summary Detailed material test and design verification programs were performed to evaluate the laser welded sleeve and some programs are currently in progress.

The purpose of the programs is to verify the ability of the sleeve to span the degraded region in a steam generator tube and maintain the steam generator tube

. primary-to-secondary pressure boundary under normal and accident conditions.

These programs include assessment of the structural integrity and corrosion resistance of sleeved tubes.

The objectives of the progre.:, are to :

Verify the selection of Alloy 690 in the thermally treated condition as the sleeve material.

, - Establish the process parameters required to achieve

,atisfactory installation and performance. (See Section 4.)

Verify the structural strength of the sleeved tube under normal and accident conditions.

Verify the fatigue strength of the siteved tube under I

, transient loads considering the design objective of the

~

plant.

Establish the corrosion performance of the as-installed laser 4 welded joint.

The data base from previous test programs, in addition to the laser we'd qualification program, verifies the adequacy of the design.

3-9

-o , -

The sleeve material is Alloy 690 (UNS N06690) manufactured to the requirements of ASME SB 163 with supplemental requirements of Code Case N 20. The material

- has been thermally treated (TT) to further enhance resistance to corrosion in steam generator primary and secondary side environments. - This material has exhibited good resistance to degradation in extensive laboratory corrosion

[ tests and, in the mill annealed condition, as implant samples in operr* N penerators. Alloy 690 TT is currently being used as steam generator tubes and has been used in previous sleeving programs. ,

The mechanical performance evaluations use acceptance cdteria based on plant technical specification leak rates. Testing encomp u ses static and cyclic pressures, temperatures and loads. Joints fabricated from Alloy 690 TT sleeves laser welded into Alloy 600 tubes are tested to verify that structural requirements are met.

Extensive proof-of-principle and acceptance testing of laser welded joint establish that the laser welds meet dimensional and general w.ld quality requirements. The weld process qualifications verify that the process is capable of repeatedly producing sound welds.

l The corrosion resistance of the heat treated laser welded joint is verified through extensive testing under accelerated corrosion conditions.

Details of the various verification test programs and the results- are provided below.

3.3.2 Corrosion and Metallurgical Evaluation The objectives of tM w rosion and metallurgical evaluations are: (1) to verify that thermally treated Alloy 690 (690 TT) is a material of choice l

for use in steam gen..4 tor environments and, (2) to verify that the l sleeving process does not have a detrimental effect on the serviceability l

I of the existing tube or sleeve.

l 3.3.2.1 Materials i

l The material of construction for the tubing in Westinghouse designed Series 51 Steam Generators was Alloy 600 in the mill annealed (MA) condition. Alloy 600 3-10

is a nickel base alloy with a minimum 72 percent nickel, 14-17 percent chromium and 6-10 percent iron. The selected sleeve material is nickel base Alloy 690 in the thermally treated condition. The Alloy 690 is also a nickel base alloy with minimum 58 percent nickel, 27-31 percent chromium and 7-11 percent iron.

.' For steam generator tubing applications the carbon is limited to 0.05 percent maximum and cobalt is limited to 0.10 percent maximum. The sleeve material is manufactured to the requirements of ASME SB-163 and the supplemental requirements of ASME Code Case N 20. '

The primary concern with the materials used in the steam generator tube pressure boundary is the resistance to stress corrosion cracking (SCC). Alloy 600 tubes have on occasion experienced some degradation when stress and environmental conditions were unfavorable. Laboratory testing has shown Alloy 690 TT to have a degree of corrosion resistance superior to that of Alloy 600 in the mill annealed condition and equal to or better than that of thermally treated Alloy 600. The higher chromium content of Alloy 690 contributes to the enhanced corrosion resistance. In addition, the thermal 1

~

treatment further enhances the resistance to stress corrosion cracking. This er,hancement is related to the distribution of carbides in the grain boundaries as illustrated in Figure 3.3.2-1. When the carbides are precipitated as a continuous or semi-continuous network in the grain boundaries the corrosion

- resistance is much higher than for microstructures characterized by a relatively low density grain boundary carbide.

The resistance to stress corrosion cracking in Alloy 690 has been studied extensively in various stressed test configurations including stressed reverse U-bends, C-ring samples, constant extension rate tests and model

. boiler tests. Field experience was also gained from irrplants in Diablo Canyon 1 and Farley Unit 1. Alloy 690 TT is currently recommended for tubing in steam generator applications.

The laboratory stress corrosion tests in both off-chemistry secondary side and

. primary side environments have continually demonstrated the additional stress corrosion cracking resistance of thermally treated Alloys 600 and 690 as

.- compared to mill annealed Alloy 600 material. Direct comparison of thermally treated Alloys 600 and 690 has further indicated an additional margin of SCC resistance for thermally treated Alloy 690. (Table 3.3.2-1).

3-11

I i

l l

l b

a Cumulative Percent Stress corrosion cracking M I I -

100 H #

. go -

Microstructure 80 -

Process ,D ,1 ,C, -

M A + 5T + P e ~

70 A M A + TT .

60 - M A + TT -

M A + 5T OO

+ P + TT - -

50 -

40 -

30 -

g.,, -----.

20 -

10 -

0 r-- @ O 'O O-1000 2000 3000 ,4000 Excosure Time, n MA= Mill Annealed ST= Solution Treated

. P= Polished TT= Thermal Treated D= Discontinuous S= Semi-continuous C= Continuous-

~

Figure 3.3.2-1 Influence of Process and Microstructure on SCC-Resistance of Alloy 690 3-12 1

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

The caustic SCC performance of mill annealed and thermally . treated Alloys 600 and 690 was evaluated in 10 percent NaOH solutions at temperatures from 288'C i to 343*C. Since the test data were obtained over various exposure intervals ranging from 2000 to 8000 hours0.0926 days <br />2.222 hours <br />0.0132 weeks <br />0.00304 months <br />, the test data were normalized in terms of i average crack growth rates determined from destructive examinations of the C-ring test specimens. The results integrate the initiation and propagation rates

[

The crack growth rates presented in Figure 3.3.2 2 indicate that thermally treated Alloys 600 and 690 have cdditional caustic stress corrosion cracking resistance to that of Alloy 600 in the mill annealed condition. The performance of thermally treated Alloys 600 and 690 is approximately equal at temperatures of 316'C and below and in both cases the indicated rates are low.

At 332'C and 343*C, the additional SCC resistance of thermally treated Alloy 690 is observed. in all cases the SCC morphology was intergranular, which is typical for these alloys.

C-ring specimens were tested in 10 percent Na0H solution at 332'C to index the reistive intergranular attack (IGA) resistance of Alloys 600 and 690.

- Comparison of the IGA morphology for these C-rings stressed to 150 percent of the 0.2 percent yield strength is presented in Figure 3.3.2-3. Mill annealed

Alloy 600 is characterized by branching intergranular SCC extending from a 200 pm front of uniform IGA. Thermally treated Alloy 600 exhibited less SCC, and IGA was limited to less than a few grains. Thermally treated Alloy 690 exhibited no SCC and only occasional areas of intergranular oxide penetrations that were less than a grain deep.

! . The enhancement in IGA resistance can be attributed to two factors: heat treatment and alloy composition. A characteristic of mill annealed Alloy 600 C-rings exposed to a deaerated sodium hydroxide environment is the formation of intergranular SCC with uniform grain boundary corrosion (IGA). The relationship between SCC and IGA is not well established but it appears that l- IGA can occur at low or intermediate stress levels. Therma? treatment of Alloy 600 provides additional grain boundary corrosion resistance along with l

In the case of A119y 690, the composition provides I -., additional SCC resistance.

l' an additional margin of resistance to IGA and the thermal treatment further enhances the SCC resistance.

1 3+13

Table 3.3.2-1 SV", MARY OF C0RROS10N COMPARISON DATA FOR MILL ANNEALED ALLOY 600 AND  !

i THERMALLY TREATED ALLOYS 600 AND 690

~

1. Thermally treated Alloy 600 tubing exhibits greater resistance to SCC and l

IGA in both secondary-side and primary-side environments than mill annealed Alloy 600.

2. Thermally treated- Alloy 690 tubing exhibits additional SCC resistance i

compared to thermally treated- Alloy 600 in caustic, acid sulf:te, and primary water environments.

3. The alloy composition of Alloy 690 provides additional intrinsic resistance to caustic induced IGA compared to Alloy 600 MA or TT. Moreover, use of the thermal treatment confers even greater resistance to corrosion.

G E'

S 4

a 9

k 3-14

, . 1

~~

1 ,

t j -l '

l 1

)

i i

1.000 I o Ri loy 600 MR a Alloy 600 TT H

o Allay 690 TT 0 E

p,D.100

$1 N r/ o

'^

$3 23

[6 /^ g ob0.010 -

Lui

~

7 X ^'

C

,7 0.001 '

280 290 300 310 320 330 340' 350 360 TEMPERRTURE (Degrees C) i Figure 3.3.2-2 SLC Growth Rate for C-Rings (150 percent YS and TLT) in 10 percer.t NaOH 3-15 l

l T

t c , . . , . . '. / k sk.

. , -.7 h

- . , - y., -

.e .,,J' . .. - % ,

-( , . 's I r. \ -

. % ]'s.t

• N' 7..

. ., l .. N .

~,.' % s r,

~ -

.. .s y . , .

1 g m _- .

1 I.- A, '

s .t

'A I

l 1

h- ,

g L., . /

s a

r I

,p *c '

P . q' '

. 1 ,. .

L..

. 1h .

.1 600 MILL ANNEALED 4

.y . .

d .,...

.. . s. d' .

.. . ..~. . ,. ..

~~ '

. lb 600 THERM AL TR EATED

. _.- e.  ;

.....v... ..i

-y

+.:

• 9.> . ' yt . . % . .-

.,j,.;,

. /.

7..w . ,. t,'d .. $, q:.py #.g.c:

. J3. s . . .< ,- ,,,

v.%'.1 .s , 4 r.

i

... .'\. . .; e . .; . , . . . . . . :.,

i . .s,* . .: ; w_;_y.; .< . . .

+ .. ., . . . ,7 .,. v .-.

.. .. ~., -;.. .:- . .,  ;

o. . ..:, .. ,. ,0..,..~ , x+.. .:.. .,.,: .:, x

<.,a,..- ..e tm 690 THERM AL TREATED f

~

Figure 3.3.2-3 Light Photomicrographs Illustrating IGA 7 '

After 5000 Hours Exposure of Alloy 600

, and 690 C-Rings to 10% NaOH at 3320C (6300F)

, 3-16 s

,,,a- - - - - , . .

The addition of oxidizing species to deaerated sodium hydroxide environments results in either no effect or a deleterious effect on the SCC resistance of thermally treated Alloys 600 and 690. Depending on the specific oxidizing species and concentration (Table 3.3.2-2). The addition of 10 percent copper oxide to 10 percent sodium hydroxide decreases the SCC resistance of thermally treated Alloys 600 and 690, and it also can modify the SCC morphology with the generation of transgranular cracks in Alloy 690. The mechanism for this change

, is not known, but it may be related to a change in electrochemical potential.

The specific oxidizing species and the ratio of oxidizing species to sodium hydroxide concentration appear to effect the cracking mode. The apparent deleterious effect on SCC resistance does not appear at lower copper oxide or sodium hydroxide concentrations.

Mill annealed and thermally treated Alloys 600 and 690 were also evaluated in 8 percent sodium sulfate environments. The room temperature pH et the beginning of the test was adjusted using either sulfuric acid or ammonia. As the pH was lowered, the SCC resistance for mill annealed and thermally treated Alloy 600

, was decreased. In comparison, thermally treated Alloy 690 did not crack even at a pH of 2, the lowest pH tested (Figure 3.3.2-4).

Primary water SCC test data are presented in Figure 3.3.2-5. For the beginnina-of-fuel-cycle water chemistries,10 of 10 specimens of mill annealed Alloy 600 exhibited SCC, while only 1 of 10 specimens of thermally treated Alloy 6(0 exhibited SCC in exposure times of about 12,000 hours0 days <br />0 hours <br />0 weeks <br />0 months <br />. In the end of-fuel cycle water chemistries, 9 of 10 specimens of mill annealed Alloy 600 exhibi;ed SCC, while 3 of 10 specimens of thermally treated Alloy 600 exhibited SCC. In comparison, no SCC was observed in the mill annealed or thermally treated Alloy 690 specimens in either test environment after 13,000

,' hours of testing.

The conclusion from the extensive tests on the thermally treated Alloy 690 in both primary and secondary environments, including caustic and sulfur bearing environments, is that the material is highly resistant to stress corrosion cracking. The resistance of Alloy 690 TT to steam generator service conditions is also verified through the field implants where no degradation has been noted after up to nine years of service ( Farley Unit 1 ).

3-17

I Table 3.3.2-2

- EFFECT OF OX1DlZING SPECIES ON THE SCC SU,SCEPTIBILITY OF THERMALLY 1REATED ALLOY 600 AND 690 C-RINGS IN DEAERATED CAUSTIC Temperature Exposure Alloy Alloy _

Time ' Hrs) 600 TT 690 TT Environment (OC) 4000 increased Increased 10 Percent NaOH + 316 Susceptibility Susceptibility

• 10 Percent Cu0 332 2000 No effect No effect 10 Percent NaOH +

1 Percent Cu0 332 4000 No effect No effect 1 Percent Na0H +

1 Percent Cu0 316 4000 No effect No effect 10 Percent Na0H +

10 Percent Fe304 316 4000 No effect No affect 10 Percent Ha0H +

10 Percent Sg02

• Intergranular and transgranular SCC.

e S

3-18

C pm

-0.040 1000 - c O ALLOY S00 MR A RLLOY 600 'T O ALLOf 000 TT - 0.032 l 800 l g .

0.024 1 l600 0 i g $ .l g 0.016 400 6  ;

g 0.008 200

~

~

O.000 0 8 9 10 11 1 2 3 4 5 6 7 ROOM TEMPERATURE. pH

-l Figure 3.3.2-4 SCC Depth for C-Rings ( 150 percent YS )

in 8 percent Sodium Sulfate 3-19

I l

i REVERSE U-BEND TESTS RT 360 DEGREES C C680 DEGREES F)--l

' BEGINNING OF FUEL CYCLE PRIMARY WATER  :

seem 12 -

im 10 8

- /  !

( -

30

- 8 4

' ' ****- o h0-2 '

10000 12000 14000 16000 m 2000 4000 6000 8000 m O

H I E

o m m z

d m END OF FUEL CYCLE PRIMRRY WRTER

!' " 12 - - roo g 00 l' 91 -

6

- M n eoo

~

g4 -

e f n eso, m eco

-2 10000 12000 14000 16000 O 2000 4000 6000 8000 EXPOSURE TIME CHOURS) l l

.' l 1

l. -

l l l

Figure 3.3.2-5 Reverse U-Bend Tests at 3600C (6800F) 3-20 i

l

l 3.3.3 1.aser Welded Joints Test programs were conducted to demonstrate that the joint formation processes do not degrade the integrity of the tube or sleeve. The testing considered the individual effects of the hydraulic expansion and laser weld process on the sleeve tube assembly. In addition, the effect of welding in tubes that have deposits on the outside surface was also evaluated.

The data used to evaluate the effect of hydraulic expansion are based on tne testing conducted to demonstrate the corrosion resistance of the Mechanical or Hybrid Expansion Joint (HEJ) Sleeve. The data are appi. cable since a hydraulic expansion is utilized to bring the sleeve into contact with the tube prior to the welding operation. The parameters used in the HEJ sleeve configuration bound the parameters used in the laser welded sleeve joint, making the data and the conclusions reached relative to the effect of hydraulic expansion on corrosion performance applicable to the laser welded sleeve. The data is not reproduced within this report, the results are discussed later.

The assessment of the effects of welding is based on corrosion testing of prototypical upper joint samples in an accelerated 7500F steam environment.

This testing establishes a basis for comparison of the corrosion rates of a

- a sleeve installed by laser welding to a typical roll transition in a steam generator tube .

3.3.3.1 Expansion Processes s

The formation process for the lower joint on the full length tubesheet sleeve uses a combination of hydraulic expansion and roll expansion followed by a single pass laser weld. If only a roll expansion were applied, the stresses in the roll expanded transition region of the sleeve would be about 40 ksi as deduced from polythionic acid tests using conventional tubesheet specimens. If the sleeve were only hydraulically expanded, these stresses would be about 20 ksi. To retain the low residual stresses inherent in hydraulic expansion with the load carrying capability of a roll expansion, a combined

hydraulic plus roll expansion is utilized to bring the sleeve into contact with the tube prior to welding. The formation process for the unoer joint is hydraulic expansion.

3-21 ,

Stress levels in the outer tube are influenced by the expansion technique. The absolut; magnitude of these stresses will depend on the specific diametral expans on. An extensive program was previously conducted to establish the accept.oility of the Hybrid Expansion Joint (HEJ Mechanical Joint) for use in the installation of sleeves in steam generaters. The program used a number of testing techniques to verify the total joint t rformance in various operating environments. The conclusion based on these tests is that hydraulic expansion of the tube during joint formation has little impact on the corrosion

.' resistance of the sleeve / tube.

3.3.3.2 Outside Diameter (OD) Surface Condition r The laser welding sleeve process and the subsequent post-weld heat treatment (PWHT) is typically applied to tubing that has been in service. Since the sleeving operation is conducted from the primary side, no operations are l conducted on the OD surface. In operational steam generators, the outside I surfaces of the tubes can collect botier water deposits and scales. These are typically oxides or minerals that are in the thermodynamically stable form of the constituent elements, magnetite being the most prominent deposit. At the l." temperatures of the OD of the tubing during the sleeve welding or during the PWHT, these compounds are typically stable and do not thermally decompose. All

- such compounds have molecular structures that are too large for diffusion into the lattice of the Alloy 600 tubing. Reactions between these stable oxides and minerals and the alloying elements of the Alloy 600 tubing are thermodynamically unfavorable. Consequently the presence of boiler sludge / scale species on the 00 surfaces of tubes that receive the temperatures

- associated with LWS is not expected to produce deleterious tube-sludge / scale

. interactions.

Westinghouse has implemented in-situ heat treatment of Row I and Row 2 U bends in service exposed tubing in a number of plants as well as heat treatment of laser welded sleeves at a European utility. The temperatures achieved in these heat treatments are in the ranges of interest in the LWS operations. There have been no reported instances of tube distress in thermally stress relieved

- U-bends or sleeve joints that could be attributed to the presence of sludge and/or scale on the tube OD during the stress relief.

1 3-22

Three tests performed as a part of the development of a sleeve brazing technique applied previously in the field also support the preceding discussions. The first test involved a laboratory evaluation in which a braze cycle was applied to tubing in contact with simulated plant sludge from Southern California Edison (SCE). The braze cycle involved (

la,c. Bend tests of longitudinal sections removed from the brazed area showed no embrittlement as a result of the thermal cycle. A second test involved microprobe analyses of polished metallographic cross sections. Results indicated the presence of Fe, Ni, Cr, Cu and Zn on the tube OD surface, but no evidence was found of diffusion into the tubing. A third test involved removal of a tube from San Onofre which was brazed in the region of sludge. The pulled tube was analyzed for the presence of contaminants on the 00 surface and beneath the OD surface. The microprobe results indicated Fe, P, Si, Cu, Ca and Na on the tube 00, but no evidence was found of diffusion into the tube.

• In addi+ ion to the above tests, archive tubes from Beznau 1 and Turkey' Point 4 were welecd and a microanalytical examination was made for contaminant ingress

before and after welding. Before welding, ['

l l-ja,c, A final test involved metallographic observations of three areas on a U-bend of l

- Alloy 600 tubing which was coated with sludge and heat treated in air to 13500F for 10 minutes. The analysis of the sludge used was: 62 wt.% Fe3 04, 30% Cu, 3% Cu0, 2% Zn0, 2% CaSO4 and 1% MgCl2 . The sludge was attached to the 0.D. of

- the U-bend using acrylic paint as a binder. Microscopic observation of the tubing following heut treatment indicated that the presence of the sludge did not cause any significant intergranular or general corrosion of the tube

l. surface.

l-

- To summarize, several observations have been made for a variety of Alloy 600

' samples heated to temperatures from 1350*F to above 2000'F in the presence of typical secondary side chemical spec.es. No significant diffusion, corrosion, or embrittlement of the tubing has been found.

3-23 i

1 3.3.3.3 Stress Corrosion Testing of 1.aser Welded Sleeve Joints The resistance of the laser welded sleeve joint to in-service corrosion is related to the resistance of the Alloy 600 tubing to intergranular stress

.' corrosion cracking (IGSCC). The sleeve material, Alloy 690 TT, has been demonstrated to be highly resistant to IGSCC under steam generator conditions Stresses in the tubing, either service imposed or residual, (Section 3.3.2.1).

are a major factor determining the response of the material in terms of IGSCC.

Two sources of resi A:31 stresses in the laser welded sleeving process are a) minor stresses related to the hydraulic expansion during sleeve placement and b) residual stresses that occur as the r,siten weld pool solidifies.

This section summarizes results of a testing program to evaluate the PWSCC resistance of laser welded upper sleeve joints used to install sleeves in degraded steam generator tubing. The program included both as-welded and as heat treated jc,ints. The testing was conducted under conditions which accelerate corrosion in steam generator materials that may be susceptible to stress corrosion cracking in long term steam generator service. Some of the laser welding processes included in these corrosion tests are not identical to

- the process to be used at J.M. Farley but are considered to be representative.

One, the CO2 laser process, has been used previously in field sleeving applications.

3.3.3.3.1. Corrosion Test Description An accelerated corrosion test ueveloped by Westinghouse is used as a means to evaluate the resistance of steam generator materials to degradation in steam generator primary water environments. The test produces the same type of degradation through intergranular stress corrosion cracking that '

has been observed in some mill annealed Alloy 600 steam generator tubing.

The test has also been found to provide the same relative ranking of material resistance to IGSCC that has been observed in service.

The accelerated test is conducted in an autoclave operating at 750*F (4000C) l:

I with steam at 3000 psig. The steam contains ( _

Ja,c,e The 10 of the specimen is exposed to the 3000 psi doped steam while the OD sees undoped steam at 1500 psi. The accelerating effect of the test is 3 ?4

1 i

related to hydrogen at 3000 psi and also to the dopants. Reverse U bend tests have indicated the pressure effect acceleration to be a factor of four.

1 The configuration of the laser welded specimen used in this corrosion program is a full upper joint as illustrated in Figure 3.3.3 1. The sleeve joints were l fabricated using equipment and practices typically used in field sleeving I operations. The doped steam environment is introduced to the inside of the sleeve and has access to the ID of the sleeve and, on one side of the weld joint, to the OD of the sleeve, the ID of the tube and the weld. The other side of the weld joint and the outside of the tube are exposed to the 1500 psi steam environment. The 1500 psi differential across the tube wall simulates the active loading that is present in operating steam generators. In this way l it is possible to test the weld under stress conditions similar to those in the generator.

l The corrosion performance of the sleeve weld joints is compared with the performance of mockups of roll transitions exposed in the same test. The roll transition control samples illustrated in Figure 3.3.3-2 are representative of l the transitions found at the top of the tubesheet in full depth, hard rolled steam generator tubes. The inclusion of the potentially IGSCC susceptible I.

configuration (the roll transitior,' in the test provides verification of the ,

- aggressiveness of the corrosion tess environment. Any variability in the aggressiveness from one autoclave run to another can be accounted for by having roll transition controls in each run.

The time to cracking of the test sample is measured in the accelerated test.

- For both weld samples and roll transitions, cracking time is defined by the appearance of through wall cracks which is reflected in a loss of the 1500 psi differential pressure across the weld and tube.

. 1 l

l 3-25 l i

4 2.0 a.o swagelom

. s tainle s s tube me%,. m en. . - pressure' l

. steel serial weld Titting

. plug numbers u-

, 2 YA liH

= _

v c

i sleeve i 4.0 inches 0.5

. Inches .

l

.. l 6.5 inches 9.0 inches l

l l

~

Figure 3.3.3.-l ACCELERATED CCRR0SION TEST SPECIMEN

' WELDED JOINT CONFIGURATION 1

3-26

collar l

. swagelock ,

stainless tuloe pressure l

/

steel fitting )

- s WA h

." 3.0 inches 8.0 ' aches >

9,0 inches l

l l

I Figure 3.3.3.-2

. I ACCELERATED CORROSION TEST SPECIMEN ROLL TRANSITION CONFIGURATION l

3-27 l

i 3.3.3.3.2 Corrosion Resistance of laser Welded Joints

?

Most of the welded joint corrosion samples and all the roll transition sections were f abricated from mill annealed Alloy 600 tubing from Heat NX-1019. This is a high carbon heat (0.04% C) which previous testing has shown to be sensitive to PWSCC and has been used in a variety of corrosion test programs over the laser welded samples was also fabricated from past several years. A set of CO2 a lower carbon (0.02% C) mill annealed Alloy 600 tubing from Heat NX-9621 which was also susceptible to PWSCC. The lower carbon heat was included to determine if the carbon difference produced adverse metallurgical changes during welding.

A third set of CO2 laser welded sleeves were included from Heat NX-2721 which-was more resistant to cracking than Heat NX-1019 when tested as reverse U-bends.

The response of laser aided joints to the accelerated corrosion conditions is shown in Figures 3.3.3-3 and 3.3.3-4 for C02 laser welds and in Figure 3.3.3-5 for Nd:YAG laser welds. These figures are Weibull distribution plots of the cumulative percentage of samples exhibiting cracking as a function of time.

The as welded joints generally exhibited times for through wall IGSCC in (

]a,c than that of the roll transitions. One tubing heat, Heat NX-2721, exhibited [

The

' la,c in the welded joint as in the roll transition.

overall cracking response in these tests is consistent with expected behavior since both fabrication processes, i.e., hard rolling and welding, introduce residual stress in the Alloy 600 tubing, and in the case of the welded sleeve joint, in the weld and in the Alloy 690 sleeve also. The magnitude of the

' residual stress is potentially different for each process which may explain the difference in the times for cracking between the roll transitions and the as welded joints.

The cracking in the as-welded joints occurs adjacent to the weld in the Alloy 600 base material as shown in Figure 3.3.3-6. It is intergran ar and typical

- of IGSCC. Hinor cracking was also noted at the surface of the weld on a YAG 1 laser welded corrosion test sample. No cracking or other corrosion effects I

were noted in the Alloy 690 sleeve material which is consistent with the good corrosion resistance of the 690 TT as discussed previously.

3-28 1

- a,b,e 4.

8 i

l .  !

l 1

Figure 3.3.3-3 l Cumulative Percent Cracking for 002 Laser Welded Sleeves in 7500F Accelerated Steam Corrosion Test 3-29 l

a,b,e l

-l 1

1 1

l n

Figure 3.3.3-4 l l

Cumulative Percent Cracking for CO2 Laser Welded Sleeves in 7500F Accelerated Steam Corrosion Test l

1 1

3-30

1 l l

-- - a,b e  ;

I i

l 4

j l*

t.

l i

1 Figure 3.3.3-5 Cumulative Percent Cracking for YAG Laser Welded Sleeves in 7500F Accelerated Steam Corrosion Test 1

3-31 l

Since the cracking is related to a large extent to residual stresses, a reduction in the residual stress level will enhance the corrosion resistance of the welded joint. During the CO2 laser weld program, extensive development of a post weld heat treatment was done and a local stress relief treatment that

. encompassed the weld and the heat affected zone was developed. The development program determined that (-

Ja,c.e would reduce the level of residual stresses without

, significant microstructural changes.

The effectiveness of a stress relief is evident in figure 3.3.3-3 mea

(- la,c in the time to cracking in heat treci.ed welds over as fabricated welds can be seen. The beneficial effect of stress relief is also evident in the Nd:YAG laser welds in both the conduction limited weld i

(CLW ) and continuous molten pool ( CMP ) weld regimes. The test of the stress relieved CLW joints [.

,]a,c . This represents more than (. Ja,c in time to cracking over that of the as-welded joint. The corrosion test on the stress relieved continuous molten pool weld was continuing with no indication of cracking after [ ]a,c hours.

The effect of the stress relief can also be seen in the cross section of the heat treated spot weld shown in Figure 3.3.3-7. [

l ja,c In addition there was no evidence of the minor corrosion at the weld surface noted previously in the as-welded, corrosion test sample.

Based on these data, the stress relieved, laser welded sleeve joints will have good resistance to IGSCC. In the accelerated corrosion tests, they exhibit corrosion rates that are (

-]a,c than top of l tubesheet roll transition. Considering-the stress dependency difference i between the 7500F steam tests and steam generator environments, the corrosion i

rate of the sleeve joints is predicted to be ( ']a,c than the top of the tubesheet roll transitions.

i

{

l i 3-32

~

8

~

• l r

t

[.

l Figure 3.3.3-6 IGSCC in Alloy 600 Tube of YAG Laser Welded

Sleeve Joint after 109 Hours in 7500F Steam Accelerated Corrosion Test 3-33

O l

. j l

"I

• 1 L

4 l .

t Figure 3.3.3-7

1.-

' Minor IGSCC in Alloy 600 Tube of Stress Relieved YAG Laser Welded Sleeve Joint after 1000 Hours 1

in 7500F Steam Accelerated Corrosion Test 3-34 l

l l 3.3.4 Mechanical Test Program for the Laser Welded Joints-N

[ 3.3.4.1 Description of the Laser Weld Joint Test Specimens l

.' There are three types of laser welded joints. One type is a joint formed out of the tubesheet. This joint is identical for the uppcr and lower joints of a tube support plate sleeve, the upper joint of an elevated tubesheet sleeve, and -

the upper joint of a full length tubesheet sleeve. A second type is the lower joint for the elevated tubesheet sleeve. The third type is the lower joint for i

a full length tubesheet sleeve. A description of each type of joint follows.

3.3.4.1.1 Description of the Out of Tubesheet Test Specimens The test specimen fabricated for the out of tubesheet LWJ testing is shown in Figure 3.3.4.1.1-1. The tube is honed and the sleeve hydraulically expanded using the field process. The joint consists of a 360 degree partial penetration-laser weld within the hydraulically expanded region. End plugs are welded to the two ends of each specimen. Each end plug is threaded so that the specimen 5 can be axially loaded during the various tests. One end plug contains a 1/4 inch OD tube so that the specimen can be hydrostatically pressurized to simulate the primary-to-secondary pressure difference. Provision is made for collecting leakage if it issues from the anulus between the tube and sleeve.

t 3.3.4.1.2 Description of the Elevated Tubesheet Sleeve Lower Joint Specimens l

The elevated tube /tubesheet sample is manufactured so that it is representative of the Series 51 steam generators which have full depth rolled tubesheets (Figure 3.3.4.1.2-1). The tube is installed in the collar using the representative roll expansion process from steam generator manufacture. The tube ID is honed using the field process. The preformed sleeve is then inserted into the tube and the lower joint formed by application of the field

- hydraulic expansion and weld process. The final step in the test specimen f 1

I preparation consisted of welding end plugs and caps to the specimens so that

,: they can be internally pressurized and axially loaded.

3-35 l

a'b,e 1

i I

lO f

l .

4 l

1

. 1 l

1

)

l l

l i l

l 1.

1 Figure 3.3.4.1.1-1 Out of Tubesheet Laser Welded Joint Test Specimen 3-36 l

L ,

- a,b e-O l

l l

l l

l l

I 1

Figure 3.3.4.1.2-1 Elevated Tubesheet Sleeve Lower Joint Test Specimen -

3-37

3.3.4.1.3 Description of Full length Tubesheet Sleeve Lower Joint Specimens

.' The tube /tubesheet mockup is manufactured so that it is representative of the Series 51 steam generators (Figure 3.3.4.1.3-1). The tube is installed in the collar using the representative manufacture roll expansion process. The tube ID is honed per the sleeve installation procedure. The preformed sleeve is then inserted into the tube, hydraulically expanded, roll expanded and the

~

lower joint formed by welding. The final step in the test specimen preparation consisted of welding end plugs and caps to the specimens so that they can be internally pressurized and axially loaded.

l 3.3.4.2 Description of Tests for the Laser Welded Joints Test specimens for representative out of tubesheet and in tubesheet LWJ's are tested in the sequence described below.

1. Initial leak test: The leak rate is determined at room temperature, 3110 psi and at 600*F, 1600 psi. These tests establish the leak rate of the laser welded joint af ter it has been installed in the steam generator and

! prior to long-term operation.

2. The specimens are fatigue loaded for 5000 cycles.

3. The specimens are leak tested at 3110 psi room temperature and at 1600 psi 600'F. This establishes the leak rate after a simulation of 5 years of normal operation (plant heatup/cooldown cycles) produced by step 2.

Following the intermediate leak test of the fatigua cycle specimens, they are then run for an additional 30,000 cycles followed by leak testing.

4. The specimens are subjected to simulated SLB conditions.

5. The specimens are leak tested as in Step I to determine the post-accident l

leak rate.

(

6. Specimens are subjected to axial tensile and compressive loading to failure.

3-38

i

- a,b.e.

9 l

1 l

l A

~

Figure 3.3.4.1.3 f Full Length Tubesheet Sleeve Lower Joint Test Specimen 3-39

3.3.4.3 Leak Test Acceptance Criteria l The leak rate acceptance criteria for the laser welded sleeve joints is that the sleeve shall have 'ero leakage during the testing described above. The

leak test results ca. se compared to the leak rate criteria to provide verification that the sleeve exhibits no leakage under simulated normal operating conditions or under umbrella postulated accident conditions. Leak rate measurement is based on the number of drops counted during a 10 20 minute l .

period.

3.3.4.4 Results of the Mechanical Tests of the Laser Welded Joints )

[

Ja,c,e

]a,c.e

[

Therefore, the laser welded joints are confirmed to be adequate for their l

purpose in isolating tube degradation.

3.3.4.5 Rewelding Under some conditions, the initial attempt at making a laser weld may be i interrupted before completion. If the sleeve / tube has not been perforated by the interrupted weld, an additional weld, having the same nominal characteristics as the original weld, will be made closer to the end of the sleeve than the initial weld ( Figure 3.3.4.5-1 ). Performing the reweld l

. outboard of the initial weld maintains the incomplete weld within the pressure 4

boundary formed by the sleeve, and so, in the case of no perforation, the pressure boundary of the sleeve and tube is not compromised and the sleeve installation is acceptable. If the sleeve / tube were perforated during l I

interruption of the initial weld, the tube would be removed from service.

(

3.4 Post Weld Heat Treat t l Westinghouse has extensive experience in stress relief processes from prior ,

work on U-Bend and support plate heat treat programs. The objective of the l laser weld post weld heat treatment is to relieve residual stresses in the l 3-40

3

/ w/

C I

}

l REVELDs 4 >

l

! . L ti 4 >,

- i l

• \ UPPER. LASER VELD 5

ex f,-

I ,,

~N _/

. 7

f,/ f J ,/

I

/

Y __

.[. x v +

y 9 f

/ ih #'

/\

LOVER LASER

'N VELD l

~

Figure 3.3.4.5-1 Laser Welded Sleeve with Reweld 3-41

sleeve / tube that may be introduced by application of the welding process. The length of sleeve / tube heat treated spans the weld and the adjacent heat l

( affected zone. l I To satisfactorily relieve the residual stresses, it was necessary to develop ,

the optimal heat up, soak, and ramp down power cycles. Several physical factors affect the engineering task of maintaining the tube temperature within- l j

', the required temperature band.

. 1. The tube is predominantly cooled by radiation, with minor effects of conduction and convection.

2. The physical configuration (power density) of the heat source affects heat distribution within the tube.

l

3. The heat source and the heated portion of the tube cannot be excessively long as this could result, under certain boundary conditions of tube fixity, in excessive compressive stresses within the tube which could~

produce bowing or barreling of the tube.

4. The process has to account for weld axial positional tolerances as well as heater axial positional tolerances.

To address these factors, the heat source was sized such that it heated the area of interest with sufficient margin to allow for axial position variations.

Given the heat source, a multitude of laboratory tests were performed which addressed the following issues:

l l

a. Nominal heat source power.

! b. Initial heat source power profile to expedite the time required to achieve I acceptable tube temperatures.

l c. Acceptable soak powers and temperatures.

d. Effect of varying tube emissivities.

e. Effect of a misplaced heater.

f. Circumferential tube temperature profile.

g. Axial tube temperature profile.

h. Sleeve to tube temperature gradient.

3-42

The test mockup shown in Figure 3.4 1 was used for stress relief process testing. The initial sleeve / tube samples are shown in Figure 3.4-2. The

', expansion zone and the weld were approximately centered in the sleeve. The initial test samples were outfitted with thermocouples centered around the centerline of the weld throat. These test samples included a sample with a nominal weld throat size and a sample with no weld at all. The test samples for final testing were made with twa types of welds to bound the temperature distribution profile and the power te?'ile. [

• ]a,c, The sleeve / tube samples used for final process development were prototypic of l the field sleeve / tube joint configuration, shown in Figure 3.4-3. The weld centerline was positioned 1.50" below the top of the expansion zone and the samples were equipped with thermocouples. A temperature measurement probe, which included a heater and a fiberoptic filament, was used to heat the test samples as well as to measure the sleeve wall temperature at the center of the heat zone. A control system was used to control the power input to the heater, i

~

and to monitor the power and temperature parameters on a chart for permanent l record.  ;

The results of the above laboratory testing led to a typical power profile as shown in Figure 3.4-4. This figure represents a typical profile, for a tube with a particular emissivity. In the field application, the power level would be adjusted as a function of field tube emissivities, determined through a capping process.

l 3-43

a ..2. p A .. #.. a. J .,A_ .m-. 4- _a -A _ ;mA2 e a 4

4 e

4 9

e 4

9 G

e 1

1 O

I e

9 FIGURE 3.4-1 VERTICAL TEST STAND MOCK-UP 1 3-44

~

4eb.e j j

l

. l l

G e

4 4

e r

f e

9 P

m FIGURE 3.4-2 y-INITIAL TEST SAMPLES DETAILED 3-45

I

l. I f

L

l

" 4 bie e

9 e

S l-

i.  :

l*

e l

l l

0 FIGURE 3.4-3 FIELD PROTOTYPIC TEST SAMPLES DETAILED

.3-46

am, y 4 h

e a e .i.

i dgb' e l 4

I E

9 0

9 e

e 0

9 e

h

" FIGURE 3.4-4 TYPICAL STRESS RELIEF POWER PROFILE 3-47

3.5 Analytical Verification.

3.5.1 Introduction This section contains the structural evaluation of the full length tubesheet  !

I sleeves, the elevated tubesheet sleeves, and the tube support plate sleeves (each type with Laser Welded Joints (LWJ) and Alloy 690 sleeve material),  !

performed in accordance with the rules of the ASME Boiler and Pressure Vessel Code, Section 111, Subsection NB, 1986 Edition (Reference 1)( References are

- listed on page 3-68). The analyses include the primary stress intensity

- evaluation, the primary plus secondary stress intensity range evaluation, and the fatigue evaluation for mechanical and thermal conditions which umbrella the loading conditions specified for the Joseph M. Farley Units 1 and 2 Series 51 steam generators.

3.5.2 Component Description The general configurations of the sleeve-to-tube assemblies with LWJ's are presented in Figures 3.2-1, 3.2-2 and 3. 13. The critical portions for each

- assembly are the upper LWJ, the lower LWJ, and the straight unexpanded sections

~

of the sleeve and tube between the two joints.

3.5.3 Material Properties The sleeve material is Alloy 690 procured to SB-163 and Code Case N-20 (Reference 2). The tube material is SB 163 (A11oy-600) (Reference 3). All material properties are as specified in the ASME Boiler and Pressure Vessel Code,Section III, Appendices (Reference 3).

3.5.4 Code Criteria The calculated stresses due to various loadings must comply with the criteria set forth in the ASME Code (Reference 1). These criteria are given in Tables 3.5.4-1 through 3.5.4-4.

3-48

Table 3.5.4-1

$ CRITERIA FOR PRIMARY STRESS INiENSI:- .ALUATION (SLEEVE) e )

8,C,e e

0 4

l O

?

b l

e l

l l

)

(

3-49

• Table 3.5.4 2 I CRITERIA FOR PRIMARY STRESS INTENSITY EVA1.UAT10N j (TUBE)

! - = ,

a,c.e 9

e i

i e

9 e

e 4

e s

4 e

3-50

Table 3.5.4 3 E

CRITERIA FOR PRIMARY PLUS SECONDARY AND TOTAL STRESS INTENSITY EVALUATION '

o.

• (SLEEVE) 1 8 ', C , e 9

e a

e 9

4 e

M e

e e

e 3

i Table 3.5.4-4

, CRITERIA FOR PRIMARY PLUS SECONDARY  !

AND TOTAL STRESS INTENSITY EVALUATION ,

( (TUBE) 8,C,e ,,

l

'I e

I I

e s

t i

O I

I 3-52

i 3.5.5 Loading Conditions Evaluated

. The generic loading conditions are specified below:

$ 1. Design cond'itions .

l

a. Primary side design conditions P = 2500 psia

. T 6500F i

~

b. Secondary side design conditions P - 1100 psia T 6000F i

c. Maximum primary to secondary pressure differential - 1600 psi, T - 6500F

d. Maximum secondary to primary pressure differential - 670 psi,  ;

T - 6500F

2. Full load steady state conditions are:

4 Primary side pressure, psia 2250.0 Hot leg temperature, OF 616.8 Cold leg temperature, OF 552.2 Secondary side pressure, psia 720.0 Feedwater temperature, OF 427.3 j Steam temperature, OF 506.3-Zero loed reactor coolant temperature, OF 547.0 Other operating conditions are specified in Tables 3.5.7.1-1 and=3.5.7.2-1.

l .

ja c The generic operating transients have been selected from the applicable Westinghouse Design Specifications. They are conservative and envelope the J.

M. Farley Series 51 steam generators.

3-53

3.5.6 Methods of Analysis

( Structural analysis of the sleeve to-tube assembly includes model development, heat transfer and stress analysis, primary stress intensity evaluation, primary C plus secondarf stress intensity range evaluation, and fatigue evaluation utilizing the conditions described above.

3.5.6.1 Model Development Three finite element models are used in this analysis. They correspond to the three types of sleeves being analyzed: full length tubesheet sleeves,  ;

elevated tubesheet sleeves, and tube support plate sleeves. l The tolerances used in developing the finite element model are such that the maximum sleeve outside diameter is evaluated in combination with the minimum sleeve wall thickness. This allows maximum stress levels to be developed in the model. Characteristics of the models are as follows: ,__,a,c,<.

1 l

e l

3-54

I a,c.e i

o

q e

1 m

All the clement types are quadratic, having a node placed in the center of each l surface in addition to nodes at each corner.

3.5.6.2 Thermal Analysis The purpose of the thermal analysis is to provide the temperature distribution I needed for thermal stress calculations,

! Thermal transient analyses are performed for the following events:

1 Small step load increase l Small step load decrease-

'Large step load decrease 3-55 i

Hot standby operations Loss of load 7 Loss of power Loss of secondary flow j

! Reactor trip from full power l l

l The plant heatup/cooldown, plant loading / unloading and steady fluctuation events are considered under thermal steady state conditions.

An air gap is included between the tube and sleeve. Although this space may be filled with secondary fluid, assuming the physical properties of air for these elements is conservative for the thermal analysis. Priary fluid physical l

l properties are used for the gap aedium above the LWJ.

i in order to perform the WECAN thermal analysis, boundary conditions consisting of fluid temperatures and heat transfer coefficients (or film coefficients) for I the corresponding element surfaces are necessa*y, Special hydraulic and thermal analysis has been performed to define the primary and secondary side fluid temperatures and film coefficients as a function of time. Both boiling f and convective heat transfer correlations are taken into consideration.

3.5.6.3 Stress Analysis A WECAN (Reference 4) finite element model is used to determine the stress levels in the sleeve / tube configuration. Elements simulating the medium between the tube and the sleeve are considered as dummy elements.

Based on the results demonstrating the applicability of a linear elastic analysis, thermally induced and pressure induced stresses are calculated l

separately and then combined to determine the total stress distribution using l the WECEVAL computer program (Reference 5),

e Pressure Stress Analysis P For superposition purposes, the WECAN model is used to determine stress distributions induced separately by a 1000 psi primary pressure and a 1000 psi secondary pressure. The results of these " unit pressure" runs are then scaled to the actual primary side and secondary side pressuras corresponding to the 3-56

loading condition considered in order to determine the total pressure stress distribution.

The end cap pressures due to the axial pressure stress induced in the tube away from discontin'ulties are taken into consideration for undented tubes. For i assumed dented conditions, the end cap load is not required.

Thermal Stress Analysis

. The WECAN model is used to determine the thermal stress levels in the sleeve / tube configuration that are developed by the temperature distribution calculated by the thermal analysis. Thermal stresses are determined for each

- steady state solution as well as for the thermal transient solutions at those times during the thermal transient which are anticipated to be limiting from a stress standpoint.

Stresses Induced By the Tubesheet Rotation The tubesheet rotation is developed by the primary and secondary side pressures, temperature and different coefficients of thermal expansion of the steam generator components, and resulting interactions among the tubesheet, channel head and the stub barrel. The effect of the tubesheet rotation on the sleeve / tube assembly is modeled by applying the displacements or stresses developed by this tubesheet motion to the tubesheet ligament outside boundary.

Generic results of the tubesheet motion displacement calculations for Series 51 l

steam generators conditions are used in this analysis.

I l

3-57

l l

Combined Stresses The total stress distribution in the sleeve-to-tube assembly has been L determined by combining the calculated stresses as fcilows:

o Ppg (o) unit primary pressure total ,1000 P

+ SEC (o) unit secondary pressure 1000

+ (o) thermal

-+ PR (o) unit primary pressure for tubesheet 1000 motion effect

+ " (o) unit secondary pressure for tubesheet 1000 motion effect The combining procedure is performed by the program WECEVAL (Reference 5).

3.5.7 Results of Analyses Stress and fatigue evaluations of the sleeved tube assembly are completed in accordance with the requirements of the ASME Boiler and Pressure Vessel Code,Section III. The results are presented below; all requirements are met.

l 3.5.7.1 Primary Stress Intensity l

Primary stresses in the structure are developed by the primary and secondary

• pressure acting on the sleeved tube assembly.

The umbrella loads for the primary stress intensity evaluation are given in Table 3.5.7.1-1.

3-58

i TABLE 3.5.7.1 1

• UMBRELLA PRESSURE LOADS FOR DESIGN, FAULTED, AND TEST CONDITIONS PRESSURE LOAD, PSIG SECONDARY - a,c,e PRIMARY

. CONDIT10h5 S

G 3-59

The analysis demonstrates that the primary stress intensities for the laser welded sleeved tube assembly are within the allowable ASME Code limits. Design condition loads are found to be limiting for primary membrane and membrane plus l bending stress intensities. The largest magnitudes of the ratio " Calculated j Stress Intensity / Allowable Stress Intensity" are 0.97 for Primary Membrane Stress Intensity, and 0.71 for Primary Membrane Plus Bending Stress Intensity, respectively. For the algebraic sum of principal stresses, faulted conditions are limiting and the ratio of " Calculated Stress Intensity / Allowable Stress Intensity" is 0.30.

3.5.7.2 Range of Primary Plus Secondary Stress Intensities and Fatigue Evaluations Primary plus secondary stresses in the assembly are developed by the pressure acting on the sleeve and tube, by the thermal stress, and by the deformations imposed by the tubesheet rotation. Table 3.5.7.21 contains the pressure and temperature loads for maximum range of stress intensity evaluations at well as for fatigue evaluation. Maximum range of stress intensity and fatigue evaluations are performed by the program WECEVAL (Reference 5). WECEVAL is a multi-purpose code which performs ASME Code,Section III stress evaluations. At any given point or section of the model, the program WECEVAL is used to determine the total stress distribution per the Subsection NB requirements.

That is, the total stress at a given cross-section through the thickness, so-called analysis section, is categorized into membrana, linear bending, and non linear components which are compared to Subsection NB allowables. Complete

. transient histories at given locations of the model are used to calculate the total cumulative fatigue usage factor per ASME Code Paragraph NS-3216.2 (Reference 1). Sased on the sleeve design criteria, the fatigue analysis considers a design objective of 40 years for the sleeved tube assemblies.

Because of postible opening of the interface between the sleeve and the tube along the hydraulic expansion regions, the maximum fatigue strength reduction factor of 5.0 (NB-3222.4(e2)) (Reference 1) is applied in the radial dirdction I at the " root" interface nodes of the welds, which is conservative when compared to the recommeeded maximum fatigue strength reduction factor of 4.0 for l welds ( NG 3352-1 ).

9 3 60 I

1

l

?

TABLE 3.5.7.2 1 5 PRESSURE AND TEMPERATURE LOADINGS FOR MAXIMUM RANGE OF STRESS INTENSITY AND FATIGUE EVALUATIONS PPESSURE, PSIG Time, sec/

Thermal CONDITION CYCLES PRIF%RY SECONDARY Conditions a,c,t; l

I O

h 3 61

TABLE 3.5.7.2 1 (contd)

PRESSURE AND TEMPERATURE LOADINGS FOR MAXIMUM RANGE '

OF STRESS INTENSITY AND FATIGUE EVALUATIONS PRESSURE, PSIG i

Time,sec/ Thermal CYCLES PRIMARY SECONDARY Conditions ,a,

. CONDITION r

l 1

l

• Umbrella transient l TR = transient, ST Steady state, UF Uniform temperature Note: Tharmal Conditions: 1 3-62 1

l l

The analysis considers both undented and dented tubes. In the analysis of dented tubes, only one tube is considered to be locked-up at the first tube i support plate at 100% power conditions. In addition to the undented tube conditions, the following effects on stress components of the dented tubes are analyzed:

- effect of thermal conditions in the tube and wrapper /shell regions

- effect of pressure drop across the tubesheet

~

- effect of pressure drop across the tube support plates

- effect of interaction among the tubesheet, tube support plates, shell/ wrapper, stayrods, and spacer pipes The effects of pressure drop across the tubesheet and the tube support plates, as well as the tubesheet tube support plate assembly interactions, are taken into account for central dented tubes, while they are neglected for the outermost tubes.

The results of maximum range of stress intensity and fatigue evaluations are summarized in Tables 3.5.7.2-2 through 3.5.7.2-5. The requirements of the ASME Code, Paragraph NB-3222.2 (References 1) are met.

l E

3 63

l Table 3.5.7.2 2 MAXIMUM RANGE OF STRESS INTENSITY AND FATIGUE 5 Full Length Tubesheet Laser Welded Sleeve

! Normal interfacial sleeve / tube weld width of ( Ja.c Upper LWJ: hydraulically expanded Lower LWJ: hydraulically expanded Calculated Allowable Ratio S.I. S.I. Calculated S.I.

KS! KS1 Allowable S.I.

Maximum Range of Stress Intensity

- ~ a,c.e Straight Sections: sleeve Upper LWJ: sleeve tube weld Lower LWJ: sleeve tube weld -

Cumulative Fatigue Usage Factor a,c

[ ,

l 3 64

r Table 3.5.7.2 3 MAXIMUM RANGE OF STRESS INTENSITY AND FAT!GUE Full Length Tubesheet Laser Welded Sleeve Normal interfacial sleeve / tube weld width of ( Ja,c Upper LWJ: hydraulically expanded Lower LWJ: hydraulically expanded and hard rolled in the tube Calculated Allowable Ratio

[

S.I. S.I. Calculated S.I .

KSI KS! Allowable S.I.

Maximum Range of Stress Intensity a,c,e Straight Sections: sleeve Upper LWJ: sleeve tube weld

(* Lower LWJ: sleeve

! tube weld l

Cumulative Fatigue Usage f actor

[ Ja,C f .

O

Table 3.5.7.2 4  ;

MAXIMUM RANGE OF STRESS INTENSITY AND FATIGUE Elevated Tubesheet Laser Welded Sleeve

. Normal interfacial sleeve / tube weld width of ( Ja,c Calculated Allowable Ratio S.I. S.I. Calculated S.I.

KSI KSI Allowable S.I.

Maximum Range of Stress Intensity

- 8 C ee Straight Sections: sleeve Upper LWJ: sleeve tube weld Lower LWJ: sleeve l' tube I weld

(

Cumulative Fatigue Usage Factor

[ ja,c l

o 3-66

Table 3.5.7.2-5

- MAXIMUM RANGE OF STRESS INTENSITY AND FATIGUE Support Plate Laser Welded Sleeve Normal interfacial sleeve / tube weld width of ( Ja,c 4

Calculated Allowable Ratio- ,

S.I. S.I. Calculated S. I . i KS! KSI Allowable S.I.

Maximum Range of Stress Intensity

- a c.e ,

Straight Sections: sleeve Upper and sleeve Lower LWJ tube ,

weld Cumulative Fatigue Usage Factor 1a,c

[

l .

I o

l l

l 3 67

3.

5.8 REFERENCES

1. ASME Boiler and Pressure Vessel Code, Section Ill, Subsection NB, 1986 Edition, July 1, 1986,

2. ASME Boiler and Pressure Vessel Code, Code Cases, Case N 20, 1986 Edition.

July 1, 1986.

3. ASME Boiler and Pressure Vessel Code, Section !!!, Appendices, 1986 Edition, July 1, 1986.

4. WECAN, WAPPP and FIGURES 11, F. J. Bogden iditor, Second Edition, May 1981, Westinghouse Advanced System Technology, Pittsburgh, PA 15235.

5. A. L. Thurman, J. M. Hall, "WECEVAL, A Computer Code to Perform ASME BPVC Evaluations Using Finite Element Model Generated Stress States", Revision 1 January 1989, Westinghouse, WCAP 9376 O

F 3-68

3.6 Special Considerations 3.6.1 Flow Slot Hourglassing f

Along the tube ~ lane, the tube :upport plate has several long rectangular flow slots that have the p6tential to deform into an " hourglass" shape with significant denting. The effect of flow slot hourglassing is to move the neighboring tubes laterally towards the tube lane from their initial positions. The maximum bending would occur on the innermost row of tubes in

', the center of the flow slots.

3.6.1.1 Effect on Burst Strength I l

The effect of bending stresses on the burst strength of tubing has been studied for both axial and circumferential crack orientations. The results of'the study show the effect of bending stresses on tube burst strength of tubes with axial cracks to be negligible.

3.6.1.2 Effect on Stress Corrosion Cracking (SCC) Margin j Two long term modular model boiler tests have been conducted to address the

- effect of bending stresses on SCC. Based on the results of this test program on mill-annealed tubing, the bending stress magnitude due to flow slot hourglassing is judged to have only a small effect, if any, on the SCC resistance margins. It is to be noted that thermally treated Alloy 600 and 690 have additional SCC resistance compared to mill annealed Alloy 600 tubing, j -

l

, 3.6.1.3 Effect on Fatigue Usage Factor

- In addition to the above twu considerations, the analysis also considers the effect of hourglassing induced bending stresses on fatigue usage of the sleeve.

The effect on the fatigue usage is negligible since the bending stress is a

. mean stress and the maximum mean stress has been accounted for.

1 3 69

\

3.6.2 Tube Vibration Analysis i

Analytical assessments have been performed to prtdict modal natural frequencies l and related dynamic bending stresses attributed to flow induced vibration for

sleeved tubes. The purpose of the assessments was to evaluate thf effect on the natural frequencies, amplitude of vibration, and bending stres, due to installation of sleeves of various lengths.

• Since the stresses resulting from flow induced vibration are significantly below the endurance stress of the tube material and since higher natural (

frequencies result from the use of a sleeve / tube versus an unsleeved tube, it is concluded that the sleeved tube will not experience any significant fatigue from flow induced vibration.

3.6.3 Sludge Height Thermal Effects i

In general, with at least 2.0 inches of sludge, the tubesheet is isothermal at the bulk temperature of the primary fluid. The net effect of the sludge is to reduce tube /tubesheet thermal stren effects.

3.6.4 Allowable Sleeve Degradation 3.6.4.1 Minimum Required Sleeve Thickness The minimum required sleeve wall thickness, tr, to sustain normal and accident condition loads is calculated in accordance with the guidelines of USNRC Regulatory Guide 1.121. In this evaluation, the surrounding tube is assumed to be completely degraded; that is, no design credit is taken for the residual strength of the tube.

Since Regulatory Guide 1.121 constitutes an operating criterion, it is permissible to derive the allowable stress limits based on expected lower bound material properties, as opposed to the Code minimum values. Expected strength properties are obtained from statistical analyses of tensile test data of

, actual production tubing. Lower bound statistical tolerance limits, LTL, for l yield and ultimate strength values are computed in accordance with the accepted industry practice such that there is a 95 percent probability that 95 percent of the sleeve / tubes will have strength greater than LTL values.

3-70

i j

l USNRC- Regulatory Guide 1.121 Criteria

" 1. Normal Operations j 4

l

- PPR

• 2250 psia PSEC = 720 psia AP = 1530 psi '

.t Criteria: Pm < Su

• 3

[ Ja,c Minimum required sleeve wall thickness, tri APx Rg tr "

Su 0.5 (PPR+ PSEC) 3 where, RJ = inner radius of the sleeve, (= 0.333 inch) tr

• I la,c

. 2. Upset Conditions PPR = 2600 psia l.

PSEC = 1035 psia

~

aP = 1565 psi Criteria: Pm < Sy

. [ ja,c Minimuin required sleeve wall thickness, tr AP x R}

tr

• Sy 0.5 (PPR + PSEC) tr " I l

. 3. Accident Condition Loadings a.LOCA + SSE The major contribution of LOCA and SSE loads is the bending stresses at the top

- tube support plate due to a combination of the support motion, inertial l,

loadings, and the pressure differential across the tube U bend resulting from the rarefaction wave during LOCA. Since the sleeve is located at the sixth tube support plate or below, the LOCA + SSE bending stresses in the sleeve are quite small. The governing event for the sleeve, therefore, is a postulated secondary side blowdown, either FLB or SLB.

3-71

b. FLB + $$E The maximum primary to secondary pressure differential occurs during a postulated feedline break (FLB) accident. Again, because of the sleeve

location,' the SSE bending stresses are small. Thus, the governing stresses for the minimum wall thickness requirement are the pressure membrane stresses.

PPR = 266; psia PSEC = 15 psia AP = 2650 psi Criteria: Pm < 1 esser of 0.7 Su or 2.4 Sm

(_ la c Pm <. 0.7 Su " (

Minimum required sleeve wall thickness tre APxR g t

r 0.7 S 0.5 (PPR + PSEC) tr " I la,c in summary, the minimum required sleeve wall thickness is (

ja,c remaining wall for nominal operating conditions.

e 3-72

Leak Before-Break Verifi 311QD

> The leak before break evaluation for the sleeve is based on leak rate and burst pressure test data obtained on 7/8 inch 00 x 0.050 inch wall and 11/16 inch OD x 0.040 inch wall cracked tubing with various amounts of uniform thinning simulated by machining on the tube OD. The margins to burst during a postulated SLB (Steamline Break Accident) condition are a function of the mean

- radius to thickness ratio, based on a maximum permissible leak rate of 0.35 spm

- due to a normal operating pressure differential of 1530 psi.

Using a mean radius to thickness factor of 9.5 for the nominal sleeve, an allowable leak rate of 0.35 gpm, a SLB pressure differential of 2560 psi, and the nominal leak and burst curves, a ( .]a c percent margin exists between the burst crack length and the leak crack length. For a sleeve thinned Ja.c percent through wall over a 1.5 inch axial length, a ( Ja,c

[

percent margin to burst is demonstrated. Thus, the leak before break behavior is confirmed for unthinned and thinned conditions.

3.6.4.2 Determination of Plugging Limits The minimum acceptable wall thickness and other recommended practices in Regulatory Guide 1.121 are used to determine a plugging limit for the sleeve.

This Regulatory Guide was written to provide guidance for the determination of a plugging limit for steam generator tubes undergoing localized tube wall thinning and can be conservatively applied to sleeves. Tubes with sleeves which are determined to have indications of degradation of the sleeve in excess of the plugging limit, would have to be repaired or removed from service.

- As recommended in paragraph C.2.b. of the Regulatory Guide, an additional thickness degradation allowance must be added to the minimum acceptable tube wall thickness to establish the operational tube thickness acceptable for continued service. Paragraph C.3.f. of the Regulatory Guide specifies that the basis used in setting the operational degradation allowance include the method and data used in predicting the continuing degradation and consideration of l

l eddy current measurement errors and other significant eddy current testing parameters.

As outlined in Section 6.0 of this report, the capability of eddy current inspection of the sleeve and tube in the sleeve area has been demonstrated.

3 73

l The ( JC d eddy current measurement uncertainty value of (

Ja c.e of the tube wall thickness is appropriate for use in the

> determination of the operational tube thickness acceptable for continued service and thus determination of the plugging limit.

Paragraph C.3.f of the Regulatory Guide specified that the basis used in j setting the operational degradation analysis include the method and data used in predicting the continuing degradation. To develop a value for continuing degradation, sleeve experience must be reviewed. Westinghouse designed sleeves have had up to 8 years of operation. No degradation has been detected to date on Westinghouse designed mechanical joint sleeves and no sleeved tube has been ,

removed from service due to degradation of any portion of the sleeve. This result can be attributed to the changes in the sleeve material relative to the tube and the lower heat flux due to the double wall in the sleeved region.

Sleeves installed with the laser weld joint are expected to experience the same performance.

> It is the position of Westinghouse Electric that since no degradation has been detected in the sleeves, presently any allowance for continuing degradation

~

other than zero would be an arbitrary value not supported by the data and would represent a conservatism in addition to the safety factors implicit in the determination of minimum acceptable tube wall thickness using Regulatory Guide i 1.121 recommendations. However, the recent practice of the NRC has been to require an arbitrary value of 10% of the sleeve wall as allowance for continued degradation. That allowance is used in this evaluation.

In summary, the operational tube thickness acceptable for continued service includes the minimum acceptable tube wall thickness (( la,c.e of wall thickness (see 3.6.4.1), the combined allowance for eddy current uncertainty, and operational degradation (( Ja,c plus 10% of wall thickness as required by f theNRC). These terms total to 63% resulting in a plugging limit as determined by Regulatory Guide 1.121 recommendations of 37% of the tube wall thickness.

l l

l The plugging limit for the tube, where applicable as defined below, is as 1 specified in the Technical Specifications for the non sleeved portions of the tube, currently 40% of the tube wall thickness.

3-74

l 3.6.4.3 APPLICATION Of" PLUGGING LIMITS Sleeves or tubes which have eddy current indications of degradation in excess of the plugging limits must be repaired or plugged. Thoss portions of the tube and the sleeve for which indications of wall degradation must be evaluated are summarized as follows:

1) Indications of degradation in the length of the sleeve between the weld joints must be evaluated against the sleeve plugging limit.

2) Indication of tube degradation of any type including a complete break in the tube between the upper weld joint and the lower weld joint does not require that the tube be removed from service.

3) At the weld joint, degradation must be evaluated in both the sleeve and tube.

4) In a joint with more than one weld, the weld closest to the end of the sleeve represents the joint to be inspected and the limit of the

, sleeve inspection.

5) The tube plugging limit continues to apply to the portion of the tube above the upper weld joint and below the lower weld joint.

3.6.5 Effect of Tubesheet/ Support Plate Interaction Since the pressure is normally higher on the primary side of the tubesheet than on the secondary side, the tubesheet becomes concave. Under these conditions.

the tubes protruding from the top of the tubesheet will rotate from vertical.

This rotation develops stresses in the sleeved tube assembly. Analysis results show these stresses do not significantly affect the fatigue usage factors already reported.

3.6.6 Structural Analysis of the Laser Welded Sleeves with Interfacial Weld Widths from ( Ja,c The results of structural analysis presented in Section 3.5 are for sleeves with interfacial weld widths of ( la c.e for the upper and lower i LWJ's. Sleeves with interfacial weld widths of ( Ja, cinch 3 75 l

l (full length tubesheet sleeves) and ( ja.c.e inch (elevated tubesheet sleeves) have also been analyzed and are reported in this subsection. The finite element model used in Section 3.5 was revised to accommodate the new weld widths. The case of hydraulically expanded upper LWJ and hydraulically I expanded and roll expanded into the tube icwer LWJ is considered. The change l in the weld widths does not affect the results of the Primary Stress Intensity l Evaluations presented in Section 3.5. Therefore, only Maximum Range of Stress

- Intensity and Fatigue Evaluations are performed. The pressure, axial force and

• some thermal stress runs are recomputed with the revised finite element models.

The thermal stress runs recomputed are those giving the largest contributions to maximum range of stress intensity. The results of maximum range of stress intensity and fatigue evaluations are presented in Tables 3.6.6-1, 3.6.6-2, and 3.6.6 3. It is demonstrated that the laser welded sleeves with interfacial Ja.c inch satisfy the sleeve / tube weld widths between (

requirements of the ASME Code, Section !!!. These results are also applicable for the tube support plate sleeves.

IN l-e 1

1 3-76

l Table 3.6.6 1 o MAXIMUM RANGE OF STRESS INTENSITY AND FATIGUE Full Length Tubesheet Sleeves

- (Interfacial Sleeve / Tube Weld Width of ( Ja.c)  ;

i

- Calculated Allowable Ratio

- S.I. S.I. Calculated S.I. l

• KS! KSI Allowable S.I.

- 1 I

Maximum Range of Stress Intensity

- a,c,e Straight Sections: sleeve Upper LWJ: sleeve tube ,

- weld

~

Lower LWJ: sleeve tube weld -

Cumulative Fatigue Usage Factor

[ ja.C l

4 3-77

l Table 3.6.6 2 MAXIMUM RANGE OF STRESS INTENSITY AND FATIGUE o Full Leng'/n Tubesheet Sleeves (Interfacial Slr' ave / Tube Weld Width of ( Ja.c)

Calculated \llowable Ratio S.I. S.I. Calculated S. I . j

- KSI KSI Allowable S.I. l

\

L

• Maximum Range of Stress Intensity ,

- a,c.e Straight Sections: sleeve j Upper LWJ: sleeve tube weld Lower LWJ: s M ye

~

weld -

Cumulative Fatigue Usage Factor

[ Ja,c

• ) Thermal bending stresses removed per NB-3228.5 (a) l r:

3-78

l 3.6.7 Minimum Sleeve Wall Thickness Nominal and minimum slaev .<all thickness cases are analyzed.

, Allowin( u.r 3:: . ...

0.003 Inch for corrosion / erosion, the recommended sleeve wei th' - es:

, . Nominal Sleeve Wall Thickness ( .)a,c

. \

- Minimum Local Sleeve Wall Thickness [ Ja,c

. \

3.6.8 Effect of Potential Gap Openings at the Welds The welding process can stimulate some localized shrinkage. As a result, the sleeve pulls away from the tube, opening a gap between the sleeve and the tube. l The shrinkage due to the weld occurs symmetrically about the weld, both above l and below the weld. This gap opening is simulated by decoupling the

'. appropriate nodes above and below the upper and lower welds. The length of the modeled gap openings is in the range of 0.030 inch to 0.060 inch.

F The pressure, axial force and thermal stress runs which give the largest contributions to maximum Range of Stress Intensity and Fatigue have been recomputed. The results of analyses show that the requirements of the ASME Code are met.

3.7 EVALVATION OF OPERATION WITH FLOW EFFECTS SUBSEQUENT TO SLEEVING 3.7.1 Safety Ana. lyses and Design Transients The emergency core cooling system (ECCS) performance analysis being performed for Farley Units 1 and 2 supports operation at up to 20 per cent equivalent steam generator tube plugging (SGTP) in the most plugged steam generator with an average plugging level of 15%. For the evaluation of acceptable number f of sleeves, a uniform plugging level of 20% is considered. This analysis and the corresponding non LOCA evaluation are considered applicable for the steam generator sleeving program with a combination of plugging and sleeving flow restriction equal to or less than the restriction due to the acceptable plugging level. In addition, in support of the steam generator sleeving 3-80

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

program, Westinghouse has done an evaluation of selected LOCA and non LOCA transients to verify that use of sleeves resulting in a plugging equivalency of a up to 20 per cent in the most plugged will not have an adverse affect on the thermal-hydraulic performance of the plant. For the accidents as evaluated, o the effect of'a combination of plugging ano sleeving up to the limits of the existing analysis would not result in any design or regulatory limit being exceeded.

- The items listed below were evaluated for a sleeving and plugging combination equivalent to the existing tube plugging limits and the results indicated no adverse effects.

Large Break LOCA Small Break LOCA LOCA Hydraulic Forcing Functions Post LOCA boron requirements Time to switch over the ECCS to hot leg recirculation The steam generator tube rupture (SGTR) accident is analyzed to ensure that the o

offsite doses remain below 10CFR100 limits. The primary parameters affecting the conclusion are the extent of fuel failure assumed for the accident, the amount of primary to secondary break flow through the ruptured tube, and the mass released to the atmosphere from the ruptured steam generator. The amount of fuel failure assumed for the Farley FSAR SGTR analysis is 1% which is assumed to be independent of the transient conditions. The primary to secondary break flow and the mass released to the atmosphere are primarily dependent upon the RCS and secondary thermal hydraulic parameters.

- An evaluation was performed for Farley Units 1 and 2 which demonstrated that the effect of up to 20% steam generator tube plugging on the SGTR analysis would be acceptable. The SGTR evaluation was based on a uniform plugging level of 20%. The evaluation bounds the effect of non-uniform plugging with the mos' plugged steam generator at or less than 20% plugging. Thus with the combined sleeving and plugging, up to the limit based on the LOCA evaluation, the operating RCS temperature and steam pressure will not be reduced below the values for the evaluated tube plugging level. On this basis, the evaluation i performed for the previausly evaluated tube plugging level limit is applicable for the combined tube plugging and sleeving, and it is concluded that the 3-81

sleeving will not change the previous conclusion that the SGTR analysis will remain acceptable.

^

The effect of sleeving on the non-LOCA transient analyses has been reviewed.

Since the effect of the reduced RCS flow rate at the tube plugging limit has been evaluated for the non LOCA safety analyses, these analyses bound the  ;

equivalent effect of steam generator tube sleeving. Therefore, the steam generator sleeve installation up to the equivalent of plugging limit would not invalidate any non LOCA safety analyses.

Evaluations of the level of sleeving and plugging discussed in this report have shown that the Reactor Coolant System flow rate will not be less than that for the analyzed plugging level. The effect of the reduction in RCS flow rate for the analyzed plugging level on the design transients has been evaluated and has no impact. Any combination of plugs and sleeves which does not i'sult in an RCS flow rate less than that for the analyzed plugging level would not have an adverse effect on the previous evaluation of the design transients.

Any smaller number of sleeves would have less of an effect.

? 3.7.2 Equivalent Flugging Level For the Series 51 steam generators in Farley Units 1 and 2, 20 percent of the total tubes (3388 tubes per S/G) vquals 678 tubes in a steam generator.

Inserting a sleeve into a steam generator tube results in a reduction of primary coolant flow. For the purposes of this section, it is assumed that any l number of tubes per steam generator may be sleeved, up to the equivalent I

. plugging limit. The evaluation of flow effects for sleeving at Farley Units 1 and 2 assumes the use of ( Ja,c.e inch long laser sleeves which are expected to be long enough to span the degraded areas in the tubesheet region and to place the upper joint above the sludge pile in either the hot or cold leg side of the steam generators. The flow effects of this sleeve length bound a range of tubesheet sleeve lengths (( Ja,c.e inches) which could be used in the 5

sleeving of the Farley Units 1 and 2 steam generators. The sleeves at each of the tube support plate locations are ( ,)a,c.e, P

The flow reduction through a tube due to the installation of a sleeve can be considered equivalent to a portion of the flow loss due to a plugged tube, ine hydraulic equivalency ratio of the number of sleeved tubes required to result 3 82

i in the same flow lost as that due to a plugged tube can be used to determine I the allowable number of plugs and sleeves in combination. The hydraulic  !

a equivaleccy ratio determined at nominal conditions is independent of the fuel in the reactor. The hydraulic equivalency ratio for LOCA fluid conditions was

- established using flow rates at the peak clad temperature based on the analysis with Westinghouse supplied fuel. In all cases the hydraulic equivalency ratio l for normal operation is more limiting than for postulated LOCA conditions

• The flow reduction through a tube due to the installation of a sleeve and tia

insulating effect of the double wall at the sleeve will reduce the heat transfer capacity of the tube. An evaluation of the loss of heat transfer at  ;

normal operating conditions found that the per:entage loss of heat transfer due to sleeving is less than that the percentage loss for fluid flow. In other words, the heat transfer equivalency number is larger than the hydraulic equivalency number. Thus, the hydraulic equivalency number is limiting.

The specific LOCA conditions used to evaluate the effect of sleeving on

- the ECCS analysis occur during a portion of the postulated accident when the analysis predicts that the fluid in the secondary side of the steam generator

~

is warmer than the primary side fluid. The reduction in the heat transfer capacity due to installation of sleeves would have a beneficial reduction on a

the heat transferred to the fluid on the primary side of the tubes. Thus the hydraulic equivalency number for normal operation is also limiting compared to LOCA heat transfer effects.

Many combinations of tubesheet and tube support plate sleeves have been evaluated for flow reduction and hydraulic equivalency to plugged tubes. Since l

the temperature of the reactor coolant changes with elevation in the tube, the

• effect of a sleeve on the flow is different at each elevation. Additionally, the effect of a sleeva '. the flow rate is also dependent on any other sleeves present. Table .<-l presents a summary of hydraulic equivalency numbers for the limiting combinations of tubesheet sleeves and the specified numbers of

'ube support plate sleeves. For example, the hydraulic equivalency number in Table 3.71 for no tubesheet sleeve and four tube support plate sleeves (9.1) is the value for the four tube support plate elevatins which result in the n

' smallest hydr' 'ic equivalency number. Table 3.7-2 presents the information used to devebp the summary Table 3.7-1. The values in Table 3.7-2 may be used if the equivalent plugging level is near the plugging level limit and more accurate, less conservative values are required to support increased sleeving.

3 83

WESTINGHOUSE PROPRIETARY CLASS 2 Additionally, if actual installed conditions are not covered in Ttbie 3.7-2, refined analyses can be conducted using the same methodology as in Table 3.7-2.

To establish the equivalent plugging level, the number of tubes with a given l combination of sleeves is divided by the hydraulic equivalency number. This gives the equivalent number of plugged tubes for that combination: The total 5 equivalent number of plugs for all sleeve combinations combined with the number of plugged tubes is then the equivalent plugging level.

The method and values of hydraulic equivalency and flow loss per sits tube

,' outlined above can be used to represent the equivalent number of sleeves by the

. following formula:

p- ,[ ) b,d,e e

where P = Equivalent number of plugged tubes.

e P, - Number of tubes actually plugged.

Sg - Number of active tubes with a sleeve combination.

N = Hydraulic equivalency number for the sleeve combination,

, hyd,i g

P - Equivalent number of plugged tubes due to other sleeve designs.

?

3.7.3 Fluid Velocity As a result of tube plugging ent heving, primary side fluid velocities in the steam generator tubes will ir./ ease. The effect of this velocity increase on the sleeve and tube has beer ev luate/ ming a conservative limiting condition in which 20 percen' re are plugged. As a reference, normal flow velocity through a tub. ,- . au hir .y [

. ]cft/sec, for the unplugged condition. With 20 percent f tae> ,aupged, the fluid velocity through an non-plugged and non-sleeved tuus n ( ]b,c Ft/sec, and for a tube with a

'. single tube support plate sleeve, the local fluid velocity in the sleeve region is estimated at [ ]b,c,e ft/sec. Because thes i fluid velocities are less than the inception velocities for fis!d impacting, cavitation, and erosion-corrosion, the potential for tube degradation due to these mechanisms is low.

l 3.7.4 Flow Effects Summary E

l The effects of sleeving on LOCA and non-LOCA transient analyses have been 4

3-84

reviewed. No adverse result is indicated for sleeve and plug combinations up to an equivalent of the analyzed steam generator level of 15 per cent average

• and 20 per cent in the most plugged steam generator. The ECCS performance analysis and the corresponding non-LOCA evaluations are considered applicable

for the steam generator sleeving program with a combination of plugging and sleeving flow restriction equal to or less than the acalyzed tube plugging level. Steam generator sleeve ipt!,ellation up to the equivalent of the

- analyzed plugging level would 96; inr>iidate any non-LOCA safety analyses or

' the evaluation of design transtuott,,

The results of evaluations show that any combination of sleeving and plugging may be utilized at Farley Units 1 and 2 as long as the effective analyzed plugging level, using the hydraulic equivalency number for normal operation'. is not exceeded.

Accordingly, using the assumptions stated in this Section, sleeve installation at J. M. Farley Units 1 and 2 Nuclear Power Plant with up to the limit of the

- equivalent plugging level using laser welded sleeves in the tubesheet and at the tube support plates will not have an adverse effect on the normal operation, design transients, and postulated accident conditions.

a e

e 3-85

Table 3,7-1 Summary Hydraulic Equivalency Numbers

( Normal Operating Conditions b

I Total Number of TSP Sleeves Nhyd-a,c.e l.

e i

s -

- -I Welded Sleeve Lengths: Tubesheet[ la,c.e Inches Tube Support Plate [ la,c,e Inches l 3 86 1

l' 1

Table 3.7 2

(

Detail Hydraulic Equivalency Numbers a , c.e

(

a t

e I

I r

h e

d i

I N

t I

l' .

t I

l l

3 __.

4.0 PROCESS DESCRIPTION

'. The following description of the sleeving process pertains to current processes used. Westinghouse continues to enhance it's tooling and processes through

development programs. As enhanced techniques are developed and verified they will be utilized. Use of enhanced techniques which do not materially affect the technical justification presented in Section 3.0 is considered to be within

~

the bounds of this Steam Generator Sleeving Report for Laser Welded Sleeves.

The sleeves are fabricated under controlled conditions, serialized, cleaned, and inspected. They are typically placed in plastic bags, and packaged in protective styrofoam trays inside wood boxes. Upon receipt at the site, the boxed sleeves are stored in a controlled area outside containment and as required moved to a low radiation, controlled region inside containment. Here the sealed sleeve box is opened and the sleeve removed, inspected and placed in a protective sleeve carrying case for transport to the steam generator platform. Note that the sleeve packaging specification is extremely stringent and, if unopened, the sleeve package is suitable for long term storage.

The sleeve installation consists of a series of steps starting with tube end preparation (if necessary) and progressing through tube cleaning, sleeve

• insertion, hydraulic expansion at both the lower and upper joint, hard rolling the lower joint locations (if applicable), welding, UT inspection, post weld stress relief and eddy current inspection. The sleeving sequence and process are outlined in Table 4.0-1. These steps are described in the following sections. .

( -

. 4.1 Tube Preparation

- There are two steps involved in preparing the steam generator tubes for the sleeving operation. These consist of rolling at the tube mouth and tube cleaning. Tube end rolling is performed only if necessary to insert a sleeve.

(

4.1.1 Tube End Roll M1 ( Contingency )

A light mechanical rolling process will be performed if the tube mouth is deformed such that it would restrict the entry of a sleeve. This process has l

j been used in previous sleeving programs and has been demonstrated to be acceptable.

4-1

TABLE 4.0 1 l SLEEVE PROCESS SEQUENCE

SUMMARY

l TUBE PREPARATION 1) Light Mechanical ~ Roll Tube Ends .l

( If necessary')

2) Clean Tube Inside Surface SLEEVE INSERTION 3) Insert Sleeve / Expansion Mandrel Assembly

4) Hydraulically Expand Sleeve Top  !

and Bottom Joints LOWER JOINT FORMATION 5) Roll Expand Lower Sleeve End

( If necessary )

/, WELD OPERATION 6) Weld Upper Joint

7) Weld Lower Joint INSPECTION 8) Ultrasonically Inspect STRESS RELIEF 9)- Post Weld Stress Relief INSPECTION 10) Baseline Eddy Current-4 9

5 4-2

4.1.2 Tube Cleaning 3

• The sleeving process includes cleaning the inside diameter _ area of tubes to be sleeved to prepare the tube surface for the upper and lower joint formation by removing frangible oxides and foreign material. Evaluation has demonstrated that neither of these processes remove any significant fraction of the tube Cleaning also reduces the radiation shine from the tube

, wall base material.

inside diameter, thus contributing to reducing man-rem exposure.

• The interior surface of each candidate tube will be cleaned by a tungsten carbide brush with a flushing water orifice. The hone brush is mounted on a flexible drive shaf t that is driven by an pneumatic motor and carries reactor The hone brush is driven to grade deionized flushing w?.ter to the hone brush.

a predetermined height in the tube that is greater than the sleeve length in order to adequately clean the joint area. The water jet from the brush is above the hone; it directs the water downward on the tube surface, thus rinsing away residue from the surface. This water flush also lubricates and cools the hone during operation. The Wet Hone End Effector mounts to a tool delivery robot and consists of a guide tube sight glass and a flexible seal designed to A flexible conduit surround the tube end and contain the spent flushing water.

is attached to the guide tube and connects to the tube cleaning unit on the The conduit acts as a closed loop system which steam generator platform.

serves to guide the drive shaft / hone brush assembly through the guide tube to the candidate tube and also to carry the spent flushing water to an air driven diaphragm pump which routes the water to the radioactive waste drain.

4.2 Sleeve Insertion and Expansion When the candidate tubes have been wet honed, the wet hone end effector will b removed from the tool delivery robot and the Select and Locate End Effector The SALEE consists of two pneumatic camlocks, dual (SALEE) will be installed.

pneumatic gripper assemblies, a pneumatic translation cylinder, a motorize drive assembly, and a sleeve delivery conduit.

f The tool delivery robot draws the SALEE through the manway into the channel head.

It then positions the SALEE to receive a sleeve, tilting the tool such that the bottom of the tool points toward the manway and the sleeve delivery At this point, the platform worker pushes a conduit provides linear access.

sleeve / mandrel assembly through 4-3 the conduit until it is able to be grippe l

the translating upper gripper. The tool delivery robot then moves the SALEE to the candidate tube. Camlocks are then inserted into nearby tubes and pressurized to secure the SALEE to the tubesheet.

I Insertion of the sleeve / mandrel assembly into the candidate tube is accomplished by a combination of the translating gripper assembly and the motorized drive assembly which pushes the sleeve to the desired axial elevation. For support plate sleeves, the support plate is found by using an i

eddy current coll. The sleeve is positioned by using the grippers and translating cylinder to pull the sleeve into position to bridge the support plate. For tubesheet sleeves, the sleeve is positioned by use of a positive stop on the delivery system.

The bladder style At this point, the sleeve is hydraulically expanded.

hydraulic expansion mandrel is connected to the high pressure fluid source, the Lightweight Expansion Unit (LEU), via high pressure- flexible stainless tubing.

The Lightweight Expansion Unit is controlled by the Sleeve / Tube Expansion l

C'ontroller (S/TEC), a microprocessor controlled expansion box which is an The S/TEC expansion system previously proven in various sleeving programs.

activates, monitors, and terminates the tube expansion process when proper expansion has been achieved.

The one step process hydraulically expands both the lower and upper expansion zones simultaneously. The computer controlled expansion system automatically applies the proper controlled pressure depending upon the respective yield The contact strengths and diametrical clearance between the tube and sleeve.

' forces between the sleeve and tube due to the initial hydraulic expansion are sufficient to keep the sleeve from moving during subsequent operations. At the end of the cycle, the control computer provides an indication to the operator In addition, a

• that the etxpansion cycle has been properly completed.

l pressure-time trace of each expansion is recorded and is retained for review.

When the expansion is complete, the mandrel is removed from the expanded sleev The SALEE is then repositioned to by reversing the above insertion sequence.

receive another sleeve / mandrel assembly.

4-4

l l

4.3 Lower Joint Hard Roll ( Full Length.Tubesheet Sleeves )

At the primary face of the tubesheet, the sleeve is joined to the tube by a ,

! raechanical hard roll (following the hydraulic expansion) performed with a roll expander which extends approximately 2 inches into the tube. The control of the mechanical expansion is maintained through a torque setting. The tool automati: ally shuts off when it reaches a preset torque value. The roll expander torque is calibrated on a torque calibrator prior to initial hard

[ ,

rolling operations and periodically verified during and at the completion of tool operation. This control and calibration process is a technique used throughout industry in the installation of tubes in heat exchangers.

4.4 General Description of Laser Weld Operation Welding of the upper and lower joint will be accomplished by a Westinghouse developed laser beam transmission system and rotating weld head. This system employs a Nd:YAG laser energy source located in a trailer outside of

~

containment. The energy of the laser is delivered to the steam generator platform junction box through a fiberoptic cable. The fiberoptic contains an intrinsic safety wire which protects personnel in the case of damage to the fiber. The weld head is connected to the platform junction box by a prealigned fiberoptic coupler. Each weld head contains the necessary optics, fiber termination and tracking device to correctly focus the laser beam on the interior of the sleeve.

1 The weld head / fiberoptic assembly is precisely positioned within the hydraulic l '

expansion region using the SALEE (described earlier) and an eddy current coil located on the weld head. At the initiation of welding operations, the shielding gas and laser beam are delivered to the welding head. During the welding process the head is rotated around the inside of the tube to produce f* the weld. A motor, gear train, and encoder provide the controlled rotary i motion to deliver a 360 degree weld around the sleeve circumference.

! The welding parameters, qualified to the rules of the ASME code, are computer controlled at the weld operators station. The essential variables per code case N-395 will be suitably monitored and documented for field weld acceptance.

4-5

---

___- --m

4.5 Post Weld Heat Treatment The tooling required to perform the stress relief process consists of four basic items:

I 7

a. A fibtroptic probe ]

l

b. A heater (production) probe

c. A pop-up end effector

d. A production end effector The fiberoptic probe is used in conjunction with the pop-up end effector. The end effector places a probe within the proper zone to perform the stress relief operation. To verify the temperature achieved, a fiberoptic filament is used to obtain the inside sleeve temperature by means of an optical pyrometer. After determining the proper power profile, the production probes are usud to stress relieve the balance of the sleeve / tube interfaces. This is done by using the-ROSA robotic arm and the SALEE to sequentially place production probes at the proper welded sleeve / tube interfaces, followed by application of the stress rel',ef process.

1 i

i l

i 4-6

i 1

4.6 Inspection Plan

( in order to verify the final sleeve installation, an eddy current inspection will be performed on all sleeved tubes to verify that all sleeves received the required hydraulic and roll expansions and to establish a baseline for future eddy current examination of the sleeved tubes. The basic process check on 100 percent of the sleeved tubes will be:

. 1. Verify presence of lower hydraulic expansion zone.

2. Verify lower roll joint location within the lower hydraulic expansion.

3. Verify presence of upper hydraulic expansion zone.

4. Verify the weld location and quality of the upper and lower welds.

5. Check for the presence of any anomalies.

In addition to eddy current process verification and baseline testing, the weld joint will be assessed through the use of an ultrasonic inspection technique.

The ultrasonic technique provides for overlapping circumferential scanning of the weld zone from the inside diameter of the sleeve and provides verification of the weld interfacial width and weld quality. The details of the ultrasonic inspection process are covered in Section 5.0. The ultrasonic inspection is performed only upon initial sleeve installation. It is not necessary to ultrasonically inspect the weld joints during future inservice inspections. In service inspection requirements are discussed in Section 6.0.

If it is necessary to remove a sleeved tube from service as judged by an evaluation of a specific sleeve / tube configuration, tooling and processes will be available to plug the tube.

1 i

4-7 ,

^ - - - ' _ . _ . _ . _

i l l

! 5.0 NDE Inspectability

(

The non-destructive examination (NDE) davelopment ef fort has concentrated on two aspects of the sleeving process. First, a method is required to confirm

$ that the laser welds meet critical process dimensions and acceptable weld T quality. Secondly, it must be shown that the sleeve / tube assembly is capable of being evaluated through subsequent routine in-service inspection. The ultrasonic inspection techniques are used during sleeve installation to confirm

,' weld acceptance and eddy current inspection technology is used to establish

. baseline inspection and perform subsequent in-service inspections.

5.1 Ultrasonic Weld Inspection 5.1.1 General Process Overview The ultrasonic inspection process is based on further refinements of past well-known and field-proven techniques used on brazed and CO2 laser welded

~

sleeves installed by Westinghouse.

I The inspection process developed for application to the laser welds incorporates the basic idea of transmission of ultrasound to the interface region (i.e., the sleeve OD/ tube ID boundary) and analyzing the amount of reflected energy from that region. An acceptable weld joint should present no acoustic reflections above a calibrated limit at the weld interface. It must also produce reflection from the tube OD that is above a calibrated limit.

Appropriate transducer, instrumentation and delivery systems have been designed and techniques established to demonstrate detectability and resolution of relevant defects at the interface. The main goal is to verify the width of a sleeve / tube fusion zone present in the joint. The entire weld interface (100%

of the axial and circumferential extent) will be examined with adequate overlaps to provide full coverage. The final acceptance criteria of the welds

(

are confirmed by comparison to ASME flat bottom hole defect standards and actual weld workmanship specinens. These are derived from the

• structural / functional requirements of the weld and satisfy the intent of the inspection.

The examination equipment and procedures meet all applicable codes, standards and regulations.

5-1

5.1.2 Principle of Operation and Data Processing The ultrasonic inspection of a laser weld is schematically outlined in Figure 5.1.2-1. An ultrasonic wave is launched by the application of a pulse to a piezoelectric transducer. The wave propagates in the couplant medium Ultrasonic energy is both transmitted and (water) until it strikes the sleeve.

reflected at the boundary. The reflected wave returns to the transducer where it is converted back to an electrical signal, which is amplified and displayed

- on a UT instrument oscilloscope.

The transmitted wave propagates in the sleeve until it reaches the outside  !

surface of the sleeve. If weld material is present, the wave continues to propagate through the weld joint into the tube. This wave then reaches the outer wall (back wall) of the tube and is reflected to the transducer. The resulting UT instrument display from a sound weld joint is a large signal from the sleeve-couplant interface, followed by a back wall " echo" spaced by the time of travel in the sleeve weld-tube assembly (T1 ,2,3). If no weld material is present, another pattern is observed with the large signal j

from the sleeve ID followed by a reflection from the sleeve 0D (T1 ,2). The spacing of these echoes depends upon the time of travel in the sleeve alone.

If there are some void regions in the weld, a complex' combination of these two signal patterns will result. Thus, by observing the patterns in the reflected pulse, a quality can be assigned to the weld joint.

The condition of the surface at the entry point of the sound energy as well as subsequent grain structure of the weld fusion zone determines the level of energy that reaches the back wall of a " fused" sleeve / tube section. To provide 1 the required resolution and ability to maximize energy input to the interface appropriately focused transducers have been chosen. The response of the focused transducer in the weld region is shown in Figure 5.1.2-2. The resultant wave form for a good weld is characterized by the sleeve couplant interface echo (at position 4 on the time basis) followed by a single echo from the tube OD (position 7). On the other hand, the UT trace shown in Figure The no-weld condition produces a decaying l

5.1.2-3 illustrates a lack of bond.

i pattern ( beginning at position 6 on the time basis) of sleeve OD (interface) echoes, repeating in time. The UT trace of Figure 5.1.2-4 illustrates the condition observed when there is a lack of bond that does not provide sufficient surface for multiple echoes.

5-2

l l/

/

/ INPUT SIGNAL ' I

/[ -

1 /

=3 / =a =1 f' RETL'RN SIGNAL

/

/ ,

l /

A SLEEVE

/' JOINT TUBt. j i

10E ALIZED VAVEFORMS I_ 't.2

_i ND JOINT l \ l

' L 2.3 ' =

. ' GOOD' -

JOINT Figure 5.1.2-1 Ultrasonic Inspection of Welded Sleeve Joint 5-3

a,b.e 4

wi Figure 5.1.2-2 UT Signal in Good Weld Region t

5-4

l i

l

(

f a,b,e f

i Figure 5.1.2-3 1

UT Signal in Lack of Weld Region 5-5

?

a,bse i

)

p 1

Figure 5.1.2.4 UT Signal in Small 1.ack of Bond Region 5-6

An automated system of processing the UT data from time gates (windows) is used. The gate position, threshold, and width are optimized to capture an echo from " weld lack of bond" signal and translate it via some electronic processing into an oscilloscope go/no go based trace. The traces (C-scans) map the positions of good and bad signal conditions relative to the tube geometry.

Position is provideo by encoders in the UT delivery device. Several repeated tests on many scanned weld samples have demonstrated the correspondence of C-scans to cross-sections of welded samp'.es.

5.1.3 Equipment and Tooling The probe system is delivered by the Westinghouse ROSA zero entry system. The various subsystems include the water couplant, UT, motor drives, electrical sy-tems and data display / storage.

The probe motion and feedback is done via rotary and axial drive modules which allow a range of speeds and axial advance per 3600 scan of the transducer head.

The axial advance allows for sufficient overlap to provide a high degree of overlapping coverage without sacrificing resolution or sensitivity.

The controls and displays are designed for trailer mounting outside containment. The system also provides for easy periodic calibration of the UT subsystem on the steam generator platform.

The permanent record of the inspection is either a polaroid picture of the oscilloscope C-scan with the continuous video record of the split image UT A-scan and concurrent oscilloscope C-scan. An example of a laser weld C-scan

'. is shown in Figure 5.1.3-1.

The UT instrument is used with the gate modules synchronized to the front wall (sleeve 1.0.) signal .

The transducer itself is a custom made 20 MHz narrow band, 0.125" diameter focussed transducer.

i 5-7

a,b,e

( _

l 1

1 l

l l

l l

1 l

l

(

1 1

' Figure 5.1.3-1 C-Scan from UT Examination of a Typical Laser Weld 5-8

5.1.4 Laser Wald Test Sample Results ,

.1 7he calibration standards consist of (a) Equipment setup standard -solid Alloy 690 thick. walled tube (wall thickness 0.100"). l l

(b) A sensitivity / resolution check " workmanship" standard, a typical laser welded sleeve / tube assembly.

The UT techniques were developed to assure that the flat bottom holes and notches of the setup standard (described in Figure 5.1.4-1) were detectable and measurable. The polaroid picture of Figure 5.1.4-2 shows the C-scan of the setup standard. Hence notches of 0.040, 0.030, 0.020,_0.010 inches wide, and flat bottom holes of 0.040, 0.030, 0.020, and 0.015 ' aches diameter can be detected. The notch lengths can be sized.

The " workmanship" standard was prepared using the typical weld process. The sample was inspected before further processing was done. A set of two notches

( was introduced in the outside diameter across the weld. These notches extended across the width of the weld. The notches simulate 'a " breach" or leak path across the weld. The notch widths are 0.020 and 0.030 inches wide, their depths are an average 0.050 inches and their lengths are 0.200 inches.

A Polaroid picture of the C-scan of this notched workmanship standard is shown in Figure 5.1.4-3. The equipment is set up using the thick-walled tube .

f l - standard to allow the operator ease in identifying and setting the UT instrument gates and gain. The setup standard presents uniform signals and is repeatable for every A-scan.

j 5.1.5 Summary The UT laser weld inspection system can confirm that there is a metallurgical bond between the sleeve and the tube. The system is used to determine any existence of leak path across the weld and a minimum acceptable weld width for 360 degrees around the circumference.

5-9

l-I 1

-1 i

a ,b,e ,

i 1

I 1

1 l

l 6

4 9

Figure 5.1.4-1  ;

UT Setup Standard 5-10 i

$ a,b.e I

1 1

i 1

( t i

l Figure 5.1.4-2 l

C-Scan from UT Examination of Equipment Setup standard 5-11

e a i

a ,b ,<

e Figure 5.1.4-3 C-Scan from UT Examination of Workmanship Sample of a laser Welded Sleeve 5-12

5.2 ED0Y CURRENT INSPECTION 5.2.1 General Process Overview Upon conclusion of the sleeve installation process, a final eddy current inspection is performed on every installed sleeve to provide interpretable baseline data on the sleeve and tube. This information is pathered by an eddy current process which utilizes a double cross wound coil. The double crosswound coil is designed to minimize the effects of geometry and weld zone changes that are 3600 in nature, i.e.: upper and lower hydraulic expansion.

transition areas, roll expansion transition areas, top of sleeve, the band of good weld material, etc.

5.2.2 Principle of Operation The eddy current inspection equipment, techniques, and restits presented herein apply to the proposed Westinghouse sleeving process. Eddy current inspections are routinely carried out on the steam generators in accordance with the Farley Nuclear Plant's Technical Specifications. The purpose of these inspections is to detect at an early state tube degradation that may have occurred during plant operation so that corrective action can be taken to minimize further degradation and reduce the potential for significant primary-to-secondary leakage.

The standard inspection procedure involves the use of a bobbin eddy current '

probe, with two circumferential1y wound coils which are displaced axially along ,

the probe body. The coils are connected in the so-called differential mode;

- that is, the system responds only when there is a difference in the properties of the material surrounding the two coils. The coils are excited by using an eddy current instrument that displays changes in the material surrounding the coils by measuring the electrical impedance of the coils. Presently, this involves simultaneous excitations of the coils with several different test frequencies.

The outputs of the various frequencies are combined and recorded. The combined data yield an output in which signals resulting from conditions that do not affect the integrity of the tube are reduced. By reducing unwanted signals, improved inspectability of the tubing results (i.e., a higher signal-to-noise 5-13

=r-- ::-- . .

ratio). Regions in the steam generato? such as the tube supports, the tubesheet laser weld area, and sleeve transition zones are examples of areas where multifrequency processing has proven valuable in providing improved inspectability.

After sleeve installation all sleeved tubes are subiected to an eddy current inspection which for establishment of a baseline to which all subsequent inspections will be compared.

There are a number of probe configurations that lend themselves to enhancing the inspection of the sleeve / tube assembly in the regions of laser weld as well as configuration transitions. The crosswound coil probe has been I

selected since it provides an advancement in the state of the art over the conventional bobbin coil probe, yet retains the simplicity of the inspection procedure.

The inspection for degradation of the sleeve / tube assembly has typically been performed using crosswound coil probes operated with multifrequency excitation. ,

For the weld free straight length regions of the sleeve / tube assembly, the inspection of the sleeve and tube is consistent with normal tubing inspections.

In sleeve / tube assembly joint regions, data evaluation becomes more complex.

The results discussed below suggest the limits on the volume of degradation that can be detected in the vicinity of the laser weld and geometry changes.

5.2.3 Transition Region Eddy Current Inspection The detection and quantification of degradation at the transition regions of

• the sleeve / tube assembly depend upon the signal-to-noise ratio between the degradation response and the transition response. As a general rule, lower frequencies tend to suppress the transition signal relative to the degradation signal at the expense of the ability to quantify the degradation. Similarly, the inspection of the tube through the sleeve requires the use of low frequencies to achieve detection with an associated loss in quantification.

Thus, the search for an optimum eddy current inspection represents a trade-off between detection and quantification. With the crosswound coil type inspection, this optimization leads to a primary inspection frequency for the sleeve on the order of 700 kHz and for the tube and transition regions on the order of 100 kHz.

5-14 l

l I

Figure 5.2.3-1 shows a typical 700 kHz calibration curve for the sleeve from which 00 sleeve penetrations can be assessed.

For the tube sleeve combination, the use of the crosswound probe, coupled with a multifrequency mixing technique for further reduction of the remaining noise signals significantly reduces the interference from all discontinuities (e.g. a diameter transition) which have 360 degree symmetry, providing improved visibility for discrete discontinuities. As is shown in the accompanying

. figures, in the laboratory this technique can detect OD tube wall penetrations with acceptable signal-to noise ratios at the transitions when the volume of metal removed is equivalent to the ASME calibration standard.

The response from the sleeve / tube assembly' transitions with the crosswound coil i is shown in Figures 5.2.3-2, 5.2.3-3 and 5.2.3 4 for the sleeve standards, tube standards and transitions, respectively. Detectability in transitions is enhanced by the combination of the various frequencies. For the crosswound probe, two frequency combinations are shown; the 50-150-kHz combination provides the overall detection capability while the 150 400 kHz combination  !

l

! provides improved sensitivity for the sleeve and some quantification capability l for the tube. Figure 5.2.3-5 shows the phase / depth curve for the tube using this combination. As examples of the detection capability at the transitions, I Figures 5.2.3-6 and 5.2.3-7 show the responses of a 20 percent 00 penetration in the sleeve and 40 percent 00 penetration in the tube, respectively.

For the inspection of the region at the top end of tk.9 sleeve, the transition response signal-to noise ratio is about a factor of four less sensitive than  :

that of the expansions. Some additional inspectability has been gained by tapering the wall thickness at the top end of the sleeve. This reduces the end-of-sleeve signal by a factor of approximately two. The crosswound coil, however, again significantly reduces the response of the sleeve.end. Figure 5.2.3-8 shows the response of various ASME tube calibration standards placed at the end of the sleeve using the cross-wound coil and the 50-150 kHz frequency combination. Note that under these conditions, degradation at the top end of the sleeve / tube assembly can be detected.

5-15

i Q.C.G h.-

I l

1 l

l i 1

l l

, [ la,c e calibration Curve l

l I l r

l Figure 5.2.3-1 1

! 5-16 l

, 1 l

a.c.a i

(

( l l

1 l

1

(

Eddy Current Signals from the ASTM Standard,-Machined on the Sleeve O.D. of the -Sleeve / Tube Assembly Vithout Expansion (Cross Vound Coll Probe).

Figure 5.2.3-2 c L

5-17

l a

l Eddy Current Signals From the ASTM Standard, Machined on the Tube O.D. of the Sleeve / Tube Assenbty Vithout Expansion (Cross Vound Coll Probe).

Figure 5.2.3-3 5-18

l l "

Eddy Current Signals from the Expansion Transition Region of the Tube / Sleeve Assembly (Cross Vound Cell Probe).

Figure 5.2.3 4 l

5-19

- Eddy Current Calibration Curve for ASME Tube Standard at t  : n c' and a Mix Using the Cross Vound Colt Probe Figure 5.2.3-5 5 20

A Eddy Current Signal fron a 20% Deep Hole, Half the Volume of ASTM Standard, Machined on the Sleeve D.D. In 1;he Expansion Transition Region of the Steeve / Tube Assembly (Cross Vound Cell Probe).

Figure 5.2.3 6 5 21

h Eddy Current Signal from a 40% ASTM Standard, Machined on the Tube O.D. In the Expansion Transition Region of Sleeve / Tube Assembly (Cress Vound Colt Probe).

Figure 5.2.3 7 5 22

nr.

i 4

Eddy Current Response of the ASME Tube Standard at the End of the Sleeve Using the Cross Vound Coll l Probe and Multifrequency Combination.

Figure 5.2.3-8 l

l 5 23 1

5.2.4 Laser Weld Region Eddy Current Inspection The cases considered in Section 5.2.2 cover the inspection of laser welded sleeves pressure boundary in four areas:

(i) The sleeve below the weld (ii) The sleeve between the hydraulic expansion transition and the weld (iii) The expansion transition (iv) The tube above the sleeve expansion (behind the free unexpanded sleeve) up to the top end of sleeve The only zone not covered is the zone where the laser weld exists.

The test sample used for this study was a prototypically made laser weld in an expanded sleeve zone of a sleeve / tube assembly.

The weld was inspected before and after the introduction of a 40% thru wall 3/16 inch diameter flat bottom hole placed on the outside surface of the tube at the centerline of the weld.

The weld presents an axisymmetric condition similar to the transition geometry which is demonstrated by the low phase angle signal similar to transition signals. The weld also displays a material disturbance by its distinct lobes which can be successfully mixed out.

Figure 5.2.4-1 shows the 50 kHz response from the weld zone and Figure 5.2.4 2 shows the successful 50 150 kHz mix response using cross wound coil.

The 50 150 kHz combination has proven to be optimum for detection in the weld zone, particularly at the tube I.D./ sleeve 0.0. interface. Figures 5.2.4-3 and Figure 5.2.4-4 show the response of the 40% FBH using 50 kHz and mix, respectively.

5 24

r a c.e 4

• Crosswound k[d'[ Current Basetti of L%5ER V'LW Figure 5.2.4 1 5 25

l

- e ,c .(

~

i Crosseownd Wx Eddy Current Response Baseline of LASER Veld Figure 5.2.4 2 1

1 5 26

a.c E

B a.C.e '

] Eddy Current Response Aftt,-

Crosswound[

Flat Bottomed Hole Vas Placed in D.D. of Tuki 9)

Center of Veld Figure 5.2.4 3 5 27 j , . . . . . . . _ . . . . , , , , , , , . . . . . _ _ _ _ . . . . . . . _ . _ . . . . . . _ _ . . . . . .

8,C.

I l

i Crosswound Mix Cdey Current Response Af ter 40X Flat Bottomed Hole Vas Placed in 0,D of Tulee at Center of Veld Figure 5.2.4-4 5 28

5.2.5 Summary Conventional eddy current techniques have been modified to incorporate the most recent technology in the inspection of the sleeve / tube assembly. The resultant inspection of the sleeve / tube assembly invcives the use of a cross wound coil

for the straight regions of the sleeve / tube assembly and for the transition regions. The advent of digital E/C instrumentation and its attendant increased

. dynamic range and the availability of eight channels for four frequencies has expanded the use of the crosswound coil for sleeve inspection. While there is

. a significant advancement in the inspection of portions of the assembly using the cross wound coil over conventional bobbin coils, efforts continue to advance the state of-the art in eddy current inspection techniques. As enhanced techniques are developed and verified, they will be utilized after' 10CFR50.59 review. For the present, the cross wound coil probe represents an inspection technique that provides additional sensitivity and support for eddy current techniques as a viable means of assessing the sleeve / tube assembly.

~

e 5-29

6.0 INSERVICE INSPECTION PLAN FOR SLEEVED TUBES ,

The need exists to perform periodic inspections of the supplemented pressure boundary. The inservice inspection program will consist of the followinc. The

.- sleeve will be eddy current inspected upon completion of installation to obtain a baseline signature to which all subsequent inspections will be compared.

Periodic inspections to monitor sleeve and tube wall conditions will be performed in accordance with the inspection section of the plant Technical Specifications.

The inspection of sleeves will necessitate the use of an eddy current probe that can pass through the sleeve 10. For the tube span between sleeves, this will result in a smaller fill factor than is optimal. The possibility for tube degradation in free span lengths is extremely small, as plant data have shown that this area is less susceptible than other locations, in the event that an indication is detected, any tube indication in this region that has not been previously identified on the prior to sleeving tube baseline will require further inspection by alternate techniques prior to acceptance of that indication.

e 61