Pressurized-Water Reactor Internals Aging …Abstract Ibis report is a summary of the results of a Phase 1 study on the effects of aging degradations in pressurized-water reactor (PWR)

NUREG/CR-6048ORNU-TM-12371

Pressurized-Water ReactorInternals AgingDegradation Study

Phase I

Prepared byK. H. Luk

Oak Ridge National Laboratory

Prepared forU.S. Nuclear Regulatory Commission

AVAIL.ABIUTY NOTICE

Availability of Reference Materials Cited hI NRC Publications

Most documents cited hI NRC publications will be available from one of the foflowing sowes:

1. The NRC Public Document Room, 2120 L Street, NW, Lower Level, Washington. DC 20555-0001

2. The Superintendent of Documents, U.S. Government Printing Office, Mal Stop SSOP, Washington,DC 20402-9328

3. The National Technical Information Service, Springfield, VA 22161

Although the lsting that follows represents the majority of documents cited hI NRC publications, It Is notIntended to be exhaustive.

Referenced documents available for inspection and copying for a fee from the NRC Pubic Document Roominclude NRC correspondence and Internal NRC memoranda; NRC Office of Inspection and Enforcementbulletins. circulars, information notices, Inspection and nvestigation notices; Ucensee Event Reports; ven-dor reports and correspondence; Commission papers; and applicant and licensee documents and corre-spondence.

The following documents In the NUREG erles are available for purchase from the GPO Sales Program:formal NRC staff and contractor reports, NRC-sponsored conference proceedings, and NRC booldets andbrochures. Also available are Regulatory Guides, NRC regulations hI the Code of Federal Regulations, andNuclear Regulatory Commission Issuances.

Documents available from the National Technical Information Service Include NUREG series reports andtechnical reports prepared by other federal agencies and reports prepared by the Atomic Energy Commls-sion, forerunner agency to the Nuclear Regulatory Commission.

Documents avalable from public and special technical lbrarles Include all open literature Items, such asbooks. journal and periodical articles, and transactions. Federal Register notices, federal and state legisla-tion, and congressional reports can usually be obtained from these libraries.

Documents such as theses, dissertations, foreign reports and translations, and non-NRC conference pro-ceedings are avalable for purchase from the organization sponsoring the publication cited.

Single copies of NRC draft reports are available free, to the extent of supply, upon written request to theOffice of Information Resources Management. Distribution Section, U.S. Nuclear Regulatory Commission,Washington, DC 20555-0001.

Copies of Industry codes and standards used hI a substantive manner In the NRC regulatory process aremaintained at the NRC IUbrary. 7920 Norfolk Avenue. Bethesda, Maryland, and are available there for refer-ence use by the public. Codes and standards are usually copyrighted and may be purchased from theoriginating organization or, If they are American National Standards, from the American National StandardsInstitute, 1430 Broadway, New York. NY 10015.

DISCLAIMER NOTICE

This report was prepared as an account of work sponsored by an agency of the United States GovernmenLNeither the United States Government nor any agency thereof, or any of their employees, makes any warranty,expresed or Implied, or assumes any legal liability of responsibility for any third party's use, or te results ofsuch use, of any nformation, apparatus, product or process disclosed in this report, or represents that its useby such third party would not infringe privately owned rights.

NUREG/CR-6048ORNL-TM-12371RV

Pressurized-Water ReactorInternals AgingDegradation Study

Phase I

Manuscript Completed: August 1993Date Published: September 1993

Prepared byK. H. Luk

Oak Ridge National LaboratoryOperated byMartin Marietta Energy Systems, Inc.

Oak Ridge National LaboratoryOak Ridge, TN 37831-6285

Prepared forDivision of EngineeringOffice of Nuclear Regulatory ResearchU.S. Nuclear Regulatory CommissionWashington, DC 20555-0001NRC FIN B0828Under Contract No. DE-ACO5-84OR21400

Abstract

Ibis report is a summary of the results of a Phase 1 studyon the effects of aging degradations in pressurized-waterreactor (PWR) internal components. Westinghouse (WE),Combustion Engineering (CE), and Babcock & Wilcox(B&W) reactors are included in the study.

Stressors associated with the operating environment insidethe reactor pressure vessel provide conditions that arefavorable to the development of aging-related degradationmechanisms. The dominant stressor is flow-induced oscil-latory hydrodynamic forces generated by the reactor pri-mary coolant flow. Results of a survey of the componentfailure information identified three major aging-relateddegradation mechanisms: fatigue, stress corrosion crack-ing, and mechanical wear.

between stressors and aging degradation mechanisms.Flow-induced vibration problems are resolved by conven-tional engineering practices: by the elimination of excita-tion sources or by de-tuning the structure from input exci-tations. Uncertainties remaining in the assessment of agingeffects on PWR internals include long-term neutron irradi-ation effects and the influence of environmental factors onhigh-cycle fatigue failures.

An effective plant in-service inspection (ISI) program willensure the structural integrity of reactor internals. Reactorinternals can be replaced if it is deemed necessary. There-fore, an inspection method with early failure detectioncapability will further enhance the safety as well as theefficiency of plant operations.

Strategies for controlling and managing aging degradationsare formulated based on the understanding of the linkage

iii NUREG/CR-6048

Contents

Page

Abstract ............................................................................... iii

List of Figures ........ ........................ ................................................ vii

List of Tables ........ ......................... ............................................... ix

Acknowledgments ................................................................................. xi

Summary ......... ........................... .................................................... xiii

1 Introduction ................................................................................ 1

Reference ........ ........................................................................ 2

2 PWR Internal Components ................................................................................ 3

2.1 Westinghouse (WE) Internals ............................................................................... 3

2.2 Babcox & Wilcox (B&W) Internals ............................................................................... 11

2.3 Combustion Engineering (CE) Internals ....................... ........................................................ 16

Reference ............................................................................... 22

3 Primary Stressors ......... ....................................................................... 25

3.1 Applied Loads .......... ..................................................................... 25

3.2 Eivironmental Stressors ................................................................................ 25

3.3 Manufacturing Stressors ............................................................................... 26

References............................................................................................................................................................... .. 27

4 Aging-Related Degradation Mechanisms ............................................................................... 29

4.1 Corrosion........................................................................................................................................................ .. 294.2 Fatigue ........ ....................................................................... 31

4.3 Erosion ........ ....................................................................... 324.4 Mechanical Wear ............................................................................... 32

4.5 Embrittlement ............................................................................... 32

4.6 Creep and Stress Relaxation ............................................................................... 33

4.7 PWR Internals and Potential Aging-Degradation Mechanisms .................................... I................................. 34

References ........ ..................................................... I. ................ 35

5 Survey of Aging-Related Failures ............................................................................... 37

5.1 ISI Program for Reactor Internals ............................................................................... 37

5.2 Failure Information Summary ............................................................................... 37

53 Failure Information Survey Results ............................................................................... 43

References............................................................................................................................................................... .. 45

6 ISI and Monitoring Programs ............................................................................... 47

6.1 LPMS ........ ....................................................................... 47

v V ~~~~~~~NUREG/CR-6048

6.2 Vibration Monitoring and Trending Studies ................................................ .................................... 48

References ........ ............................................................................ 49

7 Discussion and Conclusions ..................................................................................... 51

7.1 FIV ........ ............................................................................. 51

7.2 SCC ..................................................................................... 52

7.3 Other Potential Aging-Related Degradation Mechanisms ............................................................................. 52

NUREG/CR-6048 vi

List of Figures

Figure Page

2.1 Westinghouse PWR internals ......................................................................... 4

2.2 Westinghouse PWR lower core support structure . ........................................................................ 5

2.3 Westinghouse PWR core barrel cross sections ......................... ............................................... 6

2.4 Westinghouse PWR radial key-keyway system .......................... .............................................. 7

2.5 Westinghouse PWR thermal shield flexure support system ........................................................................ 7

2.6 Westinghouse PWR secondary core support assembly . ......................................................................... 8

2.7 Westinghouse PWR upper core support structure ............................... .......................................... 9

2.8 Westinghouse PWR thimble and guide tube ..................... ................................................... 11

2.9 B&W PWR internals ........................................................................ 12

2.10 B&W PWR eral shield lower support ........................................................................ 14

2.11 B&W PWR internal vent valve ......................................................................... 5

2.12 CE PWR internals ........................................................................ 17

2.13 CE PWR core support assembly ........................................................................ 18

2.14 CE PWR core support barrel ........................................................................ 19

2.15 CE PWR core support snubber assembly ........................................................................ 20

2.16 CE PWR upper grid assembly ........................................................................ 21

2.17 CE PWR in-core instrumentation support structure ...................................... .................................. 23

5.1 Westinghouse PWR downward bypass flow scheme ............................................ ............................ 39

52 Westinghouse PWR baffle water-jet impingement patterns ........................................................................ 40

5.3 Westinghouse PWR guide tube support pins ...................... .................................................. 41

5A KWU-built PWR core baffle and bolting scheme ............................... ......................................... 42

5.5 KWU-built PWR core baffle bolt failures ........................................................................ 42

vii NUREGICR-6048

List of Tables

Table Page

4.1 PWR internals primary stressors and aging-related degradation mechanisms ...................... .............................. 35

5.1 Summary of WE internals failure information ............................................................................. 44

5.2 Summary of B&W internals failure information ............................................................................. 45

5.3 Summary of CE internals failure infonnation . ............................................................................ 45

ix NUREG/CR-6048

Acknowledgments

The guidance and direction provided by D. A. Casada,W. S. Farmer, and R. D. Cheverton in dhe conduct of thisstudy are greatly appreciated.

The assistance of C. C. Soutbmayd and E. W. Carver inpreparing this report is gratefully acknowledged.

The author also wishes to hank A. E. Cross and W. P.Poore m for their help in obtaining the component failuredata for PWR internals.

xi NUREG/CR6048

Summary

Reactor internals operate in the environment inside thepressure vessel; this is favorable to the development oftime-dependent or aging degradations. The main objectiveof this study is to assess the effects of aging degradationson pressurized-water reactor (PWR) internal components.The assessment Includes an evaluation of the effectivenessof the plant in-service inspection (ISI) program in detectingfailures in nternals before they can affect the safety andefficiency of plant operations. Westinghouse (WE),Combustion Engineering (CE), and Babcock & Wilcox(B&W) reactors are included in the study.

Reactor internals selected for the study serve three basicfunctions: they support the core; they provide housings forcontrol rods, control rod drive mechanisms, and in-coremonitors; and they direct and guide the reactor coolantflow through the reactor vessel. Most internals are made oftype 304 stainless steel. Although they are located insidethe reactor vessel, fuel assemblies, control rods, control roddrive mechanisms, and in-core monitors are not includedas internals in the present study.

Stressors are conditions that can initiate and sustain thegrowth of aging-related degradation mechanisms. Reactorinternals are subjected to stressors generated by appliedloadings (thermal and mechanical), by contacts with high-tempemture flowing water, and by exposures to high-energy (E > I MeV) neutron fluxes. The applied loadingsof primary concern are flow-induced oscillatory hydro-dynamic forces because they can excite structural compo-nents into vibrations. The reactor cooling water providesan environment that may contain conditions that are con-sidered as favorable to the development of stress corrosioncracking (SCC). Neutron irradiation effects are importantstressors to internal components located in close proximityof the core. Manufacturing processes may also imposestressors on nternals.

Aging-related degradation mechanisms generally associ-ated with the primary stressors for reactor internals arecorrosion (including SCC), fatigue, mechanical wear, ero-sion, embrittlement, creep, and stress relaxation. Thesedegradation mechanisms may develop at different rates,and as a result they are not of equal mportance in thedesign life expectancy of the reactor. However, agingdegradations, if they are not mitigated, will eventually leadto a failure in the affected component. Reported failureinformation and laboratory testing results can identify themore significant aging-related degradation mechanisms.SCC, fatigue, and mechanical wear are the major aging-related degradation mechanisms for PWR internals.

Major reported failure cases for PWR internals includebolting failures in core support structures caused by fatigueand SCC, fuel assembly damages caused by high-speedleakage flows through enlarged gaps in core baffle plates,excessive thinning of flux thimble guide tubes caused byflow-induced vibrations, and crevice-assisted SCC in con-trol rod guide tube support pins. Some of these failureshave resulted In extended outages for repair work, but cur-rently available information did not indicate that they havecompromised the safety-related functions of reactorinternals.

The plant in-service inspection ([SI) program calls for thevisual inspection of accessible areas of reactor internalsduring refueling outages. Limitations of the visual Inspec-tion method are well known. Reactors licensed since 1978are equipped with loose parts monitoring systems(LPMSs). The main objective of the LPMS is to alert plantoperators that loose parts are present in the reactor primarysystem so that appropriate actions can be taken to limit fur-ther damages to other reactor components and systems.New technologies have been developed, and they have thecapability of early failure detection in key core-supportinternal components. They involve the use of neutron noisevibration measurements and trending studies. These prac-tices, while common in France and Germany, have notbeen formally incorporated into the ISI program for U.S.nuclear plants. They have been used on a voluntary basis.Visual inspection, supplemented by ultrasonic and eddy-current inspection methods, remain the major tools forinspecting reactor internals in the U.S.

Most flow-induced vibration and SCC problems have beenresolved by conventional engineering practices such aselimination of excitation sources, detuning the structurefrom external excitations, design changes, and use of mate-rials That will make the components less susceptible to cor-rosion attacks.

Small-amplitude flow-induced vibrations in reactor inter-nals are difficult to eliminate, and high-cycle fatigueremains as an active degradation mechanism. Internalsclose to the core are exposed to high-energy neutron fluxesand neutron irradiation effects, and related degradationmechanisms may become more prominent in the remaininglife of the reactor. Thermal aging in cast austenitic stainlesssteel (CASS) internal components is in the same category.In the presence of active aging degradations, a vigilantinspection program is needed to ensure the structuralintegrity of reactor internals. The Incorporation of an effec-tive early failure detection method will further enhance thesafety and efficiency of plant operations.

xiii NUREGICR-6048

1 Introduction

Systems, structures, and components of a commercialnuclear power plant are subjected to time-dependent oraging degradations during operations. Effects of agingdegradations, if they are not mitigated, will eventually leadto failures that could adversely affect plant safety and per-fornance. The Office of Nuclear Regulatory Research(RES) has established the Nuclear Plant Aging Research(NPAR) Programl to increase basic understanding ofaging-related degradations and their effects on reactor sys-tems, structures, and components. The NPAR approach isto perform in-depth studies on selected reactor systems,structures, and components that are judged to be vulnerableto aging degradations. Understanding the interrelationshipbetween stessors and aging-related degradation mecha-nisms is also the basis for the formulation of strategies forcontrolling and managing aging effects. One of the reactorsystems selected for aging study is reactor internals, andthe study is assigned to the Oak Ridge National Laboratory(ORNL). The effects of aging on boiling-water reactor(BWR) internals has been addressed in a previous report.This report will concentrate on the aging assessment ofpressurized-water r (PWR) internals. Operating his-tories of Babcock & Wilcox (B&W), CombustionEngineering (CE), and Westinghouse (WE) reactors pro-vide the majority of the information for the aging assess-ment process. Westinghouse has licensed Its PWR technol-ogy to European and Japanese vendors, and appropriateaging-related failure information of overseas reactors willalso be included in the study.

The term "internal" is generally applied to reactor compo-nents that are located inside the reactor pressure vessel.Fuel assemblies, control rods, control rod drive mecha-nisms, and in-core monitoring equipment are routinelyreplaced and are excluded from this study. Housings forthese components are considered as internal components.Reactor Internals perform many functions; the primary oneis to provide structural support and orientation to the coreand control rod assemblies (CRAs). Other internal compo-nents direct and guide the coolant flow through the coreregion and provide shielding to the pressure vessel wall.

At the present time there are 73 PWRs licensed for com-mercial operation in the United States. Using the commer-cial operation starting date as the reference for countingreactor ages, 6 reactors or about 8% of the total are over 20years old; 42 reactors or 58% are between 10 and 20 yearsold, and 25 reactors or 34% are less than 10 years old.There is a total of 907 reactor-years of PWR operations,and the accumulated operating histories of these reactors

1K H. Lk, EMcling Water Ram Iteals Aing DegadaionStudy-A Phas 1 Reor4" USNRC Report NUREGACR-S754(ORNITM- 1 176), to be pubished.

provide the Information for studying aging effects inselected reactor components. The plant ISI Program s themajor source of information on aging-related failure forreactor systems.

The aging assessment is performed in a multiple-step pro-cess. The first step is the identification and description ofreactor internal components included in the study. The sec-ond step is to identify stressors that are presented in theoperating environment inside the pressure vesseL The thirdstep Is to establish linkage between stressors and aging-related degradation mechanisms. The final step is the iden-tification of the more significant aging-related degradationmechanisms based on a review of the operating histories ofPWRs and reported component failure information. Theestablishment of the proper linkage between an aging-related degradation mechanism and the associated stressorscan be used as the basis for formulating strategies for con-trolling and managing aging effects.

Selected reactor internals are identified in Chap. 2 of thereport The information provided in Chap. 2 includes abrief description of each selected component, the functionsit performs, and the material of construction. Primarystressors Inside the reactor pressure vessel are discussed inChap. 3. Stressors generated by applied loads, environmen-tal conditions, and manufacturing processes are included inthe discussion. Chapter 4 identifies aging-related degrada-tion mechanisms associated with the primary stressors.Potential aging degradation mechanisms include corrosion(including SCC), fatigue, mechanical wear, erosion,embrittlement (thermal and radiation induced), creep, andstress relaxation. Chapter 5 is a summary of the more sig-nificant reported aging-related failures of PWR internals. Itincludes discussions in the thermal shield support bolt fail-ures in B&W, CE, and WE reactors; baffle plate water-jetting problems In WE units; WE control rod guide tubesplit pins failures; and flux thimble tube thinning prob-lems. The core baffle bolt failures in Kraftwerk Union(KIWU)-built PWRs of the WE design are also included inChap. 5. The reported aging-related failure informationidentifies three major aging-related degradation mecha-nisms: fatigue, SCC, and mechanical wear.

This study also addresses issues concerning the Inspectionand maintenance methods used to control and manageaging effects in reactor internals. The effectiveness of thevisual inspection method is discussed in Chap. 6 of thereport, which also provides information on the develop-ment of new technologies in these areas, such as loose-partmonitoring, neutron noise vibration measurements, and

I NUREGICR-6048

Introductiontrending studies. Chapter 7 is a summary of importantresults in this Phase 1 aging assessment of PWR internals.

Reference

1. J. P. Vora, Nuclear Plant Aging Research (NPAR)Program Plans, USNRC Report NUREG-1 144, Rev. 1,September 1987.*

Available for purchwe fron National Technical Information Service,Springfield. VA 22161.

NUREG/CR-6048 2

2 PWR Internal Components

lbree domestic PWR vendors in the United States are theWestinghouse Electric Corporation (WE), the CombustionEngineering Company (CE), and the Babcock & WilcoxCompany (B&W). Of the 73 domestic PWRs in commer-cial operation, 51 or about 70% of the total are WE, 15(20%) are CE, and 7 (10%) are B&W reactors. The threereactor designs share some common features, but the inter-nals are sufficiently different that using "generic" compo-nents foraging studies is not feasible. The three reactorinternal systems will be treated separately.

The primary function of reactor internals is to providestructural supports to the core and to properly positionCRAs under normal and accident operating conditions.Reactor internals that perform such functions are compo-nents of the core support system. Other internal compo-nents direct and guide the coolant flow through the coreregion and help to generate a uniform flow distribution toenhance core heat transfer. A third type of core internals isdesigned to provide gamma and neutron shielding to thereactor pressure vessel. Housings for in-core instrumenta-lions are also considered as reactor internals. Even thoughthey are also located inside the reactor vessel, aging effectsin fuel assemblies, control rods, control rod drive mecha-nisms, and in-core monitors are not addressed in this study.

The majority of reactor internals are made of type 304stainless steel. Unless it is specifically stated, It can beassumed that the material of construction for an internalcomponent is type 304 stainless steel. Reactor internals aredesigned in accordance with the requirements of Sect. mof the American Society of Mechanical Engineers (ASME)Boiler and Pressure Vessel (B&PV) Code.I For reactorsthat were designed and built before the establishment ofSect. HI of the ASME B&PV Code, analyses were per-formed to ensure that calculated stress values for reactorcomponents met the intent of the Code under specifieddesign conditions.

Information on internal components is obtained from vari-ous plant Final Safety Analysis Reports (FSARs) and tech-nical reports published by the Electric Power ResearchInstitute (EPRI). The components included in the study arethose for a "typical" reactor made by the vendors. Reactordesign is an evolving process, and many design featuresare plant specific. The report will attempt to identify majordesign changes in internal components, but no attempt willbe made to account for all design changes. Reactor inter-nals are identified by their names commonly used in plantFSARs. They may be referred to by different names inother reports, and confusions can be avoided by referringto the detailed descriptions of the component in question.

2.1 Westinghouse (WE) Internals

WE PWR internals are divided into three structural units:the lower core support structure, the upper core supportstructure, and the in-core instrumentation support structure.A simplified sketch of the arrangement of WE reactorinternals is shown in Fig. 2.1.

2.1.1 Lower Core Support Structure

The lower core support structure is the principal core sup-port structure. It consists of the core barrel, the core baffle,the lower core plate, the lower support columns, the bottomsupport plate, the intermediate diffuser plate, the thermalshield, and the secondary core support assembly. A sketchof the lower core support structure Is shown in Fig. 2.2.

In addition to its core support functions, the lower coresupport structure also directs and guides the coolant flowthrough the core. The coolant enters the vessel throughinlet nozzles and flows down the annular region betweenthe core barrel and the vessel wall. The thermal shield,when it is used, is located in the annular region. The maincoolant flow goes into a plenum at the bottom of the vesselwhere it is turned around and then flows up into the corethrough perforations of the bottom support plate, the dif-fuser plate, and the lower core plate. After passing throughthe core, the coolant enters the upper core barrel regionoccupied by the upper core support structure. The flowthen turns radially outward and leaves the core barrelthrough outlet nozzles. The core barrel outlet nozzles directthe coolant flow into the pressure vessel outlet nozzles.

In addition to the main coolant flow described above, thereare also secondary flows in the reactor coolant system. Asmall amount of cooling water is diverted into the regionbetween the core baffle and the core barrel, and this bypassflow provides additional cooling to the core barrel. Coolantflow is also directed into the vessel head plenum for cool-ing purposes, and it exits through vessel outlet nozzles.

2.1.1.1 Core Barrel

The core barrel is the main structure of the core supportsystem. Other core support structures are attached to thecore barrel, which transmits the weight of the core to thereactor vessel. It is also used to position the fuelassemblies.

3 NUREG/CR-6048

PWR

ORNL-DW 93-2884 EM

* Control Pl dtove shat

Control rod cust (withdrawn)

Outlet noose

* Stun

D BafU radl suppon

Lower core plate

Fowimft plateCore euppon columns

InsuISmentItln thumble quide"

Figure 2.1 Westinghouse PWR Internals

The barrel is a long, cylindrical, one-piece welded structuredivided into an upper and a lower region. The upper flangeof the core barrel rests on a ledge in the reactor vessel beadflange, and the lower end of the barrel is welded to thebottom support plate, which in turn is restrained by a radialsupport system attached to the wall of the pressure vessel.The outlet nozzles are made by welding forged rings toopenings in the upper part of the core barrel.

In some of the older WE reactors, the core barrel is a two-piece structure with the bottom of the core barrel connectedto the bottom support plate by tie rods. The upper andlower barrel sections are connected by a bolted joint.

NUREG/CR-6048 4

PWR

ORNL-OWG 92-3460 ETD

Figure 2.2 Westinghouse PWR lower core support structure

2.1.1.2 Core Baffle

The core baffle forms the boundary of the core, and it alsodirects and guides the coolant flow through the core region.

The core baffle is made of vertical baffle plates and hor-zontal former plates (also known as baffle radial supportplates). The horizontal former plates are bolted to theinside surface of the lower part of the core barrel. The ver-tical plates are bolted to the inner edges of the horizontalplates, forming the boundary of the core. The bolts aremade of type 316 stainless steel. Holes in the horizontalformer plates provide a flow path for the bypass coolingflow in the region between the core barrel and the verticalbaffle plates. A cross section of the core showing the corebaffle and other components of the lower core supportstructure is shown in Fig. 2.3.

2.1.1.3 Lower Core Plate

The. lower core plate positions the fuel assemblies andtransmits their weights to the core barrel. This perforatedplate is located at the bottom of the core below the core

baffle assembly. The plate thickness is -5.08 cm (2 in.). Itrests on a ledge on the Inside of the lower core barrel. Theholes distribute the coolant flow to the core. Fuel assemblylocating pins, two per assembly, are inserted into the lowercore plate. They provide proper alignment as well as lateralsupport to the fuel assemblies.

2.1.1A Lower Support Columns

The weight of the fuel assemblies Is -113,276 kg(250,000 lb), and to prevent excessive deformations of thelower core plate, it is stiffened by the placement of shortcolumns between the lower core plate and the bottomsupport plate (Fig. 2.2). The lower support columns arealso referred to as the core support columns. The top endsof the columns are bolted to the under surface of the lowercore plate. The bolts are made of type 316 stainless steel.The column lower ends are inserted into the bottomsupport plate. A major portion of the fuel assemblies'weight is then transmitted to the core barrel through thebottom support plate.

5 NUREG/CR-6048

PWROANLWM ZN5 ETO

THERMALBAFFLE

ET NOZZLES

REACTOR VESSEL'

FUEL ASSEPABLY

Figure 2.3 Westinghouse PWR core barrel cross sections

2.1.1.5 Bottom Support Plate

The bottom support plate supports the lower core platethrough lower support columns. It transmits a major por-tion of the fuel assemblies weight to the bottom of the corebarrel.

The bottom support plate, also known as the bottom disk,is a perforated plate welded to the lower end of the corebarrel. Typical plate thickness is -20.32 cm (8 in.). Lowerends of the core support columns are inserted into the bot-tom plate. Perforations in the plate distribute coolant flowto the core.

The bottom support plate is an integral part of the radialsupport system at the lower end of the core barrel. Thesupport system consists of keys and keyways attached tothe bottom support plate and the reactor vessel wall.Inconel clevis blocks are welded to the inside of the reactorvessel at six uniformly spaced locations around the circum-ference. Six Inconel insert blocks are bolted to these clev-ises and act as keyways. Six keys, also uniformly spaced,are welded to the outside surface of the bottom supportplate. When the core barrel is lowered into the pressurevessel, the keys engage the keyways in the axial direction,and lateral motions of the core barrel lower end arerestrained. A sketch of the radial key-keyway support sys-tem is shown in Fig. 2.4.

The bottom support plates in some of the older reactors aremade of Grade CF-8 cast austenitic stainless steel. Whenthey are casted, the bottom support plate is also referred toas the bottom support casting. The bottom support platesare also machined from type 304 stainless steel blocks.

2.1.1.6 Intermediate Diffuser Plate

The intermediate diffuser plate is a perforated plate locatedbetween the lower core plate and the bottom support plate.It is attached to the lower support columns. The intermedi-ate diffuser plate, also called the flow mixer plate, gener-ates a uniform coolant flow to the fuel assemblies. Not allWE units are equipped with intermediate diffuser plates.Perforations in the lower core plate and the bottom supportplate are used to provide the uniform flow distributionwhen the diffuser plate is not used. Intermediate diffuserplates in some of the older reactors are made of GradeCF-8 cast austenitic stainless steel. Others are machinedfrom type 304 stainless steel plates.

2.1.1.7 Thermal Shield

The thermal shield protects the reactor vessel wall fromexcessive gamma heating and radiation damages. It shieldsthe vessel wall from fast neutron fluxes and gamma radia-tions. It is located in the annular region between the corebarrel and the reactor vessel wall.

NUREGICR-6048 6

PWR

ORNL-DWG 03-288 ETM

LEVIS

LEVIS INSERT

ADIAL KEY

VESSEL

Figure 2.4 Westinghouse PWR radial key-keyway system

The early thermal shields are made of shell segmentsassembled inside the reactor vessel. The shell segments(typically three) were fastened together to form a cylindri-cal shell by the use of pins or bolts. The lower end of theshield was keyed to support lugs attached to the bottom ofthe vessel The top end was free. One WE PWR wasequipped with radial spacer pins at the top of the thermalshield.

The intermediate thermal shield design is a one-piececylindrical structure and supported In one of two ways. Inone support system, the shield is rigidly bolted to the loweredge of the core barrel, and a flexure support system isused at the top. The flexure support is basically a structuralelement with one end attached to the thermal shield and theother end bolted to the core barrel. Six flexure supports arespaced unifornly on the top of the shield. The flexure sup-port system allows limited displacements at the top end ofthe shield. A top-mount flexure support system is shown inFig. 2.5. Support conditions are reversed for the secondtype of shield support system; the top has a rigid support,and a flexure support system is used at the bottom edge.Material surveillance samples are inserted into surveillancespecimen holder tubes that are bolted and pinned to theoutside of the thermal shield.

ORNL4.WG 9232? EMD

COREBARREL

FLEXURESUPPORT N

IN

THERMALSHIELD -

Figure 2.5 Westinghouse PWR thermal shield Ilexuresupport system

NUREG/CR-60487

PWR

The latest thermal shield design uses shielding pads. Fourshielding pads, also called neutron shield pads, are boltedand pinned to the outside of the core barrel. These pads arelocated in areas with high neutron fluence levels to offerthe maximum protection to the reactor vessel wall. In thisdesign, material surveillance samples are located in speci-men holder tubes bolted to the outside of the neutron shieldpads.

2.1.1.8 Secondary Core Support Assembly

The secondary core support assembly, often referred to asthe core catcher, is an energy-absorbing device used tolimit the vertical displacement of the core and to absorbsome of the impact energy in the event of a catastrophicfailure of the core support system. Figure 2.6 is a sketch ofa typical secondary core support assembly. The number ofenergy absorbers required is plant specific, and it is deter-mined by the condition that the maximum stress value inreactor internals (except in the absorbers) is below theyield stress of the material of construction for the affectedcomponent during a postulated core drop. Yield stress val-ues can be found in SecL III of the ASME B&PV Code.

The energy absorbers, cylindrical in shape, are attached toa base plate that is contoured to fit the bottom surface ofthe reactor vessel. The top ends of the absorbers are con-nected to columns that are bolted to the bottom supportplate. In the event of a core support failure, the energyabsorbers will limit the fall of the core as well as absorbingsome of the kinetic energy of the dropped core assembly.This is accomplished through plastic deformations of avolume of stainless steel, initially loaded in tension, in theabsorber. The plastic strain in the stainless steel piece islimited to -15%, and then a positive stop is used to limitthe fall and transmit the additional loads to the bottom ofthe reactor vessel.

2.1.2 Upper Core Support Structure

Tle upper core support structure is located in the upperregion of the core barrel. The assembly consists of the topsupport plate, hold-down spring, deep beam sections, sup-port columns, upper core plate, and guide tube assemblies.

ORNL-DWO 93-2887 Ero

CORE BARREL

8cTTON SUPPORTPLATE

LP D

ENERGY ABSORBING DEVICE

Figure 2.6 Westinghouse PWR secondary core support assembly

8NUREG/CR-6048

PWR

The upper core support assembly is defined by the top sup-port plate on the top and the upper core plate in the bottom.Proper spacing between the two plates is maintained bysupport columns. Deep beam sections are attached to thebottom of the top support plate to increase its stiffness andto minimize deflections. Guide tube assemblies providehousing for CRA drive shafts and rod control clusterassemblies (RCCAs). The upper core support structure ismoved as a unit during refueling operations. A sketch ofthe upper core support structure is shown in Fig. 2.7.

2.12.1 Top Support Plate

The top support plate, also called the upper support plate,serves as the primary structure for the attachments of otherupper internal components. The plate transmits the weightof these components to the reactor vessel.

The top support plate is a perforated flat plate that is stiff-ened by deep beam sections at its bottom surface. The edgeof the plate rests on top of the upper core barrel flange. Theplate is held in place by the hold-down spring, which is Inturn held down by the vessel head flange. Perforations onthe plate provide access for the RCCAs and in-core ther-mocouple conduits.

The top support plate is also the dividing boundarybetween the upper plenum and the reactor vessel headregion. The coolant flow, which enters the upper plenumthrough the upper core plate, is turned to a radial direction

TOP SUPPORTPLATE

DEEP BEAM .

by the top support plate. The coolant flow exits through thecore barrel outlet nozzles.

In some WE units, the flat top support plate is replaced byan upper support assembly. The upper support assembly isin the form of a shallow top hat or an inverted shallow tophat. In the top hat configuration, perforations on the topsurface provide access for the RCCA and in-core thermo-couple conduits. The rim of the top hat serves as a supportflange, and it rests on the upper flange of the core barrel.The undersurface of the top hat is stiffened by deep beamsections. In the inverted top hat configuration, the bottomis a thick plate, and perforations on the plate provideaccess for the RCCA and in-core thermocouple conduits.The rim of the inverted top hat rests on the upper flange ofthe core barrel. The Inverted top hat upper support assem-bly is not stiffened by deep beam sections.

2.1.2.2 Hold-down Spring

The hold-down spring limits the axial movements of coresupport structures. It is a large annular or ring spring. Thecore barrel upper flange, the annular hold-down spring, andthe top support plate all rest on a ledge on the reactor ves-sel head flange. When the vessel head is installed on thepressure vessel, the hold-down spring is compressed by thetightening of the closure nuts, and the components restingon the ledge form a rigid joint that would restrict axialmovement of the core support structures. The hold-downspring is made of type 403 stainless steel.

ORNL-DWG 92S461 ETD

SUPPORTCOLMNS

GUIDETUBE

PLATE

Figure 2.7 Westinghouse PWR upper core support structure

9 NUREGICR-6048

PWR

2.1.2.3 Deep Beam Sections

Deep beam sections are welded to the bottom surfaces ofthe top support plate and the upper support assembly with atop hat configuration. The beam sections increase the platestiffness, and loadings acting on the plate are transmitted tothe reactor vessel without excessive deformations. Deepbeam sections are not used in the upper support assemblywith an inverted top hat configuration.

2.1.2.4 Upper Support Columns

Upper support columns are tubular structures bolted to thetop support plate and the upper core plate. The bolts aremade of type 316 stainless steel.

The upper support columns maintain the proper spacingbetween the top support plate and the upper core plate.They transmit loadings from the upper core plate to thestiffened top support plate that, in turn, transmits the load-ings to the reactor vessel. Some of the support columnsalso provide structural supports to in-core thermocoupleconduits.

2.1.2.5 Upper Core Plate

The primary function of the upper core plate is to establishproper alignment for the upper core support structure, thelower core support structure, fuel assemblies, and controlrods. The upper core plate is a perforated plate bolted tothe lower ends of the upper support columns. The bolts aremade of type 316 stainless steel.

The upper core support structure is positioned with respectto the lower core support structure by a pin-slot alignmentscheme. Four unifornly spaced flat-sided pins are weldedto the inside surface of the core barrel at the elevation ofthe upper core plate. Slots are milled into the upper coreplate at the corresponding positions. When the upper coresupport structure is lowered into the core barrel, the slotsengage the flat-sided pins in the axial direction, and properalignment is achieved. The pin-slot alignment scheme willalso restrict lateral displacements of the upper core supportstructure.

The top ends of fuel assemblies are aligned by locatingpins protruding from the bottom of the upper core plate.The pins engage the fuel assemblies when the upper coresupport structure is lowered into place.

2.1.2.6 Guide Tube Assemblies

Guide tubes provide housings for control rod drive shaftsand RCCAs. They shield these components from effects of

cross-flows in the upper plenum region of the reactor pres-sure vessel. Guide plates are used to maintain proper spac-ing of the RCCA rodlets in the guide tubes.

The guide tube assembly is a tubular structure and isdivided into two parts or assemblies, with the upper sup-port plate serving as the dividing line. The upper assembly,often referred to as the control rod shroud tube, is fastenedto the top of the upper support plate. RCCAs enter theupper plenum through openings in the reactor vessel head,and they are inserted into the top openings of the controlrod shroud tube. The lower part is called the control rodguide tube. The top end of the control rod guide tube is fas-tened to the bottom of the top support plate, and the lowerend is held in place by split pins inserted into the uppercore plate. The split pins are bolted to the guide tube bot-tom; the support will allow limited axial movements of thelower end of the guide tube, but lateral displacement willbe restrained.

2.1.3 In-core Instrumentation SupportStructures

In-core instrumentation support structures are stainlesssteel tubular structures that provide housing and support toin-core instrumentation such as in-core thermocouples andthimbles of the reactor flux-mapping system.

Thermocouple housings penetrate the vessel through thevessel head, while the flux thimble guide tubes enter thevessel through the bottom. In a few of the older reactors,all in-core instrumentations penetrate the vessel from thevessel head.

2.13.1 In-core Thermocouple Housing

In-core thermocouple housings house and guide in-corethermocouples before their insertion into the core. Thehousings start as port columns that penetrate the vesselfrom the vessel head. They are also referred to as upper in-core instrumentation port columns. These port columns areslip-connected to in-line columns fastened to the top sup-port plate. Thermocouples are inserted through these portcolumns and the top support plate to positions above theirsensor locations in the core.

2.13.2 Flux Thimble Guide Tubes

Flux thimbles are components of the reactor in-core neu-tron monitoring or flux-mapping system. They are insertedinto the reactor pressure vessel through the bottom. Fluxthimbles are retracted during refueling and maintenanceoperations. Guide tubes provide housing and support tothese thimbles along most of their lengths. Because they

NUREG/CR-6048 10

enter the reactor vessel through the bottom, they are alsoreferred to as lower in-core instrumentation.

A thimble guide tube is divided into two parts. The upperpart is located inside the fuel assembly, while the lowerportion is located between the reactor vessel bottom andthe lower core plate. The segment of the flux thimblebetween the bottom of the fuel assembly and the top of thelower core plate is not protected by the guide tube, and it isexposed to the reactor coolant flows.

When they leave the reactor vessel, flux thimbles arelocated inside high-pressure conduits, which penetrate thepressure vessel from the bottom. The high-pressure con-duits, containing the flux thimbles, continue on to the sealtable. A simplified sketch of the flux thimble and guidetube arrangement is shown in Fig. 2.8.

PWR

by a mechanicil seal at the seal table. The flux thimbletube and the high-pressure conduit are considered as a partof the reactor primary pressure boundary.

2.2 Babcock & Wilcox (B&W)Internals

B&W reactor internals are grouped into two main struc-tural assemblies: the core support assembly and the plenumassembly. In-core instrumentation guide tubes are a part ofthe core support assembly, and CRA guide tubes belong tothe plenum assembly. A sketch of the arrangement ofB&W reactor internals is shown in Fig. 2.9.

The core support assembly is the primary core supportstructure. The plenum assembly is used to maintain aproper alignment of the control rod guide tubes.

2.2.1 Core Support AssemblyThe thimbles are sealed at the reactor end. The regionbetween the flux thimble and the high-pressure conduit ismaintained at the reactor coolant pressure, and it is sealed

The core support assembly is the major core support struc-ture, and it consists of the core support shield, the core

ORL-MG 93-28 ETD

con

SIALTABLE

UNGWIDD THIMILE lUBE

Figure 2.8 Westinghouse PWR thimble and guide tube

11 NUREGtCR-6048

PWR

ORNLDWC3 93-2889 ErD

PLENUM ASSEMBLY -

OUTLET NOZZL -

CORE BARREL-

LOWER GR'

FLOW OISTRUUTOR

Figure 2.9 B&W PWR Internals

barrel, the lower grid assembly, the flow distributor, thethermal shield, surveillance specimen holder tube, in-coreinstrumentation guide tubes, and internal vent valves. Thecore support shield and the core barrel are the major com-ponents of the core support assembly. They provide struc-tural support and attachment points to other internalcomponents.

In addition to its core support functions, the core supportassembly also directs and guides the coolant flow in thecore region. The flow enters the vessel through inlet noz-zles, and the core support shield turns the coolant flowdownward in the annular region between the core barreland the vessel wall. The coolant flow is turned upward atthe bottom of the vessel, and it goes up to the core throughthe flow distributor and the bottom grid.

NUREG/CR-6048 12

The core barrel guides the coolant flow through the core.The main flow is upward through the fuel assemblies out-lined by the core baffle. A small quantity of the coolant isalso diverted to flow upward in the region between the corebaffle and the core barrel. The coolant pressure in the corebaffle-core barrel region is maintained at a lower valuethan the main coolant pressure in the core. The resultingpressure differential helps to prevent the formation of ten-sion stresses in the bolts attaching the vertical baffle platesto the horizontal former plates. In the event of flow leak-ages through gaps in the baffle plate joints, the pressuredifferential will force the coolant to move outward andaway from the fuel assemblies. This bypass flow schemecan prevent fuel assembly damages caused by baffle platewater-Jetting problems.

2.2.1.1 Core Support Shield

The core support shield provides structure supports to otherinternals and transmits loadings to the reactor vessel. Theshield is a cylindrical structure with flanges at the ends.The top flange is forged, and it rests on a circumferentialledge in the reactor vessel closure flange. The core supportshield lower flange is bolted to the top flange of the corebarrel. The plenum assembly is located inside the coresupport shield. The position and orientation of the plenumassembly with respect to the core support shield are main-tained by fixtures located on the shield inside surface. Insome B&W units the lower end of the core support shieldis welded to the top of the core barrel.

The core support shield wall contains two types of open-ings. The first type consists of outlet nozzles used for reac-tor coolant flow. The exact number of outlet nozzles isplant specific. The flow nozzles are formed by weldingforged rings to the wall openings. The rings mate withinternal projections of the reactor vessel outlet nozzles. Thering seal surfaces are sized and finished so that a clearancegap is maintained at cold condition to facilitate the installa-tion and removal of components of the core supportassembly. When the reactor is heated to the operating tem-perature, the differential thermal expansions of the stain-less steel core support shield and the carbon steel reactorvessel will close the gap and form the necessary seal sur-face. The cold gap is sized so that when contact is made,maximum stresses in the reactor vessel and internal com-ponents remain below allowable values as stipulated inSect. M of the ASME BPV Code.

The second types of opening in the shield wall are internalvent valve openings. Mounting rings are welded to theseopenings for the installation of the internal vent valveassemblies. The exact number of vent valve openings isplant specific, but most B&W plants are equipped withfour to ight internal vent valves.

PWR

2.2.1.2 Core Barrel

The core barrel supports the fuel assemblies, the lower gridassembly, flow distributor, and in-core instrumentationguide tubes. It guides the primary coolant flow through thecore.

The core barrel is a cylindrical structure with flanged ends.The upper flange is bolted to the lower flange of the coresupport shield, and the lower flange is bolted to the lowergrid assembly. In some B&W units, the top end of the corebarrel is welded to the lower end of the core support shield.

The core baffle is considered as an integral part of the corebarrel. The baffle is formed by bolting a series of horizon-tal former plates to the inside surface of the core barrel.Vertical baffle plates are then bolted to the inside edges ofthe horizontal former plates. The vertical baffle plates formthe boundary of the fuel assemblies.

2.2.1.3 Lower Grid Assembly

The lower grid assembly provides structural supports to thefuel assemblies, the thermal shield, and the flow distnbu-tor. Fixtures attached to the lower grid assembly also alignthe in-core instrumentation guide tubes with fuel assem-blies' instrument tubes.

The assembly consists of two grid structures connected byshort tubular columns and surrounded by a forged cylinderwith flanges at the ends. The upper grid structure is a per-forated plate attached to the top flange of the forged cylin-der. Pads bolted to the perforated plate are used to align thefuel assemblies and in-core instrumentation guide tubes.The lower grid structure is formed by welded intersectingplates. In some units, the lower grid structure is a machinedforging.

The lower and upper grid structures are connected by tubu-lar columns. Also a perforated plate located midwaybetween the two grid structures is used to generate a uni-formly distributed coolant flow to the core. The lower gridstructure rests on and is also bolted to the lower flange ofthe core support shield. The top flange of the forged cylin-der is bolted to the lower flange of the core barreL

22.1.4 Flow Distributor

The flow distributor is a perforated dished head with anexternal flange that is bolted to the bottom of the lowergrid assembly.

13 NUREGICR-6048

PWR

Perforations in the flow distributor produce a uniformcoolant flow to the core entrance. The flow distributor alsoprovides support to the in-core instrument guide tubes.

2.2.1.5 Thermal Shield

The thermal shield protects the reactor vessel wall fromexcessive gamma beating and radiation damages. It shieldsthe vessel wall from fast neutron fluxes and gamma radia-tions. The shield is located in the annular region betweenthe core barrel and the reactor vessel wall.

The thermal shield is a cylindrical structure. The upper endof the shield is restricted against radial vibratory motionsby restraints bolted to the core barrel cylinder. The lowerend of the shield is shrunk-fit onto the upper flange of theforged cylinder in the lower grid assembly. As an addedassurance, 96 high-strength bolts are also used to securethe thermal shield to the upper flange of the forged cylin-der. The bolts are made of grade A 286 stainless steel. Asketch of the thermal shield lower support is shown inFig. 2.10.

2.2.1.6 Surveillance Specimen Holder Tube (SSHT)

SSHTs are cylindrical tubes mounted on the outside sur-face of the thermal shield. Each tube can hold two surveil-lance capsules, and the exact number of tubes in a reactoris plant specific. There is an off-set for the tubes, and they

are located at a short distance from the thermal shield out-side surface. The span of the tube typically extends fromthe top of the core support shield to the lower end of thethermal shield. There are usually three support mounts foreach holder tube.

2.2.1.7 In-core Instrumentation Guide Tubes

In-core instrumentation guide tubes provide housing andsupports to in-core instruments from penetrations in thereactor vessel bottom head to instrument tubes in the fuelassemblies. Guide tubes are segmented tubular structureswith different diameters for the various sections. The lowerend of the tubes are connected to instrument penetrations atthe vessel bottom head. The tubes then go through open-ings in the flow distributor and the lower grid assemblybefore being connected to instrument tubes in the fuelassemblies.

2.2.1.8 Internal Vent Valves

The use of internal vent valves is a unique design featureof B&W PWRs. The primary function of the internal ventvalve is to release the pressure buildup in the reactor upperplenum region following a cold-leg pipe break accident.

The internal vent valve assemblies are attached to mount-ing rings welded to openings in the wall of the core support

ORNL-DWG 92-3328 ET

THERMALn oEL

0.00 TO 0.060INTERFERENCEFIT

LOWER GRIDFORGING

96-0.8' BOLTS* LOCXINGCLIPS

Figure 2.10 B&W PWR thermal shield lower support

NUREG/CR-6048 14

shield. The mounting rings also contain devices and fea-tures that are used to align and position the valve assemblyfor proper operation of the valve disk. The internal ventvalve assembly consists of a hinge assembly, the valve diskwith sealing surfaces, split-retaining rings, and fasteners.The design of the internal vent valve is similar to that forswing check valves. The vent valves are uniformly spacedaround the circumference of the core support shield wall,and the exact number of valves required is plant specific. Asketch of an internal vent valve assembly is shown inFig. 2.1 1.

Under normal reactor operating conditions, the coolantpressure in the annular region between the core supportshield and the reactor vessel wall (downcomer) is higherthan the coolant pressure in the core. The magnitude of thispressure differential during normal operations is-0.29 MPa (42 psi). It will generate a large force that willpress the valve disk seal face tightly against the tilted valvebody seal face. Thus internal vent valves are closed duringnormal operations.

In the event of a cold-leg pipe break accident, the suddenloss of pressure would lead to rapid phase changes in theprimary reactor coolant. The coolant becomes a two-phasemixture of steam and water. The pressure in the core regionwill drop from the operating pressure of 15.5 MPa(2250 psi) to -10.6 MPa (1541 psi), which is the saturatedpressure at the operating temperature of 316C (6000F).The large thennal inertia of the core would prevent a rapidchange in the fluid temperature in the core region. A largepressure differential now exists between the core regionand the pipe break location, which is at the ambient pres-sure. The pressure differential will drive the reactor coolantfrom the core region to the pipe break location through two

PWR

potential flow paths. The first flow path is through the hotleg and the steam generator. The second flow path isthrough the core and the cold leg. The flow rate througheach path is determined by hydraulic resistances in eachflow path. In any event, reactor coolant is transported fromthe core region to the pipe break location. If the process isnot mitigated, the water level in the core will drop, and aportion of the core may be uncovered. Note that the pro-cess is dynamic, and the event could take place in a veryshort time interval.

The internal vent valve is designed to achieve a quick pres-sure equalization between the core region and the cold-legpipe break location and to prevent the lowering of the wa-ter level in the vessel below the top of the core. After thepipe break occurs, decompression waves propagate fromthe cold-leg pipe break location into the reactor vessel. Thefluid pressure of the core region becomes higher than thepressure in the annular region between the core supportshield and the vessel wall. The direction of the resultingpressure force will now oppose the valve closing forcegenerated by the hinge assembly. The valve closing forceis overcome when the magnitude of the reversed pressuredifferential attains a value of 1034 Pa (0.15 psi). The valvewill be fully opened with a pressure differential of no morethan 2069 Pa (03 psi). When the internal vent valves areopened, the major reactor coolant flow path, the one withthe least flow resistances, will be from the upper plenum ofthe core region to the cold-leg pipe break location. Theopening of internal vent valves together with the activationof the emergency core cooling system can prevent theuncovering of the core in a cold-leg pipe break accident.

Because of their important safety function, internal ventvalves are inspected and tested during refueling outages.

ORNL4-WG 929 ETD

rI-Z

-EXERCISE WG

SECTION Z-Z

- REACTORVESSELWALL

L-Z

Figure 2.11 B&W PWR Internal vent valve

1S NUREG/CR-6048

-

PWRThe valves become accessible to visual inspections andmechanical testings after the vessel head and the plenumassembly are removed. A hook tool is used to engage theexercise lug, and the freedom of the movement of thehinged valve disk is tested. When the valve disk is raised,a remote visual inspection is also performed on the valvebody and the sealing faces.

2.2.2 Plenum Assembly

The plenum assembly positions and supports the CRAguide tubes. It also provides a path for the reactor coolantto flow to the reactor outlet nozzles.

The assembly is made up of the plenum cylinder, theplenum cover, the upper grid, and CRA guide tubes. It islocated inside the core support shield and directly abovethe core. The structural unit is formed by attaching theplenum cover and the upper grid to the upper and lowerflanges of the plenum cylinder. CRA guide tubes connectthe plenum cover and the upper grid. Lifting lugs arewelded to the plenum cover, and the plenum assembly isremoved as a single unit during refueling operations.

The plenum assembly is aligned with the reactor vesselclosure head, control rod drive penetrations, and the coresupport assembly. When the bolts of the vessel closurehead are tightened, a large clamping force is producedbetween the reactor vessel, vessel closure head, the coresupport shield upper flange, and the plenum cover assem-bly upper flange. The clamping force helps to form a rigidjoint between the components involved, and it will limitmovements of these components. Proper positioning isattained by the locking of keyways in the plenum assemblycover flange to keys in the reactor vessel flange. The bot-tom of the assembly, which is the upper grid, is restrainedby the inside surface of the core support shield.

2.2.2.1 Plenum Cylinder

The plenum cylinder positions and supports CRA guidetubes and guides the reactor coolant flow to the outlet noz-zles. It is a cylindrical structure with flanges at the twoends. The side wall contains holes as outlets for the reactorcoolant flow. The plenum cover is attached to the topflange, and the upper grid is connected to the lower flangeof the plenum cylinder.

2.2.2.2 Plenum Cover

The plenum cover aligns the upper ends of CRA guidetubes and provides support points for the lifting of theplenum assembly. The cover is formed by welding parallel

intersecting plates to form square lattices, and the latticestructure is covered by a perforated flat plate with an inte-gral flange at the periphery. The inner edge of the integralflange is attached to the upper flange of the plenum cylin-der, while the outer edge of the flange is attached to theupper flange of the core support shield. Holes in the perfo-rated plate are positioned to match the upper ends of theCRA guide tubes. Lifting lugs are welded to the plenumcover.

2.2.2.3 Upper Grid

The upper grid aligns the lower end of CRA guide tubes tothe upper ends of fuel assemblies. The grid is a perforatedplate bolted to the lower flange of the plenum cylinder, andit is also guided by the inside surface of the lower flange ofthe core support shield. Perforations in the plate positionthe lower ends of CRA guide tubes to the upper ends offuel assemblies.

2.2.2.4 CRA Guide Tubes

CRA guide tubes provide housing and support to theCRAs. The tubes also shield the CRA from the effects ofcross-flows in the plenum assembly. Guide tubes are alsoconsidered as structural components connecting theplenum cover to the upper grid.

A CRA guide tube is a tubular structure with a mountingflange welded to one end. The mounting flange is bolted toan opening of the upper grid, while the top end of the guidetube is welded to a perforation in the plenum cover plate.CRAs are located inside the guide tubes.

2.3 Combustion Engineering (CE)Internals

CE internals are divided into four structural units: the coresupport assembly, the upper guide assembly, the flow skirt,and the in-core instrumentation support system. A sketchof the arrangement of CE reactor internals is given inFig. 2.12.

2.3.1 Core Support Assembly

The major core support structure is the core supportassembly. The assembly consists of the core support barrel,the core support plate, the lower support structure, the coreshroud, thermal shield, and the core support barrel to pres-sure vessel snubbers. Because of flow-induced vibrationproblems, thermal shields have been removed from severalCE reactors. A typical core support assembly is shown inFig. 2.13.

NUREG/CR-6048 16

PWR

ORNo 38W ED

UN-CORE INSTRUMENTATIONUPPORT PLATE

Figure 2.12 CE PWR Internals

The core support assembly also directs and guides thereactor coolant flow through the core region. The coolantenters the pressure vessel through inlet nozzles and isturned to flow downward in the annular region betweenthe core support barrel and the reactor vessel wall. Thecoolant flow is turned upward in the lower plenum. Itflows through the flow skirt, the bottom of the core supportbarrel, and the lower support structure before entering thecore through flow distribution holes in the core supportplate. At the cow exit, the coolant flow is turned to a radialdirection in the upper region of the core support barrel andleaves the bawrel through outlet nozzles. A small portion ofthe main coolant flow Is diverted into the gap regionbetween the core shroud and the core support barrel. Thissecondary flow is used to provide a more effective coolingof the core shroud. Other minor bypass flows are leakageflows through key-keyway alignment systems, instru-mentation guide tubes, and nozzle clearances.

23.1.1 Core Support Barrel

The core support barrel transmits the weight of the fuelassemblies to the reactor vessel. A typical total weight

of the fuel assemblies and claddings for a CE unit is-133,333 kg (-300,000 lb). The barrel also provides properalignment for the core support assembly, the upper guidestructure assembly, the reactor vessel, and the closure head.

The core support barrel is a long cylindrical structure withan external ring flange at the top and an internal ring flangeat the bottom. The top flange rests on a ledge in the reactorpressure vessel. Four equally spaced alignment keys arepress-fitted into the top flange of the core support barrel.The reactor vessel, closure head, and the upper guide struc-ture assembly flange are slotted to fit into the alignmentkey locations. The keys are used to align the core supportassembly and the upper guide assembly with respect to thereactor pressure vessel. A cut-away view of the core sup-port barrel is shown in Fig. 2.14.

Two outlet nozzles in the upper section of the core supportbarrel are fitted to internal projections of the reactor vesselexit nozzles. Four equally spaced guide pins are attached tothe inside of the core support banmel just below the outlet

NURECR-604817

PWR

ORNL-OW 93oa EMD

OUTLET

III

I IFigure 2.13 CE PWR core support assembly

nozzles, and the pins are used to align and to limit themovement of the upper guide assembly.

The core support plate rests on a ledge in the inside surfaceof the core support barrel. The ledge is located near thebottom of the barrel. The lower support structure rests onthe internal ring flange at the bottom of the barrel.

The lower end of the core support barrel is restrained bysnubbers located on the outside surface of the barrel. Thesnubbers allow radial and axial expansions of the core sup-port barrel, but they will restrict lateral or circumferentialbarrel displacements.

2.3.1.2 Core Support Plate

The core support plate positions and supports the fuelassemblies. It is a perforated flat plate, and the perforationsdistribute coolant flow to the fuel assemblies. The fuelassemblies rest on the core support plate and are positionedby locating pins (four for each assembly) shrunk-fit to theplate. The core support plate rests on a ledge located in theinside surface of the core support barrel near the bottomend. The periphery of the plate is pinned, bolted, and lockwelded to the ledge. The core shroud is also attached to thecore support plate.

NUREGICR-6048 18

PWR

ORNUWM 034M Io

Figure 2.14 CE PWR core support barrel

bolts. The gap in the region between the core boundary andthe core support barrel is maintained by seven tiers of hor-zontal centering plates. These centering plates are bolted tovertical core sMiud plates and centered during assemblyby adjusting bushings located in the core support barrel.All bolted joints in the core shroud are lock-welded. Thelocation of the core shroud in the core support barrel isshown in Fig. 2.14.

Most of the coolant flows through the core. Holes aredrilled in the horizontal plates to provide some coolantflow through the region between the core shroud and thecore support barrel. The flow in the gap region will elimi-nate stagnation pockets and provide a more effective cool-ing to the shroud. Improved shroud cooling would reducetemperature gradients and minimize thermal stresses in theshroud and the core support barrel.

2.3.15 Core Support Barrel to Reactor VesselSnubbers

The upper end of the core support barrel is clampedbetween the pressure vessel flange and the closure head.From a structural analysis standpoint, the core supportbarrel has a cantilevered support at the upper end, and noadditional support would be needed. However, the coresupport barrel is submerged in a turbulent flow and is sus-ceptible to flow-induced vibrations. The amplitudes ofsuch flow-induced vibrations would be larger at the lowerend of the barrel when it is not restrained. Snubbers areinstalled at the lower end of the core support barrel to limitand reduce the amplitude of potential flow-induced vibra-tions. The locations of the snubbers on the core supportbarrel are shown in Fig. 2.14.

The snubber system consists of six equally spaced lugswelded to the core support barrel outside surface. The coresupport barrel lugs act as the groove of a tongue-and-groove assembly. Mating lugs attached to the inside sur-face of the pressure vessel wall serve as tongues. Anexploded view of the snubber assembly is shown inFig. 2.15.

When the core support barrel is lowered into the pressurevessel, the pressure vessel lugs slide into the grooves of thecore support barrel lugs. Shims are bolted to the side sur-faces of the vessel lugs, and the corresponding surfaces ofthe barrel lugs are hard faced to minimize wear. There isno direct connection between the lugs. The gap clearancebetween mating surfaces will impose a limit on the barreldisplacements. The snubber system will not impede ther-mal expansions in the axial and radial directions. It wil,however, restrict lateral or circumferential expansions. The

23.13 Lower Support Structure

The lower support structure provides additional stiffness tothe core support plate and also transmits a part of the coreassembly weight to the bottom of the core support barrel.The lower support structure is a welded assembly thatconsists of a cylinder, columns, support beams, and a bot-tom plate. The core support plate is supported by columns,and the bases of these columns are welded to supportbeams. The bottoms of the support beams are welded to thebottom plate, which is perforated for coolant flow passage.The ends of the support beams are welded to the cylinder.The lower end of the cylinder is welded to the perforatedbottom plate. The whole assembly rests on the internal ringflange of the core support barrel. The outer edge of thecore support plate is also supported by the top end of thecylinder. The cylinder guides the main reactor coolant flowto the core and limits the core shroud bypass flow bymeans of holes located near the bottom of the cylinder.

23.1A Core Shroud

The core shroud forms the boundary of the core and con-trols the coolant flow through the core region. The shroudis formed by bolting vertical shroud plates to horizontalcentering plates. The vertical plates, of varying widths,become the core boundary. The bottom of the verticalplates are attached to the core support plate by anchor

19 NUREGR-6048

-

PWRORNL 2S3 EMD

CORE SUPPORT BARREL

.1ra PIN14 REQ'D PER ASSEMBLY I

SOLT 4 REQ'D PER ASEMBLY )

HIM I 2 REQ'D PER ASSEMBLY

Figure 2.1S CE PWR core support snubber assembly

snubber system is not a fixed support system, and the coresuppoit barrel is not immune from small-amplitudevibrations.

23.1.6 Thermal Shield

The thennal shield is a cylindrical structure located in theannular region between the core support barrel and thereactor vessel wall. The upper end of the shield is sup-ported by nine uniformly spaced lugs attached to the out-side of the core support barrel. The lugs restrict the axialand tangential movement of the shield. A preloaded posi-tioning pin under each lug is threaded radially through theshield and butts against the core barrel. The lower end ofthe shield is held in place in a similar manner by 17 radialpositioning pins.

Because of flow-induced vibration problems, thermalshields had been removed from several CE reactors.Analysis results indicated that it is not necessary to replacethese shields in the remaining design life of these reactors.Thermal shields are not used in the newer CE units.

2.3.2 Upper Guide Assembly

The upper guide assembly consists of the upper guidestructure support plate assembly, the fuel assembly align-

ment plate, control element assembly (CEA) shrouds, and ahold-down ring. The assembly is handled as a single unitduring refueling operations. A sketch of the upper gridassembly is shown in Fig. 2.16.

The upper guide structure (UGS) support plate assemblyprovides support and alignment to CEA shrouds, whichshield the control rods from cross-flow effects in the upperplenum region. The CEA shrouds also provide support toin-core instrumentation. The fuel alignment plate positionsthe upper ends of the fuel assemblies. The hold-downspring holds down the fuel assemblies during normal oper-ation and prevents the fuel assemblies from being lifted outof the core during accidents.

23.2.1 UGS Support Plate Assembly

The UGS support plate assembly is a welded structure. Theassembly is constructed by welding a support flange to thetop of a cylinder, and a support plate is welded to thecylinder inside surface at the cylinder midsection. In someCE units the support plate is located near the top of thecylinder, just below the support flange. The top of a gridarray structure, made of welding intersecting deep beams,is welded to the bottom of the support plate. The ends ofthe deep beams are welded to the cylinder inside wall. Thesupport plate and the deep beam grid array structure posi-tion and support the upper ends of CEA shrouds. Four

NUREGCR-6048 20

PWR

ORNL-DWG 03-2894 EM

MEL AsSEMBLY -ALIGNMENT PLATE

Figure 2.16 CE PWR upper grid assembly

emay spaced keyways are machined into the top supportflange, and they engage the core support barrel alignmentkeys. The key-keyway alignment scheme ensures a properalignment of the core with respect to the reactor closurehead and CEA drive mechanisms.

machined Into the support barrel ring flange and the hold-down ring, and they will engage core support barrel align-ment keys. The bottom plate of the support barrel providessupport and alignment to CEA shrouds.

2.3.2.2 Fuel Assembly Alignment Plate

In the new CE umits, the UGS support plate assembly Isreplaced by a UGS support barrel assembly. The UGS sup-port barrel assembly consists of a ring flange welded to thetop of a circular cylinder. A circular plate is welded to hebottom of the cylinder. The ring flange rests on the hold-down ring that, in tn, sits on the core support barrelupper flange. Four uniformly spaced keyways are

The fuel assembly alignment plate positions the upper endsof the fuel assemblies and also provides support to thelower ends of CEA shrouds. Locating holes are machinedInto the plate to engage posts on the fuel assembly upperend fittings. Four equally spaced slots or keyways are alsomachined into the outer edge of the fuel assembly align-ment plate, and they engage pins protruding from the core

NUREG/CR-604821

PWRsupport barrel. The pin-keyway arrangement restricts lat-eral displacements of the upper guide assembly duringoperations. The fuel assembly alignment plate pressesdown on the fuel assembly hold-down ring, and the upwardreaction forces are transmitted via the alignment plate andCEA shrouds to the flanged UGS support plate.

23.2.3 CEA Shrouds

CEA shrouds in most CE reactors are tubular structures.They extend from the fuel assembly alignment plate to anelevation above the support plate of the UGS assembly.The shrouds protect CEAs from cross-flow effects in theupper plenum.

The majority of CEA shrouds are the five-element type,and they are made by welding a cylindrical section to abase; the base is bolted and lock-welded to the fuel assem-bly alignment plate. Flow channel inlets are machined intothe cylindrical section at the base, and they serve as a pas-sageway for the coolant flow through the fuel assemblyalignment plate. The upper ends of the shrouds areconnected to the UGS support plate by spanner nuts, whichwould allow the shrouds to expand in an axial direction.

Four-element shrouds are located at the periphery of theUGS support plate, and they consist of a cylindrical sectionwelded to a base; the base is bolted and lock-welded to thefuel assembly alignment plate. The upper section of theshroud is welded directly to the UGS support plate.

In the older reactors, the CEA shrouds have a cruciformconfiguration, and they extend from the fuel assemblyalignment plate to an elevation just above the reactor vesselflange. The shroud is fabricated by welding four formedplates to four end bars to complete a cruciform-shapedstructure. The shroud ends are fitted with support pads.The bottom ends are bolted and lock-welded to the fuelassembly alignment plate and the top ends to the UGS sup-port plate. CEAs located inside these crucifonn shroudsalso shield them from cross-flow effects.

2.3.2.4 Hold-down Ring

The hold-down ring restricts axial displacements of inter-nal components. Differential thermal expansions, fuelgrowth, and rotations of the closure head during bolt tight-ening and pressurization are major causes of internal com-ponent axial displacements. The hold-down ring is alsoreferred to as the expansion compensation ring.

ring segment contains plungers supported by Bellevillewashers, and the compression of these devices will resultin an axial hold-down force acting on the upper guideassembly. The ring segments are fabricated from type 403stainless steel.

A shim plate is inserted into the space between the UGSand the core support barrel flange to accommodate internalcomponents' axial expansions.

2.3.3 Flow Skirt

The flow skirt is a perforated right circular cylindricalstructure with stiffening rings at its top and bottom. Theskirt is supported by nine equally spaced machined sec-tions welded to the bottom head of the reactor pressurevessel. There is no connection between the flow skirt andother internal components. The skirt is made of Inconel.The flow holes are designed to provide a uniform inletflow to the core.

2.3.4 In-core Instrumentation SupportStructure

The in-core instrumentation support structure is a part ofthe in-core neutron flux monitoring system. It consists ofan instrumentation support plate that fits in the recess sec-tion of the UGS assembly and is supported by four bearingpins. CEA shrouds extend through perforations in theinstrumentation plate. The in-core instruments are guidedand protected by in-core instrumentation guide tubes thatroute the instruments to various locations in the core. Theguide tubes are bent and grouped together to form clusterassemblies above the instrumentation plate. The clustersare supported by frame-type structures bolted to the instru-mentation plate, and they extend into the reactor vesselhead instrumentation nozzles. A sketch of an in-coreinstrumentation support structure is shown in Fig. 2.17.

Reference

1. American Society of Mechanical Engineers, ASMEBoiler and Pressure Vessel Code, Section III, NuclearPower Plant Components, Div. 1.*

Available from American National Standards Institute, 1430 Broadway,New York, NY lOOS, copyrighted.

The hold-down device is a segmented circular frame, andeach ring segment is bolted to the flange of the UGS. Each

NUREG/CR-6048 22

PWR

ORNL4OWG 932895 EM

4- WGUIDE TUBE

Figure 2.17 CE PWR In-core Instrumentation support structure

23 N2R3G/CR 6048

3 Primary Stressors

In the context of aging studies, stressors are conditions thatwill promote the development and sustain the growth ofaging-related degradation mechanisms. Applied loads,environmental conditions, and manufacturing processescan Impose stressors on reactor internals.

Applied loading is an important stressor for internal com-ponents. The majority of the internals are not parts of thereactor coolant primary pressure boundary and are notsubjected to static pressure differential loadings. Flow-generated oscillatory hydrodynamic forces and preloads inbolts are the applied loadings of concern to reactorinternals.

The operating environment inside the pressure vessel alsoimposes many stressors on internal components. Reactorinternals are submerged in the reactor primary coolantlow. The coolant temperature in the core region, where

most internals are located, is -3160C (6000F), and theaverage coolant flow speed is -49 ms (16 fWs). The nomi-nal system pressure in the reactor vessel is -15.5 MPa(2250 psia). These conditions can generate many aging-related stressors. Normal reactor operating conditions forthe three types of PWRs may vary, but In general they donot deviate significantly from conditions mentioned above.

Because of their proximity to the core, some internals areexposed to stressors associated with fast neutron fluxes andthe heating of gamma radiation.

Accidents can impose much more severe thermal andmechanical loadings on reactor components. However, theemphasis of the present study is on effects of normal plantoperations, which constitute the great majority of the oper-ating history of the reactor.

3.1 Applied Loads

Thernal and mechanical loads are the major applied loadsacting on reactor internals during normal steady-state andtransient (startup or shutdown) operations. Thermal loadsare produced by temperature gradients in a component, bythermal expansions of different materials, and by restrictedthermal expansions. Mechanical loads are generated bystatic pressure differentials, preloads in bolts, and fluid-flow-generated cyclic forces.

Reactor internals are designed to accommodate thermalexpansions; as a result, constraint-induced thermal loads

are kept at a low level. The existence of temperature gradi-ents and rapidly changing temperatures in a component areimportant thermal stressors. They can lead to thermalcycling and fatigue crack initiation.

The two major sources of static applied loads are differen-tial pressure loads and preloads in bolts. Most internals arenot a part of the reactor primary pressure boundary and arenot subjected to large static differential pressure loads.Preloads in bolts can produce tensile stresses that mayreach the yield stress value of the bolt material. Tensilestresses of such a magnitude are considered as an essentialfactor to the development of stress corrosion cracking(SCC).1

The applied loads of primary concern to reactor internalsare flow-induced oscillatory hydrodynamic forces. Thevolume flow rate for PWRs ranges from 250 to 380 Us(4000 to 6000 gal/min), and the coolant is forced throughthe reactor vessel by reactor coolant pumps. Major sourcesof flow-related excitations are the pump-generated pressurepulsations at the pump rotational speed, blade passingfrequencies, and their harmonics. Pressure pulsations canact as periodic forcing functions on reactor internals, andtheir effects are most pronounced at the entrance regions ofthe vessel inlet nozzles.

Two types of hydrodynamic forces act on a blunt objectwhen it is placed in a flow stream. The first type is a staticload and is usually referred to as a drag force. The secondtype is a time-dependent force induced by flow separa-tions. The weights of the components and structural sup-ports are sufficient to counterbalance the drag forces. Thecyclic or periodic flow-generated hydrodynamic forcesmay force a structure to vibrate. Flow-induced vibrationscan lead to fatigue failures and mechanical wear.

A third source of flow-related stressors is the highly turbu-lent flow generated by the forced flow through gaps andother small openings. A structure located in the wake ofsuch high-intensity turbulent flows can undergo vibrationscaused by the time-dependent hydrodynamic forces. Baffleplate water-jetting is an example of such flow-inducedvibration problems.

3.2 Environmental Stressors

The primary environmental stressors for reactor internalsare related to the operating environment inside the reactorpressure vessel. They include contact with the primary

25 NURBGtCR-60-48

-

Pimary

reactor coolant flow and exposure to fast neutron fluxes.Long-term contacts with a high-temperatute fluid mediumand exposures to fast neutron fluxes may lead to physicalchanges and deterioration in some of the materials of con-struction for reactor internals. Neutron irradiation effectsmay cause embrittlement and irradiation-assisted SCC instainless steel components. Parts made from cast austeniticstainless steels (CASSs) are susceptible to thermal agingeffects.2

The coolant is borated water of high purity. The corrosive-ness of the coolant is determined by the quantity of dis-solved oxygen content, concentrations of impurities, andboric acid present in the flow stream. The dissolved oxy-gen is a product of the radiolytic reactions in the core, andthe process is generic to reactor operations. The hydrogenoverpressure system in the volume control tank of thechemical and volume control system adds hydrogen gasesto the flow stream, which act as scavengers and removemost of the dissolved oxygen. However, it is possible thatlocally high concentrations of dissolved oxygen and otherimpurities may exist in crevices in some internal compo-nents. PWR internals are susceptible to corrosion attacks.

Chlorides and fluorides are the two impurity componentsin the reactor primary cooling water that are of concern toreactor internals. They can be introduced into the flow sys-tem by condenser leakage and as impurities in the reactormake-up water. Impurities may be trapped in crevices, andtheir concentrations may reach such a level that corrosioncracks can be initiated in the affected components.

Boron, as boric acid dissolved in the cooling water, is thepreferred nuclear poison used for reactivity control inPWRs. Some reactor components, usually made of high-nickel alloys, are susceptible to corrosion attacks in boricacid solutions under a stagnant condition. Reactor internalsare submerged in a flowing fluid medium, and they areusually not sensitive to boric acid corrosion attacks.

An important environmental stressor for some internals isthe exposure to fast neutron (E > I MeV) fluxes. The neu-tron irradiation effects are most pronounced for compo-nents located in the immediate vicinity of the core.Prolonged bombardments by neutrons can change themechanical and physical properties of the materials.Specifically, they will increase yield and ultimate strengthsand reduce the uniform elongation to fracture and fracturetoughness. Irradiation effects can also lower the tempera-ture at which creep can become a significant deformationmechanism. Exposure to neutron fluxes can also lead to alowering of the threshold stress level that is considered asnecessary for the development of SCC.3 These changes

NUREG/CR-6048

could have potentially adverse effects in the structuralintegrity of reactor internals.

3.3 Manufacturing Stressors

The processes used to fabricate reactor components mayintroduce stressors to the finished parts. Welding, bolting,cold working, and casting are four common processes usedin the making of reactor internals. Stressors are associatedwith each of these processes.

Welding is a common method for attaching componentstogether to form an integral structural unit. Austeniticstainless steel such as type 304 may be sensitized in awelding process. The chromium depletion sensitizationprocess can make the finished products susceptible to SCC.Residual stresses in weldments, if not properly heatrelieved, may also contribute to the development of SCC.

Many reactor core support structures are joined together bybolts. Gaps and crevices in bolted joints can create a localenvironment that is conducive to the development of corro-sion attacks. Preloads in bolts and the resulting tensilestresses are stressors that can aid the SCC process.

Cold working is used in the making of some internal com-ponents. The component is formed to a predeterminedshape by a bending operation. Plastic strain accumulationand surface flaws are the stressors associated with a coldworking operation. They can lead to crack initiations andaccelerated crack growths.

Some stainless steel internals are cast in one piece. Whilecasting may eliminate many of the stressors associatedwith welding, bolting, and cold working, it is not astressor-free manufacturing operation. CASS componentsare susceptible to thermal aging effects.

Stressors imposed on reactor internals by applied loads,environmental conditions, and manufacturing processes arethe basic ingredients for the development of aging-relateddegradation mechanisms. Stressors that are present in asystem, whether acting independently or in conjunctionwith others, will initiate the aging degradation process.Degradation mechanisms develop at different rates, and,unless correction or preventive measures are taken, willeventually lead to failures in the affected components. Areview of the failure history of reactor internals can pro-vide useful information on the relative development ratesof potential aging-related degradation mechanisms.Understanding of the synergistic relationship between

26

stressors and aging degradations is also the basis for theformulation of strategies for managing aging effects.

References

1. R. L. Cowan and C. S. Tedmon, -Intragranular Cor-rosion of Iron-Nickel-Chromium Alloys," Advances inCorrosion Science and Technology, Vol. 3, M. S.Fontana and R. W. Staehle, Eds., Plenum Press, NewYork, 1973.

2. 0. K. Chopra and H. M. Chung, "Long-Term Aging ofCast Stainless Steel: Mechanisms and Resulting

Primary

Properties," in Proceedings of the Fifteenth WaterReactor Safety Research lqfornation Meeting,Gaithersburg, MD, USNRC Conference ProceedingsNUREG/CP-0090, October 1987*

3. A. J. Jacobs and G. P. Wozaldo, "Irradiation AssistedStress Corrosion Cracing as a Factor in NuclearPower Plant Aging:' J. Material Engineering 9(4)(1988).t

Avaiale for purchase from he National Technical Information Service,Springfield. VA 22161.

tAvailable in public tecnical libraries.

27 NUREG/CR-6048

4 Aging-Related Degradation Mechanisms

The development of an aging-related degradation mecha-nism is a time-dependent process. The process is initiatedwhen the necessary stressor or stressors are present in theoperating environment of the affected components. Oncestarted, an aging degradation mechanism will remainactive, and the deteriorating effects are cumulative. Theaffected components will eventually fail unless mitigatingmethods are used to remove the responsible stressors fromthe operating environment When it is not feasible to elimi-nate the stressors, then it is essential to replace thedegraded components before failure. One of the objectivesof an aging study is to Identify potential aging-relateddegradation mechanisms associated with the primary stres-sors of the system.

Aging-related degradation mechanisms generally associ-ated with the operating environment of reactor internals arecorrosion, fatigue, erosion, mechanical wear, embrittle-ment, creep, and stress relaxation. Corrosion includes gen-eral wastage and SCCs. Erosion is the loss of materialcaused by the flow of an abrasive and/or high-velocityfluid medium. Fatigue is vibration induced, and the respon-sible oscillatory forces can be either mechanical or thermalin nature. Vibration can also lead to mechanical wear andfretting. Embrittlement decreases the fracture toughness ofa material and is caused by thermal aging and/or irradiationeffects. Creep and stress relaxation may lead to changes inmaterial properties and structural deformation mechanisms.They are usually caused by prolonged exposure to hightemperatures and irradiation effects. The operating envi-ronment inside a reactor pressure vessel contains stressorsthat can activate all these potential aging-related degrada-tion mechanisms.

4.1 Corrosion

Corrosion is a term applied to a class of aging effects inwhich the structural integrity of a component is weakenedas a result of material deterioration caused by electro-chemical reactions with the surrounding medium. Theeffects can be highly localized, or they can cover a largeportion of the structure. The localized effects usually takethe form of crack initiation and development, while themore global effects are general corrosion and wastage.Operating conditions, corrodents, and alloy compositionswould determine the dominant corrosion mechanisms in aparticular situation.

4.1.1 General Corrosion and Wastage

When a structure is submerged in a corrosive medium, thesurface of the structure can become oxidized, and corrosion

products can be removed by the fluid. The corrosion pro-cess occurs more or less uniformly over the entire contactsurface. A moving fluid medium can increase the corrosionproduct removal rate. Austenitic stainless steels, specifi-cally type 304, have good resistance to general corrosionand wastage.1 General corrosion and wastage are not con-sidered as significant aging-related degradation mecha-nisms for reactor internals made of stainless steels.

4.1.2 Corrosion Cracking

Materials that provide good resistance to general corrosionand wastage, such as stainless steels, are often susceptibleto corrosion cracking. These failures are localized, andthey have the appearance of microscopic brittle fractures.There is no visual indication of the presence of corrosionproducts, and if the cracks are not detected, failures canoccur with little or no advance warnings. Most SCC mech-anisms require the presence of tensile stresses in the struc-ture as well as a corrosive medium. Other types of corro-sion crackings can develop without tensile stresses. As ageneral rule, the presence of tensile stresses may acceleratethe crack growth rate.

SCC is a major aging-related degradation mechanism forreactor internals. Crevices and Irradiation effects can assistthe SCC process.

4.1.2.1 SCC

It is generally accepted that the simultaneous presence ofthree conditions are necessary for the development of SCC:a susceptible material, a corrosive environment, and tensilestresses. Elimination of any one of the three conditions willstop the SCC process. Depending on the alloy composi-tions and corrodents involved, cracks can develop alongboundaries between grains; such failures are known asintergranular stress corrosion crackings (IGSCC). In othercases, cracks propagate along certain crystallographic slipplanes within the grains. These failures are referred to astransgranular stress corrosion crackings (TGSCC). Moreinformation on the fundamentals of SCCs can be found intexts by Logan2 and Romanov. 3

Austenitic stainless steels, such as type 304, have goodresistance to general corrosion and wastage, but they canbe made susceptible to SCCs by a sensitization process. 4

When a component made of type 304 stainless steel isheated or cooled slowly through the temperature range of482 to 8160C (900 to 15000F), carbon in the steel will pre-cipitate out as chromium carbide along grain boundaries.Chromium, which is a key alloying element for providing

29 NUREGICR-6048

Aging-Related

corrosion resistance, is depleted in regions adjacent tograin boundaries, and the chromium-depleted regions aresusceptible to corrosion attacks. Slow cooling after anneal-ings, prolonged stress-relieving operations, and weldingcan provide the temperature condition that is needed tosensitize stainless steels. For reactor internals made of type304 stainless steel, welding is a likely cause of the sensiti-zation process, and weld heat-affected zones (HAZ) arecommon locations for SCCs.

It should be emphasized that a susceptible material byitself, such as a sensitized stainless steel, cannot causeSCC. Other conditions must also be present in the systembefore SCC can occur.

The second requirement is a corrosive environment. Undernormal PWR operating conditions, the primary reactorcooling water contains small quantities of dissolved oxy-gen, chlorides, fluorides, and other impurities. The impuri-ties may contribute to the development of SCC when theyare trapped in crevices in internal components.

Dissolved oxygen is a product of radiolytic reactions in thecore. In steady-state PWR operations, the NRC StandardTechnical Specification stipulates that the dissolved oxy-gen content in the reactor primary cooling water be kept at<100 ppb. The plant primary cooling water chemistryrequirements, as stated in the Final Safety Analysis Report(FSAR), meet and exceed the Standard Technical Specifi-cation requirements for dissolved oxygen. The reactorchemical and volume control system, using hydrogen gasas a scavenger for the dissolved oxygen, can maintain adissolved oxygen content at <5 ppb in the bulk of the pri-mary cooling water system during steady-state power gen-eration. At this concentration level, dissolved oxygen is nota factor in the development of SCC. However, pockets ofhigh concentrations of dissolved oxygen may exist increvices and can contribute to the development of SCC. Inthe presence of impurities such as chlorides, SCC may beinitiated when the dissolved oxygen content is >40 ppb.5

SCC in an oxygenated water environment is intergranular.

The Standard Technical Specification for PWR primarywater chemistry also requires that the chloride and fluoridecontents be kept at <150 ppb. Chlorides are of special con-cern. Under favorable conditions, the presence of chloridesand dissolved oxygen in the reactor primary cooling watermay promote the development of SCC in austenitic stain-less steel components. 6 In general, the quantity of chlo-rides required for the development of SCC decreases withincreasing dissolved oxygen content. Experimental resultswith water in the range of 204 to 3160C (400 to 6000F)indicated that in the absence of dissolved oxygen, SCC will

NUREG/CR-6048

not occur even when the chloride content attains a level ashigh as 20,000 ppm. During steady-state PWR operations,the dissolved oxygen content in the bulk of the reactorcooling water is <5 ppb, and it is not likely that SCC candevelop under such a low level of dissolved oxygen.Crevices, which can trap and create high local impurityconcentrations, are needed to initiate SCC in PWRinternals.

Sulfide is another conrodent that can cause SCC. It is notpresent in the reactor cooling water. However, molyb-denum disulfide (MoS2 ), often used as a thread lubricant,can create locally high sulfide concentrations in reactorinternals with crevice conditions such as those that existedin bolted joints.7 MoS 2 reacts with borated water, and oneof the reaction products is hydrogen sulphide (H2 S), whichis highly corrosive. H2S can cause SCC in componentsmade of stainless steel. Reactor internals that use MoS2 asa lubricant are susceptible to sulfide-induced SCC.

Small quantities of fluoride are present in the reactor pri-mary cooling water, and they may induce SCC in stainlesssteel components. However, information on fluoride-induced SCC in PWR conditions is very limited.

The development of SCC in reactor internals is only one ofmany concerns that need to be taken into consideration inestablishing an optimum primary cooling water chemistryprogram for PWRs. Cracking problems in steam generatortubings and out-of-core radiation control are equallyimportant factors. More detailed discussions concerningPWR primary water chemistry control can be found inRefs. 8 and 9.

The third requirement for the development of SCC is thepresence of tensile stresses in a structural component. Inthe absence of significant irradiation effects, the magnitudeof the tensile stresses must exceed a threshold value beforeSCC can occur. A generally accepted value for the thresh-old stress is the yield stress of the material of construction.Bolt preloads and weld residual stresses can generate thenecessary tensile stresses in a reactor internal component.

Based on this information, it is reasonable to suggest thatcrevices in PWR internals are likely locations for thedevelopment of SCC. A "crevice condition" is a generalterm that is used to describe a small region in which highconcentrations of corrodents may be trapped. Crevices arecreated by small holes, surface deposits, narrow gaps ingasket surfaces, lap, and bolted joints. The small region isusually filled with a stagnant liquid, and a differential aera-tion cell is established within the stagnation region. The

30

differential aeration cell can create locally high concentra-tions of anionic species such as chlorides in the crevice.The presence of bolt tensile stresses and dissolved oxygenwould supply the necessary conditions for the developmentof SCC. Most crevice-assisted SCC is intergranular.

In addition to the crevice condition, exposures to fast neu-tron fluxes can also assist the SCC process. SCC has beenobserved in reactor components made of nonsensitizedstainless steels. In most cases, stresses in these componentsare not high, and they are much lower than the yield stressof the material. A common factor in these nonsensitizedstainless steel components is the exposure to high energyor fast neutron fluxes. These observations seem to suggesta new mechanism for the development of SCC. This formof SCC is known as irradiation-assisted SCC (IASCC);IASCC is intergranular.

Basic understanding of IASCC is not complete. Manytheories have been proposed, but no one single theory cansatisfactorily account for the iradiation effects on SCC.General agreement is that a threshold neutron fluence levelexists below which IASCC is not likely to occur. The bestestimate of the threshold neutron fluence levels for stain-less steel is -5 x 102° neutrons/cm 2 (E > 1 MeV). Theexpected lifetime neutron fluence levels for internalslocated in close proximity to the core, such as the core baf-fle, in-core monitor housings, and the core support plate,can exceed the threshold value, and such components aresusceptible to IASCC.

One of the proposed theories suggested that the materialmay be weakened by the formation of bubbles in the solidcaused by a transmutation process. Austenitic stainlesssteels contain trace quantities of boron and can react withthermal neutrons to form lithium and helium.9 Hydrogengases are also produced by transmutation reactions involv-ing fast neutrons and elements such as nitrogen, nickel,iron, and chromium.8 These insoluble gases precipate fromthe solid and have a tendency to migrate to dislocations orto grain boundaries.9 Gas bubbles are discontinuities, andthey can weaken the structural material. The solid may dis-integrate if the bubbles are fused together, even when nosignificant stresses are acting on the system. Basic under-standings on bubble sizes, their weakening effects, and thedriving forces behind bubble movements are not complete.

The presence of impurities such as phosphorus (P) and sili-cone (Si) increases the susceptibility of unirradiated stain-less steels to IGSCC. This seems to suggest that grainboundary segregation of impurities could be the responsi-ble mechanism for IASCC. Therefore, high-purity stainlesssteels, with lower impurity contents, may be less suscepti-

Aging-Relatedble to IASCC. However, there is not sufficient informationto correlate grain boundary impurity concentrations withneutron fluence level and IASCC susceptibility. Testingresults are inconclusive; some high-purity stainless steelsare more resistant to IASCC under PWR operating condi-tions while others (including type 304) performed no betterthan commercial-grade stainless steels.

There is no agreement on the primary mechanism forIASCC, and there are many active research works in thefield. It is generally accepted that IASCC would require theexcedence of a threshold neutron fluence level. Also, thethreshold tensile stress value for IASCC is much lowerthan the yield stress of the material. Stresses incurred dur-ing manufacturing and handling operations may be suffi-cient for the development of IASCC. It is also acceptedthat the critical flaw size decreases with increasing neutronfluence level. The reduction of the critical flaw size canaccelerate crack growth rates.

The operating environment inside the pressure vessel of aPWR can produce conditions that are favorable to thedevelopment of SCC. SCC, in one form or another, isexpected to be a major aging-related degradation mecha-nism for reactor internals.

4.12.2 Pitting

Pitting corrosion is a localized phenomenon. Holes or pitsare etched into the metal surfaces and are filled with corro-sion products. The pits are created by crevices or other fab-rication flaws in the form of small cuts and nicks. They actas differential aeration cells in which the pit is an anodeand the surrounding surface is a cathode.I The anodic reac-tion will create locally high concentrations of anionicspecies such as chlorides and sulfides in the pits. The cor-rosion process is the result of high concentrations ofaggressive corrodents. Pitting can occur without the pres-ence of significant stresses in the cornponent. A stagnantcondition will promote the growth of pitting corrosion.

Pitting is a concern for reactor internals when they are instorage during extended plant outages. Unless the reactorhas a history of long outages, pitting is not considered as aprimary aging-related degradation mechanism for reactorinternals.

4.2 Fatigue

In addition to static loads, reactor internals are subjected todynamic or time-dependent forces. A structure willundergo some form of vibratory motions as a response todynamic loads. Vibrations can lead to crack initiation and

NUREGICR-604831

Aging-Relatedsubsequent crack growth. Structural failures caused byvibrations are classified as fatigue failures. Fatigue failuresare further divided into two types: low- and high-cyclefatigues. The determining factor is the number of vibrationcycles that a component experienced before the develop-ment of a crack. The counting process begins with theonset of vibrations and is terminated with the initiation of afatigue crack in the structure. There is not a precise divid-ing line separating low- and high-cycle fatigue failures.Generally speaking, when a crack is initiated between 103and 104 cycles, the failure is considered as a low-cyclefatigue failure. Plastic strain accumulations are associatedwith low-cycle fatigue failures. High-cycle fatigue failuresare characterized by large numbers of cycles and elasticstresses. There are usually little or no plastic strain accu-mulations in a high-cycle fatigue failure.

Fatigue failures are environment dependent Understandingof the interactions between fatigue and environmental fac-tors such as corrosion is limited. Most fatigue analyses areperformed without inputs from environmental conditions.Fatigue design curves, such as those provided by the ASMEB&PV Code,10 are developed for dry air where the envi-ronmental corrosion effects are insignificant.

Some dynamic loads, such as those caused by hydraulictransients and seismic excitations, are treated as parts ofthe design loadings for the reactor system. Stresses, strains,and the dominant frequencies associated with the structuralresponse to these events can be calculated, and their con-tributions are added to the fatigue life estimation for theaffected components.

The primary cause of fatigue failures in reactor internals isflow-induced vibrations. The amplitude of the vibrations isdetermined by the amount of damping in the system aswell as the closeness between he structural natural fre-quency and the dominant excitation frequency. When thesefrequencies are well separated, the amplitude of the vibra-tions is small. Large-amplitude resonant vibrations maydevelop when the input excitation frequency is close to thefundamental natural frequency of the structure. Dampingcan reduce the amplitude of such vibrations. The develop-ment of large-amplitude resonant vibrations in a reactorcomponent will quickly lead to a low-cycle fatigue failure.When they are detected during reactor preoperational test-ings, corrective actions will be implemented to eliminatesuch large-amplitude vibrations. Therefore, in an agingassessment study, high-cycle fatigue failures caused bysmall-amplitude vibrations is the more important and rele-vant aging-related degradation mechanism.

4.3 Erosion

Erosion is the removal of materials from a metal surfacewhen it is submerged in a moving fluid medium.I The pro-cess is best illustrated by the sandblasting operation inwhich a gas stream laden with fine solid particles literallygrinds away a metal surface. Reactor cooling water is fil-tered, and the quantity of suspended solid particles in theflow stream is kept at a very low level (<1 ppm). Erosioncaused by the actions of suspended solid particles is notexpected to be a major aging-related degradation mecha-nism for PWR internals.

Cavitation can also cause erosion on a metal surface sub-merged in a liquid flow stream. In regions where the flowspeed is high, the local static pressure may fall below thevapor pressure of the liquid, and a phase change takesplace. Bubbles are formed and can become attached to themetal surface causing erosion problems. There is no phasechange in the primary reactor coolant in normal PWRoperations, and the flow speed in the core region is not suf-ficient to produce cavitations. Cavitation-induced erosionis not expected to be an aging issue for PWR internals.

4.4 Mechanical Wear

Mechanical wear is the loss of material as a result of rela-tive motions between two contact surfaces. The presence ofa corrosive medium may accelerate the material loss rate.Most mechanical wear problems in reactor internals arevibration-induced.

Mechanical wear can also develop in components that areheld fixed initially. Examples are wears observed inflanges and bolted joints. Preloads are applied to theseparts to prevent slippage between adjacent contact sur-faces. Vibrations and creep can reduce the forces holdingthese parts together, and slippage can occur. The resultingrelative motions can cause mechanical wear.

Mechanical wear can be minimized by hardening the con-tact surfaces or by the use of special wear pads. Periodictightening of bolts in bolted joints can prevent mechanicalwear caused by slippage.

4.5 Embrittlement

Embrittlement is the loss of ductility and fracture tough-ness in a material. It is usually accompanied by increases

NUREG/CR-6048 32

in the material yield and ultimate strengths. The processcan transform a normally ductile material to one that issusceptible to brittle fractures. Prolonged exposures to fastneutron fluxes and thermal aging can cause embrittlementin reactor components made of stainless steels.

4.5.1 Radiation Embrittlement

Effects of radiation embrittlement are controlled by severalparameters; the more important ones are neutron fluence,irradiation temperature, and material compositions. Theexperimental data base is obtained from testings usingsurveillance specimens from commercial power reactorsand other test reactor experiments. Reductions in uniformelongations are used as indications of the decrease in duc-tility in the material after exposure to fast neutron fluxes.Changes in fracture toughness can be deduced from resultsof Charpy V-notch impact testings as well as other types offracture-toughness testings.

At a temperature of -316°C (6000F), testing results1 1 withtype 304 stainless steel indicate that the effects of radiationembrittlement are becoming noticeable at a neutron fluencelevel of -S x 1020 neutrons/cm2. The estimated neutronfluence for reactor internals is determined by the distanceof the component from the core. In 40 years of poweroperations, for those components located in close proxim-ity to the core, the expected maximum neutron fluencet2 is-I x 1022 neutrons/cm 2. Such reactor internal componentsare susceptible to the effects of radiation embrittlement.

At specified neutron fluence levels and using annealed type304 stainless steel, experimental resultst3 showed that thematerial hardens and there is a corresponding decrease inductility, as indicated by a reduction in uniform elongation.The experiments were conducted at temperatures rangingfrm room temperature to -7600C (1400'F') and at neutronfluence levels (E > 1 MeV) up to 6 x 1021 neutrons/cm2.At 299°C (5700F) the unirradiated uniform elongation hasa value of -38%. The uniform elongation at a neutron flu-ence level of 1.5 x 1020 neutrons/cm 2 is -22% at the sametemperature. At -1.5 x 1021 neutrons/cm2, the uniformelongation dropped to -0.5%. The decrease seemed tolevel off at higher neutron fluence levels. Assuming theseresults are representative values for 304 stainless steels, itcan be speculated that when the expected neutron fluencelevel does not exceed 1.5 x 1020 neutrons/cm 2, reactorinternals should retain sufficient ductility that brittle frac-ture can be prevented. When the neutron fluence level foran internal component approaches the maximum value ofI x 1022 neutrons/cm2, the decrease in ductility wouldmake a component susceptible to brittle fractures (reason-able combination of critical fracture toughness, stress, andflaw size). The core baffle, in-core instrumentation guide

Aging-Relatedtubes, and possibly the core support plates are internalcomponents that can experience neutron fluence levelsclose to the maximum value,12 and they are vulnerable toradiation embrittlement effects.

4.5.2 Thermal Embrittlement

Thermal embrittlement is an aging-related degradationmechanism for reactor components made of cast austeniticstainless steels14 (CASS). CASS is a two-phase alloy con-sisting of austenite and ferrite. The austenite is ductile, andits ductility is not affected by thermal exposure. The ferritecan become embrittled by thermal exposures, and its con-tent is the controlling factor for thermal aging effects inCASS. When the ferrite content is <20%, the thermalembrittlement of CASS components is low.

When the ferrite content is >25%, CASS components canbecome embrittled when they are aged in the PWR operat-ing temperature range of 280 to 3200 C (535 to 6100F).Thermal-aging effects under these conditions can lead to abrittle cleavage fracture of the ferrite. Failures can alsotake the form of a separation of the ferrite/austenite phaseboundary, which provides a convient crack propagationpath. In either case, the fracture toughness of the materialis reduced, and CASS components are susceptible to brittlefractures.

4.6 Creep and Stress Relaxation

Creep and stress relaxations can cause a load-bearingstructure to lose its structural integrity when it is exposedto high temperatures for an extended period of time. Expo-sures to fast neutron fluxes can lower the temperature inwhich creep and stress relaxation can become significantdeformation mechanisms.

Creep is the progressive deformation of a structure under aconstant internal stress. Stress relaxation is the reduction ofinternal stresses in a structure with a constant deformation.Creep and stress relaxation are caused by the same mecha-nism, but they differ in the constraints imposed on thestructure. Creep can lead to brittle fractures in structuralcomponents that can undergo deformations. Stress relaxa-tion is a concern for constrained structures such as boltedjoints, where deformations are held fixed. The decrease instresses in a bolted joint can cause the joint to lose its ini-tial tightness, and leakage and slippage may occur. Thedevelopment of creep and stress relaxation is influenced bythe operating temperature and neutron fluence level.

33 33 ~~~~~NUREG/CR-6048

Aging-Related

Creep and stress relaxation are considered as significantstructural deformation mechanisms when the operatingtemperature is higher than half of the melting-point tem-perature of the material. The melting points for austeniticstainless steels are -14270 C (26000F). The normal operat-ing temperature range for PWR internals is between 280and 3200C (535 and 6100F), and it is much below 714 0C(13000F), which is half the melting-point temperature forstainless steels. The creep analysis temperature limit foraustenitic stainless steels, as specified in the ASME B&PVCode, is 4270C (8000F). Creep analyses are not requiredfor reactor components operating at the normal PWR oper-ating temperature. When the neutron fluence level is low,thermally induced creep and stress relaxation are not con-sidered as major aging-related degradation mechanisms forPWR internals.

In-pile experimental results at -288 0C (550F) indicatedthat stress relaxation can occur in bolts made of type 304stainless steelsl at a neutron fluence level (E > 1 MeV) of-6 x 1019 neutrons/cm2. At the temperature range from 60to 3160C (140 to 6000F), significant stress relaxation hasbeen observed in type 304 stainless steels at neutron flu-ence levels >5 x 1020 neutrons/cm2. The expected lifetimeneutron fluence levels for reactor internals located near thecore can exceed this value. The core baffle, in-core instru-mentation guide tubes, and core support plates are suscep-tible to the aging effects of irradiation-assisted creep andstress relaxation.

4.7 PWR Internals and PotentialAging-Degradation Mechanisms

A reactor internal component is exposed to many stressors,and it is subjected to the effects of more than one aging-related degradation mechanism. Aging-related degradationmechanisms develop at different rates. In most cases, adominant degradation mechanism emerges and wouldeventually cause a failure in the affected component.Actual operating conditions may favor the development ofcertain aging-related degradation mechanisms. As anexample, it is well understood that impurity contents andtensile stresses have a strong influence in the developmentof SCC in stainless steel components. In PWR operatingconditions, crevices are needed to trap high-impurityconcentrations. Preloads in bolts can produce internaltensile stresses that are close to the yield stress of the bolt

material. The combination of crevice conditions and highbolt tensile stresses indicates that PWR internals withbolted joints and other tight-fit connections are probablelocations for the development of SCC.

The location of an internal component can have someinfluence in the development of a dominant aging-relateddegradation mechanism. A major stressor for reactor inter-nals is flow-generated excitations, which may take theform of pump-generated pressure pulsations, vortex shed-dings, and high-intensity turbulent flows. Pump-generatedpressure pulsations are at their strongest level in regionsaround inlet nozzles. Reactor internals located in theseregions, such as core support shields, thermal shields, sur-veillance specimen holder tubes, and core barrels, are sus-ceptible to flow-induced pressure pulse excitations. As thecoolant flow travels deeper into the core region, fluiddamping and friction losses reduce the intensity of thepressure pulsations, and they would become less of a factorin exciting internal components.

Reactor internals located close to tube-bank-like structures,such as in-core instrumentation guide tubes and shrouds,are susceptible to the effects of cross-flow generated vortexsheddings. Flow-induced vibration is a major stressor forthese components.

The effects of fast neutron fluxes are most pronounced inregions around the core. Reactor internals such as core baf-fle, core barrel, thermal shield, surveillance specimenholder tubes, core support plates, and in-core instrumenta-tion guide tubes are susceptible to irradiation-assisted SCCand radiation-induced embrittlement.

Reported aging-related failure information for PWR inter-nals is needed to assess the relative importance of the dif-ferent aging-related degradation mechanisms associatedwith the many stressors that may exist inside a reactorpressure vessel.

Primary stressors and associated aging-related degradationmechanisms for PWR internals are summarized inTable 4.1.

NUREG/CR-6048 334

Aging-Related

Table 4.1 PWR Internals primary stressors and aging-related degradation rnechanisms

Stressor Source Aging-related degradationmechanisms

Oscillatory hydrodynamic forces Reactor coolant flow Fatigue and mechanical wear(dynamic stresses)

Preloads in bolts (static stresses) External applied loads Contributing factor to SCCa(torquing during installation)

Residual stresses (static stresses) Welding Contributing factor to SCC

Thermal stresses (static and dynamic) Reactor coolant and gamma heating Fatigue, contributing factor to SCC

Dissolved oxygen and other impurities Reactor coolant Contributing factor to SCC

Local concentration of corrodents Crevice condition in reactor coolant Contributing factor to SCCflow

Chromium depletion in grain Welding Enhance potential for SCCboundaries (sensitization)

High operating temperature Reactor coolant Thermal embrittlement in CASS[-3160C (600*F)J components

Exposure to fast neutron fluxes Nuclear reaction in the core Irradiation-assisted SCC andembrittlement, irradiation-enhancedcreep and stress relaxationb

4SCC is most likely t occur as a result of high tensile stresses, local concentration of orrodents (crevices), and a susceptible material.bradition effects can make creep and ses relaxation major degradation mechanisms at reactor eperating temperature 3160C (6000F)].

References Type 304 Stainless Steel:' Corrosion 35, 288(197 9).t

1. "Component Life Estimation: LWR Structural Mate-rials Degradation Mechanisms:' EPRI NP-5461, aninterim report prepared by Structural Integrity Asso-ciates, Inc., for the Electric Power Research Institute,Palo Alto, Calif, September 1987.*

2. H. L Logan, The Stress Corrosion of Metals, JohnWlley & Sons, New York, 1966.

3. V. V. Romanov, "Stress Corrosion Cracing ofMetals," the Israel Program for Scientific Transla-tions and the National Science Foundation,Washington, D.C., 1961.

4. R. L. Cowan and C. S. Tedmon, "Intergranular Cor-rosion of Iron-Nickel-Chromium Alloys," Advancesin Corrosion Science and Technology, Vol. 3, M. S.Fontana and R. W. Stachle, Eds., Plenum Press, NewYork, 1973.

5. M. E. Indig and A. R. Mcllree, "High TemperatureElectrochemical Studies of the Sress Corrosion of

6. B. M. Gordon, "The Effects of Chloride and Oxygenon the Stress Corrosion Cracking of Stainless Steels:Review of Literature," NACE 36(8) (1980).1

7. "Degradation and Failure of Bolting in NuclearPower Plants" Vol. 1, EPRI NP-5769, a final reportprepared by Applied Science & Technology for theElectric Power Research Institute, Palo Alto, Calif,1988.

8. A. J. Jacobs and 0. P. Wozadlo, "Irradiation-AssistedStress Corrosion Cradcng as a Factor in NuclearPower Plant Aging," . Materials Engineerig 9(4)(1988).1

9. 1. Gittus, Irradiation Effects In Crystalline Solids,Applied Science Publishers, Ltd., London, 1978.

10. American Society of Mechanical Engineers, ASMEBoier and Pressure Vessel Code, Section I, NuclearPower Plant Components, Div. 1.

NUREGCR-6D4835

Aging-Related11. R. E. Robbins, J. J. Holmes, and J. E. Irvin, "Post

Irradiation Tensile Properties of Annealed and Cold-Worked AISI-304 Stainless Steel," Trans. AmericanNuclear Society, November 1967.t

12. V. S. Shah and P. E. MacDonald, Eds., "ResidualLife Assessment of Major Light Water ReactorComponents-Overview:' USNRC Report NUREG/CR-4731, November 1989.*

13. 0. K. Chopra and H. M. Chung, "Long-Term Aging ofCast Stainless Steel: Mechanisms and Resulting Prop-erties," in Proc. of the Fifteenth Water Reactor Safety

Research Information Meeting, Gaithersburg, MiL,USNRC Conference Proceeding NUREG/CR-0090,October 1987.*

14. 0. K. Chopra, Argonne National Laboratory, "Estima-tion of Fracture Toughness of Cast Stainless SteelsDuring Thermal Aging in LWR Systems," USNRCReport NUREGICR-4513, June 199.t

Available fr purchase from National Technical Infornation Service,Springfield, VA 22161.

tAvailable in public technical libraries.

NUREG/CR-6048 36

S Survey of Aging-Related Failures

Reactor Internals are subjected to stressors that can activatemany potential aging-related degradation mechanisms.Aging-related component failure information is needed toproperly identify the more Important aging-related degra-dation mechanisms. The identification of major aging-related degradation mechanisms may provide informationthat can be used to formulate strategies for controlling andmanaging aging effects.

The reactor component failure information is compiledfrom results of the plant in-service inspection (ISI) pro-grams. The program specifies requirements and proceduresfor inspecting reactor systems or components to ensuresafe plant operations. The requirements and proceduresmay include orders, rules, criteria, and guidelines estab-lished by regulatory or licensing agencies as well as indus-try standards developed by technical organizations. ISIprograms for U.S. commercial nuclear power plants areestablished under the rules and regulations of Sect. XI ofthe ASME Boiler and Pressure Vessel (B&PV) Code.I Theinspection of some reactor internals is included as a part ofthe plant ISI program.

5.1 ISI Program for Reactor Internals

The plant ISI program requires visual inspections for ac-cessible areas of reactor internals. A complete inspectioncycle is 10 years, and selected internal components are in-spected at refueling outages.

Section XI of the ASME B&PV Code specifies threclasses of visual inspections: VT-I, VT-2, and VT-3. AVT-i visual examination is conducted to determine thecondition of the part, component, or surface examined, in-cluding such conditions as cracks, wear, corrosion, erosion,or physical damage on the surface of the part or compo-nent. The examination can be performed either directly orremotely. A VT-2 examination is used to detect leakagefrom pressure-retaining components. Most reactor internalsare not pressure-retaining components, and VT-2 inspec-tions are seldom used on internals. A VT-3 examination isconducted to determine the general mechanical and struc-tural conditions of components and their supports, such asverification of clearances, settings, physical displacements,loose or missing parts, debris, corrosion, wear, erosion, orthe loss of integrity at bolted or welded connections. VT-3inspections can be performed either directly or remotely.

During a refueling outage, some internal components areremoved from the pressure vessel and stored in a pool.VT-3 examinations are performed on the accessible areasof these components by remote television cameras under

proper lighting. Internals that remain in the pressure vesselare also inspected by renote television cameras. Note thatthe ISI plan only calls for the inspection of accessible areasof the internals. Details of the inspecting procedures can befound in SecL XI of the ASME B&PV Code.

When a failure is detected, information concerning thefailure such as its location and suspected causes isrecorded. The information is sent to NRC for inclusion inthe Licensee Event Report (LER). The LER may also con-tain information on corrective actions taken to repair thefailure or damages. For these reasons, LERs are consideredas reliable sources for reactor systems or components fail-ure information.

5.2 Failure Information Summary

The majority of PWRs in the United States started com-mercial operations in the 1970s, and since that time therewere many reported aging-related failures of reactor inter-nals. Most of the reported failures can be attributed to threeaging-related degradation mechanisms: (I) fatigue,(2) SCC, and (3) mechanical wear.

Most of the reported aging-related failure cases involveddomestic PWRs. Many overseas PWRs of theWestinghouse design have experienced aging-related fail-ures similar to those of the domestic units. However, theonly overseas case discussed in details in this report is theSCC failures in core baffle bolts detected in some Germanand Swiss PWRs. This failure case is included because ofits safety significance.

The following is a brief summary of the reported failurecases for each of the three major aging-related degradationmechanisms.

5.2.1 Fatigue

Most fatigue failures in PWR internals are caused by flow-induced vibrations (FIVs). Important excitation sourcesinclude pump-generated pressure pulsations, cross-flowvortex sheddings, and highly turbulent flows. There is usu-ally a dominant frequency associated with these distur-bances. Large-amplitude vibrations can develop when thestructural natural frequency matches one of these excitationfrequencies. The component will be subjected to small-amplitude vibrations when the frequencies are well sepa-rated.

37 NUREG/CR-6048

Survey

5.2.1.1 WE Thermal Shield

The failure of thermal shields in older WE PWRs havebeen attributed to FIV.2 Thermal shields are located nearinlet nozzles and are subjected to pump-generated pressurepulsations. Most of these failed thermal shields were of thesegmented-shell design with shell segments (usually three)keyed to support lugs at the bottom of the reactor vessel.The top end was either free or was equipped with radialspacer pins. The support condition can be characterized asa cantilevered support with limited displacements at thefree end. The shell segments were fastened together byvertical pins at the intersections. Coolant flow-induced ex-citations caused the assembled shield to vibrate in a shellmode. During vibrations, some of the shell segments cameinto contact with the core barrel. In these older reactors, thecore barrels were bolted together, and the repeated impactloadings caused failures in the core barrel support bolts.The impact loadings also damaged the thermal shield.

One of the reported failure cases involved a one-piececylindrical thermal shield and occurred during the reactorfunctional hot testing. The shield was clamped to the corebarrel at the bottom. The top was free except for the pres-ence of radial limiter pins, which fit into a keyway in thecore barrel. The pins were shrunk-fit into the thermalshield and lock welded in place by light fillet welds. FIVcaused the pins to come into contact with the sides of thekeyway in the core barrel, and repeated impacts eventuallyled to the cracking of the fillet welds that locked the pins tothe thermal shield. After the failure was detected, the deci-sion was made to replace the pin-keyway system by aflexure support system.

There was also one reported case of failure involving aone-piece cylindrical thermal shield with a flexure supportsystem. The top-mounted flexure support system failed.The cause was attributed to high-cycle fatigue caused bysmall-amplitude FIV of the thermal shield.

There are no reported failures in thermal shields using theneutron-shield pad design.

5.2.1.2 CE Thermal Shield

CE thermal shields are located near inlet nozzles where ef-fects of pump-generated pressure pulsations are strong.Two CE units equipped with thermal shields reportedproblems with their support system.2 ISIs during refuelingoutages revealed missing support and positioning pins. Allremaining pins showed signs of excessive wear or damage.Lugs welded to the core barrel to support the top end of thethermal shield were also damaged. In one unit the damaged

NUREG/CR-6048

lugs caused a through-the-wall crack in the core barreL Nodamage to the reactor vessel was observed.

The damaged thermal shields were removed. Analysis in-dicated that removal of the thermal shield did not lead toany significant changes in the core thermal-hydraulic oper-ating conditions. The removal of the thermal shields is notexpected to have any undesirable effect on the reactors dur-ing their remaining design life.

5.2.1.3 CE Hold-Down Ring

During inspection of a maintenance outage, excessive me-chanical wear was observed in the hold-down ring of a CEunit.2 Failure was attributed to FIV caused by insufficienthold-down spring force. The hold-down ring was replaced.The new ring was fabricated with type 403 stainless steelinstead of type 304, which was used to make the old hold-down ring. Additional hold-down spring force was appliedduring the installation of the new hold-down ring. The in-ternal hold-down spring force was also increased for otherCE reactors. There were no other reported hold-down ringproblems.

5.2.1.4 WE BaMe Plate Water-Jetting

Fuel rod damage caused by baffle plate water-jetting hasbeen reported in a number of WE reactors. 2 '3 The corebaffle outlines the boundary of the core. A bypass flow isestablished in the region between the core baffle and thecore barrel, and it is used to provide more effective coolingto the core barrel. In some of the older WE reactors, thebypass flow is introduced into the region between the corebaffle and the core barrel by holes located in the upper corebarrel. The bypass flow moves in a downward directionthrough holes in the horizontal former plates, and it isturned around at the bottom of the core barrel and thenmerged with the main flow going through the core. Whenthe bypass flow is in a downward direction, as illustrated inFig. 5.1, the bypass flow pressure is higher than that of themain coolant flow in the core. A pressure differential is es-tablished between the bypass flow and the core, and it willpush the coolant into the core if gaps exist between thevertical baffle plates. The jetlike leakage flow will impingeon fuel rods in the vicinity of the gaps and set the rods intowhirling motions and vibrations. Excessive fuel rod mo-tions will eventually lead to cladding degradations andfailures.

Two types of baffle plate water-jet impingement patternshave been observed and are illustrated in Fig. 5.2. The jetfrom a center-injection joint impinges directly on a fuelrod, while that from a corner-injection joint will have amore sideways, impact. They can both set the affected fuelrods into vibrations and whirling motions.

38

Survey

ORNL-OWG 93-2896 ETM

CORE PLATE

CORE PLATE

Figure 5.1 Westinghouse PWR downward bypass flow scheme

Several remedies have been tried to reduce the severity ofthe baffle plate water-jetting impingement problem. Oneremedy that has been tried is the peening of an entire jointto reduce the gap width between the vertical plates.Subsequent inspections revealed that peening the jointswas not effective because damaged fuel rods were foundaround the peened Joint. There was also evidence to sug-gest that peening a center-injection joint will have the ef-fect of enlarging the gap width in nearby coner-injectionjoints. A more effective remedy is to replace the fuel rodsin the water-jet impingement region with solid stainlesssteel rods of the same diameter. The insertion of partialgrids, which will serve as midspan supports for fuel rods inthe impingement region, has also been effective in reduc-ing fuel rod damage. However, uses of solid stainless steelrods and partial grids are not considered as the permanentsolution to the baffle plate water-jeting problems. A moreeffective solution Is to reduce or eliminate the driving forcebehind the water-jetting flow.

The bypass flow rate is small when compared with themain coolant flow rate through the core, and the coolant

pressure distribution in the core can be regarded as con-stant during normal reactor operations. The variable thatcan be changed to affect the pressure differential betweenthe bypass flow region and the core is the pressure of thebypass flow. A decrease in the bypass flow pressure re-duces the pressure differential and the driving force behindthe water-jetting flow. The bypass flow pressure can belowered by increasing the pressure loss in the main coolantflow before the diversion of a small portion of the mainflow into the bypass flow region. This is accomplished byplugging the inlet holes near the top of the lower core bar-rel. The pressure in the bypass flow is reduced because ofthe added pressure loss in the flow down the annulardowncomer region between the core barrel and the vesselwall. The bypass flow is then diverted into the core baffle-core barrel region through the lower core plate. In themodified flow scheme, the bypass flow is in an upwarddirection. Ideally, pressures in the core and the bypass flowregion should be balanced, and there will be no drivingforce to produce jetting flow through gaps. In practice, abalanced pressure system is difficult to attain, and theseverity of the water-jetting problem is reduced by lower-ing the pressure differential between the bypass flow and

39 NUREGICR-6048

Survey

ORNL-DM 93-2897 ETM

COOLANT FLOW THROUGHJOINT ENLARGED GAPI

Pi

P3 000000

FUEL RO pi), Pa

Figure 5.2 Westinghouse PWR baffle water-jet Impingement patterns

the core. The effectiveness of the modified flow schemecan be further improved by designing the flow system insuch a manner that the core flow pressure is higher than thebypass flow pressure. In this situation, potential water-jet-ting flows will be from the core into the bypass flow regionand away from the fuel rods.

One WE unit that has a history of baffle plate water-jettingproblems has modified its downward bypass flow schemeinto an upward bypass flow scheme. Other plants with adownward bypass flow have also developed plans to con-vert to an upward bypass flow scheme.

In the new WE reactor design, the bypass flow enters thecore baffle-core barrel region from the bottom region ofthe core barrel, and the bypass flow travels upward throughholes in the horizontal fonner plates.

Discussions of the baffle plate water-jet impingementproblem in WE reactors are based on information providedby NRC E Information Notice No. 82-27.3 There are noreported baffle plate waler-jetting impingement problemsin CE and B&W PWRs.

5.2.2 SCC

A crevice condition is a common feature in the develop-ment of SCC in PWR internals. Bolted joints and tightly

fitted connections are likely sites for SCC attacks. Mostobserved SCC is intergranular. Internals with creviceconditions that are also exposed to high-energy neutronfluxes may also be susceptible to irradiation-assisted SCC

5.2.2.1 B&W Thermal Shield Support Bolts

The B&W thermal shield is a cylindrical structure with itsupper end bolted to the upper flange of the core barrel. Thelower end of the shield is shrunk-fit into the upper flangeof the forged cylinder in the lower grid assembly and se-cured by 96 high-strength bolts. B&W thermal shield sup-port bolts were made of a nickel alloy stainless steel (gradeA-286).

ISIs in several units revealed missing bolts from the ther-mal shield lower end support joint. 2 The majority (-80%)of the remaining bolts were loose, and several bolt lockingcups were also missing. Cracks were also detected in boltsat the shield upper end and in the SSHT mounted on theoutside wall of the thermal shield. The more serious fail-ures were bolts at the lower end of the shield.

Failures were attributed to IGSCC at the bolt-head-to-bolt-shank transition. The replacement bolts were designed toreduce the tensile stress level in the bolt, and this was ac-complished by redesigning the shank region, peening thesurface of the bolt, and reducing the preload used to installthe bolts. The material of construction also was changedfrom A-286 stainless steel to Inconel X-750.

NUREG/CR-6048 40

A review of LERs did not show any reported failure of thenew thermal shield support bolts.

5.22.2 B&W Core Barrel-to-Core Support ShieldBolts

The core barrel and core support shield are flanged cylin-drical structures. The upper flange of the core supportshield rests on a circumferential ledge of the vessel closureflange. The lower flange of the core support shield isbolted to the top flange of the core barrel. The bolts weremade of a nickel alloy stainless steel (grade A-286).

Ultrasonic (L inspections in two B&W plants showedindications of cracking in a number of the bolts joining thecore barrel to the core support shield. The results wereverified when the bolt heads became separated from thebolt shanks when the locking clips were removed. Failureswere attributed to IGSCC. Visual inspections failed to de-tect these failures.

The cracked bolts were replaced by bolts made of the samematerial, grade A-286 stainless steel. The new bolts weremade by machining, while the old bolts were made by ahot-headed operation. The torque applied to the new boltswill be significantly reduced. A review of LERs showed noreported failure of the new core barrel-to-core supportshield bolt.

522.3 WE Control Rod Guide Tube Support Pins

The guide tube assembly is a part of the upper core supportstructure, and it houses the control rod drive shaft and theRCCAs. The assembly is a tubular structure divided into

Survey

two parts: the upper part, called the control rod shroudtube, and the lower part, the control rod guide tube. Theupper support plate is the dividing boundary. The top ofthe control rod guide tube is fastened to the bottom of theupper support plate, and the lower end of the tube is held inplace by split pins inserted into the lower core plate.

Several WE units had experienced steam generator failurescaused by loose parts in the reactor primary cooling sys-tem. The loose parts were identified as parts from failedsplit pins in control rod guide tubes. The cause of the fail-ures was identified as IGSCC.

The split pins are made from a nickel alloy (Inconel X-750)and are bolted to the bottom of the guide tube column. Thesupport pins are then inserted into the upper core plate. Thepins support the guide tube against hydrodynamic forcesand also align the tubes with respect to the upper core plateand the fuel assemblies. A sketch of the guide tube supportpin is shown in Fig. 5.3.

Crevice conditions are created when the pins are insertedinto the upper core plate and the pins are exposed to a lo-cally corrosive fluid medium. Improper heat treatment andovertorquing of the nuts during installation of the pins mayhave contributed to the development of IGSCC in the bot-totn region of the shank. As a remedy to the IGSCC prob-lems, WE now recommends a solution heat treatment at ahigher temperature, increasing the size of the pins, peeningthe nuts, and reducing the preloads during installation. Theobjective is to reduce the susceptibility of the material tothe corrosive environment and to reduce tensile stressesthat are necessary to the development of IGSCC.

ORNUO 3G- ETD

m 1 a,, -U

GUICE TUBEOT T0M

GurDE TUBEBOTTOM

PLATE

*. STRAIGHT PIN b. TAPERED PIN

Figure 5.3 Westinghouse PWR guide tube support pins

41 NUREGICR-6048

SurveyThe guide tube split pin IGSCC problems at PWR plantswere addressed in a NRC IE Information Notice No.82-29.4 Many plants have initiated programs to replace theexisting guide tube split pins with pins of the new design.

5.2.2.4 Core Baffle Bolt Failures In KWU-Built PWRs

The Kraftwerk Union (KWU) AG has built many PWRplants of the Westinghouse design in Germany andSwitzerland. Five of these plants have reported SCC prob-lems in the core baffle bolts.

The core baffle maintains the geometry of the fuel assem-blies and controls the coolant flow through the core. It is abolted structure composed of horizontal former plates andvertical baffle plates. The outer edges of the former platesare bolted to the inside of the core barrel, and vertical baf-fle plates are bolted to inner edges of the former plates. Inreactors built by KWU before 1979, the core baffle boltswere made of a nickel-alloy (Inconel X750). After theywere installed, the bolts were secured by tack welds. After1979, the core baffle is a welded structure in KWU-builtunits. A simplified view of the core baffle and its boltingscheme are shown in Fig. 5A.

ORNL-DWG 92-3331 ETD

IGSCC X

IiIIIi

IGSCCFAILURE

wIWI

Figure S5 KCWU-built PWR core afte bolt failures

Routine UT inspections detected signs of cracks in corebaffle bolts in five KWU-built PWRs. A typical bolt failurepattern is shown in Fig. 5.5. The problem was first reportedin 1987.5 Failures were attributed to IGSCC. The sus-pected cause was the sensitization of the nickel-alloy bolts

ORNL-DWG 92-3330 ETD

by the welding process, coupled with crevice conditionsand high tensile stresses in the bolts.

The nickel-alloy bolts were replaced by 1.4571 austeniticsteel bolts, and the replacement bolts were secured by amechanical scheme. There is no reported failure of the re-placement core baffle bolts.

The core baffle in domestic WE PWRs is a bolted struc-ture. The bolts are made of type 316 stainless steel. Asearch of the LER data base did not locate any reportedfailure of the core baffle bolts in U.S. reactors.

5.2.3 Mechanical Wear

The primary cause of mechanical wear in reactor internalsis FIV. Parts that are in close proximity to each other maycome into contact during vibrations, and the rubbing wouldlead to excessive thinning of the contact areas. Boltedjoints, in-core instrument lines protected by tightly fittedshields, and guide tubes are probable sites for mechanicalwear problems.

5.2.3.1 WE Reactor Flux Thimble and Guide TubeThinning

In-core neutron monitors, which are parts of the reactorflux-mapping system, are located inside retractable thimble

Figure 5.4 KWU-built PWR core baffle and boltingscheme

NUREG/CR-6048 42

tubes. The thimble tubes extend from selected locations in-side the core, through the bottom of the pressure vessel,high-pressure conduits, and seal tables to a ten-path trans-fer device. The thimble tubes, in turn, are supported byguide tubes in the lower region of the reactor pressure ves-sel. The guide tube is divided Into two parts: the upperpart, located inside the fuel assembly, and the lower part,extending from lower core plate to the reactor vessel bot-tom. The portion of the thimble tube between the fuel as-sembly and the lower core plate Is unprotected and exposedto reactor coolant flows. The thimble tubes are susceptibleto FIV in the unshielded region. Contacts between thimbletubes and guide tubes can lead to mechanical wearproblems in both structures. A sketch of the flux thimbleand guide tube arrangement is shown In Fig. 2.7.

Excessive thimble and guide tube thinnings and leakageswere reported in a U.S. reactor in 1981.6 Fretting-inducedfailures of guide tubes have been reported in foreign WEunits. The consequence of a thimble tube leakage is seriousbecause it can result in a breach of the reactor coolant pri-mary pressure boundary. The thimble tubes are opened atthe ten-path transfer device for the insertion of the neutronmonitor, and the development of tube leakages could resultin a nonisolable coolant leak.

Tube thinnings were detected by the eddy-current inspec-tion method. When tube thinning is detected, the currentremedy is to retract the thinned segments to move themaway from the vibrating region. Thicker-walled tubes alsohave been used. Tubes that were seriously degraded will beremoved from service by closing the isolation valves.Many utilities have instituted inspection programs to moni-tor thimble-tube thinnings. At the present time there Is nopermanent solution to the thimble-tube thinning problems.

Flux thimble and guide-tube thinning problems are dis-cussed In JE Information Notice No. 87-44.6

5.2-3.2 B&W SSHTs

Excessive wear was detected in SSHTs in B&W units inthe mid-1970.1 The tubes were mounted on the outsidesurface of the thermal shield and exposed to pump-gener-ated pressure pulsations. The failure was considered asgeneric in nature and was attributed to FIV. The tubes weremodified so they would be less sensitive to the pressureexcitations; this was accomplished by changing the stiff-ness of the tube and/or adding additional structural sup-ports. No problems have been reported since the correctivemeasures were implemented.

Survey

5.3 Failure Information Survey Results

More specific information will be provided on major inter-nal component failures. In addition to the major aging-related degradation mechanisms, the total number ofreactors reporting the failure and their unit age will also beidentified. The unit age of a reactor is measured incommercial operation years, which is the differencebetween the commercial starting date and the event date ofthe first LER on the reported failure. Unit age is aparameter that can be used to represent the age of thereactor when the first failure occurred. Corrective actionstaken will also be included in the survey results. LERsfrom 1980 to 1990 provide the bulk of the failureinformation. Useful information was also obtained fromEPRI reports on nuclear unit operating experiences.

5.3.1 WE Internals

53.1.1 Thermal Shield Support Bolts

Major aging-related degradation mechanisms: bolt fail-ures caused by fatigue. Pump-generated pressure pulsationsand oscillatory hydrodynamic forces are the primarystressors.

Number of units reporting failure: 6, mostly early reac-tors with the old thermal shield design.

Unit age at firt failure: not known for four units. Of theremaining two units, one reported thermal shield supportbolt failures after 12 years, and the other unit after 21 yearsof commercial operation.

Corrective actions: The pin-keyway support system wasreplaced by a flexure shield support system. New reactordesign replaces thermal shields with neutron shield padsattached directly to the core barreL

53.1.2 Baffle Plate Water-Jetting

Major aging-related degradation mechanisms: High-cycle fatigue. Baffle plate gaps enlarged by FIV. Fuel rodsdamaged by vibrations caused by high-intensity turbulentflows through gaps.

Number of units reporting failure: 6

Unit age at rt failure: 4, 5, 6, 13, and 20 years.

Corrective actions: Several remedies have been tried withvarying degrees of success, Including peening the baffleplate joints, replacing fuel rods in the impingement areaswith solid stainless steel rods, and using partial grids for

43 NUREGICR-6048

Survey

additional supports to the fuel rods. One unit has convertedthe downward bypass flow into an upward bypass flow inthe core baffle-core barrel region. New reactors use theupward bypass flow scheme.

53.1.3 Control Rod Guide Tube Support Pins

Major aging-related degradation mechanisms: IGSCCcaused by crevice conditions, improper heat treatment, andovertorquing of the split pin nuts.

Number of units reporting failures: 5

Unit age at first failure: 4. 7, 7, 8, and 14 years.

Corrective actions: Heat treatment at a higher tempera-ture. New split pins are larger in size. The nuts are peenedand installed with a reduced torque.

5.3.1.4 In-Core Thimble and Guide Tube Thinning

Major aging-related degradation mechanisms: mechan-ical wear caused by contacts with flux thimbles and adja-cent guide tubes. Motions induced by FIV of the un-shielded portion of the thimble between the fuel assemblyand lower core plate.


Unit age at first failure: 4, 4, 8, 11, and 17 years

Corrective actions: Tube sections with excessive mechan-ical wears are retracted and moved away from the vibratingregions, and thicker-walled tubes are used. There is nopermanent solution at the present time. Utilities have set upinspection programs to monitor thimble-tube thinnings.

5.3.1.5 Core Baffle Bolts in KWU-Built Units

Major aging-related degradation mechanisms: IGSCCcaused by welding-induced sensitization of nickel-alloybolts, coupled with crevice condition and high bolt tensilestresses.


Unit age at first failure: Not known.

Corrective actions: Nickel-alloy bolts were replaced byaustenitic steel bolts. Replacement bolts are secured by amechanical means instead of tack welds.

Table 5.1 Summary of WE internalsfailure infomation

Failure Aging-relatedComponent description degradationmechanisms

Thermal shield Cracked bolts Fatiguesupport bolts

Core baffle plate Enlarged gaps be- Fatiguetween baffleplates

Flux thimbles and Excessive thin- Mechanical wearguide tubes ning

Control rod guide Cracked pins Crevice-assistedtube support pins SCC

Core baffle bolts Cracked bolts Crevice-assistedSCC

5.3.2 B&W Internals

5.3.2.1 Thermal Shield Support Bolts

Major aging-related degradation mechanisms: boltingfailure caused by crevice-assisted IGSCC.

Number of plants reporting failure: 7

Unit age at first failure: 6, 7, and 5 units at 8 years

Corrective actions: using replacement bolts that featurenew boll design with reduced stresses in the shank region,peening the bolts to reduce surface tensile stresses, reduc-ing the torque used in the installation of the bolts, and us-ing Inconel X-750 instead of A-286 stainless steel as thematerial of construction.

5.3.2.2 Core Barrel-to-Core Support Shield Bolts

Major aging-related degradation mechanisms: crevice-assisted IGSCC.


Unit age at first failure: 7 and 8 years

Corrective actions: Replacement bolts are made with thesame material, grade A-286 stainless steel. New bolts weremachined, while old bolts were made by a hot-headed op-eration. The torque applied to the new bolts is also signifi-cantly reduced.

5.3.2.3 SSHT

Table 5.1 is a summary of the reported component failureinformation on WE reactor internals.

Major aging-related degradation mechanisms: mechan-ical wear caused by FIV.

NUREG/CR-6048 44


Unit age at first failure: 1 2 2, 2 2 and 3 years

Corrective actions: New SSHTs are structurally detunedto flow-induced excitations; this is accomplished bychanging the stiffness of the tube and adding new supports.

53.2.4 SHHT Bolts

Major aging-related degradation mechanisms: boltingfailure caused by crevice-assisted IGSCC.

Number of plants reporting failure: I

Unit age at first failure: 6 years

Corrective actions: SSHT bolts were replaced with studand nut fasteners made of Inconel X-750.

Table 5.2 is a summary of the reported component failureinformation on B&W reactor internals.

5.3.3 CE Internals

S.3.3.1 Thermal Shield Support Bolts

Major aging-related degradation mechanisms: high-cycle fatigue bolt failures caused by FIV.


Unit age at Srst failure: 7 and 8 years

Corrective actions: Thermal shields and support lugswere removed from reactors.

Table 5.2 Summary of B&W internalsfailure Infomation

Survey53.3.2 COe Support Barrel

Major aging-related degradation mechanisms: fatiguecaused by FIV. Failure of thermal shield support pins ledto increased loadings on the core barrel support lugs, and athrough-the-wall crack was detected at two lug locations.



Corrective actions: The through-the-wall crack was ar-rested by drilling a bole at each end of the crack. The coresupport barrel was put back into service.

5.3.33 Core Internal Hold-Down Ring

Major aging-related degradation mechanisms: mechan-ical wear caused by FIV. FIV was attributed to insufficientbold-down spring force.



Corrective actions: New hold-down ring made of type403 stainless steel was installed with an increased internalhold-down spring force. No new failures were reported.

Table 5.3 is a summary of the reported component failureinformation on CE reactor internals.

Table 5.3 Summary of CE internalsfailure infomation

Failure Aging-relatedComponent description degradation

mechanisms

Thermal shield Cracked bolts Fatiguesupport bolts

Core support bar- Through-the-wall Fatiguerel crack

Hold-down ring Excessive wear Mechanical wearFailure Aging-related

Component description degradationmechanisms

Thermal shield Cracked bolts Crevice-assistedsupport bolts SCC

Core barrel to Cracked bolts Crevice-assistedcore support SOCshield bolts

SSHT Excessive Mechanical wearthinning

SSHT bolts Cracked bolts Crevice-assistedSCC

References

1. American Society of Mechanical Engineers, ASMEBoiler and Pressure Vessel Code, Section XI, Rulesfor Inservice Inspection of Nuclear Power PlantComponents.

2. "Characterization of the Performance of Major LWRComponents:* EPRI NP-5001, a frnal report prepared

45 NUREG/CR-6048

Survey

by the S. M. Stoller Corp., for the Electric PowerResearch Institute, Palo Alto, Calif., 1987.

3. "Fuel Rod Degradation Resulting from Baffle Water-Jet Impingements," IE Information Notice No. 82-27,USNRC Office of Inspection and Enforcement,Washington, D.C., August 1982.

5. Nucleonics Week 28(10),1-3 (March 5, 1987), aMcGraw-Hill publication.t

6. "Thimble Tube Thinning In Westinghouse Reactors:IE information Notice No. 87-44, USNRC Office ofInspection and Enforcement, Washington, D.C.,September, 1987.

4. "Control Rod Drive (CRD) Guide Tube Support PinFailures at Westinghouse Plants," IE InformationNotice No. 82-29, USNRC Office of Inspection andEnforcement, Washington, D.C., 1982.

Available friom the National Standards Institute, 1430 Broadway, NewYork, NY 10018, copyrighted.

tAvailable in public tdhnical liraries.

NUREG/CR-6048 46

6 ISI and Monitoring Programs

The ISI program stipulates visual Inspections for reactor In-ternals during refueling outages. Visual inspections, whilerelatively simple to perform, have obvious limitations.They are used mainly to detect surface flaws in accessibleareas of a conponent. Visual examinations are not effec-tive in detecting subsurface or partial through-the-wallcracks and cannot detect failures in Inaccessible areas.Alternate methods have been tried, on an experimental ba-sis, to improve the effectiveness of detecting failures in re-actor internals. UT inspections have been used with somesuccess in detecting cracks in bolts and in areas that are notaccessible to visual inspections. However, UT Inspectionsalso have limitations and access problems. In addition, theinterpretations of UT examination results are much morecomplicated and difficult than those for visual inspections.The eddy-current method is another inspection method thathas potential applications in examining reactor internals. Ithas been used to inspect long tubular structures for exces-sive wall thinnings. The establishment of an effective in-spection program for reactor components is an evolvingprocess. When a new inspection method has been usedsuccessfully, the responsible organization will issue bul-letins and letters to inform plant operators of the latest de-velopments and their potential applications to reactor com-ponent inspection programs.

A major concern in reactor operations is the presence ofloose parts in the reactor primary coolant system. Reactoroperation is disrupted when loose parts are lodged in criti-cal locations such as inlets to pumps and heat exchangers.Loose parts were generated in some of the reported aging-related failures. Although none of these incidents had en-dangered the safety of reactor operation, the possibilitycannot be ruled out In future occurrences. In addition tosafety considerations, these failures have resulted in eco-nomic losses caused by extensive outages for repair work.For these reasons, NRC and plant operators are interestedin the development of inspection and monitoring methodsthat can be used to detect failures in reactor components.Research and development work to detect loose parts1 hasled to the issuance of NRC Regulatory Guide 1.133,2which outlines the operating requirements for loose partmonitoring systems (LPMS) in U.S. commercial nuclearpower plants. LPMSs are required for reactors licensedsince 1978. Many plants licensed before 1978 have in-stalled LPMS on a voluntary basis.

There is also a considerable amount of research and devel-opment work In the areas of preventive maintenance meth-ods for reactor components.-An effective preventive-main-tenance program can enhance the safety and efficiency ofplant operations. Vibration monitoring and trending studies

of key reactor components can provide the basic informa-tion needed for the decision-making process in an effectivepreventive-maintenance program. These studies are usedextensively in ISI programs for French and German reac-tors.3 They have been used on an experimental basis indomestic reactors, but vibration monitoring and trendingstudies have currently not been formally incorporated IntoU.S. plant ISI programs. Some individual plants are re-quired to make vibration measurements periodically as acondition of their operating Technical Specifications.

6.1 LPMS

The LPMS Is an in-service detection system designed toindicate the presence of loose metallic parts in the reactorprimary system. The system can provide diagnostic infor-mation to the plant operator concerning abnormal condi-tions in the reactor so that appropriate actions can be takento ensure safety in plant operations. Early warnings canalso minimize the risks to other reactor components andsystems.

The collision of a loose part, carried along by the reactorcoolant flow, with a stationary reactor component willgenerate sound waves, primarily bending waves, whichwill propagate to other structural components. The effec-,tiveness of an LPMS will depend on the system's capabilityto detect, capture, and interpret these structure-borne soundwaves.

A typical LPMS consists of a series of sensors (piezo-electric accelerometers) mounted on the outside of thereactor pressure vessel and steam generators to detectcollision-generated structure-borne bending waves. A ringof three sensors is mounted around the top and the bottomof a PWR pressure vessel. Two sensors, separated by atleast 3 ft typically, are recommended at the primary inlettube-sheet for steam generators. Inputs from these sensorsare amplified and then fed to a monitor that records andanalyzes the signals. When iterprted properly, looseparts collision signals can provide information on the massand energy of the moving object as well the Impact loca-tion. The input signals contain information that can be usedto evaluate the dominant frequency and amplitude of thestructure-borne sound waves detected at the sensor loca-tions. Differences in arrival time at the various locationscan also be determined. Using the dominant frequency andthe amplitude of the structure-borne sound waves and apredetermined calibration curve, the mass and energy ofthe loose part are estimated. The location of the Impactpoint can be determined by using differences in arrival

47 NUREGCR-W8

IsItime at the various sensors and a triangulation process.Many uncertainties are associated with the signal process-ing procedures. In general, the uncertainty level is lowwhen the loose part is small, and the sound wave propaga-tion path is simple and straightforward. The uncertainty in-creases with the size of the loose part and the complexityof the sound wave propagation path. More detailed infor-mation on LPMSs can be found in reports by Kryierl andMayo. 4

The performance of LPMS is mixed. This is not unex-pected because of the complexity and difficulties in inter-preting the structure-borne sound waves generated in acollision process. EPRI has conducted a research projectwith the goal of improving the performance of LPMSs.Key results and recommendations of the project can befound in an article by Weiss and Mayo.5

Note that a LPMS is not considered as a safety-related sys-tem. LPMSs can provide signals indicating the presence ofloose parts in the reactor primary system. These signals,when properly processed and analyzed, can lead to actionsthat could minimize damages to other reactor componentsand systems. Minimizing the damage in a loose-part inci-dent would reduce the outage time for repair work. Whenviewed in this manner, the incentive of improving the per-formance of LPMSs is economic.

6.2 Vibration Monitoring andTrending Studies

The safety of reactor operations and plant availability canbe improved with a monitoring system that can detect sys-tem degradations at an early stage so that failures or mal-functions can be effectively prevented. Vibration monitor-ing and trending studies, which are common practices inpreventive maintenance programs for rotating machineries,have potential applications in reactor inspection and main-tenance programs.

The hostile environment inside a reactor vessel would pre-clude the use of sensors attached directly to internal com-ponents for long-term vibration measurements. Instead, vi-brations of selected reactor internal components can be in-ferred from neutron noises measured by ex-core detectors.Neutron noises are fluctuations in the neutron flux arounda mean value. The neutron flux is moderated by the waterlayer between the core and the pressure vessel. Vibrationsof some reactor internals (e.g., the core barrel) can changethe thickness distribution of the water layer surroundingthe core. Changes in the water layer thickness lead to vari-ations in the moderating effects and can be correlated toneutron noises as measured by ex-core detectors. Ex-coreneutron noise measurement is considered an effective

NUREG/CR-6048

method for studying vibrations in reactor internals such ascore barrels and thermal shields. For more information onthe theory and application of neutron noise analysis for re-actor diagnosis, consult Ref. 6.

Results of an ex-core neutron noise analysis at a givenpower level and a known sensor location are usually pre-sented in the form of a noise spectrum, which is a plot ofthe normalized power spectral density (NPSD) curve overa specified frequency range. The ability to correctly asso-ciate characteristic features or spikes on the noise spectrumwith natural frequencies of reactor components is essentialto the success of the method. The structural frequencyidentification process can be accomplished by correlatingin-core and ex-core vibration measurements during preop-erational vibration testing or by analytical modeling.Temporary in-core sensors are often used during reactorpreoperational testing and can provide information on thestructural natural frequencies of major reactor internals.Neutron noise analysis is of limited value for reactor com-ponents whose natural frequencies do not form identifiablefeatures or spikes on the noise spectrum.

When the component natural frequencies are clearly iden-tified on a noise spectrum, trending studies can be per-formed with noise spectrums at different time intervals.Typically, three to five measurements are made in one fuelcycle. Deviations of an identifiable point in an actual noisespectrum from that of a reference spectrum can be inter-preted as indications of degradations in the component.The process will require the establishment of a detailed andaccurate reference noise spectrum for a specific reactor.The reference noise spectrum may be obtained from resultsof preoperational vibration testings. The ability to correlatedeviations in the noise spectrum with the severity of a sys-tem degradation is the key to the success of trending stud-ies. The establishment of a knowledge-based expert systemon normal and abnormal behaviors of reactor componentscan aid in the decision-making process.

Neutron noise analysis can be used on many reactor com-ponents and systems including reactor internals. One of thesuccessful applications of the method is the vibration as-sessments of core barrels in PWRs.7 The core barrel isspecially suited for neutron noise analysis because of itssize and the effects of its vibrations on the distributions ofthe water layer thickness between the core and the vesselwall. The vibrational characteristics of the core barrel arealso well understood. Inspections of the core barrel duringrefueling outages confirmed the general correctness of theneutron noise analysis results. There is practically nopublished information on the results of trending studies onreactor internals in U.S. reactors. One of the reasons couldbe because results of vibration monitoring and trending

48

studies may contain information that is considered as pro-prietary in nature by reactor vendors and plant operators.Proprietary information is not published in the open litera-ture.

Vibration monitoring and trending studies have been in-corporated into ISI programs for French and German reac-tors. They have been used on an experimental basis inmany domestic plants. The ASME Operation andMaintenance (OM) Committee is currently studying thepossibility of requiring in-service vibration monitoring forPWR internals.

References

1. R. C Kryter et al.,'Loose Parts Monitoring: PresentStatus of the Technology, Its Implementation in U.S.Reactors and Some Recommendations for AchievingImproved Performance," Progress in Nuclear Energy1(2-4), (1977).*

2. "Loose Parts Detection Program for the PrimarySystem of Light-Water Cooled Reactors," USNRC

IsI

Regulatory Guide 1.133, Washington, D.C., May1981.

3. D. Wach, "Vibration, Neutron Noise and AcousticMonitoring in German LWRs," Nucl, Eng. and Des.129 (1991).*

4. C. W. Mayo, "Loose Part Signal Theory," Progress inNuclear Energy 15 (1985).*

5. J. M. Weiss and C. W. Mayo, "Recommendations forEffective Loose Part Monitoring," Nucl. Eng. and Des.129 (1991).*

6. J. A. Thie, "Power Reactor Noise," American NuclearSociety, 1981.

7. D. N. Fry, J. March-Leuba, and F. J. Sweeney, UnionCarbide Corp. Nucl. Div, Oak Ridge Nail. Lab., "Useof Neutron Noise for Diagnosis of In-VesselAnomalies in Light-Water Reactors," USNRC ReportNUREG/CR-3303 (ORNLITM-8774), January 1984.t

Available in public technical libraries.tAvflabl. fr purchase fom National Technical Infcmation Service,

Springfield VA 22161.

49 NUREG/CR-6048

7 Discussion and Conclusions

Reactor Internals operate Inside the pressure vessel and aresubjected to many environmental as well as physical stres-sors. Primary stressors for PWR internals are related to thereactor primary coolant flow and exposures to fast neutronfluxes (E > I MeV). Stressors can initiate and sustain thegrowth of aging-related degradation mechanisms.

Aging-related degradation mechanisms develop at differentrates, and conditions inside the pressure vessel may favorthe development of selected dominant mechanisms in in-ternal components. Fatigue, SCC, and mechanical wear arethe major aging-related degradation mechanisms for PWRinternals. They are identified based on a review of reactoroperating histories and reported component-failure infor-mation.

Flow-induced cyclic hydrodynamic loads are importantstressors for PWR internals. These loadings include pump-generated pressure pulsations, vortex-shedding oscillatoryforces and high-intensity turbulent flows. These cyclicloads can excite internal components into vibrations. Whenthey are not properly mitigated, Fly can lead to fatigue andmechanical failures.

Because of Its low dissolved oxygen content, the waterchemistry of a PWR is not favorable to the development ofSCC in the bulk of the reactor cooling system. However,narrow gaps and other small regions in some internal com-ponents can trap a stagnant fluid medium with locally highconcentrations of corrodents. These crevices create a cor-rosive environment and can cause SCC when tensilestresses are also present in the system. Preloads in boltscan generate tensile stresses needed for the development ofSCC. Crevice-assisted SCC is found in bolted joints andother tightly fitted connections in reactor intemals.

Long-term exposure to fast neutron fluxes is a stressor thatcan lead to embrittlement and radiation-assisted SCC. Abasic understanding of these potential aging degradationmechanisms has not yet been achieved. There are manyactive research works whose main goal is to increase theunderstanding of neutron Irradiation effects on the aging ofreactor components.

When they are not mitigated, aging effects will eventuallylead to a failure in the affected components. Understandingthe stressors associated with these aging degradationmechanisms can lead to the establishment of effectiveprograms to control and manage these aging effects.

7.1 FIV

PWR internal components have failed as a result of FIV.Fatigue and mechanical wear are the two important failuremechanisms. FIV problems are generally resolved byeliminating the excitation sources or by changing thevibrational characteristics of the affected structuralcomponents. The practical choice for most reactor internalsis the second option where the component is detuned fromthe external excitations.

Some of the FIV problems were detected very early in re-actor operations and can be regarded as a part of the reactor"debugging" process. Excessive mechanical wear observedin CE reactor internal hold-down rings is an example of aproblem of this nature. The problem was resolved by in-stalling a new hold-down ring and increasing the hold-down force during installation. The SSHT in B&W units isanother successful example of detuning the structure frominput excitations.

Other FIV problems are more difficult to resolve. Thermalshields in many early CE and WE units have encounteredFIV problems. Fatigue cracks In support bolts were themost common reported failure. Detuning the shields by in-creasing the stiffness of the support systems did not proveto be effective in minimizing the effects of flow-inducedexcitations. Thermal shields had been removed from someCE units. Neutron-shield pads, instead of thermal shields,are now used In the new WE reactors. However, the re-moval of the thermal shield without an effective replace-ment may increase the neutron fluence at the wall of thereactor pressure vessel and adversely affect its long-termstructural integrity. It should be noted that analysis resultsindicated that removal of the thermal shields from the CEreactors will have no undesirable effect in the remainingdesign life of the affected reactors.

A more recent FIV-related problem is the excess mechani-cal wear observed in flux thimbles and thimble guide tubesin WE units. The problem has been traced to FIV in theunshielded portion of the thimble between the fuel assem-bly and the lower core plate. The current practice is to re-tract the thinned thimble sections away from the vibratingregion. No permanent solution has been proposed, and theflux thimble and guide tube thinning problem is still underactive investigation.

51 51 ~~~~~NUREG/CR-6048

Discussion

The baffle plate water-jetting problem observed in severalWE units is an example in which the coolant flow fieldwas modified to eliminate flow-generated excitations. Thedriving force behind the jetting problem was eliminated bychanging a downward bypass flow to an upward bypassflow in the core baffle-core barrel region. New WE reac-tors use the upward bypass flow scheme.

Most of the serious FIV problems in PWR internals havebeen resolved by well-established engineering practices.Flow-induced mechanical wear in flux thimble and guidetubes is an unresolved problem. Many plants have initiatedinspection programs to keep track of the thimble and guidetube thinning problems.

Due to the relatively high flow speed and high-intensityturbulence level in the reactor coolant flow, it is not possi-ble to eliminate all FlVs in PWR intemals. Some internalcomponents are subjected to small-amplitude vibrationsduring reactor operations. They may be susceptible tohigh-cycle fatigue failures. The effects of a corrosive envi-ronment on high-cycle fatigue failures are not well under-stood. Research results in this area have not reached thestate that they can be incorporated into design codes suchas the ASME B&PV Code. High-cycle fatigue may becomea more important factor in determining the useful workinglife for reactor intemals.

7.2 SCC

Most SCCs in PWR internals were detected in bolted jointsand tightly fitted connections. Crevice conditions at theselocations create a highly localized corrosive environmentthat is conducive to the development of SCC. The use ofmolybdenum disulfide (MoS2) as a thread lubricant cancontribute to the corrosiveness of the crevice condition.The susceptibility of the material and the corrosiveness ofthe cooling fluid are not sufficient to cause SCC. Preloadsin bolts and the tensile stresses that they generate are alsoneeded for the development of SCC.

Failures have been detected in thermal shield support boltsin B&W units and in control rod guide tube split pins ofWE reactors. The B&W thermal shield support bolts weremade of A-286 stainless steel, and the split pins are madeof Inconel X-750 alloy. The responsible aging degradationmechanism is SCC, specifically IGSCC. Improper heattreatment of the split pins may have also contributed to thecorrosion problems in guide tube split pins.

A two-prong approach has been taken to resolve SCCproblems in PWR internals. The first approach is to reduce

NUREGICR-6048

the susceptibility of the material, and the second is to de-crease the tensile stress level in the component. The mate-rial of construction for support bolts in the B&W thermalshield was changed from A-286 stainless steel to InconelX-750 alloy. A higher temperature is now used for the beattreatment of WE guide tube split pins. In both cases thesize of the bolt shank region was enlarged, and preloadswere reduced during installation. These design and fabri-cation changes should lower the tensile stresses below thethreshold level in the bolts. Together with the use of lesssusceptible materials, they should provide adequate protec-tion against IGSCC problems in support bolts.

7.3 Other Potential Aging-RelatedDegradation Mechanisms

Aging-related degradation mechanisms develop at differentrates, and, in addition to the three identified major degrada-tion mechanisms, other aging mechanisms may emerge infuture reactor operations.

Neutron irradiation effects may manifest themselves in theforms of embrittlement and IASCC. The expected lifetimeneutron fluence level for some reactor internals can attainthe value in which neutron irradiation effects are important.This is of special concern to components in the immediatevicinity of the core, such as the core baffle, in-core instru-mentation guide tubes, and possibly the core supportplates. The existence of crevice conditions, bolt preloads,and exposure to fast neutron fluxes would make the corebaffle a prime candidate for embrittlement and IASCC.

The exposure to fast neutron fluxes may also lower thetemperature at which creep and stress relaxation can be-come important aging-related degradation mechanisms forreactor internals. At the PWR, the normal operating tem-perature ranges between 288 and 3160C; the best estimatefor the threshold neutron fluence level for creep and stressrelaxation in stainless steels is about 6 x 1019 (neutrons/cm2). The expected neutron fluence levels for reactorinternals located in the core region can exceed the thresh-old value and are susceptible to the effects of radiation-induced creep and stress relaxation.

Reactor internal components made of cast austenitic stain-less steel (CASS) are susceptible to thermal aging and em-brittlemlent. PWR CASS internal components have ferritecontents in excess of 20% and operate at a temperature of-3160 C (600TF). CASS components can becomes brittleunder these conditions and may be susceptible to thermalembrittlement. Many CASS components have been re-placed by machined parts.

52

I -

Discussion

In-service inspections are relied upon to detect cracks andother signs of aging-related failures in PWR internals thatare susceptible to neutron Irradiation id thermal agingeffects. The adoption of a preventive maintenance programwith the capability of detecting system degradations willenhance the safety as well as the efficiency of plant opera-tions. Such a program can also provide useful informationto the decision-making process concerning the repairand/or replacement of a degraded component.

Note that most reactor internals are not components of thereactor primary containment system. Failures of reactor In-ternals may iesuli in conditions that can challenge the in-tegrity of the reactor primary containment system, but suchfailures cannot weaken the system itself. Key internalcomponents can be replaced if necessary. Therefore, a vig-orous and comprehensive in-service inspection programwith the capability of detecting and monitoring systemdegradation will ensure the safe working of reactor inter-nals components.

53 NUREG/CR-6048

NUREG/CR-6048ORNLITM-12371Dist. Category RV

Internal Distribution

1. D. A. Casada 41. W. P. Poore m2. D. D. Cannon 42. C. E. Pugh3. R. D. Cheverton 43. 1. S. Rayside4. D. F. Cox 44. T. L. Ryan5. C. P. Frew 45. C. C. Southmayd6. R. H. Greene 46. J. C. Walls7. H. D. Haynes 47. ORNL Patent Office8. J. E. Jones Jr. 48. Central Research Library9. R. C. Kryter 49. Document Reference Section

10. J. D. Kueck 50-51. Laboratory Records Department11-40. K. H. Luk 52. Laboratory Records (RC)

External Distribution

53. G. Sliter, Electric Power Research Institute, P.O. Box 10412, Palo Alto, CA 9430354. J. W. Tills, Institute for Nuclear Power Operations, 1100 Circle 75 Parkway, Atlanta,

GA 30339-306455. R. J. Lofano, Brookhaven National Laboratory, Bldg. 130, Upton, NY 1197356. R. P. Allen, Battelle-PNL, MS P8-10, P.O. Box 999, Richland, WA 9953257. J. P. Vora, U.S. Nuclear Regulatory Commission, Office of Nuclear Regulatory

Research, Electrical and Mechanical Engineering Branch, 5650 Nicholson Lane,Rockville, MD 20852

58. G. H. Weidenhamer, U.S. Nuclear Regulatory Commission, Office of NuclearRegulatory Research, Electrical and Mechanical Engineering Branch, 5650 NicholsonLane, Rockville, MD 20852

59. E. J. Brown, U.S. Nuclear Regulatory Commission, Office for Analysis and Evaluationof Operational Data, Reactor Operation Analysis Branch, Maryland National BankBuilding, 7735 Old Georgetown Road, Bethesda, MD 20814

60. C. Michelson, Advisory Committee on Reactor Safeguards, 20 Argonne Plasma Suite365, Oak Ridge, TN 37830

61. M. Vagins, U.S. Nuclear Regulatory Commission, Office of Nuclear RegulatoryResearch, Electrical and Mechanical Engineering Branch, Division of Engineering,5650 Nicholson Lane, Rockville, MD 20852

62. J. J. Bums, U.S. Nuclear Regulatory Commission, Division of Engineering Safety,Office of Nuclear Regulatory Research, 5650 Nicholson Lane, Rockville, MD 20852

63. M. J. Jacobus, Sandia National Laboratories, P.O. Box 5800, Division 6447,Albuquerque, NM 87185

64. H. L. Magleby, Idaho National Engineering Laboratory, MS 2406, P.O. Box 1625,Idaho Falls, ID 83415

65. V. N. Shah, Idaho National Engineering Laboratory, P.O. Box 1625, Idaho Falls, ID83415

66. Office of Assistant Manager for Energy Research and Development, Department ofEnergy, Oak Ridge Field Office, Oak Ridge, TN 37831

67-68. Office of Scientific and Technical Information, P. 0. Box 62, Oak Ridge, TN 37831

55 NUREG/CR-6048

NRC FORM 335 U.S. NUCLEAR REGULATORY COMMISSION 1. REPORT NUMBER12-891 (AsIoned Irv NRIC. Add Val., Swap., Pev..N2C1302 BIBLIOGRAPHIC DATA SHEET NUREG/CR-6048

See instructions on the ORNL/TH-123712. TITLE AND SUBTITLE

Pressurized-Water Reactor Internals Aging Degradation Study 3. DATE REPORT PUBLISHED

MONTH I YEAR

Phase 1 September 19934. FIN OR GRANT NUMBER

B08285. AUTHOR(S) 6. TYPE OF REPORT

K. H. Luk7. PERIOD COVERED anchww Dates,

B. PERFORMING ORGANIZATION -NAME AND ADDRESS IN NRC. pfvide Division. Office erlion, U.S Nuckar Rguleto Cou nisvion and mWilig addred mifconwsct, provideowne end mailing adibusr

Oak Ridge National LaboratoryOak Ridge, TN 37831-6285

9. SPONSORING ORGANIZATION -NAME AND ADDRESS Iii NRC. type maes eb we ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~: if con rwsos~~~~~~~~~~~ provide NRC Divleion. Office es Region. U.S Nuclear Regulasosy Commission.~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~9 SPONSOR ING ORGANIZATION -NAME AND ADDRESS (if NRtyeaeo / etpeviwNCOvin Offk* Or Iionw U&S NaulrarltvuhotwCommtnnon,

and mailihn addre.)

Division of EngineeringOffice of Nuclear Regulatory ResearchUS Nuclear Regulatory CommissionWashington, DC 20555

10. SUPPLEMENTARY NOTES

11. ABSTRACT 1200 was ar less)

Tkis report documents the results of a Phase 1 study on the effects of aging degradations on pressurized-waterreactor (PIR) internals. Primary stressors for internals ar generated by the primary coolant flow in the reactorvessel, and they include unsteady hydrodynanic forces and pump-generated pressure pulsations. Other stressorsare applied loads, manufacturing processes, impurities in the coolant and exposures to fast neutron fluxes. Asurvey of reported aging-related failure information indicates that fatigue, stress corrosion cracking (SCC) andmechanical wear are the three major aging-related degradation mechanisms for PWR internals. Significantreported failures include thermal shield flow-induced vibration problems, SCC in guide tube support pins andcore support structure bolts, fatigue-induced core baffle water-jet impingement problems and excess wear in fluxthimbles. Many of the reported problems have been resolved by accepted engineering practices. Uncertaintiesremain in the assessment of long-tenn neutron irradiation effects and environmental factors in high-cycle fatiguefailures. Reactor internals are examined by visual inspections and the technique is access limited. Improvedinspection methods, especially one with an early failure detection capability, can enhance the safety andefficiency of reactor operations.

12. KEY WORDS/DESCR PTORS Listwotdsorphramrharw.,isistefhev in locatng. the teaOn. I 13. AVAILABILITY STATEMENT

NPAR, reactor internals, stressors, aging-related degradation unlimitedmechanisms, component failure information, inservice inspections I4. SECURITY CLASSIFICATION

pressurized-water reactor (PWR), stress corrosion cracking (SCC), unclassifiefatigue unclassified

Iaes Regotti

unclassified15. NUMBER OF PAGES

16. PRICE

NRC FORM 335 1249

Federal Recycling Program

NUREGICR-6048 PRESSURIZED-WATER REACTOR INTERNALS AGING DEGRADATION STUDY SEPTEMBER 1993

UNITED STATESNUCLEAR REGULATORY COMMISSION

WASHINGTON, D.C. 20555-0001

SPECIAL FOURTH-CLASS RATEPOSTAGE AND FEES PAID

USNRCPERMIT NO. G-67

OFFICIAL BUSINESSPENALTY FOR PRIVATE USE, S300