.•.

\

..•

SESSION t

TECHNICAL SESSIONS

Technlca' program Information Is tentative and subjectto revilion

SESSiON 2 (continued)

Improvement of Strel. CotTo.lon Cracking Re.lstanceof Inconel Weld Metals by Mean' or StabilizationParameter Control

K. YAMAUCHI, 1. HAMADA. lind A. NISHIOKA, KureResellrch LAborlltory. Babcock·HltllChl K. K.. Kure. JAPAN,T. YOKONO lind T. OKAZAKI. Kure Works. Bllbcock·HltllChi K. K.. Kure. JAPAN

Temper Embrlttlement and Hydrogen Attack of 2 1/4Cr.IMo Steel, In High Pressure and High TemperatureHydrogen Atmo.phere.

T.IMANAKA lind J. SHIMOMURA. Kllwuakl Steel Corpora­tion. Kurashlkl. JAPAN

Bk. 1'(0. 000242

PIPIHQ DESIGK-STRES8 ArtAlY81S

09.00- t 2.00 hour.Tuelday. September 1 J

Volume t

Flanged Benda Under Combined Pre••ure and In·PlaneBending .

J. SPENCE. University of Slrllthclyde. Glasgow,SCOTLAND, lind G. THOMSON, Femml1. Edinburgh.SCOTLAND

Thermal RatchetUng CriterIa and Behavior of PipingElbo..,

M. UEOA, Hitachi losen Corp.. Saklli. JAPAN, T, KANO.Power Reactor lind Nuclellf Fuel Development Corporation.Tokyo. JAPAN. lind A. YOSHITOSHI. Power Reactor lindNudellr Fuel Developmenl Corporation. IblIrllki·ken,JAPAN

Analyals of Bimetallic Clamp Type Connectora Sub­Jected to Thermal Tran,lents

D. YOONG and J. NUTT, Harry J. Sweel (, "",odates. Inc..Houston. TX lind F. ADAMEK. Grlly Tool Compllny.Houston. TX

Fatigue or PIping. Under Out-or·Phase LoadsL ISSLER

DI'cus,lon of Stre.. limit. for CIa,. 2'3 PipIngG. SLAGIS. G.c. SI"9IS Associlltes, Walnut Cr~k. CA

SESSION 2

ENVIRONMENTAL EFFECTS ON PRESSURE VESSELMATERIALS

Tue,day, September 11 09.00·12.00 houn

Volume 2 Bk. No. G00243

SES8101"l3

STRESS ANALYSIS TECH"IQUES-PART I

Tuesday. September 1 t 09.00·12.00 hour'

Volume 1 Bk. No. 000242

Fillet Weld. Under Bending and ShearH. FESSLER. The University of Nottingham, Nottlnghllm,U. K.. and C. PAPPALETTERE, University of Barl. Barl.ITALY

A New Technique for E,tlmatlng the Stre.. ln Pad TypeNozzle. Attached to a Spherical Shell

T, OIKAWA and T. OKA. Nippon Koklln K.K.• Kawasaki,JAPAN

The Stress Analy.l. of Rotationally Symmetric Bellow.or Arbitrary Section

J. T. BOYLE and J. SPENSE, University 01 Strllthclyde.Glasgow. SCOTLAND

Autofrettaged Calculative Method of a Thlck.WaliedCylinder ror the Open End Ca.e

X.·Y. CENG. South Chinll Institute of Technology.Guangzhou. PEOPLE'S REPUBLIC OF CHINA

On Radial Spring Con.tanLa at the Juncture of a RadialNozzle and a Spherical Shell

B. BATRA and B. SUN. New Jersey In,tltute of Technology.Newark. NJ

Stress Corrosion Cracking Susceptlblllty of Low AlloySteel' Used for Reactor Pressure Vessel In HighTemperature OlCygenated Water

J. KUNIYA.1. MASAOKA.llnd R. SASAKI. Hitachi ResearchLAborlltory. HitllChi Ltd.. Ibaraki·ken, JAPAN. H. ITOH.Hitachi Works. HitllChi Ltd.• lborllki-ken. JAPAN. lind T.OKAZAKI. Kure Works. Babcock Hitachi Ltd., Hlroshlmll'ken. JAPAN

Hydrogen Damage of 2 114 Cr-IMo Steel Under Con­,tant Loading In High Pre..ure Hydrogen at ElevatedTempentures

K.'YOKOGAWA. S. FUKUYAMA. K. KUDO. and M. ARAKI.Government Industrial Research Institute. Hiroshima.JAPAN

\

Integrity Assessment for a Zlrcaloy·2 Pre.,ure Tubewith Frettlng Corro.lon Defect.

F. AUCINO lind G. ZAMPINI. NIRA. S.P.A. GenovlI. ITALY(conlinued)

SESSIO" 4

MANUFACTURING TECHNIQUES FOR PRESSUREVESSELs AND PIPING_PART I

Tueaday. Se;ptember 11 09.00-12.00 houn

Volume 2 Bk. No. G00243

Manufacturing Supervision and Inspection ofMultilayer Pre.sure Vessel. for High Pre.,ureInltallatlon.

K. O. PRESSEL. Rhelnis<:h·Westfall1scher TUV. eV Essen.GERMANY

(con/irlUed)

10 11

."

INTEGRITY ASSESSMENT FOR A ZIRCALOY-2PRESSURE TUBE WITH FRETTING CORROSION DEFECTS

F. ALICINQ

.,

•

'.ABSTRACT

In pressure tubes for he..... water reactors. fretting cor.rolion defects can rise owina to the contact with the fuelcladding bearin& pads..

These defecu, wttich may be' "single" or "multiple" in• tube belt, could $ignUlCirltly c:n1wlcc: the stresses in thepreuure tube.

For this reason, to have high confidence in tube relia­bility, • specifIC tube integrity assessment ..auld be per·formed taking into account the different modes of failurewhich the component can undergo, Le.:• static collapse (bunting)• buckling• creep• non.<fuctilc fracture• fatigue

To complete such an assessment, an evaluation for thechanges in the wety ractors throughout the life ot theplant may be performed. This evaluation can permit plantoperators to dcflnc suitable surveillance and in·aervicc in­qJection programs with greater reliability.

NOMENCLATURE

ad ,. Depth of each defect COfUtitutin& • circumfcrcn·tially eroded tone

If .. Depth of the uniformly eroded zone equivalent toa set of circumferentWly distributed defecu("multiple" defecu case)

" '"' Depth of a continuous eroded zone (in general),,(N) '"' Depth of the fretting defect after N refuelingsA .. Area of the cross section of the pressure lubt:Ce ;; (2 JT2 E{Oy)1I1

Cs '"' KL/R

G. ZAMPINI

d ., Width of the fuel cladding bearing padsD ;; Diameter (in general).0;(0);; Inner (outer) diameter .of the pressure tubee ., Neperian basis£ '"' Elasticity modulusFe '"' Materill parameter'"' (O.25/(n+O.221»)·(e!nfFs ;; Safety factor (pB(pO)Fz ,., Axial shape faclorH '" Axial coordinateJ .. Moment of inertia'"K ;; Material constant or sfendemess ratio (see expres-

sion of C,)L ;; Length of the tubeLp = Fuel cladding bearing pads lengthM ;; Folias' factorn .::: Number of prinlS lefl by the pads on the tube or

strain hardening exponent (sec expression Fe>N .. Number of refuelings already occurred (N =. o .... )P '" Applied pressurePB '" Bursl pressure for undefected or defected tubePn .. Desi&n pressurePL .. l...ocal primary membrane stress intensityPm '"' General primary membrane stress intensityqe '"' Number of circumferential defects (qe CO; Q)Q '"' Minimum number of defects around a circumlcr·

enee which are necessary to obtain an equJvaJentcontinuous defect

, = Inertia radius of section (= ../JIA)R" ;; Ratio fur axial buckling (R,,:::: Erl fcrit>Rm ;; Mean radius of the pressure tubeRr ;; Ratio for shear buckling (R t ;; Er!rcrit)Sm ;; Allowable suess intensitySI/ ;; Ultimate strength of the maleri:1l of Ihe prc-ssuII:

'"'"

\

=-t------------------·-- -- IJI,' ·:iI'::olt~r>\LI r-• ~ • Thickness of the prcssure tube wall II So, to perform a complctc structural verification of a r

• Tcmperature prtssure tube, it is nccessaty to evaluate how frelting corolZ • Ratio betweeo depth and thickness of the fretting rosion defects can influence the above failure modes.

defect (Z-Il/I) According to the ASME Boiler.and Pressure Vessel CodeZ(N) - Increucd Icagth due to acep after N refuelinp definition, such defects have to be c1.assif"lCd .as "IfOSS dis- ,II - Axialanilotropk: creep paruneter cootinuitieso"6 - Height or fuel cladding bearing pads , On the other haDd. they differ from the "grOIS discon- IA. - la~uo in fretting defect depth after each refuel- tlnuities- of the components covered by the Code, because •

10& (I- 1..N) their dimcasions oonnally lncreuc throughout the life of'dzi - 1Dcreue iD. dcfClCtl kOlth dl.IC to each refueling the plant.

0-1- N) - Therefore, a specific tube integrity usessment which~z - Creep axial rtram addresses the effects of fretting corrosion defects should~tfWI..- Muimum QCutron flux be performed .accordinl to the following guidelines: 1feZ) • Fut IlCUtroo flux (E > 1 MeV) (Il) Application of the ASME Code, Sectlon III Ow l," - Poisson's ratio - design methodology, with .:tme suitable modifica-0" .. AppUcd u.ialsuess lions to account the differcnt behavior of the cono ,oQil - CriUc.a.l axial stress delered material (Zircaloy·2) compared with RU-

06 - Hoop wess terials adopted by the ASME Code itself;08,£ - Loca1hoopltrea (b) cocucnoatiYe theoretical evaluation of defect sizes'D6p • Hoop stress for. defective pressure tube . and shapes and their change VI time, on the buisDy - YIeld strelJ8lb. of the material of both short tenn experimental tests and analyticala; • Modifled applied axial suess considerations;a-cd - Modif.cd c:ritical axial stress (e) identifICation of the failure inodes - among thosedT - Interval of time betweeo two refuelings . normany considered in designing pressure tubes _,. - Applied shear sUess or iourval of time from one I which are IfCaUy influenced by the presence ofsuch,

R{ueliDg to IDOther defects;,.ail • Crit.ical shear stress (d) defwOOn of suitable ""refere.nce" or "design" de·~ - Mod.iCICd applied shear stress . fectl, from the point ofvjew of both shape and size;t - Safety f.actor (depending 00 loading conditions: (~) execution of the structunl verifIcation according to

normal, upset, emergency, fauJud) the ASME Code.t4. - Ratio betweeo defective and undefective cross sec· I Furthermore, the reliability aspect for nuclear compo-

tiona! areu -. 'nents being of primaly importance. the propoJed integritytlx - Ratio betweea defective and u-ndefectivc axial mo- I assessment will address also the changes of safety factors

ments of inertia I throughout the life of the pbnt, to demonstrate that thetip - Ratio between defective and undefective polar mo-' plant o.n operate with sufficient safety margins.

ments of inertia Tho approach followed to purS1le the above items isIDusf:rated in the flow chart of Fig. I.

I IINTRODUCTION I,

Owing to the contact with the fuel cladding bearingpads., heavy water ~eactor pressure tubes may be IU~

jected to the rise of fretting conosion defects (see theaketch of a prasure tube cross section shown in Fig. 14).

These defects in general increase the state of stress inthe premue tube. and therefore can cause its failure. The Ifailure may be due to the (oUowing failure modes with

respect to which • prellUJ"e tube Is designed: I• static co11.apwe (burst1n&)• bucklin&• creep• non-ducti1e fracture

• fatigueII

GENERAL METHOD

In the Fig. 1 flow chart the method followed to per-Iform the integrity assessment of fretting corrosion de·fecttd pressure tubes is shown.

The steps shown in Fi&- 1 are briefly described below:(a) The fust step consists in identifying all the POf­

sible shapes and ws of defects which the freltingcorrosion phenomenon is able to develop.

It is not possible to define these parameters on I

an exact basis, sincc long term experimental resultsare not available, Furthermore. real types and di·mensions of these defects have a probabilistic fU·

ture, depending on normal forces between fuel clad·ding bearing pads and pressure tube, frequency of

\

r.d .:. I:! "1'

DETERMINATION OF TYPOLOGIESAND DIMENSIONS OF

FRETTING CORROSION DEFECTS

It has already been outlined that shapes and dimensionsof frettin& c:orTOlion defeeu ClD be estimated only on atheoretical basil. owio&: to lack of long term experimentalresults and the large number of different parametcn whichcan prodlolCC the. defects.. I

Two families of defects an after all be recognized:"lingle" and ""m\lltiplc" defects. -

this means that two different situations can be imag·l&ted: one .tUch produ.c:es only one defect which becomesdeeper during the life of the plant. and another in whichmany defects, doc enough to each other 10 that they maybe conJidercd lntcnatna1, are able to arise. -

Keeping in mind that the aim is to evaluate in a con·servatiYe way the maximum poSSlble dimensions of fret·ling corroaion defects, it is important that the events whichkad to one or lbe other of the above situations are con­scrvatlYely taken into account and their effecu are notoverestimated..

In such • way • methodology will be defIned to eyalu·Ite the defect', dimensions, taking into ,ccount crackIlowth. even If this effcct is really neg1ig.ible due to thedefect', lhapc and the number of cycles.

'Z" ~o8K(T-I6fJ)HZ)6r (4)

Parameters tI and L Y1 time assume the trends qualita·tively shown in FigL 2 and 3, tespcctively_

When examining It:ngth, it lw to be considered that Zir·caloy pressure tubes are subjected to aeep. so thai itseffects on length must be taken into account. Therefote,the following equation for the "lingle" defect length maybe written:

(2)L " Lp+Z(N)

preuure tube is left.(b) Fuel cladding bearing pads maintain the same posi­

tion during each refueling2.(c) Bearing pads and pressure tube are equally con·

,umed by erosion. The amount of erosion aftereach refue!lna: is assumed to be equal to the actualhalf height of the PolIch.

(d) The effects due to the contact between pressuretube and fuel cladding arc not considered.

(c) ~fu.ellnp follow each other at constant intervalsofume.

EVALUATION OF THE -SINGLE" DEFECT TYPEDIMENSIONS

Depth and length have to be aiaJl.ated.On the basis of the previous hypothesis. it is possible 10

write the foUowinl e.xpression about depth:

tI(JI) • AiI•• .6.«I" ---- .. .6.«N;

• • • leN"2+<+-----2[2-(2) I (I)

This equation states that the actual defect length (L), aftetthe N refueling, is equal to the pad's le03th (Lp ) plus theinaeue UN in defect length due to creep times the num·ber N ofrefueUngs [Z(N) .. N·~ZNI. The term ~ZN h.asthe following expression:

6ZN" fll"dZ""'fH ~(,)dZ""'~F,'H(3)o 0

with3 :

".

the relative motion, ctc.Therefore. a theoreUtal conxrvative assumption,

(or dupes and sius, is the only possible and prom­Wnllpproach.

(b) The next step consists of ev:aJ.uating what failuremodes aR i.nlluenoed by DlCh defects.

(e) Then I strudu.n.l veriflCl.tioo according to ASMECode Section III [IJ bas to be perfonned. "Refer­e.ace" or "deUp" defects are assumed fOI this

""po.(d) EnlltUa1ly. it Is convenient to show that the yery

blah reliability of nuclear components. assured bythe hiab Code Ilfety (acton. is not diminished bythe plelCOCO olsuch defeeu.

-I

SITUATION PRODUCING A "SINGLE DEFECT"

To COl1ICrvatiwly evaluate the "dngJe" defect type di­mensions. the following hypothesis may be used:

(a) Erosion velocity is assumed to be very high. so that1mmcdiatcly during each refueling a print on the, ,

J lntc,actiocl IRq .. defiDal on tho bail of ASWE SectioQ IIICod4i .cqulrtmc_u.

lThb". t'C1Y peu.iaUltil:: uaumption.

_ ) For Uw mc.aninl of the oUler amouatl, -= Homerw:;blure.

SITUATION PRODUCING A "MULTIPLE" OEFECT

To conservatively evaluate the "multiple" defect typedimensions. the foUowmg hypothesis may be used (forclariflQoon, see Fig. 14): I

(tI) a ct of defecta lying on different meridional planes,but In corcespondence to the same transverse pl:me,is considered;

(b) erosion velocity is ycry high, just as for sing.k de­fects;

\

(8)

(9)

if" <Qf2

if.>Qf2

if n <Qj2

if n > fJJ2

(a)

(b)

(a) ,. 7[,t (2 - (t)"]

(b) •• ~ (2 _ (!)"]

., '.1 ... - ...- ...•-- - - - - _... _. II

~ it hu been assumed that each refueling produces I ­

two prints upon the tube, when N refuelings have occurred .the number of defects is: I

then:

Since each defect hu the depth:

'd·t (2-4)N].

For simplicity fmally it may be assumed that the length \for "multiple" defects grOWl following the same law ofthe "single" defect.

In Fig. 4 the qualitative variation of parameter Ii VI time• "'own for the ''multiple'' defect type.

STRUCTURAL VERIFICATION BASEDUPON ASME SECTION III CODE

..' ." ~ ;. r

(e) a unif~ <:orrosion around the circumference is ItakeD 11I10 account. Such • lituation can be caused Iby,(I) • real uniform crosion uound the circumfer- !

cocc; I(2) • let'ot many prints very c;lose to each other,

10 that it 11 possible to coOJidcr them u inter­ac1in& 00 .ltructunJ basiL

By means of. unifoon erosion. lbaUower defects thaAthose of the KCODd IituatJon may be obt&ined.

For this reuon. refel'Cnce wiI1 be made in the followin&to this lut lIituatiolL

then it will be auwned u uniform erosion depth. thodepth due to • Jet of defects IYlnB on the same transverseprcUl.Ire- tube belt.

- (d) It is aaumed that initially. aU the defects around Icircumference arc formed and afterwards their di­.merWons begin to incrcliC.

(e) Finally. It is Illumed that not only one pa.d is incontact with the pressure tube durin8 each refuel­in& 10 that more than ODe print 11 formed. Twoprints of klentical depth are assumed u • COllie­

qucncc ofeach refudina-

!--,_ ..,,,

-REFERENCE" DEFECTSEVALUATION OF THE "MULTIPlE"_OEFECT TYPEDIMENSIONS

4 Cup has been taJr;cn inlO aa:ollnt bl' mcreaan, the derecuknlth and dimmilhinllhe lube thiocl:ne•.

When the number of circumferential defects isqe <Q.each baring depth lid, an equi¥alent uniform eroded zone Imay be defmed. This has a depth ii equal to the depth lidof each defect averaged on the circumference: I

, •• 'dfa (7) 'I• ,

At fust it is nccnsary to stlte when more defects lyingon diffe,nnt meridional planes can be considered u inter·K&a- _

In this context, reference is made to ASME III criteriaregarding local stresses (Art. N~32I3). On tbe basis ofthis criterion, two 1oca1 stressed zones can be considcredas interacting if their circumferential distance is less than1.0 VRm,.

This criterion pennits us to calculate the minimumnumber Q of defects around the circumference whichcauses the development of. uniform eroded z.one. Thcn:

MODES OF FAILURE TO CONSIDER

To perform the structural verification of a defcctlvepraAlle tube. it is coD¥t:nicDt to introduce the con(:ept ofa "tcfercnoe" dcfect.

Such a "reference" or "design" dcfect will have its di·mension and shape referenced at the time of verification.

Thus, the verification in the initial condition. beforevuiation of mechanical characteristics caused by ncutronfluence. shall be conducted referring to a "singlc" dcfectwith a dimension not greater than the initial one (Fig. 2).To verify, instead. the defective prcssure tube subjectedto irradiated conditions. it is most proper to consider acircumferential (continuous) defect. supposing that a i"single" defect grOWl in Icngth rather than in depth durolog the Ufe of the componenL In fl~t it is most probablcthat a 5Ct of close defects can rise, leading to a situationof continuous defect only after sevcral cycleL

From the strUCtural poinl of view. such I defect hu tobe considered u a "gross gcometrical discontinuity",according to ASME Seclion III (Arl. N~3220).

Fretting corrosion defects can mainly influence the fol­lOwing failure model:• stalic collapse• buckling• non·ductile fracture• creep4

(S) I

(6)

Qd ... Q.,JRmt - ·~I

a• ••,/(d • VRmr)

\

, -

(II)

(12)

(14)

(13)

(IS)

aa

Urnt

_t_ '" p -..!..-/-(1 m Il

I- t

tilt" 0.33.

when:

It followa:

A plccwre tube an buckle due to the combined effectsof dlear (torsion.a.l moment) and &Jtia1 suess (axial load).For an uftdefective tube, the condition which the compo­nent shall satisfy will be:

Therefore. the statk collapse of the defected tube isverified in accordance with the Code if the undefe.ctivctube is verified and if the maximum expected defect depthilless than 0.33 t.

when:

Id>I-18

(16)

to being a parameter depending on:• geometry of defective and undefective pressure

lub<• applied axw and shear stresses'• critical axial and shear messes1

(b) The same conclusion may be obtained for lhemultiple defect.

BUCKLING

In fad:

In a defective tube the critical and applied stresses arecharged by the defects, because geometrical properties ofClOSS Jections ue changed (Fig. 12). When evaluating theinfluence of thc defects on critiClJ and applicd suesses,the main conclusions which can be rcached arej;:

(a) In the cue of "singlc" defect, the defective pres­sure tube is verified against buckling, if the unde·fective one is verified in regiolU far from discon­tinuities with. margin:

Then, wuming PL '" 0B.L and 1.5 Sm being the liinit forPL- the 10cal primary membrane stress intensity alwayssatisflCS this limit, if in the regions far from the discon­tinuity the general primary membrane stress intensity Pm I,~ verifted. providing that: Glt'" 0.33. I:..- J . 1Sne- ...........Irioog are "a1id o"'Y In the caJoC wben the pl'Clalnl \

tube ~.. dl_....' -* that tonioaa1 budIi:A& (alb lA theplutk fWd, (FJe,. 13).

'Appliod uiIl.trcM .. cnlwlccd by tIM pre.nee of the defoet in

lbc ...ount IItA <_ HomcndltwC); applied 1bear ttreD 11.~ ill tM: unount IIfJp <toe NOIllllDdatutll).

7 Qitkal uialnreu ,ben bY (_ (S I and (11);

. . -lOY II-C;J.fC:1 if c, < Cc (11)

CR .' EtC: 1f C, > C('

..~d. to thc defect. to the n.hac:

.CR~ • .Jfl4fA tlCR (11)

Fw: c:ritieal shear Itrcu.ee Fie- 13 [6-81; it can be oblr:rved thaiIn the plattic rldd critical Jhcar s""'" Is IcCI unchanacd by thedeCoct.

-T--. STATIC COLLAPSE

To verify that • defective pressure tube cannot fail bystatic collapse in both initial and end~(·lifc conditions, it11 ncceu:uy to determine the state of Illcu induced bymechanical loads and pttSlUlC. The most signiflC&ll.t loadis pressure; therefore, ill the foUowinl discussion, onlypreuwc will be considered.

11 has boeD mentiooed already that the "'reference" de­fect in tho Initial condition is "ainaJo" while the defect incnd-of·llfc conditions is "'multiple." Ncycrthelcu. the dateof mea iodut:cd by preuurc aD a defective pressure tube11 most COD'tCnicntly aDd easily determined usuming con­tinuous defects wb.k:h have • depth corresponding to the

. life considered.. .:-.Applying the preau.n: as load, stress intensities VI axIal

c:oordinatc d.Lagrams can be obtained, as shown in Fig. S.On the basil of Article NB-3210·13 of ASME Section

III •• -grail geometrical discontinuity" will be Nch that:(a) membrane Wesse, can be considered as local if

Pm > 1.1 Sm tluoug,h a meridional length notpeater than ..JRmr;

(b) bcndinc rtrean always have to be considered as10 cal. .

Centrally the maximum membrane stress is less thanthe value obtained, considering the thickness of the tube ,k:ss the defect depth: I

pD -. .aB,L • 2(t-a) (10) I

\

pressure, which is the most signifICant load. In such a cue, 'the "safety factors are deftnCd as:

Furthermore, to take into account safety factor changesdUring the Ufe of the plant, variations in mechanical charac·teristics of Zr·2 have to"be considered. I

BURST PRESSURE FOR AN UNOEFECTIVEPRESSURE TUBE

The PYRe (3] equation fOl isotrop~ materialshas beenconsidered;

(19)

(20)

This equation is valid for DolDj <: 1.4 and it gives thelower bound ofPB among other correlations.

this correlation would not be applicable to Zr·2 pres­sure tubes, due to the anisotropy of the latter material.

However, using for Su the lowest value between axialand transverse directions, Eq. (20) gives a conservativeevaluation ofPB (4].

BURST PRESSURE FOR A DEFECTIVE TUBE WITHSINGLE AXIAL DEFECT

1he bunt pressure for a partial through thickness de­fect can be enluated u foUoWl:

NON.()UCTILE FRACTURE

, This failure mode is influenced only by a "si.ngSc" de­fect and it refers essentially to a prelSUre load.

Analytical approadles to deftne critical 10adJ; in theput hIVe not fUmiJhed acceptablt results, 10 that loadexperimental teits pesformed on different sizes and rna.terial condiUons have been uJed.

In alch a way the tube critical defect IenglhJ at dif.ferent prcssure stress conditions can be obtained.

Figure 6 &hows the critical stress trend VI temperaturefor • defmed length and the hoop Itress to temperaturecolIelation for a typical heavy ....ter reactor, wilh hoopatrea dependinl 00 atl.ln.ted steam pressure.

Curve I ahows t.he correlatjon between the hoop Itreuand the tempentute r:I the cooli.na. bated upon the factthat in lhe aaturatioa condition, pressure, and then hoopJtrel&, are related to temperature.

OJrve 2 outlines bow the critical hoop stress changes I

.. temperature iD the caae of. defect of Uligncd Length..It is emphuized the transition from the ductile behavior(the largest critical hoop stress) to the brittle ont (thesnuIlest critical hoop stress) which hu been experienced.t a value of 2OQ°C for temperature, in the cue of irradi­ated and hydrided Z,·2 tubes.

Figure 7 shows the critical length VI hoop stress..The tube integrity will be satisfaed if, for the muimwn

""reference" defect kngth., the critical stress will be ade·quately greater than the applied one.

CONSIDERATIONS ABOUT VARIATIONOF SAFETY FACTORS °8 d l-Z

a~ "" 1 -Z/M(21)

BURST PRESSURE FDA A MULTIPLE-DEFECT TUBE

TIlis situation can be conservatively estimated using theEq. (20~ where for r the corroded uniform thickness isassumed.

This law has been confumed by experimental resullS.Substituting Eq. (20) in Eq. (21), the burst pressure for

a sin&lt defective tube can be found:

l-Z .PS.d • Ps (, _Z/M) •

(22)L'_c_4Rmt

Folias' factor'" ~ + 1,61

with:

The approach to takc fretting colIosion defects' struc·lUral effecu into account las been shown. However, evepif the defects satisfy the structural verifications, a differ· I

ence in strength of a defective tube vs an undefective one Iwill ansc. ~

It is therefore useful and meaningful to calculate the Irul safety margins of a defective lube. especially apinRthe pressure bunt mode of collapse.

In additaon, because the fretting colIosion defects canchange their dimensions with time, the variation of safetyfacton VI time is of interest. In fact, such an evaluationpermits:

(a) addressing the safety margins VI tube burst for theUfe of the plant;

(b) defining a periodic in·service inspection to check:fretting corrosion defect growth, if necessary. !

The burst analysis will be performed with regard 10

GENERAL

Dj ·106.1 nun

t • 3.15 nun

a • I.I0mm

L • 2S nun

__.J ~ _

SAFETY FACTORS EVALUATION DURING THE LIFEOF THE PlANT

To consider the variation of safelY Caclon through thelife of the plant. at lean the following three different situa­tions have to be ,considered:

(-) lrUtiaillfe with unimdiatcd material;(b) intermediate We with irradiated material;(e) end~(-life with irradiated materiaJ.

For thele different IiluationJ and (ot the two type. of de­fect&" intermediate aad cnd-o{.ij{e defect dimensioos (fromFigs.. 2. 3 and -4) are associated.

On this basis. ","m;ng • known embritUcment of Zr·2due to Deutroa OuCQCt, CUlTe$ U thole Ihown 1D Fip. 8and 9 can be found.

NUMERICAL RESULTS

The above procedure has beeo applied to • referencedprcsaure tube characterized by:

inner diameter

DOminal thickness

fuel bearing padsbc;gbt

Cud bearing kngth

Itrain lwdcningcoefHc::lcnt " • 0.120 I unundiatcd mao

ultimate strength Su ·370 MP, r {erial

atnin hardening - irradiated rna-cocfncicnt n • 0.06 l terial at cnd-of.

ultimate suength Su • 480 MPa ~ life flucnee

The numerical results arc shown in Figs. 8 Ind 9 for"single" and ""multiple" (uniform eroded tube) defects,respectively.

To realize how safety bctoll cha.qge with respect tothe undefecLive situation, the safety faetq.n curves VI timefor the undefective situation ue shown, in FigL 10 and 11.

.0

CONCLUSIONS IA general methodology to account for fretting corrosion

defecu in a pressure tube for heavy water reactors has Ibeen ouilined.

Verification according to ASME Code Section IJI hasbeen mown, considering the defecuugross discontinuities.,particularly addressing buckling due to axial and torsional""'sseL

A main point is that, if the defect dimensions uc withindefUled limilJ, the II.tWactioa of Code rule. far from thediJcontinuiUelautomal.ica1ly USUlel meeting the allowablelimitata the local defective repon. --

Safety marpna IpinIt burst pressure Nove been def"med.1also takiD& into account ilTadiation embritUemcnt. On rthis bub, considerations can be made about the wrveil.lance program to control the embrittlement of the ma­terial. and the in-service inspection program to weck thefrettin& cnck growth can be done.

Additionally. indications regarding the minimum safety Ifactor at en~f·lifecan be obtained.

REFERENCES

(ll ASME Boiler MIl Prrrswr Y~~I Cod~. &criotr /II _Nucleu Power F'b.nt Componeou.K

(2) R. a Wu_U - CRNL. "Wcdlanicat Properties and Cor­rown Rewtanoe - Zirconium Alloys." Put 2, Rcv. 1 - Dee.1970.

(l) It'RC Bufletill M9$, "PYRe IIltel'P~ta.th·eReport of P~..lUre V~I RoelCarch."

(4) ASME Boiler fWi Prr$lUU "~ucl Cod~. &cdOll HI - ,"Nudcar Power Plant Components." Division J, Appcndi.x XVII. i

(S) Timoshenko, G.• Theory 0/ Ek1.sdc Stllbillry. Second Ed~

don. McGraw Hill Book Company, 1961.(6) BNsh, Almocoth, SuckliJlt 0/ SIlTS, PlJIlt:$ lind SIId/s,

McGuw Hill Book Company. 1975.(7) Column Rescacdl Committee of Japan,HlUUlbook o/urue­

nmll u.bOlry. 1971.

\

..

\

~l)'·tm,. l'r Al

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

FnrnlNG connOSIONDEFECTS: types and di-mensions deli nition..

1Two defects repre-sental ive of

~~-situation arefied.

~Single_deject;, ~~U~l~.df'i~L

[!]a unl orm re-duced Ihick - bdness_Jone 'IScons I ered.

Failure modes influenced:aj Slatic Collapseb Buckling

,/- c Non-Duel ill' Fraclure ,

Creep is considered ~Xes /I he reducIi on of thick /I hat i I produces.

/. .L'Tleference.. or "Design"defecls defi nilion ''1 or-der 10 perform the slruc-. tural verification follow-ing the ASME III Code re

quirements.

4,I

I

i,,­I ,. "-!.

--l. ------.--.__ . . __'.L

I

IFIG.1 FLOW CHART ILLUSTRATING THE APPROACH TO PURSUE INTEGRITY IASSESSMENT fOR A PRESSURE TlJBE WITH FRETTING CORROSION aeFECTS I

I

..•-_..

\

.-....-- ---.-- ..--... --.---~i=II

------ ep ------- -._--

t •_Sir.J~lLdete.d..:. _ .MJ.J1Up~_ defed.:..define .standi. define.il.. am .L

• - • -

Verification in . - Verification U1corrispondence corrispondenceto the initial life to the interme-situation (no ra- diate and end-of-dialion l. life situatons (ra-

diation consider-ed) .

FAILURE MODESto consider:

- al Static Collapse- b) Buckling

- •-RESULTS

-

tVariat ion through t he life of the Iplant of the real safety factors,considering: .-only pressure ,as most significantload·

-safely factors, as: burst pressureover design pressure.

._,""-,,i

, ,FIG. 1 (CON'T)

_ J

\

.---t­. --II

Symbols:defect depthbearing pads height

RUIINI,.t 11:- .),/-

a (m m)

.. ,,,

1.5 6 f-

0.5 6 f----

1.06

­,,.-",'f'------J

/,/

II

I/

i

\\, ,I 'i\ ', I

\:

o,

r, , ,

S' 6"flefuelings

,

, ,FIG.2 QUALITATIVE TREND VS TIME OF THE DEPTH FOR THE CASE·

OF SINGLE aEFeCT

\

,

,., .,

,- .. -I"

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

L (mm)

60

40 .

20

S.'i-mbol 5:

Defect's length

_..__ .. - .

i----+,

\

\ I

\I

I I, ,I '

iI,

,

"

o 1° 2° 3° 4° 5° 6°Ilefueti ngs

7°

•

IFIG.3 QUALITATIVE TREND VS TIME OF THE LENGTH FOR THE CASES

OF SINGLE AND MULTIPLE DEFECTS

\

t ._...

•

! 1.56

1.06

0.56

a (mm) Symbols:£ defect depth.Q. 003r ing pa::Js heigH.

\\,,i\

\

I

\-¥---I----+---j-,--+---+--+---+----+a 8 16 24 32 40 48

[(efuelings56

I :fiG.. QUALITATIVE TREND VS TIME OF THE DEPTH fOR THE CASE OF MUL­

TIPLE DEFECTS

\

Pm (MPa)

-/'

, I

"

11 5 f • • -'ue!' .__________ - _" - """ - DC JIl"'O:; _[<;:J m.g _

.'

.;... . -

----- -----,----

a~0.5s

1.1 Sm- J,QLlJi _re.fuelio9-_.

-z

l I I I I I5 10 1S 20 25

Distance (mrn)

Schernati:: representation of a defectivepressure tube

~..,

__ . __ . __ . __ . -'--' 0_' _._ _.-._---:,- -'._'-- -&.... _ _ --"-'t_

.~

FIG.6 MEMBRANE STRESS INTENSITY PATTERN IN A DEFECTIVE TUBEAFTER DIFFERENT REFUELINGS

i ,.. _--_.- ----_..._-+-'-I

,~-- - -- ~_. "-_.._-- -i .--T---- ...-.--. ----

6e (MPa).;

I

.- -- ~- - q'I"i~.·

Curve l

~ 'IS T wren P=Psalur.:lted steam (T)

Curve 2

CritiCdI hoop stress 'IS terrperalurefor a defirecJ defect lergth

50

r------------ -

2

I_____________________ J

1•

\III

oo 100

,200

T(DC)

300

I . .

IFIG.6 CORRELATION BETWEEN HOOP STRESS AND TEMPERATURE FOR A

TYPICAL HEAVY WATER REACTOR

\

I-T--"'- - _ .. :,,'~':. . .. ~.

,.

!

III,,,

400

300

200

100

o

66 (MPa)

•

50 100

L (mm)

150

\\.\

,i

FIG.7 CORRELAliON BETWEEN HOOP STRESS AND THE CRITICAL LENGTHOF THROUGH THICKNESS DEFECTS (200°C <; T <; 300°CI

\

4

3.04. 3

2

3.42with rcdiation

3.08

2,23

,,, .-

1\e fuel ing 51

o 4 8 12 16 20 24 28FIG.8 QUALITATIVE TREND OF THE SAFETY FACTOR FOR THE CASE OF A

PRESSURE ruBE WITH A SINGLE DEFECT (VS TIME)

\

4

SF: (p (p ). b d

3·71

with radiation

,\,

3·043

282WIThout radiation

1

2·04

,

r<efueli n g5

o 4 8 12 16 20 24 28,I

FIG.9 QUALITATIVE TREND OF THE SAFETY FACTOR FOR THE CASE OF APRESSURE TUBE WITH A CONTINUOUS DEFECT (VS TIME)

SF.=(p Ip )b d

4 61

337

3

2

448 ,. \,

1

,•

Ilefuelings

o 4 8 12 16 20 24 28FIG.10 QUALITATIVE TREND VS TIME OF THE SAFETY FACTOR OF AN UN­

DEFECTIVe PRESSURE TUBe SUBJECTED TO NEUTRON FLUX

SF;(p Ip )b d

4

3. J71------:3-:.:..:..3::::4~ ~

3.243

2

,,

,\

,,I

f(efuelings1

o 4 8 12 16 20 24 28

fIG.11 QUALITATIVE TREND VS TIME OF THE SAFETY FACTOR FOR ANUNDEFECTIVE PRESSURE TUBE WHICH IS NOT SUBJECTED TO

NEUTRON FLUX

\

defect

1(.

,

FIG. 12 GEOMETRY OF A HOLLOW CIRCULAR CROSS SECTION WITH A SINGLEiFRETTING CORROSION DEFECT

\

2 0

1 -5

K= 'tcril,L1:cr it.2

-Symbols-

~'t -t 1=Critical stress inen j

'C -t 2; Critical stress Inen .

plastic

elasticfield

field

)

1-0

0-5

00

k<~Lmit• ' .• •1: ;'( •

crit crit.2 ~:I

.(tm)U

--0... 'C _;'t----... Glt crit 1•

tm

FIG.13 BUCKLING DUE TO SHEAR STRESS; QUALITATIVE TREND OF THE

FACTOR K - TClit,1Ircrit.2VS (l/Al

\

---,,,

1.9Zc::

,,

a)

Zircaloy - 2pressuretube

. .::". Spacers

::::~~:?::::::::. ···.::::Fuel Elements

Fuel Element

<f)o({

~c::« :Cll ;'

II,

I, ~.-

,c-.1~(::7I,I

b)

Spacers

FIG. 14 i) SKETCH OF AGENERIC FUEL ELEMENTS ASSEMBLY (TRANSVERSAL)b) SKETCH OF A FUEl ELEMENT (LONGITUDINAL)