ML20087A473

From kanterella
Jump to navigation Jump to search
Rev 0 to Summary of Entergy Position Re Deferral of Permanent Nozzle Repair for Arkansas Nuclear One,Unit 1 Pressurizer Level Tap Nozzle, Engineering Rept
ML20087A473
Person / Time
Site: Arkansas Nuclear Entergy icon.png
Issue date: 12/20/1991
From:
ENTERGY OPERATIONS, INC.
To:
Shared Package
ML20087A472 List:
References
91-R-1017-01, 91-R-1017-01-R00, 91-R-1017-1, 91-R-1017-1-R, NUDOCS 9201090168
Download: ML20087A473 (25)
v • d • e


Text

_- ._ _ _ __

FORM 6020.005A

, REV, 0 ENGINEERING REPORT FOR ARKANSAS NUCLEAR ONE RUSSELLVILLE, ARKANSAS t

m _

0 D ~$Y y NO. DATE l REVISIONS BY CH'E APPR.

d;>,

aSu%mehe9 ein e r, En+
)m Reviewed:N/A Chk'd:Cleve Reasoner (IN /2 -204/ Approved:Cowper Chadbourn MCC @ZO-t (Print Name) (Initials) (Date) (Print Name) (Initiale) (Date)

Check if Additional Revisions:

l, .

EXECUTIVE

SUMMARY

At Arkansas Nuclear One-Unit 1, a pressurizer level tap instrument nozzle failed on December 22, 1990. The failure was attributed to Pure Water Stress Corrosion Cracking. The- 1R9 repair method-involved leaving a small area of the carbon steel pressurizer

~ ~

vessel exposed to reactor coolant. Due to the carbon steel exposure, ANO committed to the NRC to replace the nozzle during 1R10.- However, after further review ANO Design Emjineering recommended deferring a permanent repair to at ic6st 1R11. ANO Design Engineering also recommends deferring a repair for additional cycles past 1R11 if either inspections can be performed for corrosion of the carbon steel vessel or corrosion data is obtained which verifles that carbon steel degradation of the pressurizerE vessel is not a concern. By deferring the nozzle repair / replacement at least three ~ benefits may be gained: (1) improved repair / replacement methods (installation), (2) improved repair / replacement design, and (3) reduced radiation exposure.

i i

1

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

TABLE OF CONTENTS

-EXECUTIVE

SUMMARY

......... .................. ............. 1 2

1.0 PORPOSE OF DOCUMENT....................................... ,

3

2.0 INTRODUCTION

3.0 CODEWORTHINESS...........................................

6 4.0 IMPLICATIONS REGARDING MATERIAL DEGRADATION...............

6 A. CREVICE CORROSION..........................................

7 B. STRESS-CORROSION CRACKING..................................

7 C. BULK STEAM ENVIRONMENT.....................................

D. INSPECTION TECHNIQUES FOR A, B, AND C FOR FUTURE CYCLES.... 8 5.0 GENERIC IMPLICATIONS OF OTHER SENSING LINES............... 14 6.0 REASONS FOR LEAVING THE EXISTING NOZZLE INSTALLED......... 15-7.0 STATUS OF ANO'S ALLOY 600 ISSUES AND DEFERRAL OF THE NOZZLE REPAIR ............................................ 17

^

19

8.0 CONCLUSION

S...............................................

21

9.0 REFERENCES

ATTACHMENTS A. ANO-1 PRESSURIZER N9ZZLE REPAIR DRAWING B. STATUS REVIEW OF RELATED ACTIVITIES ON ALLOY 600 NOZZLE ISSUES I

i l

1 1

1 Page 1 of 21 1 l

l 1

~ . _ . . _ . - _

1.0 PURPOSE OF DOCUMENT ,

.The purpose of this document is to summarine Entergy's position regarding deferral of a permanent nozzle repair both short term and

-long term.

l i

l

)

l k.

i d

'I Page 2 of 21

2.0 INTRODUCTION

A Reactor Coolant System (RCS) leak was discovered on an upper level instrumentation nozzle. The leak was discovered at the annulus of the pressurizer shell and nozzle. Refer to the conceptual drawing, figure 1, below.

~! R

[

OUTS!DE P2R ll/l j INSIDE

/ P2R Axussn%unknux - s s

// / E

/3 LINEAR CRACK WHICH PROPACATED OUTWARD THROUCH THE NO22LE OR WELO.

A%%%uN.Xxx x x x x n'chq N022LE 0.0. : 1.**5*

/'/' ' / -

(/ // NDZ2LE TH!C(NESS

  • 0.320* l CLAODING THICXNESS
  • 3/16' Non. I'0" N!N, l

/ .

\Cle00!NC

.' Z R SHELL TH!CKNESS e 6.188* j

( // /, /

J

// /1 l P2R SHELL FIGURE 1 - NOZZLE FAILURE Page 3 of 21

a The repair method included removing a portion of the existing nozzle and instal).ing s new nozzle (alloy 600) into the prosauricor shell. A portion of the failed nozzlo remains ireside the pressurizer. The pressure boundary was moved to the outsido of the vessel. Refor to the onceptual drawjng, figure 2, be aw.

~ ~

l Matteint tx!$TINC NC71([ $916G ALLOY 681 Ntu N0Zitt 59360 PLLOY 500 IN510t PZR LHELL $A516 CR 70

& P2s WILD PAD b LtLD (N ite r. *3 RLLOW 600

/ 1 e u dsxx immm.

exisitwC Nor:Lt vis. n:w, f tp go.xw pmvesm,cs.

/

Nrw NezzLe s // ,//

6

( jl l g Pre SHtLL FIGURE 2 - IR9 N0ZZLE REPAIR This repair method allows the reactor cool.?nt be in contact with carbon stool shel.1 material (A-516, Grudo 70) via the existing crack and the gao between the old and new nozzloa Since the nozzlo no longer meets the original design intent b 1:ause of the exposed carbon stool ANO committod to repair /replae. the nozzle during 1R10. The following pages justify deferring thc repair to at least 1R11 and provide information on generic Alloy 600 issues.

Pago 4 of 21

e 4

3.0 C CEWORTN1NESH A series of evaluations and calculations qualify /just.if y the IR9 repair. Those ovaluations woro performed by Babcock and Wilcox Nuclear Services (BWNS). Isbstracts of key evaluations which '

qualify / justify the nozzle for operation until 1R11 are shown

~

below:

1. BWNS document # 51-1202276-01: The evaluation of potential matorial degradation due to prossurizor steam being in contact with the carbon stool base material concludes only minimal

- corrosion /orosion and no hydrogen ombrittlement will occur.

2. BWNS document # 32-1202246-00 This structural calculation -i concludes that the replacement nozzle and now wold moet ASME i Codo requirements and are equal in strength to the original l

' design. i

! 3. BWNS document # 51-1202275-00: Leak-Defore-Break evaluation of ANO-1 pressurizor instrument nozzles reviews similar f ailures ,

in U.S. and Europe. Failure modo is leakogo versus catastrophic failure.

4. BWNS document .# 32-1202270-01: This fracturo mechanics / "

Section XI evaluation of an assumed flaw in the prosaurizor shall concludes that the flaw, if ono exists, will not grow significantly. Wido margin is demonstrated.

5. 'BWNS document # 51-1202274-01: Nonlo ovaluation summary concludes that the pressurizer structural integrity should be main *.ained for the design life- of the plant provided the recommendea examinations of the repaired nozzlo and adjacent baso material are performed and that no significant corrosion la detected in thoso exatpinations. If ANO is unsuccessful in developing NDE techniques for - examining the. repaiced penetration by 1R10, DWNS concludos that operating with the 1R9 pressurizer nozzio repair for one additional fuel cycle wos1d not comprotsise the structural integrity of the component 1 and would not be a safety concern <
6. BWNS document # 33-126f648-00: Stress report for_ANO-1 Prosaurizer lovel sensing nozzle.

Those and other ovaluationu amend the original stress calculations and stress report.

Pago 5 of 21

f l

l 4.0 IMPl.ICATIONS HEGAllDING MATEitI AL DEGRADATION j t

A. GENERAL /CHEVICE CORH0d10N The BWNS material degradation ovaluaticn justifios/quallfles exposure of the carbon stool shall material f or. two refueling with

~

cycles without inspections and for additional cyclos inspections. Addressed in the ovaluation is crovice corrosion, boric acid corrosion, general and pitting corrosion, galvanic corrosion, hydrogon ombrittlement, orosion corrosion, and stress corrosion cracking (SCC) of the now and old Alloy 600 nozzlos. The evaluation considers crevice corrosion and boric acid corrosion to be the most important types of carbon stool material degradation  !

for consideration. BWNS predicted that a maximum corrosion rato of 35 mila depth por operating cycle and shutdown combination, but BWNS also concluded that the actual corrosion rato should be significantly less.

The crevice corrosion and boric acid corrosion concerns woro also discussed with other Engineering firms, such ar, MPR, Structural Integrity, and Combustion Engineering. Based on those discussions, DWNS ovaluation, and CEOG corrosion testing the following corrosion scenario is expected:

Due to tho'.high temperature and the limited supply of oxygen the corrosion rato is expected to be minimal. The crevice will most likely fill up with corrosion particles (iron oxido) and soal the crovico, therefore, 011minating or significantly reducing any further corrosion. The experiments that BWHS referred to had borated water flowing through the crovico, but for ANO's caso the crevice would be in a relatively stagnant environment. Thorofore, the corrosion deposits would not be subject to washing (ut of the crevice area. In conclusion, the crovice would be filled with '

corrosion particles much quickor, and the corrosion rate would be much lower than if a flow path existed.

The conclusion of the above evaluation can be further verified by ,

the CEOG's corrosion testing (Task 637 -

Corrosion And Corrosion / Erosion Testing of the Pressurizer Shell Material Exposed to Borated Water). The test involved making mock-ups of a nozzlo and shell. The crack length ranged from .094 to .469 inchos. The leak path was in the annulus betwoon the simulated shell and nozzle. The test was performed at 600 degrees and 2250 poi for up to 27 days. The conclusions worth noting for AN0's nozzio are:

1. The corrosion rate at the cracked area inside the annulus was negligible. In fact tho machino marks could still be soon on somo of the carbon steel shell material.
2. The maximum corrosion occurred on the outsido diameter of the shol) in a localized area.

' age 6 of 21

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

4

3. The leak rate was reduced by the corronion deposits.
4. Alloy 600 was not significantly attacked by this type of leakage.

It is apparont f rom conclusion number 1 that the lack of oxygen and

~

the high temperaturo reduced the amount of corrosion to almost zero, and from conclusion number 2, that the oxygon in the atmosphere and the lower temperature on the outsido of the shell accolorated the corrosion rato in the localized area.

In summary, if the carbon steel material does corrode in the crevice, the crovice will eventually b9 filled with iron oxide and stop or greatJy reduce any further reactions. The reactor coolant being in contact with the carbon stool base material of the shell should not be a concern for the romaining life of the plant.

Attoinpts will be made to verify this conclusion with UT inspectione. Should significant material degradation bo detected, the nozzia will be repaired or replaced. If the UT inspection or othot inspecting methods are not credible or it can not be proven that corrosion is not a concern via corrosion testing or other means, the nozzle will be replaced / repaired during 1Rll.

H. STRESS CORROSION CRACKING Stross corrosi >n cracking is considered to bo incignificant for the carbon stool s." ell . However, if it is assumed that all proporties and fabricatiot, methods are the samo, the now Alloy 600 nozzle could be more susceptible to PWSCC than the original nozzle since the now nozzle wasn't stress relieved after installation Ln 1990.

B&W is the only vendor who stress relieves their pressurizar vessel after nozzle instellation. Thorofore, as far as PWSCC is concerned, this repair is equivalent to the original design of other vendors pressurizor's with Alloy 600 material. Refer to attachment B for industry activities on Alloy 600 nozzlo issues.

C. BULK (STEAM ENVIRONMENT)

AN0's RCS specification requires the oxygen content to be less than 100 ppb, but the oxygen content is usually below the detectablo limit, which is approximately 5 ppb. Two samples in the steam space have boon taken in the past. These samples were taken during a hydrogen peroxido flush due to the concern of a high oxygen lovel content. The oxygen content of one of the samples was above the detectable limit but below tho specification limit. The environment at the gaped sroa between the now and old nozzlo is expected to be the same as the bulk steam since the gap is loss than 7 inches away from the ID of the pressurizer. Sinco thoro is only a very limited supply of oxygen, the corrosion caused by the boric acid is expected to be negligible.

Page 7 of 21

D. INSPECTION TECilNIQUES FOR A, D, AND C, FOR FUTURE CYCLES

  • Inspection Techniques for A (general corroulon)

Southwest Research Instituto. is prosently in the process of

~

developing an Ultrasonic Inspection Technique to attempt to datormino if the carbon stool base metal has corroded due to contact with the pressurizor steam space which contains some boric acid. Although the amount of corrosion is predicted to ,

be negligible, Entergy Operations bo11ovos it is prudent to validate this prediction. In order to develop an inspection technique a mock-up of the IR9 repaired nozzle was built.

The mock-up is of the pressurizer vessel wall, inconel pad, and the structural weld. The pressurizor vessel wall was simulated by using a 12 by 16 inch block of the same type of material, SA516 Grado 70 carbon stool, 6 inchos thick. An Inconel weld pad was deposited on the carbon steel block using the same type of wold rod, ENiCrFo-3 (Alloy 182), with a similar wold technique as used in the IR9 repair. Each layer was ground suitable for dye ponotrant testing (PT) for ,

indications. All PT indications woro removed by grinding before proconding to the next layer. Three UT inspections woro performed at various stages of the pad build-up to insure that no reflectors woro present. Figuros 3 and 4 on pago 9 are cepios of pictures taken of the IR9 repair wcld pad and the mock-up wold pad respectively.

I Page 8 of 21 l

. , - _ _ . . _ -- . - __ , _ _ , _ x., _ . _ _ , _ _ _ . . . . , _ _ _ _ . . _ _ _ _ . _ _ . - _ . . - . . _ , . , . - . _ _

,._.I

m m. , ng,7, , ,g y

\' 4:_;

, s *8 -c Si .#.,. . . ., ,-

p .

3- .4 A. 4, 3

.$ 't?'

3 i\..

'I.

.r3 4 L, $ $

", rij yw. .%

t. %w s

% w+. . y/. 4

-)>

Q h

% , h*$ *).

.s e w wes

.. 'g/,

u u ;,l d w-so, _

,y l'IGURE 3 - IIt9 INCONI'L 1%D

~- ,.e% ,

. 1 1

o

  • - a f' .

/ '

s w

. 4.g,, ..

.:, ..~.

y f'l.',,,

k . .

, J. . n

&:j ~

-n2 '

i. a _,.. i . ., x _ --

I'IGullE 4 - MOCK-UD INCONEL PAD Page 9 of 21

The J ytouve for the htzucturcil weld (see figure 5) was not Itachined litt o the wcld pad of the inoc k - u p since it wii r. riot riecussary to create a borid between the pad and no ctle, f

'f:

6

+ ,

, . s 4

(

V

f. ' .
i-

.s 3 1' , f%

g

.)$ ; R$

' Sy[f

+3 x

h. \;

44 t N*- r o

+

s in e- a as.we . v .- -. . j ., s .., ally FIGUllE 5 - J GitOOVE ON lit 9 1(EPAllt A hole was machined t h r ou .;h the weld pad and carbon steel material the same diameter as the original bore on t. h e pressurl:er (see figure 6 page 1I).

t E

Pago 10 of 21

r r .',.

(# # .'

q s .

% Ap N  %.

" y Y

,a c yx

..a x .

t l'IGUlti; 6 - MOCK-UP 110RI A stainless stee] tube was placed into the hole approxirnately 1 inch, and a f1110t weld was made on ihe pad the c arne height and width as the IR9 repair. The tube was then cut off flush wi t h the iillet weld (see figure 7).

r e

nq ,

!((

i

.us u -;.

l'IGUHE 7 1'I LIET WELD MOCK-UP The diarneter of the hole in t.h e carbon steel vessel was Pogo 11 of 21

l_ _ _ _ _ _ . _ _ _ _

i slightly larger ior the replacen.ent nottle than the original

! nozzle. Therefore, a hole tho came diamotor and depth was machined into the structural weld, weld pad, and carbon stool .

A radius of curvature was machined into the int ersection of l

the two difforent diamotor bores the same as the original

repair. The stainless steel t ubing was then removed f rom the i

boro. The mock-up was then cut in half, so that, I ndica t.l ons l representing corrosion flaws cauld be inserted into the carbon i stool near the gapped aree where the two nozzlos meet (retor to figure 8).

l 1

l 6

l -o y,  ::

3 3; J e d l 6 ' [ J: .

g i .

'! f

! 1 i

! , o-  !

T;

'l 5tk f l\

4 )y %', J,,

g , l-.j ; ? y yn l

a .f l .

! i 1 '

l

( i i I t

i '

! i i FIGURE 8 - PRESSURIZER SilELL MOCK-UP

\

l ,

l The mock-up was delivered to Southwest Research Institute on j December 2, 1991. Preliminary testing shows that the sound i i

waves distort when attempting to pass through the Inconal wold pad build-up and structural wold. Thoroforo, it is highly un11 holy that it will be possible to perform 100% inspection ,

on the entiro carbon steel vessel surrounding the no: le (soo  !

Attachment A). An alternativo method of UT examination will ,
l. be performed by placing approximately a 75 degreo angle  !

transducer on various positions on the pressuricor vossal wall .

outsido the wold pad. By using this method, a small area f above the gap will not be tested, but a reasonable assumption is the worst area whero corrosion could occur will bo in the  !

gap area. Therefore, only one half inch abovo and below the

! gapped arca will be inspected. Due to the angle at which UT l oxamination will be performed thoro is somo question as to the Pago 12 of 21 j r

l l

accuracy with which the corrosion depth could be measured, but based on preliminary results well below a 1/4" corrosion depth can be detected around the outor diamotor of the nozzle.

To overlap any potential detection limit, BWNS has proven that a conservativo 1/2" corrosion depth 360 degreco around the

~

nozzle and the entire longth of the nozzle would not cause the nozzlo to blow out (DWNS document 32-1206678-00).

Ef forts are being made to be able to dotect a 20 mil corrosion depth and to determine the rango of the depth betwoon 20 and 100 mila at 20 mil increments. Due to a possibility of some unknown variables in tho field, such as, if wator condonson in the annulus area during shut down, the noise lovel could cause some level of difficulty in reading the UT results.

Thorofore, ability to perform the UT inspection will not be fully proven until the inspection technique is attempted under actual field conditions.

If attempts to quantify and validato the expected low corrosion rato during 1H10 or 1R11 aro unsuccessful, then ANO will install /ropair the nozzio in 1R11, unless more experimontal data on boric acid corrosion under those conditiono is located or developed.

  • Inspection Techniques for us (stress corronion cracking)

A visual inspection with the RCS pressurized will be used to datormino if a crack has occurred due to PWSCC. In addition, the Doric Mid Inspection Program is in the process of being revised to insure the pressurizar instrument nozzios in the water space were specifically identified as areas requiring inspection overy refueling outago. Current plans are to inspect steam spacs nozzios at pressure at the start of overy refueling outage. The steam spaco nozzles are not being included in the Boric Acid Inspection Program due to concern that the concentration of boric acid in the steam spaco may be too low for this to be a rollable leak detection method.

  • Inspection Techniques for C: (steam sp6;o chemistry)

As f ar as testing the bulk steam environment, two samples have benn taken in the past. One of the tests was above the detectablo limit, 5 ppb. However, since these tosts woro taken when the o 'gon concentration should have boon at a peak, Entorgy Opo.ations bollovos no further testing should be required because the oxygen content is always hold below 100 ppb, is of ton below the detectablo limit of 5 ppb, and is thoroforo not considered adequato to cause crovico corrosion.

Even if the content of oxygen in the steam space is significantly higher, it should still bo below any levol that would result in significant corrosion occurring.

Page 13 of 21

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

i i

5.0 GENERIC IMPl.lCATIONS FOR OTilER SENSING 141 Nim The generic implications are being studied through the B&W Owner's Group (B&WOG) activities. To dato, it is not known if PWSCC is a gonoric problem for H&W plants. ANO-1 has the '

~

first B&W pressuri>nr to experienco a nozzlo failuro due to PWSCC. Prior t o ANO's failuro, it was bolioved that the stress relieving process used by D&W after the nozzlo installation substantially reduced the probability of f ailuro duo to PWSCC. Even after AN0's failuro it is still belloved and proven in tests that stress ro11oving reduces the susceptibility to PWSCC. The industry boliovos that if the material is susceptible to PWSCC then stross relieving will prolong the time beforo SCC occurs. Whether this increased time is 5, 10, or 15 years is not known. Though it is not known at this time if there is a gonoric problem with AN0's other sensing nozzlos, Entorgy is troating the other nozzles as if there is a potential generic problem. All of the instrument nozzlos were visually inspected at pressure during the 1R9 and the 1M91 outages, and plans for future insp6ctions during 1R10 are being formulated.

1 I

Page 14 of 21

i 6.0 REASONS PCR LEAVING Tile EXISTING N0ZZLH INSTALLED j

a. The material and process used (besides B&W stress relioved pressurizers) in the IR9 repair is typical of nuclear industry l Alloy 600 penetrations (with the exception of the carbon stool j vessel being exposed to reactor coolant). f
b. This repair should last for at least soverni cycles, and .

possibly for the life of the plant.  ;

c. Soveral industry activities are attempting to rank the suscoptibility of various inconal components to PWSCC failure. ,

Should this repaired nozzle and/or other consing nozzlos be determined to be highly susceptible to PWSCC, it would be more cost offective to replace several nozzlos at a timo.

d. DWNS's and CE's repair method places a seal wuld on the ID of the pressurizer shell and nozzle as well as a structural wold on the OD of the pressurizer shell and nozzle. This repair method may increase the residual stressos imposed on the nozzle due to the shrinkage of the two wolds. Alloy 690 ,

which would be used for the replacement material is loss susceptibio to PWSCC, but increasing the stressos increases  ;

the susceptibility to PWSCC. BWNS believes that a repair method of t.his typo would last the life of the plant.

liowever, there is very little operating experience for Alloy 690, and using a one wo3d design versus a two wold design 3 would reduce.the susceptibility to PWSCC. 1

o. Should the corrosion rate be minimal as expected, ANO's temporary repair may be a viable alternativo to current BWNS and CE permanent repair methods. Using a repair method that places a weld only on the outside of the vossol as in the IR9 repair method requires less radiation exposure.  ;
f. Other replacement methods involvo entering the prosaurizer with one or two mon in order to install a structural wold on the ID of the vessel. A repair method of this type would be a significant ALARA concern.
g. Thoro is a possibility of damaging the heater bundlesany or cladding by dropping tools or debris during repair / replacement process,
h. Detter repair methods are expected in the future; such as, a nickel plating technique which has been-used in France for steam generator tubing repairs. Also, botter tooling for machining, cutting, and welding are being developed, which will reduce the man-rem costs of this repair.
1. By not replacing the nozzle this outago, the total radiation exposure for 1992 would be reduced. If it is later Page 15 of 21 3

e--, g g,-- 7p---- - r,--- , ,,- , ~ , - , ,-w-+ - - , - ,- e , a

i, i

l decided to replace several nozzlos , then the cumulativo dose will be lower than if this nozzlo is replaced separately in IR10 or 1R11.

r

i i

i I

t

-I

. I 6

?

l I

i l

t I

Page 16 of 21  ;

) ' . j a c , . - .

...:..--.-.-..---..o ...--. -- - -- - i u.--u..-. i.-- - - -

7.0 STATUS OF AN0'S ALLOY 600 ISSUES AND DEFERRAL OF Tile N0ZZLE REPAIR ANO's current position is that the Alloy 600 issue applies to our B&W unit as well as to our CE unit.

Activities completed:

Completed visual inspections of steam space nozzles at pressure 6Jring 1R9 and 1M91 Prscured Alloy 690 replacement material, and developing a generic seal weld procedure Issued recommendations f or revisions to Boric Acid Inspection Program Attended and made presentation at EPRI Alloy 600 meeting Completed revisions of evaluations and calculations to reflect the acceptance of an additional operating cycle without mandatory inspection for corrocion or several operating cycles with inspection for corrosion Fabricated mock-up for use in UT qualification process

}

Developing an Ultrasonic Inspection technique and procedure 1992 Plan Attempt UT or other inspection methods to quantify corrosion depths Perform visual inspections of this and other Alloy 600 nozzles Attend additional industry meetings on Alloy 600 issues, and utilize industry information to continue development of long term resolutions to the Alloy 600 issues Complete qualification of seal weld procedure Page 17 of 21

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

1993. Plan  ;

i Expected opt. ions to be available in 1993:

Caso It Inspection techniques provo unsuccessful during 1R10 and -

1R11 l g

-Install a new nozzle in IR11, unless more experimental  ;

data on boric acid corrosion under those conditions is located i or developed .

e l Caso II: Inspection techn quos prove acceptable and test results 4 are favorable i

-Dovelop periodic inspection plan in lieu of immediato nozzlo  !

replacement  ;

i S

Caso III: Inspection techniques prove acceptable and test results '

are unfavorable

-Repair or replace nozzlo during 1R11 i

F 1

B l

i I

Page 18 of 21 .

w& rw m~w-w,e--eve- +-

O.0 CONCbdSIONS A reactor coolant leak occurred on an ANO-1 pressurizor instrument level tap nozzle on December 22, 1990. The cause of failure was attributed to Puro Water Stross Corrosion Cracking (PWSCC). The method of repair caused a small area of the carbon stool

^

pressurizer vessel to be exposed to reactor coolant. Sinco the nozzio no longer meets the original design intent due to the exposed carbon stoel ANO committed to the NRC to replace the nozzle IR10. However, following further review of the IR9 repair, ANO Design Engineering recommended deferring the permanent repair for at least one additional refueling cyclo. The documents and items considered in the review consisted of evaluations / analyses, corrosion inspection techniques, corrosion tost data, Alloy 600 issues, ALARA, and current repair methods.

The ovaluations/analysos support deferral to 1R11 without inspection for corrosion as well as for additional cycles with corrosion inspection. Doferral of a permanent repair for additional cycles past 1R11 will be dependent on inspection results or other corrosion test data. Inspection techniques will be attempted initially, but if inspection techniques fall then add.tional corrosion test data may be located or developed to justify further cycles. Preliminary results of an ultrasonic inspection' technique performed on a mock-up indicate that the pressurizer che11 can be inspected for corrosion. However, the nozzio will be replaced during 1R11 if inspection techniques provo unacceptable and/or corrosion results of test data prove unacceptable.

The IR9 Alloy 600 replacement nozzle was fabricated and installed with similar. methods used throughout the industry during initial pressurizor fabrication (with the exception of Br.W's stress re11oving process). The Alloy 600 material is also at least  ;

equivalent to the average matorial used in the industry during initial fabrication. Therefore, thero is a possibility that the now nozzio may f all due to PWSCC, but t ae time in which f ailure may occur is possibly 10 to 20 years from now. The nozzle will most likely be replaced / repaired before-this time. Even if a the nozzle were to fall due to PWSCC it is not a safety concern since the nozzlo would crack in the axial direction causing only minimal reactor coolant leakage and would not result in a catastrophic failure.

Based on the above considerations, repair or replacoment-of the nozzle is fully justified-to be-deferred at-least until 1R11 and possibly several _ additional cycles. By deterring the nozzio repair / replacement the total radiation dose will be reduced in 1992 and if it is later decided to replace several nozzles, the cumulative dose will be lower than if this _ nozzlo is_ replaced separately in 1R10 or 1R11.

Page 19 of 21

t BWNS's permanent. repair method involves placing a seal wold on the ID of tho ' pressurizer and nozzle as well as a structural wold on tho OD of the pressurizer and nozzlo. The curient BFNS repair mothed may induco higher residual stresses due to shrinkage of the two wolds. Alloy 690 is loss susceptibio to PWSCC than Alloy 600, but increasing the stresses on the nozzle will also increase the susceptibility to PWSCC. Therefore, other repair methods may be proferred. At present the other repair methods would involve sending one to two m<in in the pressurizer in order to place the structural wold on the inside of the pressurizer. A repair method of this type would be a high ALARA concern. Though thoro are methods to replace the nozzlo it may be beneficial to wait until other technology can be developed which could reduce the residual and thermal stresses, and ALARA concerns.

In- summary- based on the evaluations /analysos, corrosion inspection techniques, corrosion test data, Alloy 600 issues, TLARA, and current repair methode; Engineering recommends deferring the permanent repair / replacement of the nozzle. By deferring the nozzle repair / replacement at least throo bonofits may be gained:

(1) improved repair / replacement methods (installation), (2) improved repair / replacement design, and (3) reduced radiation exposure.

e Page 20 of 21

__ __- = ____

1 9.0 HEFERENCES

1) BWNS document. No. 51-1202276-01, ANO docuinent No. 86-E-0074-88:

Materials Evaluation of the ANO-1 Interim Lovel Tap Hopair

^ ^

2) BWNS_ document 32-1202246-00, ANO document No.86-E-0074-85: ANO  !

Pressurizor Level Sonsing Nozzio Evaluation  !

3) BWNS document 51-1202275-00, ANO document No.86-E-0074-87 : ANO-1 Pressurizer Nozzle Lovel Tap LDB Assessment
4) BWNS document 32-1202278-01, ANO document No.86-E-0074-90: AND PZR Lovel Sonning Nozzio FM Evaluation
5) BWNS document 51-1202274-01, ANO document No.86-E-0074-86: ANO-1 Pressurizer Lovel Sensing Nozzle Leak Evaluation Summary ,
6) BWNS document 33-1206648-00, ANO document No.86-E-0074-101:

1 Stross Report for ANO-1 Pressurizer Levol Sensing Nozzio

7) BWNS document 32-1206678-00, ANO document No.86-E-0074-102: PZH Level Sensing Nozzio Cavity Assessment
8) -CE doc'umont CE-NPSD-648 P: Corrosion and Corrosion / Erosion Testing of Pressurizor Shell Material Exposed to Borated Water I

i I

Page 21 of 21 m 7-'erege-+-e r-'W g*Wri-wy ww- in-iMggyWW-----Pg 71*wmy "epn,+pg

. ' .e.M4h44=.-S.*4A + 4.mwA A 4 _ M.enfu k k A4 uA p.~ + M A4 m.w 4 % s. 4 44.e 4 A.o.-+A44CJ d 0 -aw 44 Ad 4dM,.4d.4.Aq

'g

,.-,----y-- l \ f~*7

/'/'// l / / /,,  !

,/ / // -

.<~

O

~ ~ ,

. 4

. .- s T

. w.,

f E

Y v i

=

/ '

f .,

c

\ hk mm J/ /hh.!b' , 4 .

[y- _ _ .

. g'lglll

,] {l Il= [ - - - - -

l .

tuusuuuus  ; i a

/

i g ,

//

1 '

fi

  1. ')/'

/

/ /ew/, ,9/

e

-!/  ;

$ 73NGHJ *13551A V g.g3nciat ow;ooyt t .vwe 3v ,,itt

. - - , . - . . . . . . ~ _ , . .,._,~...,._,__,._,,,,._.,,.._,..,_,,,-.,.n.,n,-_,-,.,-,.,,,,,.-

STATUS REVIEW OF RELATED ACTIVITIES ON ALLOY 600 N0ZZLE ISSUES A. D&W OWNERS GROUP The B&WOG Materials Committoo is in the process of  !

preparing for a nozzle replacement should another nozzlo failure occur. The committee has procured enough Alloy 690 material to fabricate 10 nozzle assemblies and mock-ups. BWNS is developing a generic seal wold proceduro l which could be used by all B&W plants. The wold attachos  ;

a sloovo to the insido of the pressurizor. BWNS is also '

evaluating their half-bead /tempor-boad wold procedure to soo if it could be used for all the B&W plants at all the ,

pressurizer ponotration locations. Several other '

activities related to the Inconal concern are being discussed for 1992, but these activities are not yet funded.

B. CE OWNERS GROUP The CE Owners Group developed an Alloy 600 Working Group to evaluate PWSCC of Alloy 600 primary prosauro boundary penetrations in CECt.1 plants with the respect to busceptibility, saf ety, operations, and maintenanco. The Working Grov;p. compiled all the data concerning f abrication proconses .and material characteristics of those compononts. CE conducted a review of the design, materials data, and fabrication history of the Alloy 600 heater sleeves. CEN-393-P was submitted to the NRC November 17, 1989 which concluded that three olomonts were necessary for PWSCC to occurs an aggressivo environmont, a susceptibio metallurgical material, and high tensiin stressos. The report also ranked the susceptibility for PWSCC to occur in the heator sleoves based on the above critoria. CE has performed further failure, susceptibility, and safety significance analyses. The results of these analysos were submitted to the NRC May 31, 1991 in report CEN-406-P.

l ATTACHMENT B 31-R-1017-01 Pago 1 of 2 l

i t

=

i l ,

4 C. EPRI ACTIVITIES ON PWSCC OF alloy 600 PENETHATIONS (Hof. Charlot.Lo, NC mooting Octe 1991)

Completed activiticut f Initiated a now research project (RP 3223) on sontrol of PWSCC of alloy 600 penetrations in PWRs e s av e a survey on fiold exporlonco with PWSCC of eluss J00 panotrations /EPRI NP-7094) 9 Nadt.od a Coordinating Group to exchango 1 irddrmation and coordinate rosearch on this issuo l 1%1 Phin

@otturviino residual stresses in soveral alloy 600 moch-ups experimentally to confirm the analytical

  • mruita Initiat.ed accolorated SCC tests on alloy 600 mock-ops to evaluato materials susceptibility and to qualify remedial methods Organizo a utility workshop on PWSCC of alloy 600 penetrations (October 9-11, Charlotto, NC)  ;

(Note: Entergy was represented at this mooting.)

Consolidated available information on alloy 600 penetrations in a data baso 1992 Plan Con',inue the SCC test program to evaluate material susceptibility and to qualify remedial methods Develop a methodology to assess- the PWSCC tausceptibility of alloy 600 panotrations in PWRS Coordinato EPRI activities with those of CE, B&W and Westinghouse Owners Groups as well as EDF and I other Foreign utilitios ATTACHMENT B 91-R-1017-01 Page 2 of 2 l

i

Retrieved from "https://kanterella.com/w/index.php?title=ML20087A473&oldid=2737185"