ML20153E148

From kanterella
Jump to navigation Jump to search
Forwards Info Re Byron-Jackson Co Recirculation Pump Shaft & Cover Cracking,Per 880629 Request Re Relief Requests for Inservice Insp Program.No Maint Action Should Be Required Before Unit 2 Startup
ML20153E148
Person / Time
Site: Browns Ferry  Tennessee Valley Authority icon.png
Issue date: 08/29/1988
From: Gridley R
TENNESSEE VALLEY AUTHORITY
To:
NRC OFFICE OF ADMINISTRATION & RESOURCES MANAGEMENT (ARM)
References
NUDOCS 8809060224
Download: ML20153E148 (15)


Text

^

i TENNESSEE VALLEY AUTHORITY CHATTANOOGA. TENNESSEE 37401 SN 157B Lookout Place AUS 291988 U.S. Nuclear Regulatory Comission ATTW: Document. Control Desk Washington, D.C.

20555 Centlement in the Matter of

)

Docket Nos. 50-259 Tennessee Valley Authority

)

50-260 L

50-296 i

BMOWNS FERRY WUCLEAR PLANT (oFN) - BYRON JACKSON COMPANY RECIRCULATION PUMP SHAFT AND COVER CRACKING

]

During ' teleconference with merbers of your staff on June 29, 1988, concetm..) relief requests for the Inservice Inspection program, there was a j

discusslu. of the status of the recirculation pump shafts and covers.

The staff requested that TVA provide the'Ceneral Electric SIL Wo. 454 entitled

[

"Byron Jackson Recirculation Pump Shaft and Cover Cracking" dated December 15, 1987. TVA was asked to report on the operatinf, history of the i

2 reeleculation ' pumps and to show evidence that the pumps could operate without l

additional risk untLL the pump shafts could be inspected and/or replaced.

Records indleate that BYU's unit 2 has 57,399 hours0.00462 days <br />0.111 hours <br />6.597222e-4 weeks <br />1.518195e-4 months <br /> tied-to-line.

By adding a i

10 porcent factor for other hours, where the punp would be at or near i

i operating temperature, and an additional 18-month operating cycle, total hot.rs gre approximately 70,000.

Unit 2's recirculation pumps would still remain below the lovels stated for recomended inspection af ter the next operating cycle.

1 l

Based upon the stated operating time and the attached report, no malntenance i

action should be required before unit 2 startup. TVA is evaluating both vendor and utLitty findings to develop our plan of action af ter unLt 2 startup.

i Ple4Je refer any questions regarding this matter to Pattlek Carter at (205) 129-2689.

4 Very truly yours, TENNESSEE VALLEY AUT110RITY

\\

1 id1h at A..: at Licensing and Regulatory Affairs Enclosure i

cc: See page 2 8G09060224 880929

(

1 PDR ADOCK 05000259 I

1 p

PDC An Equal Opportunity Cmployer

l

?\\t)8 291968 U.S. 'Wuclear Regulatory Commission cc (Rnclosure):

Ms. 3. C. Black, Assistant Director for Trojects TVA Projects Division U.S. Nuclear Regulatory Commission l

One White Flint, North 11555 Rockville Pike Rockville, Maryland 20852

(

Mr. F. R. McCoy, Assistant Director for Inspection Programs e

TVA Projects Division U.S. Nuclear Regulatory Commission Region 11 101 Marietta Street, NW, suite 2900 Atlanta, Georgia 10323 l

l Browns Ferry Resident Inspector l

Browns Ferry Nuclear Plant Route 12, Box 637 hthens, Alabama 35611 l

j i

, ( j". 7 4

1 BYRON JACKSON REC lRCULATION PUMP SHAFT & COVER CRACKING

REFERENCE:

GENERAL ELECTRIC SIL NO. 459 i

Background

3 j

Browns Ferry's reactor recirculation pumps were manufactured by the Byron Jackson Company (B J). The mechanical seal design used on Browns Ferry's pumps is common to that used on a number of PWR reactor coolant i

pumps and BWR react &r recirculation pumps manufactured by B J. Some have been disassembled and inspected during the past few years because of the length of time the pumps have been in operation, ASME Section XIISI requirements, and some operational problems. Thermal fatigue cracking has been observed in several pump shafts and covers, i

The initiation of the thermal fatigue cracks is due to.the mix!ng of cold mechanical seal injection water with hot reactor system water near the

)j bottom of the thermal barrier. In this turbulent mixing region, the material surfaces are exposed to rapidly varying water temperatures I

1 which cause alternating stresses. The frequegcy of this dynamic action

!j can vary from about 1 Hz to 25 Hz. Thus a large number of stress cycles can occur in a short operating time and fatigue cracking of the material l

surface can result.

)

i

)

The problem was first noted in 1980 and at the time was considered to be relatively benign. Calculations indicated the cracks would exhibit virtual i

ar.'est or show very small growth for depths exceeding 0.2 inches. Cracks were also expected to be limited to a one inch band at the bottom of the a

i thermal barrier (See Figure 1).

1

)

Discussion

-- - Recent ISI activities have shown that some cracks have penetrated to

~

depths greater than previously anticipated. For one BWR pump cover the cracks were found with a depth of 0.310 inches. On this cover and a second 2

cover for which depth measurements wrv lot available, crack propagation i

extended upward into the cover bore te u alevation coincident with the j

bottom of the water jacket holes (above md one inch band; See Figure 12 &

i 14). Material thickness between the cover bore and waterJacket hole is 0.280 inches.

4 A

If these axial cracks were at the same azimuths as the wa.ter jacket holes (See Figure 13) RBCCW could then be contaminated and possibly subjected to higher pressures. While initialleakage wo'uld be small, rate of leakage and rate of increase would not lend itself to analytical predictions. No such occurrences have been reported.

Recent destructive examination of one BWR 4 shaft, showed axial fatigue cracks on the surface of the grooved region (See Figure 17 & 18). At a depth between 0.100 - 0.200 inches, these cracks turn in circumferential directions. On this shaft portion circumferential crack lengths up to three quaders of an inch were measured. Circumferential cracks of this nature are of concern because they are not readily observed and growth potential is increased by normal operating loads. No such failures have been reported.

Recommended Corrective Action From General Electric SIL NO. 459(copy attached), GE recommends:

" 1. Review pump and system design, period.of operation and maintenance history. Correlate with other fie!d experience tb assess the potential Impact on the pump or pumps in question. For pumps with a high potential for significant cracking, the following actions should be considered:

Install shaft probes to monitor shaft vibration response to protect against consequential damage in the event crack propagation changes from thermally driven to mechanically driven.

Monitor CCW effluent for contamination by reactor coolant to provide early warning and to minimize contamination of external CCW system and equipment in the event communication develops between these systems.

1

2. Review ASME Section XI ISI plans and commitments and establish plans which consider the recent findings described in this SIL. Schedule inspections of pumps with greater than 80,000 hours0 days <br />0 hours <br />0 weeks <br />0 months <br /> of pump operation to detect cracks and implement corrective actions before excessive crack growth occurs.
3. Prepare an inspection plan and develop the following:

4

Methods for examining pump shafts and covers Criteria for return to service Repair methods and alternatives considering ALARA Plans for rept... ont of existing parts or use of improved designs" 5

GE's definition of "pumps with a high potential for significant cracking" are those that are due for 10 year ISI (which is equal to approximately I

80,000 hours0 days <br />0 hours <br />0 weeks <br />0 months <br />). Records indicate Browns Ferry's Unit 2 has 57,399 hours0.00462 days <br />0.111 hours <br />6.597222e-4 weeks <br />1.518195e-4 months <br /> tied to line. By adding a 10% factor for other hours at or near operating temperature and an additional 18 month operating cycle, total hours is j

approximately 76,000. Unit 2 recirculation pumps would still be below i

the levels stated for recommended inspe.ction after t,he next operating cycle (See Figure 16 ).

Based on the above and the fact that no related failures of either the pump covur or shaft have been reported, even with (nany plants with similar pumps in operation longer than Browns Ferry such as Dresden with 18 years, Mechanical Maintenance recommends the pumps be operated the next l

cycle without inspection.

l For further info contact John Woodward x3168.

i Sources for above information include "Technical Information Exchange Thermal Fatigue Cracking on Byron Jackson Recirculation Pump Cover and

{

Shaft October 1987;GE SIL NO 459.

1 l

l b

i

/

ltr~~ j' N

T s.1 b

[

]

~

n '

1 7

}

Moto'r Hall Coupling

,t-L Y < l'

<l,; L.

N KEY:

~

3 Primary Loop Water N

s

@ Temporalwre. {

l

~

s 3 Coolec. Primary Lsep Water s

s s

Cl component coolog W ate'

)

(

Coupling Spacer N

M Q{' t $'N

-Pump Hall Coupling N

(

W '11 p"

  • <]

'?

l y

Shatt N

N y

'1 l

g j', J

,l

- Coohag water owilet Cocio, vater inie t w

g, D h b $

mg",

,b a

m n=

~

r L..)

E3JII _y L

,f Qi lU integral Heat I

j O

G y( h'w<

I

. I

,~

i

. i Stud ~l 6

y 0

Exchanger OQs> G-Mechanical { 'Mensioner. Gl,- i b,.,'1' O f@) l g ( b _T gi .,40 bpp1 s S l'\\ io!,'/[ h 5 -3 t I \\ O kx. . p, _. s o e. N j %d k'c 0L f w ['\\ n s-y g j f / ]" ' \\ \\g \\ ?} h 3-Cover. Thermal r N p. N '/ Barrier ~ %m w n'7 7 l/ / / ':G/ u.m %:p,n P /s' k,/ l /, ??, hb ,W '? ///. S 17 l ~ J -$ / TiQW Q@?nw= r w we e~ ape >,c, i.U'v/ y,/ {v a.,$y'f. y, *, xt, T

  • ~5

$ '/ e ?' ) '*h Q gV "3 /!j!pf4lw~o t-Q4 l i x. y^,,/ //,//1 : 5,Lg W)) i 3 /+J 4 tu t //,i yn.wsc..w 2,. y%. 7 ,/ w / ar ... m u g. a ,) y /- a Pump Case .z" k r ,@n.,..+udssME.4R + ~ ngure 1 i ro n,wi n qv N' mi po q pHM P

.1[ i Sh l (c:') .) 5 e ,m 8 gj s s o s 'o l

o

\\ i b 0 4: sg [ [ ~ ~l ] i THERM AL B ARRIER DR AIN HOLE ~ 180 270 0 90 180 t i i l l l 1 I J I L a ? n l _ D' ~ = 1 L L 1 E r I t i i y I i COVER BORE (EXP ANDED VIEW) Pigure 12 a COVER THERM AL F ATIGUE CR ACKS (TYPIC AL FOR PUMPS WITH INJECTION) s 1 J

  • /

45* ~, is \\ s i s 3 \\ S 2h 45 s o-~ 45 /] i , a i 1 4 E' ~ 1 1/2 DIA. 5'MN 5' i ] 1 1/2 DlA.- $f4_ { .l. l 1 ,d l _ _f \\,,i s. i A l l.: ! ! \\ H )' ! zl l -l i 6 r / 1 -5/8 / ) J l l' 7 3/8 i I i j l / / t l - i t 1 / lp- \\

/

\\ h / i \\, \\ \\; 1\\ A. \\ / 1 4 ] Figure 13 a COVER W ATER J ACKET DET All (TYPIC AL) i1 O ed i

s..,

l ,I I I I i 4 I i i i I i I I 3 I I l 0.28" i = i l I II 1 II I I i i i 1 l 'i I i i, s I 3/32" WIDE l 1.028" i l 1/16" DEEP I o 1 I l I iN crmk i -[ r gop.3d. u ..n2. 0 %7 i >2 0.597" (m.t.) l n n 8.0635" DI A. rigure u

u u u u u u u u u a u o u o o u u u 4 ~.... 9 0.5 6 2 0.4 C ay Ig 0.3 / ( A IO t h. = 3.0 KsiQin 2AM-uj g g g o 6 e6 y 0.2 40 0 o o o I / o 0.1 SHAFT u 0 20 40 60 80. 100 120 140 160 ~180 HOURS OF OPERATION (1000 s) l FATIGUE / CRACK GROWTH vs. TIME Figure 16 t I

a ,\\ - i s SHAFT 1 i t ARR ER v' ACME -2 THREADS eg [ J 3 FILLET WELD v r # ~ i ~

  1. w, %

\\ 1/4, SMOOTH HYDROST ATIC p* / i 4 TR AWSITION BE ARING j / <3g ~ AREA E N D P L A T E Y [' g / v l A t i u a Circumf erential cracks vary in smooth tr ansition ar ea up t o 2*-3' in length, joining axial cracks in threaded area and branching down into weld. One crack transverses across weld into bottom toe of weld and runs 2700 around shaf t. i All dimensions a ppro ximat e. Indications are typical f or f ull circumf erence of I above lis t e d areas. 1 a r Figure 17 ...m. ++o

  • ~ * *
  • SH AFT THERM AL F ATIGUE CR ACKS

~ (TYPIC AL FOR UNITS WITH INJECTION) i y i j .,l

  • j l

e, J

l-1 T F A L H S E H C L 4 T A O F 2 OR L MU t SS 2 1 = 1 s e s 1 o p 1 r u p S e K v Cl i t t a A O W. r I R T t O s A ~A u 8 L N u l t l I C I 1 A M J rN , d e M A r o r t u R X f L g E E i l e F i TE t a V 7f n l r F I e g O T T ? g C F a NU A [. x OR L

3 e

H T = I T S ? S h A E P t d G D M t .r, i S w A D U PM ? A k P c OO E f, l a RR F R r PF O C H 1 T L e

EI t

M o 1 N L_ C A 1 L 1 L 1 i !i

i

l f

' Li;l f' 7' 4 0k 1 NO 1!NMi W\\% if 0 kW Di OH H M Oi Wf %,4,;, i P kP 1_.M'""5 Jb v~DW NC W-~60MWinw~E WMMM_,?MMi I SAN JOSE, CALIFORNIA i Dece=ber 15, 1987 SII.!!o 459 File Tab 3 Category 1 - Please note that this SIL is a follovup to RZbSIL No. 003 issued by GE on May 21, 1986. BYRO:( JACKSC:t RECIRCULATIO:t PUMP SHAFT & COVER CRACKIdC

  • s eb mu~i A nu=ber of F'.tR reactor coolant pu=ps and B'.7R reactor recirculation pu=ps =an-ufsetured by the Byron. Jackson Company (B J) have been disasse= bled and in-spected during the past few years because' of the length of ti=e the pu=ps have been in operation. ASME Section XI ISI require =ents and some operational prob.

le=s. Ther=al fatigue cracking has been observed in several pu=ps. CE !!uclear Energy issued RICSIL tio. 003 on May 21, 1986, which, in part, ad- 'ressed ther=al fatigue cracking in pu=p shaf;s and covers. d B.J issued a re-laced advisory (Tech : ote 8701 80-005) in January 1987. Both docu=ent:s con-tained the conclusion that although the observed ersching was undesirable, it was considered to be relatively benign based on analyses and observations. Cracks were expected to exhibit virtual arrest or show very s=all growth for depths exceeding 0.2 inches (shafts are 5 to 8 inches in dia=eter). The Janu. ary 1987 B J Tech !!ote recc== ended that utilities disasse=ble and inspect the pu=ps at the first convenient opportunity. Recent findings, however, indicate that the cracks =ay represent a threat to pu=p reliability. Dircussie,n Recent observations and =aasure=ents of the pu=ps in operating installations show that so=e cracks bnve penetrated to dopths greater than 0.3 inches in both shafts and covers. These depths are grester than previously anticipated based on the expected frequency of ther=al cycling. In the cover, cracks are axial and could be of sufficient depth to penetrate the closed cooling water (CC) circuit if cracks and flow passages are at corresponding sti=uths. This represents a risk. of.conta=insting nor: ally clean system and-possibly sub. jecting it to higher pressures, tio such occurrences have been reported. In the shaft, so=e erseks change orientation fro: axial to circu=ferential. This represents an increased risk of shaft failure under nor:sl opersting con. 1 GENERAL h ELECTRIC Ctst4 4 titcTac cewe.%,.553,15 k AtSNs.5-eoTv 8 cA O*80TV CA C'w'Ct "*G w'.-....,.,.. m,,., m.

m. o. _.cl _...,:,,o,

1._ c c o-m s m c. _, - _... 814,85044,mt Vit o8, alp =8 C**** f *0N

\\ I SIL No. 459 Category 1 ditions if the cracks propagate to sufficient depth or under abnor=al condi. tions if the shaft is subjected to increased dyna =ic loading as could occur from cavitation or by a foreign object becoming lod ed in the pump. No such 5 failures have been reported in SVR pu=ps. During the period in which these recent field observations were cade, B.J per-for:ed an analysis of the crack ini:iation and growth =echanis:s. This work included analysis of recent test data from a full scale operating temperature pc=p shaft and cover cock up. The data verified previous conclusions that mixing cold seal purge fiev vith hot system water initiates cracks. The data also showed that previously unknown low frequency flue:uations occur. Shaft and cover crack growth analyses considering these low frequency fluctuations support the field observatiens and indicate that such ther:a1 cracks can pen. e: rate to depths beyond the observed depth of approxt=ately 0.3 inches, but at low propagation races. Cracking in pu=ps without seal purge can occur but is expected to be less severe and to occur'at a highe'r posi: ion on.the shaft and

cover, pece--ard d ae-ten CI reco== ends that EVR owners consider these ' findings and develop concingency plans to preven: negative 1: pact on power production and to minicize effec:s on outage maintenance activities.

C T. r e c c== e nds tha: su:h contingency plans include the folleving. 1. Review pump and system design, period of cperation and =aintenance histo. Correlate with other field experience to assess the potential impact ry. on the pu p or pu=ps in question. For pu=ps with a high potential for significant crackins. the following actions should be considered: Install shaft probes to =oni:or shaft vibration response to protect against consequential da= age in the even from ther: ally driven to mechanically driven. crack propagation changes Monitor CCW effluent early varning and to mini =izefor contamination by reactor coolant to provida and equip =ent in the even: contamination of external CCV system ce==unication develops between these sys. te=s. 2. Review A$d! Sectien 5 ISI plans and ce==it=ents and establish plans ~ which consider the recent findings described i t, this SIL. Schedule in. spect',ons of pu=ps with greater than 80,000 hours of pu:p operation to deteer cracks and i=ple:ent corrective actions before excessive crack grovtt occurs. 3. Prepare an inspection plan and develop the following: iethods for examining pump shafts and covers ( 2-J

.r. e l b SIL ::o, 459 Cate gorf 1 Criteria for return to service Repair =ethods and alternatives considering ALLM Plans for replacement of existing parts or use of i= proved designs l'.7, utilities are requested to infor= CE of the results of pu=p exa=inations. This infor=ation vill be used to characterite crack growth behavior and can be used to develop inspection and maintenance actions for other pu ps. I If you would like additional infor=ation on this subject, please centact your local CE :uclear Energy Service Representative. Technical Source: J. H. Cates e t ll M Ih!, O!V j Issued by: J. G. Moore ,Harketing Services Manager. Product Co:::unications 6 l Product

Reference:

331: Reactor Recirculation System 1 l 1 i a I s l l C 1 3 -}}