ML20126J621

From kanterella
Jump to navigation Jump to search
Forwards LER 88-009-01 Issued on 921113,discussing Increased Thrust Bearing Loading for RHR Pumps When Pumps Run at Reduced Flows.Concern Raised That Inservice Testing May Not Detect Problem Soon Enough to Avoid Failure
ML20126J621
Person / Time
Site: Seabrook NextEra Energy icon.png
Issue date: 12/24/1992
From: Rosenthal J
NRC OFFICE FOR ANALYSIS & EVALUATION OF OPERATIONAL DATA (AEOD)
To: Norberg J
Office of Nuclear Reactor Regulation
References
NUDOCS 9301060229
Download: ML20126J621 (1)


Text

_ - _ _ _ _ - - _ _ _ _

o ..-

DEC 2 41932 MEMORANDUM FOR: James A. Norberg, Chief Mechanical Engineering Ilranch

, Division of ' engineering Technology Office of Nuclear Reactor Regulation FROM: Jack E. Rosenthal, Chief ,

Reactor Operations Analysis Branch Division of Safety Programs  !

Office for Analysis and Evaluation t of Operational Data t

SUBJECT:

GENERIC IMPLICATIONS OF TilRUST 11 EARING LOADING DURING LOW FLOW SURVEILLANCE TESTING OF PUMPS Seabrook Station licensee event report (LER) 443/88 009 01 issued November 13, 1992 discusses increased thrust bearing loading for the residual heat removal system pumps l when the pumps are run at reduced Dows. I am concerned that inservice testing may:  !

(1) not detect the problem soon enough to avoid the concern of a common mode failure of the_ pumps or, (2) reduced flow may actually increase damage to the pumps. The same pumps are currently installed at Byron, Braidwood, Catawba, Vogtle and San Onofre Units 2 and 3. The concern may be applicable to other safety related pumps.

It is my understanding that this issue is being reviewed as a Part 21, Log Number 92-256.

The item has been assigned to the Reactor Systems Branch for evaluation.

A copy of the LER is attached for your information. If you have any questions, please call Earl Brown (492-4491) of my staff.

Originalsigned by Jack E.Rosenthal Jack E. Rosenthal, Chief Reactor Operations Analysis Branch Division of Safety Programs Office for Analysis and Evaluation of Operation'al Data

Enclosure:

As stated 060097 cc w/ enclosure:

$ Gordon E. Edison, NRR ro n7ny";rp)dugimV-f0'TDfd.c 18' Walter P. Ilaass, NRR: n, U, b u 4.u vi.mu T:. Ml M ,

, r8 ~

Distribution:

W -PDR LGundrum VBenaroya Dross hg Ceni91 File'

  • _ROAB R/F EBrown ,

JRosenthal TNovak EJor n G Millman, RES .

MVagins, RES

% j

.ph ROAB g ~

ROAB- :ROAB [$)p L) ~. g

/-

LGundrum:mmk pEBrow' y JRosenthal 12/0/92 12MJ/91- 12/IT92 l'

. . __. _ _ _ _ _ _ _ _ _ _ . .._-.._._._..___-.____m_ _

__.._...__m___ . . . _ - __m.._ . _ _ .

2 d ~

W .A. kucktaa 8t&Vlatcas guewission

. AM40vtD out no 3,weios UCENSEE EVENT REPORT (1.El& oran: sma

, Acipt, =*=4 o,e ca ntav..taui , . a ,2, SEAbkOOK STATION o l$ l o j o l o l41413 1 lorl0 l 5 -

RESIDUAL llEAT. R120 VAL (RllR) PUMP MODIFICATIONS TO REDUCE TilRUST BEARING I.0ADING tytert pati (H 46 A huwst h 16e etPDAf DAf t 171 OtMO R 7 Attttf all twygtyl0 til l posiin 4ay YtAA V t A 81 ., ha" '

Ngj WDas tet ' DAV VEAR 74C8Utt masta 00ca t t Wwe t hip 018101010 1 1 I

~ ~

l_l2 0l 8 _8 8 8l 8 0l0l9 0l1 1l 1 1l3 9l2 o isio ic i o g i l

,,,,,,,,,, 1,5 n,oa, a tv.uinie evalvakt t0 t t n 0 vin utun De ie u n o ecA . . - .. . ,.,,, ,, nu asoot tel 5 re Aosto re *o*iai **.' Haimi statin Po.'t a 39 40steHtHil 60 3HeHl) Y 64J31.lt3 H.) 333t tot T.P o oi 0 r. mi.in Hu 30 aogleH1Hal

[ ==m H.13(eluHil

[ w ni.iuH.o 5433LalDH.dlltAl

[ g3 <rg,.;;, ,

J6ddl 39 stalteHile.1 H 331s HIHal 64)3talt3H 41181 P0.406 Be H1 H.1 Sents H2H ill 6033Ialt3Hal llCthst t CONT ACt PCA iMit kla titi NAus titte o%l huustm Mr. Jamen M. Peschel, Regulatory Complance Mariager, ext. 3772 .at A ccot  ;

61013 417141-19151211 COMPLif t CNt itNS f OR (ACM COMPQNtWT F AILDAt pg&Cgggtg lgg yMjg pgpggy gjgg .

CAU18 Ivettu (Curcht ki

" 0 " '

tO h ,n CALil (YTflW CCuPChtwt 64 'n Cate t p

Ie I O 7 5 B B P 1 Pl t i /

I i 1 N

i l I I i l I cf m-13 li P g

M 0, W 1 2 0 N s

9 tupet t ut wt a t s t POm f l RP4 C1 t D H el goesty gay ygan Sve ustsiON G ,t s rio . .-, v >rroro svownwon cd tuo T ~0 j l l

. . n . . n ,v . . , ~ . , . . .- . -. . ,, . - ... - . -- . . .o n o On December 8,1988, testing indicated that premature degradation of the Residual llent Removal (RilR) pump thrust bearing could occur which would reduce its specified design life. This condition could result in the unanticipated shutdown of a system required to remove residual heat.

Long term corrective action for this concern involved modification of the pumps to reduce the loading on the thrust bearings. This ruodification was developed by Westinghouse and Ingersoll Rand based on testing of a Unit 2 pump in the original and modified configurations. The modification required a reduction in suction side wcar ring inner diameter and impeller wear area outer diameter.

Maintenance schedules have been developed to ensure thrust bearing replacetnent prior to the end of the current predicted service life.

There were no adverse safety consequences as a result of this condition. The RilR-pumps remain capable of performing their safety ' function, despite decreased bearing -life. The administrative.

controls ensure that sufficient bearing life remains to provide long term post accident cooling.

Additionally, vibration monitoring will identify the onset of bearing deterioration prior to actual bearing' f ailure.

This is the first occurrence of this type at Seabrook Station. ,

,  % w ~- j' m s a

j% V$ G Y 4#p 1 s.

l ,l ' ;

~

uxe r 3.e

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

u usti.e eau i... sw. . .

LICENSEE EVENT REPORT ILER) TEXT CONTINUATION .mo, s e ... .o .. .

rsPitt& vivQ ama t t sevas s a m sateut t a.4 m 418 18 8 '#4:4m

.... , u........ ...... *

  • SEABROOK STATION c l6 jo lc l0 l414 ]3 8l 8 0l0l9 0l 1 0; 2 or 0l5 m w , ,s m , m .*co sman.m On November 23, 1987, Residual llent Removal (IlllR) pump Ril p 8A was removed from service due to excessive vibration at a frequency range indicative of excessive thrust bearing wear. This occurrence was premature when compared to the expected life of the bearing (7300 hours0.0845 days <br />2.028 hours <br />0.0121 weeks <br />0.00278 months <br /> versus 100,000 hours0 days <br />0 hours <br />0 weeks <br />0 months <br />). Each Seabrook Station RilR pump consists of a Westinghouse vertical, 400 lip, frame $009P39, motor installed on a close coupled Ingersoll Rand model 8X20WDF pump. The affected bearing was one of two thrust bearings located in the lower bearing housing of the motor, floth bearings are 40 degrees angular contact ball bearings mounted f ace to face. The failure occurred in the uppermost of the two bearings, i.e., the downward loaded thrust bearing.

In situ testing was performed to determine the actual thrust on both the A and 11 train RilR -

pumps. The measured thrusts significantly exceeded the specified maximum down thrust and closely correlated to the loads required to produce the observed bearing dcgradation. As a result of this testing, a decision was made to return one of the Seabrook Station Unit 2 RilR pumpr. to ingersoll Rand for further testing.

On December 8,1988, testing was performed at Ingersoll Rand on the Unit 2 RilR pump original design and on a modified configuration. This testing indicated that premature degradation of the Residual llent Removal (RllR) lilP) pump thrust bearing could occur which would reduce its specified design lite. This condition could result in the premature shutdown of a syst e m required to remove residual heat. Corrective actions discussed below _ were developed by Westinghouse and Ingersoll Rand to reduce RilR pump thrust bearing loading.

As a result of these modifications, RilR pump bearing life is predicted to be 33,017 hours1.967593e-4 days <br />0.00472 hours <br />2.810847e-5 weeks <br />6.4685e-6 months <br /> at 3500 gpm.

EAFETY SIGNIFICANCE RilR pump performance is monitored quarterly as part of the ISI program. A review of past performance data indicates that RilR pump characteristics (i.e., head and flow) did not degrade-over time. pump vibration monitoring provides additional assurance that the RilR pumps will continue to be capable of performing their required safety function. The

  • alert
  • and ' required action limits
  • in the ASME Code Section XI insure that vibration will be corrected prior to severe pump damage.

Assuming the single f ailure of one train of RilR, the remaining R11R pump would not catastrophically fail as a result of bearing degradation. Rather, increased vibration in the operating pump would provide warning of degradation long before any impact on system performance would occur. Therefore, adequate time would exist to restore the alternate train or take other remedial measures prior to actual failure of the pump.

Degradation is also not likely to result in simultaneous failure of both trains. The lower thrust values recorded on the il pump resulted in considerably longer bearing life for this pump. The differences in thrust values, point in service -life, and previous operating history make a simultaneous failure unlikely.

The safe shutdown capability of the ECCS is not impaired as a result of reduced bearing life.

Since beaiing degradation is a slow process, the engineered safety features of the plant would perform normally. After transfer to recirculation or long term RilR cooling, the RllR pump

~...... a ,oe,m..w. m .I

m ... .a es weunso m.u, UCENSEE EVENT 11EPORT ILER) TEXT COP MUATION . neem . ,, . ,.

. cmis. sw.

, u,g .. u. oi = " *" a in u ..n . ,a o.

"" "*=."  ! 7/',7:

SEABROOK STAT 10!I e ls l0 l0 lc l4 l 4l3 88 g ._ O09 gg ,,,,

0;1 O3 g o, 0;5 nn u e m **--a **r o. . ama en.n bearing may begin to degrade. Sufficient time would exist to take corrective action to ensure that the redundant pump is operable and capable of performing the long terrn cooling function while repairs are made to the degraded purnp.

CORRl!CTIVi! ACTIONS The RilR pumps have been modified as discussed below to reduce the loading on the thrust bearings, The modifications were developed by Westinghouse and Ingersoll Rand based on testing conducted at Ingersoll Rand in December 1988 on a Unit 2 R11R pump. The Ingersoll Rand testing was perfortned on the Unit 2 pump original design and on a modified design. The modified design resulted in a significant reduction in thrust bearing loading. The Unit 1 RilR pumps were modified based on the recommendations of Westinghouse and Ingersoll Rand and in situ measurements of thrust bearing loading were performed. The in situ thrust bearing measurements indicated that loading was not reduced to the extent determined by ingersoll Rand, however, loads were sufficiently reduced, extending the predicted life of the thrust bearings to 33,017 hours1.967593e-4 days <br />0.00472 hours <br />2.810847e-5 weeks <br />6.4685e-6 months <br /> at 3500 gptn. Replacement of the thrust bearings will be inade prior to exceeding 24,257 hours0.00297 days <br />0.0714 hours <br />4.249339e-4 weeks <br />9.77885e-5 months <br /> of service at 3500 gpm so as to ruaintain a one year reserve life in the unlikely event of an accident at the predicted end of service life. A detailed discussion of the corrective actions is provided below.

The Rllit pump design requirements are found in Westinghouse E Spec. 952470, Rev. 2 (FP50279, issue 12). This document specifies that the bearings are to have a Rating Life of 1110 as calculated in AFilM A Standard 9, dated June 1972, and be suitable for a minimum of 100,000 hours0 days <br />0 hours <br />0 weeks <br />0 months <br /> of service.

1 To achieve design compliance, utilizing the existing bearing, (type 7222 11EAGY). externally applied loads which contribute to downthrust must be reduced to 1700 lbs. These externally applied loads are the net result of hydraulic forces acting upon the impeller during operation-of the pump, in the original pump design, the net force has been lowered by employment of-five. 9/16' diameter, balancing holes drilled in the impeller which reduce the pressure being applied. Close tolerance running clearances are also employed to help maintain pressure differentials applied across the impeller.

Ingersoll Rand was contracted.by Westinghouse to measure the impeller sial thrust using load cells between the motor and the motor support head. The results of the tests performed before and after innpeller and wear ring modifications are presented in Ingt.rsoll Rand Technical Report .

TR 8808, dated 12/22/88. A copy of ingersoll Rand Technical Report TR 88081s available for review at Seabrook Station. The test results were provided to Westinghouse and forrned the basis for issuance of Westinghouse Field Change Notice FCN 10588, and Westinghouse-supporting analysis documented in Safety Evaluation Check List (SECL) No. 89 001 A copy of these documents are available for review at Seabrook Station.

The Ingersoll Rand test results verify that a reduction in the front ring diameter was effective.

Peak hydraulic ~ thrust (occurring at 2300_ gpm) was reduced from 5300 lbs. -to 3100 lbs.

Dependent on flow, a dramatic reduction in hydraulic thrust load was seen, at and beyond 3000 gpm. The modified confi;turation thrust at 3000 gpm as measured by Ingersoll Rand was recorded to be 1376 lbs.

74,.. nu .u:c. coin.+. m si i

\

er... -

    • ' e m a. . . . . .. . - -

LICENSEE EVENT REPORT (LERI TEXT CONTINUATION .n... . . c .. e 3, .

(IPtAls v31/ss +

f as.su.s hast tu M4 s ti sevang e a up 488 kWit 86 p'att G

"" "

  • 2." .

"a',i: -

SEABROOK STATION 0 pl0l0l0l4l4l3 8l 8 0l0 l 9 0l1 0l 4 oi 0l$

sw a . m av w.nm Pump performance was monitored and recorded during the Ingersoll Rand test. Westinghouse has reviewed this data against the manufacturers certified performance test curve, and has determined pump performance to be virtually unchanged. System flow and NPSit calculation therefore, were not affected by this change.

Westinghouse in conjunction with Ingersoll Rand developed the design modification specified la FCN 10588. The modification reduces downthrust by increasing forces acting up on _the impeller. To accomplish this, the suction side wear ring inner diameter and impeller wear area outer diameter were reduced, thus increasing the surface area (outside the wear ring) over which the higher discharge pressures are applied. The corresponding reduction of area inside the suction wear ring reduces uplift applied by suction pressure. Since the magnitude of force applied by discharge pressures are significantly greater a net increas: in uplift will result ' Peak hydraulic thrust loads were not reduced to the original design value of 1700 lbs; thus, Westinghouse / Ingersoll-Rand recommended that the specified bearing life of 100,000. hours be revised to reflect the expected bearing life with the modified impeller / wear ring.

Westinghouse /Ingersoll Rand predicted that when operating at peak thrust (approximately 2300 gpm), a bearing life of 28,560 hours0.00648 days <br />0.156 hours <br />9.259259e-4 weeks <br />2.1308e-4 months <br /> is predicted and that operation at flow rates greater than 2800 gpm can estend predicted life to 74,700 hours0.0081 days <br />0.194 hours <br />0.00116 weeks <br />2.6635e-4 months <br /> due to the reduction of loads in this range.

These predictions are bas:d on prototype modification test data. Actual bearing life predictions to be used in determination of bearing replacement at Seabrook Station were determined based ,

on post implementation in situ testing discussed below.

Westinghouse recommended that the RHR pump flow controller nominal setpoint be increased from the current 3000 gpm to a value between 3200 and 3300 gpm. 11ased on in situ test results operation at a flow controller setpoint of 3500 gpm is acquired to offer an appreciable reduction in thrust loads. Westinghouse concurred that operating at 3500 gpm is acceptable, except during mid loop operation and that adequate NPSil will exist at the higher flow.

Ilearing replacement will be made at the predicted bearing life less one year post accident reserve life. Although SECL 89 Ov1 utilizes a 100 day reserve in calculation of the replacement interval example, the thrust bearing replacement schedule utilized by North Atlantic will ensure RilR pump operability for a postulated post accident durntion of one year.

The predicted bearing life, established in SECL 89 001, is based on thrust load data collected during prototype testing of the RilR pump modification. Ucaring replacement intervals to be used at Seabrook Station are based on post implementation in situ test data collected on the installed pumps. These post implementation thrust data results aie documented in Teledyne Report TR 20834 (885), showing less reduction in thrust load than that determined - by Ingersoll Rand. The Teledyne Reports (September 7,1989, for RilR Pump 8A and January 5, 1990 for RHR Pump 8U) are available for review at Seabrook Station.

Operation of the RilR pumps at 3500 gpm will provide the lowest achievable thrust load within an acceptable operating band for the pumps. Operating procedures have been revised to-require operating flowrates of 3500 gpm during normal RHR sy tem operation, except during mid loop operation.

To predict the service life of the thrust bearing, North Atlantic utilized the highest hydraulic load established during the in situ test of 2775 lbs (extrapolated from pump 8A data at 3500

, a~ ~ .n e o m.u.m4

=,... .m.....mi..,u.

    • " LICENSEE EVENT REPORT (LERI TEXT ' 'ifvTINUATION .

... . . . n ,,. .,e 3,

. DPets 4111s facsytt aus tis W a t t am.se t s (33 g,, gg "em." .

SEABROOK STATIO!i 0l5l0jojol4l4l3 8l8 _

0j 0l 9 Og 1 0l5 or 0l5 rtri n . - uc w am. .mn gpm) and added the test data tolerance (400 lbs) and rotating element weight (950 lbs) to establirth a total bearing load of 4125 lbs. North Atlantic calculated the predicted bearing life (using equations and bearing data from SKF llulletin 213 715, Sept. 1985) as 33,017 hours1.967593e-4 days <br />0.00472 hours <br />2.810847e-5 weeks <br />6.4685e-6 months <br />.

The predicted life (33,017 brs.) has been reduced by one year to ensure reserve life in the unlikely event of an accident at or near the end of predicted life.

North Atlantic has determined that replacement shall be made after 24,257 hours0.00297 days <br />0.0714 hours <br />4.249339e-4 weeks <br />9.77885e-5 months <br /> of service at 3500 gpm and has established the controls necessary to determine when replacernent is necessary based on a cumulative account of cperating hours. Should mid loop operation become necessary with its lower required flow rates, North Atlantic will determine the bearing life expended during that period using the in situ test data and reestablish remaining life.

There were no adverse saIcty consequences as a result of this condition. The Rlill pumps r e in ain capabic of performing their safety function, despite decreased bearing life, Administrative controls ensure that sufficient bearing life remains to provide long term post-accident cooling. Additionally, vibration monitoring will identify the onset of bearing deterioration prior to actual bearing failure.

,;'"- au . v s e., i..mi.u...I

Seab o k JH 03874 Telephone (003)474 9521 fj hhffC 3 y Facsimile (603)474 2987 Ener0y Servico Corporation Ted c. reigenbaum Senior Vice President and Chief Nuclear Officer NYN 92160 November 19, 1992 United States Nuclear Regulatory Commission Washington, D.C. 20555 Attention: Document Cont *ol Desk

References:

(a) Facility Operating License No. NPF 86, Docket No. 50 443 (b) North Atlantic Letter NYN 89006 dated January 9,1989,

  • Licensee Event Report (LER) No. 88 009 00: Premature Residual lleat Removal (RilR)

Pump Thrust Bearing Failure,' G. S. Thomas to USNRC

Subject:

Licensee Event Report (LER) No. 88 009 01: Residual IIcat Removal (RilR)

Pump Modifications to Reduce Thrust Bearing Loading Gentlemen:

Enclosed please find Licensee Event Report (LER) No. 88 009 01 for Seabrook Station. This submittal provides supplementary information regarding modifications to the Scabrook Station Residual llent Removal (RI{R) pumps which have been implemented to reduce loading on the RilR pump thrust bearings to extend their service life.

Should you require further information regarding this matter, please contact Mr. James M. Peschel, Regulatory Compliance Manager, at (603) 474 9521 cxtension 37'l2 %

Very tr ty yours, id.~ f-/ kkhkLb #

Ted C. Fei TCF: A LL/act

Enclosures:

NRC Forms 366, 366A l

l l

400040 9211200262 921113 PDR .ADOCK 05000443 nomber of the Northeast Utilities system //) 8 5- PDR.

' 'lL'

4 I

i United States Nuclear Regulatory Cornmission November 19,1992 L -!

, Attention: Document Control Desk Page two .j l

I cc: Mr. Thomas T. Martin _;

Regional Administrator U. S._ Nuclear Regulatory Commission .. ,

llegion 1 j 475 Allendale Road l King of Prussia, PA 19406 i Mr. Gordon E. Edison, St. Project Manager Project Directorate 13 Division of Reactor Projects U.S. Nuclear Regulatory Commission Washington, DC 20555 Mr. Noel Dudicy 1

- NRC Senior Resident inspector -

P.O. Ilox 1149 Seabrook, Nil 03874 -i INPO '

Records Center 1100 Circle 75 Parkway Atlanta, G A . 30339 ,

1

?

2

[

_7

)

-1

)

u

.. < ~

, ,,am., -w r[ , , (. [ 4%. ,, y ._y..,- c- 1y,,. -,,gg-. -m, p4L.-9 0 -y , y