05000324/LER-1986-001, :on 860107,discovered That 10 of 11 Safety Relief Valves Tested Exhibited Symptoms of higher-than- Allowed Setpoint Drift.Caused by Pilot disc-to-seat Sticking & High Friction.Maint Program Revised

:on 860107,discovered That 10 of 11 Safety Relief Valves Tested Exhibited Symptoms of higher-than- Allowed Setpoint Drift.Caused by Pilot disc-to-seat Sticking & High Friction.Maint Program Revised

ML20197D058
Person / Time
Site:	Brunswick
Issue date:	05/07/1986
From:	Dietz C, Pastva M CAROLINA POWER & LIGHT CO.
To:	NRC OFFICE OF ADMINISTRATION (ADM)
References
BSEP-86-0555, BSEP-86-555, LER-86-001, LER-86-1, NUDOCS 8605140115
Download: ML20197D058 (9)

LER-1986-001, on 860107,discovered That 10 of 11 Safety Relief Valves Tested Exhibited Symptoms of higher-than- Allowed Setpoint Drift.Caused by Pilot disc-to-seat Sticking & High Friction.Maint Program Revised

Event date:
Report date:
3241986001R00 - NRC Website

text

.

NR P*'* 384 U S. NUCLEAM E EluLATO4Y COMME 5StON APPROvtD O*A8 NO 3150 0104 LICENSEE EVENT REPORT (LER)

' " a ' S

• 8' "

84CILITY NAME (13 DOCKET NUMGER (21 P AGE 'la Brunswick Steam Electric Plant Unit 2 VITLE tel o i s t o l o l o l 3 l 2 l4 1jorl0p Safety Relief Valve (SRV) Setpoint Drift E% ENT DATE til LER NUMBER 16)

REPORT D4TE (7)

OTHER F ACILITitS INVOLVED 18 MONTM OAV VEAR vfAR 5'y U (

[f[

MONin DAv vfAR r acetsf y NawEs COCRET NUYSER 51 olslol0lol ; ;

Oll 0l7 8

6 8 l6 0l0ll 0lI 0l5 0l7 8l 6

~

o,5, o, o o,

OPER U M TMIS REPORT is SusMITTED PURSUANT TO THE REQUIREMENTS OF 10 CPR $ (Caece og er more of rae rose.*gi (til "ooi *'

5 20 402.i 20 40 iei so73.n2a.

73Fimi R

20 40steitt:(4 50 36seH11 50 731 ell 2Hv8 73 718e4 (10n l

20 4054eH1Het 50 381sil21 X

SO 73teH2Houl OTH oee a a rrect 20 40$deH1Hol 90 73 sH2sl4 50.73(a H2 Hemil Al J66AJ 20 406 del (11:sl 60 73:eH2Hal 60 73teH2Hv=ltS6 20 4056eH1 Hv6 50 73. ell 2H=1 50 73 eH2Hal LICENSEE CONTACT FOR TMfS LER 1121 Nave TELEPwCNE Nvv0f a anta COO 4 M. J. Pastva, Jr., Regulatory Technician 9ll19 4l 5l7 1 2l3 1 1 l5 t

COMPLETE ONE LINE FOR EACM COMPONENT F AILURE DESCRsBE0 aN TMis REPORT 113)

{E O TA CAvst 5vsTEv COvPC%gs, W ANLP AC-R E *CR Ta g 6E

,M s,E Iwata rv Neava CAwis a v ai tw cvwo% tN T X

A ;D

] 1R ;V T 0 2 ;0 Y

X A;D

, iR,V T;0,2 p Y

i 1 X

A ]D l lR lV T 0l2 l0 Y

X AlD l lR V T; 0 ]2 l0 Y

1 SUPPLEMENTAL REPORT E MPECTED itse VONT=

DAv viaR Sv e v ' 55 'O N v EE fr9yee comgere fMfCTfD $LGw$$oON CA Tfi

%O l

l l

ABSTR ACT fler ra f e00 seeces e e, ecensweroy hheea srag e roece typeweere heep 114e During a Unit 2 refuel / maintenance outage, testing of safety relief valves (SRVs) in accordance with Technical Specification 3.4.2 revealed 10 of the 11 valves tested exhibited symptoms of higher than allowed setpoint drift. The test results are attributed to pilot disc-to-seat sticking and high friction due to inadequate clearances in the labyrinth seal area. The ADS function of the valves was not affected.

The SRVs, Target Rock Corporation Model No. 7567F, were refurbished, recertified, and returned to CP&L for reinstallation.

Pilot valve discs in six of the SRVs were replaced with discs manufactured of PH 13-8Mo as recommended by the BWR Owners Group (BWROG) materials selection panel.

SRV setpoint drift resulting from friction in the labyrinth seal area has been significantly reduced through implementation of the BWROG improved maintenance.

program. The installation of PH 13-8Mo pilot valve discs is part of the effort to eliminate or reduce SRV setpoint drift resulting from pilot valve disc-to-seat sticking.

/

./i a 8605140125 860507

'[

DR ADOCK 05000324 3

PDR N# C Form 364

'9 63+

i l

NRC Fsem 3000 U $ NUCLEAR [EGULATOIY COMMIS8104 LICEN!EE EVENT REPORT (LER) FAILURE CONTINUATION

* E o o *8 *o 2 ' w-*

• EMPIRES 8<3185 F ACILITY NAME 03 DOCKET NUMSER 12)

LER NUMSER ISI PAGE (31 YEAR

. II g 'm

]3 Brunswick Steam Electric' Plant Unit 2 0 l 5 l 0 [o l01312 h 816 O l011 0l1 0l2 0F O l 8_

COMPLETE ONE LINE FOR E ACH COMPONENT F AILURE DESCR10E0 IN THIS REPORT Hal

CAUSE

Sv 8TE U COMPO"4 E NT

"*s$ C.

R,TA8 E RE v^%g3c.

R EpR,7 OE

Cause

sysuv ccMPesEsT X

AfD I ] RIV Tl Ol210 Y

l l l l l l l X

AID I I R IV Tl 0]2l0 Y

I l l l l l l X

AID I l a ly T, q 2l0 y

)

[ [ l l l l X

AfD l I R lV Tl 012l0 y

l l l l l l l X

AlD I ! RIV Tl Ol2l0 y

l l l l

- l l l X

AID l-I R IV Tl 01210 Y

l i l l l l J l

i I I I I I I

I I I I I I I

I I I I I I I

I I I I I I I

I I I I I I I

I I l I l l I

I I I I I I I

I I I I I I I

I I I I I I I

I I I l l I I

I I I I I I I

I I I I l I I

I I I I I I I

I l l I I I I

I I I I I I I

I I I I I I I

I I I I I I I

I I I I I I l

l I I i l l I

l I l l I I I

I I I I I I I

l I I i 1 1 I

I I I I I I l

I l I I I I I

I l !

I I i l

l I I I I I I

I I I I I I I

I I I I I I I

! I I I I I I

I I I I I l

+-

1 I I I I I I I

I I I I I I I

I I I I I I I

I I I I I I I

i 1 I l ! I I

I I I

! I I I

I I I I I I l

i I I I I I l

l I I I l I I

I I I I I l upgfOaMaus

N_RC Form 386A U S NUCLEAL 5.E1ULLTORY COMMIS$10N LICENSEE EVENT REPORT (LER) TEXT CONTINUATION maovfo ove NO 3 iso-oio.

EXPIRES: 8 3188 FactLI) y NaasE til DOCKET NUMSER (21 l

LER NUMeER (s)

PAGE (3a "t:.r.

c=3:

naa Brunswick Steam Electric Plant Unit 2 ol5jolol0[3l2l4 8 l6 0l0[1

- 0l0 0l 3 0F 0 l8 raxr a-

-.v, w uc i ass 4.nm EVALUATION OF UNIT 2 SRV TEST DATA General Description During the Unit 2 1985-1986 refueling / maintenance outage, testing of the Unit 2 safety relief valve (SRV) pilot valves at Wyle Labs showed that 10 of 11 had incurred disc-to-seat bonding. Subsequent investigation shows that this condition did not create the potential for exceeding allowable safety limits on Unit 2, nor does it indicate the existence of a similar situation on Unit 1.

Nature of Problem Target Rock two-stage SRVs have exhibited high setpoint drift, both in-service and during as-received testing at Wyle Labs. This drift is attributed to two unrelated phenomena. A major contributor is high friction due to inadequate clearance in the labyrinth seal area. This problem resulted in setpoint drift of greater than 3% in six Unit 1 SRVs and four Unit 2 SRVs during as-received testing at Wyle in 1985 and 1984, respectively. The greatest drift caused by this phenomenon was 6.5% on Unit 1 and greater than 11% on Unit 2.

Data from utilities using these type valves shows the frequency of high setpoint drift due to labyrinth seal friction is 0.144, while experience at the Brunswick Steam Electric Plant (BSEP) has been at a 0.455 rate.

The amount of drift caused by labyrinth seal friction is such that the plant is below any safety limits. A preventive maintenance program was implemented by the vendor during the 1984 and 1985 test cycles at Wyle, which significantly reduced the labyrinth seal friction problem. Only one Unit 2 SRV exhibited setpoint drift of greater than 3% due to this problem during the current test cycle.

It was measured at 4.5%.

The second contributor to this phenomenon is corrosion-induced disc-to-seat bonding, which affected one valve on Unit 1 and two on Unit 2 during the 1985 and 1984 testing at Wyle. The setpoint drift ranged from 6.2% to near 18%,

depending upon the degree of bonding.

Industry data from two-stage Target Rock SRVs shows a frequency of high setpoint drift due to bonding of 0.083, and our past experience has been at a 0.136 rate. While the magnitude of drift due to bonding is much greater than that due to labyrinth seal friction, it has not caused a safety problem because of its low rate of occurrence. New pilot valve discs made of PH 13-8Mo are being installed, as per BWROG recommendation, to correct this problem. Six of the Unit 2 valves had these new discs installed while at Wyle during the present outage.

~. c,o.

u..

19 $38

WRC Pere 384A U $ NUCLEL3 KETULATORY COMMIS$104 LICENSEE EVENT REPORT (LER) TEXT CONTINUATION APPRovio oMe =o siso-oio4

( APtRES. G 31. 88 FJ.Cittiv NAME til DOCKET NUMSER (2)

LER NUMSEP tes PAGE 13)

"t'?,P.

%*,0

's^a Brunswick Steam Electric Plant Unit 2 0 l5 l0 j o l0 l3 l 2l4 8 l6 0 l0 l 1 0l 0 0l4 0F 0 [8 TEXT (# more asoce e escured, use sammenaf 4#C Fem JEE4'st (1M To differentiate between these two phenomena, Wyle now performs a diagnostic test in which the spring force is removed from the pilot disc and nitrogen is introduced below the disc to lift it.

If no bonding is present, 5 psig nitrogen will lift the disc of f the seat.

If this does not occur, the nitrogen pressure is slowly increased to determine the force required to break the bond.

The test is terminated when the nitrogen pressure reaches 200 psig or when the disc lifts, whichever occurs first.

During the 1985-1986 Unit 2 refueling / maintenance outage, testing of the SRV pilot valves at Wyle Labs showed that 10 of 11 valves exhibited some degree of bonding of the pilot valve disc to the seat. This 0.909 rate is well above industry experience.

Setpoint drift varied from 1% to nearly 18% per valve, with a calculated average of just under 13%. Four valves opened between 5 psig and 200 psig, while six others had not lifted when the test was terminated at 200 psig nitrogen.

One valve opened at 5 psig and went through all testing with no indications of sticking or labyrinth seal friction.

See Table 1 for specific results of the as-received tests.

Evaluation of Problem General Electric (GE) Company performed an analysis of the as-found condition to determine its impact upon the safe operation of both Units 1 and 2.

The setpoints used in this analysis were derived from the as-received test data and the retest data on the six valves that did not open during the diagnostic test.

The nitrogen pressure required to lift the disc was added to the first steam actuation pressure to determine the as-found setpoint. For those valves that did not open during diagnostic testing, 200 psig was the pressure added.

Discussions with GE determined that the addition of 200 psig gives a reasonable approximation of the actual setpoint drift due to disc-to-seat bonding.

TABLE 1 SRV OPENING SETPOINTS DERIVED FROM TEST DATA Valve Serial Opening Setpoint Number (psig) 1091 1142 1099 1307 1101 1145 1102 1216 1103 1182 1104 1353 1105 1338 anaC FOaw 364A 49 835

NRC Feren 304A U S NUCLEAR REIUL1 TORY COMMISSION LICENSEE EVENT REPORT (LER) TEXT CONTINUATION Anemovfo oMe No sisaato4 EXPIRES 8 31'88 F ACILIT V NAME (1)

OOCKET NUMGER (2)

LER NUMeER (Si PAGEE31 3E I*',',

,(I[y*,y,"

VEAR Brunswick Steam Electric Plant Unit 2 01510 l 0 l 0 l 3 l2 l 4 816 0]Ill 0l0 015 OF 0l8 TEXT I# nuwe ansce e msguest use estaaener Mec Form JE4'al(17)

TABLE 1 (Continued)

Valve Serial Opening Setpoint Number (psig) 1106 1116 1107 1338 1108 1376 1109 1305 Two test runs were made using lift pressures calculated to achieve an average of 13% drift. Run 1 assumed that the three 1125 valves lifted 1% high and the other eight drifted 17.5%.

Run 2 assumed all valves lifted 13% above the highest nameplate set pressure of 1125 psig. The use of average drift is more conservative for BSEP since it does not take credit for those valves that lift earlier. The results of these runs are as follows:

Peak Pressure (psig)

Run 1 Run 2 Unit 1 1378 Unit 2 1380 1382 Since these results were slightly over the vessel upset limit, an additional run was made to remove the conservatism in the averaging technique and use the derived opening setpoints for each valve. Peak vessel pressure for both Unit 1 and Unit 2 was under the upset limit of 1375 psig.

Many conservatisms are built into the model and the assumptions used for the peak overpressure transient. These'conservatisms are as follows:

1.

The GE ODYN-M code used for this analysis was verified against actual data from the Peach Bottom nuclear station. The code overpredicted the peak pressure by 30 psig. The event used was a turbine trip with bypass.

More severe transients are expected to produce greater overpredictions of peak pressure.

2.

The model assumed 105% steam flow.

3.

The model assumed SRV capacities of 90% of design.

4.

The model assumed other parameters were at the technical specification limits rather that at nominal values.

5.

The model assumed main steam isolation valve (MSIV) closure with failure of the MSIV position scram and reactor scram on high flux.

% AC FORW 364A C els

NRC Form 3eSA U $ NUCLE 42 8'EIULATOIY COMulSSION LICENSEE EVENT REPORT (LER) TEXT CONTINUATION Awaoveo oue no mo_cio.

ERPIRES. S/31'8B F ACluTV NAME (1)

DOCKET NUMSER (21 ggggg

" C.' '

7;*,0 va^a Brunswick Steam Electric Plant Unit 2 0 ]5 j o lo j o l31214 8 l6

-- Oli l 1

-- Ol0 0l6 0F 0 l8 TtxT IM man anece e requeenL anee.

'NMC Fame Jinn'ai m) 6.

The model assumed worst case end-of-cycle scram times and fuel exposure.

Although no credit can be taken for more nominal values, Carolina Power &

Light Company (CP&L) made estimates of the overprediction that would be expected from some of the above assumptions. GE made an additional run using 100% steam flow and 100% SRV capacity. This produced a peak pressure of 1355 psig on Unit 2.

This was not run for Unit 1 since it is bounded by Unit 2 and could be expected to be six to eight psig lower.

Data in the plant Final Safety Analysis Report (FSAR) indicates that scramming on tue r.02ary signal (MSIV position) would reduce the peak pressure by approximately 75 psig.

Therefore, there is approximately 100 psig overprediction inherent in the i

model and the assumptions used in the analysis as compared with a more realistic situation. No estimate was made of the overprediction associated with conservative assumptions 4 and 6 above.

In addition to the GE evaluation, CP&L conducted an independent analysis to quantify some of the concervatisms inherent in the GE calculations. The RETRAN program used for this analysis is not under CP&L Quality Assurance procedures; however, benchmarking against GE results show this model to produce results in acceptable agreement with those of GE.

Six runs were made with the CP&L model.

One run was performed comparable to the one GE made using 100% steam flow and 100% SRV capacity (except CP&L used 105% steam flow). The peak pressure predicted for Unit 1 was 8 psi below that for Unit 2 (as expected), and 28 psi below the upset limit. Another run showed a margin of 23 psi from upset with the six highest setpoint SRVs disabled. The final run predicts an 89 psi conservatism associated with failure of the MSIV position scram. The analysis results show that these conservatisms contribute significantly toward predicted peak vessel pressure.

The high rate of occurrence of stuck discs (well above past BSEP and industry experience) during the recent testing at Wyle Labs indicates some unusual conditions or events during the last Unit 2 operating cycle. Furthermore, since the three valves exhibiting the least disc-to-seat bonding were located on main steam line (MSL) C, the problem appears to be location dependent.

Consequently, the operation of Unit 2 during the last refueling cycle is being scrutinized, particularly the period from 10/20/85 to 11/23/85, when MSL C was isolated. The possible effects of high drywell ambient temperatures as well as frequency and extent of thermal cycling are being examined. During the summer of 1985,'a temporary technical specification change raised the allowable average ambient temperature of the Unit 2 drywell from 135'F to 140 F.

Following this, the installation and operation of the drywell chiller allowed continued full power operation through the remainder of the summer.

This problem is unique to Unit 2 at Brunswick. Due to differences in HVAC ducting design, Unit 1 is not hampered with seasonal drywell high ambient temperature problems. Reactor water chemistry was examined to determine whether water chemistry might have promoted corrosive conditions. The results NRC f 0 AW M4A 19 83

NRC Form 3B&A U $ NUCLEI 44 [EGULATORY COMMISSION LICENSEE EVENT REPORT (LER) TEXT CONTINUATION A**aOvEo oms 40 siso-oio.

EXPtRES 8 31 BB F ACILITY NAME tis DOCKET P" UMBER (21 LER NUMGER IG)

PAGE(33 "t?7

"'#,70 Brunswick Steam Electric Plant Unit 2 o l5 l 0 l 0 l o l 3l 2 l4 8l 6 0l ll 1 0]O O [7 0F 0l8 nxten m..

e.,-

unc v.~.anu nm of this study show that due to the retubing of the main condenser, improvements in reactor water cleanup (RWCU), etc., Unit 2 exhibited excellent reactor water chemistry during the operating cycle.

Prior to this time, excursions of reactor water chlorine levels above 50 ppb were not uncommon.

During this past fuel cycle, chlorine levels below 5 ppb were normal, with infrequent excursions to slightly higher levels.

In light of this, reactor water chemistry appears unlikely to have contributed significantly to the problem. Unit 1 underwent condenser retubing prior to Unit 2 and has subsequently maintained excellent water chemistry. Other possible factors considered include steam temperature drop in MSL C due to isolation, physical characteristics such as location and 1

position within the drywell, and abnormal conditions within the drywell (e.g.,

a HPCI valve in the vicinity of the SRVs had a packing leak for seme time).

For factors common to both units, Unit 1 operation was reviewed to identify any areas of concern. Nothing was identified in Unit 2 that might correlate with an increase in Unit 1.

The mechanism of bonding must be understood to determine if any of these factors actually contributed to disc-to-seat sticking. To accomplish this, six discs were sent to Harris E&E Center for testing. The metallargists there are currently investigating the cause and magnitude of bonding.

Resolution Continued operation of Unit 1 is justified based on the following:

1.

This event is more severe than any previously experienced at Brunswick or any other BWR and thus indicates an unusual condition on Unit 2.

Investigation has revealed nothing to date that might point toward a similar situation in Unit 1.

2.

The GE analysis predicts a peak pressure which is under the upset limit.

CP&L studies support this and identify significant conservatisms associated with the analysis.

3.

The conservatism in the GE model and in the assumptions used in the overpressure transient provide a substantial margin between the predicted result and any credible event.

4.

The GE analysis demonstrated that fuel thermal limits would not be exceeded, there was negligible impact on peak centerline temperature, and no impact on ECCS performance or drywell temperature. The potential increased loadings on the SRV discharge piping and torus were found to be within allowable limits. Potential increased pipe support loadings were conservatively analyzed.

.c. o.

x..

19 838

hmC Poem JBGA U $ NUCLi&A KEGULATO3Y COMwissioN LICENSEE EVENT REPORT (LER) TEXT CONTINUATION anemovie cwe ~o 3 iso-aio.

EXPtRES 8<3188 F ACILITV NAME til DOCKET Numeam (2e gg, gyggg,,,,

,, g,,3, "tO!'."

JJ'.O

aa Brunswick Steam ElectriE Plant Unit 2 ol5lo[ol0l312[4 81 6 -0 D lI 0 l 1.0 l 8 OF g lg TaxT m =

.===: -

._=nci ass 4.,nn 5.

Potential contributing factors will be monitored in Unit 1 to identify any condition which might affect SRV setpoint.

6.

During the next refueling outage on Unit 1, SRVs will be removed for testing at Wyle Labc.

l 1

i

% AC FO4W 344a e9 836 e

CA&L Carolina Power & Light Company Brunswick Steam Electric Plant P. O. Box 10429 Southport, NC 28461-0429 May 7, 1986 FILE: B09-13510C SERIAL: BSEP/86-0555 NRC Document Control Desk U.S. Nuclear Regulatory Commission Washington, DC 20555 BEUNSWICK STEAM ELECTRIC PLANT UNIT 2 DOCKET NO. 50-324 LICENSE NO. DPR-62 SUPPLEMENT TO LICENSEE EVENT REPORT 2-86-001 Gentlemen:

In accordance with Title 10 to the Code of Federal Regulations, the enclosed Supplemental Licensee Event Report is submitted. The original report fulfilled the requirement for a written report within thirty (30) days of a reportable occurrence and was submitted in accordance with the format set forth in NUREG-1022, September 1983.

Very truly yours, Y;

C. R. Dietz, General Manager Brunswick Steam Electric Plant MJP/mbh Enclosure cc:

Dr. J. N. Grace