ML20069E348

Rev 0 to Root Cause Analysis Rept Rcar 91.0001, Root Cause Analysis of Automatic Switch Co (Asco) Model L206-832-RVF Solenoid Operated Valves Associated W/2-G16-F003 & 2-G16-F004 Failure to Close Event

ML20069E348
Person / Time
Site:	Brunswick
Issue date:	07/28/1991
From:	Groblewski T CAROLINA POWER & LIGHT CO.
To:
Shared Package
ML20069E342	List: ML20069E348
References
RCAR-91.0001, RCAR-91.0001-R, RCAR-91.0001-R00, NUDOCS 9406070134
Download: ML20069E348 (41)

Text

_ _ _ _

/*7 h W f r

RCAR No. 910001 g.

a ROOT CAUSE ANAIJSIS REPORT j

Rey, so, o der,,

Page No.1 gop CAROLINA POWER & LIGHT

%h' gl BRUNSWICK STEAM ELECTRIC PLANT

(

(,

\\.

i hs ROOT CAUSE ANALYSES OF 1

i AUTOMATIC SWITCH COMPANY (ASCO)

MODEL L206 832-3RVF SOLENOID OPERATED VALVES ASSOCIATED WITH THE 2.G16-F003 AND 2.G16.F004 FAILURE TO CLOSE EVENT Dated July 28,1991 Prepared By The BNP ASCO Task F Tony Groblewski r,#

Bob S. Hams

}EA

~

David H. Hinds d

LeeJ. Kalkofen d

a Tun N. King S

Eric Rydzewski e,

H. Allen Walker

/ M S

Lury W. Wheatley

[

kk d

9406070134 910903 ct'60 O e/3 #

PDR REVGP NROCROR MEETING 209 PDR

/

f..,.,..-..

~.

1 t

n I

i RCAR No. 910001 Rev.No. 0

+

Page No. 2 of 19 1

TABLE OF CONTENTS 2

5-i I

1 Section 1

Pages..

I Cover Page 1

Table of Contents 2

References 3 4.

Executive Summary S.

I Root Cause Anahses 6.

Problem Desenprion 68 Scope of Root Cause Evaluation 89 Root Cause Evaluation i

. 9 16 Conclusions 16 17 Corrective Actions -

18 19 4

Attachments:

I l Failure Prevention Report

[40 Sheetsj 2 Chan ASCO SOV CoilHeat Rise 12 Sheetsj 3 Chart. Dow Corning 550 Silicone Lubricant Projected Life

{ 1 Sheet }

4 - Chart. Dow Corning Data Extrapolation i

[1 Sheet l m

)

RCAR No. 91 0001 Rev. No. O Page No. 3 of 19 References Adverse Conatuon Pyn 91290 LCO A2-91-1068 3.

LCO A2 911069 4.

ASCO Test Repon No. AQR-67368, Rev.1 (CP&L DR 5.2) 5.

ASCO Test Repon No. AOR-21678/TR, Rev. A (CP&L DR-5.1) 6.

Farwell & Hendricks Test Repon No.20226. Rev. O

%ermal Endurance Test Repon On ASCO Dual Coil Solenoid Valves Prepared For The Cleveland Electne 111uminating Company's Perry Nuclear Power Plant."

7.

AEOD Case Study Repon No. C 90-01 " Solenoid Valve Problems At U.S. Light Water Reactors.*

3.

USNRC E Bulletin 75-03, Dated 3/14/1975 'lacorrect Lower Disc Spring and Clearance Dimensions in 8300 and 8302 ASCO Solenoid Valves.'

9.

USNRC IE Bulletin 78-14, Dated 12/19/1978 'Detenoration Of Buna N Components in ASCO Solenoids.'

10.

USNRC1E Bulletin 7941A, Dated June 6,1979

• Environmental Quali5 cation Of Class 1E Equipment (DeSciencies In Tbc Environmenta1 Qualification Of ASCO Solenoid Valws1.*

1L USNRC IE Bulletin 8014, Dated 5/12/1980

• Degradation Of BWR Scram Discharge volu Capability.'

12.

USNRC IE Bulletin 80-17. Dated 7/3/1980 ' Failure Of 76 Of 185 Control Rods To F Insen During A Scram At A BWR.*

13.

USNRC IE'BuDetin 8017, Dated 7/18/1980, Supplement 1. ' Failure Of 76 of 185 Control Rods To FuDylasert During A Scram At A BWR.'

14.

USNRC IE Bulletin 80-17. Dated 7/22/1980, Supplement 2, ' Failures Revealed By Subsequent To FaDure Of Control Rods To Insert During A Scram At A BWR.*

15.

USNRC IE Bulletin 80-23, Dated 11/14.1980,

• Failures of Solenoid Valves Manufact Vakor Engineering Corporation.'

16.

USNRC IE Bulletin 80-25, Dated 12/19/1980, ' Operating Problems With Target Rock Relief ValWS At BWRs.*

17.

USNRC IE Notice 85 08, Dated 1/30/19&S,

• Industry Experience On Cenain Matenals Used in Safety-Related Equipment.'

i RCAR No. 910001 Rev. No. O Page No. 4 of 19 References tCont'd) 18.

USNRC IE Nonce 85-17, Dated 3/1/1985. *Possible S&bng Of ASCO Solenoid Vahes.'

19.

USNRC IE Notice 85-17, Supplement 1. Dated 10/In985. *Possible Sticking Of ASCO Solenoid Valves.*

20.

USNRC E Notice 85-47,6/18/1985, ' Potential Effect Ofline Induced Vibration On Certam Target Rock Solenoid-Operated Valves.'

21.

USNRC IE Notice 85 95, Dated 12/23/1985, ' Leak Of Reactor Building Caused By Scram l

Solenoid Valve Problem.'

22.

USNRC IE Notice 86-57, Dated 7/11/1986. ' Operating Problems With Solenoid Operated Valves At Nuclear Plants.*

23.

USNRC IE Notice 86-72. Dated 8n9/1986, ' Failure Of 17-7 PH Stainless Steel Springs In Valcor Valves Due To Hydrogen Embntilement.'

i 24.

USNRC IE Notice 86-78, Dated 9/2/1987, " Scram Solenoid Pilot Valve (SSPV) Rebuild Kit Problems.*

l 25.

USNRC E Notice 87-48, Dated 10S/1987, *Information Concerning the Use Of Anaerobic l

Adhesive / Sealants.'

l USNRC IE Notice 88-24, Dated 6/13n988, ' Failures Of Air. Operated Valves Affecting 26.

Safety Related Systems."

27.

USNRC E Notice 88-43, dated 6/23n988,' Solenoid Valve Problems.'

28.

USNRC IE Notice 88 51, Dated 7/21/1988,' Failure Of Main Steam isolation Valves."

29.

USNRC IE Notice 88-86, dated 3/31/1989, " Operating With Multiple Orounds in Direct Current Distnbutton Systems.'

30.

USNRC IE Notice 89-66, dated 9A1/1989,' Qualification IEe Of Solenoid Valws.'

31.

WPSC Report

• Failure Analysis Program For 'Ibe Automatic Switch Company (ASCO)

Model NP12314C28E Solenoid Valves (SOVs) For Use in Kewaunee Nuclear Power Plan 32.

WPSC Kewaunce LER 88-007 l

33.

NUREOKR 5141 ' Aging And Qualification Research On Solenoid Operated Valves.'

34.

Engineering Work Request (EWR) No. 07641VR 35 Rfwr Bend LER 88-023 l

l 36.

General Electne SIL 481 l

l 1

l

I 1

RCAR No. 91000i Rev. No. O Page No. 5 of L9 i

Executive Summary On June 30.1991 Containment isolation valves not closing. After proper reporting per pl established to determine the root cause of failure.

Based upon the information assimilated dunng this investigation, the root cause determination of the Unit 2 solenoid valve failures were primanly attnbuted to a gelling of the Dow Corning 550 Silicone Lubricant and some foreign particulate. The lubricant was used in the manufacturing process to facil assembly of the solenoid valve core subcomponents and to reduce chatter caused by the 60 cycle hum. Hi s

lubncant migrated to the surface of the core subassembly and solidified following long periods of rele valve energization. When solidified. the lubricant formed an amber colored residue which was respo the core assembly adhenng to the solenoid valve base sub assembly.

De failed solenoid vahes haw beca replaad with the same ASCO L206-832 3RVF model A Action has been established to weekly cyde all ASCO Model L206-832 normally energized solenoid valws This compensatory action is consistent with other utility correcuve actions which have previo this deficiency. De ASCO Task Force recommends that effected SOV's be replaad. Replacement s

' staggered" whereby the redundant compoecat SOV's woaM always be a recent replacement wh increase the safety and reliability of the system function. This additional defense through redu diversity is a recom=*adation per AEOD Case Study Report " Solenoid Valve Problems At U.S.

Reactors

• to elimmate rodeadaat hilures.

Additionally, the Root Cause Analyses Report provides recommendations for corrective actio to significantly reduce the occurrence of SOV failures.

1 i

1

R CA R No. 91 0001 Rev.No. O Page So. 6 of 19 ROOT CAUSE ANALYSES REPORT 1.0 Problem Description Containment Isolation valves The 2.G16.F003 and.F004 valves are located in the Drywell Floor Drain System piping and are components of the Primary Containment isolation Svstem (PCIS). The valves are three inch.150 lb.,

Anchor Darling cast carbon steel gate valves. De valves are opened by an air actuated cylind controlled by an Automatic Switch Company (ASCO) solenoid valve. The valves are closed b force upon loss of air pressure. Since the drain valves are open to the containment atmosphe redundant. automatic isolation valves are located outside the primary containment (i.e., Reactor Building).

Safety Function ne solenoid operated valves evaluated by this RCAR are the ASCO Model No. L206-832 3RVF assoaated with 2.G16.F003 and 2.G16.F004. Each SOV is located at Reactor B,rilding Elev. 30 ft.

Per BNP Safety Classifications, the solenoid operated valves safety function is considered 'Actrve) is ' required to close the Drywell Floor Drain Isolation Valve on a Group 2 Isolation Signal pnmary containment isolation *. FSAR Accident Sections 15.2 6,15.44 15.6.4, and 15.6.6 and Technical Specafications Sections,3.6.3,4.63, and B3/4.63 apply to these solenoid operated va A (manmum) normal ambient temperature of 104 'F has been assumed for this area for e qualification purposes.

This value is documented in Section 33 of the Reactor Building Environmental Report (RBER), UE&C Report No. 9527 058.S-MS401 (Rev3). This eq required to perform an active safety function following any High Energy line Breaks (HELB postulated to occur inside the Reactor Building. Additionally, the SOVs must remam opera through temperature fluctuauons resulting from a LOCA in the Primary Containment

.UE&C Report RBER, Figure 4-2, shows a 133 *F peak temperature occurring in this nea over the postulated 30-day post.LOCA duration.

De Total Integrated Dose (TID), over the SOVs 40. year normal and 30aiay HF1 R/LOCA service conditions, for this area is 1.O r 1r rads gamma.

Event Desertotion. June 30.1991 De following event desenption was summartzed from the Shift Foreman and Control Room Operatorslog:

05:19 Outboard Drpell Floor Drain Isolation Valve (2.G16.F004) failed to close on demand.

05:26 1 hour1.157407e-5 days <br />2.777778e-4 hours <br />1.653439e-6 weeks <br />3.805e-7 months <br /> phone report made to USNRC due to failure of 2-G16-F004.

~_

1-l.

t RCAR No. 910001

_Rev.No.0 Page No. 7 of 19 12:35 Techmcal Specification (3.6.3)'Line tsolation* required the redundant valve in line (2-G16 F003) be closco ensunng ime isolation De 2 G16-F003 sv.& was taken to 'close* and the valve rematned open (failed to change state). The switch was cycled 4 times and the valve closed at 12:55.

13:43 i bour phone report made to USNRCdue to failure of 2 G16 F003.

i I8.90 Stroke tested valves 2-G16-F019. 2 G16-F020. 2 Cl2-CV-F010,2 Cl2-CV-F0ll. 2 C12 V139.and 2 I

C12-V140 sausfactortly 22:10 SOV replacement of 2 G16-SV-F004 and 2 016-SV-F003 was completed.

23:00 Valves 2-G16 F004 and 2-G16-F003 were declared operable and returned to service.

Event Discussion On June 30,1991 at approximately 05:19, BSEP Unit 2 was operating at appronmately 95 following a peer reduction to perform Main Turbine valve testing.

A Control Operator was performing a weekly surveillance using an operating procedure to check the operation of the Manu (DC) and Auto (AC) Main Generator Voltage Regulator. After completing a step of the proce which places the Voltage Regulator Mode Selector in the Manual position, the Control Ope observed a significant deDection gDhe transfer voltmeter (TVM). De TVM monitors the ou the Manual (DC) and Auto (AC) voltage regulators and is used to match the outputs of the two regulators so the generator voltage does not change when the regulator is transferred. he observed deDection of theTVM indicated that generator excitation had decreased upon transferring from to the Mansal mode. De Control Operator asta! te Ser.!;r Control Operator if he had seca the TVM deGecnon and be replied that he had. De Control Operator looked again at the TVM and also observed that the generator megavais had decreased from 80 to 10. An alarm came in for a 250 volt Battery B ground folloned by an alarm for 250 volt Battery A ground. The Control Operato to receive alarms indicating a loss of Reactor Protection System Bus B and Emergency Bus Bus). At approdmately 05:19, the Bus 2C to Emergency Bus E4 master and slave brea due to actuation of 2 out of 3 E4 Bus degraded voltage relays. His initiated an auto start of the No. 4 Emergency Diesel Generator (EDG). One relay associated with the auto start logic for the No. 3 EDO actuated.

ECCS systems and Emergency Diesel Generators were operable and in standby.

Results of the loss of the E4 Bus included!

E*

Autostart of the associated Emergency Diesel Generator which re-energim! the E4 Bus.

Trip of the 2B Reactor Protection System (RPS) Motor Generator (MG) Set which resulted a4 in the following:

1/2 SCRAM-Division 11

RCAR No. 910001 Rev. No. O Page No. 8 of 19 Primary Containment isolation Sprem:

GROUP 2 ISOLATION Dinston 11 (Transverse incore Probe and D Equipment Dratn varses)(2.G16-F004 failed to closes GROUP 3 ISOLATION (Reactor Water Cleanup)

GROUP 6 ISOLATION (Contatament Atmospherte Control)

At approumately 05:29, the Control Operator manually raised the Main Generat with the manual voltage regulator and transferred the voltage regulator to the automade m Expected actuations resulting from a loss of the E4 Bus occurred as required e Floor Drain Outboard Isolation Valve (2 G16-F004) which failed to cl signal. After vertfication of actuations, the 2B MG SET was restarted.

Due to the failure of 2-G16 R)04, an eight bour active Limiting Condition of Opera accordance with Technical Specificatiorts (T.S.) was initiated. This Technical Speci that with only one operable valve, the affected penetration line be isolated within eig use of one descuvated automatic valve secured in the isolated position. ~1 bis required th the Drywell Floor Drain inboard Isolation Valve (2-G16-F003). At approximately 12:

45, the Control Operator attempted to close 2.G16-F003 valve using the control switch. At 12:55, after four attempts, the valve closed.

Although, the penetration line was already isolated, problems encountered in the closing of 2-G16-F003 required declaring the valve inoperable. With the Operational Leakage Surveillance required per Technical Specif After solenoid valve replacement on 2-G16-F003 and 2-G16 F004, the valves were declared operable and re service at apprtmmately 22:10 and 23:00 respecuvely.

Adverse Condition Report 91290 was tssued to ensure event notification per 10C 2.0 Scope of Root Cause Analyses / Evaluation ne scope of the root cause is directed at the ASCO solenoid valves (Model L20643 This is based upon field venfication during testing after the event of the SOVs failure (change state) venficauon ruled out any concern with the valve, valve actuator, or limit switch.

The following will be evaluated per this Root Cause Analysis Report:

a.

Root cause of failure.

Was the equipment known to be deficient prior to the event?

W-E Did the equipment history indicate that the equipment had either been histori or if maintenance or modifications had been recently performed?

Was any equipment vendor involvement prior to or after the event?

E' m.

He Pre event status of surw.dlances, testing, and /or preventatiw maintenance.

ne extent to witich the equipment was covered by custing correcuve action p 3;

the implicanon of the failures with respect to program effecuveness.

.. ~

n.

RCAR No. 910001 Rev. No. O Page No. 9 of 19 L206432 3RYF Description ne ASCO.1 ev solenoid operated valve. Model No. L206-832 3RVF is primarily used as operators on larger control valves. The followmg information is applicable to the BNP SOVs w 3

failed:

Pipe Size:

1/4' Orifice Size:

1/4" Max. Continuous Ambient:

140 *F Mar. Operating Pressure:

150 PSI i

Max. Fluid Temperature:

180 *F CV Flow Factor:

.45 Solenoid enclosure:

Explosion / Water Proof Body Material:

Brass Watt Rating:

20 AC Weight:

4 lbs (approximate shipping Weight) l 3.0 Root Cause Analysis /Evaluados Root Cause initial Valve Insnections and failure mode identification 1

2 G16.F003 A gelled Silicone lubricant was found to completely coat the top of the solenoid core. ' Die lubria i

was identified as Polymethyl Siloxane (i.e., consistent with Dow Corning 550 or one of the Neolub products) by infrared analysis. A few light scratches were found on the SOVs upper and lower ste A patch of copper bearmg material was found adhenng to the upper stem in a region tha throuth a brass bushing. A miculation was performed to demonstrata that if all of the co i

scaled by Dow Corning 550 that the resistant force from air pressure would cause the core to re in iu energized position after the solenoid coil deenergines. After solenoid decaer*=Haa. the a pressure induced resistant force gradsally disappeared as the sealing broke beams of the relashe l

thermal contraction between the core housing and the cote. A almlntion demonstrated that wh the inbriant gets over a period of time (i.e In a Idgh temperature environment), it will preve air pressure resistant force (62.51bf. inch) from allowmg the core to change position.

2.G16-F004 A gelled Silicone lubricant was found to coat about 50% of the top of the solenoid core. The lubricant was identiSed as Polymethyt SGazanc (i.e., consistent with Dow Cornag 550 or one Neolube products) by infrared analysm. Many 5 to 30 Miaon wide scratches were found on th stem. Additionally, a long thread machining burr was found between the upper seat bushing main valve body. This==*=g burr was still attached to the threaded region of the valves port. A calculation was performed to demonstrate that if a portion (Le.,50%) of the core outer radius was gelled by Dow Corning 550 that the resistant force from air pressure would not cau core to remain in its energized position aher the solenoid coil deenergizes. A calculation indicated air pressure resistant force (.3.27 lbf-inch), from slight gelling of thelubricant in a high environment, would not prewnt the core from changing position. This low value was in the area calculational uncertainty, therefore gelling will be considered as applicable and/or contrib I

i

~

- - J

l RCAR No. 91-C001 Rev. No. O Page No.10 of 19 failure mode, A calculation of the friction force on the lower disc guide stem due to the 5 to 30 micron scratches found on the lower disc guide was performed. The scratches (indication of high fnction forcess todicated a presence of a hi? strength foreign matenal between the stem and the bushing interface. It was determined that the scratches could not have originated from the softer brass' bushings and had to onginate from a hard foreign material. De total friction force of the -

scratch marks was calculated to be 1.14 lbf. Since the core can move freely downward, the i

gravitational force of the core becomes the force to open the lower exhaust valve. This is due to ',he lower disc spring force being almost zero when the lower valve starts to be pushed open by the Icw.

A calculation of the resistana force (0.314 lbf. inch) results indicates the SOV would not change state. De binding of the lower core with the partial gelling of the lubricant is the most probable failure mode.

Outstandinn Eauipment Deficiencies PHor To ne Event-There were no existing deficiencies identified with the 2.G16.SV.F003 and 2.G16-SV.F004 pnor to the event.

BNP ASCO SOV Failure History Ex.gineering has perfonned a review to determine the failure history of various model ASCO solenoid valves installed at BSEP Units t & 2. An EDBS data query was initially performed to identify the entire population of ASCO solenoid valves utilized at BSEP (2395 valws); this query identi6cd valve model number, tag number, quality classification, and description. Further research was performed to identify the components which are ideati6ed as andene safety related and could be cuompass the failure mode determination for global plant impact. De following cnteria was utilized to ide

)

tbese components:

ASCO Solomold Valve QualityClass T CoS NormallyEnergimi De soleaoid valves (appror. 628) which meet the above noted criteria were identi8ed by review Environmental QnaHArntion (EQ) data, previous research incinded within EER 880(D6, and information supplied by various Systems Engineers.

Of the 628 solenoid valves which met the faGure mode criteria,548 of these valves are identi6ed as the Scram P0ot Valves for the Control Rod Drive system (CRD); these valves are ASCO Model

1. HVA904052A.

General Elecxtic Company (GE) generated an operating experience report for faDure history of these components (Ref. NEDE.22292); this review gathered data on 5490 valves installed at 23 different plants which had beca operating as early as 19tio. Relative to the number of valves conadered, and reactor years of service (esthanted at 198 years),less than 50 BUNA.N related i

nive failures were reported. Of this number,10 failures were known to be attributed to core disit i.

failure. Due to the research performed by GE, engineenng deteramed further investigation of thes model valves was not warranted; the subject solenoid valves are replaced on a regular scheduled basis to ensure the service life is not exmeded. Herefore, the CRD scram pilot solenoid valves were not included in the ASCO failure history investigation.

l l

- ~. -. -.,.

RCAR No. 91-0001 e

Rev. No; o Page No.1i of 19 Due to the large number of solenoid vahts idenufied (80), a sample lot of vanous model solenoid i

valves has been researched to determtne the failure history of these components. The sample tot us linuted to 25 percent of a subject model number and size. Where four or ten -Ives were idenufied-3 the sample lot increased to 50 percent. The sample valves researched were randantly selected from j-the list of components which met the failure mode criteria, if a solenoid failure was attnbuted to a

' sticking

• problem, the sample lot was increased by one sample lot for the affected model number and

{~

stze.

i once the sample lot was determined, a historical WR/JO review was performed, utilizing AMMS l

(Automated Maintenance Management System), to determine failure history. The AMMS query was j

performed for each component identified by researching the actual solenoid valve tag number, i

process valve tag number, and process valve actuator tag number (e.g.,2 G164V F003,2-G16-F003, i

and 2-016-F003-AO): this extended research was performed to better ensure valve failures were captured. The AMMS query also included review of the General Inquiry Menu (actim plus 3 year j

WR/JO history) and Ar: hive inquiry Menu (completed WR/JO >3 years); reviewing AMMS General t

inquiry and Archive Inquiry will capture work orders petformed since November 1985.

3

{

Engineering also performed a NPRDS query to capture identided solenoid valve. failures pnor to i

November 1985. Due to the methodology utilized by NPRDS, the query was performed to identify j

valve operator failures at BSEP Units 1 & 2. NPRDS does not list solenoid valves as a individual components but rather as a piccc-part of the process valve operator. Engineering performed a review j

of the NPRDS data received and has included the identified solenoid valve failures in the sample lot.

Below is a listing of sample solenoid valves researched including model number / size, tag number, and failure history:

\\'

1 Model Number' Tan Number Failures 4

)

Filt345E11 2 CAC.-CV-2890 SV4 N/A

)

1 HTB302B25RU 1 E41 F0284V3 N/A e

2-E41 F028-SV3 N/A 4

i i

HT8321A5 1-SW-V137 SV3 N/A j

2 SW V123 SV3 N/A 1.

RIlt321A6 1 SW-PY 116 N/A 1 SW-PY-136 Replaced due to corrosion f.

24W PY 118

_ Replaced due to corrosion j

Kr8345E11 2-CAC-CV 28894V4 Airleak Solenoid body 4

j L206432 2RF 1832-SV-F019 N/A 2 832 SV F019 i

Poaalble solenoid sticking Root canaeindeterminate (1988)

L206-832 2RVF 1 B32 SV-F020 N/A 2 B32-SV F020 N/A l

4 4

---.,_,--.-_m._.____,--..

~...

~.

RCAR No. 910001:

Rev. No. o 4

Page No.12 of 19 4

L206-832 3RVF 2 G16-SV F003 Solenoid sticking (1986. 87,8& & 91) 4 2 G16 SV F004 Solenoid stickJng (1987. 88;& 9 t) 2 G16 SV F019 Solenoid sticking (1988) 2 G16-SV F020 Solenoid sticklog (1988 & 90) 1-016-SV FD03 N/A 1016-SV F004 -

N/A 1 G16-SV F019 Sluggish stroke (1991)-

{

t G16.SV F020 Solenoid sticking (1990)

NP12316A57V'

'l CAC SV VS IS N/A 2 CAC-SV V61S N/A i

NP18316A65V 1-C11 SV-F009A

'N/A 2 C12 SV F009B N/A 4

NP12321A2V 2 SW V129 SV3 N/A NPL8321A6E 2 SW-V141 SV3

-N/A NP832093V.

1 CAC-SV-4409 '24 N/A 2 CAC-SV-4410-24 N/A NP8321A2E 1-SW V124-SV3 N/A 1

l NP8321A6V 1-SW V136-SV3 '

N/A 1

- WPlfIll321A1 1-SW-V123 SV3 N/A 2-SW V128-SV3 N/A' NP12323A36E 1821-SV F028A 1 Defective Coll i

i 1 B21 SV F02881 N/A 1 B21 SV F02801 Valve bodyleak 4

i 1 B21 SV F028D 1 N/A 1

NPm;%36V t 821-SV F922A 1 Solenoid sticidag (1985) 1 B21 SV R)2281 N/A 1 B21 SV F02201 N/A 1

i 1 B214V F022D 1 N/A 2 B21 SV R)22A 1 Defective con i.

2.B21 SV R)22B 1 Defective solenoid 2 B21 SV F022C 1 Defective coH 2 B21 SV F022D-1 Defective cod j

2 R21 SV Fe28A 1 Solenoid sticidag(1985) 2 B21-SV F028B 1 Defective con 2 B21 SV F028C 1 Defective coll / Solenoid sticking (L985) 2 B21 SV F028D-1 N/A A historic review of the BNP ASCO solenoid vane failures indicate the highest failur occurnng on the ASCO Model # l206-832 solenoid valves Other solenoid valve models, f histoncal search, do not indicate failure trends or common mode failures at a leve easts. The ASCO Model #8323A36E/V also reflects a high faQure rate; this model v the Main Steam isolation Valves (MSIVs). Review of the failure date ure

~

i.

RCAR No. 910001 j-Rev. No. 0 -

Page No.13 of 19 nys have not reoccurred since 1985. this change in the failure rate is due to a design c

.mplemented for triese components. Since the MSIVs have not expenenced solenoid valve

.ccent years the design change impamented appears to have corrected the root cause p i

therefore, these model solenoid valve., are not considered high fatture rate valves pe j

investigauon.

l industry Failures s.

The history of industry SOV failures is extensive and has been musidered in this root caus t

evaluation. This report is not intended to reveal all SOV failure data as presented in large 4

research propets, but to show similarity to the lubricant (i.e., residue. FUSS [ foreign unident i

sticky substance}, amber colored sticky material. organic deposit, sticky deposit, sti

{

involved and/or identified in numerous SOV failures.

Additionally, most failure analyses had 4

concerns with the degradation of the Ethylene Propylene Diene Monomer (EPDM) clastomers tended to focus on the clastomers in the final root cause determination.

1 Franklin Research Center (NUREG CR/S141)

A research program was conducted for the Division Of Engineering OlBce Of Nuclear j

Research on the aging of solenoid operated valves.

During this test program. ASCO SOVs failed to transfer (chante sute) when j

one fourth of the accelerated thermal aging had been completed and it was tim the Erst increment of 500 operational cycles. hs'aiaations of all six new ASCO SOV i

undergoing acmierated thermal aging revealed an organic deposit at the top of the assembly. De deposit appeared to have acted as a sticky substance that prevented I

Additionally test specunens from the Fort Calhoun Plant were found to have deposit. A detailed analysis (e.g., using IR techniques) of the sticky substance was not i

possible within the budgetary restraints of the test program. Tbc material is used in th masefacturers assembly process of the solenoid valves.

Grand Conf 1

1985-MSIV Failures related to solenoid valve problems were related to foreign s ASCO feh cenaan that the valve failures resulted from high temperature sticking mre to plag not interfaces resulting from a foreign substance or combination of saw===

conected at this interface. There was not enough residue for deanttwe identificatio nature of the foreign substana.

KewauneeIER 88 487 i

Two redundant containment isolation valves failed to transTer (change state) a mansfecr-ds use of a lubricant, resulting in degradation of co=*ala'arat integ ASCO failure was attributed to a lubricant P 80. However, sigm6 cant amounts of D i

Coratag 550 SGicone Lubricant were prescat. De P 80 lubricant has been remov i

ASOD manufacturing process. After the testing, there were concerns raised abo of ASCO introducing contaminants during assembly. systems

RCAR No. 91-0001 Rev. No. O Page No.14 of 19 3

Perry (CED -

3 October 29.1987.Three or eight MSIVs failed to close due to solenoid ulve failures.

3 November 3.1987.Two of eight MSIVs failed to closa

{

et to solenoid valve problems.

November 29,1987.One of eight MSIVs failed to close due to solenoid valve pro

)

i nermal endurance tesung performed by Farwell & Hendricks for CE! concluded that i

EPDM elastomers and lubncants were the cause of failure to shift. During t i

foreign substance desenbed as an amber colored sticky residue was identified. His i

i j

was similar in composition to the lubricant applied to the valve internals (i.e., Siliconc lubricant).

1, -

~

3 1Mialle (Com. Ed.)

s j

December 17.1987. ne failure analyses by LaSalle demonstrated that the cohesive / a 4

force caused by a foreign sticky substance between the plug nut and the core )

j

{

significant and could have caused the failure. After the core assemblywas held v i

plug nut was pressed against the core assembly. He plug out was then let go. De adhes l

forces from the foreign substance between the two surfaces were able to sup 9

of the plug nut to prevent it from falling.

1 j

River Bend i

J l

]

September 30,1988. Two MSIVs fsfed to close due to solenoid valve faGure to shift.

root cause was determined to be gelling of the lubricant (Dow Corning $50) (Referen j

IER-88423).

General Electric I

s-4 SIL No. 481 identified testing of the Dow Corning $50 Lubricant which dried

)

i

)

with time. General Electne could not contribute the failures of MSIV l

solenoid valve failures were attributed to the clastomer materials.

San Onofre (SCE) i 4

]

1987. San Onofre experienced five SOV failures of the ASCO Model L206 380 3R!

(Reference IER 87 016).

deternuned that the valve design did not recognize the j

core top when the valve is used in a nonnally energized sernce application. D i

identifies Dow Conung $50 to be the ideati6ed material which gelled k

DOW Corwine 550 Evaluation i

i.

i FaGure Prevention Inc. (FPI) identi6ed gelling of the silicone lubricant by the

}

report (Reference Attachment 1). Dow Corning Data published literature identi5cs that Silicone Labricant gets at a temperatore of 200 'C (392 *F) at 14-smooths.

Note:

ASCO has not identi6cd any age related temperature information concer DowCorning $50 Silicone lubricant.

3 Further investigation into other BNP ASCO solenoid valves which could be gel 4ng concerns was performed.

4

^

4 i

Using ASCO solenoid valve temperature heat rise data for normally energued sole BNP a heat rise vs BNP ambient temperature plot (Reference Attachment 2) w l

best rise w ambient temperature plot was used to identify any ASCO solenoid models i

RCAR No. 91-0001 Rev. No. O Page No.15 of 19_

potentially be most suscepuble to the Dow Corning 550 lubncant gelling. Discussions rese.11cd that the ASCO published heat nse data is based on using a reststance calc Mus 5 *C marg 1n for the coil hot spot.

Evaluation of the heat nse data n BSEP smbient temperatures (Reference Attac i

extensive heat nse temperatures for the ASCO SOV model L206432 vs other ASCO S BSEP Units 1 & 2.

De data supplied by FPI provided a temperature plot of the Dow Corning 550 Silic thermal capabilities. De data was analyzed for heat rise effects of the normally en SOVs and the Dow corning 550 Lubricant gelling. Using the Arrhenius Meth spect5c temperatures (25 *C. 40 *C. and 65 *C) were graphed (Reference Att energized ASCO solenoid vahe models. Dey were graphed to identify any other ASCO nhes which would potentially have a gelling condillon.

Evaluation of the Arrhenius data revealed signi6 cant reduction in life erpece*7 due heat nse temperatures for the ASCO SOV model number L206432 vs other ASCO BSEP Units 1 & 2.

j ne area, where the lubricant gelling was identified (Top Core Sub Assembly) was temperature would have been less than the coil hot spot used la the Arrhenius quah of a similar solenoid valve 2-016-SV.F019 which is co the BNP calculations. De temperatures were 295 T (146 'C) for the Top Core i

VAC at an ambient of 105 T (40 *C). De temperatures as espected were diroc the voltage apptred to the coit Dese temperatures were considerabV less th temperature of 400 T (204 *C). These measurements were takes wit; thermocoui vanous metal interfaces to prevsat convecuve losses. Using the three Dow Cora sed to i

Lubricant to various BSEP temperatures. trend graph (Reference Attach lower than that published by the manufhee rer Dow c matenals testing by Dow Corning on the 550 Sillmne Lubricant in a circulating air ei ASCO application is is a non-arculating air environment until the valve is cyded. A contributing factor in accelerating the aging process of the Dow Corning SDicone Im

{

A&litionally, the amoest o(lubricant testest at Dow Corning versas the am

'j the accelerated gelling,to the ASCO component (Le., thin film) in the manuf r

i Note:

Dunng post ewat discussions, ASCO stated that their procedure for ap Corning 550 SiUcone Lubricant is non specific on how the lubricant is ap the quantity to be used.

a At m

J,i-4A ea

.j*

RCAR No.910001 t

- Rev. No. O i

Page No.16 of 19 f

Ecoiement Vendor (ASCO) Involvement (Prt.Esent) nere has been previous eg::~aent vendor involvement associated with SOV fai I

MSIV incident). Additionsdy, the equipment vendor has participated through the Nucl Group On Equipment Quali6 canon (NUGEQ).

There has not been any vendor involvement assoctated with the L206-832 model or tubricant degradation, except transmission of heatI which has been incorporated into the BNP EQ Program. He SOV vendor has n informadon assocsated with the Dow Corning 550 Silicone lubricant and age related fi mechanssms.

Eauloment Vendor (ASCO) Involvement (Post Event)

ASCO was invited to parucipate in the Root Cause Analysis at the site or Harris E & E ASCO did not participate but indicated they would provide the root cause analysis if th:

components were provided.

Phone question and answer sessions have been held with the SOV sendor.

Pre. Event Status Of Surveillances. Testine. And /Or Preventadve Malatenance.

Surveillances and testing was performed in the time required by plant schedules u event. Testing performed (i.e., stroke testing) indicated no wh of solenoid degr the failure.

j De vehes stroke time data for all drywell drain valves were reviewed and graphed fro present time. Based upon this review, it was determined that the valve stroke time i

indicauon of the imminent valve failure.

increase in stroke time in the test prior to failure, however this data scatter for this valve. Dere was no increase in the stroke time data prior to the va SOVs were functionally tested prior to installation by the maintenance group. His t connecting to the Service Air System and cycling the valve for proper operation.

Post Event status of systems affarW by ASCO SOY I M = Gelline R z The tuo SOVs involved in the fanure have been replaced with the same ASCO Mod 832-3RVF.

Instrument Air System was analyzed downstream of the Elter for contamina the following results:

Sample Volume 1/2 Cubic Foot Hydrocarbons

<1 ppm Dew Point

-45F Particulate

.5 micron

@ 5000 parncies 2 micron.

@ 20-30 parncies 3 micron

@2 Insigni6 cant numberlarger (170 micron @ 1)

I m..

... _~..,, l

i

• R CAR No. 910001 j

Rev. No. O Page No.17 of 19 Semcc AJr Splem (air used for functional test) was analyzed for air quality with the fo Hydrocarbons

< 1 ppm i

De voltage measurements at the failed solenoid valves were as follows:

LG16-SV-F003 116.63 Vac 2-G16-SV F004 115.95 Vac I.'

Ambient temperature (T) measured at the failed solenoid valves was 102 *F.

i j

Radiation environment surveyed at the failed solenoid valves was 400 to 500 mr/hr gamma-i i

For Compensatory Action normally energized ASCO L206-832 solenoid valves are b weekly intervals to reduce potenual gelling effects. This compensatory acuon is consistent with j

utilities correcuve actions which experienced this deficiency.

4 i

i ne extent to which the eovioment was covered by existing cos.dve action sir..ms and the implication of the failures with respect to proeram effectiveness.

1>

i Previous corrective action programs were performed associated with ASCO solenoid vatve i

i vatws were modi 5ed with an elastomer preference (Viton) when the industry had problems i

Ethylene Propylene Diene Monomer (EPDM) elastomers. This correctrve action has pr{

eNective in reducing problems with ASCO SOVs assocasted with the MSIVs. Additio participates in the Nuclear Utilities Group On Equipment Quali6 cation (NUGEQ) which discus many plant failures nuaa=ted with age degradation and EQ related issues.

]

i EWR 06741VR (inidated on 8M/90) directed review of the NRC P:stheimmsy cas Solenced Yahefrobismus At U.S.I.Jght Water Remenos". Disposhion antamisial =q na=====dasi=== of EWR OM41VR included perdonning tratming andAar proceda approprinsa, so initimes a Root Chase Amatyeis for amy SOV dnBuss klassiSee during s tesdag, to assure adhssents to mammiscturers tobrecasica tastrucdoms, and co mainismanewassalag, and eqdma*==== of redmadent SOV't on a staggesed hem i

implementados of the W actions had not been completed at the thee of this cuest.

)

Note:

BSEP currently does not disassemble ASCO solenoid valves which is in accordan with ASCO recommendations Currently failure trending is not performed for ASCO solenoids at BNP.

i j

3.0 Condusloos 1

Based upon the information that was assembled per this root cause analysis the fol

?

aad resolutions were reached.

)

i The Dow cornmg $50 Lubricant showed evidence of age related degradation (

5*

i normally energized ASCO Solenoid Operated Valve Model No. L206-832 top surfac I

core assembly and contributed to the failure of the SOV to change state.

All normally energized ASCO Model No. L206-832 Solenoid Valws need to continue cyded weekly as compensatory action.

r- -,-e,

-w

--w-,

---

p-,m.p e-m

~_

?

RCAR No. 910001 Rev.No. 0 Page No.18 of 19 I

All normally energized ASCO Model No. L206 832 solenoid valves should be replaced. In j

redundant spterns the inboard and outboard solenold valves should be replaced o staggered basis per AEOD Case Study Report No. C.90 01 " Solenoid Valve P W ms A U.S. Light Water Reactors *.

i Perform further testing of the Dow Corning 550 to identtfy any synerpsms associ the ASCO application.

i Additionally, include the ASCO heat rise data (Le., actual measurements vs resistive calculation method). His should be performed on an basis.

Perform feasibility study for design modification options for a different type of S model No. which could have a lower operating temperature and submit to p i

j E

{

The ASCO Model No. L206-832 solenoid valve has exhibited an abnorma (Le., sticking). If a SOV Fallure Trending Program had been in place, the subje l

j solenoid valves would have been previously identified as a problem model valve. T more in depth root cause analysis would have been performed to establish corrective

(

to reduce /climinate subsequent valve failures, i

t i

SOV Failure Trending Program should be implemented to monitor / trend BNP S performance. SOV Trend data can provide indicators through evaluations / resol failures to ensure that failures will decline.

I 4

E The ASCO solenoid valve testing does :.ot demonstrate conditions of a soleno i

only periodically cycled. ASCO testing performed cyclic aging (20,000 a

{

which may not be representative of actualinstalled conditions.

1 i

Periodic cycling is to be investigated for ASCO L206-832 SOVs wb2h m routine basis (i.e., < 1 month) to alleviate concerns of sticking.

E haa===d vahes are fhacnonally tested prior ao field i=arait**ian to use Service Air systas (particulate iDiered but not oQ Sitered) wts pracdos k 4

i internal=====a==rtaa of a SOV. FPis soport did not identify cone==enant have beca induced due to air quality at BNP.

(P*

jl Although a sanan preinstallation test could lead to a came==ie==a b p

6

=a====eaa psocedars br beads testingshould be i ' f and/or scrised br useo systent (equhatest.to the plaats instrustat air or other engiacestag appetne e supplysynes), which k tree of-->==== which could contrhete no SOY senarequha j'pj n

y4 Based upon the above RCAR evaluation and compensatory actions,it can be Dow Corning 550 materials' physical endurance properties to the postulated LOCA, and post LOCA environmental mnditions wouldL got impair the safety 5

v during the compensatory time penod natR the vahes can be replaced with a acordsace with ABOD Chse Study Repost No. C 9H1

• Solenoid Valve P Water Reactors' recomenendations.

His RCAR provides the root cause analysis assessment and assures the e operability cornpensatory actions currently taken.underits expect

RCAR No. 910001 Rev. So. O

{

i Page No.19 of 19 De RCAR corvrctive actions pronde reasonable assurance that postulated common mode equipment failure resulting from a normal and Design Basts Event (Dd' E) exposures would n m

degrade its safety fu.5'mt. the components would ny cause the failure of any other satety related funcuon or equipment, and the components would n_ot result in misleading information to the operator dunng anyinstallation.

3 Steps to orevent recurrence are addressed per this RCAR assessment by providing the followin

]

recommended correctrve acuons.

4.0 Recommended Corrective Actions l

).

The following short term and long term corrective actions are recommended per this RCAR.

1 i

Short Term recommended corrective actions 4

1 1)

Develope a staggered schedule for SOV replamments for all normally energized model ao.

L206432 ASCO SOVs per AEOD Case Study Report No. C-9001 " Solenoid Valve i

Problems At U.S. Light Water Reactors

• recommendations.

)

2)

Replam all Normally Energized L206-832 ASCO SOVs yearly (12-months)in accordance with plant approved procedures. This is to be performed ce a staggered basis (6.moesks) for rdandaan systeestualves taldag into consideration the AEOD Case Study Report No. C.90 01"Solenote ValveProblems At U.S. Ught Water Reacsors' raaa===daeia==

3)

Continue to perform weekly (7-days) cycling of normally Energized L206432 ASCO SOVs.

This isttrbe performed on a staggered basis (3 to 4. days) with redundant systems taken into 4

i consideration. Operations procedures for stroke time associated with ASCO SOVs shoeM be seviewed and sevuodjp include local venScation of sa8=*t performanos and staggesing of strata asses per ABOD Ossa study Report No. C-9001 "Solomoid Vahe Problems At U.S.

1.JghtWater Raatsass'r-Maeinaa 4)

Reviour and revins anissoid e' procWores/ practices to ensare that coaramineans are not introduced daring inaranatina andAor shop testing.

Imag Term -MM corrective actions

}

1) i Perform feasibility study for design modi 5 cation options for a di8ferent type of SOV or model No. with lower heat rise values and submit to plant for approval.

2)

!=,-: --st a solcooed vahe failure trending program to further identify SOV failure trends and to initiate root cause evaluations with corrective actions.

3)

Perform testing on the Dow Corning 550 Silicone Lubricant for gelling characteristia and or synergisms nuannted with the ASCO SOVs. This should be done on an industry basis.

a 4)

Revise EQ dxumentation to reflect the replacement times established for the ASCO SOVs l

asaresultof thislavestigation.

_ _ _. _,a

1 9

a 4

1 to the Minutes of CRGR Meeting No. 209 ProDosed Generic Letter on NRC Upgrade of the Emeroency Telecommunications System (ETS) i August 27, 1991 TOPIC R. Wessman (AEOD) and T. Kellam (IRM) presented for CRGR review a proposed 3

i generic letter alerting power reactor licensees to the forthcoming NRC effort to upgrade the agency's emergency telecommunications system (ETS).

The major expense of this effort is to be borne by NRC; but some level of licensee i

i effort and cooperation will be required for successful implementation of this program, so this action is considered as a backfit.

The staff invoked the compliance exception of 10 CFR 50.109 in justifying the proposed action.

Briefing slides used by the staff to guide their presentations and discussions with the Committee at this meeting are enclosed (see Attachment).

i BACKGROUND a

The documents submitted for review by CRGR in this matter were transmitted by i

memorandum, dated August 12, 1991, j

package included the following items:R.L. Spessard to E.1. Jordan; the review 1.

Draft Generic Letter (undated), " Emergency Telecommunications", and i

attachments as follows:

1, 3

3

a. -

" Essential Emergency Communication Functions",

b. -

" Licensee Support for' Upgrade to the Emergency 1

Telecommunications System",

4

c. - " Schedule for FTS 2000 Installation",

I

d. -

" Documented Evaluation" (providing the basis for

)

invoking the compliance exception).

i 2.

Commission Paper (SECY-91-149), dated May 11, 1991, " Upgrading the NRC Emergency Telecommunications System",

3.

NRC Staff Responses to CRGR Charter Section IV.B., "Information Require for Packages Submitted to CRGR".

4 4

5 i

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

COMMENTS / RECOMMENDATIONS 1.

As a result of their review of this matter, including the discussions with the staff at this meeting, the Committee recommended in favor of issuing the proposed generic letter, subject to incorporation of revisions to clarify and strengthen the justification provided in the package for proceeding under the compliance exception of 10 CFR 50.109, e.g.:

Highlight the fact that after May 1992 the _ Emergency Notification a.

System (ENS), required explicitly by NRC regulations, can no longer be practicably maintained in its present configuration due to unavailability of key components.

Therefore, in the absence of an acceptable alternative, licensees will be unable to demonstrate compliance with the requirement for that essential communications link after that date.

The NRC has determined that FTS-2000 is sufficiently reliable to be an acceptable alternative for ENS purposes; and it is_ less costly than other available alternatives (e.g., satellite link, microwave link, foreign exchange lines).

b.

Note explicitly that-upgrading of the Emergency Response Data System (ERDS), also required explicitly by regulation, is already underway using FTS-2000; and the proposed overall ETS upgrade is consistent-with that agency action.

Cite the extensive background (e.g., regulations, approved _ guidance, c.

Commission Papers, etc.) that has established the agency position that providing the seven essential ETS communications links identif-ied in this package (i.e., ENS and ERDS plus the RSCL', HPN, PMCL, MCL, and LAN links) is an acceptable means of complying with the overall requirement for reliable emergency commications capability.

Also state that the NRC has determined in_ connection with this pro-posed action that all seven essential ETS communications pathways should be upgraded by use of FTS-2000 in order to ensure the needed reliability for ETS, and complete compatibility among all of the component parts of ETS.

All changes made to the package in response to this recommendation should be closely coordinated with the CRGR staff prior to final issuance of the i

generic letter.

2.

As a separate matter, to be pursued on a basis not to interfere with expeditious issuance of the generic letter and implementation of the needed ETS upgrades identified in this package, the Committee recommended that the staff examine whether adequate consideration was given previously to the possible need for an eighth essential emergency communications link, dedicated to use by NRC/ licensee safeguards officials in the case of an emergency situation involving significant safety and safeguards elements.

simultaneously.

PRESENTATION OF DRAFT GENERIC LETTER ON EMERGENCY TELECOMMUNICATIONS TO CRGR 1

l i

1 BY:

R. WESSMAN, AEOD l

T. KELLAM, IRM t

4 x

%1 3 o

+

n n sh i

a i : ;

AUGUST 27,1991

3 t

1,,

i

'u N

o s' 9 e

b

BACKGROUND SIGNIFICANT COMMISSION DISCUSSION t

SECY-87-290 (NOVEMBER 27,1987) l

-PROPOSED 3 ALTERNATIVE SYSTEM UPGRADES

-DESIGNS FEATURED REDUNDANCY AND DIVERSITY

-SATELLITE AND TERRESTRIAL NETWORK SECY-89-340 (NOVEMBER 2,1989)

=

-UPDATED COMMISSION ON STATUS AND SCHEDULE

-DISCUSSED NEED FOR RULEMAKING SECY-91-149 (MAY 21,1991) RECOMMENDATIONS l

-lNSTALL FTS 2000 l

-EVALUATE RISK /NEED FOR REDUNDANT AND l

DIVERSE PATH l

l I

i i

l

..._..,..-.-u---_-_---__._--_---__-___--n___--_-_--___----__

BACKGROUND CURRENT EMERGENCY TELECOMMUNICATIONS CIRCUlT SERVICE PROVIDER PROBLEMS ENS NRC/ DEDICATED OBSOLETE HARDWARE HIGH COST HPN NRC/PSN PSN BLOCKAGE 1

RSCL LICENSEE /PSN PSN BLOCKAGE PMCL LICENSEE /PSN PSN BLOCKAGE 1

MCL LICENSEE /PSN PSN BLOCKAGE LAN ACCESS (E-MAIL)

LICENSEE /PSN PSN BLOCKAGE ERDS NRC/PSN/FTS2000 PSN BLOCKAGE i

t 9

i

DISCUSSION CURRENT STATUS ENS SERVICE LIFE EXTENSION FTS 2000 SATELLITE SYSTEM STATEMENT OF WORK i

RISK ANALYSIS l

i I

l 1

l 4

1 i

i i

4 I

i 1

4

j DISCUSSION GENERIC LETTER 1

I

• UPGRADES NRC EMERGENCY TELECOMMUNICATIONS TO AVOID OBSOLESCENCE AND ENSURE AVAILABILITY

+

k NECESSARY TO MAINTAIN COMPLAINCE WITH CURRENT 1

EMERGENCY TELECOMMUNICATIONS REQUIREMENTS

-10 CFR 50.47 (b)(6)

-10 CFR PART 50 APPENDIX E, IV.E.9d 4

REQUIRES LICENSEES TO:

I

-PROVIDE INSIDE WIRING

-lNSTALL NRC PROVIDED TELEPHONES

-ESCORT NRC CONTRACTOR FOR SERVICE INSTALLAT!ON

-REVISE PROCEDURES AS APPROPRIATE FOR ENS DIALING

-lN SOME CASES PROVIDE SPACE AND DEDICATED POWER FOR MULTIPLEXING EQUIPMENT i

i l

DISCUSSION COSTS-LICENSEE COSTS

-$2.5K FOR INSIDE WIRING

-CABLE INSTALLATION FROM. CENTRAL OFFICE NOT INCLUDED

-SPECIAL INSTALLATIONS (i.e. PENETRATION

-OF VITAL AREAS OR INTER CAMPUS WIRING)

NOT INCLUDED NRC COSTS

-$1.3M INSTALLATION COSTS.

-1 FTE (SUPPORTED BY CURRENT STAFFING)

-$300K SYSTEM INTEGRATION CONTRACT

-$23.1K ANNUAL OPERATING COST i

NRC COSTS MORE THAN OFFSET BY CURRENT $4.4M i

ANNUAL COST OF EMERGENCY TELECOMMUNICATIONS i

i e

i

4 i-i to the Minutes of CRGR Meeting No. 209 Procosed Supplement to Generic Letter 88-01 to Modify 4

Augmenteo Insoection Reouirements Relating to IGSCC in 8WR Pioing

)

l 1

August 27, 1991 TOPIC i

B.D. Liaw (NRR) and W. Koo (NRR) presented for CRGR revin i proposeu i

generic letter to modify some aspects of the augmented inspe: tion require-ments contained in GL 88-01 for detection of intergranular stress corrosion i

{

cracking (IGSCC) in BWR. austenitic stainless steel piping.

The proposed changes include relaxations of existing staff positions on IGSCC inspections,

)

and clarifications of other IGSCC positions.

The staff requested waiver of I

formal review of this package by CRGR; but the Committee reviewed the item j

because of possible safety implications of proposed relaxation of IGSCC inspection frequencies.in reactor water cleanup piping, and because some of the clarifications arguably involved backfitting.

Copies of the briefing-i i

slides used by the staff.to guide their presentation and discussion with the Comittee on this item are enclosed (Attachment 1).

BACKGROUND l

1.

The documents submitted for CRGR consideration in this matter were trans-mitted by memorandum, dated August 1, 1991, F.J. Miraglia to E.L. Jordan-l the package included the following documents:

1 Revised cover letter for the review package.

(An earlier version of a.

this letter was submitted to CRGR previously to transmit proposed relaxations to IGSCC inspection requirements; however,'the earlier package was withdrawn by the staff in order to include new staff-i positions on IGSCC inspection as well as relaxations, as reflected in the current package.)

t b.

Draf t Generic Letter No. 88-01, Supplement 1 (undated), "NRC Position on IGSCC in BWR Austenitic Stainless Steel Piping" i

NRC Staff Responses to CRGR Charter, Section IViB. (Contents of c.

Packages Submitted to CRGR) i 2.

CRGR members were also provided copies of a letter, dated July 31, 1991, BWR Owners' Group to E.L. Jordan, regarding schedule of CRGR review of the proposed Generic Letter Supplement (Attacnment 2).

COMMENTS / RECOMMENDATIONS As a result of their review of this matter, the Committee recommended in favor of issuing the proposed supplement, subject to the minor revisions given below.

The Committee agreed specifically with the staff view that the proposed generic letter contains only relaxations and clarifications of the existing staff L

positions in GL 88-01; it does not impose new requirements or staff positions on affected licensees.

l' 5pecific changes recommended by the Committee in the proposed Supplement are as follows:

1.

At p.3 of ne draft Supplement, change the last sentence under subparagraon (1) to read as follows:

"Thus. RCS leakage measurements should be taken at least once per shift, not to exceed 12 hours1.388889e-4 days <br />0.00333 hours <br />1.984127e-5 weeks <br />4.566e-6 months <br />."

2.

At p.3 of the draft Supplement, rearrange / revise subparagraph to read as follows:

"The staff found that the radiation level associated with the RWCU system outboard of the containment isolation valves l

is very high; and this portion of piping is designed to be isolable and is generally classified as nonsafety piping.

Affected licensees requested that they be exempt from GL 88-01 with regard to the inspection of this piping.

However, the service-sensitive stainless steel RWCU system piping is subject to the most aggressive environment with regarc to IGSCC; therefore, until the actions associated with GL 89-10 on MOVs is completed by licensees, the staf f determined that an inspection of the subject piping on a sampling basis...should be performed...to ensure structural integrity of the piping."

3.

At pp.3-4 of the draft Supplement, revise the last sentence of subpara-graph (3) to read as follows:

"Therefore, the staff finds that manual leak rate measure-ments can be acceptable alternatives during the period (30 days) when the drain sump monitoring system is being restored, provided the licensee demonstrates their suit-ability with regard to accuracy and inspectability."

4.

At p.4 of the draf t Supplement, add a new phrase to begin subparagraph (5) as follows:

" Consistent with Code rwuirements and the licensee's written commitments, wl,an weld overlays...."

All changes made to the Supplement in response to these recommendations should be closely coordinated with the CRGR staff prior to issuance of the final Generic Letter Supplement.

l l

,y.

WP

-e,

--=.u m

BACKGROUND

- SUPPLEMENT TO GL 88-01

\\

• DISCUSSED STAFF POSITIONS IN 7 ITEMS

• 3 ITEMS: -MODIFIED STAFF POSITIONS

-H ARDSHIP CONSIDERATIONS

• 4 ITEMS: -CLARIFICATIONS & GUIDANCE

• IDENTIFIED DURING IMPLEMENTATION OF GL 88-01

• RESPONSES TO GL 88-01

• RESULTS OF IGSCC INSPECTION

• RESOLUTIONS OF HARDSHIP ISSUES

• BWR OPERATORS

• BWR OWNERS GROUP

• NO ADDITIONAL ACTIONS REQUIRED FROM LICENSEES

• NO CHANGE IN THE BASIS OF BACKFITTING oM q

- REGULATORY COMPLIANCE

/

ha 3 n j (

• BENEFITS:

0

-REDUCE BURDEN ON LICENSEES WITHOUT LOSS OF

! {t I

s-3AFETY BENEFITS

\\

n Q!

y

• TO FACILITATE IMPLEMENTATION OF GL 88-01' t

i BY GENERIC ACTION

\\s' i

1 A

J j

m m

m

PROPOSED SUPPLEMENT TO GL 88-01

• UNIDENTIFIED LEAKAGE MONITORING FREQUENCY 1

• INSPECTION OF RWCU SYSTEM PIPING

• OUTBOARD OF THE CONTAINMENT ISOLATION VALVES

• OUTAGE OF LEAKAGE MEASUREMENT INSTRUMENTS

• SAMPLE EXPANSION OF CATEGORY D WELDS

• ASSESSING SHRINKAGE EFFECTS

• RESULTING FROM WELD OVERLAY REPAIR & SI

• TECH SPEC AMENDMENT

-ISI STATEMENT

• TECH SPEC AMENDMENT

-LEAKAGE DETECTION I

I I

RELAXED STAFF POSITIONS t

• UNIDENTIFIED LEAKAGE MONITORING FREQUENCY GL 88-01 SUPPLEMENT l

EVERY 4 HRS EVERY 8 HRS OR EVERY SHIFT

• INSPECTION OF RWCU SYSTEM PIPING i

-OUTBOARD OF THE CONTAINMENT ISOLATION VALVES GL 88-01 SUPPLEMENT 100X 10X EACH REFUELING OUTAGE i

• OUTAGE OF LEAKAGE MEASUREMENT INSTRUMENTS G L 8 8 -O'1 SUPPLEMENT 24 HRS 24 HRS, WHEN MANUAL METHODS ALSO FAILED i

l i

CLARIFICATION AND GUIDANCE

• SAMPLE EXPANSION OF CATEGORY D -WELDS

> SAMPLE EXPANSION WILL APPLY TO CATEGORY D WELDS IF INSPECTED ON A SAMPLING BASIS

• ASSESSING SHRINKAGE EFFECTS RESU LTING FROM WELD OVERLAY REPAIR & STRESS IMPROVEMENT (SI)

> STRESS ANALYSIS / DESIGN BASIS IN PIPING SYSTEM

~ PIPE WHIP RESTRAINTS & SUPPORTS

• EFFECT OF INCREASED DEAD WEIGHT & STIFFNESS FROM OVERLAY i

• TECH SPEC AMENDMENT RELATING TO l

ISI & LEAKAGE DETECTION i

i

• ISI SECTION WILL STAY IN IMPROVED TS.

1

• INCORPORATION INTO AN ADMINISTRATIVE DOCUMENT IS NOT ACCEPTABLE.

I C^.

l I

i

(B) i AUGMENTED RWCU INSPECTION i

i o

NRC PROPOSED POSITION:

100% WELD INSPECTIONS REDUCED I

TO 10% INSPECTION i

)

o BWROG PROPOSED INSPECTION :

0% WELD INSPECTIONS WITH:

1)

GL 89-10 COMPLETION ON RWCU VALVES j

- OR -

l 2)

SECTIONS REPLACED WITH IGSCC RESISTANT MATERIALS

- OR -

i

]

3)

STRESS IMPROVED SECTIONS

~

I j

o CONCERN:

AUGMENTED INSPECTION REQUIREMENTS

~

i l

1 i

5

..m.-.-...

...._.__......._m._....m.~..---.m_...~a...-m.m..

m.__._-..___..m.m_

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

m.m..m.

r REACTOR l

NON HE Giht H ATIVE REGENERATIVE HE AT EXCHANGERS HE A T E XCHANGE HS g

y$

I t

TO RE ACTOR TO RE ACTOR 4

6 RECIRC RECIRC PUMP PUMP RSCCW gg H

W CRO HYDR AUllC L

l SYSTEM RETURN a i l

RCIC g

m m

10 S AMPL L DRAIN m

i t

ST A TION

+

j 1

II H/L AH l

ta g_

l--

i i

=

AH "S~ FEEDWATER LINE g

A

>=

((

d aps.

t

~~

N CONTAINMENT FROM g

FILTER DEMtN l

UNIT l

UNgT A g

a se "I

l TO-MAIN SAMPLE HOLO 4NG f,

=

l sTAnON l PUMP v

i e CONDENSER 2 j

L. _ _

._ _ _ _ _ _._ _ _J To unit

^

i

- gCi]

,.____.i H

~~

~~

FROM SAMPLE STA

- =

1_I L___-J 3

WASTE COLLECTOR _

ANO SURGE TANet ARE A LEAK ARE A F LOOR DETECTION DR At;$ L E AK DETECTION

.l i

i 1

FIGURE 4.1-1 REACTOR WATER CLEAN UP SYSTEM L

3 i.

i i

1 4

1 b

5

u sa i,g 3 a4 a x

-x.s.... ~..,.

r..

s a,\\

a s'

+'~,p,%.

s

/

L....

.m. w a

. ~

Ng8 a,

N s

[J J

g.....'

., s

.v

.c,

.'.s.v.

)

/

t.v -

p.y

........ A

. u.is s

.g

.x s

FOR INFORMATION ONLY I

CPL P.O. BOX 1551 R ALEIGH,N.C.

27602 A..,...........,

- - -.. ~ ~

- m g g pgggut g,g y

.AM 4 M.D.La 444taug h

/

1. e.

- east s e m. sums.no as M i 8 Mi i ( *i i

( 8'

)

r ka

~ "J"E 8 * "'" *"8

= =.... _

y

-,.-.-g..

w

. mal ra n s

- }<

ers s eivn i a r r : i i i....

.e ii s e i., i..

0 4

014 Ai le de WAii ae. Ii AtJ t #1 -.

3 3 t.g

.a t.* hi l( l h all W i t i n i 6.e A t h er J r.

Al'11 H IVI I 6 Ilv *

• I I ( ( 'MI I I Y I.,'le. IP W 8

^ ria. i r.-

i s.

u, i.... a.

.c i.4..

s e.... i.......i t i i 14.

f

(

NUTE[H t May 29,1991 ENEINEERS CHF-91-019

~

1 i

Mr. Frank J. Miraglia, Jr.

i Associate Director for Projects Office of Nuclear Reactor Regulation U.S. Nuclear Regulatory Commission Washington, D.C. 20555

Subject:

Effects of IGSCC Mitigation and Repair Techniques on Piping System Supports

Dear Mr. Miraglia:

It has recently come to our attention that the cumulative application of intergranular stress

{

corrosion cracking (IGSCC) mitigation and/or repair techniques may affect the piping support components of a treated / repaired system. The application of a significant number of weld overlay repairs at three domestic Boiling Water Reactors appears to have caused upward movement of the Reactor Recirculation system ring headers. The owner of these plants has asked us to inform you that cumulative weld overlay repair axial shrinkage on each of the vertical risers has changed the air gaps of pipe whip restraints and the normal set-points of variable spring hangers (a more detailed discussion is presented in attached NUTECH Document j

No. COE-45-098).

Subsequent analyses have found that these air gap and set-point changes have not affected the j

operability of these systems, though hardware adjustments were required to both whip restraints and spring cans in order to restore their full design margins. It is recommended that any plant i

that has applied or will be applying repairs or stress improvement techniques which can change the axial length of significant numbers of piping system components address these effects. Both the affected piping system supports and the sustained stress level at each unrepaired and/or stress-improved weldment should be evaluated.

If you need additional information, please contact Jim Brown in NUTECH's Westmont, Illinois office at (708) 789-2800 or me at (408) 629-9800.

Very truly yours, i

w Carl H. Froehlich, P.E.

Engineering Manager Fracture Mechanics &

IGSCC Resolution Services

/

()

CtSDk NUTECH Engineers. Inc.

145 Martnva4 Lane San Jose. Cakfomia 95119 (408) 629-9800 Fax (408) 0814186

.\\lt. Frank J. Miragita. Jr.

Office of Nuclear 5eactor Regulation M*y29'199I

~

CHF-91 019 cc:

All GL 88-01 Respondees W. H. Koo (USNRC)

EPRI (Palo Alto / Charlotte) 1 l

NUTEEH ENGHEERS

a i

EFFECTS OF IG8CC MITIGATION AND REPAIR TECHNIQUES ON PIPING SYSTEM BUPPORTS Prepared by:

C.

H. Froehlich

& J.

A.

Brown' Abstract It has recently been discovered that the cumulative application of intergranular stress corrosion cracking (IGscC) mitigation and/or repeir techniques may affeA the piping suppvrt ' components of a treated / repaired system.

The application of a significant number of weld overlay repairs at three domestic Boiling Watsr Reactors-appears to have caused upward movement of the Reactor Recirculation system ring headers.

Cumulative weld overlay repair axial shrinkage on each of the vertical risers has changed the air ge.ps of pipe whip restraints and the normal set-points of variable spring hangers.

subsequent analyses have found that these air gap and set-point changes have not affacted the operability of these systems.

However, it is recommended that any plant that has applied or will be applying repairs or stress improvement techniques which can change the axial length of significant numbers of piping system components address these effects.

Not only should the effects on the sustained stress level at each unrepaired and/or stress-improved weldsent be evaluated, but also the affected piping system supports.

Background

As discussed in Appendix A of United States Nuclear Regulatory Commission (USNRC) Document NUREG-0313, Revision 2 (Reference 1),

the magnitude of both residual and service (applied) stresses present during normal operation must be known or assumed in order to perform a flawed pipe evaluation.

To predict IGScc growth, a

" sustained stress" load combination including deadweight, internal pressure, and restraint-of-free end displacement thermal expansion loads must be determined.

Since at least 1984, the USNRC has required that this load combination also-include the effects of weld overlay repair axial shrinkage on a piping system receiving an overlay repair (s).

COE-45-098 N RS

.~

. _.. _ _ - - _ =

j,a!

l l

I Recent Industry Enerience I

Since late 1989, it has become apparent that axial length change l

effects must also be addressed relative to their impact on the i

support systers of repaired piping systems.

Quad Cities Nuclear i

Power station Units 1 & 2 and Dresden Nuclear Power Station Unit 2 l

each have weld overlay repairs on at least one weld of almost every j

vertical riser on the pump discharge-side of their Reactor Recirculation systems.

The overlays were applied during 'three j

dif ferent refueling outages at each of these units since late 1983.

During a piping system support walkdown performed at the Quad i

Cities Unit 1 198C outage, it *wns oLaervad that the normal set-points on variable spring hangers and the air gaps between the ring j

header and its pipe whip restraints were just beyond their maximum j

allowable tolerance. This same out-of-tolerance condition was also j

observed at the Quad Cities Unit 2 and Dresden Unit 21990 outages.

Although a root cause analysis has not yet been completed, it j

appears that these out-of-tolerance conditions were caused by i

upward movement of the ring header due to weld overlay axial

{

shrinkage on the vertical risers.

Piping system analyses performed for all three plants show that the affected piping systems met j

FSAR/UFSAR stress criteria during their operation even with these j

out-of-tolerance supports.

However, to bring the subject supports back to their original "as-analyzed" condition, pipe support j

adjustments have been/will be made at each unit.

i t

j Generie Issue i

The nuclear power industry has been aware of the effects of weld overlay repair axial shrinkage on flawed pipe evaluations since 1984 and stress improvement effectiveness evaluations for the past j

few years.

However, due to the application of a large number of j

weld overlay repairs at Quad Cities Units 1 & 2 and Dresden Unit 2, i

performed on a,pi,eceneal basis over the past six to seven years, cumulative shrinkage effects on the supports of repaired piping j

systems were not recognized until recently.

It is, therefore, 1

recommended that any plant that has applied or will be applying significant numbers of weld overlay repairs address their effect i

not only in flawed pipe and stress improvement offactiveness j

evaluations, but also on the supports of repaired piping systems, i

i i

J 1

1 COE-45-098

-3 NUTEEH ENGNEERS

. ~ -

n#

i d

1 i

i d

i 4

4 1

i i

TYPICAL WELD OVERLAY REPAIR DATA SHEET e

l gm

.LL.W _.

%iest asa i

l f

R j

l.

f h

ih i

i m

at as to W

C D

4t M

d 8E' EWE #

m

)

i 1

=_

1 i

3 i

l ammeesnumeen===

• *~

ahumage e Samus t

oansuung emansummee

~~

TT GREur

)

N i

E*

M-een 1

i amme..

. m.

a m.m..nmm a.an.mmmenemm eem j

niiimmune summus

.m mm..co me.=

Du uuume Rumumb m

l m

& C.Gh.uun.g._m m.

v.--.am.eu.sesman =_,3 F

.msnsar 3

t

... w.w.

=

813 5mm -

j

. g.,-...m.-.-

y

%,,,g;;;m==

8

...._.m, a-mm.massapursessnesus arves

~

l..

• • =="

• =

A y

w a

mmm.

sWW W

06 1

i, 1

=

1

• • =**'

t.

b N

l

.l l pensam imene Tamsen m

I h

N m e.

4 I

t 4

(

COE-45-098 5-i NUTEIH i

ENEINEERS 4

,,,._..,_w,,._,._.,__.r.-_

,y

.,,,,,m,.,.,,.

, _ -,.