ML20137D072

Forwards Copy of Final ASP Analysis of Operational Condition Discovered at Plant,Unit 2 on 950125,reported in LER 336/95-002

ML20137D072
Person / Time
Site:	Millstone
Issue date:	03/21/1997
From:	Mcdonald D NRC (Affiliation Not Assigned)
To:	Carns N, Laudenat R NORTHEAST NUCLEAR ENERGY CO.
References
NUDOCS 9703250251
Download: ML20137D072 (15)

Text

. __

g es a%g

[ ,y 1 UNITED STATES

, s* g NUCLEAR REGULATORY COMMISSION

'2 WASHINGTON, D.C. 2006H001 k .. . . . ,o/ March 21, 1997 l

Mr. Neil S. Carns Senior Vice President and Chief Nuclear Officer i Northeast Nuclear Energy Company j c/o Mr. Richard T. Laudenat i Director - Regulatory Affairs P.O. Box 128 Waterford, CT 06385

SUBJECT:

FINAL ACCIDENT SEQUENCE PRECURSOR ANALYSIS OF OPERATIONAL CONDITION  !

AT MILLSTONE UNIT 2 l l

Dear Mr. Carns:

Enclosed for your information is a copy of the final Accident Sequence Precursor (ASP) analysis of the operational condition discovered at Millstone

, Unit 2 on January 25, 1995, which was reported in Licensee Event Report (LER)

No. 336/95-002. This final analysis was prepared by our contractor at the Oak Ridge National Laboratory (ORNL), based on review and evaluation of comments on the preliminary analysis received from the NRC staff and from our independent contractor, Sandia National Laboratories (SNL). The preliminary analysis of this event was transmitted to you for review and comment on a voluntary basis via my letter dated September 20, 1996. Initially you indicated your intention to provide comments on the analysis. However, you 1ater informed us that, because of the current situation at the Millstone  !

plant and the extent to which your resources have been committed to the l

. resolution of higher priority licensing issues, you would not be able to '

respond to our request in a timely manner. As a result, the final analysis l was prepared without your comments. Resolution of the comments we did receive i resulted in revisions to the analysis and associated documentation. The l results of the final analysis indicate that this event is a precursor for 1995.

I NRC FILE CENTER COPY -

>/d J

1 9703250251 9703'21^ i PDR ADOCK 05000336 P PDR l l

\ \

\

Neil S. Carns March 21, 1997 Please contact me'at (301) 415-1408 if you have any questions regarding the enclosure. We recognize and appreciate the effort expended by you and your staff in' reviewing the preliminary analysis.

Sincerely, Original signed by:

Daniel G. Mcdonald, Senior Project Manager Special Projects Office - Licensing Office of Nuclear Reactor Regulation Docket No. 50-336

Enclosure:

ASP Summary - LER No. 336/95-002 cc w/ encl: See next page DISTRIBUTION:

' Docket File PUBLIC SP0 Plant WTravers PMcKee LBerry DMcDonald OGC ACRS JDurr, RI p':UMENT thi.G:\

.iv. . copy orNAME: MCDONALD nocom.ni,indic... i. \ ASP.

c FIN c9,y withoui .et. chm ..

in. 6..: ur. t - copy wein .et. chm.ntioncio.or. u - No copy

[FICE lSPO-L:PM l6 SP0-L:L/bQJ SP0H/:8DI l l l MIE DMcDonald (TlJh .LBerry 7 PMpKee

[TE

~

03/2V97 03/AO/97 03hk/97 OFFICIAL RECORD COPY

Neil S. Carns Please contact me at (301) 415-1408 if you have any questions regarding the enclosure. We recognize and appreciate the effort expended by you and your staff in reviewing the preliminary analysis.

Sincerely, c-s h 7 st Daniel G. Mcdonald, Senior Project Manager Special Projects Office - Licensing Office of Nuclear Reactor Regulation Docket No. 50-336

Enclosure:

ASP Summary - LER No. 336/95-002 cc w/ encl: See next page

I i

l i

Appendia B LER No. 336/95-002 l

. i B.6 LER No. 336/95-002 Event Desoiption: Contamment Sump Isolation Valves Potentially Unavailable due to Pressure Lockmg Date of Event: January 25,1995 l 1

i Plant: Millstone 2 i

l B.6.1 Event Summary Northeast Utilities determmed that the containment sump isolation valves at Millstone 2 were subject to pressure locking, which could preclude their operating following a loss of coolant accident (LOCA). This analysis assumes the pressure locking problem would impact plant response to a large- and medium break LOCA, and estunates an increase in the co e damage probability (CDP) over a one-year period of 3.1 x 10-5, over a nominal value for the sae period of 4.6 x 10-5 Uncertainty in the frequency oflarge- and medium-break LOCAs (neither of which have been observed) and in the impact of the pressure locking problem l contribute to a substantial uncertainty in this estimate.

l B.6.2 Event Description On January 25,1995, Northeast Utilities determmed that the containment sump isolation valves (valves 2-CS-16.l A and -16.lB) at Millstone 2 were subject to pressure locking, which could preclude their operating for sump recirculation following a LOCA (see Fig. B.6-1). Pressure locking is a phenomenon where water trapped in the bonnet cavity and in tne space between the two disks of a parallel-disk gate valve is pressurized above the pressure assumed when sizing the valve's motor operator. This prevents the valve operator from

opening the valve when required. Water can enter a valve bonnet during normal valve cycling or when a i differential pressure moves a disk away from its seat, creating a path to either increase fluid pressure or fill the bonnet with high-pressure fluid. A subsequent increase in the temperature of the fluid in the valve bonnet

will cause an inen

ase in bonnet cavity pressure due to thermal expansion of the fluid.

Valves 2-CS-16.l A and -16.lB had initially been reviewed for pressure locking and thermal binding issues in D ber 1989. That review, in response to an Institute of Nuclear Power Operations (INPO) Significant Operatmg Events Report (SOER), was conducted by Stone and Webster and concluded that these valves were not susceptible to pressure lockmg or thermal binding.

In 1994, following an inspection of the Millstone 1 motor operated valve (MOV) program by the NRC and in order to address a planned NRC Generic Letter (GL) on the pressure locking / thermal binding issue

! (GL 95-07), Raytheon Corporation was contracted by the licensee to perform a second analysis of all safety-related "GL 8910" MOVs. Raytheon's final report, issued in October 1994, concluded that valves 2-CS-16. l A and -16.lB were susceptible to pressure locking.

I B.6-1 NUREG/CR-4674, Vol. 23 1

Enclosure

j .  !

l LER No. 336/95 002 Appendix B !

t 4

A subsequent analysis by the valve vendor, Anchor Darhng, detenmned that the maxunum pressure, lock -

that valves 2-CS-16.lA and -16.lB could overcome and still open was approxnnately 150 psi. Based on  ;

i informaten provided in NUREG 1275, Vol. 9 (Ref. 2), an mcrease in bonnet temperature of 5'F, if the [

bonnet were waar solid, would cause this pressure.' Following a LOCA, water in the contamment sump  ;

1 could reach 289'F. Durmg the .yy-.... aly 44 min before the initiation of sump recirculation, the hot i I water in the enatamawat sump would easily heat the valve bonnet by the 5'F required to pressure-lock  ;

the valve. 1 l The bcensee noted that the two containment sump isolation valves, as well as two dw i.e. check l i valves (2-CS-15A and -15B) are subject to periodic surveillance testing. As a consequence of these tests,  ;

l water tends to accumulate in the piping between the isolation valves (-16.1A/B) and the sump. His water l 1 may serve as an insulator, preventing hot sump water fnxn reachmg the isolation valves and mmunizmg 4 the impact of the pressure locking condition. At the time Ih Event Report (LER) No. 336/95 002 was written, check valve 2-CS-15A was being overhauled because the train A sump piping was found to 4

be full of water. Evidence of water was also found in the train B piping. Further analysis by Raytheon concluded that if the sump piping was full of water, the temperature increase would only occur along the first foot of the fdled pipe on the containment side.

i j The report of a special NRC inspection that was performed at the time of the event (Ref. 3) provides j additional information concermng the testing ofisolation valves 2-CS 16.l A and -16.lB and check valves 2-CS 15A and -15B, plus the arrangement and condition of the valves. Valves 2-CS-16.lA and -16.lB l were cycled monthly to verify operability and measure valve openmg time. When cycled, water that j collected in the piping betwe.. the isolation and check valve (2 - 3 ft of 24 in. pipe) would flow into the i longer containment piping, equalmng the water level upstream and downstream of the valves. Prior to l

} each test, a vent valve between the isolation and associated check valve was opened to confirm that the j'

check valve was not leaking excessively. If it was, the Refueling Water Storage Tank (RWST) was isolated from the Emergency Core Cooling System (ECCS) header during the valve test to nummize the amount of water that flowed into the contamment.

Every three months, valves 2-CS-16.1A and -16.lB, and check valves 2-CS-15A and -15B were also  !

tested to meet the mservice testing reqmrements of ASME Section XI. De piping between the two valves was filled with borated water by hose-jumpering around the check valve using two vent  ;

connections. A hose and rotameter was then maw from the Prunary Makeup Water (PMW) system '

to the vent connection between the two valves and the piping was pressurized to the normal operating pressure of the PMW system (125 - 150 psi). Flow through the rotameter indicated the partial stroking of the chack val *e. This test, by filhng and pressunzmg the fluid volume between the valves, would tend to j move the downstream (pressunzed) isolation valve disk away from its seat, and fill and pressurize the valve bonnet The isolation valves were also stroked, but the testing sequence (before or after the check i valve te.st) was not paart if the isolation valves were stroked before the check valves were tested, the I piping between the two valves would have been left full of water followmg the check valve test.

i "Preliaunary c  ;

• T-; dets from flexible wedge gate valve tests at the Idaho Nations! I:agmeeting Laboratory indicates that bonnet c 1 = also occurs if small quantities of air are trapped in the bannet, although at higher temperatures than for a water-solid condition (NRC memorandum from M.E. Mayfield to R.H. Wensman June 25.1996).

NUREG/CR-4674, Vol. 23 B.6 2 j i

1 .

1 l l

Appendix B LER No. 336/95-002 l j

j A walkdown in conjunction with the ia=amiaa =wlumM that packing leakage had occurred on valves 2-l CS-16.lA and -16.lB since the last packing change in 1992. This was indicated by boric acid crystals i and rust in the packing gland areas of both valves. This leakage indicated that the valve bonnets had been l filled with water at least some of the 6me since 1992. A closed 1984 work order for valve 2-CS 16.lB to i be cleaned of boric acid crystals and for the valve packing to be tightened provides evidence of earlier

water fd!od valve bonnets  !

I In order to preclude pressure lockmg of valves 2-CS 16.lA and -16.lB, licensee personnel drilled a 1/8 i in. diameter hole through the containment-side disk center line on these valves. These holes will prevent t- the volume between the two disks from pressurizing.

B.6.3 Additional Event-Related Information t

! Following a LOCA, the contmnment sump collects water from the break in the reactor coolant system, l (RCS), the s.fety injection systems, and the containment spray system. AAer water in the refueling water l storage tank (RWST) is depleted, the contamment sump provides a source of recirculation water for 1 ce=W decay heat removal and contamment spray. Upon receipt of a sump recirculation actuation 4-signal (SRAS), at 9.5% RWST level, the high-pressure and low-pressure safety injection (HPSI and LPSI) pumps and the containment spray pump suctions are automatically transferred to the contamment l

l sump by the opemng of valves 2-CS-16.lA and -16.lB. The SRAS also trips the LPSI pumps to l maximize the net positive suction head to the HPSI pumps and the contamment spray pumps. HPSI pump l flow provides decay heat removal following sump switchover.

I

. The containment " sump" at Millstone 2 is actually the floor of the contamment, and not a separate pit

below the contamment floor. Two 24 in. pipes, which protrude approximately 11 in. above the floor, drop vertically about 5 A, and then run almost horizontally, with a slight downward slope, about 20 ft to i valves 2-CS-16.lA and -16.lB. Check valves CS-15A and -15B are loca'ed 2 - 3 A downstream of the 1 isolation valves (Fig. B.6-2).

(

l B.6.4 Modeling Assumptions i-l This analysis assumes that sump isolation valves 2-CS 16.1A and -16.lB would be unavailable due to l pressure lockmg followmg a large-break LOCA, and possibly followmg a medium-break LOCA, for those priods during wiuch the valve bonnets were filled with water prior to drdhng holes in the

contamment-side disks. Both of these LOCAs rapidly deplete the RWST and provide little time for thermal equilibration and valve bonnet depressurization which woidd permit initially pressure-locked isolation valves to operate However, switchover to sump recirculation following a small-break LOCA t occurs aAer about 6 h; this time is assumed to be d~p* to allow the valve bonnets to depressurize to the point that the valves will operate correctly.

l

The valve bonnets for valves 2-CS-16.lA and 16.lB were assumed to be filled with water following the

quarterly tests of 2-CS-15A and -15B, during which the PMW system was used to fill and pressurize the j pipe section between each isolation and check valve. Reference 3 notes that the isolation valve vendor, Anchor / Darling had concluded that a closing thrust of 91,000 pounds would be required to seal the l

B.6-3 NUREG/CR-4674, Vol. 23 l.

i

d

, LER No.336/95 002 Aa=== dix B i

) isolation valve disk against a pressure of 45 psig (a conservative estunate of the pressure the RWST bead I would impose on the valve disk). With a valve actuator maxunum thrust of 45,000 pounds

(.yye...mely one-half of the required closing thrust for 45 psig) and an actuator torque switch trip point 1

one half again of the required value, it is reasonable to assume that the isolation valve disk would move

[ away from its seat during the 125 - 150 psi PMW system pressunzaton i-:=4 during the tests on check

! valves 2-CS-15A and -15B. This disk movement would allow water to enter the space between the

isolation valve disks and, with leaking packing, the isolation valve bonnet. Once a bonnet was filled with i water, mmor leakage past the normally closed check valve would be sufficient to mamtmin the bonnet full ofwater, j Filled valve bonnets were assumed to remam fdled until an isolation valve was cycled. When the valve was cycled, water in the bonnet and between the check valve and the isolation valve would flow into the

- sump piping in the contamment, v= Wing the water levels on both sides of the isolation valve.

Evaporation

• would gradually reduce the volume of water in the sump piping. ' mis would explain why

- only evidence of water (boric acid crystals) was found in the B train piping, and not a large volume of water (the large quantity of water found in the A train piping was the result of excessive check valve i leakage). Because check valves 2-CS 15A and -15B were tested quarterly and valves 2-CS 16.lA and -

16.lB were cycled monthly, the valve bonnets would normally (without the excessive check valve leakage) be expected to be filled with water one-third of the time if the isolation valves were always

cycled before the check valves were tested. Although no requirement was placed on the sequence of

testing, the isolation valves (2-CS-16.lA and -16.lB) were typically cycled after the check valve tests (2-

! CS-15A and -15B) most of the time? Assuming this occurred in 75% of the tests reduces the estimated i fraction of time during which the sump isolation valves were subject to pressure locking to one-twelfth.

(If the sump valves had always been tested after the check valves, the potential for pressure locking would j be significantly reduced, or perhaps eliminated. Pressure locking of the isolation valves would only be possible under these conditions if a substantial leak path existed past both the check valve disk and the isolation valve bonnet.)

i l The impact of partially-filled sump piping (evidence of past water'in the B train sump piping was

! described in the LER), caused by monthly isolation valve cycling, on reducing the potential for pressure

, locking was considered to be minor and was not addressed. The piping arrangement at Millstone 2 (see l Additional Event-Related Information) prevents water from entering the sump piping until the l contamment water level reaches 11 in. At this point, hot water will drop into the vertical sump piping l segment and rapidly fill the piping. The turbulence that results is -Wed to mix the hot containment i water with the water in the partially-filled sump pipes, exposing the sump isolation valves to tenores i

i 1 "Fogowns the discovery of the possenal for pressure loclung and the drilling of bones in the downstream disks, the hcensee

, began to bypass the check valve and fill the sump piping to provide an insulating barrier agamst the i.ot sump water that would exist l foDowing a LOCA. 'Ihis process is perfonned monthly to ensure that evaporation to the an=h= ment will not significantly reduce the volume of weser in the sump piping [ personal commuratma D. Beaulieu (NRC) and J. Mannck (SAIC), July 26,1996j.

• Personal comununcanon, M. Buckley (NRC) and J. Minarick (SAIC), February 21,1997. j i

i J

NUREG/CR-4674, Vol. 23 B.6-4 1 I

]

,a - - -

1. .

i Appendia B LER No. 336/95 002 i well above ambient.' However, the excessive amount of water that apparently existed in the train A sump piping due to leakage past check valve 2-CS-15A is believed to have provided an insulating barrier for MOV 2-CS-16.lA for part of the one-year period prior to discovery of the pressure locking problem (one year is the longest unavailability paiod used in an ASP analysis). Durmg the time that water in the train  :

A sump piping insulated 2 CS-16.lA, the potential impact of sump valve pressure lockmg was muumal I (an increase in CDP less than the ASP screenmg value of I x 104 is calculated).

Reference 3 notes that 2-CS 15A leaked excessively during a test on December 8,1993, when water completely filled the train A sump piping, as eviden-i by an acrease in contamment sump level. No further incidante of gross check valve leakage were identified. In early 1995, when the sump valves and piping were i=5-:-%i the water level in the train A pipe was found to be at the top of the 2-CS-16.lA disk (because of the slight upward slope of the sump lines towards contamment, the water only partially filled the horizontal pipe). The reduction in water level since December 8,1993, is assumed to have been ,

l caused by evaporation. This reduction is reasonably consistent with a licensee estimate of the evaporation rate in the sump piping, performed to support a strategy of maintakung the sump lines full of water to prevent pressure locking (described in footnote b), and submitted as a part of Reference 4 (calculation GL89-10-1243 M2, Rev.1).

Following the December 8,1993, check valve leakage, when the sump piping was completely filled, MOV 2-CS-16.lA would be protected from heatup during a LOCA. At some time thereafter, as water in the . sump piping evaporated, the MOV would become susceptible to pressure lockmg. Licensee memorandum NE-95-SAB-297, dated July 26,1995, and also Wluded in Reference 4, concluded that the water level in the sump piping should be maintamed above -24 n (2 ft below the top of the vertical sump piping shown in Figure B.6-2) to elimmate the potential formation of thermal diffusive currents, which could cause sump valve heatup and pressure locking following a LOCA. The licensee estimated that water in a completely full sump pipe would evaporate to -24 ft in about 106 d. Applying this estimate in this analysis,2-CS-16.lA, and therefore both 2-CS-16.lA and -16.lB, were assumed to be potentially vulnerable to pressure locking for [1 - (106/365)], or 0.71 of the one-year period prior to discovery.

Combining this value with the fraction of time the valves were considered subject to pressure locking i based on the testing regunen at Millston: 2,0.083, results in an overall estimate of 0.059.

The Accident Sequence Procursor (ASP) Program typically considers the potential for core damage following four postulated initiating events in pressurized water reactors. transient, loss of offsite power l (LOOP), small-break LOCA, and steam generator tube rupture (SGTR). Supercomponent-based linked l fault tree models are available for each of these postulated initiating events. The two initiating events that are of concern in this analysis (i.e., large- and medium-break LOCAs) are not currently modeled.

However, for both of these initiating events, unavailability of sump recirculation is assumed to result in core damage in all probabilistic risk ==se==narits.

The significance of this event was estunated by first considering the increase in core damage probability (CDP) over the unavailability penod Since a nonrecoverable failure of valves 2-CS 16.lA and -16.2B

! *A simple taixing calculatian, without consuleration of the sump piping as a beat sink, supports an assumption that the sump f

piping wimid beve to be initiaDy ahmost full to preclude a substantial temperature increase.

i.

B.6-5 NUREG/CR-4674, Vol. 23

\

i d

)

l

LER No. 336/95-002 Appendis B will fail both high- and low-pressure recirculation, and recirculation is required following a medium _ or I large-break LOCA, the significance of the event can be estunated directly from the change in the

!- - probability of recirculation failure and the probability of'a medium- and large-break LOCA in the

unavailability penod.

i I Dunng the time period that valves 2-CS-16.lA and -16.lB were assumed to be subject to pressure- '

locking, this pressure-lockmg condition was assumed to prevent the valves from opening for sump I recirculation following a large-break LOCA (LBLOCA). Sump recirculation would be initiated by an l i

SRAS signal about 44 min aAer the break, allowing little time for bonnet depressurization. The RWST

! would be empty about 4 min later, allowmg little time for corrective action

{ Followmg a medium-break LOCA (MBLOCA), the condition of the valves is considered mdetermmate l

! Additional time (well over I h) exists before the RWST is depleted, and this time may allow for l l depressurization of at least one of the valve bonnets to the point that the valve will operate. For a

]

1 medium-break LOCA, a sump isolation valve failure probability of 0.5 was assumed, given that pressure-j locking condition existed.

I ne frequency of a large-break LOCAs is estimated to be 2.7 x 10d/yr, while that of a medium-break i j LOCA is estimated to be 5.0 x 10d/yr. These values are based on a survey of large- and medium-break

! LOCA freque,ncies performed in support of the analysis of Turkey Point LER 250/94-005 in the 1994 )

precursor report (see Appendix H to Ref. 5 for additionalinformation). 4 The increase in core damage probability due to LBLOCAs and MBLOCAs based on these initiating event frequencies and the assumed probability of pressure locking are given below. The probability of core )

damage due to a LBLOCA is d

2.7x10 fprob of a LBLOCAL x 0.059 prob that pressure- } x lover r.1-yr period a Uock condition existsi 1.0 jprob ofsump recirc failureL - 2.6x10" /nommal failure prob' Idue to press-locked valves < for two sump valves.

= 1.6x10-5 ~ in core damage rob due to a LBLOCA }. J De probability of core damage due to a MBLOCA is 5.0x10d fprob of a MBLOCAL x 0.059 prob that pressure- ( x Lover a 1-yr period , tiock condition exists.

0.5 fprob ofsump recire failure' -

2.6x10d pommal failure probl Idue to press-locked valves 1 tror two sump valves ,

= 1.5x10-5 reasein core damage '

due to a MBLOCA .

NUREG/CR.4674, Vol. 23 B.6-6

Appendix B LER No. 336/95 002 4  !

1 .  !

B.6.5 Analysis Results  !

Combining the estimates for large- and medium-break LOCAs results in an wi==W increase in the CDP i because of the sump isolation valve pressure locking over a 1 year penod of 3.1 x 10-5 The dominant  ;

core damage sequences for these event involves

]

a a postulated large-break or medium-break LOCA, and i )

[ . the failure of high-pressure recirculation. I

} De daminaar sequence for a large-break LOCA is highlighted on the event tree in Figure B.6-3. A  :

j sindar sequence exists for the medium-break LOCA.

l j- A greater than usual uncertamty is associated with this estimate. His uncertamty is' donunated by the ,

j uncertainties in the frequencies oflarge- and medium break LOCAs (neither of which have occurred) and l

l by uncertainties in the assumptions regarding the inoperability of the sump isolation valves because of the pressure-lockmg problem.

l L

j The nominal CDP over a one-year period estimated using the ASP Integrated Reliability and Risk Analysis

. System (IRRAS) models for Millstone 2 is 4.6 x 10-5. The increase in the CDP because of the unavailable sump isolation valves (3.1 x 10~5) was added to the nommal CDP for a one-year period (4.6 x 10-5) to estimate a CCDP of 7.7 x 10 for the one-year period prior to discovery of the pressure locking problem. For earlier

! one-year periods, when 2-CS 15A was not leaking excessively, the CCDP for the event would not be affected by the flooded train A sump piping. For such periods, a CCDP of 8.9 x 10~5would be estimated i

j Section 6.3.3.1 of the Millstone 2 FSAR states that during normal operation the contamment sump i reci culation lines between the sump isolation valves and the HPSI pumps will be filled with stagnant water, wh'le portions between the sump inlets and the sump valves will not be filled with water. As described in i the Event Description, piping downstream of the isolation valves was not maintamed full of water following

valve testing, and water was allowed to accumulate in the sump piping upstream of the isolation valves. If 4

the 1=maar had taken measures to adhere to the FSAR statement, the isolation valves would most likely have j been subject to pressure locking at all times. This would have increased the CCDP associated with the event to 5.7 x 10d.

i B.6.6 References i

1. LER 336/95-002, Rev.1, "Contamment Sump Isolation Valves - Potatially Subject to Pressure .

IMiai" July 7,1995.

\ '

l 2. Operating Experience Feeback Report - Pressure Locking and Thermal Binding ofGate Vahes, i NUREG 1275,Vol. 9, March 1993.

1 i

j 3. " Millstone 2 Motor-Operated Valve Inspection," NRC Special Inspection Report 50-336/95-08, U.S.

i Nuclear Regulatory Commission, March 22,1995.

i i

i.

i B.6-7 NUREG/CR.f 674, Vol. 23

.I 1

h

1

=

9 LER No. 336/95402 Appendix B j 4 4. " Generic Im 95-07, ' Pressure Locking and Thermal Binding of Safety-Related Power-Operated Gate Valves,' Response to Request for Additional Information," letter from T. C. Feigenbaum, ,

Northeast Utility Services Company, to U. S. Nuclear Regulatory Commission, August 7,1996.

i

5. Precursors to Potential Sewre Core Damage Accident Sequences

1994, A Status Report,  !

! NUREG/CR- 4674, Vol. 21, December 1995, l

! I t

i l

1 i

NUREG/CR-4674, Vol. 23 B.6-8

. . . .. . .. .- . - _ . . .=_ _ . - -. . . - . . . .

o 4

4 Appendix B LER No. 336/95-002 4

'l J

I i

RWST i

Containment Pressure M

>1/\/1I' O SUMP CS-16.1A CS-13.1 A CS-13.1B

CS-16.1 i CS-15A CS-14A N  ? iA CS-15B CS-148 l ECCS Pumps CVCS SFCS

)

l 5

Fig. B.6-1. Cmain=nt Sump /RWST Configuration for Millstone 2 (Source: Mllstone 2IndividualPlant Examination).

B.6-9 NUREG/CR-4674, Vol. 23

1

. LER No. 336/95-002 Appendix B 1

i a

4 i

l l Standpipe (with Grating)

! UI 111n I

- 1 4 ft - 2 to 3 ft l Containment Floor q g I- 16 ft -

l 24 in SS piping MOV Check (slight rise in piping towards sump) 16.1 A/B Valve i

15.1 A/B l

i i

. I i

1 1

a

^

l i

Fig. B.6-2. Millstone Unit 2 Contanment Sump Piping (Source: " Millstone 2 Motor-Operated Valve Inspection," NRC Special Inspection Report 50-336/95-08, March 22,1995).

NUREG/CR-4674, Vol. 23 B.6-10 1

su. L,i 4 h /._4 mn4 1 =4. M ,ima. 4- a,_.h.= -4L_.

_. '5_._ ,gt,. . ,.- -, ha+ -.1 .g. m.,%,a 4 s4 9 $

i i' =

w a

h 9

!l 5:

E g unos ^ tow won a

f saem LOCA

"[ "J88 8Py ItOECTION as messuae RECIRC SEQUENCE NO.

END STATE a

.o 1 OK R

w $

R 2 CD r

tT1 PC E CD e

m h 4 CD g

9 LER 336/95 - 002 E

=

9 7 k u

Northeast Nuclear Energy Company Millstone Nuclear Power Station

..' Unit 2 cc:

Lillian M. Cuoco, Esquire Mr. F. C. Rothen Senior Nuclear Counsel Vice President - Nuclear Work Services Northeast Utilities Service Company Northeast Nuclear Energy Company P. O. Box 270 P. O. Box 128 '

Hartford, CT 06141-0270 Waterford, CT 06385 1 Mr. John Buckingham Ernest C. Hadley, Esquire '

Department of Public Utility Control 1040 B Main Street  !

Electric Unit P.O. Box 549 10 Liberty Square West Wareham, MA 02576 New Britain, CT 06051 Mr. D. M. Goebel {

Mr. Kevin T. A. McCarthy, Director Vice President - Nuclear Oversight Monitoring and Radiation Division Northeast Nuclear Energy Company Department of Environmental Protection P. O. Box 128 79 Elm Street Waterford, CT 06385 Hartford, CT 06106-5127 l Mr. M. L. Bowling, Jr. I Regional Administrator, Region I Millstone Unit No. 2 Nuclear U.S. Nuclear Regulatory Commission Recovery Officer 475 Allendale Road Northeast Nuclear Energy Company King of Prussia, PA 19406 P. O. Box 128 Waterford, CT 06385 First Selectmen Town of Waterford Mr. J. K. Thayer Hall of Records Recovery Officer - Nuclear Engineering i 200 Boston Post Road and Support Waterford, CT 06385 Northeast Nuclear Energy Company P. O. Box 128 Mr. Wayne D. Lanning Waterford, CT 06385 Deputy Director of Inspections Special Projects Office Deborah Katz, President 475 Allendale Road Citizens Awareness Network King of Prussia, PA 19406-1415 P. O. Box 83 Shelburne Falls, MA 03170 Charles Brinkman, Manager Washington Nuclear Operations Mr. B. D. Kenyon ABB Combustion Engineering President and Chief 12300 Twinbrook Pkwy, Suite 330 Executive Officer Rockville, MD 20852 Northeast Nuclear Energy Company P. O. Box 128 Senior Resident Inspector Waterford, Connecticut 06385 Millstone Nuclear Power Station c/o U.S. Nuclear Power Station Mr. Allan Johanson, Assistant Director P.O. Box 513 Office of Policy and Management Niantic, CT 06357 Policy Development and Planning Division Citizens Regulatory Commission 450 Capitol Avenue - MS# 52ERN ATTN: Ms. Susan Perry Luxton P. O. Box 341441 180 Great Neck Road Hartford, CT 06134-1441 Waterford, CT 06385

.