ML17139C892

Discusses Potential Generic Issue Re UHS (Spray Pond) During Cold Weather Operation.Resident Inspectors at Facility Following Util Safety Evaluation & Corrective Actions

ML17139C892
Person / Time
Site:	Susquehanna
Issue date:	03/06/1984
From:	Starostecki R NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION I)
To:	Eisenhut D, Jordan E NRC OFFICE OF INSPECTION & ENFORCEMENT (IE), Office of Nuclear Reactor Regulation
Shared Package
ML17139C891	List: ML17139C890 ML17139C892
References
NUDOCS 8502110651
Download: ML17139C892 (25)

Text

~p,R RECT

~

4p

+>>*<<+

Docket Hos VIEYiORAHOUN FROM:

UjillTEDSTATES NUCLEAR REGULATORY COMMISSION REGION I 63'I PARK AVENUE KING OF PRUSSIA, PENNSYLVANIA19406 MAR 6 1984 FOR:

50-387/50-388 1

7 E. Jordan, Director, Division of. Emergency Preparedness and..~'"-

Engineering'esponse D.'isenhui,- Di,rector, Division of Licensing Y

R. Starostecki, Director, Division of.Project and Resident

Programs, Region I SU""JECT:

POTENTIALLY GENERIC,ISSUE ULTINATE HEAT SINK (SPRAY POND)

COLD MEATHER OPERATIOH P

'" Region I Da ly Reports'n 1/12 and '1/l8/84 described an event at Susquehanna Steam Electric S ation (SSES) where, following ESF flow balancing using ulti-mate heat sink sprays,"43 spray nozzle arms came off and fell into the pond.

Pbou 75'. of ihe nozzle arns on both loops were later determined to have in-sufficierit thread engagement, ano some nozzles were found obstructed by ice.

This event was caused by a leaking spray network supply valve which,.though shut, filled the spray network faster

.han drain. pump could empty it.

A heavy 1

e accumulation developed, hanging down from ihe nozzle end of each spray

arm, and the 43 arms came off because of insufficient hread engagement.

The licensee=reirieved the fallen nozzle arms, cut the ihreads from all nozzle

arms, and welded the arms to the risers.

A sl etch of the riser-noz=le arm configuration 'is enclosed.

In addition to welding the previously-threaded connections, each nozzle arm was sloped slightly, back towards the riser, to facl liiate drainage.

Section 10 o'f HRC Inspection Report 50-387/83-29; 50-388/83-32 describes

his event.

r I

'wmA OHQ OIO OO i, lrIO

')IA O

~K

) lniM OQ MQ WCC

, OlOC i IAA

', mlLL Susouehanna FSAR Figure 9.2-24 depicts tha spray pond layout and pipino net-work.

The eioh acre pond has a capacity of 25 million gallons and is shared by Units 1

arid 2.

Principal post-accident ultimate heat sink loads include the RHR hea. exchanoers, the standby diesel generator lube oil and jacket wa er

coolers, arid ESF eouipment such as ECCS room coolers and Control S ructure chillers.

A 36-inch re urn header from RHR and ES'8 systems supplies each spray loop network; branched-off this header is a bypass line for direct discharoe in o the pond.

A small pump allows for drain-down of exposed (above-pond) piping iollowing spraying.

Cold weather operations are 'described in FSAR Sub-section 9.Z.7.2.3.

Region I has two principal concerns with the modifications to'usquehanna s

'1 ima-e heat sink: (1) the e fec of nozzle arm inclination on pond thermal performance:

and, (2) the potent',al for ice formation and nozzle blockaoe.

The subject of porid therr<<a I'er<ormance was recently addressed (prior o this everit) in the Deceriber Z1, 1983 letter PLA-2007 fro-PPLL to NRR (Curtis to Schwencer) which fixed operating lin>its for pond operation at 8

F maximum tempe.

a VIE and 23 riiillion ga Ilons minimum inventory for two-unit opera ion bas d on pon" oerformiance tests conducted in July 1983.

Region I believes that sli ic~= incl-enation of he nozzle EIms will no'ionif'.cantly alter ei.her the cata o

con" lusions presented iri the 12/21/

3 letter; -his c'onclusion's

-'ased cn c.'scussions wi:h "Fine<pal NRR Hydrologic Engineerinq.Section re;iewers;or 'h-S'S":S u'.t'mate 'neat

sink, and with ihe nanufaciurer of the

r Lr

'> C)

C

< f j'1

~./p>

., r 1

4 ~ '>

~ ~

1

~

J '1 1

4>i

'4

~

~

~

~ H

. ~ a I I Lr

~ '

~

~ >

>5

~ 5

~ *> 4'

~ 1

~ ~ I

~ P

>>.> ii

.>>1 2 I

)

1

~ >

~ 1 5

~ '>

J >."

~

~

~ >

~ ~

~

~

>5 I'

5>,

~

5

'\\

'4 4 '>M rt r

~

~

L 5

'I

~I ~ )

~ "r 5

~

I r

5 5

~

1

~ 1 I

>r,

~ i

~

1 "

~

~

1

Director, Division of Emergency.

2 Preparedness and Engineering'esponse Director, Division of Licensing t

SPRAYCO Model I75IA nozzle.,

However, the potential for ice blockage of spray nozzles at Susquehanna and other plants with'.spray ponds as ultimate heat sinks (e.g.,

Limerick,'WPPSS-2, Palo Verde, North Anna, and R'ancho "Seco)

. warrants generic consideration.

Susquehanna has provided Operating Procedure precautions to commence drainage of.spray network piping whenever ambient temperatures fall below 41'F, but no instrumentation is provided to detect supply leakage into.the. network piping.

Further, no drainage holes at the nozzle exist..

The issue has.been discussed with R. Ballard and Yi. Fleigel of NRR's Geotech-nical Engineerino Branch; and R. Codell (currently with the Office of NMSS),

..author'f NUREG-0733, "Analysis of Ultimate-Heat-Sink Spray Ponds".

Senior Resident Inspectors a: Limerick, Rancho

Seco, and WPPSS-2 were also contacted.

-- The resident inspecto".s a

Susquehanna are fo'llowing PP5L's safety evaluation

'nd corrective ac.ions.

Unit 1 is in an outage expected to end in mid-February; Unit 2 is scheduled to load fuel mid-Harch.

, This memo is provided to IE for potential generic notifications; to NRR for Susouehanna licensino implications; and to AEOD for information.

Region I' contact is E.'Kelly (FTS 8-488-1132).

I Richard W.

ki D

.arostec, irector Division of Project and Resident Programs Enclo'sures:

Nozzle Arm Sketch 2.

Pic.ure from Jan.

30th PPE L "Nuclear Notes" 3.

FSAR Figures 9.2-5 and 24

- 4.

F'SAR Subsec ion 9.2.7.2.3 5.

Section 10 from Combined Region I Inspection Report 50-387/83-29; 50-388/83-32 cc w/enc',s:

DPRR D-:rectors, Reoions II, III, IV, V

C.

Heltem s, AEOD R. Bal'.ard, Chief, Geotechnical Engineering Branch R.

Perch, Project

Yianager, Susquehanna R. Jacobs,

SRI, Susquehanna

I I

P PP h r

~t I r

~ r I

r Ir

r

~P

)h P II rll

~

.J' I

V C

I i~

h h\\

v\\

'I

'1 h

IP

'"tg.( (g~J Q))(c'Ark S'gc~cg

~

b

~ ~

P,q fi~YQ~ 0L!J

(

Tffj)g-il '0 < Q

/'~0+.J 7

l~'.i '< D &b

~

I I

i g0

< V.,

(7'

,r t""

jf~, SATE(m~na<)py

~c-b

(> 5rtop C 'g)

I

> ~&6

)~g~ p 6

(g.

gPYht gj(

Q4 J"'l~

/~i

~ (c-<

I I

I I

Cg)~~0 WQ

/au.6&

I

<(78 ~

4

~

~

e I

4

'" n C

~ 44, 44

<<le 4

4 l

4'S V

~'

~

g 444 44 4 r

~<<l e

l 4

4

• el'l 444 4

4\\

4:

e.e 4

4

~

44 l

~

l7 j 4

e 4

7

~

~

C ~

Exce~ from.PPKL.Nucl'ear. Dept. heekW'="

("Nuclear ~es" Jan.

30, 5984; Yol.'V, Mue 5)

E~~(or~~ (~).:

t tS

'LS '

~ 'g

~

~

tS

-~l t

. th)%

't L

l

~

1'

~,

C JtL L

@St 4

'4

~ '~~r.

4VCL 7

1 S

~S i~~

~f

~' ~

I

'. as" week work continued on repair of the ESP spray pond nczzles that vere da-aged during the recent cold weather.

See-story on next page.

(

V

~

>> ~

r

~ "

'(

r>>

(

'I ip(

I

~~:i l 7>>

.)

I P

/

I t

>> ~

~ W

~

'4a(7 /r I

~ 7

'(

~

~

('

,/'

I~ 'Il 11 (I

r,

~

0 I

(

pP C>>

I

, 'N(.

'Yl

(

((/

~'

r

(

m'>>>>.

>> r" i">>I( ~ I I

1\\

I

~ i

~

lg

~ p

(

I

(

~

~

~

S 1'

~ I( ~

A I

~

~

~ ~

~

~

~

~

~rv h

I

\\'

~

~

C 4~%~

~

~t r

t

~

h

~

r t

y'

~

t ~

III) lf

'v to l/l/I/IRhc'uP Qgr.ft Ee'p bl fgl I

II(a~

'V&II((0fr.

g (g (E,. g-(aug 2).

+(C//A /

Cher/ ~ f-i~@

~Z(t -mg N C'g I oo(I

.~cg< ~'

)

tttr rt

'IIVC~VI.

U( I ICIC

( +J'v/

~

IA( ta(4tr($

((%(

Ds( ~

~ \\I, IIII (C(.CC

~ tti)

M.~

~-"

• tl III~(tt).lt i

1 'lilt.t)l. ~I Ilttlt)II V

~

tilt(Ilt lttl(i(it) irlf00(

(t 0ttft g IJ 4 llf t)D( ~ I 1 2.1 ICl ~ 1.1 t.

4 I~ I;r.l.~ ~

IM

~ ~ I, ~ ~ '1 tC V

)I(iVIVCOI I

I f;(ff.aiij.ii 4 littil 20 I(t( (lttf

)Il0(tt

~I 10 I

)

tt III,C)f)0I 12'C")(I toe(~I MC I,CICOOI tt0 P f. A tf. SP IlAY P 0 tl 0 8 AQa.eD

/( I~~((m e~f~+

r

~ I

~'tlt tt 4

I r

~ ~I

~

~ ~

~ r C,

~

~

D

( I ~ IC.(201) it vt( 2 I

Drt

/

r w pD 6') Ai /i'g J.

/4.

IVC Itttc)tt DIPtl)

C(t It4tt C Dl II p

~

t QBY' I

I II't(P g)

PCC

~(pre %32 udB Ni'gksu(

twyI I

~

telJI'OO

~

~

~

~

~ ~

~ I I

I f)IIgO/(IQi ut( tt pygmy F(t'(N~

.%sw~q',z;zq s/~g~

": (.) A~(. i'-4 3)(>>,g I('(I/r/

EIDER QGS~I(tl/~i'

~

~

S I

1

~

~

r~

'T" J

~1 pt I

I

~*i

~,l C

L I

h

~

~

h

'a

~

~

~ ~

r

~

e

(

~

~

ah

~ e a>>

e

~ (,laeaQ)(

~ W

<<a

~e<<I'\\ f 1

,I Ie aa e

I e

$ h

~

\\(\\a (v e

r~

~ wwt y ero ar

~ ~ e'

~

~

~

IIa>> at ~ ae

~ a>>tg<<e yee ae a~I

~

>>a>>e

~ rr

(~>>

ee h e

h

~r >>>><< ~ a<<V

~ ee ~art

~ 'e

~ 'ICV 1

6/78

~ M a

4 ~

( ~ Ae>>>>e>>r H<<e<<a CUTOVCIIANNACTCAM CLCCTIIICTTATION UNITT I AND C flkALCAFCTYANALVTIC(ICE(IT STIIAYCOr<O NCTTTOIIK

~ (leO<<9 0

SIOUhC T.C-TC TNCCTXOF O

P t

f i

1 V

I Ao C

~ 'I

~ ~

~'

W l

I

~

P I

q (

I ~

~

'I 1

I ~(I'I 1',

~

a v

r>> a>>,

I

~

~

~ a

~

H I

N

~ a ~ag(l)

H

~ ~ ~ ~ a ~ 'I La P a

a'r<<<<gg

'"K a

~ V

~

.. ~ ~ ar ~

aa aaavaaa~aa

~

~

'aF a~:,.>r--

/a

~ ~

. p./

/Ir I!I v aa

~ a

~

~ a

<<VVarr

~ v<<vva er

~v v v>> aa vrr

'Y~ Iaavg!hl

~v 8ITK SUSOVKIIANNASTKAM KLKCTIIICKTATIOH UNITK I ANO 2 FINALKAFKTYANALY1ITlaKFOIIT

'FIIAYFONO NKTWOIIK I

FIOUIIK 4.2 24 5HKKT 2 OF 2

~

~

~

~

C

\\

~ 0"

~

~ c ~

~

=..SSES-FSAR 4 c.los~~ (Wf in the aakeup ~My sYstem.

This Line is used to fillthe pond initially and to refill it follovinq its use as the result of an eaerqency.

R'hen the pond,.is not, in use the only loss is.by evaporation, vhich is made up by rainfall and a continuous flov of vater throuqh a

0.in bypass line around the closed isolation valve in the 18 in makeup line Any excess water.in the pond flows over a veir and back to th'e river.

Ne teoroloqy for the area indicates that rainfall is expected to add more vater to the pond than is lost by evaporation.

9. 2.7.2 3

System Operation Summer Startu2 and Operation "The conditions that could. exist in the pond durinq summer startup vill be less.severe than the conditions that could exist during summer operation..Ho attempt vill be made to start up eithez

unit.if the pond is not full.

During periods of summer operation with no sprayinq but high ambien.

wet bulb tempezatu"es, the pond temperature vill approach the equilibrium temperature vhich is that temperature a staqnant.

body of water will reach after prolonged'xposure to the ambient

'condit'ons.

The maximum Susquehanna pond temperature vas calcula+ed to be 89~F under these conditions.

Under normal plant operatinq conditions, the maximum pond temperature vill approach the equilibrium teaperature.

Technical Specification limits have been established to 3.iait plant operation if the pond bulk temperature reaches 88~P.

This temperature has been calculated to be the max'iaum allowable startinq temperature if the -ESSV tempezature is to be.'limited to 95~7 'unde-the worst meteroloqical'nd plant accident conditions.

The otal pond vater volume is assumed to start at the highest pond temperature reached after.exposure to the worst ambient conditions.

This assumption is conservative, since at times of qeneratinq unit staztup, the pond temperature vill actually be below this assumed value.

A lover pond temperatu e will inczease the quantity of.sensible heat that can be absorbed by the water volume.

Summer startups vill therefore impose no limitation on the ulti ate heat sink capability sainte-Sta~tuo and ODeration Star+ up of ei'ther unit vill not be implemented unless the spray

pond, spray network.

and pumpinq system are available for opera tion.

P ev.

30, 5)82

9. 2-26

g r 0

~

~ hI

- ~

= ~

l

%e v

1

'SS ZS-FS AB At times of su'bfreezinq tern peratures, procedures vill be enforced to prevent zcinq of the spray system.

These consist primarily of th e foilovinq:

Simultaneous shutdovn of bot.h units for refueling or extend ed maintenance vould, if possible', 'be scheduled to avoid midvinter conditions, thus avoiding the possibility of.freezing the ESSM pump suctions after the beat source is exhausted.

The total return flow of both the RHRSR and ESR pumps vill be first discharqed directly into the pond, thzough a.bypass line, without passing through the spray netvork.

This vill permit the operation of the pond if nozzle's become covered vith ice from, 'for example, a

freezing rain..

As the vatez. tehperature in the pond increases, conduction of heat to the nozzle vill melt any accumulated ice.

The, bypass lin'es enter above the pond level so that they'-drain to prevent freezeup and therefore alvays assure a flov path for the ESV C RHRSR..

The bypass lines are located betveen the tvo spray

netvorks, approximately 400 feet,away from the pump suctions.

The physical distance betveen the pump suctions (which are kept ice free) and the retur n.lines makes the pzoba.bility of increasinq the va ter temperature of the pump suction above the design maximum temperature due to short circuiting negliqible, even if the mond surface does not thaw significantly.

An overviev of the pond. pipinq and bypass lines is shovn on Figure

9. 2-24 Sh.

1.

The bypass lines are numbered as 36" HRC-1 an d 36" HBC-2.

c)

Portions of the nozzle header and riser system that are located above pond water 1evel ca.n be drained vhen not in service.

Draininq is accomplished by pumping the water out of the p'ond pipinq a t, the lov point.

The resultinq water level in the pipe is more than 2 ft belov the pond vater level The majority of the vater distribution system associated with the ultimate heat sink vill be either buried belov the frost line oz located inside heated buildinqs and therefore not exposed to freezinq problems.

e)

Any sections of the pipinq vhich are either not within buildinqs, or drainable vill be electrically traced to protect aqainst freezinq The electrical supply for the'racinq is not supplied from the diesel generators

since, in the event of auxiliary power loss, heated water vill be flowinq in the piping that is traced.

?ev..'.0, 5y82 9 2-27 P

0

(

1 K 3.l JQ.;b 1/

t

'IJ

'b 4

r

'I f

~ The maximum expected ice thickness, assuainq there is no heat load on. the spray pond, is estimated to be 22 inches, vhich aqrees closely vith the maximum expected ice thickness based on probability studies that used field data for colder regions of Horth America.

The extreme veather conditions used for the above analysis vere obtained from meteorological zecords and vere based on the month havinq the lovest averaqe dry bulb temperature This averaqe temperature va.s used for the analysis and the resultinq estimate of maximum ice thickness is therefore conservative-With the extreme (cold) meteoroloqical conditions considered, no

-pzovision is made to prevent freezing of the spray pond surface if both units vere shut dovn at the same time Hovever, freezing

'f the pond vhen the units are shut dovn is n'ot a safety concern.

9. 2.7 3

Sa fet'va'3uation.

The ultimate heat sink spray pond is capable of providing enough coolinq vater to safely shut dorn and cool dovn both reactors, vithout the addition of makeup vater., for 30 days concurrent,

~vith any of the follovinq postulated design basis events:

a)

SSE, flood or drouqht.

Any sinqle site related event c)

A reasonably probable combination of less severe natural phenomena and/or site related events.

d) san-made structural features of the spray pond aze desiqned considerinq all conceivable failure mechani stEs, includinq the SSE and desiqn basis tornado effects.

Conservative allovances are added to the spray pond rater volume to account for seepage through the liner.

(See Table 9. 2-8)

%here "he above desiqn events could result in the loss of offsite

povez, such a loss is assumed.

In addition, a'ingle failure is postulated; T'h e ESSV intake structure is located directly adjacen t to the spray pond; thezefoze, no canals,

conduits, or vatervays are associated vith or required to ensure positive vater flov to the suction of the RH'HSR pumps ana ZS%

pumps The pumps for each "loop are in separate closed rooms vithin the PSSi oumphouse.

There are no communication pathvays be".veen pump rooms Internal floodinq due to a leakage crack in

. he mod~rate R~v 30, 5/82

9. 2-28

'll

~

~

t 9 "I

Excerpt From Region I Inspection Report 387/83-32;388/83-29 10.

Ultimate Heat Sink The licensee found on January 6,

1984 that the Ultimate Heat Sink (Spray Pond) loop 'B'pray distribution network was incapable of performing its safety. function.

Each spray loop consists of 132 risers, with four spray arms on each riser.

Forty-three (43) spray distribution arms in the

'B'oop spray network had decoupled from thei r spray riser'nd fell into the pond due to a combination'of the weight of accumulated ice and insuffi-cient thread engagement to the spray riser nipple.

Initial evaluation by-the licensee indicated that the thread engagement deficiency involved about seven y-five percent of the spray network'arms.

Also, the potential existed for blockage of the spray nozzles due to ice accumulation.

No drain holes exist at the base of.he nozzles to allow drainage of the arms and spray nozzles.

The ice accumulation occurred during a flow balance of the Emergency Ser-vice Mater (ESM) System coincident with extreme cold weather.

On December 22,

1983, the flow balance was

secured, and an attempt was made to..pump-down the isolated spray network piping.

Due to supply isolation valve

leakage, the pumpdown could not be completed.

On December 25,

1983, the spray network was placed in service to.aid in nozzle deicing, since all attempts to pumpdown the system had failed.

Attempts to tightly seat the butterfly isolation valve again fai led on December 28, 1983.

The damage to the network occurred on, January 6,

1984.

The licensee immediately declared the 'B'oop spray network inoperabl,e, and issued:an Equipment Release Form.

The 'A'oop network was not subjected to the flow balance test, so no ice accumulation occurred.

The.'A'oop,

though, also had to be modified since the potential exists for similar degradation.

'ivers were used to retrieve the fallen spray arms.

The inspector obser-ved modification work on the spray headers on January ll and 17, 1984.

The modification consisted of removing the di stribution arm and shop weld-ingg the distribution arm to a coupling.

A field weld was then made on 'the coupling to riser nipple connection.

All of the nozzle arms were tilted slightly from horizon al to permit drainage of the arms into the riser.

(The arms were more nearly horizontal before, and some were oriented sliohtly downward, thus preventing water drainage.)

The leaking spray network isolation butterfly valves were also repaired.

The exi s ino design could allow this situation to recur if a spray network isolation valve were to leak.

ESM, upstream of the spray network isola-tion valves, is.normally pressurized by the keep-f Ill system'.

- ~

I ski j

~

'I I

C t

3

FSAR Section 9.2.72.3 states'hat neither unit will be started up unless the spray pond and spray networks are operable.

The FSAR also describes procedures to prevent icing of the spray system.

One meth'od is to dis-charge the total return flow of both the Residual Heat Removal Service Water (RHRSW) and ESW pumps directly into the pond, through a bypass line, without passing through the spray network.

The FSAR also states that "

nozzle header and riser system parts located above pond water level can be drained when not in service by pumping the water out of the spray piping

,at the low point, and tha.,sections which are not within buildings or not drainable are electrically heat traced t'o protect against freezing.

Ho provisions are made.

o.prevent freezing of the spray pond surface'when units are shut down.

Such freezing is not a safety concern because maxi-mum ice cover is calculated to be 22 inches.

'In. Operational Condition 5, the RHRSW and ESW systems must be declared inoperable if the spray pond is inoperable.

Licensee safety evaluation concluded that,.with Unit 1 in Operational Condition 5, the spray network is not requi red. if the spray pond temperature i s less than 40'F.

On January 17, 1984 the i.nspector noted that 21 spray distribution arms on the 'A'oop had been.,removed for modification.

Both spray'networks were then inoperable, and the RHRSW and ESW spray pond network isolation valves HV-01224A1, A2, Bl and B2 were closed, deenergized,i and tagged.

When this condition was identified, Unit 1 was shutdown in Operational Condition.'5.

The division II systems (i.e.,'B'oop) were already inoperable due to outage work.

'Because spray pond temperature was 36~F, the spray pond

'function remained operable.

The inspector reviewed the licensee's operating procedures and discussed hem with.he Operations Supervisor.

to determi ne if adequate precautions were being taken duri no extreme weather condi ions to prevent freezing of.

he spray network and if the procedures described in the FSAR were being implemented.

Opera.i ng Procedure OP-116-001, Revision 0,

RHRSW, provides for pumpdown of the spray piping if the outside ambient temperature approaches freezing.

It also states that the temperature in the spray pond should be main ained between 55'-88 F (one-unit operation).

During winter periods with both units shutdown, heat load is insufficient to maintain the pond above 55~F.

Operating Procedure OP-142-001, Revision 0,

Circulating Mater System and Cooling Tower Operation, describes normal spray pond deicino using letdown from the cooling tower basin.

Opera.ing Ins.ruction OI-AD-016, Revision 7, Operators

Rounds, Attachment G, is a

plant log for outside buildings.

This log requires the. operator to drain the. spray piping and deice the spray pond as necessary if the outside

.empera ure is less than 41'F.

The i em is unresolved pending completion of modifications and testing, and review of the impact of the modifications on system performance.

(387/83-29-.03; 388/83-32-02)