ML20054M964

Requests Proposed Schedule for Submitting Program to Address J Humphrey Concerns Re Adequacy of Design Margins of Mark I & Mark II Containment Sys

ML20054M964
Person / Time
Site:	Shoreham File:Long Island Lighting Company icon.png
Issue date:	07/08/1982
From:	Tedesco R Office of Nuclear Reactor Regulation
To:	Pollock M LONG ISLAND LIGHTING CO.
Shared Package
ML20054M963	List: ML20054M962 ML20054M964
References
NUDOCS 8207150199
Download: ML20054M964 (12)

Text

.L l 9AL

/

,(#

,\\

UNITED STATES

, 3 e c (,p, NUCLEAR REGULATORY COMMISSION y%,.bgyc j WASHINGTON, D. C. 20555

%' @ f e o,,

/

Jul.

8 1982 Docket Nos.: 50-322 Mr. M.S. Pollock Vice President - Nuclear Long Island Lighting Company 175 East Old Country Road Hicksville, New York 11801

Subject:

Concerns Regarding the Adequacy of the Design Margins of the Mark I and II Containment Systems A number of concerns regarding the adequacy of the General Electric Mark III containment design have been raised by Mr. John Humphrey, a fonner employee of GE.- When these concerns cane to the attention of the NRC staff on the Grand Gulf docket, we initiated a series of telephone conference calls and meetings.

While these concerns were first directed by Mr. Humphrey towards Grand Gulf, in subsequent discussions he extended his concerns towards other facilities.

Af ter considering Mr. Humphrey's concerns, we concluded that this infomation is potentially generic and may have applicability to all dockets with boiling water reactors which use the GE pressure suppression containment systems (i.e., the GE Mark I, II and III containment systems).

The enclosure to this letter contains 22 categories into which Mr. Humphrey's concerns were coalesced on the Grand Gulf docket as of June 21, 1982, less those items which are clearly associated with a Mark III contaiment.

In some instances, we have modified an original concern to make it more applicable to the Mark I and II contaiments and have added some footnotes which generalize Mr. Humphrey's concerns. We request that you respond within 14 days of receipt of this letter with your proposed schedule for submitting a program I

to address those concerns which we have identified as being potentially l.

applicable to your facility.

If you have any questions on this matter, please contact the assigned project manager. Mr. Byron Siegel (301-492-7534) is the contact for operating Mart I facilities and Mr. M.D. Lynch (301-492-9793) is the contact for all other facilities.

This infonnation is being requested under Section 50.34(b)(2)(i) of 10 CFR Part 50 and was approved by CNB under a blanket clearance 3150-0011 which expires April 30, 1985.

Sincerely, S O c > 4, Me L. Tedesco O

8207150199 820713 Assistant Director for Licensing PDR ADOCK 03000322 c

PDR Division of Licensing

Enclosure:

As stated cc: See next page

Shoreham Mr. M. S. Pollock Vice President - Nuclear Long Island Lighting Company 175 East Old Country Road Hicksville, New York 11801 cc:

Howard L. Blau, Esquire MHB Technical Associates Blau and Cohn, PC.

1723 Hamilton Avenue, Suite K 217 Newbridge Road San Jose, California 95125 Hicksville, New York 11801 Stephen Latham, Esquire Mr. Jay Dunkleberger Twomey, Latham & Shea New York State Energy Office Post Office Box 398 Agency Building 2 33 West Second Street Empire State Plaza Riverhead, New York 11901

~

Al ba ny, New Yo rk 12223 Matthew J. Kelly, Esquire Energy Research Group, Inc.

Staff Counsel 400-1 Totten Pond Road New York State Public Service Commission Waltham, Massachusetts 02154

' Three Rockefeller Plaza Albany, New York 12223 Mr. Jeff Smith Shoreham Nuclear Power Station Ezra I. Bialik, Esquire Post Office Box 618 Assistant Attorney General Wading River, New York 11792 Environmental Protection Bureau New York State Department of Law W. Taylor Reveley, III, Esquire 2 World Trade Center Hunton & Williams New York, New York 10047 Post Office Box 1535 Richmond, Virginia 23212 Resident Inspector.

Shoreham NPS, U.S. NRC Ralph Shapiro, Esquire Po3t Office Box B Cammer & Shapiro Rocky Point, New York 11778 9 East 40th Street New York, New York 10016 Herbert H. Brown, Esquire Kirkpatrick, Lockhart, Hill, Mr. Brian McCaffrey Christopher & Phillips Long Island Lighting Company 1900 M Street, N.W.

175 E. Old Country Road Washington, D.C.

20036 Hicksville, New York 11801 Lawrence Coe Lanpher, Esquire Honorable Peter Cohalan Kirkpatrick, Lockhart, Hill, Suffolk County Executive Christopher & Phillips County Executive / Legislative Bldg.

1900 M Street, N.W.

Veteran's Memorial Highway Washington, D.C.

20036 Hauppauge, New York 11788 Karla J. Letsche, Esquire David Gilmartin, Esquire Ki rkpatrick, Lockhart,, Hill, Suffolk County Attorney

~

Christopher & Phillips County Executive / Legislative Bldg.

1900 M Street, N.W.

Veteran's Memorial Highway Washington, D.C.,20036 Hauppauge, New York 11788

](

l N.EY.,f41CA.I2p.ID. C CONCERNS 1.

If fee:s of Leezi Ir.creach=enis en ? col Svell Leads

~ 1.1 1.2 a

3.3 N/A for Mark I and Mark II Containments 1.4 s'

1.5 1.6 1.7 2.

Safety Relief falve Discharge Line Sleeves 2.1

\\

N/A for Mark I and Mark II Containments g

m.

2.3 e

+ir

-r.

.-n

~ - %

y -,

y

3.

ECCS P.elief Valve Discharge lines 3elev the Suteressien pool lese -

3.1 The desig= of the STR1DE plant did sc: consider vent cleartigl cendensatien oscillation and..cF.ugqig1 cads which migh: be ):oduced by the actuatics of these relief valves.

3.2 The STRIDE design previded only nine inches of sub=e:ge=ee above :he F2?.

relief valve discharge lines a: lov suppression pool levels.

3.3. Discharge f:c= the FIP. relief Valves =ay produce. bubble. discharge or other sub=e:ged s::ucture 1 cads c= equip =en: in the suppressien pool.

3.4 The 72R hea: exchaager relief valve discharge lines are previded vi:h vacuu= treakers to preve : negative pressure in the lines when discharging stea= is condensed i= the pool.

If the valves experience repeated actuation, the vacuu: breaker si=ing =ay not be adequate to'

~

preve=: drav1=g slugs of water back th cush the discharge pipi g.

These slugs of water =ay apply i= pac: loads to :he relief valve or be discharged back Tnto the pool at thu (ex: relief valve actuarien and

. apply i= pact loads to sub=erged structurps.

3.5 N/A forMark I and II Containments.

3.6 If the 722 heat exchanger relief valvo discharge 'stea= to the upper levels of the suppressien pool folleving a design basis accident, they vill significantly aggravate suppressics pool'te=perature s::atificaties.

3.7 The conce=s related to the 72R heat excha=ger relief valve discharge lines should also be addressed for all other relief lines that exhaus:

l i= o pool. (p.132 of 5/27/82 ::anscrip:)

4.

Superessie'= pool Te=cerature Stratifica' tion 4.1 The preses: ce=:ai==ent respense analyses fer dryvell break accidents.

assu=e that.the ECCS syste=s transfer a significas: quantity of va:e:

fre= the suppression pool to the lever regiens of the dryvell th cugh the break.

This results i= a pool i= the dryvell which is essentially isolated fre= the sup'pressien pool at a te=perature of app cxt=ately 135'F.

The centai==en: response analysis assu=es that the d:yvell peel is thereughly -Ned with the suppressien pool.

If the inventery in the I

dryvell is assu=ed to be isola:ed and the re_ainder of the heat is discharged to the suppression pool, an increase is bulk pool te=perature cf 10*F =ay occur. /1 4.2 The cris:ence of the dryvell peel is predicated upon centinueus operatics i

of :he ECCS.

The curren: e.=ergency procedure guideli=es require the

~

l operators to thre::le ECCS eperatic: to =aintain vessel level beTev level l

8.

Conseque::1y, the dryvell pool =ay sever be fe==ed. L2_

4.3 All P. ark III analyses presently assd:e a perfectly =ixed unifer=

suppressien pool.

These analyses assu=e that the te=perature of the I

sue:ien to the p2R heat exchzugers is the sa=e as the bulk peel-l te=perature.

In actuality, the te:pera:ure in the lever part of the pool g

-*wm

-,,-.3m.

. s*

s where the suction is located win be as =uch as 7h*F cooler than the bulk pool te=perature, nus, the heat-transfer through the In heat. exchanger vin be less than expected.

4.4 ne long te:= analysis of'contiii=ent, pressure /,te=perature respodse assuses that the vervell airspace is in ther=al equilibrtum with the suppression pool water at all tines.

De calculated bulk pool te=perature is used to dete=ine the at space te=perature. If poo1~

ther=al stratification vere considered, the surface te=perature, which is in direc: cc.,ntact with the airspace, vould be highet..narefore the airspace te=perature (and pressure) would be higher.

4.5 A nunber of faotors =ay aggrava:e suppression pool ther=al stratdfica:Lon.

ne chuggi g produced through the first row of hori ental vents viu no: produce any mixing fro = the suppression pool layers below the vent rev. An upper pool du=p =ay contribute to' f

additional suppression pool te=perature stratification.

ne large volu=e of water fro = the upper pool further submergesi R R beat exchanger effluent discharge which vin decrease =1xing of the botter, uppe:

regions of the pool.

Fi= ally, operation of thi contai==ent spray el4 dnates the heat exchanger effluent discharge jet which contributes to 4 * " g. / 3, 4.6 The initial suppressien pool te=perature is assu=ed to be 95*7 vhile the '

=axi== expected service water te=perature is 90*F for an GGNS accident.

analyses as noted in TSAR table 6.2-50.

If the service water te=perature

..is consistently higher than expected, as occurred at Kuosheng, the P a syste= =ay be' required to operate nearly centinuously in order to

=aintain suppression. pool te=perature at or below the eh pe=issible value.

4.7 All analyses ce=pleted for the Mark HI are generic in nature.and do not consider. plant specific int'eradtions,of t'oe Pa suppressien pool suction and discharge.

i 4.8 Operation of the Fa.systen in the conta - ent spray =ede vin decrease d

the heat transfer coefficie c through the pa heat exchangers due to decreased syste= flow, ne 75AR analysis ass =es a constant heat transfer ra:e fro = the suppression pool even with operation of the co:1:ain=ent spray.

4.h ne effect en the long tem contai=ent respense and the operability of the spray syste= due. :o cycling the contai=ent sprays on and off to

-wi=ize pool cooling needs to be addressed.

Also provide and justify the cri:eria used by the operator for svitching fre= the contain=en:

spray =ede ~ o pool cooling =ede, and back again.

(pp.147-148 of 5/27/82 transcript) 4.10.7ustify that the current strangenent of.the discharge and suc':ien points l

of the pool cooling syste= =axi=izes pool =ixing.

(pp.150-155 of 5/27/82 transcript) i

1. 9 a#%

e

--n.

n

-r

5.

Drvueli to contai=ent Ev ass Leakare 5.1 The vers: case of dryvell to centa4 est bypass leakage has been '

established as a s=all break accident.

An inter =ediate break acciden:

.l vill actually produce the =os: sdm.ificant dryvell to ec=:ai:=en: leakage prior to ini:iatics of con:ai =ent sprays.

5.2 Under Technical Specifica:ic, id-4:s, bypass leakage correspe=di:g to.

M(IC = 0.1 f:.2 cc s itute acceptable operating conditie:s.

5 aller-than-IBA-sized breaks can =aintain break flev i=:n :her dryvell fer leng ti=e periods, however, because the P2'{ vould.be.decressurized ever a 6 hour6.944444e-5 days <br />0.00167 hours <br />9.920635e-6 weeks <br />2.283e-6 months <br /> period.

Gives, for exa=ple, an 53A vith A/[f = 0.1, proj ected ti=e period for cc tad--== pressure to reach 15 psig is 2 hours2.314815e-5 days <br />5.555556e-4 hours <br />3.306878e-6 weeks <br />7.61e-7 months <br />.

In the IM;ter 4 hours4.62963e-5 days <br />0.00111 hours <br />6.613757e-6 weeks <br />1.522e-6 months <br /> of the depressurization the centa4-a::

vould presu= ably erperience ever-increasing overpressurizatics. f4 5.3 leakage frc= the dryvell to cc :ai=en vill increase the te=perature.and pressure in the centa4 ent. The operators vill have to use :he ce::aie=ents spfay in order to =.adhra*in ecstai--a'st te=perature a=d pressure ec=:rol.

Given the decreased effective =ess of the p_=F. syste= in ace==plishing this obj ective,.in the ce= tai =ent ipray = ode, the bypass leakag, =ay increase the cyclical duty of the contad- :: sprays.

5.4 Direct leakage fre= the dryvell to the ee::a4--ant =ay dissipa:e hydrogen.

curside the regien where the hydregen rece=Siners take suctics. The a::icipated leakage exceeds the capacity of the' dryvell purge ce= pressors.

This could lead to pocketi=g of hydrogen which e=ceeds the cencentration 1'-4: of 47. by volu=e. f 5

~

5.5 Equip =en =ay be exposed to local conditions which exceed the environ = ental qualificatien ecvelope as a result of direct aryvell to contai==ent bypass leakage.

5.6 N/A for Mark I and Mark II containments 5.7 1

I

{

5.8 The pessibility of high te=peratures i= the dryvell vithout reachi=g the 2 psig high pressure scra= level because of bypass leakage threttgh th,e l

drf.: ell vall should be addressed.

(pp.168-174 of 5/27/S2 transcript) l l

i l

i I

\\

6.

RER Per=issive en Cc: tai =ent Scray l

6.1 We understand that GE' has recommended 'for Mark III containments that the combustible gas control systems be activated if the reactor vessel water-level drops to within one foot of the top of the active fuel.

Indicate what your facility is doing in regard to this recommendation.

• 6. 2 General Electric has recc ended tha: an 5:erlock be provided to require co::ab=es: spray prior to starting the rece=biners because of the large quantities of hea: input to the contat=ent.

In* correct' i=ple=enta:1cs of this in:erlock could result in inability to operate the rece=biners without contai==e:: spray. f.5 6.3 he rece=tiners =ay produce " hot spots" near the rece=biner exhaus:s which ight ex:eed the e:vire== ental qualifica:ics e:valope or the c==:ain=ent design te= pere:ure, g 6.4 For th'e cente ent air =o=itoring,syg:e= furnish d by General Electrie, the a:alyzers are not capable of =essu 1 g hydroge concentra:Len at 3

volu=e:ric stea= cencentrations above 60..

Effec:ive =casure=e:: is precluded by c=densation of stea= in the equip =e==.

6.5 Discuss the possibility of local te=pera:=res due to rece biner opera: ion being higher than the re-perature qualifica:ic: profiles for equip =en: in the regie: around a=d above the rece=biners.

State what instrue:1cns, if any, are available to the opera:or to actuate contad-en: sprays to keep this te=perature below design values.

(pp.183-185 of 5/27/82 transcript) g 7.

Contai==ent Pressure Resnense 7.1 The wetwell is assu=ed to be in ther=al equ' librium with a perfectly-

=1xed, unif,or= te=perature suppression pool. As noted under tepic 4, the surface te=perature of the pool vill be higher than the bulk pool T11s =ay produce higher tha= expected centai=ent te=perature.

1 te=peratures and pressures.

7.2 The ce=puter code used by General F.lectric to calculate envire== ental qualification para =eters considers heat transfer f c= the suppression pool surface to the 'ce= tai==en: atmosphere. This is not in accordance vidh the ext:.ti=g licensing basis for Mark III. envire= ental qualifica:icn.

Additionally, the bulk suppressien pool temperature was used in the analysis i= stead of the suppressien pool surface te=perature.ff, 7.3 The analysis assu=es that the wetwell

. airspace is in ther al.

ecuilibriu= with the suppressics pool.

In the sher: ter= this is non-censervative for Mark III due to adiaba:ic ce=pression effects a=d '

finite ti=e required for heat.and = ass to be transferred between the pool and cen ad-es: volu=es.f6 8.

Centain=ent Air Mass Effects 8.1 This issue is based en c sidera: ion that se=e Tech Specs allev eperatie a: para =e:e: values that differ fre= the values used in assu=p:1 ens for FSAA ::an'sie : analyses.

Normally analyses ',are dene assu=ing a ne=inal

centa' est pressure equal to a: bien: (0 psig) a ta=perature near =ax1=u=

opera:ing (90*T) and do not li=it the dryvell pressure equal to the conta'==es: pres sure. The Tech Specs operatic under conditions such as a posi:ive ces:ai =e=: pressu

=axi=u= (60 or 70*F) and dryv'n (1,Sagpis), te=peratures less thatell'p the ec :ain=es: (-0.5 psid). All of thes~e differesces venid result in transie : resp::se different than the FSA.R descriptions.

' 8. 2 The draf: CSSS technical specifica:icus per=it operatics of the, plant vith centai=e:t. pressure ranging betwee O and -2 psir, 2 itia:1c ef centainment spray at a pressure of -2 psig =ay reduce the ecstai =ent pressure by as additie:al 2 psig which could lead to buckling and fcilures 1: the cc:;:Ai==ent liner plate.

8.3 If the contai-- :: is =ai= ained at -2 psig, the top rev of vents could-ad 1: bicuders :o the suppressics pool duri=g as SEA vithout a LOCA signal being developed.g 8.4 Describe all of the possible nethods b):h before and after an accident of

. creating a e,c=dition of icv air = ass inside the conta'- est. ' Discuss the effects c the centa'- en: design exte:nal pressure of actua:1=g the cc :ai==en: sprays.

(pp.190-195 of 5/27/82 transcript) 9.

Tinal D:vvell Air Mass 9.1 The curren: TSAR analysis i-hsed upen e==tinueus injectics of rela:ively cool ECCS vater into the dryvell through a broken pipe -

folleving a design basis accident.

Since th'e. operator is directed to throttle ECCS operation to maintain 'the reactor. vessel water level tocabout th'e level of the 3 team lines, the break will *be re-leaking satur,ated-steam instead of releasing relatively cool E'CCS water _.

Therefore, the~ dry'well air which would have been purged and then drawn back into the drywell, will remain in the wetwell and higher pressures than anticipated will result in both the. wetwell and the drywell.

9.2 The ces:i=ucus s:

d g produced by throttling the ICCS flow vill eause increased direct leakage fro = the dryvell to the wetwell. This could

-~

resul: in increased wetwell pressures.

~

9.3 'I:. appearc th'i: se=e' cesfusien exists as to whether 53A's and stuck epe SRV accidents are treated as transie :s or design basis accidents.

Clarify hev they are treated and indicate whether the initial c ditions I

vere se: a:===inal er licensing values.

(pp. 202-205 of 5/27/82 transeript)

~

10.

Dryvell Fleedict Caused by Deeer ?oel Duro l

10.

N/A for Mark I and Mark II Containments

10.2 N/A for Mark I and Mark II Containments 11.

coera:1enal Control of Drvuell to Centain=ent Diffirentiri-Pressures Mark III lead definitions are based upon the levels in the suppression pocl and the dryvell vet: annulus being the same..

The CGNS technical specifications per=it elevatien differences betxeen these-pools. This r_zy effect load definitien for vent clearing.18

~

12.

Sueoression Pool Makeen LOCA Seal In N/A for Mark, I and' Mark II Containments

!}inetySecondSprayDelay 13.

a N/A for Mark I and Mark II Containments 14.

RER 3ackflow Threugh Contai=ent Soray A failure in the check valve in the IPCI line to the reactor vessel could result in. direct leakage from the pressure vessel to the contain=ent at=esphere. This leakage =1ght occur as the LPCI =ctor operated isolatics valve is closing and the motor operated isolation valve in the,.

centa'e-me spray line is opening. This could produce unanticipated increases in the conta1=nent sp. ray.

15..

Se endary Centainment Vacuum 3:eaker Plen== Reseense the STRIDI plants had vacuu: breakers between the centa1=ent and the secondary contain=ent. With suf ficiently high flows through the vacuu=

breakers to contain=ent, vacuun could be created in the secondary centain=ent.

16.

Effect of Suveressien Pool level on Te=perature MeasuYe=ent Sc=e of the suppression pool te=perature sensors are located (by SE reco=endation) 3" to 12" below the pool surface to p cvide early varning of high pool te=perature.

Ecuever, if the suppression pool is drawn down belov the level of the te=perature sensors, the operato: could be = isled by erronecus readings and required safety action could be delayed.

_______________.__________a

17.

E=ergenev Procedure cuidelines [9 The IpCs contain a curve which specifies limitations on suppressidn~ pool 1evel and reactor pressure vessel ytessure. The curve presently does not adequately account for upper pool dt p.

At present, the operator would be required to initiate auto =atic depressurization'when the only action required is the opening of one additional SRV.

18.

If f eets of Insulation Debris /10 18.1 Tailures of reflective insulation in the dryvell may lead to blockage of the gratings above the weir a=ulus. This may increase the pressure required in the dryvell to clear the first row of dryvell vents and perturb the existing load definitions.

18.2 Insulation debris =ay be transported through the vents in the dryvell vall into the suppression pool. This debris could then cause blockage of the suction strainers.

19..Sub=ergence Ef fects en Chugging Lesds 19.1 N/A for Mark I and Mark II Containments 19.2 N/A for Mark I and Mark II Containments 20.

Loads en Structures Piping and Eouipnent in the Drvve11 During Reflood N/A for Mark I and Mark II Containments

~

21.

Containment Makeup Air For Backup Purge Regulation Guide 1.7 requires a backup purge H re: val capability.

..o 2

. This backup purge for Mark III is via the dryvell purge line which Bischarges to the shield a=ulus which in turn is exhausted through the standby gas treat =ent systen (SGTS). The contai=ent air is blevn into the dryvell via the dryvell purge conpressor to provide a positive purge.

The co: pressors draw fro = the contaip=ent, however, without hydrogen lean air =akeup to the contain=ent, no reduction in centa k ent hydrogen concentration occurs.

It is necessary to assure that the

~'

shield a=ulus volu=e contains a hydrogen le.an =ixture of air to be admitted to the containment via contain=ent vacuu= breakers.

For Mark I and II facilities, discuss the possibility of purge exhaust being mixed with the intake air which replenishes the containment air mass.

l

.t.

22.

Miscellane.ous E=ergency Procedure Guideline Concerns f 2, The EPCs currently in existen:e, have been prepared with the intenf 6f coping t.-ith degraded core accidents.

They r.ay contain require =ents' conflicting with design basis. accident conditions.

So=cene needs to carefully reviev the I?G's to assure : Fat they do hot co:fiict with the expected course of the design basis accident.

O 9

• e e

O m

3 l

e d

l 1

1 i

l l

e 4

g t.

b

• e TABLE OF FOOTNOTES APPLICABLE TO MARK I AND MARK II CONTAINMENTS Footnote Ccmment i

This concern is related "to the trapping of water in the drywell.

2 This issue applies only to those facilities for which EPG's are in effect.

3 For Mark I and II facilities, confine your response on this issue to those concerns which can lead to pool stratification (e.g.,

operation of the containment spray).

4 For Mark I and II facilities, refer to Appendix I to Section 6.2.1.lc of the Standard. Review Plan (SRP).

5 This concern applies to those facilities at which hydrogen recombiners can be used.

6 This issue as phrased applies only to a Mark III facility. However, the concern can be generalized and applied to the earlier containment types.

For Mark I and II facili-ties, indicate what methodology was used to calculate the environmental qualification parameters including a discussion of heat transfer between the atmosphere in the wetwell and the suppression pool.

7 Not applicabl'e to Mark II facilities.

8 For Mark I and II facilities, consider the water in the downcomers.

9 This issue as phrased applies only to a Mark III facility. However, the concern can be generalized. Accordingly, discuss what actions the reactor. operator would take in the event that the limitations on the suppression pool level and the pressure in the reactor vessel are violated.

10 This issue as phrased applies only to a Mark III facil ity.

However, the concern can be generalized.

Accordingly, discuss how the effects of insula '

tion debris could perturb existing load defini-tions or could block suction strainers.

In responding to this issue, you may refer to existing generic studies; e.g., the study done for the Cooper facility.

.