ML17158A287

Forwards Info Re NRC RAI Concerning Loss of Spent Fuel Pool Cooling Issues & Standby Gas Treatment Sys Performance Following Seismic Event

ML17158A287
Person / Time
Site:	Susquehanna
Issue date:	05/17/1994
From:	James Shea Office of Nuclear Reactor Regulation
To:	Lochbaum D, Prevatte D AFFILIATION NOT ASSIGNED
Shared Package
ML17158A288	List: ML17158A287
References
RTR-REGGD-01.003, RTR-REGGD-1.003 TAC-M85337, NUDOCS 9405240117
Download: ML17158A287 (28)

Text

W/AO/> Sn4 14: '77 lU A-79FiA PR sir.

i Tea C'AR=

A4 Enclosure to PLA-4133

.9405240117

'VALUATION Or SGTS RESPONSE TO A SEISMIC EVENT Pennsylvania Power & Light Company May 2, 1984

0

AR/A4/1994 14:77 A1A-7 79FiA 4

PPI n!IIV. i 'HW

~A PAAF'Fi Attachment to PLAA199 i.0 OBJECTS E Thc purpose ofthis report is to document a suiniruuy of the evaluations performed ofthe SSF5 plant response to a postulated

'dual fuel pool boil event assumed to bc caused by a seismic event.

ALOOP is also assumed to occur coincident withthe seismic event.

Specifically evaluated are the oFsite dose consequences, impact on thc operability of SGTS, Imp+1 on rcac(vr bulldog aud control structure cquipmcnt.

Several analyses are required to describe all of the consequences.

These require expertise in the radiological, structural, systems and HVAC analyses

areas, Separate. analyses have been performed in each of these areas.

The referenced documents contain the details ofeach ofthese analyses.

These analyses are suaunarizcd hcrcin.

Should more specific details than thcsc provided herein be required, these documents should be reviewed.

i Bd 1

BU<<h "d'~

• i'.~

h scenario postulated in FSAR Appendix 9A.

FSAR Appendix 9A assesses, in a conservative manner, the'f a dual fidel pool event, Inputs and assumptions utiiizul in the Appendix 9A analysis were chosen to maximize impact on offsitc dose.

Thc inputs/assuinptions and sequence ofevents'werc chosen solely to conservatively exhibit that the pffft~consequences of a dual fuel pool bciHng event are weB within radiological hmits.

Reference 9 contains additional discussion of the applicable SSES limits and Licensing requirements.

Itis not appropriate to utilize the same set ofinputs/assumptions and event sequence to perform a plant espouse evaluation ofthe impacts ofa postulated dual fuel pool boiling event caused by a seismic event, Use of the same set of inputs/assumptions and sequence for an assessment of plant response would result in unrealistic results, Therefore, the evaluations comprising the basis fnr the p)ant response evaluations summarized herein were not based on the sct utilized for the Appendix 9A analysis.

Thc inputs/assumptions and the event sequence which are utilized for this evaluation are dctaBcd herein. The bases and conscrvatisms arc idcntificd.

2.0 CONCLUSION

S dk RRCQMMENDATCONS Section 5,0 ofthis report contains the details supporting the cunvlusions identified herein.

The offsite doses resulting &om this event would be weil within the Accident Dose limits ofboth 10CFR100 and SRP 9.4.2 without t May credit for SGTS operation durIng the event.

Operation of SOTS for various times was also evaluated to assess thc impact. It is concluded that ithas little impact.

AS/A4/1cIR4 1d: 97 A1A-7RRR PPI Nl I( ~

I TC" PARF Aa Attachment to PLA4133 Reactor building tempcmturcs have bccn detcanincd.

Tho conditions have been used as input to assess the affect on reactor building equipment and SGTS.

The followingis concluded:

The recirculation plenum water accumulation rate will cause the plenum accumulation to begin spilling over into plenum connected ducting, including the SGTS duct, before the SGTS duct condensation reaches its analyzed limit. The spillover occurs 51 hours5.902778e-4 days <br />0.0142 hours <br />8.43254e-5 weeks <br />1.94055e-5 months <br /> after event occurrence.

This is cnnsidered to be the aaalyzed limit. Spillover is uIuutalyzed and considered to be undesirable.

2.2.2 The reactor building conditions created by the assumed event as analyzed only cause equipment located on the refueling Qoor area to exceed the previously evaluated environmental conditions.

The limiting equipment has been evaluated and determined to be capable ofperforming its design function for up to 30 days.

The water ac<<umulatiun rate for the limiting SOTS duct section is determined to reach the analyzed limit 36 hours4.166667e-4 days <br />0.01 hours <br />5.952381e-5 weeks <br />1.3698e-5 months <br /> a6er start of'pool boiling, and 71 hours8.217593e-4 days <br />0.0197 hours <br />1.173942e-4 weeks <br />2.70155e-5 months <br /> (approximately 3 days) afbr event occurrence.

The SGTS room temperatures have been determined.

The conditions have been used as input to assess thc affect on the equipment in the room. It is concluded that the resulting temperature of the SOTS room would be within the ma1dmum normal operating temperature ofthe SGTS room at the time inwhich the recirculation plenum reaches the ana1yzed limit.

3.0 ASSUMPTIONSIINPUYS

'Ihc following assumptions)input data are appHcable to a11 aspects of Ihe analysis..

3.1.1 It is assumed for this analysis that loss of spent fuel pool cooling to both units occurs as the result of a seismic event.

Additionally, a Loss-Of-Afrite Power (LOOP) is assumed to occur coincident withthe seismic event.

3.1.2 RHR fuel pool cooling mode is assumed to bc unavailable.

This assumption is necessary to assure dual fuel pool boiling occurs.

With RHR fuel pool cooling mode available, the fiIelpools are nnt expected to boil.

A9/A4/> 994 14: 97 FT A-7 79rR PPI NI I(;

I T('Ft 8

pa(~F A7 Attachment to PLAA133 3.].3 This analysis is based on sequential unit refueling outages which are much shorter than past and current schedule practice.

This shortens the time tn boil and increases the boiloff rate.

Umt 1 is assumed to be in an outage, since this maximizes the condensation rate in the SOTS duct due to lower assusnol temperatures in the unit during an outage.

'IMs is due to the duct being rout@1 through the Unit 1 reactor building (Zone I).

The period of time in which the plant is in this configuration is small, It mould more likely be in a configuration where both units are operating.

With both units operating, the event would progress at a slower rate than determined herein. Also, future operation is planned tn be with the pools crosstied and not isolated as is evaluated herein.

nte time sequence ofevents forthe scenario analyzed is as follows:

T 0

Unit 2 Shutdown for. Befiielingk lTIspectinn Outage (RIO), 1/3 core oEoad fillsUnit2 pool, T

100 days Unit 1 Shutdown for RIO, Unit2 at 100% Power Unit 1 A 2 spent fuel pools iso)atmo with FPCACU systems operating with 550 GPM demineralized flow for each pool.

Normal deminerallzer Qow ls 600 OPM per reference 2, thus Ihc assumption of550 GPM is conservative.

Thc voter volume for Unit 1 &el activity leakage is the reactor coolant system volume. (10,100 ft3)

Unit 1 RWCU is operating with 135 GPM flow through each demineralizer for a total 270 GPM demineralim Qow. This is a typical Qow value.

T = 102 days Unit 1 Reactor Pressure Vessel (RPV) Head Removed.

(This is a typical time duratinn fnr head Temnval.)

The refueling water volmne for Unit 1 fuel activity leakage is thc reactor pressure

vessel, moisture separator pit and the reactor cavity. (85,326 fthm)

T = 105 days Open Unit 1 Fuel Pool Gates (Typical)

The refueling water volume for Unit 1 fuel activity leakage is the reactor pressure vessel, moisture separator pit, the reactor cavity and Unit 1 spent fuel pool. (134.016 fP)

AR/A4/1994 14:'77 818-7

-7988 PPI NI IC.'

Tr'F A

&4C.".'8 Attachment to PLAA133 Unit I 4 2 FPC4CU systems still operating with SSO GPM demineralized flowfor each pool, T = 106 days Remove Cask Pit Gates, Begin Unit 1 Core Offload (Typical)

The refueling water volume for Unit 1 fuel activity leakage is the reactor prcssuro vessel, moisture separator pit, the reactor cavity, cask pit and Unit I and 2 spent fuel pools. (192,504 ft3)

Ave FPC.'8cCU system operating with one demineraluer and 550 GPM flowfox refueling water volume as noted above.

T 111 days Isolate Unit 1 (Zone I) from the recixc plenum. Zones II 8h III are aligned to the recirc plenum.

T-115 days Complete Unit I Core ONoad (Typical)

RWCU system shutdown.

T 127 days Core Reload Begins (Currently this occurs later.

Starting core reload earlier in the outage sequence results in higher decay heat loads when the pools are isolated later in the sequence.

This results in higher boilo8 rates and offsite doses, as compared to those which would occur had the sequence been based on a current outage schedule.

This adds conservatism to this analysis.)

T 133 days Unit 1 RWCU placed into operation with 270 GPM demineralized low.

T-135 days Unit 1 Core Reload Complete; Cask Pit and Fuel Pool Gates Installed and SI'P's are Isolated. (Typical)

The refueling water volume for fuel activity leakage is the Unit 1 spent fuel pool. (48,690. fP)

SEISMC EVENT OCCURS WHICH CAUSES THE LOSS OF BOTH UNITSFPCACU SYSTEMS, A LOOP IS ALSO ASSUMED TO OCCUR.

NORMAL HVAC SYSTEMS ARE NOT FUNCTIONAL. THE REACTOR BUILDINGRECIRCULATION SYSTEM AND SGTS AUTOMATICALLYSTARTS AS RESULT OF THE LOOP MIINGZONES II AND III ONLY. (ZONE I IS ISOLATED PROM THE RECIRCULATION PLENUM DUE TO

, THE OUTAGE,)

4

Afi/Ad/T9'9d 1 d'7 IilA-"

79fiR PPI Mal I T('F ti PAIiF A9 Attachment to PLA-4133 AtT-135 days+ 35 hours4.050926e-4 days <br />0.00972 hours <br />5.787037e-5 weeks <br />1.33175e-5 months <br /> Unit 1 Pool Begins To Boil (Utilizes conservative Time-to Boil

Methods, thus adds conservatism to this assumed sequence).

Recirculation system fans are turned offprior to start ofboiling.

AtT ~ 135 days+52 hours Unit 2 Poo} Begins To Boil (Utilizes conservative Time to Boil Methods thus adding conservatism to this'assumed sequence).

Notes:

(1)

There is effectively no difference relative to SFP cooling capabilities between a unit at power with an isolated SPP and one in an outage with an isolated SFP and a Sdl core in the RPV. It will be necessary to provide cooling to two separate locations in each unit.

Thus, with the above assumed scenario, Unit 1 is in an outage and Unit 2 is operating at MO %

powers (2)

The above sequence provides for a conservative analysis with regard to SFP heat load, time-to-boil, and boil-offrate to maximize the impact ofthe event.

(3)

Attachment 1 contains the time to boil and evaporation rate calculation, 3.1.4 It is assumed that makeup water is available for the spent fuel pool after boiling begins and the volume of water in the boiling pool remains constant.

ES% is a seismically designed system which willbc accessible in this scenario.

The reactor building source term will be no more than that occurring during normal plant oper'ation because ofthe essentially noncxistcnt source tenn created by this event.

Cooling effects ofthe ESV Qow arc accounted for in the HVACanalyses and not in the radiological analysis.

3,1.5 All heat in thc fuel is assumed to bc transferred to the water; none is lost to the environment, This is conservative because this elevates the boiloffrate.

3.1.6 Outside daily average design temperature air has been assumed when assessing the "cooling" impact ofreactor building inleakage on refueling floor temperatures.

No credit for inleakage cooling is taken for Zone II. These conditions are 73'F and 98% RH and are appropriate A>r long term Lrunsicnt mudyscs which is consistent with other SSES design base analyses.

Zone IIinleakage is assumed to be 800 cfm and Zone III 1845 cfm, ofwhich only 1000 c5n is credited for cooling the re6Ieling Boor.

Assumption of daily average conditions is appmpriate since P

AFi/A4/1994 t 4: '77 Rt A-"

79FiR PPI Nl IC.

I TCS' PAhF

'IA Attachment to PLAA133 this is a t'ransicnt 30 day analysis in which the temperature would cycle between night and day conditions.

3.1.7 The outside air supply to the SGTS train is assumed to.bc at the design basis peak condition for the plant of92'F db and 78'1'b due to the short term nature ofthe RATH train operability analysis.

Note that this is not used in a transient analysis in contrast to the usc of assumption 3.1.tI and thus use of peak temperature conditions is appropriate.

'Re following assumptions)input data applicable to the radiological analysis were used.

These assumptions are deemed to be conservative, bounding, anil consistcn( with design basis accident assumptions identified in Regulatory Guides {c.g,, Regulatory Guide 1.3, 125 ctc.),

3.2.2

, Based upon the assumed sequence of events identified above, the iodine activity contribution to the refueling water volume &om the Unit 2 fuel and fuel pool water willbe negligible compared to the activity from Unit 1.

Since the Unit 2 shutdown occurs 100 days before the Unit 1 shutdown, and since the longest-lived iodine isotope is I-131 with a half-life of 8.05 days, the iodine activity source terms &om the Unit 2 outage willbe negligible due to radioactive decay.

Thus, only the most recent unit refueling activity source terms (defined as Unit 1) are considered for thc dose analysis.

Activity leakage &om the spent fuel after shutdown is based on the full power design fhc1 Icakngc rate decayed for the time after shutdown.

The following leakage rates are used and are based on a design offgas release rate of 100,000 pCi/scc ofnoble gases aAer 30 minutes decay:

Isotope Fuel Design Leakage Rate AtShutdown (4Ci/sec)

I-131 I-132 I-133 7>>] 34 I-135 700 9400 4900 28,000 7900 Since the temperature of the fuel during spent fuel pool boiling is expected to bc well below reactor operating temperature, the use ofthe above "at power" 1cakagc rates " is considered to be extremely conservauvc.

It is assumed that all of the defective fuel rods in the core are transferred to the spent fuel pool.

Ah/A4/1'RR4 1d: 97 A1A-"

7RFiR PPI Nl IA I TC'F Ki PAhF 11 Attachment to PLAAD3 3.2.3

'Ihe activity released

&om the fuel is assumed to be uniformly mixed in the refueling water volume. Thc refueling wato'olume for fuel activity leakage will change throughout the outage depending on whether the fuel pool and cask pit gates are installed or not.

The reke1ing water volumes used are indicated in the assumed event sequence described above.

When the caskpit gates are opened, activity leakage Bom Unit I spent fuel pool is mixed with the Unit 2 spent fuel pool water. As a result, both fuel pools willcontribute to o8site dose once boiling beglnse 3.2,4 Credit is taken for activity reduction in the reiiieling water due to operation ofthe Reactor Water Cleanup and Fuel Pool Cooling and Cleanup systems.

A dccontamnation factor of 10 fax the RWCU and SFPCACU deminexalizers is assuiued for iodincs, 3.2.5 An iodine spiking factor nf 100 fnt spent fuel leakage is assumed to result Rom boiling of the spent fuel pool.

It is conservatively assumed that the activity concentration in the refueliiig water at the time fuel pool boiling begins instantaneously increases by a factor of 100.

3.2.6 The activity release rate &om the pool depends on the evaporation rate and the iodine carryover fraction at the pool surfhcc, Mc evaporation rate during heatup ofthe spent fuel pool water (after loss of cooling but prior to boiling) is bounded by the evaporation rate at initiation of boiling. It is conservatively assumed that thc evaporation rate during heat up is the same as that during boiling.

The evaporation rate at the start of boiling is conservatively assumed for all subsequent time periods during which boiling occurs.

No credit is taken for a reduction in the evaporation rate after boiling begins due to decreases in thc spent fuel decay heat rate.

3.2.7 The iodine canyovcr Rom pool surface to steam is assumed to be Z%; this is the design basis iodine carryover assumed for reactor coolant to steam at full power operation.

3.2,8 The activity released Born the spent fuel pool water is assumed to be released immediately to the environment. No credit is taken in the refueling area for iodine platcout on walls and equipment or for washout by condensing water vapor, No credit is taken for radioactive decay of turbomc activity due to holdup in the zefbeling area, An iodine filter efficiency of 99'/o is assumed fox the SGTS filter forthe time it is in operation.

asja~/is~d 1d:'77 fi1A-7 79sA PPI NIIC.

I TCFNRT C~

PAAF 1'7 Attachment to PLAA133 3.2.9 The start of this event is assumed to be when fuel pool cooling is lost (sequence day 135). The dose is determined for as long as thirty days after the event occurs (sequence day.165).

Atmospheric dispersion factors and breathing rates used in the accident method analysis are the same as used in FSAR Chapter 15 and are as follows:

Post-Accident Time Period (hrs)

Breathing, Rate (m~/sec)

Atmospheric Dispersion Factor (sec/m3) 0-8 8-24 24-96 96 - 720 3.47-04 1.7544 2.32.04 2.32-04 2,18-0$

2.8246 1A346 1.0&46 The reactor recirculation fans are turned offsometime prior to boiling, SGTS is restarted.

3.3,2 3.3.3 Condensation occurs on the refheling Qoor, walls, and rouE only. Tins is a conservative assumption since equipment in Zone IIIwould provide surface area on which condensation would also occur.

'I Zone I air temperature (air outside of the SGTS duct) is assumed to be 80'F and remains constant for the duration for the duct condensation calculation, The air is stagnant due to the LOOP vkich causes the loss of the HVAC system.

This is considered to be a conservative value as summer conditions have been assumed wl111e SOS Scheduled outages occur in the. spxins and fa11.

The temperature would more likelybe on the order of90-100'F since Zone I cooling has been lost and since thc duct are located high in the building whcrc the temperature would bc the highest.

3.3A Condeilsatlon which wcts the we113 of tho duct willnot impact the condensation rate.

This is a conservative assumption since the water film would add a Qlin codBcient term resisting heat loss.

3.3.5 To be conservative, the second pool is assumed to begin boiling 5 hours5.787037e-5 days <br />0.00139 hours <br />8.267196e-6 weeks <br />1.9025e-6 months <br /> dtex the first rather than 17 hours1.967593e-4 days <br />0.00472 hours <br />2.810847e-5 weeks <br />6.4685e-6 months <br /> es dcoeribcd in the sequenoe.

T

AR/Ad/1 qRd 1 d: 97 lilA-7 7'RfiA PPI Nl k; l TCFN.

PACiF Attachment to PLAA133 4,0 METHOD The SPPBOIL computer code was used to evaluate offsite doses for this event and is the same methodology Qmt is used to evaluate the fuel pool boiling scenario currently in Appendix 9A of the PSAR. The SFPBOIL computer code is a single node activity balance code which evaluates activity releases and thc resulting offsitc doses for a boiHng spent fuel pool, The code is capable ofevaluating activity production due to spent fuel leakage and activity losses due to radioactive decay, cleanup and water boil-oK The input data for the SFPBOIL program is based on the data provided in the timing sequcucc of events given in Section'3.1.3 of this rcport, A separate, analysis is performed for the Unit 1 and Unit 2 spent fuel pools since each pool will have different activity levels after Rel pool cooling is lost. Cases were evaluated with and without the RWCU system operating in determining the activity source terms in the ~pent fuel pool water at the time boiling begins.

This analysis did not address in-plant doses and the accessibility of equipment located in the reactor building. Such an evaluation is not necessary since thc in-plant doses &om this scenario would not provont personnel access to the reactor buildings.

The results ofthis analysis were compared to the dose limits of 10CFR100 and SBP 9.4.2.

Calculation ofthe reactor building, control structure and refueling floor conditions during a dual fuel pool boiling event was accomplished by use of the Compartment Transient Temperature Analysis program (COTTAP) code, The average refuel floor temperature is calculated as described in rcfcrcnce 10.

Condensation modeling was calculated utilizing the Uchida Vorrelalloll.

Reactor building (reference 25) and control structure (reference 26) temperature conditions were evaluated for impact on the equipment therein.

Cooling affects of inieakagc on the refueling Qoor was e;cvuutul for in a conservative manner.

In the recirculation plenum, Zone IIand Zone IIIairstreams are mixed and the resulting airstrctun properties are those assumed to enter the SGTS suction duct, Total flowinto thc SOTS duct is the sum of the Zone II and illinleakage and the volumetric flow occurring offof the boiling pools minus the rainout and condensation rate occurring on the refueling floor.

AR/A4/1994 14:'P7 61A-7 7'9AR PPI I4 Ir, I TCÃN9 (i

PARF 14 Attachment to PLA4133 A significant amount of moisture will accumulate in the recirculation plenum.

The amount of moisture that accumulates in the plenum was determined.

The rainout occurs due to the miiCing of the warm meist rcfheling floo auetream and the relatively cool Zone II airstream.

Any condensation which may occur on the plenum walls would be relatively insignificant as compared to the rainout.

Moisture accumulation in the SGTS ductwork is also of concern.

The source of moisture is condensation and to some cxtcnt mist.

Moisture accumulation in the reactor building sump as a result of the collection of the cond~ion ncciirring nn the refueling floorwas also assessed.

1 To assess the impact on SGTS, duct drawings werc reviewed to determine the duct location(s) in which duct condensation mill collect.

These were conRnned by a plant walkdown. Volumes of the low point duct sections were determined. Ae rate ofcondensation collection was calculated (as described above) and correlated to a duct low section volunielriu tdl rate. Duct water level as a functiori oftime was then determined.

Duct structural capability was reviewed and compared to the condensation weight. An assessment ofimpact ofcondensation on duct prcssure drop was also made, With the above developed information, an assessment ofimpact on SGTS was made.

Thc filtertrain conditions werc evaluated and assessed for impact on SGTS operation.

Environmental Qualification Assessment Reports (EQARs) are used te identify the 100My post-accident room temperature limits for Class lE equipment. Ifa room temperature during fuel pool boiling during a seismic event scenario does not exceed the maximum room temperature used in the EQAR, the cquiymcnt in that room is assumed to bc operable for at least 100 days.

Previous analyses have shown that aH CIass 1E equipment is opemble for at least 30 days following a fuel pool boiling with a LOCA/LOOP. Ifa room tempcraturc for fuel pool boiling during a seismic event scenario docs not exceed the room temperature used in these previous analyses, the equipment in that room is considered to be opcrablc for at least 30 days.

s.0 RESULTS This evaluation is contained in reference 19, Activityconcentrations in the spent fuel pool water at the time lms offuel pool cooling occurs are as follows:

PPI Nlk. l Tl'FN.

6 PACiF

'l 5 Attachment to PLA4133 Isotope ActivityConcentration In Spent Fuel Pool Water AtThc Time OfLoss OfFuel Pool Cooling (CLtm3)

I-131 I-132 1-133 I-134 I-135 RNCU Operating 8.2644 0.

8,31-14 0,

0.

RWCU Not Operating 1.1043 0.

1.12-13 0.

0.

lt can be seen &om the above results that operation of the R%'CU system results in a small decrease in the activity levels in the spent fuel pool @ken loss offuel pool cooling occurs.

OfBite doses were evaluated for activity releases Rom each pool and fox various SGTS operating times ranging Bom no operation to operation for7 days after fuel pool boiling begins.

Results of these calculations are sutnmarized in the followingtable, 30 Bay Thy1'Old InhahtiOa DOSe AtLoW PopulatiOn Zone (Rem)

SGTS OPERhTlNG TIME (FRoM KYENT 0CCLRRENCE)

DOSE DUETo garxruxs FROM UNrr 2 pooL Dosz DUETo RELEhSKS PROM UNrr 1 PooL 35 Hours (Start of Unit 1

Fuel Pool Boiling) 51.7 Hours (Rccirc.

Plenum Limit Reached) 30 Days S.64-E06 3.7-H6

'3.71-F06 8.64-808 2.96-E03 2.95-803 2.73-E03 2.96>>EOS 2.97-203 2,95-803 2.73-E03 2.97-E05 Note that these results exhibit that thc non-outage pool {second pool to boil) effectively provides no contribution to ot?site dose.

-11 <<

PA(iF 1 Fi Attachment to PLAA1M Regulatory doso limits for this event are as folio@+:

HLCEEl9Q (Design Basis Accidents)

Site Boundary(2 hours2.314815e-5 days <br />5.555556e-4 hours <br />3.306878e-6 weeks <br />7.61e-7 months <br />):

25 Rem whole Body 300 Rem Thyroid Inhalation Low Population Zone(30 day):

25 Rem Whole Body 300 Rem 'Ihyroid Inhalation (Fuel Pool Area Ventilation Systems)

Allportions ofthe systems whose failure may result in the release ofradioactivity which causes an of&ite doe'. of mnre than 0.5 Rem to the whole body or its equivalent to any part ofthe body shall be classified seismic Category I and safety related.

As can be scen &omthe above table, the doses are well withinthe regulatory Hmits.

5.2.1 The HVACanalysis assumes the followinginitial zone temperatures:

Zone I 80'F Zone II 100-104'F Zone IO

.100'F The Zone II and IIItemperatures are based on normal operating maximum temperatures.

These initial cuiulilivns arc assumed to exist fmm initial isolahon and aligmncnt of Zones II and IIIto the recirculation plenum with the recirculation fans on until shutdown of the recirculation fans sometime prior to boiling, Once boiling of the first pool begins, the conditions calculated are as tabulated below. Zone I cnnservativcly willremain at 80 F since it is isolated &om the other two zones.

0,

Ah/A4/1 RR4 1 4: 97 Fi1A-774-7c1fiR F'PI Nl II.

I 1CFN5"fN(i PdhF 17 Attachment to PK A4133 TIMEAFTER BOILINGBEGINS (HOIJRS)

ZONE III ZONE li RECIRC PLENUM 127 110 123 10 157'47 20 177 110 30 187 110 173 19n 110 175 260 198 120 The above results are determined in references 11, 20 and 21. Note that the conditions corresponding to 4, 30 and 36 hours4.166667e-4 days <br />0.01 hours <br />5.952381e-5 weeks <br />1.3698e-5 months <br /> were utilized as input to the SGTS duct and train evaluations.

5,2.2 Water accumulation in the recirculation plenum was determined.

The water accumulated is primarily due to rainout. Rainout is the difference in mass ofwater at supersatumted conditions and saturated conditions. Thc SOTS duct and Zone I and ll supply duct come up through the bottom of the recirculation plenum, A 4 inch lip exists which will keep the accumulated water in thc plenum until the plenum water lcvcl reaches 4 inches. Using thc 4 inch level, it is determined in reference 21 that it willtake 16.7 hours8.101852e-5 days <br />0.00194 hours <br />1.157407e-5 weeks <br />2.6635e-6 months <br /> after the first pool begins to boil for 4 inches to accumulate in the plenum, thus 35 +16.7 - 51.7 hours8.101852e-5 days <br />0.00194 hours <br />1.157407e-5 weeks <br />2.6635e-6 months <br /> after event occurrcIlcc.

5,2,3 Based on reference 22, the temperatures used in reference 15 for a LOCA/LOOP envelop thc temperatures for the seismic event, with the exception ofthe refueling floor(Room 1-810). The maximum temperature on the refueling floor is 197,63'F for a seismic

event, compared to 171 1'or a LOt.'A/LOOP.

'l'hc higher temperature reduces the post-accident qualification for prcssure diQercntial transmitters FDT07554A1, A3, Bl, 83 &om 100 days to 30 days, pcr reference 16.

The qualified life for thc PDTs is 40 years at 100'F for both post-accident temperatures.

PARF lR Attachment to PLA4133 The Functional Evaluation of the PDTs was discussed in reference

15. No other equipment was affected by higher temperatures from boiling fuel pools caused by u scisoiic cvcnt.

5,2.4 Reference 11 determines the condensation rates occurring on the refueling floor.

The condensation rate varies with time during the event due to thc heat up ofthe floor airspace and subsequent increase in temperature ofthc condensation surface.

The maximuni condensation rate occurs when thc rcfucling floor airspace temperature is 157'F. Thc maximum condensation rate is 8600 ibm/hr.

~ i Reference 17 evaluates condensate drainage during fuel pool boiling scenarios.

For a given refueling Qoor condensation rate, condensation collection rate in the reactor building sumps anti subsequent ovcrQow tu adjoiiiing areas is assessed, lt determines the height of water and thus time for a given collection rate in which the water tight doors preventing spillover to adjacent rooms would be reached.

Thus, to assess the impact of this event with tbe shove refi]cling Rnor condensation rate, reference 17 methods willbe used, 157'F;. 'v =.01638 fP/ibm per reference*18 8600 ibm/hr (.01638 83/ibm) (lhr/60min)(7,48 gal/ ft>)- 17.56 gpm This is equivalent to the "boiljtngrate" in reference 11-thu's entering the chart in thc reference identifies that. itwilltake.45 days in Unit 1 and 38 days in Unit 2 to reach the water tight door limit. Note that reference 11 assumed all ofthc boiloE

&om the pool(s) was available to condense whereas this evaluation is based only on.the portion, ofthc boiiofFwhich condenses on the refueling floor,

• i 'i "i

i n.

Based on the above, it.is-deemed that condensation collection in the reactor building sump area willbe contained such that itwillhave no impact on the plant.

II

ii/i i

~

r \\

~ ~ q The event creates both high temperature and high moisture conditions in the St 'TS suction duct, thus both these canditions are addressed.

-Vi The ducting contains six fire dampcrs.

'Ihcsc &c dainpcrs are rated for 285'F.

Since thc duct temperatures do not approach this value as exhibited in section 5.2.1 above, the dampcrs willnot adversely affect SOTS operation.

The SGTS duct run from thc recirculation plenum to the control structure is located in Zone I (Unit 1 reactor buildiiig) aud Zonc1II (refueling Qoor).

Three low sections exist in the duoting where duct condensation will collect. All three 1ow sections are located in Zone!.

Of these three, one is limitingin that itwillfillwithcondensation faster than thc other two. '1'his limiting duct inw section was evaluated, The evaluation is contained in reference 5.

Ah/Ad/1 994 1 4: '77 A1 A-7

-7'9AR pp~

w e t res'Ns wh PACK 1 9 Os Attachment to PLAA133 Reference 5 determines that the limiting section would not adversely affec SGTS for minimally 36 hours4.166667e-4 days <br />0.01 hours <br />5.952381e-5 weeks <br />1.3698e-5 months <br /> after the Grat pool begins to boil, thus minimally 71 hours8.217593e-4 days <br />0.0197 hours <br />1.173942e-4 weeks <br />2.70155e-5 months <br /> a8er the occurrettco nf the event.

The event creates both high temperature and high moisture content conditions at the train inlet, Thus these both are addressed.

AC 4

~

~

II Concerning the high moisture content, condensation collection inthe SGTS train willaccumulate until the loop seal is overcome at which time the condensation'ill begin to drain.to the liquid radwaste system.

The height of water in thc train at thc point when the loop seal is overcome willnot adversely affect thc SGTS train operation as documented in reference 6, The SGTS train contains.filter's, heat'crs and a charcoal bed',,Thc charcoal bed and heater section contain temperature sensors.

The heater section temperat1ue control circuit maintains iconstant differential temperature across the heaters of20'P.

This assures acceptable air. stream humidity entering the charcoal bed, thus assuring effectivenes of.'the charcoal bed.

-:The heater also contains a thermal cutout set for 250'F (reference 7). An evaluation ofthis sctpoint is contained in reference

8. The charcoal bcd tcmpcrature circuit contains a high temperature inlet alarm sct at 220'F.

A high temperature set is.also provided which is sct for 410'F which willtrip the SGTS fan in case of a dercnal bed Rre.

The airstream temperature approaches 127 F as dctcrmincd in reference 5, thus thc setpoints will-not be'pproached.

Note that only thc high high temperature set at 410'P willshutdown'he SOTS fan..

~

~ IQ

~ I It is thus concluded that the airstream conditions entering the SGTS train during this event will not adversely affect the operation of the train as temperature limits and charcoal bed relative humidity limitswillnot be exceeded.

lE ~

~

Ige'>khl

'w

(

The COTTAP room temperatures for the. fuc1 pool boiling durin~seismic.event.scenari was dctemomd.

Atthe time in which thc recirculation plenum would'each the analyzed limitof,51 hours5.902778e-4 days <br />0.0142 hours <br />8.43254e-5 weeks <br />1.94055e-5 months <br /> after event occurrence, the SUI'5 room temperature is determined in reference 2L to be less than 104'F. This temperature is less than the 104'F maximum normal temperature for the room.

Thus, the equipment would function as designed for as long as required to support SGTS operation.

6,0 SUMMARX Evaluations of the SSES plant response to a postulated dual fuel pool boil event assumed to be caused by a seismic event have been performed.

A LOOP is assumed coincident with the seismic event.

The offsitc dose consequences, impact on the operability of SGTS; impact on reactor duiiding equipment and impact on the control structure equipment have been assessed, AFi/A4/'I994

'l4:97 AlA-774-79FiA

~

A

)

'A gaia Attachment to PLAA133 The calculated oQsite doses resulting from this are well within the Accident Dose limits of both 10CFR100 and SRP 9,4,2 mthout tahing credit for SOTS operation during 8e event. Various SGTS operatin8 durations are evaluated. ',

The recirculation plenum water accumulation rate willcause the plenum uwlnnulatioa to begin spilling over into plenum connected ducting, including the SGTS duct, before the SGTS duct condensation reaches its analyzed limit, 'Thc spillover occurs 51 hours5.902778e-4 days <br />0.0142 hours <br />8.43254e-5 weeks <br />1.94055e-5 months <br /> after event occurrence and rcprcscnts the analyzed limit.

The water accumulation rate for the limiting SCiTS duct section is dctcrmined to reach the analyzed limit 36 hours4.166667e-4 days <br />0.01 hours <br />5.952381e-5 weeks <br />1.3698e-5 months <br /> after start of pool boiling, and 71 hours8.217593e-4 days <br />0.0197 hours <br />1.173942e-4 weeks <br />2.70155e-5 months <br /> (approximately 3 days) aAcr event occueence.

The reactor building conditions created by the assumed event only cause equipment located on the refueling floor area to exceed the previously evaluated conditions.

The affected equipment has been evaluated and determined to be capable ofperforming as designed during this event.

The SOTS route Lclnperaturcs arc determined to be within the. mmdmum normal operating temperature ofthe SOTS room at the time in which the recirculation plenum reaches the analyzed

limit,

8S/8a/188' 4: 97 81 A-7

-7'RFi8 PPI Nl IO I TCFN.

C PACER O'I Attachment to PLAP133

REFERENCES:

l.

2.

3.

4.

5, 6,

'7.

8.

9.

10.

11.

12.

13.

14, 15.

16.

17.

18.

19.

20.

21.

22.

23.

24.

25.

25.

26.

~,

~

i M-FPC-010 RO 8c M-FPC-014 Rl (Pool VolInnes)

OP-135-001 K l6 (Fuel Pool Cooling)

NFE-8-NA-055 Rl (Heat Loads)

M-FPC-013 RO (Fuel Pool Cooling Heat Exchanger Capacity)

DC-070-1003 Rl NCR 89-0427 SEIS BWR 83-0076 PLA4128 PLA<112 EC-035-1001 Rl not used not used OT-93-DAM-004 RO a w

~I EC-EQQL&525 Rl EC-EQQL-1003 Rl EC-035-0510 Rl ASEGUQ? PUNDAMFXTAT$,1980 BC-RADN-1007 Rl EC-070-1002 Rl

'I EC-034-1003 Rl M-FPC-015 Rl EC-EQQL-1007 RO not used BC-035-0513 R1 EC-070-1004 RO EC-034-1004 RO PARF Attachment to PLA4133 OILOFF RATE Reference 3;

(Note that this <<<<l<<ul<<tion of decay heat is based on a full fuel pool with all bundles operated under power uprate conditions for 18 months continuously.)

Unit 1; 35 days after cold shutdown assuming Nominal decay heat 8.56 MBTU/HR Unit 2; 135 days <<Ger co1d shutdown sssIImiog Nnminal decay heat 5.82 MBTU/HR Keference 1

Unit 1; T 3011131~(210- 110)/8.56E6 = 35 hours4.050926e-4 days <br />0.00972 hours <br />5.787037e-5 weeks <br />1.33175e-5 months <br /> Unit2; T ~ 3011131 "(210- 110)/5,82E6 = 52 hours6.018519e-4 days <br />0.0144 hours <br />8.597884e-5 weeks <br />1.9786e-5 months <br /> Unit 1; M 8.56K&971.6 ~ 8810 lbm/hr Unit2; M= 5.82E6/971.6 ~ 5990 Ibm/hr (1)

Values and methods <<ro from tho roferenoos a<<notod.

Q)

Initialpoo1 temperature of 110'F ls assumed. This is appruprl<<tc <<iud OIo t'u<<1 lIwlcovlins system can maintain the pools at 110'F even with the design maximum service water temperature of95'F, as scen from the results of reference 4

. Pool temperature assuming normal service water temperatures would be less thus lengthening the time to boil.

(3)

No credit is taken for the decay ofthe beat ln<<d with time in the radiological analysis but is taken credit for in the HVACanalyses, (This represents a conservatism, however, very slight.)

(4)

Time to boil calculation above is conservative as it assumes that all the heat goes to the water.

(5)

Hollow rates determined above were used for the radiological analysis, (6)

The HVACanalysis takes credit for cooling provided by ESW and takes credit for heat load decay with time. Because credit was taken for the above it also assumes the maximum decay heat instead ofthe nominal values used forthe radiological evaluation.