ML20086S622

From kanterella
Jump to navigation Jump to search
Proposed Tech Specs,Allowing Use of Credit for Soluble B in Spent Fuel Pool Criticality Analyses & Relocation of Spent Fuel Pool Operating Limits to Unit 1 COLR
ML20086S622
Person / Time
Site: Prairie Island  Xcel Energy icon.png
Issue date: 07/28/1995
From:
NORTHERN STATES POWER CO.
To:
Shared Package
ML20086S606 List:
References
NUDOCS 9508020067
Download: ML20086S622 (94)


Text

+ .

,e

~

TS.1-2 REV 111 S/10/9'-

CONTAINMENT INTEGRITY CONTAINMENT INTEGRITY shall exist when:

1. Penetrations required to be isolated during accident conditions are either:
a. Capable of being closed by an OPERABLE containment automatic isolation valve system, or
b. Closed by manual valves, blind flanges, or deactivated automatic valves secured in their closed positions, except as provided in Specifications 3.6.C and 3.6.D.
2. The equipment hatch is closed and sealed
3. Each air lock is in compliance with the requirements of Specification 3.6.M.
4. The containment leakage rates are within their required limits.

CORE ALTERATION CORE ALTERATION is the movement or manipulation of any component within the reactor pressure vessel with the vessel head removed and fuel in the vessel, which may affect core reactivity. Suspension of CORE ALTERATION shall not preclude completion of movement of a component to a safe conservative l position. I CORE OPERATING LIMITS REPORT The CORE OPERATING LIMITS REPORT is the unit-specific document that provides core operating limits for the current operating reload cycle. These cycle-specific core operating limits shall be determined for each reload cycle in accordance with Specification 6.7.A.6. The} Unit M CORELOPERATING3 LIMITS REPORT {alsojontains "theispentifue1[ poolloperat inflihi'ts("Plantf operation within these corefandfspentifuelfpooll:l operating limits is addressed in individual specifications.

950B020067 950728 PDR ADOCK 05000282 p PDR

[o ..

o

, TS.3.8-4 REV 109 9/3/93 3.8.C. Small Spent Fuel Pool Restrictions No more than 45 recently discharged assemblies shall be located in the small pool (pool No. 1).

D. Spent Fuel Pool Special Ventilation System

1. Both trains of the Spent Fuel Pool Special Ventilation System shall be OPERABLE at all times (except as specified in 3.8.D.2 and 3.8.D.3 below).
2. With one train of the Spent Fuel Pool Special Ventilation System inoperable, fuel handling operations and crane operations with loads over spent fuel (inside the spent fuel pool enclosure) are permissible during the following 7 days, provided the redundant train is demonstrated OPERABLE prior to proceeding with those operations.
3. With both trains of the Spent Fuel Pool Special Ventilation System inoperable, suspend all fuel handling operations and crane operations with loads over spent fuel (inside the spent fuel pool enclosure).
4. The provisions of specification 3.0.C are not applicable.

E. Spent Fuel Pool Storane

1. Fuel Assembly Storage
Tc be etered .14heut rectrictier 1r the spent fuci peel, the burnup nd initi:1 enrich ent cf a fuci ecccmbly ch:11 he eithir the unrectricted r:nge cf rigure TS.3.8 1.

ab. Fuel assemblies with ec=binatier cf burnup and ! :it! 1 er-ichment ir the rectricted range of Figure TS.3.8 1 shall be stored in accordance with Specif!catier 5.E . ^ .1.d thd configurations;specifisdlinitM1UnitillCORE[0PERATING;LlMITS REPORT.

yo. If the requirements of 3.8.E.1.a and 3.8.E.1.b are not met, immediately initiate action to move any noncomplying fuel assembly to an acceptable location.

44 The provisions of Specification 3.0.C are not applicable.

14 ,.

TS.3.8-5

.- *: o.,','o_

RF.'.V ' _ _ _ _ _

3.8.E.2. Spent Fuel Pool Boron Concentration .

1

a. The spent fuel pool boron concentration shall be ssistsihsd -t

. , - . . . ~ . n C

i,itt. hiiE. .+theWiki. t_i?.hp,s_Ei,f.is. d!16T,l%i.~itidiE^Eu.OR,KT.O..P.ERATI.

N. d~ ~ '

LI_MIT..$..i_ REP.O_R~T..' 1,900 pp: --her fuel 2:renh11er uith

_ .- i cenhinctier af burnup cnd initial enri:Fnent ir th: rectrfeted

ng: ef Figur: TS.3.9 1 cre etered in the pent fub1'perl end
cp:nt' fuel perl ferificatier h net beer perferned eine:

the lect n:z: :nt rf any fuel : enhly ir the ': pent fuel p: 1.  :

b. If the re ;uir:nente ef rpreificatier 3.9.E.2.2 :s 2pp1!ceb1:~ l nd the : pent fuel peel berer concentretier ! net :ithir it -

'* IfIthETspunt$fddlfp66lisbof6Hi!E6hE6WWWEish3FIhotW1this thsT. Hm1Isd$ecif

- - - 6.df._is.t_M,U..n.i.

- - - a h~1(COR.E10PE_RA_TI.

,. . .nn Nb..l!_ LIM. I.,Td.,"~~~~~

i REPORT, then immediatel.ay: i

1. Suspend movement of fuel assemblies in the spent fuel '!

pool, and i

r

2. Father itnitiate action to restore spent fuel pool b_oron i concentration to within-its limit er perfer= 2 rpent fuel y_ml ferificctier.

I

c. The provisions of Specification 3.0.C are not applicable.

.4 5

l 5

h E

f 4

.k i

l 4 -

FIGURE TS.3,8-1 REV 108 9/3/93 12000 s

f_

1 11 , ,#

t . 1, s 1 i s I /

l 10000 s s 1 f - /.

1 1 i

% RESTRICTED w 1

/ / i <

s 1 - i

^ l .F /

\

3000

,# e

/

r 1 i P s i e

$ s

- s ,J' ,/

m 7000 'ss I /

f /

l ' I /

- r 1 /

g g  %

s I/

  1. /

o

3. / /

3' & if l 502 DELETE f

- 4000 13

/f'

'/

't w

$ If s f/ 't m u' RFSTkTED l s 1

  1. / 't
  1. / \

is s 2000 e

f' s )

s I i<

r t 1000 f

e 's  %,

i 6 I s g r g 3.5 4.0 4.5 5.0 )

INITIAL NOMINAL U-235 ENRICIDCENT (w/o) i FIGURE TS.3.8-1 Spent Fuel Fool Unrestricted Region Minimum Burnup Requirements

, e.

  • e Table TS.4.1-2B (Page 1 of 2)

REV 108 9/3/93 TABLE TS.4.1-2B MINIMUM FREOUENCIES FOR SAMPLING TESTS TEST FREQUENCY

1. RCS Gross 5/ week Activity Determination
2. RCS Isotopic Analysis for DOSE 1/14 days (when at power)

EQUIVALENT I-131 Concentration

3. RCS Radiochemistry E determination 1/6 months (l) (when at power)
4. RCS Isotopic Analysis for Iodine a) Once per 4 hours4.62963e-5 days <br />0.00111 hours <br />6.613757e-6 weeks <br />1.522e-6 months <br />, whenever Including I-131, I-133, and I-135 the specific activity ex-ceeds 1.0 uCi/ gram DOSE _

EQUIVALENT I-131 or 100/E uCi/ gram (at or above cold shutdown), and b) One sample between 2 and 6 hours6.944444e-5 days <br />0.00167 hours <br />9.920635e-6 weeks <br />2.283e-6 months <br /> following THERMAL POWER change exceeding 15 percent of the RATED THERMAL POWER within a one hour i period (above hot shutdown)

5. RCS Radiochemistry (2) Monthly
6. RCS Tritium Activity Weekly
7. RCS Chemistry (Cl*,F*, 02) 5/ Week
8. RCS Boron Concentration *(3) 2/ Week (4)
9. RWST Boron Concentration Weekly
10. Boric Acid Tanks Boron Concentration 2/ Week
11. Caustic Standpipe NaOH Concentration Monthly
12. Accumulator Boron Concentration Monthly
13. Spent Fuel Pit Boron Concentration Menthly/Ueekly """

I I

L

I

. Table TS.4.1-2B (Page 2 of 2)

REV 10B 9/1/93 TABLE TS.4.1-2B MINIMUM FREOUENCIES FOR SAMPLING TESTS TEST FREOUENCY

14. Secondary Coolant Gross Weekly i Beta-Gamma activity
15. Secondary Coolant Isotopic 1/6 months (5)

Analysis for DOSE EQUIVALENT I-131 concentration

16. Secondary Coolant Chemistry pH 5/ week (6) pH Control Additive 5/ week (6)

Sodium 5/ week (6)

Notes:

1. Sample to be taken after a minimum of 2 EFPD and 20 days of POWER OPERATION have elapsed since reactor was last suberitical for 48 hour5.555556e-4 days <br />0.0133 hours <br />7.936508e-5 weeks <br />1.8264e-5 months <br />s-or longer.
2. To determine activity of corrosion products having a half-life greater than 30 minutes.
3. During REFUELING, the boron concentration shall be verified by chemical analysis daily.
4. The maximum interval between analyses shall not exceed 5 days.
5. If activity of the samples is greater than 10% of the limit in Specification 3.4.D, the frequency shall be once per month.
6. The maximum interval between analyses shall not exceed 3 days.

T' minirur spent fuel perl herer concentretier fre: Specific tier 3.8.B.I .h ch:11 he verified by che-!cci n:1 :Ir 7 rechly 9hil  : pent fuel cec' certaining fuel is lec ted ir the rpent fuel p::1 S. The spent fucI peel here cencentretier ch:11 he verified . cchly, by cherie:1 an 1 7:!:, te he eithir the li=it ef Specific: tier 3.9.E.2.2

^e- fuel nerr:blier with cenhinctier Of burnup and initial enrichment ir t'^ rectricted r:nge of Figure TS.3.8-1 cre etered ir the : pent fuel peel nd 2 rpent fuel perl verificatier her net beer perfer=:d cine the Inst reverrnt cf any fuel 22:2 51y ir th: : pent fuel peel.

. . l i

m. a h TS.S.6-1 H *-

f REV 108 9/3/93 ,

5.6 FUEL HANDLING ,

I- Criticality Consideration A.

1. The spent fuel storage racks are designed (Reference 1) and shall be ,

maintained with:

a. Fuel assemblies having a maximum U-235 enrichment of 5.0 weight percent;
b. K o rg s 0.95 if fully flooded with unborated water, which includes -f an allowance for uncertainties as described in Reference 3}; phd j
c. undW Mikimus K,gg"Cl?071f?fu11 lsinfdondif Jf166deditritEssburstsdNsts@EEGi~EpEdi?u:

sides 1 fe aA3_3_ 3_ 3_

gisnsM.g.g.h

_ _ g fi. 6Md d.i.nIR$fArenedE3 33 3,, i ,

_ n_ _ _ . . r - - - - - - - - - - - - - - - - - - - - - - - - - -

the unrectricted rang: ef Figure TS.3."-1 211eued unrectricted eter ge ir th: : pent fuel rechr; and ,

d. M .. er spent fec1 2:reshlice eith cebbinctier f burnup and  ;

initin1 enrich :. !- the rectricted rnnge cf Figure TS.3.2 1 ctered ir eenpliance eith Figure: TS.5.5 1 and TS.5.5-2.

2. The new fuel storage racks are designed (Reference 1) and shall be maintained with: ,

4

a. Fuel assemblies having a maximum U-235 enrichment of 5.0 weight percent;  ;
b. K e rr s 0.95 if fully flooded with unborated water, which includes an allowance for uncertainties as described in Reference 2; and I
c. K.tr s 0.98 if accidentally filled with a low density moderator which resulted in optimum low density moderation conditions.
3. Fuel will not be inserted into a spent fuel cask in the~ pool, unless a 1 minimum boron concentration of 1800 ppm is present. The 1800 ppm will j ensure that k o rt for the spent fuel cask, including statistical i uncertainties, will be less than or equal to 0.95 for all postulated arrangements of fuel within the cask. The criticality analysis for ,

the TN-40 spent fuel storage cask was based on fresh fuel enriched to 3.85 weight percent U-235. ,

B. Spent Fuel Storace Structure i

The spent fuel storage pool is enclosed with a reinforced concrete building having 12- to 18-inch thick walls and roof (Reference 1). l The pool and pool enclosure are Class I (seismic) structures that afford protection against loss of integrity from postulated tornado missiles. The storage compartments and the fuel transfer canal are j connected by fuel transfer slots that can be closed off with i pneumatically sealed gates. The bottoms of the slots are above the ]

tops of the active fuel in the fuel assemblies which will be stored l vertically in specially constructed racks. <

l I

i

)

1 j

1

l . TS.5.6-3 REV 1^o 9/3/93 D. Spent Fuel Storane Capacity The spent fuel storage facility is a two-compartment pool that, if completely filled with fuel storage racks, provides up to 1582 storage locations. The southeast corner of the small pool (pool no. 1) also serves as the cask lay down area. During times when the cask is being used, four racks are removed from the small pool, With the four storage racks in the southeast corner of pool 1 removed, a total of 1386 storage locations are provided. To allow insertion of a spent fuel j cask, total storage is limited to 1386 assemblies, not including those l assemblies which can be returned to the reactor.  !

l I

1 1

1 I

1 Reference

1. USAR, Section 10.2
2. " Criticality Analysis of the Prairie Island Units 1 & 2 Fresh and Spent l Fuel Racks", Westinghouse Commercial Nuclear Fuel Division, February 1993. '
3. WCAPi14416f P; f!"Wes tinghouse! Spenti FueMRack.1CriticalitiKAnalysis. 4 MethodologyJ;Junel1995; j

- FIGURE TS.S.6-1 REV 108 9/3/93 LE E <

E i 1 1 E E.

EE!

lI B LE B

?B i BB El TE PATTERN FOR CHECKERBOARD REC JN

/

imenoc. WK4 FAVA ES FAR W W9 9899 FA

! DELETE IM

.c

^~-redE y m:= beg p-1 i

I BOUNDA . BETWEEN CHECKERBOARD AND UNRES CTED REGIONS Fresh el: Enrichments up to 5.0 w/o U-235, no restric 'ons on burnup f,$j Che .erboard Region is.5.6-2.

B ted Fuel: Must satisfy minian turnup requirements of risur E

/

nrestreited Region Burned Fuel: Must satisfy minian burnup requirements of Figure is.3. -1.

Note: The Checkerboard and unrestricted regions can alternatively e  ;

separated by a single row of vacant cells on each adjacent fa .

FIGURE 75.5.6-1 Spent Fuel Fool Burned /Fres;i checkerboard Cell 1.ayout

, FIGURE TS.S.6-2 REV 108 9/3/93 30000 J

/

/ 2 2m \g

/ /

7 f FBCEPTABLEl / /

\ / /

- \ / /

\ / /

{2m g j f

!~ \ / /

\ / /

d k [ [

/ /

UI \ f f Y l ,

y /\ WIRCCEPTRBLEl l lsm

/ D LETE w / .. ,.

\

/ / \ '

/ / \

/ / \  !

/ / \  !

/ / \

/ / \

/ / \

) f \

/ \

h.[5 - 3.0 3.5 4.0. 4. 5.0 INITIAL NOMINAL U-235 ENRICHMENT (w/o)

FIGURE TS.S.6-2 Spent Fuel Fool Checkerboard Region Minimum Burnup Requirements

m

s. .

TS.6.7-4 REV 93 2/11/91 6.7.A.S. Annual Summaries of Meteorolonical Data An annual summary of meteorological data shall be submitted for the previous calendar year in the form of joint frequency distributions ,

of wind speed, wind direction, and atmospheric stability at the request of the Commission.

6.7.A.6. Core Operating Limits Report ,

a. Core operating limits shall be established and documented in the CORE OPERATING LIMITS REPORT before each reload cycle or any remaining part of a reload cycle for the following:

RM

1. Heat Flux Hot Channel Factor Limit (Fn ), Nuclear Enthalpy RTP Rise Hot Channel Factor Limit (Fag ), PFDH, K(Z) and V(Z) '

(Specifications 3.10.B.1, 3.10.B.2 and 3.10.B.3)

2. Axial Flux Difference Limits and Target Band (Specifications 3.10.B.4 through 3.10.B.9) 1
3. Shutdown and Control Bank Insertion Limits (Specification 3.10.D)
4. Reactor Coolant System Flow Limit  ;

(Specification 3.10.J) 10.*gSpentKfh61lp001(operating?limitpsha1Kb4EsstablFshsldgind documentedtinitheJUnit

" W COREnOPERATINGLL1MITSiREPORTffor

~"~~~~~ ' ~ ^ ~ " ' ~ ~ ~ ~ ~ ~ ~ ~

Ms

^" ~'

folldyi N f ~ ~

igSpsnt)Tue1[Po61tStorags[Confip,hrati6n}Vaitsti6ds ISecificationf318/E(1}

R i 2 nSpshtIFue M P E LBbydd K6H6ep Est16hlkimij JSpecificatign13iBjE/2)  ;

s. The analytical methods used to determine the core operating limits shall be those previously reviewed and approved by the NRC, specifically those described in the following documents:

NSPNAD-8101-A, " Qualification of Reactor Physics Methods for  !

Application to PI Units" (latest approved version)

NSPNAD-8102-A, " Prairie Island Nuclear Power Plant Reload Safety Evaluation Methods for Application to PI Units" (latest approved version)

WCAP-9272-P-A, " Westinghouse Reload Safety Evaluation Methodology", July, 1985 i

WCAP-10054-P-A, " Westinghouse Small Break ECCS Evaluation l Model Using the NOTRUMP Code", August, 1985 I

" ', y.

i; '

f TS.6.7-5 l J ~' REV 109 12/3/91' WCAP-10924-P-A,' " Westinghouse Large-Break LOCA Best-Estimate Methodology", December, 1988 WCAP-10924-P-A, Volume 1,1 Addendum 4, " Westinghouse Large Break LOCA Best Estimate Methodology", August, 1990 XN-NF-77-57 (A), XN-NF-77-57, Supplement l'(A), " Exxon Nuclear Power Distribution Control for Pressurized Water.

Reactors Phase II", May, 1981 WCAP-13677, "10 CFR 50.46 Evaluation Model Report:

E-COBRA / TRAC 2-Loop Upper Plenum Injection Model Updateto

h. Support ZIRL0m Cladding Options", April 1993.(approved by NRC -

SE dated November 26, 1993)

NSPNAD-93003-A, " Transient Power Distribution Methodology",

(latest approved' version)

,dETheannlyt$c'al"Inetho"di"L' s ed,to7deteratWths'is'psnt!" fuel pool operating limits shall be those previously reviewed and approved by the NRC specifically,those, described in the following dogameppa;y WCAP -14416 '- P ;' '*Wes t inghous's " Spent Pue1[ RackCriticali ty N 1 sisjMethodology,*,,

7- June 1995 K. The core {$Jppp@lfpgj[pppjoperating limits shall be determined

, such that all applicable limits (e.g., fuel thermal-mechanical-l limits, core thermal-hydraulic limits, ECCS limits, nuclear limits such as shutdown margin, end-transient and accident analysislimitsyghMy@p{fd]e@gjfijiyE1]yfjllpitjj) of the l safety analysis are met.  !

l

f. The CORE OPERATING LIMITS REPORT, including any.mid-cycle l revisions or supplements thereto, shall be supplied upon H issuance,.for each reload cycle, to the NRC Document Control' Desk with copies to the Regional Administrator _and Resident Inspector.

~i B. REPORTABLE EVENTS  !

The following actions shall be taken for REPORTABLE EVENTS: y

a. The Commission shall be notified by a report submitted pursuant
. to the requirements of Section 50.73 to 10 CFR Part 50, and
b. Each REPORTABLE EVENT shall be. reviewed by the Operations Committee and the results of this review shall be submitted to the Safety Audit Committee and the Vice President Nuclear Generation.

e

  • C' B.3.8-2 REV 409 9'3'93 f I 3.8 REFUELING AND FUEL HANDLING Bases continued During movement of irradiated fuel assemblies or control rods, a water level of 23 feet is maintained to provide sufficient shielding.

The water level may be lowered to the top of the RCCA drive shafts for latching and unlatching. The water level may also be lowered below 20 feet for upper internals removal / replacement. The basis for these allowance (s) are (1) the refueling cavity pool has sufficient level to allow time to initiate repairs or emergency procedures to cool the core, (2) during latching / unlatching and upper internals removal / replace-ment the level is closely monitored because the activity uses this level as a reference point, (3) the time spent at this level is minimal.

The Prairie Island spent fuel storage racks have been analyzed id?ac(6tdanch e, ith_?!theU. iiethodo,l6. fy3b..h. ts.i.,ned,jid, s% Reference $4} to allow for the storage of

. , .._ m _ . _ . _ -

fuel assemblies with enrichments up to 5.0 weight percent U-235 while maintaining Ko rt s 0.95 including uncertaintiesiAhdydditsf6Ms61Ub1Mbbr6h.

INSERT

,.u...

..-s, vj -.....j-.

-~-.-......,....,..vaa--....-u..- - c s. ,. ---.. . . . g-

.v -.---

.. --. gv vv...c a. p. . . . . .

r. v . . ,. -

v.

regien te encure that the cpent fuel peel vill rc=cir cuberiticc1 during the cterage of fuel acce=blice ith all peccibic ccmbinctienc cf burrup and initiel enrichment:

1 The first regien utilice a checkerbecrd Iceding patterr tc acec==cdcte neu

-. , . 1...r ...g

. .-- . .. , .c.. a. .s u.

. .. . ...s.... - . . . .s u. . . . nc e.n ..a v. n ,, e, .

-...w... ~..w v. - - -wy n t. z. ..

ecnfigurat.icr eterec " burred" cnd "frecF" fuel ccccmblice ir a 2x2 checkerbcard patterw. "uc1 acce=blice stered ir " burned" cell 1ccatienc muct 'cyc an initici enricF:ent lecc than 2." utt U 235 (nc=inal) er enticfy . mia&=um burnup require =crt The uce of c=pty ec11: ic cine an acceptabic eptier fer the " burned" cell 1ccatienc. Fuel ccccmblice ctered ir the "frech" cell 1ccatienc can have erricF=crt up te 5.0 utt U 235 .:ith

__..:.._.c-.t,.........,

..v

.s3m. ....v -- .v. . .g __t...._u.,

v. - . . ---v.u-.

t.-v.-.

2 The cecend reg'cr decc net utilice any cpecici leading patterr Fuc1 l cace blice uitF burnup and initici crricF ent: "hich fall inte the l unrectricted range es-F4gure TS.3.S 1 cc- be atcred anyuhcre ir the regler ]

vith ne cpecini picccrent rectrictienc. Fuc1 cccc=blice "hich fc11 inte the rectricted range cf Figure TS.3.8 1 muct be etered i- the checkerboard regler ir ceccrdance uit' Specificatier 5.6.^.1.d.

4 The burned /frech fuel chechcrbeard rgier car he pecitiened an7 hcre uith!- the cpent fuel rcchc, but *he beundary bet ccr the chechcrbeard regien and the unrectricted regler muct be citber:

1. ceperated by 'ccant reu cf celle, er
2. the interface cuct be configured cuch that there ic cre re carryever cf

+ke patter cf burned cccc=blic fre- the checkerbecrd reglen inte the first re af the unrectricted regler (Figure TS.5.5 1).

B.3.8-3 REV 'n" -- o. ,' _' ,' o, '-

e 3.8 REFUELING AND FUEL HANDLING Bases continued r i

. .go._. .e. n - T. .c 1. D . 1_ , , ,b. . i. n. k.

- _ .egn.mn4 - - . f .i. m . +- b.

-s . . ..- a --m. 4 n. . t.r.. i .m.. L . . ep.. . n enn.4emmmn+- * ' ' - * ' " -

fne

. . n. e. n.,. & e.4. --n e n,A-

.e, +_ n,, .e .a g .n, 4. n .

+h.,. .n,7-a m+n. .f.- .-- n 1 n. n_ n, i. , . 4 .n L.. e _e g

r '

m-

" A,r -nn ' ., e-* ~4 , h , n

  • e- . . - - f e. n_ m..

. m...

. . .. 1 . O. *,1

+n C.A, 4nh+

.m.g...-. n n_ e. n n n. . +.- T T _ 010. ,

7 -- - . T' n. .. e.t . n.

e-i.. n . , m e e b.- . .e n. .

- . - 1. .D. 7

,,s

-.g..e4mb m m e n n n. e.-

7-. - -.m.

me.

n n. n n. m. e.. ,.n. +

.. 4..- . . m i. j. . L n . . n. A m A- - - bj. , &

- - - . - b. . m m.. i. n. 4. m....-. k_ -.. . e n. . . gn. e .- n - . . .4. e. m . m n. e-

-. n,.g. f. a_ .e.

10F

. - . ..vmf.g

. a b. . e-n i, t _---S i t ...b. .4g i n-- .r 4.gn ., e. -m- - - 1 be, A 7 n.e . n m n. e. O. A A. A MT Th.,/.MT. it . Tc.1.0 h ,m .. ,s -.n.. n ,m

, .- - . T. b n- .e n. . .t -n -.e,m - --

en +- h o e- f ,s 1 1 07

-- . e. m n . 4 e m. 3-.. - --. 4 + L.

.. -.m., i n. 4 + .4 n 1

. .. - - m n. . ei.- n k. . m. n. e,. nf i n n .e.

-. - - . o.h. . .n. n. - - .m..

4 ,, h

g. . e-n f, f S.1 C. b e. . a b....n....n n.

cene.non 7-m .e n n n. t.- - . . 9. A A A -- Mf. .?N,/MT.

. - --. TY y - n e. g .- m ---va & n e. h n f. n. e.m

-- . . . . n. . e. .-- n -- .n e- v..in+mA-

~ -. o

.n.

i f. . m 1 ,411 L,

, - .al i. n. . m A ,

e. h. m- ,7 n n.n..e- .-- nnn1--

7 .. . - . - - -

T. h. ,. , e. e n. e. , .i n. e- b. . m. ,,.nm,n+-

7, . f. i n i.

--- m.

g-~a a. l n. n,. em.n.1 1. ,. ,

. n n n. +. n .i n.-

-- - .e.n....1 h. 1. m-

. k--n-. .r. ,n e , ,h. .t .n h. .e n .n . . l. e. .n

.t.. , 1-eg-nn

n. . .L.. n. v4

. - . . e_- . t ,, n.1. 4 e ,. . - n. .e.. g,. i n. . e

- n. . A n--

.n_#.-..m. . . - -

1 n n n e. n +-. 4 n. .g n

o

-- n . n. ~A l b.i. n n..n I. f. n . , m .---.--sv,

~7- .

+- b. . m PD. /,.*.

m gn.4 - . A..- m . "" 1 4> n. . mkm-mA ,. "-~~- ..unn r"*

+bm mantAnne nonA4etnn -- ' ' * * - -

in ,,binh mi l ,, n 1. .b i n nn4enn 4- - - - ' - -ane..mmA +n L ni . m L n ,. m 1* a- e- +- '- ~ '-nmn ' ' - - -4 f. ," '+1.g +-'"-"' +-b,.

14m' - - " ~ ~ -t & 4 m ,,

1, nf A OC L,--

- ' ' -- ' - ' - - - - " ' ~"

-r-"-/ o - u6 -- -'--

r .....

m. ". - 1. . n. o n i A. .

g h. . m. . .m L - n. .--n.m a, f. e n 1. . .h.. i -n--.- knenn.

- if ,s. n, . n m-- , . . +b- Am,4

-. - . pm. , . nf k._ n,. +.. b. e. mgn. in n. . .n,.

n A 4.c. b n .n. m A._ nn

+ b ,m .._nr. af-1

,k-

. . n. n .e .n. e-- . n e n .e , = =b. 4 n..b.

--- . . ,. n. . .e . n .e m .s * - k. n +- - - _ .nnb .e r. n 4 n n. 4n o._ -

4 .- b m...e.

. i. n.. &. .n. .

t

.n m-- A .. - a

_e.._n_e. .4 +- i n .a.1

.. n, a n. . A 4 o t.-

a n --

A . ..e. .ion. on. .n e m...e l. -. n

-g. n. n e. n e- 4. -. m .. .,4

+- L. . + L. . m- e,.s.g.n..,,.

,ni . . . ,

f.,

_ 1 1. j. , 1mmAnA. -- -

M m n... m ._, n 4 A ,s n. . +.- ,, a. n. A. 4. + 4 n n. . e_ n.r, e 4 e- . - -- A._ n_ - e.---.m c . 1 +. . .. - n n 4. g,. n....--

. 4 f. 4 n e n. . e 4. m.- .--n .e

., ,s 4n + L. . m e n n t

-_n n +- 4. . , 4 e j. , n .f.

. . m .- t & b. m e. a. .f

. e h. .-n+ --.n. g.

n,, n. . .e_ .

U. , n. m.. n. i.m _e

... . g - n.. .f- F. L m.,.m . . - -

an n 4. . A_ n n. e. .

n n n. , A_ t e i n n.o_

af n n,. a, l....g, 4 ., n n a,. .f onn A e. n g n. n n

.n e.m - . +- L. . a i_ n_ n e . -. + b. m

. - y .t n. .g .a .f. . m 1

-- .n. e .e.

. ..b_ l. ,. ,

nn

- &- b. . n- n .f

+.b..n,

-7 e n mb 1. , k n e. .m e n .,mi l / e n n t, y .n. , 1.,. , ._n. . A- +- t m-- .-." Aennninn af a f. . n l e

--n 1,

mnA.1n-" ' " ' - ' ' ' ' - ~ ~ - --

., n A

-'"",m"J r r' - " O ' ' ' - - - ' " - ' -

T-'-"--

A_n _,.4 o n. . ,s e n a_ l_ . . A ,s. .. e h. 4

.g 7 - - . .

., n,.n n. . A 4. e. 4. n n. . \ .

f 1.1. m. .. . - .ne

-nna- ,, 4 A n ., * -. - - . . - n a_ n.ka. -

-- n. n .e, e g --

. 1 ., +,. n,. A- o.h. . .n e-n n . 1. A- +- k n A n ., b.. i. n

-- 4 n.

., - n e. n .n. _e n. . . e n a n + 4 . , 4 +.j, .

.----.... U. n e . e.b. . n .e.- n .n a n t A n ., *.- - - . - - . . _.n. n n. . A....4 +- 4n.. . , &- L..m - .

o n. . -.

n

n. . e i n. .g ,,s n- n ,i , 7 ne. 4 m. .m.g i n.1 r-af

- A M. . c T. M. i. t_ .1. _1 0. .1 C, n, ~a n.km . -- .e_7gn n.1 4 m A. .- T. h. i _- , e .n. e m. .

  • L. . e. e.

n n..,.

4

. .n. n. , n o .- n n . 4 e. n A--

.-3-. +.a n,

.n.~., m..- m . - --

+.,,n . . n. l. 4 1., n 1,. . , 4.n. . .--

- A n n n ., A n g. -e-., -- -a n. n. . n . . e. e. n n. . +.- n , m n. & _n o

- ,,n nen+nn+ fan n n.

.. ",. . . e n' - - "

r- -~-'---'

an%

'C***-- 4n c +- '- a n e 4 +- 4 , . n l 4 & , , ' ' ' ' ' ' - - - 'maniAnne'- '

"**--""--J T. L. ,- A m. h.1. m netnnin1, 1 1 n .. , ..b1, knenn . . n A ,. e

-- - , . e. n.. n. *., 4. n. . y,. m /n.r. *a .-. - " '" r * ~ - - - ' ' ' ' - ' nenAte - ' " -fme "-~--- nni' ' - - ' ' ' ' - ' ' ' - ' - - "- - " " " " ~ ' ' ' * * " -nbnnemng - - -

g.- . n..----o n 4 A n n. . +_. n -- -n- n.A. .- 4 +. 4 m m. . .n. , e. .t --

n.... n a n n_ .,1. y. , - . 4. n. . g . m n1 - - ma n_4 A n n. +.-

.-- n. . n n_ A_ h m_

n_ o n. . _., .4-.-

A._ n m A._

no

- . . ,m..m,,

Q m.. m. Un.e. n1 &bn mn-e--- - ' ' ",. n .', 'm' e' n 4m ohm m a n 4 A n n e. ,1 m u m.m...r"8

  • m

- " ' " - a n n 4 A n m +."- ' " ' ' " ' - -nnnnetn

' - - ' ' - - ' ' -- -~ ~*-" - - " " * ' " " - ' - ' ' '

  • I
g. .,., I. n -n.. .A,_ I n. . g o
n. .f

. .n. f. . . n l.

.- ,n_ g .e. n. m.. _ j. h 1. , 4 n. . F . _an_e. .n. n l. ,. l n n a +. I ^ n - - - . . . .-. .. , In .- , ,b. 4. n b. F. b. . ^- e.n n b e. .I .a O. 4. n. n. . .e, an

i. n n n +.- 4. n. n. n n. e4 .-n -.h. m.. n n. e.- -. h . . e. n. .. . n ne aem nne m*4 Tb 4 ,.- - ' ' ' n n. 1 A n ,. o n n e t e l l , ,

r - - - ' ' ' ' ' - - - - ' ' - 'ftmA- ' ' - ' - ' ' ' ' - - -

r"*'->

4 ., Tn

.n.n f e. n..e. .. . ., - --m + k. . n e. n n n #- 4. ...,4 F,. e ..

.e, n. n n. . e_ .f. -. . n_1

.n n i.,. _n m. 4. +.- 4 a n. &. m. &. h.-. n_ _-7 n n n .. + . 1. .n. o.. n A--

g . . . --

..w . -

n, .

  • 4. > .i n .n 1 4. F . ,

j e. n 1.a. b n A--

onn4

- - - . A._ n _.n. e. .n. , .'Onnat(4nnbinn r"--'*-----

1 0 P O

' ' ' ' * * - " ' - ' - - -ann..ean- -&bn ' ** *

e. n n ., &

'r"*~

f. . n l

{ n n., i.

n a n. . o. .n. 4.

n.n., -nann-.

- , . a. t a -- .. ..

A_ 4 .,,. .,. n l. . , n A.-- b- -a- ..e. n. n .o. n. .ji , *- 4. m.. m_

- . f. n l.

-- m .e. .n.-.n ..b 1. I .n_ .n, . 4 + h.

. . .n o_-.n-.m..b i n. . ,- e-. i.-. n n nf n. . . e.n. . . 7n . - . .a-

- 4. .a. .

. 4.

1

-..f.ab.m.-..*- .n.

i

._-b..-

e n ..

e 4 n ,, A--- e. , n. . y,. n n f.

r. e ,. . . e. n y -

T. c. . . -

1 0 1

.n e. m . +.

.e...n.e n.. A- i n.

. +- b. . n.

f. . m 1

.- n7 a- n i. -..A.

an

-n - n ng. n. . e_ .f. . m 1- - ' -nan r""-

1 . . n .- 4 f i n n e i n n

' kan" " - nne' " " "

,Innn' ' * ' ' ' ' " +~b^ f. . 1 - ^ mb i , ---"---,& +ba

h. -n. n . nnefnemnA' ' ' ' ~'" 1ec&- - - - ".u - i e n - n n +-' ' ' ' - ' - 'nf ~ '-/ - - ' ' '

a.,,, a -nn. f. n l r"*-','- 4n -

"--- / -' r'"- - ' - " -

o 4 . , a na.n 7 i. T. b. n n a .n.

g -. .-

e.n. .

.a. n,, & 4 =., 4 4 - 3 .

n 4' f a. n #,. a .f t h. . e- .e, n 1. . .b l. e

- - k n .e n n. ..v n .--.A-1 a n. m..g.

. . n. n n. . .e .n. t--

n fg g.a . . 4 n. . n.e--n a _e n_ A_ e. n .n n. + 4 -

, , 4 +- . , a .r . n n A_

. .._j - __

L_ j. , - ..4. .e'r"-~-"'""'- n n n 4 +- 4 n n n A 4".

n l* - - ' ' ' - - - ' -m n - a mb1,,

s' T. h. e k. n e. -a--n.-. n n n. . a n n. . e. e. .n-.F. i n, . e. , . n . 4. v a y -. .

m.mn.+..,_. a. f. .C pa n 4. f 4. ,a n+.f.a,,.n.

-- . 1.O.P.O . _ .n. en. ,

en

i. n,, n.nne.y

( f...n.,. m..m..-m.mn..+-,, -ne- A .- ,m..- f. . 1 .mn.,. 4,m

.,g-_m.. u n .. a- . .b. n n. v

.,n n n n . . e. e .f m. a.


. . .-n,.-

..7- - .-

7-.

,,n.v..f..t.,ni. - - -

b.,nn.a..._-^

p. a .n. k .a n n n, .. a m.. g n.1 n o.- n A ,

nA

+ b. n .e. &-- n e. .n. g,. a. e. n ny-. s . 4 e. n m., a.n. e e af

. C,nnn4f4na&4nnn r " - * * * " - - " "-1 O

  • U 1 - ' ' - -

-..A.

nn

.R . f.

A

.1. . A- .m . n - -

&,. b, -. n n.

- - n.,

n.. o n

-- & n, n.

g e n . gn. . e_

-. - t l n e..e..a,..n..F.,u, 14

.C.g n. a, n_t f. 4 n n e. 4 a n

.. . - - - . - . . 1 . O . P. 9.. . a 4 n.n.+.-

I m...n. n _e. n_ A . . ,b a.n n+ n 1,. , f.. . . a l. ----n e n n mb. 1 4 n r..-- , ,4 o b. n g- -- . - -

n n .m. b. .i n..e. &. 4 m,..,

-- o,. .f b...,.en..,n. m..anA

. i n. t. + 4.- n.= 1

. n n e.t .n k. . n n. +.

4r o h. , . . n. e. ,.s . &.. *.- 4. n. +. n A--e. .n n. . y,, n nf

.g . . ..s . -.

r .i nu..

. . . e n T. c 1.

0 .1 ee.

-.m -e t. n e..- n. A 4 +

h., _- vn n n. e. .-- n ar-n i. .

f. . n l g.m . e.n n . 4. e. n.. m. n. n. . o n--af

. ~

---.l f .i n n & 4. .n_ ., 1 n. . n e e.n n . .g n.. A- f .. 1 sma

. - . . ~ 0. . P.. . O . . .e, e. n. -

3 i. n. +h...e+.- n.n._,,m. .

k_ m -

n_ _e . . , ,- . _ ,4.. + b. . m n. . i. , ,

- j ...n.

c. e n.mk.1. .f a n

. . +- b. . n e b...,.. .

b.,...-n....n

.r 'anA' ' - in .st 1*+ "4- - - - - -

n n

" * - ' 4 n k m n n +-' ' ' ' ' *fn &bn

-- . .n e n n & e 4 n- " - ~ ~ onA ' ' ' ' ' - - --*

ennnn

--- af P.4n. y,. e. n T. C . 1. . O . 1 - . . .4+.

4

.e. n. , n e-- n g-n..,-,. 4 g1. n

+n n

-.ai._,m - -

an 4

n. ..n.A. i. n e o.. ,n n, e

. . n. .e .t e. 4 n. m 1 4 o,n

, . k.. ,. , ' r ' '

  • m.. .i .e n n ,,7 . &.

. .-4 n n. . i. n. . g,. -

n

f. . n. l.

a

- c _c a. m...b i. ,,

- .n.

i e b. . n _e gn, n n. . o.

fe in 1

.-- n a n,. i.

7-

B.3.8-4 REV 108 9/3/93 3.8 REFUELING AND FUEL HANDLING Bases continued Wheni theTre<[dirementsTbf? Specifiba tion?3i8:E ITa?a r eino ti me tWimmsdiateTa 6 tion mus t{ bs ltsakehito [ moveianp[non TcomplyinglfuelTas s embly %o l an ? asce ptable (location t o) pre se rve? the f doubleicontiripucyl princ ip1 pas s;umptionj;;o;f&th~ep c ri.t isality. i accidentianalysis;eL When the requirements of Specification 3.8.E.2.a are 6otJmetinpplicab12, crd the '

sweentr: tier cf berer ir the cpe-t fuel pec1 ic lecc ther required, immediate action must be taken to preclude the occurrence of an accident or to mitigate the consequences of an accident in progress. This is most efficiently achieved by immedistely suspending the movement of fuel assemblies. The concentration ,

of boron :.s restored simultaneously with suspending movement of fuel ,

assemblies. .-^ acceptable 21 terre *!re ir te crrplete : cpent fuel pect w s4Wat er t

!! cue"^r, pPrior to resuming movement of fuel assemblies, the concencration of boron must be restored. This does not preclude movement of a i fuel assembly to a safe position. l i

^ cpe-t fuel peel "crificatie- ic required fellering t! c lect movement er feci ecce-blicc i r t : cpe-t fuci perl, if fuel acce=blic: .:ith a ce=binatie ef  !

hurnup and initial e-ric1 ment in th rectricted range of Figure TS . 3. 8 1 cre t

etered ir the cpert fuci pec1. Thic verificatier .:111 cerfir- t'et any fuel scremblic: .zith a c^ binatier cf burnup and initial enricFeent in the i rectricted range of Figure TS.3.81 cre etered ir neccrdance uith the l

require-^nt cf Specific tier 5.6.^.1.d.

l l

References

1. USAR, Section 10.2.1,2
2. USAR, Section 14.5.1 i I
3. USAR, Section 10.3.7 l 1
4. " Criticality Analysis of the Prairie Island Units 1 & 2 Fresh and Spent Fuel Racks", Westinghouse Commercial Nuclear Fuel Division. February 1993.
5(WCAP y1441 fs P p"We s tinghous e
l S p eht( Fuelf Radk;jyritisali ty[AnAlys islMe tho da165py June 319951 62 Amer i b ah;?.Nd el e a r? Soc ie ty fi "Ame ric an ' Na tional? S tahdird X Des ignJ Re'quirements; foi -

LightlWaterj Reactor Fuel l Storage' Facilities; Tat Nuclear lPowerjPlants";,{ ANSI /ANSj l 57,2fl983h.0ctoberj 7.',319831 7f; Nuclsar?Regu1'atoryJCommissich? Letter;1to?All? Power Reactor 1 LicenseesTfronif BT K?

Grimes , ' "OTz Position (for . Review ' and : Acceptance of y Spent Fuel} Storagefand ' Handling Appli ca t i.o,ns."g Ap rilf 14j fl978 ;

INSERT 7

The?Westingh66se ;SpenMFusFRsek?CritiEslity!Meth6do16gy?lsnsurisTthatstheTspentifuel racFmultiplic ati.on l factor RKdtkisdes s';;than i: 0. 9.5 ? ns i rec ommended bpfANS I[: 57i2 /1983

'(Reference (6)L(and?NRCfgdidancef(Reference (7)0 ;Thefcodessmethsdat:and{ techniques c ontAinsd);in[ths;ime thodoloj;yiarelussdito i;s atis fylthisjeritehhnionj Ketts Thuiseth6do16gffof*:theINITAWINIly XSDRNPM?SU shd KENO WsThodss{sstablish3a methodologf;:; biasiandj bias (dncertainty ( l PHOENIXpPy ainuclearfdesignidodes Used primarilygforicordjreastorfphysles3aldslationslisiused36fsimdlatchpenQfuel atpragepaskJeonetries?

Ths5Spsnt$ Fds11Rshk;;(Crihids111tp:'CsiculstI6h5?ishbi'6h'^6f %helbsth6d6165f7diEchisss the detailsYonithsiass'umpti6nskssds[tisimodslithelspenttfeslisborasel:racksfandithe '

Useief[theNesulti::Lareipre;sested $Spshifici:'detdisson7KENONas;;calediatsinnsh "

PHOENIXf Pjl; tole.ranc e[cs1culsti$nssa$dithe sfina1095/95j Khrf de tsrmindtionshinf^

discdssedQ The7 Rsac tiTityl:?Equivaleh61nF Mstho'ds16gfidisEEs sss @hs7tsshhihue s][usadit6Ial16s higherf fne1[assemblyTenrichmehtsiti6]be-[st6rsd (in%helsper tEfue1[stor.dgef racksib takingietidit1forsfuel?sssembiff burndpland{Intehrals Fuel (B'drnable;?Ab$osbersy(IFBAE -

The'soncbst{bf[reactivitif(equivalending(.nndf the;dasuspeibnsland}anbertaintidi^

l Associated withjeachKreactivityjeqdivalencin6techniqueJaret;:discussedj JThsibssI6f PHOENIX) Plinic achirea c t;1virfie.quivalene in'g {me th6do16gygisj als oldis cus s e d Q ThelPo s tuld ts d ~Ad c identRMe th6 dolo gf3de finsa s ths?p 6stnis tsd[sp en tM fds1pscWsss ident s Uhichfareiconside'redbinfths?spentif6sl:fraa16chiticalitspi.auslysiil LThaimethod616gp

~

used(to;detorminei th6)reac'ti0itpiimpactj;ofitnese;iAccidentspisidiseussedi gFihd11yli ~ ~

thel applicat ioniof Lthe'j doh61sicontiingencyfprindiplesof these(:sp'entifuelfrask postuistedJaccidentseiscdiscussedOwhichfalloksJcheditiforispentitfue1@aolSs616bli ~

boronJto"offsetYthe jotential Weactivitydinbresse}cassed M these:ioff2nbrm d ,

conditionsj  !

Ths" S 61bbid@6rohi Cre di'tiMs th6do16gyydis duss'e s1hbvMrs di'tif6rl::spsntifU e1%po61 soluble b6roniisi:usediunderf normelystorage?configuratiahic6nditichs iThe;fstofshs; configurat loniis4definedissing"thefmasimum? feasible?K o rtscsiculahlonsjtioichs6rejnhat the - spent fuelf racWK,re* villi be J1.sss ' thanl130lwith > no < solubleib;orbni dnders normal

.atorage; c6nditionsi iSoluble boron;(cr6dictisl then{usedj toioffsetiths{uncertaint-les andatolerancesiand'maintainLKrplessWthan(orsequalstef0195n:IThsidsstoffsoluble; boronl crediti fori:reac tivityl equivalenc'ingiuncertaint!ies( iMdisbusse'd pas dss 117adRhs calculation oE postulated /abcidents l crediting l soluble}. boron? gaummagofLall{of the"solbble'1b o ron ie re dit;tre quireme nts s isi als oidi s e;ds s ed f

,e *,

L '_ R -

Exhibit C Prairie Island Nuclear Generating Plant ,

License Amendment Request Dated July 28, 1995 Revised Technical Specification Pages i Exhibit C consists of revised pages for the Prairie Island Nuclear Generating ,

Plant Technical Specifications with the proposed changes incorporated. The-revised pages are listed below:

TS.1-2 TS.3.8-4 TS.3.8-5 Table TS.4.1-2B (Page 1 of 2) l Table'TS.4.1-2B (Page 2 of 2). -

TS.S.6-1 '

TS.S 6-3 TS.6.7-4 TS.6.7-5 B.3.8-2 B.3.8-3 l

l l

)

i i

1

'i I.

b

v c TS.1-2 REV ,

CONTAINMENT INTEGRITY CONTAINMENT INTEGRITY shall exist when:

i

1. _ Penetrations required to be isolated during accident conditions are either:
a. Capable of being closed by an OPERABLE containment automatic isolation valve system, or
b. Closed by manual valves, blind flanges, or deactivated automatic valves secured in their closed positions, except as provided in Specifications 3.6.C and 3.6.D.
2. The equipment hatch is closed and sealed
3. Each air lock is in compliance with the requirements of Specification -,

3.6.M.

4. The containment leakage rates are within their required limits.

CORE ALTERATION CORE ALTERATION is the movement or manipulation of any component within the reactor pressure vessel with the vessel head removed and fuel in the vessel, l which may affect core reactivity. Suspension of CORE ALTERATION shall not '

preclude completion of movement of a component to a safe conservative position.

CORE OPERATING LIMITS REPORT The CORE OPERATING LIMITS REPORT is the unit-specific document that provides I core operating limits for the current operating reload cycle. These. I cycle-specific core operating limits shall be determined for each reload cycle in accordance with Specification 6.7.A.6. The Unit 1 CORE OPERATING LIMITS REPORT also contains the spent fuel pool operating limits. Plant operation within the core and spent fuel pool operating limits is addressed in individual specifications, i

)

l l

.g= ..g TS.3.8-4 REV 3.8.C. Small Spent Fuel Pool Restrictions No more than 45 recently discharged assemblies shall be located in I the small pool (pool No. 1).

D. Spent Fuel Pool Special Ventilation System

1. Both trains of the Spent Fuel Pool Special Ventilation System '

shall be OPERABLE at all times (except as specified in 3.8.D.2 and 3.8.D.3 below).

2. With one train of the Spent Fuel Pool Special Ventilation System inoperable, fuel handling operations and crane operations with loads over spent fuel (inside the spent fuel pool' enclosure) are permissible during the following 7 days, provided the redundant  :

train is demonstrated OPERABLE prior to proceeding with those operations. ,

3. With both trains of the Spent Fuel Pool Special Ventilation System inoperable, suspend all fuel handling operations'and  ;

crane operations with loads over spent fuel (inside the spent fuel pool enclosure).  ;

r

4. The provisions of specification 3.0.C are not applicable.

E. Spent Fuel Pool Storane

1. Fuel Assembly Storage  ;
a. Fuel assemblies shall be stored in accordance with the configurations specified in the Unit 1 CORE OPERATING LIMITS REPORT.
b. If the requirements of 3.8.E.1.a are not met, immediately initiate action to move any noncomplying fuel assembly to an acceptable location.
c. The provisions of Specification 3.0.C are not applicable. l i

l 1

i i

l 1

I l

TS.3.8-5 REV 3.8.E.2. Spent Fuel Pool Boron Concentration

a. The spent fuel pool boron concentration shall be maintained within the limits specified in the Unit 1 CORE OPERATING -

LIMITS REPORT.

b. If the spent fuel pool boron concentration is not within the limits specified in the Unit 1 CORE OPERATING LIMITS REPORT, then immediately:
1. Suspend movement of fuel assemblies in the spent fuel pool, and
2. Initiate action to restore spent fuel pool boron concentration to within its limit.
c. The provisions of Specification 3.0.C are not applicable.

~_

. cl e zh  !

s .1 l

i Table TS.4.1-2B (Page 1.of 2) l REV.  ;

}

l TABLE TS.4.1-2B l MINIMUM FREOUENCIES FOR SAMPLING TESTS i t

TEST FREOUENCY f

1. RCS Gross 5/ week l

'j

~

Activity Determination-t

2. RCS Isotopic Analysis for. DOSE 1/14 days (when at power)- l EQUIVALENT I-131 Concentration i 1

~

13. RCS Radiochemistry i determination 1/6 months (1)'(when at power) '!

i 7 4. .RCS Isotopic Analysis for Iodine a) Once per 4 hours4.62963e-5 days <br />0.00111 hours <br />6.613757e-6 weeks <br />1.522e-6 months <br />, whenever  !

Including I-131, I-133, and I-135 the specific-_ activity ex -

ceeds 1.0 uCi/ gram DOSE _. _

EQUIVALENT.I-131'or 100/E ,

uCi/ gram (at or.above cold ,

shutdown), and j b) One sample between 2 and 6 f hours following THERMAL'- ..j POWER change' exceeding 15 l percent of the RATED THERMAL l POWER within a one hour l period (above hot shutdown) i

I RCS Radiochemistry (2) Monthly '

5.

6. RCS Tritium Activity Weekly f
7. RCS Chemistry (Cl*,F*, 02) 5/ Week  ;
8. RCS Baron Concentration *(3) 2/ Week (4)
9. RWST Boron Concentration Weekly -i
10. Boric Acid Tanks Boron Concentration 2/ Week l

'll. Caustic Standpipe NaOH Concentration Monthly .

r L12. Accumulator Boron Concentration Monthly l l

13. Spent Fuel Pit Boron Concentration Weekly t

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

~ ~

n ]

e 4  ;

Table TS.4.1-2B (Page 2 of 2) -

REV TABLE TS.4.1-2B MINIMUM FREOUENCIES FOR SAMPLING TESTS TEST FREOUENCY

14. Secondary Coolant Gross Weekly Beta-Gamma activity e
15. Secondary Coolant Isotopic 1/6 months (5)

Analysis for DOSE EQUIVALENT t I-131 concentration j

16. Secondary Coolant Chemistry pH 5/ week (6) '

pH Control Additive 5/ week (6)

Sodium 5/ week (6)

Notes:

Sample to be taken af ter.a minimum of 2 EFPD and 20 days of POWER 1.

OPERATION have elapsed since reactor was last suberitical for 48 hours5.555556e-4 days <br />0.0133 hours <br />7.936508e-5 weeks <br />1.8264e-5 months <br /> or longer. 1

2. To determine activity of corrosion products having a half-life greater than 30 minutes.
3. During REFUELING, the boron concentration shall be verified by chemical analysis daily.
4. The maximum interval between analyses shall not er:ceed 5 days. j i
5. If activity of the samples is greater than 10% of the limit in Specification 3.4.D, the frequency shall be once per month. i
6. The maximum interval between analyses shall not exceed 3 days. '

l l

1 l

l l

1 i

I l

l

R pe; ..

i j3 'TS.S.6-1 REV e 5.6 FUEL HANDLING -

(

A. Criticality Consideration  ;

1. The spent fuel storage racks are designed (Reference 1) and shall be maintained with D a. Fuel assemblies having a maximum U-235 enrichment of 5.0 weight . 2 percent;  ;
b. K,gg s 0.95 if fully flooded with borated water, which includes an '

allowance for uncertainties as described in Reference 3; and'

c. Ke rr < 1.0 if fully flooded with unborated water, under maximum feasible conditions as described in Reference 3.

l

2. The new fuel storage racks are designed (Reference 1) and shall be  :(

maintained with: .i

a. Fuel assemblies having a maximum U-235 enrichment of 5.0 weight -

percent;  :

i

b. K e rr s 0.95 if fully flooded with unborated water, which includes ,

an allowance for uncertainties as described in Reference 2;.and  !

c. K e rr s 0.98 if accidentally filled with a low density moderator )

which.resulted in optimum low density moderation conditions. .,

3. Fuel will not be inserted into-a spent fuel cask in.the pool, unless a minimum boron concentration of 1800 ppm'is present. :The 1800 ppm will s ensure that k o rt for the spent fuel cask, including statistical- .

uncertainties, will be less than or equal to;0.95 for all postulated I

arrangements of fuel within the cask. The criticality analysis for the TN-40 spent fuel storage cask was based on fresh fuel enriched to 3.85 weight percent U-235.

B. Spent Fuel Storare Structure The spent fuel storage pool is enclosed with a reinforced concrete building having 12- to 18-inch thick walls and roof (Reference 1) . ,

The pool and pool enclosure are Class I (seismic) structures that Jj afford protection against loss of integrity from postulated' tornado y missiles. The storage compartments and the fuel transfer canal are l connected by fuel transfer slots that can be closed ~off with 'l '

pneumatically sealed gates. The bottoms of-the slots are above the

. tops of the active fuel in the fuel assemblies which will be. stored vertically in specially constructed racks.

I

i

. )

TS.S.6-3 REV  !

i l

D. Spent Fuel Storace capacity l The spent fuel storage facility is a two-compartment pool that, if completely filled with fuel storage racks, provides up to 1582 storage locations. The southeast corner of the small pool (pool no. 1) also serves as the cask lay down area. During times when the cask is being used, four racks are removed from the small pool. With the four storage racks in the southeast corner of pool 1 removed, a total of i

1386 storage locations are provided. To allow insertion of a spent fuel cask, total storage is limited to 1386 assemblies, not including those assemblies which can be returned to the reactor, i

Reference i

1. USAR, Section 10.2
2. " Criticality Analysis of the Prairie Island Units 1 & 2 Fresh and Spent i Fuel Racks", Westinghouse Commercial Nuclear Fuel Division, February 1993. l
3. WCAP-14416-P, " Westinghouse Spent Fuel Rack Criticality Analysis l Methodology", June 1995. i i

L

.+- .

_, TS.6.7-4 REV 6.7.A.S. Annual Summaries of Meteoro1ocical Data An annual summary of meteorological data shall be submitted for the previous calendar year in the form of joint frequency distributions of wind speed,-wind direction, and atmospheric stability at the request of the Commission.

6.7.A.6. Core Operatinz Limits Report

a. Core operating limits shall be established and documented in the CORE OPERATING LIMITS REPORT before each reload cycle or any remaining part of a reload cycle for the following:

RPT '

1. Heat Flux Hot Channel Factor Limit (Fg ), Nuclear Enthalpy RTP Rise Hot Channel Factor Limit (fag ), PFDH, K(Z) and V(Z)

(Specifications 3.10.B.1, 3.10.B.2 and 3.10.B.3)

2. Axial Flux Difference Limits and Target Band (Specifications 3.10.B.4 through 3.10.B.9)
3. Shutdown and Control Bank Insertion Limits (Specification 3.10.D)
4. Reactor Coolant System Flow Limit  ;

(Specification 3.10.J)

b. Spent fuel pool operating limits shall be established and documented in the Unit 1 CORE OPERATING LIMITS REPORT for the l following: l l

1 Spent Fuel Fool Storage Configuration Limitations (Specification 3.8.E.1) ]

2. Spent Fuel Pool Boron Concentration Limit (Specification 3.8.E.2)
c. The analytical methods used to determine the core operating limits shall be those previously reviewed and approved by the NRC, specifically'those described in the following documents:

NSPNAD-8101-A, " Qualification of Reactor Physics Methods for Application to PI Units" (latest approved version)

NSPNAD-8102-A, " Prairie Island Nuclear Power Plant Reload Safety Evaluation Methods for Application to PI Units" ,

(latest approved version) )

WCAP-9272-P-A,'" Westinghouse Reload Safety Evaluation Methodology", July,1985 WCAP-10054-P-A, " Westinghouse Small Break ECCS Evaluation Model Using the NOTRUMP Code", August, 1985

7-

_o- 4.

, TS.6.7-5 REV WCAP-10924-P-A, " Westinghouse Large-Break LOCA Best-Estimate Methodology", December, 1988 WCAP-10924-P-A, Volume 1, Addendum 4, " Westinghouse Large Break LOCA Best Estimate Methodology", August, 1990 KN-NF-77-57 (A), XN-NF-77-57, Supplement 1 (A), " Exxon Nuclear Power Distribution Control for Pressurized Water Reactors Phase II", May, 1981 WCAP-13677, "10 CFR 50.46 Evaluation Model Report:

H-COBRA / TRAC 2-Loop Upper Plenum Injection Model Update to Support ZIRLOm Cladding Options", April 1993 (approved by NRC SE dated November 26, 1993)

NSPNAD-93003-A, " Transient Power Distribution Methodology",

(latest approved version)

d. The analytical methods used to determine the spent fuel pool' operating limits shall be those previously reviewed and approved by the NRC, specifically those described in the'following documents:

WCAP-14416-P, " Westinghouse Spent Fuel Rack Criticality Analysis Methodology", June 1995

e. The core and spent fuel pool operating limits shall be determined such that all applicable limits (e.g., fuel thermal-mechanical limits, core thermal-hydraulic limits, ECCS limits, nuclear limits such as shutdown margin, transient and accident analysis limits, and spent fuel pool criticality limits) of the safety analysis are met,
f. The CORE OPERATING LIMITS REPORT, including any mid-cycle -l revisions or supplements thereto, shall be supplied upon issuance, for each reload cycle, to the NRC Document Control Desk with copies to the Regional Administrator and Resident Inspector.

B. REPORTABLE EVENTS The following actions shall be taken for REPORTABLE EVENTS:

a. The Commission shall be notified by a report submitted pursuant to the requirements of Section 50.73 to 10 CFR Part 50, and
b. Each REPORTABLE EVENT shall be reviewed by the Operations Committee and the results of this review shall be submitted to the Safety Audit Committee and the Vice President Nuclear Generation.

, B.3.8-2 REV 3.8 REFUELING AND FUEL HANDLING Bases continued During movement of irradiated fuel assemblies or control rods, a water level of 23 feet is maintained to provide sufficient shielding.

The water level may be lowered to the top of the RCCA drive shafts for latching and unlatching. The water level may also be lowered below 20 feet for upper internals removal / replacement. The basis for these allowance (s) are (1) the refueling cavity pool has sufficient level to allow time to initiate repairs or emergency procedures to cool the core, (2) during latching / unlatching and upper internals removal / replace-ment the level is closely monitored because the activity uses this level as a reference point, (3) the time spent at this level is minimal.

The Prairie Island spent fuel storage racks have been analyzed in accordance with the methodology contained in Reference 5 to allow for the storage of fuel assemblies with enrichments up to 5.0 weight percent U-235 while maintaining K e rr s 0.95 including uncertainties and credit for soluble boron.

The Westinghouse Spent Fuel Rack Criticality Methodology ensures that the spent fuel rack multiplication factor, K.rt, is less than 0.95 as recommended by ANSI 57.2-1983 (Reference 6) and NRC guidance (Reference 7). The codes, methods and techniques contained in the methodology are used to satisfy this criterion on K.tr-l The methodology of the NITAWL-II, XSDRNPM-S, and KENO-Va codes establish a methodology bias and bias uncertainty. PHOENIX-P, a nuclear design code used primarily for core reactor physics calculations is used to l simulate spent fuel storage rack geometries.

The Spent Fuel Rack Criticality Calculations section of the methodology discusses the details on the assumptions made to model the spent fuel storage racks and the use of the results are presented. Specific details on KENO-Va calculations, PHOENIX-P tolerance calculations and the final 95/95 K,gt determination are discussed.

The Reactivity Equivalencing Methodology discusses the techniques used to i allow higher fuel assembly enrichments to be stored in the spent fuel l storage racks by taking credit for fuel assembly burnup and Integral Fuel Burnable Absorbers (IFBA). The concept of reactivity equivalencing and the assumptions and uncertainties associated with each reactivity equivalencing technique are discussed. The use of PHOENIX-P in each reactivity equivalencing methodology is also discussed.

The Postulated Accident Methodology defines the postulated spent fuel rack accidents which are considered in the spent fuel rack criticality analysis. The methodology used to determine the reactivity impact of these accidents is discussed. Finally, the application of the double contingency principle to these spent fuel rack postulated accidents is discussed, which allows credit for spent fuel pool soluble boron to offset the potential reactivity increase caused by these off-normal conditions.

l s .

B.3.8-3 REV i

3.8 REFUELING AND FUEL HANDLING Bases continued  !

l The Soluble Boron Credit Methodology discusses how credit for spent fuel pool soluble boron is used under normal storage configuration conditions.

The storage configuration is defined using the maximum feasible K.tr calculations to ensure that the spent fuel rack K.tr will be less than 1.0 ,

with no soluble boron under normal storage conditions. Soluble boron credit is then used to offset the uncertainties and tolerances and l maintain K.tr less than or equal to 0.95. The use of soluble boron credit  ;

fer v44ccivity equivalencing uncertainties is discussed, as well as the j calculation of postulated accidents crediting soluble boron. A summary i ef all of the soluble baron credit requirements is also discussed.  !

I When the requirements of Specification 3.8.E.1.a are not met, immediate  :

action must be taken to move any non complying fuel assembly to an  !

acceptable location to preserve the double contingency principle  !

assumption of the criticality accident analysis.

When the requirements of Specification 3.8.E.2.a are not met, immediate l' action must be taken to preclude the occurrence of an accident or to mitigate the consequences of an accident in progress. This is most efficiently achieved by immediately suspending the movement of fuel assemblies. The concentration of boron is restored simultaneously with ,

suspending movement of fuel assemblies. Prior to resuming movement of l fuel assemblies, the concentration of boron must be restored. This dors

.wt preclude movement of a fuel assembly to a safe position. I i

References

1. USAR, Section 10.2.1.2 ,

l

2. USAR, Section 14.5.1 l l

1

3. USAR, Section 10.3.7 l
4. " Criticality Analysis of the Prairie Island Units 1 & 2 Fresh and Spent Fuel Racks", Westinghouse Commercial Nuclear Fuel Division, February 1993.

i S. WCAP-14416-P, " Westinghouse Spent Fuel Rack Criticality Analysis Methodology", June 1995. l

6. American Nuclear Society, "American National Standard Design Requirements for. Light Water Reactor Fuel Storage Facilities at Nuclear Power Plants",

ANSI /ANS-57.2-1983, October 7, 1983, i

Nuclear Regulatory Commission, Letter to All Power Reactor Licensees from B. K. Grimes, "0T Position for Review and Acceptance of Spent Fuel Storage and Handling Applications", April 14, 1978.

. _ D ., * ', d ] t Lp . l

-( .

o Exhibit D

.' f Prairie Island Nuclear. Generating Plant License Amendment Request Dated July 28, 1995 I

Final Report .I Applicability of WCAP-14181 PRA Results l To Prairie Island  ;

i k

i i

i i

k I

~$ .

^

{

,p

'l .

f i

I i

i

-)

.i I

.f-l I

'. i I

'I 1

+

'I f

i

.9 FINAL REPORT APPLICABILITY OF WCAP-14181 PRA RESULTS TO PRAIRIE ISLAND Table of Contents Section Page >

1.0 INTRODUCTION

1 2.0 PLANT COMPARISON 2 2.1 Spent Fuel Pool Characteristics 2  :

2.2 Boron Dilution Initiating Events 4 ,

t 2.2.1 Reactor Cavity Pneumatic Seal Failuro 5 2.2.2 CCW Leak 5 2.2.3 Seismic Event 7 '

2.2.4 Tornado Event 8 2.2.5 Random Pipe Breaks 8 2.2.6 Demineralizer Valves / Makeup Valves Open 10 2.2.7 Initiating Event Results 15  ;

l 2.3 Boron Dilution Times and Volumes 15 t

2.3.1 Consideration of Prairie Island Dilution Sources 18

3.0 CONCLUSION

S 20

4.0 REFERENCES

21 )

i l

I l

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

..' "-. 'lP Preme latend Evaluation reuphpet .-

-t 1.0 INTRODUOT!ON I i

, in WCAP-14191, Evaluation of the Potential for Diluting PWR Spent Fuel Pools  !

(Reference 1), a generic methodology was applied to identify potential events which  !

could' dilute the soluble boron contained in PWR spent fuel pools and to quantify the frequency of these events. The methodology utilized a probabilistic risk assessment  !

i (PRA) to calculate the event frequency of a dilution event, and used a ' composite plant' to complete the PRA. The results of the PRA concluded that the event frequency remained less than the NRC Safety Goal Policy Statement target risk objective of 1E-6/ reactor year,,

Q.

I i

lll in order to reference the results of the composite plant PRA, a comparison of the i Prairie Island-specific features related to the spent fuel pool must be made and evaluated versus those of the ' composite plant.' The specific items to be compared are noted in Section 2.0. In addition, a Prairie Island-specific boron dilution event will - 'l be used to determine the available operator action times to respond to a dilution event. This event will consider nominal conditions at Prairie Island, and will be compared to the best estimate case of Reference 1. The specific dilution event to be evaluated for Prairie Island is discussed in Section 2.3.

Finally, deterministic calculations will be performed in order to define the dilution time and volumes for Prairie Island. This data will then be compared to analogous data for the composite plant. Also, the dilution sources available at Prairie Island will be compiled and evaluated against the dilution volume calculated to determine the potential of a spent fuel pool dilution event.

J 1

.- . . -- , .- . - - _ . - . - _ - - _. . . l

2 . - - -e

. I

e. -

Preme Beland Evaluation Fruf% sat l l

2.0 Plant Comparison To assess the applicability of the generic PRA results to Prairie Island, the following plant-specific features will be evaluated:

o . Spent Fuel Pool and Related System Features Dilution Sources Dilution Flow Rates Boration Source Instrumentation Administrative Procedures Piping Loss of Offsite Power impact o Boron Dilution Initiating Events o Boron Dilution Times and Volumes 2.1 Spent Fuel Pool Characteristics Table 2-1 provides a comparison of relevant spent fuel pool data for the composite plant and Prairie Island.

2

er.wi. i.i.no evwo.non Anm%xx TABLE 2-1 SPENT FUEL POOL COMPARISON - PRAIRIE ISLAND vs COMPOSITE PLANT Plant Feature Composite Plant Prairio Island Prairie Island Large Pool Small Pool Pool Water 232,000 395,000 63,000 Volume (gal)

Typical Pool Boron 2200 3250 Concentration (ppm)

Location Fuel Building, Fuel Building, Top Floor, Top Floor, Seismic I Seismic l Unborated Water CCW, Demin, CCW, Demin, Sources Reactor Makeup, Reactor Makeup, Fire Protect, Fire Protection SW Borated Water CVCS, RWST CVCS, RWST Sources Piping near Reactor Makeup, 3/4" Domin Water, Spent Fuel Pool Fire Protect, One Fire Hose Station i Domin, CCW, SW 1 Dilution 100 opm for 20 opm for Domin Pipe Rupture.

Flow Rates Domin, Reactor ]

Makeup, Fire 100 opm for CCW tube leak, ,

Protect. Domin Valve Open, Makeup 1 500 opm for Valve Open CCW,SW Instrumentation 1 Train Level 1 Train Level Alarms, No Alarms, No Safety Related Safety Related Power Power Loss of Cannot Use Cannot Use RWST Offsite Power RWST Leaks in SFP To SFP from CCW To SFP from CCW Heat Exch.

Admin Controls in effect in effect 3

e .i e

'a P,eine loland Evaluation Frapet As can be seen from a review of Table 2-1, the Prairie Island spent fuel pool characteristics are very similar to the composite plant. With the exception of the small pool volume, each of the Prairie Island features is either identical or bounded by the comg.osite plant. The small pool volume is considered because for approximately one week of tiis year, this poolis isolated from the large pool for cask handling. The small pool volume is calculated to be 63,000 gallons. For the remaining 51 weeks of the year, both pools and the transfer canal are interconnected for a large pool volume of 395,000 gallons. Note that during the times of cask handling, there is increased activity in the vicinity of the spent fuel pool and the spent fuel pool boron i concentration is verified 4 hours4.62963e-5 days <br />0.00111 hours <br />6.613757e-6 weeks <br />1.522e-6 months <br /> prior to loading the cask.

2.2 Boron Dilution initiation Event l

Based on a review of Prairie Island data (i.e., Licensee Event Reports and Deviation Reports pertaining to spent fuel pool events), no new spent fuel pool boron dilution events have been identified. In addition, the justification for removing some of the potential initiating events, as discussed in Reference 1, remain valid. The following i l

discusses the differences between consideration of the remaining initiating events for the composite plant versus Prairie Island, in addition, it will be noted which events are not applicable with respect to the small pool volume.

Note that for Prairie Island, the nominal spent fuel pool boron concentration is 3250 ppm. This would be the starting point for consideration of a dilution event. Per Reference 2, the soluble boron concentration required for criticality requirements is 1050 ppm. Thus,1050 ppm would be the final endpoint for e Prairie Island dilution event. For the purposes of this evaluation, the boron endpoint is increased to 1380 ppm; this will be a conservative representation of the Prairie Island spent fuel pool 4

p

,e-1

~..  !

- Prene telend Evoluebon ' FTudRipet

?

boron dilution event, and permit the utilization of calculations previously completed 1 for Prairie Island boron dilution events.  !

2.2.1 Reactor Cavity Pneumatic Seal Failure The initiating event frequency, event tree and top event descriptions as discussed in  ;

Reference 1 for the composite plant are applicable to Prairie Island. In addition, the j assumption of a 1000 gpm leak rate is maintained. However, operator action times j are different for Prairie Island. Specifically, if it assumed that there is 23 feet of water above the top of the fuel assemblies and approximately 37.5 feet of water in the large . j pool,21 feet of water would drain in approximately [(21/37.5)*395,000/1000 gpm)

= 221 minutes. Thus for Prairie Island, there are at least 3 hours3.472222e-5 days <br />8.333333e-4 hours <br />4.960317e-6 weeks <br />1.1415e-6 months <br /> available to begin to restore pool inventory before the water is drained to two feet above the fuel assemblies. For the composite plant, it is assumed that there is 23 feet of water above the top of the fuel assemblies and a water height of 40 feet; however , the pool -

volume is only 232,000 gallons. Thus, at a 1000 gpm leak rate, 21 feet of water would drain in only 2 hours2.314815e-5 days <br />5.555556e-4 hours <br />3.306878e-6 weeks <br />7.61e-7 months <br />. Therefore, for the composite plant, only 2 hours2.314815e-5 days <br />5.555556e-4 hours <br />3.306878e-6 weeks <br />7.61e-7 months <br /> are available for detection, diagnosis and isolation. For this initiating event, Prairie Island ~

is bounded by consideration of Reactor Cavity Pneumatic Seal Failure for the composite plant.

Since the small poolis isolated for cask handling, reactor cavity pneumatic seal failure is not considered as an initiating event for the small pool.

2.2.2 CCW Leak The initiating event frequency, event tree and top event descriptions as discussed in 5 )

r- .

.[

l .

Preene toland Evaluation 4dRupat Reference 1 for the composite plant are applicable to Prairie Island, in addition, the assumption of a 100 gpm leak rate is maintained. However, operator action times are different for Prairie Island. Specifically, the detection (DETECT LATER) and response (OPERATOR RESPONSE) times for consideration of the large pool volume are much longer for Prairie Island. Based on the calculation of dilution time for a 100 gpm dilution flowrate for the Prairie Island large pool volume, over 56 hours6.481481e-4 days <br />0.0156 hours <br />9.259259e-5 weeks <br />2.1308e-5 months <br /> are available for detection and response. This is about three times as long as the detection / response time calculated for the best estimate case of the composite plant.

Thus, for this initiating event, Prairie Island is bounded by consideration of this event for the composite plant.

For the small pool volume, the CCW leak is a'so considered as an initiating event. For consideration of the smaller pool volume, and based on a dilution flow of 100 gpm, approximately 6 hours6.944444e-5 days <br />0.00167 hours <br />9.920635e-6 weeks <br />2.283e-6 months <br /> are available to detect and respond to the dilution event.

Considering that the small pool volume configuration exists only approximately 1 week per year, and that the larger pool volume configuration exists the remaining 51 weeks of the year, the CCW leak event frequency can be calculated and compared to the composite plant. For the large pool volume, the event frequency is calculated to be j 5.5E-9/ reactor year. For the small pool volume, the event frequency is calculated to be 8.5E 8/ reactor year. Considering the pool configurations noted above, the overall frequency of this event would then be:

1/52 (8.5E-8) + 51/52 (5.5E-9) = 7.0E-9/ reactor year.

Thus the Prairie Island CCW leak frequency remains less than that calculated for the composite plant (1.5E-8/ reactor year).

l 6

l l

b y 2 l 9: ,

l

.o l

p,.,n. w.nd svu.oo - W j 2.2.3 Seismic Event  !

i For Prairie Island, there are several differences for this event. Specifically, the  !

L initiating event frequency and several of the top events are impacted. This is~ l discussed in the following paragraphs. .

For a Prairie Island seismic event, the safe shutdown earthquake is 0.12g (Reference ,

3). From Reference.4 (NUREG-1488), the mean frequency of exceedance for a 0.12g l r

earthquake is approximately 7.3E-5/ reactor year. Thus the seismic initiating event j frequency for Prairie Island is less than the seismic initiating event frequency for the .j composite plant. l t

I For a seismic event at Prairie Island, if offsite power is available, a rupture of the 3/4" j l

demineralized water piping would cause a 20 gpm dilution rate.. If offsite power is not : 2 available, the domineralized water system would not operate and thus there would be no dilution source. For the composite plant, the dilution rate is assumed to be 200

[

gpm with offsite power available, and 100 gpm with offsite power not available. j With respect to operator detection and response time,. calculations performed in Section 2.3 indicate that for consideration of the large pool volume, over 280 hours0.00324 days <br />0.0778 hours <br />4.62963e-4 weeks <br />1.0654e-4 months <br /> are available. This is much greater than the time calculated for the composite plant.

The remaining top events for the seismic event of the composite plant are applicable to Prairie Island. Based on the above, the Prairie Island seismic event with the large pool volume is bounded by the seismic event considered for the composite plant.

For the small pool volume, the seismic event is considered as an initiating event. For 7 l l

l

,I

-e  ;

l Preene latend Evolustion FedRarrt consideration of the smaller pool volume, and based on a dilution flow of 20 gpm, -

approximately 45 hours5.208333e-4 days <br />0.0125 hours <br />7.440476e-5 weeks <br />1.71225e-5 months <br /> are available to detect and respond to the dilution event. This ,

operator detect and response time remains longer than the time calculated for the composite plant. Thus, a seismic event for the small pool volume is also bounded by  ;

consideration of the seismic event for the composite plant.

2.2.4 Tornado Event A tornado event is not considered as an initiating event at Prairie Island. The tornado is modeled as causing a loss of offsite power. The only piping that could rupture in the vicinity of the Prairie Island spent fuel pool is 3/4" piping for the demineralized water system. The pumps for the demineralized water system would not operate following the loss of offsite power. Therefore, even if this piping were ruptured, the >

demineralized water system would not deliver unborated water to the spent fuel pool.

2.2.5 Random Pipe Breaks For Prairie Island, the following differences from the event considered in the Reference 1 analysis are noted.

For Prairie Island, there are only 2 pipe sections in the vicinity of the spent fuel pool.

The composite plant considered 50 pipe sections. Thus using the same methodology, the initiating event frequency is reduced by a factor of 1/25; the random pipe break frequency for Prairie Island is calculated to be 5.3E-6/ reactor year. This is much less than the 1.3E-4/ reactor year frequency calculated for the composite plant. I For the accident sequence model of the composite plant, a split fraction exists for I

8

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

W a, j F

' Praene leland Evaluation RwRsxst '

i modeling the probability of a break of safety related piping (and thus a larger dilution  ;

rate). Thus, the composite plant assumed a 500 gpm dilution flow rate for a safety i related piping break, and a 100 gpm dilution flow rate for a non-safety related piping break. For Prairie Island, there is no safety related piping in the vicinity of the spent  ;

i fuel pool, only the domineralized water system. As noted in Table 2-1, the dilution j flow rate for the domineralized water system is 20 gpm. This is much less than the _

l dilution flows assumed for the composite plant. As a result, with respect to operator detection and response time, calculations performed in Section 2.3 indicate that for .

consideration of the large pool volume, over 280 hours0.00324 days <br />0.0778 hours <br />4.62963e-4 weeks <br />1.0654e-4 months <br /> are available at Prairie Island.

This is much greater than the time calculated for the composite plant. 3 i

The remaining top events for the random pipe break event of the composite plant are l applicable to Prairie Island. Based on the above, the Prairie Island random pipe break ~  !

with the large pool volume is bounded by the seismic event considered for the )

composite plant. i I i

) For the small pool volume, the random pipe break event is considered as an initiating event. For consideration of the smaller pool volume, and based on a dilution flow of j l

20 gpm, approximately 45 hours5.208333e-4 days <br />0.0125 hours <br />7.440476e-5 weeks <br />1.71225e-5 months <br /> are available to detect and respond to the dilution l

event. This operator detect and response time remains longer than the time calculated for the composite plant. Thus, a random pipe break event for the small pool volume is also bounded by consideration of the random pipe break event for the composite plant.

l 1

1 i

l 1

'e s

~,,

g.

e.

Praene loland Evaluation FeudRsxrt r

2.2.6 Demineralizer Valves / Makeup Valves Open For Prairie Island, spent fuel pool dilution events due to the mispositioning of these valves are discussed below, and the initiating event frequencies are determined.

Misationment of Valves Interfacino with the Soent Fuel Pool Normal makeup is via the chemical and volume control system. This manual connection through a 2 inch line is on the downstream side of the boric acid blender and is isolated by 2 manual valves (VC-11-59 and SF-14-4). VC-11-59 is a normally closed valve that isolates a line that supplies the spent fuel pool system along with the RWST and holdup tank through a check valve. SF-14-4 is a normally closed valve that specifically isolates the spent fuel pool cooling system from the CVCS. Thus, there are two normally closed manual valves that must be opened and left open to allow a dilution event via the CVCS.

The CVCS also connects to the spent fuel pool cooling system at a 4 inch line from the inlet to the 121 spent fuel pool heat exchanger to the CVCS holdup tank. This connection is normally isolated and is used to transfer water from the spent fuel pool to the CVCS system. The holdup tank has a capacity of 67,081 gallons. If the normally closed valve either fails open or is inadvertently left open, then this results in a loss of inventory from the spent fuel pool and not a dilution. l l

The next connection is from the holdup tank recirculation pump discharge to the transfer canal suction / discharge piping. This is a normally isolated 4 inch line that is a second source of supply to the pools / transfer canal. SF-9-6 is a normally closed valve used to isolate the connection. The CVCS holdup tank recirculation pumps take 10

y e e-

' Prairie island Evaluation reushput suction off the CVCS holdup tanks. These tanks have a capacity of 67,081 gallons.

This connection is also a designated supply in a loss of spent fuel pool inventory i event. l l

The fourth and final CVCS connection is from the CVCS monitor tanks through a 2 inch line that has 4 isolation valves, 3 of which are normally closed. This path is specified as a makeup supply in case of a loss of inventory in the spent fuel pools.

The tank has a capacity of 10,000 gallons.

The reactor water makeup system connects to the spent fuel pool cooling system  ;

directly at the outlet of the 121 spent fuel pool heat exchanger via a 2 inch line and indirectly via the boric acid blender. The direct connection has a normally closed i manual valve and a check valve. It is used as a potential supply path to the pools if CVCS is not available.

The RWST at both units connects to the system through separate 2 inch inlet and  !

outlet lines. These connections are normally used to purify the RWST water when the purification loop is isolated from the spent fuel pool cooling system. The inlet comes from the bottom of the tank through the reactor water purification pumps, to the inlet of the spent fuel pool cooling system purification loop. Each unit's inlet to the RWST is isolated by a normally closed manual valve (SF-14-12 and SF-14-13). There is also a check valve in each line. The outlets from the spent fuel pool cooling system to the RWSTs are isolated by normally closed valves SF-14-20 and SF-14-21 and connect to the tank at the top. The tank has a capacity of approximately 275,000 gallons.

This connection is specified as a source of water in a loss of inventory event. The boron concentration of the RWST is typically 3250 ppm.

11 I

e e.

e 0

Prains toland Evaluation FedRant The domineralized water system connects directly to the spent fuel pool cooling system at the purification loop demineralizer through a 2 inch line that is isolated by a normally closed manual valve. The connection is normally used to sluice and refill the demineralizer during a resin changeout. The domineralizer has two additional connections. The first is a resin flush line that is connected through a normally closed manual valve, SF-14-16, to a 2 inch spent resin header which in turn connects to the 121 spent resin tank via a check valve. The second is a resin fill line that is a flanged connection, only opened when adding new resin to the demineralizer.

There are also 5 demineralizer water hose connections in the vicinity of the pools, only one of which is directly next to the pool and is therefore specified as a source of water in case of a loss of inventory event.

During normal operation, evaporative losses will require that makeup be provided to the spent fuel pool. The resins in the demineralizer tank may have to be flushed or changed periodically. If valves are misaligned during either operation, it is possible that unborated water could be accidently delivered to the spent fuel pool. Manual valve alignments are required for all makeup sources.

Since makeup from the CVCS holdup tanks or monitor tanks would only be used in an emergency and since the volume of water in these tanks is not sufficient to substantially dilute the spent fuel pool, only two cases are considered: the domineralizer valves open and the makeup valves open.

12

(. -'

n Pratte Island EvalustlOn WRggst l

Demineralizer Valves Ooen (Demin Valves Ooen)  :

- The first case is failing to close the valve after flushing / sluicing the demineralizers.

It is assumed that this flushing / sluicing operation occurs twice a year. During this l

operation,'one valve is closed to isolate the domineralizer tank and the valve to provide demineralizer water for flushing or sluicing is opened. A dilution path to the _ .

spent fuel pool is created after the operation is completed, if the closed valve is opened (no error), but the wrong valve is closed to isolate the domineralized water.

Failure of the operator to close one valve, after it has been opened to the spent fuel  !

poolis evaluated as 8E-4 in Appendix B of WCAP 14181 This represents the failure l to close an isolation valve (and verify that it is closed). The frequency of a valve being left open is: '

2/yr

  • 8.0E-4/ year = 1.6E-3/vear. l l

Makeuo Valve Ooen f i

The second scenario is following makeup. Makeup is generally via the blender (with reactor makeup water or borated water via the blender) which requires opening two valves to the spent fuel pool. Failure of the operator to close the first valve, after .;

both valves have been opened to the spent fuel pool is evaluated as 8E-4. This represents the failure to close an isolation valve (and verify that it is closed). There 1

are two valves, so a conditional failure probability of 0.5 that the second valve fails is used (high dependency from Table 10-4, NUREG/CR-1278, Reference 5). This is 8.0E-4

  • 0.5 = 4.OE-4. Makeup to the spent fuel poolis provided about every two I

weeks so that the valves isolating the spent fuel pool will be opened at least 24 times a year. During this time interval, the boron in the pool could only be diluted if sufficient unborated water is delivered to the pool. If an unborated water source is 1

not in use in the CVCS, then even if the valves are open, the spent fuel pool will not 13 l

^

e .  !

i Prm,ie Island Evaluation Frd%nt

- be diluted. During normal CVCS operation, unborated water is delivered to the reactor  ;

coolant system to allow load follow, or to slowly dilute the boron concentration to  !

remain critical, or to return to criticality after refueling (startup). During the two week .

time interval the valve could be open to the spent fuel pool, it is not likely that  !

sufficient undiluted water would be delivered via the reactor makeup system to dilute  !

the pool. .The valve would then be verified again when makeup is required. An l inadvertent dilution of the reactor coolant system (failure in the makeup system, etc.)  ;

would be identified and boration initiated, so the scenarios describing inadvertent

]

dilution of the RCS will be bounded by the deliberate dilutions of the RCS. Itis assumed that a major dilution of the reactor coolant system (startup after shutdown, or cold shutdown) occurs at least three times a year. The frequency of a boron  ;

dilution event following makeup is then:

3/yr* 4.0E-4 = 1.2E-3/ reactor year. i Note that this is the event frequency used in the composite plant, as well, for makeup l l

valves open.

The event tree and top event descriptions as discussed in Reference 1 for the composite plant are applicable to Prairie Island. In addition, the assumption of a 100 gpm leak rate is maintained. However, operator action time is different for Prairie Island. Specifically, for the large pool volume at Prairie Island, over 56 hours6.481481e-4 days <br />0.0156 hours <br />9.259259e-5 weeks <br />2.1308e-5 months <br /> (see Table 2-2) are available for the operator to detect and respond to the dilution event.

This is a longer time than calculated for the composite plant.

Thus it may be concluded that for the makeup valve open initiating event, Prairie Island is bounded by the results of the composite plant. However, for the demineralized valve open initiating event, the Prairie Island initiating event frequency l

14 1

l

. l Prairie telend tvetumelan W Rsot is calculated to be higher (1.6E-3/ reactor year vs. 8.0E-4/ reactor year). However, the longer operator detect and response time will result in a conditional probability on the ,

I operator detect (DETECT LATER) and response (OPERATOR RESPONDS) top events; j specifically, a factor of 0.1 is used to consider the additional time available. This l conditional probability for DETECT LATER and/or OPERATOR RESPONDS offsets the larger initiating event frequency for demin valves open at Prairie Island. Thus it may be concluded that Prairie Island is bounded by the results of the composite plant.

1 Boron dilution via operator error to close the domin or makeup valve is not considered as an initiating event for the small pool. Since the small pool would be isolated during cask handling, the cooling system would not be in operation and the piping valved out; therefore, no dilution is possible from the demineralizer system or the makeup system.

2.2.7 Initiating Event Results Based on Sections 2.2.1 - 2.2.6, it can be concluded that the Prairie Island spent fuel pool boron dilution event frequency is bounded by the frequency calculated for the composite plant using the generic methodology presented in Reference 1.

2.3 Boron Dilution Times and Volumes For Prairio Island, the normal boron concentration maintained in the spent fuel poolis 3250 ppm. Based on the Prairie Island criticality analysis, the soluble boron i l

concentration required for criticality requirements (i.e., a k,, _< 0.95) is 1050 ppm.

As noted in Section 2.2, this requiremont is conservatively increased to 1380 ppm for 1

15 i

. . - . >. , - ~ . ~ . .- .. . .~ - . . . - - , . . . . - - -

.e. .. ,

., a l

V FndRgxst Prsme intend Evolustion {

1 the purposes of this evaluation. These will be ~the endpoints considered for the i

. deterministic evaluation of dilution volume and time;' thus a (3250 - 1380), or 1870 t

. ppm boron. dilution event is considered. This amount of dilution is greater than j considered for the best estimate case (820 ppm) in the generic methodology.  ;

The dilution volumes and times for these scenarios are calculated based 'on the - .

following equation: I 9

t ona = In (C, /Con, )V/O (Equation 1)

Where: ,

C,is the boron concentration of the pool volume at the beginning of the event Cono = the boron endpoint concentration Q = dilution rate (gallons of water / minute) i V = volume (gallons) of spent fuel pool.

P The time to dilute depends on the initial volume of the pool and the postulated rate of dilution. There are two spent fuel pools, a small one and a larger one. Both pools and the transfer canal are generally opened to each other. The small pool is isolated 1

for cask handling operations and the volume in the small pool is reduced by lowering

]

the water level by 10 feet. The volumes of the pools used to calculate the dilution i times are: I both pools and the transfer canal: 395,000 gallons )

small poolisolated and volume reduced by 10 feet: 63,000 gallons )

16

-+ e i e

a' Praene leland Evaluation NW '

Equation 1 is used to calculate the dilution time for a range of dilution rates from 20 ,

gpm to 250 gpm for the two spent fuel volumes given above. Table 2-2 shows the Prairie Island dilution time and volume data for a dilution event from 3250 ppm to 1380 ppm, as discussed previously. Table 2 3 shows the dilution time and volume data calculation for the composite plant dilution event, a dilution from 2200 ppm to 1380 ppm.

TABLE 2-2 Prairie island Dilution Time & Volume Data Dilution Event: 3250 ppm to 1380 ppm Dilution Large Pool Volume Small Pool Volume Flow Rate 395,000 gal 63,000 gal Dilution Time Dilution Vol Dilution Time Dilution Vol (br) (gal) (hr) (gal)  ;

20 gpm 282 338,400 45 54,000 I

50 gpm 112 18 100 gpm 56 9 200 gpm 28 4 250 gpm 22 3 1

17

i

,a*

P,awse leland Evalustson rni%rrt TABLE 2-3 Composite Plant Dilution Time & Volume Data Dilution Event: 2200 ppm to 1380 ppm Pool Volume = 232,000 gal Dilution Flow Rate Dilution Time Dilution Volume (hrs) (gal) 20 gpm 90 108,000 50 gpm 36 100 gpm 18 200 gpm 9.0 250 gpm 7.2 500 gpm 3.6 2.3.1 Consideration of Dilution Volumes As can be seen in the summary tables from Section 2.3, a large volume of water is necessary at Prairie Island, versus the composite plant, for a spent fuel pool boron dilution event. For a dilution event from the nominal spent fuel pool boron concentration to a boron endpoint concentration of 1380 ppm, a dilution volume of nearly 340,000 gallons is required. If the boron endpoint concentration calculated in Reference 2,1050 ppm, is used as the boron endpoint concentration, the dilution volume increases to over 445,000 gallons.

To assess the potential of a spent fuel pool boron dilution event at Prairie Island, the water available to dilute the spent fuel pool will be determined and compared to the dilution volume. The Prairie Island dilution sources are summarized in Table 2-4.

18

6 a Prairee Beland Evaluation FrutReptst TABLE 2-4 Prairie Island Dilution Sources mammmmmmmmmmmmmmmmme Dilution Source Quantity Available Water Total Water Volume, gal each Volume, gal CVCS Holdup Tank 3 65,824 197,472 l CVCS Monitor Tank 3 10,000 30,000 Demin Water via 4 46,000 184,000 Reactor Makeup Tank Component Cooling 2 1000 + ' 2000 + '

Water n  ;

1 Surge Tank with normal water volume of 1000 gallons, plus system piping i

l As can be seen from the available water volumes at Prairie Island, there is no single I source of water which can provide the quantity of water necessary for a dilution event. Although the CCW system and the other tanks have makeup capability from other systems, detection of a dilution event via level alarms and/or visualinspections i would be expected long before a sufficient dilution would occur.

1 I

I l

l 19

a e.  !

'4 ,

Praine telend Evaluation Fhd%nt

3.0 CONCLUSION

S i i

Based on the above, it is concluded that the results of the PRA completed for the f composite plant bound Prairie Island; thus the spent fuel pool boron dilution event  ;

frequency for Prairie Island is less than the NRC Safety Goal Policy Statement target ,

frequency risk level objective of 1.0E-6/ reactor year. Furthermore, evaluations show .

that a large volume of water is necessary to dilute the spent fuel pool to the soluble boron concentration required for criticality requirements at Prairie island (1050 ppm),

or the boron endpoint concentration considered for this evaluation (1380 ppm). As shown in Section 2.3, there is no single source of water which can provide this '

quantity of water necessary for a dilution event. Furthermore, since such a large water volume turnover is required, the dilution event that is taking place would be l

readily detected and terminated by plant personnel. Evaluations indicate tens of hours are typically available to detect and respond to such an event. ,

f r

i I

l 20 l

l l

4

n L+. ,,

,l ,

4 P,mne telend Evolustion FedRunt r

4.0 ' REFERENCES '

_1. ' WCAP-14181, " Evaluation of the Potential for Diluting PWR Spent Fuel Pools,"

~ July 1995.' ]

2. Northern States Power Prairie Island Units 1 and 2. Spent Fuel Rack Criticality

[

c-Analysis with Credit for Soluble Boron, Westinghouse Commercial Nuclear Fuel )

Division, June 1995. i

3. Prairie Island Updated Final Safety Analysis Report. ') ,

l

4. NUREG-1488, " Revised Livermore Seismic Hazard Estimates for 69 Nuclea'r -

Power Plant Sites East of the Rocky Mountains," US - Nuclear Regulatory ,

Commission, Draft Report, October 1993.

3 1 5. NUREG/CR-1278, " Handbook of Human Reliability Analysis with Emphasis of'

~

j Nuclear Power Plant Application - Final Report," . US Nuclear Regulatory ;

Commission, August 1983.

1 i

l 1

d -

e

'21  :

- - _..___ ____2

. . ~

m y

- d)_'s : '

m (

1..
', M. [

Y Exhibit E Prairie Island Nuclear Generating Plant'

. License Amendment-Request Dated July 28, 1995-Information Copy Unit 1 Core-Operating Limit: Report Spent Fuel Pool. Storage. Operational Limits l

i

')

i 3

2 l

1 l

1 I

~. __ ._

p. - - - _ . . _

' g. g.

p'"4

, ..; 1 -UNIT 1 CORE OPERATING LIMITS REPORT ,

b

<Soent Fuel Pool Operatina Limits- ,

i Definitions:

i

.STD' Fuel - Westinghouse and Exxon standard fuel designs. -

E l TOPROD Fuel - Exxon TOPROD fuel design, 1

OFA Fuel - Westinghouse OFA and Vantage + fuel designs. ,

~

Spent Fuel Pool Storace' Configuration Limitations: ,

t To be stored without restriction in the spent fuel pool, the burnup and l

' initial enrichment of a fuel assembly shall be within the unrestricted range- l of the all cell storage burnup credit limits of Figure 7. The all cell  !

storage burnup credit limits are also shown in Table 1.

Fuel assemblies with a combination of burnup and initial' enrichment in the restricted .

range of Figure 7 shall be stored in accordance with Figures 8 and 9. .The.3x3.

checkerboard storage burnup credit limits are also shown in Table 2. Only=

unconsolidated fuel assemblies shall be stored in a 3x3 checkerboard storage array.

A spent fuel pool verification is required.following'the last movement of fuel i assemblies in the spent fuel pool, if fuel assemblies with a combination of burnup

.and initial enrichment in the restricted range of Figure 7 are stored in the fuel:

pool. This verification will confirm that any fuel assemblies with a combination;of i burnup and initial enrichment in the-restricted range of Figure 7 are stored in accordance with the limitations specified above.

Soent Fuel Pool Boron Concentration Limit: j The spent fuel pool boron concentration shall be 2'1,050 ppm when fuel-is. '!

stored in the spent fuel pool.

f 6

')

i 1

i

__~'

.t: f e c, Unit 1 Core Operating Licits. Report v.

Table 1 Spent Fuel Pool All Cell Storage Burnup Credit Limits I

Nominal W - OFA W - STD Enrichment - Burnup Burnup

, (w/o) (MWD /MTU) (MWD /MTU) l.85 0 0 1.95 0' 1471 2.00 687 2194  !

2.20 3360 4999 2.40 5922 7677 2.60 8381 10238 2.80 10748 12692 3.00 -13030 15051 3.20 15237 17325 ,

3.40 17379 19524 3.60 19464 21659 3.80 21502 23740 4.00- 23501 -25779 4.20 25471 n/a 4.40 27422 n/a 4.60 29361 n/a 4.80 31299 n/a 4.95 32757 n/a

e ,

, o; Unit 1 Core Operating Liinits Report s

Table 2 Spent Fuel Pool 3x3 Checkerboard Storage Burnup Credit Limits Nominal W - 0FA W - STD Exxon Enrichment Burnup Burnup TOPROD (w/o) (MWD /MTU) (MWD /MTU) (MWD /MTU) 1.38 0 0 0 1.46 0 1723 0 1.50 0 2567 949 1.60 1840 4629 3222 1.80 5361 8550 7371 >

2.00 8683 12220 11044 2.20 11821 15657 14308 2.40 14790 18881 17231 2.60 17606 21912 19878  ;

2.80 20282 24769 22316 3.00 22835 27473 24612  :

3.20 25279 30043 26833 3.40 27629 32498 29044 3.60 29900 34858 31313 3.80 32107 37144 33707.

4.00 34265 39374 36292

! 4.20 36389 n/a n/a 4.40 38494 n/a n/a 4.60 40595 n/a n/a 4.80 42708 n/a n/a 4.95 44308 n/a n/a I

-< e' Unit 1 Core Operating Limits Report 3

35000 OFA Fuel /

......... STD Fuel /

30000 ,

/

/

UNRESTRICTED >

/

, t 25000 ,'

D ,' >

/

> , t 2 '

l i5  : l

$ .* )

2 20000

,. /

/

E o' /

a >

tD

,o f x .

y 15000 ,.

/

e 8 o' /

< o

/

o a ,a

,/ .

10000 *

/

i o (

/ RESTRICTED  ;

/

5000 '

/

o' ) ,

f

' l l

o' / l 0

' I '

1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 235 Initial U Enrichment (nominal w/o)  :

l l

Figure 7 Spent Fuel Pool All Cell Storage Burnup Credit Limits i

F.

e e Unit 1 Core Operating Limits Report i I

45000 i

/ l

) i OFA Fuel >

/ I 40000 TOPROD Fuel /

........ STD Fuel .

/ i e' ,

r 35000 '

' / l

, r .i ,

ACCEPTABLE ,

e  !

^

f

./ / I F 30000 '

' ' / 1 Q

Q  ;

/

/

/

/ __

, / / i 2 / / '

' / / '

Eo 25000 ,

, f c '

/ /

Ei i / / i (D o' / /  !

I '

$ 20000 ,

', j E

o .

)

$ / / i k  ; /

/

Ts 15000 ,'/  :

a /

LL I / /

+

o, /

i' >

UNACCEPTABLE i 100.00  ;

,/ j l

t

( ,

il l 5000 ,

,v ;

'l i/

!  !) '

0 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 initial 235 U Enrichment (nominal w/o) l Figure 8 Spent Fuel Pool 3x3 Checkerboard Storage Burnup Credit Limits  !

I

- e- .e ; ,

, t Unit 1 Core Operating Limits Report  :

I i

r i

I 1

-- t E .

E ::e . .

E i L

r / ., '

/ "l 'h

-

  • l__ Y

?:v.4 , :ggg ,

~

s 2 :I : *

,  ; .. s, y. m :m 1

& 'k .

e , 4.a m .s v s  ;

;- ~h'jhl

' _:l'. i ^; _,. i.T:?:l5$. < .::Q}:-

w; . g. ,

Ag.Q

A M

~

h  :

-u.7.y s- < g g<.y I

c 1 Fresh Fuel: Must be less than or equal to 4.95 wIo 235 U

l Burned Fuel: Must satisfy the minimum burnup requirements of Figure 8 i

Figure 9 Spent Fuel Pool 3x3 Checkerboard Cell Layout i

e- *

  • a-ep--

'. U .~ 4 i e-p,.

i.

Exhibit F-O Prairie' Island Nuclear Generating Plant License Amendment Request-Dated July 28, 1995 Information Copy Northern States Power Prairie Island Units 1 and.2 Spent Fuel Rack

. Criticality Analysis With Credit for Soluble Boron

i

'b l

1 a

11 I

y.-]

t

-i 1

I

e

.l, l

1 CDB-95-094 l

Northern States Power Prairie Island Units 1 and 2 Spent Fuel Rack Criticality Analysis With Credit for l Soluble Boron l

l

)

June 1995 i W.D. Newmyer j i

Prepared : 2* >M ^ '

l I

W. D. Ne/myer7 Criticality Services Team Izader Verified: $ ' S^2^^~

S. Srinilta Core Design B Approved: N- I/A f" CE b l C. R. Savage, M&hager Core Design B

+

Westinghouse l Commerical Nuclear Fuel Division 4

6

Table of Contents 1.0 I n t ro d u ct io n . .... .... ...... ...... .. .. .... .. .... .. ........ ............ .. ........ ........ ........ .......... 1 1.1 Design Description.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..2 1.2 De s i g n Criteria . . .. .. . . . . . . . . . . . .. .. . . .. . . .. . . . . . . ... .. .. ...... . . . . .......................2 2.0 A n a l y I i ca l M et h od s .......... .... .................. .......... .................. ......................... 3 3.0 Criticality Analysis of All Cell Storage..................... .... .. ...................... 4 3.1 Maximum Feasible K err Calculations ........ ...... ... .... .......... ..... ...... ........ . . . ... . A 3.2 Soluble Boron Credit K err Calculations....... ....... .. ..................................5 3.3 B urnup Credit Reactivity Equivalencing ......... ... ................... ..... .... ..... . .. . . ... 7 4.0 Criticality Analysis of 3x3 Checkerboard Storage ...... .. .. ............ .... 9 4.1 Maximum Feasible K rfeCalc ulatio n s . .. .. . .. . . .. . . .. .. .... . . .. .. .. .. . . . . .. ... . .... . .. . .. . .. ... . . . 9 4.2 Soluble B oron Credit K err Calculations ... ......... .. ........ ............ ....................... .... 1 1 4.3 B urnup Credit Reactivity Equivalencing ......... ........ ................ ...... ......... ... .. ....12 5.0 Discussion of Postulated Accidents.... ............ .......... ......... .... .. .... .. ..14 6.0 Soluble Boron Credit Su mma ry ...... .. ...... .. ...... .. ......... ..........16  :

I 7.0 Summa ry of Criticality Results ... .... .. .. .. .. .... .. ................17 B i b l i og ra p h y .. .................. .... ...... .... .... .. .... .... .... .. .... .... ........ ...... 31 f

l l

l i

Prairie Island Spent Fuel Racks

List of Tables Table 1. Fuel Parameters Employed in the Criticality Analysis...... . . . . . . . . . . 18 Table 2. Praide Island All Cell Storage Soluble Boron Credit Ke rr .. . ... .... . .19 Table 3. Prairie Island All Cell Minimum Bumup Requiiements... . . .. .. . . . . . . .20 Table 4. Prairie Island 3x3 Checkerboard Storage Soluble Boron Credit Ke rr.. . . ..21 Table 5. Prairie Island 3x3 Checkerboard Minimum Burnup Requirements ... . . . . .22 Table 6. Summary of Soluble Boron Credit Requirements.... . .... . . . . . . . . . .. . 23 4

i Praide Island Spent Fuel Racks ii i

r, 1

. 1 t

i List of Figures  ;

i Figure 1. Prairie Island Spent Fuel Rack Layout .................. ............................... ... .... .. 24 l Figure 2. Prairie Island Spent Fuel Storage Cell Nominal Dimensions.. ..... .. .. . . . .. 25 f Figure 3. ' Prairie Island All Cell Storage Burnup Credit Requirement ..... ....................26 i Figure 4. Prairie Island 3x3 Checkerboard Layout Requirement............... . . ..... . .. . .. 27 I

' Figure 5. Prairie Island 3x3 Checkerboard Storage Burnup Credit Requirement.. ... . . .. 28 ,

Figure 6. Prairie Island All Cell Soluble Boron Worth......... ..... . . ...... ... . .. . . . . . .. . 29 l Figure 7. Prairie Island 3x3 Checkerboard Soluble Boron Worth ............ ........ .. ... . . . 30 l

'i i

i 4

'i t

I

\

i k

i i

c

. i Prairie Island Spent Fuel Racks iii

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

e .

1.0 Introduction This report presents the results of a criticality analysis of the Northern States Power Prairie Island Units I and 2 spent fuel storage racks with credit for soluble baron in the spent fuel pool. The methodology employed here is contained in the topical report, " Westinghouse Spent Fuel Rack Criticality Analysis Methodology"m The spent fuel storage rack design considered herein is an existing array of fuel racks. previously qualified (2) for storage of various 14x14 fuel assembly types with maximum enrichments up to 235 U. Two different storage configurations are currently allowed. The first configuration 5.0 w/o allows fuel assemblies to be stored in a 2x2 checkerboard pattem of " burned" and " fresh" fuel 235 U (equivalent with burnup) and 5.0 w/o 235 U (no assemblies with enrichments of 2.5 w/o burnup), respectively. The second configuration allows storage of fuel assemblies in all storage cell locations (no checkerboard) if they satisfy a minimum burnup credit requirement as a function of enrichment. The spent fuel rack Boraflex poison panels were considered in this analysis.

zed to allow storage of all 14x14 fuel The Prairie Island spent fuel racks are being assemblies with nominal enrichments up to 4.95 w/o 23 reanalyU in all storage celllocations using credit for checkerboarding and burnup. The analysis will also ignore the presence of the spent fuel rack Boraflex poison panels. The following* storage configurations and enrichment limits are considered in this analysis:

All Cell Storage Storage of 14x14 assemblies in any cell location with nominal 235 U for Westinghouse Enrichment Limits enrichments no greater than 1.95 w/o 235 U for Westinghouse 14x14 OFA fuel assemblies and 1.85 w/o 14x14 STD and Exxon 14x14 fuel assemblies. Fuel assemblies with initial nominal enrichments greater than these must satisfy a minimum burnup requirement. The soluble boron credit required for this storage configuration is 600 ppm for Westinghouse 14x14 OFA fuel assemblies and 700 ppm for Westinghouse 14x14 STD ,

and Exxon 14x14 fuel assemblies.

3x3 Checkerboard Storage of Westinghouse 14x14 OFA assemblies with nominal 235 enrichments no greater than 4.95 w/o U in the center of a 3x3 Enrichment Limits checkerboard. The surrounding fuel assemblies must have an 235 initial nominal enrichment no greater than 1.50 w/o U for Westinghouse 14x14 OFA fuel assemblies,1.46 w/o 235U for 235 Exxon TOPROD fuel assemblies, and 1.38 w/o U for Westinghouse 14x14 STD and other Exxon fuel assemblies. Fuel assemblies with initial nominal enrichments greater than these 4 must satisfy a minimum burnup requirement. The soluble boron credit required for this storage configuration is 950 ppm for Westinghouse 14x14 OFA fuel assemblies and 1050 ppm for ,

Westinghouse 14x14 STD and Exxon 14x14 fuel assemblies.

1 Prairie Island Spent Fuel Racks

u..  :.

x The Prairie _ Island spent fuel rack analysis is based on maintaining Keff < l.0 under maximum feasible conditions with no soluble boron. - Soluble boron credit is used to provide safety margin by maintaining Keft 5 0.95 including uncertainties, tolerances and accident conditions in the -

presence of spent fuel pool soluble boron.

Fuel types being considered in the analyses include the current Westinghouse 14x14 OFA design being used in Prairie Island Units 1 and 2 and the Westinghouse 14x14 STD and Exxon 14x14 fuel assembly types previously used in the reactors and currently in storage in t! e Prairie Island.

spent fuel pool.

1.1 Design Description The Prairie Island spent fuel storage rack layout is depicted in Figure 1 on page 24 and the spent .

fuel rack storage cell is shown in Figure 2 on page 25 with nominal dimensions provided on each figure.

The fuel parameters relevant to this analysis are given in Table 1 on page 18.

1.2 Design Criteria Criticality of fuel assemblies in a fuel storage rack is prevented by the design of the rack which' limits fuel assembly interaction. This is done by fixing the minimum separation between fuel assemblies and inserting neutron poison between them.

In this report, the reactivity of the spent fuel rack is analyzed such that K eg remains less than 1.0 under maximum feasible conditions with no soluble boron as defined in Reference 1. To provide safety margin in the criticality analysis of the spent fuel racka, credit is taken for the soluble boron present in all PWR spent fuel pools. This parameter provides significant negative reactivity in the  ;

criticality analysis of the spent fuel rack and will be used here to offset the reactivity increase when ignoring the presence of the spent fuel rack Boraflex poison panels. Soluble boron credit provides sufficient relaxation in the enrichment limits of the spent fuel racks to allow the racts to j be used under checkerboarded conditions with no credit for the Boraflex poison panels. If some '!

amount of Boraflex material is considered remaining, the reactivity of the spent fuel rack and tiic l amount of soluble boron required to maintain K,g 50. 95 will be reduced. l The design basis for preventing criticality outside the reactor is that, including uncertainties, there is a 95 percent probability at a 95 percent confidence-level that the effective neutron multiplication factor, K,g, of the fuel rack array will be less than or equal to 0.95. This

^ requirement as currently stated in ANSI 57.2-1983W, and NRC position paperW does not allow

. for reactivity credit due to the presence of soluble boron. This criticality analysis report will tak e exception to this and show that the effective neutron multiplication factor, K,g, of the fuel rack array is less than 1.0 under maximum feasible conditions and less than or equal to 0.95 when credit is taken for the presence of spent fuel pool soluble boron.

Prairie Island Spent Fuel Racks 2

2.0 Analytical Methods The criticality calculation method and cross-section values are verified by comparison with critical experiment data for fuel assemblies similar to those for which the racks are designed. This benchmarking data is sufficiently diverse to establish that the method bias and uncertainty will apply to rack conditions which include strong neutron absorbers, large water gaps, low moderator densities and spent fuel pool soluble boron.

The design method which insures the criticality safety of fuel assemblies in the fuel storage rack is described in detail in the Westinghouse Spent Fuel Rack Criticality Analysis Methodology topical reportU ). This report describes the computer codes, benchmarking, and methodology which are used to calculate the criticality safety limits presented in this report for Prairie Island.

As determined in the benchmarking in the topical report, the method bias using the described methodology of NITAWL-II, XSDRNPM-S and KENO-Va is 0.0077 AK with a 95 percent probability at a 95 percent confidence level standard deviation on the bias of 0.0030 AK. These values will be used throughout this report as needed.

I l

l I

1 3

Prairie Island Spent Fuel Racks

3.0 Criticality Analysis of All Cell Storage This section describes the analytical techniques and models employed to perform the criticality analysis and reactivity equivalencing evaluations for the Prairie Island spent fuel storage racks all cell enrichment limits with credit for soluble boron.

Section 3.1 describes the maximum feasible Ke g KENO-Va calculations performed for the all cell storage configuration. Section 3.2 discusses the results of the spent fuel rack Keg soluble boron credit calculations. Finally, Section 3.3 presents the results of calculations performed to show the minimum burnup requirements for assemblies with higher initial enrichments above those determined in Section 3.1.

3.1 Maximum Feasible K err Calculations The following assumptions are used to develop the maximum feasible KENO-Va model for storage of fuel assemblies in the Prairie Island spent fuel storage rack:

1. The fuel assembly parameters relevant to the criticality analysis are based on the Westinghouse 14x14 OFA and 14x14 STD designs (see Table 1 on page 18 for fuel parameters). The Westinghouse 14x14 STD design bounds the reactivity of the Exxon 14x14 fuel assemblies.
2. Westinghouse 14x14 OFA and STD fuel assemblies contain uranium dioxide at a nominal enrichment of 1.95 w/o and 1.85 w/o, respectively, over the entire length of each rod.
3. The fuel pellets are modeled assuming nominal values for theoretical density and dishing fraction.
4. No credit is taken for any natural or reduced enrichment axial blankets. This assumption results in equivalent or conservative calculations of reactivity for all fuel assemblies used at Prairie Island including those with annular pellets at the fuel rod ends.

236

5. No credit is taken for any 234 U or U in the fuel, nor is any credit taken for the buildup of fission product poison material.
6. No credit is taken for any spacer grids or spacer sleeves.
7. No credit is taken for any burnable absorber in the fuel rods.  !
8. No credit is taken for the presence of spent fuel rack Boraflex poison panels. The Boraflex volume is replaced with water.
9. The moderator is water with 0 ppm soluble boron at a temperature of 68'F. A water density of 1.0 gm/cm3 is used.
10. The array is infinite in lateral (x and y) extent and finite in axial (vertical) extent. )
11. All available storage cells are loaded with fuel assemblies.

I Prairie Island Spent Fuel Racks 4

L

~.

9 a

With the above assumptions, the KENO-Va calculations of Keg under normal conditions resulted in a K,g of 0.98104 and 0.97700 for both Westinghouse OFA and STD fuel assemblies, respectively. The reactivity bias calculated for the normal temperature range of the spent fuel pool water (50*F to 150*F) is 0.00602 and 0.00677 AK for Westinghouse OFA and STD fuel assemblies, respectively. Finally, the methodology bias associated with the benchmarking of the-

[ Westinghouse criticality methodology is 0.0077 AK.

l ' Based on the above results, the following equation is used to develop the maximum feasible Ke g

( for the Prairie Island spent fuel storage racks:

1 Kag = K,,,,,, ,1 + B,,,,,, + B,,,h,,

l where:

K,,,,m a t =

normal conditions KENO-Va Ke g l B,,,,,y

=

temperature bias from 50*F to 150*F B ,,,s,s meM Nas kend ham &ndmd scal j comparisons  ;

Substituting calculated values in the order listed above for Westinghouse OFA fuel, the result is: i i

I K,g = 0.98104 + 0.00602 + 0.00770 = 0.99476  !

Substituting calculated values in the order listed above for Westinghouse STD fuel, the result is:

K,g = 0.97700 + 0.00677 + 0.00770 = 0.99147 Since Keg is less than 1.0 for both fuel types, the Prairie Island spent fuel racks will remain subcritical under maximum feasible conditions when all cells are loaded with 1.95 w/o Westinghouse OFA and 1.85 w/o Westinghouse STD fuel assemblies and no soluble boron is present in the spent fuel pool water. In the next section, soluble boron credit will be used to provide safety margin by determining the amount of soluble boron required to maintain Ke g 5 0.95 including tolerances and uncertainties.

3.2 Soluble Boron Credit K err Calculations To determine the amount of soluble boron required to maintain Ke g s 0.95,' KENO-Va is used to establish a nominal reference reactivity and PHOENIX-P is used to assess the effects of material and construction tolerance variations. A final 95/95 Keff is developed by statistically combining the individual tolerance impacts with the calculational and methodology' uncertainties and .

I summing this term with the nominal KENO-Va reference reactivity.

Prairie Island Spent Fuel Racks 5

7

> _, l I f .1 v.: .  ;

The assumptions used to develop the nominal case KENO-Va model for soluble boron credit for '

. all-cell storage in 'the Prairie Island spent fuel racks are similar to those in Section' 3.1 except for  !

assumption 9 regarding the moderator soluble boron concentration. The moderator was replaced l with. 200 ppm and :250 ppm for the Westinghouse OFA and STD fuel assembly types, j respectively.

l With the above assumptions, the KENO-Va calculation for the nominal case results in a Ke g of )

J0.91893 and 0.91341 for Westinghouse OFA and STD fuel assembly types, respectively.: j

-I Calculational and methodology biases must be considered in the final Keg summation prior too i comparing against the 0.95 Ke rr limit. The following biases are included:' j Methodology: The benchmarking bias as determined for the Westinghouse KENO-Va 'l methodology was considered. j Water Temperature: A reactivity bias is applied to account for the effect of the normal range l of spent fuel pool water temperatures (50*F to 150*F). j '

To evaluate the reactivity effects of possible variations in material characteristics _ and _

mechanical / construction dimensions, PHOENIX-P perturbation calculations are performed. For the Prairie Island spent fuel rack all-cell enrichment storage con 6guration.- UO2 material-tolerances are considered along with construction tolerances related te the cell 1,D., storage cellt  ;

pitch, and stainless steel wall thickness. Uncertainties associated with. calculation and l methodology accuracy are also considered . in the statistical summation of uncertainty j components. j i

The following tolerance and uncertainty components are considered in the total uncertainty l statistical summation: -t 235 I U Enrichment: The standard DOE enrichment tolerance ofi0.05 w/o 235U about the -

nominal reference enrichments of 1.95 and 1.85 w/o 235U was considered. f UO 2Density: A 12.0% variation about the nominal reference theoretical density'(the nominal .

)

reference values are listed in Table 1 on page 18) was considered. l J

Fuel Pellet Dishing: A variation in fuel pellet dishing fraction from 0.0% to 2.0% (the nominal j

reference values are listed in Table 1 on page 18) was considered.

l Storage Cell I.D.: The 10.10 inch tolerance about the nominal 8.27 inch reference cell I.D.was -i considered.

Storage Cell Pitch: The 10.06 inch tolerance 'about the nominal 9.50 inch reference cell pitch' ]

' was considered. 1 i

Stainless Steel Thickness: The i0.01 inch tolerance about the nomina 10.09 inch reference ni stainless steel thickness for all rack structures was considered.

1 i

i i

[

l Prairie Island Spent Fuel Racks 6- ]

l

- -. = . . - . . . - . - - - -. w

ie' Assembly Position: The KENO-Va reference reactivity calculation assumes fuel assemblies are symmetrically positioned within the storage cells. Conservative calculations show that an increase in reactivity can occur if the comers of four fuel assemblies are positioned together.

This reactivity increase was considered in the statistical summation of spent fuel rack tolerances.

Calculation Uncertainty: The 95 percent probability /95 percent confidence level uncertainty on the KENO-Va nominal reference K err was considered.

Methodology Uncertainty: The 95 percent probability /95 percent confidence uncertainty m the benchmarking bias as determined for the Westinghouse KENO-Va methodology was considered.

The maximum K err for the Prairie Island spent fuel rack all cell storage configuration is developed by adding the temperature and methodology biases and the statistical sum of independent uncertainties to the nominal KENO-Va reference reactivity. The summation is shown m Table 2 on page 19 and results in a maximum Keg of 0.94769 and 0.94220 for Westinghouse OFA and STD fuel assembly types, respectively.

Since Ke g is less than 0.95 including soluble boron credit and uncertainties at a 95/95 probability / confidence level, the acceptance criteria for criticality is met for the all cell enrichment storage of 14x14 fuel assemblies in the Prairie Island spent fuel racks. Storage of fuel 235 assemblies with nominal enrichments up to 1.95 and 1.85 w/o U is acceptable for Westinghouse OFA and STD fuel assembly types, respectively, in all cells in the Prairie Island spent fuel racks including the presence of 200 and 250 ppm.

3.3 Burnup Credit Reactivity Equivalencing j 235 Storage of fuel assemblies with enrichments higher than 1.95 and 1.85 w/o U for the i

Westinghouse OFA and STD fuel types in the Prairie Island spent fuel rack all cell configuration i

is achievable by means of the concept of reactivity equivalencing. The concept of reactivity equivalencing is predicated upon the reactivity decrease associated with fuel depletion. For j

. burnup credit, a series of reactivity calculations are performe d to generate a set fo enrih c ment-fuel l assembly discharge bumup ordered pairs which all yield an equivalent Keg when stored in the spent fuel storage racks.

Figure 3 on page 26 shows the constant Ke g contour generated for the Prairie Island spent fuel rack all cell configuration. This curve represents combinations of fuel enrichment and discharge i

burnup which yield the same rack multiplication factor (K eg) as the rack loaded with 1.95 and 1.85 w/o fuel (at zero burnup) for Westinghouse OFA and STD fuel assemblies, respectively, in all celllocations.

Uncertainties associated with burnup credit include a reactivity uncertainty of 0.01 AK at j j

30,000 MWD /MTU applied linearly to the burnup credit requirement to account for calculational and depletion uncertainties and 4% on the calculated burnup to account for burnup measurement l

\

l l

7 Prairie Island Spent Fuel Racks l

't W; l uncertainty The amount of additional soluble boron needed to account for these uncertainties in - l the burnup requirement of Figure 3 is 200 and 250 ppm for the Westinghouse OFA and STD fuel-assembly types, respectively. This is additional boron above the 200 and 250 ppm required for  !

' Westinghouse OFA and STD fuel assembly types, respectively, as calculated in Section 3.2. This . )

results in'a total soluble boron credit of 400 and 500 ppm for the Westinghouse OFA and STD fuel assembly types, respectively.

-It is important to recognize that the curve in Figure 3 is based _on calculatio'ns of constant rack -

- reactivity, in this way, the environment of the storage rack and its influence on assembly reactivity. I is implicitly considered. For convenience, the data from Figure 3 is also provided in Table 3 on

. page 20. Use of linear interpolation between the tabulated values.is acceptable since the curve -

shown in Figure 3 is linear in between the tabulated points. t The effect of axial. burnup distribution on assembly reactivity has been considered in the development of the Prairie Island burnup credit limit. Previous evaluations have been performed i to quantify axial burnup reactivity effects and to confirm that the reactivity'equivalencing  ;

methodology described in Reference I results in calculations of conservative burnup credit limits.  !

The evaluations show that axial burnup effects can cause assembly reactivity to increase only at burnup-enrichment combinations which are beyond those calculated for the Prairie Island burnup credit limit. Therefore, additional accounting of axial burnup distribution effects in the Prairie Island burnup credit limit is not necessary.

i l

i t

I; 1

J I

i 8-Prairie Island Spent Fuel Racks -

1 '

4.0 Criticality Analysis of 3x3 Checkerboard Storage This section describes the analytical techniques and models employed to perform the criticality analysis and reactivity equivalencing evaluations for the Prairie Island spent fuel storage racks 3x3 checkerboard storage enrichment limits with credit for soluble boron. The purpose of the 3x3 checkerboard storage configuration is to allow the most reactive fresh fuel to be stored in the Prairie Island spent fuel racks. The most reactive fresh fuel for Prairie Island has a nominal enrichment of 4.95 w/o 235 U in a Westinghouse 14x14 OFA fuel assembly.

Section 4.1 describes the maximum feasible Keg KENO-Va calculations performed for the 3x3 checkerboard storage configuration. Section 4.2 discusses the results of the spent fuel rack Ke g soluble boron credit calculations. Finally, Section 4.3 presents the results of calculations performed to show the minimum burnup requirements for assemblies with higher initial enrichments above those determined in Section 4.1.

4.1 Maximum Feasible K err Calculations The following assumptions are used to develop the maximum feasible KENO-Va model for j storage of fuel assemblies in the Prairie Island spent fuel storage rack: ,

i

1. The fuel assembly parameters relevant to the criticality analysis are based on the Westing- i house 14x14 OFA, STD and Exxon TOPROD design (see Table 1 on page 18 for fuel param-eters). The Westinghouse 14x14 STD design bounds the reactivity of the other Exxon fuel assemblies currently stored in the Prairie Island spent fuel pool.
2. Westinghouse 14x14 OFA fuel assemblies stored in the middle of the 3x3 checkerboard con-tain uranium dioxide at a nominal enrichment of 4.95 w/o over the entire length of each rod.
3. Westinghouse 14x14 OFA, STD, and Exxon TOPROD fuel assemblies surrounding the center of the 3x3 checkerboard contain uranium dioxide at nominal enrichments of 1.50,1.38 and  ;

1.46 w/o, respectively, over the entire length of each rod. l

4. The fuel pellets are modeled assuming nominal values for theoretical density and dishing frac-tion. l
5. No credit is taken for any natural or reduced enrichment axial blankets. This assumption results in equivalent or conservative calculations of reactivity for all fuel assemblies used at Prairie Island including those with annular pellets at the fuel rod ends.

236 U

6. No credit is taken for any 234 U or in the fuel, nor is any credit taken for the buildup of tission product poison material.
7. No credit is taken for any spacer grids or spacer sleeves.
8. No credit is taken for any burnable absorber in the fuel rods.
9. No credit is taken for the presence of spent fuel rack Boraflex poison panels. The Boraflex volume is replaced with water.

Prairie Island Spent Fuel Racks 9'

10. The moderator is water with 0 ppm soluble boron at a temperature of 68'F. A water density of 3

1.0 gm/cm is used.

11. The array is infinite in lateral (x and y) extent and finite in axial (vertical) extent.
12. Storage cells are loaded with fuel assemblies in a 3x3 checkerboard pattem as shown in Figure 4 on page 27. The center of the 3x3 checkerboard is always a fresh 4.95 w/o Westinghouse OFA assembly. The surrounding assemblies are Westinghouse OFA, STD or Exxon 14x14 fuel assemblies with the specified enrichment limits.

With the above assumptions, the KENO-Va calculations of Keg under normal conditions resulted in a K ge of 0.98663, 0.98421, and 0.98029 for the Westinghouse OFA, STD, and Exxon TOPROD fuel assemblies, respectively. The reactivity bias calculated for the normal temperature range of the spent fuel pool water (50*F to 150*F) is 0.00476,0.00532, and 0.00535 AK for l Westinghouse OFA, STD and Exxon TOPROD fuel assemblies, respectively. Finally, the methodology bias associated with the benchmarking of the Westinghouse criticality methodology is 0.0077 AK.

Based on the results above, the following equation is used to develop the maximum feasible Ke g for the Prairie Island spent fuel storage racks:

K,y = K,,,,,, g + B,,,, + B, ,,,,,

where:

K,,,,,,,, -

normal conditions KENO-Va Ke g B ,,,,

= temperature bias from 50*F to 150*F B ,,,g,g meM Nas &&d & &&M Mal comparisons Substituting calculated values in the order listed above for Westinghouse OFA fuel, the result is:

K,g = 0.98663 + 0.00476 + 0.0077 = 0.99909 Substituting calculated values in the order listed above for Westinghouse STD fuel, the result is:

l K,g = 0.98421 + 0.00532 + 0.0077 = 0.99723 l Substituting calculated values in the order listed above for Exxon TOPROD fuel, the result is:

K,g = 0.98029 + 0.00535 + 0.0077 = 0.99334 10 Prairie Island Spent Fuel Racks

3 3 1

L Since Kerr is less than 1.0 for all fuel types considered, the Prairie Island spent fuel racks will i remain suberitical under maximum feasible conditions when cells are loaded in a 3x3 ,

checkerboard as specified in Figure 4 with a 4.95 w/o Westinghouse OFA fuel assembly l surrounded by any combination of 1.5,1.38 or 1.46 w/o Westinghouse OFA. STD or Exxon

. TOPROD fuel assemblies, respectively, in the next section, soluble boron credit will be used to provide safety margin by determining the amount of soluble boron required to maintain Ke g s 0.95 including tolerances and uncenainties.

4.2 Soluble Boron Credit Kg Calculations ,

To determine the amount of soluble boron required to maintain Ke g 5 0.95, KENO-Va is used to ,

establish a nominal reference reactivity and PHOENIX-P is used to assess the effects of matenal and construction tolerance variations. A final 95/95 Keff is developed by statistically combining l the individual tolerance impacts with the calculational and methodology uncenamties and summing this term with the nominal KENO-Va reference reactivity.

The assumptions used to develop the nominal case KENO-Va model for soluble boron credit for .

3x3 checkerboard cell storage in the Prairie Island spent fuel racks are similar to those in Section l 4.1 except for assumption 10 regarding the moderator soluble boron concentration. The moderator was replaced with 300,350 and 350 ppm for the Westinghouse OFA, STD and Exxon TOPROD fuel assembly types, respectively.

With the above assumptions, the KENOda calculation for the nominal case results in a eK rr o f 0.90886,0.89908, and 0.90072 for Westinghouse OFA, STD and Exxon TOPROD fuel assembly types, respectively.

Calculational and methodology biases must be considered in the final Ke rr summation prior to comparing against the 0.95 K,g limit. The follow-ing biases are included:

Methodology: The benchmarking bias as detemtined for the Westinghouse KENO-Va  ;

methodology was considered.

Water Temperature: A reactivity bias is applied to account for the effect of the normal range e of spent fuel pool water temperatures (50*F to 150*F).

To evaluate the reactivity effects of possible variations in material characteristics and ,

^

mechanical / construction dimensions, PHOENIX-P penurbation calculations are performed. For _

the Prairie Island spent fuel rack 3x3 checkerboard storage configuration, UO2 material tolerances  ;

are considered along with construction tolerances related to the cell I.D., storage cell pitch, and  ;

stainless steel wall thickness. Uncertainties associated with calculation and methodology accuracy are also considered in the statistical summation of uncenainty components.

The following tolerance and uncertainty components are considered in the total uncertainty statistical summation: ,

B 11 Prairie Island Spent Fuel Racks

,(*

235 U Enrichment: The standard DOE enrichment tolerance of 0.05 w/o 235 U about the nominal fresh reference enrichment of 4.95 and nominal burned reference enrichments of 1.50.

235 1.38 and 1.46 w/o U was considered.

UO 2Density: A 2.0% variation about the nominal reference theoretical density (the nominal reference values are listed in Table 1 on page 18) was considered.

Fuel Pellet Dishing: A variation in fuel pellet dishing fraction from 0.0% to 2.0% (the nominal reference values are listed in Table 1 on page 18) was considered.

Storage Cell I.D.: The 10.10 inch tolerance about the nominal 8.27 inch reference cell I.D.was considered.

Storage Cell Pitch: Tb: 20.06 inch tolerance about the nominal 9.50 inch reference cell pitch was considered.

Stainleus Steel Thickness: The i0.01 inch tolerance about the nominal 0.09 inch reference stainless steel thickness for all rack structures was considered.

Assembly Position: The KENO-Va reference reactivity calculation assumes fuel assemblies are symmetrically positioned within the storage cells. Conservative calculations show that an increase in reactivity can occur if the comers of four fuel assemblies are positioned together.

This reactivity increase is considered in the statistical summation of spent fuel rack tolerances.

Calculation Uncertainty: The 95 percent probability /95 percent confidence level uncertainty on the KENO-Va nominal reference K eg was considered.

Methodology Uncertainty: The 95 percent probability /95 percent confidence uncertainty in the benchmarking bias as determined for the Westinghouse KENO-Va methodology was considered.

The maximum K eg for the Prairie Island spent fuel rack 3x3 checkerboard storage configuration is developed by adding the calculational and methodology biases and the statistical sum of independent uncertainties to the nominal KENO-Va reference reactivity. The summation is shown in Table 4 on page 21 and results in a maximum K eg of 0.94627, 0.94114. and 0.93511 for Westinghouse OFA, STD, and Exxon TOPROD fuel assembly types, respectively.

Since Keg is less than 0.95 including soluble boron credit and uncertainties at a 95/95 probability / confidence level, the acceptance criteria for criticality is met for the 3x3 checkerboard configuration storage of 14x14 fuel assemblies in the Prairie Island spent fuel racks when cells are loaded in a 3x3 checkerboard with a 4.95 w/o Westinghouse OFA fuel assembly surrounded by any combination of 1.5,1.38 or 1.46 w/o Westinghouse OFA, STD or Exxon TOPROD fuel assemblies, respectively, including the presence of soluble boron as specified above.

4.3 Burnup Credit Reactivity Equivalencing .

Storage of fuel assemblies with enrichments higher than 1.5,1.38 and 1.46 w/o 235U for the Westinghouse OFA, STD and Exxon TOPROD fuel types in the Prairie Island spent fuel rack 3x3 ,

checkerboard configuration is achievable by means of the concept of reactivity equivalencing. f

)

Prairie Island Spent Fuel Racks 12

The concept of reactivity equivalencing is predicated upon the reactivity decrease associated with fuel depletion. For burnup credit, a series of reactivity calculations are performed to generate a set of enrichment-fuel assembly discharge burnup ordered pairs which all yield an equivalent Kerr when stored in the spent fuel storage racks.

Figure 5 on page 28 shows the constant Ke rr contour generated for the Prairie Island spent fuel rack 3x3 checkerboard storage configuration. This curve represents combinations of fuel enrichment and discharge burnup which yield the same rack multiplication factor (K err) as the rack loaded with 1.5,1.38 or 1.46 w/o fuel (at zero burnup) for Westinghouse OFA STD and Exxon TOPROD fuel assemblies, respectively.

Uncertainties associated with imrnup credit include a reactivity uncertainty of 0.01 AK at 30,000 MWD /MTU applied linearly to the burnup credit requirement to account for calculanonal and depletion uncenainties and 4% on the calculated burnup to account for burnup measurement uncertainty. The amount of additional soluble boron needed to account for these uncertainties m the bumup requirement of Figure 5 is 300 ppm for Westinghouse OFA and 350 ppm for Westinghouse STD and Exxon TOPROD fuel assembly types. This is additional boron above the 300,350, and 350 ppm required for Westinghouse OFA, STD cnd Exxon TOPROD fuel assembly types, respectively, as calculated in Section 4.2. This results in a total soluble boron credit of 600, 700 and 700 ppm for Westinghouse OFA, STD and Exxon TOPROD fuel assembly types, respectively.

It is important to recognize that the curve in Figure 5 is based on calculations of constant rack reactivity. In this way, the environment of the storage rack and its influence on assembly reactivity is implicitly considered. For convenience, the data from Figure 5 is also provided in Table 5 on page 22. Use of linear interpolation between the tabulated values is acceptable since the curve shown in Figure 5 is linear in between the tabulated points.

1 The effect of axial bumup distribution on assembly reactivity has been considered in the l development of the Prairie Island bumup credit limit. Previous evaluations have been performed to quantify axial burnup reactivity effects and to confirm that the reactivity equivalencing ,

methodology described in Reference I results in calculations of conservative burnup credit limits.

The evaluations show that axial burnup effects can cause assembly reactivity to increase only at burnup-enrichment combinations which are well beyond those calculated for the Prairie Island burnup credit limit. Therefore, additional accounting of axial burnup distribution effects in the Prairie Island burnup credit limit is not necessary.

1 Prairie Island Spent Fuel Racks 13

,3 -

5.0 Discussion of Postulated Accidents Most accident conditions will not result in an increase in K eg of the rack. Examples are:

Fuel assembly drop The rack structure pertinent for criticality is not excessively deformed

- on top of rack and the dropped assembly which comes to rest horizontally on top of the rack has sufficient water separating it from the active fuel height of stored assemblies to preclude neutronic interaction.

Fuel assembly drop Design of the spent fuel racks is such that it precludes the insertion of a between rack fuel assembly in other than prescribed locations.

modules or between rack modules and spent fuel pool wall However, two accidents can be postulated for each storage configuration which would increase reactivity beyond the analyzed condition. The first postulated accident would be a loss of fuel pool cooling system and the second would be a mistoaded of an assembly i.,to a cell for which the restrictions on location, enrichment or burnup are not satisfied.

For the loss of fuel pool cooling system accident, calculations were performed for teth all cell storage and 3x3 checkerboard storage to show the reactivity increase caused by a rise in the Prairie Island spent fuel pool water temperature from 150*F to 212*F. The reactivity increase for all cell storage is 0.00543 and 0.00585 AK for Westinghouse OFA and STD fuel assembly types, respectively. The reactivity increase for 3x3 checkerboard storage is 0.00295 and 0.00300 AK for Westinghouse OFA and STD fuel assembly types, respectively. The Westinghouse OFA and STD fuel assembly types conservatively bound the Exxon TOPROD fuel assembly type.

For the mistoaded assembly accident, calculations were performed for both all cell storage and 3x3 checkerboard storage to show the largest reactivity increase caused by a 4.95 w/o ,

Westinghouse OFA fuel assembly misplaced into a storage cell. The reactivity increase caused by misplacing a fuel assembly in the storage cell will bound the reactivity increase caused by placing a fuel assembly into the cask loading area. The largest reactivity increase for all cell storage is 0.04695 and 0.03746 AK for Westinghouse OFA and STD fuel assembly types, respectively. The largest reactivity increase for 3x3 checkerboard storage is 0.05561 and 0.04851 AK for Westinghouse OFA and STD fuel assembly types, respectively. The Westinghouse OFA and STD fuel assembly types conservatively bound the Exxon TOPROD fuel assembly type.

For an occurrence of the above postulated accident condition, the double contingency principle of ANSI /ANS 8.1-1983 can be applied. This states that one is not required to assume two unlikely, independent, concurrent events to ensure protection against a criticality accident. Thus, for these postulated accident conditions, the presence of additional soluble boron in the storage pool water (above the concentration required for normal conditions and burnup credit) can be assumed as a realistic initial condition since not assuming its presence would be a second unlikely event.

14' Prairie Island Spent Fuel Racks

The reactivity change due to the presence of soluble boron in the Prairie Island spent fuel pool has been calculated with PHOENIX P and is shown in Figure 6 on page 29 for all cell storage and Figure 7 on page 30 for 3x3 checkerboard storage.

The amount of soluble boron required to offset each of the postulated accidents can be determined i by using Figures 6 and 7. The additional amount of soluble boron needed for accident conditions is shown below:

Soluble Boron Total Soluble Storage Fuel Assembly Reactivity Required for Boron Required Configuration Type Increase (AK) Accidents (ppm) (ppm)

All Cell W - OFA 0.04695 200 600 Storage W - STD 0.03746 200 700 3x3 W - OFA 0.05561 350 950 Checkerboard W - STD 0.05015 350 1050 Storage ,

Based on the above discussion, should a loss of, spent fuel pool cooling accident or a fuel assembly misload occur in the Prairie Island spent fuel racks, Ke rrwill be maintained less than or equal to 0.95 due to the presence of at least 1050 ppm of soluble boron in the spent fuel pool water.

i 15 Prairie Island Spent Fuel Racks

b 6.0 Soluble Boron Credit Summary Spent fuel pool soluble boron has been used in this criticality analysis to offset storage rack and fuel assembly tolerances, calculational uncertainties, uncertainty associated with burnup credit

. and the reactivity increase caused by postulated accident conditions. The total soluble boron concentration required to be maintained in the spent fuel pool is a summation of each of these components. Table 6 on page 23. summarizes the storage configurations, fuel types and corresponding soluble boron credit requirements.

i l

l 1

I, 16 Prairie Island Spent Fuel Racks

i 1

7.0 Summary of Criticality Results Fcr the storage of fuel assemblies in the spent fuel storage racks, the acceptance criteria for criticality requires the effective neutron multiplication factor, K g, e to be less than or equal to 0.95, including uncertainties. This report shows that the acceptance criteria for criticality is met for the Prairie Island spent fuel racks for the storage of 14x14 fuel assemblies under both normal and accident conditions with soluble boron credit, no credit for the spent fuel rack Boraflex poison panels and the following storage configurations and enrichment limits:

All Cell Storage Storage of 14x14 assemblies in any cell location with nommal Enrichment Limits enrichments no greater than 1.95 w/o 235 U for Westinghouse 14x14 OFA fuel assemblies and 1.85 w/o 235 U for Westinghouse 14x14 STD and Exxon 14x14 fuel assemblies. Fuel assemblies with initial nominal enrichments greater than these must satisfy the minimum burnup requirement shown in Figure 3. The soluble boron credit required for this storage configuration is 600 ppm for Westinghouse 14x14 OFA fuel assemblies and 700 ppm for Westinghouse 14x14 STD and Exxon 14x14 fuel assemblies.

3x3 Checkerboard Storage of Westinghouse 14x14 OFA assemblies with nominal 235 Enrichment Limits enrichments no greater than 4.95 w/o U in the center of a 3x3 checkerboard. The surrounding fuel assemblies must have an  :

initial nominal enrichment no greater than 1.50 w/o 235 U for 235 Westinghouse 14x14 OFA fuel assemblies,1.46 w/o U for l I

Exxon TOPROD fuel assemblies, and 1.38 w/o 235U for Westinghouse 14x14 STD and other Exxon fuel assemblies. Fuel assemblies with initial nominal enrichments greater than these must satisfy the minimum bumup requirement shown in Figure 5.

The soluble boron credit required for this storage configuration is 950 ppm for Westinghouse 14x14 OFA fuel assemblies and 1050 ppm for Westinghouse 14x14 STD and Exxon 14x14 fuel assemblies.

The analytical methods employed herein conform with ANSI N18.2-1973, " Nuclear Safety )

Criteria for the Design of Stationary Pressurized Water Reactor Plants," Section 5.7 Fuel {

Handling System; ANSI 57.2-1983, " Design Objectives for LWR Spent Fuel Storage Facilities at )

Nuclear Power Stations," Section 6.4.2; ANSI N16.9-1975, " Validation of Calculational Methods for Nuclear Criticality Safety"; and the NRC Standard Review Plan, Section 9.1.2, " Spent Fuel Storage". Exception is taken to the requirement that no reactivity credit may be taken for the presence of soluble boron in the spent fuel pool as stated in ANSI 57.2-1983W and the NRC l position paperW and shows that the effective neutron multiplication factor, Keg, of the fuel rack j array is less than 0.95 with the presence of spent fuel pool soluble boron.  ;

17 Prairie Island Spent Fuel Racks

'4 Table 1. Fuel Parameters Employed in the Criticality Analysis Westinghouse Westinghouse Exxon 14x14 OFA 14x14 STD TOPROD Number of Fuel Rods per Assembly 179 179 179

' Rod Zirc-4 Clad O.D. (inch) 0.400 0.422 0.417 Clad Thickness (inch) 0.0243 0.0243 0.0295 Fuel Pellet O.D.(inch) 0.3444 0.3659 0.3505 Fuel Pellet Density (% of Theoretical) 95 95 94 Fuel Pellet Dishing Factor (%) 1.1926 1.1870 1.0000 Rod Pitch (inch) 0.556 0.556 0.556 Number of Zire Guide Tubes 16 16 16 Guide Tube O.D. (inch) 0.526 0.539 0.541 Guide Tube nickness (inch) 0.0170 0 0170 0.0185 Number ofInstmment Tubes 1 I I Instmment Abe O.D. (inch) 0.399 0.422 0.422 Instmment hbe Thickness (inch) 0.0235 0.0240 0.0240 i

e 1

i Prairie Island Spent Fuel Racks 18 l

l

. ~ . .

Table 2. Prairie Island All Cell Storage Soluble Boron Credit K,g W OFA W - STD i

Nominal KENO-Va Reference Reactivity: 0.91893 0.91341 Calculational & Methodology Biases: $

Methodology (Benchmark) Bias 0.00770 0.00770 .

Pool Temperature Bias (50'F - 150*F) 0.00613 0.00679  ;

TOTAL Bias 0.01383 0.01449 Tolerances & Uncertainties:

235 UO2Enrichment Tolerance (iO.05 w/o U) 0.00837 0.00864 UO2Density Tolerance ( 2%) 0.00412 0.00385 ,

Fuel Pellet Dishing Variation (0 to 2%) 0.00219 0.00203 Cell Inner Diaineter ( 0.10 inch) 0.00026 0.00031 Cell Pitch (10.06 inch) 0.00564 0.00550 Cell Wall Thickness ( 0.01 inch) 0.00762 0.00798 Asymmetric Assembly Position 0.00491 0.00072 Methodology Bias Uncertainty (95/95) 0.00300 0.00300 Calculational Uncertainty (95/95) 0.00282 0.00272 TOTAL Uncertainty (statistical) 0.01493 0.01430 f9

[ ( (tolerance;...or... uncertainty;) 2)  ;

$i=t {

Final K,g including Uncertainties & Tolerances: 0.94769 0.94220 ,

i 1

Prairie Island Spent Fuel Racks 19' i

, f ;

t- .

Table 3. Prairie Island All Cell Minimum Burnup Requirements Nominal W OFA W - STD Enrichment Burnup . Burnup (w/o) (MWD /MTU) (MWD /MTU) r 1.85 0 .0' l.95 0 1471 2.00 687 2194 2.20 3360 4999 2.40 5922 7677 2.60 8381 10238 2.80 10748 12692 3.00 13030 15051 3.20 15237 17325 3.40 17379 19524 3.60 19464 21659 3.80 21502 23740 4.00 23501 25779 4.20 25471 n/a

! 4.40 27422 n/a 4.60 29361. n/a 4.80 31299 n/a I 4.95 32757 n/a i

f 20 Prairie Island Spent Fuel Racks L

p5 ,

u. .

r #'

.+ 7 I

Table 4. Prairie Island 3x3 Checkerbeard Storage Soluble Boron Credit K,g W - OFA W - STD T P OD a

9 Nominal KENO Va Reference Reactivity: 0.90886 0.89908 0.90072 Calculational & Methodology Hiases:

Methodology (Benchmark) Bias 0.00770 0.00770 0.00770 l Pool Temperature Bias (50*F - 150*F) 0.00492 0.00545 0.00527 TOTAL Bias 0.01262 0.01315 0.01297 Tolerances & Uncertainties:

235 0.01138 UO2Enrichment Tolerance ( 0.05 w/o U) 0.01124 0.01201 UO2Density Tolerance ( 2%) 0.00461 0.00429 0.00453 Fuel Pellet Dishing Variation (0 to 2%) 0.00273 0.00230 0.00197 Cell Inner Diameter (i0.10 inch) 0.00023 0.00020 0.00012 Cell Pitch (10.06 inch) 0.00454 0,00444 0.00430 Cell WallThickness (i0.01 inch) 0.00704 0.00723 0.00712  ;

Asymmetric Assembly Position 0.01929 0.02404 0.01475 1 Calculational Uncertainty (95/95) 0.00285 0.00297 0.00300 $

Methodology Bias Uncertainty (95/95) 0.00300 0.00300 0.00300 TOTAL Uncertainty (statistical) 0.02479 0.02891- 0.02142 f9 , i

[ ( (tolerance;...or... uncertainty;) 2) hi=1 I

Final K,g including Uncertainties & Tolerances: 0.94627 0.94114 0.93511 5

f l

1 f

21 Prairic Island Spent Fuel Racks

]

p, 0; Table 5. Prairie Island 3x3 Checkerboard Minimum Burnup Requirements Nominal W - 0FA W - STD Exxon Enrichment Burnup Burnup TOPROD (w/o) (MWD /MTU) (MWD /MTU) (MWD /MTU) 1,38 0 0 0 1.46 0 1723 0 1.50 0 2567 949 1.60 1840 4629 3222 1.80 5361 8550 7371 2.00 8683 12220 11044 2.20 11821 15657 14308 2.40 14790 18881 17231 2.60 17606 21912 19878 2.80 20282 24769 22316 3.00 22835 27473 24612 3.20 25279 30043 26833 3.40 27629 32498 29044 3.60 29900 34858 31313 3.80 32107 37144-- 33707 4.00 34265 39374 36292 4.20 36389 n/a n/a 4.40 38494 n/a n/a 4.60 40595 n/a n/a 4.80 42708 n/a n/a 4.95 44308 n/a n/a t

Prairie Island Spent Fuel Racks 22-

Table 6. Summary of Soluble Boron Credit Requirements Storage Fuel Soluble Boron Soluble Boron Soluble Boron Total Soluble

  • Configuration Assembly Required for Required for Required for Boron Credit Type Tolerances / Burnup Credit Accidents Required Uncertainties (ppm) (ppm) (ppm) ,

(ppm)

All Cell W - OFA 200 200 200 600 Storage W - STD 250 250 200 700 Exxon 250 250 200 700 3x3 W - OFA 300 300 350 950 Checkerboard W - STD 350 350 350 1050 Storage Exxon 350 350 350 1050 i

23 Prairie Island Spent Fuel Racks

. i; ,

> I

. a M

1

.I -

e

/ 'L N- 1 i l o

~

~

y% ~ = .

e l

~ e

. n

.T >

  • =

. S

. . l .

= ,

I

= .

l o

j d b

ey i

/

/

/

f

/

- 25000 .

/

D .' >

r 2 l 3 >

$ .* )

2 20000 ,.'

CL o )/ .'

3 ,

/

E ,o' /

o  ;

CD ,o- r X ,

y 15000

/

m 8 o' f  :

o

/ \

3 ,

/  !

= ,

10000 '

/ l

)

f l

,o ,

o f

/

5000 o'

)

s' )

/

//

0 -

1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 Initial 23sU Enrichment (nominal w/o)

Figure 3. Prairie Island All Cell Storage Burnup Credit Requirement l

Prairie Island Spent Fuel Racks 26

~

~

E E E ,

/

i E E .

E ,

SI3

~

l e

/ , .

- l

_ , z 235 Fresh Fuel: Must be less than or equal to 4.95 w/o U Burned Fuel: Must satisfy the minimum burnup ,

requirements of Figure 5 l i

1 i

Figure 4. Prairie Island 3x3 Checkerboard Layout Requirement l

I Praide Island Spent Fuel Racks 27

g: - - ,

V.,:..

~

t 45000 ,

ii

/  :

,/ 1

. OFA Fuel.- /

40000 TOPROD Fuel /

/

......... STD Fuel .

2 I;

o' ,

f

}

-35000 ,

c

' ' i

, / .r  ;

c t' / .

f  ;

H '

o

/ /

30000 ,

/ J v.

a ,

/;

l g '

a' / / .;

2 / / l E 25000

, j I o .

g  ; s / j o o / )  ;

D o f i l

> s' / / ,

99999 j o

/ /

/ )

/  :

j r m.

/ / l

.l

' q Tu 15000 // j o n u , / /  !

'; / l i f f ,

10000 ,9 #

t  !

a /

't >

f,f f 'I il )  ;

i r r ,

5000 7f j j

r;

\ //

I () -i M  :

0 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 .;

Initial 235 U Enrichment (nominal w/o) l Figure 5. Prairie Island 3x3 Checkerboard Storasp Burnup Credit Requirement 4 1

.l Prairie Island Spent Fuel Racks 28 1

_ _ . - _ _ _ . - . . _ _ _ _ _ , , _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ i

g

.q:

l;'.E,- :.-  : ,.

- ' H .

~

O.00 s

l 1 t  : r: \'  :

l l T l l -0.05  !

v, i

---3 g '

OFA Fuel 6

-0.10 STD Fuel L 3.,

T i.

\',

W x -0.15 3 ',

< 3,, ,

.c \ '

t O 3' ' ,-

3 -0.20 h- 's ',

y -

s 5 x -

o 3 ,

m e 'x ,

C -0.25 '

( ..

N '- m x ,

x ,

\ 4,

-0.30 x '

s ., k x .

s

-0.35 g '

N s

-0.40 0 250 500 750 1000 1250 1500 1750 2000 Soluble Boron Concentration (ppm)

Figure 6. Prairie Island All Cell Soluble Boron Worth Prairie Island Spent Fuel Racks ,9

s: --

'% 4.,y { 7 .

.g4 w:

t .'g ? .:_!

'; i:

  • 0.00 i 3

E n

\ ,

\ ,__

-0.05 T i

L V#

r OFA Fuel V

q , ......... STD Fuel

-0.10 N

(',

\ '.

V',

2 \N ,

~

<1

-0.15 N '. '

.c \ .- -

t o \ ,,

3 \

,, j

~@

\ '.,

.2:

U 3 .

m -0.20 N - .

@ w .

m 'N N '. ..

N N

-0.25 sw -

y .

N,

h.
  • 3 .

-0.30 '

s

-0.35 0 250 500 750 1000 1250- 1500 1750 2000 Soluble Boron Concentration (ppm)

Figure 7. Prairie Island 3x3 Checkerboard Soluble Boven Worth Prairie Island Spent Fuel Racks 30'

g

+ :.

O-p '

t Bibliography  ;

1. Newmyer, W.D., Westinghouse Spent Fuel Rack Criticality Analysis Methodology,

.WCAP-14417, June 1995.  ;

2. Newmyer. W.D., Criticality Analysis of the Prairie Island Units 1 & 2 Fresh and Spent Fuel l Racks, February 1993.  ;
3. American Nuclear Society, American National StandardDesign Requirementsfor Light Water ,

Reactor Spent Fuel Storage Facilities at Nuclear Power Plants. ANSllANS-57.2-1983.

October 7,1983.

4. Nuclear Regulatory Commission, Letter to All Power Reactor Licensees from B. K. Grimes.

OT Positionfor Review and Acceptance ofSpent Fuel Storage and Handling Applications. l April 14,1978. ,

f i

t i

i

.t i

l i

l i

31 Prairie Island Spent Fuel Racks l

m __, =_ _ _ _ .

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