ML20091E338

From kanterella
Jump to navigation Jump to search
Rev 1 to Davis-Besse Nuclear Power Station,Unit 1,Cycle 8 -- Reload Rept
ML20091E338
Person / Time
Site: Davis Besse Cleveland Electric icon.png
Issue date: 10/31/1991
From:
FRAMATOME COGEMA FUELS (FORMERLY B&W FUEL CO.)
To:
Shared Package
ML20091E333 List:
References
BAW-2137, BAW-2137-R01, BAW-2137-R1, NUDOCS 9111190252
Download: ML20091E338 (52)


Text

_ _ _ _ . _ _ _ _ _ _ _

1, J

l ENcLosugt It i

jAW 2137 Rev.1 i October 1991 lI I I l lt i DAVIS BES$E NUCLEAR POWER STATION f ju UNIT 1. CYCLE 8 REIDAD REPORT i I i iI I iI 4

I i

lI !I i ,I i iK i,

!I k

lI B&W Fuel Comoan#y i 41111._ 1' + > .52 4 y11112 r 4I pts; p ADOCK 05000336 FDei

i L y ISLlll7 ltY l October 1991 I l I I I DAVIS tiESSE fiUC1.EAlt 1%'Eli STAT 10!1 I!!;1T 1, CYC1.E t! It01 DAD l'J.I'Ol',T l l I I s I

                   !!,t.W l'uel Company l>,  o, !\ox 10935 1,ynchburg, Virginia 24506 0935 l

B&W FuelCompany l l

g Rev. 1 5 2o/92 I I CONTENTS l'fie.

1. INTRODUCTION AND SUPJiARY . . . . . .. 11
2. OPERATING ll1 STORY . . . 21
3. CENERAL DESCRIPTION . . . . 31
4. FUEL SYSTEM DESIGN . . . . . . . . . . 4-1 _

I 4.1. 4.2. Fuel Assembly Hechanical Desir,n Fuel Rod Design 4.2.1. Cladding Collapse 4-1 42 4-2 4.2.2, Cladding Stress . . . 43 4.2,3. Cladding Strain . 4-3 4.3. Thermal Design . . . . . 44 4.4 Material Compatibility . . 44 4.5. Operating Experience . . 4-4

5. NUCLEAR DESIGN . . . 51 5.1. Physics Characteristics . . 51 5.2. Changes in Nuclear Design . . 51
6. TilERMAL-llYDRAbLIC DESIGN . . . 6-1
7. ACCIDENT AND TRANSIENT ANALYSIS . . . .... 71 7.1. General Safety Analysis . . . . . . .. /l 7.2. Accident Evaluation . . . . . . . . . . ... 71
8. PROPOSED MODIFICATIONS TO CORE OPEP>. TING LIMITS REPORT , . . . B1
9. STARTUP PROCIW1 PilYSICS TESTING , . . 91 9.1. Precritical Tests . . . . 9-1 9.1.1. Control W d Trip Test . . . . . . .. . 91 9.1.2. RC Flow 91 I

9.2. Zero Power Physics Tests , , . . . . . . .. . 91 9.2.1. Critical Boron Concentration . . . . 91 9.2.2. Temperature Reactivity Coefficient . . . . . 9-2 9.2,3. Control Rod Group / Boron Reactivity Vorth . . 92 I I 111 B&W FuelCompany l

Rev. I g 10/91 5 CONTFNTS (Cont'd) faII 9.3. Power Escalation Tests . . . . .. 93 9.3.1 Core Symn:etry Test . . . . . 93 9.3.2. Core Power Distribution Verification at Intermediate Power Level (IPL) and ~100erP R Vith Nominal Control Rod Position . 93 3 9.3.3. Incore vs. Excore Detector 1mbalance Correlation Verification at the IPL . . . . 9-4 9.3.4 Temperature Reactivity Cocificient at -100tFP .. 95 9.3.5. Power Doppler Reactivity Coefficient at -100trP 95 9.4. Procedure for Use 11 Acceptance Criteria Not Met . . 9-5

10. REFERENCES . . . . . . . , 10 1 list of Tables Table 4 1. Fuel Design Parameters . . . 46 l
51. Davis Besse Unit 1, Cycle 8 Physics Parameters 53 g 5-2. Shutdown Margin Calculation for Davis Besse. Cycle 8 .. 5-5 5 6 1. Maximum Desirr Conditions, Cycles 7 and 8 . 6-3 7 1. Compariron ni <ey Parametets for Accident Analysis 7-3 7-2. Bounding Values for Allowable LOCA Peak Linear Heat Rates 7-3 8 1. Quadrant Power Tilt Limits . . . 8-18 B 2. Negative Moderator 'lemperature Coefficient Limit . . 8-18 LLs t oL Fi ru re s g Figure 3-1. Davis Besse Cycle 8 Core Loading Diagram . . . , . 33 l 3 2. Davis-Besse Cycle 8 Enrichment and Burnup Distribution . .34 W 3-3. Control Rod Locations for Davis-Besse 1, Cycle 8 . 3-5 3-4. Davis Besse Cycle 8 BPRA Enrichment and Dis.tribut ion . 36 5-1. Davis-Besse Cycle 8 Relative Power Distribution at BOC (4 EPPD),

Full Power, Equilibrium Xenon, All Rods Out, APSRs Inserted . . . 5-6 6-1. Regulating Group Position Limits, O to 200 110 EFPD, four RC Pumps Davis Besse 1, Cycle 8. . , . . . . . .... 82 8 2. Regulating Croup Position Limits, 200 110 to 400 110 EFPD, y Four RC Pumps Davis-Besse 1, Cycle 8 . , . . . . . . . B3 8-3. Regulating Group Position Limits, After 400 110 EFPD, Four RC Pumps -- Davis Besse 1, Cycle 8 . . . .. . . 84 I iv B&W FuelCompany 5 a

    .__.______-_...m,                                     _ , _ _                  . . - - , _ _ _ _ _ . _ . _ . . - . _ . . , _ _ _ _ _ _                                                                                                       _ _ .               . _ . _ _ _ _ _ _ .

I  : 1 I iltLALDr_.bwi LCrn' u 2 3 l'aLL E 6-4, Regulating Cloup Position Liivif.s. O te 200 710 ETPD,

 ;                                    Three RC Putops , - Davis-Bosse 1, Cycle 8                                                                                                                    .                                     .              .        . 85
  , l                          8-5. Regulating Group Position Litnits, 200 j)O to 600110 Erl'D,
  !W                                  Three RC Puttps ,                            -   Davis besse 1, Cycle B                                                          . . . .                                                                               .         8-6 8 6. Regulating Group Position Limits, After 400 110 ErPD, Three RC Pumps,                                  Davis 1455.e 1, Cycle 8                                                                                                                                                         67 I

8 7. Control Rod Locations icr Davis Besse 1, Cycle 8 , 68 8 8. APSR Pot,i t ion Liwits , 0 u. 400 110 Erit, Tour RC Purips Dwls Nse 1 Cycie 8 , . . 89 i l 8 9. APSR Position Limit s Af ter 40v 110 ETPD, Three or

       =                              Tour RC Purtpr., APSRs Withdrown                                                    DaviF Besse 1, Cycle 8                                                                                          ..                           8 10 8 10. APSR Position Limits, O to Loa 110 ElrD, Three I                              RC Puttps - Invis-Besse 1, Cycle B 6 11. AX1 AL POWER IMBA1ANCE 1.irn!t s O to 200 110 ETPD, l our RC Purops                          -

Davis Besse 1, Cycle 8

                                                                                                                                                   ,                                               .                                                                   8-11 8 12

, g 8 12. AX1 AL POWER IMPA1 ANCT Limi t s , 200 410 t o 400 + 10 ETPD. I r""r "' ""*"' - '"" "' 8 1? . AX1 AL POWER 1!!bA1ANCE Limits , After 400 110 ETPD, c'c2* 8 - - " 3 Tour RC Pumps -- Davis-bess.e 1. Cycle 8 . 8-14 E 8 14. AX1 AL POWER IMBALANCE Limi t s , O to 200 110 ErPD, E Three RC Pumps - Davis-Btsse 1, Cycle 8 . . . . 8 15 8-15. AX1 AL POWER IMBALANCE LJ mi t s , 200 110 to 400 110 ErPD, Three RC Puttps 8 16 I Davis besse 1, Cycle 8 , . . , 8 16. AX1AL POWER IMBALANCE Limits, /siter 400 1 01 EPPD, ! Three RC Pumps -- Davia-Be-se 1, cycle 8 , 8-17 I I I I l I l I l l " B&W FuelCompany

I Rev. I 10/91 I i

1. INTRODUCTION AND SUtDiARY I

Thir report justifies operation of Davis Besse Nuclear Power Station Unit I at the rated core power of 2'72 MWt f or cycle 6. The required analyses are included as outlined in t.he Nuclear Regulatory Concission (NRC) document, Gu i d a nc e for I- Proposed License Amendments Relating to Refueline,,* Junn 1975. This report utilizes the analytical techniques and design bases that have been submit ted to the NRC and approved by that agency. Toledo Edison has an objective of operating with zero fuel defects. To acconplish this objective, all once- and twice-burned fuel assenblics were ultrasonienlly tested tot- leaking rods at the end of cycle 7 As a result of this inspection, five fuel assemblies were dctermined to have one questionable or def ective fuel rod. Three of the f uel assen.blies were in batch 8 and two were in htch 9. Th batch 8 a s s enibl i c s did not have reconstitutable mechanical features, and were discharged (in order to prevent leaking f uel f rom beirg used in cycle 8) along with symmetrically locat ed assemblies. The batch 9 assemblies have the reconstitutable nicchanical f eatures that provide easy repair canability, lloweve r , one of the batch 9 a s s en:bl i e s had one fuet rod that was severely damaged. That assettbly was discharged along with three other symmetrically located batch 9 asseniblies. The remaining batch 9 assembly was repaired, as described in section 4.1, with a stainless steel rod and reused in cycle 8. The changes described above resulted in a revis. ion to the core loading thac was presented and analyzed in the original cycle 8 reload report.) The balance of this toport provides the description and analysis of the revised cycle 8 design. Cycle 8 reactor and fuel parameters related to power capability are summarized in this report and compared to those for cycle 7. All accidents analyzed in the Davis Besse Updated Safety Analysis Report a (USAR), as applicable, have been l I reviewed for cycle 8 operation, and in cases where cycle 8 characteris, tics were conservative when compared to previous values, new analyses were not perf ormed. In all cases, the cycle 8 parameters are bounded. 1l B&W FuelCompany g

 ..-. ,       -_.. -.. - .- .---.-            . _. -               ..   - . _ - - - - - . . . -             . - . - _ - . . ~ . - . - . . _
I NRC Cenc ric Lett er 88 1(. allows t he placen>ent of numeric values of certain cycle I-specific paratteters in a Core Operating Limits Report ( C012 ) . On September 20, 1990 Toledo Edisot, issued a r evision to tb- plant Technical Specifications which irtplertenced the C012 concept for Davis Besse 1, cycle 7. The C012 chan6es for cycle 8 are included in section 8 of this report, I

u j The Technical Sg ccifications have been reviewed for cycle 8 operation. Isased on g the analyses performed, taking into account t.b e emert.ency core cooling system = (ECCS) pinni Acceptance Criteria and postu'.ated fuel dent.iiication effects, it is concluded that Davis Bes.se t'n i t 1, cycle S can be operated safely at i t r.  ; licensed core power Icvel of 2772 MVt. I. I I I E I I I I 12 I I B&W FuelCompany g

Rev. 1 4 10/91

 'I 45

, 2. OPIRATit;G lilSTORY I The ref erence cycle for the nuclear and thermal hydraulic analyses of Davis Besse l Unit 1 is cycle 7, which achieved cri 'cality on July 1,1990. Power escalation l began on July 3,1990 and full power (2772 MVt ) was attainet, a July 10, 1990. I Cycle 7 was shutdown f or refueling af ter 405.4 effective full power days (EFPD) of operation, During cycle 7 operation, no operating anomalies occurred that would adversely af fect fuel perf ormance during cycle 8. The scheduled duration of cycle 8 is 469 l EFPD. Cycle 8 was analyzed to 479 EFPD and the applicability of the cycle 7 reactor protection system (RPS) limits and setpoints to cycle 8 has been verified to 479 EFPD. The cycle 8 operating limits have also been verifled to 479 EFPD. The cycic 8 design includes an APSR pull and power coastdown. l The cycle 8 design minimizes the number of fuel anemblies that are cross core shuffled to reduce the potential for quadrant tilt artpl i f ica t ion. The cycle 8 , shuf fle pattern is discussed in section 3.

 'I
I

'I I I I ' 2-1 'g B&W FuelCompany

l Wl Rev, I 10/91 I I

3. GENERAL DESCRIPTION I

The Davis-Besse Unit I reactor core is described in detail in chapar 4 of the USAR E f or the unit. The cycle 8 core consists of 17 7 f uel assemblies (FAs), each l of which is a 15x15 array norn. ally containing 708 f uel rods, 16 cont rol rod guide I tubes, and one incore instrument guide tube All FAs in batches 8, 9, and 10 have a constant nominal f uel loading of 468.25 kg of uranium. The fuel consists of dished-end cylindrical pellets of uranium dioxide clad in cold worked Z1. caloy

4. The undensifled nominal active fuel lengths, theoretical densities, fuel and f uel rod dimens, ions , and other relat ed f uel parameters may be f ound in Table 4-1 of this report.

Figure 3 1 is the enre loading diagram f or Davis-Besse Unit 1, cycle 8. Fifty-three batch 7B asseniblies,, 7 bat ch BA assemblies, and 4 batch 9A assemblies were discharged at the end of cycle 7. The f uel assenblies in batches 8h and 9B were shuffled to their cycle 6 locations, with the core periphery locati<>ns occupied by both batch 8B and 9B f uel assemblies. Batches 8 and 9 had init ial enrichments l I of 3,13 and 3. 38 wt t, respectively. The feed batch, consiiting of 64 batch 10 assemblies with uranium enrichment of 3.69 wt 4, was inserted in the core interior in a symmetric checkerboard pat tern with the batch 8B and 9B FAs. The - shuffic scheme is a partial very low leakage core loading. Figure 3-2 is a quarter core map showing each assemblv's burnup at the beginning of cycle (BOC) B a-d its initial enrichment. Cycle 8 is operated in a feed and-bleed mode. The core reactivity is controlled by 53 full length Ag In Cd control rod assemblies (CRAs), 56 burnable poison rod assemblics (BPRAs), and soluble boron. Eight of the BPRAs were reinserted from I cycle 7. In addition to the full length control rods, eight inconel-600 axial power shaping rods (gray APSRs) are provided for additional control of the axial power distribution. The cycle 8 locationn of the control rods and the group designations are indicated in Figure 3 3. The core locations and the rod group I I B&W FuelCompany l

I group designat. ions of the 61 control rods in cycles 7 and 8 are the same. The cycle 8 locations and enrich: tents of the BI' ras are shown in l'iguro 3-4. I I I I I I C I I I E I I I I 3-2 I I B&W FuelCompany g!

Rev. I 10/91 Figure 3-1. Davis-Besse Cycle 8 Core loading Diagram i *- North SB OB 9B 96 SB Bl A----------------- P10 K4 10 K12 MB 8B CB 9B 10 BB 10 9B 8B B----------- C5 L3 K2 F P11 F nl4 9)13 C11 L 8B 10 46 10 9B 10 9B 10 9B 10 SB j C-------- 111 4 F K6 F L5 F L11 F K10 F PS 8 10 GB 10 BB 10 BB 10 8B 10 9 10 8 l D----- E F E4 F A7 F A8 F A9 F D1 F i 3  ; I E----- 9B C10 98 F9 10 F Sp D2 10 F SB L14 10 F 8B G8 10 F 8B B12 10 F 9B F7 9 Ce?> SB 98 10 8 10 8B 10 9B 10 8 10 BB 10 88 F-- H11 B9 F Gp F f% F OB F Dp4 F G15 F 99 B L2

 !             OB           10              9b               10                BB              10    8B    9B   8B     10    BB     10  9B            10     9B G--     D9            F              E10                 F              H7                F   A6    04   F15      F   P6       F E6               F   D7
   =           98          (.B              10               8B                10              9B    98    EB   OS     9B    10     EB  10            BB     9B
       !! - - 012          E14                 F             H1                   F            H13   N13   P5   D3     H3      F    H15  r            M2     C4 9B               10                                10          9B   8B     10    BB     10  9B            10     4B

'5 K -- 9B N9 10 F M10 F EB B10 F 8b L1 C12 kl0 F 119 F M6 F N7 I i I L-- 8B F14 9B P9 9B 10 9B F 8B K1 10 10 F CB N2 10 10 8B F 48 C8 10 10 F BB BB P12 10 10 GB F 88 K15 10 10 9b F 9B P1 9B 8B H5 BB l M - - - -l- 010 L9 F F4 F K8 F F2 F N14 F L/ 06 lI 1 8B 10 9B 10 BB 10

                                                                                                            'B  10     8B    10     oB  10            88 I

l i M3 F NS F R7 F M8 F R9 F M12 F M13 N----lI- l l l 8B 10 9B 10 9B 10 9B 10 9B 10 BB l l ! O-- B8 F G6 F F5 F Fil F GJO F H2 ' l l l l l l 8B 9B 9B 10 BB 10 9B 98 8B l l l 05 F3 G2 F BS F G14 F13 011 ! P----lI I I I I I l l l l EB 9B 9B 9B SB l l l l l E8 G4 D13 G12 B6

R----lI I I I I I I I I I i

i l I I I I I I I I I I I I  :

;l              1 1

2 I 3 I 4 I 5 I 6 I i 7 8 l 9 l I 10 I 11 l 12 i 13 I 14 15 I i t Kev: _ xxx xxx - Batch ID ". yyy yyy - Location in Previous Cycle l I 3-3 l B&W Fuel Company

I Rev. 1 E 10/91 5 Figure 3-2. Davis-Isesse Cycle B Enrichment and liurnup Distribution 1 8 9 10 11 12 13 14 15 m l g 3.13 3.38 3.38 3.69 3.13 3.69 3.13 3.38 ' " 24534 13334 17635 0 23757 0 24529 13362 3.38 3.13 3.69 3.13 3.69 3.38 3.69 3.38 3' 13334 22398 0 26977 0 17902 0 17633 3.38 3.69 3.13 3.69 3.13 3.69 3,38 '.13 t 17636 0 22588 0 24106 0 15233 27961 3.69 3.13 3.69 3.13 3.69 3.35 3.38

                                                               "                                      0                  27735                                                    0                  22590                                      0 17661  16H12 ea=.,-
                                                               "                   2 753                                                                       24b                                                                         1757          28b (

l l O 3.69 3.38 3.69 3.38 3.69 3.13 g

O 17859 0 17645 0 30466 m E

E P 24529 1 24 18 28b63 I i 3.38 3.38 3.13 g 13362 17624 27?22 g l ! I x.xx Initial Enrichment g

yyyyy is0C Burnup M'a'D/MTU g i

l I l l 3-4 B&W Fuel Company g 4 ,,

I Tigure 3-3. Control Rod Locations for Davis-Besse 1, Cycle 8

  • North X

l A I B 4 6 4 l C 2 5 5 2 I D 7 8 7 8 7 E 2 5 5 2 j F 4 8 6 3 6 8 4 G S 1 1 5 11 W--- 6 7 3 4 3 7 6 -Y K 5 1 1 5 I L 4 8 6 3 6 8 4 M 2 5 5 2 f N 7 8 7 8 7 ,I 0 2 5 5 2 j P 4 6 4 i

!                                                 l 1 2    3     4     5       6   7         9  10    11    12       13 14  15 Group      No. of Pods     function x        Croup Number                   1             4          Safety 2             8          Safety 3             4          Safety 4             9          Safety

[I 5 12 Control 6 8 Concrol I 7 8 Total 8 _8, 61 Control APSRs B&W Fuel Company g 33

Rev. I g 10/91 g Pigure 3-4, Davis-Besse Cycle 8 BPRA Enrichnient and Distribution 8 9 10 11 12 13 14 15 g 2.0 2.0 y, 1.7 1,7 0.2t I g 1.7 1.4 1.1 g 2,0 1.4 1.7 I g 1.7 1.7 g 2.0 1,1 E_ E 0,2t 7 I R t These BPRAs are reinserted from eycle 7, I x.x Initial EPRA C.'ncentration, wt% B.C in Al 20 3. I 36 I I B&W FuelCampaq g

                                                                                                                      =

Rev. I ~ 10/91 H 4 FUEL SYSTEM DESIGN 4.1 Fuel Arserbiv Mechanical Desira The types of fuel assert.blies and per tinent f uel parameters for Davis-Besse cycle 8 are listed in Tabic 4 1. Batch BB is the Mark B5A design, while batches 9 and 10 are of the Mark BBA design. Batch 10 fuel incorporates all of the features of the batch 9 fuel but includer, a slight reduction in pin propressure to I increase the similarity in mechanical and thermal perfortrance to that of the Mark BSA design. I Eight gray ApSRs and 53 f ull length Ag In Cd control rods will be used in cycic

8. Forty eight new Mark B5 BPRAs were introduced into the core along with 8 once burned Mark-B5 BpRAs for a total of 56 BPRAs. In terms of creep collapse, stress, strain, and corrosion, the Mark B5 BPRAs were found to be mechanically i.d qu:. for irradiation up to 1,000 EPPD.

One fuel assembly (NJ0542) was reconstituted. This consisted of removing the upper end litting and removing one leaking fuel rod which had been identified by UT. The leaking fuel rod was replaced by another fuel rod from within that assembly. That rod in turn was replaced by a SS304 solid rod or pin. Then the upper end fitting was t eattached to the fuel assembly. When a fuel assembly is reconstituted, failed rods are repInced with solid stainless steel (SS304) rods. The steel rods are designed and analyzed to ensure that there is no adverse impact on fuel assembly performance. The dummy steel rod design is determined such that; a) The steel rod engages all spacer grid stops under all conditions, b) Clearance is maintained between the dummy steel rod and the top grillage under reactor temperature and irradiation conditions, c) Thermal expansion of the steel rod will not cause any set of the spring stops on the spacer grids, and 41 B&W Fuel Company

l l Itev. 1 i i 10/91 d) The difference in mass between the s.t eel rod and t he fuel pins will not affect f uel assen:bly lift. i The ef f ect of dif fererc '.al tereperature expansion between the stainless steel pin i and the Zircaloy guide tubes was analyzed over the operating and shutdown I temperature range. Additionally, the irradiation growth of the guide tubes was analyzed with r(spect to the stainless steel rod. l The thermal expansion of the stainless steel pin in the radial direction vill be 4 , about three times roorc than a Zircaloy 4 clad fuel pin. Vith thermal expansion the stainless steel pin will compress the spring stops on the spacer grids about E 0.002 inches more than the fuel rods. That compression is within the clastic W l i i range of the syring stops and will not cause any set of the spring stops. The mass of the steel rod is slightly less than that of a fuel rod. The W difference is about one 1%. For one dummy rod the ef f ect on fuel assembly lift , and response is insignificant. 4.2 Fud Rod Desirn The frr1 rod design and mechanical evaluation are discunsed below. I' 4.2.1 Claddinn Collapse The most limiting power nistory for each of the three fuel batches was determined. These histories were compared to generic and previous creep collapse analyses. Both the generic and the previous analyses are based on the a,ethod , from reference 3 and are applicable to the batch BB design for cycle 8 operation. I g The analyses predicted a creep collapse life in excess of 35,000 effective full E power hours (EFPH) . This is longer than the maximum batch 8B residence of 29,983 ! EFPH. For batches 9 and 10, the creep collapse analysis followed the method from reference 4 The operational conditions and mechanical characteristics of the I batch 9 and 10 fuel assemblies were compared to an envelope formulated by BWFC . and approved by the NRC S All values for the Mark BBA fuel assemblies are l encompassed by the corresponding parameters of the limiting envelope. The as-built data for the batch 10 fuel assemblies was approximated from partial batch g 9 as built data for Mark-BBA assemblies and then compared to the limiting u envelope. This is reasonable as the tolerances affecting these as-built values I, 4-2 I. B&W Fuel Company ,

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

It e v . 1 10/91 , i have not changed f roin previous bat ch f uel desigini. The creep collapse life of 1 I the batch 9 and 10 f uel rods % sed on ref e ence 5 is 5 5. 000 !%'d/ int U . 4.2.? Cl a dd i ne s_ tun l The stress parar.seters f or the two f uel rod designs are enveloped by conservative generic fuel rod stress anclyses. Por design evaluation, certain stress intensity limit s for all condit ion I nnd 11 events toust be itet . 1.init s are based I on ASMC criteria. Stt ess intens! ties are calculated in accor dance with the ASME Code, which includes both normal and t.h e a t :; tress effects. These s t r e t.s intensities are compared to Sm. Sm is equal to two thirds of the minimum specified unirradiated yield strength of the material at the operating t ertpe ra t u re range (650 deg r). The stress intensity limits are as. f oll. .rs : Pm < 'i.0 Sm P1 < 1.5 Sm Pm 4 Pb ( 1.5 Sm Pm 4 lb + Q < 3.0 Sm Pm: General Primary Membrane Stier,s Intensity Pl: hocal Primary Membrane Stress Intensity Pb: Priraary Bending St ress, Intensity Q: Secondary Stress Intensity Stress intensity calculations combine stresseu so that the resulting stress intensity is maximized. For both fuel rod designs the margins are in excess of 18.7%. The f o .' owing conservatisms were used in the stress analyses to ensure that all condition I and II operating parameters were enveloped:

1. how post-densification internal pressure, or as built I 2.

p r e p r e s.s u re . High system pressure.

3. liigh thermal gradient across the cladding.
4. Minimum specified cladding thickness.

4.2.3 Claddine_ Strait) The fuel design criteria specify a limit of 1.0% on cladding plastic tensile circumferential strain. The f( pellet is designed to ensure that plastic I cladding strain is less than 1% at design local pellet burnup and heat generation 1 43 1 g B&W Fuel Company

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

! Rev. 1 l, 10/91 5 rate. The design values are higher than the worst caso values Davis Besse Unit 1, cyc.c 8 fuel is expected to experience. For the hatch BB, Mark B5A fuel a s s e n:bl i e s , a generic strain analysis was reviewed conservatively based on the upper tolerance valueu for the fuel pellet diameter and density and the lower i tolerance limit for cladding inside diameter. For the Mark BBA fuel assemblies from batches 9 and 10, a strain analysis was done utilizing the method of reicrence 6, l 4.3 Therral DesirD All fuel in the cycle 8 core is thern.nlly similar. The design of the hatch 10 Mark B8A assemblies is such that the thermal nerfortnance of this fuel is equivalent to the fuel design used in the remainder of the core. The Mark BBA fuel rod design includes a grippable upper end cap and a bullet nose lower end cap. These f uel tod design f eatures have no ( f f eet on thermal performance One batch 9B fuel assen.bly was reconstituted; this has no effect on its thermal performance Fuel thermal analyses were performed with the TACO 2' fuel pin performance code. Nominal undensified input paramet ers used in the analysis are E' presented in Table 4 1, bensification etfects were accounted for in the TACO 2 code densification model. The results of the thermal design evaluation of the cycle 8 core are summarized in Table 4-1. Linear heat rate ( UIR) to fuel melt capability for all fuel was de termined with TACO 2. The analyses performed f or cycle 8 demoant rate that 20.5 kW/f t is a conservative limit to preclude centerline fuel melt (CPM) for all fuel E E batches. The maximum fuel pin burnup at FOC 8 is predicted to be less than 46,840 mwd /mt.U l (batch EB). Fuel rod internal pressure has been evaluated with TACO 2 for the highest burnup fuel rod and is predicted to be less than the 2200 psia reactor coolant pressure at the core outlet, = 4.4 Material Cottentibility The compatibility of all possible fuel-cladding-coolant-assembly interactions for hatch 10 fuel assemblics is identical to that of present fuel assemblies. 4.5 Operatior Experience Babcock 6 Wilcox operating experience wit.h the f irk B 15x15 fuel assembly has 4-4 I' B&W PuelCompany g _- - . s,_ . _- - . , . _ _ . _ _ _ . . _ _ . _ - - _ _ _ _ ,

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

M L F verified t.h e adequacy of its design. The following experience has been ac c unzula t e d for eight B&W 177 fuel ar.sembly plants using the Mark B fuel Cumulative Current Max TA Purnuo. MVd/mtlW net cluctric s Ecactor cvele Intrag Discharred outou d 'h_ C Oconce 1 13 39,438 58,310 85,555,367 Oconee 2 12 36,393 42,820 79,956,383 Oconee 3 12 45,795 39,701 79,420,910 Thret Mile Island 8 30.366 33,975 47.252,376 Arkansas !!uclear One, Unit 1 10 34,421 57,318 64,615,141 Rancho Seco 7 H) 38,266 43,20B,042 Crystal River 3 8 33.857 40,600 51.714.542 Davis Eesse 7 36,220 40,300 39,295,898 (*) As of February 28, 1991. N As of February 28, 1990.

4) Plant Shutdown in June , 1989 and core unload,d.

I I I 4-5 B&W FuelCompany

                                                                                                          .\

Rev. 1 10/91 lb E 10ble 4-1. Fuel Dnsirn Parn'reters Es t ch F B Bnt ch 4B(*) Patch LQ l Fuel assembly type Mark-BSA Mark B8A Mark-BBA No. of assemblies 57 56 64 I Fuel rod OD, in. 0.430 0.430 0.430 Fuel rod ID, in. 0.377 0.377 0.377 Tubular spacer 2r 4 NA NA Undensified active fuel length, in. 143.2 143.2 143.2 Fuel pellet (mean) diameter, in. 0.3686 0.3686 0.3686 I c Fuel pellet initial e density, %TD mean 95.0 95.0 95.0 Initial fuel batch enrichment v/o U235 3.13 3.38 3.69 Average burnap BOC, mwd /mtU 25,766 16,595 0 s Expos.re Time, Erpli 29,983 20,856 11,256 Cladding collapse g g

                                                                                                             ~

time, EFPti >35,000 NA NA Maximum essembly burnup, mwd /mtU 44,250 36,530 21,530 Cladding Collapse Burnup, mwd /mtU UA 55,000 55,000 Nominal linear heat rate at 2772 MWt, kW/ft 6.14 6,14 6.14 Minimum linear heat rate to melt, kW/ft 20.5 20.5 20.5 (* Includes one reconstituted assembly (NJ0542) which has a SS304 salid pin in place of a fuel rod. 4-6 I B&W Fuel Company g

Rev. 1 - 10/91 t

5. NUCLEAR DESIGN I

L1. Phystes Charactei d stics Table 5-1 compares the core physics parameters for the cycle 7 and 8 designs. The values for cycles 7 and 8 were both generated with the N00DLE code, i 7 Dif fercrees in core physics parametet s are to be expected between the cycles due to the changes in fuel and burnable poison enrichments which create changes in radial flux and burnup distributions. Figure 5 1 illustrates a representative

 ,                            relative power distribution for BOC 8 at full power with equilibrium xenon, all h                             rods out and gray APSRs inserted.

The ejected rod worths in Table 5-1 are the maxim < calculated values. Calculated ejected rod worths and theJ r adherence to criteria are considered at all times in life and at all power levels in the development of the rod position lim ts presented in section 8. The adequacy of the shutdown margir with cyc!e 8 rod worths is shoorn in Table 5-2. The following conservatisms were applied for the shutdown calculations:

1. Pcison n.aterial depletion allowance. *
2. 10% uncertainty on net rod worth.

I - A maximum flux redistribution penalty, i T maximum power deficit uith minimum boron. The maximum flux redistribution was taken into account to ensure that the effects l of operational maneuvering transients were included in the shutdown analysis. 5.2. Chances in Nuclear Desi;,n There are no significant core design changes for cycle 8. The calculational models and the methods used to obtain the important nuclear d sign parameters for this cycle were the same as those used for the reference cycle, except the calculations performed at end of cycle (EOC) for the Doppler and moderator coefficients as discussed in the footnotes to Table 5-1. An additional parame*er, temperatura coefficient at steam line break conditions, is includeri 5-1 B&W Fuel Company g

Rev. I 10/91 'a for cycle 8 (see footnote (f) of Table 5-1) . The purpose of this new pararnerer is discussed in section 7. No cignificant operational or procedural changes exist with regaro to axial or radial power shape, xenon, ot tilt control. The stability and control of the core with APSRs withdrawn has been analyzed. The calculated stability index without APSRs, is 0,0439h4, which demonstrates the l axial stability of the core. The operating limits (COLR changes) for the reload ' cycle are given in section 8. I I I I I E I I I I 52 I I B&W Fuel Company g

I Rev. 1 10/91 I

                                                                                                                                                                                                                              )

Table 5-1. .navis-P. esse Unit 1. Cve l e El'hys i c s Pa ran e t e r s l I Cycle,2 Cvele 8(*) Cycle length, EFPD 415 4 7 9 "') Cycle burnup, mwd /mtU 13,880 16,020 Average core burnu3 - EOC, mwd /mtU 26,806 29,568 l Initial core loading, mtU 82.9 82.9 Critica' boronk) - BOC, No Xo, ppm HZP 1,506 1,737 IIFP 1,323 1,515 Critical boron") - EOC, Eq. Xe, ppm , ilZP 200 207 l ilFP 10(d) 10(d) Control rod worths - IIFP, BOC, tak/k Grou;, 6 1.14 1.02 Group 7 1.03 1.02 Croup 8 0,19 0.15 I Control rod worths - 1:f P. E')C , tak/k Croup 7 Croup 6 1.09 NA 1.13

                                                                                                                                                                                          !!A l

Hax ejected rod worth IlZP, tak/k BOC, Groups 5-8 inserted (L-10,N-12) 0.32 0.30 EOC, Groups 5 7 inserted (L-10,N-12) 0,38 0.35 Max stuck rod worth - IlZP, %Ak/k BOC (M 13,N-12) 0.60 0.77 , a EOC (M-11) 0.81 0 68 4 l ll Power deficit - HZP to HFP, Eq. Xe, tak/k BOC (4 EFPD) -1.71 -1 74 I EOC *2.55 -2,71 Doppler coef f liFP,10-3 gAk/k/oF l F,0C, No Xe") Croup 8 inserted 1.58 -1.'? l l EOC, Eq. Xe, 0 ppm, Group 8 withdrawn 1.86 -1,90 l Moderator coeff - IlFP,10-2 tak/k/*F ll = BOC, No Xe N EOC, Eq. Xe, 0 ppm m

                                                                                                                                                                     -0.70
                                                                                                                                                                     -2.93
                                                                                                                                                                                   -0.66
                                                                                                                                                                                   -3.33                                 I I

5-3 I B&W Fuel Company g

m . . _ _ _ _ _ _ _ . _ _ . . _ _ _ _ _ . . . _ . . _ _ . . . . . _ . . . ._ . . . . _ _ _ _ . . _ _ _ _ _ . _ _ . . _ _ _ _ _ _ _ . . . . _ . . . _ . _ . . . _ Rev. I 10/91 Table 5-1. Davis-Besse Unit 1. Cycle 8 Physics Parwerers (Continued) Cycle 7 Cvele 8 Boron worth - HFP, ppm /%Ak/k g BOC (') 129 13B g EOC 110 114 Xenon worth - HFP, tak/k BOC (4 EFPD) 2.61 2.63 EOC (equilibrium) 2.78 2.77 l Effective delayed neutron fraction - HFP BOC 0.00618 0.00623 i EOC 0.00520 0.00514 (*) Based on cycle 6 length of 380.3 EFPD (actuali and cycle 7 length of 400 EFPD; I however, all boron concentrations have been adjusted to the actual cycle length of 405.4 EFFD. N All end-of-cycle values calculated at 479 EFPD; the design cycle 8 length is 469 EFPD. (*) Control rod group 8 is inserted at BOC and withdrawn at EOC. l 5 (d) Power coastdown to EOC at 10 ppm. (*) Cycle 8 values wero calculated at 1549 ppm; cycle 7 values were calculated at 1323 ppm.

         ") The EOC 8 value was calculated with the control rods at the insertion limit. The tamperature coef ficient calculated at the steam line break conditions is - 2,6 7 x 10-2 % Ak/k/ F and is discussed in section 7.0.

I, E< I I L I 1 I l 5-4 Ii l B&W Fuel Company gl

                                                                                                                                                                                           =:

1 l

Rev. I 10/91 l l Table 5-2 Shutdown Marrin Calcu}.ation for Davis-Besse. Cvele 8 I BOC(') 410 EFPD(') 479 EFPD(") [ Available Rod Worth Total rod worth, llZP 6.92 7.01 7.05 I Vorth reduction due to burnup of poison material Maximum stuck rod, ll2P

                                                          -0.16
                                                          -0.77 0.18
                                                                                 -0.68
                                                                                                               -0.19
                                                                                                               -0.68 Net worth                                         5.99                  6.15                         6.18 Less 10% uncertainty                           -0.60                  -0,67                         -0.62 Total availabic worth                             5.39                   5.53                        E.56 Reouired Rod Worth Power deficit, ilFP to 112P                       1,74                   2.66                        2.71 Max allowable inserted rod worth                 0.33                   0.51                         0.50 Flux redistribution                              0.53                    1.02                        0.46 Total required worth                              2.60                  4.19                         4.17 Shutdown Marc!n Total available minus total                       2.79                   1.34                        1.39 required rod worth I         (*)          Croup 8 is at the nominal position at BOC and 410 EEPD and is out at 479 EFPD.

, Note: Required shutdown margin is 1.00tak/k. I I 5.s B&W FuelCompany g

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

Rev, 1

                                                                                                                                              =

10/91 Figure 5 1. Davis Besse Cycle 8 Relative Power Distribution at BOC (4 EFPD), Full Power, Equilibrium Xenon, All Rods Out, APSRs Inserted (*) 8 9 10 11 12 13 14 15

            "         0.966         1.180        1.204      1.255            1.063    1.304     0.904    0.567 I-1.184        1.048        1.280      1.000            1.294    1.244     1.250    0.524 8

1.209 1.283 1.066 1.288 1.041 1.292 0.881 0.298 I 1.259 0.998 1.285 1.083 1.299 1.061 0.612 8 I i

            "          1.066        1.295        1.040      1.297            1.182    1.140     0.332 l

0 1.309 1.249 1.295 1.062 1.142 0.428 l E 5 0.909 1.257 0.887 0.614 0.333 0.571 0.530 0.307 I x Inserted Rod Group Number l l x.xxx Relative Power Density I (*) Calculated results from the two dimensional pin-by-pin PDQ07. 5-6 B&W Fuel Company 5-

l Rev, 1 l W 10/91 I i l 6.0 THERMAL HYDRAULIC DESIGN The Mark BBA fuel assemblies inserted for cycle 8 have Zircaloy intermediate spacer grids. Zircaloy intermediate spacer grids were first inserted on a full batch basis in cycle 7 of Davis-Besse Unit 1. The thermal hydraulic design evaluation supporting cycle 8 operation utilized the same methods and models as cycle 7 (described in references 8, 9 and 10 as supplemented by reference 11) which implemented the BWC (reference 12) CHF correlation for analysis of Zircaloy-grid fuel assemblies. The analyses presented in section 5 of reference 11 demonstrated that changes in the flow parameters resulting from the l , incorporation of Zircaloy spacer grids do not significantly impact the thermal-hydraulic characteristics of a Zircaloy grid core relative to the standard Inconel-grid (Mark-B) core, Implementation of the Zircaloy-grid fuel assemblies l into existing reactors, however, is performed on a batch basis, with the l I transition cycles having both Zircaloy grid and standard Mark-B fuel assemblies. l The Mark-B8A fuel assembly has a slightly higher pressure drop than the standard l l Mark-B assembly due to the higher flow resistance of the Zircaloy spacer grids. t l The presence of Mark-BBA fuel assemblies in a mixed core, will, therefore, tend j to divert some flow from the more restrictive Mark BBA assemblies to the Mark-B fuel. As a result, the Mark B8A fuel assemblies in a mixed core wil? experience slightly less coolant flow than in a homogeneous Mark-B8A core. This reduced flow results in a reduced thermal margin for the Mark-B8A assemblics relative to a full Mark-B8A core. The amount of coolant flow reduction is dependent on the I- number of Mark-B8A assemblies (with the smaller number of Mark-B8A assemblies being more limiting) . A " transition core penalty" is, therefore, consioered for a mixed Zircaloy-grid and standard Mark-B core. For cycle 8 of Davis-Besse Unit 1, this transition penalty is offset Ly the consideration of a core bypass flow fraction in the thermal-hydraulic model that is higher than the actual value. The Mark-B8A fuel rod design includes a grippable upper end cap and a bullet nose lI l l 6-1 1 l B&W Fuel Company

i 4 Rev. I

;                                                                                                                        10/91 lower end cap. These fuel rod design features have an insignificant effect on
,              thermal-hydraulic performance.

i f The reconstituted fuel assembly contains one stainless steel replacement rod that is surrounded within the fuel rod array by heated fuel rods. The BWC CHF ! correlation and associated licensing methodologies are approved for this E geometry, Calculations show there is no DNB penalty associated with the i placement of the stainless steel rod in this configurati n. The reconstituted

fuel a sembly has over 1001 DNBR n.argin during ncrmal operation. If the core should approach the DNB-based core safety limits, the reconstituted fuel assembly l would still retain over 40t DNBR margin relative to the limiting fuel bundle Therefore, adequate DNBR margin is available to justify operation of the core l g f with the reconstituted fuel assenbly. E
B!N-de signed reac tors , including Davis-Besse Unit 1, currently operate without orifice rod assemblies in the control rod guide tubes (CRGTs). The core bypass l

fraction is dependent on the number of unplugged guide tubes, which is in turn dependent on the number of burnable poison rods (BPRAs) and control rod j assemblies (CRAs), since these components restrict flow through the (CRCTs) . For thermal-hydraulic analysis, the most limiting case is that with the higher bypass

flow fraction, or smaller number of BPRAs.

The design basis chosen for cycle 8 thermal-hydraulic analyses vas a full l Zircaloy grid core, containing 37 BPRAs, for which the core bypass flow is 8.9%. 4 This design configuration was used to calculate the 1.54 DNBR (112% FP) shown on 5 ! Table 6-1 for cycle 8. The actual cycle 8 core configuration consists of 64 E fresh Mark-B8A fuel assemblies, 56 second cycle Mark-B8A fuel assemblies, 57 standard Mark-B fuel assemblies and 56 BPRAs. The core bypass flow for the cycle 8 configuration is 8.5%. The DNBP for this configuration, using the same core conditions presented in Table 6-1, is greater than 1.56. A comparison of the DNBRs for the design and the actual core configuration shows that the design configuration is conservative for cycle 8 DNBR analyses, Therefore, a transition core penalty is not necessary. Table 6-1 provides a summary comparison of the DNB analysis parameters for cycles 7 and 8. I 6-2 J B&W Fuel Company

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

1 1  ; i i !I Tf.bje 6 1. Maximum Desien conditions. cycles 7 and 8 i Cycle 7 cvele 8 ! Rated thermal power level, MVt 2772 2772 Nominal core exit pressure, psia 2200 2200

Minimum core exit pressure, psia 2135 2135

! Reactor coolant flow, gpm 380,000 380.000 l Core bypass flow, t (*) 8.9 8.9 I DNBR modeling Crossflow Crossflow i l Reference design radial-local l power peaking factor 1.71 1.71 i jg Reference design axial flux shape 1.65 chopped 1.65 chopped

g cosine cosine i

{ Hot channel factors i Enthalpy rise 1.011 1.011 j Heat flur 1.014 1.014 i Flow area 0.97 0.97 Active fuel ler.gth, in. 143.2 143.2 ! Avg heat flux at 100% power, i 105 Beu/h-ft2 1.86 1.86

Max heat flux at 100% power,
105 Btu /h ft 3 5.25 5.25 CHF correlation BWC BWC
CHF correlation DNB limit 1.18 1.18 i
f. M. imum DNBR

! c 102% power 1.78 1. 78 (b) 1 at 112% power 1.54 1. 54 (b) (*) Used in the analysis. (b) Calculated for the instrument guide tube subchannel which is limiting for the Mark-B8A fuel assemblies. I 6-3 B&W Fuel Company g

W Rev. I 10/91 I l l 7, ACCILENT AND TF.ANSIENT ANAINSIS I 7.1 General Srfety Analn ),3 Each USAR accident analyst: has been examined with respect to changes in the cycle 8 parameters to determine the of fects of the cycle 8 teload and to ensure that thermal performance durins hypothetical transients is not degraded. The effects of fuel densification on the USAR occi6cnt results have been evaluated and are reported in reference 13, l The radiological dose cont.equences of the Usto Chnpter 15 accidents have been evaluated using conservative radionuclide source terms that bound the cycle specific source term f or Davis-Besse Unit I cyclo 8. The dose calculations were

performed consistent with the assumptions described in the Davis-Besse Unit 1 USAR but used the more conservative source terms (which bound future reload cycles). The results of the dose evaluations showed that offsite radi ologi c t.1 doses for each accident were below the respective acceptance criteria values in the current NRC Standard Review Plan (NUREC 0800).

7.2 Accident Evaluation The key parameters that have the greatest effect on determining the outcome of a transient can typically be classified in three major areas: (1) core thermal, (2) thermal-hydraulic, and (3) kinetics parameters including the reactivity feedback coefficients and control rod worths. Fuel thermal analysis parameters from each batch in cycle 8 are given in Table 4-1. The cycle 8 thermal-hydraulic maximum design conditions are presented in Tabl e 6 - 1. A comparison of the key kinetics parameters from the USAR and cycle 8 is provided in Table 7 1. The EOC moderator temperature coefficient listed in Table 7-1 for cycle 8 is the 3-D, hot full power (HFP) temperature coefficient. An evaluation wa.s performed to verify the acceptability of the more negative cycle 8 moderator temperature I coefficient for all USAR accidents excluding steam line breaks. The results of

   -                                                                          7-1 l

l B&W FuelCompany

Rev. 1 l 10/91 = the evaluation were acceptable for all USAR accidents , excluding steam line breaks, for a moderator temperature coefficient as negative as -4.0 x 10-2 uk/k/* F. .he steam line break accident was evaluated based on a combined moderator and Doppler temperature coefficient from 532*F to the minimum temperature reached during the event. The combined temperature coefficient used in safety analysis of the steam line break is the sum of the EOC moderator and Doppler coef ficients ( 3.10 x 10-2 uk/k/*F) . The combined temperature coefficient for EOC 8 is shown in section 5, Table 5-1, footnote (f) as -2.67 x 10 2 ut/k/* F . Since the safety a;,alysis value for the combined temperature coef ficient is more negative than the = - cycle 8 value, the steam line break analysis remains bounding for cycle 8. A generic loss-of-coolant accident (LOCA) analysis for B&W 177-FA raised-loop I nuclear steam system (NSS) has been performed. The final acceptance criteria B6W ECCS evaluation model techniques and assumptions, as described in BAW-10104P, Rev. 53, were used in this analysis. The application of the evaluation model" l includes the impacts of the NUREC-0630 fuel pin rupture curves and FLECSET reflooding heat transfer coefficient calculations. In addition, the BWC C11F correlation was used to determine the time of DNB. The combination of averag fuel temperatures as fur.ctions of linear heat rate (UIR) and the lifetime pin pressure data for the Mark BSA fuel used in the LOCA lllR limits analyses is bounded by those calculated for the B&W 177-FA raised loop plant for previous reload evaluations" A tabulation showing the allowable LOCA UIRs for Davis- I E Besse Unit 1, cycle 8 fuel is provided in Table 7-2. g It is concluded by the examination of cycle 8 core thermal, thermal-hydraulic, and kinetics properties, with respect to acceptable previous cycle values, that this core reload will not adversely affect the ability to safely operate the Davis-Besse Unit 1 plant during cycle 8. Considering the previously accepted design basis used in the USAR and subsequent cycles, the transient evaluation of cycle 8 is considered to be bounded by previously accepted analyses. The initial conditions of the transients in cycle 8 are bounded by the USAR, the fuel densification report, and/or subsequent cycle analyses. I 7-2 B&W Fuel Company g a

Rev. 1 i

 "                                                                                       10/91
     ~

( Iable 7-1. Comparison of Mey Pa._rpeeters.lo.r Accidenl_Annlysis I USAR and de ns,1 f ' n report Cycle 8 Parameter value value BOL ") Doppler coeff, 10'3, tak/k/*F 1.28 -1.59 l EOL *) Doppler coeff, 10~3, *Ak/k/*F - 1. 4 5 " ) 1.96 EOL moderator coeff, 10~ 2 , t ak/k/* F 40.13 -0.66 EOL moderator coeff, 10'2, %Ak/k/*F -3.0 3.33(d) l EOL temperature coeff, ($32F to 510F), 10 ~ 2 , VAk/k/"P -3.10 -2.67 All rod bank worth (llZP), tak/k 10.0 6.92 i horon reactivity worth (llFP), ppm /ltak/h 100 138 Max ejected rod worth (liFP), vAk/k O.65 0.17 l Max dropped rod worth (llFP), kok/k O.65 $0.20 Initial toron cone (llFP ) , ppm 1407 1515 l I

4) BOL denotes beginning of life
        *)  EOL denotes end of life
        ")  -1.77 x 10'3 tak/k/*F war, used for steam line failure analysis.

(d) Moderator coefficient is bounded by generic plant analyses value of

            -4.00 x 10-2 tak/k/*F at HFP.

Table 7-? Enundinn Values ter Allowable LnCA Peak Lipear lient Potes Allowable Allowable Core peik IJIR, peak LilR, 0 40,000 mwd /mtU ufter 40,000 mwd /mtU I, elevation, ft kVjft kW/ft 2 16.0 16.0 4 15.75 15.75 6 16.5 18.0 8 17.25 17.25 10 17.0 17.0 7-3 I l B&W Fuel Company

L f~ L l'

8. PROPOSED MODIFICATIONS TO CORE OPERATING LIMITS REPORT The Core Operating Limits Report (COLR) has been revised for cycle 8 operation to accommodate the influence of the cycle 8 core design cn power peaking, reactivity, and control rod worths. Revisions to the cycle-specific parameters were made in accordance with the requirements of NRC Generic Letter 88 16 and Technical Specification 6.9.1. 7. The core operating limits were determined f rom i a cycle 8 specific power distribution analysis using NRC approved methodology provided in the references to Technical Specification 6.9.1.7.

The core operating limits are b.. s.e d on an ECCS bounding analysis that was performed to determine the allowable LOCA linear heat rate limits for the B&W 177 fuel assembly raised loop plant. The analysis inccrporated the NUREG 0630 cladding swell and rupture model, TACO 2 code, the IQC CHF correlation, and the E6V modified version of FLECSET reflooding heat transfer coef fielent correlation. Figures 8-1 througt 8-16 are revisions to the previous cycle operating limits contained in the COLR. Table B-1 presents the quadrant power tilt limits for I cycle 8 and Table B 2 provides the negative moderator temperature coef ficient for cycle 8. Based on the analyses and operating limit revisions described in this report, the Final Acceptance ECCS limits will not be exceeded, nor will the thermal 'esign criteria be violated. I 8-1 1 1 B&W Fuel Company

I I figure 8-1. Regulating Group Position Limits 0 to 200:10 EFPD I Four RC Pumps -- Davis-Besse 1, Cycle 8 (200,102) (277,102) (300,102)

                                                                                                                                                     ^

POWER LEVEL t 100 - CUT 0FF=1001 (266,92) SHUTOOWN N 80 (250,80) f{ 4 b OPERATION { RESTRIC1ED

                                      #                                                                                                                             g 60        -

UNACCEPTABLE C OPERATION b N D 40 -

                                       .a                                                                                             ACCEPTABLE E                                                                                             OPERATION E                                                                                                                          E
                                      $                                                           (98,28.5)                                                         E h20                     -

g f 5 0 ' i i l i I I I I I I i 0 100 200 300 mm Rod Index (% Withdrawn) g GR Si 1 1 0 75 100 GR 6 I I I I O 25 75 100 g GR 7 I I 3 0 25 100 I 8-2 B&W Fuel Company g

                                                                                                                                                                      =

I I l 1 l Figure 8-2. Regulating Group Position Limits, 200:10 to 40010 EFPD Four RC Pumps -- Davis-Besse 1 Cycle 8 (279,102) (300,102)

                                                                                                                ^

100 POWER LEVEL 0  ; CUT 0FF=100% SHUT 00WN UNACCEPTABLE MARGIN g 80 OPERATION LIMIT (250,80) I W 2 E (250,60) 60 - (218,60) p I i cc E T 40 - 8 ACCEPTABLE E OPERATION E (178,28.5) b1 e 20 - (0,3)' 0 I e i l i e i l i I I i .g 0 100 200 300 E Rod Index (% Withdrawn) GR 5 I I I O 75 100 GR 6 I I I I ,g -0 25 75 100 g GR 7 e i 1 0 25 100 I 8-3 B&W Fuel Company g

I' E Figure 8-3. Regulating Group Position Limits After 40010 EFPD four RC Pumps -- Davis-Besse 1, Cycle 8 (273,102) (300,102)

                                                                                                     ^

100 POWER LEVEL CUiOFF=1001 80 - NACCEPTABLE

                                                             /                                    (250,80)

I OPERAT10N L DilT f ' s 60 - (218,60) ,(250,60) I OPERATION s RESTRICTED E

          ~

ACCEPTABLE { OPERATION j a. (178,28.5) E 20 - (0,3)' 0 ' ' ' l ' ' ' ' ' ' ' ' O 100 200 300 Rod Index (% Withdrawn) GR 5 i f I O 75 100 GR 6 I I l I , 0 25 75 100 GR 7 ' ' ' E N 0 25 100 I

                                                      *~'

B&W Fuel Company 3

I I I 3 I Figure 8-4. Regulating Group Position Limits,0 to 200110 EFPD, Three RC Pumns -- Davis-Besse 1, Cycle 8 I 100 - UNACCEPTABLE OPERATION h 80 - (200,77) (277,77) (300,77) I ' SHUT D0'n'N MARGIN (266,69)

                                                                                                                                      ~
               $                                                        LIMIT I             5 e                    60   -

I (250,60) W l E w OPERATION RESTRICTED

                 $                                                (130,45.5)                                           8(250,45)
                   $                 40   -

I D I 20 - (98,21.9) ACCEPTABLE OPERATION I (0.2.8) 0' ' i I I I I I I I I I I I O 100 Rod Index (% Withdrawn) 200 300 GR 5 8 I I 0 75 100 GR 6 I I i i I O 25 75 GR 7 i 0 100 25 i 100 i I 8-5 g B&W Fuel Company

I I

                                 -: 'i.%

Figure 8-5. Regulating Group Position Limits, 200!10 to 400!10 EFPD Three RC pumps -- Davis-Besse 1, Cycle 8 I 100 - I (279,77) (300,77) S 80 - 2 a 2 g UNACCEPTABLE SHUTDOWN y OPERATION MARGIN

  • 60 -

LIMIT (250,60) o E is (218,45.5) >(250,45) I

                ~

OPERATION 5 RESTRICTED S E b E b 20 (178,21.9) I ACCEPTABLE OPERATION O' I I I I I I I I I I I I O 100 200 300 g Rod Index (% Withdrawn) E GR 5 I I i 0 75 100 GR 6 1 1 I I 0 25 75 100 GR 7 I I I 0 25 100 I a-6 B&W FuelCompany

I I I Figure 8-6. Regulating Group Position Limits, After 400!10 EFPD, I Three RC Pumps -- Davis-Besse 1, Cycle 8 I 100 - 'g UNACCEPTABLE E OPERATION b 80 - (273,77)_(300,7,7)

                                                                                                                    ~

E  ? E I cc E { 60 SHUTOOWN MARGIN (250,60) LIMIT I W E

      'B                                                                      (218,45.5)              ,(250,45) j 40      -

E i OPERATI0tl RESTRICTED { 2 20 - (178,21.9) ACCEPTABLE OPERATION I- (0,2.8)' 0 i e i l i i I I i i i i 0 100 200 300

.I                                                           Rod Index (% Withdrawn)

GR 5 I I I 0 75 100 GR 6 I t i I I O 25 75 GR 7 I 0 100 25 I 100 i I B&W Fuel Company g

I Figure 8-7 Control Rod Locations for Davis-Besse 1 Cycle 8 x I I A B 4 6 4 C 2 5 5 2 D 7 8 7 8 7 E 2 5 5 2 F 4 8 6 3 6 8 4 G 5 1 1 5 H W- 6 7 3 4 3 7 6

                                                                      -Y I 5 K              I     5            1        1l L                4       8      6      3 l6        8       4 il        ,        l2         5              l      5      2 N         l        l     7      8      7     l8        7 0         l    l              2   5        5l       2 P         l        l            4      6       4 R         l            l    t I                               =

2 E 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 I X Group Number Group No. of Rods Function 1 4 Safety E 2 8 Safety 5 3 4 Safety 4 9 Safety 5 12 Control 6 8 Control 7 8 Control 8 _8 APSRs Total 61 I 8-8 g M

I I F igure 8 -8, AFSR Position Limits, O to 400 210 EFPD, Four RC Pumps -- Davis-Besse 1. Cycle 8 I I I . 100 '- (0,102) RESTRICTED REGION (100,102)

                                                                                                                                   ~

I 80 - I E M l 5 5 60 - PERMISSIBLE a OPERATING REGION I g I I 40 - a E e I 20 - 2 I O e e i i i e i , i i 0 10 20 30 40 50 60 70 80 90 100 APSR Position (% Withdrawn)

  'I I
                                                                                                   *~'

l B&W Fuel Company

I I Figure 8-9. APSR Postion Limits After 400 10 EFPD, Three or Four RC Pumps. APSRs Withdrawn -- Davis-Besse 1. Cycle 8 I I , 100 I I 80 - e s 1 5 60 - A NS ON NOT AU.0W0 IN THIS TIME INTERVAL b a t 40 - 4 8 5 ~ k E [ 20 ~ 1 I O 0 10 20 30 40 50 60 70 80 90 100 l AP?,R Position (% Withdrawa) I I I 8-' I B&W Fuel Company g

I figure 810. APSR Position Limits, O to 400!10 EFPD, ) 'I Three RC Pumps -- Davis-Besse 1. Cycle 8 I I 100 - f 80 - RESTRICTED REGION cc tv y (0.77) (100,77) I c.

E y 60 -

! S ll 3 y 40 - PERMISSIBLE ,I 3 OPERATING REGION li S

y E b 1

g o- 20 -

.5

'I

,        O         i     e      i      i      i           i   i     i     i      i JB         0       10   20     30    40      50         60   70    80    90   100 APSR Position (% Withdrawn)

}l t I 8-11 I B&W FuelCompany f

                                    . . _        ..       . 4                ._             __ _   .m I

Figure 8-11. AXJAL POWER IMBALANCE Limits, O to 200 1 10 EFPD Four RC Pumps -- Davis-Besse 1, Cycle 8 I

                                               . 110

(-18,103) - -

                                               ~
                                                    ~100      l(14,102)

(-20,92) ( #'

                                               --     90

(-30,80)4 -

                                                   - 80             >(18,80)

I RESTRICTED -- RESTRICTED REGION cc

                                              --      70                   REGION 5

2 (-30,60) o _ , -- 60 o (18,60) I cc PERMISSIBLE

  • W -- 50 OPERATING REGION Pu
                                          $-- 40 t

t-- 30 W e E w

b. - '20 l

5

                                           $..         10 I       t      1       I     I                         I        i       1       I    i I

j -50 30 -20 -10 0 10 20 30 40 50 l AXIAL POWER IMBALANCE (%) l I I I 8-12 B&W Fuel Company g

I 5

!I lI    figure 8-12. AX1AL POWER IHBALANCE Limits, 200 10 to 400110 EFPD Four RC Pumt,s -- Davis-Besse 1, Cycle 8
                                           - 110

(-20,102) ;;; , ) gg  ; (14,102) ,I (-20,92) > t (15,92) , - 90 . (-30,80)4 -- 80 >(18,80) i cc a RESTRICTED o RESTRICTED 70

REGION REGION i N

, (-30,60) q ,

                        ~ PERMISSIBLE c
                                        $-- 60               < >(18,60)

,3 OPERATING e jg REGION ce 50 l . 40 E g._ 30 I- -

;g                                      '

g-- 20 'E

                                            -      10 l      I               I     I                I        i      f      I     i

!3E -50 -40 -30 -20 -10 0 10 20 30 40 50 AX1AL POWER IMBALANCE (%)

I il l

l 8-13 ' g B&W Fuel Company )

I I Figure 8-13. AXIAL POWER IMBALANCE Limits, After 400!10 EFPD I Four RC Pumps -- Davis-Besse 1. Cycle 8

                                       . 110

(-23,102) . (20,102)

                                         - 100

(-30,92) 4 o(20,92) 90 I-30,80) o -- 80 o(20,80) I RESTRICTED " PERMISSIBLE RESTRICTED REGION OPERATING g-- 70 REGION REGION (-30,60) o 3- - W 60 o(20,60)

                                  =

o- - 50 2 I x g g- - 40 3 4 3 g' -- 30 l g- - 20 5 a 10 l i 1 i f f I f f -50 -40 -30 -20 10 0 AXIAL POWER IMBALANCE (%) 10 20 30 40 50 l I I I! 8-14 B&W Fuel Company g

                                                                               =1

1 I I Figure 8 14 t.ilAL POWER IMBALANCE Limits, O to 200210 EFPD I Three RC Pumps -- Davis-Besse 1, Cycle 8

                                                                                    -- 110
                                                                                    -    100 I                                                                                  .. 90

(-13.5,77j -- 80 ,(10.5,77)

                                                                         ,                   v

(-15,69) g-- 70 t'(11.2,69) I (-22.5,60)q. E a. a- - 60 < >(13.5,60) RESTRICTED h RESTRICTED I REGION (-22.5,45) o W A-S 50 REGION q>(13.5,45) g- - 40 o I w y- - 0 E 30 di S. - - 20 l G;5

                                                                    $5G u

i EfW 2- - 10 l I I 1 t* i I I f I i 1 I -E; -40 -30 -20 -10 0 AY.lAL POWER IMBALANCE (%) 10 20 30 40 50 I I I 8-" I B&W Fuel Company

I I Figure 8-15. AX1AL POWER IMBALANCE Limits, 200 10 to 400!10 EFPD I Three RC Pumps -- Davis-Besse 1, Cycle 8

                                                                                                                  -     110
                                                                                                                     - 100
                                                                                                                  --     90
                                                                                                                  --     80 I

(- 15,77 ) ::  : (10.5,77) 0 (-15,69) > 4 (11.2,69) (-22.5,60)q @-- 60 m(13.5,60) ' RESTRICTED T RESTRICTED E REGION

                                                                                                             $.- 50                      REGION                         5

(-22.5,45) o E o (13.5,45) 40 cr n E-- F 30 m. E w 5 e9 E- - 20 m s GCE 2 g '

                                                                                     $sG                     ~

EWW b- - 10 e a a E g i i l i i f I I i 1

               -50                       -40    -30                      -20                             -10      0         10        20     30     40   50              g AXIAL POWER IMBALANCE (%)                                                                m I

I a-16 I\ I B&W FuelCompany gi

                                                                                                                                                                         =

l I I figure 8-16. AX1AL POWER IMBALANCE Lin.its, After 400110 ETPD Three RC Pumps -- Davis-Besse 1, Cycle 8 I.

                                                                                                                  - 110
                                                                                                                  -  - 100
                                                                                                                  --     90

(-17.2,77) , - 80 (15,77) I ( 22.5,69)q(

                                                                                                    /              --    70      ,  ,(if.69)

(-22.5,60) ti g-- 60 < >(15,60) RESTRICTED a RESTslCTED REGION REGION l-22.5.45)O E i >(15.45)

                                                                                                               $-- 40 l                                                                                                 N e>

D-- 30 l ggg rse - p- - 20

                                                                                                    @ l3 "       g,--     10 I                        i                   l                                             i       I      i  ,

e l I I I I

                          -50    -40 I                                                                                         -30     -20    -10        0 AX1AL POWER IMBALANCE (%)

10 20 30 40 50 I I I 8-17 B&W FuelCompany

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

I s I' l 1 i litltlt.hl JAldiUH1t I'_ogy r T Lil.,_Linh5 . Steady state Steady state i Llan t t for Limit for ' Quadrant l'ower Tilt The rical The rital T r ans. i e n t Ma x iinurri as steasured by: Pow r s 60t Power 2 60t Li rr i t Limit Synaretrical incore 6.83 4.11 10.03 20.0 detector system Powe r I!ange charinels 4.05 1.96 6.96 20.0 ti.i nimutt incore detector 2.80* 1.90* 4.40* 20.0* system iAs sun >e s detector s t ri ngr, with >604 depletion are excluded from the m inin;urt incore t.ystem configuration, l I I' Inble 8 ?. tiera tixe Moderat odnnterat ere Coef f irAnt Limu tiegative fioderator Temperature - 3. 73 x 10" Ak/k/*r E, , Coefficient Llinit E I (at RATED 'lHtaMAL POWER) i i I. i I I l 8 18 I I' B&W Fuel Company g

I I I 9, STARTLIP PROGPJJi - tilYSICS TLSTING I The planned startup test program ass-ociated with core performance is outlined I below. These t est s verif y that core perf ormance is within the assuttptions of the saf ety analysis and provide inf orn:at ion f or continued saf e operat ion of the unit . 9.1. Precritical Tests LLI . Cont rol PoLTrio 'lest Precritical con '. rod drop times are recorded for all control rods at hot I full flow condi,2cnp hofono n ro power physics to. ting begins. Acceptance criteria state that t he rod drop time f rom f ully withdrawn t o 754 inn.erted shall be less than 1.Sc necondr, at the conditions above I It should be noted that saf e ty analys.15. calculations at e based on a nod drop f rom fully withdrawn to two thirds inserted. Since the most accurate position I indication is obtained from the rone reference t: witch nt the 754 inoerted position, this position is ust d instead of the t wo thirds inserted posit ion f or data gathering, 0.1.2. PC Flow Reactor coolant flow with four RC pumps running will be measured at hot standby conditions. Acceptance criteria require that the measured flow be within allowable limits. 9.?. Zero Power Physics Tegn 9.? 1. Critieni P.oron Concentra.i.inD Once initial criticality is achieved, equilibrium boron is obtained and the I critical boron concentration deter.nined, The critical boron concentration is calculated by correcting for any rod withdrawal required to achieve the all rods out equilibrium boron. The acceptance criterion placed on critical boron concentration is that the actual boron concentration must be within i 100 ppm boron of the predicted value. 91 B&W FuelCompany

I 9.2.2 Tettptratyre Reactjvi t y roef ficie nt The i s ot he rttal llZP t ertpe ra t u r e coefficient is neasured at approximately the g all tods out configuration. During changes in t e n,perature , reactivity feedback E may be con:pensated by control rod moven.ent. The change in reactivity is then calculated by the sun.niation of reactivity associated with the tenperature change. Acceptance criteria state that the n>casured value shall not differ from the predicted value by more than t 0.4x10 2 g 3pfpf t y, g

                                                                                              "   L^

The moderator coeffielent of reactivity is calculated in conjunction with the t emperature coef ficient n ensuren: erit . Af ter the tenterature coef ficient has been neasured, a predicted value of fuel Doppler coefficient of reactivity is subtracted to obtain the moderat or coef ficie nt . This value must not be in excess of the acceptance criteria limit of 40.9x10'8 t t.k/k /"r. 9.7.3. Control Pod Groupp oron Rec.ctivity Worth Individual control rod group reactivity worths (groupo 5, 6, and 7) are measured at hot zero power conditions using the boron / rod swap method. This technique consists of deborating the reactor coolant system and contensating for the reactivity changes from this debo.ation by inserting individual cantrol rod groups 7, 6, and 5 in incremental steps. The reactivit y changes that occur during these nic a s u r e me n t s are calculated based on reactimeter data, and differential rod worths are obtained from the measured reactivity worth versus the change in rod group position. The different tal rod worths of each of the controlling groups are then somned to obtain it;tegral rod group worths. The E acceptance criteria for the control bank group worths are as follows: E

1. Individual bank 5, 6, 7 worth:

predicted value menruled value 100 < 15 measured value -

2. Sums of groups 5, 6, and 7:

predicted valut - meastir,til valut x 100 < 10 W measured value - I 9-2 I i B&W FuelCompany g

L The boron reactivity wor;h (dif f erential bcron worth) is measured by dividing the total insen ted rod worth by the boron change trade f or the rod wort h t est . The acceptance criterion f ar measured dif f erential baron worth is as follows:

1. predic:cd vq1ue - trensured valge steasured value x 100 "< 15 The predicted rod worths and differential boron worth are taken from the PTM.

0.3. Power Pscalation Tests 0.3.1. Cnre Symretry Test The purpose of this test is to evaluate the symmetry of the core at low po9er during the initial power escalation f ollowin;; a ref ueling. Symmetry evaltation is based on incore quadre.nt power tilts during escalation to the intermediate power level. The core symmetry is acceptable if the absolute sa hie s of the i quadrant power tilts are less than the error adjusted alarm limit. 9.3.2. Core Power Distribution Verification at I n t e rn:e di a t e rower Level I (IPL) and ~1004 TP Vit.b Non-inal Contrni Rod Position Core power distribution tests are performed at the IPL rnd approxirtately 100% full power (FP) . Equilibrium xenon is established prior to both the IPL and 1004 FP tests. The test at the IPL is essentially a check of the power distribution in the core to identif y any abnormalities before escalating to the 1004 FP plateau. Peaking f actor criteria are applied to the IPL core power distribution results to determine if additional tests or annyses are required prior to 100t I FP operation. The following acceptance criteria are placed on the IPL ano inO4 FP tests:

1. The maximum LHR must be less than the LOCA limit.
2. The minimum DNER must be greater than the initial condition DNBR limit.
3. The value obtained from extrapolation of the minimum DNBR to the next power plateau overpower trip setpoint must be greater than the calculated 112%

FP DNBR value, or the extrapolated value of istbalance must f all outside the RPS power / imbalance / flow trip envelope. 9-3 B&W FuelCompany

I

4. The value obtained f rom extrapolation of the worst case itax: rtum UlR to the next power plateau overpower trip setpoint must be less thai the fuel melt limit, or the extrapolated value of imbalance must fall o itside the RpS power /in. balance / flow trip envelops.
5. The quadrant power tilt shall not exceed the limits specifie d in the Cout,
5. The highest measured and predicted radial peaks shall b* within the fo11owin6 litni t s :

predicted value - treamred value measured value x 100 must be more positive than 5 g a

 ?       Tb highest measured and predicted total peaks shall be sithin the tollowing limits:

nLedicted value mensured value measured value 100 m h m e mi t M h 7. 5 I tems 1, 2, and 5 ensure that the ini t ial condit ion LOCA, initial conditiot DNBR, and quadrant power tilt limits respectively are maintained at the IPL anc 100% FP. Items 3 and 4 establish the criteria whcreby escalation to full power may be accomplished without exceeding the saf ety limits specified by the saf ety analys, s with regard to DNER and linear heat rate. Items 6 and 7 are established to determine if measured and predicted power distributions are consistent. g 9.3.3. Incore Vs. Excore Detector Imbalance I Correlation Verification at the IPL. Imbalances, sec up in the core by control rod positioning, ar, read simultaneously en the incore detectors and excore power range detec' ors. The excore detector offset versus incore detector offset slope must be rreater than 0.96, If this criterion is not met, gain amplifiers on the excore detector s signal processing equipn:ent are adjusted to provide the required slope. I 9-4 I I B&W FuelCompany B nn 1

l I l I 4.3.4. Te gerature Penetivity Coefficient at -1004 FP The average reactor coolant t<nperature is decreased and then increased at constant reactor power. The reactivity associated with each temperature change is obtained f rom the change in the controlling rod group position. Controlling I rod group worth is treasured by the f ast insert / withdraw method. The temperature reactivity coef ficient is calculated f rom the measured changes in reactivity and temperature. Acceptance criteria state that the moderator temperature coefficient shall be negative. L3. 5. Powe r Dopol e r Peite t ivi t y rnellic ient at -1004 FP The power Doppler reactivity coef ficient is calculated f rom data recorded during control rod worth measurements at power using the fast insert / withdraw method. The fuel Doppler reactivity coetficient is calculated in conjunction with the power Doppler coef f leient measurement . The power Doppler coe f f icicot as measured above is multiplied by a precalculated conversion factor to obtain the fuel Doppler coefficient. This treasured fuel Doppler coefficient must be more negative than the acceptance criteria limit of 0. 90 x 10 % ok/k/"F. L!. . Procedure for Use if Acceptance c r1Leria No t Mel If acceptance criteria for any test are not met, an evaluation is performed before the test program is continued. This evaluation is performed by site test personnel with participation by BW !Nelear Technologies t echnical personnel as required. Further specific actions depend on evaluation results. These actions can include repeating the tests with more detailed test prerequisites and/or steps, added tests to search for anomalics, or design personnel performing I detailed analyses of potential safety problems because of parameter deviation. power is not escalated until evaluation shows that plant safety will not be compromiscd by such escalation. I I I e.s B&W FuelCompany g 4

I Rev. I 10/91 i

10. REFERENCES I 1. Davis Besse Nuclear Power Station Unit 1. Cycle 8 Reload Report, MV -

2137, June 1991.

 #*               2. Davis-Besse Nuclear Power Station No.                              1, Updated Safety Analysis Report, Docket No. 50 346,
3. Program to Determine In Reactor Performance of B&W Fuels - Cladding Creep Collapse, EV 10084P Pev. 2, Babcock and Vilcox, Lynchburg, VA, October 1978.

I 4. Letter, J . ll . Taylor (B6J) to C.O. Thomas (NRC),

Subject:

Creep Collapse Analysis f or B6V Fuel, Jilt /86 011 A, Dated January 31, 1986.

5. Letter. Dennis M. Crutchileid (NRC) to J . ll . Taylor ( P,6V) ,

Subject:

Acceptance for Ref erencing of a Special Licensing Report, Dat ed December 5, 1986,

6. TACO 2: Fuel Performance Analyt.is, ImW- 1014.1hAAQ , Babcock 6 V11cox ,

Lynchburg, Virginia, June 1983.

7. NOODLE - A Multi Dimensional Two Group Reactor Simulator, IMW 1015? A ,

Babcock 6 Vilcox., Lynchburg, Virginia, Jutm 1985. 8, LYNXT Core Transient Thermal llydraulic Program, BAV-10156- A, February 1986.

9. Davis Besse Nuclear Power Station Unit 1, Cycle 7 - Reload Report, BAV-2006, Noven.ber 1989
10. Thermal llydraulie Crossflow Applications, IAV-1829, April 1984.
11. Rancho Seco Cycle 7 Reload Report - Volume 1 Mark BZ Fuel Assembly Det.lgn Report, BAV 1781P, April 1983.
12. BVC Correlation of Critical llent Flux, LAV - 1014 )P .- A , Babcock & Wilcox, Lynchburg, Virginia, April 1985.

Davis Besse Unit 1 Fuel Densification Report, MV - 1401, Babcock 6 Vilcox, I 13. Lynchburg, Virginia, April 1975. 14, B&W's ECCS Evaluation Model. MW- 10104 P Rev. 5, Babcock 6 Wilcox, I 15, Lynchburg, Virginia, April 1986, ECCS Evaluation of B6V's 177 FA Raised Loop NSS, BAV-10105. Rev.1, Babcock

                          & Wilcox, Lynchburg, Virginia, July 1975.

10-1 I B&W FuelCompany R __ - _-_____-__ _- ____ __ . _ _}}