ML17252B197

Second Reload License Submittal

ML17252B197
Person / Time
Site:	Dresden
Issue date:	09/30/1973
From:	General Electric Co
To:	US Atomic Energy Commission (AEC)
References
Download: ML17252B197 (100)

Text

File Cy.

'°Regulatory U..:.*- . ' . ---- -

DRESDEN 3 NUCLEAR POWER STATION SECOND RELOAD LICENSE SUBMITTAL I) ....... ~

SEPTEMBER 1973 PREPARED BY GENERAL ELECTRIC COMPANY

. NUCLEAR FUEL DEPARTMENT

TABLE OF CONTENTS Section Page

1. INTRODUCTION 1-1

2.

SUMMARY

2-1

3. MECHANICAL DESIGN 3-1

4. THERMAL-HYDRAULIC CHARACTERISTICS 4-1

5. NUCLEAR CHARACTERISTICS 5-1

6. SAFETY ANALYSIS 6-1

7. TECHNICAL SPECIFICATIONS 7-1 iii/iv

LISTOF ILLUSTRATIONS

• Figure Title Page 2-1 Dresden 3 Cycle 3 Design Reference.Core Loading 2-2 3-1 8 x 8 Reload Fuel Assembly 3-2 3-2 Dresden 3 R2 Reload Fuel Lattice 3-3 5-1 Hot Average Void Infinite Lattice K00 versus Exposure 5-2 5-2 2.50 wt% U-235 Bundles, Infinite Lattice K 00 versus In-Channel Void Fraction 5-4 5-3 ~K Void Comparison 7 x 7 versus 8 x 8 from 0.40 Void to Other Voids 5-5 5-4 200 MWd/t Doppler Coefficients Uncontrolled 5-7 5-5 10,000 MWd/t Doppler Coefficients Uncontrolled 5-8 5-6 (3 versus Exposure, Average Voids, Uncontrolled, 2.50 st% Bundles 5-10 5-7 Maximum Local Peaking versus Exposure 2.50 wt%

U-235 Bundles Average Voids 5-11 6-1 Doppler Reactivity Coefficient vs Average Fuel Temperature as a

• Function of Exposure and Moderator Condition 6-7 6-2 Accident R activity Shape Functions for Cold Startlip 6-8 6-3 Accident Reactivity Shape Functions for Hot Startup 6-9 6-4 Scram Reactivity Function for Cold Startup 6-10 6-5 Scram Reactivity Function for Hot Startup 6-11 6-6 Cladding Temperature versus Time OBA with Failure of LPCI Injection Valve (2CS +ADS+ HPCI) AEC Analysis #1 15000 MWd/t Exposure 6-14 6-7 Performance of ECCS with Failure of HPCI for a Small (.02 ft2) Break.

(4 LPCI + 2CS +ADS) AEC Assumptions 6-16 6-8 Cladding Temperature versus Time for a Small Break with a Failure of HPCI (0.02 ft2) Break 4 LPCI + 2CS +ADS AEC Assumptions 6-17 6-9 Peak Cladding Temperature Spectrum for a Single Failure Condition with AEC Assumptions 6-18 6-10 Emergency Core Cooling System versus Break Spectrum 6-19 6-11 Performance of ECCS for Main Steam Line Break inside Drywell with all ECCS Operating - AEC Assumptions 6-21 6-12 Core Flow and Pressure Following a Recirculation Line Break 6-22 v

LIST OF ILLUSTRATIONS (Continued)

• -Figure Title Page 6-13 Performance of ECCS with Failure of One Diesel Generator for the Design Basis Accident. ( 1 HPCI + 2 LPCI + 1 CS+ ADS) AEC Assumptions 6-24 6-14 Minimum Critical Heat Flux Ratio for DBA at Dresden 2/3 6-25 6-15 Cladding Temperature versus Time for an Intermediate Break with Failure of HPCI (0.1 ft2 Break) 4 LPCI + 2CS + ADS AEC Assumptions _ 6~26 6-16 Quality versus Time for DBA at Dresden 2/3 *-6-27 6-17 Heat Transfer Coefficient for a Small Break 0.02 ft2 4 LPCI + 2CS + ADS AEC Assumptions _ . - 6-28 6-18 Heat Transfer Coefficient for an Intermediate Break (0.1 ft2) 4 LPCI + 2CS +ADS AEC Assumptions 6-29 Heat Transfer Coefficients for DBA with LPCI Injection Valve Failure AEC Analysis #1 (2CS + HPCI +ADS) ;1

• 6-30 6-20 Performance of ECCS with Failure of HPCI for an Intermediate _

(0.10 ft2) Break. (4LPCI + 2CS +ADS) AEC Assumptions 6-31 6-21 Power Generation Following a Design Basis Recirculation Line Break Accident -* 6-33 6-22 Variation of Peak Cladding Temperature with Fuel Bundle Exposure, 6-34 6~_23 Fuel Rod Perforation Data 6-35 6-24 Distribution of Internal Pressure Within Rods 6-36 6-25 Percent Rod Perforation versus Break Area (AEC Assumprions) 5;37 6-26 8 x 8 Reload Fuel Rod Identification 6-44 6-27 Dresden 3 Cycle 3 RWE Response Case 2 6-45 6-28 Dresden 3 Cycle 3 RWE Response Case 1 6-46 6-29 Dresden 3 Cycle 3 RWE Respon~ Case 2 . 647 6-30 Dresden 3 Cycle 3 RWE Response Case 1 6-48 6-31 Dresden 3 Cycle 3 RWE Response Case 1 **.. **- 6-49

_6-32 Dresden 3 Cycle 3 RWE Response Case 2 50 6-33 _Dresden 3 Cycle 3 RWE Response Case 1 -* 6-51 vi

LIST OF ILLUSTRATIONS (Continued)

Figure Title* Page 6-34 Dresden 3 ABM Response to Control Rod Motion Case 1 (8 x 81 Channel A+C 6-52 6-35 Dresden 3 ABM Response to Control Rod Motion Case 1 (8 x 81 Channel B+D 6-53 6-36 Dresden 3 ABM Response to Control Rod Motion Case 2 (7 x 71 Channel A+C 6-54 6-37 Dresden 3 ABM Response to Control Rod Motion Case 2 (7 x 71 Channel B+D 6-55 LIST OF TABLES Title Title Page 2-1 Fuel Type and Number . 2-1 3-1 Initial Core and Reload Fuel Assembly Design Specifications . 3-4 3-2 Stress Intensity Limits 3-6 3-3 Summary of Leading Experience on Currently Operating Production Zircaloy-Clad U02 Pellet Fuel as of October 1, 1971 3-11 3-4 Summary of Production Fuel Experience Zircaloy-Clad U02 Pellet Fuel as of October 1, 1971 3-12 3-5 General Electric Developmental Irradiations Zircaloy-Clad 95%

TD U02 Pellet Fuel Rods 3-13 3-6 General Electric Developmental Irradiations Zircaloy-Clad 95% TD U02 Pellet Capsules General Electric Test Reactor 3-14 4-1 Results of Thermal-Hydraulic Analyses 4-4 5-1 7 x 7 to 8 x 8 Comparison of Physical Parameters 2.50 Bundles 5-3 5-2 Cold Reactivity Comparison-2.50 wt% Enrichment Zero Exposure 5-3 5-3 Nuclear Characteristics of the Design Reference Core . 5-9 6-1 Peak Cladding Temperatures 6-13 vii/viii

1. INTRODUCTION This document provides the technical basis of the license submittal for the second reload of the Dresden 3 Nuclear Power Station. Presented herein is a description of the new 8 X 8 fuel and the results of the evaluation of the refueled core for the January 1974 outage.

The design reference core loading is based on the use of as many as 60 8 X 8 bundles, comprised of reload 2 fuel assemblies having an average enrichment of 2.50 wt % U-235.

The objective of this outage is to provide reactivity augmentation for continued high load factor operation through to the next refueling anticipated in Spring 1975.

Sections 3, 4, 5 and 6 of this document, dealing with the subjects of reload fuel mechanical design and reloaded core thermal-hydraulic, nuclear characteristics, and safety analysis, present description of design criteria, methods and results from design calculations and safety evaluations and represents complete information for the review of fuel assembly and core design.

1-1 /1-2

2.

SUMMARY

the DreSden -Nuclear Power Station Unit 3 reload 2 fuel wi-11 employ an 8 X: 8 fuel a~sembly configuration instead of the previously used 7 X 7. The p~llet diamete~. pellet length, cladding diameter, and rod pitch are changed from the 7 X 7. design; however, the assembly exterior .dimensions remain unchanged. The basic materials and fuel fabrication process used for the reload 2 fuel a5semblies are the same as those _used on the 7 X 7 design.

The design reference con~ q:mfig~ration tor. this license consists of the bundles defined in Table 2-1. The relative locations of the 60 new 8 X 8 fuel buridles are shown in Figure 2-1. Cold shutdown evaluations have been made on full use of all 60 reload 2 bundles at* an incremental core average exposure of 2200 MWD/t from the spring 1973 refueling outage. :In addition, all of the remaining 140 temporary poison curtains are to be removed. Ample shutdown margin throughout the cycle has been *demonstrated. The flexibil\ty for fuel shuffling is included in this submittal.

Table 2-1 FUEL TYPE AND NUMBER Fuel Type Number Initial 612 Reload 1 52 Reload 2 60 Total 724 2-1

Figure 2-1 Dresden 3 Cycte* 3 Design Reference Core Loading 60 - - - - - - - - - - - - - - i .!  : l  :

58 - - - - - - - - - ' - - - - - ' -

56 I.

rI I

I

/ '\ I

. I I.

I

-_-1-:- -:-,---,

I I

.. I . / '

I

!° I *'

54--------*i ---:-:--l-- --: --:6- """'-:-- c.-:-- ""e_: __ --:-- '""-:-- :1 I I I I I I I I I

. 52 - - - - - . -

I I

I I

I I

I I

I I

I I . ' I I I ,.

I 1- .

I

. I i

50 ----r:-T- ""'-::e- ~-:-8 --rG~E--:0-.--:- -E~e-:-- - -r~A~e_! __ 9-:-- :-:- -"'.l

' ' I *I I ,E I I . *I *I I I ....... I I* . ,, ' I I 48 - - - I I I I I I I . I I I I **1 I

-- ""!""- -~ -- -- -- -- -- -- -- -- -- -- -- -- -- --,-- ---- -- ---- --

46 - - -  :. . :* B :G . :G  : B .. : G  : G : B :. GI G : B :. . -1 44 - - - I I I 'I I I I I I I I I I 42 I I

--r--:-~B;-r&- --:(;:A~-1-8 I I

--r(;: ~-:-e- I I

--:- ~

I

_e_: __

I I I c.-i-- E~8 :-- "GT-E~B-:-- --:-~E-,

I I 40 I I " I I ..... I ' I I I. I I ... I I ... I I -..}-f--

I I 38 --:-- __ :_B_ --:-ci --:-;

I I

--:o- -T6-I . I 1

--!~---:-::a:--

I I I I I 7;-i--c.-:--.-;r- -;:---;-:-- ,

I I __ __

1 36 I

I I

I I *' I I I I I I I I I I I I

~- -- ---- -- -- ---- -- I -- ---- -- -- -- -- ---- ----1--

I I I I I I I I

-- I I I

--,-- -- I --

34  :  ! G : B : G I B l B,,:-. i G : G: B : ,, ,.Bl G l Bl G 1 ,~

I I I ,,BI ,,B__,, I I ,,o,, I I IE,..,....,-

3320 --:--

I I

I 1 I ,_ I

• I _t-I I I I I I I I I I

,B I I I ,.. 'G I I B I I I I I I 28 , 1G 1G 1B ,B 1 , G 1 1 B 1 G 1 B 1 G 1 1

--,-- --,-- --,-.,. --,-- -- -- --,-- --,-- --,-- --,-- --,-- --i-- --i-- --,-- --,-- -:--.--

26 , 1 'B' , , 'B' I 1 c' , 1 c 1 1 D' 1 , 'D'

• 1 1 24  : l B" l G I B";J' l G lG .... : e  ; " B:

1 G: "G I Bl G: e: l

--,-- --,-- --,-- --,-- --,-- --1-- --,-- --,-- --i-- --,-- - - , - - - -,-- --,-- --,-- --,--

22 I I I I I I I I I I I I

• I I I 20 ---L __:____:_I!_ __ :_~I*1 __ :_~ __:~- __l_e___:___'!_: __ ~_: __ ~B_I ___G_l ___'!_ :____ l__ _J

_ ...._ ._ I I I I 1 I I I I I I I I 18 ' L__!_J"B' I I 1s I I c' I I c' I I 'D' I I 1

I I 16- >-->- IT"\!;.> 1 B 1 G"" 1 I G 1 B" 1G I ... G I1 e'I "G 1I G* 1 B 11 II

-L:-- --!-- --!-- ;_+-.--:-- --:----:----:--.--!-- --!-- --:--~ -- :-- --J:--

I I I I I 1 14 _ _ _ ...

12 I 1B I B 1 G IB 1 B 1 G 1 B I B I 1 10 08 06

-+---+--...

-t-+--t-T--t-~

I I ___:__

I I

I f_: ____

'c I I

I I "'c

_e~:_:_G I

I I

1 1

I I

I

'c I

___ l_~~G-~-- _B_:_: __ :__ :

I I

I I

I

'c' I I

I I

I I

"'D I 04 I I I I I I I I I I

-~,__--+--~f---+1~11--+-1 *7 --:- - - -:- -  : - -:- - - -:- - ......JI 02 01 03 05 07 09 11 13 15 17 19 I I I I I 21 23 25 27 29 31 I I I I III 33 35 37 39 41 43 45 47 49 51 53 55 57 59 LPRM LOCATION (LETTER INDICATES 15 TIP MACHINE) 14 13

• LPRM LOCATION (COMMON LOCATION FOR ALL TIP MACHINES) 12 11

@ lf!M LOCATIONS ~'l-+-+-+--+--+--+--+--+-+-+-+~1-- ......

10 t-+-+-+--+--+--+--+--t--t--t-t---tt--t--1 l::. BRM LOCATIONS 9t-+-+-T--+--+--+--t--t--t--t----tt--t--t 8 -+-+-+---l--l--+--+--+-+-+-+--11--1--t

• SOURCE LOCATIONS 7t-+-+--+--+--+--+--t--t--t--t----tt--t--t 6._t-t--11--11--11--1-t-t-t-t-t--t--t--fo....

5 G RELOAD 1 BUNDt.E (7X7, 2.30e) 4 B RELOAD 21lUNDLE l8X8, 2.509) 3 2

BLANK DRESDEN 3 INITIAL FUEL BUNDLE ABCDEFGHJKLMNPR 2-2

3. MECHANICAL DESIGN 3.1 GENERAL DESIGN DESCRIPTION The 8 x 8 fuel bundle contains 63 fueled rods and one spacer-capture water rod which are spaced and supported in a square (8 x 8) array by the upper and lower tie plates. (See Figure 3-1 .) The lower tie plate has a nosepiece which has the function of supporting the fuel assembly in the reactor. The upper tie plate has a handle for transferring the fuel bundle from one location to another. The identifying assembly number is engraved on the top of the handle, and a boss projects from one side of the handle to aid in assuring proper fuel assembly orientation. Both upper and lower tie plates are fabricated from T,ype-304 stainless steel castings.

Each fuel rod consists of high-density (95% TD) U0 2 fuel pellets stacked in a Zircaloy-2 cladding tube which is evacuated, backfilled with helium, and sealed by welding Zircaloy end plugs in each end. The fuel roe;! cladding thickness is adequate to be "free-standing," i.e., capable of withstanding external reactor pressure without collapsing a

onto the pellets within. Although most fission products are retained within the U0 2 , fraction of the gaseous products are released from the pellet and accumulate in a plenum at the top of the rod. Sufficient plenum volume is provided to prevent excessive internal pressure from these fission gases or other gases liberated over the design life of the fuel. A plenum spring, or retainer, is provided in the plenum space to prevent movement of the fuel column inside the fuel rod during fuel shipping and handling.

Three types of rods are employed in a fuel bundle: tie rods, a water rod, and standard rods. The eight tie rods in each bundle have threaded end plugs which thread into the lower tie plate casting and extend through the upper tie plate casting,. A stainless steel hexagonal nut and locking tab are installed on the upper end plug to ho_ld.the assembly together. These tie rods support the weight of the assembly only during fuel handling operations when. the assembly hangs by the handle; during operation, the fuel rods are supported by the lower tie plate. One rod in each' fuel bundle (see Figure 3-2) is a hollow water tube used to position seven Zircaloy-4 fuel rod spacers vertically in the bundle. The water .rod is a hollow Zircaloy-2 rod equipped with a square bottom end plug to prevent rotation and assure proper location of the water rod within the fuel assembly. Several holes are drilled around the circumference of the water rod at each end to allow coolant water to flow through the rod. The spacers are equipped with lnconel-X springs and maintain. rod-to-rod spacing. The remaining 55 rods in a*bundle are standard rods having a single tube of fuel pellets the same length as the tie rods. The end plugs of the standard rods have pins which fit into anchor holes in the tie plates.

An lnconel-X expansion spring located over the top end plug pin of each fuel rod keeps the fuel rods seated in the lower tie plate and allows them to expand axially and independently by sliding within the holes of the upper tie plate.

The fuel pellets consist of high-density ceramic uranium dioxide manufactured by compacting c!fld sintering uranium dioxide powder into cylindrical pellets with chamfered edges. The average U0 2 pellet immersion density is approximately 95% of theoretical density.

Four different U-235 enrichments are used in the fuel assemblies to reduce the local power peaking factor (see Figure 3-2). Fuel element design and manufacturing procedures have been developed to prevent errors in enrichment location within a fuel assembly. The fuel rods are designed with characteristic mechanical end fittings, one for each enrictiment. End fittings are designed so that it is not mechanically possible to completely put together a fuel assembly with any high enrichment rods in positions specified to receiv.e a lower enrichment. As in the 7 x 7 assembly design, the 8 x 8 bundle incorporates the use of small amounts of gadolinium as a burnable poison in selected fuel rods. The gadol inia-urania fuel rods are designed with characteristic extended end plugs. These extended end plugs permit a positive, visual check on the location of each gadolinium-bearing rod after bundle assembly.

Most aspects of the 8 x 8 bundle design are similar to the current 7 x 7 design. Specifically, the upper and lower tie plates, the fuel rod spacers, the upper and lower end plugs, and other associated bundle hardware are the same as the 7 x 7 except for modeling down in size to be compatible with the increased number of rods per bundle and the reduced rod diametral wall thickness. The 8 x 8 fuel assembly. outline dimensions are the same as the current 7 x 7 dimensions. Table 3-1 presents a summary of 8 x 8 design dimensions, and Figure 3-2 shows the location of the various fuel rod types within the reload-2 assembly.

3-1

TYPICAL FUEL ROD D

0.493 In. o~d.

UPPER TIE PLATE SPACER, LOWER TYPICAL TIE PLATE OF SEVEN CASTING SPACER DETAIL W SPACER CAPTURE/WATER ROD T TIE ROD Figure 3-1 8 x 8 Reload Fuel Assembly

WIDE-WIDE CORNER T T 4 3 2 2 2 2 2 3 3 2 1 1 1 1 1 2 T G G T 2 1 5 1 1 1 5 1 2 1 1 1 1 1 1 1 2 1 1 1 ws 1 1 . 1 . *'

T ' T 2 1 1 1 1 1 1 1 G G 2 1 5 1 1 1 5 1 T T

,,~..

3 2 1 1 1 1 1 *2 ROD ENRICHMENT NUMBER TYPE wt% U-235 OF RODS 1 2.73 40 2 2.06 14 3 1.80 4 4 1.40 1

\ 5 2.73 4 WS - 1 WS SPACER CAPTURE WATER ROD T TIE RODS G GADOLINIUM RODS Figure 3-2 Dresden 3 R2 Reload Fuel Lattice 3-3

Table 3-1 INITIAL CORE AND RELOAD FUEL ASSEMBLY DESIGN SPECIFICATIONS Initial Reload Fuel Core Fuel R1 R2 Fuel Assembly Geometry 7x7 7x7 8x8 High Enrichment Rods . 30 32 44 Medium High Enrichment Rods 16 10 14 Medium Low Enrichment Rods 3 6 4 Low Enrichment Rods . 0 1 1 Poison Rods 0 3 4 Water-Spacer Capture Rods 0 0 1 Rod Pitch (in.) 0.738 0.738 0.640 Water to Fuel Volume Ratio 2.42 2.53 2.60 Heat Transfer Area (ft2) 86.5 86.5 97.6 Fuel Rod Active Fuel Length (in.) 144.0 144.0 144.0 Gas Plenum Length (in.) 11.25 11.0 11.24 Fill Gas helium helium helium Getter no yes yes Fuel Material . sintered U02 sintered U02 sintered U02 Initial Enrichment, wt/% U-235 Average for Bundle 2.12 2.30 2.50 High . 2.44. 2.56 2.73 Medium High 1.69 1.94 2.06 Medium Low 1.20 1.69 1.80 Low . 1.33 1.40 Pellet Diameter (in.)*. 0.487 0.477 0.416 Pellet Immersion Density (%TD) 95.0 95.0 95.0 Cladding

• Material . Zr-2 Zr-2 . Zr-2 Thickness 0.032 0.037 0.034 Outside Diameter (in.) o.563 0.563 0.493 Fuel Channel Material

• Zr-4 Zr-4 Zr-4 Outside Dimension (in.). 5.438 5.438 5.438 Wall Thickness (in.) . 0.080 0.080 0.080 Channel Length (in.). 162-1/8 162-1 /8 162-1/8 Spacers Material Zr-4 with Zr-4 with Zr-4 with lnconel Springs lnconel Springs lnconel Springs Number per Bundle 7 7 7 3-4

_3.2 MECHANICAL DESIGN BASES

-- - I n*meeting the power generation objectives, tile nu-Clear- fuef-shaff be- useCf as the -initia-1 barrier to the- release -

of fission* products. The fission product retention capability of the nuclear fuel ~hall be substantial during normal modes of reactor operation so that significant amounts of radioactivity are not released from the reactor fuel barrier.

The .nuclear fuel. shall be designed

. to assure (in conjunction with the core nuclear

.. characteristics, the core thermal and hydraulic characteristics, the plant equipment characteristics, and the capability of the nuclear instrumentation and reactor protection system) that fuel damage limits will not be exceeded during either planned operation or abnormal operational transients caused by any single equipment malfunction or single operator error.

3.2.1 Basis for Fuel Damage Analysis Fuel damage is defined as a perforation of the fuel rod cladding which would permit the release of fission products to the reactor coolant.

The mechanisms which could cause fuel damage in reactor operational transients are: 1) rupture of the fuel rod cladding due to strain caused by relative expansion of the U0 2 pellet; and 2) severe overheating of the fuel rod cladding caused by inadequate cooling.

A value of 1% plastic strain of the Zircaloy cladding has traditionally been defined as the limit below which fuel damage due to overstraining of the fuel cladding is not expected to occur. The 1% plastic strain value*is based on General Electric data on the strain capability of irradiated Zircaloy cladding segments from fuel rods operated in several ,~

BWRs. 3 None of the data obtained fall below the 1% plastic strain value; however, a statistical distribution fit to the ~-

available data indicates the 1% plastic strain _value to be approximately the 95% point in the total population. This -.,

distribution implies, therefore, a small (< 5%) probability that some cladding segments may have plastic elongation less  ;~

than 1% at failure.

For design purposes, critical heat flux (the onset of. the transition from nucleate bo_iling to film -boiling) is conservatively defined as a design limit for fuel damage, although fuel damage is not expected to occur until well into the film boiling regime. Severe overheating of the fuel rod cladding is assumed to occur at a condition of minimum critical heat flux ratio (MCHFR-the minimum ratio of the critical heat flux correlation value at the corresponding fluid ,~.

conditions to the actual heat flux at a given point in the fuel assembly) less than 1.0: If MCHFR remains above 1.0 no fuel damage occurs as a result of inadequate cooling. The steady-state MCHF R and the resulting MCHFR during *~

transients are discussed in more detail in Sections 4 and 6.

3.2.2 Effects of Radiation and Fuel Swelling Irradiation affects both fuel and cladding material properties. The effects include an increased cladding strength and a reduced cladding ductility. In addition, irradiation in a thermal reactor environment results in the buildup of both gaseous and solid fission products within the U0 2 fuel pellet which tend to increase the pellet diameter, i.e., fuel irradiation swelling. Pellet internal porosity and pellet-to-cladding gap have been specified in such a way that the thermal expansion and irradiation swelling are accommodated for the worst-case dimensional tolerances throughout life. The irradiation swelling model is based on data reported in References 1 and 2, as well as an evaluation of applicable high exposure data. 3 Observations and calculations based on this refined model for relative U0 2 fuel-cladding expansion indicate that the as-fabricated U0 2 pellet porosity is adequate (without pellet dishing) to accommodate the fission-product-induced U0 2 swelling out to and beyond the peak exposures anticipated for this reload. 3 The primary purpose of the gap between the U0 2 fuel pellet and Zircaloy cladding is to accommodate differential diametral expansion of fuel pellet and cladding and, thus, preclude the occurrence of excessive gross diametral cladding strain. A short time after reactor startup, the fuel cracks radially and redistributes out to the cladding. Experience has shown that this gap volume remains available in the form of radial cracks to accommodate gross diametral fuel expansion.4 3-5

The thermal conductance across the pellet/clad gap, in theory, depends upon the gas conductivity and the distance of the pellet from the cladding when pellet and clad are not in contact, and upon the pressure of the fuel on the cladding if they are in contact. Initially, the gap is filled with helium. As the fuel accumulates exposure, a number of phenomena which can influence the pellet-clad thermal conductance can become important. Fission gases are released from the fuel and dilute the helium gas to form a mixture of He, Kr, Xe and U0 2 impurity volatiles with l.ower thermal conductivity than pure helium in the free volume within the fuel rods. In addition, it has been postulated that the phenomenon of fuel densification may tend to cause an increase in the pellet-to-clad gap with an attendant feedback on pellet-clad thermal conductance. The important observation in this regard is that there is a phenomenon which tends to counteract the adverse effects of fission gas dilution and fuel densification. Specifically, it has been observed that for high power BWR fuel rods, the fuel pellet-to-clad gap closes progressively with exposure in spite of any effect of densification on pellet diameter, with the result that the pellets and clad achieve intimate contact with increasing exposure, thus reducing the importance of the gas conductivity to good thermal conductance. 4 This qualitative discussion serves merely to describe the phenomena influencing pellet-clad thermal conductance with increasing exposure. In the integral models employed in the detailed mechanical design analysis of BWR fuel, the value of pellet-clad thermal conductance is held constant for convenience. The constant value employed is 1000 Btu/h-ft2 -°F. The use of this constant value has been found to be a conservative assumption when applied in conjunction with the integral fuel design models employed by General Electric. Specifically, the design fission gas

• release model employed in the determination of fuel rod plenum size and cladding wall thickness has been shown to*

overpredict available data on fission gas release when applied with a pellet-clad thermal conductance value of 1000 Btu/h-ft2 -° F. Similarly, the design model for relative fuel-cladding expansion (pellet-to-cladding interaction) also has been shown to be very conservative relative to available data when a value of 1000 Btu/h-ft2 -°F is used for pellet-cladding thermal conductance. The basis for these integral fuel design models is described in more detail in Reference 3.

Fission-product buildup also tends to cause a slight reduction in fuel melting temperature. The melting point of U0 2 is considered to reduce with irradiation at the rate of 32(°C)/10,000 (MWd/Te).

In the. temperature range of interest (> 500° C) the fuel thermal conductivity is not considered to be significantly affected by irradiation.

A small fraction of the gaseous fission products (approximately 20%) are released from the fuel pellets to produce an increase in fuel rod internal gas pressure. In general, such irradiation effects on fuel performance have been characterized by available data. and are considered in determining the design features and performance. Thus, the irradiation effects on fuel performance are inherently considered when determining whether or not the stress intensity limits and temperature limits are satisfied.

3.2.3 Maximum Allowable Stresses The strength theory, terminology, and stress categories presented in the ASME Boiler and Pressure Vessel Code, Section 111, are used as a guide in the mechanical design and stress analysis of the reactor fuel-rods. The mechanical design is based on the maximum shear stress theory for combined stresses. The equivalent stress intensities used are defined as the difference between the most positive and least positive principal stresses in a triaxial field. Thus, stress intensities are directly comparable to strength values found from tensile tests. Table 3-2 presents a summary of the basic stress intensity limits that are applied for Zircaloy-2 cladding:

Table 3-2 STRESS INTENSITY LIMITS Yield Strength Ultimate Tensile Categories (Sy) Strength (Su)

Primary Membrane Stress 2/3 1/2 Primary Membrane Plus Bending Stress Intensity 1 1/2 to 3/4 Primary Plus Secondary Stress Intensity 2 1.0 to 1.5 3-6

In the design of BWR Zircaloy-clad U0 2 pellet fuel, no continuous functional variations of mechanical properties with exposure are employed since the irradiation effects become saturated at very low exposure. At beginning of life, the-cladding mechanical-properties-employed are the unirradiated values. At subsequent times in life, the cladding-mechanical properties employed are the saturated irradiated values. The only exception to this is that unirradiated mechanical properties are employed above the temperatures for which irradiation effects on cladding mechanical properties are assumed to be annealed out. It is significant that the values of cladding yield strength and ultimate tensile strength employed represent the approximate lower bound to data on cladding fabricated by General Electric, i.e.,

approximately two standard deviations below the mean value.

Design analyses have been performed for the 8 x 8 reload fuel which show that the stress limits given in the above tabl!! are not exceeded during continuous operation with linear heat generation rates up to the operating limit of 13.4 kW/ft, nor for short-term transient operation up to 16% above the peak operating limit of 13.4 kW/ft, i.e., 15.6 kW/ft.

Stresses due to external coolant pressure, internal gas pressure, thermal effects, spacer contact, flow-induced vibration, and manufacturing tolerances were considered. Cladding mechanical properties used in stress analyses are based on test data of fuel rod cladding for the applicable temperature.

3.2.4 Capacity for Fission Gas Inventory A plenum is provided at the top of each fuel rod to accommodate the fission gas released from the fuel during operation. The design basis is_ to provide sufficient volume to limit the fuel rod internal pressure so that cladding stresses do not exceed the limits given in Table 3-2 during normal operation and for short-term transients of 16% or less above the peak normal operating conditions.

3.2.5 Maximum Internal Gas Pressure Fuel rod internal pressure is due to the helium which is backfilled at one atmosphere pressure during rod fabrication, the volatile content of the U0 2 , and the fraction of gaseous fission products wh!ch are released from the U0 2 . The most limiting combination of dimensional tolerances is assumed in defining the hot plenum volume used to compute fuel rod internal gas pressure. A quantity of 1.35 x 10- 3 gram moles of fission gas are produced pre MWd of power production. In fuel rod pressure and stress calculations, 4.0% of the fission gas produced is calcu.Lated to be released from any U0 2 volume at a temperature less than 3000°F and 100% from any U0 2 above 3000°F. The above basis has been demonstrated by experiment to be conservative over the complete range of design temperature and exposure conditions. The calculated maximum fission gas release fraction in the highest design power density rod is

<20%. This calculation is conservative because it assumes the most limiting peaking factors applied to this rod. The percentage of total fuel rod radioactivity released to the rod plenum is less than 20% because of ra~ioactive decay during diffusion from the U0 2

• 3.2.6 Internal Pressure and Cladding Stresses During Normal Conditions The maximum internal pressure is applied coincident with the minimum applicable coolant pressure to compute the resulting cladding stresses which, combined with cladding.stresses from other sources, must satisfy the stress limits described in Table 3-2. The maximum internal pressure generally does not exceed 1800 psia.

3.2.7 Cycling and Fatigue Limits The design basis for fuel fatigue limits consists of the linear cumulative damage rule (Miner's hypothesis) 5 and the Zircaloy fatigue design basis of Reference 6. The fatigue life analysis is based on t_he estimated number of temperature, pressure, and power cycles. During fuel life, less than 5% of the allowable fatigue life is consumed.

Cyclic Condition Estimated Cycles Room temperature to 100% power -4/yr Hot standby to 100% power -12/yr 50% power to 100% power -60/yr 75% power to 100% power . -250/yr 100% power to 116% power . -1/2 yr 3-7

3.2.8 Deflection The operational fuel rod deflections considered are the deflections due to:

1. Manufacturing tolerances

2. Flow-induced vibration

3. Thermal effects

4. Axial load There are two criteria that limit the magnitude of these deflections. One criterion is that the cladding stress limits must be satisfied; the other is that the fuel rod-to-rod and rod-to-channel clearances must be sufficient to allow free passage of coolant water to all heat transfer surfaces. Thermal hydraulic testing has demonstrated that allowing a statistical minimum clearance of 0.060 inch at two standard deviations away from the nominal clearance is sufficient to assure a very low probability of local rod overheating due to occurrence of critical heat flux.

3.2.9 Flow Induced Fuel Rod Vibrations Flow-induced fuel rod vibrations depend primarily on flow velocity and fuel rod geometry. For the range of flow rates and geometrical variations for the plant, vibrational amplitude does not exceed 0.002 inch. The maximum vibrational amplitude occurs midway between spacers due to the constraint of the spacer. The stress levels resulting from the vibrations are negligibly low and well below the endurance limit of all affected components.

3.2.10 Fretting Corrosion Fretting wear and corrosion have been considered in establishing the fuel mechanical design basis. Individual rods in the fuel assembly are held in position by spacers located at intervals along the length of the fuel rod. Springs are provided in each spacer cell so that the fuel rod is restrained to avoid excessive vibration. Tests of this design have been conducted both out of reactor as well as in reactor prior to application in a complete reactor core basis. All tests and post-irradiation examinations have indicated that fretting corrosion does not occur. Post-irradiation examination of m-any fuel rods indicates only minor fretting wear. Excessive wear at spacer contact points has never been obser'1ed with the current spacer configuration.

3.2.11 Potential for Hydriding The design basis for fuel in regard to the cladding hydriding mechanism is to assure, through a combination of engineering specifications and strict manufacturing controls, that production fuel will not contain excessive quantities of moisture or hydrogenous impurities. An engineering specification limit on moisture content in a loaded fuel rod is defined which is well below the threshold of fuel failure. Procedural controls are utilized in manufacturing to prevent introduction of hydrogenous impurities such as oils, plastics, etc., to the fuel rod. Hot vacuum outgassing (drying) of each loaded fuel rod just prior to final end-plug welding is employed to assure that the level of moisture is well below the specification limit. As a further assurance against possible fuel rod perforation resulting from inadvertent admission of moisture or hydrogenous impurities into a fuel rod, General Electric is now using a zirconium alloy hydrogen getter material in all fuel rods. This getter material has been proven effective by both in-pile and out-of-pile tests.

3.2.12 Dimensional Stability The fuel assembly and fuel components have been designed to assure dimensional stability in-service. The fuel

• cladding and channel specifications include provisions to preclude dimensional changes due to residual stresses. In addition, the fuel assembly has been designed to accommodate dimensional changes that occur in-service due to thermal differential expansion and irradiation effects: for example, the fuel rods are free to expand lengthwise independent of each other, and the channel is free to expand relative to the fuel bundle.

3-8

3.3 RESULTS FROM MECHANICAL DESIGN EVALUATIONS

• -

• 3;3.1 Steady~State Mectianical Performance Reload fuel is. designed to operate at core rated power with sufficient design margin to accommodate reactor operations and satisfy the mechanical design bases discussed in detail in Section 3.2. In order to accomplish this objective, the 8 x 8 reload fuel is designed under the most limiting conditions at 100% of rated power, to operate at a maximum steady-state linear heat generation rate of.;;; 13.4 kW/ft.

Thermal and mechanical analyses have been performed which demonstrate that the mechanical design bases are met for the maximum operating power and exposure combination throughout fuel life.

3.3.2 Fuel Damage Analysis For fresh 8 x 8 reload fuel, the calculated linear heat generation rate (LHGR) corresponding to 1% diametral plastic strain of the cladding is approximately 25.4 kW/ft. Later in life the calculated linear heat generation rate

• corresponding to 1% diametral plastic strain decreases to approximately 23.8 kW/ft at 25,000 MWd/t and approximately 21.1 kW/ft at 39,500 MWd/t. However, due to a depletion of fissionable material, the high exposure fuel has less nuclear capability and will operate at correspondingly lower powers; therefore, a wide margin is maintained throughout life between the operating LHGR and the LHGR calculated to cause 1% cladding diametral strain.

The addition of small amounts of gadolinia to U0 2 results in a reduction in the fuel thermal conductivity and melting temperature.* The result is a reduction in the LHGRs calculated to cause 1% plastic diametral strain for gadolinia-urania fuel rods. However, the gadolinia-urania fuel rods are designed to operate at lower power to compensate for this and provide margins similar to standard U0 2 rods.

.For the 8 x 8 reload fuel design analysis has shown that the power required to produce 1% plastic strain throughout life for all rod types in the assembly is equal to or greater than 180%_ of the maximum steady-state power.

3.3.3 Incipient U0 2 Center Melting For the 8 x 8 reload fuel, incipient center melting is expected to occur in fresh U0 2 fuel rods at a linear heat -_-,'f generation rate of approximately 20.4 kW/ft. This condition corresponds to the integral:

T melt J

32°F kdT = 93 w/cm where k [

6 :~.~~-~T J+ 6.02366 x 10- 12 3 (T + 460) Btu/h-ft-°F, and T is in °F.

The value of the above integral decreases slightly with burnup, as a result of the decrease in fuel melting temperature with increasing exposure.

3.4 FUEL OPERATING AND DEVELOPMENTAL EXPERIENCE 3.4.1 Fuel Operating Experience The peak linear heat generation rate design limit for steady-state operation is 13.4 kW/ft which corresponds to a heat flux of 354,250 Btu/h-ft 2

• This condition is well within the bounds of available production and developmental fuel experience.

3-9

The fuel operating limit and the fuel damage limit have been established based on operating experience and experimental tests covering the complete range of design power and exposure levels. Tables 3-3 and 3-4 present a summary of power reactor production fuel experience. Tables 3-5 and 3-6 show the ranges of development fuel irradiations which have already been completed or are in progress. This experience has been used in establishing design features and in the analysis of performance characteristics. A large volume of experience has been obtained over the past 10 to 15 years with production fuel in commercial power BWRs and numerous developmental irradiations.

The large volume of production experience, starting with the first load of fuel in Dresden 1 Nuclear Power Station in 1960, has provided feedback on the adequacy of the design for, and the effects of, operation in a commercial power reactor environment. Production fuel experience has also provided feedback on the incidence and effect of flaws and impurities which occur statistically in large volume production processes.

The production Zircaloy-clad U0 2 pellet fuel experience is supplemented by a large amount of in-pile and out-of-pile developmental work. The developmental work to date has been employed to test a wide range of design characteristics, to investigate various mechanisms affecting the performance of the fuel rod, and to extend irradiation experience to higher local combinations of fuel rod power and exposure than covered by production fuel.*

More than 25 production fuel types have been designed, manufactured, and operated in more than 19 BWRs.

When all production fuel types are considered, a total of more than 440,000 Zircaloy-2-clad U0 2 fuel rods have been operated in GE-designed BWRs. Out of this number of rods, -180,000 of which went into operation during 1970 and 1971, - 0.2% have been detected to have failed due to wall perforation, arid this includes fuel which failed failed after having exceeded design performance conditions.

Peak linear heat generation rates (LHGR) from approximately 10 to 17 kW/ft have been experienced with the production fuel. Individual fuel assemblies have achieved average exposures greater than 23,500 MWd!Te and have operated more than 9 years in-core residence. In comparison, the 8 x 8 reload fuel has the following proposed operating characteristics:

13.4 kW/ft maximum LHGR (Operating Limits),

45,000 MWd/Te maximum local exposure, and 4-6 years in-core residence time.

Fuel rod diameters in the range of 0.425 to 0.570 inch o.d. with cladding wall thickness from 30 to 40 mils and pellet-to-cladding gaps from 3 to 11 mils have been used in production fuel. Rod-to-rod pitch has varied from 0.533 to 0.874 inch, with rod-to-rod spacing varying from 0.128 to 0.213 inch. Active fuel column lengths have varied from 59.8 to 144.0 inches with fission gas plenum volume per unit of fuel volume from 0.013 to 0.100. Such fuel rods have been licensed and operated in 6 x 6, 7 x 7, 8 x 8, 9 x 9, 11 x 11 and 12 x 12 fuel bundle configurations.

In comparison, the design for this 8 x 8 reload fuel has the following physical characteristics:

Bundle geometry 8x8 Active fuel length 144 in.

Fission gas plenum volume 0.08 x fraction of fuel volume Fuel rod o.d. 0.493 in.

Pellet-to-cladding gap 0.009 in.

Rod pitch 0.640 in.

Rod spacing 0.147 in.

3.4.2 Fuel Developmental Experience The production Zircaloy-clad U0 2 pellet fuel experiences described in the previous section is stipplemented by a large amount of in-pile and out-of-pile developmental work. The developmental work to date has been employed to test a wide range of design characteristics, to investigate various mechanisms affecting the _performance of the fuel rod, and to extend irradiation experience to higher local combinations of fuel rod power and exposure than covered by 3-10

Table 3-3

SUMMARY

OF LEADING EXPERIENCE ON CURRENTLY OPERATING PRODUCTION ZIRCALOY-CLAD U02 PELLET FUEL AS OF OCTOBER 1, 1971 Design Design Fuel Pellet-to- Active Fission Gas Number of Exposure Exposure Time Max Heat Peak Rod Clad Clad Gap Fuel Plenum (Vol Segments or Peak Pellet Avg Assembly lncore Fluxla,b) LHGRla,b) Dia. Thickness (Nominal)* Length Per Unit RDHsStill Reactor (MWd/Te) (MWd/Tel (Years) (Btu/h-ft2) (kW/ft) (in.I (mils) (mils) (in.I Fuel Vol) in Core Dresden I Type 111 B 25,800 16,450 5.45 360000 15.4 0.555 35 7.5 109.0 0.040 3)80 Dresden I Type 111 F 28,200 19.480 4.45 360000 15.5 0.5625 35 10 108.25 0.048 2;592 Dresden I Type V 21,320 13,250 2.45 360000 15.5 0.5625 35 10 108.25 0.048 3.492 Garigliano Type A 26,120 15,180 7.35 252000 10.3 0.534 30 5 105.7 0.031 6,804 Garigliano Type SA *&:SB 15, 160 7,270 2.95 320000 14.6 0.593 34 11 107.0 0.030 7,936 Consumers (BRP) Type 9(e) 35,380 23,430 4.95 434000 15.0 0.449 34 8 70.0 0.048 242 Consumers (BRP) Type e(e) 15,500 8,730 2.75 410000 17.7 0.5625 40 11 70.0 0.048 1,386 Consumers (BRP) Type EG(d,e) 15,559 7,930 2.00 410000 17.7 0.5625 40 11 70.0 0.048 2,079 Consumers (BRP) Type F 0.75 410000 17.7 0.5625 40 11.5 70.0 0.048 1,771 Humboldt Type 11 21,598 13,230 4.00 325000 12.1 0.486 33 10 79.0 0.062 3,724 Humboldt Type 111 14,332 6,615 2.00 389000 16.8 0.563 32 11 79.0 0.062 3,;384 KRB 22,409 14,634 4.45 367000 15.8 0.5625 35 10 130.0 0.058 5,784

'.:: KRB-KD 11, 124 7,050 1.15 367000 15.8 0.563 32 11 130.0 0.058 648 Tarapur I 13,738 8,337 1.80 365000 15.8 0.5625 35 10.5 144.0 0.059 10.?24 Tarapur II 13,407 7,818 2.00 365000 15.8 0.5625 35 10.5 144.0 0.059 10.?24 Oyster Creek I 11,976 88,307 2.35 400000 17.5 0.570 35.5 11 144.0 0.078 27,440 Nine Mile Point 8.412 5,106 2.05 400000 17.5 0.570 35.5 11 144.0 0.o78 26,068 Dresden II 4,825 2,900 2.00 405000 17.5 0.563 32 12 144.0 0.078 34,941 Dresden 11 (reload) 1,600 970 0.50 405000 17.5 0.563 32 12 144.0 0.078 10,535 Dresden Ill 575 290 0.25 405000 17.5 0.563 32 12 144.0 0.078 35,476 Tsuruga 12,037 7,107 1.75 400000 17.5 0.570 35.5 12 144.0 0.078 14,700 Millstone 5,293 3;065 0.85 400000 17.5 0.570 35.5 12 144.0 0.078 28.420 Fukushima-I 4,410 3,300 0.85 400000 17.5 0.570 35.5 12 144.0 0.078 19,600 Monticello 2,945 1,582 0.75 405000 17.5 0.563 32 12 144.0 0.078 23,716 Nuclenor 2,070 1,290 0.60 400000 17.5 0.570 35.5 12 144.0 0.078 19,000 KKM 500 500 0.10 428000 18.5 0.563 32 12 144.0 0.11 11,172 BWR/4(c) 45,000 27,500 5.00 428000 18.5 0.563 37 12 144.0 0.11 8 x 8 Reload(c) 45,000 28,000 5.00 354000 13.4 0.493 34 9 144.0 0.08 a = at rated power b license limit c typical design as opposed to proven performance in preceding entries d includes 14 assemblies with 2 rods per bundle of plutonium e =. values as of February 11, 1971

• ~J* ...... ~ *. ~ .. ,. .:;,. . "'"

Table 3-4

SUMMARY

OF PRODUCTION FUEL EXPERIENCE ZIRCALOY-CLAD U02 PELLET FUEL AS OF OCTOBER 1, 1971 Number of Design or Average Maximum Years of Weight Number Fuel Rods Warranted Assembly Assembly Operation of Fuel of Fuel .or Segments Exposure Exposure Exposure In Identification (lb UI Assemblies (S=Segmentsl (MWd/Te) (MWd/Tel (MWD/Te) Reactor Dresden 1 Type I 132.400 534 77,184(SI 7.400 9,100 23,100 1960-1969 Type 111-B 43,500 192 6,912 14,900 16,750 20,650 1964-1971 Type 111-F 21.400 104 3,744 16,500 17,900 25,950 1965--1971 Type V 24,800 106 3,816 16,500 13,050 19,600 1967-1971 f.IWE-KAHL 14,100 100 . 7,200(S) 8,800 11,000 21,000 1960-?

Garigliano Type A 111, 100 229 16,848 12, 100 13,600 19,750 1963-1971 Type SA 29,380 66 4,224 19,300 15,200 17,900 1968-1971 Type SB 28,770 64 4,096 2,900 5,300 1970-1971 JDPR 9,800 76 5.472(SI 8,800 3,800 N/A 1963-?

Humboldt Type II 28,600 169 8,281 15.400 13,600 17,250 1965-1970 Type 111 23,702 140 5,040 20,700 4,950 13, 150 1971-Consumers Type B 8,700 30 3,630 16,500 19,800 24,600 1966-1971 Type E 12,640 42 3,234 16,500 8,700 10,800 1968-1971 Type EG 11,617 38 2,926 16,500 7,700 10,800 1969-1971 Type F 6,977 23 1,771 22,000 1971-KRB Type A 104,200 371 13,248 16,500 15,300 18,900 1966-1971 Type KO 5,130 1~ 648 16,500 7,050 8.450 1970-1971 Tarapur 1 185,120 184 13,916 16,500 8,300 10.450 1969-1971 Tarapur 2 185,120 284 13,916 16,500 7,800 9,900 1969-1971 Oyster Creek 242,900 560 27.440 16,500 5,100 6,250 1969-1971 Nine Mile Pt. 230,760 532 26,068 16,500 . 5,100 6,250 1969-1971 Tsuruga 136,200 314 15,092 16,500 7,100 9,300 1969-1971 Dresden 2 314,050 753 35.476 20,900 2,900 3.450 1970-1971 Dresden 2 Reload 87.420 215 10,535 20,900 1,000 1,300 1971-Fukushima 1 172.484 400 19,600 20,900 3,300 3.400 1971-Monticello 206,873 4 484 23.716 20,900 3,100 3,700 1971-Millstone 250,625 580 28.420 20,900 3,100 3,700 1971-Nuclerfor 172,818 400 19,600 20,900 1,300 1,550 1971-Dresden 3 314,050 724 35.476 20,900 300 350 1971-KKM 97,017 228 11,172 20,900 <500 <500 1971-3-12

Table 3-5 GENERAL ELECTRIC DEVELOPMENTAL IRRADIATIONS ZIRCALOY-CLAD 95% TD U02 PELLET FUEL RODS No. Fuel Rod Clad Wall Pellet-to Peak Heat Peak Peak of Dia. Thickness Clad Gap Flux LHGR Exposure Name Reactor Rods (in.) (in.) (mils) (Btu/h-ft 2 ) (kW/ft) (MWd/Te) Status Dresden Prototype VBWR 9 0.565 0.030 3.0-16.0 460,000 19.94 12,000 Completed Fuel Cycle (R & D)a VBWR 144 0.424 0.022 2.0-8.0 509,000 16.6 13,800 Compl,eted Dresden Prototypes VBWR 52 0.565 0.028 5.0-8.0 407,000 17.64 10,000 Completed High Performance GETR 12 0.565 0.080 4.0-6.0 630,000 27.0 1,500 Compl'etedh U02 b 1, 126,000 49.0 High Performance GETR 2 0.565 0.030 4.0-11.0 1,355,000 58.0 14,000 Completede "fl w U0 2 b SA-lC Dresden 1 98 0.424 0.022 4.0-8.0 400,000 13.0 40,000 Completed D-1,2,3d Consumers 363 0.424 0.030 7.0 434,000 14.2 30,000 Compl~ted D-50f Consumers 36 0.570 0.035 12.0 507,000 22.0 15,400 g,i D-52,53 Consumers 58 0.700 0.040 13.0 525,000 27.0 4,600 GE-Halden Halden 21 0.563 0.032-0.060 7.0-14.0 510,000 22.0 6,300 Continuing a USAEC Contract A T(04-3) - 189 Project Agreement 11 b USAEC Contract AT(04-3) - 189 Project Agreement 17 c USAEC Contract AT(Q4.J)*- 189 Project Agreement 41 d USAEC Contract AT(04*3) - 361 e Hollow Pellet f USAEC Contract AT(04-3) - 189 Project Agreement 50 g Eight fuel rods failed during second operating cycle due to abnormal i;rud and scale deposition h One rod failure@ 49 kW/ft Fuel assemblies presently out of reactor pending approval for reinsertion

Table 3-6 GENERAL ELECTRIC DEVELOPMENTAL IRRADIATIONS ZIRCALOY-CLAD 95% TD U0 2 PELLET CAPSULES GENERAL ELECTRIC TEST REACTOR Number Fuel Rod Clad Wall Pellet-to- Peak Heat Peak Peak of Dia. Thickness Clad Gap Flux LHGR Exposure Capsule Rods (in.) (in.) (mils) (Btu/h-ft2 ) (kW/ft) (MWd/Te) Status A 3 0.425 0.024-0.032 1.4-10.2 750,000 24.5 88,000 Complete 0.488 0.032 11.2 785,000 29.4 34,000 Complete B 6 0.489 0.034 7.8-11.6 504,000 18.9 65,000 Complete c 5 0.557 0.036 2.0-15.0 475,000 20.3 59,000 Complete D 5 0.557 0.036 2.0-14.0 540,000 23.0 36,500 Complete E 5 0.250 0.015 6.5 735,000 14.1 100,000 Complete F 3 0.443 0.030 3.0-13.0 480,000 16.3 29,000 Complete production fuel. The following presents a discussion of the pertinent developmental fuel experience which, in combination with the production fuel experience, provides the basis for the current BWR fuel design and operating limits.

Tables 3-5 and 3-6 present a summary of design details and performance conditions for Zircaloy-clad U0 2 pellet fuel rods and capsules* irradiated under General Electric or USAEC-General Electric development test programs. These data complement the BWR production fuel experience by providing additional data at higher local combinations of fuel rod power and exposure. Overall, more than 800 fuel pins with design characteristics similar to the current BWR fuel have been irradiated under General Electric or USAEC-General Electric programs. The irradiations have been performed with BWR environment in both test reactors and in commercial power BWRs. Test reactors employed in General Electric developmental irradiations summarized in Tables 3-5 and 3-6 are the Vallecitos Boiling Water Reactor (VBWR),

the General Electric Test Reactor (GETR) and more recently the Halden Reactor. Developmental fuel irradiations have also been performed in the Consumers Big Rock Point and Dresden Unit 1 commercial power BWRs.

The range of peak performance conditions covered by the various development irradiations goes beyond the design performance conditions for fuel in this class of reactor. The development performance conditions include:

13.0-58.0 kW/ft maximum LHGR, and 1500-100,000 MWd/Te maximum local exposure.

The corresponding operating conditions for this reload fuel are:

13.4kW/ft maximum LHGR, and

"'.'45,000 MWd/Te maximum local exposure.

The range of design characteristics and dimensions covered by the various developmental irradiations also encompasses the characteristics and dimensions employed in the current BWR fuel design. The range of design characteristics and dimensions covered by the various developmental irradiations include the following:

• A capsule, as used herein, refers to a test fuel rod, or group of rods combined with all features similar to production fuel rods except for having reduced active fuel l_ength (as low as approximately 3 in.).

• 3-14

Fuel rod o.d. - 0.250 to 0.700 in.,

Clad wall thickness - 0.025 to 0.060 in.,

Pellet-clad gap~ 0.0014 to 0.016 in., ana Pellet length - 0.3 to 0.95 in:

The corresponding fuel design characteristics for this reload fuel are:

Fuel rod o.d. - 0.493 in.,

Clad wall thickness - 0.034 in.,

Pellet-clad gap - 0.009 in., and Pellet length - 0.420 in.

Considering the range of power levels and peak fuel burnups attained in the broad base of operating and developmental fuel experience, it has been concluded that the current 8 x 8 fuel design is a conservative application of this experience, A more complete review of GE BWR fuel experience is provided in Reference 3.

3.4.3 Fuel Damage Experience Although General Electric Zircaloy-clad U0 2 fuel has experienced some failures (~ 0.2% out of more than 440,000 fuel rods), fuel has been successfully operated at Dresden Unit 1 and elsewhere with perforated cladding.

Dresden Unit 1 has operated with some failures in the Type I Zircaloy-clad U0 2 fuel, in the Type 11 stainless-steel-clad U0 2 fuel, and more recently in fuel Types 111-B, I V-F, and V. The Humboldt Bay and Big Rock Point reactors have also operated with failures in stainless steel fuel (Humboldt Type I and Big Rock Type A). The Big ROck reactor has operated with some fuel failures in both the Type B and Type E Zircaloy-clad U0 2 fuel designs as well as a number of failed high power (22 to 27 kW/ft) fuel rods in the center-melt developmental fuel assemblies~ KRB, Dresden 2, Tsuruga, and Fukushima have operated with some perforated fuel rods during initial operation. Failures have resulted

~--

from manufacturing defects, incompatibility of stainless steel as cladding material in the BWR core steam-water environment, inadequate volume for accommodation of fuel expansion and/or fission gas pressure for fuel operated beyond design exposures, cladding overtemperature caused by excessive deposits of crud on fuel rod surfaces resulting from materials in the feedwater system, fretting wear caused by foreign debris trapped in fuel rod spacers, local internal hydriding of the cladding, and local clad strains due to pellet/cladding interaction. In essentially all cases, the mechanisms causing the fuel rods to fail in service have been carefully identified. Appropriate corrections have been made to the manufacturing process and to the fuel or system design and operation to reduce the probability of future recurrence of such failures.

Operation with failed fuel rods has shown that the fission product release rate from defective fuel rods can be controlled by regulating power level. The rate of increase in released activity apparently associated with progressive deterioration of failed rods has been deduced from chronological plots of the offgas activity measurements in operating plants. These data indicate that the activity release level can be lowered by lowering the local power density in the vicinity of the fuel rod failure. These measured data also indicate that sudden or catastrophic failure of the fuel assembly does not occur with continued operation and that the presence of a failed rod in a fuel assembly does not result in propagation of failure to neighboring rods. Shutdown can be scheduled, as required, for repairing or replacing fuel assemblies that have large defects.

Evaluating the fission product release rate for failed fuel rods shows a wide variation in the activity release levels.

Designers have attempted to relate the release. rates to defect type, size, and specific power level. These data support the qualitative observations that fission product release rates are functions of power density and that progressive deterioration is a function of time.

A more detailed summary of General Electric experience with BWR Zircaloy-clad U0 2 pellet fuel, including recent production and development data, has been documented (see Reference 3).

3-15

3.4.4 Fuel Densification The amount of in-pile fuel densification in BWR Zircaloy-clad U0 2 pellet fuel has been observed to be small and is not considered to have any significant effects on fuel performance. Detailed consideration of the occurrence and potential effects of in-pile fuel densification in General Electric BWR s is reported in Reference 4. The AEC staff has recently issued a model for analysis of densification effects in BWRs. This model is considered by General Electric to be overly conservative in light of observations on BWR fuel. A separate submittal will be provided to present the results of analysis employing the AEC staff model.

REFERENCES-SECTION 3

1. WAPD-TM,263, Effects of High Burnup on Zircaloy-Clad, Bulk U0 2 Plate Fuel Element Samples, September 1962.

2. WAPD-TM-629, Irradiation Behavior of Zircaloy-Clad Fuel Rods Containing Dished End U0 2 Pellets, July 1967.

3. Williamson, H. H., and Ditmore, D. C.. Experience with BWR Fuel Through September 1971, May 1972 (NED0-10505).

4. Ditmore, D. C., and Elkins, R. B., Densification Considerations in BWR Fuel Design and Performance, December 1972 (NEDM-10735).

5. Miner, M. A., "Cumulative Damage in Fatigue," Journal of Applied Mechanics, 12, Transactions of the ASME, 67, 1945.

6. O'Donnel, W. J., and Langer, B. F., "Fatigue Design Basis for Zircaloy Components," Nuclear Science and Engineering, 20, 1964.

3-16

4. THERMAL-HYDRAULIC CHARACTERISTICS 4.1_ 'FUEL ASSEMBLY HYDRAULIC ANALYSIS 4.1.1 Core Pressure Drop, Hydraulic Loads, and Correlations The flow distribution to the fuel assemblies is calculated on the assumption that.the pressure drop across all fuel assemblies is the same. This assumption has been confirmed by measurements of the flow distribution in modern boiling water reactor as reported in References 1 and 2. The components of bundle pressure drop considered are friction, local, elevation, and acceleration. Pressure drop measurements made in operating reactors confirm that the total measured core pressure drop and calculated core pressure drop are in good agreement. There is reasonable assurance, therefore, that the calculated flow distribution throughout the core is in close agreement with. the actual flow distribution of an operating reactor.

4.1.1.1 Friction Pressure Drop Friction pressure drop is calculated using the model relation w2 fl 2gp DHA~h </>TPF :

where APf friction pressure drop, psi, w mass flow rate g acceleration of gravity, p water density, DH channel hydraulic diameter, Ach channel flow area, L length, f friction factor, and

<l>TPF two phase friction multiplier.

This basic model is similar to that used throughout the nuclear power industry. The formation for the two-phase multiplier is based on data which compare closely to those found in the open literature.3 General Electric Company has taken significant amounts of friction pressure drop data in multirod geometries representative of modern BWR plant fuel bundles and correlated both the friction factor and two-phase multipliers on a best-fit basis using the above pressure drop formulation. Checks against more recent data are being made on a continuing basis to ensure that the best models are used over the full range of interest to boiling water reactors.

4.1.1.2 Local Pressure Drop The local pressure drop is defined as the irreversible pressure loss associated with an area change such as the orifice, tie plates, and spacers of a fuel assembly.

The general local pressure drop model is similar to the friction pressure drop and is w2 K 2gp A_2" ~PL:

where local pressure drop, psi, local pressure drop loss coefficient, reference area for local loss coefficient, two-phase local multiplier, 4-1

and w, g, and p are defined the Same as for friction. This basic model is similar to that used throughout the nuclear power industry. The formulation for the two-phase multiplier is similar to that reported in the open literature4 with the addition of empirical constants to adjust the results to fit data taken at General Electric Company for the specific designs of the BWR fuel assembly. Tests are performed in single-phase water to calibrate the orifice and lower tie plate, and in both single- and two-phase flow to arrive at best-fit design values for spacer and upper tie plate pressure drop.

The range of test variables is specified to include the range of interest to boiling water reactors. Full scale 8 x 8 tests have been performed to determine the local loss coefficients for upper and lower tie plates and fuel rod spacers. These loss coefficients are in turn used in hydraulic analyses of the core for determination of local pressure losses.

4. 1.1.3 Elevation Pressure Drop The elevation pressure drop is based on the well-known relation where elevation pressure drop, psi, length, average water density, a = void fraction, and pf,pg = saturated water and vapor density, resp.

The void fraction correlation is similar to models used throughout the nuclear power industry and includes effects of pressure, flow direction, mass velocity, quality, and subcooled boiling. Checks against new data are made on a continuing basis to ensure that the best models are used over the full range of interest to boiling water reactors.

4.1.1.4 Acceleration Pressure Drop The p~essure drop component due to acceleration includes the pressure change experienced by the fluid at an area change and the pressure change resulting from density change, such as that which occurs in steam formation. The formulation for the acceleration pressure drop is as follows:

Acceleration Pressure Change due to Flow Area Change:

2 (1 - a2 ) w

• a A~

=----.

APACC 2gp A;' A1 where APACC acceleration pressure drop, A 2 = final flow area, A1 initial flow area, and other terms are as previously defined.

Acceleration Pressure Change due to Density Change:

APACC g A~h 4-2

where 1 x2 (1-x) 2

-**- = - - + - - -

PM Pg a Pf(l--0'.)

PM momentum density, x = steam quality and other terms are as previously defined. The total acceleration pressure drop in boiling water reactors is on the order of less than 5 percent of the total pressure drop.

4.2 FUEL ASSEMBLY THERMAL-HYDRAULIC EVALUATION 4.2.1 Critical Heat Flux and Minimum Critical Heat Flux Ratio The critical heat flux (CHF) condition (the onset of the transition from nucleate boiling to film boiling) is one of the important design considerations in boiling water reactors. It occurs whenever excessive heat is being transferred to boiling or evaporating water and is usually accompanied by a rapid deterioration of the heat transfer process. The critical heat flux is a function of the local steam quality, mass flow rate, pressure, and flow area geometry.

Analyses of CHF are based on the concept of the minimum critical heat flux ratio (MCHFR). The steam quality distribution, calculated by means of energy balances between the fuel and coolant, is used with the CHF *correlation 5 to "

.:,.t*

calculate the spatial distribution of CHF values. Dividing these values by actual design reactor heat ffuxes yields the design MCHFR. >>.'

4.2.2 Steady-State Thermal-Hydraulic Licensing Criteria For purposes of maintaining adequate thermal margin during normal steady-state operation, the previous *,

established license limits of MCHFR ;;. 1.9 and MCHFR.;; 17.5 kW/ft were applied to the 7 X 7 initial core and reload fuel. For the 8 X 8 reload fuel, the limits of MCMFR ;;. 1.9 and MLHGR .;; 13.4 kW/ft were employed. Results from safety analysis using these steady state operating limits as initial conditions are presented in Section 6. Results of full scale 8 X 8 CHF testing will be made available to the AEC upon completion of this ongoing test program 6 .

4.3 RESULTS OF THERMAL-HYDRAULIC ANALYSIS Analyses were performed for a variety of core loadings to fully assess the effect of the 8 X 8 reload assembly on core thermal-hydraulic characteristics. The six core configurations considered are described as follows:

1. Core loaded with 7 X 7 fuel (representative of initial core or reload 1 core loading)

2. Core loaded with 7 X 7 fuel and a single 8 X 8 reload fuel assembly.

3. One-quarter of the core loaded with 8 X 8 reload assemblies and the remainder loaded with 7 X 7 fuel.

4. One-half of the core loaded with 8 X 8 reload assemblies and the remainder loaded with 7 X 7 fuel.

5. Full core loaded with 8 X 8 reload assemblies.

The thermal-hydraulic analyses were performed for the following reactor conditions:

Reactor Power: 2527 MWt Reactor Pressure: 1035 psia (steam dome)

Recirculation flow rate: 98.0 X 106 lb/hr Inlet enthalpy: 522.7 Btu/lb Bypass flow: 10% of total core flow 4-3

The same design basis power distribution as was previously employed, was used in the analysis. The power peaking factors are as follows:

Power Peaking Factor 7X7 8X8 Radial 1.47 1.47 Axial 1.57 1.57 Local 1.30 1.22 Table 4-1 presents a tabulation of significant thermal-hydraulic characteristics calculated for the identified cases.

The results show that, irrespective of the number of 8 X 8 fuel assemblies loaded in the core, both the 7 X 7 and 8 X 8 fuel assemblies receive adequate coolant flow. The margin to CHF for the limiting assembly in an 8 X 8 core; or in a mixed 7 X 7-8 X 8 core, is always equal to or greater than the margin to CHF for the limiting assembly in a 7 X 7 core.

Furthermore, due to the increased heat transfer area and correspondingly lower operating heat flux of the 8 X 8 assembly relative to the 7 X 7 assembly, the 8 X 8 fuel has greater margin to CHF than does the 7 X 7 fuel.

Table 4-1 RESULTS OF THERMAL-HYDRAULIC ANALYSES Case Number 2 3 4 5 Core Average Void Fraction,% 27.6 27.6 27.5 27.5 27.4 Core Pressure Drop, psi 18.3 18.3 18.6 18.9 19.5 Water Rod Flow,% of Total Core Flow NA 0.0006 0.08 0.16 0.34 Assembly Type 7X7 7X7 8X8 7X7 8X8 7X7 8X8 8X8 Number 724 723 543 181 362 362 724 Hot Channel Coolant Flow, 103 lb/hr 118 118 110 119 111 121 113 116 Hot Channel MCHFR 2.05 2.05 2.33 2.07 2.35 2.09 2.38 2.44 Case

Description:

Case 1: Full core loading (724 assemblies) of 7 X 7 fuel Case 2: Same as case 1 with one 7 X 7 assembly replaced with an 8 X 8 assembly.

Case 3: One-quarter core load of 8 X 8 reload assemblies with the remainder 7 X 7 assemblies.

Case 4: One-half core load of 8 X 8 reload assemblies with the remainder 7 X 7 assemblies.

Case 5: Full core loading of 8 X 8 reload fuel.

REFERENCES-SECTION 4

1. Core Flow Distribution in a Modern Boiling Water Reactor as Measured in Monticello, Licensing Topical Report, January 1971 (NED0-10299).

2. Kim, H. T., and Smith, H. S., Core Flow Distribution in a General Electric Boiling Water Reactor as Measured in Quad Cities Unit 1, Licensing Topical Report, December 1972 (NED0-10722).

3. Martinelli, R. C., and Nelson, D. E., Prediction of Pressure Drops during Forced Convection Boiling of Water, ASME Trans., 70, pp. 695-702, 1948.

4. Baroozy, C. J., A Systematic Correlation for Two-Phase Pressure Drop, Heat Transfer Conference (Los Angeles),

AICHE, Preprint No. 37, 1966.

5. Dresden Nuclear Power Station Unit 3 Safety Analysis Report, Docket No. 50-249.

6. Hinds, J. A. (General Electric Co.) letter to J.M. Hendrie (USAEC), March 30, 1973.

4-4

5. NUCLEAR CHARACTERISTICS

5.1 INTRODUCTION

The nuclear design of the 8 x 8 reload bundles described in this section has been performed with the same analytical models and design methods used for General Electric 7 x 7 reload cores licensed by the AEC over the past several years. No changes have been made in the analytical models or in the design methods. The 8 x 8 reload bundles will be loaded into the cores that have been closely followed by GE using these same analytical models and design methods. A high degree of confidence can be expressed regarding the verification of GE nuclear models and methods for these plants. In addition, these same models and methods have been routinely used for cores having lattices in the range from 6 x 6 to 11 x 11, and in cores with mixtures of either 6 x 6 and 7 x 7 or 8 x 8 and 9 x 9.

The 8 x 8 fuel being licensed is well within the range of physical parameters of previous General Electric fuel designs, and no decrease in accuracy can be expected because of the change to an 8 x 8 fuel design.

5.2 BUNDLE NUCLEAR DESCRIPTIONS The mechanical description and physical parameters of the 8 x 8 reload fuel have been given in Section 3. This section describes the calculated nuclear parameters of the 8 x 8 reload bundles and makes comparisons to previously licensed 7 x 7 fuel designs.

There are few real "limits" on the bundle design itself. The real limits are generally expressed in terms of core parameters (e.g., shutdown margin or maximum heat flux). The results of analyses involving core nuclear characteristics are discussed in Section 5.5. The intent herein is to describe the nuclear parameters and to show that for 7 x 7 and 8 x 8 bundles of the same average enrichment, the calculated nuclear parameters are either not remarkably different or are different in a manner that would be expected. The choice of some parameter (say hot reactivity) for reload fuel is .3 ..

dependent on the environment in which the reload fuel bundle will be used: that is, the reloa'd bundle requirements would be slightly different for a very early shutdown than for an outage following a period of operation beyond full power exposure capability. Generally, for reload fuel, the enrichments and reactivities of the bundles will be higher than for initial cores. Specifically, a much higher value of reactivity is allowable for the low exposure reload bundle than is allowable for the initial core bundle (at the same low exposure) because the reload fuel bundle is loaded into an environment of highly exposed bundles of generally lower average reactivity.

5.2.1 2.50 wt"lo U-235 8 x 8 Bundle Design 5.2.1.1 Reactivity Figure 5-1 shows the hot average-void reactivity of the 2.50 wt% U-235 bundle versus exposure. On the same graph a 7 x 7 bundle of the same average enrichment is also shown. The gadolinium concentrations are the same weight percent in the same number of rods in both bundles; however, since there is a smaller volume fraction of gadolinium-containing rods in the 8 x 8 bundle, the initial reactivity of the 7 x 7 bundle is lower than the 8 x 8 bundle.

The gadolinium in the 8 x 8 bundle burns out at a slightly faster rate because the rod diameters are smaller than in the 7 x 7 bundle. Table 5-1 presents a comparison of some physical parameters for these two bundles. This 7 x 7 fuel bundle has been either used or licensed for use at Nine Mile Point-Unit 1, Millstone, Nuclenor, Fukushima 1, and Tsuruga in the recent past and has been demonstrated to be completely acceptable for use in BWR reload cores.

Table 5-2 presents a comparison of zero exposure cold reactivities of the 7 x 7 and 8 x 8 bundles. The Ak/k control-blade-strength variation is well within the range seen for a variety of 7 x 7 fuel designs. As an example, the Ak/k control-blade strength for the Nine Mile Point-Unit 1 initial fuel is 0.157 at beginning of life with curtains present.

5.2.1.2 Void Reactivity The variation of reactivity with void is of importance in the stability of the reactor core while at normal power operation. There is no design criteria placed on the void coefficient except that the overall void coefficient be negative at every point in the operating cycle. Overall void coefficients refer to the core response. It will be sufficient here to give results of infinite lattice calculation of reactivity versus in-channel void fraction and to show that the same 5-1

GADOLINIUM CONCENTRATION SAME NUMBER OF RODS; SAME WEIGHT PERCENT IN EACH ROD 1.16 1.14 1.12 1.10 1.08 1.06 NOTE: 7X7 INITIALLY LOWER AS A LARGER VOLUME FRACTION OF THE BUNDLE CONTAINS GADOLINIUM 1.04 1.02 1.oo"'-~~~.i.....~~~'-~~~.i.....~~~.i...~~~"-~~~'--~~~.i...~~~.L..~~~..i...~~--'------~"--~--"""'""

0 2 4 6 8 10 12 14 16 18 20 22 24 EXPOSURE (GWd/t)

Figure 5-1 Hot Average Void Infinite Lattice K 00 versus Exposure

Table 5-1 7 x 7 TO 8 x 8 COMPARISON OF PHYSICAL PARAMETERS 2.50 BUNDLES 7x7 8x8 Pellet Outside Diameter (in.) 0.477 0.416 Rod Outside Diameter (in.) 0;563 0.493 Rod-to-Rod Pitch (in.) .. 0.738 0.640 Water-Fuel Ratio (cold) 2.53 2.60 U Bundle Weight (pounds) 412,8 404.6 Cladding Thickness (mils) 37 34 Table 5-2 COLD REACTIVITY COMPARISON-2.50 wt% ENRICHMENT ZERO EXPOSURE Condition Controlled? 7 X*7* 8x8 Cold No 1.129 1.148 Cold Yes 0.960 0.966 Llk/k Control Strength 0.150 0.158 behavior is seen for both 8 x 8 and 7 x 7 fuel. Figure 5-2 compares the void reactivity of the same two bundles described above, and, as can be seen, the variation in reactivity with void is very close for both the controlled and uncontrolled states at zero exposure. Again, note that the value of the 7 x 7 bundle is lower than that of the 8 x 8 because an initially larger volume fraction of the bundle contains gadolinium. Figure 5-3 compares the lattice Llk 00 going from 0.40 void to other voids as a function of exposure. As can be seen, the void reactivity characteristics are very similar.

5.2.1.3 Doppler Reactivity The Doppler coefficient is of prime importance in reactor safety. The Doppler coefficient is a measure of the reactivity change associated with an increase in the absorption-of-resonance-energy neutrons caused by a change in the temperature of the material in question. The Doppler reactivity coefficient provides instantaneous negative reactivity feedback to any rise in fuel temperature, on either a gross or local basis. The magnitude of the Doppler coefficient is inherent in the fuel design and does not vary significantly among BWR reactor designs having low fuel enrichment. For most structural and moderator materials this effect is not significant, but in U-238 and Pu-240 an increase in temperature produces a comparatively large increase in the absorption cross section. The resulting nonfission absorption of neutrons causes a significant loss in reactivity. In BWR fuel, in which approximately 98% of the uranium in the U0 2 is U-238, the Doppler coefficient provides an immediate reactivity response that opposes fuel fission rate changes.

Although the reactivity change caused by the Doppler effect is small compared to other power-related reactivity changes during normal operation, it becomes very important during postulated rapid power excursions in which large fuel temperature changes occur. The most severe power excursions are those associated with rapid removal of control rods. A local Doppler feedback associated with the temperature rise is available for terminating the initial burst.

• Same gadolinium as 8 x 8 5-3

1.18 0.89 1.17 0.88 1.16 0.87 BXS CONTROLLED 1.15 7X7 CONTROLLED 0.86 1.14 0.85 8

!Iii:

0 w

..I

..I I 0

!Iii:

0 1.13 0.84 ...a:z w

..I 0

..I tJ 0 BXS UNCONTROLLED w

...za: tJ

~

0 1.12 0.83 ct tJ ..I z

> ...w z

u.

z 1.11 0.82 1.10 0.81 7X7 UNCONTROLLED 1.09 0.80 1.08 0.79 1.07 0.78 1.06 _ _ _ _ _ _ _ _ _....._ _ _ _ _ _ _ _...__ _ _ _....._ _ _ _ _ _ _ _...__ _ _..... 0.77 0 0.10 0.20 0.30 0.40 0.50 0.60 o. 70 0.80 IN-CHANNEL VOID FRACTION Figure 5-2 2.50 wt% U-235 Bundles, Infinite Lattice K= versus In-Channel Void Fraction 5-4

o.04 r-----------------------------

0.03 0% v 0.02 20% v 0.01 c

g 0

~

<l 0.01

• 7X7 A 8X8 0.02 0.03 70%V 0.04  ::----------"'-----------.l....__________J 0 5 10 15 EXPOSURE (GWd/tl Figure 5-3 AK Void Comparison 7 x 7 versusB x 8 from 0.40 Void to Other Voids 5-5

The Doppler reactivity decrement is derived directly from the lattice calculations which are performed to generate the nuclear constants. The lattice methods currently being employed in the fast and resonance-neutron-energy regions are based on the method of Adler, Hinman and Nordheim' with the inclusion of the intermediate resonance approximation. This provides an adequate calculation of both the spatial and energy self-shielding for the resonance absorbers that explicitly includes temperature, moderator density, and geometry effects. A fine group B-1 slowing-down calculation of the fast and epithermal neutron spectrum provides the proper weighting of the resonance absorption to yield effective resonance integrals or cross sections that accurately represent the BWR environment.

The Doppler decrement is determined by doing the lattice calculations at several fuel temperatures holding all other input parameters constant. This results in a change .in the neutron multiplication factor which is solely due to a change in the fuel temperature, which is the Doppler effect. From these analyses it has been determined that the Doppler defect, .:1koop* can be represented very accurately by the following expression:

'1KooP = COOP(~ -v"Til.

Therefore, the Doppler reactivity decrement increases proportionally with the square root of fuel temperature, T, and COOP is the constant of proportionality. The Doppler reactivity coefficient is derived using the same techniques described above. The following equation is used to calculate the Doppler reactivity coefficient:

1 dk COOP

---=

k dT2 [k 1 +COOP(~ - v'°Till 2.,/T; Figures 5-4 and 5-5 compare the Doppler coefficients for two 7 x 7 fuel designs to the Doppler coefficients for the 2.50 wt% U-235 8 x 8 bundle at 200 MWd/t and at 10,000 MWd/t, respectively.

It should be understood that the data presented in these figures are for an infinite lattice. In a finite reactor system the power distribution, and hence fuel temperature distribution, will vary spatially. This in turn results in a spatial variation in the Doppler feedback with larger Doppler reactivity decrements occurring in the high temperature and thus in high neutron flux regions of the reactor core. Therefore, high Doppler reactivity feedback can occur for relatively low core average power increases since the larger Doppler reactivity decrements will occur in the high flux, or importance weighting, regions of the core. Results of core calculations are reported in Table 5-3.

5.2.1.4 Delayed Neutron Fraction Given in Figure 5-6 is a comparison of the delayed neutron fraction for the 2.50 x 8 bundle and the 2.50 7 x 7 bundle at hot average void conditions. As can be seen, the differences are negligible.

5.2. 1.5 Peaking Factors The calculated maximum local peaking factors at average void for the 7 x 7 and 8 x 8 2.50 wt% U-235 bundles are given in Figure 5-7; as can be seen, the peaking is reasonably similar. Of more importance to the reactor operator is the increased heat transfer area of the 8 x 8 bundle, leading to much lower peak kW/ft, as noted in Section 3.

5.3 ANALYTICAL METHODS The analytical methods and nuclear data used to determine the nuclear characteristics are similar to those used throughout the industry for water-moderated systems. 2 The Lattice Physics Model is used to generate few-group-neutron cross sections for use in calculating lattice reactivities, relative fuel rod powers within assemblies, and averaged few-group cross sections. These cross sections and reactivities are calculated at various void and exposure conditions and are used for calculating two' and three-dimensional reactor power distributions. Local fuei rod powers are calculated for an extensive combination of 5-6

VOID 0%

40%

-0.6 70%

I&. -0.8

~

Ul 0

8

!Iii:

-1.0

- - ...,_ 0 7X7 INITIAL v

---

A 7X7 RELOAD

---

osxs ..\

-1.2

-1.4 _ _ _ _ _ _ _ _ _ _ _ _ _ _ _.....,.i...__ _ _ _ _ _ _...__ _ _ _ _ _ _..__ __

0 1000 2000 3000 4000 TEMPERATURE (OFI Figure 5-4 200 MWd/t Doppler Coefficients Uncontrolled 5.7

VOID

-0.6 40%

70%

-0.8 "6

~

0 x

j: -1.0

~

~

8 3!

J

-=

- - - 0 7X7 INITIAL

-1.2

---!:::. 7X7 RELOAD

---=-0 8X8

-1.4

-1.6 ------------------'---------"--------....!--~

0 1000 2000 3000 4000 TEMPERATURE (OF)

Figure 5-5 10,000 MWd/t Doppler Coefficients Unccntrolled 5-8

Table 5-3 NUCLEAR CHARACTERISTICS OF THE DESIGN REFERENCE CORE Core Effective Multiplication and Control System Worth (0% Voids, 20°C)

Kett BOC Uncontrolled 1..124 Fully Controlled 0.962 Strongest Rod Out 0.988 Increase in Core Reactivity with Exposure Into Cycle Reactivity Coefficients, Range of During Operating Cycle Steam Void Coefficient at 34% Voids; -10.2 x 10- 4 to (1/k)(.::lk/.::lV), 1/% Void, Range of Values - 9.0 x 10- 4 Power Coefficient at 2527 MWt and 527.27 BTU/lb Inlet Enthalpy; -0.0505 to

(.::lk/k)/(.::lP/P), Range of Values -0.0414 Fuel Temperature Coefficient at 650°C; - 1.20 x 10- 5 to (1/k)(.::lk/.::lT), lf F Fuel, Range of Values -1.30x10- 5 parameters including fuel and moderator temperatures, burnup, steam voids, and the presence or absence of adjacent control rods. These few-group calculations are performed over either single-bundle cells or groups of four bundles characteristic of repeating arrays in the loaded reactor core. The fast and resonance-energy cross sections are computed by GAM-type program. 1

• 3

• 4 . The fast energies are treated by multigroup, integral collision probabilities to account for geometrical effects in fast fission. Resonance cross sections are computed using the . intermediate resonance *'"

approximation, and the epithermal spectrum is obtained from a B-1 multigroup solution.3 Account is taken of position ~:

and energy-dependent Dancoff factors. Changes due to concentration self-shielding and spectral effects of isotopic composition are recomputed as a function of fuel exposure. THERMOS-type calculations 5 are used to determine the spatially varying thermal spectrum throughout the fuel bundle. The effects of control blades on the cross sections of adjacent materials are calculated and accounted for. Power and flux distributions, infinite multiplication factors, and material and flux-weighted cross sections are calculated using two-dimensional, fewiJroup diffusion theory on fuel assemblies and arrays of fuel assemblies. Burnup calculations are performed by integrating the secular equations describing the fuel depletion process with spatial neutron flux and energy distributions typical of reactor operating conditions. At selected burnup intervals, the nuclide concentrations are used to recalculate revised cross sections with the lattice model, and these are again recycled through two-dimensional diffusion theory.

A large three-dimensional boiling water reactor simulation code 6 providing for representation and calculation of spatially varying voids, control rods, burnable poisons, and other variables is used to compute power distributions, exposure, and reactor thermal-hydraulic characteristics at the beginning of core life and as burnup progresses. Gadolinia is distributed in a few rods within each fuel assembly for supplementary control. This feature makes it necessary to compute the radial space-time dependence of the Gd-155 and Gd-157 concentrations within the fuel rods.

Experimentally, verification of the calculated reactivity effect as well as the calculated removal rate of the high cross-section isotopes has been accomplished. Observation of the operating control rod pattern during full power operation has shown the removal of the gadolinia control to be well matched to the fissionable isotope removal. The effective rate of depletion can be monitored by observing the operating reactivity status. Thus, any trend toward an unacceptably small shutdown margin caused by faster-than-anticipated absorber removal could be detected and remedial action applied before any unsafe condition could be created. Any tendency toward slower removal rates would affect only cycle length and would be an economic problem unrelated to safety.

5-9

0.007 0.006 0.005 EXPOSURE (GWd/tl Figure 5-6 ~versus Exposure, Average Voids, Uncontrolled, 2.50 wt% Bundles

1.22 1.20 Cl 1.18 z

~

~

w

~

..J

~

u 1.16 0

<t1 ..J x

~ 1.14 1.12 1.10 1.08""-------------~~~....i..--~----------------

0 5

......--~--------~------'--------------.;...----....i..--------------------'

10 15 20 25 EXPOSURE (GWdltl Figure 5-7 Maximum Local Peaking versus Exposure 2.50 wt% U-235 Bundles Average Voids

Operating reactor and critical experiments compared to theoretical data provide the precision necessary for reactor design. 7 *8 *9 The reactivity calculation of these analytical methods is frequently compared to the actual performance of operating reactors. Specific comparisons have been made for the Oyster Creek and Dresden 2 plants.

The results of these comparisons show that the calculated and actual results agree within experimental and manufacturing tolerances. The design methods have been shown to be able to compute local powers to within +/-3%, fuel assembly segment powers to within +/-10%, Pu-U ratios versus exposure to within +/-3%, and core reactivities and cold shutdown margin to within 0.5 .1k.

Experimental tests have also been used to verify the analytical calculations of both reactivity and isotopic composition for lattices in the range from 6 x 6 to 8 x 8. These tests give results nearly identical to the comparisons with the operating plants. The most recent experimental comparison is documented in Reference 9.

5.4 EXPERIENCE WITH GE NUCLEAR MODELS .

The analytical methods described in Section 5.3 have been used by General Electric to design and follow cores having lattices in the range from 6 x 6 to 11 x 11 aside from the normal 7 x 7 reload cores. Of special interest in this regard are the Humboldt Bay and the Garigliano reactors. These cores are operating with mixed lattices and have operated successfully for some time. In the case of Humboldt, the core has operated since July of 1969 with a mixture of 6 x 6 and 7 x 7 reload fuel bundles in the core. This mixed lattice reload core has been licensed by the AEC following General Electric analysis using the same analytical methods described above. Also of note in this regard is the Garigliano reactor which.has operated since October of 1968 with a mixture of 8 x 8 and 9 x 9 fuel bundles. This reload core has been licensed by a regulatory agency comparable to the AEC following General Electric analysis. All nuclear license submittal information supplied by General Electric for the past several years has been developed using these same well proven analytical methods. There has been adequate experimental and operational verification of these methods to lattice designs of other than 7 x 7 fuel. No decrease in accuracy can be expected because of the change to an 8 x 8 fuel design.

5.5 NUCLEAR CHARACTERISTICS OF THE CORE Earlier sections have discussed the infinite lattice steady-state reactivities and reactivity coefficients of the new 8 x 8 reload bundles and have made comparisons to previously used 7 x 7 reload bundles. This section discusses the results of core calculations on shutdown margin (including the liquid poison system) and core average reactivity coefficients.

5.5.1 Core Effective Multiplication, Control System Worth and Reactivity Coefficients A tabulation of the typical nuclear characteristics of the reconstituted core is given in Table 5-3. Since the nuclear characteristics of the Reload 2 fuel bundles are similar to those previously loaded, the temperature and void dependent behavior of the reconstituted core will not differ significantly from the values previously reported.

5.5.2 Reactor Shutdown Margin The reconstituted core fully meets the criteria established for reactor shutdown margin in that it may be maintained subcritical by at least 0.25% .1k in the most reactive condition throughout the subsequent operating cycle

• with the strongest control rod fully withdrawn and all other rods fully inserted.

The core loading scheme assumed the insertion of 60 Reload 2 together with the discharge of 60 initial core fuel bundles. In addition, all of the remaining temporary poison curtains ( 140) will be removed at the refueling outage. A minimum shutdown margin of 0.012 .1k has been calculated for the assumed refueling at cycle 2 core exposure increment of 2400 MWd/T. Due to the relatively large shutdown margin calculated, the cycle 2 exposure increment can be as low as 2200 MWd/T at the refueling outage.

5.5.3 Liquid Poison System The liquid poison system is designed to provide the capability to bring the reactor dbwri from full power (2527 MWt) to a cold xenon free shutdown condition assuming that none of the control rods can be inserted. The system is

designed to maintain 720 ppm boron in the moderator water. This amount of boron and therefore the liquid poison system has been found to be adequate in that the core Keff at 20°C and xenon free has been found to be< 0.97.

5.5.4 Reactivity of the Fuel in Storage Both the new fuel storage rack and the spent fuel storage pool will adequately handle the reload fuel bundles. The basic criteria for the storage of fuel is that in a core configuration the uncontrolled k 00 of the fuel bundle must be

< 1.260 at 65°C. This limit will ensure that fuel bundles in the storage rack will have a Keff of.;; 0.90. The reload fuel bundles have koo of< 1.260 at both zero exposure and at their peak reactivity points.

REFERENCES-SECTION 5

1. Carter, J. L., Jr., Computer Code Abstracts, Computer Code-HRG, Reactor Physics Dept., Technical Activities Quarterly Report, July, August, September 1966, October 15, 1966 (BNWL-340).

2. Chernick, J., "Status of Reactor-Physics Calculations for U.S. Power Reactors," Reactor Technology, 13, 4 (Winter 1970-1971 ).

3. Wilcox, T. P., and Perkins, S. T., AGN-GAM, an IBM 7090 Code to Calculate Spectra and Multigroup Constants, April 1965 (AGN-TM-407).

4. Carter, J. L., HRG3 A Code for Calculating the Slowing Down Spectrum in the P 1 orB 2 Approximation, October 15, 1966 (BNWL-340).

5. Honeck, H. C., THERMOS-A Thermalization Transport Theory Code for Reactor Design, June 1961 (BNL-5826).

6. Crowther, R. L., Petrick, W. P., and Weitzberg, G. A., Three Dimensional BWR Simulation, ANS National Topical Meeting, April 1969.

7. Fuller, E. D., "Physics of Operating Boiling Water Reactors," Nuclear Applications and Technology, 19, November 1969.

8. Aline, P. G., et al., The Physics of Non-Uniform BWR Lattice, BNES International Conference on the Physics Problems in Thermal Reactor Design, June 1967.

9. Contained Burnable Neutron Absorber as Supplementary Control, Quad Cities Units 1 and 2 FSAR, Amendment 9.

5-13/5-14

6. SAFETY ANALYSES 6.1 MODEL APPLICABILITY TO 8 x 8 FUEL This section provides information on the applicability (to the 8 x 8 design) of existing models used for safety analysis. Where changes in fuel design affect model applicability, the capacity of the models to accommodate these changes is discussed.

6.1.1 Control Rod Drop Accident (RDA)

The postulated sequence of events for this accident involves an abnormally high worth rod becoming disconnected from its drive, being stuck in the fully inserted position, the drive being withdrawn and the control rod falling out of the core to the rod drive position. Analysis of this accident is performed at various reactor operating states; the key reactivity feedback mechanism affecting the shut.down of the initial prompt power burst is the Doppler coefficient. Final shutdown is achieved by scramming all but the dropped rod. The methods utilized to evaluate the rod 1 2 1 4 5 drop accident have been updated on a continuing basis to reflect improvements in analytical capability * * * *

• The change from a 7 x 7 to an 8 x 8 fuel lattice has no effect on the excursion model used in the analysis of the RDA or on the reactivity feedback effect due to Doppler which is used in the analysis. The number of fuel pins failed due to the RDA is dependent on the fuel pin (local) power peak 1ing factors in the bundle and final peak fuel enthalpy in the core. The local peaking factors and the peak fuel enthalpy are inherently known for an 8 x 8 lattice, the local peaking factors from the lattice design calculations} and the peak fuel enthalpy from the RDA analysis.

Homogenized bundle cross sections and nuclear constants are calculated using standard lattice design techniques as noted in Section 5. Since the bundle cross sections. which are produced from the lattice calculations and which are used in the RDA excursion model, are homogenized, the RDA excursion model does not recognize the lattice type used to produce the bundle cross sections.

A mixture of 7 x 7 and 8 x 8 fuel bundles in a reloaded core present no analytical problem. The homogenized cross sections and nuclear constants used to represent each fuel bundle in the RDA analysis are calculated using methods which have previously been used for lattice designs from 6 x 6 to 11 x 11 geometry and in cores with mixtures of either 6 x 6 and 7 x 7 or 8 x 8 and 9 x 9 (refer to Section 5). Local power peaking at RDA conditions is exp I icitly calculated.

6.1.2 Loss-of-Coolant Accident ( LOCA)

The Emergency Core Cooling System models which are used for the LOCA analysis for 8 x 8 fuel are essentially those which have been previously used for the 7 x 7 fuel designs. They are described and exemplified in Reference 6.

The specific models as applied to the 8 x 8 fuel design will be discussed in the following paragraphs in their order of presentation in Reference 6 ..

6.1.2.1 Short-Term Thermal-Hydraulic Model The significant parameters used by the short-term thermal-hydraulic model will remain essentially the same in changing fuel designs. The exceptions are:

l. Core pressure drop-the total core pre-transient pressure drop for a full 8 x 8 core is ~1 psi higher than for a full 7 x 7 core. Since maximization of the core pressure drop is conservative, a partial 7 x 7 /partial 8 x 8 core is assumed to be fu.lly 8 x 8.

2. Core heat flux-the core heat. flux versus time is consistent* with 8 x 8 fuel operating LHG R and stored energy as well as 8 x 8 geometry: As in the case of core pressure drop the effect is small. The most significant parameters,. the core thermal power, the maximum steam flow, and the recirculation flow, remain unchanged with this change in fuel design: The changes listed above result in only a small change in core flow and pressure responses.

6-1

6.1.2.2 Long-Term Thermal-Hydraulic Model The only significant change to the long-term thermal-hydraulic model is the change in bundle geometry and therefore a small change in the core total hydraulic diameter of the core. The long-term thermal-hydraulic model has the capacity to model various geometries; therefore, such small changes resulting from the change in fuel design do not represent an "extrapolation" in the model. The important parameters, e.g., core power, steam flow, recirculation flow and basic reactor geometry, remain unchanged.

6.1.2.3 Transient Critical Heat Flux Model The transient critical heat flux model will change only in that the bundle geometry and LHGR will change. Test data taken in the new ATLAS loop with full power 8 x 8 bundles will be available on a timely basis to verify the applicability of the existing model. If modification of the model is required it will be made based on the results of these extensive tests.

6.1.2.4 Core Heatup Model The core heatup model used for 8 x 8 analyses is essentially that described in Reference 6 with the incorporation of the modifications described in Reference 7 and the obvious change to the 8 x 8 bundle geometry and LHGR. The model has been used to. predict the results of a number of ECCS transient tests of a full scale stainless steel clad 8 x 8 heater rod bundle. These tests fully confirm the applicability of the Core Heatup Model as modified for 8 x 8 fuel. Full scale ECCS tests with pressurized Zircaloy heaters will be conducted in September of 1973 for further demonstration of the applicability of the Core Heatup Model.

6.1.2.5 Total LOCA Analysis The total LOCA analysis which includes the four above models will not change in procedure. The only changes in the results will be due to changes in fuel geometry and linear hea.t generation rate, which are handled by the existing models without modification with the possible exceptions noted in 6.1 .2.3. General Electric is presently discussing the applicability of current LOCA models for licensing* 8 x 8 fuel with the AEC, and confirmation is expected shortly.

6.1.3 Transient Analysis and Core Dynamics A complete range of single failure caused events which are abnormal but reasonably expected during the life of the plant were analyzed for 7 x 7 fuel as a part of the original plant licensing. Results from these analyses were included in the FSAR and subsequently reviewed for 7 x 7 reload fuel. The purpose of this section is to demonstrate the applicability of the current analytical models to 8 x 8 fuel and mixed core analysis.

6.1.3.1 Transient Analysis Model Applicability to 8 x 8 Fuel The documentation of transient analysis methods for General Electric BWRs is provided in Reference 8. This document includes not only the equations of the transient model, but also a parameter study and comparison of safety analyses applying the model to plant startup data. The mathematical model described in Reference 8 is applied to both new and reloaded cores. The model as presently constituted 1s a "lumped" thermodynamic model with single bundle representations for average and hot channels. The neutron kinetics representation is a point reactor using the point reactor kinetics equations. This brief model review serves as a basis to point out that the model, which is very generally defined, does not change or lose validity due to a mixture of 7 x 7 and 8 x 8 or a total core of 8 x 8 fuel. Parts of the model lump or average system components for computation purposes. These model parts, such as thermodynamic regions or neutron kinetics, are affected by simple input parameter changes due to fuel changes.

The most affected part of the model is the actual fuel heat transfer model. There are several objectives in transient analysis which affect the fuel model. Briefly these can be broken down as: 1) computation of fuel thermal margins, 2) conservative heat flux computation for the system transients, and 3) computation of average fuel temperature for Doppler. If the system contains a complete load of either 7 x 7 or 8 x 8 fuel, the fuel model input is straightforward because the entire core is represented by the same fuel parameters. In the case of mixed 7 x 7 and 8 x 8 core fuel loading, the average core thermal calculations are not completely characterized by either the 7 x 7 or 8 x 8 6-2

fuel type. The mixed fuel loading can be adapted conservatively to the model however. This is achieved by doing three things for input to the dynamic model: 1) using the fuel type conservative for the fuel thermal margin as the hot channel fuel type in the transient analyses; 2) since mixed load core dynamic p~rformance is the average of 7 x 7 and

__ 8 x 8 fuel, choosing the conservative fuel type for plant transients to yield overall system conservative results in the dynamic analysis; and 3) using void and Doppler coefficient input data which are conservative to the overall core design when coupled with the conservative fuel design of 2) above. The use as outlined above of the dynamic model will allow

~totally conservative dynamic analysis for any fuel loading.

6.1.4 Rod Withdrawal Error (RWE)

The rod withdrawal error reactivity insertion event is normally included in the Transient Analysis portion of

.reload fuel safety analysis submittals. However, since the event is analyzed by methods other than the transient mathematical models referred to in 6.1.3, the model applicability of analysis of this event to 8 x 8 fuel or mixed cores is discussed separately.

Analysis of the rod withdrawal error is performed on the assumption that the maximum worth rod is fully inserted and adjacent rods are withdrawn in a manner which will allow full design reactor power with operating limits attained near the inserted rod. This is an abnormal rod pattern which is not normally employed, but it maximizes the rod worth of the inserted rod for purposes of the conservative analysis. The maximum worth rod is then inadvertently withdrawn until rod block occurs, initially assuming the worst allowable LPRM bypass conditions. The results depend primarily on the capability of the flux monitors to detect the local change in the fuel around the control rod as it is withdrawn and to stop the control rod before damage limit conditions occur.

It should be noted that there are two rod block systems currently in use in GE BWRs. The fir.st is described in Reference 9 and is employed in Oyster Creek Unit 1 and Nine Mile Point Unit 1. All other GE BWRs utilize the rod block system described in Reference 10.

The Oyster Creek 1/Nine Mile Point 1 system uses the APR Ms oh a quadrant basis 9 and the other system uses the . ..:;:.*

• • r LPRM strings surrounding the control rod being withdrawn. In both cases the sensors in the system are reading neutron flux. Also in both cases, analysis of the trans_ient is performed assuming worst case allowable LPRM bypass conditions.

The total analysis of the RWE transient utilizes the three-dimensional couplec;l nuclear-thermal-hydraulic representation of the core as described in Section 5.2 for determination of neutron flux levels at instrumented locations and for determination of fuel assembly flow rate. The responses by instruments to changes in flux levels is independent of the fuel type.

6.2 RESULTS OF SAFETY ANALYSES 6.2. 1 Core Safety Analyses Use of the Hench-Levy correlation to determine the safety limit and to establish margins from the normal operating points to the safety limit was established in previous licensing submittals. The same considerations, margins, and damage limits described in detail before, have been applied in evaluating the reloaded core. The operating limit on LHGR for the reload fuel is lower than previously loaded fuel. A further discussion of these controlling factors in the core safety analyses is presented below.

6.2.1.1 Fuel Damage Limits Fuel damage from perforation of the cladding and a subsequent release of fission products can result from overheating or excessive strain of the cladding. The former is assumed to occur when MCHFR reaches 1.0 based on the Hench-Levy correlation and the latter is assumed to occur when MLHGR reaches 25.4 kW/ft (see Section 3). The mechanical design of the reload fuel is to the same design criteria and bases as the initial core fuel and the same damage limits are applicable.

6-3

6.2.1.2 Operating Limits The reload bundles are designed to operate with the same MCHFR limit as the initial fuel in the same environment and with a lower MLHGR. This is, MCHFRs will be greater than 1.9 and MLHGRs will not be greater than 13.4 kW/ft. The limiting values of MCHFR for the reload fuel during normal operation are the same as for the initial core fuel based on the similar design conditions established for the fuels. The limiting value of M LHGR is lower for the reload fuel since it has 63 rods instead of 49 rods in the initial core fuel with approximately the same bundle power.

6.2.1.3 Operating Margins With the previously given damage limits and design limits for the reload fuel, operating margins between the two limits for the reload fuel are expected to be greater than the previously loaded fuel. However, this is based on the maximum design condition. Actual reload fuel operating conditions of MCHFR and MLHGR are expected to be weli below the design limits as has been the experience for previously loaded fuel. Thus, actual operating margins will continue to be greater than the minimum allowable values used in the analyses discussed below.

6.2.1.4 Abnormal Conditions The minimum allowable operating margins described above are conservatively used in analyses of events such as abnormal operational transients and uncertainties concerning steady-state fuel operating conditions. Since these margins are not reduced with the reload fuel the results of these analyses are not expected to change appreciably with the insertion of reload fuel except where dynamic changes are occurring on the reactor and its characteristics have been changed by the reload fuel in such a way as to significantly affect the transient results. These considerations involve the transient analyses which is covered separately below.

6.2.2 ACCIDENT ANALYSES 6.2.2.1 Main Steam Line Break Accident The analysis of the main steam line break accident depends on the operating thermal-hydraulic parameters of the overall reactor, such as the pressure, and the overall factors affecting the consequences, such as primary coolant activity. Insertion of 8 x 8 reload fuel will not change any of these parameters so the previously reviewed results of this analysis will not change.

6.2.2.2 Refueling Accident The analysis of the refueling accident depends on mechanical damage caused by a fuel bundle falling back onto the top of the core while it is being removed, which will not change with the use of the reload fuel. The consequences depend on the fission product inventory in the fuel and various factors affecting the amount and kind of releases to the atmosphere. The fission product inventory is not expected to increase even with the large number of rods in the 8 x 8 reload fuel bundle. This is a result of lower fission product inventory per rod due to the lower power level of operation per rod. Thus, even if more rods were damaged, the total fission product inventory is not increased, but there will be slight changes in the relative amounts of different constituents because of the slight differences in enrichment and gadolinia concentration. The effects of these small differences will be inconsequential in terms of the releases caused, and undetectable when the various reduction factors are applied to determine offsite consequences. Therefore, the previously reviewed results of this accident analysis will not change.

6.2.2.3 Control Rod Drop Accident 6.2.2.3.1 Identification of Causes. There are many ways of inserting reactivity into a boiling water reactor. However, most of them result in a relatively slow rate of reactivity insertion and therefore pose no threat to the system. It is possible, however, that a rapid removal of a high worth control rod could result in a potentially significant excur,sion.

Therefore, the accident which has been chosen to encompass the consequences of a reactivity excursion is the control Rod Drop Accident (RDA).

6-4

6.2.2.3.2 Starting Conditions and Assumptions. Before the control rod drop accident is possible, the following sequence of events must occur:

1. The complete rupture, breakage. or disconnection of a fully inserted control rod drive from its cruciform control blade at or near the coupling.

2. The sticking of the blade in the fully inserted position as the rod drive is withdrawn (worst case).

3. The falling of the blade after the rod drive is fully withdrawn (worst case).

This unlikely set of circumstances makes possible the rapid removal of a control rod. The dropping of the rod results in a high local k 00 in a small region of the core. For large, loosely coupled cores, this would result in a highly peaked power distribution and subsequent shutdown mechanisms. Significant shifts in the spatial power generation would occur during the course of the excursion. Therefore, the method of analysis must be capable of accounting for any possible effects of the power distribution shifts.

In order to limit the worth of the rod which could be dropped, the rod worth minimized system or a second operator controls the sequence of rod withdrawal. This assures no movement of an out of sequence rod before the 50%

rod density configuration is achieved and limits movement of rods to in-sequence segments beyond the 50% rod density configuration during startups. The 50% rod density configuration occurs during each reactor startup and corresponds to the condition in which 50% of the rods are fully inserted in the core and 50% are fully withdrawn.

6.2.2.3.3 Accident Description. The accident is defined as:

1. The highest worth rod that can be developed at any time in core life under any operating conditions drops from fully inserted position to fully withdrawn position (rod increments only beyond 50% rod density)

2. The rod drops.

3. The scram is that defined in the technical specifications.

The detailed analysis of this accident is discussed in References 1, 2, 3 and 5. A continuing effort is being made

• ~ .

in the area of analytical methods to assure that nuclear excursion calculations reflect the latest "state-Of-the-art." .....

The sequence of events and the approximate times of occurrence are as follows:

Approximate Event Elapsed Time (1) Reactor is at a control rod density pattern corresponding to maximum in-sequence rod worths (2) Rod worth minimizer or operators are functioning to restrict rod withdrawals to in-sequence rods or rod increments. Maximum worth in-sequence control blade becomes decoupled.

(3) Operator selects and withdraws the control rod drive of the decoupled maximum worth in-sequence rod to its fully withdrawn position (rod increments only beyond 50% rod density).

(4) Blade sticks in the fully inserted position.

(5) Blade becomes unstuck and drops at the maximum velocity determined from experimental data (3.11 fps). 0

• 6-5 i

Approximate Event Elapsed Time (6) Reactor gaes prompt critical and initial power burst is terminated by the Doppler Reactivity Feedback. <1 sec (7) APR M 120% power signal scrams reactor.

(8) Scram terminates accident. <5 sec 6.2.2.3.4 Identification of Operator Actions. The termination of this excursion is accomplished by automatic safety features or *inherent shutdown mechanisms~ Therefore, no operator action during the excursion is required.

6.2.2.3.5 Analysis of Effects and Consequences 6.2.2.3.5.1 Methods, Assumptions and Conditions. The methods, assumptions, and conditions for evaluating the excursion aspects of the control rod drop accident are described in detail in References 1, 2, 3, and 5.

Reference (1) is the topical report on rod drop and is applicable to beginning of life conditions for curtained cores. Reference (2) is the first supplement to Reference (1) and is applicable to beginning of life conditions for Gadolinia cores. Reference (3) is the second supplement to Reference ( 1) and is applicable to exposed cores.

Reference (5) is not a supplement to Reference 1), however, the information contained therein is supplemental since it is a direct expansion of the described methods applied to a parametric study of worst cased variables resulting in a boundary approach to rod drop accident evaluation.

The technical bases which is presented in Reference (5) was used to verify that the result of a rod drop excursion in the reloaded core would not exceed the design criteria, as described below.

Although there are many input parameters to the Rod Drop Accident Analysis, the resultant peak fuel enthalpy is most sensitive to three basic conditions. These are: 1) Doppler reactivity feedback, 2) accident reactivity characteristics, and 3) scram reactivity feedback.

If all other parameters remained unchanged, the rod drop excursion for exposed cores would be less severe than for initial cold clean cores under the same set of conditions since the Doppler reactivity feedback will be more negative.

This is due to the fact that Pu-240, which has a large negative Doppler effect, builds up with exposure. Figure 6-1 shows the comparison between the actual Doppler coefficient and the Technical Bases Doppler (Reference 5) coefficient.

The accident reactivity characteristics have varying effects on the rod drop excursion results. These characteristics are accident reactivity shape, total control rod worth, local peaking factor, and the delayed neutron fraction. The total control rod reactivity worth (worth of the dropping rod) has a major effect on the accident results; this will not change substantially with the insertion of reload fuel. The local peaking factor and the delayed neutron fraction were inputs to the evaluation and were in the same range as those shown for actual plant experience) in Reference (4a). A comparison of the actual scram shape with the Technical Bases scram shape is shown in Figures 6-2 and 6-3.

The scram reactivity feedback function has a significant effect on the results of a rod drop excursion. The scram reactivity feedback shape was evaluated and shown to be above that used in establishing the boundary in Reference (4a). A comparison of the scram reactivity function is shown in Figures 6-4 and 6-5.

The evaluation of each of these parameters by comparison to the boundary values presented in Reference (5) shows that the maximum rod worth is not as great as the 1.3% .:1K derived. This verified that the consequences of a rod drop excursion from any in-sequence control rod would be below the 280 cal/gram design limit, since maximum in-sequence rod worths after this reload will be well below the 1.3% .:1K allowable.

6-6

--0.6 0 MWdlt, 0% VOIDS, COLD

-0.8 0 MWdlt, 0% VOIDS, HOT 0 MWdlt, 40% VOIDS, HOT u: -1.0 5 MWdlt, 40% VOIDS, HOT 0

Lt)

...0

)(

15 MWdlt, 40% VOIDS, HOT

"'I

'C 'C

.._ .. -1.2

-I~

zw -1.4 u

u..

u..

w 0

qi u

> -1.6

/

> I TYPICAL OPERATING u

j::

w I PLANT DATA a:

a:

w

-1.8 I Q.

Q.

I 0

0 I

-2.0 I

I

-2.2 I

I 400 800 1200 1600 2000 2400 2800 3200 3600 4000 4400 4800 0

AVERAGEFUELTEMPERATURE(~)

Figure 6-1. Doppler Reactivity Coefficient vs Average Fuel Temperature as a Function of Exposure and Moderator Condition

.016

~-------------- PL=l.2

.014

1. 3

.012

~

c IX

.010 0

31:

c 0

IX

.:...J ACCIDENT REACTIVITY SHAPE 0

IX FUNCTIONS FOR COLD STARTUP

z .008 8

- - - 280 CAL/GM PEAK FUEL ENERGY CJlNTENT

/:J = .0050

.006

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

/

/

/ ____ Dresden-3 Reload-2

.004 I In-sequence Rod I ,A= .0054 I

I I

.002. I I

FIGURE 3 0 2 4 6 8 10 12

• CONTROL ROD POSITION (ft. withdrawn)

Figure 6-2 Accident Reactivity Shape Functions for Cold Startup 6-8

.016

.014 .~--~--~~~~~~~~~-f>L=l.2

1. 3

.012 ACCIDENT REACTIVITY SHAPE FUNCTIONS FOR HOT STARTUP

~- .010

........ - - - 280 CAL/GM PEAK FUEL ENERGY CONTENT

I: ~ = .0050 I-0::

0 3:

0

---

Dresden-3 Reload-2 In-sequence Rod 0

0:: .008 tS = .0050

....J 0

0::

I-z:

0 u

.006

.t

.004

/

/

FIGURE 4

.002 /

/

/

/

/

/

~

/

0 0 2 4 6 8 10 12 CONTfllL ROD POSITION (ft. withdrawn)

Figure 6-3 Accident Reactivity Shape Functions for Hot Startup 6-9

-.08 /

-.07

-.06

~

~

-

-.03 DRESDEN 3 PERFORMANCE 280 CAL/GM PEAK FUEL ENERGY CONTENT

-.02

-

• 01 0

0 1 2 3 4 5 ELAPSED TIME AFTER SCRAM SIGNAL (SEC.)

Figure 6-4 Saam Reactivity Function for Cold Startup 6-10

"'* 16

-

• 14

-

• 12

~

io::

4 -.10 1-I-

LI.I Cl::

-.08 280 CAL/GM PEAK FUEL ENERGY CONTENT

-.06 DRESDEN 3 PERFORMANCE

-.04 "-..

-.02 0

0 1 2 3 4 5 '

ELAPSED TIME AFTER SCRAM SIGNAL (SEC.)

Figure 6-5 Scram Reactivity Function for Hot Startup 6-11

• 6.2.2.3.4.2 Fuel Damage. The fuel damage thresholds are based on both experimental and theoretical data. This information is discussed in Section 5 of Reference 11.

The rod drop accident analysis is sensitive to spatial variations in the core design such as fuel loading patterns, gadolinium distribution, etc. An estimate of the radiological exposures has been made and is based on the failure of all fuel rods above an energy content of 170 cal/gm assuming the maximum enthalpy reaches 280 cal/gm during the accident. This is consistent with the boundary approach established in Reference (5) and results in approximately twice the number of failed fuel rods and released fission products as those originally described in the FSAR. The resulting doses are still well within the 10 CFR Part 100 guidelines.

6.2.2.4 Loss of Coolan~ Accident The following evaluation is based on the 8 x B reload fuel. The results of the 7 x 7 fuel evaluation has not changed and can be found in previous submittals.

6.2.2.4.1 Design Bases. The objective of the emergency core cooling systems (ECCS). in conjunction with the containment, is to limit the release of radioactive materials following a loss-of-coolant accident so that resulting radiation exposures are within the guideline values given in published regulations.

Safety design bases and functional requirements for the emergency core cooling systems are given in the FSAR and have not changed.

6.2.2.4.2 System Design. The ECCS, containing four separate subsystems, is designed to satisfy the following performance objectives:

1. To prevent fuel clad fragmentation for any mechanical failure of the nuclear boiler system up to, and including, a break equivalent to the largest nuclear boiler system pipe.

2. To provide this protection by.at least two independent, automatically actuated cooling systems.

3. To function with or without external (off-site) power sources.

4. To permit testing of all ECCS by acceptable methods including, wherever practical, testing during power plant operations.

The aggregate of these emergency core cooling systems is designed to protect the reactor core against fuel clad damage (fragmentation) across the entire sprectrum of line break accidents.

The operational capability of the various emergency core cooling systems to meet functional requirements, and the performance objectives is as follows.

During the first ten minutes following the initiation of operation of the ECCS, the functional requirements is satisfied for all combinations of single active component failure and single pipe breaks, including pipe breaks in any ECCS sub-system which might partially or completely disable that sub-system.

After the first ten minutes following the operation of the ECCS and in the event of an active or passive failure in the ECCS or its essential support system, long term core and containment cooling is provided by any one LPCI or core spray pump delivering water to the reactor vessel and by one AHR pump supported by one AHR heat exchanger with 100"Ai service water flow.

The description and detailed design information on specific parts of the emergency core cooling system is presented in the FSAR.

6-12

6.2.2.4.3 Performance Evaluation Summary. To achieve reliability, each emergency core cooling subsystem uses the minimum feasible number of components that are required to actuate. All equipment is testable during operation. Two different cooling methods-spraying and flooding-provide diversity.

Evaluation of ECCS controls and instrumentation for reliability and redundancy shows that a failure of any single initiating sensor cannot prevent or falsely start the initiation of these cooling systems. No single control failure can prevent the combined cooling systems from adequately cooling the core. The controls and instrumentation can be calibrated and tested to assure adequate response to conditions representative of accident situations.

The emergency core cooling systems are provided to remove the residual and decay heat from the reactor core so that fuel cladding temperature is kept substantially below 2300° F. The intent of the ECCS temperature criterion is to prevent gross core meltdown and fuel cladding fragmentation. Under extreme conditions highly oxidized Zircaloy is known to fracture on cooling. Based on experimental data, cladding fragmentation on cooldown is prevented (for the time scale of interest here) if the maximum cladding temperature is limited to less. than 2300°F. This is therefore the design temperature criterion for ECCS system performance. The actual performance of the core cooling systems is such that peak temperatures much lower than 2300°F will be maintained throughout the complete break spectrum.

A summary of* peak cladding temperatures calculated to occur for the worst intermediate break and the design-basis break will be found in Table 6*1.

Table 6-1 PEAK CLADDING TEMPERATURES Large Intermediate Intermediate Break Break Break Temperature Temperature Size (oF) (o F) (Ft 2 )

Single Failure Assumed AEC Index of Acceptability* Worst Single Failure 2300 2300 x Case*

1. AEC Assumptions LPCI injection valve:# 1920 x x

2. AEC Assumptions HPCI Failuret x 1800 0.07

• Calculated metal-water reaction is less than 0.2% of cladding for all cases above. AEC acceptability index is 1%.

tFour LPCI pumps, two CS pumps, and ADS remaining.

XDoes not apply.

Evaluation Model. The performance analysis of the ECCS is based upon analytical models used to conservatively predict reactor vessel pressure, liquid inventory, and fuel cladding temperature variations with time after a break. These models are identified, exemplified, and fully explained in Reference 3. There have been no deviations from the evaluation model described in Appendix A, Part II of AEC Interim Policy Statement.

Fuel Clad Effects. Figure 6-6 shows peak cladding temperatures as a function of time for the worst single failure case which leaves 2CS + HPCI and the ADS operable. As shown, the maximum cladding temperature for this break, the most severe design-basis accident, is substantially limited by the emergency core cooling systems.

6-13

HOTSPOT REFLODDEDl 2000 ii:

2.

llC c(

llC 1500 RATED SPRAY l Q.

~

~ ...""

C1

"'" ~

0 0

c(

u 1000

/

• ROD 37

/ *SEE FIG.6-26 FOR ROD 500 IDENTIFICATION 2 5 10 20 50 100 200 500 TIME AFTER ACCIDENT (sec)

Figure 6-6 Cladding Temperature versus Time OBA with Failure of LPCI Injection Valve (2CS +ADS+ HPCIJ AEC Analysis #1 15000 MWd/t Exposure

An example of the integrated system performance is shown in Figure 6-7 for a typical small size break with failure of HPCI. Peak cladding_ temperature for this case is.shown in Figure 6-8.

F igur~ 6-9 is a break area spectrum anal;sis of the peak c.ladding temperatur.e and percent. metal-water *reaction for the wo~st single taih:ires. The single failures are th~ loss of the HPCI or the loss of the LPCI injection valve resulting in ECCS degradation to 4 LPCI + 2CS +ADS and HPCI + 2CS +ADS, respectiv~ly. Adequate cooling is maintained.

ECCS Performance.

. .~

Individual System Performance. The capability of the individual subsystems of the ECCS is shown on the bar chart (Figure 6-10). A whole bar represents the capability of an individualsystem to protect the core without assistance from another subsystem. A half bar represents the range of break.sizes for which a low pressure system must rely upon a high pressure system for additional inventory makeup and/or more rapid vessel depressurization. The ADS provides no inventory makeup and therefore cannot protect the core individually. The bar chart reveals subsystem characteristics but should not be applied to ~CCS performance evaluations. No single failure could be hypothesized that would result in only one subsystem of the ECCS being available.

Integrated Operation of Emergency Core Cooling Systems. Two different methods and at least two independent core cooling systems are provided to limit fuel cladding temperature, over the entire spectrum of postulated reactor primary system breaks, as required by the design bases.

The following discussion is direct~d-.toward the integrated performance of the ECCS; that is, how the ECCS will actually operate to provide tore cooling for the entire spectrum of loss-of-coolant accidents. The discussion is subdivided based on the two types of loss-of-coolant accidents; a break of a liquid line and a break of a steam line.

It is a convenient to classify the breaks acc~rding to the location of the penetration on the reactor vessel. The break types will fall into one of three categories. These, along with the lines that fall into these categories, are as follows:*

1. Steam Type Breaks. T~.ese are breaks in which the reactor vessel penetration is exposed to the steam regions inside the vessel.

a. Steam Lines

b. Some Instrument Lines

2. Steam/Liquid Type Breaks. These are breaks in which the reactor vessel penetration is either exposed to the two-phase regions inside the vessel or to regions which. are exposed to liquid, but are near the water level and would therefore turn into steam breaks very shortly after the break occurred. These are located above the core.

a. Feedwater Lines

b. Core Spray Lines

c. Some Instrument Lines

3. Liquid Type Breaks. These are breaks in which the reactor vessel penetration is well below the vessel water level, and below the top of the core.

a. Recirculation Pump Suction Lines

b. Outer Recirculation Riser Line

c. Drain Line

d. CRb housing

e. lncore housing

f. Jet pump instrument line 6-15

WATER LEVEL INSIDE SHROUD 0.8 40 12, IC w

ca TAF >

!w _,

w a: a:

w

> l-Cf>

O'>

aw ei:

3=:

a:

II.

0.4 20 BAF REACTOR VESSEL PRESSURE o .._____.;....________________..._______________________...,..._______________________..._______________________~o 0 200 400 600 800 TIME (secl Figure 6-7 Performance of ECCS with Failure of HPCI for a Small(.02 ft2) Break.

(4 LPC/ + 2CS +ADS) AEC Assumptions

2000 1500 HOT-SPOT REFLOOOEO .

1 ROD 23 ROD20 ROD40 1000 ROD55

-~*

500

• SEE FIG.6-26 FOR ROD IDENTIFICATION 0 ._______________..;.;;,i,_ _ _ _.;...;...;..._ _ _ _ _ _ _ _. . __ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _. __ _ _ _ _ _ _ _ _ _ _ _ _ _ ~

0 200 400 600 800 TIME (sec)

Figure 6-8 Cladding Temperature versus Time for a Small Break with a Failure of HPCI (0,02 ft2) Break) 4 LPCI + 2CS +ADS AEC Assumptions 6-17

0.6 200C>. 0.5 LPCI INJECTION HPCI FAILURE VALVE FAILURE

~ 1600 0.4

~

w

~

a: z

~ 0 I- j::

c (,)

a:

w c

w a:

~

w 1200 0.3 a:

q> w CX>

I-C),

I-c

! ~

0 0 c c_, I-w

(,)

800 0.2  ::!!:

lit cw

. A.

4.oo* 0.1 o --~~~....i~~~~~~.....~~~~.....~~~~.&;..__."-~~~--'~~~~--'~~~~....1.~~~~~~o 0.01 0.02 O.Q5 0.1 0.2 0.5 1.0. 2.0 5.0 BREAK AREA (tt2)

Figure 6-9 Peak Cladding Temperature Spectrum for a Single Failure Condition with AEC Assumptions.

I LIMIT OF STEAM BREAK SIZE STEAM BREAKS

---

AUTO-RELIEF I ..

HPCI

.I l.

• ----r-~

---

I LPCI

! \

I

~----. I CORE SPRAY I

I

\

I I

LIQUID BREAKS i

AUTO-RELIEF i

~----------- \

I LIMIT OF LIQUID BREAK SIZE I

I

---

HPCI I

-*- -------. LPCI I

II \

---

I CORE SPRAY I

I l

" I ISCALE I ICHANGES I.

I I I I I I 0.0 0.1 0.2 ~-0 2.0 3.0 4.0 5.0- 6.0 2

BREAK AREA (tt 1 Figure6-10 Emergency Core Cooling System versus Break Spectrum

For a given size break, the peak clad temperatures will be higher, the lower the line penetration is located on the vessel; i.e., the peak clad temperature for a given size break will be higher for those lines in liquid type breaks than in steam/liquid type breaks and those in steam/liquid type breaks will be higher than those in steam type breaks. In demonstrating the performance and capability of the ECCS, recirculation line breaks are analyzed since these will result in the highest ECCS peak clad temperatures for a given break size. The rupture and consequences of a main steam line break have also been analyzed since this is the most severe case with regard to containment performance.

For purposes of core performance and cladding integrity the most severe accident and the design basis accident is the loss-of-coolant accident (recirculation line break). By analyzing breaks in the main steam line, the effects of all other steam type breaks are covered. For liquid type breaks, the spectrum analysis performed on the recirculation line breaks, covers the effects of all other type liquid breaks such as the RHR suction and return lines, and recirculation riser lines.

The peak clad temperatures for the steam/liquid type breaks will be less than for the comparable size liquid breaks. This was shown in part in Millstone Unit 1, AEC Docket No. 50-245 Amendment 14, in which the effects of various size feefjwater breaks were analyzed.

Steam Line Breaks. The most severe steam pipe break is one .that occurs inside the drywell, upstream of the flow limiters. Although the isolation valves close within 10.5 seconds (10-second valve action time plus 0.5-second instrument response), such a break permits the pressure .vessel to continue to depressurize. For purposes of analysis, pre-accident conditions assumed are the reactor operating at design power, steam dome at maximum design operating pressure, scram low water level in the pressure vessel, and loss of auxiliary power coincident with the steam pipe break.

The accident sequence starts with an instantaneous, guillotine severance of the steam pipe upstream of the steam flow restrictors. The steam flow accelerates to its limiting critical flow value in the break at the pressure vessel end and at the flow-limit~ end. Steam loss exceeds the generation rate and results in rapid depressurization of the pressure vessel and steam pipes. The first 10 seconds of this accident are similar to the break outside the drywell. However, for the break inside the drywell, closure of the isolation valves reduces the blowdown rate but does not prevent the vessel from depressurizing. The vessel continues to depressurize causing sufficient voids to immediately shut down the reactor.

A mechanical scram is initiated by a position switch in each isolation valve (at approximately 10% closure) so control rod insertion begins with 1.5 seconds after the.break. Low water level or high drywell pressure also initiates' a scram.

Loss of reactor coolant through blowdown from the double-ended break consists of three intervals: first steam blowdown, then mixture blowdown, and finally steam blowdown again. As the reactor vessel depressurizes, flashing causes the water level to rise. When the level reaches the steam pipes, the break floyv changes from a steam blowdown to a steam-water mixture blowdown. Mass flow rate through break increases sharply. At 10.5 seconds the isolation valves are dosed, which reduces the blowdown rate. As coolant is expelled and pressure decreases; the water level outside the shroud drops below the steam pipe elevation and steam blowdown begins again. The long term pressure transient and level elevation transient are shown in Figure 6-11.

Approximately 40 seconds after the break occurs, both core spray systems and the LPCI system start to inject coolant into the vessel. For this analy~is the normal situation where all ECCS pumps are operating is assumed. Analyses of degraded situations in which only a portion of ECCS* operates also show that the core remains covered and cooled throughout the entire blowdown transient, with cladding integrity maintained; Liquid Line Breaks. The double-ended recirculation line break is the design basis accident for the emergency core cooling systems. The reactor is assumed to be operating at design power when a complete circ_umferential rupture instantly occurs in one of the two recirculation system suction lines. Normal a-c power supply to the recirculation pumps is assumed to fail at the time of the accident. Core inlet flow and vessel pressure following the accident are shown in Figure 6: 12.

6-20

1000 WATER LEVEL INSIDE SHROUD 800 40

. 600 - -- -- -- -- -- -- -- -- --. -- -- --**-* -- -- TAF

£

!w * . ...J w

a: >

w

J ...J REACTOR VESSEL PRESSURE

!:!w a:*..

W*

Cf> a: l-

~ e{

N

...J w

!:!w 400 20

> -- -- --. ---------- ---- --------- -- BAF

• 0 o.._-----------~----....i. ~-----------------'---~----~-~~~~-.L..-~~~~...;;,.~~~~~...Jo 50 100 150 200 TIME lsecl Figure 6-11 Performance of ECCS for Main Steam Line Break inside Drywefl with all ECCS Operating - A EC Assumptions

1.0 1000 o:e 800

~

0 I&..

0.6 600

!w w

a: a:

0  :>

u . jg en 0 w t..:J w a:

i'.J

~ FLOW ~ PLENUM A.

w c( a:

E COAST DOWN FLASHING 0 a: 0.4 400 u 0

z 0.2 200 o ....~~~~~~.....~~~~~~.....~~~~__........_~~~~~~.........~~~~~....~~~~~~--~~~~~~--~~~~~....... o 0 2 4 6 8 10 12 14 16 TIME (sec)

Figure 6-12 Core Flow and Pressure Following a Recirculation Line Break

Initially, the rotating energy stored in the pump and motor of the unbroken recirculation system line provides continuing flow into the lower plenum,'maintainirig a relatively high level of-core flow. The flow.ls. assu~*ed to ceaSe when the falling level in the downcomer reaches the jet pump suction level.

When the break flow in the severed recirculation Iine changes to steam, the associated high vessel depressurization rate causes the water in the lower plenum to vigorously and immediately flash to steam. This will force a two-phase flow up through the core and through the jet pump diffusers. As the lower plenum inventory is depleted, the mass flow rate into the core diminishes.

Calculations indicate that the reactor vessel depressurizes in approximately 50 seconds. The ECCS is initiated by either the low water level sensors in the reactor vessel or high drywell pressure sensors. The ECCS begins delivering flow to the vessel at -30 seconds after the accident. Figure 6-13 shows the vessel pressure and water inventory transient following the accident.

The transient minimum critical heat flux ratio (MCHFR) for the highest powered fuel bundle during the blowdown is shown in Figure 6-14. The axial power shape was chosen to assure t.hat the fuel bundle was initially operating at thermal limits.

As is evident from the figure, the MCHFR decreases initially after the accident occurs, increases slightly, and then decreases to less than 1.0 when core flow stagnates due to the uncovering of the jet pumps. Steam then blankets the reactor core and film boiling is established. However, this heat transfer is conservatively neglected for this analysis (i.e.,

the heat transfer coefficient is set i:o zero).

MCHFR again becomes greater than unity as a result of the high core flow rates caused by water in the lower plenum flashing to steam. This flashing forces large quantities of water through the core and jet pumps. (See Figure 6-12.) With MCHFR greater than unity, reestablished nucleate boiling would quickly cool the cladding to near saturation temperature~ However, no credit for rewetting is taken. The Groeneveld film boiling correlation (AECL-3281) is used to determine the convection coefficient as instructed by the AEC Interim Acceptance Criteria (IAC).

When the core uncovers, it is assumed to be insulated. Drywell high pressure or reactor vessel low water level signal starts the HPCI and LPCI, the LPCS, and the standby a-<: power supply. When the core spray flow reaches rated value or wh~n the core is reflooded, the appropriate coefficients are applied.

Figures 6-8 and 6-15 show the peak cladding temperatures for four rod groups for a small and intermediate break, respectively. The intermediate break size shown is one that results in high peak cladding temperature in the smaller break size range.

Figure 6-16 shows core inlet and outlet quality versus time for the OBA. The curve is shown for only the DBA, because quality affects the film boiling heat transfer coefficient. For small and intermediate size breaks, nucleate boiling is assured as long as the core is covered. Nucleate boiling heat transfer coefficients are independent of fluid quality. When the core is uncovered, the heat transfer coefficient is assumed to be zero, even though a significant steam

• cooling coefficient would exist.

Figures 6-17, 6-18, and 6-19 show the heat transfer coefficient versus time for the small break, intermediate break, and design basis accident, respectiv,ely.

Figures 6-7 and 6-20 show the reactor vessel (RPV) water level versus time for the small break and1* intermediate break, respectively. ....

• Figure 6-14 shows the minimum critical heat flux ratio (MCH FR) versus time for the design basis LOCA. Because the flow transient for the small and intermediate break sizes is mild compared to the LOCA, it is not shown. As long as the core is covered, the MCHFR for smaller breaks is always greater than unity, and nucleate boiling is always assured.

6-23

1000 50 800 40 WATER LEVEL INSIDE SHROUD TAF

-* --------.--------------- 30

£

..J w

w

..J IC w

cl:

20 :!:

BAF 10 REACTOR VESSEL PRESSURE 0 --~......--~~~~~~~~~--~~--~~~~~~~~--.....~~~--~~~~~~~~....--~~~~~~~~~~~..... 0 0 100 2oo 300 400 TIME (sec)

Figure 6-13 Performance of ECCS with Failure of One Diesel Generator for the Design Basis Accident.

(1HPCI+2 LPCI + 1 CS+ ADS) AEC Assumptions

9 8

7 0

j::

ct a: 6 x

..j II.

...ct w

J:

..j 5

ct u

E.

a:

u

E

~ 4 z

j ..

3 2

0"--------'-------....L.------...U--------.LL----~-'---~---L-------.....11 0 2 4 6 8 10 12 14 TIME (sec)

Figure 6-14 Minimum Critical Heat Flux Ratio for DBA at Dresden 2/3 6-25

2400 HOT SPOT REFLOODED 2000 iL ROD 11

£. ROD20 w 1600 a:

..... ROD23 c(

a:

w ROD55 II.

I w

C) z 1200 0

cc(

d 800

• SEE FIG.6-26 FOR ROD 400 IDENTIFICATION .

0 ...___________.....____________...__________.-.. -"'------"""!"----,.___________...

0 200 .. **"'"". . . 400 600 . . 800 1000 TJME lsecl Figure 6-15 Cladding Temperature versus Time for an Intermediate Break with Failure of HPCI (0.1 ft2 Break)4 LPCI + 2 cs+ ADS AEC Assumptions 6-26

0.9 0.8 0.7 CHANNEL OUTLET QUALITY 0.6 0.5 0.4 0.3 0.2 0.1 CHANNEL INLET QUALITY 0 .._______--1________....._________,_________.i...1111::;....__...1._________..._________,

0 2 4 6 8 10 12 *14 TIME lsecl Figure 6-16 Quality versus Time for OBA at Dresden 2/3 6-27

10,000 1000 ...........

HOT-SPOT RE FLOODED 10 1

I I I 0 200 400 600 TIME lsecl Figure 6-17 Heat Transfer Coefficient for a Small Break 0.02 ft2 4 LPCI + 2 CS+ ADS AEC Assumptions 6-28

10,000 1000 -

IL 0

c-.i

~

E 100

~

z HOT-SPOT RE FLOODED

..~.

10 1

I I I 0 200 400 600 TIME (sec)

Figure 6-18 Heat Transfer Coefficient for an Intermediate Break (0.1ft2)4 LPCI +2CS +ADS AEC Assumptions 6-29

10,000 1000 -

0

-u..

N!

.:E a

~

I- ...

z w

u u..

u..

w 100 -

0 u

a:

w u..

Cl) z

<{

a:

I-I-

<{

w

I:

1.0 ~ '

REFER TO GENERAL ELECTRIC TOPICAL REPORT NED0-10801

- RATED SPRAY

- CHANNEL WETTING I,

0 1 2 I I I 10 I

20 I

50 100 200 500

. TIME AFTER ACCIDENT lsecl Figure 6-19 Heat Transfer Coefficients for OBA with LPG/ Injection Valve Failure AEC Analvsis =1 (2CS + HPCI + ADS)

&30

0.8 40 12, ..._,..

x w

w

~

w a:

a: w qi  :::> I-w ~

.... ~

w 3::

a:

~

0.4 20 TIME (secl Figure 6-20 Performance of ECCS with Failure of HPCI for an Intermediate (0.10 ft2J Break.

(4LPCI + 2CS +ADS) AEC Assumptions

Figure 6-21 shows the assumed power generation following a design basis accident.

System Capacity. System capacity description and detailed information is contained in the FSAR and has been shown to be adequate to keep peak cladding temperatures< 2300°F.

Long Term Core Cooling. Long term cooling is defined as cooling after the initial thermal transient has been terminated until the fuel can be safely removed. Long term cooling conditions have not changed with insertion of the reload fuel.

Peaking Factors Figure 6-22 is a plot showing the typical behavior of cladding temperature versus exposure for the OBA. Peaking factors giving the highest peak cladding temperatures occur at -15000 Mwd/T exposure. These peaking factors were then used in conjunction with the IAC calculational models to determine the stated peak cladding temperatures for Dresden-3.

Fuel Rod Perforations. The mechanism of fuel rod perforation during the LOCA has been studied extensively and is well understood. A fuel rod will perforate if the cladding hoop stress exceeds the ultimate strength of zircaloy at ttie peak cladding temperature experienced during the LOCA. The number of fuel rods perforated is therefore a function of the predicted peak cladding temperatures as well as the experimentally determined internal gas pressure distribution and perforation stress data. A plot of stress at perforation and ultimate strength of zircaloy at various temperatures is presented in Figure 6-23.

The calculated fission gas internal pressure distribution within the core is shown in Figure 6-24 for a typical 7 X 7 fuel design. The distribution is obtained by integrating the expected fission gas release rate for normal operation to the end of an equilibrium cycle when the accumulated fission products are maximum. In addition, the partial pressure of volatile materials and initial gas pressure is included. Because the 8 X 8 fuel design has lower operating fuel temperatures, a smaller diameter and thicker cladding relative to the 7 X 7, the stress distribution produced by Figure 6-24 for 7 X 7 fuel can be conservatively applied to 8 X 8 fuel. These data are used with the heatup analysis to determine the maximum percentage of fuel rod perforation for any size break and the worst single failure assumption (see Figure 6-25).

Conformance With Interim Acceptance Criteria. In the analyses discussed above there have been no deviations from the evaluation model described in Appendix A, Part 2 of the AEC Interim Policy Statement.

Effects of ECCS Operation on the Core. The mechanical effects of ECCS operation on the core, reactor coolant system and ECCS are those associated with the thermal effect of injection water into these systems which is cooler than these systems and components. These thermal stresses have been considered in the design of the core, reactor coolant system and ECCS.

There are no nuelear effects resulting from ECCS operation, since all control rods are inserted and the reactor remains subcritical during the injection of the cooler ECCS water.

There are no chemical additives in the ECCS water and therefore no chemical effects on the core, reactor coolant system or ECCS.

Lag Times. The system time delays assumed in the LOCA accident analyses are as follows.

Maximum Allowable Time From Maximum Time Delay After Signal Receipt Until the Pumps Receipt of Signal Until All have Reached Rated Speed Valve Motion is Complete*

System (sec) (sec)

HPCI 30 30 cs 30 30 LPCI 43 43 ADS 120 6-32

POWER GENERATION FOLLOWING A DESIGN BASIS RECIRCULATION LINE BREAK ACCIDENT

i c

E z 0.1 I'-

0 z

0 i=

CJ cIC

~

z 0

i=

C'l c

w w

IC w

zw 0 0.01 IC w*

~

0 IL 0.001 .....--------~~...a..~~......1~~~'-~--~-'-~~...i..~~.....1.~~~~'-~~..i....~~-'-~~~--"~~~"-~--""--------'------'

0.01 0.1 1.0 10 100 1000 TIME AFTER ACCIDENT (sec)

Figure 6-21 Power Generation Following a Design Basis Recirculation Line Break Accident

!?.

w a:

c(

a:

1500 Iii Q.

~

...w Cl

<11 z w Q L--...;...

~ Q 1000 c(

u

~

c(

w Q.

500 .....___

FUEL BUNDLE EXPOSURE (GWd/tl Figure 6-22 Variation of Peak Cladding Temperature with Fuel Bundle Exposure

f:::,. APED DATA IN AIR (SINGLE RODI l:::,. T APED PREOXIDIZED (SINGLE RODI 10,000 ** 0 0

IN STEAM APED DATA IN AIR (9-ROD TEST II*

APED DATA IN AIR (9-ROD TEST Ill Q NMPO DATA 0.076 cm x 0.63 cm x 2.54 GAGE SHEET SPECIMENS, E 0.06 min-1 IN ARGON 0*

NMPO IRRADIATED Zr TUBING IN ARGON 5,000

,o NMPO IRRADIATED Zr TUBING IN STEAM SEE NEDO 10329 FOR FULL EXPLANATION

\

\

\ BEST FIT CURVE

\ OF NMPO DATA GEMP-482

! \ PERFORATION

~

w a:

t; 2,000

\

\

\

\~

1,000

'\ 0 500 CONSERVATIVE TENSILE-STRESS ~ >' '

CURVE . '

..... * . .,. .................... 0 NO PERFORATION

200 ..___...._________________.....~________________......__________________......____________......

',0 1,400 1,800 2,200 2,600 3,000 TEMPERATURE (Of)

Figure 6-23 Fuel Rod Perforation Data 6-35

86.5%

w a:

0 u

z en 8a:

II.

0 Jc B

0.2%

0.1%

0 100. i 500 1000 1500 INTERNAL PRESSURE (psial (MAXIMUM EXPECTEDI Figure 6-24 Distribution of Internal Pressure Within Rods 6-36.

LPCI INJECTION VALVE FAILURE 16

.*~ 12 HPCI FAILURE z

0

~

cca::

0

"'a::w

~

ga:: 8

//

/

4 I

I I

I 0.1 0.2 0.5 1.0 2.0 5.0 BREAK AREA (f12)

Figure 6-25 Percent Rod Perforation versus Break Area (AEC Assumptions)

6.2.3 Transient Analysis and Core Dynamics 6.2.3.1 Identification of Abnormal Operational Transients A complete range of single failure caused events which are abnormal but reasonably expected during the life of the plant were analyzed as part of the original licensing of the plant. These were included in the FSAR and a review was conducted to determine the effect of the reload fuel on the original analyses of these events.

These transient analyses can be categorized into the following nuclear system parameter variations:

1. Nuclear system pressure increases

2. Reactor vessel water (moderator) temperature decreases

3. Positive reactivity insertions

4. Reactor vessel coolant inventory decreases

5. Reactor core coolant flow increases

6. Reactor core coolant flow decreases The purpose of the transient analyses is to show that safety related fuel damage limits would not be transiently reached during any of the postulated events and to establish certain design objectives such as keeping the safety valves closed during the worst case pressurization transient. These damage Iimits are the same as those described in Sections 3 and 4 above, i.e., the 1% cladding strain and MCHFR = 1.0 limits.

Events Resulting in a Nuclear System Pressure Increase Loss of Main Condenser Vacuum A loss of condenser vacuum causes scram, turbine stop valve closure and, at a lower vacuum setpoint, turbine bypass valve closure; thus an instantaneous loss of vacuum is a turbine trip without bypass. Once initiated, all of the turbine stop valves achieve full closure within about 0.1 second. Low vacuum and closure of multiple stop valves of more than 10 percent both initiate reactor scram. Following the stop valve closures, the reactor pressure rises causing the collapse of voids in the core, resulting in increased neutron flux and further increase in pressure. The pressure increase is terminated by the opening of the relief valves.

Closure of All Main Steam Line Isolation Valves Closure of one i.solation valve at powers less than rated is permitted for testing purposes without initiating a scram signal. However, if three steam lines are closed in excess of 10 percent, it is interpreted as the beginning of a system isolation and a reactor scram signal is initiated. This event results in scram at all power levels although it is more severe at high initial power. The consequences of the main steam line isolation valve events with trip scram are less severe than the corresponding turbine trip events because the isolation valve closure times are slower (3-10 seconds) than the turbine stop valve closure times.

Turbine Trip, High Power A turbine trip will have the same transient sequence as a loss of condenser vacuum except that for a turbine trip the turbine bypass valves would open. This would result in a transient less severe than an instantaneous loss of condenser vacuum.

Turbine trips from lower initial power levels decrease in severity to the point where scram may even be avoided within the bypass capacity if auxiliary power is available from an external source.

6-38

/

1. Bypass Valves Failure Following Turbine Trip, High Power. This event is included to illustrate that single failure could prevent the turbine bypass valves from opening in-conjunction with a turbine trip. However-,

this unlikely event would produce a transient similar to; but no more severe than, a loss of condenser vacuum.

2. Bypass Valves Failure Following Turbine Trip, Low Power. This abnormal operational transient is of interest because turbine stop valve closure and turbine control valve fast closure scrams are automatically bypassed when the reactor power level is low. Turbine first-stage wessure is used to initiate. this bypass. The highest power level for which these scrams remain bypassed is about 45% of rated power. Reactor scram results from high pressure or high neutron flux and the consequences are less severe than the loss of vacuum transient.

Generator Trip, High Power A generator trip is a loss of generator electrical load which results in a speed up of the turbine-generator; The turbine-generator acceleration protection devices trip to initiate the turbine control valve fast closure and a reactor scram signal. This transient is less severe than the similar case for a turbine trip since stop valve closure is slightly faster than control valve closure. At power levels below bypass capacity the bypass system will transfer steam around the turbine and avoid scram.

scram. Above bypass capacity high pressure scram will result unless operator action can* reduce power to within the bypass capacity.

Pressure Regulator Failure - Increasing Pressure, High Power If the regulator fails, the backup regulator will function automatically producing a slight 10 psi pressure change.

Pressure regulator malfunctions that result in the turbine steam flow shutoff and a nuclear system pressure increase are similar to but of milder consequence than the generator trip described previously because turbine control valve Closure time .is slower than the fast closure time of this valve.

Events Resulting in a Reactor Moderator Temperature Decrease

• • ~

\?

Feedwater Controller Malfunction - Maximum Demand \,'

Failure of the feedwater controller in the direction of increased feedwater flow results in a moderator temperature decrease causing a reactor power increase through the effect of the negative void reactivity coefficient.

This initial power increase is not sufficient to initiate a high flux scram signal and the reactor will continue 'to operate at a slightly increased power level while water level increases. Under severe conditions, the mismat.ch between the steam line mass flow rate and the feedwater mass flow rate will cause the water level in the reactor vessel to rise at the rate of approximately 3 inches per second. A high water level turbine trip will be initiated when the sensed level has been increased by approximately 1.5 feet. This transient then reverts to that of a turbine trip.

Events Resulting in a Positive Reactivity Insertion Continuous Rod Withdrawal During Reactor Startup (See section 6.2.4).

Events Resulting in a Reactor Vessel Coolant Inventory Decrease Pressure Regulator Malfunction - Decreasing Pressure, High, Medium, Low Power If either the operating pressure regulator or the backup pressure regulator fails in an open direction, the turbine admission valves can be fully opened, and the turbine bypass valves can be partially opened. This action initially results in decreasing coolant inventory in the reactor vessel as the mass flow *rate of steam.leaving the vessel exceeds the mass flow rate of water entering the vessel. This depressurization results in the formation of voids which causes neutron 6-39'

power to decrease. The main steam line isolation valves automatically close when the pressure at the turbine decreases by approximately 100 psi. After the isolation valves begin to close, this transient reverts to the transient for closure of all main steam line isolation valves except with a reduced initial power level. The sequence of actions vary for this event depending upon the initial power level.

Loss of Feedwater Flow.

A loss of feedwater flow results in a situation where the mass flow rate of steam leaving the reactor vessel exceeds the mass flow rate of water entering the vessel, resulting in a net decrease in the vessel coolant inventory. After the water level in the vessel drops to the low level scram setpoint a reactor scram signal is initiated, and after an additional drop to the low-low level setpoint, a signal to close the isolation valves is initiated. After the isolation valves close 10 percent, a second reactor scram signal is initiated. This transient then reverts to that of an isolation valve closure.

Events Resulting in a Core Coolant Flow Increase Recirculation Flow Controller Malfunction - Increasing Flow Failure of the master controller can result in a speed increase of both recirculation pumps so that flow increase would cause the neutron flux to increase beyond initial values. The most severe case, however, is the failure of the speed controller of one of the motor-generator sets, since the speed controller rate limits are adjusted to keep the effect of master flow controller failure less severe than that of single speed controller failure. As a result the high neutron flux scram setpoint may be reached. If this setpoint is not reached, no system limits are exceeded. The bypass system can adequately handle the increase in steam flow. The increased recirculation flow will not cause system damage.

Startup of Idle Recirculation Pump The event will not raise power sufficiently to initiate scram if initial power is below - 60%. If the idle loop water is sufficiently low in temperature, a scram might be required, but even in the worst case, this transient is less severe than the above.

Core Coolant Flow Decrease Failure of a Recirculation Pump(s)

Recirculation pump failures would cause an abrupt reduction in core flow, thereby, increasing core void fraction and decreasing reactor power. The most rapid decrease in core flow and therefore the most severe event in this category would be caused by a seizure of one pump. For that event core flow and MCHFR would quickly reach their minimum values (1 to 2 seconds). however, MCHFR would stay above 1.0 for all power levels thereby maintaining adequate thermal margins. Vessel pressure would initially decrease but the pressure regulator would maintain control as the reactor settles out to the final low power conditions.

Recirculation Flow Controller Malfunction - Decreasing Flow A failure in the recirculation flow controller could cause the variable speed converter to move at its maximum speed in the direction of zero pump speed and flow. The transient would be very similar to a trip of one recirculation pump. However, the pump speed reduction would be slower than during the pump seizure discussed above so that the decrease in fuel thermal margins would be less. This transient would therefore, be similar but not as severe as the one for the pump seizure event.

6.2.3.2 Evaluation of Abnormal Operational Transients Nuclear System Pressure Increase For those postulated events resulting in a primary system pressure increase, the most severe of which is the turbine trip without bypass, the most important variable parameters affecting the magnitude of the pressure increase are the void coefficient and negative reactivity inserted from the control rod scram. The void coefficient becomes less 6-40

negative with exposure which has the effect of reducing the peak of the pressurization type transients. However. the scram reactivity rate is also slowed down as explained in the recently revised transient analyses previously submitted for Dresden-3. ( 1 2 l This more than offsets the favorable void coefficient change.

The scram reactivity rate is primarily a function of the core average reactivity effects and as such does not depend on the individual fuel bundle configuration. Therefore, the 8 X 8 fuel has no significant.effect that 1s different from 7 X 7 fuel.

. Moderator Temperature Decrease Moderator temperature decreases from such events as a feedwater controller failure in the increasing direction are also affected primarily by the void coefficient and the scram reactivity change. However. these transients are less severe because they are slower in nature and the resulting minor pressure and flux increases are of no concern for end of cycle conditions. Therefore, the 8 X 8 fuel has no significant effect that is different than 7 X 7 fuel.

Reactivity Insertion The effects of this transient are covered in Section 6.2.4;. Rod Withdrawal Error.

Decrease in Coolant Inventory These transients result in an RPV depressurization and, in some cases, a low level scram. Power l~el drops, due to  :;-*;";

void formation before the scram, and MCHFR effects are minimal. A mild repressurization on MSIV ciosure at 850 psig *~

'~

occurs on some. RPV temperature transients are the only concern on some. Insertion of 8 X 8 has a negligible effect on ~,* .

these transients.

Core Coolant Flow Increase Th is transient is not a severe one and is basically affected by the reactivity increase from the cold water addition.

which is not significantly changed from that used previously. Insertion of the 8 X 8 fuel has no effect on these transients.

Core Coolant Flow Decrease A scram does not occur as a direct result of core coolant flow decrease transients, i.e., trips of recirculation pumps, one pump seizure and others in this category, so a void coefficient change would be the principal effect changing these transients. The change in void coefficient is not sufficient to significantly affect the results. Further.

• start-up tests where actual *recirculation pump. trips were conducted demonstrated that these transients are Comparatively mild. Insertion of the 8 X 8 fuel is not expected to have a significant effect on these* transients.

Confirmatory calculations are currently in progress.

Thermal-Hydraulic Stability (To be submitted later) 6.2.3.3 Conclusion The important transient analysis parameters of the 8 X 8 reload fuel have been reviewed and compared to those used in the FSAR and reference ( 12) transient analysis and the differences were found to be insignificant. The transient analysis presented iri the FSAR remains applicable to the reloadec;I core as long as the actual operating scram reactivity insertion rate remains within that assumed in the analysis. When the curve in the FSAR is exceeded, a reduction to 97%

of full rated power for plant operation will be instituted in order to make applicable the transient analysis of Reference 12. It is possible that near the end of the cycle the scram reactivity in~ertion rate may no longer lie in the range covered by Reference 12 ... Core performance will be monitored to identify the approach of such a condition and before it occurs a further reduction in power level, or other changes, may be necessary ia order to maintain the pressure margins shown in Reference 12.

Current estimates show that the FSAR scram curve will be reached 4 months following startup from the reload outage and that the scram curve used in Reference 12 will be reached 5 months after the FSAR scram curve is reached.

6.2.4 Rod Withdrawal Error 6.2.4.1 Identification of Causes*

Starting Conditions and Assumptions. The reactor is operating at a power level above hot standby at the time tne control withdrawal error occurs. The reactor operator has followed procedures and up to the point of the withdrawal error is in a normal mode of operation (i.e., the control rod pattern, flow set point, etc .. are all within normal operating limits). For these conditions it is assumed that the withdrawal error occurs with the maximum worth control rod.

Therefore, the maximum positive reactivity insertion will occur.

Event Description. While operating in the power range in a normal mode of operation the reactor operator makes a procedural error and withdraws the maximum worth control rod to its fully withdrawn position. Due to this positive reactivity insertion, the core average power will increase. More importantly. the local power in the vicinity of the withdrawn control rod will increase and potentially could cause localized fuel failures due to either achieving critical heat flux (CHF) or by exceeding the 1% plastic strain limit imposed on the cladding as the transient failure threshold.

The following list depicts the sequence of events for this transient.

Approximate Event* Elapsed Time (1) Event begins* operator selects and withdraws at maximum rod speed the maximum worth control rod 0 (2) Core average and local power increases (3) LPRM's alarm <5 sec (4) Event ends - rod block by R BM <30 sec Identification of Operator Actions. Under most normal operating conditions no operator action will be required since the transient which will occur will be very mild. If the peak linear power design limits are exceeded, the nearest local power range monitors (LPRM's) will detect this phenomenon and sound an alarm. The operator must acknowledge this alarm and take appropriate action to rectify the situation.

If the rod withdrawal error is severe enough, the rod block monitor (ABM) system will sound alarms at which time the operator must acknowledge the alarm and take corrective action. Even for extremely severe conditions (i.e.,

for highly abnormal control rod patterns, operating conditions, and assuming that the operator ignores all alarms and warnings and continues to withdraw the control rod) the R BM system will block further withdrawal of the control rod before fuel damage occurs.

6.2.4.2 Analyses of Effects and Consequences Methods, Assumptions, and Conditions. The analysis considers the continuous withdrawal of the maximum worth control rod at its maximum drive speed from the reactor which is operating at rated power with a control rod pattern which results in the core being placed on thermal design limits (i.e .. MCHFR = 1.9 and a peak linear power of 13.4 kW/ft for the 8 X 8 fuel or 17.5 kW/ft for the 7 X 7 fuel). A worst case condition is analyzed to ensure that the results obtained are conservative ..Also, this approach serves to demonstrate the function of the ABM system.

The worst case situation is established for the most reactive reactor state and assumes that no xenon is present.

This ensures that the maximum amount of excess reactivity which must be controlled with the movable control rods is present. During a normal startup sufficient time would be available to achieve some xenon and samarium buildup, and after some short period of operation samarium will always be present. This assumption makes it possible to obtain a worst case situation in which the maximum. worth control rod is fully inserted and the remaining control rod pattern is selected in such a way as to achieve desig11 thermal I imits in the fuel bundles directly adjacent to or diagonally adjacent 6-42

to the inserted maximum worth control rod which is to be withdrawn. It should be pointed out that this control rod configuration would be highly abnormal and could only be achieved by deliberate operator action or by numerous operator errors during rod pattern manipulation prior to the selection and complete withdrawal of the maximum worth rod. -

Figures 6-27 through 6-34 show the results for this worst case condition for Dresden-3. It should be noted that the RBM set point is selected to allow for failed instruments for the worst situation. This case demonstrates that even if the operator ignores all alarms during the course of this transient that the R BM will stop the rod withdrawal while the MCHFR is still greater than 1.0 and before the cladding reaches the 1% plastic strain limit.

6.2.5 Loading Error The worst case loading error for the reference core configuration occurs when a reload bundle is rotated 180 degrees in a location near the center of the core.

Proper orientation of fuel assemblies in the reactor is readily verified by visual observation and is assured by verification procedures during core loading. Five separate visual indications of proper fuel assembly orientation exist:

1. The channel fastener assemblies, including the spring and guard used to maintain clearances between channels, are located at one corner of each fuel assembly adjacent to the center of the control rod.

2. The identification boss on the fuel assembly handle points toward the adjacent control rod.

3. The channel spacing buttons are adjacent to the control rod passage area.

• 4. The assembly identification numbers on the fuel assembly handles are all readable from the direction of the center of the cell.

5. There is cell-to-cell replication ..

Experience has demonstrated that these design features are clearly visible so that any misoriented fuel assembly would be readily distinguished during core loading verification.

If, however, through an error, fuel assembly were installed rotated 180° from the prc;>per location which is the worst case rotational error, no fuel damage would be incurred during the subsequent power operation, even if the misoriented assembly were operating at the maximum permitted power. Analysis shows that this error would result.in a MLHGR :s;;;16.3 kW/ft and a MCHFR ~1.50 for the rotated bundle. These are less than the damage limits established for this fuel.

6-43

WIDE-WIDE CORNEA 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 36 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 Figure 6-26. 8 x 8 Reload Fuel Rod Identification 6-44

24 JULY 5, 1973 MLHGR VS ROD POSITION 20

~

11:

16 CZ:

CJ

z:

~

U1

~

8X8AT(12,141 12 8

0 10 20 30 40 50 60 ROD NOTCH WITHDRAW~

Figure 6-27 Dresden 3 Cycle 3 RWE Response Case 2 I

. : * ' ":"* - " ..* .),'J'.;{

1,1

MLHGR VS ROD POSITION 20 8X8 AT (10,101 12 7.20 AT ROD POSITION 0 9._____________......____________.....,______________"-_____________._____________.....i....____________,

0 10 20 30 40 50 60 ROD NOTCH WITHDRAWN Figure 6-28 Dresden 3 Cycle 3 RWE Response Case 1

2800 MCHFR VS ROD POSITION 4.075 AT ROD POSITION 0 2400 BXS AT 112,141

..j 2000 I

I 0

I:

IC LI.

'......,.f :z:

~ 1600 1200 800 0 10 20 30 49. 50. 60 70 ROD NOTCH WITHDRAWN Figure 6-29 Dresden 3 Cycle 3 RWE Response Case 2 I

-~-. '~

1,i

MCHFR VS ROD POSITION (7X71 4.123 AT ROD POSITION O 2000 II

I 0 1600

~ .::

Cl) Ir I&.

z

(,)

E (10,121 1200 1.06 800 .....____________...._____________......______________..._____________...i...______________.______________,

0 10 20 30 40 50 60 ROD NOTCH WITHDRAWN Figure 6-30 Dresden 3 Cycle 3 RWE Response Case 1

MCHFR VS ROD POSITION (8X81 5.005 AT ROD POSITION 0 2400 iii 2000

.I:

ic ll

i 0

.I:

r:c

~

tD I&.

z (J

~ 1600 1200 aoo~~~~~--~~~~---'--""'.'~~~~..L...~~~~.....L.~~~~~.J._~~~~~~~~~......J 0 10 20. 30 40 50 60 70 ROD NOTCH WITHDRAWN Figure 6-31Dresden3 Cycle 3 RWE Response Case 1

,i

REACTOR POWER VS ROD POSITION 1040 iii ic 1030

II~

0

~

~

'fl ...

g ~

in N

..... 1020 IC w

~

~

1010 1000 --~~~~~~--~~~~~~--~~~~~~--~~~~~~--~~~~~~--~~~~~~....~~~~~~....

0 10 20 30 40 50 60 70 ROD NOTCH WITHDRAWN Figure 6-32 Dresden 3 Cycle 3 RWE Response Case 2

REACTOR POWER VS ROD POSITION 1040

.i c:

I 1030

~

..j

~

qi N in U1 N IC 1020 w

3:

f(

1010 1000

• O 10 20 30 40 50 60 70 ROD NOTCH WITHDRAWN Figure 6-33 Dresden 3 Cycle 3 RWE Response Case 1 I

i". . . . . . ii

_..... ~ * ~' *, .... -_,. . .I .. ... * **

140 (16,49) (16,41)

• ~ []

STRINGS FAILED 24,49 (24,491 124.41 I 16.49 NO

FAILURES w

w E

z I&.

0 I

Cl z

120 16,41 16,41 AND Ci 16.49 w

a:

m a:

ROD BLOCK LINE 110 100 0 2 4 6 8 10 12 14 CONTROL ROD POSITION (feet WITHDRAWN)

Figure 6-34 Dresden 3 RBM Respon5e to Control Rod Motion Case 1 (8 X 8) Channel A+C 6-52

140 (16.491 (16,411 130 (24,49)

CJ (24.411 w

w

...I

...I STRINGS FAILED

<(

E z

~

0 120 c

CD u

! 24.49 C) z Q 16.49

<(

w NO IC FAILURES

E GI IC 16,41 ROD BLOCK LINE 110 16.49 AND 16,41 100 ..,________,,_________,,_________......______....1..._________..._________..._______......

0 2 4 6 8 10 12 14 CONTROL ROD POSITION (feet WITHDRAWN)

Figure 6-35 Dresden 3 RBM Response to Control Rod Motion Case 1 (8 X 8) Channel B+D 6-53

(24,331 (32,331 130 (24,251 Cl (32,251 w

w c(

~

z II. STRINGS FAILED 0

...c 120

.I:!

C> 24,25 z

Ci 32,25 c(

.w a: NO

E FAILURES ID a:

32,33 ROD BLOCK LINE 32,33 AND 32,25 100 0 2 4 6 8 10 12 14 CONTROL ROD POSITION (feet withdrawn)

Figure 6-36 Dresden 3 RBM Response to Control Rod Motion Case 2 (7 X 7) Channel A+C 6-54

140 (24,331 (32,33) 130 (24,25) (32,25)

STRINGS FAILED

24,25 w

w

_, 32,25 c(

j:: NO z

0 FAILURES c 120 B

"Ciz c( 32,33 w

a: 32,33 AND

E 32,25 Ill a:

100 0 2 4 6 8 10 12 14 CONTROL ROD POSITION !feet WITHDRAWN)

Figure 6-37 Dresden 3 RBM Response to Control Rod Motion Case2 (7 X 7) Channel B+D 6-55

REFERENCES-SECTION 6

1. Paone, C. J., and Woolley, J. A., Rod Drop Accident Analysis for Large Boiling Water Reactors, Licensing Topical

.Report March 1972 (NED0-10527).

2. Stirn, R. C., Paone, C. J., and Young, R. M., Rod Drop Accident Analysis for Large BWRs, Licensing Topical Report, July 1972 (NED0-10527, Supplement 1).

3. Stirn, R. C., Paone, C. J., and Haun, J. M., Rod Drop Accident Analysis for Large Boiling Water Reactors Addendum No. 2 Exposed Cores, Licensing Topical Report, January 1973 (NED0-10527, Supplement 2).

4. "Technical Basis for Allowable Rod Worth Specified in Technical Specification," AEC Dkts. 50-237, 50-249, 50-254, and 50-265.

5. "Technical Basis for Changes to Allowable Rod Worth Specified in Technical Specification 3.3.8.3 (a)", to be submitted later under separate cover.

6. Slifer, B. C., and Rogers, A. E., Loss-of-Coolant Accident and Emergency Core Cooling Models for General Electric Boiling Water Reactors, Licensing Topical Report, April 1971 (NED0-10329 and NED0-10329 Supplement 1).

7. Duncan, J. D., and Leonard, J. E., Modeling the BWR/6 Loss-of-Coolant Accident: Core Spray and Bottom

/

Flooding Heat Transfer Effectiveness, March 1973 (NEDE-10801).

8. Linford, R. B., Analytical Methods of Plant Transient Evaluations for the General Electric Boiling Water Reactor, February 1973 (NED0-10802).

• I

9. In-Core Nuclear Instrumentation Systems for Oyster Creek Unit 1 and Nine Mile Point Unit 1 Reactors, August 1968 (APED-5456).

10. Morgan, W.R., In-Core Neutron Monitoring System for General Electric Coiling Water Reactors, November 1968, revised April 1969 (APED-5706).

11. Boyden, J. E., et al., "Summary Memorandum on Excursion Analysis Uncertainties," Dresden Nuclear Power Station Unit 3 Amendment No. 3.

12. Dresden-3. Special Report No. 29. Dkt. 50-249.

[

(

6-56 I.

7. TECHNICAL SPECIFICATIONS 7.1 SCOPE OF CHANGES The technical specification changes recommended herein are based on the assumption that previously recommended changes have been or are in the process of being incorporated into the Technical Specifications. The principal change of interest concerns the reduction in allowable rod worth for an in-sequence control rod.

7.2 SPECIFIC CHANGES Item Location Change Reason Basis statement for 1.1 Bott.om of pg 12 Add a reference to The re-evaluation needs this submittal to be referred to Basis statement for 2.1 E & F End of para. E & F As above As above pg 18 Basis statement for 1.2 Refs. at bottom As above As above of pg 20 Basis statement for 2.2 Third sentence As above As above on pg 48 Basis statement for 3.2 Next to last Change to read This is a result of the sentence on pg 21 " ... MCHFR is new evaluation

-1.06, ... "

Limiting condition Sec. 3.a on pg 57 To be submitted This is a result of the for operation later under new evaluation separate cover Basis statement for B.3 Sec. B.3 first To be submitted This is a result of the sentence pg 62 later under new evaluation separate cover 7-1/7-2

"' * .Co.rrections to Dresden 3 Second- Reload License Submittal-*

1. *- _ Page '+-:2, Section 4.1.1~3: Change f' :.. .t>-f. (I~) + I°~ o(

To ..

'"F -: ~ (t-Dl)-+ £5<< .. --

2. Page 4-3, Section 4-.2~2: Change _ MCHFR S 17-~ 5k W/ft to MLHGR ~ 17. 5k W/ft Change. . MCMFR -~ 1. 9

-_.* *-. _ _- . _ * *. to MCHFR:?: 1. 9.

3. Page 6-1, Section 6.1.1, *_Paragraph 1, Line 1: * /

./

After* **. sequence of events for" * /

• Insert "the worst case of" l.J. Page 6-5, Sec.tion 6.2.2.3.2_~ Last paragraph, Line 1:

Change "minimized"

. To * "minimizer" S. >{>age 6~.6.,'>S.~.p.t.i,on-.6.2.2.3.5~1, Paragraph 6, Line .5:*

• _ ~-Remove: parenthesis. mark after " *** pl?nt experience) ** "

Change "Refe-rence. (4a)"

To "Reference 5"

. ~*

6 *. ,Page 6-6, Section 6.2.2.3.5.l, Paragraph 7, Line 3:

Change "Reference (4a)"

_ To "Reference. 5"

• 1. Page 6-7, Figure 6-1: Change curve label units

• .\..:* f rorn "MWd/t" to "GWd/t" *. (5 times)_  : . 'l

8. Page 6-7,*Figure 6-2:" Delete "Figure 3" r .. -* , ;: r 9.. Page 6-8, Figure 6-3:_ Delete "Figure 4" r ,

l ,.-:

10. Page 6-13, _Table 6-1:,

_._ On .the lin*e *following * "X Does not. apply", add "#Two CS pumps, * *

. an HPCI pump, and ADS remaining."

'****. ----.:..~*.! .. _ . .=.::::-1

~-----~- .. ~~--

ii.

• Page _6~-1.7, -F-igure---5~s*:~Lab~l the vertical axis*. cladding temperature, °F' *_

.. t'

12. Page 6-25, Figtire 6,-24:

• Change " **** Dresden 2/3" {

.. To ~.~ ** Dresden 3" . i

"* " ~-- . ** .. * *1: :** : ' *.... :.. ;:.. _".: '* .., .' . . . : .. ' .

~ ~ ~:  :. *.** *.r.-:... * '*"* .~ "

.J . .. ..- : '  ;*:. ~. ... .: .' '~ .' ...*,'\.""' ,: .. '\ .. *,~ ....~ ~ t

' *' '* ..* '*. -=**. t_.. :*:

.., .";- . *' *.: '-. .._.. ,_._:, '. :: ,_._:._;. :*_- _,, _- *: ;* .... -:_ ,, :.::): ~_. '.. ,>__. *.. -, ,.. ,* -~- .:~ -'*-~.-.._,I*)

"J. *.-.:_**-~:***..

'-J-:*~ .... : =:

-*.*.l.:_~*:: .... ~ . . :~:.o---~*... .. *-.

"-- ~ . ": ., ~ ..* , .

• ~ .

..::. *~,** .

~

-2"'"

13. Page 6-27, Figure 6-27: . Change"Dresden 2/3" To *"Dresden 3n 14-. Page. 6-41, Sect~on 6.2.3.3, Line 2:

After .n:found to be- insignificant."

Insert "except*f6r the scram* reactivity curve."

.~ ':'

i \

15~ f'age 6-4-2, Section. 6. 2. 4.1, /Paragraph 1, Line 2: ..*

• • change "control :withdrawal:

To "cont~l r_c,>d withdrawal" 16~ Page 6'.'"4.5, Figure 6".'"27:

• Delete' "july 5, 1973"

-:-.. *~

• 1

. ~ .....

'i i

• I

. ,t"i

~

'-I l-- '

("

.I

\

,I

. ' ~: . . i . 5 *.

l.

• ~* .. *-.:<I" ":,.

'*~

• t. ,.:*,"..*,: c' ":" ,*. . r-;f. :, ,.,*~* .. , . I. ~. ~**

• • • .:: .. ,' :.. . ::.~: ** - ;*:

• I ~: .... .': . ~ ' ,h *. ., *.* ':.  :< . . ~..... j,.' .::. :~ *.' *

• • fl;* .

,*--** 'i** ,:*, .. ~.. . "-.; .. ..... . ....  ;~. ;. .. *:*.1*.-  : ' ~., .. " .... ..

~

~ . .~* .

i, .* ,.. ~i:: :. ; *  ;.':**r.:;:

• . *. . . :.. ~ .. ~* ...... '

'~- '.'.

,_ ..