ML19309C541
| ML19309C541 | |
| Person / Time | |
|---|---|
| Site: | Three Mile Island  |
| Issue date: | 05/01/1967 |
| From: | JERSEY CENTRAL POWER & LIGHT CO., METROPOLITAN EDISON CO. |
| To: | |
| References | |
| NUDOCS 8004080732 | |
| Download: ML19309C541 (199) | |
Text
. -.
- O 1
1 s.
l 5 ,
53 '
E
./
O l
c l
o .
32 8004080-000] 115 L - _.
f
. TABLE OF CONTENTS Section Page 3 REACTCR 3-1 31 DESIGN BASES 3-1 3 1.1 PERFORMANCE OBJECTIVES 3-1 3 1.2 LIMITS 3-1 3 1.2.1 Nuclear Limits 3-1 3 1.2.2 Reactivity Control Limits 3-2 3 1.2 3 Ther. al and Hydraulie L1=1ts 3-2 3 1.2.4 Mechanical Limits 3-3 32 REACTOR DESIGN 3-6 3 2.1
. GENERAL SLNMARY 3-6
/ 3 2.2 NUCLEAR DESIGN AND EVALUATION 3-7 3 2.2.1 Nuclear Characteristics of the Design 3-7 3 2.2.2 Nuclear Evaluation 3-19 323 THERMAL AND HYDRAULIC DESIGN AND EVALUATION 3-26 3231 Thennal ami Hydraulic Characteristics 3-26 3232 Thermal and Hydraulic Evaluation 3-35 3 2.4 MECHANICAL DESIGN LAYOUT 3-51 3 2.4.1 Interre. Layout 3-51 3 2.4.2 Fuel Assemblies 3-56 32 4.3 Centrol Rod Drive System 3-69 33 TESTS AND INSPECTIONS 3-82.
331 NUCLEAR TESTS AND INSPECTICN 3-82 3 3 1.1 Critical Experiments 3 82
! 3 3 1.2 Zero Power, Apprcach to Power, and Power Testing 3-82 3-1 0000 116
Cc:rIs:TrS (Comt'd)
Section M
332 THERMAL AND HYDRAULIC TESTS AND HISPECTION 3-82 3 3 2.1 Reactor Vessel Flow Distribution and Pressure Drop Test 3-82 3 3 2.2 Fuel Asse=bly Heat Transfer and Fluid Flow Tests 3-83 3323 Preoperational Testing and Postoperational Testics 3-84 333 FUEL ASSEMBLY, COITIROL ROD ASSEMBLY, AND CON'IROL ROD DRIVE MECFANICAL TESTS AND INSPECTION 3-84 3331 Prototype Testing 3-85 3332 Model Testing 3-85 3333 Ccmponent and/or Material Testing 3-85 O 3 3 3.u Comt=e1 Roe Drive Teste emd Imsrect1om 3-86 334 INTERNALS TESTS AND DISPECTIONS 3-90
3.4 REFERENCES
3-91 O -
3-11 'D003 117
LIST OF TABLES Table No. Title Page 3-1 Core Design, Thermal, and Hydraulic Data 3-6 3-2 Nuclear Design Data 3-8 l l
3-3 Excess Reactivity Conditions 3-9 3-h First Cycle Reactivity Control Distribution 3-9 3-5 Shutdown Reactivity Analysis 3-13 3-6 Soluble Boron Levels and Worti- 3-lh 3-7 Exterior Neutron Levels and Spectra 3-17 3-7-1 Calculated and Experimental Rod and Rod Assembly 1 Comparison 3-21 3-8 Reference Core Parameters 3-24 3-9 First Mode Threshold Dimensions and Flatness 3-24 3-10
/0 3-11 Th2:eshold Ratio and Power Flatness Coefficients of Variation 3-25 3-29 3-12 DNB Results - Maximum Design Condition 3-31 3-13 DNB Results - Most Probable Condition 3-32 3-lh Heat Transfer Test Data 3-h1 3-15 Comparison of Heat Transfer Test Data 3 kh 3-16 Hot Channel Coolant Conditions 3-h5 3-16-1 DNB Ratios in the Fuel Assembly Channels 3-50f 1 3-17 Clad Circumferential Stresses 3-61 3-18 LRD Fuel Swelling Irradiatier. Program 3-67 3-19 Control Rod Drive Design Dsta 3-72 3-20 Control Rod Assembly Design Data 3-80 l
i O : .
I 3-111 (Revised 7-21-67) /t l 0003 118
LIST OF FIGURES O (At rear of Section)
Figure No. Title 3-1 Boror Concentration versus Core Life 3-2 Axial t. 2 to Average Power versus Xenon Override Rod Insertion 3-3 Axial Power Pru.4 . Xenon Override Rods 55 Per Cent Inserted 3-h Moderator Temperature Coefficient versus Boren Concentration 3-5 Moderator Temperature Coefficient versus Moderator Temperature and Various Boron Levels 3-6 Per Cent Initial Power versus Time Following Trip 3-6-a Effect of Fuel Temperature (Doppler) on Xenon Oscillations - 1 Beginnin6 of Life 3-6-b Effect 'f Fuel Temperature (Doppler) on Xenon Oscillations -
Naar End cf Life 3-6-c Control of Ax;al Oscillation with Partial Rods 3-7 Population IncLded in the Statistical Statement Versus DNB Ratio 3-8 Power Shape Reflecting Increased Axial Power Peak for 144-Inch Core 3-9 Distribution of Fuel Rod Peakin6 3-10 Possible Fuel Rod DNB's for Maxi =um Design Conditions -
36,816-Rod cc -
3-11 PossiEe Fuel Rod DNB's for Most Probable Conditions -
36,316-Bod Core 3-12 Distribution of Population Protected, P, and 1-P versus Number Rods for Most Probable Conditions 3-13 DNB Ratios (BAW-168) versus Reactor Power 3-14 Maximum Hot Channel Exit quality versus Reactor Overpower 3-15 Thez::al Conductivity of Uo 2 3-16 Fuel Center Te=perature at the Hot Spot versus Linear Power 3-17 Number cf Data Points ve:aus
- E/*C 3-18 Hot Channel Factor versus Per Cent Population Protected -
! 3-iv (Revised 7-21-67) l 0000 119 l
l
FIGURES (Cont'd)
Figure No. Title 3-19 Bumout Factor versus Population for Various Confidence Levels 3-20 Rods in Jeopardy versus Power 3-21 Ratio of Experimental to Calculated Bumout Heat Flux 3-22 Ratio of Experi= ental to Calculated Bu=out Heat Flux 3-23 Ratio of Experimental to Calculated Burnout Heat Flux 3-24 Ratio of Dperi= ental to Calculated Burnout Heat Flux 3-25 Ratio of Dperimental to Calcule.ted Burnout Heat Flux 3-26 Ratio of Experi= ental to Calculated Bumout Heat Flux 3-27 Ratio of Dperimental to Calculated Bumout Heat Flux 3-28 Ratio of Experimental to Calculated Burnout Heat Flux 3-29 Ratio of Experimental to Calculated Bumout Heat Flux
/ 3-30 Ratio of Experi= ental to Calculated Bumout Heat Flux 3-31 Ratio of Dperi= ental to Calculated Burnout Heat Flux 3-32 Ratio of Experi= ental to Calculated Bumout Heat Flux 3-33 Ratio of Experimental to Calculated Burnout Heat Flux 3-3h Ratio of Experi= ental to Calculated Burnout Heat Flux 3-35 Ratio of Experi= ental to Calculated Bu=out Heat Flux 3-36 Ratio of Experimental to Calculated Burnout Heat Flux 3-37 Paximum Hot Channel Dit Quality versus Reactor Power 3-38 Hottest Design and Nc=inal Channel Dit quality versus Reactor Power (without .mgineering Hot Channel Factors) 1 3-39 Flow Regime Map for Unit Cell Channel at 2,120 psig 3-39-a Flow Regime Map for Unit Cell Channel
- 3-39-b Flow Regime Map for Corner Channel 3-39-c Flow Reg 1=e Map for 'n'all Channel O
V -
3-V (Revised 7-21-67) #
}
FIGURES (Cent'd) Fi.ture No. Title
; k0 Hot Channel DNB Ratio Cc=parisen 3-41 Reactor Coolant Flow versus Power 3 h2 Thermal Conductivity of 95 Per Cent Dense Sintered UO 2
Pellets 3-43 Fuel Center Temperature for Beginning-of-Life Conditions 3-kh Fuel Center Temperature for End-of-Life Conditions 3 kh-a Per Cent Fission Gas Released as a Emetion of the Average 1 Temperature of the UO Fuel 2 3 kh-b Axial Local to Average Burnup and Instantaneous Power Comparisons 3 kh-c Fiscion Gas .ielease for 1.50 and 1.70 Max / Avg Axial Power Shapes 3 hh-d Gas Pressure inside the Fuel Clad for Various Axial Burnup and Power Shapes
'O 3 kh-e Nominal hel Rod Pcwer Peaks and Cell Exit Enthalpy Rise Ratios 3-kh-f Max 1=um Fuel Rod Power Peaks and Cell Exit Enthalpy Rise Ratios 3 kh-g Calculated and Design Limit Local Eeat Flux versus Enthalpy in the Hot Corner Cell at the Nc=inal Condition 3-kh-h Calculated and Design L1=it Local Heat Flux versus Enthalpy in the Hot Corner Cell at the Postulated Worst Condition 3 h5 Reactor Vessel and Internals - General Arrangement 3-h6 Reactor vessel and Internals - Cross Section 3 h7 Core Flcoding Arrangement 3 h8 Fuel Assembly .
3 h9 Orifice Ibd Assembly , 3-50 Control Rod Drive - General Arrangement i i. . l
- l 3-51 Control Rod Drive - Vertical Section *
~^#
3-52 (DELETED) 1 e 3-vi (Revised 7-21-67) 0000 121
. . . --. - ..- _ . _ ~ - . - -.
i l FIGUPES (Cont'd) j Figure No. Title 3-53 (DELETED) 1 3-5L (DELETED) 3-55 (DELETED) i
- 3-56 Control Rod Drive Control System Block Diagram 3-57 Limit Signal and Position Indication System i
3-58 Reactor Trip Circuit i 3-59 Control Rod Assembly O 4 3-vii (Revised 7-21-67) 000] ly
. . . -_ -- - - _ . _ . - . _ _ . . _ - - ._..._-___._.___L __ - - _ . _ - - -
O V 3 REACTOR 31 DESIcN EASES The reactor is designed to =eet the perfor=ance objectives specified in 3 1.1 without exceeding the li=its of design and operation specified in 3 1.2. 3 1.1 PEFEORMANCE OBJECTIVES The reactor is designed to operate initially at 2,452 Wt with sufficient design =argins to accommodate transient operation and instru=ent error without damage to the core and without exceeding the pressure at the safe-ty valve settings in the reactor coolant sys tem. The ulti= ate operating power level of the reactor is expected to be 2,535 trit, but additional operating info mation vill be required to justify operation at this hi her6 power level. Thus, this section of the report describes only reactor op-eration at the initial power level. The fuel rod cladding is desi6ned to maintain it: integrity for the antic-ipated core life. The effects of gas release, fuel dimensional chan6es, and corrosion- or irradiation-induced changes in the mechanical properties of cladding are considered in the design of fuel assemblies. Reactivity is controlled by control rod assembliec (CRA's) and chemical poison dissolved in the coolant. Sufficient CRA vorth is available to shut the reactor down (kerr s 0 99) in the hot cordition at any time dur-ing the life cycle with the most reactive CRA stuck in the fully with-drawn position. Redundant 9quipment is provided to add soluble poison to the reactor coolant to insure a similar shutdown capability when the re-actor coolant is cooled to a=bient te=peratures. The reactivity worth of CRA's, and the rate at which react:.vity can be added, is limited to insure that credible reactivity accidents cannot cause a transient capable of da= aging the reactor coolant system or caus-ing significant fuel failure. 3 1.2 LIMITS 3 1.2.1 Nuclear Limits The core has been designed to the followin6 nuclear 11=its:
- a. Fuel has been designed for an average bumup of 28,200 WD/MrU and for a =axi=u= burnup of 55,000 WD/MTJ.
- b. The power coefficient is negative, and the control system is capable of compensating for reactivity changes resulting from nuclear coefficients, either positive or negative.
- c. Control systems vill be available to handle core xenon insta-bilities should they occur during operation, witSut..jeopar-O dizin.3 the safety conditions of the syste=.
l , b\' :. 3-1 !23 i l
- d. The core vill have sufficient excess reactivity to produce the de-sign power level and lifeti=e withcut exceeding the control capacity g or shutdown =argin. .
- e. Controlled reactivity insertion rates have been li=ited to 5.8 x 10~2 A k/k/see for a single regulating CRA group withdrawal, and 7 x 10-6 A k/k/see for soluble boren re= oval.
- f. Reactor control and =aneuvering procedures vill not produce peak-to-average power distributions greater than those listed in Table 3-1.
The lov vorth of CRA groups inserted durirg power operation li=its power peaks to acceptable values. 3 1.2.2 Reactivity Control Li=its The control syste= and the operational procedures will provide adequate control of the core reactivity and power distribution. The follovird control limits vill be met:
- a. Sufficient contrcl vill be available to produce a shutdown =argin of at least 1% Ak/k.
i b. The shutdown =argin vill he =aintained with the CRA of highest worth ! stuck out of the core. l .
- c. CRA vithdrawal limits the reactivity insertion to 3 8 x 10-3 A k/k/sce on a single regulating group. Doron diluticn is also li=1ted to a reactivity insertion of 7 x 10-0 Ak/k/sec.
3 1.2 3 Themal and Hydraulic Luits The reactor core is designed to =eet the fellowing li=iting themal and hydrau-lic conditions:
- a. ~o central =elting at the design overpower (114 per cent).
- b. A 99 per cent confidence that at least 39 5 per cent of the fuel rods in the core are in no jeopardy of ex1.,eriencing a departure fro = nu- 1 cleate boiling (D G) during continucus operation at the design over-power.
- c. Essentially 100 per cent confidence that at least 99 96 per cent of the fuel rods in the core are in no jeopardy of experiencing a DNB during continuous operation at rated power.
I d. 2.e generation of net stes= in the hottest core channels is pemis-sicle, but stea= voids vill be low encugh to prevent flov instabili?.ic::, The design overpower is the highest credible reactor operating power pemitted by the safety syste=. No mal overpower to trip is significantly less than the 1 design overpcVer. The core rateu power is 2,h52 :st. . G g,\ y (*\'t e aet.: '
; 3-2 (Revised T-21-67)
[>; . GGG0 iq4 L
O 3 1.2.4 Mechanical Limits 3 1.2.4.1 Reactor Internals The reactor internal components are designed to withstand the stresses resulting from startup; steady state operation with one, tvo, three, or four reactor coolant pumps running; and shutdown conditions. No damage to the reactor internals vill occur as a result of loss of pumping power. Reactor internals vill be fabricated from SA-240 (Type 30k) material and vill be designed within the allavable stress levels pe m itted by the ASME Code, Section III, for normal reactor operation and transients. Structural integrity of all core support assembly circumferential velds vill he as-sured by compliance with ASME Code Sections III and IX, radiographic in-spection acceptance standards, and velding qualifications. The core support structure vill be designed as a Class I structure, as defined in Appendix 5A of this report, to resist the effects of seismic disturbances. The basic design guide for the seismic analysis vill be AEC publication TID-7024, " Nuclear Reactors and Earthquakes". Lateral deflection and torsional rotation of the lover end of the core support assembly vill be limited to pre' vent excessive =ovements resulting from aeismic disturbances and thus prevent interference with control red assemblies (CRA's). Core drop in the event of failure of the no mal sup- ' pd ports vill be limited so that the CRA's do not disengage from the fuel assembly guide tubes. The structural interrals vill be designed to maintain their functional integrity in the event of a major loss-of-coolant accident as described in 3 2.4.1. The dynamic loading resulting from the pressure oscillations because of a loss-of-coolant accident will not prevent CRA insertion. l 3 1.2.k.2 Fuel Assenblies 1 j l The fuel assemblies are designed to operate satisfactorily to design bumup and to retain adequrte integrity at the end of life to pemit safe removal from the core. The assemblies are designed to operate safely during steady state and tran-sient conditions under the combined effects of flow-induced vibration, cladding strain caused by reactor pressure, fission gas pressure, fuel growth, and differential themal expansion. The cold-vorked Zircaloy-k cladding is designed to be freestanding. Fuel rods are held in place by
=echanical spacer grids that are designed'to maintain di=ensional control .
of the fuel rod spacing throughout the design life without impairing clad-ding integrity. Contact leads are limited to prevent fretting. The spacer grids are also designed to pemit differential thermal expansion i of the fuel rods without restra'at that would cause distortion of the rods. l The fuel assembly upper end fitting and the centrol rod guide tube in the internals structure are both indexed to the grid plate above the fuel as-semblies, thus insuring continuous alignment of the guide channels for the CRA's. The control rod travel is designed so that the rods are always e
.w. <r 4'
0000 125 I
k '
3-3 l l 1 e . _ _ . _ . . _ _ _ . - _ _ _ _ - - - _ _ - - - - . _ _ _
engaged in the fuel assembly guide tubes, thus insuring that CRA's can al-O. vays be inserted. The assembly structure is also designed to withstand handling loads, shipping loads, and earthquake loads. Stress and strain for all anticipated nor=al and abnor=al operating con-ditions vill be limited as follows :
- a. Stresses that are not relieved by small defor=ations of the ,
materini vill be prevented from leading to failure by not per-mitting these stresses to exceed the yield strength of the ca-terial nor to exceed levels that vould use in excess of 75 per cent of the stress rupture life of the =aterial. An example of this type of stress is the circumferential membrane stress in the clad due to internal or external pressure.
- b. Stresses that are relieved by small defor=ations of the material, and the single occurrence of which will not make a significant contribution to the possibility of a failure, vill be permitted to exceed the yield strength of the =aterial. Where such stresses exceed the material yield strength, strain limits will be set, based on low-cycle fatigue techniques, using no more than 90 per cent of the material fatigue life. Evaluations of cyclic load-ings will be based on conservative estimates of the number of l cycles to be experienced. An exampic of this type of stress is the thermal stress resulting from the thermal gradient across the clad thickness.
- c. Combinations of these two types of stresses, in addition to the individual treatment outlined above, will be evaluated on the low-cycle fatigue basis of Item b. Also, clad plastic strain due to diameter increases resulting from ther=al ratcheting and/
or creep, including the effects of internal gas pressure and fuel swelling, will be limited to about 1 per cent.
- d. Mini =um clad collapse pressure =argins will be required as fol-lova:
(1) 10 per cent margin over system design pressure, en short time collapse, at end void. l l (2) End void must not collapse (must be either freestanding or have adequate support) on a long time basis. (3) 10 per cent =argin over system operating pressure, on short time collapse, at hot spot average te=perature through the clad wall. (k) Clad =ust be freestanding at design pressure on a short time basis at = 725 F hot spot average te=perature tFxough the clad vall. O
') '
o 3u 0003 !26
3.1.2.k.3 centrol Red Assembly (CRA) The control rod clad is designed to the same criteria as the fuel clad, as applicable. Adequate clearance vill be provided between the centrol rods and the guide tubes, which position them within the fLel assembly, so that control rod overheating vill be avoided and unacceptable mechan-ical interference between the control red and the guide tule vill not occur under any operating condition, including earthquake. Overstressing of the CRA components during a trip vill be prevented by minimising the shock loads by snubbing and by providing adequate strength. 3.1.2.h.k Centrol Ecd Drive Each centrol rod drive is prov'ided with a pressure bre udovn seal to allov 1 a controlled leakage of reactor coolant va'.,er. All pressure-containing ccmponents are designed to meet the requirements of the ASME Code, Section III, Nuclear Vessels, for Class A vessels. The control rod drives provide control rod assembly (CRA) insertion and withdrawal rates consistent with the required reactivity changes for re-actor operational load changes. This rate is based on the vorths of the various rod groups, which have been established to limit pcver-peaking flux patterns to design values. The max 1=um reactivity addition rate is specified to limit the magnitude of a pcssible nuclear excursion result-ing from a control system or operator =alfunction. The normal insertion p) ( and withdrawal velocity has been established as 25 in./ min. The control rod drives provide a " trip' of the CRA's which results in a 1 rapid shutdevn of the reactor for conditions that cannot be handled by the reactor centrol system. The trip is based en the results of various reactor emergency analyses, including instru=ent and control delay times and the amount of reactivity that must be inserted before deceleration of the CRA occurs. The maximum travel time for a 2/3 insertion of a CPA has been established as 1.h sec. The control rod drives can be coup. Led and uncoupled to their respective CRA's withcut any withdrawal movement of the CRA's. Materials selected for the control rod drive are capable of c % ting within the specified reactor environment for the life of the mu tanism without any deleterious effects. Adequate clearance vill be provided be-tween the stationary and moving parts of the centrol rod drives so that the CRA trip time to full insertion vill not be adversely affected by
=echanical interference under all operating conditions and seismic dis-turbances.
Structural integrity and adherence to allevable stress limits of the cen-trol rod drive and related parts during a trip will be achieved by estab-lishing'a limit on impact leads through snubbing.
- n g< '
0003 127 3-5 (Revised T-21-67)
l l
'1 l
l 32 REACTOR DESIGN 3 2.1 GENERAL
SUMMARY
The. important core design, thermal, and hydraulic characteristics are tabulated in Table 3-1. Table 3-1 Core Design, Thema.1, and Hydraulic Data Reactor Type Pressurized Water Rated Heat Output, MWt 2,452 Vessel Coolant Inlet Te=perature, F 555 Vessel Coolant Outlet Te=perature, F 602.8 Core Outlet Te=perature, F 60k.3 Operating Pressure, psig 2,185 Core and Fuel Assemblies Total Number of Fuel Assemblies in Core 177 Number of Fuel Rods per Fuel Assembly 208 Number of Control Rods per Centrol Rod Assembly 16 Number of Incore Instrumentation Positions per Fuel Asse=bly 1 h Fuel Rod Outside Diameter, in. 0.420 Clad Thickness, in. 0.026 Fuel Rod Pitch, in. 0.558 Fuel Assembly Pitch Spacing, in. 8 587 Uhit Cell Metal / Water Ratio 0.80 Clad Material Zircaloy-k (cold-worked) l Fuel Material Form U02 Dished-End, Cylindrical Pellets 11ameter, in. l 0 362 Active Length, in. Ikh j Density, % of theoretical 95 i Heat Transfer and Fluid Flov at Rated Power Total Heat Transfer Surface in Core, ft2 ,k8,578 l AverageHeatFlux, Btu /hr-ft2 167,620 l Maximum Heat Flux, Btu /hr-ft2 543,000 Average Power Density in Core, kv/1 79 60 Average Themal Output, kv/ft of fuel rod
- 5.k Maximum Themal Output, kv/ft of fuel rod .17 49 Max 1=um Clad Surface Te=perature, F 65k Average Core Fuel Te=perature, F g Maximum Fuel Central Te.perature at Hot Spot, F 1,385 k,160 W
., . . . . . .;~ ,.
' ". ~
r 3-6 0003 128
Table 3-1 (Cont'd) 6 Total Reactor Coolant Flov, lb/hr 131 32 x 10 Core Flow Area (effective for heat transfer), ft 2 47 75 Core Coolant Average Velocity, fps 15 7 Coolant Outlet Temperature at Hot Channel, F 644.4 Power Distribution Maximum /AveragePowerRatio,radialxlocal (Fa h nucleari 1.85 Maximum /AveragePowerRatio, axial (F:nuc*. ear) 1 70 Overall Power Ratio (Fq nuclear) 3 15 Power Generated in Fuel and Cladding, 7, 97 3 i Hot Channel Factors 1 Power Peaking Factor (Fq) 1.008 1 ) Flow Area Reduction Factor (F ) 0 992 J Incal Heat Flux Factor (Fqn) A 1.013 HotSpotMaemum/AverageHeatFluxRatio (Fq nue and mech.) 3 24 DITB Data Design Overpower Ratio 1.14 l
, DNB Ratio at Design Overpower (BAW-168) 1 38 DNB Ratio at Rated Power (BAW-168) 1.60 l
l l 3 2.2 m]CI2AR DESIGN AND EVAI,UATION The basic design of the core satisfies the folloving requirements:
- a. Sufficient excess reactivity is provided to achieve the design power level over the specified fuel cycle.
- b. Sufficient reactivity ec:2 trol is provided to permit safe reac-tor operation and shutdown at all times during core lifetime.
3 2.2.1 Nuclear Characteristics of the Design 3 2.2.1.1 Excess Reactivity The nuclear design characteristics are given in Table 3-2. The excess reactivities associated with various core conditions are tabulated in Table 3-3 The core vill operate for 410 full power days for the first cycle and vill have a 310 full power day equilibrium cycle. Design limits vill be held with respect to reactivity control and power distri-bution. In: ore instrumentation vill be used to insure proper power peaking levels. Single fuel assembly reactivity infonnation is also in-cluded in Table 3-3 t ; ,, 3-7 (Revised T-21-67) 0.0#3 J 2F 1 I
Table 3-2 Nuclear Desien Data l I Fuel Asse=bly Volu=e Fractions Fuel 0.285 M:derator 0 590 l Zircaloy 0.099 Stainless Steel O.011 ' l Void 0.015 1.000 Total ID2, =etric tons 91.61 Core Di=ensions, in. Equivalent Dia=eter Active Height 128 9 144.0 Lbit Cell 141.o U Ato=ic Ratio (fuel assembly) Cold Hot 2 97 2.13 i Full Power Lifeti=e, days First Cycle 410 Each Succeeling Cycle 310 Fuel Irradiation, hk'D/MI'U First Cycle Averag6 12,460 Succeeding Cycle Average 9,410 Feed Enrich =ents, v/o U-235 First Cycle 2.29/2.64/2 90 (by zone) Equilibrium Cycle 2 94(a) Control Data Control Rod Material Cd-In-Ag l Number of Control Rod Assemblies 69 ' Total R;a Worth (.1k/k), % 95 Control Rod Cladding Material Type 304 SS (a) Average feed enrich =ent. .
. O g 4 g a u
\
D
Table 3-3 F.xcess Reactivity Conditions Effective Multiplicatien - BOL(a) Cold, Zero Power 1.302 2 Hot, Zero Power 1.247 Hot, Rated Power 1.229 Hot, Equilibrium Xe, Rated Pcver 1.192 Hot, Equilibrium Xe and Sm, Rated Pcver(b) 1.158 h ie Fuel Assembly (c) Hot 0.77 Cold (d) 0,87 ("}BOL - Beginning-of-Life. (b) Includes burnup until equilibriu= sa=arium is reached. (C Based en highest probable enrichment of 3 5 veight per cent. (d)A cer.cer-to-center sssembly pitch of 21 in. is
/ ,
required for this kett in cold, nonberated Water b with no xenen or samarium. The minimum critical mass, with and without xenon and samarium poisoning, may be specified in a variety of forms, i.e., single a.;sembly, =ultiple assemblies in various gec=etric arrays , damaged or crushed asse=blies , etc. The unit fuel assembly has been investigated for ccmparative purposes. A single cold, clean assembly containing a maximum probable enrich =ent of 3.5 vt % is suberitical. Two assemblies side-by-side are supercritical except when both equilibrium xenon and samarium are present. Three assemblies side-by-side are supercritical with both equilibrium xenen and samarium present. 3.2.2.1.2 Rer.ctivity Control Distribution Control of excess reactivity is shown in Table 3 h. l Table 3 k First Cycle Reactivity Control Discribution l 5 ak/k
- 1. Centrolled by Soluble Boron
- a. Moderator Temperature Deficit (70 to 520 F) 3. ,k f~h j
- b. Equilibriu= Xenon and Samariu=
2.'y l2 , 1 y , l 3 (Revised 10-2-67) Od03131 l
O Table 3 h (Cont'd) 5 ak/k
- c. Fuel Burnup and Fission Product Suildup 16.0 2
- 2. Controlled by Inserted Control Rod Assemblies
- a. Transier.: Xenon (normally inserted) 1.h
- 3. Controlled by Movable Control Rod Assemblies
- a. Doppler Deficit (0 to 100% rated pover) 1.2
- b. Equilibrium Xenon 1.0
- c. Moderator Temperature Deficit (0.to 15%
power at end of life) 0.6
- d. Dilution Control 0.2
- e. Shutdown Margin 1.0
- f. Total Movable Control Worth Required 4.0
- h. Available Control Red Assembly Worths
- a. Total CRA Worth 10.0
- b. Stuck Rod Worth (rod of highest reactivity value) -3.0
- c. Minimus Available CRA Worth 7.0
- d. Minimum Movable CRA Worth Available 5.6 Exclanation of Items Above
- 1. Soluble Baron Zoron in solution is used to control the following relatively slow-moving reactivity changes:
- a. The moderator deficit in going frem ambient to operating temper-atures. The value shown is for the maximus change which would occur toward the end of the cycle.
- b. Equilibrium sa=arium and a part (approximately 1.h% ak/k) of the equilibrius xenon.
g c. The excess reactivity required for fuel burnup and fission prod-
*8 . , (l . uct buildup throughout cycle life. ;
s 0003 132 3-10 (Revised 10-2-67)
+ <
Figure 3-1 shows the typical variation in bcron concentration with life for Cycle 1 and the equilibrium cycle. Control rod assemblies (CRA's) vill be used to centrol the reactivity changes associated with the following:
- 2. Inserted Centrol 2
(DELETED)
'O Sufficient rod worth remains inserted in the core during ner=al op-eration to overccme the peak xenon tran tent folleving a pcVer re-duction. This override capability facilitates the return to normal operating conditions without extended delaya. The presence of these rods in the core during operation does not produce pcVer peaks above the design value, and the shutdevn =argin of the core is not ad-versely affected. Axial pcVer peak variation, resulting frc= par-tial cr full insertion of xenon everride reds , is described fully in Figures 3-2 and 3-3. The less of movable reactivity control due to the insertion of this greup produces no shutdcun difficulties and is reflected in Table 3-5
- i 9, 3. Movable Control
- a. Pcver level changes (Deppler) and regulation.
()l] ] } l } } I {, , . l l 3-11 (Revised 10-2-67) l
O
- b. The portion of the equilibri':s xenon not co.itrolled by soluble boren, approxi=ately 15 ak/k, is held by =cvable CRA's.
- c. Between zero and 15 per cent of full pcVer, reactivity conpensa-tion by CRA's ma - be required as a result of the linear increase of reactor coolant te=perature frem 520 F to the normal operating value,
- d. Additional reactivity is held by a group of partially inserted CRA's (25 per cent insertion maximu=) to allev periodic rather than continuous soluble bcron dilution. The CRA's are inserted to the 25 per cent limit as the boron is diluted. Autcmatic withdrawal of these CRA's during operation is allowed to the 5 per cent insertion limit where the dilutica procedure is again initiated and this grcup of CRA's is reinserted.
- e. A shutdown margin of 1% ak/k to the hot critical condition is also required as part of the reactivity controlled by CRA's.
- h. Rod 'Jorth A total of h.0% ak/k(8) is required in =ovable centrol- Analysis of l2 the 69 CRA's under the reference fuel arrangement predicts a totcl CRA vorth of at least 10.0% ak/k. The stuck-9ut CRA verth was also 2 evaluated at a vahe no larger than 3.0% ak/ktb). This evaluatien included selection of the highest verth CRA under the first CRA-out condition. The minimus available CRA vorth of 5.6% ak/k(a) is l2 sufficient to meet movable control requirements.
3.2.2.1.3 Reactivity Shutdown Analysis The ability to shut devn the core under both hot and cold conditions is illus-trated in Table 3-5 In this tabulation both the first and equilibrium cycles are evaluated at the beginning-of-life (30L) and the end-of-life (EOL) for shutdown capability.
- - . ? g t. ,
("}Does not include transient control. See Table 3 h. Firsy cycle. See Table 3 h. 3-12 (Revised 10-2-67)
O V Table 3-5 Shutdown Resetivity Analysis First Cycle Eauilibrium Reactivity Effects, % ak/k EOL EOL BOL EOL
- 1. Maximum Shutdown CRA Requirement Doppler (100 to 0% Power) 1.2 1.5 1.2 1.5 2 Equilibrium Xenon 1.0 1.0 1.0 1.0 Moderator Deficit (15 to 05 Power) 0.0 0.8 0.0 0.8 Total E.? 3.3 2.2 3.3
- 2. Maximum Available CRA Worth (" -10.0 .-10.0 -10.0 -10.0 Transient Xe Insertion Worth 1.4 1.h 1.h 0. 0 .
Possible Dilution Insertion 0.2 0.2 0.2 0.2
- 3. Minimum Available CRA Worth All CRA's In -8'k ~8*h -8'k ~9'O One CRA Stuck-Out(b) -5.4 -5.h -5.h -6.8
- O V 4. Minimum Hot Shutdown Margin All CRA's In -6.2 -5.1 -6.2 -6.5 One CRA Stuck-Out -3.2 -2.1 -3.2 -3 5 5 Hot-to-Cold Reactivity Changes (#}
All CRA's In 0.0 +6.h +3.0 +8.0 One CRA Stuck-Out -09 +5 5 +2.1 +7.1 d
- 6. Cold Reactivity Condition All CRA's In -6.2 +1.3 -3.2 +1.5 One CRA Stuck-Out k.1 +3.4 -1.1 +3.6 7 PPM Boron Addition Required for k
'#f
= 0 99 (cold) .
l All CRA's In 0 170 0 190 One CRA Stu-k-Out 0 330 0 350 (a) Total vorth of 69 CRA's. ( CRA of highest reactivity value. - Includes changes in CRA vorth, moderator deficit, and
- equilibrium Xe held by soluble boron.
,,'(N. No boron, addition.
.) .
0000 135 3-13 (Revised 10-2-67) , l
O Examinatica of Table 3-5 fer Mini =u= Hot Shutdevn Margin (Ite: L) shows that, with the highest worth CRA stuck cut, the cure can be maintained in a suberit-ical condition. Normal conditions indicate a minimum het shutdevn margin of 5.1% ak/k at end-of-life. l2 Under cenditions where a cooldcun to reactor building a=bient temperature is requiren, concentrated soluble boren will be added to the reactor coolant to produce a shutdevn =argin of at least 1% ak/k. The reactivity changes that take place between the hot-zero-pover-to-cold conditiens are tabulated, and the corresponding increases in soluble boron are listed. Beginning-of-life boron levels for several core conditions are listed in Table 3-6 alcng with boren verth values. Additional soluble boren could be added for situations involving
= ore than a single stuck CRA. The conditions shown with no CRA's illustrate the highest requirements.
Table 3-6 Soluble Boren Levels and Worth 30L Boron Levels, Core Ccnditions tem
- 1. Cold, k,f; = 0.99 No CRA's In 1,820 2 All CRA's In 1,290 One Stuck CRA 1,450
- 2. Hot, Zero Power, k,gy = 0.99 I
No CRA's In 2,080 All CRA's In 1,080 One Stuck CRA 1,380 l l 3. Hot, Rated Power l No CRA's In 1,860 l k. Hot, Equilibrium Xe and Sm, Rated Power No CRA's In 1,360 Core Condition Boron Worth, T ak/k/ rem Hot 1/100 Cold 1/75 O l.-pnn }2J (.. U103 0 g.; , ,, 3-1h (Revised 10-2-67)
O 3 2.2.1.4 Reactivity Coefficients Reactivity coefficients foz.s the basis for analog studies involving nor-mal and abnormal reactor operating conditions. These coefficients have been investigated as part of the analysis of this core and are described belov as to function and overall range of values.
- a. Doppler Coefficient The Doppler coefficient reflects the change in reactivity as a function of fuel temperature. A rise in fuel temperature re-sults in an increase in the effective absorption cross section of the fuel (the Doppler broadening of the resonance peaks) and a corresponding reduction in neutron production. The range for the Doppler coefficient under operating conditions is expected to be -1.1 x 10-5 to -1 7 x 10-) ak/k/F. l
- b. Mcderator Void Coefficient The moderator void coefficient relates the change in neutron multiplication to the presence of voids in the =Jderator. f Cores controlled by appreciable amounts of soluble boron may exhibit a small positive coefficient for very small void levels ;
(several per cent void), while higher void levels produce in- l creasingly negative coefficientg. The expected range for the ! ( void coefficient is tl.O x 10- to -3 0 x 10-3 6 k/k/% void. j
- c. Moderator Pressure Coefficient I
The moderator pressure coefficient relates the change in modera- { tor density, resulting from a reactor coolant pressure change, i to the corresponding effect on neutron productior. This coef-ficient is opposite in sign and considerably smaller when com-pared to the moderator temperature coefficient. A typical ran6e of pressure coefficients over a life cycle vould be -1 x 10-6 to + 3 x 10-6 ak/k/ psi.
- d. Moderator Temperature Coefficient The moderator temperature coefficient relates a change in neu-tron multiplication to the change in reactor coolant tempera-ture. Reactors using soluble boren as a reactivity control have fever negative moderator temperature coefficients than do cores controlled solely by movable or fixed CRA's. The major tempera-ture effect on the coolant is a change in density. An increas-ing coolant temperature produces a decrease in water density j and an equal percentage reduction in boron concen* ration. The l concentration change results in a positive reactivity component !
by reducing the absorption in the coolant. The magnitude of this component is proportional to the total reactivity held by soluble boron. O The moderator temperature coefficient has been parsmeterized
., .., for the reference core in terms of boron concentration and
, .o , c 0000 137 3-15
reactor coolant te=perature. The results of the study are shown in Figures 3 h and 3-5 Figure 3-k shows the coefficient varia-tion for ambient and operating temperatures as a function of soluble boron concentration. The operating value ranges from 1.0 x 10-4 at the beginning of the first cycle approx to -3 0 x1= 10-ate 1{ a+k/k/F at the end of the equilibrium cycle. Figure 3-5 shows the =oderator te=perature coefficient as a function of temperature for various poison concentrations for the first cycle. The coefficients cf the equilibrium cycle vill be more negative than those of the first cycle since the boron concentration levels are considerably lover. The positive temperature coefficient during the initial portion of each cycle vill not constitute an operational problem. The Doppler deficit represents a much larger reactivit,y effect in the negative direction and, together with the CRA system re-sponse, vill provide adequate control.
- e. pH Coefficient Currently, there is no definite correlation to predict pH reac-tivity effects between various operating reactors, pH effects versus reactor operating time at pover, and changes in effects with various clad, te=perature, and vater chemistry. Yankee (Rove, Mass.) Saxton, and Con Edison Indian Point Station No. I have experienced reactivity changes at the time of pH changes, but there is no clear-cut evidence that pH is the direct influ-g ,
encing variable without considering other items such as clad
=aterials, fuel assembly crud deposition, system average tem-perature, and prior system water chemistry.
Saxton experi=ents have indicated a pH reactivity effect of 0.16 per cent reactivity per pH unit change with and without local boiling in the core. Operating reactor data and the results of applying Saxton observations to the reference reactor are as fol-lows: (1) The proposed system pH vill vary from a cold =easured value of approx 1=ately 5 5 to a hot calculated value of 7 8 with 1,400 ppm boron and 3 ppm KOH in solutien at the beginning of life. Lifetime bleed dilution to 20 ppm boron vill reduce pH by approximately 0.8 pH units to a hot calculated pH value of 7 0. (2) Considering the maximum system makeup rate of 70 gps, the corresponding changes in pH are 0.071 pH units per hour for boron dilution and 0.231 pH units per hcur for KOH dilution. Applying pH vorth values of 0.16% Ak/k per pH
, as observed at Saxton, insertion rates are 3 16 x unig%
10~ Ak/lc see and 1.03 x 10-5% ak/k see, respectively. These insertion rates correspond to 1.03 per cent pover/ hour and 3.h per cent pcver/ hour, respectively, which are l easily compensated by the operator or the auto =atic con-l trol system. s . 3-16 00f.lil~!38
3 2.2.1 5 Reactivity Insertion Rates Figure 7-7 displays the integrated rod vorth of four overlapping red banka as a function of distance withdrawn. The indicated groups are those used in the core during pcver operation. Using approx 1=ately 1.2% a k/k CRA groups and a 25 in./ min drive speed in conjunction with the re-activity response given in Figure 7-7 yields a maxi =um reactivity inser-tien rate of 5 8 x 10-5 a k/k/sec. forsolubleboronremovalis7x10pemaximumreactivityinsertionrate Ak/ksecond. 3 2.2.1.6 Power Decay Curves Figure 3-6 displays the beginning-of-life pcver decay curves for the two least effective CRA vorths as outlined in Table 3-3, Item No. 3 Se pcver decay is initiated by the trip release of the CRA's with a 300 msec delay frcm initiation to start of CRA motion. Ihe ti=e required for 2/3 red insertion is 1.4 sec. 3 2.2.1 7 Neutron Flux Distribution and Spectra The neutron flux levels at the core edge and the pressure vessel vall are given in Table 3-7 7 Table 3-7 v Exterior Neutron Levels and Spectra Neutron Flux Levels. n/cm2/sec(a) Interior Wall of Flux Core Edge Pressure Vessel Group (x 1013) (x 1010) 1 0.821 Mev to 10 Mev 6.0 34 2 1.230 Kev to 0.821 Mev 90 75 3 0.414 ev to 1.230 Kev 6.2 37 4 Less than 0.414 ev 71 2.1 (a)These values include the =aximum axial peak-to-average power ratio of 1 7 I The calculations vere perfo=::ed using The Babcock & Wilecx Cc=pany's LIFE code code, (BAW-29}l), TOPIC.\ ASection 4-group36 3) is edit to obtained generate frem input the dataLIFE for the transport cutput which in-cludes diffusion coefficients, abscrption, removal, and fission cross sec-i l I tions, and the zeroth and first =cments of the scattering cross section. I l TOPIC is an Sn code designed to solve the 1-dimensional transport equation ) in, cylindrical coordinates for up to six groups of neutrons. For tr.e ra-dish ,and acimuthal variables, a linear approx 1=ation to the transport i
)
Ob' 1':' 3-17 L
equation is used; for the polar angle, Gauss quadrature is used. Scatter-ing functions are represented by a Legendre series. The azi=uthal angle can be partitioned into 4 to 10 intervals en the half-space between 0 and
- . The number of =esh points in tne. radial direction is restricted by the nu=ber of these intervale. For the core exterior flux calculations, four intervals en the a:Imuthal vere used. This allevs the " v" rm nu=-
ber of =esh points (240) in the "r" direction to describe the shield cc=- plex. An option is available 'to use either equal intervals on the azi-
=uthal angle or equal interials on the cosine of the angle. Equal inter-vals on the cosine were chosen since this provides = ore detail in the forward direction of the flux (tovard the vessel). Five Gauss quadrature points were used on the cosine of the polar angle in the half-space be-tveen 0 and r.
Results frc= the above =ethod of calculatica have been ec= pared with ther-
=al flux =easure=ents through an array of iron and water slabs in the LIDO pool reactor.(2) Although this is not a direct ec=parison with fast neu-tron =easure=ents, it does provide a degree of confidence in the =ethod since the magnitude of the thereal flux in shield regions is governed by fast neutron penetration.
Results of the cc=parisen showed that fluxes predicted by the LIFE-TOPIC calculation were lever, in general, by about a factor of 2. Results of the fast flux calculations are, censequently, increased by a factor of 2 to predict the nyt in the reacter vessel. The following conservatis=s were also incorporated in the calculations: j
- a. Neutron fluxes outside the core are based en a =ax1=u= pcVer density of kl vatts/cc at the outer edge of the core rather than an esti=ated average of 26 vatts/cc over life, resulting in a safety =argin of about h5 per cent.
- b. A =ax1=u= axial power peaking factor of 1 7 was used. This is about 30 per cent greater than the 1 3 expected over life.
Uncertainties in the calculations include the folleving:
- 1. The use of only four neutron groups to describe the neutron energy spectru=.
- 2. Use of the LIFE code to generate the 4-group cross sections.
In the LIFE progra=, the 4-group data in all regions are ec=- puted frc= a fission spectru= rather than a le dage spectru=. 3 Having only four intervals, i.e., n = k in the Sn calculation, to describe the angular seg=entation of the flux. It is expected that the cc=bination cf 1 and 2 above vill conserva 1re'ly predict s high fast neutron flux at the vessel vall because it underesti-
=ates the effectiveness of the then::al shield in reducing .he fast flux'.
In penetration through water, the average energy of the neutrens in the group above 1 Mev increases above that of a fission spectru=, i.e., t)e spectru= in this group hardens. For neutrons above 1 Mev, the nenelastic 0)) ' 5-18 0000 140
cross section of iron increases rcpidly with energy. Therefore, the as-sumption of a fissien spectrum to et.spute cross sections in the thermal . shield, and the use of a fev-group model to cover the neutron energy spec-trum, vould underestimate the neutron energy less in the ther=al shield and the subsequent attenuatica by the water between the vessel and ther-mal shield. The results frca 3k-group P3MG(3) calculations show that re-duction of the flux above 1 Mev by the ther=al shield is about a factor of h greater than that ecmputed from the h-group calculations. The effect of 3 above is expected to underecti= ate the flux at the vessel ve.11. In calculations at ORNI, using the nS technique, a comparison be- l tween an Sg and an S12 calculation was made in penetration through hydro-gen. The results for a variety of energies over a penetration range of 140 cm showed the Sh calculation te be icver than the S12 by about a fac-i tor of 2 at maximum. Good agreement was obtained between the S12 and moments method calculaticas. l The above uncertainties indicate that the calculation technique should overestimate the fast flux at the reactor vessel vall. However, the ecm- , parison with thermal flux data indicates a possible underesti= ate. Until a better ecmparison with data can be made, we have assumed that the under-estimate is correct and accordingly have increased the flux calculations by a factor of 2 to predict the avt in the reactor vessel. The reactor utilizes a larger water gap and thinner thermal shield between 1 the core and the reactor vessel vall when ecmpared to currently licensed , f plants. The effect of this steel-vater configuration on (a) the neutron 5 irradiation, and (b) the thermal stresses in the reactor vessel vall, were , evaluated as follevs: i I
- a. Neutron Irradiation l j
Calculations were perfor=ed in com:ection with the reactor ves-sel design to determine the relative effects of varying the baf-fle and ther=al shield thicknesses on the neutron flux (>l Mev) at the vessel vall. These 9alculations were performed with the P1 cptica of the P3MG1 codet3) using 3h fast neutron groups. l The results showed that the neutron flux level at the vessel l vall is dependent, for the most part, en the total metal and water thickness between the core and the vessel. However, there was some variation in fluxes depending upon the particular con- I figuration of steel-vater lamina lons. Also, the gain in neu-tron attenuation by replacing water with steel dirdnishes seme-what with increasing steel thickness. l l In general however, the results showed that for total steel thickne m in the range of 3 to 6 in., 1 in. cf steel in place of 1 1.2. of water would reduce the neutron flux above 1 Mev by about 30 per cent. In pure water the calculations showed that the neutron flux would be reduced, en the average, by a factor of 6 in 6 in. of water.
^ 3ased en the above analysis a ecmparison has been =ade of the neutron attenuation in this reactor vessel with those in San l
U't'
,0003 141 -
3-19 (Revised T-21-67) - l
Onofre. Turkey Point No. 3 and h, Indian Point No. 2 and Ginna. 1 lI The total distance between this core and the reactor vessel is 21 in. This provides frc= 1.5 to as such as 5.75 in. : ore dis-tance the core and the vessel than in the other reactors units. For neutrens above 1 Mev it was found that this additional dis-tance vould provide additional attenuation ranging frc= a fac-tor of 1.1 to 5 ti=es greater than that in the other PWR's con-sidered.
- b. fhermal Stresses The ga==a heating in the reactor vessel is produced by pri=ary gn==as fro = the core and by secondary gamens originating in the core liner, barrel, ther=al shield, and the vessel itself. In this reactor design the =ajor portien of the heat ?s generated by ga==a rays frc= the core and by secondary ga==a rays frc=
the core liner and barrel. Since the ga=nas frc= each of these sources =ust penetrate the ther=al shield to reach the vessel, the vessel heating rate is dependent on the ther=al shield thickness. For designs which e= ploy thicker ther=al shields, or in which internals are to be exposed to higher neutron fluxes, ga==a rays criginating in the thermal shield or in the vessel itself
=ay govern the vessel heating rates. Since ga==a rays frc=
these sources voald have to penetrate only portions or none of the ther=al shield to reach the vessel, the vessel heating in , such cases vould be less dependent en ther=al shield thickness / than in this reactor design. A ce=parisen was made between the ga==a attenuation provided by the water and metal in this reactor vessel and that in other PWR's by assu=ing that, in each design, the vessel heating was dependent en the ga==a ray attenuation provided by the thermal shield. This approach would be conservative since, as noted above for sc=e designs, ga==a sources other than those attenu-ated by the ther=al shield may contribute appreciably to the vessel heating. The results of the eccparisen shoved that the difference in gnea attenuation between this reactor and other PWR's ranged fro- legligible difference to a factor of 5 3 less for this reactor uajign. The =aximu= steady-state stress resulting frc= ga==a heating in the vessel has been calculated to be 3,190 psi (tensien). This is a relatively low value, and no proble=s are anticipated frc= ther=al stresses in the reactor vessel vall. 3.2.2.2 Nuclear Evaluation Analytical =edels and the application of these =odels are discussed in this section. Core instabilities associated vith xenon oscillation are also =entioned, with threshold data evaluated under reference conditions.
, . . m 0000 !42
~5 3-19a (Revised 7-21-o7)
3.2.2.2.1 Analytical Models Reactor design calculaticas are made with a large nu=ber of ce=puter codes. The choice of which code set or sets to use depends en which phase of the design is being analyzed. A list of codes used in core analysis with a brief discussion follevs in 3.2.2.2.2.
- a. Reactivity Calculations Calculation of the reactivity of a pressurized water reactor core is performed in one, tvo, or three di=ensions. The geo-metric choice depends en the type of calculations to be made.
In a clear. type of calculation where there are no strona; lo-calized absorbers of a type differing from the rest of the lattice, 1-dimensional analysis is satisfactory. This type of ' problem is handled quite well by the B&W 1-dimensional deple-tion package code LIFE. LIFE is a composite of MUFT (Ref. b), KATE (Ref. 5), RIP, WANDA (Ref. 6), and a depletion routine. Normally the MUFT portion is used with 3h energy groups, an exact treatment of hydrogen, the Greuling-Goertzel approxima-tien for elements of mass less than 10, and Fermi age for all heavier elements. The KATE portion nor= ally uses a Wigner-Wilkins spectrum. In WAITDA, h energy groups are utilized. Disadv:mtage factors for input to the ther=al group are cal-culated with the
<O A '
V
.*. .i, 0000 143 3-19b (Revised 7-21-67) b .' ' i l . ,
TERMCS (Ref. 7) code. This cede set has been shown to give
, reliable results for a reactivity calculation of this type.
Recent check calculations en critical experi=ents have a stan-dard deviation of less than 0 5 .ak/k. A 1-di=ensional analysis of a sec=etric arrange =ent, where there are localized strong absorbers auch as CPN s, requires a prel1=inary 2-dimensional m 1ysis. The required properties of the 1-di=ensional syste= are then =atched to the 2-d1=en-sicnal analysis. In this =anner, it is possible to analyze the simpler 1-dimensional syste= in a depletten survey proble= vith only a s=all loss in accuracy. The 1-dimensional calculations are used as preliminary guides for the more detailed 2-di=ensional analysis that follevs. Values of reactivity coefficients, fuel cycle enrich =ents, life-ti=es, and soluble poisen concentrations can be found to i= prove the initial conditicas specified for 2-di=ensional analysis. Two-d1=ensional reactivity calculations are done with either the PDQ (Ref. 8) or TURBO (Ref. 9) diffusion and/or depletion ecdes. These codes have =esh li=itations on the size of a configuration ubich can be shcvn explicitly and are often studied with quarter core sy==etry. Sy==etry is desirable in the design, and no loss in generality occurs. The gec=etric description includes each fuel asse=bly and as =uch detail as is possible, i.e., usually each unit in the fuel asse=bly. Analysis of this type per=1ts detailed pcVer distribution studies as well as reactivity anal- / ysis. De pcVer distribution in a large P4R core which has zone leading cannot be predicted reliably with 1-di=ensional calculaticas. This is particularly true when local power peak-ing as a function of pcVer history is of interest. It is neces-sary to study this type of proble= with at least a 2-di=ensional code, and in sc=e cases 3-d1=ensional calculations are necessary. Use of the 2-di=ensional progrs=s requires the generation of group constants as s Stnetion of =aterial ec. position, pcVer history, and gec=etry. For regions where diffusion theory is valid, MUFT and KATE vith TERMCS disadvantage factors are used to generate epither-m' and ther=al coefficients. This vould apply at a distance of a few =can paths frc= boundaries or dis-continuities in the fuel red lattice. Discontinuities refer to fuel asse=bly can, water channels, instrumentation ports, and CRA guide tubes. The interfaces between regions of different enrich =ent are censidered to be boundaries as well as the outer limit of the core. To generate coefficients for regions where diffusion theory is inappropriate several =ethods are utilized. The arrange =ent of l structural =aterial, water channels, and adjacent fuel red revs can be represented well in slab gec=etry. This proble: is ana-lyzed by P3 m (Ref. 3) which is effective in slab gec=etry. The coefficients so generated are utilized in the epither=al - energy range. Ccefficients for the ther=al energy range are n .1.? , p< .
. t 3-20 a e
- U.~lly e n f;. ;44
generated by a slab THERMCS calculation. The regions adjacent to an interface of material of different enrich =ent are also well repre-sente.1 with the P3MG code. Thr. arrangement of instrumentation ports and control red guide tubes Jends itself to cylindrical geometry. DTF-IV (Ref. 10) is quite ef-fective in the analysis of this arrangement. Input to DTF-IV is from GAM (Ref.11) and THERMOS or KATE. Iteration is required between the codes. Le riux shr;s is calculated by DTF-IV and cross sections by the others. The outer boundary of the core where there is a trans-ition from fuel to reflector and baffle is also represented by the DTF-IV code. The 3-dimensional analysis is accomplished by extend-ing the techniques of 2-dimensieral representation.
- b. Control Rod Analysis B&W has developed a procedure for analyzing the reactivity worth of 1 small Ag-In-Cd rods in fuel lattices. Verification of this procedure was made by the comparative analysis of 14 critical experiments with varying rod and rod assembly configurations (13,14). Critical lat-tice geometries were similar to those of the reference core design.
Baron cencentration ranged from 1,000 to 1,500 ppm. The Ag-In-Cd rods were arranged in various geometrical configurations which bracket the reference design. Water holes, simulating withdrawn rods, were in-cluded as part of the lattice study. The resulting comparison of the analytical and experimental vorths are shown in Table 3-7-1. Details j of the critical configurations are given in References 13 and 14. Table 3-7-1 Calculated and Experimental Rod and Rod Assembly Comparison Ag-In-Cd Rod Assembly - Rod Assembly - Core Assemblies Rods per H2 O Holes Calculated Experimental No. per Core Assembly per Core Worth, 4 a k/k Worth, 4 A k/k 1 5-B 4 4 252 2.00 1 98 k-F 4 9 0 3 38 3 34 5-c 2 12 276 2 38 2 35 k-D 1 16 0 1.43 1.k2 5-D 2 16 284 2.80 2.82 4-E 1 20 0 1 54 1 52 5-E 2 20 292 3 05 3 01 The mean error in calculating these configurations is shown to be less than 1 per cent. Comparison of the power shape associated with the 16-rod reference assemblies showed good similarity. Point-to-average power had a maximum variation of less than 2 per cent with experimental data. The analytical method used for this analysis is based on straight dif-fusion theory. Thermal ecefficients for a control red are obtained from THERMOS by flux-veighting. Epithermal coefficients for the upper energy groups are generated by the B&W LIFE program. The re-O- sulting coefficients are used in the 2-dimensional code PDQ to obtain the required eigenvalues. st t . ' ' ' .t 3-21 (Revised I-21-CT) }~p
r l GAKER and LISPM are used to prepare data for THERMCS. GAKER generatec 1 scattering cross sections for hydrogen by the Nelkin technique. LIDPM uses the Brown and St. John free sac =cdel for generating the rc=aining scattering cross scetions. THERMCS is used in two steps. First, the critical fuel cell is an-alyced to obtain a velocity-veighted disadvantage factor. This is used in the hc=ogenication of fuel cells and gives a first order cor-rection for spatial and spectral variation. The ratio of flux in the moderator to flux in the fuel was analyced to within 2 per cent of
' experi= ental valaes using the velocity-veighting technique. The sec-cnd step is to use THERMCS in a calculation where the Ag-In-Cd rod is surrounded by fuel. This is used to decerate the flux-veighted con-trol rod ec11 coefficients as a function of boron concentration. As a check on the validity of the THERMCS approach, extr polation dis-tances were ec= pared to those given by the Spinks =cthod (1h-1).
The agree =ent vac within 2.2 per cent for a set cf cases wherein the number densitics of Ag-In-Cd were varied in a range up to 250 per cent. All other ccefficients are generated by LIFE in =uch the sa=e manner as with THERMCS. The data are used in a 2-di=ensional PDQ layout where each fuel red cell is shown separately.
- c. Deter =ination of Reactivity Coefficients This type of calculation is different frc= the reactivity analysis caly in application, i.e., a series of reactivity calculations being required. Ccefficients are deter =ined for moderator te=perature, voiding, and pressure, and for fuel te=perature. These are calcu-lated frc= s=all perturbations in the required para =eter over the range of possible values of the para =eter. '/
The =oderator te=perature coefficient is deter =ined as a function of soluble poison concentration and coderator te=perature, and fuel te=- perature or Doppler coefficient as a function of fuel te=perature. The coefficient for voiding is calculated by varying the =ederator concentration or per cent void. O 3-21a (Revised 7-21-67)
. r 5h00!<16
- e
3 2.2.2.2 codes for Reacter Calculattens h This section contains a brief description of ecdes =entioned in the pre-ceding sections. THERMCS (Ref. 7) - This code solves the integral fem cf the Bolt = ann Transport Equation for the neutrcn spectru= as a function of position. A diagonalized connection to the isotropic transfer = atrix has been incorporated allevira a degree of anisotropic scattering. MVPf (Ref. 4) - This program solves the P1 cr 31 =ultigroup equa-tien for the first two Legendre coefficients of the direc-tional neutron flux, and for the isotropic and anisotropic ec=ponents of the sicving down densities due to a ecsine-shaped neutron source. Ccefficients are generated with MUFT for the epithemal energy rance. KATE (Ref. 5) - The code solves the Wigner-Wilkins differential equation for a hc=cgeneous =ediu= =cderated by che=1cally unbound hydregen atc=s in the mal equilibriu=. Coeffi-cients for the the=al energy range are generated by KATE. RIP - This progrs= averages cross sections over an arbitrary group structure, calculates resonance integrals for a set of re-solved peaks, and ec=putes L-factors for input to MUFT, P1MG, and P3MG. WANDA (Ref. 6) - This code provides nu=erical solutiens of the 1-di=ensional fev-group neutron diffusion equations. LIFE - This is a 1-dimensional depletien package code which is a ec=bination of MUFT, KATE, RIP, and WAUDA. The ec=bination
=echanizes the procedures for usi C the codes separately.
GAM (Ref.11) - This ccde is a =ultigroup coefficient generation progra= that solves the P1 equations and includes aniso-tropic scattering. Inelastic scatterir4 and resonance para =eters are also treated by GAM. P3MG (Ref. 3) - The code solves the cultienergy transport equation in various gec=etries. The code is pr*-"ily used for epi-themal coefficient generations. DTF (Ref. 10) - This code solves the =ultigroup, 1-di=ensional Boltn= ann transport equation by the =ethod of discrete ordi-nates. DTF allevs =ultigroup anisotropic scattering as well as up and down scattering. PDQ (Ref. 8) - This progrs= solves the 2-di=ensional neutren diffu-sion-depletien proble= with up to five groups. It has a flexible representation of ti=e-dependent cross sections L/
=eans of fit optiens.
0000 147
TURBO (Ref. 9) - This code is si=ilar in application to the PDQ depletion program. It, however, lacks the great flexi-bility of the PDQ fit options. CANDLE (Ref. 9) - This code is similar to TURBO, but solves the diffusion equations in one dimension. TNT (Rd. 9) - This code is simnar in applicatic.*, to TURBO, but j is a 3-dicetsional code extended from DRACO. 3 2.2.2 3 Xenon stability Analysis 1 Initial studies of the initial and equilibrium cores, where realistic fuel temperatures are generated by thermal-nuclear iteration, indiente no instability at any time during the life cycle. Dese results are en-couraging, but until more detailed analyses are completed, it will be ' assumed that axial xenon oscillations are possible. Az1=uthal oscilla-tions are unlikely, and radial oscillations vill cct occur. An extensive investigation must be completed before the stability of a core can be ascertained. An adequate solution can be found by first us-ing analytical techniques in the manner of PaM an and St. John to pre-dict problematic areas, and then by analyzing these with diffusion theorf programs that are coupled with heat transfer equations. The results of the stability analysis of the reference core are presented below, followed by the methods section containin6 the details of the threshold and diffusion theory calculations employed. The closing sec-tion outlines an overall approach to the solution of the stability prob-lem in regard to additional detailed calculative programs as well as a method for the correction of unbalanced power distributions.
- a. Summary of Results (1) Threshold Analysis In the threshold analysis axial, aximuthal, and radial'oscilla-tions were investigated for beginni of life, flattened, and slightly dished power distributions. 16) The results are as follows:
(a) For a fixed dimension, the tendency toward spatial xenon oscillation is increased as the flux increases. (b) For a fixed flux, the tendency.toward spatial oscillation is increased as the dimension of the core increases. (c) The large size of current PWR designs pernits an adequate xenon description usirs 1-group theory. (d) Flattened pcVer d!.stributions are = ore unstable than nor-mal beginning-of-life distributions. Dished power distri-O butions are even vorse. P '" 9li ,- 3-23 (Eevised 7-21-67) 0000 148
(e) In a modal analysis of the reference core, =cdal coupling 1 can be ignored. In addition, the core is not large enough to permit second-har:enic instability. (f) A large, negative pcVer coefficient tends to da= pen oscilla-tions. If this coefficient is sufficiently large, oscilla-tions cannot occur regardless of core size or flux level. Current FWR designs have a substantial negative pcver coef-ficient. (g) The critical diameter for ati=uthal oscillations is larger than the critical height for axial escillations. (h) The reference core design is not large enough to excite radial oscillations. (i) Examination of the dia=eter, height, and pcuer coefficient for this reference design indicates that escillations should not occur at the beginning of life with unflattened power distributicas. Ecvever, there exists a finite probability of oscillations at sc=e later time, since core depletion tends to flatten the power distribution. (j) The period of escillation (25 to 30 hours) is long enough . l to permit easy centrol of the oscillations. (k) The =odal analysis of this core toward the end of the initial cycle (with about 80 per cent flatness) showed that axial oscillations are possible, azi=uthal esc 111ations are un- -j likely, and radial oscillations vill not occur. (2) Decletion Analysis Diffusion-depletien calculatiens coupled with heat transfer equa-ticas were employed to investigate further the exial stability of the core since the analytical study indicated that this was the , most probable = ode of escillatien. The results follow: l (a) Axial instability did not occur at any time during the ini-tial cycle. An average fuel te=perature of 1,h00 F vas
=aintained during the cycle.
(b) The threshold for axial instability near the end of the ini-tial cycle was found to coincide with a core average fuel te=perature of 900 F. + Diffusien theory was also used to exa=ine the prablem of centrol-ling the system with rods if the stabilizing pcuer Doppler was not present. The folleving was concluded: (a) Partial control reds are quite adequate in controlling asial oscillations. These rods have 3-ft-1cng poison sectio s which are moved up and devn about the =idplane of the c=re to offset oscillatory power shifts. 3-23a (Revised 7-21-67) p M03 14cl
(b) Detailed power profiles vill be available to the reactor 1 operator as output frem the instru=entation. The large period of the oscillation vill allow partial rod =ovement such that axial power peaks are held well within allevable limits.
- b. Methods (1) Threshold Julysis The method used in the threshold analysis is an extension of the 1-group treatment including power coefficient introduced by Randall and St. John. One- and 2-group treatments have been compared, sud the conclusion drawn is that a 1-group =edel is satisfactory for large cores. For all three gec=etries, data were generated as a functica of:
(a) Core size. (b) Flux level. (c) Degree of flatness in the pcver distribution. (d) Power coefficient. (e) Reactivity held by saturation xenon. l In addition, slightly dished power distributions were investi-gated to show that any dishing resulting frc= high depletion is not sufficient to require correction to data based on replacing the dished segment with a flat power distribution. The effect of modal coupling has been examined and shown to be of no consequence for cores similar to the reference reactor de-sign. Values of the critical dimensien varied no more than 1 to 2.8 per cent for the same core with an without =cdal coupling. The lever value was cc=puted with a zero power coefficient and was not conservative without modal coupling. The higher value was ec=puted with the reference pcver coefficient and was cen-servative without =odal coupling. Table 3-8 summarizes those parameters for the reference core which affect the xenon stability threshold. The parameters vere calculated at two substantially different ti=es in core life. Reference physical di=ensions are also shown for cc=parisen purposes in the folleving discussion. Table 3-9 shows the threshold dimensions for first mode insta-bility as a function of flux flattening. The percentage of flattening is defined as 100 per cent times the ratio of the flattened pcver distribution to the total physical di=ensien under consideration. The parameters of Table 3-8 at two full power days were used since they are virtually the same as l1 O :
!+{ .* 'st: 0003 150 3-23b (Revised 7-21-67)
., ' , '. p .
these at 150 days but are =cre conservative. Axial depletion studies show that power distributions are flattened by 0, 63, and 73 per cent at 2,150, and 35h full pcuer days, respectively. A =axi=u= flatness of approximately 80 per cent =ay be expected for lon6 core life. An exa.aination of the data in Table 3-9 shows that--with the =ar-i=u= flatness--axial oscillations are possible, a:,1=uthal oscilla-tiens are unlikely, and radial oscillations vill not occur. Threshold di=ensicns for second = ode oscillations were 50 per cent larger in =agnitude than those shown in Table 3-9 for the first mode. Oscillations in the second mode vill not occur in the reference core. Table 3-8 Reference Core Parameters Two Full (Rated) 150 Full (Rated) 1 Fever Days Pever Days M2 , c,2 57.0 57.0 I g, n/c=2 -sec 3.9 x 10 13 3.8 x 10 13 a x (reactivity held by saturation xenon), ak/k 0.03h 0.033 / Doppler Ccefficient, ak/k/F -1.1 x 10-3 -1.1 x 10 -5 Moderator Temperature Ccefficient Positive but Small Negative s (power coeff.), ak/k/ unit - flux =-2.2 x 10-16 =-2.3 x 10-16 Equivalent Dimensions, ft Height 12.00 Diameter 10.7h Radius 5 37 1 Table 3-9 First Mode Threshold Dimensions and Flatness l Flatness, 5 Threshold Dimensions, ft O SO 80 Threshold height (axial oscillations) 18.5 14.1 11.8 Threshold dia=eter (aci=uthal oscillation) 20.h 16.5 lk.0 Threshold radius (radial oscillation) 16.8 16.7 lb.5 g s\ I ,i'* . 3-2h (Revised 7-21-67) 0003 1M ,N
Table 3-10 shows the values of H/D versus pcVer flatness for equal likelihood of axial, azimuthal, and radial first harmonic escil3ations, i.e., if the ccre is just at the axial threshold for axial oscillations, it can also be expected that there vill be azimuthal and radial oscillations provided the value of H/D in Table 3-10 is satisfied. H/D for this reference reactor is 1.12. Table 3-10 Threshold Ratio and Power Flatness Flatness. ". Ratio 0 20 50 80 100 H/D (axial versus azimuthal) 0 91 0.87 0.86 0.86 0.85 H/D (axial versus radial) 0 55 0.h9 0.42 0.kl 0.k1 The modal methods used to exa=ine the xenon oscillation problem made use of core-averaged quantities such as flux, pcver coef-ficient, and reactivity held by saturation xenen. In addition, flux distributions were limited to (a) Geometric distributions. f /~~) (b) Partially or totally flat. V (c) Slightly dished. (DELETED) The power distribution of Cycle 1 during early life is such that no tenon instabilities vill occur. The pcver flattening effect of fuel burnup vith tims renders the core more susceptible to xenen oscillations. , (DELETED) (2) Depletion Analysis 1 Core-averaged quantities were used in the analytical analysis. For a more comprehensive investigation, it is desirable to study xenon oscillaticns with diffusion-depletion pro 6 rams including heat transfer. Such calculations, which include the important local temperature effects, allow the designer to look for xenon oscillations under actual operating conditions. For these rea-sons, the BW LUE depletion program was = edified to include axial heat transfer. The equations and iteration scheme are outlined below: (a) The average fluid temperature for each axial rekion is ecm-puted frem a previcusly kncvn power density fistribution as follows: s
, ' ATg = (Teut - Tin)1=C[C"t PD (Z) dZ (A) 5" s .4 , c -
3-25 (Revised 7-21-67) r 1 ano,157 '
l l l 1 1 where: 1 g aT 1: te=perature change in region "1" PD(Z) : power density in Z direction l Zins Zout : region "i" boundaries i ara C- (3) H PD(Z)dZ vbere H = active fuel height. Equation (A) is solved to T of region "i". Since T in is hown frem core inlet conditions,outthe average fluid te=perature is de-fined as follows: l Tout + Tin T.- fluid i 2 (b) The newly co=puted region-averaged fluid te=peratures are used to co=pute new fluid densities. These fluid densities are then used to adjust the nu=ber densities for water and soluble poison. Lo-cal or bulk boiling is not permitted. (c) The average fuel te=perature for each axial core region is then ec=puted frc= the average fluid te=peratures and power densities : fuel g i+Tfluidy l l vhere FD1: coverage power density of region "i" and K is defined by I _ fuel - T_fluid core (E) N Core 1 (d) After the new fluid te=peratures, moderator densities, and fuel i te=peratures are obtained, these quantitie.s are used as new LIFE l input to obtain a new power distribution until either a conver-gence criterion is met er a specified nu=ber of iterations is l rade. , This analysis used an exact solution in that the spectru= vas recalcu-I lated for each :ene (11 axial :enes described the reactor) for each O i ' 3-25a (?.evised 7-21-67) 0000 153
g iteration at every ti=e step. This included the effects of the =oder- 1
' ator coefficient.
This LIFE package was used to deter = ire the effects of the uncertainty in the power Doppler on the stability of the core. ~he uncertainty in the Doppler was more than ec=pensated with a : eduction in fuel te=per-ature of 500 degrees. De reference core was analyzed with core aver-age fuel te=peratures of 1,kOO F and 900 F. Figurc 3-6a co= pares the cyclic response of these two cases following the 3-ft insertion and re-
= oval (after two hours) of a 1.2% Ak/k rod bank near the beginning of life. Case 1 on Figure 3-6a depicts the behavior of the core if the heat transfer equations were not included in the calculation.
Figure 3-6b shows the effect of fuel te=perature toward the end of life. It is easily verified that the 900 F fuel te=perature case ap-preached the threshold condition for axial oscillation in this core. On the basis of the infor=ation presented, it can be said that for a realistic fuel temperature this core does not exhibit cxial instability at any ti=e during the initial cycle. The 1-D odel was used to dste: nine a =ethod of controlling ,the core without taking into account the stabil1:ing effect of the power Doppler. Nor= ally, this would produce a divergent oscillation as shown in Fig-ure 3-6c. A study was co=pleted wherein a 1% Ak/k rod bank with a 3-ft-long section of regular centrol rod =aterial was successfully =a-neuvered to control the core after a perturbation of the power shape at a point about 3/4 of the var through Cycle 1. The controlled re-sults are also shown in Figure 3-6c. The =ini=u= red =otion was one /O t foot, and the ti=e step employed was 4.8 hours. More precise rod =ove-O =ent over shorter ti=e periods vould produce a =uch c=cother power ra-tio curve. This control mechanis= appears quite adequate,
- c. Conclusions .
Instability in the radial or a:1=uthal =cde is not expected since the dif-fusion theory study showed that the core is stable throughcut lifeti=e and the L/D ratio is 1.1. The results are encouraging, but until additional analyses are completed, it vill be assu=ed that axial xenon oscillations are possible. Consequently, rod =otion vill be used to ec=pensate for un-balanced power distribution as indicated by the instru=entation. Work is underway to provide a 2-di=ensional depletion progra= which allows nuclear-ther=al iterations. A detailed quantitative analysis of core sta-l bility and control procedures is to be undertaken with the new progrs=.
- l O
V . EC
'.
- S,' i '. i! 3-25b (Revised 7-21-67) 9 Y "I 3
"#, 4 l
323 THERMAL AND HYJRAULIC DESIGN AND EVALtlAT::oN h 3231 ' Iter =al and Hydraulic characteristics 3 2 3 1.1 Fuel Asse=bly Heat Transfer Design
- a. Design Criteria The criterion for heat transfer design is to be safely belev Departure frcm Nucleate Boiling (DNB) at the design overpower (114 per cent of rated pover). A detailed description of the annlysis is given in 3 2 3 2.2, statistical Core Design Tech-nique.
The input info mation for the statistical core design technique and for the evaluation of individual hot channels consists of the following: (1) Heat transfer critical heat flux equations and data corre-lations. (2) Nuclear power factors. (3) Engineering hot channel factors. (4) Core flow distribution hot channel factors. (5) Max 1=um reactor overpower. / Thece inputs have been derived from test data, physical measure-ments, and calculations as outlined below.
- b. Heat Transfer Equation and Data Correlatien Theheattransferrelationshipusedtopredigt)'i=1tingheat transfer conditions is presented in 3A'J-168,(17 The equation is as follows:
q" = (1.83 - o.000415 P) x 90,000
'G 2%
0 3987 e 0.001036 a ,3c - 1.027 x lo-6(a ;,,c)2 q" = best critical heat flux as prpicted by the itt fom, Stu/hr-ft P = core ope. ating pressure, psia G = channel =ats velocity,1b/hr-ft2 S = channel equivalent dia=eter, ft g . .m . . , ,
. 3-2s dOOD155
O , L = length up the channel to the point of interest, ft ATesc = inlet subcooling (Tsat - Tinlet), F T sat = coolant saturation temperature corre-spending to P, F This aquation was derived frcm experi= ental heat transfer data. An analysis of heat transfer data for this and other relation-ships is described in detail in 3 2 3 2 3, correlation of Heat Transfer Data. Individual channels are analyzed to deter =ine a DNB ratio, i.e., the ratio of the heat flux at which a DNB is predicted to occur to the heat flux in the channel being investigated. This DNB ratio is related to the data correlation aJ in Figure 3-7 A confidence and population value is associated with every DNB r.: tic as described in the Statistical core Design Technique. The plot of DNB versus P shown is for a confidence of 99 per cent. The DNB and population relattenships shown are also the values associated with the single hot channel analysis for the hottest unit cell where a 1 38 Dtm ratio corresponds to a 99 per cent confidence that at least 94 5 Per cent of the population of all such hot channals are in no jeopardy of experienc1=g a DNB.. This statement is a corollary to the total core statistical statement given in 3 1.2 3, thermal and Hydraulic Limits. The criterion for evaluating the ther=al design marsh for individual channels or the total core is the confidence-population relation-ship. The DNB ratios required to meet the basic criteria or limits are a function of the experimental data and heat transfer correlation used, and vary with the quantity and quality of data.
- c. Nuclear Power Factors The heated surfaces in every flow channel in the core are exam-ined for hest flux limits. The heat input to the fuel rods comprising a coolant channal is determined frcm a nuclear anal-ysis of the core and fuel assemblies. The results of this anal-
! ysis are as folleva: (1) The ncminal nuclear peaking factors for the vorst time in core life are: Fah = 1 79 Fz = 1.TD ( Fq = 3 04 (2) The design nuclear peaking factors fer the worst time in core life are: - r>c . 0000 L56 3-27
Fa h = 1.85 F: = 1. '70 h Fq = 3 15 wherc Fah = =ax/ avg total power ratio (radial x local nuclear) F: = =ax/ avg axial power ratio (nuclear) Fq ' Fah x F: (nucleartotal) D e nc=inal values are the =ax1=u= calculated values. The de-sign values are obtained by increasing the =ax1=um calculated total pcVer ratio, F ah, frc= 1 79 to 1.85 to obtain a =cre con-servative design. The axial nuclear factor, F , is illustrated in Figure 3-8. The distribution of pcvtr expressed as F/P is shcvn for two condi-tiens of reactor veration. D e first condition is an inlet peak with a =ax/xyg value of 170 resulting frc= partial inser-tion of a CRA group for transient control folleving a power level change. Bis condition results in the =ax1=u= local heat flux and =ax1=um linear heat rate. The second power shape is a sy==etrical ccaine which is indicative of the power distribution with xenon override rods withdrawn. S e flux peak =ax/s,vg value is 1 50 in the center of the active core. Ecth cf these flux shapes have oeen evaluated for ther=al DIG li=1tations. Se 1 biting condition is the 1 5 cosine pcVer distribution. The inlet peak shape has a larger =sx1=um value. However, the posi-tion of the 15 cosine peak farther up the channel results in a less favorable flux to enthalpy relationship. This effect has been de=enstrated in DIO tests of nonunifom flux shapes. (10) The 15 cosine axial shape has been used to deter =ine individual channel DITB li=1ts and =ake the associated statistical analysis. The nuclear factor for total radial x local red power, Fah, is calculated for each rod in the core. A distribution curve of the fraction of the core fuel rods cperating above varicus peak-ing factors is shcvn in Fictee 3-9 Line B shows the distribu-tion of the =axi=u= calculated values of Fa h for nc=1nal condi-tiens with a =ax bu= value of 1 79 De distribution of peaking factors for the design condition is obtained by increasing the
=ax1=u= calculated value for all rods in the core by the ratio of 1.85/1 79 or 1.033 to provide conservative results. Determi-nation of the peaking distribution for the design condition in this =anner has the effect of increasing reactor power by about 3 per cent. This assu=ptien is conservative since the distribu-tion with a =axbum peak Fah of 1.85 vill follow a line si=11ar to Line C where the average pcVer of all rods in the core is represented by an Fah of 1.0. The actual shape of the distri-
- bution curve is dependent upon statistical peaking relationships, CRA positions, =oderator cociitions, and operating histcry. Se
' shape of the distribution curve vill be = ore accurately de-scribed during the detailed core design.
.x ..i- 3-a 0000 157
- d. Engineering Hot Channel Factors Power peaking facters obtained frem the nuclear analysis are based on mechanically-perfect fuel asse:blies. Engineering hot channel factors are used to describe variations in fuel leading, fuel and clad dimensions, and flev channel gec=etry frem per-feet physical quantities and di=ensions.
The application of hot channel factors is described in detail in 3 2 3 2.2, Statistical ore Design Technique. The factors are detemined statistically frcm fuel asse=bly as-built or speci-fied data where Fq is a heat input factor, Fqn is a local heat flux factor at a hot spot, and FA is a flow area reduction fac-ter describing the variation in coolant channel flew area. Several subfactors are combined statistica"y to obtain the final values for Fq, Fqa, and F A. 2:ese subfactors are shown in Table 3-11. The factor, the coefficient of variation, the standard deviation, and the mean value are tabulated. Table 3-11 Coefficients of Variation CV No. Description 7 E CV 1 Flev Area 0.00075 0.17625 0.00426 2 Local Rod Diameter 0.000h85 0.k20 0.00116 3 Average Rod Diameter 0.000485 0.420 0.00116 (Die-drawn, local and average same) h I.ccal Fuel Loading - 0.00687 Subdensity 0.006h7 0 95 0.00681 Subfuel area 0.000092 0.1029 0.00089 (Diameter effect) 5 Average Fuel Ioading 0.00370 1 Subdensity 0.00324 0 95 0.00341 ! Sublength 0.16181 14h 0.00112 Subfuel area 0.000092 0.1029 l 0.00089 (Diameter effect) 6 Local Enrichment 0.00323 2.24 0.001kh l 7 Average Enrichment 0.00323 2.ek 0.0014h , 1 CV CoefficientofVariation,7/E , l e Standard Deviation of Variable E Mean Value of Variable . O (Enrichment values are for vorst case no=al aesay batch; V l
, =axi=um variation occurs fc- -"*-"~ enrich =ent. )
0 , ' .* , b,> , . . ..
,. )bh..)
l E, k. l 3 29 l l l l
- e. Core Flow Distribution Hot Channel Factors The physical arrangement of the reactor vessel internals and noz:les
- results in a nonuniform distribution of coolant flow to the various fuel assemblies. Reactor internal structures above and below the active core are designed to minimize unfavorable flow distribution.
A 1/6 scale model test of the reactor and internals is being per-formed to leronstrate the adequacy of the internal arrangements. The final variations in flow will be determined when the tests are com-pleted. Interim factors for flow distribution effects have been cal-culated from test data on reactor vessel models for previous pres-surized water reactor designs. A flow distribution factor is determined for each fuel assembly loca-tion in the core. Re factor is expressed as the ratio of fuel as-sembly flow to average fuel assembly flow. The finite values of the ratio may be greater or less than 1.0 depending upon the position of the assembly being evaluated. The flow in the central fuel assemblies is in general larger than the flow in the outermost assemblies due to the inherent flow characteristics of the reactor vessel. The flow distribution factor is related to a particular fuel assembly location and the quantity of heat being produced in the assembly. A flow-to-power comparison is made for all of the fuel assemblies. The 1 worst condition in the hottest fuel assembly is determined by applying model test isothermal flow distributicu data and heat input effects at power as outlined in 3.2 3 2.hi. Two assumptions for flow distri-bution have been made in the thermal analysis of the core as follows: ) (1) For the maximum design condition and for the analysis of the hot-test channel, all fuel asse=blies receive minimum flow for the 1 worst condition, regardless of asse=bly power or location. (2) For the most probable design conditions predicted flow factors have been assigned for each fuel assembly consistent with loca-tion and power. The flow factor assumed for the maximum design condition is conservative. Application of vessel flow test data and individual assembly flow factors in the detailed core desyn will result in improved statistical state =ents for the maxi =um design condition.
- f. Maximum Reactor Design Overpower .
Core performance is assessed at the =axi=um design overpower. The selection of the design overpower is based on an analysis of the re-actor protective system as described in Section 7 The reactor trip l point is 107 5 per cent rated power, and the =axi=um overpower, which is lik per cent, will not be exceeded under any conditions. 1 1 3-30 (Revised T-21-67)
O g. Maximum Design Conditions Analysis Suz:m:ary
- The Statistical Core Design Technique described in 3 2 3 2.2 was used to analyce the reactor at the =ax1=um design condi-tions described previously. De total number of fuel rods in the core that have a possibility of reaching DNB is shown in Figure 3-10 for 100 to n 8 per cent overpower. Point A on Line 1 is the -rimine design point for 114 per cent power with the design Fah nuclear of 1.85 Line a was calculated using the meximum calculated value for F ah nuclear of 1 79 to show the margin between maximum calculated and design conditions.
It is anticipated that detailed core nuclear analyses vill per-mit a loweri::s of the maximrm design value for Fah. D e number of fuel rods that may possibly reach a DNB at the maximum design condition with an Fah of 1.85 and at n 4 per cent overpower, represented by point A on Figure 3-10, foms the basis for this statistical state =ent: There is a 99 per cent confidence that at least 99 5 per cent of the fuel rods in the core are in no jeopardy of experiencing a departure frem nucleate boiling (DUB) dur-ing continuous operation at the design overpower of 114 per cent. f Statistical results for the maximum design condition calcula-tion shown by Figure 3-10 may be summarized as fonovs in Table 3-12. Table 3-12 DNB Results - Maximum Design Condition (99% Confidence Level) Power, Possible Population Point 4 of 2.h52 MWt FAh DNB's Protected. % A n4 1.85 184 99 50 B 114 1 79 100 39 73 c 100 1.85 17 99 95 D 100 1 79 10 99 98 E n8 1 79 184 99 50
- h. Most Probable Design Condition Analysis Su=ca:/
The previous maximum design calculation indicates the total number of rods that are in jeopa:dy when it is conservatively assumed that every red in the core has the mechanical and heat transfer characteristics of a hot channel as described in O. . *.
.. 3 2 3 2.2. For example, a u channels are analyzed with F (flow area factor) less than 1.0, Fq (heat input factor) Agreater 3-31 0000 160
than 1.0, and with =ini=um fuel asse=bly flow. It is physi-cally !=possible for all channels to have hot channel charac-teristics. A = ore realistic indication of the nu=ber of fuel rods in jeopardy =ay be obtained by the application of the statistical heat transfer data to average rod power and =echan-ical conditiens. An analysis for the =ost probable conditions has been =ade based on the average condhions described in 3 2 3 2.2. The results of this analysis are shown in Figure 3-11. The e.rd-ysis =ay be su==arized as fellows in Table 3-13 Table 3-13 DNB Results - Most Probable Condition Power, Possible Population Point % of 2,k52 MWt Fah DNB'd Protected. I F 100 1 79 2 99 994 G 114 1 79 32 99 913 H 118 1 79 70 99 815 The analysis was =ade frc= Point F at 100 per cent power to Point H at 118 per cent power to show the sensitivity of the / ardysis with pover. The vorst condition expected is indi-cated by Point G at 114 per cent power where it is shavn that there is a small possibility that 32 fuel rods =ay be subject to a departure frc= nucleate boiling (DNB). This result forms the basis for the following statistical state =ent for the = cat probabla design conditions: There is at least a 99 per cent confidence that at least 99 9 per cent of the rods in the core are in no jeopardy of experiencing a DNB, even with continuous operation at the design overpower of 114 per cent. l 1 1. Distribution of the Fraction of Fuel Rods Protected l The distribution of the fraction (P) of fuel rods that have been shown statistically to be in no jeopardy of a DNB has been calculated for the =axi=um design and =ost probable de-sign conditions. The ec=puter progra=s used provide an output of (N) nu=ber of rods and (P) fraction of rods that vill not experience a DNB grouped for ranges of (P). De results for the most probable design condition are shown in Figure 3-12. The population protected, (P), and the population in jeopardy,
,(1-P), are both plotted. Se integral of (1-P) and the number of fuel rods gives the nu=ber of rods that are in jeopardy for -
,given conditions as shown in Figures 3-10 and 3-11. The nu=ber lM
, 3-32
O d of rods is obtained from the product of the percenta6e tines the total number of rods being considered (36,816). The two distributions shown in Figure 3-12 are for the most probable condition analysis of Points F ard G on Figure 3-n. The lower line of Figure 3-12 shows P ard (1-P) at the 100 per cent power cordition represented by Point F of F1 6ure 3-11. The upper curve shows P and (1-P) at the 114 per cent power condition represented by Point G of Figure 3-n. The inte-gral cf N and (1-P) of the upper curve for=s the basis for the statistical statement at the most probable design cordi-tion described in paragraph h above. J. Hot Channel Perfomance Suz:m:ary The hottest unit een with an surfaces heated has been exam-ined for hot channel factors, DNB ratics, ard quality for a range of reactor powers. The cell has been examined for the
=axtnum value of Fah nuclear of 1.85 The hot channel was assumed to be located in a fuel assenbly with 95 per cent of the average fuel assembly flow. The heat generated in the fuel is 97 3 per cent of the total nuclear heat. The re=aining 2 7 per cent is assu=ed to be generated 1:1 the coolant as it pro-ceeds up the channel within the core and is reflected as an increase in A T of the coolant.
Error bands of 65 psi operating pressure ard 2 F are re-(c flected in the total core and hot channel ther=al margin cal-culations in the direction producin6 the lowest DNB ratios or highest qualities. The DNB ratio versus power is shown in Figure 3-13 The DNB ratio in the hot channel at the maximum overpower of H 4 per cent is 1 38 which corresponds to a 99 per cent confidence that at least 94.5 per cent of the fuel channels of this type are in no jeopardy of experiencing a DNB. The engineering hot channel factors corresponding to the above confidence-popula-tion relationship are desertbed in 3 2 3 2.2 and listed below: Fq= 1.008 Fqe. = 1.013 FA= 0 992 The hot channel exit quality for various powers is shown in F16ure 3-14. The combined results =ay be su==arined as follows: Reactor Power, 1, DNB Ratio (BAW-168) Exit quality, % 100 1.60 0 , , 107 5 (trip setting) 1.h7 2.6 n4 (maxi =um pover) 1 38 5.4 149 1.00 23 0 to: e 3 33 0000 !62
3.2.3.1.2 Fuel and Cladding Ther=al Ccaditiens
- a. Fuel A digital cc=puter code is used to calculate the fuel te=pera-ture. The progra= uses unifor= volumetric heat generation across the fuel diameter, and external coolant conditions and heat transfer coefficients determined for ther=al-hydraulic channel solutions. The fuel ther=al ccaductivity is varied in a radial direction as a function of the te=perature variation.
Values for fuel conductivity wre used as shown in Figure 3-15, a plot of fuel conductivity versus te=perature. The heat trans-fer frc= the fuel to the clad is calculated with a fuel and clad expansion model prcportional to te=peratures. The te=perature drop is calculated using gas conductiyity at the beginning-of , life conditions when the gas cenductivity is 0.1 Stu-ft/hr-F-ft'. The gas conduction model is used in the calculation until the fuel ther=al expansica relative to the clad closes the gap to a di=ension equivalent to a contact coefficient. The contact ccef-ficient is dependent upon pressure and gas conductivity. A plot of fuel center te=perature versus linear heat rate in 1 kv/ft is shevn in Figure 3-16 for beginning-of-life conditions. , The linear heat rate at the =axi=u= overpever of 11h per cent l is 19 9 kv/ft. 2ne correspending center fuel te=perature shown in Table 1-2 is h,k00 F. The center and average te=peratures at 100 per cent power are h,160 and 1,385 F respectively as shown in Table 3-1. / The peaking facters used in the calculation are: FAh = 1.85 F = 1.70 F ,, = 1.03 q , F (nue. and =ech.) = 3.2k l1 1 A conservative value of 1.03 was assu=ed for the heat flux peak-ing factor, Fq r. The assigned value corresponds to a 99 per cent confidence and 99 99 per cent population-protected rela-tionship as described in the statistical technique.
- b. Clad l The assumptiens in the preceding paragraph vere applied in the l calculation of the clad surface temperature at the =aximu= over-power. Boiling condit1 9ns and Lottes reldtionship\l9) prevail for the at the hot coolant spot, and
-to-clad theboil-aT for Jens ing was used to deter =ine the clad te=perature. "he resulting
=ax1=u= calculated clad te=perature is 65h F at a system oper-ating pressure of 2,185 psig. g e v 0000 163 cd4 3
, ., 3-3h (Revised 7-21-67) t *1 9\
O 3232 mmal and Hydraulic Evaluation 3 2 3 2.1 Introduction Summary results for the characteristics of the reactor design are pre-sented in 3 2 3 1. D e statistical core Design Technique employed in the design represents a refinement in the methods for evaluating pres-surized water reactors. Con esponding single hot. channel DNB data vere presented to relate the new method vit' previous criteria. A compre-4 hensive description of the new technique is included in this section to l pemit a rapid evaluation of the methods used.
- The BAW-168 correlation is a B&W design equation. An extensive review l
of data available in the field was undertaken to derive the correlation and to detemine the confidence, population, and DNP relationships in-cluded in this section. A comparison of the BAW-168 correlation with other correlations in use is also inclW,ed. 1 A detailed evaluation and sensitivity analysis of the design has been nade by examining the hottest chan' el in the reactor for DNB ratio, i quality, and fuel temperatures. BAW-168 DNB ratios have been compared with W-3 DNB ratios to facilitate a comparison of the design with PWR reactor core designs previously reviewed. j 3 2 3 2.2 statistical core Design Technique The core thermal design is based on a Statistical Core Design Technique developed by B&W. The technique offers many substantial improveme 2ts over older metheds, particularly in design approach, reliability of the result, and mathematical treatment of the calculation. The method re-flects the perfomance of the entire core in the resultant power rating
- and provides insight into the reliability of the calculation. This sec-
{ tion discusses the technique in order to provide an understanding of its i engineering merit. The statistical core design technique considers all parameters that af-feet the safe and reliable operation of the reactor core. By consider-ing each fuel rod the method rates the reactor on the basis of the per-fomance of the entire core. The result then vill provide a good mes-sure of the core safety and reliability since the method provides a sta-tistical statement for the total core. Bis statement also reflects the conservatism or design margin in the calculation. , A reactor safe operating power las always been detemined by the ability of the coolant to remove heat from the fuel material. The criterion , that best measures this ability is the DNB, whj.ch involves the individual parameters of heat flux, coolant te;nperature rise, and flow area, and their intereffects. The DNB criterion in connonly applied through the use of the departure from nucleate boiling ratio (DNBR). This is the ^ i minimum ratio of the DMB heat flux (as ccmiputed by the DNB correlation) to the surface heat flux. The ratio is a measure of the margin between the operating power and the power at which a DNB =1ght be expected to occur in that channel. The DNBR varies over the chanr.el length, and it
. !, is thg) minimum value of the ratio in the channel of interest that is used.
0000 19 ddI ir 2 3-35 - 1
I l The calculation of DNB heat flux involves the coolant enthalpy rise and O ecolant flev rate. The coolant enthalpy rise is a function of both the heat input and the flev rate. It is possible to separate these two ef-fects; the statistical hot channel factors required are a heat input facter, Fq, and a flow area factor, FA. In additica, a statistical heat flux factor, F a, q is required; the heat flux factor statistically de-scribes the variation in surfaco heat flux. The D:ER is most limiting when the burnout heat flux is based on mini =um flov area (scall FA ) and maximum heat input (large Fq), and when the surface heat flux is large (large Fqa). The DNB correlation is provided in a best-fit form, i.e. , a form that best fits all of the data on unich the correlation is based. To afford prctection against DIG, the DNB heat flux computed by the bect-fit correlation is divided by a DNB factor (B.F. ) greater than 1.0 to W ld the design DNB surface heat flux. The basic relationship Q" DNBR = B.F. x f(F A , F )x Q Q;urface xFo q involves as parameters statistical hot channel and DNB factors. The DNB factor (B.F. ) above is usually assigned a value of unity when r*- perting DNB ratios so that the margin at a given condition is shown
- directly by a DIGR 6reater than 1.0, i.e. ,138 in the hot channel.
To find the DNB correlation, selected correlations are cc= pared with DNB data obtained in the B&W burnout loop and with published data. The comparison is facilitated by preparing histogra=s of the ratio of the ' experimentally determined DNB heat flux ($g) to the calculated value of the burnout heat flux (9C). A typical histogram is shown in Figure 3-17 A histogram is obtained for each' DNB correlation considered. *he histo-grams indicate the ability of the correlations to describe the data. They indicate, qualitatively, the dispersion of the data about the =ean value--the smaller the dispersion, the better the correlation. Since ther=al and hydraulic data generally are vell represented with a Gaussian (normal) distribution (Figure 3-17), mathematical parameters that quantitatively rate the correlation can be easily obtained for the histegr e. These same mathematical parameters are the basis for the statistical burncut factor (B.F.). In analyzing a reactor core, the statistical infor=ation required to de-scribe the hot channel subfactors =ay be obtained from data en the as-built core, from data on similar cores that have been censtructed, or from the specified tolerances for the proposed core. Regardless of the source of data, the subfactors can be shown graphically (Figures 3-13 and 3-19). All the plots have the same characteristic shape whether they are for sub-facters, hot channel factors, or burnc tt factor. The factor increases with either increasing population or ecnfidence. The value used for the statis-tical het channel and burnout factor is a function of the percentage of confidence desired in the result, and the portion of all possibilities de-sired, as well as the amount of data used in determining the statistical factor. A frequently used assumptien in statistical analyses is that the
.a 3-:e 0000 165
data available represent an infinite sample of that data. De i= plica-tiens of this assumption should be notpd. For instance, if li=ited data are available, such an assu=ption leads to a sc=ewhat opti=istic result. Se assumption also i= plies that = ore infor=ation exists for a given sample than is indicated by the data; it i= plies 100 per cent confi:ience in the end result. The Ba# calculational procedure does not =ake this assumption, but rather uses the specified sample sice to yield a result that is much more =eanin6ful and statistically rigorous. 'Ihe influence of the amount of data for instance can be illustrated easily as follows: Consider the heat flux factor which has the fo= F,=1*Kr q F. y where Fq,. is the statistical hot channel factor for heat flux K is a statistical =ultiplying facter ry g,, is the standard deviation of the heat flux fac-tor, including the effects of al3 the subfactors if r pg. = 0.03 for 300 data points, then a K fact.or of 2.608 is re-quired to protect 99 per cent of the population. The value of the hot channel factor then is [ Fq,. = 1 + (2.608 x 0.050) = 1.1304 and vill provide 99 per cent confidence for the calculation. If, in-stead of using the 300 data peints, it is assu=ed that the data repre-sent an infinite sample, then the K facter for 99 per cent of the popu-lation is 2 326. The value of the hot channel fseter in this case is Fq,, = 1 + (2 326 x 0.050) = 1.1163 whichi= plies used 100percentconfidenceinthecalcu}20) ation. The values of the K factor above are taken frem SCR-607.g The sace basic techniques can be used to handle any situation involvinc variable confi-dence, population, and number of points. Having established statistical het channel factors and statistical DIB factors, we can proceed with the calculation in the classical =anner. The statistical factors are used to deter =ine the =in1=u= fraction of rods protected, or that are in no jeopardy of experiencing a DIS at each nuclear power peakin6 factor. Sines this fraction is kncun, the maxi-
=um fraction in jeopardy is also kncvn. It should be recccniced that every rod in the core has an asscciative DIG ratio that is substantially Greater than 1.0, 'even at the design overpcVer, and that theoretically no red can have a statistical population factor of 100 per cent, no =at-l ter hev large its DIG ratio.
l Since both the fraction of rods in jeopardy at any particular nuclear acv a hac,er peaking facter and the m=:ber cf reds operating at that peskingtt:r are knc e . i LO! b' i' ' 3-37 0000 lg i i
can be obtained by simple su=ation. The calculation is =ade as a fune-tien of pcVer, and the plet of reds in jeopardy versus reacter over;cver is obtained (Figure 3 20). The su==atien of the fracticn of reds in Jecpardy at each peakinc facter su==ed over all peaking factors can be
=cde in a statistically riscrcus =anner caly if the ccnfidence for all populations is identical. If an infinite sample is not assu=ed, the confidence varies with populatica. To fer= this su==ation then, a cen-servative assu=ption is required. B&W's total core =cdel assu=es that the confidence for all reds is equal to that for the least-protected red, i.e., the =in1=u= pessible confidence facter is associated with the entire calculation.
The result of the foregoing technique, based en the =cx1=u= design een-ditions (111+ per cent pover), is this statistical state =ent: There is at least a 99 per cent confidence that at least 99 5 per cent ef the rods in the core are in no jeopardy of experiencing a DNB, even with centinuous operation at the desicn overpcVer.
"he =aximu= des 16n conditions are represented by these asst =ptions:
- a. The =ax1=u= design values cf Fab (nuclear =ax/ avg total fuel rod heat input) are obtained by increasing the =axi=u= calcu.-
lated value of Fa h by a factor of 1.033 to provide additional design =argin.
- b. The maxi =u= value for F, (nuclear =ax/av6 axial fuel red heat 1 input) is dete=ined for the limiting transient er steady-state condition.
- c. Every coolant channel in the core is assr=ed to .tave less than the nc=inal flew area represented by ergineeri 6 hot channel area factors, F g, less than 1.0.
- d. Every channel is assu=ed to receive the =ini=u= flow associ-ated with core flev =aldistribution.
- e. Every fuel rod in the core is assu=ed to have a heat input greater than the =axi=u= calculated value. This value is represented by en61 neering hot channel heat input facters, Fq and Fq , , which are greater than 1.0.
- f. Every channel and associated fael rod has a heat transfer
=argin above the experi= ental best " ' '*-d*s reflected in DNB ratics $reater than 1.0 at =sxi=u= cverpover conditions.
The statistical core design technique =ay al w be used in a si=ilar
=anner to evaluate the entire core at the =c t probable =echanical and nuclear conditions to give an indication of the =cs probable degree cf fuel ele =ent Jeopardy. The result of the technique based on the =cs:
probable design conditiens leads to a statistical state =ent which is a corollary to the max 1=u= design state =ent: 1 I (g
,. 3-38 (Revised 7-21-67) 0000 167
There is at least a 99 per cent confidence that at least 99 9 per cent of the rods in the cere are in no jeopardy of experiencing a DNB, even with continuous operation at the design overpower. 2 e most probable design conditions are assumed to be the same as the maxbum design conditions with these exceptions:
- a. Every coolant channel is assumed to have the nemical flow area (FA = 1.0) .
- b. Every fuel rod is assumed to have (1) the maximum calculated value of heat input, and (2) Fq and Fqa are assigned values of 1.0.
- c. Se flow in each coolant channel is based on core flow and pcVer distributions,
- d. Every fuel red is assumed to have a ncminal value for Fah nuclear.
The full meaning of the maximum and most probable design statements re-quires additional comment. As to the 0 3 per cent or 0.1 per cent of the rods not included in the statements, statistically, it can be said that no more than 0 5 per cent or 0.1 per cent cf the rods vill be in jeopardy, ani that in general-the number in jeopardy will be fever than [ r~g 0 5 per cent or 0.1 per cent. Re statements do not =ean to specify a U given number of DNB'c, but only acknowledge the possibility that a given number could occur for the conditions assumed. In summary, the calculational procedure outlined here represents a sub-stantially improved design technique in two vays:
- a. It reflects the perfomance and safety of the entire core in the resultant power rating by considering the effect of each rod on the power rating.
- b. It provides information on the relikbility of the calculation and, therefore, the core through the statistical statement.
32323 correlation of Heat Transfer Data The BAW-168 report (Ref.17) serves as a reference for the "best-fit" fem of the design relationship used by B&W. B is heat transfer corre-lation has been found to be the most satisfactory in the representation of both unifom and nonunifom heat flux test data. The BAW-168 correls-tion is used by comparing the integrated average heat flux along a fuel red to a DNB hest flux 11=1 predicted by the correlation. For unifom heat flux the integrated average heat flux is equal to the local heat flux. The ecmparison is carried out over the entire channel length. Re point at which the ratio of the DNB heat flux to the integrated average heat flux is a minina is selected as the DUB point, and that O value of the ratio at that point is the DNB ratio (DNBR) for that chan-U nel.
. .c. a i . *
- 0000 168 3-39
I l 1 This particular discussicn deals with the ec=parison of DUB data to three particular correlations. The* correlation in the case of EAW-168,(co-relations selected were the E&W 17)acorelationwithwhichthe industry is fa=iliar in the case of WAFD-188,121 and a correlation re-cently proposed "or)use in the design of pressurized water reacters in thecaseo*W-3.122 The data considered for the purpose of these ec=parisons were taken frc= the following sources:
- c. WAFD-id8 (Ref. 21).
- b. AEEW-R213 (Ref. 23).
- c. Colu=bia University Data (Ref. 24, 25, and 26).
- d. Argonne National Laboratory Data, A'L (Ref. 27).
- e. The Babcock & Wilecx Cc=pany Data, 3&W (Ref. 28).
- f. The Babcock & Wilecx Cc=pany Euratc= Data (Ref. 29).
The cc=parison of data to the EAW-168 correlatica is presented as histo-grams of the ratio of the experimental DNB heat flux ($3 ) to the calcu-lated heat flux (9C). The data frc= each source vere grouped by pres-sure and m 1yzed as a group; batches were then prepared including co=-
=en pressure groups frc= all sources. Altogether there are 41 different j data groups and batches censidered. Histogrs=s for only the BAW-168 correlation are presented to =in1=1:e the graphical =aterial. The in-fe mation required for the generation of histogra=s of the other two correlations was also prepared.
The cc=parison of the varicus ecrrelations to each other is facilitated through the use of tabulations of pertinent statistical pars =eters. The standard deviation and =ean value were obtained frc= the cc=puted values of ($ 7/6C) for each group or batch. A cc=parisen of staMa"d deviations is sc=ewhat indicative of the' ability of the correlation to represent the data. Ecwever, differences in =ean values frc= group to group and correlation to correlation tend to ec=plicate this type ec=parison. A relatively si=ple =ethod =ay be used to ec= pare the correlations for various data; this =ethod uses the coefficient of variation (Ref. 30) which is the ratio of the standard deviation (c) to the mean I. The coefficient of variation =ay be thought of as the standard deviatica given in per cent; l it essentially no=alizes the various cc 6ed deviations to a ec==on I mean value of 1.0. I Table 3-14 is a tabulation of ~.M q.. purce, heat flux type, a=d cer-l responding histogra: nu=bers. Ne h a,,e.p a=s are shown on Figures 3-21 through 3-36.
/
e h d ! d,' ' 0000 169 3 ho
O Table 3-14 Heat Transfer Test Data Histogram Figure Source Heat Flux Type Nu=ber Nu=ber WAPD-188 Uniform 1-9 3-21 3-22 3-23 AEEW-R-23 Uniform 10-14 3-23 3-24 3-25 Colu=bia Unifom 15-19 3-25 3-26 3-27 ANL Uniform 20 3-27 B&W Uniform 21 3-28 ( B&W-Euratcm Unifom 22-24 3-28 3-29 Ccmbined Data (500-720 psia) Unifom 25 3-29 Ccmbined Data (1,000 psia) Uniform 26 3-30 ccmbined Data (1,500 psia) Unifom 27 3-31 Ccmbined Data (2,000 psia) Unifom 28 3-32 Ccubined Data (1,750-2,750 psia) Unifom 29 3-33 B&W Euratom Chopped Cosine Nonunifom 30-32 3-34
)
B&W-Euratom and B&W Inlet Peak Nonunifom 33-35 3-34 3-35 Euratem and B&W Cutlet Peak Nonunifom 36-38 3-35 Ccmbined Nonunifo m (1,000 psia) Nonuniform 39 3-36 Ccebined Nonunifem (1,500 psia) Nonunifom M 3-36 Combined Nonuniform (2,000 psia) Nont nifom 41 3-36 O C . a.
- e.
0000 mv g_uz
me histegra=s graphically de=enstrate the distribution of (?g/oC ) IC# each data group. The Gaussian type distribution of ($3 /6C ) aoout the
=can for the group is apparent in the large data groups. Sc=e data groups are too s"11 to provide =eaningful histogra=s, but they are pre-sented in order to ec=plete this survey.
The data vere used as presented in the source for the calculation of ($E/*C); no points were discarded for any reason. A good correlation should be capable of representin6 DNB data for a full range of all per-tinent para =eters. The result of the cc=parison on this basis is de=- onstrated in Table 3-15 me data source, pressure, histogra= rigure nu=ber, heat flux type, and number of data points in the grcup are tab-ulated. For each of the three correlations the folleving data are indi-cated: e/i The coefficient of variatien based on all available data in the group. na The nu=ber of data points rejected using Chauvenet's crite-rien (Ref. 31). This criterion is statistical in nature and is applied to the values of ($ /@C}* E "P *8 ** cutside certain 11=its with respect to the =ain body of data are rejected. (e/5)' The coefficient of variation based on the original data sa=-
^
ple less those points rejected by Chauvenet's criterion, i.e., based on n-ng values of ($E[4C)* j It is unfortunate that Chauvenet's criterion =ust be applied to the values of (@E/4 C ) rather than to the original data, since application to (&E!#C) leads to the rejection of points for either of two reasons:
- a. Bad data points,
- b. Inability of the correlation to represent a particular data point.
It is not desirable to reject points for the second reason, and yet cne
=ight expect to encounter sc=e bad data. me icgical choice then is to present data both ways, i.e., with and without Chauvenet's criterion ap-plied. Of the 41 groups and batches analv:ed the folleving is observed frc= Table 3-15:
Groups and Batches of Data Groups and Batches of Data With Smallest e/i Withcut With Smallest e/E With Correlation Chauvenet's Criterien Chauvenet's criterien BAW-168 38 36 WAPD-188 2 3 W-3 1 2 l 9; l l 1
+ 0000 171 3 h2
Chauvenet's criterion rejected the following nu=ber of points for each correlation: Uniform Nonuniform Total BAW-168 (croups only) 32 1 33 BAW-168 (Batches only) 39 o 39 WAPD-188 (orcups only) 34 2 36 WAPD-188 (Batches only) 33 o 33 W-3 (croups only) 59 12 71 W-3 (Batches only) 50 9 59 Several notable peculiarities exist in the tabulat' ion of Table 3-15 The Columbia data 500 psia group contained only five data points; four were rejected by Chauvenet's criterion, leaving one point. A standard deviation cannot be computed for one point; therefore all three values of (e/T )' are shown as not available (N.A.). Neither the BAW-168 nor the WAPD-188 predicted any negative DNB heat fluxes; the W-3 predicted 93 negative values for uniform data. 'Ite fact that only 59 vere re-jected for this correlation indicates that the remaining 34 uniform points which were negative (93-59 = 34) were close enough to the body of the data to be. considered statistically significant. Table 3-15 may be consolidated somewhat as belov by tabulating the number M groups and batches of data having coefficients of variation within a specified interval for each correlation. (e/I) Interval BAW-168 BAW-168:(a) WAPD-188 WAPD-188e(a) w.3, g.3 (a) Negative o o o o 2 o o-o.1 6 8 o o o 1 0.1-o.2 24 24 13 13 1 5 0.2-0 3 8 8 7 8 3 1 0 3-o.4 1 o 3 4 1 2 0.4-o.5 1 o 5 7 5 6 0 5-o.6 o o 6 5 3 4 0.6-o.7 o o 3 2 1 1 0 7-o.8 o o 2 1 7 8 0.8-o.9 1 o o o 1 5 0 9-1.0 o o o o 1 o oretter than 1.o j o ,2_ j 16 _7, Total kl ho 41 ho 41 ho (" Chauvenet's criterion applied. - O bs g ,' 0000 1/2
!~
3-43
. *
- q} '
W 7-Table 3-15
- Comparison of Heat Transfer 'Ibst Data
- MAW-168 WAmtm W.3 mietogram seat rima punter of
- nource Pr.eeure su. t pe Det rotata eli *a deliP eli % (etir e li S le lin' WAFD-188 500 1 tharorm 57 0.2n92 o -- 0.74018 0 --
1.6785 2 1 7483 WAPD-188 600 2 thitura AM 0.24525 1 0.23373 0.605M 5 0. % o11 o.8skar 3 c.81663 WA m 188 1000 3 ths torm 164 0.n351 1 0.M755 0.53793 4 0 50000 0 75 % 7 0 -- WAFD-188 1500 4 thiform M o.13390 3 0.10537 c.30489 0 -- 0.44994 0 -- WAPD-188 1750 5 thirorm 30 0.076698 0 -- c.18176 o -- o.34816 0 -- WAm188 2000 6 thitara 371 o.13529 4 0.12480 0.23113 6 0.20482 5.1495 7 0.79051 WAFD-188 225o 7 thiform 9 0.C61572 0 -- 0.1 % 13 0 -- 0.174 % 0 -- WAPD-188 2500 8 thifom 9 0.081763 0 -- o.16477 o -- 0.23851 1 0.19 404 WAPD-188 2750 9 Uniform 9 0.057343 0 -- 0.11820 o -- c.241n 0 -- AEEw-a213 %O 10 tm.aform 148 o.M674 3 0.2370s 0.617b4 0 -- 2.9296 1 1.4097 A m -a213 720 11 thstorm 33 0.1058 0 -- 0.50684 2 0.43312 2.3964 3 1.3510 A m -a213 1000 12 thifore 322 0.2o439 3 0.19 % 6 0 50 % 1 o -- 6.37M 3 1.4589 Am.n213 1200 13 ths tom a 18 0.15915 0 -- 0.42712 0 - 0.5Ec0 0 -- A m -m213 1500 14 tharorm 104 0.129 % 2 0.0:9659 0.28924 1 0.W0% 0.28314 15 0.090e29 Colunt.1. 500 15 thiform 5 0.13704 4 E.A. o.12752 4 M.A. o.91541 4 N.A. Coluable 720 16 thiform 29 0.16300 0 -- 0.51437 0 -- 0.58437 0 -- u Columbia 1000 1} tht form 281 0.80M8 6 0.18678 12.009 6 0.43991 0.45519 0 -- e Columbia 1200 lo thiform 15 0.12211 0 - o.29242 0 -- o.M815 0 -- Coluabla 1500 19 thiform SO 0.21043 3 0.12241 0.6W65 3 o.240c9 1.5097 3 0.11183 ANI. 2tX4 20 thiform 232 0.10271 2 4.o92803 o.19348 2 0.17973 3.6745 14 0 52 % o Bau 2000 21 thifum 21 0.058701 0 -- 0.1%%7 1 0.11792 24.400 3 1.1838 Euratom 1000 22 thiform 18 0.13104 0 -- 0.47611 0 -- o.77404 0 -- turatom 1200 23 thstorm 18 0.094606 0 -- 0 30104 0 -- 0.47690 0 -- turatom 2000 24 ths torm 14 0.12106 0 -- 0.1 % 50 0 -- 1.6369 o -- Co. Linea SOA720 25 unstorm 418 0.31215 5 0.16780 0.721m to 0.65124 2.70M 2 1.40>2 ConL:ned 1000 26 thtrors 785 0.47694 9 0.24909 17.8 % 8 0. % 791 4.1325 3 0.8t632 Co.Lamot 1500 W thitom 144 0.1 % 31 4 0.14211 0.57512 3 0.31718 1.2237 3 0.317 % Comt.smed 2000 28 thirore 638 0.14W6 4 0.14251 0.24186 8 0.219fE 5.2840 21 0.81797 Combiard 1750-2750 29 thstore 695 0.18236 17 0.14913 0.24M3 4 0.23nt 5 1401 21 0.8128 Eureton 1000 30 Ctes, ped Costne 1% o.17017 0 -- 0.4818T o -- 0.72772 0 -- Eureton 1500 31 Clopped Costae 13 0.2122 0 -- 0.24251 0 -- 0.66671 0 -- Euratom 2000 32 Chos, ped Costne 13 0.1 % 52 0 -- 0.19N1 o --
-5 7922 3 0.16ces saw a rureton 1000 33 Ia1.t hak 16 0.19n3 0 --
0.bn85 0 -- o.70580 0 -- C ma.r a ruraten 1500 2000
% Inlet Peak 12 0.1%n 1 0.10To3 0.18121 0 o
0 72369 0 -- Baw & suratum 35 Inlet hak 32 0.13755 0 e -- 0.17637 -- 4.4474 5 0.81144 Q baw a Eureton Baw a rureton tono 1500 37 Outlet hah outlet hak 12 16 0.23023 0.16799 o 0 0.55501 0 30115 2 0 0.1M% 0.74323 0.4%o9 0 0 O Baw & suratom 2000 38 outlet hak 36 0.13481 0 - 0.16799 0 -- 1.o478 4 0.loe33 Combined 1000 39 mon-ths torm 42 o.20k45 0 -- o.by6% 0 -- 0.7toA2 0 -- Combined 1500 40 man-thi form 41 0.17435 0 -- 0.253M3 0 -- 0.58846 0 -- , N Contined 2000 41 Non-thirom 81 0.17fA6 0 -- t,.17621 0 -- 9.6963 9 0.46805 O O - O O
As is seen frem the tabulation the celu=n for 3AW-168 with Chauvenet's criterion applied indicates a grouping of 0.1 to 0.2, and a =ax1='u:: value of 0.28780 is noted frem Table 3-15 For WAPD-lS8 the spread is greater with a maxi =um value of 0.Th018. For W-3 the spread is still greater, and a maxi =um value of 1 7483 is noted. ne negative values of DNB heat flux predicted by the W-3 correlation are in part respon-sible for the large spread in (e/ I). The ability of the BAW-168 correlation to fit both uniform and nonuni-form heat flux data over a vide range of pertinent variables leads us to believe that it is the best DNB correlation available. 3 2 3 2.4 Evaluation of the The. al and Eydraulic resign
- a. Hot Channel Goolant Quality and Void Fraction An evaluation of the hot channel coolant conditions provides additional confidence in the the. al design. Sufficient coolant flow has been provided to insure low quality and void fractions. The quality in the hot channel versus reactor power is shown in Figure 3-37 D e sensitivity of channel cuclet quality with pressure and power level is shown by the 2,185 and 2,120 psig system pressure conditions examined.
These calculations were =ade for an FA h of 1.85 Additional calculations for a 10 per cent increase in Fah to 2.035 vere
=ade at 114 per cent pcVer. The significant results of both calculations are su==arized in Table 3-16. The effects of
'\ using an F ah of 1 79 are shewn in Figure 3-37 Table 3-16 Hot Channel Coolant Conditiens Ecit Exit Void Operating Pcver. 5 FnH Quality, 4 Fraction. 5 Pressure, psig 100 1.85 (-)2.k(b) g,5(a) 2,1115 11L 1.35 2.8 13 5 2,135 130 1.85 9.h 36 9 2, ~.85 114 2.035 8.7 2,185 100 1.85 C 35 3.80(a) 2,120 114 1.85 5.4 25 2 2,120 130 1.85 12.1 45 2 2,120 lik 2.035 11 3 L3.k 2,120 (8)Subecoled volds.
.(b) Negative indication of quality denotes subecoling of 10.2 Stu/lb.
O , bl .
..g 9 .
/\l
- f. . s
~
5 - '5 0,.000 llf
The conditions of Table 3-16 vere deter =ined with all of the hot channel factors applied. Additional calculations were =ade for unit cell channels without engineering het channel factc'rs to show the ecolant conditions = ore likely to cccur in the reactor core. Values for F Ah of 17) and 1.85 were exn=ined with and without fuel asse=bly flev distribution hot channel facters at 2,135 psig as shown on Figure 3-38. These results show that th- exit qual-ities frc= the hottest cells shculd in general be considerably lover than the =ax!=u= design conditions.
- b. Core Void Fraction The core void fractions were calculated at 100 per cent rated power l1 for the nor=al operating pressure of 2,185 psig and for the =ini=u=
operating pressure of 2,120 psig. The influence of core fuel as-se=bly flow distribution was checked by deter =ining the total voids for both 100 and 95 per cent total core flev for the two pressure conditions. The results are as folicvs: Flav. 4 Pressure, psig Ocre Void Fraction. 5 100 2,185 0.007 100 2,120 0.033 95 2,185 0.041 95 2,120 0.127 The = cst conservative condition of 95 per cent flow at 2,120 psig results in no = ore than 0.13 per cent void volu=e in the core. Conservative =ax1=u= design values for Fah nuclear described by Line A of Figure 3-9 vere used to =ake the calculation. The void progra= uses a ec=bination of Ecuring's (31-1) =cdel with 1 Zuber's (31-2) correlation between vo'd fraction and quality. The Bowring =odel considers three different regions cf forced convec-tien boiling. They are: (1) Highly Subecoled Boiling In this regica the bubbles adhere to the vall while =oving up-vard through the channel. This region is ter=inated when the subcooling decreases to a point where the bubbles break
- through the la=inar sublayer and depart frc= the surface. The highly subcooled region starts when the surface te=perature of l the fuel reaches the surface te=perature predicted by the Jens l and Lettes equation. The highly subecoled region ends 'ahen T
bulk = eV
~ '
l sat (A) i . O
, ,p 3-k6 (Revised 7-21-67) -
0000 175
- f. \ \ \; !\
6 = local heat flux, Btu /hr-ft2 ri : 1.863 x 105 (14 r o.cc62p)
, V = velocity of coolant, ft/see p = pressure, psia The void fraction in this region is co=puted in the same =anner as Maurer,(31-3) except that the end of the region is deter =ined by Equation (A) rather than by a vapor layer thickness. The nonequilibriu= quality at the end of the region is computed from the void fraction as follows:
1 x* : 1 (3)
,1+ r
- 1)
F8d g i where , xd = nonequilibrium quality at end of Region 1 a3 : void fraction at Tsat
- Ibulk
- y P, : liquid component density, lb/ft3 P : vapor component density, lb/ft3 (2) Slightly Subceoled Boiling In this region the bubbles depart from the vell and are trans-ported along the channel (condensation of the bubbles is ne-glected). This region transcends to point where the ther o-dynamic quality is zero. In general, this is the region of major concern in the design of pressurized water reactors.
The nonequilibrium quality in this region is computed from the
- following formula
x*sx*+ " P
! (6 - 6#p)d: (C) mhfg(1 + 8 ) *d vhere
- x = nonequilibrium quality in Region 2 h,g
* : latent heat of vaporization, Stu/lb 1
= fraction of the heat flux above the single 1+8 phase heat flux that actually goes to pro-ducing voids Q
(/ 6 3p : single phase heat flux, Stu/hr-ft2 0000 176
$\!
3-k6a (F.evised 7-21-67) i
m: = ass flow rate, lb/hr '1 ph = heated peri =eter, ft O z: channel distance, ft The void fraction in this region is cc=puted frc= a= * (D) 38 3 Af PE ~ 7gg, ( P, - PE) ^
- 1/'+
Cg x + pg /pf (1 - x*) + ,
~ ',
m Q . where g = acceleration due to gravity, ft/sec2 lb = ^~ g C : constant in Newton's Second Law : 32.17 lb f seed Co : Zuber's distribution parameter A : flow area, ft2 f e: surface tension Equation (D) results from rearranging equations found in Re-ference (31-2) and assu=ing bubbly turbulent flow in deter-mining the relative velocity between the vapor and the fluid. Zuber has shown that Equation (D) results in a better pre-diction of the void fraction than earlier models based on e=pirical slip ratios. (3) Bulk Boilina In this region the 'Julk te=perature is equal to the saturation te=perature, and a]l the energy transferred to the fluid results in net vapor genention. Bulk boiling begins when the ther=o-dynamic (heat balance) quality, x, is greater than the nonequi-librium quality, )*. The vcid fraction in this region is cc=- puted using Equat: en (D) with the themedynamic quality, x, re-placing x*.
- c. Coolant Channel Hydraulle Stability A flow reg 1=e =ap vas constructed to evaluate chcnnel hydraulie sta-bility. The transition frcm bubbly to annular flev at high = ass ve-locities was determined using Bakers's(32) correlatica, and the tran-sition from bubbly to slug flov which occurs at icv = ass velocities was determined with Rose's(33) correlation. The transition frc= slug i flow to annular flev vas detemined by Haberstrch's(3k) correlation.
( Bergies(35) found that these correlations, which vere developed from l adiabatic data, are adequate for locating flow regi=e transitions with heat addition, and that they adequately predict the effects of pressure. Figure 3-39 shows the flow regime =ap on which has been g e.. 4%; i V ' 3-46b (Revised 7-21-67) 0000 177
l
}
plotted a point representing operating conditions in the hot channel at 11h per cent overpower. To aid in assesJing the conservatism of the design, en additional point is plotted at 130 per cent overpower. Inspection shows that both points lie well within the bubbly flow regime. Since the bubbly flow regime is hydraulically stable, no flow instabilities should occur. This flow regime map was prepared for the hot unit cell 1 at the maximum design condition characteristics outlined in 3.2.3.1.1. The confidence in the design is based on both experimentat re-sults obtained in multiple rod bundle burnout tests and analyt-ical evaluations. Three additional flow regime maps were con-structed for nominal and postulated worst case conditions to show the sensitivity of the analysis with respect to mass flow rate, channel dimensions and mixing intensity in unit, corner, and vall-type cells. The results are shown in Figures 3-39-a, 3-39-b, and 3-39-c. The mass velocity and quality in each type of channel for the two cases are plotted on the figures. The conditions assumed for the nominal and postulated worst case are given in 3.2.3.2.k J. Data from the burnout tests performed by BW on a 9-rod bundle simulating the core geometry are also plotted on the maps. The open data points on the maps represent the exit conditions in the various type channels just previous to the burnout condition for a representative sample of the data points obtained at the O- design operating pressure of 2,200 psia. In all of the bundle tests the pressure drop, flow rate, and rod temperature traces were steady and did not exhibit any of the characteristics as-sociated vith flow instability. Inspection of these naps shown that the nominal conditions are far removed from unstable flow regimes. The evaluation also shows that under the vorst conditions that have been postulated the reactor vill be operating in the hydrodynamically stable, bubbly flow regime.
.s* ,
3 h6c (Revised 7-21-67) L
O V (DELETED)
- d. Hot Channel DNB Cemrarisons DNB ratios for the hottest channel have been deter =ined for the BAW-168 and W-3 correlations. The results are shown in Figure 3-k0. DNB ratios for both correlations are Luovn for the 1.50 axial max / avg symmetrical cosine flux shape from 100 to 150 per cent power. The BAW-168 DNB ratio at the =ax-imwn design power of llh per cent is 1.38; the ccrresponding W-3 value is 1.72. This co= pares with the suggested W-3 de-sign value cf 1.3. It is interesting to note that the calcu-lated DNB ratio reaches a value of 1.0 at about 150 per cent power with the BAW-168 equatien which' adequately describes DNB at the high quality condition of 20 per cent. The W-3 calculation is accurate to about 130 per cent power, but be-cause of quality limitations it cannot be used to examine the channel at the 150 per cent power condition.
The sensitivity of DTB ratio with Fan and F: nuclear was examined from 100 to lik per cent power. The detailed results are labeled in Figure 3-h0. A cosine flux shape with an F: of 1.80 and and Fah of 1.85 results in a W-3 DNB ratio of 1.h5 and a BAW-168 ratio of 1.33 The V-3 value is well above suggested design values, and the BAW-168 value of 1.33 corresponds to a het channel confidence of 99 per cent that
'O about 93 per cent of the population is in no jeopardy as shown la the Population-DNB ratio plot in 3.2 3 2.2, Statis-tical Core Design Technique.
The influence of a change in Fah was deter =ined by analyzing the hot channel for an Fab of 2.035 This value is 14 per cent above the mav4 mum calculated value of 1.79 and 10 per. cent above the maximum design value of 1.85 The resulting BAW-168 DNB ratio is 1.22, and the W-3 value is 1.26. Both of these values are well above the correlation best-fit values of 1.0 for the severe conditions assumed.
- e. Reactor F* *> Effects Another significant variable to be considered in the evalua-tion of the design is the total system flow. Conservative values for system and teactor pressure drop have been deter-
=ined to insure that the required system flow is obtained in the as-built plant. The experimental programs previously out-lined in Section 1 vill confirm the pressure drop and related pump head requirements. It is anticipated that the as-built reactor flow will exceed the design value and will lead to in-creased power capability.
An evaluation of reactor core flew and power capability was made by deter =ining the maxi =um steady state power rating - tm 0* 3-h7 (Revised 7-21-67) *0000 179
versus flow. The analysis was =ade by evalunting the het channel at the overpower conditions while =aintaining (a) a DIG ratio of 1 138 (BNJ-168), ard (b) the statistical core design criteria. The results of the cnalysis are shown in Figure 3 kl. The power shewn is the 100 per cent ratin6, and tne li=iting condition is 114 per cent of the rated power. An exa=ination of the slope of the curre indicates stable characteristics, and a 1 per cent change in flev changes the pcver capability by only about 1/2 per cent.
- f. . Reactor Inlet Te=perature Effects The influence cf reactor inlet te=perature on power capability at a given flow was evaluated in a similar =anner. A variation of 1 F in reactor inlet te=perature vill result in a power capability change of slightly less than 1/2 per cent.
- g. Fuel Te=cerature 1 A fuel te=perature and gas pressure ec=puter ccda was developed to calculate fuel te=peratures, expansion, densification, equiaxed and colu=nar grain growth, center piping of fuel pellets, fission gas release, and fission gas pressure. Progrs= and data ec=parisons j vere =ade on the basis of the fraction of the fuel dia=eter within these structural regions:
i (1) Outer li=it of equiaxed grain growth - 2,700 F. (2) Cuter l!.=1t of colu=nar grain growth - 3,200 F. @ (3) cuter li=it of =olten fuel (UO2) - 5,0c0 F. Data frc= References 36 through 39 vere used to ec= pare calculated and experi= ental fractions of the red in grain growth ccd central elting. t The radial expansion of the fuel peuet is cc=puted frc= the =can fuel te=perature and the average coefficient of linear expansion for the fuel over the te=perature range considered. This =odel cc=bined with the =odel for calculating the heat .-insfer coeffi-cient was co= pared with the =odel developed by Notley et al (kO) of AECL. The difference in fuel growth for the two calculation
=odels was less than the experi= ental scatter of data.
The fuel =ay be divided into as =any as 30 radial and 70 axial in- 1 cre=ents for the analysis. An iterative solution for the te=per-ature distribution is obtained, and the thermal conductivity of the fuel is input as a function of te=perature. The relative ther-
=al expansion of tne fuel and cladding is taken into account when deter =ining the te=perature drop across the gap between the fuel and cladding surfaces. The te=peratura drop across the gap is a
..' function of vidth, =ean te.perature, and gas conductivity. The conductivity of the gas in the gap is deter =ined as a function of burnup al.d subsequent release of fissica predt:t gases. In the G,, i 3 h8 (Revised 7-21-67) 0000 180
O V event of fuel clad centact, contact coefficients are deter =ined 1 on the basis of methods suggested by Ross and Stoute(hk). The ccatact coefficient is deter =ined as a function of the mean con-ductivity of the interface =aterials, the contact pressure, the mean surface roughness, the material hardness, end the conduc-tivity of the gas in the gap. The analytical model computes the amount of central void expected whenever the temperature approaches the threshold te=perature for fuel migration, and readjusts the density according to the new geometrf. The program uses a polyncmial fit relationship for fuel thermal conductivity. Three relationships were used to evaluate the effects of conductivity. A comparisen of these conductivity relationships with the reference design CVNA-lk2 (hl) is shown in Figure 3 h2. The values suggested in GEAP k62k (h2) and CVNA-2h6 (h3) are very similar up to 3,000 F, and the former values are more conservative above 3,000 F. McGrath (h3) cen-cludes that the CVNA-2k6 values are lever limits for the O d 1
- 1 l
l l 0 0 n.3 ;.a , 1 3 k8a (Revised 7-21-67)
() high te=perature conditions. Fuel center te=peratures for all three of the conductivity relationships at the peaking facters given in 3.2.3.1.2 have been calculated to evaluate the =argin to central melting at the =axt=us Overpever and to show the sensitivity of the calculatien with respect to ther=al conduc-tivity. Since the power peaks vill be burned cff with irradi-ation, the peaking factors used are censervative at end-of-life. (TElETED) The results of the analysis with the =ethods described above are shown in Figures 3 h3 and 3 kk fer beginning ar.d end-of-life conditions. The beginning and end-of-life gas conductivity values are 0.1 and 0.01 Stu/hr-ft2 -F respectively. The calcu-lated end-of-life center fuel temperatures are higher then the beginning-of-life values because of the reduction in the con-ductivity of the gas in the gsp. The effect is apparent even though a contact condition prevails. The calculatica does not 1 include the effects of fuel swelling due to irradiation. The calculated contact pressures are conservatively lever than those expected at end-of-life ccnditions in the hottest fuel reds,
~
and the fuel te=peratures shcvn in the above figures are cen-servatively higher.
, () The B&W :odel gives very good rasults when ec= pared to the re- \- '
suits of others in the field as is shown in Figure 3 hk. In the linear heat range of most interest, i.e., approximately 20 kw/ft, there is only about 300 ? difference between the =axi- ;
=u= and mini =u= values calculated. Also the s=all differences between the 3&W curve and the other curves indicate the relative insensitivity of the results te the shape of the conductivity at the elevated te=peratures.
The = cst conservative asst =ptiens, using GEAP-k62L data with relatively little increase in ther=al cendc:tivity above 3,000 F, result in central fuel =elting at about 22 kv/ft, which is 2 kv/ft higher than the =axi=u= design value of 19 9 kv/ft at 114 per cent pcwer. Further evaluatien of the two figures shows that central fuel =elting is predictel to occur between 22 and 26 kv/ft depending en the ti=e-in-Jlfe and conductivity assu=ptiens. The transient analyses at accident rad normal conditions have 1 been made using the GEAP L62L fue'. thermal conductivity curve to reflect a conservative value for the =ax1=u= average te=- perature and stcred energy in the fuel. Use of this curve re-l sults in a higher te=perature and therefore a lever Deppler coefficient, since it decreases with te=perature. Thus the re-sultant Doppler effect is also censervative. l (~% l N.Y s ;, - 3 L9 'Rev11ed 7-21-67)
*000J, 182 l .
l l
- h. Fission Gas Release The fission gas release is based on results recorted in CF.AP-4596.(k5) Additional data frc= GF.AP h31h ,(h6)' AICL-603,( h7 ) and CF-60-12-lk(40) have been ec= pared with the suggested release rate curve. The release rate curve (h5) is representative of the upper li=it of release data in the temperature region of most i=portance. A design release rate of h3 per cent and an inter- 1 nal gas pressure of 3,300 psi are used to determine the fuel clad internal design conditions reported in 3.2.h.2, Fuel As-se=blies.
The design values for fission gas release from the fuel and for the =aximum clad internal pressure were determined by analyzing various operating conditions and assigning suitable margins for possible increases in local or average burnup in the fuel. Ade-quate =argins are provided without utilizing the initial porosity voids present in the UO2 fuel. A detailed analysis of the de-sign assumptions for fission gas release, and the relationship of burnup, fuel growth, and initial diametral clearance between the fuel and clad, are su==arized in the following paragraphs. An evaluation of the effect of having the fuel pellet internal voids available as gas holders is also included. (1) Design Acsu=ntions (a) Fission Gas Release Rates The fission gas release rate is calculated as a func-tion of fuel temperature at the design overpover of lik per cent. The procedures for calculating fuel te.7eratures are discussed in 3.2.3.2.h g. The fis-sion gas release curve and the supporting data are shown in Figure 3 kh-a. Most of the data is on or below the design release rate curve. A release rate of 51 pef cent is used for the portion of the fuel above 3,500 F. The fuel temperatures were calculated using the GF.:AP h62h fuel ther=al cenductivity curve to obtain conservatively high values for fuel te=per-atures. (b) Axial Power and Burnun Assu=ntions The t3=perature conditions in the fuel are determined for :e =ost severe axial power peaking expected to occur. Two axial power shapes have been evaluated to deter =ine the maximum release rates. These are 1 50 and 1.70 =ax/ avg shapes as shown in Figure 3-8 and re-peated as part of Figure 3 kh-b of this analysis. The quantity of gas released is found by applying the tem-perature-related release rates to the quantities of fission gas produced along the length of the hot fuel red. i l s a , 3-50 (Revised 7-21-67) 0000 183 Je . .M . ,
O v The quantity of fission gas produced in a given axial 1 1ccation is obtained from reactor cora axial region burnup studies. Three curves showing the axial dis-tribution of burnup as a local to average ratio along the fuel rod are shown in Figure 3-kh-b. Values of 100, 300, and 930 days of operation are shown. The 930-day, or end-of-life condition, is the condi-tion with the maxi =um fission gas inventory. The average burnup at the end of life in the hot fuel rod is 38,150 MWD /MTU vhich has been deter::dned as follows: Calculated Hot Bundle Averwge Burnup, MWD /MTU 33,000 Hot Fuel Rod Burnup Factor 1.05 Margin for Calculation Accuracy 1.10 Hot Red Maximum Average Burnup, MWD /MTU 38,150 The local burnup along the length of the fuel rod is the product of the hot rod marimum average value above and the local to average ratio shown in Figure 3 hk-b. The resulting hot rod local maximu= burnup for the O 93o-a F. e=*-et-tire c==41:1e= 1 se== '2 ooo Mwo/ MTU. This is the marimum calculated value. However, local values to 55,000 MWD /MTU have been evaluated to insure adequate local fuel cladding 3-50a (Revised 7-21-67)
strength for possible increases in average or local burnup over the life of the fuel for various fuel =anage=ent pro-1 & T cedures. (c) Hot Rod Power Assu=ptions The maximum hot rod total power occuring at any time in the life of the fuel has been used to calculate the overpower temperature conditions. A hot rod power of 1.85 times the average rod power has been applied. This results in a =sx-imum linear heat rate of 19 9 kv/ft which corresponds to 114 per cent of the maximum linear heat (17 49) shown in Table 3-1. This is a conservative assumption when coupled with the end-of-life fission gas inventory since burdle ard individual fuel rod power is expected to decrease with fuel burnup. A study of the power histories of all of the fuel assemblies to equilibrium conditions shows that the powers in the bundles during the last 300 days of operation are not more than 1 3 ti=es the average bundle power. The peak bundle ratio of 1.69 (1.85 + hot rod ratio) vill only occur during the first tvo fuel cycles when the fission gas in-ventory is less than the =axi=um value. (d) Fuel Growth Assumptions The fuel growth was calculated as a function of burnup as irdicated in 3 2.4.2.1. Fuel pellet di=ensicas in the ther-mal temperature and gas release = cele were increased to the , end-of-life conditions as determined above. (e) Gas Conductivity and Contact Heat Transfer Assu=ptions The quantity of fission gas released is a function of fuel temperature. The temperatures are Lafluenced by three factors: (c) the conductivity of the fission gas in the gap between the fuel and clad, (b) the dia=etral clearance between fuel and clad, an1 (c) the heat transfer conditions when the fuel expands enough to contact the clad. A gas conductivity of 0.01 Stu/hr-ft -F2 based on 43 per cent release of fission gas at the end-of-life condition was used in the analysis. Diametral clearances of 3.C025 to 0.0075 in. reflecting minimum and =aximum clearances after fue'. growth were analyzed. The contact heat transfer coefficients were calculated as suggested in Reference LL. (2) Summary of Results The fission gas release rates were deter =ined in the first eve.1-untion. Rates were found for various cold dia=etral clearances and axial power peaking and burnup shapes. The results are shown in Figure 3-kh-c. The icvest curve is the expected condition for a 1.70 axial power shape with a 930-day axial burnup distributica as shevn in Figure 3-kh-b. The incresse in release rate with y}\ \; 3-50b (Revised 7 d 0000 185
.e AIb
diametral clearance results from the fact that the fuel te=perc- 1 ture =ust be raised to higher values before contact with the fuel clad is =sdc. The release rate at the =inimu= clearance of ~.0C05 ; in. is 19 per cent. This is the condition that produces the =cx-i=um clad stress due to fuel Growth with irradiation. The asse=- bly of =ax1=u= size pellets with =ini=u= internal diameter clad-ding will produce this condition after fuel Browth. In the event a few hot pellets have the max 1=um dia=cter and the re=ainder have the minimum dia=cter, then the average cold gap would be 0.0033 in. producing a slightly larger release rate. The re-lease rate of 33 per cent for the =axi=um dic=etral clearance will not occur with the maximum stress condition due to fuel growth, since the fuel can grow into the cicarance. Two additional cases were examined to chec% the sensitivity of the calculations to axial power and burnup shapes. The results are shown by the upper two curves in Figure 3 kk-c. The top curve is a plot of the release rates when *
- i- -~~"~ed that both the axial power and burnup inventory of fission gas are distributed with a 170 =cx/ avg ratio as shown on Figure 3 kk-b. Si=ilar re-sults are shown for the 1 30 =ax/ avg ratio. These curves show 2
the release rates expected are not strongly influenced by the various pcuer and burnup shcpes. The second evaluation shows the resulting internal pressures due to the release of fission product gases. Plots of pressures for I the expected 930-day cxial burnup distribution cnd a 170 =ax/ av6 axial power shcpe cre shown in Figure 3 kh-d. The 1caer curve is a plot of internal gas pressure with open pores (3 per cent of the fuel volu=e is available to hold the released gas). The upper data band is for a closed pore conditien with all released das contained outside the fuel pellets in spcces between the ex-panded " dished ends of the pellets, the radial g:ps (if any), and the void spaces at the ends of the fuel rods. ".he band of date shown reflects the effect of fuel densification and grain growth described in 3 2 3 2.L. The upper li=it is for an ideal ther=al model without grain growth or densification; the lower li=its are for the design =cdel. The calculation of the =axi=u= pres-sure is also relatively insensitive to the cxial burnup distri-bution as shown by the dashed line in Ficure 3-kk-d for a 130 maximum to average axial power and burnup shape. (This ucrre-sponds to a local burnup peak of 37,000 K4Df!3.) The allowable desi6 n internal pressure of 3,300 psi is well above the maxi =um value of' internal pressures cciculated for open or cle. M pellet peres, and the maxi =u= internal pressure should oni; occur with the =axi=u= diametral elecrance condition. A
=cdest increase in average fuel burnup ccn be tolerated within the prescribed internal pressure design li=its.
, It has been indicated in Reference LO and in AECL-1398 that the 7-s ' "
s
' ' p' h , CO2 fuel is plastic enough to ficw under low stresses when the I
( tempereture is cbove 1,300 F. That fraction of the fuel belov l . , Vnl e I
. . . (,. ,
, 3-50c (Revised T-21-67) nnnn .n UUUV -l5 i I L- _ _ .
this temperature =ay retain a large portica of the original po- 1 rosity and act as a fissicn gas holder. The hottest axial loca-tions producing the hignest clad stresses will have little if any fuel below 1,800 F. EcVever, the ends of the fuel rods will have acce fuel belev this te=perature. The apprcxi= ate fraction of the fuel belov 1,800 F at overpcVer for a 170 axial power shape is as follows for varicus cold dia=etral clearances. C1,earance , Per Cent of Fuel in. Belev 1.800 F, 4 0.0025 ho 0.005 20 0.0075 5 The retention of fuel porosity in the icv te=perature and lov burnup regions vill result in =odest reductions in internal gas pressure.
- 1. Hot Channel Factors Evaluation (1) Rod Pitch and Boving A flow area reduction facter is deter =ined for the as-built fuel asse=bly by taking channel flow area =easure=ents and ststisti-cally determining cn equivalent het channel flow area reduction i factor. A fuel case =bly hcs been =easured with the results shown in Table 3-11. In the analytical solution for a channel flow, each channel flow area is reduced over its entire length by the FA fabtor shown in Figure 3-18 for 99 per cent confidence.
With a 99 per cent confidence and 94 5 per cent population re-lationship described in 3 2 31.1 for the het channel, the arec reduction facter is 0 992. The approxi= ate li=it of this facter is cbtained by exa=inin6 the value in Figure 3-18 as the popula-tion protected approaches 100 per cent. FA at 99 99 per cent cf the population protected is 0 963 The hot channel value is shown in Table 3-1. Special attention is given to the influence of water gap varia-tion between fuel asse=blies when deter:ining rod powers. Nu-clear analyses have been =ade for the nc=inal and =axi=u= spac-in6 between adjacent fuel asse=blies. The nc=inal a:ti =ax1=u= hot asse=bly fuel rod powers are shown in Figures 3-Lh-e and 3 kh-f respectively. The hot channel nuclear power factor (Fah nuclear) of 1.85 shown in 3.p.3 1.1 is based on Figure 3 kh-f for the =axi-
=u= vater gap between fuel asse=blies. The factor of 1.85 is a product of the hot asse=bly facter :f 1.69 ti=es the 1.096 het rod factor. This pcVer facter is a" signed to the hottest fuel i rod which is analyzed for burncut u tder unit cell, vall cell, l and corner cell flev conditicas.
ll> l 1 3-5cd (Revisec 7-21-c7) 0000 187 s., a 7 .. : , U g. . i , '. ! th.
(2) Fuel Pellet Dia=cter, Density. and Enrich =cnt Factors 1 Variations in the penet sic.e, density, and enrichment are re-flected in coefficients of variation numbers 2 through 7 of Table 3-11. These variations have been obtained frc= the =ea-sured or specified tolerances and ec=bined statistically as described in 3 2 3 2.2 to give a power factor on the hot red. For the het cucnnel confidence and population conditions, this factor, Fq, is 1.008 and is applied as a power increase over the full length of the hot fuel red. The 1ccal heat flux factor, Fqa, for 99 per cent confidence and 94 5 per cent population is 1.013 Rese het channel values are shown in Table 3-1. The correspoding values of Fq and Fga with 99 99 per cen popula-tion protw ed are 1.017 and 1.03 respectively. A conservative value of Fqa of 1.03 for 99 per cent eenridence and 99 99 per cent population is used for finding the caximu= fuel linear heat rates as shown in 3 2 3 1.2.
"'hese factors are used in the direct solution for channel en-thalpies and are not expressed as factors en enthalpy rise as is often done. The coefficients of variation vin be under con-tinuous review during the final design and development of the fuel asse=bly.
(3) Flow Distribution Effects Inlet Plenu= Effects The final inlet plenu= effects win be detemined from the 1/ 6 see.le =odel flow test new in progress. The initial runs indi-cate satisfactory flow distribution. Although the final nuclear analysis and flov test data =ay show that the het bundle posi-tions receive average or better flev, it has cecn assu=ed that the flew in the hot bundle position is 5 per cent less than aver-age bundle flow under isothe:::al conditions corresponding to the
=cdel ficv test conditions. An additional reduction of flow due to hot asse=bly power is described below.
Redistribution in Ad.jacent Channels of Dissimilar Coolant Conditions The hot fuel asse=bly flow is less than the flow through an aver-age asse=bly at the same core pressure drop because of the in-creased pressure drop associated with a higher enthalpy and qual-ity condition. Bis effect is allowed for by =aking a direct calculation for the hot asse=bly flow. We ec=bined effects of upper and lover plenum flow conditions and heat input to the hot asse=blies vill result in a hot asse=bly flow of about 85 to 95 per cent of the average asse=bly flow dependin,:; on the final ple-num effects and asse=bly pcVer peaks. W e verst ec=bination of i effects has been assumed in the initial design, and the hot as-A se=bly flow has been calculated to be about 85 per cent of the g avera6e asse=bly flow at H 4 per cent overpcVer. Actual hot as-se=bly flows are calculated rather than cpplying an equivalent i hot channel enthalpy rise facter. l (, 9 '. {'* ; s ' Q ? - 3-50e (Revised T-21-67) 0 % 0 188 l l t b
Physical Mixing of Coolant Between Channels 1 The ficv distribution within the hot assc=bly is calculated with O c =ixing code that allova an interchange Of heat between channels. Mixing coefficients have been deter =ined frc= cultired mixing tests. The fuel assc=bly, consisting-of a 15 x 15 array of fuel reds, is divided into unit, vall, and corner cells as shown by the heavy lines in Figure 3-kk-e. The =ixed enthalpy for every cell is deter =ined si=ultaneously so that the ratio of cell to average asse=bly enthalpy rise (Enthalpy Rice isctor) and the corresponding local enthalpy are obtained for each cell. Typical enthalpy rise factors are shown in Figures 3-kk-e and 3 kk-f for cells surrounding the hottest fuel rod located in the corner of the asse=bly. The asstaptiens used to describe the charmels for the peaking and enthalpy rise facters shown are given in Wall and Corner Channels Evaluation, 3.2.3.2.k J, which follevs. J. Evaluation of the JUn aatics in the Unit. Unll. and Corner Cells DNB Results The DNB ratics in the hot unit cell at the maxi =u= design condition described in 3 2 3 1 are shown in Figure 3-40. The relationships shown are based On the applicati data in the BAW-168(lI) and W-3{on 22,0c{) single channel An correlations. heat transfer additional ( sensitivity analysis of the asse=bly has been =ade utilizing 9-red assembly heat transfer DNE test data that is = ore representative of the actual vall and corner cells gec=etry effectc than single channel -, data. The sensitivity cf the assembly design with respect to variations of
= ass flow rate (G), channel spacing, =ixing intensity, and local peak-ing en the DNB ratics in the fuel asse=bly channels has been evaluated by cnalyzing the nc=inal conditiorJ and a pcstulated worst case cen-dition. The su==ary results are shown belev in Table 3-16-1.
Table 3-16-1 DNB Ratics in the Fuel Asse=bly Channels l l Uc=inal Case Cell Ty;c G. lb/hr-ft2 x 10-0 DNBR Corner 1 59 2.20 Wall 1 90 2.11 Unit 2 52 2.01 Postulated Worst Case Cell Tyre 0, lb/hr-ft2 x lO-b DNER Corner 1 32 1.70 Wall 1.6L 1.65 Unit 2.29 1 7;
' ..s'* 3-50f (Ravised 7-21-67) 01100 189 hp' lj
The DNBR's above are retics of the limiting hect flux t; tne local 1 flux along the length of the channels. The limiting hert fluxes have been detemined from the 9-rod asse=bly DUB test data. The DNB ratics in all channels cre high enough to insure a confidence-population relationship equal to cr better than that outlined in 3 2 3 1.1 for the hot unit cell channel. The postulated vorst case conditions are more severe than the required =axi=u= design conditions. The results of the asse=bly tests and this evaluation show that the perfomance of the vall and corner cells is = ore sensitive to local enthalpy than to the local = ass velocities. Although the = ass flow rates in the corner and vell cells are lower than in the unit cell, the total flew in these cells is relatively higher than the = ass flow rates imply be'cause of the increased space between the outer rods and the perforated can. This results in = ore favorable power-to-flow ratios than the = ass flow rates indicate.
*he DN3 ratios were cbtained by cc= paring the local heat fluxes and coolant conditions with heat transfer data points fro = 9-rod fuel asse=bly heat transfer tests for unifom heat flux with an appropriate correction for a nonunifo= cxial pcVer shape. ?/pical results are
. shcun in Figures 3-44-g and 3 44-h for the nc=inal and worst case condi-tions in the corner cell. The line defined by a best fit of the data is shcvn on each figure cs c solid line. A design li=it line, shown
,O as dotted, has been detemined by lovering the best-fit line to account 2 for the effects of nonunifom flux shapes. ~he =cgnitude of the re-duction was dete=ined by ec=parison with the results of the Euratc= nonunifom test data Reference 18 and the results of = ore recent non-unifom tests conducted by 3E. The limiting best-fit lines were derived from a 9-rod fuel asse=bly test section 72 in. long with rod dia=eter, pitch spacing, and spacer Crids of the type to be used in the reference design. A total of 513 data points between 1,000 psi and 2,450 psi has been obtained. One hundred and sixty-two of these points were used for the limiting lines in the PWR pressure cnd = css flow ranges. The rcnges of test varicbles for the 162 data points used were: Pressure - 1,800 to 2,L50 psi , Mass Flow Raue - 1.0 to 3 5 x 10 lb/hr-ft2 Quality - -5 to +20 per cent All of the cell conditions of interest in this analysis fall within this range of parameters. Fuel Rod Power Peaks and Cell Coolant Conditions
~'he nc=inal case local-to-average rod pcVers cad the local-to-average exit enthalpy rise ratics are shown in Figure 3-kh-e for the hot corner, hot vall, and hot unit cells in the hot fuel asse=bly. Values shcvn are for nominal water cape between the hot fuel asse=bly and adjacent
, is I '
f\'i 'j..:i'. t 3-50g (Revised 7-21-67) 00 l90 l
\
l I
fuel asse=blies vi-" -~-*~' rod-to-vall spacirs, with nc=inal flow to 1 h th9 a (* )ot fuel asse=bly,
, equal to 0.03 and with a nc=inal intensity of turbulence, Additional tests cre being run to deter =ine the =cxi=u= values of in-tensity of turbulence casociated with the fuel cssc=bly. Se expected value is greater then J.03 since this value is Obtained in s= cath tubes, and the spcecr: and can panel perforaticas should induce =cre turbulence.
2e postulated vorst case iceal-te-average rod powers and exit en-thcipy rise ratics in the hot fuel asse=bly are chovn in Figure 3-kk-f. The facters were dete:::ined for this case with twice the ac=inal vcter gcps between the hot fuel asse=bly and cdjaccat fuel asse=blies with
=inimum rod-to-vcil spacing, with =ini=u= fiev to the hot fuel asse:-
bly, and with a =ini=u= assu=ed intensity of turbulence, a, equel to 0.01. In neither the ac=inc1 nor the postulated verst Occe analysis has any credit been taken for the c0clant which is flovirq in the water gaps between the fuel csse=blics and which serves to rcduce enthalpies in the peripheral cells of the het fuel asse=bly by =ixing with the cool-ant in those cells through the can panel perforctions. In both cases, hcVever, the effe.:tise rou6h ness of the can panel perforations and its effect en reducin6 the flow in the peripheral cells of the fuel asse=- bly has been acccunted for. Se =agnitwie of the effective roughness vcs abttined frc= the results of a series of flew tests perfor=ed on a =cekup of the outer two revs of fuel reds and the can panels of two ; adjacent fuel asse:ablies. 2 e red-to-vall spacing in the peripheral cells of the fuel assc=bly nas been increased to ec=pensate for the ofrects of the can panel in reducia; the flow in the peripheral cells. S e c =inal distance frc= the center of the outside r ds to the can pcncl is 0 326 in. W e correspondir4 postulated verst case di=ension vcs assu=ed to be 2 310 in. Fuci Asse=bly power and Flev Ocnditiens Se n:=iral and postulated worst cases were run c lik per cent re-ceter pcVer with the nc=inal and verst F ah factor chown in 3 2 3 1.1 c. Se 150 =cdified :csine axial power shape of Figure 3-8 was used to describe the vorst axial condition. l t - ...- ( +) 2c intensity Of turbulence, c, is defined as l
)y.2
'I t' y
vhere VI is the transve e ec=penent of the fluctucting turbulent velocity, and V 15 the cociant ve.; city in the axial direction. Ric =ethod cf cc=- putin; =ixing is described by Sandberg, R. O. , cad Bishop, A. A., CVTR
~her=ci-!'.ydrculi Design for c5 W 3ross Fission Power, CVI:A-227. .
3-5ch (Revised 7-21-57) 0000 191 3 b e
/
t
The hot assembly flow urder nominal conditions without a flow mal- 1 distribution effect is 93 per cent of the average assembly flow, and the reduction in flow is due entirely to heat input effects. De hot assambly flow under the worst postulated conditions is 85 Per cent of the average asrembly flow and considers the vorst combined effects of heat input and flow mc1 distribution. i Analysis of all BW bundle data to date indicat.ts that the BW method vill correlate data with less deviation than previous methods. In-l dications are that this is also true when considering nonunifonn axial power distributions. Additional bundle tests vill be conducted with nonunifom axial pcVer distribution to confirm that the use of a power shape :orrection factor based on sin 61e channel and annular specimens is conservative. Completion of the test programs outlined in this report and evaluation of the experimental data vill provide final design correlations ard flow relationships tnat vill give complete confidence in the conser-vatism of the design and the BW analytical procedures. It should be noted that the postulated vorst case is worse than the hot channel pe=1tted by our specifications. Even with this postu-lated vorst case, the design is still conservative, and there is very
, little difference in the performance of the various ehannals. Bis indicates that the outside cell geometries have been compensated cor-rectly to account for vc11 effects.
,t 3-501 (Revised T-21-tit) 00'00 In
3 2.4 MECHANICAL DESIGN LAY 0172 , 3 2.4.1 Internal Layout Reactor internal components include the upper plenum assembly, the core support assembly (consisting of the core support shield, core barrel, lower grid and f3cv baffle, thermal shield, and surveillance specimen holder tubes), and the incore instrument guide extensions. Figure 3-45 shows the reactor vessel, reactor vessel internals arrangement, and the reactor coolant flow path. Figure 3-46 shows a cross section throua,h the reactor vessel, and Figure 3-47 shows the core flooding arrangement. Reactor internal components do not include fuel assemblies, control rod I assemblies (CRA's), surve1 Hance specimen assemblies, or incore instru-mentation. Fuel assemblies are described in 3.~2.4.2, control rod assem-blies and drives in 3 2.4 3, surveillance specimen assemblies in 4.4 3, and incore instrumentation in 7 3 3 The reactor internals are designed to support the core, maintain fuel assembly alignment, limit fuel assembly movement, and maintain CRA guide I tube alignment between fuel assemblies and control rod drives. They also ) ! direct the flow of reactor coolant, provide gamma and neutron shielding, ' provide guides for incore instrumentation between the reactor vessel lover head and the fuel assemblies, and support the surveinance speci- , men assemblies in the annulus between the thermal shield and the reactor vessel vall. A u reactor internal components can be removed from the re- !' actor vessel to anov inspection of the reactor internals and the reactor vessel internal surface. A shop fitup and checkout of all internal components in an as-built reac-tor vessel mockup vill insure proper alignment of mating parts before shipment. Dummy fuel assemblies and control rod assemblies vill be used to check fuel assembly clearances and CRA free movement. In anticipation of lateral deflection of the lover end of the core sup-port assembly as a result of horizontal seismic loadings, integral veld-attached, deflection-limiting spacer blocks have been placed on the reac-tor vessel inside vall. In addition, these blocks limit the rotation of the lower end of the core support assembly which could conceivably result i from flow-induced torsional loadings. The blocks anov free vertical l movement of the lover end of the internals for thermal expansion through-out an ranges of reactor cperating conditions, but in the imiikely event of a flange, circumferential veld, or bolted joint failure the blocks vill limitthepossiblecoredropto1/2in.orless. The final elevation plane of these blocks vill be established near the same elevation as the vessel support skirt attachment to minimize dynamic loading effects on the vessel she n or bottom head. Pre 14 min = 7 calculations indicate the impact loading on the stop blocks for a 1/4 in. core drop would be ap-proximately 5 s's total. Block location and geometry vin be evaluated and determined to transfer this loading through the vessel support skirt to the reactor building concrete. A significant reduction in impact lead-l .ing can be achieved through proper stop block design and detailed analysis. ! \ A1/2in.coredropwillnotallowtheloverendoftheCRApoisonrods I i I u , r' " 3-31 0000 193
to disengage from their respective fuel assembly guide tubes if the CRA's are in the full-out position, since approx 1=ately 6-1/2 in, of rod length would remain in the fuel assembly guide tubes. Acoredropof1/2in. vin not result in a significant reactivity change. The core cannot ro-tate and bind the drive lines because rotation of the core support assem-bly is prevented by the stop blocks. The failure of the core support shield and core barrel upper flanges, or related flanges and other circumferential joints, is not considered cred-ible on the basis of the conservative design criteria and large safety factors employed in the internals design. The final internals design vin be capable of withstanding various combir.ations of forces and load-ings resulting from the static veight of inte.rnals (179,000 lb total), core with control rod drive line (303,000 lb total), dynamic load from trip (10 g's gives 207,000 lb), seismic (0.10 g vertical gives 48,000 lb), coolant flow hydraulic loading (230,000 lb), and other related loadings. The algebraic sum of this simplified loading case is 507,000 lb. This results in a tensile stress of about 700 psi in the core support shield shell, which is approximately 4 per cent of the material yield strength. Final internals component weignts, seismic analysis, dynamic loadings from flow-induced vibration, detailed stress analysis with consideration for themal stress during all transients, and resolution of fabrication details such as shell roll hg tolerances and veld joint preparation de-tails win increase the stress levels listed above. As a final design criterion, the core support components vill =eet the stresa requirements of the ASME Code, Section III, during normal operation and transients. The structural integrity of all core support circumferential veld joints in the internals shells vill be insured by co=pliance with the radio-graphic inspection require =ents in the code above. The seismic analysis vill include detailed calculations to determine the maxi =um structural response of the reactor vessel and internals. *his analysis vin be per-fomed as described in 31.2.4.1. In the event of a major loss-of-coolant accident, such as e 36 in diam-eter reactor coolant pipe break near the reactor vessel outlet, the fuel assembly and vesse.'. internals vould be subjected to dynamic loadings re-sulting from an oscillating (approximately sinusoidal) differential pres-sure across the core. A preliminary analysic of this postulated accident indicates that the fuel assemolles vould move upward less than 3/8 in. Some deflection of the internals structures vould occur, but internals component failure vin not occur. The occurrence of a loss-of-coolant , accident and resulting loadings vin be evaluated during the detailed de-sign period for the fuel assemblies and related internals structural com-ponents. The deflections and move =ents described above vould not nrevent CBA in-sertion because the control rods are guided by split tutes throughout their travel, and the guide tube to fuel assembly alignment cannot change reganiless of related co=ponent deflections. CRA trip could c a eivably be delayed =cmentarily as a result of the oscillativ pressure differen-tial. However, the CRA travel time to full inse. cien vould re=ain rela-tively unaffected as transient pressure oscillecions are dampened out in approximately 0 5 sec. On this basis, the CRA trip time to 2/3 insertion
- vill be approximately 1 55 see instead of the specified 1.h0 sec. Also, i,
s J. o . 3-52 0000 194 M',
this possible initial minor delay in trip initiation would not contribute to the severity of the loss-of-coolant accident because at the initiation of CRA trip, the core would be suberitien1 frc= voids. Material for the reactor internals bolting vill be subjected to rigid quality control require =ents to insure structural integrity. The bolts vill be dye-penetrant inspected for surface flai indications after all fabrication opera-tions have been completed. Torque values vill be specified for the final as-se=bly to develop full-bolting capability. All fasteners will be lock-velded to insure assembly integrity. 3 2.4.1.1 Upper Plenum Asse=bly The upper plenum assembly is located directly above the reactor core and is re-moved as a single co=ponent before refueling. It consists of upper and center grid assemblies, CRA guide tubes , and a flanged cylinder with openings for re-actor coolant outlet flow. The upper grid is a series of parallel flat bars intersecting to form square lattices and is velded to the plenum cylinder top flange. A machined upper end on each CRA guide tute is located and velded to l1 the plenum cover which is attached to the upper grid bars. CRA guide tubes l provide CRA guidance and protect the CPA from the effects of coolant cross-flow. Each CBA guide tube consists of an outer tube housing and sixteen slotted tubes which are properly oriented and braned to a series of castings. As the tubes are slotted for their full length, the brazement provides continuous guidance for the CRA full stroke travel. Design clearances in the guide tube vill ac-
/O U
cc-'vdste some degree of misalignment between the CRA guide tubes and the fuel asse=blies. Final design clearances vill be established by tolerance studies 1 and by the results of the Control Rod Drive Line Facility (CRAL) prototype tests. Preliminary test results are described in 3 2.4.3 5 The center grid assembly consists of parallel flat bars intersecting to form square lattices. The bars are attached to a flange which is bolted to the l1 plenum cylinder lover flange. The center grid assembly locates the lover end of the individual CRA guide tube relative to the upper end of the correspond-ing fuel assembly. Locating slots in the upper plenu= asse=bly top flange engage the reactor ves-sel top flange locating devices to align the upper plecu= assembly with the re-actor vessel, reactor closure head control rod drive penetrations, and the core support shield. The bottom of the upper plenum asse=bly is guided and aligned by locating blocks attached to the inside of the core support shield. 3 2.k.l.2 Core Support Asse=bly The core support assembly consists of the core support shield, core barrel, lover grid and flow baffle, thereal shield, and surveillance speci=en holder tubes. l Static leads frcm the asse= bled components and fuel assemblies, and dynamic l l. loads from CRA trip, hydraulic flow, thermal expansion, seis=le j l, pd V l l
@l f.. 3-53 (Revised 7-21-67) 0000 195
disturbances, and less-of-coolant accident considerations, are all car-ried by the core s ;ppon asse=bly. De core support asse=bly co=ponents a e discribed as follovs:
- a. Ccre Suppert Shield The core support shield is a large flang=d cylinder which =ates with the react:r vessel Opening. The tcp flange rests on a
- 1rcu=ferential ledge in the reactor ve nel top cicsure flange.
The core suppc,rt shieM lover flange is helted t the core bar-rel. The cylinder vall has t/c nozzla cpenings fcr reactor cociant outlet flow. Locating b1:cks en the inside of the cyl-inder vall nee.r the tett0= guide and align the upper plenu= cha=ber relattve to the core support shield. The reactor vessel cutlet no :les are sealed to the =ating ec=- ponents of the core support chield by the differential the.~al expanston cetween the carbon steel rea: tor vessel and the stain-less steel cora support shield. Ra not:le seal surfaces are finished and f Ltted to a predetemined cold gap p cydding clearance during core support asse=bly installaticn and re= oval. I At reactor cperating te=perature the =ating =stal surfaces are in contact to =aka a seal without exceeding allovable stresses in either the reactor vessel or internals.
- b. Core Barrel g
W, The core barrel supports the fuel asat=blies and lever grid ar.d flev baffle. and directs the reactor coolant flow through the vessel. The barrel censists of a flanged cylinder, a series of internal horizontal spacers bolted to the cylinder, and a series of vertical plates bolted to the inner surfaces cf the horizan-tal spacers to for= an irner vall enclosir4 the fuel asse=blies. Ccnstraction of the core barrel vill be similar t.? that of the reacto.- internals ce=ponent developed by B&W for the Indian Point Station Unit No. 1. Ccolant flow is devnvard aleng the outaida of the core bar el cylinder ar.d uraard through the fuel asse=blies centained in the core barrel. A e'! portion of the ecolant ficvs upward thrcugh the space tet.een the core barrel outer cylinder and the inner plate vall. l Ccolant pressure in this space is =aintained slightly lover than I the core ecclant pressure to avoid tension 1cada on the bolts attaching the pistes to the hcrizontal spacers. The vertical l plate inner vall vill be careftlly fitted together t: reduce reactor ecolant leakage to an acceptable rate. The upper flance cf the core barrel outer cylinder is bolted to I tha =ating lover flange cf the core support shieli, and the I lever flange is belted to the =ating flange of the lover grid g l and flev baffle. All bolts vill be inspected and installed as W c e , 0000 196
i O described in 3.2.h.1, and will be lock-velded after final as-sembly. (DELETED)
- c. Lover Grid and Flow Baffle
*he lover grid provides alignment and support for the fuel as-semblies and aligns the incere instrument guide extensions with the fuel assembly incore instrument tubes. The lower grid con-sists of two flat plate and bar lattice structures separated by l short tubular columns surrounded by a flanged cylinder. The top flange is bolted to the lover flange of the core barrel.
, The lover grid top flange also positions and supports the ther- 1 mal shield. The flow baffle is a dished plate with an external flange which is bolted to the bottcm flange of the lower grid. The flow baffle is perforated to distribute the reactor coolant entering the bottom of the core.
- d. Thermal Shield A cylindrical stainlest i' eel thermal shield is installed in the annulus between the cs -e barrel outer cylinder and the reac-ter vessel inner vall. The thermal shield reduces the neutron and gamma internal heat generation in the reactor vessel vall
( and thereby reduces the resulting thermal stresses.
)
The thermal shield is supported on, positioned by, and attached 1 to the lover grid top flange. Also, the thermal shield upper end is positioned by spacers between the ther=al shield and the core barrel cuter cylinder to minimize the possibility of ther-mal shield vibration. The thermal shield attachment is designed to prevent fasteners from being loaded in shear. Fasteners are , lock-velded after final assembly,
- e. Surveillance Specimen Holder Tubes Surveillance specimen holder tubes are installed on the core support assembly outer vall to contain the st rveillance speci-men assemblies. The tubes extend from the top flange of the core support shield to the lower end of the thermal shield.
The tubes vill be rigidly attached to prevent flow-induced vi-bration. Slip joints at the icver end of the core support shield vill allow the shield to be removed from the core support assembly without destructively removing the surveillance speci-men holder tubes. 3.2.h.1 3 Incore Instrument Guide Extensiccs The incere instrq blies between thk'qent guide extensions instrument guide penetrations in the the incere instrument reactor vessel bottom assem-head and the instrument tubes in the fuel assemblies. Sufficient clear- 1 p ance in the instrument guide extensions provides for =inor misalignment Q between the reactor vessel instrument penetrations and the instrument guide extension tubes. A perforated shroud tube, ce 9~or 3-55 (Revised 7-21-o7) 0000 197-
i 1 concentric vitn the instru=ent guide tube, adds rigidity to the asse=bly and reduces the effect of ecolant flev forces. Fifty-two incere in-1 h strument guide extensiens are provided. The incere instru=ent guide ex-tensions are designed so they vill not be affected by the core drop de-scribed in 3.2.4.1. l 3 2.4.2 Fuel Asse=blj g l 3 2.k.2.1 Description .
- a. General Description The fuel for the reactor is sintered pellets of lov enrich =ent uranium dioxide clad in Zircaloy 4 tubing. The clad, fuel pel-lets, end supports, holddevn spring, ard end caps fom a " Fuel Rod". Two hundred ard ei S ht fuel rods are =echanically joined in a 15 x 15 arrsy to for= a " Fuel Asse=bly" (Figure 3-h8).
Ihe center position in the asse=bly is reserved for instru=en-tation. The re=aining 16 positions in the array are provided with " Guide Tubes" for use as control rod locations. The co=- plete core has 177 fuel asse=blies. All asse=bltes are identi-cal in =echanical construction, i.e., all are designed to accept the control rod asse=blies (CRA). However, only 69 have CRA's to control the reactivity of the core urder operatin6 corditions. In the 108 fuel asse=blies containing no CRA during a given core cycle, the guide tubes are partially filled at the top by an
" Orifice Rod Asse=bly" (Figure 3-k9) in order to =1ai=1:e bypass coolant flow. These orifice rod asse=blies also terd to equal- g ize coolant flow between fuel asse=blies with CRA's ard those '
with orifice rod asse=blies. Fuel asse=bly components, =aterials, and di=ensions are listed below. Ite: Material Di=ensions, in. Fuel UO2 Sintered 0 362 dia=. ' Pellets Fuel Clad Ziresloy-k O.420CDx0368IDx152-7/8 long i Fuel Rod Pitch 0 558 Fuel Asse=bly Pitch 8 587 Active Fuel Len8th 14k Overall Length =165 Control Rod Guide Zircaloy h 0.-530 OD x 0.015 vall Tube g 0000 198 4
., {, ;,, 3-56 (Revised 7-21-67)
O Item Material Dimensions, in. Incere Instrument Zircalcy k O.530 CD x 0.075 vall Guide Extension Spacer Grid Stainless Steel, Spacedat21-7/16in. Tp-304 Can Panel Stainless Steel, 0.031 thick TP-304 End Fitting Stainless Steel, Tp-304
- b. Fuel 1 I
The fuel is in the form of sintered and grcund pellets of ura- l nium dioxide. De pellets are dished on each end face to min- ; imize the difference in axial thermal expansien between the j fuel and cladding. The density of the fuel is 95 per cent of , theoretical. 1 Average design burnup of the fuel is 28,200 MWD /MTU. Feak l burnupis55,000 mwd /MTU. At the peak burnup, the fuel growth l iscalculatedtobe9-1/2volumepercentbythemethodgiven j in Reference h9 B is growth is acecnimodated by pellet pcros- , ity, by the radial clearance provided between the pellets and ) the cladding, and by a small amount of plastic strain in the
]
cladding. j Each fuel column is located, at the bottom, by a thin-vall stainless steel pedestal and is held in place during handling by a spring at the top. The spring allows axisi differential thermal expansion between fuel ani cladding, and axial fuel ) growth. The bottom pedestal is also collapsible, thus provid-ing a secondary buffer to prevent excess cladding axial strain. Fissica gas release from the fuel is acccamodated by voids within the fuel, by the radial gap between the pellets and cladding, and by void space at the top and bottom ends of the fuel rod.
- c. Fuel Assembly Structure (1) General The fuel assembly shown in Figure 3-48 is the canned type.
Eight spacer grids and four perforated can panels form the basic structure. The panels are velded together at w the corners for the entire length. The spacer grids are velded to the panels, and the 1cwer and upper end fittings are velded to the panels to cceplete the structure. De
>'.t I.it*' upper end fitting is not attached until the fuel rods,
.c us, p:
3" 0000 199
guide tubes, and instru=entation tube have been installed. At each spacer grid asse=bly each fuel rod is supported on four sides by integral leaf-type springs. Rese springs are designed to provide a radial load on the fuel rod suf-ficient to restrain it so that flow-induced vibrational amplitudes are =in bal. However, to avoid undesirable bowing of the fuel rods, the spring loads are designed small enough to pemit the relative axial motion required to acco==edate the differential the=al expansion between the Zircaloy fuel red and the stainless steel structure. j (2) Spacer Grid These grids are composed of ferrules =ade of square tubing. The ferrule has a portion of each side fomed into spring sections which have hydrodynamically shaped " dimples" that contact the fuel rods. The ferrules are joined together by brazing to fom the spacer grids. The grids, which provide the desired pitch spacing between fuel rods, are spot-velded at intervals to the perforated stainless steel can panels. (3) Lower End Pitting The icver end fitting is constructed from Type 304 stain-Jess steel members which when joined together fem a box structure. Four deep cross me=bers serve as the position-ing surfaces for the fuel assembly when it is inserted in-to the icver core support structure. The assembly includes a grid structure which provides a support base for fuel rods while maintaining a max hum inlet flow area for the ccolant. (4) Upper End Fittirg The upper end fitting is similar to the lower end fitting. It positions the upper end of the fuel asse=bly and pro-vides coupling between the fuel asse=bly and the hanM 4"g equipment. A hollow post, velded in the center cf the as-sembly, is designed to provide a = cans of uncoupling the CRA-to-drive connection and to retain the orifice red as-sembly. In order to identify a fuel assembly under water, a serial number is milled into a flat, chrome-plated sur-face which is velded to the box frame. (5) control Rod Guide Tubes The Zircaloy guide tubes serve to guide the control rods within the fuel assembly during operation. S e tubes are restrained axially by the upper and lover end fittings in the fuel asse=bly and radially by the spacer grids in the same =anner as the fuel rods.
.m -
3- s 0000 200
- c. n , ice. .tc L -
l l l O 3 2.4.2.2 Evaluation
- a. Fuel Rod Assembly (1) General The basis for the design of the fuel rod is discussed in 3 1.2.4. Materials testing and actual operation in reac-tor service with Zircaloy cladding has demonstrated that i Zircaloy-4 material has ample corrosion resistance and sufficient mechanical properties to maintain the integrity and serviceability required for design burnup.
(2) Clad Stress Stress analysis for cladding is based on several conserva-tive assumptions that make the actual margins of safety greater than calculated. For example, it is assumed that the clad vith the thinnest vall and the greatest ovality pennitted by the specification is operating in the region of the core where performance requirements are severest. Fission gas release rates, fuel growth, and changes in mechanical properties with irradiatien are based on a con-servative evaluation of currently available data. Rus, it is unlikely that significant failure of the cladding
/ vin result during operation.
The actual clad stresses are considerably below the yield strength. Circumferential stresses due to external pres-sure, calculated using those combinations of clad dimen-sions, ovality, and eccentricity that prcduce +5e highest l stresses, are shown in Table 3-17 Se maximwn stress of 33,000 psi compressica, at the design pressure of 2,500 psi, is the sum of 22,000 pai ccupressive membrane stress plus n ,000 psi compressive bending stress due to ovality at the claa OD in the expansion void, and at the beginning-of-life. Se maximum stress in the heat-prcducing zone is 32,000 psi at design pressure, 27,000 psi at operating pressure. At this stress, the material may creep suffi-ciently to anov an increase in ovality until further creep is restrained by support frce the fuel. Contact loadsentheorderof20lb/in.oflengtharesufficient to counteract the bending stress. Creep co uapse testa have indicated a long time cou apse resistance in excess of the requirement to prevent collapse in the end void. ] As the fuel red internal pressure builds up with time, these stresses are reduced. i Late in life, the fuel rod internal pressure exceeds the - system pressure, up to a maxtnum difference of 1 no pai. The resultant circumferential pressure stress of 9,000 psi is about 1/4 cf the yield strength and therefore is not a ( l O .
*4I potential source of shcrt time burst. Me ;cesib*lity of stress-rupture burst has been investigated using finite-E s i <.
- 0000 201 g
difference =ethods to esti= ate the long the effects of the increasing pressure on the clad. S e predicted pres-sure-ti=e relationship produces stresses that are less than 1/3 of the stress levels that would produce stress rupture at the end-of-life. Outpile stress-mpture data were used, but the greater than 3:1 =argin on stress is
= ore than enough to account for decreased stress-rupture strength due to irradiation. Clad circumferential stress-es are listed in Table 3-17 The free gas content of the fuel rod is calculated by con-sidering (1) initial helium fill gas, (2) initial water vapor and at=ospheric gases adsorbed on the fuel, and (3) fission product gases. S e water vapor present initially is expected to dissociate over the life of the fuel and enter into hydriding and oxidizing reactions. The gas re-naining at the end-of-life, when the =ax bum internal pressures exist, consists of the at=ospheric gases and holium present initially plus the released fission gases.
. The fission gas production is evaluated for a range of l neutron fluxes and thg f4ssionable =aterial present over
. the life of the fuel. 601 A design value for gas produc-tion has been determined as 0.29 atec:a of gas per fission.
O 1 l l u m e
, 3., 0000 202
. c* .
Table 3-17 Clad Circumferential Stresses Ultimate Cale. Yield Tensile Stress, Stress, Stress, operating Condition psi psi psi
- 1. BoL("} - operating at Design Pressure Total Stress (toembrane t bending) Due to 2,500 psig System Design Pressure Minus 100 psig Fuel Rod Internal Pressure Average Clad Temperature - Approx 1-mately 625 F (expansion void) -33,000 46,000
- 2. EOL - Maximum overpover System Pressure - 2,185 psig Fuel Rod Internal Pressure -
3,300 psig ( Average Temperature Th::cugh Clad Thickness at Hot Spot - Approxi-mately 725 F Pressure Stress only(b) g,oco Including 4,000 psi Thermal Stress 13,000 36,000 38,000 3 EOL - Shutdevn Immediately After Shutdown System Pressure - 2,200 psig Fuel Bod Internal Pressure - 1,750 psig Average Clad Temperature - Approxi-l mately 575 F -4,000 45,000 E8,000 (a) Cladding is being ordered with 45,000 psi minimum yield strength and lo per cent minimum elongation, both at 650 F. Minimum room temper-ature strengths vill be approximately 75,000 psi yield strength (0.2 l per cent offset) and 85,000 psi ultimate tensile strength. l (b) Cladding stresses due to fuel swelling are discussed further on l another page. 1 L .x. M ' 3 ,1 0000 203
Table 3-17 (Cont'd) h
'J1ti= ate Cale. Yield Tensile Stress, Stress, Stress, Operating Condition psi psi psi 3 Hours later (50F/hrPressurizerCooldownRate)
Fuel Rod Internal Pressure - 1,050 psig System Pressure - 680 psig Avera6e Clad Temperature - Approxi-mately 425 F 3,300 52,000 55,000 The total production of fission gas in the hottest fuel red assembly is based on the hot rod average bumup of 38,000 MWD /MrU. The corresponding maximum burnup at the hot fuel rod midpoint is 55,000 MWD /!cU. The fission gas release is based on temperature versus re-lease fraction experimental data.(45) Fuel te=peratures h ', are calculated for small radial and axial incre=ents. The total fission Gas release is calculated by integrating the incremental releases. The max 1=um release and gas pressure buildups are deter-mined by evaluating the fonoving factors for the most con-servative conditions: (a) Gas conductivity at the end-of-life with fission gas present. (b) Influence of the penet-to-clad radial Bap and contact heat transfer coefficient on fuel te=perature and re-lease rate. (c) Uhrestrained radial and axial c er=al growth of the fuel penets relative to the clad. (d) Hot rod local peaking factors. (e) Radial distribution of fission gas production in the fuel penets. (f) Fuel temperatures at reactor design overpower. O l s
, >:. . 3-62 0000 204 l
O The fuel temperatures used to determine fission 6as release and internal gas pressure have been calculated at the re-actor overpower condition. Fuel te.peratures, total free gas volume, fission gas release, and internal gas pressure have been evaluated for a range of initial diametral clear-ances. This evaluation shows that the highest internal pressure results when the maximum diametral gap is assumed because of the resulting high avera8e fuel temperature. The release rate increases rapidly with an increase in fuel temperature, and unrestrained axial growth reduces the rel-atively cold gas end plenum volumes. A conservative ideal thermal expansion mcdel is used to calculate fuel tempera-tures as a function of initial cold diametral clearance. Considerably lover resistance to heat transfer between the fuel and clad is anticipated at the end-of-life due to fuel fracture, sve1Hng, and densification. The resultin6 maxi-
.mm fission gas release rate is 43 per cent.
(3) Collapse Margins Short time collapse tests have demonstrated a clad collaps-ing pressure in excees of 4,000 psi at expansion void maxi-mum temperature. Collapse pressure margin is approximately 17 Extrapolation to hot spot average clad temperature ( = 725 F) indicates a collapse pressure of 3,500 psi and a l ' margin of 1.4, which also greatly exceeds requirement. Outpile creep collapse tests have demonstrated that the clad meets the long time (creep collapse) requirement. (k) Fuel Svelling Fuel rod avera6e and hot spot operating conditions and de-sign parameters at 100 per cent pover, pertinent to fuel swelling considerations, are listed below. - Average Max 1=um 2 HeatFlux, Btu /ft-hr 167,620 543,^00 Linear Heat Rate, kv/ft 54 17 5 Fuel Temperature, F 1,385 h,160 Burnup (MWD /MU) at Equilibrium 28,200 55,000 Ncminal Values Pellet OD, in. 0 362 Pellet Density, % of Theoretical 95 Pellet-Clad Diametral Cap at Assy., in. 0.004 - 0.008 Clad Material cold-Worked Zr-4 j , Clad Thickness, in. 0.026 l N, s-r~ 3-63 0000 205 .
The capability of Zircaloy-clad U02 fuel in solid red fom to perform satisfactorily in PWR service has been a= ply demonstrated through operation of the CVTR and Shippingport cores, and through results of their supple =entary develop-ment programs, up to approximately k0,000 !SD/!CU. As outlined below, existing experi= ental infor=ation sup-ports the various individual design parameters and oper-ating conditions up to and perhaps beyond the maximum burnup of 55,000 MWD /!CU, but not in a single experiment. However, the LRD 1rradiation test program, currently in progress, does combine the items of concern in a single experiment, and the results are expected to be available to contribute to' final design confi= ation. (5) Application of Experimental Data to Design Adequacy of the Clad-Fuel Initial Gap to Accommodate Clad-Fuel Differential i Thermal Expansion l Experimental Work Six rabbit capsules, each containing three Zr-2 clad rods of 5 in. fuel length, were irradiated in the Westinghouse . Test Reactor (41) at power levels up to 24 kv/ft. The 94 per cent theoretical density (T.D.) U02 pellets (0.430 OD) had initial clad-fuel diametral gaps of 6,12, and 25 mils. No dimensional changes were observed. Central =elting oc- ) curred at 24 kv/ft only in the rods that had the 25 mil initial Sap. Two additional capsules vere tested.(31) The specimens were similar to those described above except for length and initial gap. Initial gaps of 2, 6, and 12 mils were used in each capsule. In the A-2 capsule, three 38-in.- l long rods were irradiated to 3,450 ZGD/!CU at 19 kv/ft maximum. In the A-4 capsule, four 6-in.-long rods were irradiated to 6,250 MWD /!CU at 22.2 kv/ft =aximum. No central melting occurred in any rod, but diameter in-creases up 'o 3 mils in the A-2 capsule and up to 1 5 mils in the A-4 capsule were found in the rods with the 2 =11 initial gap. Application In addition to demonstrating the adequacy of Zirealoy-clad UO2 pellet rods to operate successfully at the power levels of interest (and without central melting), these experi=ents demonstrate that the design initial clad fuel gap of k to 8 mils is adequate to prevent unacceptable clad diameter in-crease due to differential thermal expansion between the
,g plad and the fuel. A maximum local diametral increase of less than 0.001 in. is indicated for fuel rods having the minimum initial gap, operating at the maximum overpower cundition.
d b. i i6H 0000 206
,s 3-64
1 O (6) Adequacy of the Available Voids to Accommodste Differential Expansion of Clad and Fuel, Including the 'dfects of Fuel Svelling Experimental Work Zircaloy-clad, tJ02 pellet-type rods have performed success-fully in the Shippingport reactor up to approximately 40,000 wD/m. Bettis Atomic Power Laboratory (49) has irradiated plate-type IJ02 fuel (96-98 per cent T.D.) up to 127,000 ZeiD/ m and at fuel center temperatures between 1,300 and 3,800 F. This work indicates fuel swelling rates of 0.16% AV/1020 f/cc until fuel internal voids are filled, then 0 7% AV/ 1020 f/ccafterinternalvoidsarefilled. This point of
" breakaway" appears to be independent of temperature over the range studied and dependent on clad restraint and the void volume available for collection of fission products.
The additional clad restraint and greater fuel plasticity (from higher fuel temperatures) of rod-type elements tend to reduce these swelling effects by providing greater re-sistance to radial swelling and lower resistance to longitu-dinal swelling then was present in the plate-type test speci-
, mens.
I This is confirmed in part by the work of Frost, Bradbury, andGriffithsofHarwell(52) in which 1/4 in. diameter IJ02 pellets clad ir. 0.020 in. stainless steel with a 2 mil diametral gap were irradiated to 53,300 24tD/ m at a fuel center temperature of 3,180 F vithout significant dir.ensional change.
- In other testing (53) 0.150 in. OD, 82-96 per cent T.D. oxide i pellets (20 per cent Pu, 80 per cent U) clad with 0.016 in.
stainless steel with 6-8 mil diametral gaps have been irra-disted to 77,000 16iD/ m at fuel temperatures high enough to approach central melting without apparent detrimental results. Comparable results were cbtained on rods svaged
. to 75 per cent T.D. and irradiated to 100,000 MiD/m.
Application Based on the BAPL experimental data, swelling of the fuel
+
rods is estimated as outlined below. Fuel is assumed to swell uniformly in all directions. Clad-pellet differential themal ;xpansion is calculated to be about 0.00k in at the maximum linear heat rate, so that all of the minimum initial gap of 0.004 in. is filled up by thermal expansion. If the initial gap exceeds the mini-
- O. mum, the additional gap volume is assumed available to accomodate swelling. This additional void volume may g c ,- p g, initi, ally tend to be filled by pellet thermal-expansion
.m :v o on99 ,97
~
.,p. ,
3-65
because of the low contact pressure and resultant low con-tact coefficient, but as the fuel svel'.s, the contact pres-sure nust increase if the clad is to be stretched. Where fuel cracking tends to fill the radial gap, it is assu=ed that the crack voids are available to absorb svelling. The evternal effect of fuel swelling is assu=ed to occur at 0.16% av/lGT f/ccuntilthe5percentinitialvoidin the 95 pe" cent T.D. pellets is fille <1 at about 9 x 1020 f/cc. F sm that ti=e on, svelling it, assumed to take place at 0 71 AV/1020 f/ccuntilthe=aximumburnupof13.6x 1020 r/r,c (35,000 WD/MrU) is reached. Total fuel volume increase is 4-1/2 per cent, which results in a 1-1/2 per cent diameter increase in a rod with the 0.004 in. mini-mum initial gap. Clad stress is estimated at 22,000 psi, so that the elastic strain is about 0.2 per cent. Net plastic strain is 1 3 per cent. Si=11ar calculations indi-cate that fuel rods with maximtm burnup and the nominal clad-fuel gap (0.006 in. at assembly) vill have clad plastic. strains of about 0.6 per cent at the end-of-life. Based on outpile data, stress rupture should not be a problem at these strains. Qualitative information from ISER(54 suggests that swell-ing rates for this design may exceed those indicated by the BAPL data because 9f the higher fuel te=peratures. Howev the A.E.R.E. tests \52) and the General Electric tests (53)er, do not support more than a small increase in post " break-g ', away" svelling rates at temperatures of interest. Fuel Svelling Studies - LRD Irradiation Program (55) Dimensional stability of UO2 under inpile conditions simu-lating large reactor environ =ents is under investigation. This study is currently being carried out under USAEC Con-tract AT(30-1)-3269, "Large Closed-Cycle Water Reactor Research and Development Progrsm". Parameters contributing to swelling are bumup, heat rating, fuel density and grain si::e, and clad restraint. These are syste=atically being studied by irradiating a series of capsules containing fuel rods. These experiments were as-signed by the AEC to EIR/MIR. Test variables are shown in Table 3-18. O _ , ; , _y O'000 208
.i
O Table 3-18 LRD Fuel Swelling Irradiation Program Initial coal Heat Rating (b) Capsule (*) Enrichment, kw/ vatts/ Fuel Density, Burnup, WAPD-49 4 ft em 4 T.D. WD/MM A 18.7 12 394 9h and 97 38,000 B 18 7 12 394 94 and 97 38,000 c 18 7 12 394 90, 94, and 97 38,000 D 16.o 18 591 90 and 97 47,000 E 13 5 18 591 94 and 97 E7,ooo F 13 5 18 591 90, 94, and 97 47,000 o 16.0 18 591 90 and 97 E7,ooo H 17 0 24 788 94 and 97 56,000 I 18 7 24 788 94 and 97 56,000 J 20.0 24 788 94 and 97 56,000 K 20.0 24 788 90 and 94 56,000 L 20.o 24 788 94 and 97 56,000 (*)Fourrods/ capsule. ( ) Fuel center temperatures vary from 1,570 to 4,11o F. O !' oe f ; '; C ^ 3-67 0000 209
Effect of Zircaloy Creep The effect of Zircaloy creep on the a=ount of fuel rod growth due to fuel sve ning has been investigated. Clad creep has the effect of producirg a nearly constant total pressure on the clad E by per=1tting the clad dia=ner to increase as the fuel dia=eter increases. Based on cut-of-pile data,(56) 1 per cent creep vin result in 10,000 hr (corresponding appioximately to the end-of-life dia=e-tral swelling rate) fro a stress of about 22,000 psi at the n720 F avirage te=perature through the clad at the hot spot. At the start of this high swelling period (roughly thelast1/3ofthecerelife),thereactorcoolantsys-te= pressure vould = ore or less be balanced by the rod internal pressure, so the total pressure to produce the clad stress of 22,000 psi vould have to ec=e from the fuel. Contact pressure vould be 2,h00 psi. At the end-of-life, the rod internal pressure exceeds the system pressure by about 1,100 psi, so the clad-fuel contact pressure would drop to 1,300 psi. Assu=ing that irradiation produces a 3:1 increase in creep rates, the clad stress for 1 per cent strain in 10,000 hr vould drop to about 15,000 psi. Contact pressures vould be 1,800 psi at the beginning of the high swelling period, 700 psi at the end-of-life. Since the contact pressure was assu=ed to be 825 psi in calculating the contact coefficient used to detemine the fuel pellet the= al expansion, there is only a short pe-riod at the very end-of-life (assu=ing the 3:1 increase in creep rates due to irradiation) when the penet is slightly $ hotter than W.culated. The effect of this vould be a slight increase in pellet themal expansicn and therefore in clad strain. Considering the i= probability that irridia-tion vill actually increase creep rates by 3:1, no change is anticipated.
- b. Overall Asse=bly l (1) Assurance of Centrol Rod Asse=bly Free Motion I
The 0.058 in. dia=etral clearance between the control rod ! guide tube and the control rod is provided to cool the con-trol rod ard to incure adequate freedc= to insert the con-trol rod. As indicated below, studies have shown that fuel rods will not bow sufficiently to touch the guide tube. Thus, the guide tube vill not undergo defoc scion caused by fuel rod boving Q ects. Initial lack of straightness of fuel rod and guide tube, plus other adverse tolerance G nditions, conceivably could ceduce the C.083 in. no=1:al gsp between fuel rod and guide tube to a =in1=u= of about 0.045 in., including a=plification of bowing due to axial friction loads frc= the spacer grids. The =aximu= expected l1 flux gradient of 1.176 across a fuel rod vin produce a temperature difference of 12 F, which vill result in a themal bov cf less than 0.002 in. Under these conditions, h 0; s i. 3 68 (Revised 7-21-oT) 0000 210
* . j p ' . '.
for the fuel rod to touch the guide tube, the ther-a' gradient across the fuel rod dia=eter vould have to be on the order of 300 F. The effect of a DN3 occurrirq on the side of a fuel rod adjacent to a guide tube vould result in a large te=pera-ture difference. In this case, however, investigation has shown that the clad temperature vould be so high that in-sufficient strength would be available to generate a force of sufficient =agnitude to cause a significant deflection of the guide tube. In addition, the guide tube would ex-perience an opposir4 gradient that vould resist fuel rod bowing, and its internal cooling vould maintain tempera-tures much lover than those in the fuel rod cladding, thus retaining the guide tube strength. (2) Vibration The se=ie=pirical expression developed by Burgreen(37) was used to calculate the flow-induced vibratory amplitudes for the fuel asse=bly and fuel rod. The calculated a=pli-tude is 0.010 in. for the fuel asse=bly and less than o.005 in. for ,the fuel red. The fuel rod vibratory a=pli-tude correlates with the =easured a=plitude obtained from a test on a 3 x 3 fuel rod asse=bly. In order to substan- ,p tiate what is believed to be a conservatively calculated V amplitude for the fuel asse=bly, a direct =eacure=ent vin be obtained for a full-sice prototype fuel assembly during testing of the asse=bly in the Control Rod Drive Line Facility (CRDL) at the B&W Research Center, Alliance, Ohio. (3) De=onstration In addition to the specific ite=s discussed above, the overan =echanical perfez=ance of the fuel assembly and its individual co=ponents is being demonstrated in an ex-tensive experi= ental program in the CRDL. 3 2.4.3 Control Rod Drive System
-3 2.h.3 1 Control Rod Drive System Design Criteria l I
2he control rod drive system shall be designed to =eet the follovir4 per- ! for=ance criteria: l
. 1
- a. Single Failure No single failure shall inhibit the prctective acticn of the control rod drive system. The effect of a single failure shall l be limited to one control rod drive.
t 'g.. . , s , s- .. m 0000 211 2-09
- b. Uncontrolled 'Jithdrawal No single failure or chain of failures shall cause uncentrolled withdrawal of any centrol rod assembly (CRA).
- c. Equitment Re=cval The disconnection of plug-in type connecters, =odules, and sub-assemblies frc= the protective circuits shall be annunciated or shall <.ause a reactor trip.
- d. Contr31 Red Assembly (CPA) Trip The trip ec==and shall have priority over all other ce== ands.
Trip action stall be positive and nonreversible. Trip circuitry sht.11 provide ,he final protective action and shall be direct-ac:ing, incu: A nimu= delay, and shall not require external power. Circuit '.nterrupting devices shall not prevent reactor trip. Fuses, where used, shall be provided with blevn indica-tors. Circuit bri.aker position information shall also be indi-cated.
- e. CRA Insertion Insert co==and shall have priority over withdraw cc==and. The centrol rod drive vill be capable of overcoming a "sturk-rod" condition equivalent to a h00 lb veight.
- f. 'Jithdrawal s The control rod drive system allevs caly two out of four regulat-ing CRA groups to withdraw at any time subject to the conditions described in T.2.2.1.2.
- g. Position Indication Continuous position indication, as well as an upper and lower 1 position limit indication, shall te provided for each control rod drive. The accuracy of the position indicators shall be consistent with the tolerance set by reactor safety analysis.
- h. System Monitor's The control rod trive control system shall include provisions for =onitoring coaditions that are i=portant to safety and re-liability. These include red position deviation and pcver supply voltage.
- i. Drive Seeed The centrol rod drive centrol syste= shall provide for single 1 uniform speed of the =echanis=. The drive centrols, or =ech-anis = and =otor ec=bination, shall 9
. c - .
' 9 I' 3-70 (Eevised T-21-67) 0000 212
have an inherent speed-limiting feature. The speed of the mechanism shall be 25 in./ min ppis or minus 10 per cent of the predetermined value for both insertion and withdrawal. The withdrawal speed shall be limittd so as not to exceed 25 per cent overspeed in the event of speed control fault.
- j. Mechanical Stoes Each control rod drive shall be provided with positive mechani- 1 cal stops at both ends of the stroke or travel. The stops shall be capable of receiving the rull operating force of the mech-anisms without failure.
3.2.k.3.2 Control Rod Dr're The control rod drives provide for controlled withdrawal or insertion of the control rod assemblies (CRA) out of or into the reactor core to es-tablish and hold the power level required. The drives are also capable of rapid insertion or trip for emergency reactor conditions. The con-trol red drives are buffer seal, rack-and-pinion type drives under devel- 1 opment by Diamond Power Specialty Corporation. The control rod drive data are listed in Table 3-19 A control rod drive consists of a rack housing, snubber bottoming spring assembly, rack, rack pinion, coupling assembly, drive shaft housing, miter gear set, drive shaft assembly, buffer seal assembly, magnetic clutch, f( ) gear reducer, drive motor, position indication transmitters, and limit switch system. The spool piece series to join the drive assembly to the reactor closure head no::le as shown in Figure 3-50. The drive motor supplies torque through the magnetic clutch to the drive shaft-gear system to provide vertical positioning of the rack. I f (DELETED)
\
i (h
/
/
I l s s .I ,' : 3-71 (Revised 7-21-67)
)000 ;);3
(DELETED) Table 3-19 Control Rod Drive Desien Data g Ite: Data Nu=bor of Drives 69 Type Buffer Seal, Rack and Pinion l1 Location Tcp-Mounted Direction of Trip Devn Velocity of Nor=al Withdrawal and Insertion, in./=in 25 Maximu= Trip Ti=e for 2/3 Insertion, see 1.h Length of Stroke, in. 139 Design Pressure, psig 2,500 Design Temperature, F 650 The control rod drive is shown en FiCures 3-50 and 3-51. of the control rod drive are described as follevs: subasse=blies h/
- a. Rack Housing The rack housing contains the hydraulic snubber, the bottc=- 1 ing spring assembly, the rack, rack pinien asse=bly, and a rack g"ide bushing. The lover guide tube is attached to the lever end of the rack housing, and the cap and drive line vent asse=-
bly is =cunted on the upper end of the rack housing. The hydraulic snubber decelerates the moving elements of the drive at the end of travel by controlled orificing of reactor i coolant water. The bottc=ing spring asse=bly absorbs the bot-tc=ing i= pact in a stack of spring vashers. The rack is guid;d by an upper shoe attached to the upper end of the rack, a rack guide bushing located at the pinion, and a lever guide tube bushing located at the lever end of the lover guide tube. The rack pinion is carried 1y two ball bearings. The valve on the cap and drive line vent asse=bly is used to bleed air or gases fron the rack housing during reactor startup. The re= oval of this assembly provides the access for CRA coupling and uncoupling, and for securing the racks in the retradted po-sition when the reactor closure head or individual drives are to be removed. 4 ( s i', 3-72 (devia:d T-21-67) 0003 214
!t!!.
- b. Drive Shaft Housing The drive shaft housing censists of the = iter gear set, the 1 drive shafts, and their supporting ball bearings. The drive shaft asse=bly is =ade up of two shafts with an inter =ediate bearing to increase their critical speed.
The drive shaft housing is attached to the rack housing by four through-bolts. All gasketed joints are of the double Concseal-type with a pres-sure testing tap between the seals.
- c. Buffer Seal A pressure breakdown-type seal is employed to seal the drive shaft penetration in the reactor coolant pressure centainer.
Seal system vater is injected between the eighth and ninth stages of a 9-stage seal to provide a contrclled leakage of ap-proximately 5 gal /hr into the reacter coolant systa= and 20 gal / hr to the =akeup tank. The seal vater is eccled below 120 F, demineraliced, and specially filtered before injection into the seal. A conventional rotary seal is e= ployed to prevent seal vater fran entering the drive package. s d. Drive Package The drive package is a synchronous ;ype containing a SCIF locking vorm gear reducer, a =agnetic clutch, pcsition indication transmitters, and a li=1t switch syste=. In conjunction with the magnetic clutch is a undirectional =echanical clutch which vill allow the motor to drive the rod dcun to the full-in posi-tien should a " stuck-red" condition develop in the course of a trip action. The =cter has inherent braking so no separate brake is required. The self-lceking wor = gear reducer prevents torque feedback to the =otor.
- e. Pcsition Trans=itters and Limit Switches .
The position transmitters and limit switches are located between : 1 the buffer seal and the gear =oter in the pcVer package, and l supply redundant pcsiticn signals and li=it switch contacts. I i There are three separate devices included in the position and ' limit switch transmitter asse=bly. A potentic=eter generates an analog position signal, a linear variable differential trans- 1 for=er (LVDT) generates both an analog positica signal and li=it contacts, and the li=it switch =echanis: provides li=it centacts, beter to Figure 3-57
, The potentic=eter is geared directly to the drive shaft and rx ,g),, gives a continucus d-c signal prc;crtional to t' ' CRA position.
(. The LVDT trans=itter has a ecre that is =cved b, =e ans o f a ba' ' screv =echanis= geared to the drive shaft. A de=odulater located r 1 l 3-73 (Revised 7-21-o7) 0000 215 7 a.
-m -
vithin the control cabinet contains the necessary electrcnic circuitry to generate the analog d-c signal. This descdulater 1 g also has relays with adjustable set points fer position con-tacts. The limit switch assembly consists of switches operated t/ linear cams that are moved by a ball cerev. This is also geared directly to the drive shaft. By using these three transmitters, it is possible to get both redundant position and redundant limit signals.
- f. Housing Design Critgria The control rod drive assembly housings are designed to the same design criteria as is the reactor pressure vessel. Accordingly, the drive shaft and rack housings comply with Section III of the ASME Boiler and Pressure Vessel Code under classification as Class A vessels. The operating transient cycles, whfch are con-sidered for the stress analysis of the reacter pressure vessel, are also considered in the housing designs.
Quality standards relative to =aterial selectior., fabrication, and inspection are specified to insure safety function of the housings essential to accident prevention. Materials etnform to ASTM or ASME, Section II, Material Specifications. All veld-ing shall be performed by personnel qualified under ASME Code, Section IX, Welding Qualifications. These design and fabrica-tion procedures establish quality assurance of the assemblies to contain the reactor coolant safely at operating temperature and pressure. For vibratory and seismic loadings, the assemblies are restrained with a serie's of contcured plates that are bolted to the =ain sup-port structure. These plates are contoured to restrain the upper flange outside diameter of the drive shaft and rack housings as shown in Section CC, Figure 3-51, of the PSAR. The main support structure is bolted to the reacter closure head. These plates vill pa tide lateral support only. Vertical motien of the hous-ings resulting frem ther=al expansion vill not be restricted. In the highly unlikely event that a pressurc barrier ccmponent of the control rod drive asse=bly did fail catastrophically, i 1.e., a ecmplete rupture, the folleving results would ensue: l (1) Control Rod Drive Nez:le For the failure of this component, the asse=bly vould be ejected upward as a =issile until it was stopped by the re-actor building missile shield. This upward sotion would have no adverse effect en adjacent assemblies. (2) Rack Housing The failure of this cc=penent anywhere above the lever flange vould result in a =issile-type ejection into the i 3-7h (Revised 7-21-o7) b L+> e
- I $kh
missile shielding of the reactor building. There would be no adverse effect on adjacent mechanisms. (3) Drive Shaft Housing The failure of this component could result in contact with adjacent assemblies onl;r if the bolting by which the drive shaft housing is attached to the rack housing failed com-pletely. However, it should be noted that this bolting vill be designed to Section III of the ASME Boiler and Pres-sure Vessel Code, and therefore has the sa=e integrity as the housing itself. A general design criterion--that no single bolt failure vill lead to subsequent failure of the remaining bolts- has been imposed upon the design. In addition, the distance between adjacent drive shaft housings is small, approximately 1/h in., and should a , failure be postulated, the short travel distance ava-lable vill not permit the failed drive to obtain a high velocity. The design criterion for the rack housing is to accept this impe.ct load without failure. If a rod ejection accident were assumed to occur, even though design precautions have been taken to insure the integrity of the control rod d;ive assembly, r.o further da= age to the reac-ter internals or ise reactor coolant system vill occur, nor fO vould gross fuel tailures result. This analysis is presented V An 14.2.2.2. Since the control rod drive assembly is designed to minimize the probability of an accident leading to control rod *1ection, and since the consequences of such an accident (if it did occir) do not lead to a serious potential safety problem, a holddown taechanism is not required. l l t p)
,,p*
pic f .y 3-75 (Revised T-21-67) ,} } j
O f (DEIJ.TED ) i
)
i l
)
1 l l 9 3-76 (aevised 7-21-67) 0000 218
, , s .
%e
O . (DEIEi'ED) 3.2.4.3.3 Control Rod Drive Control System The control system for the control rod drive is designed to energize and position the control rod drive, indicate the control rod assembly (CRA) position in the core, and indiente malfunctions in the s*; stem. As shown on Figure 3-56, the control system consists of Power supplies and moniters. - Clock (CRA speed standard). Control rod drive grouping panel. Individual CRA control logic. Position indicator system. Travel limit system. Automatta nquence logic. Trip tystem. Position deviation monitors. The coatrol rod drive control system provides the reactor operators with the flexibility of CRA grouping, manual or autoraatic group operation, automatic CRA group sequencing, and information of CRA position in the core. A total of 12 CRA groups is available through facilities of a control red l drive grouping panel which enables up to 12 CPA's to be assigned to each ) group. Individual position indicators are provided for all 69 CRA's and are visible to the operator. The operating centrol panel includes four group position indicators. Associated with each of these four indicators is a switch which selects CPA position data from a single CBA in each i group. Three of the indierters are assigned to groups A, 3, and C, and the pt) r ,is assigned to groups D through L. In addition, individual CRA l n .- c - ., ! l Lh' i 3-77 (Revised 7-21-67) i i i u 0000 2"9 , L. j
selection is achieved through these switches for single CRA tri= by =anual switch action. CRA groups are progra==ed so that the pcVer peaking values listed in Table 3-1 are never exceeded. Autc=atic sequencing (group overlap) of groups A through D is provided and is available for autc=atic or =anual operater CRA =otien require =ents. It allovs a limited overlap of operation of nny two groups in a fixed sequence, but no =cre than two. Inputs frc= CRA position and travel li=its feed this system. Aute=atic and =e.nual centrol is provided. In "autc=atic", the selected control rod drive group receives an autc=atic cc==and signal frc= the re-acter control system. In "=anual", provision is =ade for operation of any individual CRA or group of CRA's. Manual er autc=atic operatien of four CRA groups in a preset sequence is provided as described above. Grouping is deter =ined at the control icd drive grouping panel prior to reactor ep-eration. The drive gate is part of the individual CRA control logic circuitry which 1 perfor=s the function of selection and gating. It receives inputs frc= the clocx, the IN and CUT centrol busses, =otion " Enable," and travel 11=- its. The drive gate sends pulses to the translater upon receiving (a) clock pulses, (b) " Enable" input, and (c) and In or CUT control signal. End travel li=its and the driver monitor provide inputs to step CRA otion. Output signals of the drive gate feed into the translater. This unit pro-duces the proper signals for the drive motor. Direction is deter =ined by the IN and OUT comm eds, and speed is deter =ined by the fixed clock fre-quency. , The position indication and travel limit syste=s censist of three different types of transmitters and produce two independent analog position signals and two independent limit signals. One of the devices, the IVDT, produces both positien and limit signals. Either source of signals can be used for the position and for the limit signals. Pesition output jacks are provided for a precision =eter and for cc=puter
=cnitoring. Calibration of the potentic=eter and the LVDT 10 accenplished by initial adjust =ents pri s to. installing the power package and also by making adjust =ents within the centrol cabinet. The limit switches are ad-justed prior to installation of the drive package.
A fault detection circuit =cnitors signals to provide extra protection against unwanted withdrawal and insertion motion. See Figure 3-56. The rod drive control syste= has two speed-limiting features. ,First, the
=ctor speed is limited by the frequency of the input power set by a clock er pulse generator. Second, this limit is followed by a speed-saturating circuit which has the inherent property of not responding to a frequency greater than 125 per cent of rated frequency. These features will pre-vent an over-frequency and overspeed of the drive.
In addition to speed limitation, the red groups have independent " Enable" signals and gatcs such that no =cre than two groups can be enabled , ' 3 'id (Revised T-21-67) na 00u) 202 9 .\ s
j O V simultaneously for withdrawal motion in accordance with the description 1 in 7.2.2.1. These two features, frequency linit and group " Enable" lim-its, hold the maximum withdrawal rate well bel s that analyzed in ik.1.2.3.
~
Trip is initiated by de-energi:in6 two series circuit breakers in each of two power sources (Figure 3-58). Each loss-of-voltage trip coil is fed l ' by a separate two-out-of-four relay circuit powered by four inputs from the reactor protection system. Failure of any two inputs causes trip. The manual trip pushbutton opens all trip circuit breakers. Test push-buttons are provided to test each circuit breaker action. O O 0000 221 3-78a (Revised T-21-67)
(DEC1'ED) () 3.2.b.3.h Centrol Rod Drive System Evaluation
- a. Design Criteria 1 The system vill be designed, ter:ed, and analyzed for cc=pliance with the design criteria. A preliminary safety analysis of the control rod drive motor control subsystem was conducted to de-ter=ine failures of icgic functions. It was concluded that no single failure in any CRA control vould prevent CRA insertica, nor cause inadvertent CRA withdrawal of another CRA cr CRA group.
- h. Materials Selecticn, Materials are selected to be compatible with, and operate in, the reactor coolant. Certified mill test reports containing chemical analysis and test data of all materials exposed to the reactor system fluid shall be provided and maintained for the control red drives. Certificates of ccepliance for other mate-rials and components shall also be provided.
- c. Relation to Design Temperature All parts of the control rod drive exposed to the reactor cool-ant are designed to operate at 650 F, although it is expected that all parts vill operate considerably cooler. Scme tests have been completed, and additicnal tests are planned, to close-(~' ly determine the operating temperature gradients throughout the drive mechanism during all phases of operation. These tests vill also provide an indication of the amount of convection that takes place within the water space of the mechanism. It is expected that the more significant temperature changes vill be caused by displacement of reactor coolant in and out of the mechanism vater space as the drive line is raised and lowered.
- d. Design Life The expected life of the control rod drive centrol system is:
(1) Structural portions, such as flanges and pressure housings, have an expected life of 40 years. (2) Moving parts , such as rack, pinions, and other gears , have an expected life of 20 years. (3) Electronic control circuitry has an expected life of 20 years. V 0000 222 [$\ I 3-79 (Revised 7-21-67) ."
O (DELE.) . 3.2.h.3.5 Control Red Asse=bly (CRA) Each control rod asse=bly is =ade up of 16 control rods which are coupled to a single Type 304 stainless steel spider (Figure 3-39). Each control 1 rod consists of an absorber section of silver-indiu=-cad =ium poisen clad with cold-worked. Type 304 stainless steel tubing and Type 30h stainless steel upper and lever end pieces. The end pieces are velded to the clad to for= a water and pressure-tight centainer for the poisen. The control rods are loosely coupled to the spider to per=1t max 1=um confor=ity with the channels provided by the guide tubes. The CRA is inserted through the upper end fitting of the fuel assembly, each control rod being guided by an incere guide tube. Guide tubes are also provided in the upper ple-num assembly above the core so that full length guidance of the centrol rods is provided throughcut the stroke. With the reactor assembled, the CRA cannot be withdrawn far eacugh to cause disengage =ent of the control rods frc= the incere guide tubes. Pertinent design data are shown in Table 3-20 Table 3-20 Control Rod Asse=bly Design Data Ite= Data Number of Pod Asse=blies 69 Nu=ber of Control Rods per Assembly 16 Outside Diameter of Control Rod, in. 0.hk0 Cladding Thickness, in. 0.018 Cladding Material Type 30h SS, cold-worked Poisen Material 80f. Ag, 15". In, 5". Cd Length of Poisen Section, in. 13h Stroke of Centrol Red, in. 139
- r ,,
h\ ! ..,; .*
' *, , 3-80 (Revised 7-21-67)
\,
}g} }}}
i i ! This type of CRA has been developed under the USAIC Large Reactor Develop- I ment Program and offers the following significant advantages:
- a. More uniform distribution of abscrber throughout the core volume.
. b. Shorter reactor vessel and shorter internals owing to elimina-i tion of control rod followers.
- c. Lower reactor building requirements owing to reduction of re-I actor coolant inventory.
- d. Better core power distribution for a given CRA vorth.
A CRA prototype similar to the BE design has been extensively test +riN { at reactor temperature, pressure, and flow conditions under the LRD pro - l gram. 1 . 9 The silver-indium-cadmium poison material is enclosed in stainless steel
- tubes to provide structural strength to the control rod assemblies. These rods are designed to withstand all operating leads including those re-i sulting from hydraulic forces, thermal gradients, and reactor trip decel-erstion. The cladding of the poison section also prevents corrosion and eliminates possible silver contamination of the reactor coolant.
The ability of the poison clad to resist collapse due to the system pres-j w sure has been demonstrated by an extensive collapse test program on cold-verked stainless steel rods. The actual collapse margins are higher than the requirements. ! Internal pressure and poison swelling are not expected to cause stressing or stretching of the clad because the Ag-In-Cd alloy poison does not yield a gaseous product under irradiation. Because of their great length and unavoidable lack of straightness, some slight mechanical interference between control rods and guide tubes must be expected. However, the ptrts involved, especially the control rods, are so flexible that only very small friction drags vill result. Sim- , ilarly, thermal distortions of the control rods are expected to be small because of the lov heat generation and adequate cooling. Consequently,
! it is not anticipated that the control rod assemblies vill encounter ! significant frictional resistance to their motion in the guide tubes.
Lifetime tests are being performed on a prototype CRA in the CRDL Facility 1 1 described in 3.3.3.1 and in accordance with the progrsa outlined in ! 3.3.3.h.1. Approximately 2,200 full-stroke cycles and 250 full-stroke j trips have been computed with the reference design CRA at reactor oper-I
- ating conditions of pressure, temperature, flew, and water chemistry.
This is approximately equivalent to 20 years of operation on the CRA. Evidence of contact was noticed on the lead-in tip of the control rod assembly, but no measuraL.4 amount of metal had been removed. Visual inspection of the' spider shows an insignificant amount of wear. At the,end of h10 full-stroke cycles and 50 full-streke trips (the equiv-alento,f)threeftpf4operationinoneassembly),theincoreguidetubes I dhh ((VI 3-81 (Revised T-21-67) 0000 224
ir. the fuel asse=bly were exa=ined. Wear = arks vere noted at the entrance 1 of the guide tubes, and these marks extended into the guide tubes approxi-
=ately 5 in. Approximately 7 =11s of metal had been removed longitudinally fren the guide tubes at the upper end. Since no change in the time re-quired for two-thirds insertion was noted over the duration of the testing perfor=ed to date, it is concluded that wear of the guide tubes and the CRA's vill not be of concern. These tests vill be continued to cc=pletion.
The methods and frequency of CRA in-service inspection as well as the cri-teria for replacement vill be deter =ined during the detailed design. O O - 3-81a (Revised 7-21-67) 0000 225 e ,3 3 i
33 TESTS nrD r:SPECTIONS Q 331 NUCLEAR TESTS AND INSPECTION 3 3 1.1 Critical Experiments An experimental progra=(61-63) to verify the relative reaet;vity ver: 3 of the CRA has recently been ec=pleted. Detailed testing estab.ished the worth of the CRA under various conditions similar to those fcr the refer-ence core. These para =eters include control rod arrange =ent in a CRA, fuel enrich =ents, fuel ele =ent gec=etry, CRA =aterials, act soluble bcron concentration in the =ederator. ' Gross and local power peakh g were also studied, and three-di=ensional power-peaking data vere taken as a function of CRA insertion. Detailed pea >9g data vere also taken between fuel asse=blies and around the water holes left by withdrawn CRA's. The experi= ental data are being analyzed and vill becc=e part of the experi= ental bench = ark for the analytical
=odels used in the design.
3 3 1.2 Zero Power, Approach to Power, and Power Testing Boron vorth and CRA vorth (including stuck-CRA verth) vill be dete m ined by physics tests at the beginning of each core cycle. Recalibration'of boron vorth and CRA vorth is expected to be perfor=ed at least once dur-ing each core cycle. Calculated values of baron verth and CRA verth vill be adjusted to the test values as necessary. The bcron vorth and C3A vorth at a given ti=e in core life vill be based en CRA position indication and calculated data as adjusted by experimental data. S e reactor coolant vill be analyzed in the laboratory periodically to de-temine the bcron concentration, and the reactivity held in bcron vill then be calculated frc= the concentration and the reactivity vorth of boron. The =ethod of =aintaining the hot shutdevn =argin (hence stuck-CRA =arcin) is related to cperational characteristics (lead patterns) and to the pcver-peaking restrictions on CRA patterns at power. The CRA pattern restric-tions vill insure that sufficient reactivity is always fully withdrawn to provide adequate shutdevn with the stuck-CRA =argin. Fever peaking as re-lated to CRA patterns and shutdown =argin vill be =enitored by reactivity calculations, and interlocks vill be provided to prevent CRA patterns that pruiuce excessive pover peakMg and/cr reduction of shutdown margin. Operation under all power conditions vill be =enitored by incore instru-
=entation, and the resulting data vill be analyzed and cc= pared with =ulti-dimensional calculations in a continuing effort to provi:le sufficient sup-port for further pcver escalations.
332 THERMAL AND hTDRAULIC TESTS rid EISIECTION 3 3 2.1 Reacter Vessel Flev Distribution and Pressure Drop Test A1/6-scale =odelofthereactervesselandinternalsvillbe'.testedto g measure W 9 m 0000 226' j *! A
O a. The flow distribution to each fuel assembly of the reactor core and V to develop, if necessary, devices required to produce the desired flow distribution. -
- b. Fluid mixing between the vessel inlet no::le and the core inlet, and between the inlet and outlet of the core.
- c. The overall pressure drop between the vessel inlet and outlet no:-
les, and the pressure drop becveen various points in the reactor
, vessel flow circuit.
The reactor vessel, themal shield, flow baffle, core barrel, and upper plenum assembly are =ade of clear plastic to allow use of visual flow study techniques. All parts of the model except the core are geometrically similar to those in the prototype reactor. However, the simulated core was designed to =aintain dynamic similarity between the model and prototype. Each of the 177 simulated fuel assemblics contains a calibrated flow no::le at its inlet and outlet. The test loop is capable of supplying cold water (80 F) to three inlet no::les and hot water (180 F) to the fourth. Te=perature vill be measured in the inlet and outlet no::les of the reactor =odel and at the in-let and outlet of each of the fuel asse=blies. Static pressure taps will be located at suitable points along the flow path through the vessel. This instru-mentation vill provide the data necessary to acco=plish the objectives set forth for the tests.
, 3 3 2.2 Fuel Assembly Heat Transfer and Fluid Flow Tests I /
B&W is conducting a continuous research and develop =ent progrs= for fuel as-sembly heat transfer and fluid flow applicable to the design of the reference reactor. Si 6 1e-channel tubular and annular test secticns end =ultiple rod as-semblies have been tested at the B&W Research Center. 3 3 2.2.1 Single-channel Heat Transfer Tests A large quantity of unifor= flux, single-channel, critical neat zlux data has been obtained. References to unifom flux data are given in BAW-lc8 and 3 2 3 2.3 of this report. The effect on the critical heat flux caused by non- 1 unifom axial power generation in a tubular test section at 2,C00 psi pressure was investigated as early as 1961.(28) This program was extended to includ9 pressuregof 1,000, 1,500, and 2,000 psi and mass velocities up to 2 5 x 100 D/hr-f t . (59) The effect on the critical heat flux caused by differences in the radial and axial power distribution in en annular test section was recently investigated at reactor design conditions.(60) Data were obtained at pressures of 1,000 1 ,500, 2,0 % , and 2,200 psi and at mass velocities up,to 2 5 x 10 6 lb/hr-ft b. The tubular tests included the following axial heat flux shapes where P 8 is 1 local to average power:
- a. . Unifom Heat Flux (P8) : 1.000 constant
- b. Sine Heat Flux (P/P) :
1 396 G 50% L' m { 3-83 (Revised T-21-67) 000'u ?_2 7. l
- c. Init: Peak Heat Flux (P/E) m = 1.930 9 25% L 1h
- d. Outt.t Peak Heat Flux (P/E)=ax = 1.930 3 75% L Tests of two additional nenunifor= 72 in heated length tubular tests were undertaken to obtain data for peaking conditions =cre closely re-lated to the reference design. The additional flux shapes being tested are
- a. Inlet Peak Heat Flux (P/E)=ax = 1.65 @ 26% L
- b. Outlet Peak Heat Flux (P/E)_ = 1.65 a 72". L These tests, still in progress, vill cover approximately the same range of pressure, = ass ficv, and AT as the =ultiple rod fuel assembly tests.
3.3.2.2.2 Multiple Rod Fuel Assembly Heat Transfer Tests Critical heat flux data are being obtained frc= 6-ft-leng, 9-rod fuel as-se=blies in a 3 x 3 square array. A total of 513 data points were ob-tained covering the following conditions: 0 5 al g 5 250 1,000 5 P 5 2,h00 0.2 x 106 5G 5 3 5 x 10 6 where O AT g= inlet subcooling, F F = pressure, psia 2 G = = ass velocity, lb/hr-ft 3 e gec=etry of this section consisted of nine rods of 0.k20 in. diameter en a 0.558 in. square pitch. Analysis of the last data of this set is in process. t'
', 0000'22'8 .
3-63a (Revised 7-21-67)
(DELETED) 3.3.2.2.3 Fuel Asse=bly Flev Distribution and Pressure Drep Tests Flov visuali::atien and pressure drop data have been obtained frc= a 10-times-full-scale =odel of a single red in a square flev channel. These data have been used to refine the spacer ferrule designs with respect to =ixing turbulence and pressure drop. Flow distribution in a square k-rod test asse=bly has been =easured. A salt solutica injection technique was used to deter =ine the average flev rates in the simulated reactor asse=bly cerner cells, vall cells, and unit cells. Interchannel mixing was obtained for the same asse=bly. These data have been used to confir= the flev distribution and =ixing relationships e= ployed in the core ther=al and hydraulic design. Addi-tional =ixing, flow distribution, and pressure drop data vill be obtained to i= prove the core power capability. The following fuel asse=bly gec=- etries will be tested to provide additional data:
- a. A 3 x 3 array identical to that for which critical heat flux data have been cbtained to provide additional interchannel mixing data,
- b. A h x 6 array divided in half by a perforated plate simulating adjacent fuel asse=blies to provide data en =ixing between as-se=blies.
- c. A full scale 15 x 15 rod fuel asse=bly to provide additional flow distribution, =ixing, and pressure drop infor=ation ap-plicable to a ec=plete assembly.
3.3.2.3 Precterational Testing and Posteterational Testing Ther=ccouples are included as part of the incere =enitoring system and will enable postaperational te=perature =easure=ents to be =ade at the entrance and exit of all 52 instru=ented fuel assemblies. The results 1 of these testa vill be cc= pared to the results of the =cdel tests used for design calculations. 3.3.3 FUIL ASSEMBLY, CCNTROL ROD ASSEMBLY, AND CONTROL RCD DRIVE MECHANICAL TESTS AND INSPECTION To demonstrate th .techanical adequacy and safety of the fuel assembly, centrol rod asse= c (CRA), and control rod drive, a nu=ber of functional tests have been Jormed, are in progress, er are in the final stages of preparatien. i 1 . 3-Sh (Revised 7-21-67) 0000~229'*'
3.3.3.1 Prototyre Testing A full scale prototype fuel assembly, CRA, and centrol rod drive is pres-ently being tested in the Control Rod Drive Line (CRDL) Facility located at the B&W Research Center, Alliance, Chio. This full-size loop is cap-able of simulating reactor environmental conditions of pressure, tempera-ture, and coolant flow. To verify the =echanical design, operating ecm-patibility, and characteristics of the entire control rod drive fuel as-sembly system, the drive vill be stroked and tripped in excess of expected operating life requirements. A portion of the testing vill be performed with maximum misalign=ent conditions. Equipment is available to record and verify data such as fuel assembly pressure drop, vibration character-istics , hydraulic forces , etc. , and to demonstrate control rod drive operation and verify scram times. All prototype cc=penents vill be exam-ined periodically for signs of material fretting, wear, and vibration / fatigue to insure that the sechanical design of the equip =ent meets reac-ter operating requirements. Preliminarytestresultsaregivenin3.2.h.3.5.l1 After the prototype fuel assembly has been tested under simulated reactor operating conditions, it vill be installed in the full-size, lov pressure loop to verify specific fuel assembly design data. These data include pressure drop, coolant interchannel mixing, and coolant velocity profiles. j 3 3.3 2 Model Testing Many functional improvements have been incorporated in the design of the prototype fuel assembl/ as a result of =edel tests run to date. For ( example, the spacer grid to fuel rod contact area was fabricated to 10 times reactor size and tested in a loop simulating coolant flow Reynolds numbers of interest. Thus, visually, the shape of the fuel red support areas was optimized uith respect to minimising the severity of flow Yor-tices. Also, a 9-rod (3 x 3) actual size model was fabricated (using pro- 1 duction fuel assembly materials) and tested at' 6h0 F, 2,200 psi, and 13 l fps coolant flev. Principal objectives of this test vere to evaluate fuel rod cladding to spacer grid contact wear, and/or fretting corrosien resulting from flow-induced vibration. A vide range of contact loads I (including s=all clearances) was present in this specimen. No significant wear or other flow-induced damage was observed after 210 days of loop operation. 3.3.3.3 Comperent and/or Material Testing 3.3.3.3.1 Fuel Rod Cladding Extensive short time collapse testing was perfor=ed en Zircaloy-k tube specimens as part of the B&W overall creep-collapse testing program. Initial test specimens were 0.h36 in. OD vi-h wall thicknesses of 0.020 in., 0.024 in., and 0.028 in. Ten 8-in.-long specimens of each thickness were individually tested at 680 F at sicvly increasing pressure until collapse occurred. Col'a;;se pressures fer the 0.020 in. vall thickness specimens ranged frca 1,800 to 3,200 psig, the 0.02k in. specimens ranged frem 2,800 to 3,200 ps!.g, and the 0.028 in. specimens ranged from h,500 to h,900 psig. The material yield strc=gth of these specimens ranged from 65,000 to 72,000 psi at reem te=perature, and was 35,800 ps'i at 680 F.+ l t L b. 3-85 (Revised 7-21-67) '. G B0 bs ... 2 4 n c . ! 4 6 '4" ! jr llilllLIl / j d _ . . . . _
Additional Zirealoy h short time collapse specimens were prepared with a
=aterial yield stress of T8,000 psi at roc = te=perature and 48,500 psi at 615 F. Fifteen speci= ens having an CD of 0.h10 in. and an ID of 0.365 in.
(0.0225 in, nc=inal vall thickness) vere tested at 615 F at increasing pressure antil collapse occurred. Collapse pressures ranged frc= h,hTO to h,960 psig. Creep-collapse testing was perfor=ed on the 0.k36 in. CD specimens. Twelve specimens of 0.02k in. van thickness and 30 specimens of 0.028 in. vall thickness were tested in a single autoclave at 680 F and 2,050 psig. During this test, two 0.02k in. van thickness specimens conapsed during the first 30 days and two collapsed between 30 and 60 days. None of the 0.028 in. vall thickness specimens had conapsed after 60 days. Creep-collapse testing was then performed on thirty 0.kl0 in. CD by 0.365 in. ID (0.0225 in. nc=inal van) specimens for 60 days at 615 F and 2,1h0 psig. None of these specimens collapsed, and there vere no significant increases in ovality after 60 days. Results of the 60-day, creep-collapse testing on the 0.h10 in. CD speef.-
= ens shoved no indication of incipient conapse. The 60-day period for creep-collapse testing is used since it exceeds the point of pr4-am/ creep of the =aterial, yet is sufficiently 1cug to enter the stage when fuel red pressure begins to build up during reactor operation, i.e., past the point of maxi =u= differential pressure that the clad would be subjected to in the reacter.
In order to help cptimi:e the final clat thickness, additional clad-col-lapse testing is scheduled for 1967 usiug specimens fabricated to the reference design fuel clad dimensions, material specifications, and oper-ating conditiens. 3.3.3.3.2 Fuel Asse=bly Structural Cc=ponents The =echanical design of the pr)totype can panel asse=bly is the result of an extensive can panel design and structural evaluatten program. The full-size, simulated loop, functional testing noted in 3.3.3.1 is expected to verify can panel design criteria. Prototype static and dyns=ic lead l testing is undervar to verify can panel structural adoiuacy for vibration, handling, operation, and seismic leads. l In the mechanical design of the spacer grids, particular attention is given to the ferru!e-to-fuel-rod contact points. Sufficient load =ust be applied to position the fuel rods and to minimi:e furi rod vibration, yet allev axial ther=al differential expansion, and net produce fretting wear in the fuel rod cladding. Static load and functional testing of the pro-totype grids vill de=custrate their adequacy to perfor= vithin the design requirements. 3.3.3.h Control Red Drive Tests and Inscection 3.L 3.L.1 Centrol Red Drive Developmental Tests The prototype rack and pinion, buffer seal drive is under development at 1 Itpe' BW Byarch' Center, Aniance, Ohio. , 3-86 (Revised 7-21-67) y'
i (DELEED)
\~e Wear characteristics of critical components, such as sleeve bearings, 1 pinion and rack teeth, snubber piston and sleeve, etc. , during tests to date indicate that naterial c;mpatibility and structural design of these components will be adequate for the life of the =echanism.
Subsequent to completion of the development program, the complete pro-totype control rod drive vill be subjected to environemntal testing under simulated reactor conditions (except radiation) in the Control Rod Drive Line (CRDL) Facility at Alliance. Environ = ental tests will include, but not necessarily be limited to, the following: Operational Tests Operating speeds. Temperature profiles. Trip times for full and partially withdrawn control rod assem-blies (CRA) for various flow-induced pressure drops across the CRA. Life Tests (With internals assembled to maximum misalignment permitted by drawing dimensions and tolerances. )
'+
) 2,500 partial stroke (75 per cent) cycles.
2,500 full stroke cycles. 25 partial stroke (ho per cent) trip cycles. 175 partial stroke (75 per cent) trip cycles. 200 full stroke trip cycles. Misalignment Tests 100 full strokes and 100 full stroke trips with internals tol-erances altered to 1.5 times marimum allevable misalignment. Coupling Tests Complete check of coupling operations after testing. The cycles above meet the total test requirements of 5,000 full strokes and 500 trips. The assembly vill be completely disassembled and inspected at various B&W facilities after completion of environmental tests. l e e' ,5 (
", , ,3 0000 737 3-87 (Revised 7-21-67)
3.3.3.h.2 Centrol Rod Drive Control System Developmental Tests A centrol rod drive motor control unit has been built in breadbcard form. 1 Following the testing of the breadbcard version, prototype circuits for plug-in modules vill be designed and tested. Testing vill consist of bench testing, life testing, and determining the effects of si=ulated failures. The simulated-failure testing vill be designed to verify the g safety analysis. The control rod drive centrol system vill be tested in conjunction with the control rod drive motor control to insure proper operation. Si=u-lated failure testing vill also be perfor=ed on the ec=bined system to insure that protective requirements are being =et. The positien indicator and limit switch subsystem has been built in pro-totype for= sad life-tested =echanically under expected ,enviren= ental conditions . Further testing, both =echanical and electrical, will be done under expected environemntal conditions at the S&W Research Center. Characteristics *,o be determined will include accuracy, repeatability, linearity, short term stability, and icng term stability. 3.3.3.h.3 Production Tests The finished control rod drive vill be proof-tested as a ec=plete system, i.e., =echcnis=s, motor centrol, and system control working as a system. This proof testing vill be above and beyond any develep= ental testing performed in the product development stages. l) Mechanism production tests will include
- a. Ambient Tests Coupling tests.
Operating speeds. Position indication. Trip tests,
- b. Crerational Tests Operating speeds.
Position indication. Partial and full stroke cycles. Partial and full streke trip cycles. O
'ih e 3-88 (Revised 7-21-67) 0003 2 H
O O Control system production tests will be performed as described in the fol-loving paragraphs. The finished hardware vill be systematically operated through all of.its operating modes, checked over the full range of all set points, and checked for proper operation of all patch plugs. This vill check com-pleteness and proper functionin6 of viring and components. The operating modes to be checked will include such things as automatic operation, manual group operatica, trim or single CRA operation, position indication of all CRA's, travel limit on all CRA's, trip circuit opera-tions, IN command, OUT ccamand, etc. The trip circuit or circuits will be tested by repeated operation. The overall trip time vill be measured. The accuracy and repeatability of the position indication and limit switch 1 systems vill be tested. Power supply tests vill be performed to determine the upper and lower cperating voltage and to prove immunity to switching transients. Fault conditions will be simulated to prove that no unsafe action results
' from defective ecmponents, circuits, or viring. Ability to detect unsafe fault conditions at the operating console vill be determined. Typical of faults to be simulated are (O
- a. Defective limit switch or circuit.
- b. Improper CRA group patch.
- c. Defective patch plugs.
- d. Defective group sequencer.
- e. Defective clock.
- f. Defective automatic control signal.
- g. Defective enmmand line.
- h. Defective fuses.
- i. Defective single CRA control circuit or switch.
J. Defective power supply.
- k. Defective motor translator. l1
- 1. Defective motor cable. I
- m. Defective position transmitter.
f h,hd 3-89 (Revised T-21-67)
1 l The finished hardware vin be visually inspected for quality of work =an-ship. This inspection vill include an exa=ination of'the enclosure, h cable entrances, dust-tightness, =aintenance features, drawers and cable retractors, fasteners, stiffeners, =odule counts, vire harnesses, and other s1=11ar details. 3 3.4 EGIRNALS TESTS AND INSFECTIONS The internals upper and lover plenu= hydraulic design vin be evaluated and guided by the results fro = the 1/6-scale =odel flow test which is de-scribed in detail in 3 3 2.1. These test results win indicate areas of gross flov =aldistribution and allov verification of vessel flow-pressure drop co=putations. In addition, the test results viH provide =easured pressure pulses at specific locations to aid in assessing the vibration response characteristics of the internals co=ponents. The effects of internals =1salign=ent vill be evaluated on the basis of the test results fro = the CRDL tests described in 3 3 3.h. These test results, when correlated with the internals guide tube fh n1 design, vill insure that the CRA vin have the capability for a reactor trip or fast insertion under all = odes of reactor operation in the reactor coolant en-viron=ent. These tests viH not include the effects of neutron flux ex-j posure. l l After co=pletion of shop fabrication, all internals co=ponents vill be l shop-fitted and asse= bled to final design require =ents. The asse= bled ! internals co=ponents vill be installed in a =ockup of the as-built reac-tor vessel for final shop fitting art. align =ent of the internals for the
=ating fit with the reactor vessel. Du=::ry fuel and CRA's vill be used to check out and insure that a=ple clearances exist between the fuel and in ternals structures guide tubes to allow free =ove=ent of the CRA through-out its full stroke length in various core locations. Fuel asse=bly =at-ing fit win be checked at all core locations. The du==y fuel and CRA's vin be identical to the production co=ponents except that they vin be =anufactured to the =ost adverse tolerance space envelope; even though the asse=bly veights vill be representative of the production units, the du==y co=ponents win not contain fissionable or poison =aterials.
Internals shop fabrication quality control tests, inspection, procedures, and =ethods win be s1=11ar to the pressure vessel tests described in de-tail in 4.1.k. With regard to the internals surreinance speci=en holder tubes, the =a-terial irradiation surveillance progrs= is described in 4.4 3 All internal components can be re=oved fro = the reactor vessel to allov inspection of all vessel interior surfaces (see 4.4.1). Internals ec=- ponents surfaces can be inspected when the internals are re=oved to the canal storage location.
.h e n 35 0000.235
< *d' _
1 3.h REFERENCES 1 (1) Putnam, G.E., TOPIC - A Fortran Program for Calculating Trans-port of Particles in Cylinders, IDO-16968, April 196h. l l (2) Avery, A. F., The Prediction of Neutron Attenuation in Iron- ! Water Shields, AEEW-R125, April 1962. (3) Bohl, H., Jr., eji,a_l,., P3MG1, A One-Dimensional Multigroup P-3 Program for the Philco-2000 Computer, WAPD-TM-272. (k) Bohl, H., Jr. and He.~phill, A. P., MUFT-5, A Fast Neutron Spec-trum Program for the Philco-2000, W'PD-TM-218. (5) Armster, H. J. and Callaghan, J. C., KATE-1, A Program for Cal-eniating Wigner-Wilkins and Maxwellian-Averaged Thermal Con-stants on the Philco-2000, WAPD-TM-232. (6) Marlove, O. J. and Suggs, M. C., WANDA-5, A One-Dimensional Neutron Diffusion Equation Program for the Philco-2000 Com-puter, WAPD-TM-2kl. (7) Honeck, H. C. , THERMOS, A Ther=ali::ation Transport Theory Code for Reactor Lattices, BNL-5826. (8) Cadwell, W. R., Buerger, P. F., and Pfeifer, C. J., The PDQ-5 O and PDQ-6 Programs for the Solution of the Two-Dimensional Neu-O tron Diffusion-Depletion Problem, WAPD-TM h77. (9) Marlove, O. J., Nuclear Reactor Depletion Programs for the Philco-2000 Couputer, WAPD-TM-221. (10) Lathrop, K. P., DTF-IV, A FORTRAN-IV Program for Solving the Multigroup Transport Equation With Anisotropic Scattering, LA-3373. (11) Joanou, G. D. and Dudek, J. S., GAM-1: A Consistent P1 Multi-group Code for the Calculatiot. of Fast Neutron Spectra and Multigroup Constants, GA-1850. (12) Baldwin, M. N., Physics Verification Experiments, CORE I, p28 and Initial Conversion Ratio Measurements, BAW-TM k5h. (13) Clark, R. H. and Pitts, T. G., Physics Verification Experiments, Core I, BAV-TM h55. (14) Clark, R. H. and Pitts, T. G., Physics Verifiestior Experiments, Cores II and III, EAW-TM h58. (lk-1) Spinks, N., "The Extrapolation Distance at the Surface of a 1 Grey Cylindrical Control Red," Nuclear Science and Engineering 22, pp 87-93, 1965 \
'"
- 1 3-91 (Revised 7-21-67)
(15) Clark, R. H., Batch, M. L., and Pitts, T. G., Lu= ped Burnable O Poison Program - Final Report, BAW-3h92-1. (16) Neuhold, R. J., Xenon Oscillation, BAW-305, 1966. (17) Wilson, R. H. and Ferrell, J. K., Correlation of Critical Heat Flux for Boiling Water in Forced Circulation at Elevated Pres-sures, The Babcock & Wilcox Company, BAW-168, Nove=ber 1961. (18) U.S.-Eurato= Joint R&D Progra=, Burnout Flow Inside Round Tubes With Nonunifer: Heat Fluxes, The Babcock & Wilcox Company, BAW-3238-9, May 1966. (19) Jens. W. H. and Lottes, P. A., Analysis of Heat Transfer Burn-out, Pressure Drop, and Density Data for High Pressure Water, ANL h627, May 1951. (20) Owen, ". 3., Factors for One-Sided Tolerance Limits and for Variable Sampling Plans, SCR-607, March 1963. (21) DeBortoli, R. A. , g al_. , Forced Convection Heat Transfer Burn-out Studies for Water in Rectangular Channels and Round Tubes at Pressures Above 500 psia, WAPD-188, Bettis Plant, Pittsburgh, Pennsylvania, 1958. (22) USAEC Docket 50-2hh, Exhibit D-3, entitled " Rochester Gas and Electric Corporation, Brockwood Nuclear Station Unit No. 1", , (Third Supple =ent to: Preliminary Facility Description and Safety Analysis Report, February 28, 1966). (23) Lee, D. H. and Obertelli, J. D., An Experimental Investigation of Forced Convection Burnout in High Pressure Water. Part 1, Round Tubes With Unifor: Flux Distribution, AEEW-R-213, August 1963. (2k) Mat:ner, B. and Griffel, J. , Bi=onthly Progress Report (PPR- 1 XIII-ll and 12-63), Task XIII of Contract AT(30-3)-187, Basic Experimental Studies of Boiling Fluid Flow and Heat Transfer at Elevated Pressures, for Novenber and Dece=ber 1963, January 27, 196h. (25) Mat:ner, B. and Griffel, J., Monthly Progress Report (MPR-XIII-6-63), Task XIII of Contract AT(30-3)-187, Basic Experimental Studies of Boiling Fluid Flov and Heat Transfer at Elevated Pressures, for June 1963, June 28, 1963. (26) Mat:ner, B., Monthly Progress Report (MPR-XIII-5-63), Task XIII of Contract AT(30-3)-187, Basic Experimental Studies of Boiling Fluid Flow and Heat Transfer at Elevated Pressures,
. for May 1963, May 31, 1963.
. O
. all<.
3-92 (Revised T-21-67) 003237 l
(27) Internal Me=o, Weatherhead, R. J. to Lottes, P.A., Critical Heat Flux (Eurnout) in Small Diameter Tubes at 2000 tsia, Dece=ber 29, 1958. (28) Svenson, H. W., Carver, J. R., and Kakarala, C. R., The Influ-ence of Axial Heat Flux Distribution on the Departure Fro = Nu-cleate Boiling in a Water Cooled Tube, ASME Pacer 62-WA-297. (29) Nonunifor= Heat Generation Experi= ental Progrs=, 0,uarterly Prog-ress Report No. 7, January - March 1965, EAW-3238-7, Joint U.S.- Eurato= R&D Progrs=, AEC Contract No. AT(20-1)-3238. (30) Hald, A. , Statistic t1 Theory With Engineering Atelications , John Wiley & Sons, Inc., New York, 1955 (31) Worth 1:w, A. G. and Geffner, J. , Treatment of Exterimental Data, J un Wiley & Sons, Inc., New York, 1943 (31-1) Sovring, R. W. , Physical Model, Based on Subble Detach =ent, and 1 Calculation of Stes: Voidage in the Subcooled Region of a Heated Channel, EPR-10, OECD Halden Reaktor Project, Dece=ber 1962. (31-2) Zuber, N. and Findlay, J. A., Averare Volumetric Concentrations in Two Phase Flow Systems, Presented at the ASME Winter Meeting, 1964 To be published in the ASME Transactions. ') (31-3) Mau:rer, G. W., A Method of Predicting Steady-State Soiling Vapor Fracticas in Reactor Coolant Channels, Bettis Technical Review, WAPD-BT-19. (32) Baker, 0. , Si=ultaneous Flov of 011 and Ga s , Oil and Gas Jour-nal, Vol. 53, pp 185 - 195,195k. (13) Rose, S. C. , Jr. , and Griffith, P. , Flow Properties of Eubbly Mixtures, AGMP Pacer No. 65-HT-38, 1965. (3h) Haberstroh, R. D. and Griffith, P., The Transition From the Annular to the Slug Flov Regime in Two-Phase Flov, MIT TR 5003-28, Depart =ent of Mechan 2. cal Engineering, MIT, June 1964 (35) Bergies, A.. E. and r,uo, M., Investigation of Boiling Water Flow Regimes at High Pressure, NYO-330h-8, February 1, 1966. (36) Notley, M. J. F., The Ther=al Conductivity of Colu=nar Grains in Irradiated UO 2 uel Ele =ents, AECL-1822, July 1963. (37) Lyons, M. F., et al., UO2 Fuel Rod Operation With Gross Central Melting, GEAP- U6 5 October 1963. (38) Notley, M. J. F., et al., Zircaloy-Sheathed UO2 Fuel Elements Irradiated at Values of Integral kde Between 30 and 83 v/c=,
, AECL-1676, Dece=ber 1962.
m., v o'l'. ' 0000 238 3-93 (Revised 7-21-67)
(39) Bain, A.S., Melting of UO2 During Irradiations of S1 art Duration, AECL-2289, August 1965 (kO) Notley, M. J. F., et al., The Longitudinal and Diametral Expan-siens of UO2 Fuel Elements, AECL-21h3, Nove=ber 196h. (hl) Duncan, R. N., Rabbit Capsule Irradiation of UO ,2 CVNA-1k2, June 1962. (h2) Lyons, M.F., el al., UO2 Pellet Ther=al Conductivity Frem Irra-diatiens With Central Melting, GEAP h624, July 196h. (h3) McGrath, R. G., Carolinas-Virginia Nuclear Power Associates, Inc. , Research and Develop =ent Program, Quarterly Progress Re-port for the Period April - May - June 1965, CVNA-2h6. (hh) Ross, A. M. and Stoute, R. L., Heat Transfer Coefficients Be-tween UO 2 and Zircaloy-2, AECL-1552, June 1962. O; l l l l l t \ O Q* , 1, . 3 93a (Revised T-21-67)
.s
}Qhh
(45) Hofftan, J. P. and Coplin, D. H., The Release of Fission Gases Frc= Uraniu= Dicxide Pellet Fuel Operated at High Te=peratures, GEAP 4596, Septe=ber 1964. (46) Spolaris, C. N. and Megerth, F. E'., Residual and Fission Gas Re-lease Frc= Uraniu= Dicxide, GEAP k314, July 1963 (47) Robertson, J. A. L., et al., Behavior of Uraniu= Dicxide as a Reac- . terFuel,AECL-603,19587 (48) Parker, G. W. , et al., Fissica Product Release Frc= UCo by High Te=- perature Diffusion and Melting in Heliu= and Air, CF-6D-12-14, ORNL, February 1961. (49) Daniel, R. C., et. al., Effects of High Burnup on Zircaloy-Clad, Bulk UO2 , Plate Fuel Ele =ent Sa=ples, WAPD-263, Septe=ber 1962. (50) Ble= eke, J. O. and Tedd, Mary F. , Uraniu= Fissica Product Production as a Functica of The.-C Neutron Flux, Irradiation Time, and Decay Ti=e, CRNL-2127, Part 1, Vol. I and 2. (51) Duncan, R. N., CVTR Fuel Capsule Irradiations, CVNA-153, August 1962. (52) Frost, Bradbury, and Griffiths (AERE Harvell), Irradiatica Effects in Fissile Oxides and Carbides at Lev and High Burnup Level.g Pro-ceedings of IAEA Sy=posiu= on Radiation Da= age in Solids and Reactor Materials, Venice, Italy, May 1962. (53) Gerhart, J. M., The Post-Irradiation Exa=ination of a pug -UO Fast 2 2 Reactor Fuel, GEAP-3833 (54) Atc=1c Energy Clearing Ecuse, Vol.10, No. 3, p 11. (35) Large Closed-Cycle Water Reacter Research and Develop =ent Progra= Progress Report for the Period, January 1 to March 31, 1964, West-inghouse E ectric Corporation, Fittsburgh, P3.., 1964, WCAP-3269-2. Also WCAP-3269-3 for pericd fre= April 1 to June 30, 19c4. (56) Physical and Mechanical Properties of Zircaloy-2 and -4, WCAP-3269-M, Figure 13. (57) Burgreen, D., Byrnes, J. J., ani Benforado, D. M., Vibration of Reds Induced by Water in Parallel Flev, Trans. AS4E 80, p 991,1958. (58) Iarge Closed-c/cle Water Reactor R&D P.cgrs=, Progress Report for the Period Jar m j 1 to March 31, 1965, WCAP-3269-12. (59) Burnout for Flev Inside Round Tubes With Nenuaifc= Heat Fluxes, BAW-3238-9, May 1966. (60) Nonunifor= Heat Generatica Experimental Progra=, EA'a-3238-13, July 1966. 3-9a U000 240
(61) Clark, R. H., Physics Verification Experiments, Cores IV and V, BAW-m-178,_, September 1966. (62) _ Clark, R. H., Physics Verification Experi ent, Core VI, BAW-TM-179, December 1966. (63) Clark, R. H., Physics Verification Experi=ent, Axial Power Mapping on Core IV, BAW- M-255, December 1966. (6h) Larsen, P. S., g g., DNB Measurenents for Utvards Flov of 1 Water in an Unheated Square Channel with a Single Uniformly Heated Rod at 1600-2300 esia, Proceedings of the Third Inter-national Heat Transfer Conference,. August 1966. O O 3-95 (Revised 7-21-67) f
O
/ 8 I
/
/
l _._ r [ / . a
- / Q 2
= / ,
/
S 2 M
/ /
~
i / m E2
?
! .E I /
5 3 5 0 s
/ / i i
/ / i s
i
/ /
~
i / s
/ / _ _ _
a s g g g 8
- g g s~
R s = - - -
.a oon.no.. 3 oo m m .1, io.io.2 m ..,
1 0000 242 l N- BORON CONCENTRATION VER5US CORE LIFE p ,' ', f' ,
- N FIGURE 31 THREE MILE ISLAND NUCLEAR STATION AMEND. 2 (10-2-67)
O Axial Power Profile For 55% Insertion is shown on i Figure 3-3 =l t I l l l l l l y l l s [I N l l \ I \ 0,2 1.o, .
< g El / )'
E.s. !* l l TE
*g I
' I l 15 E k I
.l 9
-y I l j< 1.u : '
i O i 13 ' 10 20 30 40 50 60 70 80 Rod Insertion, % I ( ;. ;f p - AXIAL PEAK TO AVERAGE POWER VERSUS XENOH OVERRIDE ROD INSERTION NE FIGURE 3-2
,} { tj }
THREE MILE ISLAND NUCLEAR STATION
O - - 1 I i 1.6 1.6 / N la. 1. k-
/
/ Nl \
l l te 1.2 / N 1.0 ' [ u o 0.8 ) N\ k q 0.6 l \ E / N \ O.4 N O.2 \ 0 : ,14k" h 10 20 30 40 50 60 70 80 90 100 110 120 130 140 150 Distance From Bottom of Active Fuel, Inches l 0000 244 i 1 AXlAL POWER PROFILE, XENON OVERRIDE RODS 55 PERCENT INSERTED 1' ,1 M FIGURE 3-3 THREE MILE ISLAND NUCLEAR STATION
l 1 I l l l l
+2 l
2
/
"o /
h ' 5807
/
i% g- 0 -- 7
/
/ \ 68F BE L8 1
/
l
/
- -2
! o 2 4 6 8 10 12 14 16 18 20 22 l
!bierator Boron Concentration, ppm Boron x 10-2 l
l (*g SI
. {' ,
s MODERATOR TEMPERATURE COEFFICIENTS VERSUS BORON CONCENTRATION EM FIGURC 3 4 THREE MILE 15 LAND NUCLEAR STATION 0000 245
i O
+2
.:r
,/
+1 2000 ppe. #
g p g p 00 p yp
- 1000 ppm O "
Y ( I h500 N I N oS ' N
-1 '
ff No Boron > O -2 100 200 300 koo 500 600 Moderator Temperature, F l l l 0000 246. I l I MODERATOR TEMPERATURE COEFFIC!ENTS VERSUS MODERATOR TEMPERATURE & VARIOUS BORON LEVELS MM FIGURE 3 5 1
\ ' ,'i s
-[ THREE MILE ISLAND NUCLEAR STATION
O 1 i l I
=
1
\
l I I l I l 1 I li i ;- I t 4 i
? n s
_6 $< i
?.
3 S S a = l
% *Jaened uoJ1nen i
g,4 . PERCENT INITIAL POWER YER5US TIME FOLLOWING TRIP N.E;E!T FIGURE 3-6 THREE MILE 15 LAND NUCLEAR STATION
.=~ o. 2 n >2..n 0000 247
O l 1 2.. I
- 2. . -
- 1. 4 -
- 2. 2 -
- 2. 0 -
1.5 - 1.6 - Lemer Core
-% , s /
,e'g N
,/,*
\ %
f
, *% / / / \ j
'IP, 4.4 -ks s ,/ -g / s
, s
, s s / \
sf
/
3,3 -
\
/ N/ *. des..--- _.
s s s%fs %_- ,g ' -
/ \
3.0 - 8~ s
/ \/ g
\
N #
/
/
\
/
/
g Usper ce e g
,f f
.6 - s.
4-
.2 -
.0 i i . i ,
0 1 2 3 4 5 Time (T). ears notes:
- 4. Power Ratie tamen 34 an. f ree toe one bottee of active f uel.
Casa I . se teeeerature iteration. f . l.400 7 Case 2 . Temperature storaties with i 1.1600 F. Case 3 - Teneerature iteration meth f . 300 F. -
- 2. Oscillation enetiated at 7 2 ears.
O, , EFFECT OF FUEL TEMPERATURE (DOPPLER) OH XENON OSCILLATIONS BEGINNING OF LIFE
,g t NM FICURE 3 6 o THREE MILE 15 LAND NUCLEAR STATION G ,, . ,.
4 AMEND.1 (7 2147) 0000 24g I
O
- 2. 4 2.2 -
* ~
Lower Core
- 1. 3 - --
,- #~
- 1. 6 - \
l.4 - I* 2 - P/P N l.0 -
.8 - 1
.6- ,I
.4- ! s.
s, Upper Core
.2-
.0 , ,
O I 2 ? Time (T). days Notes: l
- 1. Power Ratio taken 36 in, from top and bottom of active fuel.
Case 1 - Teeeerature iteration with 7, . l . 400 F. Case 2 - Temperature Iteration with i . 900 F. f
- 2. Oscillation initiated at T 300 days.
l EFFECT OF FUEL TEMPERATURE (DOPPLER) ON XENON OSCILLATIONS- NEAR END OF LIFE WM FIGURE 3 6.b THREE MILE 15 LAND NUCLEAR STATION AMEND.1 (7 2167)
.}
C s [. ' . l[ i
O
- 2. 6 -
- 2. 4 - ss Upper Core ,/ 's \
22- /
/
- 2. 0 - \
f f.8 - .. " % '
\
- 1. 6 -\
l \ g
'*~ \\ s g-PIP \ " i
\ , I
- 1. 2 - \p /
i \- s
'q, gs / / '
1.0 - V \ / g
\/ g
\
.s - \
I
\
\ l \
.6 - l 3
V
\
g
/ \
.4 - ,/
Lower Core f
\ / g
.2- s _ ,/ N
~'
.0 8 ;
g i 2 Time (T). days Notes:
- 1. Case I - Divergent oscillation (without temoerature iteration).
Case 2. Power ratio variation with control (without teaperatura iteration).
- 2. Oscillation initiated at T . 200 days.
CONTROL OF AXIAL OSCILLATION WITH PARTIAL ROD 1 WM FIGURE 3 6 c
,]
THREE MILE ISLAND HUCLEAR STATION g , AMEND.1 G 21671 1
O 100 90 - v. A n. Finite Sample y 9% Confidence 3U 80 - S 70 Y 9 o. 60 - O 50 -- , , , . 1.0 1.2 1.4 1.6 1.8 2.0 DNB Ratio l l 1 O POPULATION lHCLUDED IN THE STATISTICAL STATEMENT YERSUS DNB KATIO NE7"F FIGURE 3-7 { [' ' , THREE MILE 15 LAND HUCLEAR STATION
! 0000 251
1.8 l ' ll l.7 [ CK \,
- P/P = 1 70 (Pan ial Rod i
- 1. o,
, Insertion) l ,
1.5 ' '
' 'i
's [ P/ ' I= 1 50 '
! 1.h l y / (M dified Cosine)
\
1.3
/ \
1.2 ' I / l.1 \ 1.0 m l l\ \
't.
a 09 f i l
\, \
0.8 ' '
/
/ l' II
\\ \
07 I i 0.6 l f
/ l N s e
O.5
/ -'
- I' ,
I % s
's..-
0.4 - l C3 l \' O.2 I L Fuel Mid-Place f k \
' Core Core h 0.1 I
I igga l l i
)
~
, o', o i ! ! I O ' ' l
- 20 ko l 60 80 100 l 120 l 140 l 10 30 50 70 90 110 130 Distance From Botton of Active Fuel, Inches POWER SHAPE REFLECTING INCREASED AXIAL POWER PEAK FOR 144 INCH CORE N.22tF FIGURE 3-8 THREE MILE 15 LAND NUCLEAR STATION Di]00 252
[,' (
O 1 90 1.85 - 1 70 - -
\ Line A (Design) 1.60 -- y
\
1 50 -- Line B (Nominal - &ximum Calculated) 1.40 " N Line C (Typical True Distribution) u N Sm 130-- - Eh N
" .s 1.20 -- s YN
*5 1.10 -- N 5 N
\
h 1.00 --
~d
^ 0 90 --
N 0.80 - N
\
'. 0 70 --
\
0.60 -- N 0 50 - . gl) Line A = y*, x Line B (2) Line C for illustration only 0.h0 -- (3) Line B is based on detailed data N from a rod by rod printout of a PDQ (two dimensional power and 3 0 30 -- flux calculation for vorst time \ 0.2C -- in core life) s 0.10 , , , , , 0 10 20 30 k0 50 60 70 80 90 100fi Percentage of Fuel Rods with Higher Peaking Factors Than Point Values 0000 25.3 O DISTRIBUTION OF FUEL ROD PEAKING _ M__M FIGURE 3 9 THREE MILE ISLAND HUCLEAR STATION
.i.s i' ' .S
- I et 8 i
O 200 - l 180 -
@ l c
e 160
% Line 1
[ g 140 Ech Nuclear - 1.85
,T l
l
/
)
8 no 2 um a
. I la ,
loc Fah Nuclear - 1,79 g .
$ l 80
/ wm 5
$ 60
)
/ / M 'P " '
- c
/
/ / .
[ ko - l 2 20 o l 100 102 104 1m log Ho 112 uk 116 nS 120 Rated Power (2452 Wt), % l l l l l l s
' i POSSIBLE FUEL ROD DN8's FOR MAXIMUM DESIGN CONplTIONS. 36,816 ROD CORE
.s EM FIGURE 310 THREE MILE 15 LAND NUCLEAR STATION
- 0000 254 ,
O 100 j 80 E l H A 1 e 70 : D 60 l /
- I 2 I j 50 E l 7 G
ko
' /
Fob Nuclear - 1 79, a 30 E M4m m # ' 20 - F [ o lo m u f
.8
/ I o I 100 102 lok 105 108 no n2 n4 n6 n8 120 Rated Power (2452 Mit), %
0000 255 O POSSIBLE FUEL ROD VNB's FOR MOST PROBABLE CONDITIONS 36,816 ROD CORE NE F!GURE 3-11 0.' i' THREE MIL P, .I,'. AMG NUCLEAR STATION
(1-P) (P) 0.1 ! 09 i i ' I i L' l 0.01 0 99 g , , i i
\ i i e i
\ j i i i
\ ! i
\l I
\
0*001 @ 0 999 x 1 114j i , 1 -
\ f, s i i
\ \i i
\ \i i I
\ \ l l l O'
\ I\ ,
i
\
0.0001 1 0 9999
\ \
- , #\ i !
1
]
i xix I \N 0.00001 NN%%N 1 ! ~ 0 99999 0 10 20 30 40 50 60 80 TO 90 100 Percentage of Rods with a Lower Value of P l l l 5' CISTRIBUTION OF POPULATION PROTECTED P, & 1 P VER$US NUMBER RODS
, FOR MOST PROBABLE CONDITIONS
.,gi
_NM FIGURE 3-12 , THREE MILE 15 LAND NUCLEAR STATION
O 1.6
- . o.. . - - ,
1 15 s 3 a /- a a a 1.u 3 _ ws_ __A a 1 s 13 8 i 6 4 3 7 1.2 m n O u.
- 1.1 3
m a 1.0 - - - - - - - - l 1 100 110 120 130 140 150 Rated Power, (2452
- t),%
O "- OUU Sl DNS RATIOS (B AW-168) VERSUS REACTOR POWER
~
k'- ' i.. NM FIGURE 3-13
.. THREE MILE ISLAND NUCLEAR STATION
+ 24 79
+ 22 l ) 81 e
l
+ 20 83
+ 18 85 g
/
h
+ 16 88 .5 Design g Overpower g*
+ 14 91 Ey w
a
+ 12 94 w ]4%
5 j + 10 , 98 E Ey o + 8 r 102
+ 6 r
/ 108
.c w Y
ya
. 4 ,
/ m ,
e-
+ 2 r 129 Quality O 14k Subcooled
- 2
- 4 l
- 6 100 110 120 130 140 150 Rated Power (2452 wt), %
l l MAXIMUM NT C'HANNEL EXIT QUALITY VER5US REACTOR OVERPOWER NM FIGURE 314 ,i THREE MILE ISLAND NUCLEAR STATION 0000 258
i l l l O . l 4.00 l l Uo M'1t* " 2 ca I d i h
>u l j 3 00 m !
A 6 d I a t a l 1 0 l
.O l
/ I
/ I x s
/ \
Data Based on CVNA - 142 _ June 1962
- 1.
- I I i i .
o 1000 2000 3000 4000 5000 hmperature, F O 0000 259 THERMAL CONDUCTIVITY OF Uo 2 i ' _MM FIGURE 315 f. THREE MILE 15 LAND NUCLEAR STATION
, s' s
O 6000 I UO Melt 1.ng l Te::rperature l 5000 - - - -- - - - ---- -- h-- l l
~
kooo /
/ l i
5 l l 143% l Power g g u 3000 i ! 3 { l \ e I i u ' l
~ l l 100% Power l l 2@
Nj l i 9 I I uus Power I i 1000 N, I I l l l l l 0 l l 0 5 10 15 20 25 Linear Heat Rate, hv/ft
.oro.
FbEl CENTER TEMPERATURE AT THE HOT SPOT VER5US LINEAR POWER NM FIGURE 316 THREE MILE 15 LAND HUCLEAR STATION
l l 1 1 O 70 - _ Gaussian Distribution 60 - / 3 _ 3 50 - - 40 - k -
= 30 - _
20 - - O 10 - 0 - -[ [ i i i i i 6 e i i . i i i 0.6 0.7 0.6 0 9 1.0 1.1 1.2 1 3 .1.4 15 1.6 1.7 1.8 C
- O :- 0000 261 l
l , NUMBER OF DATA PolNTS VERSUS CE /DC NM FIGURE 3-17 THREE MILE ISLAND NUCLEAR STATION l {
0 1.025 - 1.020 -
$ 1.015 -
E
" 1.010 -
F h 1.005 - q 6 o 1.000 - 60 70 80 90 100 -
=
995 - 990 - 985 - F A G Population Protected, % Hot Channel Factoi' Versue Per Cent Population Protected i l [
=s ,
e U HOT CHANNEL FACTOR VERSUS P,ERCENT PQqULATION PROTECTED g NM FIGURE 3-18 g n3 9fg THRE ! MILE 15 LAND NUCLEAR STATION v UJ LUC 1 l l
O 100 Infinite Shaple 100% Confidence
# 90 - Finite Sample g 90% Confidence 3o 3 80 -
b i a 3 Finite Sample Y 99% Confidence
'5 70 - )
l o a.' '
\ ,g l
50 . . i i i 1.0 1.2 1.4 1.6 1.8 2.0 Burnout Factor (DNB Ratio) ) 0000 263 BURHOUT FACTOR VERSUS POPULATION FOR VARIOUS CONFIDENCE LEVEL 5 t\ #
,',- LMatsyry FIGURE 3-19 THREE MILE ISLAND HUCLEAR STATION
~
O 200 180 4
. l I
l _ Mo
- a l f
t c 140 l 3 , 3 1
$ no l l F
8 l 3 100 1
- I
~5 a: I 80 ;
I . 60 I [ l ko 20 ' / I l o l 1.00 1.05 1.10 1.15 1.20 Fraction of Rated INrder (2452 Wt)
')V g ),.i, , d' RODS IN JEOPARDY VERSU$ POWER NE FIGURE 3-20 THREE MILE 1$ LAND NUCLEAR STATION 0000 264
lo _ _ O __-- __ o i e i i r-f~ i i i TT i i i 6 R i i . . . . . 4 i i 0 .2 .4 .6 .8 1.0 1.2 1.k 1.6 1.8 2. o> 2 (1) */*e WAPD-188 500 psia Data and BAW-168 20 - 10- - a - S _ _ _ E __ u o m , o e i i 4 . 4 i i . . 4 4 4 i i 6 i 4 6 4 i
'* 0 .2 .4 .6 .8 1.0 1.2 1.4 1.6 1.8 2.o> 2 (2) +/*c WAPD-188 600 psia Data and BAW-168 20 7 _
- 10. _
r __ o r- 4i i
~
% , m i e i i i 6 6 4 i i e i i i a i 4 4 i 0 .2 4 .6 .8 1.0 1.2 1.4 1.6 1.8 2. o> 2 (3) og.c WAPD-188 1000 psia Data and BAW-168 lo- _
o i i i i
~
i mb i i i i i i i i e i e 4 i e e e o .2 .4 .6 .8 1.0 1.2 1.4 1.6 1.8 2.o> 2 (4) V*c WAPD-188 1500 peia Data and BAw-168 0000 265 O RATIC OF EXPERIMENTAL To 1 l CALCULATED BURNoUY HEAT FLUX NM FIGURE 3-21 ! 2ti ,
',s THREE MILE 15 LAND HUCLEAR STATION l
i l L _ - )
10 _
~_""
O r- "k, i s s s e s s o i s . > r s a s *
- s e n 0 .2 .16 .C .d 1.0 1.2 1.k 1.6 1.8 2.C> 2 (5) */Sc WAPD-188 1750 psia Data and BAW-168 70 - -
a y 60 - E o _ h 50 i
= _
i ho - G 30 - 20 - _ l 10 _ - o m r, n I. r,
, i i i i . . . i i i > i i i i i > i i o .2 .h .6 .8 1.0 1.2 1.4 1.6 1.8 2.o> 2 (6) +/*c WAPD-188 2000 psia Data and BAW-168 .~. ;
g RATIO OF EXPERIMENTAL TO CALCULATED BURHOUT HEAT FLUX h N.12"F FIGURE 3 22 THREE MILE ISLAND NUCLEAR STATION 000) 266
lo -- o
.n.-n e-n o .2 .4 .6 .8 1.0 1.2 1.4 1.6 1.8 2.o> 2 (7) og*c WAPD-188 2250 psia Data ami BAW-168 lo - I o
i i i . . . i i . . i i n A. .r-O.
. i . . i o .2 .4 .6 .8 1.0 1.2 1.4 1.6 1.8 2.o> 2 l j (8) *g*c M WAPD-188 2!00 psia Data t.ai SAW-168 !
o l
" l lo - ' O' o r r-O,-O, l
I i s i i e i i i s i i i i e i i e i i . , i o .2 .4 .6 .8 1.0 1.2 1.4 1.6 1.8 2.o> 2 (9) *g/*c ! WAPD-188 2750 psia Data and BAW-168 l 20 - 10 - - - o Ib _ _ a i e i . i i i i i . . iiiiiiiii l 0 .2 .4 .6 .8 1.0 1.2 1.4 1.6 1.8 2.o> 2 (10)
*E!'C AEEW-R-213 560 psia Data ami BAW-168 l ,'ts-0000 N7 l l RATlo 0F EXPERIMENTAL To l CALCULATED BURNOUT HEAT FLUX
(;,'
'< #M FIGURE 3 23 I THREE MILE ISLAND NUCLEAR STATION 1
r lo - d o
. . . . . i iiiiiiii 1 nn n iiiiii o .2 4 .6 .8 1.0 1.2 1.4 1.6 1.8 2.0> 2 (n) ./*c AEEW-R-213 720 psia Data and BAW-168 l
So - l 40 - B S 2
, 30 -
84 3 ~ 9
= co _ _
lo - __ _ o 1 n r, n i i e i i i i i e i i i i i e iiiiii o .2 .h .6 .8 1.0 1.2 1.4 1.6 1.8 2.0> 2 (12) +g/*c AEEW-R-213 1000 psia Data and BAW-168 lo - o r-n- i fTT1 i i i i i i e i 4 6 e i iiie i e i i o .2 .4 .6 .8 1.0 1.2 1.4 1.6 1.8 2.o> 2 (13) */*c E AEEW-R-213 1300 psia Data and 3AW-168
, i.
RATIO OF EXPERIMENTAL TO CALCULATED BURHOUT HEAT FLUX NM FIGURE 3 24 THREE MILE 15 LAND NUCLEAR STATION 0000 2f38
)
Q 30 - _ l 20 - lo - E o
~
g i i i . . .7ii. . . . . . . . . ... - u o .? .4 .6 .8 1.0 1.2 1.h 1.6 1.8 2. o> 2 (14) oge
=
AEEW-R-213 1500 psia Data and 3AW-168 10 - l O 1
, . . , i . nini , m n O- o .2 .4 .$ .8 1.0 i i i 1.2 1.4 4 i i 1.6 i i.i 1.8 2.o> 2 ,
(15) ! l *E 'C Columbia 500 psia Data and EAW-368 10 - 0 - i i i i . . . . . . 7 m. . i i i i e ii o .2 .4 .6 .8 1.0 1.2 1.4 1.6 1.8 2.o> 2 (16) *g/*e ( Columbia 720 psia Data and BAW-368 0000 269 RATIO OF EXPERIMENTAL TO CALCULATED BURNOUT H2AT FLUX o. r MM FIGURE 3 25 THREE MILE 15 LAND NUCLEAR STATION _ l_ _
50 - $ l ko _ 30 - _
~
l
~
l 3 20 - 2 2 - O u lo - i a _ o a a a e a i i i i Ei . . i i m e e . i 6 6 o .2 .h .6 .8 1.0 1.2 1.4 1.6 1.8 2.o> 2
)
(17) '[*C Columbia 1R] psia Data and BAW-168 10 - 0 -f %i l i i i e i i i i i i i i i e i e i e i i o .; .L .6 .8 1.0 1.2 1.k 1.6 1.8 2. o> 2 (18) C Columbia 1200 psia Data and BAW-168 0li , ratio 0F EXPERIMENTAL To .
~)
CALCULATED BURNOUT HEAT FLUX hm FIGURE 3 26 THREE MILE 15 LAND NUCLEAR STATION 0003 270
20 q O _ 10 - 0 [ , i i i i i i i
.h i i i i i t e i e e ,ii 0 .2 .6 .8 1.0 1.2 1.4 1.6 1.8 2. O> 2 (19) */*e Columbia 1500 psia Data and BAW-168 60 -
50 - 3 - 3 2 O
- u uo -
y - 30 - 20 - 10 - l
)
l 0 i e i i r-i c. r"r
<t i i e
! 1 iiiiiiiiie i 0 .2 .h .6 .8 1.0 1.2 1.4 1.6 1.8 2.0x (20) +3/4g
]QQ.} }[j Argonne 2000 psia Data and BAW-168 RATIO OF EXPERIMENTAL TO l
CALCULATED BURNOUT HEAT FLUX r' 3 (- NM FIGURE 3-27 THREE MILE ISLAND HUCLEAR STAT 10H
10 - _ 0 0 .2 4 .6 .8 1.0 1.2 1.4 1.6 1.8 2.0> 2 (21) 'g/'c B&W 2000 psia Data and BAW-168 3 S 2 0 10 - 5. O c-f"]
. . i
[@, , i i . . . . . . . i 0 .2 .4 .6 .8 1.0 1.2 1,4 1.6 1.8 2. C> 2 (22) og+C .\ Euratom 1 COO psia Data and BAW-168 l 10 - 0
, , . . . i I- ,
i i i i , i i i i e i i i i 0 .2 .h .6 .8 1.0 1.2 1.4 1.6 1.8 2.0> 2 C (23) Euratom 1500 psia Data and BAW-168 r . RATIO OF EXPERIMENTAL TO CALCULATED BURNOUT HEAT FLUX o AfstM FIGURE 3 28 THREE MILE ISLAND NUCLEAR STATION 0000 272
20 - 0 10 _ 0
. d m.
~
O .2 .4 .6 .8 1.0 1.2 1.4 1.6 1.8 2. 0> 2 (24) .g/6e Eurateus 2000 paia Data and BAW-168 e f2 50 - < y k0 -
~
30 - 1 20 - , r - - 10 - ._ _ _- 1 l O m
. . . . . . . . .I . . . . . . .....
0 .2 .4 .6 .8 1.0 1.2 1.h 1.6 1.8 2.0> 2 ) (25)
.g. c n1 500-720 p.t. nata maa Baw-168 ~
0000 273 0~ .- .. RATIO OF EXPERIMENTAL TO CALCULATED BURNOUT HEAT FLUX l., . , ihEFfF FICURE 3-29 THREE MILE 1$ LAND HUCLEAR $TATION
O 80_ 70-60-3 50~ A 2 u h0-sc 30- _) 20 - ~ _ 10- _
' ' ' ' ' * + i 6 4 i , i .
0 .2 .4 .6 .8 1.0 1.2 1.k 1.6 1.8 2. 0> 2 (26) V*c All 1000 pois Data ard BAW-168 i
'\
i . .i. f RATIO OF EXPERIMENTAL TO , l CALCULATED BURNOUT HEAT FLUX NE FIGURE 3-30 THREE MILE 15 LAND NUCLEAR STATION }{ ,} l
O 80 - 70 - 60 - 3 50 - 3 2 s. ho - i
=
30 - 'O - 20 - 10 - o rf"a .,., n., n s , , , , e s , e o a s , s , , n, ,o e o .2 .4 .6 .8 1.0 1.2 1.4 1.6 1.8 2.o> 2 (27)
*/*c All 1500 psia Data and BAW-168 0000 275 RATIO OF EXPERIMENTAL TO CALCULATED BURNOUT HEAT FLUX d -
NM FIGURE 3-31 THREE MILE ISLAND NUCLEAR STATION
1 100 # g 90 - 80 - 70 - a Y
- 60 -
2 u _ 50 - 40 - 30 - 20 ~ ~ 10 - _ 0 iiiiiiii>ie r-i . hn i . 4 i i n e i i 0 .2 .4 .6 .8 1.0 1.2 1.4 1.6 1.8 2. 0> 2 (28) er !'c di , All 2000 psia Data and BAw-168 RATIO OF EXPERIMENTAL TO CALCULATED BURHOUT HEAT FLUX CGM FIGURE 3 32 I THREE MILE 15 LAND HUCLEAR STATION 0(100 276 l '
.it' l
1
l 1 l 1 y- _
\
l 1@ O - I-90 - l 80 - To - _
$ 60 -
2 u - 50 a _ A V to - - 30 -
~
20 -
~
10 - _ i. i i i i .. ...1 . . . . . . . . . . . o .2 .4 .6 .8 1.0 1.2 1.4 1.6 1.8 2. o> 2 - (29) Y'c
..All 1750-2750 psia rata and BAW-168 l
l ratio oF EXPERIMENTAL To
. ;e-CALCULATED BURNOUT HEAT FLUX
#M FIGURE 3 33 G\ . o 0000 277 raaEE =ita isLAso suCLEAa sTArios
10 - O 0
. . . . . . b. . ,.n,-O. . . . . . . . . . .
0 .2 .4 .6 .8 1.0 1.2 1.4 1.6 1.8 2. 0 > 2 (30) $ge Eurstom Chopped-Cosine 1000 psia Data and BAW-168 10 - 0 5 0 l u . . . . . e b.. ii.
. ... ...i u O .2 .4 .6 .8 1.0 1.2 1.4 1.6 1.8 2.0 > 2 I
j (31)
*/*c g
l Euratom Chopped-Cosine 1500 psia Data 10 - and EAW-168 0
.2
. . . . SY. . .
.. . . . . , , , , , , , i 0 .4 .6 . 1.0 1.2 1.4 1.6 1.8 2. 0 > 2 Euratom Chopped-Cosine 2000 psia Data and EAW-168 10 -
0 i . . . . . k. . %. , . . R . . . i i i e i . # # 0 .2 .4 .6 .8 1.0 1.2 1.h 1.6 1.? P.0 > P (33) ege Earatom and B&W Inlet Peak 1000 psia Data and BAW-168 RATIO OF EXPERIMENTAL TO CALCULATED SURNOUT HEAT FLUX ME FIGURE 3 34 THREE MILE 15 LAND NUCLEAR STATION ss .,
. 00>00 278
O 2 - o PM i i . . . . . . . . . . . . . i . . . o .2 .h .6 .8 1.o 1.P 1.4 1.6 1.8 2.0 :e (34) */*C E Iuratom and B&W Inlet Peak 1500 psin Data and BAW-168 10 - a a o & i a i e i i e i i i i i . 4 4 . . . 6 . . Ji 0 .2 4 .6 .8 1.0 1.2 1.k 1.6 - 1.8 2.0 >2 E C o u (35) Euratom and B&W Inlet Peak 2000 psia Data and EAW-168
~
lo - O o ltlHnM i e a i i i i i . 6 i i e i i i i i e i i o .2 .h .6 .8 1.0 1.2 1.h 1.6 1.8 2.0 >2 (36)
*_/*C r
Eurstom and B&W outlet Peak 1000 psia Data and BAW-168 10 - o i i 4 . 1"
.iiii,i4 + 4 . i , , . . 4 o .2 .h .6 .8 1.0 1.2 1.h 1.6 1.8 2.0 >2 (37)
*/*e g
Euratom and B&W outlet Peak 1500 psia Data ami BAW-168 l i 0000 279 RATIO 0F EXPERIMENTAL To CALCULATED BURNOUT HEAT FLUX ME FloURE 3 35 r . . - I
' THREE MILE ISLAND HUCLEAR STATioH
. . - [6 f e
lo - __
~
o n E G
, , i i i i i i i . . .
4 6 . . . . , , i o .2 4 .6 .8 1.0 1.2 1.4 1.6 1.8 2.0 m (38) ogec Euratom ar.3 B&W outlet Peak 2000 psia Data and BAW-168 10 -
~
o '-l ['] lll i i e i i i i i i i i i i . i e i e i i . 0 .2 4 .6 .8 1.0 1.2 1.4 1.6 1.8 2.0 de (39)
*g/*c All 1000 paia Non-Uniform Data and BAW-168 3
a 10 , l % _ l 3 0 - q 9 i i i i i i . i i 6 i . 4 i i i i i i i
* .2 o .4 .6 .8 1.0 1.2 1.k 1.6 1.3 2.0 x (40) */$c A;.1 1500 psia Non-Uniform Data and BAW-168 20.
lo- - o n _ l
, t i i , 4 4 i i i i e i e i e i i e i i o .2 4 .6 .8 1.0 1.2 1.4 1.6 1.8 2. ode (41) *g/*c All 2000 psia Non-Uniform Data and BAW-168 os .
ratio 0F EXPERifENTAL To CALCUL*AJ I D BdR$ouY HEAT FLUX M at m FIGURE 3 34 THREE MILE ISLAND NUCLEAR STAT!oH 0000 280 P -
l O ! i
+18 88
/ i/ ,
x a f esign D ! $
+14 ower :11a; e
/ 93 o
y f
+12
/ 7 ,
/ 96
?y
+10 ,/ 100 g g;
+8 / /g/ 7 104 EE g5
* /, / 2
$ +6 '
'O -
/ /
, log "5i
+4 116
+2 r
/ ,// , 125 li x
/
/
/
/[/'
/
quality subcoolea 1"
} ~D
,3 j 5
// \ s
.g
[ - - - 2120 pais 2185 peig (Fah-1.85) f* (FS-1.85 )
- 2185 peig (F9-179)
-6 100 llo 120 130 14 150 Rated ?cver (2,kS2 .Wt) '.
0000 281 MAXIMUM NOT CHANNEL EXIT QUALITY VERSUS REACTOR POWER (.Q, NE FIGURE 3-37
"J,'
"e,
;, THREE MILE ISLAND HUCLEAR STATION l
O i lI i i I l 5% Bel w Average Assembly Flow
+12 --- Average Assembly Flow lI I I
+10 l
+8 T#h " 1*I
+g ) /
,g > > /
l l / s j /
/
Cuality 0 j Subcooled
-2 / // / s
+ Fo2 = 1 79 l/ /
l Ft.h = 1.85 j /
' .- i i
-6 / I [ Design Overpower (114%)
-8 l 100 110 120 130 140 Rated Power, (2452 Wt) %
l l l I IHOTTIST DE31GN & NOMINAL CHANNEL EXIT QUALITY VERSUS REACTOR POWER (WITHOUT ENGlHEERING HOT CHANNEL FACTOR $) NM FIGURE 3 38 THREE, MILE ISLAND NUCLEAR STATION } (jl}.] 2 h 2
3.0
\
\
lluf Power \
\
l 2.5 130% Power \,
. 2.0
'S M
O l i 1 3 1.5 '
$ Bubble to I u $1ug 3 }
(Griffith and l Rose) \ ll } Bubble to O j l.0 '. Annular g (Baker)
\
Slug to Annular
/ (Haberstroh) \
I \
.5 , I. g Bubble to Slug N
( Baker) N 3 0 10 20 30 40 Quality, (Ib vapor / total Ib). % l REY: 7- 21-67 MORE TRANSITICN DETAIL ADDED. l FLOW REGIME MAP FOR UNIT CELL CHANNEL l AT 2120 PSIG l WM FIGURE 3 39
,f [ , ,
- g. , '
THREE MILE 15 LAND HUCLEAR STATION AM END.1 (7-2167 1
a A Bundle Burnout Test Conditions where stable operations were l observed A Hot Unit Cell Worst Conditions
+ Hot Unit Cell Nominal Conditions 3.0
\
$ 4 AA A \
\
\
l
'g 2. 5
. \
" A I A A M A
$ A O 2
- 2. 0 -
4 T. _% A d E a r 2 i5 l t Bubble g; to Slug l (Grif fith and l Ro se) 1.0 l ' Bubble to Annular (Baker)
/
.5 Bubble to $1ug (Baker) 0 10 20 30 40 Quality (lb vapor / total Ib). '.
FLOW REGIME MAP FOR UNIT CELL CHANNEL NM FIGURE 3 39 a THREE MILE 15 LAND HUCLEAR STATION AMEND.1 (7 2167) 4
,., 000 284 1 ef' q ,, <'
g Bundle Burnout Test Conditions where stable operations were O - obse rved. O Hot Corner Channel Worst Conditions
+ Hot Corner Channel Nominal Conditions
- 3. 0
\
\
\
- 2. 5 i
O O
- O
.e a 2. 0 9
g g g g b 2 i.5 N o i x 0 e O r l.0 i L Bubble to . Slug Bubble to Annular * (Griffith ( Baker) 5 . andRoss) Subble to Slug (Baker) I C 10 20 30 4 Quality (Ib vapor / total Ib). % l FLOW REGIME MAP FOR CORNER CHANNEL N_gIEF FIGURE 3-39.b [} , THREE MILE 15 LAND HUCLE]R STATION AMEND.1(7 216MJ 00 0 M 5 i
E Bundle l whereBurnout Test Conditions stable operations were observed. ! 5 Hot Cell Worst Conditions
+ Not Cell Channel Nominal Conditions
- 3.0
\ ,
e 5
\ l
\
\ !
- 2. 5 g
g 8 8 8 j e 9 M n, 2. 0 L g
- 8 f $
i 1.5
/
g Bubble ' y to Slug (Grif fith and Rose) 1.0 a il Bubble to Annular ( Baker) f
.5 Bubble to Slug
( Baker) l 0 10 20 30 16 0 Quality (Ib vapor / total Ib), f. l FLOW REGIME MAP FOR WALL CHANNEL NM FIGURE 3 39 c THREE MILE ISLAND HUCLEAR STATION y .
.p"o "7 * , c 0000 286
O h , Design Overpower i
\
2.0 -N \
\ g 1.65 cosine (w-3) 1.8 -
\
\ 1.80 cosi=e (W-3) -
0 1.6 h g a s' \ BAW-168 Design 1.k -
,,,,,_ __ DNBR (1 38 )
g & - W-3 DesiEL 1.2 - D M (1 30)
@ 1.65 cosine (BAW-168) y 1.0 - \
- n 1.80 Cosine (BAW-168) N '
\
0.8 - N O 1 5o c 1 caaw-16e)
\
O.6 - g 1.50 cosine (w-3) N 0.h - 0.2 -
)
O t t l l 100 110 3,20 130 140 150 Reactor Power fi of 2452 Wt HOT CHANNEL DNB RATIO COMPARISON _ME FIGURE 3-40 THREE MILE ISLAND NUCLEAR STATION 0000 287
O 150 l l Design ?cver ! 7 (2,h52 Wt) h ' M IM M 1 130
) ,
1 k s e
/
i 110 i
\ l 2300 2400 2500 2 00 Besctor Care Power, Wt 0000 288 O
REACTOR COOLANT FLOW VERSUS POWER
- h. .. s NM FIGURE 3,41 THREE MILE ISLAND MUCLEAR STATION
O u.x I Uo H"1t* 2 l
\'\
"a T
" I j 3.m \,,
D 'A s '\ m
'\
i b k E8M Design Value (CVNA - 142) d CVNA - 246 3
/
I
$ 2.00 k" O a V ,/ i s
\ /
E
! A Y <1
/ '
]
\ / f Wli -Nl GEAP h624 -^
1.00 l , o 1000 2000 3000 4000 5000 Te:nperature, F p THE'R$4AL CONDUCTIVITY OF 95P. , DENSE SINTERED UOf PELLETS MM FIGURE 2 42 l THREE MILE,g(AND HUCLEAR STATION
}Q()-} 2hh *h -
4
O 6000 5500 -cesign Overpc er .114; f ,e L- / / 2 UO Melting Temperature ' 2 / , 5000 P-4 ' ---
- ! /
l
/ / l 4 4500
/ # !
0 // / u .
//
0 k000 A!f l O [ l 3
)
3500 i
/
3000 BAN Design Value (CVNA-142) W 42 GEAP-M24 p _ _ _ _ _ _ m 43 CYNA-2M )) 2500 i ! t i I 6 8 10 12 14 16 18 20 22 24 26 28 30 Linear Heat Rate, kv/ft O FUEL CENTER TEMPERATURE FOR BEGINNING-OF-LIFE CONDIT10NS tp' . MM FIGURE 3-43 THREE MILE 15 LAND NUCLEAR STATION i' h il h h 2 9 .)
6000 , g r
/ i 5500 - D'Si@ OV'GCV'r ( 11' 5 * '
1
/ /
100% Power /
/
/
00 Melting Temperature 2 , 5000 ----
; -f - -
~
~
/ / /
! 45 / //
/ /
/ !/
l 5 uom // l /
// e 3500
)
Bau Design value (CVNA-142)
- _ ._ -. REF 42 (GEAP k624)
REF 43 (CVIA-246) 3000 f l 25m 6 8 10 12 14 16 18 20 22 24 26 28 30 Linear Heat Fate, kv/ft FUEL CENTER TEMPERATURE FOR g END OF LIFE CONDITIONS T l shD FIGU R E 3 .14 ,, THREE MILE 15 LAND NUCLEAR STATION ' 0003 291
l O l 100.00 i l so. " p; ; 5. ,
- e -
e !ui i r
- 7. -
I I 10.00
/ s. ." .
m f
'b i .
j 5. M b p' g 1/ I + t i 3 30/ l* l l l
- . +
A
1 .
c '
~
/
0 50 , , . , , , o . > . g o , , , , ',U l /
- I ! I I I !
0 G DP - 4596 l3 ,
+
. GDP - 4314 AECL - 60 3 j - A CF-60-12.,4 (CANL) 0.05 -V 4 i i i i e i . , i 12b0 1400 1600 1000 2000 2200 2400 2600 2000 3000 3200 3400 3600 Velvestric Averale Temperature. F
' I' PERCENT FISSION GAS RELEASED AS A FUNCTION OF THE AVERAGE TEMPERATURE OF THE UO2 FUEL t (g ? r NE7FF FIGURE 3 44.a s THREE MILE ISLAND HUCLEAR STATION AMEND.1 (7 2147)000 0 2 9
1.8 l 1.7 I
' - N P/P = 1.70 (Partial Rod 1.6 ! \ t insertion?
/ I l.5 '
\ / 'T'N P/ P = 1. 50 g,g /\ l \ (Modi fied Cosine) 100 Days
, f i2 I / 300 Days g,, l % _\
NN 930 Days M-l \ i
# I \
in 1.0 0.9 0.8 '
\ t 0.7 t 0.8 0 ? l \ \\ \\
0.5 - \ l 0.4 \
/ x \u 0.2 l 0.2 1 Core 6 Bottom l Core 5 h 0.1 ,
Top \ 0.0 I I ! I i f ! 20 % 60 80 800 120 1% 10 30 50 70 90 sto g30 I Distance from Bottom of Active Fuel, in. AXIAL LOCAL TO AVERAGE BURNUP AND > INSTANTANEOUS POWER COMPARISONS . NM FIGURE 3 44 b THREE MILE ISLAND NUCLEAR STATION 0000 293
. AMEND.1 (7 2167)
~
O . 50 Design Limit i 40 m w V
/ /
/
q I I.70 BU and ///
= Axial Shape. /
/
,3 1.50 BU and
.o 20 Axial Shape j
$30 Day BU y
[f[
, and 1.70 Axial C Shape.
I 10 0 0 1 2 3 4 5 6 7 8 Cold Diametral Clearance, in. x 10 i l Fl5510N GAS RELEASE FOR 1.50 AND 1.70 MAX / AVG AXI AL POWER SHAPES . , NM F'IGURE 3-44 c
'; (, . sj < . ,egi* THREE MILE 15 LAND HUCLEAR STATION AMEND.1 (7 2167) /
0000 294
3500 Design Limit h 3000 Closed Pores , l%i .$ 2500 I / E. E G l . 2000 1.5 Axial Power g and Burnup Shape. l e ' 3 __ _
._E 1500 -
E
/
a-l.7 Axial Power And 930-Day I Open Pores
- Burnup Shape.
1 g,
/
500 I l 0 1 2 3 4 5 6 7 8 Cold Diametral Clearance, in. x 10 l GAS PRES $URE INSIDE THE FUEL CLAD FOR VARioUS AXIAL BURNUP AND POWER SHAPES . Nm ' ' W. FIGURE 3 44.d THREE MILE ISLAND HUCLEAR STATION
. n,.n..n 0000 295 4', i * , >
O I l ___ :s- 1.006 . 978 0. 96 0.963 -: 0 97< 0.997 -- 1.006 1. COO 0.986 0.983 .987 00
- 0. 778 00 R 1.020 1.022 a 1.01: 00 0.965 9A 1.02 1.022 9 00
% 98 1.02 1.Okl 1.014 9 Ce RCD 1.032 1.014 96 .970 0.998 1.008 1.012 1.011
. 99r 77 . 776 1.010 1.012 - - 1.016 1.Cid 0.997 1.001 1. 1.004 0.906 998 .010 1.0hc l l 1 / 1.,015 1.018 1.019 clear Peeking Factor 1
Enthalpy Rise Factor i l NOMINAL FUEL ROD POWER PEAKS AND O CELL EXIT ENTHALPY FISE RATIOS
.,'J NE FIGURE 3 44.e f ! . V ' .i ' ' THREE MILE ISLAND NUCLEAR STATION AMEND.1 (7 214 t
I . O P INST 3.987 0.960 0 9k9 0.9k9' O.959 0. 9T --- 0.981 0.986 0.96, 0.969 0.986 1.01 0.986 {f '.00; 1.008 fe$ :. 01; 1.022
- 3. 949 0.967 1.007 1.027 1.022 0.99 1.020 3
0.949 0.909 00 1.027 g 1.008 982 ,,, 0.959 0.99-
. 1. 2. 1.008 0.982 .973 1.022 POD l
1.02? 1.061 .11L I
) ;r71 0. M. 1.012 0.99. 0.982 0. 97: .993 1.0s; 1.061 1.097 1.1Ld L.012 1. 0L 1.02 1.020 1.017 1.022 1.04 1. 0%
l 1.,116 1.,1L8 . 19] 8 Nu: lear Peaking Factor I Enthalpy Rise Factor MAXIMUM FUEL ROD POWER PEAKS AND CELL EXIT ENTHALPY RISE RATIOS N E FIGURE 3 44 f THREE MILE ISLAND NUCLEAR STATION AME ND.1 (7*2167)
][j[].}
[Y .
,, 'i (,\)N,
- O i.-
s i ; ,
- 1. 3 \ G = 1.59 x 106 l b/ hr-f t2 -
x .
\\ l
\ Best Fit N
l.0 , o Design Limit x 0.9 5w 0.8 x \ 3 i Minimum DNBR - 2.20 g\ 0.7 \ 5 \ 0.6 " \ ' x
- N
[
\
0.4 g
\ \
Calcul ated 0.3 h i 0.2 0.1 I O 580 600 620 640 660 680 700 720 740 760 780 80 0 Local Enthalpy. Btu /lb CALCULATED AND DESIGN LIMIT LOCAL HEAT FLUX VERSUS ENTHALPY IN THE HOT CORNER CELL O
- AT THE NOMlHAL CONDITION
, . ,g 4 NEyry FIGURE 3 M g 0 11' l '
s v' e' THREE MILE ISLAND NUCLEAR STATION I AMEND. 3 (11*6 47) fj
O i.4 , , , . , 1.3 G - 1. 32 x 106 l b/ hr-f t2 _ l.2
\ '
l .1 \ '
\ Best Fit
, 1.0 g' .
'e x 0.9 Design Limit
. ',\ \ !
% \\ \ l 2 0.8 \
\
". o7 l
i x Minimum DN8R = 1.70 4 3 o$ q\ g>
- x o5 g S xx 0'4 \
/ \\
0.3 i , N'* Calculated I lh
, \
0.2 - ' y l l 0.1 0 580 600 6 20 640 660 680 700 720 740 700 780 800 Local Enthal py. Stu/lb CALCULATED AhlO 959tGN LIMIT LOCAL HEAT FLUX VERSUS ENTHALPY IN THE HOT CORNER CELL AT THE POSTULATED WOR $T CONDITION WM FIGURE 3 44.h THREE MILE ISLAND NUCLEAR STATION AMEND. 3 (11*6 67) i W0 I'99 Q (1 \, ' h
,;a[ "
O ! i b _n _ __u, , ,
$l - .
q - - 2 1.
< \
7/, n
!\
'/ \)
<! !,/ !/ lj
. ljl T=F
. \ !ptijlQj_ W5 l W I 311 ' F .,,, ...
ks O - N
)
y\ / p , > J,
.ns,unas N lA < fl l
?
f b i
.. , . as 1
N"- a :: 1 U
. 2 s m.
) b i & ,iei N s . \ / W <
b
': N N dn: /':
N N -
\.-
/
/ N N .c. ..a
.O..r. .."' D s e N
Nf x k, k s N w s as s k
/ \. i* k
/ N 5
===
- / \
<N Q j N g:_/ ru .s M N w was ll ll 11 j) ij il N3 3 0 3 jN 0 13d
\ :
- m 1 L
J :.% l l i s 8 I REV: 7 21-67 REVISED UPPER PLENUM A33048LY AND CONTROL DRIVE REACTOR VESSEL & INTERNALS GENERAL ARRANGEMENT I ' *. I
,,,, NM FIGURE 3 45 THREE MILE 15 LAND NUCLEAR STATION
- ~o. m.u.m 0000 300 I
O T FUEL ASSEMBLY 1
//'!/ // 7,
/ - '-
SURVElLLANCE
/
/' . { o ps -
7 SPECIMEN O[l j.l t { O HOLDER TUBE
)'oj 01
/
i l@i tel a isi 41 1 l l @l .i el l IIOl e i OI l LO.. [ l l@l Gi @ll 10$*101 I f l l '
/ /O f f@j jel @l 611@l ngl *lGI LO ! / CONTROL ROD ASSEMBLY LOCATION
/] NIYeii l @l 101
- 4GI IOl 4 91 i o'
', f FI l el 1*l l eillei l*' E'*f '"
I-
' : : of a '@! 0: a , 2; :7 121 0, ---as
-/, ! i ej .I@{ ll0 +l @l ' 'S ! ! @l I
,- ! oj o INCORE INSTRUMENT
.{ .I el .! el I @i I GI l @l *l Gl LOCATION T O[ l el j e! .j @l ll@l lSI ! I i O/
/ 4 j .i ei l' et .I @l '@l 1.01 I i /
,4 o i 10 .: 01 bl@l *l GI l *ilo.
'g j iel !@l I Ol I *t o
~
/
REACTOR VESSEL /
'/
THERMAL SHIELD
, ////'
I CORE BARREL l l REACTOR VESSEL & INTERNALS GENERAL ARRANGEMENT ' CROSS SECTION ME FICURE 3 t6 ilu 01 THREE MILE ISLAND HUCLEAR STATION i, e kU
I I l L_J i
- I i.1 # i I i I l
O ! l
! '\,!
\ \ N ' ,
~%
N , EiB N
\ s 4 l i =e i: ,
1 N N _n_. k : N 1: N's i, x h ~ l: !! Z,,% (; i d b I, '
'(' h M
, .-sc--
o , - [,
> \\ - .,/ ,
< x I < / CORE FLOODLT,
< NOZZLE ,
s I f ,/ x .
; c - -
- \
; l u '7, I
i ' s
/
/
s
< /
<: ) : -
s . i:) - s - l , i y -
/
O . 0000 302 l l l CORE FLOODING ARRANGEMENT NEF7F FIGURE 3 47
-s .. THREE MILE ISLAND MUCLE AR STATION i
UP9ER END m /~T 2 ', FITTING g! ,f s ,, I i l1 s
) .
h/.!
;t - g i
e a s -> , j I, 9 l 4 l
'q 4 ,, ~ __h__pr_,
,Q ~ ~
, , il SPACER GRID -- 77 ~ I-
=0 0 0 0 C-
- w. - - - - - s
.. :.o
.n onwpn oe .
c n -
= @
>=a== [Q=a == = F: M=pt dFQ1@b =q ,
I l i#
;: ; i
=l = ~ = cc=
1: _h=_@ = ;= = = - : i E __ _ fb(d _ _ _9 == =QCANh= , PANEL l 4 c--t.r--- g i j b??"W$NAEL? O:Ca p = =:= = = @ C j l-i i 3 :s = = oc::= soc === =, j ',. _C' X: 1_7,CL C ] j
~
____ - s # ' 7 - TOP VIEW f l l <
\
' W u. _mumG V.
O' f INSTRUMENTATION TU8E ' C + MnN M
- 09C O* Ot Ug n CI 1 , . .
/
f5'.U$nn [* l ' ' i '* i 4 r CCNTROL ROD , , Q ir iO ie 4/ GUIDE TUBE i O c
.*l! '
O / l
,! j .>f -
Q Q G . .' Q
l i ,, .
l C i
, . i ,,
, , , FUEL RCD ,
O' i; i ASSEM8LY l O ,C
' ,OtJ > 1 i
]
i l I CROSS SECTION i l # M 4 1m @ l F~d UV LL 4 b.h"[j c
.. i
,e FUEL ASSEMBLY e
f: r-p1 ( . mMM FICURE 3 48 FITT l !! THREE MILE ISLAND NUCLEAR STATION l r ) . 0000 303
O coum 3
;- r N,C, l '.
; ,- /,- ..a N;..s T ; sy M
o-ic = g N !
/ -
101 O N' 00 V V VV TOP view 0000 204 ORIFICE ROD ASSEMBLY
-ME FIGURE 3-49 i -
- 3 , , ,,, THREE MILE ISLAND NUCLEAR STATION
,,,.,'r t t ' '
/
f-carve worom )
. stoucta i l
G. . w/ e=a M I i L._ ! RJ h1
.-nacu Housne i
e l *,, -onvesawTnousme tl1
'I l,i i l
dl!3'/ i y
.-spea. met l &l m
,cetacTon vtsstL wan l h. s..+' i*
i l i l l l l
.3.
i I t i i i k' ; 5, . _coupuso a==a==, CONTROL ROD DRIVE GENERAL ARRANGEMENT / , N M ,, FIGURE 3 50 h- REY: 7-21-67 THREE MILE ISLAND NUCLEAR STATION cunGEo cenract ece onivt.
- . i g.u..n .
000D 305
r __ _ . . -
!?I 3;;
s -.- . i,: ._ sa;_
.il g ._.,___
s o
,3 ,
a _ . c.i .
.n t;
- n. e -- . ,.
d t m e m . 3
- n C U j.. E t
- I a ~,*
.s).} g a
L.9 1 _s u u a 3 wt l@ -,. g [Wz'4 .. x .
'1 41 -,
i v%u i- . , . v ;
; ,a _ . . .
,.c
- q __.,
.~'
uw. s ..uy ._ gu. im / ./ C
' i
- 1 j
, I j i
; e
~
.f 0 n
w,r,n 'n: 5c n . m" . eie 6' w i i s mw":' c 4% n's . l v-L. v-
..i -
, t,
, . l-
,.g c :t y .
,,m; .-- .9s.
J [ 0,-
-l [
e
^
, ('~ ll
,0M '
H s ,7
- r ele
. uay V; ri ..
r---' y-' 5
/-
AV my%&y
--+ F- w , -el
- .%, ~ ! ! 1 q%. , i , -
C,> b p,dE9
-y Ml 1d. g.
p-6
,v m,
A . ra ,1 , p- _.
, m w i!
["pI $i L NI: 12 'Fd.i d'#G
,n u, _ .
a.ss . w, r. c i, . a, -
- u. <.s. , m %m 4 e %. .
. Q<
~~
id~L Wi cm & Lt. !_p .ch. m. , sn
" !!s, Ims b, izW., , ,W j ,vr. iw J iit+9Mt i i
-m j s ,.
- p. ,
e
, !, < . . Is .,
%.g7
- . J
.y r. .
~
- s
- . _,_, fi
'.tr{' ' mg c-
,i !" .
!'+ j j g
-g -
-l.. ..e
}.. ._. q i ;
- .x.- t ,
e . . r. u w s .a b g K7i ; !.
-a S* ___. -
if- 5 l$ m - -- , ts i cJ' :l
+ .
0000 306 m
O - l l l i i (Deleted - A=end. 1) l l l l I 1 i O l l a Fisure 3-52 (T-21-6T) O 0000~3d'7
O (Deleted - Amend. 1) ; O 1 1 l l 1 1 O I l Figure 3-53 (7-21-67) p,,,.
, 0000 308 l
O l l r l i I (Deleted - Amend. 1) O l l l F16 ure 3-5h
. (7-21-67) t,'- '
0000 '!09
O (Deleted - te.end. 1) O l 1 Figure 3-55 (7-21-67)
.1 ..
0000 310
- 1
O OO
--- c. .. :. = f,
..~..
oO A A -- _W
~ -.a 1 .. . .
- =
l ...
. l .
~T
.;;;;< ~ . ~ . . ~ ':':L',~ ..._.4i
~
t.. . . -
- f. ,
~
...,4 .
i I "O,l
"'"'"la" R
; .[. r -- - ,
}- , - . W.
] -i '".. l
~
L _ _ _ _ _s, i ... . i :;;:
.. i -
i , I t
.-,- ii i i,i
~.
l y'; i t am ,.s
-- osl ii ,,,,,,. ;
8 e8 .a.....>" i8 0 __ m / im.iu. 9 m l I ! ' < .. , - ; .____j
- i. ,e '"."..
. . '.~m.."-.-
7,;- .____; [ n t, *, i, i..
.a.~. ,
. c.
8 ,I g
.= LiJ
. . , , ,-a '.-==
...... , : i,-- - -
-=
J , .o ;. , I . I ...
=
-':. ~.~
,,,, = .,. . . . . . - . .a - ._ _ _.'
.... ...... ~ i c._1 .,
ae .... _ .=6
. .c=>, p _ _ ,,,
c-.-- ; 'a
. yi~ . . _ __ _j
" 0.-
. u. ,
q
...,..... ivl C'L .m..a.......
M
, ii1' CONTROL ROD DRIVE CONTROL $YSTEM BLOCK DIAGRAM N EEF FIGURE 3-56 l ,
6 -s
-, THREE MILE ISLAND HUCLEAR STATION ,
-'- AMEND.1 (7 2167)
Iifill9. s l
- i. ,
U tJ V V J l t 1
O , MOTOR Llulf CLUTCM MM o l I I I POSITION Signal 038Tl0m _____ CRfVE I 3"'#I POTENTIOMETER g A Il LIMIT $1GNALS _____ SwlTCNES 7, POSITION StGual pgg,ggg, l l SYSTEM Lluti stGmat DE M ATOR TR&4SF0 mER ------M i L0 sic Lv0Ti i < n v l t SEAL drive REV 72147 REDIAWN TO REFLECT REVl$10N OF CONTROL 200 DalVE 0000 312 O LIMIT SIGNAL & POSITION INDICATION SYSTEM N E7FF FIGURE 3 57 THREE MILE 15 LAND NUCLEAR STATION (i ?
> 5 I- AMEND.1 (7-2167)
,e'
O PCut2100#CE & PCeti $0URCE 2 I I 1 usott v% *ast i f#lP Cal 65
- - . . - . -) . . . .) totast e oo, e I
staC109 HCitCfiv8
-----*-------d--- sataatt no. 3
$v$fru 4 '
- - . . . . .) . . . .) sa ta s t e a0. 2 1
f2 --==== ....
. . . . . . . _ . . . . . . { . . . ) ,,,,,,, ,o ,
4 1 1 I l Y Y E I l l III tofu i 2 getttn 67 et 64 L Y J CLUTCM HOL0 leg PowtR arv. 7 2147 TO 69 CouTROL 200 Otivt3 WDRAudi 70 NEFLECT
#fV111CD OF CDRN R ORIvt l
l (.e . l' l REACTOR TRIP CIRCUlf O NM FIGURE 3 58 THREE MILE 15 LAND NUCLEAR STATION AMEND, I (7 2167) ! ,
/a '
a COUPLING - j " j r
~m. .
l lb f; O ya i i l'N -. ,l l i l l-i k
; ju ae * '
ll 1 , o '^^%_ 4
+, -
T~%. g p _ _ _ _ _
- [
TOP VIEW "D % "
-. *% M~
.w- .
M y s. NEUTRON , PolSCN l SECTION l CoNrnot aoo l g
- il i
i i i m I l l L - a - HL l
\) \;
J u JL '
,! 'ul 'd \ / i O 0000 314 CONTROL ROD ASSEMBLY !
NM FIGURE 3-59 THREE MILE 15 LAND NUCLEAR STATION}}