ML20002C861
| ML20002C861 | |
| Person / Time | |
|---|---|
| Site: | Big Rock Point File:Consumers Energy icon.png |
| Issue date: | 04/15/1977 |
| From: | Bilby C CONSUMERS ENERGY CO. (FORMERLY CONSUMERS POWER CO.) |
| To: | |
| Shared Package | |
| ML20002C860 | List: |
| References | |
| NUDOCS 8101120199 | |
| Download: ML20002C861 (45) | |
Text
- ),
-f 7
%)
h CONSUMERS PC'4ER COMPANY Docket 50-155 Request for Change to the Technical Specifications License DPR-6 For the reasons hereinafter set forth, it is requested that the Technical Speci-fications centained in Provisional Operating License DPR 6, Docket 50-155, be changed as described in Section I, below.
I. ' Changes t
A.
Add or replace the followingi 1.
Table 5.1 (Page 3ha).
2.
Page klb.
3 Table 1 (Page h3).
h.
Table 2 (Page h3a).
(
5 Table 8.2 (Page 91).
1 F/o//AbIT
~
i TABLE 51 (Additional Infoz=ation)
See Page 34 General Reload G-1U Relead G-3 11 x 11 11 x 11 Geometry, Fuel Rod Array Rod Pit,:n, Inches 0 577 0 577 UO Rods 109 113 2
Cobalt - Bearing Corner Rods 4
0 Gadoliniu= - Bearing UO Rods k
k 2
Inert Spacer Capture Rod (Zr-2) 1 1
Zircaloy Rods 3
3 Spacers per Bundle 3
3 Ptel Rod Cladding Material-Zr-2 Zr-2 Wall Thickness, Inches 0.03h 0.03h Fuel Rods Outside Rod Diaseter, Inches 0.hh9 0.hh9 Fuel Stacked Density, Percent Theoretical 91.6 91.5 Active Fuel Length, Inches Standard Rod 70 70 Fill Gas Heliu 2955 Helium 955 l
f
(
3ha
k, FUEL BUNELE SCHEM.ATIC G-3 RELOAD FUEL (TO BE SUPPLIED) 9
(
h1b
r 6-A s
t TABLE 1 Reload E-G and Modified E-G Reload Relcad F, J-l & J-2 Reload G G-1U G-3 Minimum Core Burnout Ratio at Overpower 1 5*
1 5**
1 5**
1 5**
Transient Minimum Burnout Ratio in Event of Loss of Recirculation Pumps From Rated Power 15 15 15 1.5 Maximum Heat Flux at Overpower, Btu /h-ft 500,000 395,000 407,000 392,900
' Maximum Steady State Heat Flux, Btu /h-ft 410,000 324,000 333,600 322,100 Maxim:m Average Planar Linear Heat Generation Rate, Steady State, kW/ft Stability Criterion: Maximum Measured Zero-to-Peak Flux A=plitude, Percent of Average Operating Flux 20 20 20 20 Mayimum Steady State Power Level. My 2h0 240 240 2h0 Maximum Value of Average Core Power Density 9 240 MR, kW/L h6 4
h6 h6 g
Nominal Reactor Pressure During Steady State Power Operation, psig 1335 1335 1335 1335 Minimum Recirculation Flow Rate, Lb/h
-(Except During Pump Trip Tests or Natural Circulation Tests as Outlined in Section 8) 6 x 10' 6 x 10' 6 x 10 6 x 10' D
Rate-of unange-of-Reactor Power During Power Operation:
Control rod withdrawal during power operation shall be stah that the average rate-of-change-of-reactor power is less than 50 M4t per r.inute when power is less than 120 MW, less than 20 Wt per minute when power is between 120 and e
200'MW, and 10 MW per minute when power is between 200 and 2h0 MW.
l t
g
- Based on correlation given in "*Jesign Basis for Critical Heat Flux Condition in Boiling Water Reactors," by J M Healzer, J E Hench, E Janssen and S Levy, September 1966 (APED 5286 and APED 5286, Part 2).
- Based on Exxon Nuclear Corporatice Synthesized Hench Levy.
- To be determined by linear extrapo:ation from Table 2 attached.
- (
L I
h3 4
e 7
. ~. -.
1
\\
k TABLE 2 MAPLHGR (kW/Ft)
Planar Average Exposure F, E-G, Reload G UGd/STM)
Modified F J-1, J-2 and NFSDA Reload G-1U Reload G-3 6.h53 6.h91 6.55h 0
200 95 9.4 21h 6.750 6.758 6.807 216 437 6.887 6.888 6 973 hh3 6.960 88h 6.978
~
885 7 033 893 6.929 1,758 6 970 1,769 6.984 1,773-4 6.885 3,494 6.913 3,509 6.983
(
3,5k5 5,000 99 97 6.838 6,939 6.865 6,970 6 978 7,085 10,000 99 97 6.8k7 10,422 6.882 10,h82 10,690 7 019 6.867 13,938 6 9ch 14,019 7.069 1k,355 15,000 9.8 96 20,000 8.7 8.6 6.905 21,022 6 958 21,194 7 171 21,843 25,000 8.4 8.3 6.8k3 27,778 6.903 28,035 7 161 29,08k 6.703 34,013 6.923 35,147 6.958 35,322
(
h3a
i TABLE 8.2 en emelt EEI UO -
" *#~
Pu02 mediate Advanced NFS-DA Minimum Core Burnout Ratio at Overpower -
1 5"'
l.5*.
1 5" 15 Transient Minimus Burnout Ratio in Event of Loss of Recirculation From Rated Power 15 15 15 15 Maximum Heat F.'.ux at Overpower, Btu /h-Ft2 h02,000 500,000 Maximus Steady State Heat Flux, Btu /h-Ft2 h10,000 500,000 500,000 329,000 Maimum Average Planar Linear Heat (Refer to Generation Rate, Steady State, W/Ft Table 2, Page h3a)
Stability Criterion: Maximus Measured Zero-to-Peal. Flux Amplitude, Percent 20 of Average Opcrating Flux 20 t
2h0 Maximum Steady State Power Level, W 2h0 Nominal Reactor Pressure During Steady State Power Operation, psig 1,335 1
Minimum Recirculation Flow Rate, Lb/h (Except During Pump Trip Tests or Natural Circultion Tests as Outlined in 6
6 Sec 8) 6 x 10 6 x 10 Number of Bundles:
1 3
Pellet U02 Power UO2 1
2 Rate-of-Change-of-Reactor Power Dttring Power Operation:
Control rod withdrawal during power operntion shall be such that the average rate-of-change-of-reactor power is less than 50 W. per minute when power is less than 120 We, less than 20 W: per minute when power is between 120 and 200 W, and 10 Wt per minute when power is tetween 200 and 240 W.
e t
- Based upon critical heat flux correlation, APED 5286.
- No longer used in reactor.
91
\\
k S
1 II.
DISCUSSION
1.0 INTRODUCTION
AND SUldARY The purpose of this proposed change is to anov the use of an all uranium fuel with acceptable ECCS performance characteristics in the Big Rock Point reactor and to delete the Technical Specifications li=itation on the design burnup of fuel bundles.
Presently, licensed fuel types are 11 x 11 all uranium,11 x 11 mixed-oxide and 9 x 9 an uranium. The n x n all uranium fuel (denoted
]
G-lU) was used as reload fuel at the last reloading. The proposed 11 x 11 all-uranium reload fuel (denoted G-3) is very similar to the G-lU fuel assemblies with three basic differences. For G-3 assemblies, the four corner cobalt target rods have been replaced with fueled rods.
There have been changes made to the bundle enrich =ent distribution which reduces the overan bundle enrichment from 3.885 to 314%. The placement of the gadolinia poison pins has been altered for better peaking character-istics. These effects have been accounted for in the subsequent analyses
(
presented.
This submittal contains information concerning fuel system design, nuclear design, ther=al hydraulic design and accident and transient analysis as reco== ended by the " Guidance for Proposed License Amendments Relating to Refueling." Every feasible attempt to present the infor=ation requested by the guide has been made. In geraral, the major difficulty in providing this data was in for=ulating a suitable " reference cycle" -- defined in the guide. For Sections k and 6, the reference cycle used was Cycle ik, specif-ically, the Reload G-1U' fuel. For Section T, there was no single suitable reference cycle available. Thus, the latest analysis found acceptable by the Co==ission was used as the reference cycle for each specific accident or transient. In many cases this dated back to the FHSR. Finan y, for Section 5, since many of the parameters required by the guide were not rou-
~
tinely calculated or submitted in previous reload licensing sub=ittals, no reference cycle is given. It is Consumers Power Company's intent to utilize Cycle 15 as the reference cycle for Section 5 for subsequent, reload licensing sub=ittals.
2 i
g v-
,wy
t
(.
2.0 ' OPERATING HISTORY Cycle 14 power production began on July 28, 1976 following a refueling outage. The core loading consisted of 38 - 9 x 9 fuel assemblies and 46 -
11 x 11 fuel assemblies, with residual fuel assemblies relocated in the core to provide adequate shatdown margin and acceptable cycle power peak-ing. The off-gas release rate stabilized following start-up at approxi-mately 750 uCi/s (corree;ed : tor specific gravity). Tae plant has oper-ated since that time at power levels ranging from 206 to 216 l&.
This reduced power level resulted from reaching Technical Specifica-tions MAPLHGR li=its, pri=arily for the F fuel. The off-gas release rate for the latter portion of the cycle is averaging approximately 700 to 800 uCi/s. This is the lowest off-gas release rate of any cycle and is attributed primarily to the re= oval of copper based ma-terials from the pr %7 system several cycles ago, and the subsequent discharge of fuel bearing copper based crud. Cycle lh was originally
. designed for 'an energy production of 61 G'4D; however, due to the ex-1 tended operating period, energy production is now expected to exceed slightly this figure with a power coastdown at the end af the cycle.
A su==ary of Cycle lh start-up and tests performed at the beginning of the cycle is contained in Special Report No D dated Nove=ber 24, 1976.
3
-c.
4 i
l
(
30 GENERAL DESCRIPTION Cycle 15 is designed to produce a target energy of 72 G'dD.
This corres-ponds to a cycle length of approximately 325 days of power operation at 220 !G,. The pro,1ected core loading for this cycle, including cotalt distribution, is shown in Figure 3-1.
This loading sche =e is aduject to minor changes, depending upon the results of fuel sipping conducted during the June 1977 outage, with the goal of making effluent releases as lov as reasonably achievable. Figure 3-2 details the fuel rod arrange =ent, the initial fuel enrichment and the gadolinium distribution and concentration for the G-3 fuel. The gadolinium is designed to burn up in a single cycle; thus, only the new assemblies contain si'gnificant amounts of burnable poisen.
Figure 3-3 is provided to indicate the beginning of life fuel burnup dis-tribut' ion for Cycle 15 The Cycle 15 fuel loading pattern has been designed to incorporate 180 rotational sy=cetry throughout the core. The fuel 21stribution has been developed to comply with Technical Specifications limitations and safety analysis criteria. These limits and criteria include MAPLHGR, =ini==
critical heat flux ratios, maxi =um heat flux, maximu= control rod worth and mir Mus shutdown margin among others.
(e i
i
Figure 3-1 CYCLE 15
(
(
BIG ROCK POINT d
CORE CONFIGURATION A
B C
D E
F 4
Jo do
- >s'4 4 4 4
+%#%4%
s 4 4e4 40% 4 s*e% % %@%(44 4 444444 #%%
4 + 4 4 > 4 +'% o
- D'e %
% e %#
e 4 4.Oe4'O 4 4 z
+'e 4 4 4 &s+ + e ses,se
's*4 0 4+X+Ns es
+
.O
+
5 Dukened buts indicate test assemthes.
P00R BRIB R
a 1
Figure 3-2 FUEL ROD ARRANGEMENT-BIG ROCK POINT RELOAD G 3
(
L L
M M
M M
M M
M L
L hH hH hL hM M
H L
G M
M H
H H
H M
H H
H H
H H
H H
H M
hH H
[
H Z
H H
M M
G G
M H
H H
H H
H H
M H
H hH H
Z H
Z H
H M
M.
i M
H H
H H
H H
H H
H M
M M
H H
H H
H H
M M
H hH hH hM H
M L
L L
L M
M M
M M
M M
L L
t NUMBER OF RODS DESCRIPTION 3
INERT RODS 12 1.50 wt% 235U 40 2.52 wt% 235U 61 3.82 wt%235g
(
4 3.82 wt %235 U + 1.25 wt%Gd O3 2
h 12 TIE RODS h
1 INERT SPACER CAPTURE ROD 6
(
Figure 3-3 CYCLE 15 l
BIG ROCK POINT
(
CORE CONFIGURATION l
BOL BURNUP DISTRIBUTION
_g, A
B C
D li 30 0
4 's e t4444
+44444 s
s 4 4*4 See 4 s o 4 4 4 4 4 4.
4 4 4 See 4 4
(4+4e4 4 4 See O444444>
/4444440 4
/4&#e'4e44'4 4 4 4 e4 A>4 4 44 44
/ 4 4e
./ 4 4 4 4 e\\
/ 4444
./ q /q /q
./
\\
c
&J P90RBRiBIE
(
1 I{
4.0 FUEL SYSTEM DESIGN The G-3 reload fuel design for Cycle 15 has mechanical, thermal hydraulic and neutronic perfor=ance characteristics similar to the G-1U reload fuel design for Cycle ik, the reference cycle. Both G-1U and G-3 fuels employ an 11 x 11 red matrix with four inert rods; however, G-1U fuel design incor-porated four corner. cobalt rods necessitating a slightly higher enrichment than G-3 fuel. A complete description of the mechanical, thermal hydraulic and neutronic characteristics of the G-1U fuel was presented in our letter dated Octobar 13, 1975
~ - ~ ~ ~
f k.1 Fuel Design As discussed previously, the major fuel design change for Reload G-3 fuel when compared to Reload G-1U fuel is the elimination of the four 35 cobalt target rods and the reduction of the overall U er:richment.
Table k.1-3 lists the design para =eters for both the G-3 proposed fuel and the G-1U reference fuel. Table k.1-2 delineates the fuel in'ientory at BOC for Cycle 15 It consists of the fuel type, nu=ber of assemblies, number of cycles in core, and initial BOC enrichment, density and average burnup. The G-3 fuel is desigued to be free-standing throughout its life in core, which is consistent with previous I
G reload fuels and, like other G fuels, G-3 was initially filled at nominal atmospheric pressure.
4 l
9
(
8 e
a w
a
I i,
(
TABLE h.1-1 DESIGN PARAMETERS FOR BIG ROCK POINT G-3 VS G-1U All Uraniun Designs Reload G-3 G-1U Fuel Asse=bly Rod Array 11 x 11 11 x 11 Rod Pitch, In 0 577 0.577 Water-to-Fuel Volume Ratio of Lattice 2.60 2.69 2
Hest Transfer Area' Ft 80.23 77.48 Nu=ber of Spacer Grids 3
3 Rods per Buadle Cobalt Target 0
h Lov Enrichment Urania 12 16 Inte mediate Enrichment Urania h0 (Includes h 32 (Includes k (Nonpoison)
Tie Rods)
Tie Rods)
High Enrichment Urania 61 (Includes 8 61 (Includes 8 Tie Rods)
Tie Rods)
Poison (Urania-Gadolinia) h h
Spacer Capture-(Nonfueled) 1 1
Inert Rod (Nonfueled) 3 3
121 121 Fuel Rod Dia=etral Pellet-to-Clad Gap, In 0.0095 0.0095 Overall Fuel Rod Length, In 78.501
-78.501 Active Fuel Length, In 70.00 70.00 Plenum volu=e Length 3.901 3 9dl Fill Gas Helium Helium
^
1 9
l
(
-(
TABLE k.1-1 (Contd)
J Reload G-3 G-lU Fuel Rod Weights UO Rods (Total Ceramic Gra=s) 12k8 12h8 2
Poison Rods (Total Ceramic Gra=s) 1242 12h2 Fuel Pellet Material Sintered UO Sintered UO 2
2 Diameter, In 0 3715 0.3715 Length, In 0.300 0.300 Density, % Theoretical (TD = 10.96 sn/c=3) 93.5 93.5 Initial Enrich =ent
- i Low Enrich =ent Rods (Wt% U-235) 1.50 2.30 Intermediate Enrich =ent Rods (Wt% U-235) 2 52 3 20 High Enrich =ent Rods (Wt% U-235) 3.82 4.60 UO -1.20 Wt% Gd 0 18 "
2 23 (Wt5 U-235) 3.82 h.60 Average for Bundle 3.1h 3.88 Dishing Both Ends Ecth Ends Dish Volume, 5 of Undished Pellet Volu=e 2
2
- EBC of I= purities, pp=
<h
<h Poison Pellet Material 1.20 Wt%
1.20 Vt%
Gd 0
" 0 d0 n 0 23 2
23 2
Diameter, In 0.3715 0 3715
~
Length, In 0.300 0.300
- EEC = Equivalent Boron Content 10 i
. _., l
(
I
~ (
TABLE h.1-1 (Contd)
Reload G-3 G-lU Poison Pellet (Contd)
Density, % of Theoretical (TD = 10 91 gm/cm3) 93.5 93 5 Dishing Both Ends Both Ends Dish Volume, % of Undished Pellet Volume 2
2 EBC of Impurities, pp=
<T
<T Cisdd h Material Zircaloy-2 Zircaloy-2, Cold Worked and Cold Worked and Stress Relieved Stress Relieved Outside Diameter, In (After Etching) 0.hh9 0.kh9 t
Inside Diameter, In 0 3810 0.3810 Nominal Wall Thickness, In (After Etching) 0.03h 0.03h Minimm Wall Thickness, In (After Etching) 0.032 0.032 EBC, Total, Including Impurities
< 40
< h0 Insulator Pellet Material Alumina (A1 0 0
23 23 Diameter, In 0.365 0.365 Iength, In 0.200 0.200 EBC, Total, Including Impurities, Ppm
< 76
< 76 H
\\
t
\\
TABLE h.1-2 FUEL INVENTORY TABLE Cycles No of Initial Initial Fuel Average BOC
'in Assem-Enrichment (w/c)
Stacked Density Burnup
~ Core blies Fuel Tyre U
Pu
( #, Theoretical)
(!GD/ST) 5 6
F 3.52 0
94 16,641 5
2 F-Modified 3 51 0
9h 16,703
~
5 2
G 3 08 0 90 91.5" 22,851 h
12 F-Modified 3.51 0
94 13,711 3
18 G
3.08 0.90 91 5*
13,828 2
8 G
3 08 0.90 91 5*
'6,518 2
14 G-1U 3.88 0
91.6
- 5,863 i
1 6
G-1U 3.88 0
91.6*
0
~.
16 G-3
'3 1h 0
91.5*
0
' Pellets 2" Dished i
s e
6
(
12 u
t i
s
- k h.2 Mechanical Design The mechanical design of the Reload G-3 fuel is consistent with the reference G-lU fuel with the exception of the upper tie plate. This plate was modified slightly in the Reload G-3 design to provide stan-dard fuel rod location holes in each corner of the plate replacing the locking slots utilized by the cobalt target rods. By letters dated June 16, 1972 and October 13, 1975, mechanical design analyses for Reload G and Reload G-1U fuels were submitted; these analyses are applicable for Reload G-3 fuel. Table h.2-1 describes the G-3 fuel assembly components, their purpose and cosposition.
S O
e a
9 9
4 5
(
13 4
e.
1 TABLE k.2-1 DESCRIPTION OF TYPE G-3 FUEL ASSEMBLY COMPONEFIS Item Purpose Material / Rationale Upper Tie Plate and Maintains fuel rod array.
Cast SS, Grade CF-3 Handle Provides lifting fixture.
- Strength
- Corrosion Resistance Compression Springs Accommodates differential Inconel X-750 fuel rod lengths and sup-
- Corrosion Resistance ports upper tie plate.
- Strength at Operating Con-ditions.
- Springs loaded high enough to minimize fretting and lov enough not to cause excessive rod bowing.
Fuel Rod End Cap Provides high quality seal TIG - Fillet Head Welds of fuel rods.
- Excellent penetration.
- Extremely low porosity'.
i
- High strength integrity.
Plenum Spring Maintains co= pact fuel Inconel X-750 Wire colu=n during handling
- Withstand autoclave treatment.
and shipping.
- Maintain spring load during reactor operation.
Plenus Chamber Collects fission gases.
- Assures.that gas pressure Provides space for axial vill not overstress cladding.
expansion of fuel.
Cladding Contains fission gases and Zircaloy-2 keeps water from contacting
- Minimize neutron absorption.
fuel.
- Cladding is autoclaved for prefilming for corrosion resistance and to provide a corrosion-proof test.
Pellet Cladding Provides clearance between
- Designed to maxinite fuel Gap-fuel and cladding.
rod fissile content and to minimize pellet-clad inter--
action from swelling expected at high burnup.
lh
%f'
(
(
TABLE h.2-1 (Contd)
Item Purpose Material / Rationale Insulator Pellet Reduces pellet-nonpellet Al 023 interface temperature.
- Maintains te=perature below those causing excessive stress levels and below those of con-cern with metal-fuel reaction.
- Controls hydride preceiptation.
Atmosphere Heat transfer medium be-Helium tween pellet and clad.
- Good heat transfer character-istics.
- Provides an easy and reliable leak detection monitoring means.
Spacers Maintains correct rod-to-Zircaloy-b Fra=e, Inconel 718 rod spacing.
Springs
- Corrosion minimized.
- Mechanical stability, i
- Spring loads on cladding =ust be sufficient to minimize lateral and rotational move-ment of fuel rod but cust not cause excessive cladding or spring stress.
- Spacer must not cause exces-sive coolant flow resistance.
Inert Rods Displaces the highest peak Zircalcy-2 Cladding End Caps clad temperature rods under Filler LOCA conditions and provide
- Corrosion resistance, a radiation sink.
- Lov absorption cross section.
Botton Tie Plate Paintains fuel rod array Cast SS, Grade CF-3 and distributes coolant
- Strength.
to fuel rods.
- Corrosion resistance.
Spacer Capture Rod Maintains correct longi-
- Continuous clad and tudinal position of formed Zircaloy sheet
- spacers, stock connectors.
Tie Rod Provides structural skeleton Zr-2 clad fuel rods with end of assembly by securing the fittings for attach =ent to l
upper and lover tie plates.
tie plates.
k 15
i
(
I
\\,
k.3 Thermal Design The design basis for the thermal performanc? of Reload G-3 fuel is identi-cal to that described in our submittals dated June 16, 1972 and October 13, 1975 for Reload G and G-1U, respectively.
~
h.4 Chenical Design The adequacy of materials selected for the chemical fuel design has been demonstrated through the excellent performance of Exxon Nuclear Fuels to date. Past irradiation tests for assemblies similar to G-3 have produced no fuel failures or degradation due to incompatibility with the reactor water chemistry. Results of the post-irrt.diation examinations of fuel assemblies, including those of the G design, are contained in Special Report No 2k dated November 2k, 1976.
i 9
5 16
,-,=w-
.--w-,,
s,,
s,m-..
.-c-.
y
i.
t
.{
1 50 NUCLEAR DESIGN The fresh fuel to be used for Cycle 15 is Exxon Nuclear's Type G-1U and Type G-3 Fourteen bundles of Type G-1U fuel are currently used in Cycle 1k.
Important differences between the reload G-1U and reload G-3 fuel bundle designs are the replacement of the four corner cobalt target rods with lov enrichment fuel rods, a change in the gadolinia poison pin locations and changes' in the bundle enrich =ent distribution which reduce the bundle average enrich =ent from 3.88% for G-1U fuel to 3.1h% for G-3 fuel. The effects of these changes on local peaking factor and fuel bundle reactivity have been accounted for in ce=puting the core physics characteristics.
5.1 Physics Characteristics As discussed earlier, previous reload licensing submittals for the Big Rock Point Plant did not include esny of the physics para =eters requested in the " Guidance for Proposed License Amendments Relating to Refueling,"
(
thus these para =eters are not available for previous cycles and con-sequently no refereace cycle, meeting the triteria of " reference cycle" as defined in the guide, is available. It is the intent of Consumers Power Co=pany to calculate the para =eters required by the guide for Cycle 15 and to utilize them-in subsequent licensing submittals as the reference cycle for physics parameters. These are included as Table 5 1-1 and Figure 5.1-1.
Table 5.1-1 includes the full power doppler coefficient, delayed neutron fraction, void coefficient and total peaking factors for both the Beginning of Cycle and end of Cycle Conditions The maximum reactivity for in sequence rod drop vorth is 2.295 ak/k for BOC and 1.Th% Ak/k for ECC, both cases well below the Technical Specification limit of 2.55 ak/k.
The Cycle 15 core can be maintained suberitical in the most reactive con-dition throughout the operating cycle with the most reactive rod fully withdrawn and all other rods fully inserted. The Technical Specificatica concerning shutdown =argin is 0.3% Ak/k which is significantly less than the BOC and EOC values for shutdown cargin listed in Table 5.1-1.
Figure
(
5 1-1 is the full shutdown =argin curve for Cycle 15.
17
i s
I The limiting scram reactivity curve for Cycle 15 is provided in Fig-ure 5.1-2.
This curve is most limiting at EOC vhen the control rod density in the critical rod configuration is the lowest. Figure 5 1-2 is a comparison of cycle 15 scram reactivity to the bounding curve of Cycle 11. The Cycle 11 curve is considered limiting since it was uti-lized in the reference analysis of the rod drop accident. the accident most sensitive to scram insertion characteristics.
b Two I ysics'pa'rameters requested by the guide are not provided in this submittal. The moderator temperature coefficients are not considered in acy accident or transient analysis, and essentially have no meaning for a boiling water reactor other than in its initial heatup prior to power operation. However, these vill be calculated prior to power operation for Cycle 15 The other parameter not provided is the worth of the standby liquid control system. The worth of the standby liquid control fr system is being evaluated for this cycle.
TABLE 5.1-1 Parameter BOC EOC Doppler Coefficient (Ak/k/% Pover)
-7 06x10-5 -7.65x10-5 Maximu= Radial x Axial Peaking Factor 1.697 1.668 Maximum Radial x Axial x Local Peaking Factor 2.hT9 2 306 Maximum Rod Worth (5 Ak/k) 2.29 1.Th Delayed Neutron Fractio'n
.00606
.00588 2.21 6.21 Shutdown Margin (% ak/h)
Void Coefficient (ak/k/ Unit Void)
.1663
.1127 52 Analytical Intut Reactor power distributions, reactivities, reactivity coefficients, fuel burnup and margin to ther=al limits are calculated with the GROK computer program. GROK is a three-dimensional-coarse mesh reactor si=ulator with ther=al hydraulic feedback and is a derivative of the FLARE program.
(D L Delp, et al, " FLARE, A TERIE-DIMENSIONAL BOILING WATER REACTOR SIMU-LATOR," GEAP h598, July 16, 196h.) The neutronics parameters k.,, M and 4
18
i-e
.1
.I.
s.
t local peaking factor, as a function of local operating state as computed by the fuel designer, are the major inputs. Algorithms have been in-P cluded which calculate peak heat flux, MCHFR, MAPLEGR and theoretical 1
I flux vire traces for comparisen with reactor measurements.
i 53 Changes in Nuclear Design There are no changes in core design features, calculational methods, data or information relevant to determining importact nuclear design parameters, other than those mentioned above, for Cycle 15 d
t e
4 4
f e
i i
i 3
i e
4 I
19 l
4
+w,,.
, - - - +
9 v-on.
.em.
e v
,r
~-r, r
,-y
, - - +,, -,
i FIGURE 5,1-1 SHUTDOWN MARGIN VS EXPOSURE CYCLE 15 7
2
-d M Y
M Zs a,
j a:
Zk 8ga J
5 g
1 0
2 3
4 5
6
(
EXPOSURE (GWD/ 7) 20
t I
Figure 5.1-2 1
(
BIG ROC 4 POINT PLANT SCRAM CURVE REFERENCE CYCLE (CYCLE 11)
END OF CYCLE 15 o 1.0 g 0.9 r
r E
0.8 f,
t E 0.7 s
0.6
$ 0.5
/
M 0.4
/
ll a
6 0'3 9
/<,/
0.2
=
u.
O
//
2 0.1 L '
_/
v 0
0.5 1.0 1.5 2.0 2.5 3.0 m
TIME (SECONDS)
\\
[ ;
Y c
- 6.0 THERMAL HYDRAULIC DESIGN The hydraulic design of the reload G-3 asse=blies is identical to that of the reload G-lU assemblies. The thermal performance of reload G-3 fuel differs slightly from reload G-lU in that the average rod power has changed since the four cobalt corner rods have been replaced with lov enrichment UO fuel rods. This has led to a decrease in average rod 2
power, consequently'it has also led to a decrease in maximum overpower clad and fuel te=peratures and an increase in the overpower minimum i
-itical heat flux ratio. A comparison of the thermal hydraulic pa-rameters for complete cores of reload G-3 and reload G-lU fuel assem-blies is contained in Table 6-1.
Although Cycle 15 vill be predominantly 11 x 11 reload G-type fuel as-l semblies, approximately 25% of the core vill be 9 x 9 reload F fuel assemblies. However, the maximum radial peaking factor for the reload F fuel for Cycle 15 is expected to be 0.8h.
This vill result in a 122%
rated power minimum critical heat flux ratio of greater than 2.0 for
+
s=all and large orifice channel loct.tions. Therefore, mini =um critical heat flux ratio units for reload F fuels will not limit core power op-erations for Cycle 15
)
I 22 1
i
.m
._y, y
,,_y,
~.
4
'(
TABLE 6-1 Therer.1 Hydraulic Parameters (Core Contains All of Each Type Fuel) i Reload G-3 Reload G-lU Core Conditions Reference Design Themal Output,- (MW )/(Btu /h) 2h0/8.191 t
6 Total Flow Rate, Lb/h 12.3xig Effective Flow Rate for Heat Transfer, Lb/h.
9 9 x 10
. System Pressure, Nominal in Steam Dome, Psia 1350 Assembly 22*er_iption
-Rod Diameter, Inches 0.kh9 0.hh9
[
Rod Pitch, Inches 0 577 0 577
-Number of Active Rods 117 113 Total Fuel Length per Assembly, Feet.
652.5 659.2 2
Heat Transfer Area, Ft.
00.23 77.h8 2
2 0.163/23.hh 0.163/23.hk Flow Arer., Ft /In Design Power Peaking Factors Fraction Generated in Fuel, %
97 0 96.6 Fuel Asserbly Power Factor 1.h3 1.h5 Local Peaking Factor 1.20 1.20 Axial Peaking Factor 1 51 1 51 Engineering Heat Flux Factor 1.0h 1.0h y
1 Assembly Thermal Performance Maximum Heating Rate, kW/ft, at 22% Overpower 13 53 1h.0 Maximum Heating Rate, kW/ft, at Rated Power -
11.09 11.hk Average Heating Rate, kW/ft h.06 h.20 Maximum Heat Flux, Btu /h-Ft at 22% Overpower 392,900 h07,000 Maximum Heat Flux, Btu /h-Ft at Rated Power 322,100 333,600 2
Average Heat Flux, Btu /h-Ft at Rated Power 117,900 127,100 Maximum UO2 Temperasure, *F, at 22% Overpower 3879 3900
' Maximum Clad Temperature, *F, at Overpower Th5 75h MCHFR at Overpower Conditions Axial Peak at X/L =.h5 1.68 1.63 Coolant Subcooling at Core Inlet, Btu /Lb 22.8 22.8 Assembly Hydraulic Perfomance Average Assembly Flow Inner Orifice Zone, Lb/h 132,900 132,900 80,500 80,500 Outer Orifice Zone, Lb/h (2k Assemblies on Periphery of. Core) 6 6
Active Core Flow at Design Power Lb/h 9 9 x 10 9 9 x 10 Hot Assembly Flow at 122% Design Power (Reference Design Flow) 123,000 123,000 Assembly AP at Average Design Power
[
(Includes Orifice AP) 5 37 Psi 5 37 Psi j
Hot Assembly Engineering Enthalpy Rise Factor 1.10 1.10
' Crud-Free Surface
[,
L 23 i.
1
I 70 TRANSIENT AND ACCIDENT ANALYSIS In order to update this section, the NRC Standard Review Plans, Regulatory Guide 170, and the General Electric Standard Safety Analysis Report were thoroughly researched to determine what accidents, transients and liniting design criteria were necessary for a proper review. The " reference cycle" for accident and transient analysis consists of th' latest analysis run for each. In many cases these date back to the FHSR, but wherever analyses have been run subsequent to this, they have been used as the reference cycle. The references for this section are contained in Subsection 7 3 71 Transient Analysis T.1.1 Significant Reactor Kinetics and Fuel Thermal Hydraulic Design Para =eters For each reactor transiet considered in previous licensing submittals for the Big Rock Point Plant, the reactor kinetics parameters which control the reactor transient response are shown in Table 7-1.
Also shown in Table 7-1 are the reference value and the corresponding Cycle 15 value for each significant parameter. Below is a discussion of the effects that the Cycle 15 values are expected to have on the reactor transient response.
Important to the analysis of the reactor transients is the the. 21 and hydraulic design of the various fuel bundles comprising the reactor core. All Big Rock Point fuel bundles up to and including the Reload G-3 fuel bundles have been designed to =eet the following constraints.
(Refer to Section 6)
(1) Minimum critical heat flux ratio (MCHFR) at design overpower (122%) and design peaking factors must be greater than 1.50.
(2) Maximum fuel temperature at design overpower and design peaking factors must be less than the fuel melting temperature.
Given that these constraints are met, the thermal response of each fuel type (ie, MCHFR, peak fuel temperature, peak clad temperature) to a given transient vill be as previously predicted or better.
(.
2h
TABLE 7-1 Significant Reactor Kinetics Parameters Nominal Important Cycle 15 Event Latest Analysis Kinetics Parameter (s)
Reference Value Value Loss of External Load With Reference 1 Void Coefficient BOC:
.20'
.1663 and Without Turbine Bypass Page 13 EOC:. 3 0*
.112T (Bounds Main Steam Line (ak/k/ Unit Void)
Isolation Valve Closure
-5 and Loss of Condensor Doppler Coefficient
-5.h2x10- Ak/k/% Power.
-7 06x10 Vacuum)
Steam Pressure Regulator Reference 1 Void Coefficient BOC:
. 20"
.1663 Failure Resulting in Page 10 EOC:
.10F
.1127 Reduced Steam Flow (ak/k/ Unit Void)
-5 Doppler Coefficient
-5.42x10-5 ak/k/5 Power
-7 06x10
- Uncontrolled Rod With-g
'a drawal From Suberitical
-5
-5
- Cold Start-Up Reference 1 Doppler Coefficient
-1.47x10 ak/k/*F
.95x10 Page 6 Maximum Reactivity 3.9% ak/k 1 77%**
Addition S/t*
175 183
-5
-5
- Ilot Start-Up Reference 1 Doppler Coefficient
-1.37x10 Ak/k/*F
.95x10 Page 6 Maximum Reactivity 4.2% ak/k 2.29%**
Addition S/t*
175 183 Uncontrolled Rod With-Reference 1 Void Coef'icient-BOC:
.20'
.1663 f
Drawal at Power Page 12 EOC:
.10'
.1127 (ak/k/ Unit Void)
-5 Doppler Coefficient
-5.42x10- ak/k/% Power
-7 06x10 I
,ah TABLE 7-1 (Contd)
Nominal-Important Cycle 15 Event Latest Analysis Kinetics Parameter (s)
Reference Value Value
- Inactive Recirculation Reference 1 Void. Coefficient BOC:
.20'
.1663 Pump Start-Up Page 22 EOC:
.10'
.1127 (Ak/k/ Unit Void)
-5.12x10 ak/k/5 Power
-T.06x10 '
-5
~
Doppler Coefficient 6
e
- Loss'of Recirculation (This event is reevaluated for each new core loading Pumps using the methods described h Appendix B of Reference 2.)
S
' Reference T.
~
Reference value is maximum worth for cut-of-sequence rod. Cycle 15 value is maximum worth for in-sequence rod.
k l
i 7 1.2 Pressurization Events (Loss of Load, Turbine Trip, Fain Steam Line Isolation Valve Closure, Steam Pressure Reguitor Failure)
Pressurization events are characterized by a decrease in voids resulting in a power increase. The more negative void coefficient for BOC conditions tends to maximize the system pressures and powers reached in these events. Since the reference cycle BOC void coeffi-cient is more negative than expected at any time during Cycle 15, the reference cycle analysis conservativel'y bounds the upcoming cycle.
7.1.3 Inactive Recirculation Pump Start-up or Cold Wster Event Like the pressurization events, this event is characterized br a decrease in voids and a power increase; therefore, the void coeffi-cient is again the i=portant kinetics parameter for tbis event.
Because the reference cycle BOC void coefficient is more negative than expected at any time during Cycle 15, the reference analysis i.
bounds Cycle 15 7 1.h Loss of Recirculation Pu=ps This c?ent is reevaluated for every core reloading using the method de-scribed in Appendix B of Reference 2.
Results of this analysis, althou6h presently incomplete, are expected to show, as they have shown for previous cores, that the minimum critical heat flux ratio never falls below 15 and, in fact, monotonically increase throughout the critical portion of the transient. Therefore, this event is not limiting for Big Rock Point.
7 1.5 Rod Withdrawal at Power This event is characteri:ed by increases in core power level and core voids. The void and doppler coefficients are the important kinetics parameters for this event. As noted in Table 7-1 the Cycle 15 doppler coefficient is more negative (ie, more conservative) than was assu=ed in the reference analysis. In addition, the Cycle 15 '
void coefficient is more negative than the worst case (EOC) void coefficient for the reference cycle. Therefore, it is concluded
(.
that the reference cycle analysis bounds Cycle 15 for this event.
27
i
. 7 1.6 Start-Up Event The start-up event (or the uncontrolled rod withdrawal fro = sub-critical) was maaly:ed in Reference 1 for both the cold and hot standby initial conditions. The start-up event is characterized by an extre=ely rapid increase in nuclear power (to approxicately 100 times rated pover) followed by an equally rapid power reduction due to doppler feedback. The important kinetics para =eters for this event are the doppler coefficient, the ratio of BETA /t*,
and the =aximum reactivity addition due to the withdrawal of a control rod while suberitical. As noted in Table T-1 only the Core 15 doppler coefficient is significantly nonconservative as co= pared to the values assu=ed in the reference analysis. The ma dmu= reactivity addition is =uch less than assu=ed for the reference analysis, and the BETA /t* ratio is nearly the same as assumed in the reference analysis.
If, however, the event were I
reanaly:ed assu=ing the Cycle.15 value for the doppler coefficient and assu=ing the sa=e rod worths and BETA /i* ratios as in the ref-erence analysis, the consequences of this accident would still not be severe. Assuming a linear relatiouship between doppler coeffi-cient and fuel effective temperature rise, the reduced Core 15 dopp-1er coefficient would result in full effective te=perature rises of 900*F and 850 F (as co= pared to 580*F and 590*F in the reference analysis) for the cold and hot start-up events, respectively.
Thus, assu=ing a hot spot peaking factor of 3 0, the peak fuel te=peratures of 2800 F and 3100 F for the cold and hot start-up events, respectively, vould still be significantly less than the fuel =elting te=perature. This is still extre=ely conservative since a hot spot peaking factor of 3 0 is significantly greater than vill be allowed during Cycle 15 based on Eccs li=itations.
72 Accident Analysis T. 'i.1 Loss of Coolant Accident (LOCA)
The Big Rock Point Loss of Coolant Accident analysis 'for Exxen Nuclear
(
Co=pe.ny (ENC) fue) was perfor=ed with E'iC calculational =edels which 28 4
t I
5 are consistent with the requirements of Appendix K of 10 CFR 50.
The appropriate assumptions and results of the ECCS analysis for Reload G-3 all-uranium fuel vere documented in Reference 3 This report was submitted to the Director of Nuclear Reactor Regulation
~
on February 18, 1977 in support of a proposed Technical Specifica-tions change dated Dece=ber 17, 1976 for updating MAPLHGR limits for Exxon Reload G and Reload G-1U fuel. The same report also in-cluded a reanalysis of Loss of Coolant Accident for, Reload G and Reload G-lU fuel. The limiting break size for all three fuel types (Reload G-3, Reload G and Reload G-1U) was identified to be a 2
0.25 ft small recirculation line break.' MAPLEGR limits as a fune.
tion of barnup were also provided in this report. Limits for all other fuel types (General Ele:tric F and Modified F) which will be reloaded into the core for Cfele 15 vill remr.in unchanged from values approved by the NRC for previous cycles (Reference k), with
{
one exception discussed in Section 8.0.'
7.2.2 Rod Drop Ascident The control rod drop accident has been previously nnalyzed in Reference 5 The worst case (hot standby) was analyzed for both 4
an all-uranium core and a mixed-oxide core. The important kinetics parameters for the control rod drop accident are listed below along with the values assumed in the analysis and the Cycle 15 values.
Assumed Value Parareter Uranium Core Mixed-Oxide Core Core 15 Value Effective Delayed Neutron.Fractiot
.00591
.00529
.00588
~I
-Doppler Coefficient
.916x10"' ak/k/*F
.96x10 ak/k/*F
.95x10 ak/k/*F Maximum Worth of a it.5% ak/k 2.5% ak/k 2.29% ak/k Single Control Rod The Cycle.15 s alues for the important kinetics parameters are very similar to the values assuned for both cores analyzed in Reference 5 The Cycle 15 v tlues of doppler coefficient and BETA are bounded by the
(
values assumed in the two analyses. -Based on this comparison, the reference analysis is considered conservative for the upec=ing cycle.
~
i 29
(
t 7.2.3 Anticipated Transient Without Scram (ATWS)
The consequences of the most limiting ATWS event, the loss of load without turbine bypass, were previously evaluated in References 1 and 6.
These analyses assu=ed a void coefficient much more negative
(.20 Ak/k/ unit void) than expected at any ti=e during Cycle 15,
-3 and a doppler coefficient much less negative (-5.k2x10 ak/k/7.)
than expected during Cycle 15 Thus, the previous analyses are considered conservative for the upcoming cycle.
73 References 1.
APED-h093, " Transient Analysis, Consumers Power Company Big Rock Point Plant," October 1962, and/or Big Rock Point Final Ha::ards Su==ary Report.
2.
GEAP kh96, " Core Performance snd Transient Flow Testing - Big Rock Point Boiling Water React'or," July 1965 3
XN-NF-76-55, Revision 1, "ECCS Analysis for Exxon Nuclear Co=pany "
G-3 All Uranium No Cobalt Fuel for Big Rock Point (Including Reanalysis of Reload G and G-1U Designs)," February 1977 h.
" Big Rock Point Plant Loss-of-Coolant Accident Analysis for General Electric Tuel in Confor=ance With 10CFR50 Appendix K," July 11, 1975 (Submitted as Appendix A to a Technical Specifications change request from Consumers Power Company to the NRC dated July 25,1975.)
5 Technical Specifications change request from R B Sewell (CP Co) to J F O' Leary (USAEC) dated June 20, 197h.
6.
NEDE-21065, " Anticipated Transients Without Scram Study for Big Rock Point Power Plant," October 1975 7
Proposed Technical Specifications change dated January 17, 196h.
(
30
(
l 8.0 PROPOSED MODIFICATIOUS TO TECICIICAL SPECIFICATIONS The proposed Technical Specifications are contained under Section I of this submittal. In general, the changes consist of proposing specific parameters and drawings for the Reload G-3 fuel for the Bic Rock Point Technical Specifications with justification presented in Sections 1 through 7 The MAPLHGR limits, as proposed, are consistent with the pro-posed MAPLHGR limits contained in the Technical Specifications change I
request dated December 17, 1976. Justification for these limits is con-l tained in Exxon Report XN-NF-76-55, Revision 1, forwarded to the Commission on February 18, 1977 There is also a minor correction to the MAPLEGR for modified F fuel
'.t a burnup of 25,000. By letter dated July 25, 1975, we proposed a MAP'.HGR limit of 8.14.
In Amend =ent 10, dated June k, 1976, this value was incorrectly transposed to 8.7 We propose to correct this back to 8.*.
One proposed change to the Technical Specifications has not been addressed up to this point. That change is the deletion, from Section 5 2.l(b) and Table 8.2, of the limitation for the " Maxi =um !&d/T of Contained Uraniu= for an Individual Bundle." This limitation first appeared in Consu=ers Power proposed Technical Specifications for the Big Rock Point Plant dated June 1, 1962. It was contained in Section 5 2.2, " Principal Calculated Nuclear Char-acteristics of the Core." This section contained specific para =eters relevant to Cycle 1 for the Big Rock Point reactor (eg, moderator and void coefficients, topplers, reactivity balance, average igd / Ton of contained uranium, etc). It is apparent that although the other design parameters associated with the original core composition vere updated or deleted as necessary to account for the different reload fuels, the limitation on maximum fuel burnout was maintained intact for each subsequent reload licensing submittal.
By letter dated January 20, 1577, Consumers Power Company responded to a letter from Mr D L Zie= ann concerning fission gas release from fuel pellets with high burnup. In our response, we calculated the relevant parameters for all G series reload fuels for burnups ranging from 30,620 to 38,935
!Gd/MT. This is well above the design burnup of the fuels.. The results
(
of this analysis indicated that the fission gas release modQ burnup had 31 4
('
(
very little effect on the peak clad te=perature prediction (less than 1%
reduction in MAPLHGR limits at end of life) and consequently insignificant effect on the Big Rock Point LOCA analysis and, therefore, was of no safety concern. We further indicated that these results vould be consistent for the 9 x 9 fuels.
Attached as Appendix 1 to this report is an analysis of the fission product inventory change with increased burnup for the Big Rock Point core. This analysis was conducted assuming a core average burnup of 30,000 WD/STU, which is also higher than the design average burnup of the Reload G fuel. The results of this analysis indicated that the radiation dose increases due to in-creased fission product inventory were insignificant and would remain in-significant (less than 1% of total dose) until tr e fuel burnups reached 6
approximately 15 x 10 Wd/T.
~ Further justification for deletion of the maximum burnup limitation exists in the Standard Review Plans. Section h.2 states in part, "The cladding design should be such as to acco=rodate the fission gas evolved in opera-tion, so that the fuel can reach design burnup without exceeding the cladding structural design criteria." By virtue of the preceding dis-cussion and analyses, Consu=ers Power Company concludes that it adequately meets the Standard Review Plan criteria and therefore no arbitrary bumup li=itation is necessary. Also, the General Electric Standard Technical Specifications for Boiling Water Reactors makes no mention of the maximu=
fuel burnup allovable. Since Big Rock Point is in the process of conver-sion to the standard for=at, we feel that consistency would also dictate deleting this arbitrary limitation.
Thus, based on safety analyses performed concerning fuel burnup and on guidance developed from the Nuclear Regulatory Cc==ission in the fom of the Standard Reviev Plans and Standard Technical Specifications, Consu=ers Pcver Cc=pany concludes that the li=itation on maximu= fuel burnup is un-necessary and should be deleted from the Big Rock Point Technical Epeci-fications.
32
i
(
9.0 START-UP PROGRAM The testing and start-up program planned for the next refueling outage and subsequent start-up vill include:
(1) Control rod drive testing, as required by the Technical Specifica-tions, including scram times.
(2) Core shutdown margin verification with the most reactive rod with-drawn.
(3) Critical control rod pattern.
(k) Measure =ent of flux shapes during power escalation and co=parison to ce=puter predictions.
A brief explanation of these tests is contained below.
Shutdown Margin Verification:
Core shutdown margin is verified at the beginning of each cycle and dur-ing the first cold shutdown after 35,000 MWde generation. The Technical Specifications require su'ocriticality to be demonstrated with the most reactive rod withdrawn from the core as well as an im=ediately adjacent rod known to contribute.003 k,ff or more. Ar analytical determination of the highest worth rod is made, as well a.= the number of notches of an ic=ediately adjacent rod required to contribute.6% ak/k.
35 ak/k is added to account for a reactivity increase at temperatures higher than the ambient te=perature at which the test is performed (nor= ally 10d to 204). Plant procedures call for the individual vithdrawal of each control rod in the core, tius at least the nu=ber of r.otches specified in the p'hysics analysis on an sijacent rod to verif t the analytical determination.
Core monitoring is provided by gas-filled B sron 10 lined proportional counters and may be supple =ented by portalle fission chambers positioned above the core when available count rate is lov.
Rod Drive Scrsm Time Testing:
Technical Specifications requirements state the maximu= control rod drive scram time from the fully withdrawn position to 90% of insertion shall not exceed 2.5 seconds. This require =ent is verified at the beginning of each cycle by attaching leads from a strip chart recorder traveling at a predetermined rate to the position indication for the drive to be tested
[
and to the 26 volt d-c signal from the high reactor pressure input on one 33
\\
\\
I k
of the two safety channels. The control rod is fully withdrawn and a hidt reactor pressure trip is simulated by removing power from the reac-tor pressure input to the safety system logie circuMry. Scram ti=e is me asured on the strip chart from the time of safety channel trip to full Ansertion. The test is repeated for all control rods using both safety channels.
Critical Rod Pattern:
Control. rod withdraval sequences and initial critical rod patterns are analytically determined at the beginning of each cycle. La completion of core reconstitution and shutdown margin verification, an initial criticality at ambient conditions is performed. Integrated control rod worth curves and shutdown margin are adjusted based on the conditions of actual critical rod pattern. On attaining steady state equilibrium conditions at rated power, a reactivity balance is performed and the ac-tual critical rod pattern is verified to be within 15 ak/k of expected.
Figure 91-1 shovs the cold critical control rod pattern at BOC and EOC.
Power Eistribution Measurement:
Flux aistribution measurements are me.de during the escalation to rated power after the beginning of a new cycle by inserting of copper-titaniu=
alloy wires into the core and counting the copper activation. Predictions of this flux vire activation are made with GROK, a three-dimensional one group diffusion theory code, using actual operating conditions as input.
(See Section 5.2.)
Power distribution calculations are then adjusted based on the flux vire, GROK calculation comparison. In-core instrumentation is calibrated to conform with flux vire measurements. Ther=al hydraulic analysis based on the flux vire corrected power distribution calculations are compared to MAPLHGR, MCHFR and heat flux limits to insure confor:.uce with the Technical Specifications.
k
(
3k
t I
Figure 9.11 COLD CRITICAL ROD PATTERNS BOC A
B C
D E
F 1
4 6
4 6
J 2
6 0
0 0
0 4
N 3
4 0
0 0
0 6
4 6
0 0
0 0
4 5
4 0
0 0
0 6
6 6
4 6
4 BRP CONTP0L ROD POSITIONS EOC A
B C
D E
F I
23 9
23 9
)
2 9
0 0
0 0
23 N
3 23 0
0 0
0 9
4 9
0 0
0 0
23 5
23 0
0 0
0 9
6 9
23 9
23 BRP CONTROL ROD POSITIONS
(
l l
~
0 is all in l
23 is all out 35
(
(
III. CONCLUSIONS Based on the foregoing, Big Rock Point Plant Review Co:::mittee has con-cluded that this change does not involve an unreviewed safety question.
CONSUMERS POWER COMPANY By C R Bilby, Vice Pr @ ent Pr;**tetion & Transmitsion Sworn and subscribed to before me this 15th day of April 1977 L.J)
< % 0/r'1)
'k k??/[4 )
Linda R Thayer, Notary Public Jackson County, Michigan My com.ission expires July 9, 1979 g'
1 I
36
Appendix 1
(
FISSION PRODUCT INVENTORY CHANGE WITH INCREASED BURNUP - BIG ROCK POINT PACKGROUND Radiation dose analysis for the "MCA" in the FHSR assumes 1/3 core at 5000 Wd/T,1/3 at 10,000 Wd/T and 1/3 at 15,000 Wd/T. All nuclides which are significant in radiation dose contribution (Table A-1), with the exception of Kr-85 and I-129, have achieved steady state equilibrium at the lowest burnup utilized in the FHSR. Consequently, only Kr-85 and I-129 continue to increase as burnup increases.
CALCULATIONS The FHSR does not list specific quantities for core inventory but activities for the 10% release case were determined for an earlier analysis (Table A-1) to be 1.kl x 10 Curies of I-129 (10% of core inventon) and 1.88 x 10 Curies of Kr-85 (30f, of core inventory) after two years of full power operation. Two years were chosen to provide conservative inventories of Kr-85 and I-129 equal to a full core at 15,000 Wa/T. Activities are calculated for 30,000 Wd/T by Equation I:
.g Activity = (8.h35 x
) (Y) ( A) (2ho W ) (GR) (1-e
) Equation I 0
uci-se t
t where: Y = Fission Yield A = Decay Constant (Sr.c
)
GR = 0.1 for I-129, 0 3 for Kr-85 T = k Years Irradiation (1.26 x 10 Sec) 10 Kr-85 Activity = 3 53 x 10 pCi Released to Containment I-129 Activicy = 2.82 x 10' uCi Released to Containment Contributier to off-site dose from leakage of the above quantities at maxiem containment 1sak rate (FHSR Figures 13.2 and 13.3) of h.3 x 10-9/see is per-formed as follows:
Rad /h = 0 35 % (X/Q) (Q/See) (3600
)
Equation II Kr-85:
h where:
X/Q
= Diffusion Censtant for Ground Level Release at 8h2m, From Regulatory 3
Guide 1.3 (4.5 x 10 Sec/m )*
NOTE:
- 8.6 x 10 ' froc Figure 3.A divided by a wake correction factor of 19 fre: Figure 2.
1
i I*
l
~
-9/Sec) (1.88 x 10 Ci) = 8.1 x 10-5 cifs,c Q/Sec = Release Rate = (k.3 x 10 E = Average Energy per Disintegration (0.002 Mev)
Y
-0 From Equation II, Kr-85 dose rate equals 6.6'x 10 rad /h to the total body.
In comparison, FHSR Section 13.11.h indicates the maximum dose rate from the plume (all conponents) is 5 x 10-3 rad /h. Thus, the percentage due to Kr-85 is insignificant at (6.6 x 10 /5 x 10-3) (100%) = 0.00135.
I-129: Rad /h = (K) (X/Q) (Q/Sec) (B)
Equation III where: K = Dose Conversion Factor per Regulatory Guide 1.109 (5 55 Rem /uci Inhaled) 3 X/Q = Diffusion Constant (h.5'x 10- Sec/m )
-9 5
Q/See = Release Rate = (h.3 x 10 /See) (1.h1 x 10 pC1) = 6.1 x lo pCi/See 1
B = Breathing Rate per Regulatory Guide 1.3 (1.25 m-/h)
From Equatier. III, I-129 thyroid dose commitment from one hour of inhalation
-6 6
i equals 1 9 x 10 rem, or 3.8 x 10 re= from two hours of exposure. In com-parison, FHSR Section 13.1h.3 indicates a 2-hour thyroid dose of 2 ren is ex-pected from the total halogen mixture. Thus, I-129 is insignificant at (3.8 x 10 /2) (1005) = 0.00019%.
CONCLUSION Dose contributions from nuclides affected by increased fuel burnup are negligible relative to total "MCA" doses. Radiation dose increases due to increased produc-tion of I-129 and Kr-85 vill remain insignificant (less than 1% of total dose) 0 up to core burnups of approximately 15 x 10 mwd /T.
a 2
a 4
{
t
(,
-TABLE A-1 Full Core Fuel Pin Gap i
(
Radioactivity Released Gap Activity (pCi)
Big Rock Point Plant Initial Design Calculations From Quanicassee Design Data GE - Modified for 80/20 Ratioed to Big Rock
' Isotope-1(Min 1)
U-235 Pu-238 Fission Mixture Point Power Levels-I-129 T.T6E-14 1.41E+05 I-131 5 98E-05 5.62E+12 5 92E+12 I-132 5 11E-03 8.35E+12 9 02E+12 I-133 5 69E-Oh 1.37E+13 1.33E+13 I-134 1.33E-02 1.k5E+13 1.56E+13
.I-135
'1 73E-03 1.29E+13 1.21E+13
.Xe-138 4.88E-02 1.21E+12 1.21E+12 Kr-87 9 12E-03 k.2E+11 5.11E+11 4
Kr-88 k.1kE-03 6.k3E+11 T.2TE+11 Kr-85m 2.62E-02 2.39E+11 2.66E+11 I
Xe-135 1.26E-03 1.37E+12 3.TkE+11 l.
'9 12E-05 1.36E+12 1.3TE+12 Xe-143 k.33E+01 8.11E+09 Kr-9h k.16E+01 1.80E+10 Kr-93 3.22E+01 9 3kE+10 Xe-141 2.k2E+01 2.23E+11 Kr-92 2.26E+01 3.2kE+11 Kr-91 k.8kr+00 5.81E+11 Xe-1k0 3.06E+00 6.89E+11 Kr-90 1.29E+00 8.52E+11 Xe-139 1.04E+00 9.TkE+11 Kr-89 2.18E-01 8.09E+11 Xe-137
~1.81E-01 1.20E+12 Xe-135m k.k2E-02 2.12E+11 3.68E+11 Kr-83m 6.18E-03 9.82E+10 Xe-133m 2.13E-04 3.82E+10 3.kTE+10.
Xe-131a 4.03E-05 3.62E+09 k.51E+09 j
Kr-85 1.22E-07 1.88E+10 2.02E+10 Note: For I-129 and Kr-85 equilibrium is not-obtained. Hence, the activity available is.the equilibrium value at full power operati'on for two years
(
times (1 - e-At) where t is two years.
3 r
i r
4
,w--.
.y--
.w-s-,-e-y-
3-
. m
m NRC DISTRIBUTION ron (RT 50 DOCKET M ATERI Al.
T.
,i FROM:
DATE or DOCUMENT Consumers Power Company 4/15/77 Jackson, Michigan Mr. D. Davis oAtt ntcovc o David A. Bixel 4/19/77 TTEn CNOTonl2ED PnoP lNPUT FOnM NUMDEn or CoPfLS RECLIVE D (onIGIN A L JNCLASSIFl[o g
y,,; g p y y Q *j c f 9
JCOPY ENCLoSUHL DLSCnlPTioN
,Ltr. trans the following:
Amdt. to OL/ change to tech specs to allow the use of an initially all uranium fuel type as reload fuel....
!%CKNOWLEDGED.
j (2-P)
(44-P)
PLANT NAME:
i
' p0 NOT REMOVE ig Rock Point t
h NY bJ F
.O RAFETY FOR ACTION /INi-ORM ATION p yLt;O ASSIGNED AD:
ASSIGl:ED_AD
- bERAUClLCHIEI':
l 'y,._
U-h.l._Cfaft "3 M nRI.UCIL. CHIEF:
'IPRO,1ECT M'.NACt;R:
eX-QC 73 PROJECT 1'A??ACER:
d)_[of.s $
LIC, ASST. :
Ay LIC,_ A S ST., :
w i
INTERNAL DISTRIBUTION Y, iQFIIT SYSTEMS SAFETY PLANT SYSTEMS SITE SAFF.TY_A Y.l! M IM llE.IIIEl'4!!
TEDESCO EWIn0_/J!ALYSIS_
,I ilI & E hy_t
. SCllR0EDER EENAROYA DU;IIQ'LEda'LLEP
. ', ! OELD LAINAS
{COSSICK& STAFF E!!GI;;EERING IPTOLIT.0 ENV T E O._.TECE.
KIRIX00D
_ERNST IIIPC MACARRY CASE E03'.V."
11ALLARD llANAUER SI1nh,IL OPERATING REACTORS SPANCLER llARI.ESS PAWI,IC.!;I STELLO I
SITE TEC11 PPfLTECT MANAGEMENT REACTOR S/fET_y OPERATING TECll.
_ _ __CArplILL
._k EISENilUT STEPP
{'
80YD ROSS 110USTON
_S;iAQ llDLMAN P _ COL.LI]iS NOVAK ROS7. TOC 7.Y m _J3 AEg SITE ANALYSIS PETEllSON CliECK h_~Elfrl,ER ggig1;S V01.1.MER MELTZ BUNCil lil:1.TEMES AT & I
_ ([. COLLINS EKOVil01.T SA1.TZMAN
'/
Ri!TP.ERC KREGER
.I YllitN AL DISllllflUTION CONI HOL NU.'. HIE R
.hION b18ElsC.Nf5[
ik
_UR00KilAVEM. IIAT._ LAB..
7, e
TIC:
REC V.lE ULRIKSON (OltNL) y/.,
aR k<mNSIC[
1.A Pnn
~ ~ '
h 7"/""
~
A SI.B :
CONSiti.TANTS :
ACRS CYji tett.cu;/ :I[t if" ( 8 7~ [$
~
3g g{
~nc r onM m im,
,