ML20094N285
| ML20094N285 | |
| Person / Time | |
|---|---|
| Site: | Oyster Creek |
| Issue date: | 04/02/1992 |
| From: | Devine J GENERAL PUBLIC UTILITIES CORP. |
| To: | NRC OFFICE OF INFORMATION RESOURCES MANAGEMENT (IRM) |
| References | |
| 5000-92-3017, C321-92-2098, NUDOCS 9204060267 | |
| Download: ML20094N285 (12) | |
Text
,
GPU Nuclear Corporation u.. Nuclear e- -
Parsippany, New Jersev 07054 201 316-7000 TELEX '.36 482 Wnter's Direct Dd Numter April 2, 1992 5000-92-3017 C321-92-2098 U. S. Nuclear Regulatory Commission Attn: Document Control Desk Washington, DC 20555 Gentlemen:
Subject:
Oyster Creek Nuclear Generating Station (0CNGS)
Docket No. 50-219 Facility Operating License No. DPR-16 Oyster Creek Fuel Channel Bowing
Reference:
GPU Nuclear Letter C321-91-2214, "0yster Creek Fuel Channel Bowing," August 9, 1991.
In the referenced letter, GPU Nuclear committed to submit the results of the revised bow model being developed for the fuel channel reuse program at Oyster Creek.
Attachment I to this letter provides a detailed description of the revised bow model, comparisons (based on the revised model) of measured versus predicted bows, revised model predictions of end-of-cycle (EOC) 13 bows, and resulting changes in Critical Power Ratio (CPR) penalty.
The revised model.provides an improved method of predicting large channel bows resulting from fixed and variable gradient fluences of opposite channel faces.
This model is now the basis for determining the channel average and core average channel bow which is used to determine the magnitude of the CPR channel bow penalty. The new R Factors calculated with the results from this model are somewhat larger than the original and will be used for the remainder of cycle 13.
The increased R factors resulted from high bow channt.'s residing in rodded control cells or other non-limiting core locations. The cecrease in the calculated CPR results from the overall correction to the non-limiting core locations when applied to the limiting core locations. Continued operation with the higher bows is acceptable since the higher bows are located in control cells or non-limiting regions of the cere.
The R factors have been adjusted to correct CPR for the increased peaking.
Thermal margin for axial power heat generation rate (APHGR) and local linear heat generation rate (LLHGR) in this region is sufficient to account for any additional 9204060267 920402 l
PDR ADOCK 05000219 P
PDR 32192098.let gO/
GPU Nuclear Corporabon 6 a su2Ga7 of General Peut UD t es Como< anon ffj
i C321 92 2098 Page 2 uncertainty due to higher bows, in addition the bows are away from the control rod and will not interfere with control rod movements.
Therefore, the new core average bow and larger reused channel bows will not affect the safe cycle 13 operation of Oyster Creek.
At this time, the impact of ',rge bowed reused channels to the Oyster Creek channel reuse program has h9t been determined.
A study to identify a channel management program which will minimize the development of large bows hu been initiated.
GPd Nticlear will inform the NRC of fuel channel reuse at b/ ster Creek for cycle 14 and beyoni prior to the next refueling, If you have any questions or comments on this submittal please contact Mr.
e 1
Michael Laggart, Manager, Corporate Nuclear Licensing at (201) 316 7968.
h Very[trufyyours, A-J. C. DeVine, Jr.
Vice President and Director Technical functions JCD/RZ/ pip cc: Administrator, Region 1 Senior Resident inspector Oyster Creek NRC Project Manager F
l l
l 37192098,let l
l-.
4 ATTACHMENT 1 a
- )
LMLILQF_COH1Iit1TS
1.0 Background
20 Description of the NEWBOW Malol 30 Flxed Gracient Bow Calcu!3 tion 40 itild40 Gradient Dow Calculation 50 Total Bow 0.0 Averago Bow 70 Results of the NEWB44 Model DO Ettect of L.argo Prodiciud Channel Bow on Cyclo 13 Operation 90 Conclusions 10 0 References 11.0 figures n:%
4 4
i 10 Badmund The endol cycle (EOC) 12 Oyster Creek fuel channel measuroments, for the fitst time, me.asured channels having exposuros of groator than 62 GWD/MTU, Some of thoso channots wero 1ound to have channel bows greator than 200 m!!s and as high as 400 mils. Betws of this magnitude woro unexpected and appearod to originato from reusod channel reskienco in the control cells Reference 1 reported the results of the EOC 12 channel measurements and the comparison of the data to the calculated channel bow.
The original channel bow predictions were based on an ompirical model This modol corrolated Oystor Croek channel measutomonts and included only the initial cycle of control cell operation. It did not include any high exposure roused channets. This model did not prodlet the prosonce of largo bows in the control coils. Sinco the original bow model could not be readily modified to account for this behavior, it was concludod that a now model rDvorced from the measutomont data would bo developod The now rnodol would be barod on the individual channel history of Zircatoy growth resulting from calculated neutron fluence gradients LO_Delat!oftoLthe NEWJQhadej The rovised Oyster Crook fuel channel bow rnodel prodictions, are bood on the calculated differential growth of each face of the channel. The channel fat'o growth is modelod from the Zircaloy growth curve pubilshed by ABB as part of the documonction of the Oskarshamn incident (Reference 2). The bow is calcu!ated using the formula published in Referonco 3, and uses the differential grow 1h of opposing ctannel facos. It also accounts for the a& built channot bow.
The NEWDOW model calculates growth of each channel face based on (1) local fixed flux gradients and (2) variablo flux gradsents. The growth due to those flux gradients results in channel bow about each e.ls.
A negative bow Indicates the boa :s the direction of the control rod.
P 10.f1LedRadlentfoyLCAkulation The imod gradient bow is that por1 ion of the total bow which results from the difforentkil fluence originating from the design of the 'O' fuel cell lattice. The fixed gradient bow calculation is basal on the calculated inctcased length of each channel face which results from the differential fluenco at ooch face. The increased longth is calculated using the growth to burnup relationship shown in Figure 1 (Reforonce 2). The exposuro term used in the correlation is the product of the total channel exposuru and fast fluenco factor for the applicablo faco of the channel The fast fluence factors wero derived from PDO neuttonic calculations which modelod the exp'icit fuel coil geometry. The correlation for the ler.gth of each channel face is:
Increased Length of each Channel Faco a / (E, FFF)
[
where: E is total channel exposuro and FFF is the applicablo last fluence factor The channel's bow about each axis is calcu'ated using the equation found in Reference 3.
l' BOW - ( L
- a L) / ( 8 * (W) )
where: L is the channot length, AL is tho (htfotonce in opposito channel face length, and
-W is the width of the channel The delta longths used in this equation are the differences between the growth of opposito channel faces longths as calculated from the abovo corrolation.
L_.__..__.._____.,____.____
_.. _. ~ _ _., _ _ _. -. -,.. _
a M_VAflaWe._QthdMDLQQWSBkulall A 9
Tea +.ariaNe Oradient bow is that portion of the total bow which results hom the differential fluento of the channel f acos due to the presenca of controlicds. locations at or near the core edge or due to radial power differences. The variatie gradient bow correlation uses the same channel face length equation as the find gradient bow correlation, howevnt, the channel faco exposuro term is the averago of the fuel burxile containing the subject channel and the ad}acent fuel bundles summed over the nurubar of cycles the etennel resUed incoro The increase length of the channel face cortolt tion is:
Increased Length of Each Channel Faco = f (E')
where. E' is total average channii face exposuro The variable gracient channel bow about each axis is calculated as abovo using the samo equation from Reference 3 M.ht!dQRg The total bow about each axis (1 e. X and Y bows) for the requirtd cycle is the sunt of the finod gradient, variatJo gradient bows ard the addition of the average as built traw cA 15 mils.
X or Y Totat Bow - Fiwd Gradient + Variable Gradient Bows 4 As Built Bow M_AyptAgg_DM The average bow is the average of the total X and Y bows This value is used ta determine the fuol channot bow penalty.
Averago Bow - (Total s 3cw + Total Y Bow )/ 2 ZAllew!!asf heXAWRQWWdel J
The goal of the NEWBOW modet is to piovido a better a0rooment between prediction and measurements for both the magnitude and direction of channel bow. The predictions of NEWBOW were compared to the Oycter Creek channel measutoments. Bows at higher exposures were of patticular interest because they had previously been under predictod, The measured versus predicteo bows are shown in Figure 2 The data includes channels with varied coro location histories including ed0e core ard control colt (rodded) operation This figuro also shows the one slgrr:a error band of 53.7 mits assocbted with the 107 data points This data includos all tho Oyster Creek channels which were measured more than once Figures 3 an t 4 are examples how the NEWBOW correlation can pmdict the change in direction of the channel bow as compared to the measured bow. Keep in mind that this data was not used to cortotate the mcdel.
The projected EOC 13 core averago channol bcu as calculated by NEWDOW is 58 mils. The orror assaciated with this prediction is, as stated above,63.7 mils The error associated with the previous model which includes the EOC 12 measurements is M 9 mils The new model is an improvement over the previous model due to the uso of the differenttal fluence to ca!culato edge core and control cell operating history
- ~. -.. _--
.. ~ -. -._.. - _
4 a
bows.
Figuro 6 is a coro map which shows the prodicted average channel bow at the EOC 13. The high bow channds resulted from channel historios which resided in low power and high differential flux coro location such as control cell and (dge coto locations.
g 0 The Effetlefhrrge.PadicRLCh0DeLuoroo_Cyck 110melisn The EOC 13 average channel bows wero evaluatod to determine the offoct on the channot bow CPR penalty.
Avorago channel bows of greater than 200 niils woro analyzed. It was determinnd that the sael in the colIs containing the largo average bows dd not becomo limiting following the additional CPR penalty. The fuel bundlos with the largest channot bows were located in rodded control colhi or in othol non lim!!ing coro locations. The now R.lactors resulting from the newty calculated bows will be usod in cycle 13 oporation.
The new CPR penalty will add another 2%, for a total of 6% additional CPR in the limiting core locations resulting from channel how.1ho CPR penalty for non-limiting coro locations is higher.
LQ_Csnt!n!ons The NEWBOW model prov6 dos an improved rnethod of predicting <.tgo channel bows resulting from fuod and varlable gradient fluences of oppos!!o channel facca. This modelis now the basis for dr irrnining the channel averago and coro averago channel bow which is used to determine the magriitudu of the CPR i
channel bow penalty. The new R Factors calculated with the results from this mm' are somewhat larger than the original and wl!I be used for tho tomaindor of cycle 13. The incrcased R fcctors resulted from blah bow channels residing in rodded control cells or other non limiting core location:, The decreaso in tho -
calculated CPR results from the overall correction to the non limiting core locations when applied to the limiting core locations. Continued operation with the higher bows is acceptable sinco tha higher bows are iccated in control cells or non-limiting regions of the coro, The R factors havo been adjustrxf to correct CPR for the increased peaking Thermal margin for APHGR and LLHGR in this region is suificient to account for any additional encertainty due to higher bows. In addltion the bows are away fro.n the controi rod and will not interfore with control rod movements, in conclusion, the now coro average bow and larger toused channel bows will not affect the safo cycle 13 operation of Oyster Crook.
l
. -.,, -.,,. ~
~, -
. _,. -.-. -.,,.,,.~
\\
i l
10 M tLetcoces
- 1. GPU Nuclear Letter C321912214 (to NRC) ' Oyster Creek Fuel Clunnel Bowlng". August 'J.1101, 1
- 2. OKG AB and ABD ATOM AB Poster Puuentation. Subject *In-Reactor Mechanical Beiavior of BWR Fuel Clunaels*, Ako Johnson. Lars Hallstadius. Ulf Sundstrom,1990
110_Elmrtes Figure 1 - Zircaloy Growth Curve Figure 2 - Measured vs Predicted Bows Measured and Predicted Bow Ctannel 621C Figuro 3 Figure 4 - Measurta and Prodicted Bow Channel 60039 Figure 5 Predicted EOC 13 Average Channel Bows i
0 1k 8
0 7
0 E
6 D
V E
R T
A U
L U
U C
0T C
' 6 M L
H )2
/
A T
D C
e W
Wc 1
G
+
e On S
0 r
Re E
4 E u
r R
g Gfe H
i U
F C
e N
S N
YR O
O I
O(
0P T
I L
H 3X A
A T
E U
C W
Q E
R O
R l
0 Z
G
=
2 E
C A
F L
0 E
1 N
N A
H C
O 6
6 4
2 O
0 0
0 0
e O
O le
\\"
8, N
O\\
5 Q.$
WQ aW o
Ok
\\
R5 a
a$
2 0 o
W -
'O
\\
a-
$W 2
- g om
"\\
5s o
W1 N
N Ok 2
J gj
\\
Oo
~
W>
9 Z
o \\
a.0 m
ZQ H
S 9
<W Im o%
o O @D O
d fg Og d
Z<
2 a
\\
t 2
DW ht
[
I B
Q-2 5
(k1 0
b 2
0
'E i
W O
at t
a w
0 b D "
'.O
$j T.
J D
S W J g
\\a 2
cL.
O O
E 1
.1 1
J.---
J-_
o O
O O
O O
O O
07 O
O O
O O
O O
O T
M N
N I
t
i t
o l
GPUN CHANNEL BOW MODEL I
i i
i MEASURED AND PREDICTED BOW
~
i CHANNEL 521C i
i l
4 i
CHANNEL BOW - MILS 400
+
l i
t
/7 300 3
l
/
i t
/
t
/
l 200
-/ 7---
l j
100 ~-
- - ~ ~ _ -._ c w -
-- - - - - - -- - - - - - -'1 l
P i
l f-i O'
l l
10 20 30 40 60 60 i
t I
i CHANNEL EXPOSURE - GWD/MTU l
i x
MEASURED XBOW
- - CAL XBOW l
t i
i i'
l Figure 3 i
i e
,k
.i
~
GPUN CHANNEL BOW MODEL 3
i MEASURED -AND PREDICTED BOW l
CHANNEL 50039 E
i CHANNEL BOW - MILS i
250 j.
I l
A----------I L
200'---
e 1
s l
/
f
?
b I
....... ~.......................... A.............Q........
3gg!
i
\\
\\
L i
............y...........,
3 og u.
.....,.e...
L
\\"l l
s p.
4M-
...........................................k......
N
['
0 I
20 30 40 50 60 i
CHANNEL EXPOSURE - GWD/MTU i
i i
I I
+ MEASURED YBOW
- - CAL YBOW i
i 1
2
?
Figure 4 1
i D
- . z PREDICTED EOC 13 AVERAGE CHAfii4El BOWS (MILO
-3s 16
-38 13 31 31 13
. 40 16
-1s '
2 107 44 too 1ss 97 sa sa es-too too 44 ss s7 l si 3
32 s7 do 2s ' O 2s 2s 24 2s 2s 24 2s 27
-0 2s 4
t 4
186 21 ssi s2 's os 17 21 l 143 'so 17' is 4s 1s3 21 17 ss ss s7 l 21 1ss 1
5
-11 2s 172 ss 2
ss 47 72 12 17 7s 7s is 2s 72 47 as
-2 iss 11s 2s
-12 6
161 s7 Ss ss 20 62 ss 53 27 s
81 61 s
2a s3 ' as s2 se !es ss s7 155 i
7
-s 77
-3 '
is 4s 77: 24 131 ss 3s 17 17 3s ss 3s ! 24 77 da is'
-3 es
-s j.
8 ss 27 Ss 36I s3 77 40 1s3 23 S4 ssss 1ss ss 34 23
-s 40 77 73 '3s ' es 27 ss 9
-1.
43 O f is 47 : so Is 32 4s '2s7 20 as ss ' ss ss zo sr 4s 32 2s so 47 is 0 43
-16 10 37 too 21 20 100 ' s4 ss :24 e4
-1 1ss 2s 1s7 112 2s so
-1 e4 24 as 54 too 20 es too 37 11
-42 107 26 14s 12 27 ss 187 20 43 17 46 2s 2s as 17 32 20 171 se 2s 12 144 27 107 44 a
12 a
ss 2s so 17 s
32 17s 4s 2s s2. 112 ts2 1s7 ss as 2s 4s s4 32 s
is so 24 s2 44 13
-14 71 24 18 ass 237 17 1st es 34s 27 324 :s3 s2 2s1 2s asal ss as 17 242 20s is 24 71
-14 14
-14 71 24 1s 17s 24s 17 as ss ssa 2s 2s7 53-53 324 27 s2 ' ss 'da 17 721 32211s 24 71
-14 4
15 44 63 2s so 17' s
32 s4-124 23 4s ss 1s4 1s3 61 4s 2s as 100 32 s
1s-.so 24 ss 34 i
1S 44 1oe 27 1st 12 2s ss 245 20 32 17 as 2s i2s its 17 32 20 53 ss 2s 12 iss; 27 107
-3s 17 37 too 21 20 100 54 as 24 s4
-1 174 2s 333 sss 2s 1st
-1 e4 24 ss 54 too 20 21 100 sa 4
18
-1s 43 0
1s 1 57: so 2s 32 4s s7 o-s3 ss ss as 20 s7 4r 32 2s so 47 1s 0 7s' -15 l
p s8 27 ss-3s as' 77 ao las 23 1ss ss 101 ' as 133 34 23 17s 4o 77 l so 3s se
~27 ss:
j p
-s 72
-3 '
1s: da 7s 24 3si ss 3s is 1s 3s ss 3s 24 77 es 1s
-3 as
-s M
77 s3 ss ss 21 se es s3 23 s
at 52 s
27 s3 6%
s3 20 ss ss s7 7'
22
-12 2s 172 ss
-2 3s 47 72 12 1s 7s 75 s7 51 72 47 s2
-2 ss,13s 2s
-12 s7!22 23
-- - -l 1s7 21 ss 72 ss 17' 21 iss 4o 17 17 so 34 21 17 ss 71 1s7 s7 l 2 24 31-sa l 4 2s 0
2s 2s 24
- s 2s as 2s 2s O
s2 4
25 es 44 101 12e sa 20s 1os ss 12s tot 44 ss
'26
-se is 40 12 31:
31 13
-1s is
-ss 1l213!415i6l7l8 9
10 11 12 13-14 15 16 17 18-19120121122l 23 l 24l 25126]
4 i
FIGURE 5 1i l
9 i
e
-