ML20058L809
| ML20058L809 | |
| Person / Time | |
|---|---|
| Site: | Fort Saint Vrain  |
| Issue date: | 11/30/1985 |
| From: | GENERAL ATOMICS (FORMERLY GA TECHNOLOGIES, INC./GENER |
| To: | |
| Shared Package | |
| ML20058L807 | List: |
| References | |
| GA-C18103, NUDOCS 9008080174 | |
| Download: ML20058L809 (105) | |
Text
0.h- $$ S V)
N -
9 k, yA
'- J oo g
'O
=
@hm- _ _ _ _
o _ _
=
G A-C18103 FORT ST. VRAIN CYCLE 3 CORE PDtFOR4ANCE Work Done By: Report Written By:
FSV Engineering Staff R. Hackney l
i l
GA TECHNOLOGIES PROJECT 1900 November 1985 ;
i l
9008080174 900803-PDR ADOCK 05000267 P PDC
-._._______j
J l
. ABSTRACT The Fort St. Vrain (FSV) Nuclear Generating Station, owned and operated by Public Service Ccapany (PSC) of Colorado, is a plant employing 1 the high-temperature gas-cooled reactor (HTGR) concept. Initial criti- .]
cality was achieved in January 1974 and cycle 1 operation continued through February 1979. Throughout cycle 1 operation the maximum power was limited. l by license restrictions, to $89 MW(t) or 705 of full power. Cycle 1 core performance was, in general, close to predictions. One exception to this good agreement with expected performance was the region exit gas tempera- l ture fluctuations which were detected early in the cycle. At the end of cycle 1 operation, the cause of the fluctuations was not positively identified but the most fikely explanation was small movements of reactor components such as fuel elements, reflector elements, and/or core support floor, l The first refueling was completed in April 1979, and initial
. criticality of the cycle 2 core occurred in May 1979. The core was then operatec at power levels up to 705 of rated power until Octooer 1979 when -
i the plant was shut down for installation of region constraint devices
( RC Ds ) . These RCDs were installed to stabilize gap flow areas at the top of the core to near nominal values so as to minimize block movement and, therefore, eliminate the temperature fluctuations. Cycle 2 operation i continued at power levels up to 70% of rated power with core performance as predicted and no fluctuations were detected. In March 1981, the Nuclear Regulatory Commission (NRC) issued a release to test above 705 power.
These tests demonstrated successful operation up to 905 of rated power
.before the testing was terminated for plant maintenance and' for the se tond refuelir.g.
t$
11 l
v
i Initial criticality of the cycle 3 core occurred in July 1981.
Testing in the fall of 1981 demonstrated successful operation, without fluctuations, at power levels up to 100% of rated power and at pressure drops up to 5.0 paid. Results of these tests led to the NRC release, in October 1982, for 1005 rated power operation. Cycle 3 operation continued until January 1984 when the plant was shut down for the third refueling.
Throughout the cycle the core performance was, in general, as predicted. I 1
This report discusssa the core performance during cycle 3 operation.
I 1
1 1
l l
4 8
111
I 4
l 1
APT Axial power factor BISO Two-coating fuel particle BOC Beginning-of-cycle BOL Beginning-of-life BWR Boiling water reactor 1
CS8 Core support block DL Data logger .
DOE Department of Energy EFPD Effective full pcwor day '
EOC End-of-cycle FDDM . Fuel Design Data Manual FIMA Fissions per initial heavy metal atos FSV Fort St. Vrain FTE Puel test element GA GA Technologies High enriched uranium (935 U-235)
HEU HSF Hot service facility HTGR High-temperature gas-cooled reactor ICRD Instrumented control. rod drive LBP Lumped burnable poison LCO Limiting condition for operation LWR Light-water reactor MEU Medium enriched uranium HOC Middle-of-cycle mwd. Megawatt day NRC Nuclear Regulatory Commission NW Northwest PCRV Prestressed concrete reactor vessel PIE Post irradiation examination iY
i PPS Plant protective system PSC Public Service Company of Colorado PWR Pressurized water reactor QC Quality Control R/B Release / birth rate ratio RCD Region constraint device RPF Region (power) peaking factor ,
RSS Reserve shutdown system '
RT Request for test RWP Rod withdrawal prohibit SAR Safety Analysis Report ,
S/N Serial number SOP System Operating Procedure Surveillance requirement SR TRISO Four-coating fuel particle <
WAR Weak acid resin '
t Y
E CONTENTS
- 1. .UMMARY , . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 -1
- 2. CORE DESCRIPTION ....................... 2-1 3 POWER AND ENERGY HISTORY ................... 3-1
- 4. NU C LE AR P ER FO RM ANC E . . . . . . . . . . . . . . . . . . . . . . 4-1 4.1. Initial Criticality .................. 4-1 4.2. Control Rod Group Reactivity Worth . . . . . . . . . . . 4-1 4.3 Temperature Detect . . ... . . . . . . . . . . . . . . 4-4 4.4 Nuclear Detector Decalibration . . . . . . . . . . . . . 4-4 4.5. Fuel Accountability .................. 4-7 4.5.1. Heavy Metal Inventory . . . . . . . . . . . . . 4-9 4.5.2. Region Peaking Factors (Two Versus Three Dimensional Calculation) . . . . . . . . . . . . 4-9 4.5 3 Fractional Absorptions and Excess Reactivity . . 4-l '.
,4.5.4 Particle Burnup ............. . . . 4-11 4.5.5. Axial Power Data . . . . . . . . . . . . . . . . 4-13 4.5.6. Power Correction Factors for Fuel Test Elements. 4-16 4.5.7. Burnup Data for Cycles 1, 2. and 3 . . . . . . . 4-16 4.6. R e a c t i v i ty Di sc r ep an c y . . . . . . . . . . . . . . . . . 4-16 4.7. Control Rod Management . . . . . . . . . . . . . . .-. . 4-26
- 5. THERMAL / FLOW PERFORMANCE ................... 5-1 5.1. Region Peaking Factor ... . . . . . . . . . . . . . . 5-1
- 6. FUEL PERFORMANCE ....................... 6-1 6.1. ' Calculational Methods .......... . . . . . . . 6 6.2. Results ........................ 6-3 6.2.1. Fuel and Graphite Temperature . . . . . . . . . 6-3 6.2.2. Fuel Particle Failure . . . . . . . . . . . . . 6-6 6.2 3. Gaseous Fission Product Release . . . . . . . . 6-10 6.2.4 Metallic Fission Product Release . . . . . . . . 6-13 6.3. Conclusio ns ...................... 6-17 Vi
- 7. SPECIAL TESTS AND SURVEILLANCES . . . . . . . . . . . . . . . . 7-1 7.1. Chemistry Surve111ances ................ 7-1 7.1.1. Coolant Impuritie s . . . . . . . . . . . . . . . 7-1 7.1.2. Noble Gas Release / Birth (R/B) . . . . . . . . . 7-1 7.1 3. Tritium concentrations . . . . . . . . . . . . . 7-2 7.1.4 Iodine Monitor . . . . . . . . . . . . . . . . . 't - 2 7.1.5. Plateout on Circulator . . . . . . . . . . . . . 7-3 7.1.6. Gauma Activity of PGX Surveillance Samples . . . 7-3 7.1.7. Gamma Scanning of Circulator . . . . . . . . . . 7-3 7.2. Plateout Probe ......... . . . . . . . . . . 7-4 7.3. Core Support 'slock oxidation . . . . . . . . . . . . . . 7-5 7 . '4 PGX Graphite Ourveillance . . . . . . . . . . . . . . . 7-6 7.5. Fuel Test Elements . . . . . . . . . . . . . . . . . . . 7-7 7.6. Post Irradiation Examination of Fuel and Reflector E l e me n t s . . . . . . . . . . . . . . . . . . . . . . . . 7-8 7.6.1. Se g me n t 2 El e me n t% . . . . . . . . . . . . . . . 7-8 7.6.2. Se gme n t 3 El e me n ts . . . . . . . . . . . . . . . 7-10 7.7. Gamma Scanning of Fuel Elements . . . . . . . . . . . . 7-12 7.8. Scratcher Plenus Elements . . . . . . . . . . . . . . . 7-12
- 79. Temperature Fluctuations . . . . . . . . . . . . . . . . 7-13 7.9.1. Background . . ................. 7-13 7.9.2. Cycle 3 Testing ................ 7-16 7.10. Ra dia t io n E xposure . . . . . . . . . . . . . . . . . . . 7-17
- 8. UNUSUAL OCCURRENCES . . .................... 8-1 8.1. Inadvertent Insertion of Reserve Shutdown Material into Region 27 . . . . . . . . . . . . . ... . . . . . . 8-1 8.2. Cracked Fuel Element Web . . ... . . . . . . . . . . . . B-3 8.3. Powe" Level Uncertainty .. . . . . . . . . . . . . . 8-3
- 9. REFERENCES .......................... 9-1 e
Yii
FIGURES 2-1. Cycle 3 core layout . . . ................. 2-2 3-1. Cycle 3 thermal power history ............... 3-2 3-2. Cycle 3 thermal energy history . . . . . .......... 3-3 3-3. Cycle 3 thermal and electrical energy history .. ..... 3-4 3-4 Cumulative thermal energy from BOL . . . . . . . . . . . . . 3-5 3-5. Cumulative thermal and electrical energy from BOL .... . 3-6 4-1. Cycle 3 temperature defect . ................ 4-5 4-2. Axial power distribution - FTE-2 . . . . . . . . . . . . . . 4-17 4-3. Axial power distribution - FTE-3 . . . . . . . . . . . . . . 4-18 4-4 Axial power distribution - FTE-4 . . . . . . . . . . . ... 4-19 4-5. Axial powe.' dis tribution - FTE-5 . . . . . . . . . . . . . . 4-20 4-6. Calculated minus measured K,77 . . . . . . . . . . . . . . . 4-25 '
5-1. Comparison of measured and calculated RPF cistribution . . . 5-4.
6-1. Peak fuel C/L temperature distribution . . . . . . . . . . . 6-4 6-2. Time averaged fuel C/1. temperature distribution ...... 6-5 6-3 Fast neutron fluence distribution ............. 6-7 1 6-4 Failed particle percentage - fissile particle ....... 6-8 6-5. Fa11ec particle percentage - fertile particle ....... 6-9 ,
4 6-6. Comparison of measured and predicted Kr-85m release .... 6-11 6-7. Comparison of measured and predicted Kr-85m release .... 6-12 TABLES 2-1. Cycle 3 control rod withdrawal sequence .......... 2-6 4-1. Comparison of measured and calculated control rod group reactivity Worth . . . . . . . ............... 4-3 4-2. Comparison of measured and calculated detector decalibration for sequential control rod group withdrawal . . . . . . .-. 4-6 viii
- ( ,
4-3. Total gore heavy metal' inventory at EOC3 . . . . . . . . . . 4-10 -l l
4-4 Maximum fissile particle FIMA .... ... ........ 4-12 4-5. A x ia l po ver da t a a t . E0 0 3 . . . . . . . . . . . . . . . . . . 4-14 l
4-6. Heavy metal core loadings' for cycles 1, 2, and 3 . . . . . . 4- 21 !
4-7. Fractional fissions for cycles 1, 2, and 3 . . . . . . . . . 4-22 .:
1 4-8. E5egment buraup for cycles 1, 2, and 3 ........... 4-23 6
4-9. Fast neutron exposure of control rod clad ....... .. 4-27 .
4-10. Thermal neutran exposure of control rod clad . ..... .. 4- 29 4-11. Control rod-bu.*nup . . . . . . . . . . . . . . . . . . . . . 4- 31 6-1. Sr-90 release u . . . . . . . ...............
6-14 1 t
6-2. Cs-134 release . . . . . . . .... . .. ......... 6-15 ;
6-3; Cs-137 rele ase . . . . . . . . . . . . . . . . . . . . . . . 6-16 7-1. Comparison of radittion history'- FSV vs LWRs
..... .. 7-18
- l l
i f
ix i
4
_________'i_.________________
?
i i
5 tI, p
- 1.
SUMMARY
Initial criticCcy of tne cycle 3 core occurred in July 1981. Prior to January 198h. 4 tan no plant was shut C*n for tho' third refuelinga th e -
cycle 3~ core nad operated for a total of 295" effective rull power crys
-f (EFPDs).- The cumulative burnup for' fuel segments 3 through 6, at thi end' q
'of cycle 3 was 658 EFPDa#, 3 During the initial' cycle 3. rise to power, tests were conductcd by PSC !
to verify that the core performance was as predicted. In addition, ' core performance data have been monitored throughout the cycle. Although nach of these data have been previously reported-(Refs.1 through -8 ), this report summarizes the overall cycle 3 core performance.
.l The nuclear performance was, in general, as predicted. Reasonable 4 agreement between measurec and calculated data was obtained for the .
temperature coefficient and for control rod worths. Initial criticality '
and reac'tivity behavior = with burnup was also predicted with reasonable ,
accuracy. .!
c
. The thermal / flow performance was basically unchanged from previous )
. cycles. Significant region peaking factor (RPF) discrepancies continued to j exist in the northwest boundary regions, the result of region exit tempera- '!
ture measurement; errors due-to cool gas flowing inside the thermocouple
-sleeve. In order to compensate for these measurement errors, Technical Specification Limiting Condition for Operation (LCO) 4.1.7 was revised to-include- special operating procedures. '!
'Throughout this . report the burnup (in EFPDs) is given to the nearest whole number.
1-1 ]
j The fuel performance for previous cycles grossly overpredicted fuel I failure. Therefore, the models were refined and all three cycles were ' -i re-evaluated resulting' in generally good agreement between measured and
~
calculated fuel performance data. .
1 I
Further' fluctuation testing demonstrated that the region constraint devices (RCDs) were successful in preventing temperature fluctuations at power levels up to 1005 of rated power and at core pressure dropa up to 5.0 pai. As a result of these tests, confirming resolution of the 1
temperature fluctuations observed during cycles 1 and 2,. the NRC approved j unrestricted full. power operation. l The overall oore performance of cycle 3 was, in general, as expected.-
l
.l 1
i i
s I
l!
1-2
l'
- l. <
l 1-l t
- 2. CORE DESCRIPTION The active core consists of six layers of fuel elements with 247 fuel elements per layer. .The fuel elementa are grouped into 37 separate refueling regions surrounded by a graphite reflector as shown in Fig. 2-1.
Each of the 37 refueling regions consists of either five or seven fuel columns and contains an orifice flow control assembly and a pair of control rods.
The fresh fuel materials are high enriched uranium (HEU)* and fertile-thorium in carbide form. The uranium and thorium carbide particles are coated with layers of pyrocarbon and silicon carbide and bonded into fuel rods within the hexagonal graphite elements. The particle coatings provide the prime barrier for fission product retention. The core is designed to produce 842 MW(t) at a power density of. 6 3 kW(t)/ liter.
The thorium / uranium fuel cycle specifies that approximately one-sixth of the core be replaced with fresh fuel at each refueling cycle. This means that six refueling regions are reloaded at each refueling except for the. fifth reload, at which time the central refueling region is also replaced. Each reload segment consists of 204 standard fuel elements and 36' control rod fuel elements, except as noted for the- fifth reload. In i
- Isotopic content of uranium:
U-234 - 0.0073 s-235 - 0.9315 U-236 - 0.0028 i U-238 - 0.0584 1.0000 2-1 1
, . . . . . . -.-. - . . - - . . - - - . . . . . - - . _ . ~ . . ~ . . - .. -
1
.j g ,
.l l
i 4
I tot MTICToa
,Q- i r ...
g s-iism I
'l
(.
- .'.., i toe ometoe -
p :... .
, E'w*mItwones IoYa'tYa'"nes J:1.t .
g-i k == -i Fig. 2-1 ' FSV core layout and locations of ex-core detectors I
2-2
each reload segment five of the refueling regions contain seven fuel columns and one contains five columns. The refueling region sequence was L chosen so that freshly refueled regions are never adjacent to each other, l'
except when the central region is refueled.
The reactor is controlled by a pair of cylindrical control rods in ,
each of the 37 refueling regions, each with an independent control rod I drive.- A reserve shutdown system, in the form of B.C graphite balls, is ,
provided for backup shutdown _ capability.
The core power. level is measured by six wide-range neutron flux detectors located symmetrically around the core,'in the prestressed concrete reactor vessel (PCRV), at about the core midplane.
' The core contains 84 region constraint devices (RCDs), installed in-the core 'in late 1979 as a means of eliminating the temperature fluctua-tions. These RCDs mechanically interlock fuel regions at the top plenum . I layer of the core to prevent development of nonuniform clearance gaps between the internally keyed refueling regions. Details of the RCD design are given in Ref 9.
l l
Instrumented control rod drives (ICRDs) are located in refueling. '
regions 5-(S/N 20) and 35 (S/N 43) to provide in-core data as an aid in understanding the fluctuation phenomenon. - These. ICRDs are standard control rod assemblies in which one control rod of the pair was removed and replaced with an instrument package. (These ICRDs were removed from the core 'at the end of cycle 3 and replaced with a standard control rod pair.)
Scratcher plenum elements are located above refueling regions 18 and
- 35 to permit measurement of relative motion between adjacent refueling 3 reg io ns. [These scratcher elements were removed at the end of cycle 3 (see ,
Section 7.8).]
2-3
1 l
l 1
During.the loading of segment 7 (reload 1), .elght fuel test elements (FTEs), FTE-1 through FTE-8, were loaded into layer six of regions 25, 22, ;
' 30 27, 24 10, 5, and 5, respectively. '(Details of the FTEs are given in l
-Ref. 10.)' Thesc FTEs are intended to demonstrate acceptable performance
. and safety of proposed future FSV fuel and processes prior to full-scale application (see Section 7.5).
At. the end of cycle 2 operation, segment 2 fuel was replaced with ;
segment 8 fuel in refueling regions 4, 8, 15, 25, 32, and ' 36, 1. e. , the I cycle 3 core contained fuel segments 3 through 8. . The reloaded regions are shown as the shaded areas in Fig. 2-1. A detailed-description of the segment 8 fuel design 'is given in Ref.11.
Other differences between the cycle 3 and cycle 2 cores are discussed below:-
o Segment 8 fuel fabrication differed from previous segments in that (1) leftover segment 7 fuel rods were used in some fuel a i elements, (2) several fuel rod lots contained high microporosity particles loaded in the bottom layer of -the core, and (3) several elements with fuel rods containing uranium loadings higher than L specified were loaded into special core locations (Ref.12).
o Fuel elements with a " thin" buffer were replaced with a " thick" ,
buffer, i.e. , the entire fuel element (Ref.11).
o A new concept for the lumped burnable poison (LBP) design- was.
used in segment 8. This new concept used a single LBP type for all reloads. The LSP rods used in this new approach are shorter in length (-2 in. ) than the previous designs which used 28-in.
[ long LBP rods. The shorter length LBP rods are stacked along with " dummy" graphite spacer rods in a specified order into each
( LDP hole. (This design was modified slightly for segment 8 so 1
L L
i.
'i' V , ,
~!
that,. leftover segment 7 LBP rods cou,d be cut into approximately -
2-in. lengths and used.)- A' description of the segment 8 fuel design and predicted performance is given in Refs.11 and 13 f
'i o The control rod withdrawal sequence' was changed to maintain the 1 maximum control rod worth and the region peaking' factor require-ments within LCO 14.1 3 limits. The' withdrawal sequence is given in Table 2-1.
1 s
l l
l
.l l
1 I
l i
t e
i 1
l .; -
l s.
l' !
L{
g ,
2-5
.n .;
f 1
y TABLE 2-1 CONTROL ROD WITHDRAWAL SEQUENCE J
f Control Rod- ,
Group Re gions .
+
28 3,5 ,7 -
4A 20,26.32 4C 22,28,34 t;
- 1. (0 + 115 in. ) I t 4E 24,30,36 4F 25,31,37 2A 2,4,6 4D 23,29.35.. [
3A 8,12,16-P 3C 10.14,18 4B 21,27.33 1 30' 11,15,19 3B- 9,13,17 1 (115 + 190 in. ) 1
-i
-i 1
i e
2-6 i
1
. -- --. ~ . . - . - ..
. ;- i 3 POWER AND ENERGY HISTORY Cycle 3 operation began in July 1981. Prior.to January 1984' when the plant was shut down for the third refueling, the cycle 3 core had operated for a total of 295 effective full power days (EFPDs). The cumulative burnup for fuel segments 3 through 6, at the end of cycle 3 was 658 EFPDs.
- The thermal energy generation for cycle 3 was 5.95 x 10' MWh(t) for a total-of 13 3 x 10' Mvh(t) since the beginning of life (Bot).. The gross - electri-cal energy generated during cycle 3 was 1.96 x 10* MWh(e) for. a tota'l of 4.29 x 10' MWh(e) from BOL. During cycle 3,100% of rated power was
-achie ved for the first tism. The thermal power history throughout cycle?3 operation -is ,shown in Fig. 3-1.
- The thermal energy and EFPDs as a function of time in cycle 311s shown it. .?ig. 3-2, and Fig. 3-3 shows the thermal and electrical energy throughout cycle 3. The thermal energy and EFPDs, as a function of time since the BOL, is shown in Fig. 3-4 ' Figure 3-5 shows the thermal- and electrical. energy history from BOL.
l l
l l
l 1
3-1 l l
l 1
i t
FORT ST. VRAIX OPERATION HISTORY 100 100 ;
UNOFFICIAL THERMAL POWER
_ _ ELECT,R_1 CAL, POy_ER_ _
~E q 80 - - 80 1
b 70- -- -
7 F
{[ A
< i
- k. -70 Z-m J m N 60 - -60 N o ,-
o m -
w.
i T
~
E- 50- -
E-+
Z '
Z.
N m -
U 40- -40 0
- N -
' N i EL ;- O, 30 - , ;: -30 i 20- :: : -20 1
10 -
f f l ._10
~-
0 ;,,,,,,ii,,,,,,,,,,,,,,,,,,,-,,,,,,i ,
.ba , ;
r , i ,
.0 12212 2 2314 4 2516 6 271710 312112 2 2314 4 2515 6 27 f7 8 2919 9 2 2313 4 2515 6 27 f7 7 2819 9 302111 J A S O N D J F-M A M'J J.A S O N D J P M'A M J'J'A S.O N D 1981- 1982 1983 j Fig. 3-1' !
.,..+
m .. y w
FSV CYCLE 3 THERMAL ENERGY HISTORY a
s s 300 O
t 5- - - - - - -
-+- --- -- - - --
-250
. 4 t
2 I c, k 0-E 4 . .. . . .....;.... .............+.. . . . . . .. . . . .
-200 c a .-
3 .
W 2 -
3
" o O :
w c a. ,
g3 ..+.. . .
..+.. ...........+.............. .
..+. -
-'50 d l5 2 a 5 w i a
4: F f 2
a:
2- -
--+--
. ---i-----------.
-mQ .
u W
- lA T 1L F : .
W >
l 1
g_ _ ..;.. .. . . . . .. . .. ; . . . . . . . . . . ..;. .
-50 a
- 5 i I t
I
- I o , , O l
1981 1982 1983 1984 <
WE(YR) !
Fig. 3-2 :
, , . + a .y m
, m , . . w
- , .7' w i - .
U L
FSV CYCLE 3 ENERGY-HISTORY l
6 . .
9 5- -- - - - - - - - --- - - - - - - - - - - - - - - - - --
THERMAL l
t 4_ .. ... . . . . . . . . . . . . . . . . .
u
.C .
N Y=
2 a
o 3_ .
t c .
[
I >- '
O 5
5 2- - -- -- - --- - . - - - -
- ELECTRICAL =/i i / -
t
/ :
g_ .. . . . .. . . . . . . . .. .. s' .
I :
f .
Y M ~
.\
0 ,_
f l
1981- 882 1983 1984 -i l
3 Fig. 3-3
. _ _ _ - __ _____-_z__._______ _ ___ __________________1____._____=__ - _ _ _ _ _ . -
- . . . _ _ . . , _ . .:,__.___ -- ._ _ __...~. __
.i FSV CUMULATIVE THERMAL ENERGY HISTORY 14 . 700 :
12 - - - - - - - - - - - - - - - - - -
.----4-- - - -
w I
? -
o vs
-m g p to- - - - - -
i---- - - - - - - - - - - -
e--
.c -
3' o .:
2 m w -
o 8- - - - - - - -
+-- i--------+--- --i - - + .-<oo o 3- .
y C l n_ a w >- a O s E
y 6- - -
-300 h-D !
l taJ . . : . W '
-J :
p-4 I
- ~
o w
e 4_ ; . . . - .. - .. . :
La
~
3 .
-200 jr w y a
2 e CYCLE 1 ?' CYCLE 2 c CYCLE 3 = -soo I
. . .$ '(
.[
O . . . . . . . . o ;
1976 .1977 1978 1979 1980 1981 1982 1983 1984 l MN
'I i
~ Fig. 3-4
. _ _ - . , _ . . . . - ._ -s..~ . . . . _ _. .- - _ . . .-- 1
4 FSV CUMULATIVE ENERGY HISTORY 14 gg _ . . . .. .i . . . . . . . . . . . . . . '......-.....
i
.i L 10 - * --
--E '
t
.c w = :: CYCLE 2 CYCLE 3 =-
1 a l CYCLE 1 { : :-
B 40 ?
e O- -
t w -
}
3 .. .. . . . . . . . . ; . . . . . . . . . . ; . . . . -
THERMAL- -
i o .
x w
i i
b .
i .. . .
4_ . i .. ..
..- ELECTRICAL -- i yT i
i 4- :
. .s
!. . :. : ......i...... .! !
2_ . .. . . . . . . . . . .
3 j ,, !.. A.s, i-i : i !
. ,,7_ .
. # p i.
p . . . 1 0 , , , .. , , .
1976 1977 1978 1979 1980 1981 1982 1983 1984 !
E .
.. 1 f
i Fig. 3-5 l
,+-
.u _, ,_ _ _ _ _ _ _ _
4R ,
-kl h
I 1 i ,
4 NUCLEAR PERFORMANCE The following sections discuss the nuclear performance of the cycle 3 n core. j 1
1 4.1. INITI AL CRITIC ALITY '
Initial criticality at the beginning of cycle 3 (SOC 3) was cchieved on - q July 15, 1981. .ie critical rod position was 99 in. on control rod group - j 4F. The aver y> temperature was estimated to be about 230*F.'
Initial " cold" criticality wts calculated with the two-dironsiont.1 'I OAUGE model (Ref. 14). ' The effective multiplication factor calculated with the 4-group version was 0.9992, and with the 7-group version was 1.0060.- A i review of GAUGE model~ calculational reactivity bias' for the 7-group '
version at the end of cycle- 2 indicated that the bias was 0.0065 0.0015 Ak (Ref.15). The calculational reactivity bias at the beginning of cycle 3 was essentially the sa'me as during cycle 2 operation, indicating that the core behavior wasias expected. ,
4.2. CONTROL ROD GROUP REACTIVITY WORTH Surveillance specification SR 5.1.5 requires measurement of the reactivity worth of control rod groups as they are withdrawn from low power ,
to the operating conditions at the beginning of each refueling cycle. The method used for measurement of control rod reactivity worth during cycle 3 '
' Reactivity bias is defined as the difference between the calculated k,ff and unity (the measured k,77).
4-1
- - . .- _ - . _ . -~ . ~ - - - . _ _ - - -
%+ \W
- Yi '
.fi ,
was the same as that used during the initial rise to power for cycles 1 and 2. The calibration technique, partially developed during the operation of the Peach Bottom HTGR, is referred to as the double-bump technique. An li extension of.the method was developed during FSV cycle 1 ' operation and is
, referred to as the substitution technique.
- 4) These calibration techniques have made.use of an- analog reactivity computer with input from the average of three of the plant nuclear 3g ' detectors located outside the core at the midplane in the PCRV. The m; , double-bump technique makes use of data obtained from the reactivity y
,g computer (and rod position indication) as any control rod'is steadily withdrawn a distance of approximately 5 in. (or a reactivity worth of l between five and ten cents)# and then immediately reinserted the same i distance. The' opposite motion was also used. The double-bump technique essentially eliminates the effect of temperature feedback on;the measure-ment and can thus be performed during operation. These measurements )
. previde differential rod (or group) reactivity worths ( Ap/ inch) at' various
.... l g control rod positions. A least squares polynomial fit was then made to the
,+u measured data to obtain the differential worth as a function of withdrawal dis t ance.
Reactivity worth measurements, for control rod groups 4F, 2A, 4D, and 3A were made by PSC during the initial cycle 3 rise-to power,: 1. e. , from 25 op to 655 power. calculated control rod group worths were obtained~ using the g1 m GAUGE code where control rod groups were withdrawn, in sequence, at the beginning of cycle 3. The calculated data are. reported in .Ref. - 16.
Table 4-1 ~ ahows a comparison of the measured and calculated integral
,m reactivity worth for the four control rod groups measured. It is seen that the agreement between the measured and calculated integral worth is within
- 1
- 0ne cent (1 )) is equivalent to a Ap of -5.5 x 10-8 at B0c3 and
-4.8 x 10 5 at EOC3 4-2 g
_O_L__o____________.__ _ - - - -
l TABLE 4-1 .
COMPARISON OF MEASURED AND C ALCULATED CONTROL R0D GHOUP REACTIVITY WORTH Integrei Worth ( Ap )
Control Rod Q*oup Me asured Calc ulated 5 Difference #
4F 0.0220 0.0237 7.2 2A 0.0187 0.0183 -2.2 4D 0.0063 0.0064 1.6 3A 0.0170 0.0183 7.1 Total 0.0640 0.0667 4.0
'((ap cale - 00 me as OP eale
'00.
4- 3
n p
q approximately:75 for any single group, while .the difference between the
" measured and calculated cumulative worth is only approximately- 4%
(0.001 60).
3' These data are consistent with cycle-1 and 2 data in that the cumula-tive integral worths are predicted within a few percent while individual rod groups are predicted with somewhat less accuracy (Refs.17 and 18).
All measured rod group worths are, however, within the acceptance criterion-specified for the test.
O- ,
4.3 TD(PERAWRE DEFECT .
Surveillance requirement SR 5.1.3 specifies that a measurement of the reactivity change as a function of fuel temperature be made at the beginning of each refueling cycle. These measurements were performed by j +- PSC during- the initial cycle 3 rise-to power.
i
?, The temperature defect was calculated using the 7 group GAUGE model as described in Ref.16.
r Figure 4-1 shows a compariso'. of the measured and calculated tempera-
-ture defect. It is seen that '.no measured temperature defect, fr om 220 *F ,
_a- to 1500*F, is approximate 1v 65 lower than the calculated. This is consistent with cycle 1 and 2 data and within the expected range.
a
- , 4.4. NUCLEAR DETECTOR DECALIBRATION a
Power-range nuclear detector signals are used to monitor core power during steady state and transient conditions and in the automatic control system to initiate plant protection (PPS) action.
=, During cycle 1 and 2 operation, significant decalibration of these j power range detectors, due to motion of control rod groups, was predicted and measured. This decalibration results from the location of the six 4-4 t
7
__ axn:v_,.
l- _ _
a
. - ,- . , - - m..c: -
. . .. g w - y, t'
^
CYCLE SITEMPERATURE DEFECT-AS A FUNCTION OF FUEL TEMPERATURE.
-0.06
-- /
/s
/ .
/ .
/
0.05- - CALCULATED --
~ . /
n
/ t
.l
-+~
. /- .
l 0.04- ---- - -
? /- - - - - - - - - - - - - - -
s -- .
f '
h /i
/-
g o : /
i w g.n3 .
-....-.4.. . -- .;
u}f- . .. . . . . .
g(ASyggg +- . . .. . . . . .
E -
~ / i h
E
/ i g /~ .
7-y 0.02- - -- - - - - - - - - - - - - - - - - - - - ;
N ./
/
~
l 0.01- -
i----- -
- + --- -
--+- +- -
- 1 . .
i .
1 -
i : :
t 0.00 , , , , , ,
200 400 600 800 1000 1200 MOO 1600 -
FUEL TEMPERATURE (*F) .:
i i
I Fig. 4-1 :
4m-,-%=i .
m- . , _ . , 74em%. -
,_ui,.a_,.__f _
j,, .___ _ _ _ _ _ _ _ _ _
- T 1
- detectors symmetrically around the core, in the PCRV (see Fig. 2-1). This means that each detector " sees" neutrons from principally a few fuel colunris near .the core boundary. Thus, changes in the distribution of: power throughout. the core, due to control rod position changes, can cause the detector signal to change more or less than the corresponding actual. change in total ~ core power. As a result of this detector decalibration, a
" floating" trip point was recommended to assure the reactor trips at or. I below 1405 thermal power (Ref. 19). The floating trip point hardware is
- not yet installed so the detector decalibration is accommodated by a reduction in the fixed PPS trip setpoints and frequent calibration of the nuclear channels. 'The fixed PPS setpoints are intended to activate a rod withdrawal prohibit-(RWP) at 120% of rated power and to ensure that a I reactor trip is always initiated before true power reaches.the trip point of 1405 rated power as required by the Technical Specifications.
l 1
Because of the different control rod withdrawal sequence in cycle 3 and the different core fuel loading distribution, it was necessary to re-evaluate the detector decalibration. Results of these detector -
decalibration analyses are reported in Ref.16. Basically, the detector ;
1 decalibration, i.e. , the ratio of the detector indicated power level to the j
- true power -(heat balance), is calculated using the GAUGE code (Ref. 14) l along with ' nfluence coefficients representing the contribution of adjacent
. I 1
columns to each detector response. Fixed PPS trip setpoints as a function' J of power were specified for cycle 3 operation, based on these analyses. :l l
Measurements made by PSC during December 1983, using test procedure l
T-174, provided data on the measured detector decalibration for verifica- j tion of these fixed trip setpoints. The voltage to each of the six linear .I channels 'was-measured and this voltage is proportional to reactor power.
Valid measured data were obtained for only two control rod groups (3C and
- 48) as the groups were fully withdrawn. A re-calibration of the detectors was-performed both before and after each rod group withdrawal. The measured decalibration factors were then defined as the ratio of the voltage before calibration to the voltage after calibration.
4-6
_. - .. . - - ~ . , . . - . . -
y Table 4-2 gives a comparison of .the measured and calculated detector decalibration factors for sequential withdrawal of control rod groups 3C and 4B.. It. is seen that the measured detector decalibration factors are generally lower than the calculated for rod group 3C and higher than the calculated for rod group' 48.
The worst caso decalibration factor (decalibration which most delays the PPS trip) is typically calculated to occur when a control rod group is inserted after a control rod pair runs out. The calculated decalibration factor for the rod pair out was combined with the measured decalibration factor for the group insertion to determine the worst case decalibration. 1 factor for control rod groups 3C and 48. Then, assuming opposite detectors are- grouped and- the signals auctioneered, and assuming that' one channel fails in a nontripped mode (a conservative assumption since this is a-two-out-of-three channel logic) the worst case decalibration factor determines the fixed PPS setpoint which will ensure a trip before the true ' l power exceeds 1405 of rated power. Comparison of this worst caso decali-bration with that using all calculateo data indicated only a slightly lower .
.setpoint for control rod group 48. However, the specified reduced PPS !
' fixed setpoints were determined from rod groups which cause significantly I l
more decalibration than groups 3C and 48. ;
l Therefore, based on the . limited valid measured data, there is no
-indication that the PPS setpoints based on calculated data are not valid. l However, the valid measured data are for control rod groups which do not cause a significant delay. in the PPS trip and, therefore, do not determine the PPS fixed reduced setpoints.
.l 4.5. FUEL ACCOUNTABILITY l
l Fuel accountaoility calculations are done semiannually using the three-dimensional, 4 group GATT model (Ref. 20). These analyses provide fuel composition accountability information on a fuel element basis, 4-7 j
~
TABLE 4-2 CCNPARISON OF MEASURED AND CALCULATED ;
DETECTOR DECALIBRATION: FOR SIQVENTIAL CON 1ROL ROD' I OROUP WITHDRAWAL -
t!
Control Rod Group 3C Control Rod Group 4B ,
Nuclear #
Channel- Meas.** Calc. Meas./ Calc. Meas.*** Calc. Meas./Cale.
)
III' O.70 0.81 0.86 1.04 1.07 0.97 IV 0.73 0.81 0.90' 1.09 1.00 1.09 ,
1 V 0.82 0.82 1.00 1.11 '1.04 1.07 VI 1.12 1.17 0 96 0.95 0.98. 0.97 VII 1.06 1.18 0.90 1.19 0.99 -1.20 l
1 VIII- 1.01 1.23 0.82^ 1.01 1.03 0.98 l l
Ave. 0.91. 1.00 1.07 1.02' :l I
'See Fig. 2-1 for nuclear channel location
- Using linear channel voltage - variation in linear channel power too large to be valid.
- Using linear channel voltage - linear channel V power not valid at 190 in.
1 1
Y 4-8
^
r <
including accountability information for the fuel in the fuel test' elements (FTEs). ' Since GATT calculations are quite costly, the actual operating power history is modeled with only a few time steps. (The modeling of the complete cycle 3 power history fis given-in Ref. 21.) Then, a 7 group, two-dimensional GAUGE model is used to model tha power history and to check the more complicated GATT analysis. Comparisons of the end of cycle -(EOC)
" total core" results from the two calculations models 'are discussed in-Ref. 22 and summarized- below. [The detailed information for each fuel element is stored on magnetic tape (Ref. 23).3 4.5.1. Heavy Metal Inventory s
The heavy metal inventory of the core at the end of cycle 3 (EOC3) from the .two calculational models is given in Table 4-3 The inventory from the GATT model is 'slightly (0.015) less than that from the GAUGE model. This is primarily because the GAUGE model does not model the FTEs, whereas the GdTT model represents them explicitly. Note that with about the same Th-232 and Pa-233 inventories in both models, the GAUGE model is
. consistently higher in U-233 content but is lower in U-235 content by about the sand amount. Furthermore, the inventory of 'U-234 is higher in the GATT model as .lf the capture cross section of U-233 is also higher. The reason for this1 discrepancy is not known, but since U-233 is more reactive than U-235, the lower inventory of U-233 explains the lower reactivity of the
- GATT model (see Section 4.5 3). The difference between the total. core- ;
inventory calculated with GAUGE and GATT is essentially the same as during the previous cycles.
4.5.2. Region Peaking Factors (Two- versus Three-Dimensional Calculation)
Comparisons of the region peaking factor (RPF) distribution from the t GAUGE'and GATT calculations were made throughout the cycle and, in general, the differences were saml1 (approximately 45 to 55). Somewhat larger differences may occur in partially rodded regions, due to the inability of the GATT model to represent control rods in any position other than by 4-9 i
l
,, , n:. .
' '- t ,
..M . f l;{
q _,
i
, , TABLE 4-3 MTAL CORE HEAVY METAL INVINMRY. AT THE IDC3 (kg)
Nuclide GAUGE' GATT ;
Th-232 14,509 35 14.508.58
-Pa-233 17.09- 17.01
.t X U-233 215.27 208.36 9, U-234 21.8 6 d 23.04 U-235. 410 35 415.84.
U-236 89.02 -87.92' !
o U-238 51.12 51.06 Pu 3 15 3 23 I Total 15.'317.20 15.315.04 i
l
'l l i i
o I
9 h
4-10 s
I
-[l f [.
t 7
l integer number of fuel layers, and to the apparent difference in the "S-curve" used in the GAUGE calculations and the one implicit in the GATT calc ulatio ns. The "S-curve" used in GAUGE is based on measurements made at [
the beginning of cycle 3 and, therefore, does, not account for the burnup effect throughout the cycle. ya A discussion of the comparison of the RPF distributions from GAUGE and ][
GATT is given in Ref. 22. $f F
4.5 3 Fractional Absorptions and Excess Reactivity r
The fractional absorptions -by the major absorbers, the effective and infinite multiplication factors, and the total neutron leakage from the 7
. core calculated with the GAUGE and GATT models at the end of cycle 3 :are- _
compared in Ref. 22._ The results were, in general, as predicted.
The total neutron leakage in GATT was somewhat higher than in GAUGE - b throughout the cycle. At the end of cycle 3 ths ditterence reached about 0.003 Ak. ~ Since .GATT calculates the total leakage dir,20tly, while GAUGE relies on an estimated input value for the axial le akage, it is' reasonable to conclude that this difference is real.
4.5 . 4. Particle Burnup i
The maximum fissile particle fissions per initial heavy metal atom (FIMA), as.a function of active core layer and reload segment, is given in Table 4-4. As previous analyses have shown, the maximum FD4A generally occurs in the fourth layer of the active core. ' The maximum burnup of any segment is about 658 EFPDs out of a potential total of approxiantely 1750 EFPDs, with the maximum fissile particle FIMA of 14.15 cut of a projected 205 for the initial core segments. The maximum fertile particle ;
FIMA was. calculated to be 2 3% out of a projected 75. The GATT results indicate that, in general, FIMAs increase somewhat faster with burnup than 4-11
a: '
TABLE 4-4 MAXIMUM FISSILE PARTICLE FIMA (5) core Layer Reg. 1 Se ss. 3-6 se g. 7 se g. 8 1 (top). 8 to 6- 4 2 11- 12 9 6 3 13 13 10 7 4 14 14 12 8 5 13 13 11 7 6 . to 10 9 5 o .
In 4-12 l n
t i
predicted.with '.ae GAUGE results based on the precalculated axial dit rit-tion profiles. However, for the present level of burnup the discrepancy ;o no t ~ s ignific ant.
(
, 4.5.5. Axial Power Data The end of cycle CATT calculations were done with the control rod shim i group 30 665 withdrawn. The power fracdon in the top half of the core, at the end of cycle, is shown in Table 4-5 for each region of the core. These results indicate that the power fraction in some of the oldea fuel regions -;
is lower than the desired range' of 0.55 to 0.60. However, this does not 'l
\
l L "
create a problem-since the Technical Specification LCO 4.13 limits the magnitude of the axial power factor * (APF) in the bottom fuel layer. These .
-1 APFs, also given in Table 4-5, indicate that the maximum unrodded region 1 APF (region 30)- is approximately 65 lower than the LCO limit of 0.90.
q Although these data indicate a significant flattening of the axial l power distribution with burnup and partial rod insertion,' the fuel in the -l bottom core layer was depleted to such an extent that the power tilting toward the core bottom waa' mitigated and thus the LCO limits are met in - i every; region with a substantial margin. However, by comparing the present' q results to those ' earlier in the cycle. when- the-shim bank was 505 withdrawn, I it is noted that the APFs at EOC3 are somewhat higher. This-is contrary to the earlier projection that APFs had reached their highest levels, and that in the subsequent burnup an improvement in the core response to a partial control rod bank insertion would result. Therefore, a close monitoring of APFs in the next cycle is essential.
' Axial power factor is defined as the relative power in the bottom layer of a region to the relative power in the region.
1 4-13
1 i
j TABLE 4-5 j AXIAL POWER DA7A AT EOC3 l J
l Power Fraction i Region Control Rod in Top Fuel APF in j No. Ins ertion' Zone Botton Block i l
1 1 2 0.464 0.838 2 0 0 518 0.765 l 3 0 0 516 0.788 4 0 0.543 0.724 5 0 0.550 0 738 ,
6 0 0.611 0.781 7 0 0.514 0.776 8 0 0.564 0.678
- 9. 6 0 543 0.726 to 0 0 563 0.707 +
11 2 0.472 0.818 ;
12 0 0.527 0.743 f i3 6 0.545 0.725
-14 0 0.522 0.758 1$ 2 0.490 0.808 16 0 0.522 0.748 17 6 0 580 0.686 f 18 0 0.539 0.721 19 2 0.476 0.817-20 0 0.540 0.713 21 0 0 589 0.645 22 0 0.548 0.706 23 0 0.532 0.7 40
'O - Control rod fully withdrawn.
2 - Control rod inserted two core layers, i.e., withdrawn from four core layers.
6 - Control rod fully inserted.
4-14
TABLE 4-5 (Continued)
Power Fraction in Top Fuel Region Control Rod APF in No. Ins ertion Zone Bottos Block 24 0 0.519 0.769 25 0 0.572 0.652 26 0 0.537 0 712 27 0 0.543 0 728 28 0 0 579 0.671 29 0 0.534 0.730 l
30 0 0 502 0.805 31 0 0.506 0 798 32 0 0.586 0.626
- 33. , 0 0.549 0.7 01 34 0 0.551 0.703 35 0 0 578 0.648 36 0 0.579 0.632 37 0 0.526 0 747 Averugs 0.535 0 733 I
4-15 .
1 I
j l
4.5.6. Power Correction Facters for Fuel Test Elements I The axial power distribution in the columns containing fuel test l
. elements (FTEs) was compared to the region average axial power distribution at EDC3. These data are shown in Fiss. 4-2 through 4-5 for the coluims containing FTss 2 through 5. (The axial power distribution for the columns containing FT,E-6, FTE-7, and FTE-8 are not given since these FTEs are essentially the same as the elements they replaced, i.e. , only the type of graphite was changed and, therefore, the axial power d',stribution is J essentially unchanged.) These results indicate tht* ;) the power generation of the FTEs is about 105 to 155 lower th.a in the element they replaced, and (2) the presence of FTEs has only a small effect on the axial .
power distribution in regions in which they are located and essentially no effect on the power distribution in adjacent regions. These data also illustrate the flattening of the axial power distribution and the increase in the APF in the bottom fuel layer that was discussed above.
4.5.7. Burnup Data for Cycles 1. 2. and b ,
The heavy metal oore loadings and the f.*actional fissions at the beginning and end of cycles 1, 2, and 3 are given in Tables 4-6 and 4-7, ,
respectively. These data show that at IDC3 tt.e not enrichment is approxi-mately 805 with approximately 415 of the fissions occurring in U-233. .
Table 4-8 shows the segment burnup at the end of cycles 1, 2, and 3 From these data it is seen that at EDC3 the maxinua segment burnup was 36,417 Wd/ tonne, and the core average burnup was 31,278 Wd/ tonne.
4.6. REACTIVITY DISCREPANCY The reactivity discrepancy, at any time during operation, is the reactivity difference between the critical position of the control rods and the critical position predicted for the specific operating conditions. i 4-16 ,
FSV -CYCLE 3- 294.5 EFPD -
FTE-2 (22.06.F.06) 2.0
- 1. 8 -
g ,POKR/ REG 22, COL 6 x- REGION HVERAGE POW i 1.6 1
- 1. 4 -
" "J '
l.2 M 3 I
g 1.0 /2 i
% #m ~g I O l
CL 0.8 !
g lb
- s d
tr 0.6 -
V '
0.4 0.2 0.0- , . ,
1 0 40 80 120 160 200 240 260 320 360 400 440 480 S20 l RXIRL DISTRNCE FROM TOP OF CORE--CENTIMETERS
"~ '
L ----.-- -- _.-. --. _ - -- - - - - - ~-l------_------_----------------------- -
FSV -CYCLE 3- 294.5 EFPD -
FTE-3 (30.04.F.06) 2.0 0 - POWER / REG 30, COL 4 .
M - REGION RVERAGE POW !
1.6 1.4
_. Y
$7
^
1
<= m 1 .0 > & fs+==
F
=m g r
u O_ i U 08
- I I d
e
- 0. 6 -
0.4 0.2 ;
i 0.0 . ,
0 40 80 120 160 200 210 260 320 360 400 440 460 S20 RXIRL DISTANCE FROM TOP OF CORE- CEf1Tir1CTCRS Fig. 4-3
l FSV -CYCLE 3- 294.5 EFPD -
FTE-4 (27.02.F.06) l 2.0
- 1. 8 -
O - POWER / REG 27, COL 2
- M - REG 10fi flVERf1GC l@4 1.6 -
- 1. 4 -
1.2 m , , ,
1 b 3 1. 0 -
O U 8 e
0.8 -
x g 0 .6-Y t
- I 0.4
- 0. 2 -
i l
l 0. 0 - , , , ,
O 40 80 120 160 200 240 260 320 360 400 140 180 520 :
OXIBL DISTANCE FR0t1 TOP OF CORE- CENTir1ETERS
! . . - ._. ~ . e4 . .
FSV -CYCLE 3- 294.5 EFPD -
FTE-5 (24.03.F.06)
~
2.0
~
O - POWER / REG 24, COL 3 M - REGION RVERAGE POW l.6 l 1.4 Al y m, gjlf lh " *% N s
d
.0 0.6 -
x 0.4 I
0.2 0.0- i i , ,
0 40 80 120 160 200 240 260 320 360 400 440 460 520 l RXI AL DISTANCE FR0f1 TOP OF CORE- CErJTit1ETERS rig. 4-3
1 TABLE 4-6 HEAVY METAL CORE LOADING BCC1 EOC1 BOC2 EDC2 BOC3 EOC3 Th-232, kg 15,905 15,785 15,276 15.151 14,697 14.510 U-233', kg 0 103 87 168 139 232 U-235, kg 721 552 655 498 624 410 Uranium, kg 774 740 529 785 881 804 Plutonium" , kg 0 1.0 0.9 1.8 1.6 32 ,
U-233 enrichment, 1 0 14.0 10.5 21.5 15.8 28.9 U-235 enrichment, 5 93.1 74.5 79.0 63 5 70.9 51.0 No t e nric hmen t * * * , 5 93 1 88.5 89.5 d4.9 86.6 79.9
- Assumes full decay of Pa-233.
" Assumes full decay of Np-239.
ese(Mass of U-235 + U-233)/ mass of uranium, t 4-21
1 J
l TABLE 4-7 f FRACTIONAL FISSIONS '
80C1 EDCI 80C2 EOC2 BOC3 EOC3 U-233 0.0 0,19 s o,,$o n,3n3 0 ,,5 0 #08 U-235 1.0 0.798 0.846 0.688 0.769 0.580 Pu-239 0.0 0.004 0. 002 0.006 0.003 o,oo7 5
6 k
7 1:
4 22
g ..
i TABLE 4-8 SEGMINT BURNUP Burnup (mwd / tonne)#
Fuel Segment EOC1 EOC2 EOC3 1 6,352 -- --
2 6,496 13,000 --
3 8,398 17,312 29,400 4 10,750 21,134 36,267 5 10,396 21,024 34,383 6- 10,622 20,975 36,417 7 -- 12,345 30,308 8 -- --
20,895
)
Core Ave. 8,836 17,632 31,279
' Megawatt days per initial metric tonne of thorium plus uranius.
- Assumes all segments are equal volume.
1 4
34 - 2 3
---e -we w -
l:
The reactivity discrepancy is monitored throughout the cycle (see Refs. I to 8). Calculations are done using the 7 group CAUGE model and j
during cycles 1 and 2 these calculations, in general, overpredicted the
! core reactivity by approximately 0.0065 Ak 2 0.0015 ak. This 0.0065 Ak is j the calculational reactivity bias. Then, the reactivity discrepancy is defined as the difference between the calculated k,ff and 1.0065, i. e. , the deviation from the calculational bias.
Figure 4-6 shows the calculated k,ff minus the measured k,ff (i.e. ,
1.00) as a function of EFPDs for cycle 3 These results indicate that the j prediccion of the initial cold criticality was as expected (i.e. , con-firming that the "as-built" segment 8 fuel and L8P loadings were very close to the design values). However, with burnup, the variation of the cold ;
critical predictions has a larger spread than in previous cycles, especially near the middle of the cycle. The cause for the larger spread in the predic,tions can be at least partially explained by larger uncer-tainties in the measured data. A review (by PSC and GA) of the measured data recorded for the cold critical conditions throughout cycle 3 indicated that there were significantly larger u1 certainties in the measured data -,
recorded near the middle of the cycle than at the beginning and end of the .
cycle. Additionally, the uncertainties in the measured data (in particular, uncertainties in the recorded oore inlet and outlet temperatures) for the middle of cycle cases were sufficient to cause the I=
observed deviations from the bias. Therefore, it is concluded that the calculational reactivity bias of 0.0065 ok is still valid for cold critica11 ties.
The data in Fig. 4-6 show that the calculational bias for the hot operating oonditions is increased to approximately 0.0090 Ak with an uncertainty, except in a few cases, of approximately 0.0015 Ak.
The increase in reactivity discrepancy (i.e. , the deviation from the calculational' bias) at about 50 EFPDs was shown to be due to the inadver-tent release of the reserve shutdown material into region 27, and the
-inecease at about 140 EFPDs was due to an actual core power that was up to 4-24
e ,-_ma-ed .m_.memmm.a,_
9 j
t I
)
i a.
h=
D n
m N
c b
w C
5_
4 b
=
k a.
M
$ w j w
t a.
2
= w 2 E a 3 M
w z d M l D E
- \
x ,
c W
>=
.$ 1 4 ,
5 !
b 5
m e . u emmuhe (W) ADNYd213S10 AIIAI13Y3H 4-25 .
..-,...n ,, . _ , , , , _- - , - - . .,
I l
l i
105 higher than the official power level reported. At 200 EFPDs the decrease in reactivity discrepancy was the result of including the overburn i i
in the calculation. '
It is concluded that the cold reactivity bias for cycle 3 is unchanged from the previous value of 0.0065 Ak 2 0.0015 Ak, but the hot reactivity bias increased to approximately 0.0090 Ak 0.0015 Ak. This increase is j due, in part, to uncertainties in core temperature. Since both the cold ,
and hot reactivity bias remained relatively constant throughout cycle 3, )
the reactivity discrepancy remained small.
1 4.7. CONTROL ROD MANAGEMENT l l
a The irradiation history of control rod pairs was calculated based on 1
'l the flux distributions calculated with the GATT model (Section 4.5). The l
cumulative fast neutron fluence of the absorber clad (canister) is given in Table 4-9. the cumulative thermal neutron fluence of the clad is given in j Table 4-10, and the burnup of absorber compacts is given in Table 4-11.
The calculated results are in good agreement with the predictions based on the GAUGE model given in Ref. 24 Note that the irradiation history was compiled based on the assumption that individual rod pairs have remained in the same core region since the beginning of the initial cycle. It is known, however, that some rods have been interchanged within the core and some rods have been exchanged with spare rods. Therefore, until a clear record of all such rod exchanges is available, it is not possible to make realistic assumptions about individual rod history. However, at this stage of irradiation, parameters are so much lower than the limits that the exact
- knowledge of the exposure of each rod is not necessary.
4-26 ,
._-m______________________________ -_____m m . ._ ______.-__
TABLE 4-9 FAST NEUTHON EXPOSURE OF CLADUI (TOTAL FOR 3 CYCLES)
CONTROL ROD SEGMDIT(2)
Control Rod Pair A B C D E F Avg I3) MaxI *}
1 0.00 0.00 5.27+18 2.46+20 9.20+20 1.49+21 4.43+20 1.49+21 2 0.00 0.00 0.00 0.00 0.00 3 00+20 5.00+19 3 00+20 3 0.00 0.00 0.00 0.00 2.13+18 2.83+20 4.75 + 19 2.83 + 20 4 0.00 0.00 0.00 0.00 0.00 2.86+20 4.77+19 2.86+20 5 0.00 0.00 0.00 0.00 1.73 + 18 3 035,D 5.08+19 3 03+20 6 0.00 0.00 0.00 0.00 0.00 2.97+20 4.95+19 2.97 + 20 7 0.00 0.00 0.00 0.00 2.81+18 3 07+20 5.16+19 3.07+20 8 2.83+20 3.87+20 3 96+20 3 24+20 2.40+20 3.56+20 3 31+20 3.96+ 20 9 4.62 + 20 6.09+20 6.22+20 5 39+20 4.63 + 20 5.36+20 5 38+20 6.22+20 7 10 1.97+20 3.02+ 20 4.12+20 8.25+20 4.08+20 5.17+20 3 77+20 5.17+20 0 11 6.43+20 8.98+20 9.48+20 8.49+20 8.93+20 9.29+20 8.60+20 9.48+20 12 3.21?20 4.26+20 4 32+20 3.55+20 2.67+20 3.71 + 20 3.62+ 20 4.32+20 13 4.22+20 5.56+20 5.83+20 4.92+20 4.19+20 4.70+ 20 4.90 + 20 5.83+20 14 1.69+20 2.61 +20 3 54+20 3.82+ 20 3.67+20 4.99 + 20 3 39+20 4.99 + 20 15 6.23+20 8.68+20 9 34+20 8.42+20 8.84+20 9.06+20 8.43+20 9.34+20 16 3 11+20 4.24+20 4.40+20 3 56+20 2.66 + 20 3.7 4 + 20 3.62+ 20 4.40+20 17 5.19+20 6.91+20 7.02+20 5.56+20 4.72+ 20 5.11+20 5.75+20 7.02+20 18 1.57+20 2.44+20 3 37+20 3 55+20 3.44+20 4.51+20 3.15+20 4.51+20 19 6.55+20 9.09+20 9.53 + 20 8.54 + 20 8.95+20 9.30+20 8.66 + 20 9.5 3 + 20 ,
20 7.15+19 1.02+20 1.08+20 8.59+19 7.47+19 2.05+20 1.08+20 2.05+20 t 21 1.9 2+ 19 2.53+19 3 39+19 5.88+19 I.07+20 4.32+20 1.13+20 4.32+20 !
22 1.60+20 2.28+20 2.46+20 2.01+20 1.7 4 + 20 3.12+20 2.20+20 3.12+20 23 2.82+ 18 4.19+18 4.28+18 3.01+18 1.98+18 1.60+20 2.93 + 19 1.60+20 24 2.86+19 4.44+19 5.83+19 6.65+19 8.43+19 3.34+20 1.03+20 3.34+20 25 0.00 0.00 0.00 0.00 0.00 2.37+20 3.95+ 19 2.37+20 26 7.83+19 1.12+20 1.19+20 9.59+19 8.46+19 2.06+20 1.16+20 2.06+20 27 1.65+19 2.10+19 3 00+19 5.82+19 9.50 + 19 3.91 + 20 1.02+20 3.91+20 28 1 34+20 1.9 2+ 20 2.10+20 1.75 + 20 1.54+20 3.26+20 1.98+20 3.26+20 29 2.83+18 4.20+18 4.28+18 3 01+18 1.98+18 1.48+20 2.7 4 + 19 1.48*20 30 3.24+19 4.99+19 6.46+19 7.27+19 9.17+19 3 38+20 1.08+20 3 38 20
TABLE 4-9 (Continued)
Control Rod Pair A B C D E F Avg I3) MaxI " )
31 0.00 0.00 0.00 0.00 0.00 2.58+20 4.29+19 2.58+20 32 6.70+ 19 9.60+19 1.03+20 8.26+19 7.26+19 2.0$+20 1.04+20 2.05+20 33 1.53+19 2.00+19 2.81+19 5.34+19 9.00+19 3 78+20 9.75+ 19 3.78+ 20 34 1.72+20 2.45+20 2,65 + 20 2.18+20 1.90+20 3.41+20 2.38+20 3.41+20 35 2.60+18 3 86+18 3.93+ 18 2.75 + 18 1.82+18 1.66+20 3.02+19 1.66+20 36 2.24+19 3.44+19 4.41+19 4.80+19 5.73+ 19 2.8 4 + 20 8.17+19 2.84 + 20
- . 37 0.00 0.00 0.00 0.00 0.00 2.57+20 4.29+19 2.57+20 (1) Flux > 0.18 MeV (exposure in nyt).
(2)A + F - top segment to bottom segment.
(3) Average exposure of all 6 segments.
(g) Maximum segment exposure.
T l .
I l
l
TABLE 4-10 THERMAL NEU1RON EXPOSUME OF CLADII)
(TOTAL FUN 3 CYCLES)
CONTHOL HOD SEGMENTI 2)
Control Avg I3) MaxI "I Rod Pair A B C D -
E F 1 0.00 0.00 0.00 1.59+19 6.87+20 1.13+21 3.05+20 1.13+21 2 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 3 0.00 0.00 0.00 0.00 0.00 7.36+18 1.23+18 7.36+18 4 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 5 0.00 0.00 0.00 0.00 0.00 7.10+18 1.18*18 7.10+18 6 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 7 0.00 0.00 0.00 0.00 0.00 9.25+18 1.5 4 + 18 9.25+18 8 1.95+20 2.6 4 + 20 2.89 + 20 2.81+20 2.14+20 1.63 + 20 2.34+20 2.89 + 20 9 2.77 + 20 3 56+20 4.03+20 4.14*20 3.50+20 3.32+20 3.56+20 4.14+20 y 10 8.27+19' 1.14+20 1.83+20 2.28+20 2.36+20 2 34+20 1.79 + 20 2.36+20 g 11 3.62+20 4.79 + 20 5.23+20 5.51+20 5.89+20 6.27+20 5.22+20 6.27+20 12 1.76+20 2 36+20 2.60+20 2.54 + 20 1.97+20 1.49+20 2.12+20 2.60+20 13 2.72+ 20 3.52+ 20 4.00+20 4.25+20 3.61+20 3 35+20 3 58+20 4.25*20 14 9.17+19 1.26+20 2.00+20 2 39+20 2.50+20 2.42+20 1.91+20 2.50+20 15 3.69+20 4.89 + 20 5.28+20 5.52+20 5.75+20 6.03+20 5.19+20 6.03 + 20 16 1.7 2+ 20 2.33+20 2.56+20 2.53+20 1.96+20 1.51+20 2.10+20 2.56+20 17 2.55+20 3 31+20 3.81+20 4.23+20 3.53+20 3 33+20 3 46+20 4.23+20 18 9.38+19 - 1.32+20 2.10+20 2.54 + 20 2.63 + 20 2.52+20 2.01+20 2.6 3 + 20 19 3.73 + 20 4.92+ 20 5 37+20 5.62+20 6.00+20 6 39+20 5.34+20 6.39+20 20 5.91+19 7.69+19 8.55+ 19 8.34+19 7.44+19 6.82+19 7.47+19 8.55+19
- 21 8.44+18 1.11+19 1.25+19 3.02+ 19 - 3 26+19 1.12+20 3.44+19 1.12+20 22 9.31+19 1.21+20 1 36+20 1.40+20 1.18+20 1.04+20 1.19+20 1.40+20 23 2.02+18 3 03+18 3.11+18 2.75 + 18 1.98+18 1.28+18 2.36+18 3 11+18 24 1.49+19 1.93 + 19 2.82+ 19 3.43+19 4.65+19 6.49+19 3.47+19 6.49+19 25 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 26 6.55+ 19 8.5 3+ 19 9.56+19 9.46+19 8.52+19 7.93+19 8.42+19 9.56+ 19 27 9.81+18 1.25+19 1.38+19 3.69+19 4.12+19 1.25+20 3.99+ 19 1.25+20 28 1.05+20 1.37+20 1.55+20 1.62+20 1 39+20 1.25+20 1.37+20 1.62+20 29 2.14+18 3.20+18 3 29+18 2.91 + 18 2.09+18 1.35+18 2.50+18 3.29+18 30 1.69+19 2.20+19 3.14+19 3.00+19 5.10+19 7.17+19 3.85+ 19 7.17 19 1
- _-_,-c. -_ -~ _ _ - _ _ _e __m__ _e _u _z_____._.____e___._..__ m ___,._m _m _ .___m.
_ _ _ _ _ _ _ _ _ _ _ _ . _ _ _ .a ...w-. _ ___ .-______r .m . - . . _ .
__. .m~ - _ _.,. - __..._
~
- ~
. - 3,.y
-~
. TABLE 4-10 (Continued)
Control Hod Pair A B C D E F Avg (3) n,x(4) l 31 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 32 6.84+19 8.90+19 1.00+20 9.88+19 - 8.83+19 8.17+19 8.77+19 1.00+20 33 9.37+18 1.20+19 1 32+19 3.54+19 3.93+ 19 1.47+20 4.27+19 1.47+20
, 34 8.94+19 1.16+20 1.31+20 1 35+20 1.15+20 1.02+20 1.15+20 1.35+20 l 35 2.29+18 3.42+18 3.52+ 18 3.10+18 2.23+18- 1.44+18 2.67 + 18 3.52+18 36 1;71+19 2.23+19 3 16+19 3 78+19 4.76 + 19 6.32+19 3.66+ 19 6.32+19 37 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 (1) Flux < 2.38 eV (exposure in nyt).
(2)A + F - top segment to bottom segnant.
(3) Average exposure of all 6 segments.
i , g) Maximum segment exposure.
l I
l
TABLE 4-11 CONTHOL kOD BURNUPII)
(TOTAL FOR 3 CYCLES)
CONTHOL ROD SEGMDIT(2)
~
Control Bottom (D)
Rod Pair A B C D E F Avg (3) nax(4) camp ct--
1 0.00 0.00 8.50-04 4.12-02 8.26-02 1.32-01 4.28-02 1.32-01 1 32-01 2 0.00 0.00 0.00 0.00 0.00 5.04-02 8.40-03 5.04-02 8.45-02 3 0.00 0.00 0.00 0.00 3.00-04 4.77-02 8.01-03 4.77-02 7.95-02 4 0.00 0.00 0.00 ~0.00 0.00 4.46-02 7.43-03 4.46-02 7.50-02 5 0.00 0.00 0.00 0.00 2.85-04 4.51-02 7.57-03 4.51-02 7.55-02 6 0.00 0.00 0.00 0.00 0.00 4.95-02~ 8.25-03 4.95-02 8.31-02 7 0.00 0.00 0.00 0.00 3 78-04 5.10-02 8.56-03 5.10-02 8.46-02 8 2.48-02 3.37-02 3.57-02 3 24-02 2.46-02 4.63-02 3 29-02 4.63-02 6.68-02 9 3 68-02 4.78-02 5.15-02 4.94-02 4.32-02 5.99-02 4.81-02 5.99-02 7.59-02 ;
10 1 31-02 2.06-02 2.87-02 3 24-02 3.19-02 5.66-02 3 06-02 5.66-02 7.60-02
, 11 4.92-02 6.77-02 7.41-02 7.45-02 7.44-02 8.07-02 7.01-02 8.07-02 8.43-02 2, 12 2.42-02 3 24-02 3.43-02 3.12-02 2.40-02 4.66-02 3 21-02 4.66-02 6.75-02 e 13 3 52-02 4.61-02 5.01-02 4.89-02 4.27-02 5.79-02 4.68-02 5.79-02 7.32-02 14 1.29-02 2.03-02 2.81-02 3 16-02 3.15-02 5.63-02 3.01-02 5.63-02 7.56-02 15 4.95-02 6.77-02 7.39-02 7.44-02 7.31-02 7.85-02 6.95-02 7.85-02 8.20-02 16 2.37-02 3 22-02 3.43-02 3 11-02 2.40-02 4.72-02 3.21-02 4.72-02 6.84-02 17 3.68-02 4.83-02 5.19-02 5.05-02 4.37-02 5.77-02 4.82-02 5.77-02 7.24-02 18 1.27-02 2.03-02 2.84-02 3.20-02 3 17 5.59-02 3.02-02 5.59-02 7.49-02 19 5.07-02 6.91-02 7.53-02 7.56-c2 7.54-02 8.16-02 7.13-02 8.16-02 8.53-02 20 5.45-03 7.58-03 8.19-03 7.40-03 6.36-03 3 73-02 1.20-02 3 73-02 5.82-02 '
21 1.16-03 1.53-03 2.62-03 3 88-03 9.06-03 3.89-02 9.53-03 3.89-02 5.57-02 22 1.00-02 1.40-02 1.54-02 1.38-02 1.19-02 3.59-02 1.68-02 3.59-02 5.35-02 23 2.03-04 2.99-04 3 06-04 2.43-04 1.69-04 3.42-02 5.90-03 3.42-02 5.67-02 '
24 1.69-03 2.55-03 3.47-03 4.27-03 5.35-03 3.89-02 9.37-03 3.89-02 6.08-02 25 0.00 0.00 0.00 0.00 0.00 3 26-02 5.43-03 3.26-02 5.49-02 26 6.02-03 8.40-03 9.14-03 8.35-03 7.27-03 .3.60-02 1.25-02 3.60-02 5.55-02 27 1.19-03 1.52-03 2.86-03 4.37-03 9.10-03 3.92-02 9.70-03 3.92-02 5.62-02 28 1.01-02 1.41-02 1.58-02 1.44 1.25-02 3 45-02 1.69-02 3.45-02 5.09-02 29 2.11-04 3 10-04 3.18-04 2.53-04 1.75-04 3.27 5.66-03 3.27-02 5.42-02 30 1.92-03 2.88-03 3.86-03 4.70-03 5.87-03 3.89-02 9.69-03 3.89-02 6.04-02 y-. a-- n n. u. . , , . - - , , , - . ..e e ..ne + r ,,- ,--
- - -m.
~
TABLE 4-11 (Continued)
Control -
Bottom (S)
Rod Pair A B C D E F AvgI3} Max I) Compact-31 0.00 0.00 0.00 0.00 0.00 . 3.47-02 5.78-03 3.47-02 5.84-02 32 5.87 8.20-03 8.93-03 8.21-03 7.10-03 3.46-02 1.22-02 3.46-02 5.35-02 l- 33 1.13-03 1.47-03 2.69-03 4.15-03 9.10-03 4.04-02 9.82-03 4.04-02 5.67-02 34 1.01-02 1.40-02 1.55-02 1 39-02 1.20-02 3 62-02 1.70-02 3.62-02 5.41 35 2.13-04 3.12-04 3 19-04 2.56-04 1.78-04 3 23-02 5.60-03 3 23-02 5.38-02 l
36 1.67-03 2.52-03 3 33-03 3.96-03 4.66-03. 3 52-02 8.56-03 3 52-02 5.55-02 37 0.00 0.00 0.00 0.00 0.00 3.48-02 5.80-03 3.48-02 5.86-02 g
Burnup in fraction of initial boron.
<3,A . ,_ top to m e. seg.ent.
, ,) Average burnup of all segments.
7 ($) Maximum segment burnup.
iM Bottom boron compact of bottom segment.
4 i
e s __- ___r __4__ *--r-- A_ -__._._m_, _amm-_._____m._.__ __m____ ______.,____-_.__=__._mm-.___m._ -
- . . . ,-_.m __.--._..-____,m. -
-m.
l, i1 J
l 1
l 1
- 5. THERMAUFLOW PERFORMANCE l
5.1. REGION PEAKING FACTOR Throughout cycle 1, 2, and 3 operation, the distribution of power I generation among the 37 refueling regions of the core was monitored. The power distribution is characterized by the region peaking factor (RPF),
which is the ratio of the average power in a region to the core average power. The RPFs are monitored to demonstrate that the power distributions are within the limits stated in the bases of Technical Specification l LCO 4.1.3. !
l Analyses of measured and computed RPF distributions during cycles 1, 2, and 3 have shown significant RPF discrepancies # (>105) for the northwest (NW) boundary regions 20 and 32-37. The discrepancies in these regions are typically negative (indicating a measurement lower than the calculation) and increase with core pressure drop. Other regions exhibit sraller dis-crepancies and are essentially independent of core pressure drop. The l major cause of the RPF discrepancies in the NW boundary regions is a transverse flow of relatively cool helium from the core-reflector interface along the region exit thermocouple sleeve (Type II flow). This flow passes over the region exit thermocouple assemblies of these regions and depresses ;
the indicated region exit temperature. The driving potential for Type II
- The RPF discrepancy is defined as the percentage difference betwer..
the measured and calculated RPF, i. e ,
z seas. - R PF,,yn,
( RPF Discrepancy = 100
( RPF e ,3,,
/
5-1 >
uniform fuel temperature. (See Ref. 3 for details of this calculational procedure.) This calculation was done for each control rod group, as the groups are withdrawn in sequence.
These data were then interpolated to determine the RPF distribu-tion for any specific operating control rod group configuration.
This procedure was repeated throughout the cycle au required by SR 5.1.7 o The measured RPF distribution, used in determining the RPF discrepancy distribution, was generated with the FSVCOR code (Ref. 29) using the calculated region exit temperature for regions 20 and 32-37 and the measured exit temperature for all other regions, as discussed in Ref. 28. This measured RPF distribution is then compared to the calculated distribution to.
determine the RPF discrepancy distribution. The calculated RPF distribution was generated with the 7-group GAUGE code, using fuel temperatures generated above with the FSVCOR code and atom densities from the most recent GAUGS ruel accountability calculatio ns. Detalls of these calculational procedures are described in Refs. 28 and 60. This calculation is repeated as required by SR 5.1.7. !
A typical measured and calcu;ated RPF distribution and the RPF discrepancy distribution is shown in Fig. 5-1 for a power level of -705 and a cycle 3 burnup of 264 EFPDs.
1
.j i
k i
5-3 i
- _ . -_ . . . _ _ _ _ _ . _______________________________J
o n o u ,
j,04 'M
- l.e 4 .45 j.31 , l.40 *
" j,p V l.17 n ,, g 3 lolb " "
lJY n u /. 34 n s .qq * 'go .A3 .
- q3 ' 8 . fy
.40 ,
'77
.72 n
,g /
n
.gy
.,g n n 4g u (f a u fo '
i.
' ' 'l# *II
- j, /f 'IS
,Q0 4g /. te ,il ,
.fg pg n .. y u u
'y u u ,gc n
- N j.17 *05 >
,, p j Ib4 f.ts! I'# I "
/ 37 j,pt n
/. JI n f, g3
- 3. j,f 9 "
l.2 %
" /, /f " )
" f, g/
.if ..
/.12
/ 2f ,9f n
j,p3 ;
n *yg .
" 70 t.09 n
.qg n as i
,f? , ,gs ,41 , *4, n
,"f 2 1.06
.f9 .go I. I 4
- q,4 j
\
i
\
(
Measured RPF Calculated RPF i 1
Distribution (December 16, 1983)
Distribution e n a n u ~2 n a
u _z .
" n -lf n u g o n u jos ,
n 3
a -p o s
- /9 o
gy "
y a
n u g /2 i.
' / . */7 l A/ , ,j
- 2
- " /2 "
" " /f ..
u lif n u jy
- 5 i s g 10 s
,93 u 3 u u n 3 "
" 3 n
/03 n 4 s.
4
-9 '
IN o. ,, ll z.
n
" -0 u ./3 n n .g I 3 -1 RPF Discrepancy Control Rod Distribution ,
Positions (in.) l Fig. 5-1 COMPARISON OF MEASURED AND CALCULATED RPF DISTRIBUTION 5-4
)
- 6. FUEL PERFORMANCE Fuel performance calculations were performed to provide predictions of fission product activity using the SURVEY and TR AFIC comptter codes (Refs. 62 and 63). Comparisons of measured and calculated fission product release have significant implication since they validate, either directly i or indirectly, many of the methods currently used in FSV reload fuel segment design and core performance monitoring. These methods involve calculation of core power distributions, fuel temperature distributions, i fuel particle failure, and fission product release. Cycles 1 and 2 were reanalyzed, along with cycle 3, since significant improvements have ,
recently been made to the fuel failure models.
6.1. CALCULATIONAL METHODS The - fuel performance models defined in Refs. 62, 63, 65, 66, and 67 were used in these analyses. These models differ significantly 0Fom the ones (essentially the FDDM/B sodels) used in the previous cycle 1 and 2 analyses (Refs. 68 and 69). The main differences are discussed below.
o F. (the probability of an OPyC coating failure upon a sic coating -
failure) is now explicitly modeled. It is given in FDDM/E as a function of burnup and temperature. This modeling of partially failed particles has a significant impact upon fission product ,
release, since it allows for retention of fission gases for particles with failed sic but intact OPyC coatings. In previous analyses, bounding values for F. (F, = 0 and F, = 1) had to be assume d.
6-1
o The model for fission gas release from as-manufactured heavy metal contamination was revised so that contamination fractions .
for thorius and uranium are now input, and temperature dependence of the release from contamination is now decoupled from the one for failed particles. The thorius and uranium contamination fractions used in this analysis were taken from QC records for various service-limit fuels and reload segments (Ref. 59).
o The model for fission gas release from failed fuel particles was completely revised in FDDM/E. Separate formulations were
(, provided for. hydrolyzed and unhydrolyzed fuel, also allowing for different temperature dependence of the release. The analysis j
- was first performed by assuming that failed fuel particles were !
l completely hydrolyzed as this corresponds to the standard design i practice. However, since this assumption may be too conserva- I tive, the analysis was also performed for the unhydrolyzed cond$1 tion. ,
o Fission product sic corrosion was not considered in previous fuel performance analyses since this reaction has been assumed to be nonexistent in the FSV fuel. However, recent reexamination c
. the FSV fuel samples indicated some presence of this reaction (Ret 15). For this reason it was recommended, in Refs. 66 i and 71, that this failure mode be included in FSV fuel perfor-mance analyses. This recommendation was followed in these analyses.
In order to predict nominal fuel particle failure, all the calcula-tions were performed at 505 confidence level. The calculated fission gas ,
release, based on the FDDM/E fuel performance models, substantially exceeded the measurements. A review of the FDDM/E fuel performance models indicated that there was excessive conservatism in the f.U ; "o model for particles with missing buffer coatings. In the FDDM/E mo+c 4 the proba-bility of failure for particles with missing buffers increased linearly 6-2
i from zero at zero burnup to one at a relative burnup (F1MA/FIMA,g) of 0.2.
In order to provide a more realistic appraisal of failure in fuel with l missing or defective buffers, the FDDM/E model was revised in Ref. 71 for the FSV fuel. The revised model gives the failure probability as a function of the relative burnup, fuel temperature and kernel porosity: the physical assumption is that kernel porosity provides a limited reservoir ,
for the storage of fission gases. As recommended in Ref 71, a kernel ;
porosity of 75 was used, and the analysis with the revised model was ,
performed.
Tts metallic release calculations were performed for the key nuclides Sr-90, Cs-134, and Cs-t 37. Direct release calculations for Sr-90 and Cs-137 were done with the TRAFIC code. In the previous analysis (Ref. 69),
which was based on the FDDM/B performance models, a constant transition concentration (from Hencian to Freundlich sorption isotherms) was used, while in this an& lysis, based on FDDM/E models, the transition concen-l tration is a function of temperature. Also, the sorptivity of graphite for l
metals is now calculated as a function of fluence.
l For Sr-90 and Cs-137, the contribution from precursor decay was determined by calculating the R/Bs for Kr-90 and Xe-137 from the predicted l R/Bs ter the reference isotopes Kr-85m and Xe-138. j o
6.2. RESULTS l
6.2.1. Fuel and Graphite Temperatures ;
1 L
Figures 6-1 and 6-2 show the volume distributions of the peak and time averaged fuel temperature for segment 6, which is assumed to be a typical segment. The maximum and time-averaged fuel temperatures of 1262*C (2304'F) and 918'c (1684'F) were predicted. For the entire core the peak l,
and time-averaged maximum fuel temperatures of 1440*C (2624'F) and 1128'C (2063'F) were predicted. These maximum temperatures were predicted to occur'in segment 7, region 28, colum 5, local point 5, which is in a l
l 6-3
_ _ _ _ _ _ _ _ _ - _ _ _ . _ _ _ _ _ _ _ _ _ _ . _ _ _ _ - - _ _ _ _ _ - - _ _ _ +
Peak Fuel C/L Temperature Distribution Segment 6 g_
LEGEND D CYCLE 1 n..O. . . . . . .C. YC..I.A... 2...... . .
i P J_n b
e u -
d>
u i-e 4
i e
! . o. -
I 6 .
, e -
1 b 4 >
1 a
! . N O
-n e 3- :
NU"4 >
t 54 4
e
- n. ^g- . .... . ............... .... ................ ...................... .......
. ... ... . ,,_ ,,,,,, _ ,,,,,,,,,_ ,,,,,,,,,,,,,, _ _ ,_ ,_ ,,,,,,,_ ,,,,, u 1
- g. , , . . i o 20 40 so 80 100 l ,. , g . 6 t- Percent Fuel with Temperatures Greater than T.
l .
\
~
o8a.
6~. 6e :<>eLeaeo t n
":se oN bebc,i & H 8o 8 4 28 $_8 k g o -
o ~ g g
P .
T i
e .
r .
m e
c ai e .
n
_A t ,
. v
- e
_. F .
r
- u a e g
- l e d
_ w
_ i F u
F t i
g h4 a i e
S l -
_ 6
_ T c
e g C
_2 m/
_ m _
e L p
e -
r
_ n T a .
t
. e t .
m u
6p r .
. e
. e6 .
r
_ s0 a t.
G
_ r ~ u e . r a . e t
D e
r .
. i s
. t t
_ h
. r i
_ a a .
b n o. .
u t
T
_ i o
, O O n
C C L Y Y E CG C
L L E EE 2 1 N
D
_ 1 0i 0 va
+t; . ,Ei r, .t .!, 1: i >t: . >.* : :
thin-buffered coluim. Previous analyses of such partially buffered wlumns have indicated that the predicted intercolumn power tilts can be exc3s-sively conservative, resulting in large overpredictions of fuel tempera-tures, as was the case for the above point. Esen though the above maxitsua temperature is considered excessive, it was not discarded from the analysis since it is a result of the current methodology of analyzing the partially buffered columns. The peak graphits temperature, predicted to oncur at that location, was 1394'c (2542'F) and 1092*C (1998'F) on a time-averace basis. These high graphite temperatures have a large impact on the predicted cosien release as discussed in Section 6.2.4 6.2.2. Fuel Particle Failw'e At the end of cycle 3 the maximum fissile and fertile burnups were calculated to be less than 155 and 2.55 FIMA, respectively. These values are lower than the design values of 205 and 75 FIMA for segmenta 1-6.
I Fuel volume distribution of fast neutron fluence for sognent 6 is shown l in Fig. 6-3 The maximum fast fluence at the end of cycle 3 was less than 3 5 = 1088 n/ca8 , which is lower than the design value of 8 = 108 8 n/ca8 fo r-segments 1-6.
Time histories of fuel particle failure by different mechanisrJe are ,
shown in Figs. 6-4 and 6-5 for the fissile and fertile fuel. These failures are based on the revised manufacturing defects failure modol.
The maximum predicted fissile fuel particle failure from all mechanisms was 0.0725 with 835 of the total failure from the annufacturing defects. For the fertile particle the maximas predicted failure from all mechanisms was 0.0235 with 485 of the total failure tros the sic-fission product inter-action. This failure is due to high fuel temperatures in some parts of the core and particularly in the buffered fuel blocks, which dominates the manufacturing defects failure as the latter is strongly burnup-dependent t
i and burnup was rather low in the fertile fuel during the first three l cycles. It should b6 noted that in the earlier analyses the sic-finston 6-6
4.
n Nggs Zess4oc p
3 : AacCOeWwo"s- .
o N n 9 ,
o _ 1 l u -
- 4 E y g b:: ,'
s'
, ,~
P e
~ ,-
ra F
c oI e .- ~
~ -
a s
n ~
- _t t
,- N
~
F ,
- e u ~
u .
e .~ t l .- r
- o w ,
~
n -
i - ~
t4 F
h0 I ~ - -
F S
i l g
e u 6 F -
,- g e n
mc l :
3 u '
e ' e e n .~
. n c N t Di e g
- , _ s G ~
6 tr r -
' i e s- b a o s
t g
u e .-
- t i
r ,% o t -
' n
. h. .,- b a ' , y n -
~
~
,- ;Y Y ,
e a
A mg -
r x
i
- g
. ~
s ~ _
.O O .
L
_ C Y CE Y
C L L CG
.E EE
~
,t _
2 1 N
~
D -
' t . _
3 .% .
0y_
0 3
w 7
l l
'lI,i i ;,> '
. . "w~ ,
- ,, 5 Failed . Particle Percenta'ge byL Failure. Types .
Particle 1, yn/uicz ~
Core Average 8-d- . . .
LEGEND o Amoeba Migration g ~ ~o Pressure Vessel e.-
~
d- _522lUE_~5? Nil 5[U_I555555EEli
+ Manufacturing Defects ,
x Totaf Failures 8 ~
d-
- /-
.f e8 f 1
46-Y j . .
e N
T
. so
.d e i O
E8 m 6- ( ./
. /
, ) .
[ ,
6- .-
g ,,____g_________A-______-_-_-A-A d- .
.t 5
8:
6m
- 5 5 Ni i "M' m j =-
-I I=
-7
- 8 ' M- M 'e e eM 3 I= I. p. -
I. T g. -
I I I- :
i 0 100 200 300 400 500 . 600.. 700 800 900 1000- 1100 .1200 1300 1400 Tirr.e in : Days.
y Fig. 6-4 '
m -
...-j-
-u~
y - _
g $.
- 1. :, .y.
~
_. ~C--
Failed 1 Particle: Percentage by Failure Types Particle 2, nc, -
~
Core Average
' LEGEND l 0 Amoeba Migration o Pressure Vesse1 <
~
l_16f_7hrii.53h6fii))((hlE
+ Manufact.uring Defects g '
.9-o x - Tota _l Failure _s. _ __
i
/ --M -X ...,
I .x f_ (
I
_.g G) s L *o # 1 N $ . _ ,, ,,
2 . . - - - o e
e 0
A -
$ -e.)
'\'
C i
* -A g) -A________g___.
O? n- - -
,'~
k
[ Q- 0 =
a y' ,
i i
o o-o S
e a.
a ! .
- = = =a o
e a
8
-I e
l ~~O .. ...... ... -OO.
^
- g ....... - O ....5 . . . o . . .... ,._,
g .a
....- -- O ,,,,.. o . .. : . o ..
- " , ,, ___4__.-
~
<n.- ;
^ "I I -I" 5 . 5 I I 5- 5 I I g g 0- 200 B
100- 300 400 500 .600 700 800- 900 1000 1100 1200 1300 1100 i
. Time . - in Days .-
l _ , _ . --. _ - --
F i e . ' f> 5
- _ - - - - _ _ _ _ _ _ _ _ _ _ - - . _ - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -- - - ' ~-
Y
+.
1 product interaction was assumed to be nonexistent in the FSV fuel
~
par ticles. The fissile particle failure, which is far more dominant than the fertile failure, is predicted to be rather low up to approximately the middle of cycle 2 as a result of the revised manufacturing defects model
- where, due to low burnup, the failure due to missing buffers is low.
However, during the remainder of cycle 2, the burnup in most of the fissile i fuel (in . conjunction with fuel temperature) exceeds the threshold value beyond which a rapid increase in manufacturing defects failure' is predicted
]
according to the revised model for missing buffers.
1 6.2 3. Gaseous Fission Product Release [
A comparison of the predicted and measured fission gas release for the key nuclide Kr-85m is shown in Figs. 6-6 and 6-7. Figure 6-6 shows the predictions based on both the original FDDM/E and the revised manufacturing defects models with the assumption of completely hydrolyzed failed fuel. A comparison of the predicted release rate-to-birth rate ratio (R/B) with measured data (Ref. 32) indicates that the R/B based on the original FDDM/E manufacturing defects model exceeds the data early in cycle 1 due to large predicted manufacturing defects failure, due primarily to missing buffer layers. As discussed earlier,c this large overprediction prompted the ad revision of the manufacturing defects model. With the revised model, there ,
is good agreement between the predicted R/B and measured data up to g{ approximately the middle of cycle. 2 when the predictions start to exceed th's data, resulting h1 a factor of five overprediction at the end of
" cycle 3 This overprediction can be the result of either or. both of the following two effects: (1) in spite of the revision in the particle failure model, the failure fraction is still overpredicted at moderate and large burnup, and (2) the in pile effect of fuel hydrolysis is less than observed in tests; although hydrolysis has been shown to dramatically increase the R/B from failed carbide fuel in laboratory tests and in short-term TRIGA tests (Ref. 72), irradiation and thermally Lnduced sintering mitigates substantially the long-term in-reactor consequences.
In order to assess these two effects, the analysis was performed for
$.' I 6-10 !
~
'COMPARISION OF' MEASURED Kr-85m RELEASE ,
' DATA AND: PREDICTIONS FOR FORT ST. VRAIN .,
107. . .
. : : : : : : t l CALCULATED FDDM/E i j j i
. . . i . .
i i i i !
10-4_ : : : , , q : :
m : : : : :
T )
_ ,.i e hvi __/ 6 F LI. #'
~
- ! i i. I
" .: i jfl
~
E ; i ; :
u) - .
03 - - -
r u
- i. -
. -CALCULATED FDDM/EiREVISED F i
i 3e . : :
, _J- : . : : . .
10~,: i - ,_
-e
. e: !'
e e Oee e *
~* ' io e, i e., i [ g*%
~.# ee e-i ,1 #*
- i i
_ 3 - -
- J
! ! _ ! l MEASURED [ VALUES -i 4 CYCLE 1 .
1 I CYCLE'2 .
. CYCLE 3 .> '
10~ -
0 200 406 600 800 1000 1200 1400 DAYS I
n .x , . .,
.. . .. - ~ . . - . . .. . . . . .. . -
COMPARISION OF MEASURED Kr-85m RELEASE DATA AND PREDICTIONS FOR FORT ST. VRAIN
-S I 10 _ : : : : : :
4 :cycu 1 : : : : cycu 2 :: : : cycu 3 : >
~
~4 .
10 . . .
an -
N -
E - .
' I
~ I i ! !
T E -
i i HYDROLYZED .: :
-, :~b C in :
lL-Y r"#
no : !
I. -
- i. i. i. .
J .i %gl u
i i x i
- m g F : tr-,5 -
Mk '*kai4,
-s
..+
fl =
[>
- :, ,... k. i
~- i r =e "
i 1 a l
. , AP* j . *i. l a
MEASURED VALUES CONTAMINATION ONLY UNHYDROLYZED
[ I !
10-1 5 [ .
1200 1400 200 400 600- 800 1000 0
DAYS F ig. 6-7
o
)
) ..
-(1) the' case with no particle failure where the only contribution to fission gas release is 'due to as-1nanufactured heavy metal contamination, and (2) the unhydrolyzed condition. The results, shown in F1;*. 6-7, l indicate. that excellent agreement' with data was obtained throughout the first three cycles for both cases. Consequently, the cause of the overprediction remains ambiguous. Long-term in pile hydrolysis tests with carbide fuel would be required to resolve this. issue.
6.2.4 Metallic Fission Product Release The results of the metallic fission product release analysis for l' Sr-90, Cs-134, and Cs-137 are shown in Tables 6-1, 6-2, and 6-3, respectively. Also shown are the plateout probe data (Ref. 33) and the results of the previous analysis for the first two cycles (Ref. 69).
A comparison of. the results of this analysis with data indicates' that, in general, the agreement is reasonably good. For Sr-90 the predicted
_7 release of 0 37 C1 compares well with the measured values of 0.2 to-0.66 C1. The direct release of Sr-90 as calculated with the 1RAFIC computer code was negligible, and the only contribution was fromithe precursor (Kr-90) decay.
A comparison of the Sr-90 predicted release from- ,
this and the-previous analysis indicates that' EDC1 and EOC2 values
. predicted in this analysis are. lower due to the lower predicted fuel I
particle failures resulting.from the use of.. the new failure models. For Cs-134 and Cs-137 the agreement with data isL not'as good as for Sr-90. For j
-Cs-134 the_ analysis overpredicted the data by a factor of 6.8. For Ca-137 the total predicted release was 5.0 C1 while the measured values ranged from 1 3: to - 1.9 C1. This agreement could be improved by making a correc-' !
tion for the excessive release predicted for the maximum temperature point in the core. As discussed in Section 6.2.1,- even though the temperatures predicted for this maximum temperature point (the buffered portion of the -
region 28, colum 5, axial block 6) are considered to be in excess of 3 actual temperatures, they were not removed from the analysis. Due to excessively high predicted graphite temperatures, the graphite attenuation 6-13 ,
..i.
a t
TABLE 6-1 Sr-90 RELEASE (C1) '
Plateout Probe EOC1 EOC2 Remoyal Time- EOC3 This Previous This Pr evious This This !
Analysis Analysis Analysis Analysis Analysis Data Analysis-Direct 0 0 0 0 0' 'O release (from I TRAFIC code)
Precursor 0.)4 'O.59 0 32 1.44 0 37 0.60.
( Kr-90 )
Decay contribu- ,
1- tion ' -
Total 0.14 0.59 0.32 1.44 0 37 0.66* _0.60
'" Tota'l circuit" value' from Table 4-1'1 of Ref. 33: the "on dust" value-of O'.2, C1' was reported.
Y q
i 6-14
- e4
- 3 ; !
,. j f E'.'
n
' r_
TABLE 6-2 .
' Cs-134 RELEASE (C1)
Plateout' Probe EOC1 EOC2 . Removal Time .EOC3 Analysis Analysis Analysis Data Analysis Release 0.01 2.2 3.4 0.5' 8.0
?
'From Table 4-11 of Ref. ' 33
\\.s .
i l i l i
- i. ,
t I
)
t 6-15
3
~
4 ,
i
.i j . i -i 2 l
- TABLE 6 CS-137 RELEASE (C1) -.
t
\
Plateout Probe EOC1 EOC2 Removal Time EOC3 This Previous Tnis Previous This This i Analysis- Analysis Analysis Analysis Analysis Data Analysis l
Direct 0.05 35.0 9.9 212.9 11.8 19.0~
release (from TRAFIC' code) .
.i Direct 0.02 12.1- 3.4 73 5 4.1 6.6 i
-release (after correcting for mixed isotope i and mixed ,
Lapecies)- .
Precursor 0 33 0.4~ 0.8 1.1 0.9 1'. 5 <
- ( Xe-137 ).
decay . t
'contribu-
' tion ci' Total 0 35 12.5- 4.2 74.6 ~ -5.0- 1 3* , 8.1
'i I
'*" Total circuit" value from Table 4-11 of Ref. 33: the "on dust" value '
.ofl 1'.9 Ci was reported.
. 1
?
i
.si. \
6-16
___ = = _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ __ _
t 9
l of. the cosius release is'substantially reduced at this point resulting in l excessive predicted release.- The release at this point is predicted to j amount to over 90% of the direct release from the entire core. If this-
~
release' is replaced with the release from a similar partially buffered element- where the predicted fuel and graphite temperatures are considered -
reasonable, the total direct release would be substantially reduced. Thus for Cs-137 the corrected total release would be 1.1 ci which is in excellent agreement with the data. A similar improvement can be obtained for the cs-134 predicted release. _ A comparison of the predicted cs 134 release in this and the previous analysis indicates that the release predicted in this snalysis is considerably lower due to the lower predicted
-fuel particles failures as a result of the use of the new fuel failure models .
a 6 3. CONCWSIONS
- l The results of this analysis, based on the revised PVDM/E performance .i models indicate that excellent agreement was obtained between the' predicted and measured fission gas release for the key isotope' Kr-85m during the I first'. cycle and approximately half of the second cycle. Af ter this time the predictions exceed the data, and at the end of cycle 3 the data is overpredicted by a factor.of five. This overprediction can be the result- q of either or both of the following two effects:- -(1) in spite of the revision in the fuel particle failure model . the failure fraction is still }
overpredicted at moderate and large burnups, and (2) the in-pile :effect of '
hydrolysis is less than- observed in laboratory and 1RIGA tests. In order to assess these two effects further analysis was performed for (1) the case
- j. - with no particle failure where the only contribution to fission gas release
. is due to asenufactured heavy metal contamination, and (2) the lunhydrolyzed. condition.
The results indicate that either hypothesis is m
plausible. For the release of the fission metals Sr-90, cs-134, and Cs-137, reasonably good agreement was obtained between the predictions and data: from the plateout probe. The agreement was excellent for Sr-90.
6-17
However, the predictions for cesium exceeded the data. For. Cs-134 the analysis overpredicted the data by a factor of 6.8. For Cs-137 the predicted release was 5.0 Ci while the measured values ranged from 1 3 to 1.9 C1. This agreement could be improved by making a correction for the excessive release predicted for the maximum temperature point in the core..
This point is located in a partially buffered fuel element where fuel and graphite temperatures were overpredicted due to excessively high predicted intercoluun . tilts. Due to these high fuel and graphite temperatures, the-cesium release at the maximum temperature point is predicted to be over 905 f of the total release from the entire core. If this release is replaced with the release .from a similar partially buffered fuel element, where the l-predicted fuel and graphite temperatures are considered reasonable, the ,
1 predicted Cs-137 release would be 1.1 Ci which is- in excellent agreement with the data. A similar improvement can be made for the Cs-134 predicted ,
rele ase ,
f
' A comparison of the gaseous and metallic fission product release 5
. predicted in this and the previcus analysis for the first two cycles-
- indicates that the release- predicted in this analysis is considerably l lower. This is due mainly to lower fuel particle failure' fractions predicted as a result of the use of the new fuel particle models. New a models for the fission gas and metal release have also been used in this analysis.
Based on the results of this analysis, the following conclusionax can be made regarding the fuel performance methodology. !
o The manufacturing defects fuel particle failure model should De further. examined regarding the failure of particles with missing.
buffer coatings at moderate and high burnup to determine if there is still excessive conservatism in the model.
1 6-18
c '
- - =i E 4 j
i - \
4 fi !
o-' In order' to provide a definitive answer to the. perennial q'uestio'n ' I of the effect of. fuel hydrolysis on fission gas release during: j l
reactor. operation, in pile hydrolysis data is needed.. Planned- -1 hydrolysis tests (Ref. 70) any-provide some of~these answers, although these testa are for a different fuel (UCO). (
u
- n. '!
i e
, i.
t k f s
A ik
?
r
}
n 1
3 l
4 6-19 ,
)
'I
s i
- 7. SPECIAL TESTS AND SURVEILLANCES' 1
Throughout each _ cycle, special tests and surveillances are performed to provide additional data for core performance evaluation. Results of I i
cycle = 2 and 3 tests and surveillances are summarized below '(cycle 2 results I are those results which were not available for inclusion in the cycle 2: l l
L performance report). l
(-
7.1. CHEMISTRY SURVEILLANCE l
Chemistry surveillance tests were performed throughout cycle 3
' operation. Details of these test results- are given in Refs. 30 through 32 q l
and brief summaries of these results are given in the following sections.
1 7 .1.1. - coolant Impurities r
Surveillance data showed that at steady state 705 power operation the i-coolant .-impurities were low, totaling approximately_2 to 3 ppe co + co, +
~H 0 throughout the cycle. These results are similar to those data obtained in previous cycles.
7.1.2. . Noble Gas Release / Birth (R/B)
The circulating noble gas activity was low with Kr-85m R/B at 6 to !
7 x 10-* throughout the cycle (at approximately 705 power) and virtually. !
identical to previous cycles. This is significant in that -it shows that ;
fuel failure remains undetectable.
l 7-1
7.1 3. Tritium Conoentrations Primary- ooolant- tritium concentrations during 1982- and 1983 were extremely low, in the 10~' to 10~' pCi/ca' range which is below the expected concentration of >3 x 10~' pC1/ca' . It was suspected, therefore, i that the tritium monitor and/or the procedures were faulty (i.e. , giving erroneously low results). Accordingly, tests at GA on a mock-up of the FSV tritium monitoring system were conducted.
, Results of these tests show that the monitor concept installed at FSV
-is basically sound and. capable of producing repeated tritium values which- l differ by no more than t205. Some refinements in procedure were required to obtain this reproducibility, in particular (1) using long purge times before and between samples,- (2) ensuring that the Cu0 bed is functioning, and (3) using a slightly higher Cuo bed temperature.
The main conclusion of the tritium analysis is 'that the FSV data are
-correct. The explanation for the low tritium concentration throughout the c.ycle is that tritium is continuously being chemisorbed by the 1.2 million
-pounds of graphite'in the' core and reflector. This acts as a form of constant purification for hydrogen species, including H, and HT.
s 7.1.4. Iodine Monitor Analysis of an lodine monitor test showed that a small amount.of I primary He contamination in the cold trap could account fo.- the apparent high iodine concentrations. (relative to the plateout probe results). The contamination could explain the results, depending on when the contamina - 1 tion occurred, i.e. .. at the start or end of the test. Further tests were performed near the end of cycle 3 in which special precautions were taken 4 t'o exclude primary ooolant contamination from the gamma counting trap. 'l This test showed that the apparent circulating lodine concentrations were 5105 of that found earlier, suggesting that the early iodine monitor- :
7-2
f I
i i
results were erroneous (too high) due to contamination with primary loop
-helium.
I 7 .1. 5 . Plateout on Circulator .
1 The gaman and beta scan of circulator C-2101, removed at the end of j cycle 2, showed low total activities permitting virtual hands-on main- U tenance. Much of the activity was due to activation products that apparently are transported in the circuit on dust particles. The concen-tration of these radioactive particles appeared to be constant in cycles 1 l
and 2. Rather vigorous wiping of the surface, however, could effect a I removal portion of only 125. Iodine and Ca plateout on the circulator was lower'in cycle 2 than in cycle 1.
4 7.1.6. Gamma Activity of PGX Surveillance Staples 1 The gamma activity of the PGX surveillance samples < (located in the lower reflector during cycle 2) was low, with activation products i
- dominating the total radioactivity.- Only 0.013 uct of cs-137 was found on the 23 g sample. This equates to approximately 50 uci cs-137 plateout per lower reflector element per fuel cycle.
7.1.7. Gamma Scanning of- Circulator L . ;
The circulator C-2104 was removed to' GA for refurbishment and repair -
after operating for three . cycles. The inlet duct was analyzed for radio-
-active plateout, and showed that 'the concentrations of radionuclides were !
roughly equal to those found on circulator C-2101 after the first two' cycles of operation. All of the circulator analyses to date show low ,
levels of plateout, in general, ranging in the 1 to '1000 ' nanocuries per em8 .
l- !
l.
i L l 7-3 l
t
7.2. PLATBOUT PROBE The platecut probe is a device _ that samples the hot primary coolant -
gas for condensible fission products. It allows the amount of circulating and condensed key fission products in the circuit to be calculated.
Technical Specification LCO 4.2.8 limits the concentrations of I-131 and
]
- Sr-90, and the probe is intended to be. the primary means of demonstrating S. compliance with LCO 4.2.8. l 1
i Radiochemical analysis of the plateout probe showed that the radio -
active contamination in the primary circuit was remarkably low, with ;
"t activation product concentrations much greater than that of fission-products. , The analysis demonstrates that the concentrations of the key 1
't l LCO 4.2.8.- 'This conclusion is reached even though the probe may not- have j quantitatively sampled lodins. Other available data on iodine behavior, including the iodine monitor and grab sample xenon R/B measurements can be used at present to show compliance with the iodine specification Another important finding is that, to date, little or no barium, L ' cesium, or strontium was directly released from the core. Rather, these-volatile metals were released only via their gaseous precursors xenon and kry pton. Once in the primary circuit, a large fraction of these metals i
apparently affix to sen11 dust or aerosol particles and are thereby trans-ported around the circuit. In addition, the concentrations of Ba-140 and
- Sr-90 show that the R/B of short. half-life xenon and krypton isotopes can
! be accurately estimated by straight-line extrapolation of the R/B versus half-life (t1/2) curve. Finally, the concentration of H S in the helium coolant.is apparently below that required for deleterious sulphidation corrosion.
1 l- Details of the plateout probe radiochemical analyses are reported in ,
- Ref.- 33 i
7-4 L
- - - - - - ,,u .,
7.3. CORE SUPPORT BLOCK OXIDATION Core support block (CSB) oxidation is periodically monitored to determine the effect of graphite burnoff on the stresses in the CSB. The stresses considered in the analysis are due to the weight of the core, the differential pressure across the ccre, and the vertical acceleration loads.
resulting from an operational basis earthquake. CSB graphite burnof f is permitted until the stresses developed in the CSB graphite reach one-third the ultimate tensile strength of PGX graphite. At the end of the design exposure time, the stresses in the CSB will be approximately one-third of the ultimate tensile strength and the life of the block will have been consumed.
Previous oxidation studies of the 37 CSBs were evaluated for cycle 1 and 2 operation (Refs. 34 and 35). Region 27 was found to have the largest burnoff in each cycle. The maximum'CSB life consumed after cycle 2 was -
12.15, which la well below the ' allowed service life. This study was then extended through December 1982, i.e. , through the middle of cycle 3 Burnoff accumulated during the new period was calculated and added to previous burnoff profiles. However.- the northwest boundary regions of the core wer,e reanalyzed from the beginning of cycle 1 in order to correct the previous burnoff calculations for the effect of region exit temperature measurement errors (see Section 5).
Burnoff profiles for all 210 CSB coolant channels were predicted using the CSBBo-3 code (Ref. 36) and data from the data logger (DL) history tapes. Results of these analyses'show that the column .that experienced tr.e most burnoff was in region 27 and this column had used 21.9% of its useful life at' the middle of cycle 3 (MoC3). Details of these analyses are, reported'in Ref. 37.
These studies were then extended to the end of cycle 3 by estimating the oxidation using parametric rate. curves derived for region 27, i.e. , the region which experienced the most oxidation since the begLnning of cycle 1 7-5
.. ~. - .,_ - - - .- . . - - - .
p.
ll
,_ (Ref. 38). These results show that the cumulative oxidation for region 27 (the worst osse location) had consumed approximately 28% of the useful life of the CSB at the end of cycle 3.
L.-
Approximately 40% of the burnoff for region 27 occurred at helium gas f temperatures above 1400'F and moisture levels between 1 to 5 ppm. Also, the. rate of burnoff during cycle 3 increased over past cycles. This is probably due to increased plant operation at high reactor power, and therefore, high gas temperatures.
Based on the average CSB oxidation rate from beginning of life through the end of cycle 3, the worst case CSB has approximately 22 years of remaining life. Since the CSBs are 8 years old, the 30-year CSB lifetime estimate remains valid.
7.4. PGX GRAPHITE SURVEILLANCE A PGX graphite surveillance test element, inserted Dito the bottom :
reflector (in column 7 of; region 25) during the first refueling, was .;
L removed.at the end of cycle 2. Post-irradiation examination' (PIE) of this- i element'provided data on the oxidation of PGX graphite. Comparison of the -
measured oxidation to that predicted for the surveillance specimens then) provides ,a basis for assesstng the oxidation of the core support blocks.
The-surveillance specimens were exposed to reactor conditions similar to .
"that experienced by the PGX graphite core support blocks.
The modified CSBBo-3 computer code (Ref. 39) was used to predict the i
oxidation in use PGX surveillance specimen based upon actual reactor performance data. The predicted oxidation of the PGX specimens is reported in Ref. 40. Predictions of PGX oxidation were based on a PGX oxidation
~ rate which is 1000 times the H-451 graphite oxidation rate. Since only one PGX oxidation rate was assumed, the predicted oxidation varied only with temperature and moisture level.
7-6
5 g., 'l l
4 & i
, l I
1he PIE of the surveillance specimens .was performed according to l l
. specifications given in Ref. 41.
l The results of the PIE of the surveillance samples and the comparison-with predictions are given in Ref. 42. These results show that the PGX is- \
surveillance specimens exhibited significantly less oxidation than !
predicted. Overall oxidized weight' loss averaged only 20% of the predicted -l
?
values. Actual penetration depth of oxidation into the specimens was-only
- about 50% of. the predictions and the level of burnoff in the -oxidized volumes was considerably lower than predicted. . In all cases, the CSBBO-3 code predicted higher levels of oxidation and more severe oxidation profiles than observed in the surveillance specimens. l L
7.5. FUEL TEST ELEMENTS '(FTEs) l
! During the loading of segment 7 (reload 1), eight test elements. FTE-1 i l-through FTE-8, were loaded into layer six of regions 25, 22, 30, 27, 24, 10, 5, and 5, respectively.
These FTEs were manufactured from near-isotropic H-451 graphite instead of the needle-coke H-327 graphite used in the reference elements.
I
. This results in structurally stronger fuel elements 'with better heat transfer and dimensional change characteristics than- those 'of 'the reference elements.
4 V ,
1
-Reference fresh and recycle fuel for large HTGRs, as well as improved
- .. , fuel contemplated for use .in future FSV reload segments, was included in the FTEs. . Improvements in the fuel include (1) Th0 fertile kernels, (2) HEU UC, and weak acid resin (WAR) fissile kernels, (3) cure-in-place fuel rod process, (4) a saml1 number of fertile particles with BISO coatings, and (5) a small number of fuel rods containing LEU (Th,U) On TRISO medium enriched uranium (MEU) fuel (19.55 enriched).
Details of these eight FTEs are given in Ref.10. ,
I l ,
7-7 l
l
/
L. j p
L ,
The FTEm have- a -very small effect on the operation of. the' core under all conditions. The eight FTEs represent only a small percentage of the core (0.545) and only six of the elements contain new fuel types, which represents only 'O.40% of the total fuel.- It was shown in Section 4.5.6 that these -FTEs have an insignificant effect on the axial power i- distribution.
FTE-2 was removed from core region 22 at the end of cycle 3. - An.
extensive post-irraciation examination (PIE) program was planned.for this FTE (Amendment 2 of Ref.10). However, with the exception of the graphite block metrology, this. PIE program has not yet been performed. -A reduced PIE program of.FTE-2 fuel samples is: now planned (funded by PSC). The results of the metrology examination indicate that' PTE-2 underwent very little dimensional change as a result of irradiation. The visual examination. provided observable and concrete evidence of the H-451 graphite q structural' integrity and performance. There were small scratches but .no j significant structural damage. Details of the FTE-2 examinations are reportod- in' Re f. 51.
)
'7.6. POST-IRRADIATION EX AMINATION OF FUEL AND REFLECTOR ELDIENTS -)
1 7.6.1. Sesnent ' 2 Elements .
Fifty-four fuel and reflector elements from core segment 2, removed.
t from the core during the second refueling, underwent nondestructive metrological and visual inspections in April 1982 in. the hot service facilities at the reactor site. These inspeccions were part of the FSV .
' Fuel Surveillance Program sponsored by the Department of Energy (DOE). The results of these inspections are reported in Ref. 43 w
- Manufacturer's serial number.
- r. 7-8
.. ~ .. . - - . - - - . ,. _ - - -.
1: 1 i;
During these inspections, two of the standard fuel elements, serial numbers (S/N) 1-2415' and 1-0172a, were discovered to have similarly.
aligned hairline' cracks extending the entire axial length 'in the face !
adjacent to the large single dowel. The cracked web in fuel element 1-2415 was discovered during the on-site surveillance, whereas the crack in~ i element 1-0172 was discovered via viewing video tapes and photographs from the surveillance and was later confirmed by visual examination performed in 1 the GA hot cell (Ref. 44). Preirradiation inspection reports indicated that neither element was cracked prior to insertion into the core, and there was no record. of- any damage during handling. In addition, the fact that the cracks were colinear suggests that they developed during irradiation.
Both elements 'were manufactured from graphite grade H-327 and contained fuel rods consisting of (Th,U)C TRIS 0 fissile particles and ThC TRISO fertile particles bonded together ey a carbonaceous matrix. The nominal preirradiation dimensions of the fuel rods were 12.5 mm (0.49 in.) I in diameter and 49.3 mm (1.94 in.) in length. l
~1 Both elements were irradiated in core refueling region 8, column 5.
. Element 1-2415: was located in axial layer 6 (active core layer-3): and-element 1-0172 was located in axial layer 7 (active core-layer 4). These
~
elements, placed in the core in1 January 1974, were irradiated during i scycles 1 and 2. During this time the average core burnup was 363 EFPDs.
Element 1-2415 was selected for destructive post-irradiation examination (PIE) because it exhibited the larger crack and experienced :
t Llarger fast fluence and metrological changes than element 1-0172. The purpose of the destructive examination was to provide experimental- data on the irradiated element, to verify its performance, and to acquire in-pile ,
performance data. Tensile strength, thermal expansivity, and thermal' diffusivity samples were taken from the irradiated graphite block. The fuel rod graphite interaction was determined by measuring fuel stack pushout force. Gamma scans of a central fuel stack in each face were taken 7-9
.3:
a l
so that. power and burnup could be characterized.: 'These data were analyzed
, and made available for other scoping analyses that were done to develop possible scenarios leading to the cracked web; A section of graphite from the emptied element was cleaned to do a qualitative study on the feasibility of cleaning spent graphite blocks for burial as low level waste. The results of the destructive examination of fuel element 1-2415 are reported in Ref. 45.
Extensive analytical investigations were made to evaluate the
-p consequences of the cracked' elements (Refs. 46, 47, and 48). The consequences of hypothetical fuel element damage causing coolant flow obstruction *.:::< reviewed with the NRC in response to questions raised in 1978 concerning core temperature fluctuations. The discussions; also included the ' potential for misalignment of fuel elements in the event of dowel failure and the resulting effect on insertability of control rods and reserve shutdown balls.
A conservative assessment of the consequences of fuel element damage
. in terms of fuel particle failure, fission product release, and reactivity 1
. control was performed. -This assessment is applicable. in thatt it envelopes the' consequences in the unlikely event thatilimited cracks of, the type discovered were to propagate through a block. ~
.l l
, Based on the above assessment and,the NRC licensing- review, it was-concluded that there was no safety or operational limitation on continued reactor operation and/or fuel handling (Ref. 49).
7.6.2. sessent 3 Elements 1 Sixty-two fuel and reflector elements from core segment 3 were nondestructively examined in the FSV hot service facility (HSF) during June 1984. - These examinations were performed as part of the DOE-sponsored' 1
surveillance program. The time- and volume-averaged graphite irradiation )
temperatures and volume-averaged fast neutron fluences for the elements (
7-10 4
.if.
i f' ;
x 1: 1 ranged from approximately 380'C to 670*C and from 0.96 to 2.82 x 108' n/m8 , ,
respectively. The examinations were intended to verify the structural integrity and dimensional stability of the elements and to obtain dimen-sional change data and gamma-dose rates to verify HTOR design code J/? ' calc ulat io ns. .
The axial and radial dimensions of nearly all the inspected H-327
~
graphite fuel e.'eaants shrank as. a result of irradiation, but the dimen-sional changes ' sere relatively small. The maximum element-average axial (62/2) and radial (4X/X) strains were -0.735 and -0.575, respectively. The -i maximut bow was 0.69 mm. The-examined reflector elements underwent very little dimensional. expansion (less than 10.48 mm). The measured strains in segment 3 elements were reasonably close to or greater than the calculated dimensional changes. The measured bow was consistently higher' than the calculated bca. l The visu'ai examinations of the elements showed the elements to be in good condition. No . cracks were observed. There was little evidence of-graphite oxidation. The evidence of mechanical interactions (rub marks) was no more severe than in the-two previous surveillances which suggested that the core movement was very minimal or had stopped. Minor dowel pin -
7 and dowel-socket damage was observed on two elements, but the damaged areas did not affect = the handling, storage, or the performance of the elements. y A few chips, ~ nicks, and other handling scratches were noted, but these , ;j i
blemishes did not affect the performance of the core components. ,
b Strain calculations were performed with the HTGR code SURVEY / STRESS :
- (Ref 50) and the 'results compared with measured strains. The results of the comparison showed= the calculated strains to be underpredicted in almost .
every case. . The differences between the calculated and measured axial
. strains were, however, ses11, i. e. , 50.45.
?
Details of these segment 3 elements examinations are reported in Ref. 51.
7-11
t 7 . 7 '. GAMMA ' SCANNING-OF FUEL ELEMETS !
+
The gamma scan rocot' was used to measure fission product isotopic distributions in ' fuel elements removed from the core at the end of cycle 3. ;
Fourteen regular fuel elements and'3 oontrol rod elements were samma .
scanned to' measure axial and radial profiles of selected fission products.
Cesium-137 activity profiles were in excellent agreement with calculated time-averaged power profiles. Similarly, Zr-95 activity =
L profiles agreed with end-of-life power factors. Calculated and measured 1 l
- profiles agree statistically. if a 155 error in. the calculated value is assumed.
1
- Absolute Cs-137 activity was used to measure fuel element burnup. l
' ( F IM A) . The peasured burnup in 12 of the 14 regular fuel elements agreed with the calculated burnup. In the two regular elements and the three- l centrol elements which did. not agree, the measured burnup was significantly larger than the calculated value. Although the reasons are unclear, it appears that the measured value is In error since there is no recson for I the particular elements to have a. high4F burnup than the other elements in the region..
These data are described in detail in Ref. 73 7.8. SCRATCHER PLMUM ELEMENTS In 1979, as part of the fluctuation investigation, e pair of special 1
metal-plenum elements were installed on adjacent columns of regions 18 and
, 3r the active core. One.of these two elements was equipped with a l "4 " and one with a " stylus. " The stylus was designed to reach gap between_ the two regions and bear upon the surface of the .)
6 Ad so that any relative motion between the two refueling regions .
wou.. _ recorded by scratching a pattern on the pad. (See Ref. 9 for a j l
detailed description of the scratcher plenum elements.) j i
l i
7-12
)
i i
- These special elements were removed during the third refueling. LA {
visual examination of the scratch pad revealed a tear-shaped area of matte appearance caused by relative motion between the stylus and pad. This area -
I
- was about 0.25 in. long and 0.06 in, wide at its maximuw. Relative' motion 1 a
of the order of 0.25 in. is consistent with the relative motior, between i regions postulated to explain fluctuatiors (Ref. 52). The scratched pattern on the scratch pad constitutes the only direct measurement of l
relative motion between components of the core (i.e. , adjacent columns )
and/or refueling regions). - A video tape recording of the scratched e appearance was made as part of the metrology examination.
1 7.9. TEMPER ARJRE : FLUCTU ATIONS l
,7.9.1. Ba ckground i
During the initial rise-to power program in October 1977, at -605 power, temperature- fluctuations were observed. in the primary coolant
- circuit.at -individual core region outlets and at the steam generator module inlets. This, in turn, led to1 fluctuations in the main steam and reheat l steam temperatures.
4 Fluctuations were observed during cycle 1 operation under a variety of.
core conditions atL power levels from -30% to -705. The effects of tne fluctuations were observed in the nuclear channel signals,: the region exit heliue temperatures, and the steam generator module helium inlet tempera-turer. - Data obtained during both planned fluctuation testing and fluctua-tions which occurred as a result of plant upsets or spontaneote events-which occurred during normal power operation led to the identification of-core pressure drop as an important operating parameter relative to the fluctuation threshold.' It was demonstrated that maintaining a low core i
'The threshold line was defined as that line above which fluctuations had been ~ initiated and below which fluctuations had not been initiated.
I 7-13
. - .- - - - - - . . ~ _ . - - .- -_ _ _ - -
tr q
s pressure' drop. ?otained by opening the variable orifice valves) raised the
' threshold of fluctuations' to at least 705 usermal power .
. During the ' course of the cycle 1 investigation of the fluctuations-there were many postulated theories for explaining the observed behavior.
The most likely explanation of the temperature and neutron flux fluctua-tions was sen11 movements of reactor components, such as fuel elements,
' reflector elements, and/or the core support floor. These small movements
- change the distribution of the gaps between the columns of graphite blocks which, in turn, cause changes in the bypass flow distribution in these .
gaps. This motion is most likely induced by pressure gradients in the gaps and- thermal gradients in the core components causing component deformation and bowing. The movements are sustained by the feedback coupling of these two phenomena. Details - of these cycle 1 tests and analyses are given in Ref. 53 During the first refueling in early 1979 a comprehensive in-core inspection program indicated that no damage to the core components had I occurred as a result of the fluctuations.
. l Cycle 2 operation began in May 1979 and the fluctuation testing l l
program was continued. Fluctuations were again initiated at power levels 1 H
. from approximately 405 to 705. Data from these testa demonstrated that - '
. while the fluctuations observed during cycle 2 were somewhat more regular and widespread throughout the core, they were basically the same phenomenon t-
\
as: ex perienced during . cycle 1. The cycle 2 fluctuations indicated a I similar dependence on core pressure drop to that of cycle 1, but the
- fluctuations occurred below the cycle 1 thr eshold. Details of the cycle 2 g
fluctuation testing and analyses, before . installation of the region 4
constraint devices (RCDs), ara given in Ref. 52. j i
l i
7-14 t
a In October-1979 the plant was shut down and region constraint devices -
(RCDs) were installed as a means of eliminating,the temperature fluctua-t io ns. These RCDs provide inter-region keying and, therefore..~ preclude .the accumulation of large gaps. (See Ref. 9 for a description of the RCDs.)
In late 1980 tests were conducted to evaluate the effect of RCD installation on the fluctuation threshold. No fluctuations were detected. l s
throughout these tests, at power levels up to approximately 705 and at core pressure drops up to approximately 4.2 pai, even though most of the testing was performed at conditions above the previously established fluctuation threshold line. However, during a power increase to approximately 595, a resistance # of approximately 56 and a core pressure drop of approximately .
3.4 pai, a redistribution in region exit temperatures was observed. In general, this redistribution resulted in the interior region exit tempera-tures increasing more than expected and the boundary region exit tempera -
l tures either not increasing as much as expected, or decreasing. Evidence i
of the temperature redistribution was also observed in the gap thermocouple (T/C) and. instrumented control rod drive (ICRD) data, and in the nuclear detector response. This temperature redistribution was shown to be repeatable and. in each case the temperatures returned to their expected ,
values as' the core power and pressure ' drop were reduced.
Analyses of diese data indicated that the observed region exit temperature redistributions were the result of a redistribution of gaps (i.e. , small block displacement) . These displacements were similar in .
0
' nature to the initial motion which occurred during fluctuatic .s but the displacements showed no cyclic behavior.
6
' Resistance (R) = 2 x 10 " ( A pP )/(Tm * * * ) 8 where Ap = measured core pressure drop (paid) f P = PCRV pressure (paia)
T = average circulator inlet temperature (*R) m = total circulator flow rate (1bm/h) 7-15
l As a result of the fluctuation testing after installation of the RCDs, j the Nucisar Regulatory Commission (NRC) issued a release to test above 705 j power. These tests were performed in early 1981, during cycle 2 operation, at power levels up to approximately 905, using the procedures of RT-500K' (Ref. 26). Basically, this test procedure specified 35 power increases (at i J
-35/ min) from 405 to 1005 power. Throughout the testing periods, special restrictions on the region exit temperatures were specified. In addition to the temperature limits of Technical Specification 140 4.1 7, RT-500K required (1) a margin on the allowable region exit temperature misantch",
(2) regions with temperature measurement errors be operated by comparison )
region, and (3) adjusted margins for operation after a temperature redistribution. Details of RT-500K and of these region exit temperature restrictions are discussed in f.ef. 27.
7emperature redistributions again were observed during these tests, but no temperature fluctuations were evident.
P 7.9.2. Cycle 3 Testina Cycle 3 operation began in July 1981 and RT-500K was again performed in late'1981. During these tests, no fluctuations were detected at power levels up to 100% of rated power but, again, regios. exit temperature i redistributions were observed. Deta11a of the testing and analyses for
,. cycles 2 and 3, after installation of the RCDs, are given in Refs. 27 and 54
'RT-500 ws originally performed in November 1978, during cycle 1 operation, as a part of the fluctuation test program. It has undergone numerous revisions as a result of test experience and for compatibiJity
' with cycle 2 operation.
- Temperature mismatch is defined as the deviation of the region exit temperature from the core average e tit temperature.
7-16 i
D .2 -- ,_ -- . _ _ _ _ _ _ _ _ _ _ _ _ _ _ . _ _ _ . _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ . _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
L Evaluations of the region exit temperature redistribution data indicated that these events involved no unreviewed safety questions. As a result of these tests, a method was developed for operating the core which accounts for region exit t6wperature measurement discrepancies both before and after a redistribution. Under this operating procedure, the sevon northwest (NW) boundary regions which are susceptible to exit temperature measurement errors, i.e., regions 20 and 32 through 37, are operated by comparison regions in a manner similar to that' employed in the test-procedure. For the other 30 regions in the core, indicated changes in the region exit temperature which occur during a temperature redistribution are assumed to be correct. These temperature changes can be accommodated and corrected as desired by routine orifice valve adjustments. In support of this operating procedure, revisions to the Technical Specifications were developed (see Section 5.1 and Ref. 28). These revised Technical Specifications (LCO 4.1.7 ) were approved by the NRC in September 1982.
Af ter that time, the core was operated per these revised procedures.
Cycle 3 fluctuation tr , ting demonstrated that RCDs were successful in preventing fluctuations at power levels up to 1005 and at oore pressure drops up to 5.0 paid.
7.10. RADIATION EXPOSURE HIS1 DRY The radiation exposure history for FSV is compared to that from light .*ater reactors (LWRs) in Table 7-1. The LWR data were obtained from Refs. 55. 36, and 57. These data show that the FSV radiktion exposure per j megawatt year has never exceeded 10% of the average of all United States LWRs .
4 l
7-17
l TABLE 7-1 COMPARIX)N OF RADIATION EXPOSURE AT COMMECIAL F0WE REACTORS 1978 1979 1980 1981 1982 1983 i
Aversas Collective Doses Per Resctor FSV HTOR 0.9 30 13 05 0.4 0.95 All U.S. LWRs 497. 597. 791. 773. 705. 753. ;
All U.S. PWRs 429. 516. 578. 652. 57; 592.
All U.S. BWRs 604 733. 1136. 980. 940. 1056. !
Aversas Gross Electric Generstion Per Reset;or (MW-Years )
FSV HTGR - 76. 28. 83 94 73. 94 l All U.S. LWRs 494 447. 429. 449. 443.
All U.S.' PWRs 509. 434 435. 467. 578.
All U.S. BWRs 471.- 467. 418. 419. 410.
Aversae Man-Res Per MW-Yr FSV HTGR' O.012 0.11 0.016 0.006 0. 005 0.010 All U.S. LWRs 1.0 1.3 1.8 1.7 1.6 1.7 All U.S. PWRs 0.8 1.2 .3 1.4 1.3 1.3 All U.S. BWRs 1.3 1.6 2.7 23 2.3 2.8 t
s 7-18
, i
! i
- 8. UNUSUAL OCCURRENCES ,
l 1
8.1. INADVERTENT INSERTION OF RESERVE SHUTDOWN SYSTEM (RSS) ttATERIAL INTO ,
RE0 ION 27 j l
Surveillance of core performance data from June and July 1982 )
1 indicated that the reserve shutdown system (RSS) material may have been j inadvertantly released into region 27. The first evidence of the RSS .)
material in region 27 came from examination of the region exit temperature l and RPF distributions after the control rod was withdrawn from region 27 (i.e., control rod group 4B). It was noted that both the exit temperature and power in the north regions of the core was considerably higher than expected and'the power in the south regions of the core was considerably lower than expected, with the largest effect in region 27.
Further examination of the measured data indicated:
o The orifice in region 27 was the most fully closed orifice in the core and the exit temperature was about the same as when the orifice was normally much more open, o Steam generator module B-2-1 inlet temperature was lower than expected, and this module is fed by region 27. '
o Fuel accountability analyses indicated a enange in reactivity disertpancy of about 0.002 ak. (This was subsequently determined to be the approximate reactivity worth of the RSS material in unrodded region 27.)
8-1
a .- . .
Subsequent RSS rupture disk pressure testing by PSC confirmed that, indeed, the RSS material had been inadvertently released into region 27.
A preliminary assessment of the consequences of continued short-term (4 to 6 wee:(s) operation with the RSS material in region 27 identified ne immediate concerns. However, since it was about seven months before the next scheduled shutdown, a second assessment was made to evaluate the effects of operation for that period of time with the RSS in region 27.
This evaluation included the followings b
o Potential for boron migration out of RSS material, o Potential for swelling of RSS material, o Potential for mechanical stresses due to RSS material behaving as rig 1d column.
o Potential for agglomeration (due to formation of boric acid crystals) of RSS satorial.
o Effects of a moisture ingress incident on RSS material and impact on core performance.
o Evaluation of stresses in region 27 due to perturbed tempera'.ure distribution, swelling of RSS or mechanical binding (ratche',ing).
o Effect on core reactivity, o Long-term effect on power distributions and compliance with LCO t.1 3.
The results of these analyses indicated that it was acceptable to continue plant operation at approximately 705 power with the RSS material a
8-2
};
- l in region 27 for up to about 7 months. This conclusion was based on the assumption that the reactor was operated with low moisture levels.
While it was concluded that continued operation with the RSS material in region 27 was acceptable, one constraint was identified. This con-straint resulted from the effects of the perturbed power distribution on the maximum worth control rod pair (region 11). The worth of the maximum worth rod exceeded that calculated without RSS material in region 27. The 90nsequences of this increased control rod worth on the rod withdrawal accident were assessed and it was found that there were certain control rod configurations (positions) which could result in potential reactivity i?tertions greater than those analyzed in the design basis rod withdrawal accident. Therefore, limits on operating conditions were established to avoid the the potential for a reactivity insertion exceeding that analyzed in the design basis rod withdrawal accident. PSC operated the plant in accordance with these operating conditions until the plant was shut down on September 30,'1982, and the RSS material removed from region 27.
Details of these analyses are reported in Ref. 58 t 8.2. CRACKED FUEL ELEMENT WEB The cracked coolant hole web in two segment 8 standard fuel elements, detected during the post-irradiation examination of segment 2 elements, is discussed in Section 7.6.
8 3. POWER LEVEL UNCERTAINTY Surveillance of core performance data during cycle 3 indicated that the reported thermal power was lower than the heat balance power. Further ^
-investigation, in cooperation with PSC, revealed that periodically during the cycle the core was inadvertently operated at higher than reported thermal power. Investigations and analyses by PSC (Ref. 61) and GA 8-3
a concluded that the core overburn was the equivalent of about 7.5 EFPDs.
These 7.5 EFPDs were subsequently incorpo.'ated into core performance analysis, i
'}
a l
l l
8-4
i
)
i
- 9. REFERENCES 1
- 1. FSV Fuels Engineering Staff, " Fort St Vrain Cycle 3 Core Performance !
Quarterly Report for Period Ending September 30,1981," 0A-C15822, I September 1981.
- 2. FSV Fuels Engineering Staff, " Fort St. Vrain Cycle 3 Core Performance j Quarterly Report for Period Ending December 31, 1981," GA-C15822, l December 1981. l i
- 3. FSV Fuels Engineering Staff, " Fort St. Vrain Cycle 3 Core Performance j Quarterly Report for Period Ending June 30,19A2," GA-C15822, l June 1982. j 4 FSV Fuels Engineering Staff. " Fort St. Vrain Cycle 3 Core Performance Quarterly Report for Period Ending September 30, 1982." GA-C15822, ,
September 1982.
- 5. FSV Fuels Engineering Staff " Fort St. Vrain Cycle 3 Core Performance Quarterly Report for Period Ending March 31,1983," GA-C15822, March 1983
- 6. FSV Fuels Engineering Staff, " Fort St. Vrain Cycle 3 Core Performance Quarterly Report for Period Ending September 30, 1983," GA-C15822 September 1983.
- 7. FSV Fuels Ergineering Staff " Fort St. Vrain Cycle 3 Core Performance l Quarterly Report for Period Ending December 31, 1983," GA-C15822.
December 1983.
- 8. FSV Fuels- Engineering Staff, " Fort St. Vrain Cycle 3 Core P6rformance Quarterly Report for Period Ending March 31, 19 84. " GA-C15822, March 1984
- 9. Malek, G. J., et al. , " Investigations of the Fort St. Vrain Reactor Fluctuations " GA-C15369, September 1979.
- 10. "SAR for Fort St. Vrain Reload 1 Test Elements FTE-1 through FTE-8,"
GLP-5494, June 1977.
9-1
J7 T '!
- 11. Malakhof, V., et al. , " Segment 8 Design Support Document - Fort St. Vrain HTOR," GA-D16304, March 1961.
- 12. Malakhof, V., " Fuel Fabrication Acceptance Report - FSV Segment S."
GA-C16257, March 1981.
- 13. " Safety Analysis Report for Fuel Reload 2 - Cycle 3 " GA-015812, April 1980.
14 Wagner, M. R., " GAUGE, A Two-Dimensional Few Group Neutron Diffusion-Depletion Program for a Uniform Triangular Mesh," GA-8307, March 1968.
- 15. FSV Fuels Engineering Staff, " Fort St Vrain Cycle 2 Core Performance Quarterly Report for the Period Ending June 30, 1981," GA-C15822, June ;
30, 1981.
- 16. Hackney, R., "FSV Cycle 3 Data for Reactivity calculations and for l Physics Surveillance Tests," GA Doc. 906047, July 1, 1981.
- 17. FSV Fuels Engineering Staff, " Fort St Vrain Cycle 1 Core Performance,"
GA-C15560, September 1979. 4
.. t
- 18. Hackney, R., " Fort St. Vrain Cycle 2 Core Performance," GA-C16743, ;
April 1982.
- 19. Hoppes, D. F., et al. , "The Ef fect of Nuclear Detector Deca 11bration on the Fort St. Vrain Reactor and Suggested Corrective Measures,"
GA-A13954, January 1978.
- 20. Kraetsch, H., and M. R. Wagner, "GATT - A Three-Dimensional Few Group Neutron Diffusion Theory Program for Hexagonal-Z Mesn," GA-8547, ;
I January 1969.
21 Malakhof. V., "FSV Cycle 3 Power History," GA Doc. 907731, November 19, 1984
- 22. Malakhof. V., and W. Lefler, " Fuel Accountability at the End of
- l. Cycle 3," GA Doc. 907402, May 1984.
- 23. "End of Cycle 3 Fuel Accountability," Magnetic Tape Archive No.
RPSD 3842, May 1984.
l l 24 " Control Rod Management," EE-12-002, Issue A, December 1981. '
- 25. Warembourg, D. W. (PSC) letter to G. Kuzzycz (NRC), " Fort St. Vrain Unit No. 1 Meeting, February 16,1982 - RT-500K Testing," P-82036, February 9, 1982.
- 26. " Fluctuation Test Procedure, RT-500 Revision K," March 1981.
9-2 4
I i
l
- 27. Hackney, R., " Fort St Vrain Testing and Analyses in Support of Long Term Full Power Operation," CA-C16937, to be issued.
- 28. Alberstein, D., and K. Asmussen, " Technical Specifications for )
Operation of FSV with Region Outlet Temperature Measurement ;
Discrepancies," GA-C16781, June 1982. ']
- 29. "FSVCOR Input Instructions," GA Memo CNE 209: SM: 80, November 10, 1980. j
- 30. Burnette, R., " Test Status Report On Fort St. Vrain HTGR Coolant and Radiochemistry: JJ1y 1981 through June 1982," GA Doc. 906601, l August 1982.
- 31. Burnette, R., 9 Test Status Report on Fort St. Vrain HTGR Coolant and Radiochemistry: July 1982 through June 1983," GA Doc. 907053, August 1'sS3
- 32. Burnette, R., " Test Status Report on Fort St. Vrain HTGR Coolant and Radicchemistry: July 1983 through June 1984, GA Doc 907650, j September 1984 33 Burnette, R. D., " Radiochemical Analysis of the First Plateout Probe i from the' Fort St. Vrain HTGR." GA-A16764 June 1982.
34 Dunn, T. D., " Fort St. Vrain Core Support Block Oxidation Analyses for Plant operation from June 1976 to February 1979," GA-015571 September 1979. <
- 35. Hoot, C. G. , and S. C. Bachelor, " Fort St. Vrain Core Support Block r
oxidation Analyses for Plant Operation fra June 1976 through June 1980," GA-C16158, January 1981.
- 36. Dunn, T. D., et al., "CSBBO-3: A Computer Program for Analysis of FSV Core Support Block Oxidation Distribution - Program Description and Users Manual," GA-D15577, October 1977.
p 37. Boltinghouse, S. T., et al. , " Fort St. Vrain Core Support Block -
Analysis for Plant Operation from July 1980 to Decembe.c 1982," GA Doc.
907410, April 1984
- 38. Richards, M., "FSV CSB 0xidation Due to Water Ingress Transient from l 6/22/84 to 6/23/84 " GA Doc. 907622 September 1984
- 39. Dunn, T. D., " Modification of the CSBBO-3 Computer Code," GA Doc.
906172, August 1981.
i: 9-3
. 4 i
l l
- 40. Dunn, T., and J. Battaglia. " Prediction of Graphite Burnoff in FSV Surveillance Samples - Cycle 2." GA Doc. 906223, September 25, 1981.
- 41. " Test Specification for PIE of PGX Surveillance Specimens j (Region 25),' GA Doc. 906223/A, September 25, 1981.
- 42. Beanan, L. A., "Results of PGX Graphite Surveillance Fort St., Vrain !
Region 25," GA Doc. 906624, October 21, 1982. j 43, Saurwein, J. J., " Nondestructive Examination of 54 Fuel and Reflector I Clements from Fort St. Vrain Core Segment 2," GA-A16829, Octetter 1982.
44 Ketterer, J. W. , " Visual Examination Results of Segment 2 FSV Fuel Elements 1-2415, 1-0172, 2-2693, 1-0108 and 5-0801," GA Doc. 906577/2, April 1983
- 45. McCord, F., "Postirradiation Examination and Evaluation of FSV Fuel j Element'1-2415," GA Doc. 907079, November 7, 1984 j
- 46. Yu, H., et al. , " Seismic Strength of the FSV Cracked Block," GA Doc. I 907429, July 10, 1984
- 47. Ho, F., "FSV-Structural Test of H-327 Elements with Simulated Cracks," i GA Doc. 907428, April 24, 1984.
- 48. Kapernick, R., " Operating History of the FSV Cracked Olocks," GA Doc.
906811, February 25, 1983.
- 49. " Structural Integrity of HTGR Fuel Elements," GA Doc. PC-000045, June 30, 1982.
- 50. Smith, P. D., " SURVEY / STRESS A Model to Calculate Irradiation-Induced ;
Stresses Strains, and Deformations in an HTGR Fuel Block Using Visoelastic Beam Theory," GA-A13712, October 20, 1975.
- 51. McCord, F., " Nondestructive Examinations of 62 Fuel and Reflector Elements from Fort St. Vrain Core Segment 3." GA Doc. 907785.
January 24, 1985.
- 52. Hackney, R., and J. Saeger, " Investigations of the Fort St. Vrain Cycle 2 Reactor Fluctuations Through October 26, 1979," GA-C15767, 1-March 1980.
53 Hackney, R., and W. Breher, " Investigations of the Fort St Vrain Reactor Fluctuations - January-March 1979," GA-C15536, August 1979.
54 Asmussen, K. , et al. , " Core Testing and Operation of Fort St. Vrain Up to 1005 Power," GA-C16701, June 1982.
9-4
i i
- 55. NUREG-0173 Vol. 3 ,
- 56. Brooks, 8. G. , NRC. '!
- 57. Inside NRC, Vol. 18, September 1984
- 58. Asmussen, K. , " Effects on FSV Core of Operating. with Reserve Shutdown Material in Region 27, GA Doc. 906618. June 1983. l
- 59. Disselhorst, B. F., et al., " Fort St. Vrain Fuea Element Quality
.]
Control," GA-013772 December 31, 1975.
- 60. Lefier W.. " Support Analysis for FSV SR 5.1.7 - December 1983,"
GA Doc. 907240, February 17, 1984
- 61. November Operating Report for Period 11/1/83 thru 11/30/83. Docket No.59-267. December 12,.1983.
- 62. Hudritsch, W. W. , V. Jovanovic, and D. L. Georghlou, " SURVEY, A Computer Code for the Thermal and Fuel Performance Analysis of High-Temperature Gas-Cooled Reactor," GA-C17554, March 1984
- 63. Smith, P. D., "TRAFIC, A Computer Program for Calculating the Release ,
of Metallic Fisaaon Products from an HTGR Core," GA-A1h721 February 1978.
64 Jovanovic, V., "FSV Fuel Performance Analysis for the First Three Cycles," GA Doc. 907715 June 28, 1985.
- 65. Myers, B. F., " Fuel Design Data Manual," GA Doc. 901866/E, '
September 28, 1984 "
- 66. Goodin, D. T., "FSV Fuel Performance," GA Doc.18-R-59/3, August 1, 1984 ,
- 67. Ketterer, J., " Reference HTGR Fuel Compact Thermal Conductivity Data Base and Model," GA Doc. 907591 Issue N/C, August 29, 1985.
- 68. Wan, M., " Core Performance Calculations for the First Two Cycles of !
l the FSV Core," GA Doc. 906274/1, September 25, 1981.
l 69.- Wan, M., " Cesium and Strontium Release Calculations for the First Two . .
' Cycles of the FSV Core," Memo CNE 194:MW: 81, October 9, 1981.
- 70. Ketterer, J. W. , and K. E. Partain, " Capsules MB-17/~18 Preirra-diation Report," GA Doc. 906975/1, August 1983
- 71. Goodin, D. T., "FSV Fuel Performance," GA Doc.18-R-59. Issue N/C, May 13, 1985.
9-5
l i
I
- 72. Montgomery, F., " Test Report on Hydrolysis of HTOR Carbide Fuels:
Chemical Resctions and Consequences," CA Doc. 904929/1, September 26, ,;
1980.
73 Montgomery, F., " Gamma Scanning of TSV Core Segment 3 Fuel Elements,a CA Doc. 908043, July 8, 1985. l
(
l
.l .
O
[
9-6
.) & , ~ .-- , , . . .