ML19210A087
| ML19210A087 | |
| Person / Time | |
|---|---|
| Site: | Crane  |
| Issue date: | 04/10/1978 |
| From: | Herbein J METROPOLITAN EDISON CO. |
| To: | Reid R Office of Nuclear Reactor Regulation |
| References | |
| GQL-0658, GQL-658, NUDOCS 7910240808 | |
| Download: ML19210A087 (29) | |
Text
I k fHY fg of l
l w~
- CJ.:LAiORY INFi.'. fatAT ION O I ?TRIEUT un r: TEN : F I D':.)
DIS"IRIEUTION COR Ii4 COMING MATERI AL 5 0 -28
FE : REID R
OFO HERF IN J r, DOCDATE Oa '1
NRC NETROPOL EDISON DATE FCVD 04'!2'~~
DOCTYFE: LETTEF NOTARIZED: tJ COPIEE ECEIVE0 5 00 _!ECT :
LT ' i E:.C L 1 PE5FONS.'- TO NFC LTR DTD 04/07/'?
F0FWARDINO AN.5WERS TO THE CYCLE 1 CELOAO OUE?TIONS AND ADDL CONIERN5 AS RE01'ESTED BY NRC.
PLCNT NAME: THFEE MILE ISLAND - UNIT 1 REVIEWER IN1TIAL Y -r' DISTRIEUTOR INITIAL.01
-&*++*-&********-t**
DISTRIEUTION OF THI? MATERI A'.
Is A5 FOLLOWS 4 4 4 4 4 4 -5 + 5 + + + -5 + + +
- OEr.ERAL DICTRIEUTION FOR AFTER I??UANCE OF OPERATING LICEif5E.
(DISTRIGijTION COEE AOO1)
FOR ACTION:
ER CHIEF REIDt- -W/7 ENCL INTERNAL.
C c rQ_ wl: wCAND.
NRC FOR^*W./ ENCL I ', E *-* W /.. ENCL GELD-t->LTR ONLi HAN AUERK-*W/ ENC L CHECK **W< ENCL EISENHUT**lJ/ ENCL SHAO**U/ ENCL E AER t& t'W/ ENC L BUTLER ++W RNCL EEB+*W/ ENCL J COLLINS *:FW/ ENCL J.
MCO:_.UCH4mW/ ENC L EXTERNAL.
LPDP's HARRISEURG, PA**W/ ENCL TICpcW/ ENCL pD g [g NOICi'tW/ ENCL
$. '. i \\{\\ ; 7g ACPS CAT Oc*W/Ic. ENCL
\\
'g 0
'.A
-, (( \\ $ k
ps
)
3 1489 009 D I'9TP T ?f !T I ON :
L'R 30 ENCL 20 CONTFOL NER-
~ 210 2 ' 53 c t.a.*.r. + c. gS -= c..
u, ai 1" d'- / h 1 - t} =
- J [. '"= i' U 4 d' b
\\.
d.4. 5. Ji 4% J4 4s f /. b. by #.f {. gs JC JC P. ("e O 4 J%.* *
"'s n.:. ga J",e e*". n.*. g..
k' *f b U* h k ( h O (- *h h M N [a k J.'* kE **5 d'-
=
h&
s r
YN 12"3
7f';:ge,,
- w..,
n.a.
- m METHOPOLilAN EDISON COMPANY PCST OFFICE BOX 542 READING, PENNSYLVANI A 196c3 TELEPHONE 215 - 929 3601 April 10, 1978 GQL C658 Director of Nuclear Reactor Regulation v'
Attn:
R. W. Reid, Chief Operating Reactors Branch No. 4 U. S. Nuclear Regulatory Cc=sission Washington, D. C.
20555
Dear Sir:
., n
_, y r;, (
Three Mile Island Nuclear Station, Unit blJiil' II.,..
'os Opersting License No. DPR-50 fLS..
' ~
Docket No. 50-239 Attached please find the ansvers to the Cycle h Reload questions and additional concerns identified in your letter of April 7,1978.
Members of my staff remain available at your convenience to discuss any additional concerns that you may have in regard to this submittal.
Sincerely
/ f
/r.G!Herbein
/
/Vice President-Generation
/
JGH:RJS:c,jg Attachment
.g.,70053 p\\
1489 010
Cuestion No. 1 Describe the changes to the CVCS necessary to use the feed-bleed mode of operation.
Resnonse:
No changes to the TMI-l Makeup and Purification System vere recuired to support operation of the unit in the unrodded or feed-and-bleed mode beginning in Cycle h.
All B&W nuclear power plants are designed with the capability to conduct feed and bleed operations, independent of wh3ther the core is operated in the rodded or unrodded mode. The design letdown flow-rate for all B&W 177 FA units is the sa=e (1ho gpm). In the codded plants,
feed and/or bleed operations are necessary to cetpensate the following reactivity changes:
- excess reactivity required for fuel burnup and fission product buildup over the fuel cycle (depletion effects).
moderator temperature reactivity effects due to RCS coolant temperature changes at startup and shutdown.
- buildup of equilibrium xenen and sa=arium reactivity.
- beration to shutdown requirements specified by Technical Specificiations.
- deboration fran shutdown or refueling cencentration requirements during startup.
For operation in the unrodded mode, the required feed and bleed capabilities are the sane as stated above with the addition of adjusting the RCS boren concentration to maintain the regulating control rods within specified taneuvering control bands during power level changes or load follow. Both boration and deboration are accomplished manually to keep the ecntrol reds in a predescribed operating band within the red position limits of Technical Specifications, he maneuverability of the plant is then limited only by the ability of tue vaste precessing system to handle the vaste generated, as indicated in tha attached Table 1.
TMI-l has operated at end of cycles 1 and 3 in the "all rods cut" or feed and bleed mode of operation.
)kh9
Resyonse to Cuestion !!o.1 TABLE 1 INTERVAL BETWEE'T PUSH-P JLL LOAD CHA!TGES Last Time In Life Interval Load Change Last FFD P ce ed Days Percentare Cone. (PPM)
FFD Gallons 1
70 900 30 10 1h,500 2
To h50 186 60 29,000 3
70 320 220 71 h2,500 h
70 250 2hh 79 58,000 5
70 230 250 81 72,500 1
50 510 168 Sh 14,500 2
50 265 239 77 29,000 3
50 185 26h 85 h2,500 h
50 150 279 90 58,000 310 100 5"
w.
1489 012
t Questian No. 2 The Tech. Spec. changes presented in the April 3,1978 submittal appear to be based on cross-core shuffle of the fuel even though this refueling con-figuration is no longer being proposed. Describe in detai3 the effects of non-cross-core shuffle on the para =eters contained in the January 9,1978 sub=ittal. Revise or verify all tables presented in the January submittal to reflect the effect of the additional cycle 3 burnup and the non-cross-core shuffle.
Response
Met-Ed's submittal of April 3,1978, addressed both the non-cross core shuffle scheme and the extended cycle 3.
However, since the changes, due to the extended cycle 3, were more restrictive *,han the non-cross-core changes, the non-cross core changes were not indicated. For example, only a single setpoint required change as a result of the non-cross core shuffle scheme, i.e., the power i= balance negative limit at 102% power decreased from -30.80 (January 9,1978 sub=ittal) to -28.9h for operation frem o to 125 + 5 EFPD. The extended eyele 3 further decreased the power bdbalance negative Jimit at 102% power to -23.ho.
Therefore, the change submitted April 3,19'8, was due to the extended cycle 3 and not to the non-cross core shuffle scheme. With respect to the Reload Report, which v.11 be revised as indicated in the April submittal, t - only changes resulting from the non-cross core shuffle are,1) a revised Figure 3-1, Core Loading Dicgra= for TMI-1 Cycle h (submitted to NRC April 3, 1978); 2) a revised Figure 5-1 (attached); and 3) ravised calculated nuclear peaks e.
Q11ovs:
Margin to Radial-Local Peak Ref. Desien Criginal Cycle h submittal 1.637 (EOL) 8.2% (BOL)
(January 9, 1978) 1.h21 (EOL) 20.3%
(EOL)
Non-Cross Core Shuffle 1 596 (BOL) 10.55 (BOL) 1.h07 (E0L) 21.15 (EOL)
Non-Cross Core Shuffle and 1.547 (BOL) 13.2% (EOL) extended Cycle 3 (287.1 EFFD) (To be 1.h03 (EOL) 21.3% (EOL) included in revised Reload Report)
It should be noted that revised Figure 5-1, attached in respense to this question, is superseded by revised Figure 5-1 attached in response to Question No. 5 1489 013
-Devision 1 (2/28/78)
Resoonne to Questien uo. 2
' 1.
BOC (4 EFPD), Cycic 4 Two-Dimensional Relative Figu r -
Power Distribution -- Full Power, Equilibrium Xenon, APSRs Inserted 8
9 10 11 12 13 14 15 H
0.96 1.09 1.25 0.99 1.23 0.91 0.82 0.76 K
1.29 1.10 1.70 1.07 1.17 0.84
_0.76 8
L 1.36 1.06 0.92 1.14 0.66 M
1.00 1.16 0.94 0.96 N
1.02 1.11 0.68 0
0.54 P
R X
Inserted Rod Group Number X. Y.X Relative Power Density O
1489 014
Question No. 3:
The beginning of cycle (.EOC) boron concentration for cycle k reported in Table 1 of the April 3,1978 submittal is less than that in the FSAR.
Provide available operator response times for a boron dilution event occurring (1) during refueling, and (2) during startup, cold shut'ovn, hot standby, and pcver operation.
Response
The 30C boron concentration for Cycle h is less than that given in the FSAR.
This means the reactivity insertion rate due to a moderator dilution event at power is less for Cycle h than that given in the FSAR..iince the refueling boron concentration requirement vill remain the sa=e for Cycle h, the FSAR analysis for the margin to critical for a shutdown condition remains valid.
The conditions of startup, cold shutdevn and het standby were not addressed in the ?SAE and as such were not enaly::ed for Cycle h censistent with licensing by ecmparison to the FSAR.
1489 015
Questien Fo. k Tech. Spec. Change Request No. 75, dated March 1..,1978, is for a change to allev a h5 uncertainty between the excore measured power and the pcVer obtained by a plant heat balance. In view of the assumed 2% error in measured power required to be used in accident and transi
.t analyses, explain how the h5 uncertainty has been accounted for in the accident analyses and the protection system setpoints. If the additional uncertainty in power has not been acccunted for in the accident ana2yses, provide new analyses, including ECCS, which properly reflect the additional 2% uncertainty.
Resconse:
The 25 heat balance error assumed in the Safety Analysis is retained in the setting cf the Tech. Spec setpoints. The h5 neutron power measurement error is also acccunted for in the Safety Analysis. The following break devn of assumed errors is presented to further clarify this issue:
112% Safety Analysis Setpoint
- 2% Heat Balance Error
- h% Neutron Power Measurement Error
.55 Eistable Setting Error LOS.5% Tech. Spec. Setpoint Value Tech. Spec. Change Request No. 75 was submitted to account for the full h5 neutron pcue! =easurement error accounted for in the Safety Analysis. Of this h5 error, 2% is for steady-state measurement and 2% for transient effects.
Part of Change Request 75 was a daily check cf the pcver measurement, requiring a heat balance calibration whenever the heat balance exceeds indicated neutron pover by more than 25.
In effect, this change request limits the nlant to 25 steady-state neutron pcver error, with margin to a total h5 errst i=nediately following maneuvering transients.
Questien No. 5 Provide an updated pcVer map which reflects the additicnal cycle 3 burnup and the non-cross-core shuffle for cycle k.
Restonse:
See attached Figure 5-1.
1489 016
Reccense to Question : o.
4/10/73-FICURE 5-1.
Doc (4 EPPD), CYCLE 4 TFO-DIMEN3IONAL RELATIVE POWER DISTRIDUTION-rUI.L POWER, EQUILIDRIUM XENON, APSRS INSERTED B
9 10 11 12 13 14 15 i
0.94 1.05 1.23 0.97 1.21 0.91 0.83 0.79.
H, I
1 1.24 1.06 1.16 1.05 1.18-
,0.85 0;79 '
~!
a
[
1 i
t_:!
1.32 1.05 1.05
- 0.93 1.18 0.69 i
i l
[
I i
i i
I
,I ;
0.99
?. 16 0.95 0.99
=
5' l
I, a; i:
st:
F 1.03
'1.13 0.70-i h
h l
l l
3 l
~
o[
i
'c.55-t; 1
I
\\
t l
2 Pj J
+-
j j
i g
l Updated power map reflecting q* :
additional Cycle 3 turnup
~
(287.1 ETPD) and non-cross-l core shuffle.
O b
e
[ REI *CIVE POh'ER DENSITY jX.XX
^
(16 Ouestion No. 6 Provide or reference the bounding transient and accident analyses during bleed and feed operation.
Restonse:
The parameters having tie greatest effect on the Safety Analysis are the core-thermal parameters, thermal hydraulic parameters, and kinectic (including feedback coefficients) parameters. As shown in Tables h-2, 6-1, and T-1 and discussed in Section 7 of the Reload Report, the FSAR Safety Analysis is still a bounding analysis.
T'.e one exception is a slightly less initial boron concentration and the effects of that are discussed in the response to Question 3.
1489 018
Question No. 7:
Provide an explanation of the increase in quadrant tilt from 3.kl to 4.92%
being proposed in the Technical Specifications. What kind of a penalty is taken in the calculation of peaking factors in order to account for the allow-able h.92% tilt? Provide the basis for the adequacy of this penalty.
,F espons e :
As indicated in Item 2 of Section 8 of the TMI-1, cycle h Reload Report, the quadrant tilt limit for cycle h was returned to the original limit value of h.92% actual core tilt used in cycles 1 and 2.
The reason for the tighter limit, 3.kl% in cycle 3 was that in order to preserve flexible operating vindcvs for i= balance and control red position, a smaller peaking penalty (5.1%) for allowable quadrant tilt was used to offset the required peaking penalty due to potential fuel rod bov. Thus, the allowable tilt limit was correspendingly reduced. The TMI-1, Cycle 3 Reload Report discussed these items in Section 8.
For Cycle h, a trade-off of this type was not necessary due to the use of a statistical combination of peaking fact 5rs (Section 8, Item 3), the removal of the densification power spike from ccusideration in setting ECCS-dependent Technical Specification limits (Section 8, Item h), and the reduced peaking behavior of the Cycle h core design. Thus, the original h.92%
limit on quadrar.t tilt and its associated peaking penalty (7.36% or a peaking factor of 1.0736) were reinstated.
The peaking factors quoted in the preceding paragraphs were derived frem the relationship established for the increase in the peak power due to a quadrant pcVer tilt. The following discussion describes the calculations which have been performed to investigate this behavior.
The data on calculated power peak increase due to quadrant power tilt are presented in the attached figure. These data are from both Figure 3-5 of 3AV-10076 and recent investigations of the Oconee I, Cycle h tilt behavior.
The following discussion characterizes the method of tilt inducement used in the various calculations.
The calculations were performed in both 2-D and 3-D full core geometry using the PDQOT and FLAME 3 computer codes. Two dimensional geometry was used whenever the tilt effects were unifor= axially. In the'e cases the radial peak change conservatively reflected the total peak change. This fact was confirmed by selected 3-D check cases. The value of tilt against which the peak increase was plotted was obtained by integrating the mesh block or nodal pcVers to get the power produced in ^ach quadrant. The expressicn for tilt is:
0"EdT*"t peyer
)
-1 x 100,
% Quadrant Tilt = (--.Average Quadrant Power and for the attached figure represents what can be called the " actual" quadrant tilt.
1489 C19
t Fesnense to Question No. 7 centinued:
Folleving the legend in the attached figure, the first tilt type considered was that due to multiple rods out of sequence (symbol x). Two of these values are frcm Figure 3-5 of BAV-10078, and one from recent 3-D FLAIE inveatigations of potential Oconee I =ultiple misaligned rods. These three represent from 2-6 rods misaligned. In the Ocenee I case, rods in m
ciegoually opposite quadrants were moved in opposite directions. The core was modeled with 2h axial nodes of 6" each. Bank 7 was =isaligned such that one rod (~en a minor axis) was one node above the bank average and the diagonally opposite rod was one node below the bank average.
The r-xt type of tilt, shown with the symbola, was that caused by a dropped rod.
In addition to the four cases from Figure 3-5 of B/W-10078, eleven additional cases were calculated for the Oconee I, Cycle h.
Every potential dropped red location, including those on the major axes, was lavestigated.
Tne third tilt type was that caused by a single red out of sequence (sy=boltD).
These ten cases were a ' reported in BAW-10078. The results are all clustered at 1cv tilt and peak. increase values. These were 3-D PDC.07 cases.
he fourth tilt efpe shown (sy=bol O) was that due to various numbers of individual APSR fingers (1-3) assumed to be broken off and resting on the bottom in three different asse=bly locations. Three-dimensional FLAME calculations for the beginning of Oconee I, Cycle h were rin at LO5 FP, and without xenon, to a=plify peaking effects.
~ne fifth tilt type was generated assuming several (3-6) misleaded asse=blies (symbole). Enrichment deviations of from +.01 v/o (6 locations) to
.90 v/o (3 locations) were investigated. Again, the beginning of Cycle h of Ocence I was the configuration analyzed.
"he sixth and final tilt ' type investigated (sy=bol 0) was that caused by a non-sy==ec aic burnup distribution in two fuel batches being carried over into Cycle 4 of Oconee I.
Partial results of these calculations a e given in EAW-lh77 FLAME was used to si= alate an end of Cycle 3 burnup asynnetrf of. +2%
in one core quadrant and -25 in the diagonally opposite quadrant. The fuel was then shuffled into the Cycle h pattern and depleted in full core sec=etr/
- c 50 EFPD. The pcver level was set at h0", FP to 4 EFFD, at 75% FP frem h to 23 EFPD, and a 1005 FP frca 23 to 50 EFFD. A total of 26 variations of power level and burnup supplied data for the points plotted.
As can be observed from the figure, all of the over 60 data points fall below the line which has a slope of 1.h95 This was the value assumed in assigning a T.365 peak increase to an allovable tilt of h.92% for the TMI-1, Cycle h Technical Specificatien.
1489 020
..... f.
X MULTIPLE RODS OUT OF SEQUENCE 6
DROPPED R00 24 8
SINGLE ROD OUT OF SEQUENCE
's 5
O BROKEN APSR FINGERS ON BOTTOM iD MISLOADED ASSEMBLIES O
k
. !O BURNUP GRADIENT 20 n
uO 4
o:
$ 16 i
o c.
X y
a E
g@
g N-
= 12 g
l AO Oh i
~
a a4 Ai g
i
=
l i
<t 2
58 i
g O
ao e
X b
O O
a O
O O
O g
4 OD O
O O
4g
+fN i
X i
i e
i i
i i
O 0
2 4
6 8
10 12 14 16 18 ACTUAL QUADRANT TILT, 5 e
Ouestion No. 8 How =any orifice rod assemblies vill be present during cycle h?
Where vill they be located? What are the peaking factors and flow proble=s associated with removal of orifice rod assemblies?
Restonse:
There vill be 62 orifice rod asse=blies present during cycle k.
Orifice rod (0XX) locations are indicated on the attached TMI-1 Cycle h Core Loading Plan.
The re= oval of orifice rod asse=blies does not affect core peaking distributions; furthermore, no crifice rod asse=blies have been re=oved relative to previous reload cycles. The absence of hh orifice red asse=blies has been factored into core thernal-hydraulic analyses by a reduction in the reactor coolant flow available for heat transfer.
The core thermal hydraulic analyses presented in the TMI-l FSAR (Reload Report, reference 1) and Fuel Densification Report (Reload Report, reference h) vere based upon a =axi=u= core bypass flow of 6.0h5 of system flov. The current ther=al-hydraulic analysis basis, as used for licensing of cycles 2, 3, and h, includes a core bypass flow of 8.3h5 of system flow, with the additional 2.3% bypass being a result of the absence of hk ORA's.
The actual core coolant flow available for heat transfer is greater than that which had been assumed for FSAR analyses by virtue of the fact that the RCS flovrate is approximately 109% of design flow. This is reflected in part by the use of 106.5% of design flow as the basis for thermal-hydraulic analyses of cycles 2, 3, and L.
~
<~
m o
e
gr [
Restonse to (. :estion No. 8 i
o jd juY o a g ut' Ts:t.1 CycIc 4
Core Le: ding Plan Rev. 2 TUEL TR/2;3FER CA';AL D>
X l.
CLE M,, i 1 W
. OI.Z 14P
.a_
345 13Z 0:0!
OAG OAH DAX 13!:
13V 14]
C16 C14 Rocop5 3
OXX C58 OEL 03, OAU 1B43 OLG IA22 Or.V 1T:29 ODA J3F OBG C
0:CC C29 OXX C09 OXX C17 OXX C45 OXX 134 130 0;J OL:t OLA 0/04
'OLG OSC 0;:lt OLX OA::
Ot.R 142
~
F-OXX C55 OXX A04 OXX C43 OXX A05 ou C57 OXX 1.d,L 0/.3 OLP 1A35 OAQ OMF '1A03 0::X, 0/J 1A34 0;10 OAX 13x E
C41 OXX C30 OXX C44 CE C46 0:G C33 OXX C32 0:f.!
135 1331 0::9 OAT OLT OAY 0::7 OAS CLit l02.3 OLF 1330 0:t4 OLC r
1 C60 i
COB OE A03 0:CC C13 07X C31 OXX C19 10XX A06 OXX 1,35 OEM OLY 032 OXit OAS OMA On6 0:12 035 0 2:11 OAS CLU OA9 Ci::
-}
C C11 OXX C42 OXX C12 0XX.
CIS OXX C48 0.u CIS l
13Y 01;E 1AUS OXS 1A13 OX0 CAA
- All 03?.
OL9 LA02 0:14 1A50 035 g'
~
i E
1:10 CE C39 OXX C2S OXX C61 OXX -
C34 OXX C47 on C22 u.r OAi Ou TAD 0:!J OBO C: 1
':53 0::S 03:3 0?2 OAG OM6 OAW I3:
C50 OXX C23 T
C03 Ou C40 OXX COG aXX C24
- 0 n g
1.[
0.%
1325 L:s 0E4 TL'1 OAP 3L7 039 0:IT 3B!!
OLD 1D27 II.O 14A C56 OXX A02 OXX
'C07 OXe.
C25 c:C' Col On AG'l tXY-C70 j
138 033 0' J 1A18 CaC OXL LA32 0::s GEQ 1A14 OMC 05?
139 v
C26 OXX C27 OXX C33 JE C52 uXX C36 OXX C49 143_
13A OAZ OLQ 0:rf ' CLi-
l OAI,. 3 3 OEB 1L36
'JI. i lIn21 OLS IS33 3A?.
O LL' 031 0
tOXX C3'7 OXX C05 2.u C21
-r a
- 35 OXX L
OLS DAT 037 C1,7 13T 13:t--
0:G 136 0:0:
~
E0000:
CO2 CO:
C54 OXX 1/.6 133 13r:
13:1 OLB I
l I
X 1.' 2 3
4 5
6 7
.C 9
10 11 12 13 14 15 i
l fuel 1.D. (It0G-3c.:ch 1; it::X-Ea::5 2; C u or 0.u-2.t:n 4 c:ncrs-Ba:ct.m T !. C E
orifice (0:C') or Car.::al sad I.:. (C:G - CJ,/.. MX - Ars:.1)
Source 1.3. (R000 X - Scar.d.' v}
"a:05 4. 5. ced 'i F A I.D ' 5-g:;,:,. : ::10 PrcC1xes 1489 023
5 Ouestien No. 9:
What is the maximum impact energy (in ft-lb) corresponding to the alars setpoints currently used in the Loose Parts Monitoring Systen? Also, briefly dese:ibe the location of the accelerometers.
F.esponse:
The nc=inal impact energy corresponding to the alarm setpoints currently used in the Loose Parts Monitcring System is 0.5 ft lb.
The location of the accelerometers is as follows:
a) Lever reactor vessel-incere guide tube 5 b) Lover reactor vessel-incore guide tube 13 c) Upper reactor vessel-reactor vessel head shrcud d) Upper reactor vessel-reactor vessel head shroud e} Steam generator "A"-upper tube sheet north side f) Steam generator "B"-upper tube sheet south side g) Steam generator "A"-upper tube sheet south side h) Steam generator "B"-upper tube sheet north side
~
~
1489 024
Questien No. 10:
Provide the following information regarding measurement.s made during cycle 3 a) Provide a lov and high power XY power map for BOC 3.
Both measured and predicted asse=bly powers should be given.
b) Provide the =easured and predicted BOC 3 red bank vorths by bank.
c) Provide the BOC 3 =easured values for critical boro 2 concentration and =oderation tc=perature ccefficient.
State the power and xenon conditions under which each measurement was taken.
d) Provide the measured and predicted ejected red verth for 30u 3 State the condition under which the test was done.
Resrence:
a) Power distributions frc= BOC-3 physics testing are provided en Figures 1, 2, 3 and h attached. Figures 1 and 2 provide radial and total peak-ing factors at 41.8% full power. Figures 3 and k provide radial and total peaking factors at 99.2h% FP.
b) The measured and predicted red bank vorths frc= BCC-3 zero pcuer physics tests are as follows:
Predicted Measured Group 7 0.73% AK/K 0.76% AX/K Group 6 0.96% AK/K 1.01% AK/K Group 5 1.08% AK/K 1.13% AK/K Group 1 - k 5.82% AK/K 5.h8% AK/K c) The critical bcron concentration at 30C-3 was measured at zero power and xenon free conditions. The All Reds Out Concentrations are as follevs:
Predicted Measured ARO Boren 1280 pp=
12k9 pp=
The results of the three =oderator coefficient tests performed during BOC-3 testing are as follevs:
Pre dicted Measured
-3 a = (:ero power, Xenon free,1255 pp=B)
<5n0 % AK/K/ F
+2.' :in0~ % AK/K/ ?
-3
-3 o
a = (zero power, Xenen free,1005 pp=B)
- h. 7no % AK/K/ F
-4. 8n0 3 37jgj y
-3 o
-3 a = (75% FP, 3-D equilibriu= Xenon,
-11x10 % AK/K F
-9X10 % AK/K'F 818'pp=B) d) The zero power maximu= ejected red verth measurement was made at BCC-3 zero power, no Xenen, 532 F Tave with Contrcl Ecd Groups 5, 6 and 7 at 0% vithdrawn. The results are as follows:
Predicted Measured Ejected Ecd Worth 0.3h5 AK/K 0.3h% AK/K 1489 025
Respor.se to Question ?io. 10 FIGURE 1 RADIAL PEAKING FACTORS 8/H 9/b 10/F 11/E 12/D
.13/C lli/B 15/A-1.013 1.208 0 967 1.12 1.289 1.219 o.608 o.757 H/8 1.e 1.21 1.02 1.20 1.20 1.16 o.57 o.77 0 97h 0 959 1.008 1.377 1.25 1.088 o.916 K/9 1.03 o.99 1.o8 1.39 1.22 1.lo o.86 o.982 1.090 1.192 0 9??
1.389 0.821 L/10 97 1.06 1.18 o.95 1.32 o.73 1.21h 1.102 o.939 o.971 1.28 1.16 o.9h 1.01 p,jf LEGEND X.XXX Measured Value o.59 o.687 0 5hT N/12 o.5T o.66 o.61 X.XX Calculated Value o.h85 0/13 o.h9 j0 llVl; 0 mo0 ,; m LJ md ( oc m % ud. 100 1.389 (Irlb) cp, 1_4 Ma,:inur !!easured Value 1.39 (K-12) 100 Maxieua calculated Value cp. 5 86 Maxinur Error (%): Cp. 6 9 3'3.N' x loo = -o.075 G. 7 P 1.:: 31 Gp. 8 Potter Letel kl.8 %pp E.f f e:t1 Full Potter Days o.56 EFPD 148^ 026
Figure 2 p.espor.c. to cuestia
- 1 RADIAL PEAKING FACTORS 8/H 9/G 10/F 11/E 12/D
.13/C lil/B 15/A 1.292 1.49 1.191 1.316 1.502 1.38h 0.683 0.872 H/g 1.31 1.47 1.25 1.42 1.45 1.35 0.66 0 92 1.17 1.131 1.381 1.611 1.h59 1.2h3 1.094 K/9 1.27 1.25 1.28 1.67 1.hh 1.29 1.e4 1.15 1.278 1 525 1.11E 1.6hl 0 988 1.18 1.29 1.60 1.1h 1 58 c.88 L/10 1.k52 1.301 1.056 1.132 1 56 1 kl 11 1-W11 LEGEND 'X.XXX Measured Value 0.723 0.766 0.651 0 72 0.76 0.71 N/12 X.XX Calculated Value 0.567 57-0/13 %MNh 9 D D d !it J e jd \\ 6' E % wd. gyj 1.6hl (L-1k) Gp. 1-4 100 Maximu a Measured Value 100 1.67 (K-12) Gp. 5 Maxicus Calculated Value Ob Maximur Error (%): Cp. 6 9 ~U .: 100 = -1 745 Cp. 7 _.c: 31 Cp. 8 Potcr Level Ll.3 %FP Effecti.- Tull Potter Days 0.56 EFPD 1489 027
Restense to Question IM.10 Figure 3. PADIAL PEAKING FACTORS 4 8/H 9/G 10/F 11/E 12/D .13/C II4/B 15/A-1.028 1.215 1.026 1.123 1.285 1.22 0.624 0 772 H/0 1.03 1.19 1.00 1.18 1.18 1.15 0 58 0.79 pg 0 985 0 975 0 961 1.368 1.2h6 1.082 0 905 1.02 0 98 1.06 1.36 1.21 1.10 0.89 s 0.991 1.097 1.187 0 962 1.351 0.811 96 l';
- 1. 7 97 1.32 75 L/10 1.206 1.111 0 938 0 966 1.26 1.15 0 95 1.03 j.)fg LEGEND X.XXX Measured Value 0.609 0.70 0.559 N/12 0.58 0.68 0.63 X.XX Calculated Value 0 h95 0 52 0/13 o *
- i))
70T v y//L g, oofu JJL 1 1 l % ud. 1.368 Gp. 1-4 100 }Iaxi=un Measured Value 1.36 100 Maxinus Calculated Valie Gp. 5 00 G, 6 Maxizen Error (%): P 13 1.365 '.36 x 100 = +0.59f, GP-7 g Gp. 8 Power Level 99.2h %FP Effecth : Full Power Days 5.L2 EFPD Icbr.icnn - 2.23 1489 028
TLg<re h aesper.se to deSti" " l RADIAL PEAKUtG FACTORS 8/H 9/G 10/F 11/E 12/D 13/C lll/B 15/A-1.297 1.h9 1.19 1.286 1.h98 1 36 0 771 0.881 H/8 1.30 1.h7 1.26 1.hu 1.h8 1 37 0.75 o.95 pg 1.163 1.113 1.31 1.632 1.h31 1.281 1.0h8 1.27 1.27 1.31 1.70 1.h5 1.29 1.06 1.131 1.276 1.h56 1.112 1 591 0 959 L/10 1.20 1.32 1.6h 1.18 1.60 0 91 1.391 1.278 1.056 1.125 1.55 1.h3 1.13 1.22 LEGEND X.XXX Measured Value 0.842 0 769 0.664 N/12 0.81 0 79 0 7L X.XX Calculated Value ~ - 0 556' J o % wd. Maximum Measured Value 1.632 Gp. 1-4 100 Maximun Calculated Value 1.70 Gp. 5 100 Maxinun Error (%): Gp. 6 88 1" x 103 =-h.05 Cp. 7 _ 13 O Gp. 8 20 Power Level 99.2L gyp r f f cc t i'..- Fe7.1 Pot.er Days 5.h2 EFPD 1489 029 I ba3Lc.:+ - 2.23
4 Cuestien No. 11: The startup physics test progra= ns given in Section 9 lacks the necessary depth of discussion. A significant amount of additional detail vill be rwquired in order to make clear the accept ability of the =ethods, procedures and acceptance criteria used for the various tests. Specifically, the fol-leving questions.re submitted en the test progra=s. Pes;cnse: The methed; detailed procedures and acceptance criteria for the BCC Physics Testing Progra= at TMI-l have been reviewed in detail by the NRC Region I Office of Inspectien and Enforce =ent staff thrcughout Cycle 1 (initial start-up-ind =id cycle red swap progra=) Cycle 2 and Cycle 3 The methods and procedures used for physics testing and adherance to acceptance criteria have been noted to be acceptable. These methods have not changed for BCC-k Physic, Testing. Centrolled copies of the detailed procedures describing t'rs methods and acceptance criteria for each test used for ECC physics testing are available on site for your review along with all data analysed to date. Su=saries of the test methed and acceptance criteria for each of the tests identified in your enclosure are as follevs: Ouestien Uc. Il(a): Describe in detail the tests being done to check for a misloaded asse=bly.
- T.lat assurances are there that the core is as expected before going to powers r
> 55 rated pover? Eescense: After cc=platien cf the fuel shuffle, prior to install'aticn of the reactor vessel heat, a video =ap is made of each fuel ass _e=bly identificaticn. This video tap is then ec= pared to the Cycle loading plan to assure that each fuel asse=bly is in its designated core position. Ouestien Uc. ll(b): Describe the procedures for the control red-trip test. Include the acceptance criteria and the procedures to be folleved if the acceptance criteria are not tet. Rescense: Tt.e centrol red trip times are measured in accordance with Technical Specifi-cation 4.7.1 by =casuring the time frc= deenergicing the undervoltage trip device until the 255 withdrawn (3/k insertien) reed switch is actuated. The acceptance criteria is 1.66 seconds for hot full flow or 1.h0 seconds for het no flev conditions. If the acceptance criteria is not met for a specific red, the red is declared inoperable until the problem is resolved. 1489 030
Questien 270. Il(C I. : Provide the details of the precedures for the critical boren concentration tests. Discuss how corrections are =ade to the measured data and how the =easured data is ec= pared to the predictions. What are the acceptance criteria and what are the procedures if the acceptance criteria are not met? Restense: Initial criticality following a fuel reload is achieved by withdrawal of centrol rods in Group 1-6 to 100% and Group 7 to 75%, folleved by deboration of the reactor coolant. Once an equilibriu= boren sa=ple is obtained at the initial critical rod position (nor= ally 75% vithdrawn on Group 7) the All Rods Out Critical Beren Cencentratica is obtained by fully withdravi:1g Group 7 control rods and =easuring the doubling ti=es due to the reactivity addition. This reactivity is converted to an equivalent boren concentration and is added to the equilibriu= boren concentration obtained at initial criticality to obtain an actual-all rods cut equivalent beren concentration. The predicted results for 30C-h start-up is 1250 pp=. If the acceptance criteria for this test (I 100 pp=) is exceeded the reactor would be placed in hot shutdown (Keff < 0 99) and the results vould be evaluated in depth prior to regaining criticality. Question To. ll(d): Describe in detail the procedures and =ethods used for the te=perature re-activity coefficient tests. Also provide the acceptance criteria and the procedures to be fellowed if the acceptance criteria are not =et. Rearense: The te=perature ccefficient of reactivity is =easured during 30C Zero Power Physics testing at two boren concentrations (All R0ds Out and at the Mini =u= Red Insertion Index). With the reactor just critical at equilibrium reacter cociant syste= cegditions, the reactor coolant syste= average te=perature (Tave) is varied -5 F. The change in net core reactivity due to the variatien in Tave is =easured by the Reactimeter (a reactivity calculater which uses input frc= an inter =ediate neutron range detector). The control rods are not =cved during this test at cero power. The reactivity change per change in F is calculated and extrapolated to 100% full power. If the extrapolated value shows that the moderator coefficient would be positive at hot full power, the te=perature coefficient test vill be repeated at 75% full pcVer and again extrapolated. If the extrapolation reveals the =oderator coefficient vill be negative at hot full power, the temperature coefficient tect is re-pated at 100% full power to verify that the acceptance criteria has been =et. Te=perature coefficient ceasure=ents at power are perfer=ed by varying Tave and observing the change in centrol red position while maintaining cen-stant reacter power. Thus the change in reactivity based on differential red verth per change in reactor coolant syste= average te=perature is calcu-lated. The predicted result of the 30C Zero pcver isother=al te=perature eccfficient is - 5 3 X 10"% AK/K/ F at 1230 rp= boren. The =oderator coef-D ficient shall be less than +0.5 X 10"% AK/K/ F at cero power to assure a non-positive =cderator coefficient above 95% full power. Results cf each test vculd be evaluated if acceptance criteria vere not =et and reactor power uculd not te increased above 95% full rover until it could be shown that a non-positive =oderator coefficient e'xisted. 1489 031
mm Cuestien 50. lif e) : O m M- 'f A c N. - ( cW Provide the details of the regulating centrol rod group reactivity verth tests. Give the predicted verth of each group to be ceasured, and the stuck rod worth and the predicted total vorth for all rods. Also provide the acceptance criteria and the procedures to te followed if the acceptance criteria are not cet. Restense: Centrol rod group reactivity measurements are performed at hot zero power conditions using the boren/ rod swap method and the rod drop method. The bcr:n/ red swap =ethod is used to measure the differential and integral re-activity worths of control red groups 5, 6 and 7 The total reactivity verth Of the safety red groups (Groups 1 k) is measured by the rod drop =ethod. The beren/ red swap =ethod consists of establishing a deboration rate in the reacter ecolant syste= and cc=pensating for the reactivity changes of this deb: ration by inserting centrol red groups 7, 6 and 5 in incre= ental steps. The reactivity changes that occur during these =easurements are calculated based en reacti=eter data and differential red worths are obtained frc= the kn wn reactivity verth versus the change in rod group position. The dif-ferential red worth of each of the controlling groups are then su==ed to obtain integral rod group vorths. For the red drop =easurement of the vorth of Groups 1 L, critical equilibrium ceniitions are established with centrol red groups 1 L vithdrawn frc= the core to the minitu rod index. The centrol rod groups being =casured are then dropped into the core. The reactivity inserted into the core is then calculated by analyzing data obtained frc= the reactiteter. The total reactivity worth of groups 1-4 is measured using the rod drop method. The predicted group vorths for BOC-4 testing are as follows: Group 7 1.37% AK/K Group 6 0 95% AK/K Group 5 1.39% AK/K, Group 1-h 5.00% AK/K The verst case predicted stuck red worth Cycle h is 2.03% AK/K. The acceptance criteria for total vorth is + 10% for Groups 5-7 and +- 15% for Groups 1-h. The total rod verth derived frc= these measurements is used to deter =ine available shutdown =argin. Shutdcyn margin must be greater than 15 i /K considering the cost reactive stuck rod out of the core. Ouestien No. 11(fl: Cescribe in detail the procedures for the ejected centrol red reactivity verth test. State the methods used to co= pare the =easurements with predicitions and the acceptance criteria. Also, include procedures if the acceptance criteria are not cet. ?estense: Ejected centrol rod verth is ceasured at hot zero power ccaditicas with the centrclling rod groups at the mininum allevable red index ucing two techniques. 1489 032
Resrense to Questien No.11(f) continued: The first technique is the boren swap method during which the boren concen-tration of the reactor coolant system is s10vly and centinuously increased. The ejected rod is withdrawn in quick steps to ecmpensate for the reactivity inserted by the boration and the reactivity change is ceasured by a reactivity calculator. The sus of the incre= ental reactivity changes gives the total verth of the ejected rod. In the second technique (red swap =ethod), critical equilibrium conditions are established with the ejected rod withdrawn to 100%. The ejected rod is then inserted into the core by swapping reactivity with another rod group. The =easured instantaneous worth of the rod (using react-ivity calculator) is taken as the worth of the ejected red. These =easured values are then error adjusted for uncertainty associated with the use of predicted rod worth data and uncertainty associated with the use of the boren swap =ethed. This error adjusted maxi =um ejected rod worth is then c0= pared with acceptance criteria. If the acceptance criteria of this procedure was not satisfied, the reactor vould be taken to hot shutdevn condition and the results vould be evaluated. 1489 033
0lest.cr :io. 11(c): 4 TMI-1 had a quadrant tilt at the beginning of Cycle 3 How did this tilt change during the cycle? How was the presence of this tilt used in the predictions of the power distribution fc r Cycle h? ?.escens e : The TMI-l Cycle 3 indicated tilt re=ained below the error-adjusted. Technical Specification limit throughout the entire cycle. The indicated tilt at the beginning of the cycle was near 1% and it remained steady for approximately 120 IFFD's. After returning to pcver following an outage, the indicated tilt increased to 2.2%. It gradually decreased to 1.2% by the end of the cycle. Because of the apparent enhancement of tilt in another plant due to cross ccre shuffling the original cycle h design was revised to a shuffle philosophy which generally moves the fuel frc= a given quadrant into both of the adjacent quadrants. This shuffle tends to =inimi:e possible carry-ever effects of any burnup assymetry that =ight be present in the previous cycles. Because of the ice value of indicated tilt at end of cycle 3, any carry-over effects of the tilt would be reall and should be essentially eliminated by the revised fuel shuf fle. Censequently, the presence of tilt in cycle 3 was not used in the prediction of -he power distributien for Cycle 4. e 1489 034
a ,uestien No. Il(h): Provide the details of the core power distribution tests. Desceibe in detail the =ethods used to predict the asse=bly by assembly power as vell as the analyses of the data obtained during the =easure=ents. What tre the assembly by assembly acceptance crlteria? How are tilts accounted for in the ana2ysis of the data? If a 1/4 or 2/8 core map is the result of the measure =ent, ; hat method is used to determine he asse=bly power for those assemblies having their sy==etric a::sse=blies i' tru=ented? For exa=ple, are the =easured assembly powers averaged, or is only one of the sy==etric =easure=ents used? Pesrense: Core Power Distritution Tests are perfor=ed at h0, 75, and 100% FP. The test at h0% FP is essentially a check on pcver distribution in the core to bring attention to any abnor=alities before escalating to the 75% FP plateau. Rod index is established at a nc=inal full power configuration which is where the cere power distribution calculations are perfor=ed. APSR position is established to provide a core pcVer i= balance ccrresponding to the i= balance vc.ere the core power distribution d ealations are perfcreed. The following acceptance criteria are placed on the h0% FP test: (1) The worst case maxi =u= linear heat rate must be less than the LOCA limit specified in Technical Specifications Figure 3.5-2J. (2) 'he =ini=um DU3R =ust be greater than 1.30. (3) T te value obtained from the extrapolation of the mini =u= DN3R to tae next power plateau overpower trip setpoint must be greater than 1.30 or fall catside the RPS power / imbalance trip envelope. (2) 'Ihe value obtained from the extrapciation of the vorst case maxi =un linear heat rate to the next power plateau overpcVer trip setpoint must be less than the fuel melt li=it or fall outside the RPS power / i= balance trip envelope. (5) The quadrant pcver tilt shall not exceed the limits specified in Technical Specifications. (.6) The highest =easured radial peak shall not exceed the highest predicted radial peak by = ore than 85. (7) The highest =easured total peak shall not exceed the highest predicted total peak by more than 125. Items 1, 2, 5, 6, and 7 above are established for the purpose of verifying core nuclear and ther=al calculational =odels, thereby verifying the acceptability of data fro = these models for input to safety evaluations.
- ems 3 and h establish the criteria whereby escalation to the next pcVer plateau may be acec=plished without exceeding any safety limits specified by the safety analysis with regard to DNER and linear heat rate.
1489 035
s Resnonse to Cuestion 11(h) Continued The tests are also performed r.t 75 and 100% FP and in the same manner as the L0i FP test with one exception. At 75 and 1005 FP, three dimensional xenon equilibrium is required; whereas, at h0% FP there are no equilibrium xenon requirements. The same acceptance criteria apply with the exception that the highest measured radial and total peaks shall not exceed the highest predicted radial and total peaks by more than 5 and 7.55, respectively, for both 75 and 1005 FP testing. The more restrictive limits are due to the equilibrium xenon require =ents at 75 and 1005 FP. Predicticns for the radial and total peaks at LO, 75, and 100% FP are calculated using the FLME-3 with thermal-hydraulic feedback code (BAV-1012k). Radial peaks are calculated frc= the predicted power output for each assembly in a 1/8 core. Total peaks are calculated frc= the predicted power output of the =axi=um seg=ent for each assembly in a 1/8 core. Assenbly and seg=ent power representatiens are calculated by the on-line ec=puter based ca current-signal cutputs frc= the 52 incore detector strings. Any tilt which exists.in the core is inherent in the =easurement of neutron flux by the incere detector syste=. Only instrumented asse=blies are utilized in the analysis of the data to calculate measured radial and total peaks for ec=parison to predicted radial and total peaks. Sy==etric instrumented locations are averaged to provide a single value for the asse=bly or segnent power in the 1/8 core locatien. Radial and total peak are then calculated. As previously stated, the maximum measured radial and total peaks are ec= pared to max 1=u predicted radial and total peaks. There are no crite tia for comparisons on an assembly by assembly basis. Tilt effect.: sre accounted for in the calculation of DNBR and linear heat rate. If a tilt dces exist, a routine in the on-line computer adjusts the s?g=ent pcver representatiens of an instru: ated asse=bly in order to provide aegnent power representations of a sy==etric, non-instrumented assembly. DNER and linear heat rate are calculated by the on-line computer for the maximum assembly in each of the four core flow regions. These values are then'ec= pared to acceptance criteria previously diseassed. In addition, a hand calculation of linear heat rate is performed in order to obtain values fer ec=parison with LOCA acceptance criteria which are level dependent. 1489 036
r Q2estion ifo.11(i): Provide a co=mitment to prepare a brief su==ery report of the Cycle h physics startup tests and to submit this report to NRC vithin h5 days of the com-PAetion of the startup tests. Thic report should include both measured and predicted values. If the difference between the mearured and predicted values exceed the acceptance criterion, the report should discuss the adequacy of the actions taken. Res;cnse: Met-Ed vill provide the information requested above for the Cycle h physics startup tests, and vill submit the information to NRC within 90 days of the co=pletion of the startup tests, consistant with other Tech Spec reporting require =ents. 1489 037}}