ML20206S423
| ML20206S423 | |
| Person / Time | |
|---|---|
| Site: | Maine Yankee |
| Issue date: | 09/15/1986 |
| From: | Whittier G Maine Yankee |
| To: | Thadani A Office of Nuclear Reactor Regulation |
| References | |
| 8049L-GDW, GDW-86-212, MN-86-118, NUDOCS 8609220211 | |
| Download: ML20206S423 (15) | |
Text
..
MAME HARHEE MOMICPOWERCOMPARUe
,uougr,y,in"gys h
(207) 623-3521 O
September 15, 1986 MN-86-118 GDH-86-212 Director of Nuclear Reactor Regulation United States Nuclear Regulatory Commission Washington, D. C.
20555 Attention:
Mr. Ashok C. Thadani, Director PHR Project Directorate #8 Division of Licensing
References:
(a) License No. DPR-36 (Docket No. 50-309)
(b) SECY-83-472 (c) Draft Regulatory Guide, "Best Estimate Calculations of Emergency Core Cooling Systems Performance," September 12, 1985
Subject:
Maine Yankee LOCA Analysis Gentlemen:
As discussed with members of your staff on September 2, and 9, 1986, we have determined that certain flatter axial power shapes could be more limiting than the top peaked axial shapes used in the approved Appendix K LOCA analysis.
This has potential implications for current Cycle 9 operation.
LOCA results appear to be more limiting as the axial power shape is progressively flattened (and the radial power shape peaks) relative to the traditionally analyzed top peaked shapes.
He believe the sensitivity of peak clad temperatures and resulting linear heat generation rate limits to variations in axial power shapes can be attributed primarily to the conservatisms inherent in the approved Appendix K models.
He have further concluded that this is not a significant safety issue but rather a product of particularly conservative assumptions in the reflood phase analysis. The LOCA limits currently in the Technical Specification, when combined with the higher radial peaking factors now being obtained and flatter axial profiles, will result in computed peak clad temperatures that exceed 2200*F. However, it does not necessarily follow that the safety function of the ECCS has been compromised.
8609220211 860915 PDR ADOCK 05000309 i
i P
(
8049L-GDH i\\
MAINE YANKEE ATOMIC POWER COMPANY United States Nuclear Regulatory Commission Page Two Attention: Mr. Ashok C. Thadani, Chief MN-86-ll8 The evaluation models used to calculate peak clad temperatures contain many conservatisms required by Appendix K to 10 CFR 50.
These conservatisms were designed to bound the uncertainties that the NRC felt existed in the l
technology at the time the rules were finalized in 1974.
Since then, safety research has clearly shown that large margins exist between the results of a LOCA predicted by the Appendix K approach, and results predicted by an analysis based on the results of that safety research.
This also appears to be recognized by the NRC, since they have written two documents, References (b) and (c), which have proposed alternatives to the conservatism of Appendix K based on the results of the safety research. Appendix K over-predicts peak clad temperatures by as much as 1000*F.
Therefore, even if Maine Yankee had operated with peak LHGRs at the LOCA limit, a more realistic analysis would show peak clad temperatures well within the 2200*F limit. For these reasons the safety function of the ECCS would not have been lost even if the plant has operated at the Technical Specification limits.
Notwithstanding, until the NRC staff has had the opportunity to review and approve changes which we intend to propose to our LOCA methodology, we have instituted the administrative controls depicted in the attached Figure (Attachment A).
This Figure was developed using the current NRC approved LOCA l
methods with the more limiting axial flux shape.
l The current LOCA analysis of record for Maine Yankee was performed in support of Cycle 5 operation. He have determined that from Cycle 5 to the present we did not operate outside of the administrative limits which are currently imposed. He would not expect to exceed these limits operating at full power for the remainder of the current cycle.
Recovery of this apparent lost margin is possible through the approved revisions to the LOCA codes in the following three areas:
1.
A revised core model for use in the reflood phase of LOCA could be developed to more accurately characterize expected core response.
2.
Credit could be secured for revisions associated with the emergency core cooling delta p injection model.
3.
The steam cooling models could be updated to better reflect experiment results.
8049L-GDH
MAINE YANKEE ATOMIC POWER COMPANY United States Nuclear Regulatory Commission Page Three Attention: Mr. Ashok C. Thadani, Chief MN-86-ll8 As we discussed during our meeting on September 9,1986, we plan to submit changes to our LOCA methods which incorporate a revised core model during reflood and a reduced delta p injection penalty by December 1,1986. He plan to use these revised methods in analyzing Cycle 10 which is scheduled to begin on approximately June 1, 1987.
A copy of the slides we used during our September 9, 1986 meeting are attached as Attachment B.
Please feel free to contact me if you should have any questions in this matter.
Very truly yours, MAINE YANKEE ATOMIC P0HER COMPANY bh AV G. D. Whittier, Manager Nuclear Engineering and Licensing GDH/bjp Enclosures cc: Dr. Thomas E. Murley Mr. Pat Sears Mr. Cornelius F. Holden l
8049L-GDH
ATTACHMENT A FIGURE 3 i
COMPARISON OF RCCEPTRBLE LHGR TO EXPECTED LHGR S-h-
T.E 3.10.C.1
/ 3 2; ;%,
%g E-x-
%c a
%g m
H kUh A
a 2
\\
PROPOSED RDt1IN.
s M
ENVELOPE OP Y
EXPECTED MRX. LHGR'S
'N /
LIMIT x n~
\\
CD d I
'g
-1
\\
\\
w"
\\
\\
\\
\\
S-i h
m-I
@ am i
I i
3 0
20 40 60 80 100 CORE HEIGHT (%)
.,e.
..,,,----,,_a-
l ATTACHMENT B i
MY LOCA ANALYSIS AT YAEC (BACKGROUND)
YAEC METHOD WAS FIRST USED FOR MY CYCLE 5 ANALYSIS CE ANALYSIS HAD IDENTIFIED TOP SKEW POWER SHAPE CONSERVATIVE MY SYMMETRIC 0FFSET LIMITS WERE USED TO DEFINE TOP SKEW POWER SHAPES TOP SKEW POWER SHAPE USED FOR LHGR FOR PEAK AB0VE CORE MID PLANE CH0PPED COSINE USED FOR LHGR FOR PEAK BELOW CORE MID PLANE VERIFIED THAT LOCA DESIGN SHAPE IS VALID EACH CYCLE BY REEXAMINING POWER SHAPES RECENT STUDIES WITH YR SHOWED FLATTER POWER PROFILES MORE LIMITING.
(
EXAMINATION OF MY CYCLE 10 SHAPES REVEALED FLATTER POWER PROFILES DO EXIST CYCLE 9 LIMITS WERE RE-ANALYZED WITH FLATTER I
POWER SHAPES AND ADMIN. LIMITS IMPOSED ON l
PLANT OPERATION l'
l AH: SEPT. 9, 1986
Figure 1 Maine Yankee Design Power Distributions 2.0 FSAR Design 1.68 1.41 1.5 -
LOCA Design Normalized 1.0 Axial Power Distribution Design
--Symmetric Offset--
LCO i
Shape Assembly Peripheral Power 0.5 Case Shape w/ Uncertainty Level FSAR
+0.36
+0.30 0-100%
+0.23
+0.11 100%
i 0
0.1 0.2 0.'3 0.'4 0.'5 0.'6 0.7 0.'8 0.'9 1.0 0
Fraction of Core Height 1
l i
4 I
Figure Maine Yankee Cycle 5 LOCA Axial Power Distributions compared to 3
Highest Core Height Peaking Case (Steady-State) l l
l l
l l
l 20 00 10.00__
LOCA 16.0 kw/ft 16.00__
r Presumed LOCA Limit Core Height
~
N
~
Dependence 14.0 kw/ft, #
..e, - %
14 00__
3 LOCA
/
12.00_
n D
/
E10 00__
[/
Highest Core Height Peaking, Steady-State o,.
$ B.00_
11.3 kw/ft 4
a S 00__
i t 00__
I 2.00; i
0 00 i
i 0..]O fC
.d0
,.10
.LO 40
.dC
-10
.SG
-50 1.b0 RELATIVE CORE HEIGHT
Figure Maine Yankee Cycles 5-9 LOCA Axial Power Distributions compared to Tech. Spect Limit 4
and Envelope of Measured Plant Operation with Uncertainties 20-00 i
i i
8 i
1 I
I I
I I
I l
5 19 00__
Tech. Spec. 16 kw/ft
~
15.00 Tech. Spec. 14 kw/ft 14 00__
,/
._12.00__
p m
Envelope of Measured N
610 00__
O eration with Uncertainties P
e
[.
y (includes controlled power
/
reductions, coastdown and S 9.00__
full power operation) a b 00_
A 00_.
2.00; i
0 00 i
i i
i 0 JD 1' C
.d0
. :.0
.t0 40
.50
-10
. r. 0
-i0 1.b0 REL ATIVE CORE HE!GHT
PIGURE 5 COMPARISDN OF ACCEPTABLE LHGR TO EXPECTED LHGR 2-h-
T.S 3.10.C.1
/
2; D
%g
%g' g-
%g
%g I-N - c,-
3 M
%g kUh A
a 2
\\
M ENVELOPE OP s
PROPOSED ADMIN.
EXPECTED MRX. LHGR'S
'\\ /
LIMIT ali a-
\\
x
\\
\\
l L
=-
i
\\
\\
\\
=
\\
h m-W-
h a
i i
8 0
20 40 60 80 100 CORE HEIGHT (%)
MY CYCLE 10 LOCA ANALYSIS ENI Av0!D POWER DERATE j
Av0ID UNWARRANTED OPERATING RESTRICTIONS HIGHER LOCA LIMITS NEED FOLLOWING MODEL CHANGES APPROVED INJECTION aP PENALTY AVERAGE CORE REPRESENTATION FOR FLOODING CALC.
REVISED STEAM COOLING CALCULATION AH/ SEPT. 9, 1986 n
-.,--.,,-.-,..,-.,,,,,,w
,r,,
,.--,n--,--,,
n
-r-
INJECTION aP PENALTY CURRENT MY MODEL (BASED oN XN-75-41)
AP PENALTY ANGLE OF INJECTION WITH WITHOUT Acc Acc 0
90 1.8 0.8 0
75 1.5 0.35 60 0.4 0.35 45 0.6 0.30 XN-NF-78-30 MODEL o
BASED ON EPRI-294 " MIXING 0F ECC WATER WITH STEAM" 1/14& 1/3 SCALE TESTS o
AP = 0.15 PSID FOR PUMPED FLOW o
GENERIC APPROVAL SER 3/30/79 YANKEE R0WE O
AP = 0.15 FOR PUMPED FLOW SER 11/27/85 PROPOSE FOR MY APPLICATION o
AP = 0.15 FOR PUMPED FLOW I
t J
i AH/ SEPT. 9, 1986 1
~
t REVISED CORE MODEL DURING REFLOOD CURRENT MODEL HYDRAULICS:
ALL ASSEMBLIES CONSIDERED H0T ASSEMBLY POWER = 1.02 x POWER x FASS.
TEMPERATURE OF CORE AT HOT ASSEMBLY CONDITIONS HEAT-UP:
HOT R0D WITH T00DEE2-EM PROPOSED MODEL i
HYDRAULICS:
CORE MODELLED WITH REPRESEN-TATIVE ASSEMBLIES POWER = 1.02 x POWER TEMP 0F CORE WITH REPRESEN-TATIVE ASSEMBLY HEAT-UP:
H0T R0D WITH T00DEE2-EM
{
AH/ SEPT. 9, 1986
i
?
i 1.
STEAM COOLING FLECHT-SEASET TEST RESULTS SHOW o
NO DISCONTINUITY IN HEAT TRANSFER OCCURS AT 1 IN/SEC o
CORE HEAT TRANSFER ENHANCED UNDER BLOCKAGE SITUATION STEAM-COOLING MODEL WILL BE REVISED o
LITERATURE ON REFLOOD HEAT TRANSFER WITH BLOCKAGE WILL RE REVIEWED o
T00DEE2-EM MODEL WILL BE REVISED o
REVISED MODEL WILL BE BENCHMARKED AGAINST EXPERIMENTAL DATA AH/ SEPT. 9, 1986
CYCLE 10 ANALYSIS (PHASE I)
CURRENT METHOD WITH FOLLOWING CHANGES WILL BE UTILIZED o
aP PENALTY OF 0.15 PSID ( ALREADY USED FOR YR) o REFLOOD WITH AVG. CORE CONDITIONS (PHYSICAL REPRESENTATION)
BREAK SPECTRUM SENSITIVITY WILL BE PERFORMED o
CYCLE 5 RESULTS WILL BE USED FOR THE BLOW-DOWN PERIOD o
MODIFIED REFLOOD MODEL WILL BE USED FOR THE REFLOOD PERIOD.
o WORST BREAK SIZE WILL BE IDENTIFIED B0UNDARY CONDITIONS FOR WORST BREAK SIZE WILL BE USED TO GENERATE LHGR FOR VARIOUS POWER SHAPES RESULTS WILL BE SUBMITTED WITH THE CPAR 90 DAYS BEFORE CORE STARTUP AH/ SEPT. 9, 1986 t
k h
CYCLE 10 ANALYSIS (PHASE II)
CYCLE 10 ANALYSIS WILL BE REDONE WITH A REVISED STEAM C0OLING MODEL MODEL CHANGE SUBMITTED 2/1/87 TO NRC SER FROM NRC 5/1/87 SUBMITTAL 0F ANALYSIS 10/1/87 SCOPE SIMILAR T0 (PHASE I)
SER ON SUBMITTAL 12/1/87 l
AH/ SEPT. 9, 1986
. -. - -. - -.. - - - -.