ML20198Q570
| ML20198Q570 | |
| Person / Time | |
|---|---|
| Site: | San Onofre  |
| Issue date: | 11/06/1997 |
| From: | Rainsberry J SOUTHERN CALIFORNIA EDISON CO. |
| To: | NRC OFFICE OF INFORMATION RESOURCES MANAGEMENT (IRM) |
| References | |
| NUDOCS 9711120254 | |
| Download: ML20198Q570 (5) | |
Text
.
'Q*'
s<,u,,,, - o, -
EDISON L C "; t h n, Am ib M)V l%71RA47IOML" Ownpny November 6, 1997 U.S. Nuclear Regulatory Commission Document Control Desk Washington, DC 20555 Gentlemen:
Subject:
Docket Numbers 50-361 & 50-262 ABB-CE LOCA Evaluation Model San Onofre Nuclear Generating Station, Units 2 & 3
Reference:
Letter,Mr.R.W.Krieger(SCE)toNRC,datedOctober9,1997.
On October 20 ard 22, 1997, telephone conversations were held between NRC and Southern Califor.41a Edison (SCE) person.iel to discuss additional details regarding SCE's October 9,1997,10 CFR 50.46 subnnttal (ABB-CE errors in Loss ofCoolantAccident(LCwA) analysis),
this letter serves to document the discussions.
Methodology used to determine the effect on Peak Cladding Temperature (PCT).
ABB-CE selected St. Lucie Unit 2 as the typical plant with a 16x15 fuel configuration. ABB-CE ran the STRIKIN code for the hot fuel rod analysis to determine the effect of the change in Peak Cladding Temperature (PCT) as a result of a change in the Energy Redistribution Factor (ERF).
The initial PCT from this code was 2134'F. The ERF was increased by 1%, and PCT changed to 2171'F, a 37'F increase.
The ERF was again increased by another 1% and PCT increased to 2211'F, a 40'F increase. Over this range of ERF increase between 0 to 2%, the 40'F increase in PCT was selected as the worst case for a 1% ERF increase.
Because the code, basedeck, and significant trends during the event are very similar for all 16x16 plants, and the results of the LOCA analysis between St. Lucie and the other 16x16 plants are similar, it holds that the increase in PCT for a 1% increase in'ERF would be ns more that 40'F.
f i
fD lQ 9711120254 9711 %
DR ADOCK O 31 San uclear Getwrating Station San Ckmmte, CA 9?t>7441128 714 3t+1420.
Document Control Desk Penalties for Core Operating Limit Supervisory System (t,0LSS) "UNCERT" and for CoreProtectionCalculator(CPC)BERR3andBERR4, At the beginning of each cycle an assessment of expected cperating margins is performed, and when appropriate and at management discretion, the required uncertainty factors are compensated to minimize the impact on plant operation for any unidentified analysis, operational, or maintenance errors.
The introduction of conservatism is implemented through several constants.
These constants do not typically change over the course of the cycle, but do provide some flexibility for operational considerations, where the penalties may be reduced.
-The COLSS calculation of peak operating KW/FT is cow..vatively biased by increasing the uncertainty multiplier "UNCERT" above the required value.
UNCERT is a multiplicative penalty applied to the COLSS calculation of peak operatingKW/FT.
UNCERT is calculated using approved methods and is consarvative over all allowed operating bands at a 95/95 probability /
confidence level. The value of "UNCERT" required to support Unit 2 and Unit 3 Cycle 9 operation and the values actually installed in Unit 2 and Unit 3 for Cycle 9 are provided below.
The installed values were compensated to introduce approximately a 1.8% conservatism to the calculated peak operating KW/FT. As a result of this conservatism, COLSS will annunciate reaching the KW/FTlimitwhentheuncompensatedvaluewouldhavereached12.77KW/FT.
This factor of conservatism proved to be more than adequate to compensate for the error subsequently discovered in treatment of Energy Redistribution Factors used in the LOCA analysis.
For COLSS:
Unit 2 Unit 3 Required by the cycle atelysis to maintain 13 KW/FT (Technical 3pecification (TS) Limit)
UNCERT 1.11 1.12 Installed values to provide an aaministrative margin for operating limits.
UNCERT-1.13 1.14 Conservatism 1.0180 1.0179 (installed / required)
r Document Control Desk WhenCOLSSisoutofservicetheKW/FTvaluescalculatedbyCPCsareusedto monitorcompliancewiththe13.0KW/FTLOCAlimit. As described above, CPC uncertainty requirements are also evaluated and appropriately increased at management's discretion to minimize potential impacts on plant operation from analysis, operation, or maintenance errors. The CPC uncertainty constants relevant to the calculation of KW/FT are BERR3 and BERR4.
I CPCsusetheseconstantsasabias(BERR4)andagain(BERR3)tothecore power level used in calculating peak operating KW/FT.
This compensation is performed by biasing the maximum of thermal or neutron powers by BERR4 and multiplying the sum by BEPR3.
The required and installed values for BERR3 and BERR4 applicable to Unit 2 and Unit 3 Cycle 9 operation are provided below.
As can be seen from the relative effect of conservatisms, the CPC r.alculated peakoperatingKW/FTisapproximately2.4%largerthanrequiredbydesign analyses. As was the case with COLSS in service, this voluntary CPC-conservatism proved to be more than adequate to compensate for the error subsequent 1v discovered in treatment of Energy Redistribution Factors used in the LOCA analysis.
For CPC:
Unit 2 Unit 3 l
Requiredbythecycleanalysistomaintain13KW/FT(TSLimit)
BERR3 1.104 1.117 BERR4 8.0 8.0 Compensation 1.1923 1.2064 l
Installed values to provide an administrative margin for operating limits.
1 BERR3 1.12 1.134 BERR4 9.0 9.0 Effect 1.2208 1.2364 Conservatism 1.0?39 1.0246 (installed / required)
t Document Control Desk ERF values oe.% and after error discovery.
These values of ERF are a furation of the pin / box factor, as follows:
pin / box Old ERF New ERF 1.0 0.97 0.98 1.1 0.938 0.943 PCT ',alues before and after the error.
PCT for:
Unit 2 Unit 3 Current Cycle before error 2148 2153 Current Cycle afte. error 2188 2193 WithCOLSS/CPCpenalties 2120 2125 Large Break Loss of Coolant Accident (LBLOCA) analysis.
Several years ago (1992) the LBLOCA analysis was rerun to support a Technical SpecificationchangeinPeakLinearHeatGenerationRate(PLHGR)to13KW/FT.
Currently, SCE has no plans to rerun the LBLOCA analysis.
Each cycle the new core input values are compared to the LBLOCA analysis input valves to assess theimpactontheTechnicalSpecificationPLHGRof13.0KW/FT.
10 CFR 50.46 requirements regarding this error.
10 CFR 50.46 (a)(3)(1) requires an estimate of the changes in PC1 values to evaluate and report significant (>50'F) changes and to report the results of the estimates annually. SCZ has evaluated the information provided by ABB-CE and determined the estimate of 40'F is valid and not significant, and therefore, the requirements of 10 CFR 50.46 have been met.
i t
f
' Document Control Desk :
Should you have any further questions or need additional information, please.
let me know.
Sincerely,
)
cc:
E. W. Merschoff, Regional Administrator, NRC Region IV K. E. Perkins, Jr., Director, Walnut Creek Field Office, NRC kegion-IV J. A._Sloan, NRC Senior Resident inspect 0r, San Onofre Units 2 & 3 M.-B. Fields, NRC N oject Manager, San Ono'.re Units 2 and 3 1
3 f
s i
31
(
.-n e,_. -
.w_.__
..,-,---,,,.-7,.....m-w
...,_c.,-
.,m
...mm
.-