ML20101R165
| ML20101R165 | |
| Person / Time | |
|---|---|
| Site: | Mcguire, Catawba, McGuire  |
| Issue date: | 04/09/1996 |
| From: | Tuckman M DUKE POWER CO. |
| To: | NRC OFFICE OF INFORMATION RESOURCES MANAGEMENT (IRM) |
| References | |
| NUDOCS 9604160079 | |
| Download: ML20101R165 (7) | |
Text
-
~~
._.. ;. 11 l
Dukekeer Cornpany MSDawn i
P.0 Box 1006 Senior Vice hesident Daarlotte, NC282011006 NuclearGeneration
\\
^
(701)3822200 OWice Vs (704)3824360 Tax
~
e--
i April 9, 1996 U. S. Nuclear Regulatory Commission Washington, D.
C.,
20555 l
Attention: Document Control Desk l
Subject:
McGuire Nuclear Station Docket Numbers 50-369 and -370 Catawba Nuclear Station Docket Numbers 50-413 and -414 Duke Power Company Topical Report DPC-NE-1004; Supplemental Information to Support Minor Revision
)
1 l
By letter dated December 12, 1995, Duke Power Company t
i proposed a minor revision to Topical Report DPC-NE-1004, l
" Nuclear Design Methodology Using CASMO-3/ SIMULATE-3P."
The NRC staff. requested that additional information be provided, by letter dated February 28, 1996.
Attached are responses to questions in the staff's RAI.
1 If there are additional questions, or more information is needed, please call Scott Gewehr at (704) 382-7581.
l i-Very truly yours,
, S.
E a
M.
S.
Tuckman l-f i
}
150053 9604160079 960409 i
PDR ADOCK 05000369 P
{l
~
'+rwm=rmov-
l-U.
S. Nuclear Regulatory Commission April 9, 1996 Page 2 cc:
Mr. V. Nerses, Project Manager Office of Nuclear Reactor Regulation U. S. Nuclear Regulatory Commission Mail Stop 14H25, OWFN Washington, D. C.
20555 Mr. R. E. Martin, Project Manager Office of Nuclear Reactor Regulation U.
S. Nuclear Regulatory Commission
)
Mail Stop 14H25, OWFN Washington, D. C.
20555 Mr. S. D. Ebneter, Regional Administrator U.S. Nuclear Regulatory Commission - Region II 101 Marietta Street, NW - Suite 2900 Atlanta, Georgia 30323 l
Mr. G. F. Maxwell Senior Resident Inspector McGuire Nuclear Station Mr. R. J.
Freudenberger
)
Senior Resident Inspector Catawba Nuclear Station
i l
I i
i Page 1 of 6 Response to NRC Questions l
- 1. What is the reason for the increase from 12 to 24 axial nodes? DPC's letter of December 12, 1995 mentions several factors potentially involved with the change, but the reason for the l
change is not explicitly discussed.
i e
Answert 'Ihe choice of a 12 axial node SIMULATE-3 model was originally chosen as a l
compromise between the desim for more axial nodes and that of computer mn time. Since initially submitting DPC-NE-1004, the increase in computational efficiency has allowed Duke to increase the number of nodes without a substantial increase in computer cost. Additional reasons for wanting to increase the number of axial nodes in SIMULA'm-3 from 12 to 24 are discussed below.
- a. SIMULATE-3 has a coding limitation on the number of axial regions which can be modelled within one node. For axial blanket fuel with bumable poisons rods, the nodal tengdi must be decreased to less than 12 inches in order to satisfy this criteria. Axial I
node segments of 6 or 8 inches satisfy this mquirement. A 6 inch axial node segment was chosen because this causes the nodal boundaries to match up exactly with the j
transition from the blanketed fuel region to the non-blanketed fuel region. Modelling l
these boundaries exactly increases calculational accuracy in this region of the core since I
l cross sections of two diffemnt enrichments are not averaged.
- b. To better account for the axial dependence in predicted and measured power distributions of current generation vendor fuel designs which are not axially l
homogeneous (eg. axial blanket fuel).
l i
I I
i l
i i
4' 5
Page 2 of 6 l
- 2. Discuss all of the input involved in the changes shown in the Observed Nuclear Reliability Factors (ONRF). Are the changes the result of an analysis of actual fuel cycle data versus l
previously assumed values? If so, which is more conservative? Or, are the changes the result of the axial noding change.
Answer The ONRF's for bein 12 and 24 axiallevel SIMULATE-3 models are the insult l
of analysis of actual fuel cycles. The changes in the calculated ONRF's are the result of using power distributions from more recent fuel cycles, increasing the number of axial nodes from 12 to 24 and accounting for the axial dependency of the power to reaction rate ratio in the measured power distribution. The original ONRF's for Westinghouse plants l
were developed in DPC-NE-1004 based on the analysis of the McGuire 2 Cycle 4. -
McGuire 2 Cycle 5, Catawba 1 Cycle 3 and Catawba 2 Cycle 2 core designs. 'Ihc 24 axial level ONRF's were developed based on current generation core designs in order to l
reflect the more aggressive reload design strategies reflected in current core designs.
Specifically, the power distribution database used to develop the 24 axial level ONRF's included the effects of higher enriched fuel, longer cycle lengths, higher burnup, higher BPRA loadings, and axial blankets. Characteristics of the fuel cycles analyzed are shown in Table 1.
1 Table 1 Fuel Cycle Characteristics i
No. of Feed Dggiggg
.Q11t and Enrichment EFPD Cycle Characteristics
]
L MIC09 64 feed at 3.45 w/o 340 Iww number of feed and BP loading M2C09 76 feed at 3.65 w/o 395 High Enr. and BPloading CIC07 72 feed at 3.45 w/o 350 Typical C2C06 76 feed at 3.75 w/o 380 First Transition cycle to Mk-BW C2C07 40 feed at 4.0 w/o +
430 Axial Blankets and High No. of feed j
8 feed at 3.60 w/o+
l 40 feed at 3.50 w/o j
+ Axial Blanket fuel with 6 inch natural uranium blankets The 24 axial level Fq and Fz ONRF's decreased by 2.0 and 2.2%, respectively, relative to DPC-NE-1004 values. However, the 24 axial level model ONRF for FAH increased by 0.3%.
over the DPC-NE-1004 value. This increase is considered statistically insignificant and is attributed to the selectior. of more challenging, and radially heterogeneous core designs for the creation of the 24 axial level statistical data base relative to the core designs which were available when the DPC-NE-1004 FAH ONRF was developed. The significance of the 0.3%
increase in the FAH ONRF is addressed in the answer to question 3.
i I
Page 3 of 6 Answer to Question 2 (Continued)
'Ihe decrease in the Fq and Fz uncertainties is attributed to the increase in axial resolution of the core model resulting from the axial node size reduction and from the use of axially dependent power to reaction rate ratios (which better characterize the spectral dependency and axial geometry of the fuel) to process measured reaction rates. These factors result in the j
reduction in the bias term included in the ONRF derivation. Note also that the variability of the statistical population as measured by the standard deviation of the Fq and Fz populations remain similar to previously calculated values. 'Ihcrefore, the reduction in the Fq and Fz uncertainty factors is almost entirely due to the reduction in the predicted to measured bias.
The statistical dats used to develop the 12 and 24 axial level ONRF's is provided in Table 2.
'The equations used to develop the ONRF's are contained in Section 4.2 of DPC-NE-1004.
l Table 2 12 and 24 Axial Level Observed Nuclear Reliability Factors l
12 AxialLevelONRF's:
l Parameter N
E E
k SID)
_QNBE FAH 1455 1.145 0.000 1.713 0.011 1.017 Fq 1998 1.257
-0.027 1.703 0.026 1.057 l
Fz 2520 1.148
-0.027 1.697 0.020 1.053 i
24 AxialLevelONRF's:
Parameter N
E E
h SID)
ONRF FAH 2516 1.168 0.0034 1.6973 0.016 1.020 Fq 3162 1.281
-0.0055 1.6911 0.025 1.037 Fz 4200 1.138
-0.0103 1.6854 0.015 1.031 i
I i
t
)
i
{
I
Page 4 of 6 3.
Please identify which transients and accidents are affected by each change. Pmvide supporting analysis results for the limiting transients and accidents demonstrating how their acceptance criteria (DNB, kw/ft, pressure, etc. ) am met.
1 Answer: Future safety analyses will use FAH, Fq and Fz uncenainties with values greater than or equal to the values of the uncenainty factors shown in Table 3. The uncenainties j
calculated in Table 3 were developed using the same statistical data used in the development of l
ONRF's and are based on the FAH, Fq, and Fz uncenainty factor equations developed in i
Section 5.1 through 5.3 of DPC-NE-10(M. The difference between the ONRF's shown in Table 2 and the uncenainties shown in Table 3 is the statistical combination of the pin uncenainty with the assembly uncertainty.
l i
Table 3 I
DPC-NE-1004 and 24 Axial Level Uncertainties I
Without Engineering Hot Channel Factor Assembly Pin Total DPC-NE-1004
. Parameter Ms Uncertainty Uncertainty Uncertainties Uncertaintics +
FAH
-0.0029 0.0234 0.02 1.028 1.026 4
Fq 0.0(M3 0.0329 0.02 1.043 1.061 Fz 0.0091 0.0216 1.031 1.053
+ 12 axiallevel uncertainty
)
l The Fq and Fz uncenainty factors used in current and previous accident analyses bound values calculated for the 24 axial level model. Therefore, there is no impact to past, present or future safety analyses in which only Fq and Fz uncenainties are used. The increase in the FAH uncertainty over the topical value is of no safety concem for future analyses because a value greater than or equal to the 24 axial level FAH uncenainty will be used in these safety analyses.
Since the increase in the FAH uncertainty factor may be a result of the analysis of more complex reactor cores and not a result of transitioning to a 24 axial level model, the impact of l
this increase was assessed by performing a review of FSAR Chapter 15 accidents and their j
appropriate acceptance criteria which could be affected by an increase in radial (FAH) uncertainty factor. The following calculations are affected:
I
- Pin Pressure Creep collapse
}
1 The calculation of peak fuel enthalpy, Linear Heat Rate to Melt (LHRTM) kw/ft limits and i
primary and secondary peak pressures are not affected by the increase in radial uncertainty for the following reasons.
4
Page 5 of 6 Answer to Question 3 Continued Significant margin exists to the peak fuel enthalpy limit of 280 cal /gm, such that this parameteris not limiting. For the calculation of LHRTM kw/ft limits, a total (Fq) uncertainty and not a radial uncertainty is applied since this is local phenomenon. Primary and secondary system peak pressure response calculations are based on a balance between energy removed by the steam generators, and energy added from the reactor core. Since the pressure response is dependent on the rate of energy deposited from the scactor core, independent of the peaking within the core, local peaking uncenainties are not important. Therefore, the calculation of l
accident acceptance criteria and confinnation oflimits for peak fuel enthalpy, LHRTM and j
primary and secondary side peak pressure are unaffected.
l Peak pin pressure and creep collapse calculations assume a bounding radial uncenainty of j
l 1.036. Since this uncertainty bounds the 24 axial level value of 1.028, it can be concluded that -
the current and past analyses are unaffected.
L For the FSAR Chapter 15 accidents in which DNB is a concem, thermal analyses are performed to ensure that fel cimi integrity is maintained by ensuring that the minimum DNBR remains above the 95/95 DNBR limit. Two types of DN3R analyses are performed. Thermal i
analyses which are based on the Statistical Coir Design (SCD) methodology descdbed in j
l reference 1, and thermal analyses which are not based on this methodology (non-SCD DNBR analyses). The FSAR Chapter 15 accidents which are based on the SCD methodology are i
unaffected by the change in uncertainty factors. This is because radial and axial uncertainty factors of 1.04 and 1.053 are assumed in these accident analyses, which bound the 24 axial level uncertainty factors.
'Ihe FSAR Chapter 15 accidents which are based on the non-SCD methodology, along with the radial and axial uncertainty factors assumed in the analysis of each accident, are summarized in Table 4.
)
Table 4 Non-SLD 14 8 Uncertainty Factors 1
Accident Radial Uncertainty Axial Uncertainty Startup of an inactive RC N/A N/A i
Pump at an incorrect Temp.
l Steam Line Break 1.036 1.053 j
tocked Rotor 1.026 1.053 Rod Ejection 1.026 1.053 i
j From the data in Table 4 it is not immediately evident that past DNBR analyses performed based on non-SCD methodology were conservative. Therefore, each of these analyses are i
j discussed below.
i i
I
7 -
~ -
- -. ~
)
's Page 6 of 6 Answer to Question 3 Continued The startup of an inactive coolant pump at an incorrect temperature transient is a non-limiting transient which is bounded by the analysis of other FSAR transients. He FAH uncenainty assumed in the steam line break accident bounds both the 12 and 24 axial level uncenainties.
For the Locked Rotor and Rod Ejection accidents, the FAH uncenainty assumed in the accident analyses does not bound the FAH uncenainty calculated for the 24 axial level model.
However, since DNB is a function of both the radial and axial power distribution, the decrease in the 24 level axial uncenainty factor more than offsets the slight increase in radial uncertainty, resulting in an increase in DNB margin. Herefore, it can be concluded that previous accident analyses performed are conservative and the consequences of FSAR accidents previously evaluated and the rnargin to safety as defined in Technical Specifications is not reduced for the Locked Rotor and Rod Ejection accidents. In addition, it should be noted that margin retained between the 95/95 correlation and design DNBR limit used in accident analyses, (which is retained to account for unanticipated non-conservatisms) could also have been used to account for the slight increase in radial uncenainty.
In summary, the 24 axial level calculational uncenainty factors will be used in all safety related analyses in which a 24 axiallevel SIMULATE-3 model will be used. Ec slight increase in the FAH uncertainty factor relative to the topical value does not increase the consequences or reduce the margin to safety of accidents previously evaluated. His is because either the accident was non-limiting, the accident analysis employed conservative FAH uncenainty factors, or the tradeoff between the decrease in Fz uncenainty more than compensated for the increase in FAH uncenainty. The calculation of accident acceptance criteria for LHRTM, peak fuel enthalpy and peak pressure are also unaffected by I
the increase in radial uncenainty factor. Herefore,it can be concluded that the consequences of FSAR accidents previously evaluated and margin to safety as defined in the bases to Technical Specification is not decreased.
I l
l References i
- 1. " Duke Power Company Thermal-Hydraulic Statistical Core Design Methodology". DPC-NE-i 2005P-A, February 1995.
f i