ML20034D462
| ML20034D462 | |
| Person / Time | |
|---|---|
| Issue date: | 08/25/1992 |
| From: | Rosalyn Jones Office of Nuclear Reactor Regulation |
| To: | Tritch S WESTINGHOUSE ELECTRIC COMPANY, DIV OF CBS CORP. |
| References | |
| NUDOCS 9209030298 | |
| Download: ML20034D462 (6) | |
Text
d August 25, 1992 N
Mr.
S.
R.
Tritch, Manager Engineering Technology Westinghouse Electric Corp.
P.
O.
Box 355 Pittsburgh, PA 15230-0355
Subject:
Request for Additional Information for The Review of the Topical Report WCAP-12472-P, " BEACON"
Dear Mr. Tritch:
We have completed, with the assistance of our consultants at Brookhaven National Laboratory, the initial review of the subject topical report submitted by Westinghouse by letter dated May 21, 1990.
We find that some additional information is required from Westinghouse in order to complete the evaluation. This information is indicated in the enclosed list of questions developed by our consultants during the initial review.
If you have any questions with regard to this request for information, please contact Howard Richings of my staff at 301-504-2888.
Sincerely,
@ Signed i '
Robert C. k.
Robert C.
Jones, Chief Reactor Systems Branch Division of Systems Technology
Enclosure:
As stated DISTRIBUTION:
Central Files SRXB R/F /
RJones LPhillips HRichings HRichings R/F SRXB: DST T
- DST RJ'NES HRICHINGS:jh HI IPS O
8 M /92 8/.
/92 8/t.y/92 mJh DOCUMENT NAME: LTR TRITCH.HR
.,j
(
g g o,1 9 2I E
170C44 e
i 7
1 I s y,
ADDITIONAL INFORMATION REOUIRED FOR THE REVfEW OF THE WESTINGHOUSE TOPICAL REPORT WCAP-12472-P 1.
How will the parameters and component uncertamties used in determining the BEACON enthalpy-rise and power-peaking factor uncertamties be evaluated for plant and cycle-specific application? Which BEACON parameters will be plant / cycle-specific?
2.
Do the selected plants / operating-states and thermocouple data (Figure 5-1) used to determine the Um and U uncemmties bound the expected BEACON applications?
n 3.
In Step-2 of the BEACON calibration, how is the three-dimensional nodal power distribution adjusted to reproduce the measured axial off-set?
I 4
Describe how the " axial off-set rod" is used to insure agreement between BEACON and the measured axial off-set.
5.
In the stttistical evaluation of Section-5.2.3, how were the 20 assemblies used in determiniag the Un assembly power uncertainty selected and does this selection include both monited and unmonitored locations?
6.
Is the one-dimensional SPNOVA model used in determining margin to care limits or setpoints? If so, describe the method used to collapse the three-dimensional model and provide appropriate model qualification for the collapsed model.
7.
How is the effect of changes in the inlet temperature on the thermocouple calibration factors accounted for?
3.
In the Monte Carlo uncertainty pr,opagation using Equation (5-1), are the two occurrences of the mixing factor MF (r7c,6c) varied independently? If not, please justify.
9.
Discuss the method used to determine the two-dimensional tolerance factor H(L,h) of Equation (5-3).
10.
What input uncertainty values are used in Equation (5-1) for 3.uc and how wil' these be evaluated in plant / cycle-specific applications?
11.
Equation (5-7) is based on the conclusion that the assembly power uncertainty Un is proportional to the square root of the thermocouple uncertainty a.
However, at large values of a Un is believed to be proportional to a which implies that Equation (5-7)is nonconservative. Provide data to demonstrate that Equation (5-7) is valid for the larger values of a expected between calibrations, at reduced powers and/or.with high instrumentation failure rates.
1
4 How are Equations (3-20) and (3-21) used in the BEACON methodology?
12.
How is the noneservative BEACON underprediction of Figure 5-4 accounted for?
13.
How do the P and P power distributions of Equation (5-15) compare, and are the 14 3
i differences consistent with the value of Un determined in Section-5.37 7
Is the Equation (5-18) axial uncertainty allowance applicable for Bank-C insertion? If not, 15.
how will the Bank-C correction be determmed?
16.
The Equation (5-19) for U.m and U is not applicable to the case of systematic (non-n random) biases. How will these be included in the BEACON limits calculatio In the application of Equation (5-1), how is the error introduced by the spline 17.
interpolation to unmonitored locations accounted for?
The reduction in assembly power uncertainty Un that results when the tolerance factor 18.
K increases from K=0 to K=10 is not believed to be due to an increase in BEAC accuracy, but rather due to a decreased sensitivity to the individualinstrument readings.
When the modeling parameter K = 0, the BEACON solution agrees exactly with the
" measured values" at the instrument locations. As K increases the BEACON no longer forced to agree with the measured data and becomes a weighted average of the i
neighboring measurements. As K tends to infinity BEACON approaches a least-squares planar fit of the measured data. Consequently, BEACON is most sensitive to variation This l
in the measurements when K = 0 and decreases in sensitivity as K increases.
reduced sensitivity with increasing K results in a reduced statistical variance in the as,ernbly powers and a reduced uncertainty Ug. The selected value of K = 10 provides a close to minimal sensitivity to individual thermocouples, and results in a factor of
- 1.5-2.0 reduction in the 95/95 assembly power uncertainly (Figure 5-2).
1 The calculations of Section-5.3.2 and Figure 5-2 demonstrate that a BEACON solution that is not constrained to reproduce the local point measurements is less sensitive to the individual measurements, and has a smaller variance when the measurements are varied.
These calculations do not demonstrate that the K = 0 BEACON solution which agrees exactly with the measured data is less accurate than the K = 10 solution which is an average of the local flux measurements. It is recommended that the K = 0 95/95 upper tolerance limit be used in the BEACON uncertainty analysis, or additional data be i
provided to support the use of the reduced uncertainty allowance.
l As the temperature rise up an assembly decreases with decreasing power,' the 19.
thermocouple uncertainty increases substantially. How do the Equations (4-3) and (4-4) '
account for uncertainty in the extrapolation to lower power?
20.
Are the thermocouples used at power levels higher than the power at calibration? If so, discuss the application of Equation (5-4) to this situation.
De Un defined by Equation (5-7) includes a statistical allowance for the finite number 21.
of Monte Carlo trials, but does not include an allowance for the required 95 % confidence i
level on the input standard deviations. How will this additional uncertainty allowance be included in Um and U ?
n 22.
In the BEACON criticality tests of Section-6.1, was the model normalized to match the measured axial off-set? If so, how was this normalization performed and what effect does it have on-theralculation-to-measurement comparisons?
23.
Do the thermocouple mixing factor calibration errors have any non-random dependence, for example, on core location, fuel assembly / spacer design or assembly power, which results in a systematic underprediction of the limiting assemblies? If so, how will this bias be accounted for?
24.
Is the thermocouple calibration data and statistics presented in Chapters 4-6 typical of all W plants? What plant-to plant variation is expected and how will this be accommodated?
25.
Are the normalized calibration factors defined by Equation (6-1) distributed normally?
26.
What value of b will be used in Equation ('4-4)? Will this value be plant / cycle-specific?
27.
Were the thermocouple mixing factors at time-t, MF(N,t), determined by the method of Section-3.3.2 using an incore flux trace measurement? If not, how does the ratio R(t) of Equation (6-1) determine the mixing factor uncertainty?
28.
How will the rod-bowing penalty be incorporated in the new technical specifications?
29.
Describe the BEACON calculation of DNBR. Has this DNBR evaluation method been approved for the intended BEACON applications?
30.
Describe the methods that will be used to confirm that operation is within the LOCA and LOFA limits and the adequacy of the OPAT and OTAT setpoints, in order to accommodate the wider range of initial conditions that result from the relaxation of the axial-offset control and quadrant tilt limits. If these models and methods have not been approved, provide the appropriate qualification.
31.
The Standard Technical Specification-3.1.3.1 allows operation with one (trippable) inoperable rod (due to causes other than excessive friction or mechanical interference) or one (trippable) misaligned rod, provided that: (1) a shutdown margin analysis and reevaluation of accidents is performed and (2) the power level and flux setpoint are reduced.
The proposed BEACON Technical Specification-3.1.3.1 allows operation with one or more (tnppable) inoperable rods at reduced power, but does not require a shutdown margin analysis or accident reevaluation. Also, the BEACON Technical Specihina does not explicitly address the case of trippable but misaligned rods. Please justify these differences.
Technical Specification changes should only be directly related to BEACON requirements.
Other changes, not clearly solely related to BEACON, should be avoided.
32.
In Technical Specification-3.1.3.2, justify the change in the applicability of this specificanortfrom Mode-1 and Mode-2 to Mode-1 above 50% of rated thermal power.
33.
In Technical Specification-3.1.3.2 concerning the rod positions, justify the replacement of the movable incore measurements (Action-al) and the rod position indication operability (Action-b1), by the verification that the BEACOM parameters are within the limits of Technical Specification-3.2.6.
34.
Justify the elimination of the reduction-in-power action items (b2 and b3) of the Standard Technical Specification-3.2.4 when the quadrant power tilt exceeds 1.09.
Does a comparison of BEACON local predicted-to-measured assembly power provide a more accurate power distribution diagnostic than quadrant power tilt?
35.
The proposed inclusion of action items in the Core Operating Limits Report (COLR) is inconsistent with the accepted use of the COLR. It is suggested that the current technical specifications be included in the proposed BEACON specifications, with the applicability statement indicating that they are only applicable when BEACON is inoperable. The proposed BEACON technical specifications will require some revision as a result of this change.
36.
During the period between incore flux calibrations, the core-average axial power distribution is measured using the excore detectors. The core-average axial is taken to be the same as the axial power distribution of the peripheral assemblies, which is inferred from the excore detector measurements. Since the core-exit thermocouples only provide (axially-integrated) radial power measurements, no local or assembly-wise axial measurements are made during this period between calibrations. In order to provide local monitoring of the axial power distribution that is comparable to the present incore surveillance, it is recommended that the proposed calibration interval be reduced from 180 to 31 EFPD.
37.
How are failed thermocouples identified? If failed thermocouples are returned to service, how willit be assured that the performance of these thermocouples is consistent with the uncertainty value assumed in the BEACON analysis?
38.
Technical Specification Surveillance Requirement-4.3.3.12.4 and COLR Criteria-2.4.4 imply that BEACON may be operable with less than 25% of the thermocouples operable, or with the lack of a che*dw=rd's knight move distribution of thermocouples. Please justify the application of BEACON under these minimal conditions.
39.
A thermocouple detector system that only provides 25 % availability does not provide the level of confidence required in order to relax the RAOC/CAOC power distnbution
{
controls and the conservatism implicit in these methods. It is therefore recommended that the number on operable thermocouples required for BEACON operation be increased from l
l 25% to 50% (with a knight's move distribution).
40.
How does the BEACON methodology treat core loadings which include fuel designs from i
multiple fuel vendors?
Discuss the impact of these types of fuel loadings on the BEACON uncertainty analysis.
r i
i l
f r
I f
1 4
I l
'l