ML20056E396

Safety Evaluation of RXE-90-006-P, Power Distribution Control Analysis & Overtemperature N-16 & Overpower N-16 Trip Setpoint Methodology. Methodology Acceptable

ML20056E396
Person / Time
Site:	Comanche Peak
Issue date:	08/05/1993
From:	Office of Nuclear Reactor Regulation
To:
Shared Package
ML20056E389	List: ML20056E388 ML20056E396
References
NUDOCS 9308230306
Download: ML20056E396 (4)

Text

.

j'* " l

[ # 2 UNITED STATES 2%  ! NUCLEAR REGULATORY COMMISSION

%kY/

WASmNGTON. D.C. 2554m SAFETY EVALUATION BY THE OFFICE OF NUCLEAR REACTOR REGULATION RELATING TO TOPICAL REPORT RXE-90-006-P. "P0WEP DISTRIBUTION CONTROL ANALYSIS i

AND OVERTEMPERATURE N-la AND OVERPOWER N-16 TRIP SETPOINT METHODOLOGY" TEXAS UTILITIES ELECTRIC COMPANY >

COMANCHE PEAK STEAM ELECTRIC STATION. UNITS I AND 2 DOCKET NOS. 50-445 AND 50-446 4

1.0 INTRODUCTION

By letter dated February 28, 1991 Texas Utilities Electric Company (TV Electric) submitted topical report RXE-90-006-P, " Power Distribution Control Analysis and Overtemperature N-16 and Overpower N-16 Trip Setpoint Methodology." This report contains the design bases and the methodology for ,

, the calculation of the Overpower N-16 and Overtemperature N-16 setpoints, including the treatment of the core power distribution effects and trip channel n. certainties. This methodology is needed to support reload design licensing and operation of the Comanche Peak Steam Electric Station (CPSES) i Units 1 and 2. Additional information was submitted by letters dated  :

December 20, 1991 and May 28, 1993. >

\

Power distribution control is utilized to assure that the axial power distributions and associated peaking factors during normal operation are  :

maintained within the limits assumed within the safety analyses. This power i

distribution control analysis will allow TU Electric to establish axial flux  ;

difference target bands and define normal operating power. distribution shapes i for input to loss of coolant accident (LOCA) and non-LOCA transient analyses.  ;

The basic concept of the power distribution control procedures is that the  !

variation in the core axial power distribution during reactor operation can be ,

controlled by maintaining the axial flux difference with predefined target r bands. The Overpower N-16 and Overtemperature N-16 trip setpoints are j

designed to protect against fuel centerline melting, departure from nucleate boiling (DNB), and hot leg saturation during postulated transients. T'u Electric uses a method that is very similar to a method which was approved by the NRC.

2.0 EVALUATION

! The basic concept of the methodology, which relies on establishing axial i target bands in order to define normal power distribution shapes, is well established and has been used throughout the PWR industry for many years. TU Electric adopted the principal attributes of Siemens approved methodology for generating axial power distributions for use in establishing the Overpower N-16 and Overtemperature N-16 trip setpoint and power distribution control analysis. The power distribution control analysis is used to establish the axial flux difference bands, the Fq augmentation factor, T(z) and the normal 9309230306 930805 PDR ADOCK 05000445 P PDR

i

. l operating axial power distributions required as input to LOCA and DNB '

analyses.

Since TU Electric methodology described in the topical report and the additional information differed from the Siemens methodology in some aspects, this review could not be simply a benchmarking type review. Many aspects were i identical to the Seimens methodology, and in those cases documenting the TV

~

Electric results against measured data or against Seimens predictions was <

sufficient. In other areas more evaluation was necessary. }

The augmentation factor T(z) is developed in the identical manner as Seimens.  !

The TV Electric methodology for determining the "one-hour" axial offset operating band, for operating outside the target band, is also basically identical to the method used by Seimens. The difference between the two approaches are discussed below. ,

i TV Electric uses CASMO-3 to generate fuel assembly cross sections. TV Electric's use of the CASMO-3 code was reviewed as part of RXE-89-003-P,

" Steady State Reactor Physics Methodology," which was approved by the staff. .

Thus, this use is acceptable. TU Electric normalized the core physics models to either SIMULATE-3 calculations or measured data while Siemens uses primarily normalization to measured data. TU Electric has demonstrated capability of using SIMULATE-3 for licensing basis calculations in approved ,

topical report TXE-89-030P. The TV Electric methodology uses an expanded set  :

of load follow cases for determining Eq. After calculating the augmentation l factor T(z), TU Electric adds an arbitrary 1% conservatism to determine the  !

limiting T(z) distribution, while Seimens uses a variable conservatism of at l least 1% to produce a limiting distribution consisting of straight line  ;

segments. The 1% conservatism is acceptable. The TV Electric methodology  ;

will use interpolation on the exposure dependent t(z) distributions instead of '

using one distribution throughout the cycle. This is consistent with other methodologies and is therefore acceptable.

To validate their use of the methodology, TU Electric has presented T(z) l l distribution calculations for CPSES Unit 1 Cycle I and limited comparisons to ,

calculations performed by Siemens. Additional benchmark comparisons to i support the TU Electric methodology application were preser.ted in Reference 3.

These include: (1) direct comparison of measurements obtai plant transient to calculated results using the one dimens,ned ional during XTGPWR CPSES model, (2) determination of the sensitivity of the calculated t(z) distribution to a variation of 100 ppm in critical boron concentration, (3) direct comparison of a TV Electric calculated t(z) distribution to a Siemens calculated V(z) ,

distribution for a plant without natural uranium axil blankets (NUAB), and (4) direct comparison of a series of TV Electric calculated T(z) distributions to Siemens calculated V(z) distributions for a plant containing NVAB. '

The power distribution comparisons of measured vs. predicted show excellent '

agreement in all cases. Likewise comparisons of measured and predicted rod worths were good. Comparisons of measured and predicted boron letdown curves did not show as good agreement as might be expected. The results were much >

better for Cycle 2 than for Cycle 1, however. To evaluate the effect of a 100

• t h

(

ppm increase in boron concentration on the determination of the composite T(z) distribution, the methodology described in the topical report was applied to Cycle 4 of CPSES Unit I. A base case was calculated and a composite T(z) ,

distribution was determined. The calculations were rerun with the critical boron concentration 100 ppm higher. and the composite T(z) was generated. A comparison of the composite T(z) distributions for the two cases was made.  !

The maximum difference was 0.002, thus demonstrating that the composite T(z) distribution is insensitive to changes in the boron concentration of up to 100 ppm.

Rather than do a side-by-side comparison in which both TU Electric and Seimens -

calculated the same core, TV Electric did a comparison of their composite T(z) distribution and that of Seimens for similar cores and included the sensitivity to the presence of NUAB. The TU Electric calculations demonstrated that the sensitivity of the T(z) distribution to the presence of <

NUAB is consistent with the sensitivity studies presented by Siemens. In the final benchmark TV Electric compared their T(z) distribution and Siemens V(z) distribution for similar cores with NVAB. The character having an effect on the distribution were identified by Siemens as power density, active core height, control rod worth and number of assemblies with NUAB. TV Electric compared the H.B. Robinson Unit 2 Cycle 12 and the CPSES Unit 1 Cycle 4 cores.

Both contain a significant number of assemblies with NUAB. They are 12 foot cores and have similar power density. However, the cores.have different  ;

control bank worths. TV Electric performed the calculations for CPSES assuming a range of control rods worths and therefore their results could be compared with the Siemens calculations for H.B. Robinsor.. When the TV Electric and Siemens results were compared they showed good agreement. TU Electric intends to perform the analyses on a cycle specific basis to account for small effects. Furthermore the 1% conservatism will be applied to the limiting T(z) distribution. 1

3.0 CONCLUSION

l Based on the staff evaluation in Section 2.0 above, the staff concludes that the Topical Report TXE-90-006-P " Power Distribution Control analysis and i Overtemperature N-16 and Overpower N-16 Trip Setpoints" is ;tteptable.

4.0 MFERENCES

1. Letter, William J. Cahill, TU Electric to NRC dated February 28, 1991,

" Reload Analysis Program, RXE-90-006-P, Power Distribution Control Analysis and Overtemperature N-16 and Overtemperature N-16 Trip Setpoint Methodology."

2. Letter, William J. Cahill, TV Electric to NRC, dated December 20, 1991, i

" Response to Request for Additional Information Regarding Topical Report i RXE-90-0006, Power Distribution Control Analysis and Overtemperature N-16 l and Overpower N-16 Trip Setpoint Methodology " l 1

3. Letter, William J. Cahill, TV Electric to NRC, dated May 28, 1993, i Additional information on RXE-90-006.

l I

i 1

4. Letter, William J. Cahill, TU Electric to NRC, dated July 13, 1989, i transmittal of Topical Report RXE-89-003-P, " Steady State Reactor Physics Methodology." - '

Principal Contributor: H. Chatterton Date: August 5, 1993 i

b

. , _~'

b l

t

___ _ _ _ _ _ _ _ _ _ _ - _