ML060800503

Discovery and Correction of Errors in Topical Report

ML060800503
Person / Time
Site:	Palo Verde
Issue date:	03/09/2006
From:	Mauldin D Arizona Public Service Co
To:	Document Control Desk, Office of Nuclear Reactor Regulation
References
102-05435-CDM/TNW/GAM
Download: ML060800503 (8)

Text

i, David Mauldin Mail Station 7605 Palo Verde Nudear Vice President Tel: 623-393-5553 PO Box 52034 Generating Station Nuclear Engineering Fax: 623-393-6077 Phoenix, Arizona 85072-2034 102-05435-CDM/TNW/GAM March 09, 2006 U.S. Nuclear Regulatory Commission Attn: Document Control Desk Washington, DC 20555-0001

Subject:

Palo Verde Nuclear Generating Station (PVNGS)

Units 1,2 and 3 Docket Nos. STN 50-528, 50-529, and 50-530 Discovery and Correction of Errors in Topical Report The purpose of this submittal is to inform the NRC of recently-discovered and corrected errors in the analytical method described in topical report "Arizona Public Service Company PWR Reactor Physics Methodology Using CASMO-4/SIMULATE-3, September 1999." This report is listed in PVNGS Technical Specification 5.6.5.b as document no. 11 containing a description of analytical methods used to determine the core operating limits. This topical report was submitted to the NRC by APS in a Technical Specification amendment request, letter no. 102-04455, dated June 8, 2000, and approved for use at PVNGS by the NRC in Amendment No. 132 for Units 1, 2, and 3, dated March 20, 2001.

The topical report errors, caused by a data entry error, were the reported Doppler Power Coefficient (DOPC) bias, bias-based uncertainty, and unbiased ufcertainty, and the Fuel Temperature Coefficient (FTC) bias and uncertainty. The DOPC bias-based 95/95 uncertainty went from 20.6% to a corrected value of 11.0%, while DOPC unbiased 95/95 uncertainty went from 23.12% to 26.93%. The FTC uncertainty, which is based on the unbiased DOPC, went from 16.4% to a corrected value of 19.0%.

Corrected non-proprietary pages of the "Arizona Public Service Company PWR Reactor Physics Methodology Using CASMO-4/I;lMULATE-3" topical report are attached. None of the redacted proprietary information on these pages is affected by the corrections.

A member of the STARS (Strategic Teaming and Resource Sharing) Alliance Callaway

• Comanche Peak

• Diablo Canyon

• Palo Verde

• South Texas Project

• Wolf Creek Ade

U. S. Nuclear Regulatory Commission ATTN: Document Control Desk Discovery and Correction of Errors in Topical Report Page 2 The DOPC bias and bias-based uncertainty are not used in any safety related analysis.

The FTC bias and uncertainty are also not used in any safety related analysis. The DOPC unbiased uncertainty is one of the uncertainties currently used to calculate the minimum required boron concentration to satisfy lower mode shutdown margin requirements. Evaluation of the DOPC unbiased uncertainty error found that existing conservatism used in calculating the minimum required boron concentration bounded the DOPC unbiased uncertainty error, hence there have been no shutdown margin violations. Additionally, no core operating limits specified in the Core Operating Limits Report (COLR) were impacted by the error.

The data entry error that resulted in the topical report errors was made when inferring the Unit 1 Cycle 1 20% Rated Thermal FPower (RTP) measured DOPC. To determine the measured DOPC, control rods are inserted to add negative reactivity which is then balanced by a corresponding decrease in reactor power. The reactivity change associated with the control rod movement is equal to the reactivity added by multiplying the change in reactor power with the DOPC, after accounting for slight changes in the moderator temperature distribution. When entering the data, the rod worth associated with a 120 inch withdraw was skipped causing a mismatch between rod position and worth. This data entry error lead to the formulation of incorrect DOPC and FTC bias and uncertainty. This condition has been entered in the PVNGS corrective action program as CRDR 2803384.

No commitments are being made to the NRC by this letter. If you have any questions, please contact Thomas N. Weber at (623) 393-5764.

Sincerely, CDM/TNW/GAM

Enclosure:

Corrected Non-Proprietary Pages for Topical Report "Arizona Public Service Company PWR Reactor Physics Methodology Using CASMO-4/SIMULATE-3, September 1999" cc:

B. S. Mallett NRC Region IV Regional Administrator M. B. Fields NRC NRR Project Manager G. G. Warnick NRC Senior Resident Inspector for PVNGS

ENCLOSURE Corrected Non-Proprietary Pages for Topical Report "Arizona Public Service Company PWR Reactor Physics Methodology Using CASMO-4/SIMULATE-3, September 1999" Pages:

17 76 77 78 79

Table 1-2 List of 95/95 Tolerance Limits (Bias +/-Uncertainty)

Parameter Bias 95/95 Units*

Uncertainty HFP Core Reactivity (pcm) 331.5 226.8 Absolute HZP Core Reactivity (pcm)

Bias = (0.157 :< a) - (60.136 x 3) +

621.7 Absolute (BOC only) 322.427

Where, a = Number of Fresh Erbium Rods

= BOC Core Average BU in GWD/MT Isothermal Temperature

-0.28 1.52 Absolute Coefficient (pcmPF)

Control Rod Worth Relative

- Bank Worth 0.8%/

8.3%

- Total Worth 1.0%

7.1%

- Dropped Worth

[

I%

[

]

- Ejected Worth

-1.3%

[

]

- Net (N - 1) Worth 1.0%

7.1%

Inverse Boron Worth (ppm/

-3.16%

13.49%

Relative

%Ak/k)

Doppler Power Coefficient Bias-5.7014-1.115*CAB(GWDA/T) 20.6%

Relative (pcm/% power)

+ 3.87E 03*rW+

11.0%

Bias = -0.186 + 9.02E-03*CAB(GWD/

MT) + 1.65E-03*P(%)

where CAB is the core average burnup Fuel Temperature Coeffi-

-08%

-6.4 Relative cient (pcmP°F)

-1.8%Yo 19.0%

Local Pin Power

[1

[

] Relative (Pin-to-Box)

Calculated Assembly Relative Peaking

- Fq (box) 0 O 5.34%

- Fr (box) 0 YO 3.25%

-Fxy (box) 0 3.69%

I 17

Statistical Analysis The Wtest for normality was performed on the DOPC observed differences, then a determination as to whether or not the DOPC is a function of power level or power level and core average bur-nup. Once that was determined a bias and uncertainty (K(95/95*S)) was calculated.

Table 4-20 shows the average, standard deviation (S), and K(95/95)*S.

Table 4-20 shows that the normally distributed DOPC relative difference data has a K(95/95)*S of 23.12% 26.93%. Since this is quite large a determination was made as to whether or not the DOPC is a function of power level or power level and core average burnup.

The data showed that the functionalization with the smallest uncertainty is against core average burnup, but it won't be used since the data is basically made up of beginning-of-life (BOL) data and is not entirely appropriate for reload cycles.

The next best was a combination of power and core average burnup. This one will be used since

[

]. Table 4.21 shows the statistical results for DOPC.

By using the combination of power and core average burnup the uncertainty is reduced from 23 1 % 26.9% to 20.6% 11.0%.

Statistical Results The DOPC tolerance limits (from Table 4.21) are:

5.704

.1. 15 >

5*,AB(G3ADT),

3.87E 04P.(3/4) -2 2O.6%,

-0.186 + 9.02E-03*CAB(GWD/MT) + 1.651-E03*P(%) + 11.0%,

where CAB is the core average burnup 76

Table 4-19 Doppler Power Coefficient Comparisons Unit Cycle Core Power Measurement SIMULATE-3

% Difference Average

(% HFP)

(pcm/%power)

Calculation 100*(M - C)/C Burnup (pcm/%power)

(GWD/MT) 1 1

0.2

1.

54i3

-15.01 0.99%;

-0125 18.4

-12.82

-14.590%

1 1

0.255 53.3

-11.60

-12.52

-7.348 1

1 0.988 80.5

-10.15

-10.56

-3.883 1

1 2.142 95.3

-9.11

-9.17

-0.654 1

2 12.718 95.0

-10.98

-9.86 11.359 2

1 0.541 50

-10.89

-12.61

-13.640 2

1 1.080 96.3

-7.86a

-9.71

-19.053 2

2 9.543 94.4

-10.46

-10.29 1.652 3

1 0.658 96.0

-9.28

-9.43

-1.591 a.This point is abnormally low. Compare -7.86 to the other two Cycle I HFP data points of -9.11 and -9.28. This point is not negative enough and is eliminated from the data base.

77

Table 4-20 Doppler Power Coefficient Statistics for Relative Differences pcm/%power Mean 1.663%

100*(Meas - Calc)/Calc

-3.587%

Standard Deviation (S) 8.450%

Number of Data Points 8

Degrees of Freedom 7

WValue 0.964 0.954 Critical Value(s) 0.818 Normal Distribution?

Yes K95/95 a (95/95 Tolerance Factor) 3.187 K95/95*c; 23.12%

26.93%

I

a. Reference 27 for n = 8 and 95/95 eenfidenee tolerance interval.

Table 4.21 Functionalization for the DOPC Relative Differences I

Functionalization R2 Standard Degrees of K(95/95)a Uncertainty Error (%)

Freedom 95/95

(%)

Power and Core 04° 5

3.70811.0%

Average Burnupb 0.91 2.97 11.0%

a.Reference 27 b.Bias (%) -

5.701 + 9.02E53GAB(G3Aq).\\T)

+

3.87E 03*P(%)

Bias (%) = -0. 186 +I 9.02E-03*CAB(GWDAMT) + 1.65E-03*P(%)

I 78

4.7 FUEL TEMPERATURE COEFFICIENT Statistical Analysis The fuel temperature coefficient (FTC) is related to the DOPC by the relationship:

dp = dp X dT dP dTf dP where p = core reactivity Tf= fuel temperature P = core power The term dp/dP has been assigned a bias and an uncertainty, but neither dp/dTf nor dTjdP can be evaluated separately. One way of assigning biases and uncertainties is to assign biases and uncer-tainties equally to dp/dTf and dT/dP. The data base of DOPCs is used, without regression analy-sis versus power and core average burnup. The bias and uncertainty become:

FTC Bias = Average /2 FTC Uncertainty = K(95/95)*S /X2 Another method is to assign a bias to dTjdP and uncertainty to dp/dTf (FTC) using the fit of (Meas - Calc)/Calc with respect to power and core average burnup. The FTC bias becomes zero and the uncertainty becomes K(95/95)*S of the fit.

Assigning biases and uncertainties equally to dp/dTf and dTpdP is slightly more conservative, and [

]. Therefore biases and uncertainties were assigned equally to dp/dTf and dTfjdP.

Assigning biases and uncertainties equally to dp/dTf and dTj'dP yields the following:

FTC Bias = Average /2-1.66 3.591 /2 -

0.839

.80%

FTC Uncertainty = K(95/95)*S /.F2 = 2342-26.931 /X2/ = 16.35% 19.04%

Statistical Results The tolerance limits for fuel temperature coefficient are:

-0.84 16.4%

-1.8

+/- 19.0%

1. From Table 4-20.

79