ML20215C422

Rod Swap Methodology Rept for Startup Physics Testing

ML20215C422
Person / Time
Site:	Mcguire, Catawba, McGuire, 05000000
Issue date:	08/31/1985
From:	Randles J, Tomonto R DUKE POWER CO.
To:
Shared Package
ML20215C375	List: ML20215C373 ML20215C402 ML20215C422
References
DPC-NE-1003, TAC-62981, TAC-62982, NUDOCS 8610100274
Download: ML20215C422 (18)

Text

- _

4 4 1

I McGuire Nuclear Station Catawba Nuclear Station i

Rod Swap Methodology Report for Startup Physics Testing DPC-NE-1003 August 1985 a f

J. H. Randles R. J. Tomonto t

Duke Power Company Nuclear Production Department Nuclear Engineering f

i I 8610100274 861001

. PDR ADOCK 05000369 p PDR l

l l

e

1. Introduction This report describes the calculational procedures used to develop the rod swap constants and describes the measurement procedure used to deter =

mine the inferred bank worths. This paper also presents a comparison between the calculated and inferred bank worths for McGuire 1 Cycles 2 and 3, and McGuire 2 Cycle 2. g Inorodetoperfersthe"ControlRodWorthMeasurement-RedSwapTest Procedure" (2), the following information must be provided to the station.

This information shall include the bank worths, critical heights and a's.

The critical heights and o's are used to calculate the inferred bank worth of each control and shutdown bank, as reduced from information following the iso-reactivity interchange with the reference bank.

This r2 port presents the calculated procedures used to derive these parameters. The calculations as performed in this pcocedure utilize the approved physics codes and methodologies described in reference (1).

et

,e 9

2. Definitions The following is a list of the constants needed by the plant, to perform the rod swap procedure. These include:

• [-*

Predicted reactivity worth of each control and shutdown bank, when inserted individually into an otherwise unrodded core.

• h - Predicted critical position of the' reference bank after interchange with bank x, starting with the reference bank at 0 steps and bank x fully withdrawn.

• a * -

A correction factor which accounts for the effect of bank x on the partial integral worth of the reference bank, equal to the ratio of the integral worth of the reference bank from h to the fully withdrawn position with and without x in the core.

In addition, included is a list of constants and their definitions as used in this report.

• W - Measured rod bank worth of bank x from rod exchange

• (g - Measured rod bank worth of reference bank

• (Ap), - The measured integral worth of the reference bank from the measured critical position (h") to the fully withdrawn position.

• h"* - The measured critical position of the reference bank after interchange with bank x.

3. Measurement Procedure With an initial configuration of all rods out, hot zero power, the integral worth of the reference bank is measured using the standard boration/ dilution technique. The reference bank is the bank that is predicted to have the highest integral worth. All other banks are then individually exchanged with the reference bank at constant boron conditions.

The worth of each bank is then the amount of reactivity change caused by the withdrawal of the reference bank to its new critical height.

The rod bank worth is inferred from the measured reference bank worth and the measured reference bank height using the following equation:

W[=W"f, - a, M ,

where the above terms are defined in Section 2.0 of this report.

This test procedure was performed consistent with the approved Westinghouse Rod Swap Methodology.

d-6

?

i l

l l

i

4. Calculational Procedure This calculation is performed using EPRI-NODE-P to model core conditions during the rod swap procedure. The following procedure describes the method of data generation:

1. Calculate the integral bank worth at HZP, ARO critical boron. Insert one bank at a time with no overlap and calculate the bank worth as the difference between ARO and the bank fully inserted condition.

(The calculated highest worth bank will be considered the reference bank.)

2. With the reference bank fully inserted, calculate the critical boron concentration. (The reference bank in boron concentration is used in predicting the predicted rod worth - )

3. Using the above calculated critical boron concentration for the reference bank, the new integral bank worths at HZP are determined.

. These values correspond to the predicted worth for each bank ( ).

The reference bank should be inserted in approximately six (6) step increments such that a plot of the integral worth of the reference bank can be obtained. (As should be noted, the K,gg with the reference bank inserted, is referred to as the base K,gg).

4. In order to calculate the critical height, the. core is modeled with the measured bank fully inserted. The reference, bank is then inserted in approximately six (6) step increments. The critical height (h ) of the reference bank is then calculated by plotting the steps inserted versus K,gg. When K,ff equals the base K,gg, the critical height is found. (This can also be done by linear interpolation between two data points.)

5. In order to calculate a for each bank position, the following expression is used:

Integral Worth of the reference bank from h to the fully withdrawn position with bank x inserted in the core a=

Integral worth of the reference bank from h to the fully

, withdrawn position without bank x inserted in the core

5. Results Table 1 presents a comparison between Duke's predicted and inferred bank worths. A review of the available data from McGuire 1 Cycles 2 and 3, and McGuire 2 Cycle 2, identifies a mean difference of -7.52 pcm or

-2.27% between Duke's predicted and inferred bank worths. A similar comparison using Westinghouse's predicted and inferred bank worths [

identifies a mean difference of 34.63 pcm or 7.32%.

Table 2 presents a comparison between Duke's and Westinghouse's predicted minus inferred total bank worth difference. In reviewing this information, Westinghouse appears to overpredict on the predicted worth, while Duke's trend appears more evenly scattered.

Table 3 identifies a comparison between Duke and Westinghouse measured minus predicted total critical heights. When evaluating the sum of the absolute value of the differences, Duke was 110 steps closer to the measured critical heights than Westinghouse. When considering the sum of the differences, Duke was 146 steps c.'.eser to the measured critical heights than Westinghouse. The standard deviation of the differences between the measured critical heights and Duke's calculated critical

, heights is 8.82. The standard deviation of the differences between the measured critical heights and Westinghouse's calculated critical heights is 10.46. Duke's predictions appear to be more consistent than Westing-house's prediction.

Table 4 presents some typical a values as calculated for McGuire 1, Cycle 3. ,

i l

l l

l l

l l

6. Conclusion Reference to the Rod Swap Test Procedure (2) identifies the specific

. acceptance criteria. In order to satisfy this procedure the following conditions must be met:

(a) The absolute value of the percent difference between the measured and predicted integral worth for the reference bank is 1 15%.

(b) The absolute,,value of the percent difference between the sum of the measured and predicted integral worth for all the banks is 110%.

(c) For all RCC banks other than the reference bank, either:

(i) the percent difference between the inferred and predicted worth for each individual bank is 1 30%

EE (ii) lW - W l 5 200 pcm for each bank, whichever is greater.

These criterias were found acceptable using both Duke's and Westinghouse's predicted values.

Based upon these measurements, the procedure developed to predict critical heights, a's and bank worths for rod swap yielded respits that were as good as the results obtained using Westinghouse critical heights, a's and bank worths.

Table 1 Duke Predicted and Inferred Bank Worth Duke Predicted Duke Inferred Difference

• Difference **

Unit / Cycle Bank Worth (PCH) Worth (PCM) (PCM) (%) _

1/2 CA 289 301 -12 -4.0 CB 557 606 -49 -8.1 CC 786 788 -2 0.3 CD 616 566 50 8.8 SA 473 546 -73 -13.4 SB 443 479 -36 -7.5 SC 370 354 16 4.5 SD 362 . 374 -12 -3.2 SE 223 237 -14 -5.9 4

Total 4119 4251 -132 -3.1 4

Hean - -

-14.67 -3.17 Standard Deviation - -

35.94 6.80

- i 1

1 J

l i

I

e Table 1 (Cont.)

Duke Predicted Duke Inferred Difference

• Difference **

Unit / Cycle Bank Worth (PCH) Worth (PCH) (PCH) (%)

1/3 CA 311 305 6 2.0 CB 657 609 48 7.9 CC 789 745 44 5.9 4 CD 488 466 22 4.7 SA 269 303 -34 -11.2 l SB 856 779 77 9.9 l SC 394 373 21 5.6 SD 395 383 12 3.1 SE 429 392 37 9.4 1

! Total 4588 4355 233 5.4 Hean - -

25.89 4.14 Standard Deviation - -

31.16 6.34

. h

{

e

1 Table 1 (Cont.)

4 Duke Predicted Duke Inferred Difference

• Difference **

Unit / Cycle Bank Worth (PCM) Worth (PCM) (PCM) (%)

2/2 CA 437 459 -22 -4.8 CB 413 452 -39 -8.6 CC 858 871 -13 -1.5 CD 654 664 -10 -1.5 SA 327 430 -103 -24.0 SB 425 480 -55 -11.5 SC 354 375 -21 -5.6 SD 355 374 -19 -5.1 SE 270 292 -22 -7.5 Total 4093 4397 -304 -6.9 Mean - -

-33.78 -7.79 Standard Deviation - -

29.42 6.87 i

l

Table 1 (Cont.)

Evaluation of All Available Data Duke Calculated Westinghouse Calculated Difference Differences Difference Differences

/

(PCH) (%) (PCM) (%)

Hean -7.52 -2.27 34.63 7.32 l Standard Deviation 40.07 8.13 43.81 8.59 NOTE:

1- Di f ference = Predicted - Infer:cd P - I x 00

• • Difference (%) = g W

> s

Table 2 Comparison of Differences Between Duke and Westinghouse Predicted and Inferred Bank Worth Duke Westinghouse Total Rod Worth Total Rod Worth Difference -

Difference '

Unit / Cycle (PCM) (%) (PCM) (%)

1/2 -132 -3.1 220 5.1 1/3 233 5.4 473 10.8 2/2 -304 -6.9 242 5.5 NOTE:

• Difference = Predicted - Inferred x 100

• • Difference (%) = g W ,

~

i l

l l'

t

(

E b Table 3 Duke Predicted and Measured Critical Heights Critical lieight (Steps) Duke Calculated Westinghouse Calculated Unit / Cycle Bank Measured Predicted Difference (Steps) Difference (Steps) 1/2 CA 83 88 -5 -14 CB 197 195 2 +15 CD 183 196 -13 -5 SA 191 187 4 14 SB 156 157 -1 3 SC 144 158 -14 3 SD 147 156 -9 4 SE 86 92 -6 -11 I - -

-42 9 I of absolute value - -

54 69 Standard Deviation - -

6.63 10.60

. s s

Table 3 (Cont.)

Critical lleight (Steps) Duke Calculated Westinghouse Calculated Unit / Cycle Bank Heasured Predicted Difference (Steps) Difference (Steps) 127 117 10 9 1/3 CA CB 180 172 8 23 CC 224 201 23 -4 CD 163 156 7 18 SA 127 111 16 21 SC 139 133 6 21 SD 141 133 8 23 SE 132 126 6 12 1 - - 84 123

- - 84 131 I of absolute value Standard Deviation - - 6.00 9.36 a . >

t J

Table 3 (Cont.)

Critical IIcight (Steps) Duke Calculated Westinghouse Calculated Unit / Cycle Bank Measured Predicted Difference (Steps) Difference (Steps) i 2/2 CA 153 146 7 13 CB 190 191 -1 4 CD 202 205 -3 2 SA 198 186 12 16 SB 194 183 11 15 SC 185 182 3 14 SD 184 182 2 13 SE 149 141 8 18 I - -

39 95 I of absolute value - -

47 95 Standard Deviation - -

5.49 5.74

.l t

p b e

d a

1

i .

I Table 3 (Cont.)

Evaluation of All Available Data Duke Calculated Westinahouse Calculated i I (Differences) 81 227 1

3 j I (Absolute Value of Differences) 185 295 i ,

i j Standard Deviation 8.82 10.46 l (of the Differences) 4 l

l

.I I

l l

NOTE:

• Difference = Measured - Predicted ,

I h i

4 i

1 l

1

Table 4 a's Unit / Cycle Bank Duke Calculated 1/3 CA 1.042 CB O.877 CC 0.870 CD 1.161 SA 1.060 SC 1.052 SD 1.050 SE 0.903 a

a I

t l

l l

\

l

-. g Reference

1. Duke Power Company, " Nuclear Physics Methodology for Reload Design",

DPC-NF-2010A, June 1985.

2. Duke Power Company, McGuire Nuclear Station, " Control Rod Worth Measurement: Rod Swap Test Procedure", PT/0/A/4150/11A, April 1984.

e d-

.a I

0

. . - . -- , , - . - - , - - - - . - - - - - - - , , , _ - - . - . - , - - - - - . , , . . . - , . , , - - .- . _ , - - - - - - - , - - - , , - - , . -- - - , - - - _ . - . . - - , ,