05000282/LER-1982-006-01, /01T-0:on 820423,during Surveillance Test Sp 1104, Measured Reactor Coolant Boron Concentration Higher than Originally Predicted Value.Caused by Miscalculation of Predicted Worth.Analysis Performed to Monitor Disagreement

/01T-0:on 820423,during Surveillance Test Sp 1104, Measured Reactor Coolant Boron Concentration Higher than Originally Predicted Value.Caused by Miscalculation of Predicted Worth.Analysis Performed to Monitor Disagreement

ML20051A956
Person / Time
Site:	Prairie Island
Issue date:	05/07/1982
From:	Hunstad A NORTHERN STATES POWER CO.
To:	NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION III)
Shared Package
ML20051A795	List: 05000282/LER-1982-006, Forwards LER 82-006/01T-0 05000282/LER-1982-006-01, /01T-0:on 820423,during Surveillance Test Sp 1104, Measured Reactor Coolant Boron Concentration Higher than Originally Predicted Value.Caused by Miscalculation of Predicted Worth.Analysis Performed to Monitor Disagreement
References
LER-82-006-01T, LER-82-6-1T, NUDOCS 8205140500
Download: ML20051A956 (6)

LER-1982-006, /01T-0:on 820423,during Surveillance Test Sp 1104, Measured Reactor Coolant Boron Concentration Higher than Originally Predicted Value.Caused by Miscalculation of Predicted Worth.Analysis Performed to Monitor Disagreement

Event date:
Report date:
2821982006R01 - NRC Website

text

..

LICENSEE EVENT REPORT CONTROL BLOCK: l l

l l

l l lh (PLEASE PAINT OR TYPE ALL REQUIRED INFORMATION) 1 6

lO l t l l M l N l P l I l N l 1 l@150 l 0 l-l 0 l0 l 0 l 0 l 0l -l0 l 0j@l4 l1 l1 l1 l 1 l@l l

l@

_7 8 9 LICENSEE CODE 14 LICENSE NUMBER 25 26 LICENSE TYPE JO SF CAT b8 CON'T lOlil

$URCE l L Ql 0 l 5 l 0 l 0 l0121812 l@l 014 l 2 l 3 l 8 l 2M[0 l 5 l 0 l 7 l 812j@

7 8

60 61 DOCKET NUMBER 68 69 EVENT DATE 74 75 REPORT DATE 80 EVENT DESCRIPTION AND PROBABLE CONSEQUENCES h 10121 l Measured reactor coolant boron concentration is higher than the originally l

10 l 31 l predicted value by about 120 ppm at a cycle burnup of about 6200 MWD /MTU.

l pl Not repetitive. Tech Spec 4.9 applies. No effect on public health and safety.

l log3l l See attached details.

l l

0 6 l I3TTl i i

l 10181 l OE CODE S BC E COMPONENT CODE CUB DE S

E [7171 i a Ic l@ lx l@ W@ l z l z l z l z l z l zl8 Lz_J@ W @

7 8

9 10 11 12 13 18 19 20

.AEVISION SEQUENTIAL OCCURRENCE REPORT l

_E VE NT YE AR REPORT NO.

CODE TYPE N O.

(g p,po

@,Rg 18l2l b

l 0l 0l 6l y

l0l1l W

[-J W

21 22 2J 24 26 27 18 29 30 31 32 K N AC 0 0 PL NT ME HOURS SB IT FOR B.

SUPPLI MANUF CTLRER [xjgW@

[_zjg W @

l0l0l0l0l l Yl@

W@

W@

l zl9 l9 l 9]@

34 35,

36 3/

40 41 42 43 44 47 33 CAUSE DESCRIPTION AND CORRECTIVE ACTIONS h ltl0ll Apparent miscalculation of the predicted worth and/or depletion rate of I

li ;i j l gadolinium in the fresh fuel assemblies.

See attached details.

I

_l 1

2 l

I [TTTl l l

i 4 l

80 7

8 9 ST S

% POWER OTHER STATUS 015 O RY OtSCOVERY DESCRIPTION NA l

(_]Ll@l Surveillance Test l

[E_j@ l 1 l 0 l 0 l@l i 5 ACTIVITY CO TENT RELEASED OF RELEASE AMOUNT OF ACTIVITY LOCATION OF RELE ASE NA l

l NA l

l1l6l W @ W @l PtRSoNNEL EXPOS ES NU9BER TYPE

DESCRIPTION

NA l

y l0 l 0 l 0 l@l z l@l PERSONNE L INJURIES NilV8 E R

DESCRIPTION

l l0l0l0j@l NA i a 80 7

el 9 11 12 LOM OF OR DAVAGE TO F ACILITY TYPE

DESCRIPTION

l y@l NA i 9 8205140500 820507 NRC USE ONW PDR ADOCK 05000282 E S.

PDR l

lllllllllllll;

,ssu 2 O 68 69 80 g 7

8 9 to 612-388 1121

{

nStad PHONE:

NAME OF PREPARER

May 7,1982 Attachment (Page 1 of 5)

NORTHERN STATES POWER COMPANY PRAIRIE ISLAND NUCLEAR GENERATING PLANT Docket No. 50-282 50-306 LER 82-006/0lT-0

Event Description

on April 23, 1982 surveillance test SP 1104 Reactivity Anomalics, was performed.

This surveillance compares the measured RCS full power equilibrium baron con-centration as a function of burnup to the normalized predicted boron concentra-tion. The comparison resulted in a deviation between predicted and measured RCS boron concentration FSich, within the limits of measurement error, reached the equivalent of 1% t.~a/k.

The disagreement on April 23 was 114 ppm at a burnup of 6221 MWD /MTU or the equivalence of 0.98% Ak/k or 3 ppm from the limit. The disagreement has been determin~ed to be a modelling concern and not a deviation in core performance.

Cause Description and Corrective Actions Three apparent causes were considered. These are:

1.

a miscalculation of burnup; 2. a miscalculation of the predicted gadolinium worth and/or depletion rate, and; 3.

RCS soluble boron with a reduced B-10 abundance.

Further in-vestigation has eliminated possibilities 1 and 3 as major contributors to the disagreement. Additional independent analysis by both NSP*and the fuel vendor definitely indicate that the dicagreement is related to the analytical descrip-tion of the 4 w/o gadolinia assemblies.

The predicted RCS boron concentration as a function of burnup was supplied by Exxon in their Unit 1-Cycle 7 Startup and Operations Report.

NSP also calculates the predicted boron concentration va burnup; however, the Exxon curve is used for the comparisons in SP 1104.

(Note:

By. comparison to NSP's original normalized predictions the deviation is less than half.) The predicted Exxon curve was normalized to measured data at about 10% of cycle burnup as per T.S. 4.9.

This normalization point was at 1339 MWD /MTU and a measured RCS boron concentration of 896 ppm. The predicted boron concentration at this time is 914 ppm for a correction of -18 ppm.

Subsequent comparison of normalized predicted vs measured boron concentration (Figure 1) has shown a gradual increase in the deviation between measured and predicted up to a burnup of approximately 6000 MWD /MTU where the disagreewent stopped diverging and leveled off at about 1% Ak/k.

The most recent measurements (up to May 5, 1982) indicate the disagreement has started to converge and is presently at 0.9% ak/k corresponding to a burnup of approximately 6630 MWD /MTU.

Burnup is calculated using the FOLLOW code where the hourly power history, as recorded by the Nuclear Instrumentation System (NIS), and core MTU are used as input. The FOLLOW output burnup calculation is checked by comparing the unit capacity factor based on monthly burnup to the monthly electrical capacity factor. The N1S is calibrated daily using a secondary system calorimetric. All primary, secondary and electrical plant parameters show the NIS is indicating the correct power level.

Based on this evidence a miscalculation of burnup has been dismissed as a cause of the boron disagreement.

See NSP Topicals (NSPNAD-8101P and NSPNAD-8102P) submitted 2/12/82 for review.

May 7,1982 Attachment (Page 2 of 5)

The RCS boron concentration is measured approximately daily by titration of a sample of the RCS coolant. The measurement uncertainty is < 1% of the measured value. Only full power equilibrium xenon data points are used in the comparisons.

Soluble boron with a reduced B-10 abundance was considered as a contributor to the disagreement but was dismissed because:

1.

Chemistry data on lithium (a by-product of B-10, neutron reactions) removal from the RCS could account for less than 30% of the disagreement (using very conservative calculations) and 2.

The Unit 2 comparison of measured and predicted RCS boron concentration agree very well (Unit 1 and 2 share boron reprocessing equipment and much intermixing of boric acid supplies exists). Concentrated samples of the Unit I coolant have been prepared and sent offsite for an isotopic analysis of boron.

Results of this analysis are not yet available.

Analytical analysis of the disagreement by Exxon and NSP has shown that the devia-tion can be partially in Exxon's case or more closelyin NSP's analysis, accounted for by adjustments in the modeling of the 4 w/o gadolinia assemblies.

Both analyses indicate the disagreement should start to converge at about 6200 M'JD/MTU and disappear by the end of the cycle.

Power distribution measurement using the incore instrumentation system results in good agreement with predictions with margin to Tech Spec limits. No axial shifts in the power distribution have been noted. The RCS equilibrium coolant activity has remained stable since the first month of the cycle.

Surveillance frequency of the predicted and measured boron concentration measure-ments and power distribution measurements has been increased until the disagree-ment starts to converge. Analyses by Exxon and NSP have resulted in improvements in gadolinia modeling.

Continuing analysis is being performed to monitor the disagreement. Reviews of the disagreement indicate no safety related problems and demonstrate adequate shutdown margin.

l l

t l

1 2

F1GURE 1

>M n a MM ca O%

PRAIRIE ISLAND UNIT 1 CYCLE 'T t

2 e o cow m

Q a.

m i

o Le end

\\

g

~.

PREDICTED BORON u

o Q

NORM. PREDICTED BORON

_~

u b

. _i...

_1_7.__D_K./.< H__I_G_H

• N_O_RW______.

2

_1_7_. _ _D_K_ /_ _K__L_ _O_W

'_N_ _O R_M_ _*_____

O MEflSURED BORON x

~.

t

.s.....

a

% N

'v h

't**'.***.

• • E****

• **1****

K

.r**

Y.

W

.... % T'.

\\

c s.

Ag...g.....

g q..

Ul N

• N'-

"; [ " R * : 9-.:&;g * * " " ~ ~ * " ' " "" "

6

• '" " * "* ~

s sue

...... Px. w O

. +.,

..... N.....

g

...,.,\\

N,

...4..,..

N.

.......'....s O

\\.

.........................s...,..:....... K.....

..s..j............

s..

N....

..........................., s.....

N D

l l

l l

l l

0 2000 4000 6000 8000 10000 12000 14000 CORE BURNUP mwd /mtu

_.a

- y m

May 7, 1982 Attachment (P2ge 4 of 5)

. Fig s,. 2.

M L

K J

1 H

G F

E D

C B

A 9

G4 9

IOS G10 101 1

9 H4 9**

C3 9**

F4 9

2 114 G29 132 H09 125 G14 115 9

9*

J2 H2 G13 F2 D2 9*

9 3

112 136 H14 H40 H23 H33 H15 133 109 9

9*

DIO J3 12 Gil E2 03 J10 9*

9 113 137 G20 G37 H31 G33 H32 G19 G24 138 116 4

J6 L4 K4 M6 HS 9**

F5 A6 C4 84 06 5

G13 H13 G25 H07 G05 121 Gl8 H02 G38 H16 G30 9

9**

L6 L5 16 F13 G6 H13 E6 B5 86 9**

9 107 131 H39 H30 G17 H04 G09 H05 G26 H25 H34 126 102 6

J7 K3 A7 C7 9**

H7 E10 F7 9**

K7 M7 C11 D7 G01 H12 H22 G02 124 G36 F34 G34 122 G03 H24 H10 Gil 7

9 9**

L8 L9 18 F1 G8 H1 E8 89 88 9**

9 106 130 H38 H29 G28 H01 G04 H08 G06 H26 H35 127 103 8

J8 L10 K10 M8 H9 9**

F9 A8 C10 810 08 G32 H2O G40 H06 G23 123 G27 H03 G07 H17 GIS 9

9 9*

04 J11 112 G3 E12 D11 J4 9*

9 120 140 G22 G35 H28 G12 H27 G39 G21 139 117 10 9

9*

J12 H12 G1 F12 012 9*

9 i

111 135 H19 H37 H21 H36 H18 134 110 11 jl 9

H10 9**

K11 9**

F10 9

t Il9 G16 129 Hll 128 G31 I18 12 i

i 9

G10 9

105 G08 104 13 h

8 pins of 4 w/o gadolinia per assembly 9

a l

12 pins of 4 w/o gadolinia per assembly h

- Batch Id (new fuel) or core locati.on in Cycle 6

- Assembly Serial number

+

Figure 4.1 Prairie Island Unit 1, Cycle 7 L

Loading Pattern Startup and Operations

)

4 j g i L i h 2 '

y i

i b;

N5 gx nwI_

TABLE 1 Prairie Island Unit 1 Cycle 7 fuel Assembly Design Parameters 9

Catch v.

5 7

8 9

9*-

9" o

0 Enrichn.ent, wt '4 U-235 3.30 3.40 3.47 3.56 3.52 3.50 TOPROD Fuel Assemblies Central 11 Feet, wt % U-235 3.82 3.82 3.82 UO -Gd 0 ' wt % U-235 2.90 2.90 2

23 -

Upper ard Lower 6 inches, wt % U-235

.71

.71

.71 tJumber of Assemblies 1

40 40 20 8

12 Pellet Density, % TD 94.5 94.0 94.0 94.0 94.0 94.0 Pellet to Clad Diametrical Gap, mil 7.5 7.5 7.5 7.5 7.5-7.5 I

Fuel Stack Height, inches 144.0 144.0 144.0 144.0 144.0 144.0 Batch Average Burnup at BOC7, MWD /MT 27,434 24,680 7,604 0

0 0

Fuel with 8 pins of 4 w/o gadolinia per assembly

• • fuel with 12 pins of 4 w/o gadolinia per assently_