ML20024J344

1 Cycle 12 IPC Assessment & Projected EOC-13 Slb Leakage.

ML20024J344
Person / Time
Site:	Farley
Issue date:	10/05/1994
From:	Keating R, Pitterle T, Wepfer R WESTINGHOUSE ELECTRIC COMPANY, DIV OF CBS CORP.
To:
Shared Package
ML20024J342	List: ML20024J341 ML20024J344
References
SG-94-08-006, SG-94-08-006-R01, SG-94-8-6, SG-94-8-6-R1, NUDOCS 9410140077
Download: ML20024J344 (95)

Text

SG-94-08-006, Rev. I NSD-SGD-1007 Rev.1 FARLEY-1 CYCLE 12 IPC ASSESSMENT AND PROJECTED EOC-13 SLB LEAKAGE October 5,1994 Prepared By:

T. A. Pitteric R. F. Keating R. M. Wepfer Westinghouse Electric Corporation Nuclear Services Division P.O. Box 158 Madison, PA 15663 C 1994 WESTINGHOUSE ELECTRIC CORPORATION ALL RIGHTS RESERVED 9410140077 941004 8 PDR ADOCK OSCOt .

P l __

. . w;. ... ; -g- y mg. ry - . .

, , .pv n , - - m .

~a s

., .n ,

c 3'

3 .

. . f q, [ - ) 1, Q. f \ )1

.m.'x')

..y :n

, w .5.;

, m, -

.. x%

s, m

e,

, i

4. '1 -

v ,, *

.s 1 -

.r, 7 o.c 5 t,e ;j.) r 8-..... ,

v. .:- ,

, w

- , . e,m; . , , .

t  :..

n . ,

e

,n - c

. SG-94-08 006, Rev.1 1-. . ..  ?

a.3 e * '

, i NSD-SGD-1907, Rev.1

,4

..af' s

d '.

s

, FARLEY-1 i

TCYCLE 12 IPC' ASSESSMENT AND PROJECTED

, .. EOC-13 SLB LEAKAGE i

/ -l. ' -

g 3

4 5

'l h * '

w- r, p 'I .' I 1 .- ., ..

m 7 ,

EUctober 5,1994

m. ,

4 4

y

\ \

• sv i

e w

1

' A o '

n' ')-

a

a # . . a

? * , i ~ < 9, 3,. o ., ~.

s y ~,

Prepared Byh 4

ha . m. ,

ts', e.

ser .

,s .y -_

. ~ . . . . . , . .

, s ,

f * , ,m* , -

H,

1. ... <iTfA; Pitterlef;.

m - - ,

f,^,,

e

, ., , ', R. F.:Kesting ~

,: #, w'.

ax iR;M. Wepfer? >

m +

, s

/y,

.. , a ;. L e- e  ; ,

+

y ,. s.. , ,

l-) ,

y n l{ ,')';" ,

1 , .

, , ~ , . ,w : ,

c

,,:'b. * '

, ' I. '.'

,, J

, . _ . , < .,.g',. [' l 2 , y.. g6 .

I f ( ' 4 0

'4 ,. ' t (. . , y i -

a Q

v. ~ r ,

. m +

,-s ,

,i  %;m  :- s.

p.o~ ,. p .

e.

, 3- ,

4

< 1

.=o .. 9...j s m s r e., p

_ -1 . , 9m v e y , ,

ce

{ % ;. '?$f  ::d Y$ '{ t y ::Y ~ , . , i  : '

ge, ,

> 3 , m, i ' . 1

_! m d' a.a :> . ,.

e. y lm' s c c'.. #

s- '9... u ::f *. ...t g 1 g ,, 5 p g d' ._4[bh'

< 3Westinhouse E'lectric' Corporation 9 M 69 g, g y q* 'g, v;

"

• 2

, q, , JNaclear' Services.'Divisi.ons 4

4 m , - -

m s .

na g. . ~  ; ce, ,, " ,:' M , W. /.P.O. Bor 158 i , ,

L-C -

^

g ~ , , qw s A,m, ,

W 3MadisemiPA N15663

- J M > i

. w= j, 7 .4 , , >

uz o ,

1 3

e ,, ry. .

>w/. a; m. ) .

+

s v. u ...

4 ';p' a h:,: . r: . i

c+

L 1 p:.pF .f 4[1. r ~ , 9 i

4 q  ;! > ..

, g ,

f J f.p 1

, p.

eta s >

(

i RyWQQ3; ppi

,, s - - m 4, s ,, ,

y .

f, M

m a av-M. .; v. j y v- ,r,.s, s..,.w.m...t. u a , wc . -A.,  :. . .;. / -

9 ,.s. .s,

'- , e.h.2p y r.p, m py;< $

4 , y, , 1. g.; , ,, q ,9 r y , v -; ,

f .ie -

3; i.p; y .n p. . , n, w , , . i 3

. ;3,1. ,.. .

i 1T

-.o . w. t - ~

• ' 't #:.;

>,' " -.p.....-.yr_ .

1. , a., t . ,,

. [ ,, " 1,, , ..7-61* ff#Y .*0 - < 5 -

. - , " / .i r S .' -,. ,.-',

J ~,f

." , g ?' '

67,.

n < t . M.. ,

f ~' a C.19'9,4LW'E, . STING HOUSE ELECTRIC' CO'R' PORATIONj 4 t <

n WL.. i.. ,u,p**%' ? T m,.

r. s * < - , . , -/C ?ALL,.. RIGHTS <y, RESERVED 1 t

,  ?. .,, ,

, e

) fi 'y + l.&y. f,0)9 m' s +Q, . .- x ,:

.6 i~"

1}^  ; c,' ' .

l

, a w$ ' /

s', g'

+ **

>.v

,l s - r: i t 'a p qg[':.i '

1

,, e ,i 4 -

,e ,;

w 7..

(

s _ , e u>. >

b h $ > '_ Y * ,.,f.

w :m. y .. ,. , ^ ; m[, ' ' ,

'q. " ,. , m.y' f.pj_'h  ; i Q'

• v 4

.Qf ^} , ,

,I . r.p QS .,

s .;g : . , ., r  : 4 .e  ? 'M - i y y i  ?

p; I', ~.

'a

,[ }'[ [ i {.} '

I,

.I 'r b  ?

t. j N i/j%,

I.

7 7.,b.

) d# - - ' '

'm-1 ,( p 1 l c"'r ...if...F r- r '! *

, s  !. b

<3, 1

[ y. < 's, a y  ;:f.

,M7

'v g' i i 5. n:m. ; Q ,* ~ . %m, / N^) .L:. ' < s r -;p;  ; .o 4 c

/ 4 sew i

um ,-

p - .,

s.t

' d. -:p y 3 s w r ~-

, ,  ;) >< , _

g <. y i 2(;j u-h ~Q,W ,

k, . E

. d+d. b

< ~

.h w m$4 :Qb x- $. y c... ,. ,4

. . . . . .-n-AIMm ,A:,9@M u - ,a.zhh*M m ,Ik; ',

_f

~

• ? j ~ ^ ' 0' ft m . jy? N 05, ,. . ,._'b.,.0 -~s

, ,h' f..

A lf.\r\

.,5mf'f' a... r .

,, f., ,, [a'% - a ';;: f. w..C _ . .;, , " , ,. vyg a, , y + ..w

\-

n ....g, w r s , v%., , , a m,.

L w s, .w ., , 6.~a

. . . . ~ - . . ... -.

.~

SG-94-08-006, Rev.1 NSD-SG D-1007, Rev.1 l

FARLEY-1 CYCLE 12 IPC ASSESSMENT AND PROJECTED EOC-13 SLB LEAKAGE i

l October 5,1994 i

Prepared By: l T. A. Pitterle R. F. Keating R. M. Wepfer r

l l

l Westinghouse Electric Corporation Nnclear Services Division P.O. Box 158 Madison, PA 15663 l

l I

C 1994 WESTINGHOUSE ELECTRIC CORPORATION

ALL RIGHTS RESERVED

.I l

l l

i l

FARLEY-1 CYCLE 12 IPC ASSESSMENT AND PROJECTED EOC-13 SLB LEAKAGE TABLE OF CONTENTS SECTION PAGE 1.0 Introduction 1-1 2.0 Summary and Conclusions 2-1 2.1 Overall Conclusions 2-1 2.2 1994 EOC-12 Inspection Results 2-1 2.3 Voltage Growth 2-3 2.4 Comparison of Projected and Actual EOC-12 Bobbin Voltage Distributions 2-3 2.5 SLB Leak Rate Analyses 2-5 2.6 Tube Burst Probability Assessments 2-6 2.7 Assessments of NDE Uncertainty for Analyst Variability 2-6 3.0 EOC-11 and EOC-12 S/G Inspection Results 3-1 3.1 EOC-11 Inspection Results 3-1

3.2 EOC-12

Summary of Indications at TSPs 3-1

3.3 EOC-12

Assessment of RPC Inspection Results 3-3

3.4 EOC-12

Assessment of New Indications 3-4 3.5 EOC-12 Distributions for Comparisons with Projections 3-5 4.0 Bobbin Voltage Indications Left in Senice 4-1 4.1 BOC-12 indications Left in Senice 4-1 4.2 BOC-13 Indications Left in Senice 4-1 5.0 Voltage Growth Rates 5-1 5.1 Cycle 11 Voltage Growth Rates 5-1 5.2 Cycle 12 Voltage Growth Rates 5-1 6.0 NDE Uncertainties 6-1 7.0 Projected EOC Voltage Distributions 7-1 7.1 Projected EOC-12 Voltage Distributions 71 7.2 Comparison of Projected and Actual EOC-12 Distributions 7-1 7.3 Projected EOC-13 Voltage Distributions 72 8.0 SLB Leak Rate Analyses 81 8.1 Database Supporting Alternate Repair Criteria 8-1 8.2 Alternate Leakage Analysis Methods 8-3 8.3 Projected EOC-12 SLB Leak Rates 8-5 8.4 Comparison of Leak Rates for Projected and Actual EOC-12 Distributions 8-5 8.5 Projected EOC-13 SLB Leak Rates 8-6 i

. < s.

4 FARLEY-1 CYCLE 12 IPC ASSESSMENT AND PROJECTED EOC-13 SLB LEAKAGE .

TABLE OF CONTENTS (CONTINUED)

SECTION PAGE  ;

9.0 SLB Tub'e Burst Probability Analyses 9-1  ;

9.1. Projected EOC-12 SLB Burst Probability 9-1 9.2 Projected EOC-13 SLB Burst Probability 9-1 1

i 10.0 Alternate Method for Defining BOC Indications Left in Service 10-1 l 1 10.1 Allowance for Undetected or New Indications 10-1 10.2 Considerations for RPC NDD Indications 10-3 10.3 = Comparison of Projections Applying Alternate Method with Actual EOC-12

} Distributions - 10-3 j 10.4 Projected EOC-13 Voltage Distributions and SLB Leak Rates 10-4 j i

j 11.'0 Assessments of NDE Uncertainty for Analyst Variability for EOC-Il Inspection 11-1 i

12.0 References 12-1 i

h

r

• ?

l )

4  !

a J

4 i

ii

1 l

l

1.0 INTRODUCTION

This report provides a Cycle 12 interim plugging criteria (IPC) assessment and projected end-of-cycle 13 (EOC-13) SLB leakage for the J. M. Farley Unit-1 steam generators. Included in this report is information requested in the NRC Safety Evaluation Report (Ref.1) for application of the IPC for Cycle 13. This Mformation includes comparisons of projected EOC-12 bobbin voltage distributions with actud .41ues found in the EOC-12 inspection, projections of EOC-13 voltage distributions based on indications left in service at BOC-13, projected potential steam line break (SLB) leak rates at EOC- e l

13, and tube burst probability at EOC-13. l l

The Farley-1 EOC-12 S/G eddy current inspection is the second inspection following the implementation ofIPC repair limits. Thus the EOC-12 inspection results provide an opportunity to i compare actual voltage distributions with projected values. Alternate methods of defining BOC l indications left in service are used to project EOC voltage distributions for comparison with the actual l EOC-12 inspection results. The methods of defining BOC indications include: the NRC recommended

( method of draft NUREG-1477 (Ref. 2) which includes RPC NDD and adjustments for a probability of l detection of 0.6; a distribution including RPC NDD with no adjustment for detection probability (POD

= 1.0); and an alternate method for defining the BOC distribution. The alternate method assumes that a fraction of the RPC NDD indications may be undetected flaws and accounts for undetected BOC indications based on the prior cycle voltages for new indications found in the latest inspection.

The Monte Carlo methods of the Farley-1 APC WCAP-12871 Rev. 2 (Ref. 3) are applied to the BOC .

voltage distributions to project the EOC distributions. This is consistent with the NRC guidance given I in the Farley-1 SER. SLB leak rates are calculated using the NRC methodology of draft NUREG.

1477 as well as the IPC/APC methods described in the EPRI alternate repair criteria (ARC) l documentation (Ref. 4) which utilizes methods applying the APC correlations for probability of I leakage and SLB leak rate. S/G-C is the more lindting S/G for SLB leakage and burst considerations i

for both Cycles 12 and 13 as this S/G has the largest number ofindications left in service. The latest j EPRI correlations are applied in this report. Tube burst probability analyses apply the WCAP l

methodology and the EPRI burst pressure versus bobbin voltage correlations using both the EPRI  !

database and the database recommended by the NRC in the Farley-1 SER. j Consistent with the request of the Farley-1 SER, analysis data such as voltage distributions and growth I

rates are provided in both graphical and tabulated form. Table 1-1 relates the data requested in the l Farley-1 SER to the report section, table and figure which provides the data.

i l

l l 1-1 l

l

- =. - . - - - . . _ . - - . . ~. . - . -

TABLE l-1 CORRESPONDENCE BETWEEN SER REQUIREMENTS AND REPORT SECTIONS NRC Subject Section Table Figure SER  ;

Section 4.5 [

l EOC 11 voltage distribution - all bobbin ind. 3.1 4-1 3-1 2 Cycle 11 growth rate 5.1 5-1,3 5 3 EOC 11 repaired indications' 4.1 4-1 -

4 BOC 12 voltage distribution - all bobbin ind. 4.1 4-1 4-1

! BOC 12 voltage dist. - draft NUREG-1477 4.1 4-1 4-1  :

l BOC 12 voltage dist. - alternate method 10.3 10-2 -

! 5 BOC 12 voltage dist. - RPC confirmed plus 4.1 4-1 4-5 l not RPC inspected 6 Cycle 12 NDE uncertainty 6.0 - -

7 Proj. EOC 12 voltage dist.-WCAP-12871 R.2 7.1 7-1 7-lb Proj. EOC 12 voltage dist.-NUREG-1477 7.1 7 7-la Proj. EOC 12 voltage dist.-alternate method 10.3 7-1 10-1,2 l

8 Actual EOC 12 voltage dist.- all bobbin ind. 3.2 3-1 3-2,3 i

Actual EOC 12 voltage dist.- RPC confirmed 3.5 7-1 7-1 plus not RPC inspected ,

! 9 Cycle 12 growth rate 5.2 5-1,4 5-2 10 EOC 12 repaired ind. 3.2 3-1 3-2, 3-3 l

l 11 BOC 13 voltage dist.-all bobbin ind. 4.2 4-1 4-3 l BOC 13 voltage dist.-draft NUREG-1477 4.2 4-1 4-3 BOC 13 voltage dist.-alternate method 10.4 10-2 -

12 BOC 13 voltage dist.-RPC confirmed plus 4.2 4-1 4-6 not RPC inspected 13 Cycle 13 NDE uncertainty 6.0 - -

14 Proj. EOC 13 voltage dist.-WCAP-13187 7.3 7-2 7-5 Proj. EOC 13 voltage dist.-NUREG-1477 7.3 7-2 7-5 Proj. EOC 13 voltage dist.-altemate method 10.4 7-2 10-3 i

I 1-2 i

l l

2.0

SUMMARY

AND CONCLUSIONS 2.1 Overall Conclusions l

EOC-13 projections were made, consistent with the Farley-1 SER for Cycle 13, by applying the NRC model of Draft NUREG-1477 to define the BOC distribution with a probability of detection (POD) l adjustment factor of 0.6, the NRC SLB leak rate model and the correlation of burst pressure with

! voltage to obtain the SLB tube burst probability, it is shown in this report that applying the POD =

l 0.6 adjustment leads to a substantial overprediction of the EOC-12 voltage distribution and associated j leakage and burst probabilites. The resulting EOC-13 SLB leak rate is estimated at 0.61 gpm which

! is well below the allowable limit of 22.6 gpm for Farley-1 Cycle 13. When the EPRI leak rate versus voltage conelation is applied, the projected SLB leak rate is only 0.081 gpm by Monte Carlo analyses.

The projected EOC-13 SLB tube burst probability by Monte Carlo analysis is 4.3 x 10" using the EPRI database and 1.2 x 10 2 using the NRC recommended database of the Farley-1 SER. The EPRI l database applies the outlier evaluation to exclude selected very high burst pressure data from the burst l pressure versus voltage correlation while the NRC recommended database includes these high burst indications. Both databases result in projected EOC-13 burst probabilities below the acceptance guideline of 2.5 x 10-2 based on NUREG-0844 analyses. Overell,it is concluded that SLB leakage and burst probability acceptance limits were satisfied in Cycle 12 and will be satisfied with large margins for Cycle 13. The Cycle 12 voltage growth rates at J. M. Farley-1 are smaller than those of l Cycle 11, with a maximum voltage growth of 0.93 volt for Cycle 12 compared to 1.90 volts for Cycle 11.

Comparisons between projections and actuals are provided in this report for the bobbin voltage distributions at EOC-12 and the SLB leak rate and burst probability calculated from the actual EOC l distribution. The best agreement, while retaining conservatism, between projections and actuals is

! obtained for the alternate or new indication method of defining BOC distributions described in Section 10 of this repon. The draft NUREG-1477 methodology with a POD = 0.6 adjustment to the BOC distribution results in excessive conservatism compared to the actual voltage distributions and associated SLB leak rates and burst probabilities.

The projections for Cycle 12 were based on S/G C which had the most limiting BOC-12 voltage distribution. Based on leak rates and burst probabilitics calculation from the actual EOC-12 voltage distributions,it is found that S/G A is slightly more limiting than S/G C. This results as S/G A had two larger voltage, RPC confirmed indications than S/G C. The largest EOC-12 RPC confirmed indication was 4.8 volts. The largest voltage projected for S/G C at EOC-12 ranged from 3.3 to 3.8 volts depending on the method of defining the BOC distribution. Even though the projections underestimated the maximum volte.ge found in any S/G, the projected SLB leak rates and tube burst probabilities are casentially equal to or higher than the values calculated from the actual EOC-12 voltage distributions for S/Gs A crvi C. This conclusion applies for all three methods (POD = 0.6, POD = 1.0 and new indication mtthods) evaluated for defiming the BOC voltage distributions and demonstrates the conservatisms in the projection methods.

l 2.2 1994 EOC-12 Inspection Results

! The number of potential flaw indications at tube support plates found in the 1994 inspection was 1683 which includes 509 in S/G-A,481 in S/G-B, and 693 in S/G-C. The number of RPC confirmed 2-1

i a

i indications was 67 of the 339 potential indications RPC inspected. For indications above 1.0 volt, j RPC examination confirmed 64 of the 334 potential bobbin indications RPC inspected for a RPC

confirmation rate of 19.8%. These low confirmation rates arc indicative of conservative bobbin

) indication calling criteria. The largest bobbin indications confirmed by RPC were 4.80,3.02 and 2.45 4 volts. A potential bobbin indication at 2.92 volts was not confirmed as a flaw indication by RPC. Of I the 19 bobbin indications with EOC 12 voltages greater than or equal to the 2.0 volt repair limit in

place at EOC-12, only three (vntages given above) were confirmed by RPC and repaired; the balance

were RPC NDD and did not roquire repair. No indications were found in the augmented RPC 1

inspection program which intluded all dents > $ volts (146 TSP intersections) and 10 intersections j with artifact signals. Also no circumferential indications were found at TSP intersections and no

, indications wer: found te extend outside of the TSPs.

i l

To assess the necJ for including RPC NDD indications left in service in the SLB leakage analyses, an assessment was made af the RPC NDD indications in service at BOC-12 that became RPC confirmed 1 indications at EOC-12. Out of 217 RPC NDD indications left in senice and RPC inspected at EOC-

) 12,24 of these or 11% were confirmed as flaws at EOC-12. For indications above 1.0 volt,9% were j confirmed. The highest rate of confirmation for BOC-12 NDD indications was 13% in S/G-C. Draft j NUREG-1477 methodology requires that all RPC NDD indications,in addition to a POD adjustment j of 0.6, be included in the SLB analyses. The low Farley-1 confirmation rate for RPC NDD '

j indications left in service indicates that the NRC methodology is 0.assively conservative and that about 20% of the NDD indications should be included in the SLB leak rate and tube burst analyses.

]

] New indications were found in the 1994 inspection that were not reported at the 1992 EOC-11 j inspection. Also, a number of potential bobbin indications reported at the 1992 inspection were not i found in the 1994 FOC-12 inspection. Out of a total over 3 S/Gs of about 1084 indications left in j service at BOC-12,54 or 5% of the BOC potential indications were not reported at EOC-12. A total l of 653 new indications (not reported in EOC-11 inspection) were reported at EOC-12 which leads to a j net number of new indications of 653-54=599. The RPC confirmation rate for the new indications 1

was 34% over all S/Gs which is somewhat higher than the confirmation rate for all bobbin indications.

4 Thus the EOC-12 inspection results indicate significant numbers of new indications and of prior j indications that do not remain flawlike at EOC. The three largest bobbin indications confirmed by  ;

i RPC inspection were new indications at 4.8,3.02 and 2.45 volts. These indications had small growths j (< 0.35 u it) from '92 to '94. The 4.8 volt indication had an ID phase angle which did not require

identification as a potential bobbin indication (PI) by the cddy current guidelines in place in 1992.

j Similarly, the 3.02 voit indication had a very large phase angle corresponding to 0% depth and was k

not identified as a Pl. The Appendix A cddy current guidelines were revised prior to the EOC-12 inspection to require that indications with ID phase angles and OD 0% depth be called Pls (i.e.,

indications with phase angles from 5* to 170* are now classified as Pls) and these indications were reported in the 1994 inspection. Thus, reoccurrence of the larger voltages for new indications is not expected in the future. The 2.45 volt indication had highly distorted bobbin responses in both the '92 and '94 inspections and had only a 0.64 voit RPC indication which is n. ore typical of bobbin indications about one volt. Consistent with the analysis guidelines, disterted indications are assigned a conservative voltage and the 2.45 volt assignment is judged to be veiy conservative for this indication.

The actual indication not reported in '92 likely has degradation typicai < about a one volt, undistorted bobbin indication.

2-2

)

A detectability parameter called the probability of prior cycle detection (POPCD) is evaluated in this report. POPCD is defined as the probability that an indication found by bobbin inspection and confirmed by RPC was detected and reported at the prior cycle outage. The probability of detection (POPCD) at EOC-Il for indications found at EOC-12 was evaluated. Only RPC confirmed indications are significant for considerations of SLB leakage / burst and were used for the evaluation.

The resulting POPCD for the 1992 inspection was 64% for indications < l.0 volt in 1992, 82 % l betweer 1.0 and 2.0 volts and 95% between 2.0 and 2.5 Volts. Above 2.5 volts, the POPCD is 100%

l 1 l if the 1994 Appendix A guidelines on phase angle are applied to the 1992 inspection, or 71% (5 out i l of 7 reported in 1992) for the 1992 guidelines. It is expected that the new Appendix A guidelines l implemented for the '94 inspection will result in detection probabilities approaching 100% above 2.5 volts in the future. The overall average POPCD was 82%. These results show that the draft NUREG-1477 POD = 0.6 is too low for IPC applications and that a constant POD should not be assumed for all voltage levels (i.e., the POD should be higher at high voltage levels).

The RPC conErmation rate at EOC-12 was separately assessed for three categories ofindications. i Prior bobbin indications that were not RPC NDD indications (RPC confirmed plus not RPC inspected and left in service) had a confirmation rate of 46.7%, prior RPC NDD indications had a confirmation rate of 11.1% and new indications had a confirmation rate of 33.6%. Thus prior cycle RPC NDD indications have a low likelihood of contribution to the EOC confirmed indications and should not be evaluated the same as RPC confirmed indicctions such as required by draft NUREG-1477 methodology. As would be expected, the highest RPC confirmation rate occurs for the prior cycle RPC confirmed plus not RPC inspected indications left in service.

2.3 Voltage Growth The largest voltage growth of 0.93 volts found for Cycle 12 was found in SG-A R17C77 4H which l had an EOC-12 bobb.n voltage of 1.62 volts with a bobbin 0.69 volt indication in 1992, and was RPC NDD at EOC 12. TI'e largest growth found for an RPC confirmed indication was 0.77 volts. For Cycle 11, the largest growth was 1.90 volts. The Cycle 12 growth rates are lower in average and

, maximum value than prior Farley-1 cycles and are among the smallest found for plants evaluated for IPC applications. The average Cycle 12 voltage growth was negligible or ~0% of the average BOC voltage, compared to 0.22 volt (26%) growth for Cycle 11.

l l

l 2.4 Comparison of Projected and Actual EOC-12 Bobbin Voltage Distributions The 1994 Farley-1 inspection provides an additional full cycle of operation following IPC implementation. Thus tl e resulting data permits comparisons of projected and actual EOC voltage distributions. These comparisons were made applying the NRC draft NUREG-1477 methodology including a POD = 0.6 adjustment to detected indications, a method including RPC NDD indications without a POD adjustment (POD = 1.0 for all bobbin indications left in service) and an alternate method that includes provisions for undetected indications based on new indications found in the inspection and for a fraction (conservatively 20%) of the RPC NDD indications assumed to develop to confirmed flaws. This alternate method, called the new indication method, has been now evaluated for four plants following IPC inspections against actual EOC distributions and has consistently provided the best agreement with actual EOC distributions. This method provides adequate conservatism while applying more realistic, plant specific adjustments for undetected indications and 2-3 l

l _ _ .

includes recognition that only a fraction of RPC NDD indications are likely to develop into significant flaws in the subsequent cycle. The projection methods are compared with the actual EOC-12 distribution for RPC confirmed indications plus indications not RPC inspected. Since the primary purpose of the voltage projections for IPC/APC applications is to estimate tube leakage and burst probabilities, only indications found to bc RPC confirmed (not inspected are conservatively assumed to be confirmed) are used for comparisons with the projected distributions as RPC NDD indications would not have any significant probability of leakage over the prior cycle.

At BOC-12, S/G C was the most limiting S/G and projections to EOC were made for S/G C. For the actual EOC-12 RPC confirmed distributions, S/G A was found to have slightly larger leak rates (0.25 gpm versus 0.18 gpm for draft NUREG-1477 methods) and tube burst probabilities (2.4 versus 1.1 x 10") compared to S/G C. This results since S/G A had two larger RPC confirmed indications at 4.8 and 2.45 volts compared to the largest of 1.9 volts in S/G C. The maximum projected EOC-12 voltages for S/G C ranged from 3.3 to 3.8 volts depending on the method for defining the BOC distribution as described below. Even though the S/G C projections underestimated the maximum l

i voltage found in S/G A, the projected leak rates and burst probabilitics for all projection methods are l equivalent or higher than that obtained from the actual distributions for both S/Gs A and C. The j comparisons given below compare the S/G C actuals with projections with the recognition that only the 4.8 volt indication found in S/G A is underestimated by the projection methods.

For S/G C with the largest number of indications left in service at BOC-12, it is found that the draft ,

f NUREG-1477 methodology considerably overestimates the number and magnitude of the EOC-12 l indications. The projected maximum voltage is 3.8 volts compared to the actual RPC confirmed value of 1.9 volts (maximum unconfirmed bobbin indication of 2.6 volts). It is found that this methodology

introduces unnecessary conservatism in the voltage distributions and also results in SLB leakage a i factor of 2 higher than found for the actual distribution. Results of this evaluation are provided in Section 7.2.

The POD = 1.0 method including RPC NDD indications left in service but without a POD adjustment yields good agreement with the actual distribution. This method resulted in a maximum EOC voltage of about 3.6 volts compared to the RPC confirmed 1.9 volts found by inspection. The higher voltage tail of the distribution above 1.1 volts is bounded by the projections while the projections are non-i conservative below 1.1 volts. This method, however, applies the RPC NDD indications at BOC to compensate for potential new indications as well as potential growth to confirmed indications at EOC.

l Results of this evaluation are provided in Section 7.2.

The new indication, alternate method for EOC voltage projections, which accounts for undetected indications based on prior cycle experience and a fraction of RPC NDD indications growing te detcetable flaws, results in good agreement between projected and actual EOC-12 voltage distributions i over thc entire voltage range while conservative above about 0.9 volts. The maximum projected 1 EOC-12 voltage for this case is 3.3 solts compared to the RPC confirmed 1.9 volts inspection result.

This method accounts for plant specific detectability and is a preferred method over POD = 0.6 until an acceptable voltage-dependent POD is developed for comparisons between projections and actual l distributions. This methodology is further developed in Section 10.0.  !

l The new indication method, as applied in this report, utilizes 50% of the new RPC confirmed indications summed over all three S/Gs and evaluated at prior cycle voltages (EOC-ll for EOC-13 projections)in the projections for the next cycle. With this method, the largest new indications in all 2-4 i

l l

l S/Gs are included in the projections with at least half of an indication. Since the largest three EOC-12 J

RPC confirmed indications at EOC-12 were new indications in other than the limiting S/G C, the new indication method includes these indications while the POD adjustment methods do not include them.

l This results in the EOC-13 projections for S/G C with the new indication method leading to a maximum voltage of 4.5 volts compared to 3.3 volts for the POD = 0.6 adjustment. Even with the higher maximum EOC voltage projection, the new indication method results in lower EOC-13 leak rstes, as described below, than the POD = 0.6 method. The higher leak and burst probabilities for the l l

POD = 0.6 methods result from the highly conservative assumptions that POD is independent of voltage and that all RPC NDD indications are real indications equivalent to a confirmed indication.

For Farley-1, the RPC NDD indications left in service can be expected to progressively increase in j future cycles and thus lead to continuing increases in conservatism for the draft NUREG-1477 POD = l 0.6 methodology. It is proposed for the new indication methodology that indications RPC NDD in two successive cycles need not be included in the projected distribution.

2.5 SLB Leak Rate Analyses  !

l

! SLB leak rate analyses were performed to compare values based on projected EOC-12 voltage i distributions with that obtained from the actual distribution as well as to project EOC-13 leak rates for

( comparison with allowable limits. Applying NRC draft NUREG-1477 methods for calculating leakage  ;

j given a voltage distribution, the actual EOC-12 distributions (RPC confirmed plus not RPC inspected)  ;

I resulted in a 0.18 gpm SLB leak rate for S/G C and a 0.25 gpm leak rate for S/G A. Although S/G C 1

! had the most limiting BOC distribution, larger RPC confirmed indications were found in S/G A. For l the NRC method of defining BOC voltages, which includes a POD = 0.6 adjustment and RPC NDD l indications, the predicted S/G C EOC leakage was a factor of three higher at 0.78 gpm than even the

! actual S/G A leak rate. Methods excluding the POD adjustment (all bobbin indications left in service) I

and the alternate method for defining BOC distributions yielded 0.50 and 0.35 gpm, respectively i which conservatively bracket the actual distribution leakage by up to a factor of two. The j conservatism of the POD = 0.6 adjustment thus results in excessively conservative leak rates as well l as EOC voltage projections. Even though the largest leak rate from the actual EOC-12 distributions was obtained in S/G A rather than the projected S/G C, all methods including the more realistic new indication method include adequate conservatisms to bracket the S/G A result. Application of the EPRI SLB leak rate versus voltage correlation leads to leak rates about a factor of 15 lower than draft NUREG-1477 methodology.

The draft NUREG-1477 methods were applied to project EOC-13 voltage distributions (including POD

= 0.6 and RPC NDD indications) to obtain a projected SLB leak rate of 0.61 gpm for S/G C. To confirm that S/G C was limiting for EOC-13 given the larger maximum voltage in S/G A, EOC-13 projections were performed for the POD = 0.6 adjustment method. The projected leakage was 0.45 gpm such that S/G C is the limiting S/G for Cycle 13 projections. The projected leakage for the draft NUREG-1477 leak rate analysis method applied to the BOC distributions for POD = 1.0 and the new

, indication method are 0,44 and 0.39 gpm, respectively. All projected leak rates are well below the Farley-1 allowable leakage limit of 22.6 gpm. Thus potential SLB leakage is well within acceptable l limits for Cycle 13 operation. When the EPRI leak rate versus voltage correlation methodology is

[ applied to the POD = 0.6, EOC-13 voltage distribution, the projected SLB leakage is only 0.081 gpm l or a factor of almost 8 lower than the draft NUREG methodology. This demonstrates the I

conservatism in the draft NUREG methods which apply the leak rate database independent of bobbin voltage.

2-5 l

l

2.6 Tube Burst Probability Assessments Burst probabilities calculated for the actual EOC-12 voltage distribution can be also be used to assess the different methods for defining the BOC indication distribution. The burst probability calculated with the EPRI database for the RPC confirmed plus not RPC inspected distribution at EOC-12 was 1.1 x 10" for S/G C and 2.4 x 10" for S/G A. The probability obtained after applying a POD = 0.6 4

adjustment to obtain the BOC distribution was 1.3 x 10 while the BOC distributions for all bobbin indications with POD = 1.0 yielded 6.8 x 10" and the new indication method yielded 3.8 x 10" for the burst probability. Thus all projection methods result in burst probabilities greater than that obtained l

from the actual EOC-12 voltage distributions for both S/Gs A and C. These results again shows the excessive conservatism resulting from a POD = 0.6 adjustment to obtain the BOC distribution.

The tube burst probability at EOC-13 was estimated by applying the draft NUREG method with POD

= 0.6 for defining BOC voltage distributions and the WCAP Monte Carlo methods for burst probability analyses. This is the methodology recommended in the NRC Farley-1 IPC SER. The resulting tube burst probability for S/G C at EOC-13 is 4.3 x 10" for the EPRI database and 1.2 x 10 2 for the NRC database. For projections of S/G A, the corresponding burst probabilities were about 35% lower than S/G C such that S/G C is the limiting S/G. When bobbin voltage distributions based on all bobbin indications or the new indication method are applied, the burst probabilities are about a factor of two lower than obtained for POD = 0.6. All results are below the guideline of 2.5 x 10-2 based on NUREG-0844 analyses. Thus the tube burst probability is also within acceptable limits for Cycle 13 operation.

2.7 Assessments of NDE Uncertainty for Analyst Variability I

j The laboratory reevaluation of the 1992 bobbin voltages performed as part of the growth study in j 1994 can be compared with the original 1992 voltage evaluation to asses. the analyst variability in sizing indications which is a part of the NDE uncertainties in the EPRI methodology. An NDE uncertainty distribution was developed as the difference between reevaluated and original voltage calls for the 1992 EOC-12 inspection. The resulting NDE uncertainty can be compared to the allowance of a 10.3% standard deviation for analyst variability uncertainty included in the EPRI ARC documentation and applied for the Farley 1 analyses. For EOC-12, which was the second IPC inspection implementing Appendix A analysis guidelines, the voltage differences resulted in a standard deviation of 8.7% for all indications and 8.5% for indications above 0.75 volts. Based on these evaluations,it is concluded that the 10.3% standard deviation for analyst variability in EPRI report TR-100407, Rev.1 (draft of August 1993)is adequate and no revisions are necessitated by the Farley-1 data.

l 2-6

3.0 EOC-11 AND EOC-12 S/G INSPECTION RESULTS I 3.1 EOC-11 Inspection Results EOC-11 inspection results are required in this report to define the BOC-12 indications left in service that are used as the starting point for projecting EOC-12 voltage distributions for comparison with the i

actual results from the EOC-12 inspection. The 1992 inspection at EOC-Il was the first Farley-1 l inspection implementing an IPC and applied eddy current data collection and analysis guidelines given in Appendix A of WCAP-12871 R.2 (Reference 3). This included use of ASME calibration standards normalized to the reference laboratory standard and probe wear standards.

Figure 3-1 shows the S/G C and all S/G bobbin indications found at EOC-Il and the indications confirmed as flaws by RPC inspection. The number ofindications found in S/Gs A and B are smaller than S/G C. For Figure 3-1 and all figures given in this report, the voltage values given on the x axis of the plots represent the upper or right-side value for the voltage bin. For EOC-11, only indications above the IPC repair limit of 1.0 volt were 100% RPC inspected, while a sample ofindications below 1.0 volts were inspected. S/G C has been applied for projected EOC-12 voltages and leak rates based on comparative leak rate calculations perf3rmed in 1992 that showed S/G C to be the most limiting.

EOC-12 inspection results show comparable leak rates for S/Gs A and C as described in Section 8. A l total of $85 bobbin indications were found in S/G C. Above 1.0 volt,317 indications were found in l

S/G C, of which 70 were RPC confirmed and removed from service principally due to confirmation as flaws by R.PC. A total of 1269 bobbin indications were found in all three S/Gs; 591 of these indications had bobbin voltages > 1.0 volt and 129 of the 591 indications were confirmed by RPC.

S/Gs A and B had 343 and 341 indications, respectisely. The indications left in service at BOC-12 are discussed in Section 4.1.

The largest bobbin voltage found in the inspection was 3.27 volts at S/G B R14C80 2H. RPC inspection confirmed the flaw mdication on this tube. The next largest bobbin voltages were each 3.26 volts at S/G A R19C41 IH and R38C72 IH which were also confirmed as flaws by RPC inspection.

No indications at TSPs were identified that had cracks extending outside the TSPs by either the bobbin or RPC inspections. In addition, no abnormal indications such as circumferentially oriented indications were found in the 1992 inspection.

3.2 EOC-12

Summary of Indications at TSPs The 1994 inspection at EOC-12 was performed in March,1994. This inspection was the second J. M.

l Farley-l inspection following implementation of an IPC. However, the repair limit was increased from 1.0 volt at EOC-11 to 2.0 volts at EOC-12, so that TSP indications of 2.0 volt or less could remain in service at BOC-13. The results of the EOC-12 inspection are therefore evaluated for IPC methodology considerations including comparisons between projected and actual indications in this report. The 1994 bobbin indications were classified as possible indications (PIs) and indications not reportable (INRs). The Pls are considered potential bobbin flaw indications. If a PI was RPC inspected, the indication was finally classified as a PCN if confirmed by RPC or a PIN if not confirmed by RPC. INRs are not considered as bobbin flaw indications and are not included in the bobbin flaw distributions of this report since this classification is typically assigned to bobbin Pls from 3-1

a prior inspection that are not considered as a potential flaw indication in the 1994 inspection. This classification is made to permit tracking of these signals between inspections. The IPC inspection requirements for this inspection required RPC inspection of all Pls above 1.5 volts.

Appendix A guidelines revised for the EOC-12 inspection require that indications in the ID phase angle range be called Pls, since prior RPC inspections have indicated that some ID phase angle indications are confirmed as flaws by RPC inspection. In addition, the Appendix A guidelines were l changed for the EOC-12 inspection to include OD phase angles up to 170*, which corresponds to 0%

depth (about 150* phase angle), as Pls. These Appendix A modifications were made following the 1993 Farley-2 inspection where ID phase angle indications were confirmed as flaws in the RPC l sampling plan. As noted later, these revised guidelines led to the reporting of two higher voltage indications that were RPC confirmed and not reported at the EOC-ll inspection.

A summary of the EOC-12 inspection results is given in Table 3-1. Figure 3-2 shows the total for all three S/Gs of all bobbin indications, RPC confirmed indications and repaired indications. Figure 3-3 shows the corresponding data for S/G C. S/G C with 691 indications left in service at BOC-13 is the limiting S/G for SLB leakage analyses as shown by comparative analyses with S/G A in Sections 8 and 9. Only a fraction of the bobbin indications below 1.5 volt were RPC inspected. Nineteen indications exceeded the approved 2.0 volt IPC repair limit for this inspection; all of these were RPC inspected and three were confirmed and repaired. A few tubes with indications at TSPs were removed from service due to other causes. From Table 3-1, the total number of potential flaw indications summed over all three S/Gs is 1683 which includes 509 in S/G A,481 in S/G B, and 693 in S/G C.

The total number of RPC confirmed potential flaw indications above 1.0 volt is 64 of 334 indications, for an RPC confirmation rate of about 19%. Below 1.0 volt,3 of 5 indications RPC inspected were confirmed. Table 3-1 includes the distribution of RPC confirmed indications ar.d repaired indications as well as the total bobbin voltage distribution. A total of 6 indications were repaired including 2 in the limiting S/G C. A more detailed assessment of the high fraction of unconfirmed RPC indications is given in Section 3.3 below.

Table 3-2 provides a summary of the largest bobbin voltage indications (all indications greater than or equal to 2.0 volts) found in EOC-12. RPC inspection results, voltages at the prior inspection and whether or not it is a new indication not reported at the last inspection are also included in Table 3-2.

The largest bobbin voltage indications found in 1994 were 4.80 volts in S/G A,3.02 and 2.92 volts in SiG B and 2.56 and 2.46 volts in S/G C. The largest bobbin indication of 4.80 volts,in SG A at i R37C25 4H, was RPC confirmed (as a 2.49 volt SAI). In 1992 this location had an ID phase angle, and not called per the 1992 EC guidelines. The bobbin data for the '94 and '92 inspections are given  !

in Figure 3-5. The phase angle remained ID (typically < 35 )in the '94 inspection. Revisions made I to the Appendix A analysis guidelines prior to the 1994 inspection require that phase angles in the ID plane from 10 to 35* be called as Pls and this indication was reported in the latest inspection. In S/G B the largest bobbin indication,3.02 volts at R38C33 2H, was also RPC confirmed in 1994. In the prior inspection, this location was a zero percent depth signal (Figure 3-6), and was also not called per the 1992 EC guidelines. Although the '94 phase angle also implies 0% depth, the Appendix A guidelines were revised prior to the '94 inspection to require all phase angles up to 170* to be identified as Pls and the indication was reported in '94. The third largest RPC confirmed indication in S/G A, R17C45, TSP 1 had a bobbin indication of 2.45 volts and also represented a new indication not reported in the '92 inspection. The bobbin data for this indication is shown in Figure 3-7. This indication has a highly distorted bobbin response and is somewhat more indicative of a flaw in the '94 inspection The bobbin voltage of 2.45 volts is a very conservative assignment for this signal which 3-2 l

follows the guidance of Appendix A. The low RPC 0.64 volt amplitude is more typical of a response for a bobbin indication nearer to one volt. For detectability considerations, this previously unreported indication would be moa appropriately considered as an indication below the repair limit of 2.0 volts.

l All three of the remaining six largest indications (2.92,2.56 and 2.46) were RPC NDD in 1994, and all had been reported as bobbin indications and found RPC NDD in 1992. All of the remaining 14 bobbin indications above 2.0 volts in Table 3-2 were RPC NDD with many of these indications also RPC NDD in '92. Farley-1 has an increasing number of consecutive inspection RPC NDD indications which by draft NUREG-1477 guidelines must be included in leakage and burst analyses. This issue is j funher addressed in Sections 3.3 and 10.0. Of the 10 new indications in Table 3-2,3 were RPC I confirmed and 7 were RPC NDD. A more detailed assessment of the new indications found in the l 1994 inspection is giver in Section 3.4 below.

The augmented RPC inspection program included all dents > 5.0 volts which included 62,40 and 44 TSP intersections in S/Gs A, B and C, respectively. In addition,10 TSP intersections with artifact signals that could potentially mask bobbin calls are included in the augmented program. No RPC indications were found in this augmented inspection. No cirv.imferential indications or indications extending outside of the TSPs were found in the base or augmented inspections.

3.3 EOC-12

Assessment of RPC Inspection Results The largest 1994 RPC confirmed indication was at S/G A R37C25 4H with a bobbin voltage of 4.80 l volts. RPC inspection of this location in 1994 showed a 2.49 volt SAI, however, this was NDD by bobbin inspection in 1992 and therefore leflin service. All but three of the largest 20 bobbin indications in Table 3-2 were RPC NDD. From Table 3-1, only 67 of 339 or 20% (19% for bobbin indications >l.0 volt) of the total bobbin indications that were RPC inspected were confirmed as flaw indications. For S/G C, which had the largest number of RPC inspected indications, the RPC confirmation rate was 17% for all indications as well as 17% for indications above 1.0 volt. S/G A l had the highest RPC confirmation rate of 32%. Even if the bobbin indications not confirmed by RPC l are assumed to be real flaws, the indications would be too small to contribute significantly to leakage and burst probability. Thus, only RPC confirmed plus not RPC inspected bobbin indications should be included in SLB analyses for EOC-12 conditions or used for comparisons with projected EOC voltage distributions.

For considerations of including the RPC NDD indications left in service for projecting SLB leak rates l at the end of the next operating cycle, the relevant question is whether or not the RPC NDD indications tend to remain as NDD at the end of the next cycle. Table 3-3 provides a summary of the l EOC-12 inspection results for indications that were RPC NDD at EOC-Il and left in service at BOC-

12. Quite a few (222 indications) of the RPC NDD indications were not inspected at EOC-12 since the bobbin voltages were less than 1.0 volt. The results support the small fraction of RPC NDD indications that become flawlike at the end of the subsequent operating cycle. Of the 217 indications RPC NDD in 1992 and RPC inspected in 1994,11% of the 1992 RPC NDD indications became confirmed flaws in 1994. S/G C had the highest fraction at 13% (12% for indications > 1.0 volt) of 1992 RPC NDD indications confirmed as flaws in 1994. Thus the majority of RPC NDD indications l continue to be RPC NDD and are most likely indicative of false bobbin calls. Similar trends of RPC l NDD indications remaining RPC NDD were found in the IPC assessments for the Cook-1 1994 and l

Farley-2 and Catawba-1 1993 inspections. A rate of 20% for RPC NDD indications becoming confirmed indications at the next cycle bounds all IPC assessments performed to date.

3-3 l

I

r Based on the large fraction (>81% above 1.0 volt) of RPC NDD indications and the fact that the majority (89% of all indications) of RPC NDD indications remain RPC NDD at the end of the next operating cycle, it is concluded that including all RPC NDD indications left in service in the population of tubes for projecting SLB leakage is excessively conservative. This conservatism would further increase in successive IPC cycle applications, since the number of RPC NDD indications grows from cycle to cycle. For example, bobbin indications that are RPC NDD for two successive inspections should be excluded from the SLB analyscs and only a plant-specific fraction of the new RPC NDD indications should be included in the SLB analyses. This topic is further addressed in the discussion of Section 10 which provides an alternate model for defining BOC indications left in service. It is noted that the conservatism for RPC NDD indications becomes further magnified when a POD correction, such as required by draft NUREG-1477,is applied to all bobbin indications found in

! the inspection. In this case, all RPC NDD indications left in service become 1.67 indications left in service for a POD of 0.6.

l l 3.4 EO C-12: Assessment of New Indications Given the occurrence of new indications in the largest bobbin voltages of Table 3-2, the 1994 inspection results were evaluated to categorize the 1994 indications as prior or new (not reported 'n

! 1992) indications. Tables 3-4a and 3-4b summarize this evaluation for the sum of all S/Gs and fo.

j S/G C, respectively. A number of bobbin indications reported in the EOC-Il inspection and left in l service were not found (INR in '94)in the EOC-12 inspection. Figures 3-4a and 3-4b show the l distributions for all S/Gs and for S/G C of the '92 bobbin indications which were not reported (mostly INRs)in the '94 inspection, and the distribution of new indications in '94. The figure shows that many low voltage, bobbin potential indications are not found in the subsequent inspection and can be integreted as false bobbin calls. Similar results were found in the Cook-1, Farley-2 and Catawba-1 IPC cvaluations. This type ofindication is given in Tables 3-4a and 3-4b as the column called Prior Ind. Not Found. The net number of effective new indications is also given in Tables 3-4a and 3-4b and is obtained as the total number of new indications mmus the number of prior indications not found. This net number of new indications is more relevant to POD considerations than the total number of new indications when conservative bobbin calling criteria are applied. As would generally be expected from application of conservative bobbin calling criteria, a significant number of the EOC-12 indications are new indications.

Tables 3-4a and 3-4b also include the number of new indications that were RPC inspected and the number that were RPC confirmed. Again, the fraction ofindications that are confirmed as flaws by RPC inspection is small. For S/G C and the sum of all S/Gs, the net number of new indications represent about 23% and 29%, respectively, of the total bobbin indications. Eighteen of 67 or 26.9%

in SG-C, and 36 of 107 or 33.6% for all S/Gs, of the new indications RPC inspected were confirmed by RPC.

Table 3-5 provides a summary of the RPC inspection results for C/G C and t>c sum of all S/Gs. In Table 3-5, the RPC confirmation rate is separately identified for prior bobbin ndications left in service that were RPC confirmed plus not RPC inspected (not RPC NDD), for Prior 1 PC NDD indications left in service, for new indications and for the sum of all indications. It is seen that the RPC confirmation rates are low (11.1 and 13.1%) for prior RPC NDD indications. For ptior bobbin indications left in service that were RPC confirmed or were not RPC inspected, the RPC confirmation rate is higher at 46.7% for all indications and 25.0% in S/G C. The RPC confirmation rate for new l i

3-4 I

indications is similar to that found for prior indications left in service. Thus the prior RPC NDD j indications, which are a small fraction of the total bobbin indications, contribute a small portion (24 of 67 indications) of the total RPC confirmed indications.

The EOC-12 inspection results can be used to estimate the detection probability of significant indications at the prior cycle inspection. For IPC/APC applications, a significant indication for leakage and tube burst considerations at the EOC would be RPC confirmed indications. Indications that could have significantly contributed to leakage in the prior cycle would be expected to be detected by both bobbin and RPC inspections at the EOC. Thus the appropriate population of indications for assessing detectability is the population of RPC confirmed indications. The current or l latest inspection includes both indications detected at the prior inspection plus new indications -

(indications not reported at the prior cycle inspection). An appropriate detection probability for the population of RPC confirmed indications is the probability of detection at the prior cycle inspection.  !

l This is called the probability of prior cycle detection (POPCD) as described in Table 3-6. POPCD is defined as the probability that an indication found by bobbin inspection and confirmed by RPC was detected and reported at the prior cycle inspection. Since some RPC confirmed indications may have been repaired at the prior inspection, it is appropriate that the population include indications RPC l confirmed in the latest inspection plus those RPC confirmed and repaired at the prior inspection. If no l indications were repaired at the prior inspection, the population would be all RPC confirmed indications m the current mspection.  !

l Since the population of RPC confirmed indications is small for J. M. Farley-1, the POPCD was developed based on the sum ofindications over all three S/Gs and is given in Table 3-7. The POPCD is seen to increase from about 64% below 1.0 volt to about 95% at 2.5 volts. Above 2.5 volts, the POPCD is alTected by the guidelines that did not require calling of ID phase angle indications and 0%

depth indications as discussed above. If the two subject indications are not included in the POPCD evaluation, the detection probability above 2.5 volts would be 100% based on detecting all five of the RPC confirmed indications in the 1992 inspection. When the two indications are considered as not detected in the '92 inspection, the POPCD would be 71.4% for indications above 2.5 volts. It is expected that the new Appendix A guidelines implemented for the '94 inspection will result in detection probabilities approaching 100% in the future. The results for the population of 193 RPC confirmed indications support high detection probabilities for significant indications at J. M. Farley-l.

l The POD = 0.6 proposed in drafl NUREG-1477 appears to be excessively conservative for J. M.

l Farley-l .

The number of new indications implies that an allowance must be included in projected SLB leakage analyses for new indications. The allowance for new indications should include considerations of the expected RPC confirmation rate. A proposed model for including new indications in the SLB analyses is described in Section 10. This proposed model applies a voltage distribution for the new l indications rather than applying a POD or POPCD adjustment to the detected indications. The use of l a constant POD independent of voltage is not justified and the proposed method is based on the bobbin voltage at the prior inspection (from reanalyses of prior eddy current data for voltage growth analyses) for new indications found in the current inspection. The Section 10 model applies the number of new indications as given in Table 3-4b at the EOC-Il volts to represent undetected indications. This assumes that the distribution of undetected indications at the EOC-12 inspection is the same as found at the EOC-Il inspection.

! 3-5 l

3.5 EOC-12 Distributions for Comparisons with Projections Voltage projections from BOC to EOC conditions for IPC/APC applications are performed to cetimat:

SLB leakage and burst probability. Thus the desired EOC voltages are those having significant potential for throughwall cracks or leakage. Indications that are RPC NDD at EOC can be confidently expected to have no potential for leakage over the prior cycle as RPC detectability (as well as bobbin detectability) approaches 100% for ODSCC at TSPS with near throughwall or throughwall indica: ions.

It can be noted that the confidence for no leakage over the prior cycle for RPC NDD is much higher than the judgement for leakage potential over the next cycle from RPC indications left in service (i.e.,

the NRC requirement to include RPC NDD in projected leakage analyses). Therefore, for comparisons of projected EOC-12 voltage distributions to be used in leakage analyses with actual distributions, the appropriate actual EOC-12 vohage distribution is the sum of RPC confirmed indications and bobbin indications not RPC mspected (below 1.0 volt for J. M. Farley-2). These distributions are shown in Figure 7-la for S/G C which is the limiting S/G for SLB leakage analyscs for both Cycles 12 and 13. li,s S/G C actual distributions of Figure 7-la for EOC-12 are discussed with projections from BOC-12 to EOC-12 in Section 7.

1 4

k l l

3-6

Table 3-1: Farley 1,4/94 Outage, Comparison of PIs Found at TSPs, ConFrmed by RPC, and Indications Repaired Steam Generator A Steam Generator 11 Steam Generator C All Steam Generators 94 Volts No of No of No of No of No of No of No of No of No of No of No. of No. of No. of No of No. of No of EMbin In& cations In& cations inacations Bobbin indications inacations inacations Hobbin Indications Indications Indications Bobbin In& cations in& cations Indications in& cations ruamined Confirmed Repaired in& cations Examined Confirmed Repaired Indications Esammed Confirmed Repaired In& cations Examined Confirmed Repaired 00 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 01 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 02 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 03 0 0 0 0 1 0 0 _ 0 0 0 0 0 1 0 0 0 04 8 0 0 0 6 0 0 0 5 0 0 0 19 0 0 0 05 28 0 0 0 21 0 0 0 16 0 0 0 65 0 0 0 06 40 0 0 0 40 0 0 0 47 _0 0 0 127 0 0 0 07 61 1 1 0 58 0 0 0 64 0 0 0 183 1 1 0 08 73 1 0 0 64 0 0 0 81 0 0 1 218 1 0 1 09 62 I I O 54 0 0 0 84 0 0 0 200 I I O 10 57 0 0 0 52 1 0 0 83 I I O 192 2 I O 1.1 41 1 1 0 51 0 0 0 81 10 0 0 173 Il 1 0 12 32 0 0 0 36 0 0 0 52 52 7 0 120 52 7 0 ~

1.3 30 3 0 0 23 0 0 0 51 51 11 0 104 54 11 0 I4 22 3 0 0 26 0 0 0 29 29 2 0 77 32 2 0 1.5 13 5 2 0 12 1 0 0 38 38 5 0 63 44 7 0 I6 9 9 2 0 10 10 2 0 19 19 8 0 38 38 12 0 17 11 11 2 0 7 7 0 0 15 15 4 0 33 33 6 0 18 8 8 3 0 9 9 3 0 7 7 1 0 24 24 7 0 19 6 6 2 I I I O O 8 8 3 1 15 15 5 2 20 4 4 2 0 4 4 1 0 4 4 0 0 12 12 3 0 2.1 0 0 0 0 1 I O O 2 2 0 0 3 3 _ 0 0 22 2 2 0 0 I I O O 2 2 0 0 5 5 0 0 2.3 0 0 0 0 1 1 0 0 2 2 0 0 3 3 0 0 24 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2.5 1 1 I I I I O O 2 2 0 0 4 4 I I 2.6 0 0 0 0 0 0 0 0 1 I O O I I O O 2.7 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 28 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2.9 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 3.0 0 0 0 0 1 I O O O O O O I 1 0 0 31 0 0 0 0 I I I I O O O O I I I I 48 I I I I O O O O O O O O I i 1 I lotal 509 57 18 3 481 39 7 1 693 243 42 2 1683 339 67 6 l otal > l V 180 54 16 3 185 38 7 1 313 242 41 1 678 334 64 5 ALA_ DATA XLS Table 3 I R.2 3-7

Table 3-2. Summary of Largest Farley-1 EOC-12 Bobbin Voltages l l f EOC-12

BOC-12 I Bobbin RPC Bobbin New SG Row Col TSP Volts Volts Volts Indication l A 37 25 4H 4.80 2.49 4.47 Yes '"

B 38 33 211 3.02 1.38 2.83 Yes '2' B 16 6 3H 2.92 NDD 2.72 No

C 33 40 2H 2.56 NDD 2.84 No ")

[ C 41 49 2H 2.46 NDD 2.47 No

! A 17 45 111 2.45 0.64 2.12 Yes C 11 9 5H 2.45 NDD 2.89 Yes

! B 42 32 611 2.44 NDD 2.18 Yes )

1 C 41 67 311 2.26 NDD 2.25 No )

B 14 3 5H 2.25 NDD 1.84 Nt (')

l C 26 81 2H 2.24 NDD 2.56 Yes C 41 67 4H 2.19 NDD 1.98 e No )

A 43 50 4H 2.18 NDD 2.17 *. Yes i

B 34 75 4H 2.15 NDD 2.04 ' No (3)

A 19 81 IH 2.11 NDD 2.21 Yes C 34 17 3H 2.11 NDD 2.12 Yes C 22 77 3H 2.08 NDD 2.17 No ")

B 37 23 311 2.06 NDD 1.81 No

C 33 74 3H 2.06 NDD 1.80 No )

A 40 64 2H 2.00 NDD 1.83 Yes ,

i 1

(U 1992 signal was an unusual ID phase angle and not called per the 1992 EC guidelines. l

<2 >

1992 signal was a zero percent depth signal, and not called per the 1992 EC guidelines. I

RPC NDD at last inspection.

") This was reported as an INR in '92, and was also RPC NDD in '92.

ALA_ALL.XLS T3-2 3-8

i Table 3-3. EOC-12 INSPIX. TION REdULTS FOR RPC "NDD" INDICATIONS 1. EFT IN SERVICE AT 110C-12 Steam Generator A Steam Generator 11 Steam Generator C All Steam Generators NDD in 92 and NDD in 92 and NDD in '92 and NDD in 92 and Voltage NDD in NDD in Degradstion NDD in NDD in ikgradation NDD in NDD in Degradation NDD m NDD in Degradation Range" 92 92 and 94 found in 94 N2 92 and N4 found in 94 92 92 and 94 found in 94 92 92 and '94 found in '94

$ 0.4 0 0 0 0 0 0 0 0 0 0 0 0 0.11-0.20 0 0 0 0 0 0 0 0 0 0 0 0 0.21-0.30 0 0 0 0 0 0 0 0 0 0 0 0 0.3I-0.40 0 0 0 0 0 0 0 0 0 0 0 0 0.41-0.50 0 0 0 0 0 0 0 0 0 0 0 0 0.51-0.60 I O O O O O O O O I O O 0.61-0.70 1 0 0 3 0 0 1 0 0 5 0 0 0.71-0.80 3 0 0 3 0 0 1 0 0 7 0 0  ;

0.81-0.90 7 0 0 5 0 0 8 0 0 20 0 0 0.91-1.00 11 0 0 6 0 0 15 0 0 32 0 0 1.01-1.10 12 0 0 14 0 0 46 8 0 72 8 0  !

1.11-1.20 13 0 0 15 0 0 33 29 2 61 29 2 1.21-I.30 13 2 0 15 0 0 33 27 6 61 29 6 1.31-1.40 6 1 0 17 0 0 22 19 I 45 20 1 1.41-I.50 7 2 0 8 1 0 26 22 4 41 25 4 1.51-l.60 4 3 0 7 7 0 16 11 5 27 21 5 1.61-1.70 7 7 0 6 6 0 11 9 2 24 22 2 1.71-1.80 1 I O 6 4 2 6 6 0 13 11 2 1.81-1.90 4 4 0 1 1 0 7 5 2 12 10 2 1.91-2.00 1 1 0 3 3 0 4 4 0 8 8 0 2.01-2.10 0 0 0 1 I O 2 2 0 3 3 0 2.11-2.20 0 0 0 1 1 0 1 1 0 2 2 0 2.21-2.30 0 0 0 I I O I I O 2 2 0

~

2.31-2.40 0 0 0 0 0 0 0 0 0 0 0 0 2.41-2.50 0 0 0 1 I O 1 1 0 2 2 0 2.51-2.60 0 0 0 0 0 0 I I O I 1 0 I Ol'AL 91 21 0 113 26 2 235 146 22 439 193 24 Tol'AL > l V 13 13 0 20 18 2 34 30 4 67 61 6

~~

Notes: 1. All soltages are '94 soltages.

~ ~ ~ - '

2. 1994 INR signals are included.

FARLT3_3.XLS Table 3-3 3-9

\

l l

l Table 3-4a: Prior Indications and New Indications at EOCl2 for All SGs EOCl2 Voltages Lesser of EOCl2 and EOCll Voltages EOCl2 RPC New Net New ind's Indications also Confirmed or Prior ind ind NewInd. RPC Confirmed Found in New Cycle 12 New Ind's RPC New Ind's RPC New C)cle 12 Not Tested of Not Found RPC RPC or Not RPC Volts EOCll ' Indications 2 Inspected ConGrmed Indications ' Previous Col. NDD' Confirm'd ' Exarmned '.

o.1 0 0 0 0 0 0 0 0 0 0 0.2 0 0 0 0 0 0 0 0 0 0 0.3 1 0 0 0 1 I O O O I 0.4 8 11 0 0 26 26 0 0 0 26

0. 5 31 34 0 0 50 50 5 0 0 45 0.6 58 69 0 0 74 74 11 0 0 63  ;

O7 106 77 0 0 88 87 11 1 0 75 0.8 129 89 1 0 91 87 6 3 0 78 0.9 127 73 I I 84 81 6 3 1 72 l .0 120 72 2 1 70 69 4 0 3 65 1.1 118 55 2 0 41 38 5 3 0 30 1.2 h 76 44 16 4 29 16 2 13 5 5 1.3 64 40 19 5 37 22 2 15 8 8

( l .4 51 26 10 1 16 8 0 8 1 1 1.5 43 20 13 2 19 10 1 9 7 7 1.6 26 12 12 7 7 3 0 4 3 3 1.7 25 8 8 4 5 2 0 3 2 2 1.8 15 9 9 4 5 3 0 2 3 3 1.9 14 I i 1 1 0 0 1 0 0 2.0 8 4 4 3 0 0 0 0 0 0 2.1 3 0 0 0 0 0 0 0 0 0

[ 2.2 2 3 3 0 5 1 0 4 I I 2.3 2 I i 0 1 0 0 1 0 0 24 0 0 0 0 0 0 0 0 0 0 2.5 1 3 3 1 1 0 0 1 0 0 2.6 1 0 0 0 0 0 0 0 0 0 2.7 0 0 0 0 0 0 0 0 0 0 2.8 0 0 0 0 0 0 0 0 0 0 2.9 0 0 0 0 1 1 0 0 1 1 30 1 0 0 0 0 0 1 0 0 0 3.1 0 l I i 0 0 0 0 0 0 3.2 0 0 0 0 0 0 0 0 0 0 l 3.3 0 0 0 0 0 0 0 0 0 0 34 0 0 0 0 0 0 0 0 0 0 l 35 0 0 0 0 0 0 0 0 0 0 3.6 0 0 0 0 0 0 0 0 0 0 37 0 0 0 0 0 0 0 0 0 0 4.5 0 0 0 0 1 1 0 0 1 1 48 0 1 1 1 0 0 0 0 0 0 l Totals 1030 653 107 36 653 580 54 71 36 487

' Pl. PIN, and PCN voltages from EOCl2 ('94) inspection which were P1 in'92. (Excludes INR) 2 PI, PIN. and PCN from EOCl3 ('941 inspection which were not P1 in 92. (Excludes INR) l

' The smaller of the 92 or 94 reanalysis voltages of Pl, PIN, and PCN reported in EOCl2 (94) but not P1 in'92 l

' The EOCl1 field voltage for indications found in EOCil but not in EOCl2 (i c., INR plus '92 not reported in '94).

l

' The smaller of EOCl2 and EOCil reanalysis voltages for new EOCl2 indications which were also RPC NDD in EOCl2.

l

• The smaller of EOCl2 and EOCil reanalysis voltages for new EOCl2 indications which were also RPC confirmed (MAI, SAI)in EOClz. '

' The larger of new indic'ns RPC confirmed and (new ind's (RPC NDD or not RPC insp.) minus prior ind's not found minus new ind's RPC NDDj ALASUM.XLS ALA-All 3 10 j

1 Table 3 4h: Prior Indications and New Indications at EOCl2 for SG-C EOC12 Voltages lesser of EOCl2 and EOCll Voltages EOCl2 New Net New Ind's Indications also RPC Confirmed lnd New Ind RPC Confirmed Found in New C)cle 12 New Ind's RPC New Ind's RPC New Cycle 12 or Not Tested of- Prior ind RPC RPC or Not RPC 2

Volts EOCll ' Indications Inspected Confirmed Indications ' Previous Col. NotFound' NDD' Confirm'd

• Examined '

O. ! 0 0 0 0 0 0 0 0 0 0 0.2 0 0 0 0 0 0 0 0 0 0 0.3 0 0 0 0 0 0 0 0 0 0 04 1 4 0 0 6 6 0 0 0 6 0.5 9 7 0 0 10 10 0 0 0 10 0.6 23 24 0 0 26 26 5 0 0 21 0.7 36 28 0 0 34 34 4 0 0 30 08 45 36 0 0 32 31 2 1 0 28 0.9 58 26 0 0 32 30 3 2 0 25 1.0 55 28 I i 25 25 0 0 3 25 1.1 65 16 2 0 10 7 3 3 0 l 3.2 36 16 16 4 18 5 1 13 5 5 1.3 33 18 18 5 17 5 2 12 5 5 14 21 8 8 1 7 0 0 7 0 0 1.5 27 11 11 1 9 3 1 6 3 3 1.6 16 3 3 3 1 1 0 0 1 1 1.7 11 4 4 2 3 1 0 2 I i 1.8 6 I I I O O O O O O 1.9 8 0 0 0 0 0 0 0 0 0 2.0 4 0 0 0 0 0 0 0 0 0 2.1 2 0 0 0 0 0 0 0 0 0 2.2 1 1 1 0 1 0 0 1 0 0 2.3 i  ! l 0 1 0 0 1 0 0 2.4 0 0 0 0 0 0 0 0 0 0 2.5 1 1 1 0 1 0 0 1 0 0 2.6 1 0 0 0 0 0 0 0 0 0 2.7 0 0 0 0 0 0 0 0 0 0 28 0 0 0 0 0 0 0 0 0 0 2.9 0 0 0 0 0 0 0 0 0 0 3.0 0 0 0 0 0 0 0 0 0 0 Totals 460 233 67 18 233 184 21 49 18 161

' P.I. PIN, and PCN soltages from EOCl2 ('94) inspection which were PI in '92 (Excludes INR)

' Pl. PIN. and PCN from FOCl3 094) inspection which were not PI in '92. (Excludes INR)

'The smaller of the '92 or '94 reanalysis voltages of Pl. PIN. and PCN reported in EOCl2 094) but not P1 in 92.

'The EOCli field voltage for indications found in EOClI but not in EOCl2 (i c.,INR plus '92 not reported in '94).

• The smaller of EOCl2 and EOCll reanalysis voltages for new EOCl2 indications which were also RPC NDD in EOCl2.

' The smaller of LOCl2 and EOCil reanalysis soltages for new EOCl2 indications which were also RPC confirmed (mal, SAI)in FOCl2.

' The larger of new indic'ns RPC confirmed and [new ind's (RPC NDD or not RPC insp ) minus prior ind's not found minus new ind's RPC NDD).

ALASUM.XLS ALA-C 3-11

Table 3-5. Summary of RIC Confirmation Data for S/G-C and All S/Gs Group Total '94 Total '94 Total '94 Percent of Bobbin RPC RPC '94 RPC Indications Indications Inspected Confirmed Confirrmd S/G-C l All Voltages "I t

Prior Bobbin Ind. not RPC NDD in '92 231 8 2 25.0 %

l .

Prior Bobbin Ind. RPC NDD in '92 229 168 22 13.1 %

1 i New '94 Indications 233 67 18 26.9 %

Sum of All Indications 693 243 42 17.3 %

Sum of All S/Gs 1 All Voltages")

Prior Bobbin Ind. not RIC NDD in '92 600 15 7 46.7%

Prior Bobbin Ind. RPC NDD in '92 430 217 24 11.1%

New '94 Indications 653 107 36 33.6 %

Sum of All Indications 1683 339 67 19.8 %

0) RPC inspection was performed primarily for indications >l.0 volt, so no distinction has been made between all indications and indications >1.0 volt, as in previous studies.

l 3 - 12 i

l

l l

l Table 3-6. Description of Pmbability of Prior Cycle Detection (POPCD)

OBJECITVE Define a bobbin coil detection pmbability for significant (RPC confumed) indications that is applicable to APC applications and that can be developed fmm inspection data i

POPCD - Probability of Prior Cycle Detection Probability that an indication found by bobbin inspection and confumed by RPC was detected and reported at the prior cycle inspection Pepulation includes RPC confirmed indications in latest inspection (prior indications left in service plus new indications) plus RPC confirmed and repaired indications at the prior inspection Developed as e function of vol: age using voltage bins Voltages based on values at prior c> ele inspection New indications (not reported at prior cycle inspection) include:

Indications at detectable levels and not rqx>rted Indications below detxtable level::

Indications initiated since last inspection Application considerations for eew indication voltages at prior inspection:

- Prior cycle new inaication vo'tages obtamed from growth evaluation

- Growth study conservatively calls a flaw if any suggestion for a flaw call exists

, - If a nev/ indication is found NDD at prior cycle, it would be included as a 0.5 volt l indication which is typical ofindications masked by noise, etc., unless the eddy current data j suggests a basis for a higher threshold l

Development of POPCD

)

POPCD = Prior Cycle Reported Ind. Confumed by RPC: Confirmed in l

bitest Inspection Plus Confumed and Repaired in Prior Inspection Numemtor Plus New Ind. RPC Confumed in Latest Inspection 1

3-13

[

t <

I l

l Table 3-7.

J. M. Farley-1: Probability of Prior Cycle Detection (POPCD) at EOC-12 l

l

'92 Indications '92 Indications New '94 Prior Cycle Confirmed by Confirmed by Indications Voltage ('92 RPC in '94 RPC and Confirmed by Group Volts) Inspection Repaired in '92 RPC in '94 Bin POPCD pol'CD 0.6 0 0 0 ---

0.7 0 0 0 - 63.6 %

0.8 1 0 0 100.0 % 7/11 Ind.

0.9 1 0 1 50.0 %

l.0 5 0 3 62.5 %

1.1 2 13 0 100.0 %

1.2 3 11 5 73.7 % 80.7 %

1.3 4 13 8 68.0 % 88/109 Ind.

1.4 2 25 1 96.4 %

1.5 5 10 7 68.2 %

1.6 4 6 3 76.9%

1.7 2 11 2 86.7 % 83.0 %

1.8 0 3 3 50.0% 39/47 Ind.

1.9 2 7 0 100.0 %

2.0 0 4 0 100.0 %

2.1 0 8 0 100.0 %

2.2 0 4 1 80.0 % 95.2 % <

2.3 0 4 0 100.0 % 20/21 Ind.

2.4 0 1 0 100.0 %

2.5 0 3 0 100.0 %

2.6 0 1 0 100.0 %

2.7 0 1 0 100.0 %

2.8 0 0 0 ---

71.4 %

2.9 1 "' 0 0 ---

5/7 Ind.

3.0 0 0 0 --

3.1 0 0 0 ---

100.0 %

3.2 0 0 0 --

5/5 Ind.

• 3.3 0 3 0 100.0 %

3.4 0 0 0 --

3.5 0 0 0 ---

4.5 l <2' 0 0 ---

• SG-B R38C33 2H was not called in '92 since it was reported as 0% depth.

• SG-A R37C25 4H was not called in '92 since it was a UIA.

• Expected POPCD if 1994 Appendix A Guidelines had been applied for 1992 inspection.

ALA_ DATA.XLS POPCD T3.7 3-14

Figure 3-la. J. M. Farley-1: EOC-11 Bobbin Voltage Distribution for All SGs and RPC Confirmed Indications 200 1

180 160 g 140 - -

l 5 120 - -

.2

] 100 - - -

80 - - - - - - -

p 60 - - - - - -

40 - - - - - - - - -

20 -- - - - - ~

0 . 1 1 1 1 1 ,mLLott ___- - _

0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0 2.2 2.4 2.6 2.8 3.0 3.2 Volts

[u All Indications O RPC Confirmed Indication [

Figure 3-lb. J. M. Farley-1: EOC-11 SG-C Bobbin Voltage Distribution and RPC Confirmed Indications 80 I

70

, 60 - --

5 j 50 - -

.2

] 40 l

T 30 - - - - - -

d Z 20 - - - - - - - - -

10 - - - - - - -- -

0 I l l l l l bmbmm _ _

0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0 2.2 2.4 2.6 2.8 3.0 3.2 Volts l m All Indications O RPC Confirmed Indications]

ALA_ALL.XLS Figure 3-1ab 3-15

! l k

4  ;

I i

1

) Figure 3-2. Farley-1: EOC-12, All S/Gs Bobbin Voltage Distributions a

i 250 i

i 1 .._

1 i 200 i

{

i j _ _ _ _. ..

4 3 _ _ _ _

1 j 150

! E o _ _ _ _ _

1 =

1 a _ _ __ _

i E 9 n== -_ _ _ ._ ._ _ .__

1 O f _ __ _ _

Z

100 l 4 ,

l t

{ 50 4

4 _ ._ _

3 -.._. _. _ _ _ __ _ _ _ _ _

_ _g 0 - - - - - - l ' l l l b'I' L- - --

I 0.30.40.50.60.70.80.91.01.11.21.31.41.51.61.71.81.92.02.12.22.32.42.52.63.03.14.8

) Volts e

[s All Bobbin Ind. O RPC Confirmed Ind. O Repaired Ind.

ALASUM.XLS Figure 3-2 Chart i

f I

i l i l l

Figure 3-3. Farley-1: EOC-12, SG-C Bobbin Voltage Distributions 1

i

(

90 i

l 80 1 i __ _ _

! 70 i - -

l _ _ _-

j _ _ _ _

t 60 t

t _ _

s, _ _ _ _

p _ _ _ _ _

1 /t 50 j  ;; _ _ _ _ _ _

j g _ _ _ _ _ _ _

= _ _ _ _ _ _ _

3

! o

? . 40 l } _ - -

30 20 3

j _ _ _ _ _ _ _ _ _ _ _ _ __

10 j _ _ _ _ _ _ ._ _ _ _ _ _ . ._

j _ _ _ _ _ _ _ _ _ _- _ _. _ ._ _ -

_ l _ g _ g _g ____ , s

! 0.40.50.60.70.80.91.01.1 1.21.31.4 1.51.61.71.81.92.02.12.22.32.42.52.6 i

4 Volts i  ! s All Bobbin Ind. O RPC Confirmed Ind. O Repaired Ind. l 1  !

! l 1

3-17 j ALASUM.XLS Figure 3-3 Chart 1 i

5

l i

i 1

l Figure 3-4a. J. M. Farley-1: EOC-12 New Indications l and EOC-11 Indications Not lleported at EOC-12 for All SGs t

l 90 j 80 - ---- - ------- - -

i 70 l

l E 60 - - - - -- - - - - - - - - - - - - - - - .~ - -

8 i 5 50 --- - .- - -- - - - - -- -- - - -

l E

~

j 40 - - -- - - - - - - - - - ---- ~ -

e C j 30 - - - - - -- - - - ---. - .--- --

l 20 -- - - - - - - - - - -- - - . - -- -- - --

1 1 10 - - - - _ _ - - - - -_ -- - - -- - --

0 l l - 1 ' ' - E E - a _ _ _

0.40.5060.70.80.91.01.1 1.21.3141.51.61.71.81.92.02.12.22.32.42.53.03.148 Volts j E eE'94 Ind.I'9450its[b52 InUNot IkhdIn '94 ('92 Volts) -

4 j Figure 3-4b. J. M. Farley-1: EOC-12 New Indications and EOC-11 Indications Not Reported at EOC-12 for SG-C l

4 40 f

35 l

30 4 E 3 25 ---

5

< 3 20 - ---

.5 1 i I j . 15 ----- ---- - - -

l t z j 10 -- -----

l 5 -l--- - - - - - -

i 0 iI i I. . . .

,f 04 0.5 0.6 0.7 08 0.9 1.0 1.1 1.2 1.3 1.4 1.5 1.6 1.7 1.8 1.92.02.1 2.22.32.4 2.5 Volts j l m New '94 Ind. ('94 Volts) O'92 Ind. Not Rep'd in '94 ('92 Volts) l ALASUM.XLS Figure 3-4a 3-18 l

l

es AIN 19 CM 6 9 5.19 1:5 nts 1 23 17 H 2ee sJu Cn 3 41 ggag g REZ1M 01la M SIDE E

^ "

5/5 M@elT M OUTL11

~

et l +4 00 l:l lim CITENT TEM l f(C E g ,,

SPf1D 21.2%l in/see Thip X --

og .

~~

vpp 4 30 M 29 tit vpp 1 99 M 4 943 g -- ( S k $

f d X -r h

%] 1 R

M

)

C R

6.34 ese us CN 3 29 11.34 su luw CM 5 52

]

'l  % x

/  ;

=

~

p  !

i arteises sta.: S Sef t s .9 30 Set 1. w as es:esres em vpp 4.12 MM 953 vpp 4.22 M 44 f k h rart 3 . s 7 s f

_ [ D IN --

,' 7 3

< e W - >

f

g.  ; w 3 a a > -

W mia i V CM 6 V 6 22 1:5 nlE I N 19.19 290 Eh, CM 3 el ggagg f uL i EELL Mal $E Ml10E LC

-' - s/s MalTM ouTLt1 4 -- - en I .e se l:l lim x _ ,__

ExTrwt ren l ftC 1 l c - setto 22 win / Irmin N

.C _-

g _._

l g __

E --

vpe 4.41 M 11 554 vpp 5.11 ElH 31 S2a

"' I.

1 R j R m  % C- > >

A --===~ e. C w

I M k R

1. Cn i x i. o im m, Cn s m

f-m j \

%{

M n-

,a L-l.,.

I _. .,,,

5n ,,, 2 ,i M n ,= -an.>~'=

,,, . E n

} R 5 t ... . ,>

> W P + ~

A < < - -

t

> 2* >

Figure 3-5. SGA R37C25 TSP 4, Bobbin Data at EOC-12 (top) and EOC-11 (bottom) 3-19

I ee MI. i V Os 4 y . it 1:5 mill i .. 6.12 ees EPs 0, i 31 ggagg y .,

.mmot u msini us MsintM ontfLIT Enttdf M l TEC

~ W N_ spt1D 22.2Si nn/see TSAIN g .-

4C --

g ..-

i Upp 3.42 M .54 . vpp 3 . 3. M i64 la

~ ~

} N f $

x - -

,% _ < > _]

N - _- Y -.

l h_

io2 5

144 ths 015

-=2" se L 22.79 2ee Eha De 3 if 50 m I--

m - - p ,

q_ -

. _ ~

- x, m .pp o.

. , , e. .. . . . . . i.i l

i. _L- } 3 1 3 ... , , . . . . , .

ru -

d ==~

P A

- == ~

P

-s 4-t '\ f' c' F c' e an i . 0. . . ...  :. ..i ..u .a. o. O, i u , , , , , , , ,

rac ~

,m g,,, g ,,,,

ac ., ses misiiin amri

! a I .4 as I:

g

= --

e +* MT m u.

.pm r ie u=w w.ltt lin imi.

u

.c _.-

K E i wpp 2.0 M ile . v,, 3 es FJ iu sa x J -===r - J 5 3 _ -== - N _J

"' T C __ f <

n.n I h 5 @

-, 2. = 0 3 in i. . i. O. . .

W h q- >

= .

I 1

-o, -

1 i.

= ...e...-.-

( ,,, 3.n e io ,,, , m in e.

@ P l 5 1 5 -... ...!";.,.

I w  ; a -s,.  ::: :  :

--==~ < -===~  % . .

r 2 r  ? " ' '

Figure 3-6. SGB R38C33 TSP 2, Bobbin Data at EOC-12 (top) and EOC-11 (bottom) 3-20

.s .i.i, o. . . .u i.. mi o ..u . i.

r$ A E atnminMsia ses E E ati m airui

.. i .e e. o ,s.

K -, -

tritmTI itn I rIC E g _ _

. 13,tro tw.271 tv.= taaim

.c __ __

w ___ _

.,, . .s em s. .,, e ., um .> u.

'C - -

k m d Y d

= r -3 i N $

, .. _ a i. ,s ,. e n.u i. = o. . ,

1

= .-

m .u 1N

,,,, . wtaru.n to .,, s . ia EE .: 11 .,, t .51 mm sn Tsu w w rs m; .;ir im f

L M d M ~d I

'l!-

~~~

-P % c

. , , , , ... ... . . , ...i ,, .... o. . . . . , _ .

rx M  % u n g m usu R E ssw E gg _, _ ses MaitM ausut i'C -' ~

in i .e se H inn tritxt stm I isc i J g , _

s,tra it .se! ia.n iruin

.c __ _

g -- . 1 g __ ,

.,, 3. i um 3 na .,, a . i. m se een i T ( \

'C ~~

k d N d

=

[ ~ 2 E 2

/ ( l 7 i i. . ). ren ou o. 3 x it.u see ou en 5 7s q

m __ .

I a . w M

1 ==., en.

to J

,y .,, . ,, s . ,,, i .i um =

i

.= -

, ". >. .=;-

i  ; ( > t ... , . , .

L d 2 M lll ;

4 4 f 4 "; ;  ;

i i i t Figure 3-7. S/G-A R17C45 TSP 1, Bobbm Data at EOC-12 (top) and EOC-11 (bottom) 3-21

4.0 BOBBIN VOLTAGE INDICATIONS LEFT IN SERVICE 4.1 BOC-12 indications Left in Senice The indications left in service at BOC-12 are used to project the EOC-12 voltage distribution for comparison with the actual distributions found at the EOC-12 inspection, as described in Section 3.2.

The selection of the more limiting S/G with regard to the number and size of indications left in i service for potential leakage considerations in Cycles 12 and 13 is discussed in section 8.0. However, l this section provides the BOC information for the two steam generators ofinterest in the selection, l S/G-C and S/G-A. Three representations of the BOC-12 distributions are utilized in this report; two  !

are presented in this section and an alternate methodology is presented in Section 10.

The J. M. Farley-1 NRC SER (Ref.1) requires that the indications found in the inspection be divided by a POD of 0.6 and then reduced by the indications repaired to define the BOC distribution. Bobbin indications found to be NDD by RPC inspection are conservatively included as indications left in service. The NRC-required distribution with a POD of 0.6 applied is shown in Table 4-1 and Figure 4-1 for S/G-C and Table 4-2 and Figure 4-2 for S/G-A. A second distribution (in the "All Bobbin BOC-12" columns of the tables) is also included in each of these tables and figures, which is the same as the NRC-required distribution except that there is no adjustment for POD. That is, the second distribution includes all detected bobbin indications not repaired including RPC NDD indications The largest RPC NDD indication left in senice in SG-C at BOC-12 is 2.9 volts (this indication was RPC NDD in both '92 and subsequently in '94 in the 2.6 volt bin), however, the POD adjustment results in 1.7 indications at 2.9 volts left in service. In S/G-A, the largest RPC NDD indication left in service was a single 2.5 volt indication, however, the POD adjustment results in 3.7 indications left in service at 2.5 volts, even though 3 of 4 indications at this voltage level were repaired. Both the NRC SER and the "All Bobbin BOC-12" distnbutions include RPC NDD indications left in service.

Since the POD adjustment is applied before the distribution is reduced for repaired tubes, the POD l adjusted distribution also includes a greater number ofindications than is obtained by dividing all bobbin indications left in service by 0.6. For example, the adjusted distribution for SG-A includes 1.3 indications at 3.3 volt, which is obtained by dividing the two indications by 0.6 to obtain 3.3 indications and reducing this by 2.0, since both of the indications were repaired. The effect of this conservative methodology is equivalent .o assuming that about one and one-third (1.3) of a 3.3 volt l

indication was missed, even though the largest indications were found and repaired.  ;

l 4.2 BOC-13 Indications Left in Senice '

Bobbin voltage distributions for BOC-13 were developed in the same manner and for the comparable two distributions as desenbed above for BOC-12. As mentioned above, the selection of the more limiting S/G with regard to the number and size ofindications left in service for potential leakage considerations in Cycles 12 and 13 is discussed in section 8.0, and this section provides the BOC-13 l information for S/G-C and S/G-A.

The BOC-13 bobbin voltage distributions are given in Table 4-1 and Figure 4-3 for S/G-C and Table 4-2 and Figure 4-4 for S/G-A. It is seen that the POD = 0.6 adjusted distribution includes additional tubes in service that are not present in the unadjusted data. The largest bobbin indication left in l service at BOC-13 for S/G-C was in the 2.6 volt bin (between 2.51 and 2.6 volts), and for S/G-A in the 2.2 volt bin. As was the case for BOC-12, the magnitude of the largest BOC-13 indication in 41 l

i

) S/G-C left in service (2.6 volts) is the same with and without the POD adjustment since this indication

was not repaired. In S/G-A, the POD adjustment results in 0.7 indications at 4.8 volts, although this indication was repaired at EOC-12. The distribution of Figure 4-3 is applied in Section 7 to obtain i the projected EOC-13 voltage distributions for leakage and burst probability analyses as described in

, Sections 8 and 9.

i, Figures 4-5 provides the BOC-12 distributions of RPC confirmed plus not RPC inspected indications j for both S/G-C and S/G A. Except for the inclusion of new indications, this provides the distribution

+

nearest to the actual distribution ofindications returned to service with regard to leak rate and burst

potential, since the NDD indications greater than 1.0 volt in magnitude which are returned to service are expected to have negligible effect on leak and burst over the next cycle. Figure 4-6 provides a 4

similar plot for the BOC-13 distributions, however. confinned indications up to 2.0 volts are returned i to service in this case.

3 Figures 4-7 and 4-8 provide distributions of repaired indications at EOC-11 and EOC-12 for S/G-A and S/G-C, respectively. The significant reduction in the number of repaired indications at EOC-12 compared to EOC-13 (2 vs.103 in S/G-C; 3 vs. 49 in S/G-A)is due to both the increase in the IPC limit from 1.0 to 2.0 velt and the sig nficanily reduced growth rates.

)

4-2

Table 4-1: Distribution ofIndications for Farley-1 SG-C at BOC-12 and BOC-13 BOC-12

• BOC-13

• RPC RPC '

Confumed POD = 0.6 Confumed POD = 0.6 l Plus Not BOC-12 Plus Not BOC-13 All Repaired All Bobbin RPC (as per NRC All Repaired All Bobbin RPC (as per NRC l Volts Indications Indications BOC-12 Inspected SER) Indications Indications BOC-13 Inspected SER) l 0.1 0 0 0 0 0.0 0 0 0 0 0.0 1

0.2 0 0 0 0 0.0 0 0 0 0 0.0 i 0.3 0 0 0 0 0.0 0 0 0 0 0.0 I 0.4 0 0 0 0 0.0 5 0 5 5 8.3 0.5 4 0 4 4 6.7 16 0 16 16 26.7 0.6 26 1 25 25 42.3 47 0 47 47 78.3 0.7 43 1 42 42 70.7 64 0' 64 64 106.7 I 0.0 62 2 60 58 101.3 81 1 80 80 134.0 l 0.9 55 4 51 51 87.7 84 0 84 84 140.0 l

l .0 78 7 71 66 123.0 83 0 83 83 138.3 1.1 39 10 29 0 55.0 81 0 81 71 135.0 1.2 55 10 45 s 81.7 52 0 52 7 86.7 1.3 40 10 30 0 56.7 51 0 51 11 - 85.0 1.4 45 10 35 0 65.0 29 0 29 2 48.3 1.5 30 8 22 0 42.0 38 0 38 5 63.3 l 1.6 32 8 24 0 45.3 19 0 19 8 31.7 1.7 20 8 12 0 25.3 15 0 15 4 25.0 1.8 9 2 7 0 13.0 7 0 7 1 11.7 1.9 11 5 6 0 13.3 8 1 7 2 12.3 2.0 8 2 6 0 11.3 4 0 4 0 6.7 )

2.1 14 8 6 0 15.3 2 0 2 0 3.3 2.2 8 4 4 0 9.3 2 0 2 0 3.3 2.3 2 1 1 0 2.3 2 0 2 0 3.3 2.4 2 1 1 0 73 0 0 0 0 0.0 2.5 0 0 0 0 00 2 0 2 0 3.3 2.6 0 0 0 0 0.0 1 0 1 0 1.7 2.7 1 1 0 0 0.7 0 0 0 0 0.0 2.0 0 0 0 0 00 0 0 0 0 0.0 2.9 1 0 1 0 1.7 0 0 0 0 0.0 ,

3.0 0 0 0 0 0.0 0 0 0 0 0.0 Totals $85 103 482 246 872.0 693 2 691 490 1153.0

' 92 Reanalysis voltages of P1, PIN, and PICS reported in 1992.

8 '94 voltages of P1, PIN, and PCN reported in '94.

I 1

i ALASUM.XLS SGC-BOC,EOC T4.1 4-3 R.1 930%

l

l l

l 1

Table 4-2: Distribution ofIndications for Farley-1 SG-A at BOC-12 and BOC-13 ,

BOC-12 '" BOC-13 (23 RPC RPC ,

Confirmed POD = 0.6 Confirmed POD = 0 6 l Plus No* ' BOC-12 Plus Not BOC-13 All Repaired All Bobbin RPC (as per NRC All Repaired All Bobbin RPC (as per NRC Volts Indications indications BOC-12 Inspected SER) Indications Indications BOC-13 Inspected SER) 0.1 0 0 0 0 0.0 0 0 0 0 0.0 0.2 0 0 0 0 0.0 0 0 0 0 0.0 03 0 0 0 0 0,0 0 0 0 0 00 04 3 0 3 3 5.0 8 0 8- 8 13 3 0.5 12 0 12 12 20 0 28 0 28 28 46.7 l 06 30 1 29 29 49.0 40 0 40 40 66.7 0.7 32 0 32 32 533 61 0 61 61 101.7 0.8 34 0 34 34 56.7 73 0 73 72 121.7 0.9 49 0 49 49 81.7 62 0 62 62 103 3 1.0 49 4 45 44 77.7 57 0 57 57 95.0 1.1 19 6 13 0 25.7 41 0 41 41 68 3 1.2 24 5 19 0 35.0 32 0 32 32 53 3 13 22 4 18 0 32.7 30 0 30 27 50.0 1.4 26 12 18 0 313 22 0 22 19 36.7 1.5 7 2 5 0 97 13 0 13 10 21.7 ,

1.6 11 2 9 0 163 9 0 9 2 15.0 1.7 8  ! 3 5 0 103 11 0 11 2 183 1.8 4 0 4 0 6.7 8 0 8 3 13 3 1.9 2 0 2 0 33 6 1 5 2 9.0 2.0 0 0 0 0 00 4 0 4 2 6.7 ,

I 2.1 1 1 0 0 0.7 0 0 0 0 00

! 2.2 1 1 0 0 0.7 2 0 2 0 33 23 1 1 0 0 0.7 0 0 0 0 0.0 2.4 0 0 0 0 0.0 0 0 0 0 0.0 2.5 4 3 1 0 3.7 I I O O 0.7 2.6 1 1 0 0 0.7 0 0 0 0 0.0 l

2.7 0 0 0 0 0.0 0 0 0 0 0.0 l 2.8 0 0 0 0 00 0 0 0 0 00 2.9 0 0 0 0 0.0 0 0 0 0 0.0 3.0 I i 0 0 07 0 0 0 0 0.0 33 2 2 0 0 13 0 0 0 0 0.0 4.8 0 0 0 0 0.0 1 1 0 0 0.7 ,

1 Totals 343 49 294 2M $22.7 509 3 506 468 8453 l 1

l

' 92 Reanalysis voltages of P!, PIN, and PICS reported in 1992. )

1

' '94 voltages of P!, PIN, and PCN reported in 94. I

! ALASUM.XLS SGA-BOC EOC T4.2 4-4 )

t I

l l

t i

Figure 4-1. BOC-12 Distribution of SG-C Indications j NRC SER Distribution with POD = 0.6 vs. All Indications Returned to Service l

140 l

l l 120 - - - - - - - - - - - - - - - - - . - - - - - - - - - - - -

l 100 --- -- - --- - - - - - - - - - - - - - - - . - - - --

E

=

a 1 y 80 - -- - - - . . -- - -- -. -- - - - .

t 60 - - - - - - - - - _

.=

E o

l E

1 40 --- - ---

)

l l

20 - - - - - - -- _ _ _ _ . _ _

! ! ! mm 0 I - -

)

0.1 0.3 0.5 07 09 11 1.3 1.5 1.7 1.9 2.1 23 2.5 2.7 2.9 Voltage a All Bobbin Indications ONRUSER Distribution ALASUM.XLS Fig 4-1 4-5

l

\

l l

Figure 4-2. BOC-12 Distribution of SG-A Indications NRC SER Distribution with POD = 0.6 vs. AllIndications Returned to Service 90 l

80 - --

70 - - - - - - ---- - - - . - - - -- - - -

m g 60 - - -- - - - - - - - - --

=

8

.5 50 - --

b

! [ 40 - - --

= 30 -~ - -

Z 20 - -- . . _ - - - -

l

! 10 0  ! E --- - n 01 0 2 0.3 0 4 0 5 0 6 0 7 O B 0.9 1.0 1.1 1.2 13 1.4 1.5 1.6 1.7 1.8 1.9 2 0 2.1 22 2.3 2 4 2.5 2.6 2.7 2 8 2.9 3 0 3.3 4 8 l Voltage

,a AllIndications ONRC SER Distribution -

l l

l l

ALASUM.XLS Fig 4-2 4-6 I

d 4

i i

s i

4 4

f Figure 4-3. BOC-13 Distribution of SG-C Indications i NRC SER Distribution with POD = 0.6 vs. All Indications Returned to Service 4

160 4

140 - - - ~ - - - - - - - - - -

{

i 4

120 - - - - - - - - - - - - - -- - - - _ - .

l E o

j 100 - - - - - - - - - - - . - _ ._

.2 b E j Co 80 - ~ _. -. --- -_

a

.' D I 9 60 --- - - - - - - - --. -_ - __- ---

1 5

z j 40 - - - - - . - - - . . . - - - . _ . _ .

)

20 - - .-- - - - -

a 0 I ! Eema e-i 0.1 0.3 0.5 0.7 0.9 1.1 1.3 1.5 1.7 1.9 2.1 2.3 2.5 2.7 2.9 i ,

4 Voltage i

. _ - - - - . . . ~ _ . _ . _ . _ _ _ - - - . . . .

]  ; e All Bobbin Indications O NRC SER Distribution ;

i e

i Y

4

' ALASUM.XLS Fig 4-3 4-7

Figure 4-4. BOC-13 Distribution of SG-A Indications NRC SER Distribution with POD = 0.6 vs. All Indications Returned to Service 140 120 100 E

'il j

80 - - .-

t 60 -- -

.c E

=

Z 40 -- -

20 0 l I e - -

0.10.20.30.40.50.60.70.80.91.01.11.21.31.41.51.61.7I.81.92.02.12.22.32.42.52.62.72.82.93.03.34.8 Voltage lM AllIndications ONRC SER Distribution l

)

i ALASUM.XLS Fig 4-4 4-8

l l

l 6

4 Figure 4-5. BOC-12 SG-C and SG-A Indications RPC Confirmed Plus Not RPC Inspected 70 60 I

50 - -

l 8

'ii y 40 -- -

l j 30 - -

1 5 i 5

l E

1 20 -

l 10 - - -

0 0.1 0.3 0.5 0.7 0.9 1.1 1.3 1.5 1.7 1.9 2.1 2.3 2.5 2.7 2.9 Voltage E S/G-C O S/G A ALASUM.XIS Fig 4 5 4-9 g 3 973ny

a i

d l

i l t

i l

i i

i  ;

! l

Figure 4-6. BOC-13 SG-C and SG-A Indications RPC Confirmed Plus Not RPC Inspected 4

! 90 80 -- ._

i =

1 70 __

i

,j 60 =_ _ __

e

! E 50 - _ _ _

1 3

w

, [. 40 .- - . .

=

4 S 4 E ,0 i  : 3 -

2; -

20 _ . __ __

10 . - - _

O ( . l Lu_m 0.10.20.30.40.50.60.70.80.91.01.1 1.2 1.3 1.4 1.5 1.6 1.7 1.8 1.9 2.0 2.1 j

Voltage a S/G-C D S/G-A JI l i

1 ALASUM.XLS Fig 4-6 4-10 I

I i

l l

i i

i 1

l Figure 4-7. EOC-11 and EOC-12 SG-C Repaired Indications l

11 l

l 10 - - - - - - - - - - - . - - - - - - - - - - - - -

l 9 _.- - . _ . . _ -____ - .._ . - . . -

8 --- - - - - - . - - - - -

\-

E

.o 7 - _. . ~ _ . . .

E

.I y 6 - - - - - - -

Co 5 - - -- - - - - - - - - - -

1 E 4 - - - - - - - - - - - - - - - - . - - - ---- - -

o E

3 --- -- - - -- - -..

2 -

1 0

0.1 0.3 0.5 0.7 0.9 1.1 1.3 1.5 1.7 1.9 2.1 2.3 2.5 2.7 Voltage

OEdCdA NEOC-12, ALASUM.XLS Fig 4-7 4-1I

8 l l

1 1

1 1

1 Figure 4-8. EOC-11 and EOC-12 SG-A Repaired Indications 14 12 -- -- - - - - - - - . -

10 - - - - - - - - -..- - - - - . - . - - - - - - - -

E

.o E

.M 8 - - - - - . - . - - - - - .-- -- -

2 w

e b 6 -

.=

E 3

Z 4 _. ._..

2 - - - . . . . - _ _ _

0 1 0.607080.9101.11.21.3141.51.61.71.81.92.0212.22.32.42.52.62.7282.9303.348 l Voltage l

\

O EOC-11 m EOC )

l I

l l

1 i

ALASUM.XLS Fig 4-8 4-12 ,

1 l

1

5.0 VOLTAGE GROWTII RATES 5.1 Cycle 11 Voltage Growth Rates Voltage growth rates for Cycle 11 were developed at EOC-Il by a single analyst. The evaluation included calibration corrections for normalizing ASME standard voltages to the reference laboratory standard used in the APC database development.

These data were used to develop growth distributions. No cycle length adjustments were made to the Cycle 11 growth rates; the Cycle 11 operating period was 471 EFPD. Figure 5-1 shows the Cycle i1 voltage growth distribution. The largest growth value was 1.90 volts and found for only one indication. Thirteen indications had voltage growth values >l.0 volt.

Table 5-1 summarizes the average growth rates for Farley-1 Cycles 11 and 12. The Cycle 11 (1991 to 1992) average growth was 0.22 volt or 26% of the BOC-11 average of 0.85 volt. The average growth for BOC indications 5 0.75 volt was 37% while the average growth for indications >0.75 volts was 21%.

5.2 Cycle 12 Voltage Growth Rates The Cycle 12 voltage growth rates were developed utilizing the 1994 field analysis data and a single analyst to reevaluate the 1992 data for the same indications. The use of a single analyst provides consistency to the analysis and enhances the accuracy of the growth data. Reference 3 Appendix A guidelines were used in the 1994 inspection including cross calibration of the field ASME standards to the reference laboratory standard, Growth rates were developed for all potential flaws (PI, PIN, PCN classifications). INR signals, which are not considered to be flaw indications, were also analyzed but not included in the growth rate analysis.

The Cycle 12 growth distribution is shown in Figure 5-2. The largest growth value is 0.93 volts which is lower than the value of 1.90 volts found for Cycle 11. Cycle 12 shows only 4 growth values

>0.70 volt, compared to 32 for Cycle 11. The median bin for Cycles 12 growth is 0.00 volts while the median bin for Cycle 11 is 0.20 volts. The Cycle 12 growth distribution of Figure 5 2 is used to project BOC-13 indications left in service to EOC-13 conditions, as described in Section 7. The Cycle 12 length was about 442 EFPD, although no cycle length adjustments are made to the growth values shown in this section.

Table 5-2 shows the bobbin and RPC results for indications with a growth rate >0.50 volts. Also shown is whether the indication was a new indication reported in the 1994 inspection. Three of the

! 17 indications with the largest growth rates were confirmed by RPC, and six of the 17 indications were RPC NDD; the remaining were not RPC inspected. Ten of the 17 indications were new indications in the 1994 inspection.

Table 5-1 summarizes the average voltage growth rates for Cycle 12 (1992 to 1994). The average growth rate is -0.01 volts or ~0% for Cycle 12 which is very small compared to the 26% found for Cycle 11. The average BOC voltage for Cycle 12 was 0.98 volts, compared to 0.85 volts for Cycle l 11.

5-1 l

l

f Figure 5-3 provides a comparison of the growth rates for Cycles 11 and 12. The 50% cumulative

probability value for Cycle 11 is about 0.19 volts, while the 50% cumulative probability value for Cycle 12 is slightly negative, at about -0.01 volts. The Cycle 12 distribution displays a smaller range of values, from -0.8 volts to 1.00 volts, and the Cycle 11 growths range from -0.4 volts to 1.90 volts.

The Cycle 12 duration was about 94% (442 vs. 471 EFPD) of the Cycle 11 duration. The decreasing growth rate trends for the Farley S/Gs strongly indicate the increasing effectiveness of plant operations in implementing enhanced chemistry specifications, boric acid addition, molar ratio control, enhanced chemistry monitoring, and cleaning efforts (PPC, tubesheet sludge lancing) to minimize the growth of ODSCC at TSP intersections.  !

I l

i 5-2

l

)

Table 5-1. Average Voltage Growth Per Cycle for Farley-1 Number Average BOC Average AV Percent Cycle Indications Voltage Growth / Cycle Growth Prior Cycles:

1985 to 1986 123 0.45 0.20 45 %

1986 to 1988 274 0.48 0.28 $9%

1988 to 1989 431 0.62 0.22 36%

1989 to 1991 All Indications 499 0.70 0.23 33%

V,x < 0.75V 306 0.51 0.24 48%

V,x 2 0.75V 193 1.01 0.08 8%

1991 to 1992 All Indications 1267") 0.85 0.22 26%

V,x < 0.75V $46 0.57 0.21 37% l Vax > 0.75V 721 1.08 0.23 21% 1 Cycle 12: '92 '94 SG-A All Indications 507(2) 0.94 0.01 - 1%

V,x < 0.75V 168 0.60 0.05 Vnx 2 0.75V 339 1.12 -0.01 SG-B All Indications 481 0.93 0.03 3%

' vow < 0.75V 144 0.58 0.06 V 337 1.08 0.01  !

- ax 2 0.75V l SG-C All Indications 693 1.05 -0.04 -0%

Voc<

a 0.75V 154 0.63 0.02 V,x 2 0.75V $39 1.18 -0.06 All SGs All Indications 1681(2) 0.98 -0.01 -0%

Vox < 0.75V 466 0.60 0.04 7%

V,x 2 0.75V 1215 1.13 -0.03 ~0%

") 1269 total indications reported; growth evaluations were performed for 1267 indications.

<23 1683 total indications reported; growth evaluations were performed for indicated numbers.

5-3

Table 5-2. Summary of Bobbin Voltage Growth Rates >0.50 Volts for Farley-1 Cycle 12 (Of 1681 Growth Measurements)

EOCl2 BOCl2 Voltage Bobbin RPC Bobbin Growth New SG Row Col TSP Volts Volts Volts (Volts) Indication?

A 17 77 4H 1.62 NDD 0.69 0.93 Yes B 41 32 4H 1.77 NDD 0.85 0.92 Yes B 2 77 IH 1.75 0.42 0.98 0.77 No B 40 24 4H 1.95 0.37 1.23 0.72 Yes B 2 85 1H 1.09 N.I. 0.46 0.63 Yes A 43 39 3H 1.27 N.I. 0.66 0.61 Yes A 36 70 6H 1.98 2.03 1.38 0.60 Yes B 6 49 3H 1.25 N.I. 0.65 0.60 Yes A 43 33 2H 1.13 N.I. 0.57 0.56 Yes A 43 39 4H 1.45 N.I. 0.89 0.56 Yes B 3 56 4H 1.93 NDD 1.37 0.56 No A 8 2 4H 1.53 NDD 0.98 0.55 No B 23 24 4H 1.03 N.I. 0.48 0.55 No C 2 16 3H 1.94 NDD 1.40 0.54 No B 13 7 5H 1.56 NDD 1.04 0.52 No B 26 68 3H 1.16 N.I. 0.64 0.52 Yes B 24 74 2H 1.40 N.I. 0.88 0.52 No Notes:

m N.I. = Not RPC Inspected.

l ALA_ALL.XLS T5-2 5-4 l

Table 5-3 Farley-1 Cycle 11 Growth Distribution - All SGs Voltage No. Cunt % Cumulative

-0.40 1 0.1% 1  !

-0.30 2 0.2% 3

-0.20 2 0.4% 5

-0.10 18 1.8% 23 0.00 98 9.6% 121 0.10 256 29.8 % 377 l

0.20 275 51.5 % 652 ,

0.30 263 72.2 % 915 0.40 177 86.2 % 1092 0.50 83 92.7 % 1175 0.60 31 95.2 % 1206

, 1 O.70 29 97.5 % 1235 l 0.80 11 98.3 % 1246 l l

0.90 8 99.0 % 1254 l 1.00 5 99.4 % 1259 1.10 2 99.5 % 1261 1.20 1 99.6 % 1262 1.30 1 99.7 % 1263

, 1.40 2 99.8 % 1265 l l

1.50 0 99.8 % 1265 1.60 0 99.8 % 1265 1.70 1 99.9 % 1266 1.80 0 99.9 % 1266 1.90 1 100.0 % 1267 Note: Negative growths are not included in applications for l projecting EOC distributions.

5-5 l

l l .

l i

I Table 5-4 Farley-1 Cycle 12 Growth Distribution - All SGs Voltage No. Cum.% Cumulative

-0.80 1 0.1% 1

-0.70 1 0.1% 2 l -0.60 6 0.5% 8 l

! -0.50 6 0.8% 14 l -0.40 23 2.2% 37

-0.30 47 5.0% 84 l -0.20 93 10.5 % 177

-0.10 237 24.6 % 414 0.00 501 54.4 % 915 0.10 432 80.1 % 1347 0.20 191 91.5 % 1538 l

0.30 80  %.3% 1618 0.40 28 97.9 % 1646 0.50 18 99.0 % 1664 0.60 11 99.6 % 1675 0.70 2 99.8 % 1677 0.80 2 99.9 % 1679 l

l 0.90 0 99.9 % 1679 1.00 2 100.0 % 1681 l

l Note: Negative growths are not included in applications for l projecting EOC distributions.

l l

l l

l

I

{

4 i

j Figure 5-1

] Farley-1 Cycle 11 Growth Rates (All SGs) j (1267 Growth Values) l 300 ...: : : : : : 100.0 %

i 275 256 90.0 %

4 250

{

~

80.0 %

i j -

70.0 %

i 200 -

., /177 l!!

@ /

60.0 % 8 l

e i i- 150 - 50.0 % .i:

J -

! Y -

! j f - 40.0% !

98 U 100 -

83 i -

30.0 %

1 i

i 20.0 %

i 50 -

1 31 29 4

18 -

10.0 %

'8 122

" E" 5 2 1 12oo1o1 i

0 .:. - - '-

0.0%

j -0.4 -0.2 0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 f Volts (V)

M No. --*- Cum. %

i i

j l

4

(

i 1,

i t

4 I

! ALADATA.XLS

• ALA Cyclel2 Chart 18 5-7

i a

i

'I l

j Figure 5-2 l Farley-1 Cycle 12 Growth Rates (All SGs) l (1681 Growth Values Excluding INRs) 4 l

h l 600  :  : : : 100.0 %

i i

90.0 %

{ 501 i 500 -

l -- 80.0%

l A2 l -

70.0 %

j 400 - E 1 ,,, 0 l 8 -- 60.0%")

i

=-

=aa o

.I .N "i 300 --

50.0% ,E_

C E

, 237  :

g -

40.0% V 200 -

30.0 %

100 --

0 1

-0.8 166

-0.6 23

-0.4 47

-0.2 0.0 I 80 0.2 Volts (V) 28 0.4 gg E

11

"=2 2 0 2 0.6 0.8 1.0

-- 10.0%

0.0%

M No. -+- Cum. %

ALADATA.XLS

• ALA Cycicl2 Chan 1 5-8

i i

l I

Figure 5-3 Comparison of Farley-1 Cycle 11 and Cycle 12 Growth Rates i

100.0 % -

90.0% - - - - - - - - - - - . - - , - - - . - -

(

• 80.0 % ,--- '--- .-- -

--.---- L -v - - - - -

l I

70.0 % - ---------- - - -- a + + - .--- a . - -

j 1 i

p, 60.0 % - . - - - - - - - ~ l, -- - - --. . . - - , - - - . --. -

o u .

?

.$ 50.0 % . - - - - . . --

! - . . - L.--- ,

-d .

3 ll

, E 1 i i j j 40.0 % - - - . -- - -.

-- L m .- 7 --

l ,

, e i .

l .

! 30.0 % - - - - - -

-.-l - - - --. . .

7 20.0 % -

-[E .. -

L 1

10.0 % - - " . r--/ . .- -.~u L.

l . I  !

I 0.0%

-0.8 -0.6 -0.4 -0.2 0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 i

Voltage Growth

- - - - - . Cycle 11 Cycle 12 l

l l l l l  !

l l

l l

l ALADATA.XLS .ALA Cycle 12 5-9

6.0 NDE UNCERTAINTIES NDE uncertainties for voltage measurements were developed for the Farley units in WCAP-12871 Rev. 2 (Ref. 3). The NDE uncertainty developed in this WCAP has been updated in Reference 6, EPRI Report TR-100407, Revision 1 (Draft of August 1993). While the WCAP and EPRI reports i differ slightly in the components of the net NDE uncertainty and in the development, the net NDE uncertainty is not significantly different (12% standard deviation in the WCAP versus 12.5% in the EPRI report) and the EPRI NDE uncertainty is applied for both Cycles 12 and 13 for Farley-l.

The NDE uncenainty is principally due to probe wear with a standard deviation of approximately 7%

about a mean of zero and analyst variability with a standard deviation of 10.3%. These distributions are applied as normal distributions and combined to obtain a net NDE uncertainty of 12.5% for one standard deviation. The upper bound on the probe wear uncertainty is limited to 15% by the Farley IPC requirement to replace the bobbin probes when measurements on the probe wear standard for a worn probe differ from that found for the new probe by 15%. The upper bound on the analyst l variability uncertainty would be limited to 20% by the eddy current analysis guidelines which require lead analyst resolution of bobbin voltages (with one or more reported above 1.0 volt) differing between analysts by more than 20%.. However, as of the February 8,1994, NRC/ Industry meeting the NRC has not accepted the 20% cutoff on the analyst variability uncenainty. Pending a further ,

resolution of this issue with the NRC, the analyst uncertainty is applied without a cutoff for the EOC  !

voltage projections performed in this report. For EOC voltage projections, separate distributions of l 7% with a cutoff at 15% for probe wear and 10.3% with no cutofT for analyst variability are applied in this report.

1 l

l 6-1

7.0 PROJECTED EOC VOLTAGE DISTRIBUTIONS 7.1 Projected EOC-12 Voltage Distributions Consistent with the Reference 3 WCAP methodology, Monte Carlo analyses are applied to develop projected EOC distributions from the BOC distributions. The BOC voltages are increased by l allowances for NDE uncertainties (from Section 6) and voltage growth (from Section 5) to obtain the '

EOC values. In the Monte Carlo analyses, each voltage bin of the BOC voltage distributions (Figure 4-1 for example)is increased by a random s:unple of the NDE uncertainty and growth distributions to obtain an EOC voltage sample. Each sample is weighted by the number of indications in the bin.

The sampling process is repeated for each BOC voltage bin and then repeated for a large number of samples across the BOC distribution. In the present analyses,10,000 samples for each bin of the BOC ,

distribution were applied. Monte Carlo analyser for tube burst probabilities used 100,000 samples of I each bin. The EOC projections were performed for SG-C, and are compared to both SG-C and SG-A actual distributions.

The projected EOC-12 bobbin voltage distributions are shown in Figure 7-1 (for SG-C), Figure 7-3 I (for SG-A), and Table 7-1 for the two categories of BOC distributions discussed in Section 4.1 and I shown in Figure 4-1. Table 7-1 also includes results for the alternate method of defining the BOC distribution that is discussed in Section 10. For the POD = 0.6 adjusted distribution, the maximum projected EOC-12 voltage is 3.8 volts. Since the Monte Carlo analyses yield a cumulative probability distribution of EOC voltages, a method must be defined to obtain a discrete maximum EOC voltage j value. The method adopted in this report is to integrate the tail of the Monte Carlo distribution over the largest 1/3 of an indication to define a discrete value with an occurrence of 0.33 indications. For N indications in the distribution, this is equivalent to evaluating the cumulative probability distribution of voltages at a probability of(N-0.33)/N. The largest voltages for all distributions developed by Monte Carlo in this report have been obtained with this definition for the maximum EOC discrete voltage. The largest EOC-12 voltage for the distribution of all indications without POD adjustment (POD = 1.0) is 3.6 volts. In the 1992 SLB leak rate evaluation for EOC-12 (Ref.10) using only bobbin indications < l.0 volt left in service, a maximum EOC-12 voltage of 2.1 volts was reported for all indications without POD adjustment. The difference between 2.1 volts and the 3.6 volts of this report is due to the current analyser including RPC NDD indications in the analysis and the method (1/3 of an indication in this report versus one indication in the prior analysis) used to define the reported maximum voltage. i 7.2 Comparison of Projected and Actual EOC-12 Distributions In comparing projected and actual distributions,it is necessary to consider the purpose of the l projections in order to define the appropriate actual distributions. The Farley-1 EOC inspection results  !

include many RPC NDD indications. The projected EOC voltage distributions for IPC applications are applied to project SLB leakage and tube burst probability. As discussed in Section 3.5 in

, developing the EOC-12 distribution for comparisons with projections, RPC NDD indications have a negligible likelihood of potential SLB leakage over the prior cycle and should be ignored in comparing IPC projections with the actual distributions. Thus the comparisons ofIPC projections with actual distributions are made for EOC-12 RPC confirmed indications summed with indications not

! RPC inspected since it cannot be stated with confidence that the latter indications would not have leaked over the prior cycle. The appropriate EOC-12 actual distribution for comparison is the RPC l confirmed and untested (not RPC inspected) indication given in Figure 7-la for SG-C, Figure 7-3a for i

7-1 l

l

l SG-A, and listed in Table 7-1. There are 492 and 471 indications in the actual EOC distributions for SG-C and SG-A, respectively. For information only, projected voltage distributions with POD = 1.0 are also compared with the distribution of all 693 and 509 indications for SG-C and SG-A, respectively, at EOC-12 including RPC NDD indicstions in Figure 7-lb and 7-3b.

Figure 7-la shows the comparison of projected and actual EOC-12 voltage distributions (excluding RPC NDDs). The upper figure shows the NRC model (Figure 4-1 for BOC-12 distributions) which includes a POD = 0.6 applied to the EOC-Il inspection results. It is seen that the NRC model leads to excessive conservatism both in the number of indications (872 versus actual 492) with leakage potential and in the largest EOC voltages. The NRC model leads to projected voltages as high as 3.8 volts and overestimates the number ofindications in all voltage bins except < l.0 volt. Figure 7-lb shows the comparison for the projection of all indications without a POD adjustment (POD = 1.0).

l This projection is also conservative above 1.1 volts although significantly less conservative than the NRC model. As expected, the projection based on Monte Carlo analyses distributes the indications l more continuously over the voltage range compared to the discrete actual indications, and represents a j very conservative number ofindications above 1.1 volts. This is because the number of RPC NDD indications included in the BOC distribution exceeds the number of new RPC confirmed indications.

An alternate approach to defining BOC indications including considerations for undetected indications is described in Section 10. Similar trends to those discussed above are seen in Figun s 7-3a and 7-3b that compare the EOC-12 projections for S/G C with the actuals for S/G A. In this case, the projections are conservative with the exception of not predicting the maximum 4.8 volt indication found in S/G A. In Section 8,it is shown that the SLB leakage calculated from the actual RPC confirmed distributions for S/G A, as shown in Figure 7-3 is slightly higher than calculated for the actual S/G C distribution of Figure 7-1. This results principally from the two RPC confirmed indications in S/G A above 2.0 volts. However,it is shown also in Section 8 that the leak rate l calculated from all S/G C projections is higher than that obtained from the actual S/G A distribution, l which demonstrates the conservatisms in the projections.

l i

A comparison of projected distributions with the EOC-12 distribution for all indications independent of RPC confirmation is shown in Figure 7-2 for SG-C and 7-4 for SG-A. As noted above, the actual distribution includes RPC NDD indications which would not be expected to leak at EOC-12. For this case, the NRC model (POD = 0.6) still significantly execeds the distribution. The results for SG-C with a POD = 1.0 show fair agreement with the actual distribution above 1.2 volt and underestimates the number of small indications less than 1.1 volt. It is shown in Section 10 and Table 7-1 that the best agreement with the actual distribution for all indications independent of RPC confirmation is

( obtained with the aPernate method of defining the BOC distributions. However, attempting to project l' the distribution for all mdications would lead to excessive conservatism if applied to SLB leakage analyses.

7.3 Projected EOC-13 Voltage Distributions The BOC-13 voltage distributions are described in Section 4.2 and Figure 4-2. Monte Carlo methods ,

are applied conservatively using the Cycle 12 voltage growth distribution of Figure 5-1 to obtain the l projected EOC-13 distributions. As above for Cycle 12, projections to EOC-13 have been made for j two distributions including the NRC model with a POD = 0.6 and all bobbin indications left in service including RPC NDD.

Figure 7-5 shows the projections with and without the POD = 0.6 adjustment. The conservatism of 7-2

,, . . - . . . - - - - - . - . . . .- . - --- . - ~ . ~.

the NRC model, as reflected in the BOC distribution of Figure 4-2,is seen in the EOC 13 distributions. The NRC model projects to an EOC maximum voltage of about 3.3 volts compared to 1 3.2 volts for all indications left in service. The reason for the close agreement between the two 1 maximum voltage projections is seen in the BOC-13 voltage distributions in Table 4-1; none of the larger indications (>2.0 volt) at EOC-12 were repaired for indications at the TSP intersections. Thus the NRC model, while having a larger number ofindications left in service, does not result in significantly higher voltage indications left in service than that for all bobbin indications left in service. The EOC-13 distributions given in Figure 7-5 are applied to SLB leakage analyses in the following section.

i As described in Sections 8 and 9, SG-C was determined to be the more limiting steam generator for i

. tube leak and burst. Because of the presence of a relatively large indication in SG-A at BOC-13, as described in Section 4, an EOC-13 projection was made for the SG-A voltage distribution utilizing the i BOC-13 distribution with a POD =0.6. This distribution is shown in Figure 7-6. The leak and burst  !

calculations with this distribution were less limiting than SG-C as shown in Sections 8 and 9. 3 l

l I

I i

{

7-3

)

i Table 7-l. J. M. Farley-1: Projected EOC-12 Distributions vs. SG-C and SG-A Actuals 2

Actual S/G-C EOC-12 Actual S/G-A EOC-12 Projected EOC-12 Distributions All RPC All RPC All ind. All Ind. Alternate Volts Ind. ConfJUntested Ind. Conf / Untested POD =0.6 POD =1.0 Method 0.4 5 5 8 8 0 0 4 0.5 16 16 28 28 1 1 13 0.6 47 47 40 40 7 4 29 O.7 64 64 61 61 21 12 49 4

0.8 81 81 73 72 39 23 68 0.9 84 84 62 62 57 34 80 l 1.0 83 83 57 57 74 43 83 l 1.1 81 71 41 41 82 47 74

, 1.2 52 7 32 32 83 48 63 f 1.3 51 11 30 27 79 44 47 1.4 29 2 22 19 73 40 34 1.5 38 5 13 10 65 36 22 1.6 19 8 9 2 57 31 17 1.7 15 4 11 2 48 26 12 1.8 7 1 8 3 41 21 8 1.9 8 3 6 3 33 17 7 2.0 4 0 4 2 26 14 4 2.1 2 0 0 0 21 10 4 2.2 2 0 2 0 16 8 3 2.3 2 0 0 0 13 6 2 2.4 0 0 0 0 9 5 1 2.5 2 0 1 1 8 3 1 2.6 1 0 0 0 5 3 1 2.7 0 0 0 0 4 2 1 2.8 0 0 0 0 3 1 0 2.9 0 0 0 0 2 1 I i 3.0 0 0 0 0 2 1 0 3.1 0 0 0 0 1 0 0.7 32 0 0 0 0 0 0 0 3.3 0 0 0 0 1 0 0.3 3.4 0 0 0 0 0 0.7 0 3.5 0 0 0 0 0 0 0 3.6 0 0 0 0 0.7 0.3 0 3.7 0 0 0 0 0 0 0 I 3.8 0 0 0 0 0.3 0 0 4.8 0 0 1 1 0 0 0 Totals 693 492 509 471 872.0 482.0 629.0 I

ALA_SIMS.XLS Table 71 7-4  !

i i

Table 7-2. J. M. Farley-1

Projected EOC-13 Distribution (Limiting SG),

All Ind. All Ind. Alternate Volts POD =0.6 POD =1.0 Method 0.4 3 2 7 i 0.5 16 10 25 l

O.6 43 25 52 0.7 77 47 83 0.8 106 63 100 0.9 120 72 113 1.0 128 76 111 1.1 125 76 96 f

l 1.2 111 66 76 1.3 95 57 '4 1.4 78 47 1.5 63 38 1.6 50 29 20 1.7 38 23 16 1.8 28 17 10 1.9 21 12 8

)

l

) 2.0 14 9 6

' ' l 2.1 11 7 ~4 22 8 4 3 2.3 5 3 2 2.4 4 2 1 2.5 2 2 1 2.6 2 1 1 2.7 2 1 0 2.8 1 1 1 2.9 1 0 0 l

l 3.0 0 0 0 l

3.1 0 0.7 0 3.2 0.7 0.3 0 3.3 0.3 0 0 3.4 0 0 0.7 i 4.5 0 0 0.3

== :=s 1153 691 857 IM ALA_SIMS.XLS Table 7-2 7-5

I I

t Figure 7-la. J. M. Farley-1: Comparison of S/G C EOC-12 Actual RPC Confirmed / Untested Voltage Distribution vs. EOC-12 Distribution Projected with POD = 6.6 i 90

! go . - _ _ _

l

70 - - - -- -

l 1

l'

,, 60 - - - - - -- --

.i j 50 - - - - - -- - -

1 y 40 - - - - - - - - - -

6 Z 30 - - - - - - - - -- --

j 20 - -- - -- - - - -- - - -

I 10 - _ - - _ _ _ - .- _ _ _ _ _ _

0 , a l . I Onnn- - _

O.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0 2.2 2.4 2.6 2.8 3.0 3.2 34 3.6 3.8 I Volts

's I

Actual EOC-12 O Projected EOC-12, POD = 0.6 I

._J Figure 7-lb. J. M. Farley-1: Comparison of S/G C EOC-12 Actual RPC Confirmed / Untested Voltage Distribution vs. EOC-12 Distribution Projected with POD = 1.0 f

1

. 90 -

1

! 80 - - -

i

70 - - -

. 60 - - - -

.E.

? S0 - - -

5 40 - - - - -.

& l z 30 - -

.h

,' 20 - - -- - - - -. --

10 - - - - -- - - - - - - -

0 =l n  ! I - I O fl il n - - - _

O4 0.6 0.8 1.0 1.2 1.4 1.6 f .8 2.0 2.2 2.4 2.6 2.8 30 3.2 3.4 3.6 3.8 Volts 1 W Actual EOC-12 O Projected EOC-12, POD = 1.0l

ALA_SIMS.XLS 7,g i

Figure 7-2a. J. M. Farley-1: Comparison of S/G C EOC-12 Voltage Distribution of All Indications vs. EOC-12 Distribution Projected with POD = 0.6 90 80 - - - -

70 - - - - -

. 60 - - - - - - - --

5

,3 50 - -

i y 40 -

d E 30 _ - _ _ _ _

20 -

10 - - - -- - - - -

0 0 ' ' ' ' 00""" " -

0.4 ' 0.6 0.8 LO 1.2 1.4 1.6 1.8 2.0 2.2 2.4 2.6 2.8 3.0 3.2 34 3.6 3.8 Volts l 5 Actual EOC-12 O Projected EOC-12, POD - 0.6 l Figure 7-2b. J. M. Farley-1: Comparison of S/G C EOC-12 Voltage Distribution of All Indications vs. EOC-12 Distribution Projected with POD = 1.0 90 80 - l 70 - -

, 60 - - -

3 50 - - - - - -

E a

y 40 - - -

d Z 30 - -

20 - -

10 - - -

0 , 1 , u l a a 4An--- _

0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0 2.2 2.4 2.6 2.5 3.0 3.2 3.4 36 3.8 Volts f 5 Actual EOC-12 O Projected EOC-12, POD = 1.0 l ALA_slus.ns 7-7 l

1 I

1 I

Figure 7-3a. J. 51. Farley-1: Comparison of S/G A EOC-12 Actual RPC Co:afirmed/ Untested l Voltage Distribution vs. EOC-12 Distribution Projected with POD'= 0.6 )

l 90 80 - -

70 - - - --

. 60 - - - - -- -

.5 5 50 - - - - - - -

Ee

}e 40 - - - -- - - - -

Z 30 - - - - - - - - -

20 - - - - - - - - - -

10 - - - - - - - - - -- -

0 , s a i I a . OIlnnn - _ _

040.506070.8091.01.11.21.31.41.31.61.71.81.92.0212.22.32.42.52.62.72.52.93.03.13.23.33.43.53.63.73848 Volts IN Actual EOC-12 O Projected EOC-12, POD = 0.6 l P,gure 7-3b. J. 51. Farley-1: Comparison of S/G A EOC-12 Actual RPC Confirmed / Untested Voltage Distribution vs. EOC-12 Distribution Projected with POD = 1.0 80 70 60

$ $0 - - - - - ----

=

5 5 40 _ - - - _.

2 j 30 - - - - - -

20 - - - - - -

10 - - - - - --

0 ' ' ' I I ' EO"""" - - -

040.50.60.70.80.91.01.11.21.31.41.51.61.71.81.62.02.12.22.32.42.52.62.72.82.93.03.1323.3343.53.63.7384.8 Volts (5 Actual EOC-12 O Projected EOC-12, POD =]

Al.A_SIMS.XI.S 7.g

Figure 7-Ja. J. M. Farley-1: Comparison of S/G A EOC-12 Voltage Distribution of All Indications vs. EOC-12 Distribution Projected with POD = 0.6 90 80 -- -

70 - - - - -

. 60 - - - - -

8 3 50 - - - - - - - -

5 c

}o 40 -- - - - - - - --

Z 30 - - _ - - - -

1 I

l 20 - - - - - - - - -- - l l

10 - - - - - - - - - -

0 , I e . Il Il n n - - _ -

040.50.60.70.80.91.01.11.21.31.41.51.61.71.81.92.02.12.22.32.42.5262.72.82.93.03.13.23.33.43.53.63.73.848 Volts l m Actual EOC-12 O Projected EOC-12, POD = 0.6 l l

l Figure 7-4b. J. M. Farley-1: Comparison of S/G A EOC-12 Voltage Distribution of All Indications vs. EOC-12 Distribution Projected with POD = 1.0 80 70 --- -- -

60 - - --

i 50 - - -

'ii

.2

'g 40 .- - - - - - -

C j 30 -- - - - - - -

20 - - - -- -

10 - - - - - - --

0 ' ' 0"""" - - "

040.50.6070.80.91.01.11.21.31.41.51.61.71.81.9202.12.22.3242.52.62.72.82.93.0313.23.33.43.53.63.73.848 Volts Im Actual EOC-12 O Projected EOC-12, POD = 1.0 l A1.A_SIMSES 7-9

1 1 l

1 l 1

i i

i i

I

(

l 2

, Figure 7-5. J. M. Farley-1: S/G C Projected EOC-13 Voltage Distributions

, All Indications i 140-J j 120 -- ---

100 - - - -

i .

= 80 -- -- - - --

I 5

1 Ec Co 60 - -- -- -

o E

40 -- -

, 20 -

-i O N b4=-- --

j 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0 2.2 2.4 2.6 2.8 3.0 3.2

{ Volts l !OPOD=0.6 m POD =1.0 ;l i

I i

4

?

i ALA_SIMS.XLS 7-10 t

I

l l

l i

Figure 7-6. J. M. Farley-1: S/G A Projected EOC-13 Voltage Distribution with l POD = 0.6 1

100 90 - -

l j 80 - - -

4 70 - - - - -

60 - - - - -

I

.5 50 - - - - - -

]

o j 40 - - - - - - - -

30 - - - - - - - - -

20 - - - - - - - - - - - - -

10 - -

l 0 -m - - - - - - - - - - - -Ilus..____-

) 040.50.60.70.80.9 1.01.1 1.21.31.4 1.51.61.71.81.92.02.12.22.32.42.73.65.0 i Volts

, I i

i 1

i 1,

ALA_SIMSXLS 7.] ]

l l

8.0 SLB LEAK RATE ANALYSES This section provides the results of SLB leak rate analyses for EOC-12 and projected EOC-13. Leak rrtes are presented using the draft NUREG-1477 methods and using the EPRI leak rate versus voltage correlation. The J. M. Farley-1 IPC NRC SER requires that the projected EOC-13 leak rates be calculated using the draft NUREG-1477 (Reference 2) leakage model and six probability ofleakage (POL) correlations. The six conelations include log and linear voltage correlations for logistic, normal  !

and Cauchy distributions. The leak rates for these POL correlations are presented in this section. The  !

NRC SER also specifies that the analyses are to be based on a leakage database that excludes outlier; l only on the basis of bad data This guidance is followed for the SER required analyses, although the resulting NRC SER database differs from the EPRI database in the treatment of outliers. In some cases such as the POL correlations, the differences between the NRC and EPRI databases are  ;

negligible. To assist convergence of the NRC and EPRI databases consistent with the NRC guidance i of the February 8,1994, NRC/ Industry meeting (Reference 7), the EPRI criteria (Reference 5) for deleting outliers from the database have been applied to update the database and leak rate versus voltage correlation. The SLB leak rates for the EPRI database are correlated with voltage within the statistical confidence guidelines presented by the NRC at this meeting.

The differences between the EPRI and NRC databases are described in Section 8.1 below. Develop-ment of the burst, leak rate and probability ofleakage correlations for 7/8" diameter tubing and the associated databases has been given in WCAP-13990 (Reference 11). The resulting correlation

)

parameters are also given in Section 8.1. WCAP-13990 also describes leak rate analysis methods for l draft NUREG-1477 as well as Monte Carlo and deterministic methods for applying the EPRIleak rate  !

correlation. These leak rate analysis methods are also applied in this report.

8.1 Database Supporting Alternate Repair Criteria This section describes the database used for the SLB analyses. The differences in the databases currently approved by the NRC based on the Spring,1994, J. M. Farley-1 SER and the EPRI ARC database are described below. The EPRI database is more conservative than the NRC database for SLB leak rate analyses using draft NUREG-1477 methodology and can be used for leak rate analyses based on draft NUREG-1477. The SER also specifies that the latest database excluding outliers only on the basis of bad data is to be used for probability of burst analyses, which are provided for infortnation only since these analyses do not impact the repair limits or the number of tubes to be repaired. For this report, burst probability results are provided for both the EPRI database, which excludes outliers based on atypical morphology as well as bad data, and the NRC database.

8.1.1 EPRI ARC Database i

The database for 7/8 inch diameter tubing is described in EPRI Report NP-7480-L, Volume 1, Revision 1 (Reference 4). However, at the February 8,1994, NRC/ Industry meeting (Reference 7),

the NRC presented resolution of industry comments on draft NUREG-1477. The NRC identified guidelines for application ofleak rate versus voltage correlations and for removal of data outliers in the burst and leak rate correlations. The EPRI evaluation applying the NRC guidance on removal of outliers to update the database for the 7/8 inch tubing correlations has been submitted to the NRC by Reference 5 as a draft Appendix E revision to Reference 4.

8I

In Reference 4, data were removed from the EPRI database based on less explicit criteria than that of Reference 5. The updated criteria led to a change in one data point from Reference 4. Plant D-1 tube R1IC60, TSP 1, which was previously deleted for atypical morphology,is added back into the database. No new data for 7/8 inch diameter tubing has been obtained since the preparation of Reference 4. The data of Reference 4 are updated for the present application based on the outlier evaluation of Reference 5.

At the time of this report, the NRC has not yet provided concurrence with the EPRI ARC database transmitted by Reference 5. The latest NRC documented database acceptance is that provided in the Farley-1 SER. The differences between the current NRC accepted database and the EPRI database described above at::

• NRC Additions to Burst Pressure versus Voltage Correlation The following data points are revised to be included in the NRC database burst correlation:

- Plant D-1: RI1C60-TSP 2, R18C16-TSP 1, R18C21-TSP 1 PlantF: R13C42-TSPs 1 & 2, R16C29-TSP 1, R16C42-TSPs 1 & 2 These data points were excluded from the EPRI database per Criterion 2a of Reference 5.

This criterion excludes data from the EPRI correlations based on crack morphology atypical of the ARC database. In these cases, the lack of remaining uncorroded ligaments in the shallow indications resulted in high voltages for the given degradation levels and associated burst pressures, which led to very high (burst pressure) outlier behavior in the burst pressure versus voltage correlation. In principal, these indications should also be included in the probability of leakage (POL) correlation. However, since all of these indications are non-leakers, it is conservative to not include them as applied for the EPRI ARC database. This conservatism is applied for the POL correlation to climinate the need for multiple POL correlations.

NRC Additions to SLB Leak Rate Database and Correlations The current NRC database would include the following indications in the SLB leak rate analyses as applied to the draft NUREG-1477 methodology or the EPRI SLB leak rate versus voltage correlation:

Model boiler specimen $42-4, with a bobbin voltage of 50.3 volts and a SLB leak rate of 0.861/hr.

Plant J-l pulled tube R8C74 - TSP 1, with a bobbin voltage of 30.9 volts and a SLB leak rate of 0.131/hr.

These data points were excluded from the EPRI database per Criterion 3 of Reference 5. The criterion excludes data from the EPRI correlations based on SLB leak rates less than 10%

higher than normal operating conditions (Specimen 542-4) or a factor of 50 less than the leak rate of specimens containing similar through wall degradation and voltage responses (R8C74).

In these cases,it is expected that the cracks likely became plugged by deposits during leak 8-2

testing and the resulting leak rates are much lower than expected as represented by the rest of the database.

8.1.2 Summary of Burst and Leak Rate Correlations Table 8-1 provides the regression parameters for the six probability ofleakage correlations. The log logistic correlation, as recommended by EPRI, provides the reference SLB leak rates while the other five correlations are provided as sensitivity analyses. Table 8-2 gives sample probability ofleak values for each of the six correlations as a function of voltage.

Table 8-3 shows the regression parameters for the EPRI SLB leak rate versus voltage correlation using the EPRI database. Leak rate calculations using the voltage correlation are only reported herein using the EPRI database.

A summary comparison of the various correlation parameters used as input to the analyses for both the NRC and EPRI recommended databases is provided in Table 8-4. For the leak rate analysis methodology described in draft NUREG-1477, the use of the EPRI database is seen to be conservative in that the mean arithmetic average, or expected, leak rate from the leak rate data is greater than that obtained from the NRC recommended database. In addition, the standard deviation of the data is greater for the EPRI database. However, the differences are minor and would not be expected to significantly affect the outcome of the SLB leak rate analyses. It is also seen that the probability of leak correlation parameters are not significantly different for the two databases. Differences in the burst correlation between the NRC and EPRI databases is, however, significant. While the intercept and slope parameters are not substantially difTerent, the standard error is much higher for the NRC database. This resulting increase in the uncertainty band for the NRC database results in significantly higher tube burst probabilitics for the NRC database as shown in Section 9. The larger uncertainty band results from inclusion of the high burst pressure outliers, as noted above,in the NRC database.

8.2 Alternate Leakage Analysis Methods Two methods for leak rate analyses are applied in this report. These include the draft NUREG-1477 methodology as required by the J. M. Farley-1 NRC SER and the application of the leak rate versus voltage correlation developed in the previous section.

Draft NUREG-1477 Methodology The NRC methodology of draft NUREG-1477 obtains the number of indications that are to be considered as being returned to service,N, as:

Ne N = N, + N "# - N = N, + 1 - POD y# _g,

_y, '

(8.1)

POD POD where, N, = the number of detected bobbin indications, N, = the number of repaired indications, N , = the number of indications not detected by the bobbio inspection, POD = the probability of detection (0.6 for NRC methodology).

8-3

The above adjustments for POD have been incorporated in the BOC and EOC voltage distributions so that no further adjustments are required for the leakage calculation. Section 3.3 of draft NUREG-1477 states that the total leak rate, LR, should be determined as:

LR = p P + Z cP+pP 2 2 p 3 (y,p,2) ,

2 (8.2) where, p = the mean of the leak rate data independent of voltage, o = the standard deviation of the leak rate data independent of voltage, P, = the probability that a tube leaks for the f" voltage bin, N, = number ofindications (after POD adjustment) in the th voltage bin, P = E(N,P,) - expected number of indications that Icak summed over all voltage bins, -

2 = standard normal distribution deviate (establishes level of confidence on leakage).

For the total leakage, the first term of the above equation represents a mean expected leak rate while I the square root term is an effective standard deviation for the total leakage based on the variance of the product of the probability ofleak and the predicted leak rate. Draft NUREG-1477 recommends that Z be taken as 2, corresponding to an upper confidence bound of 98%

Application of EPRI Leak Rate CorrcIntion The leak rate versus voltage correlation can be simulated in conjunction with the EOC voltage distributions obtained by Monte Carlo methods, or by applying the POL correlation and leak rate correlation to the binned EOC voltage distribution obtained by Monte Carlo methods as applied for the draft NUREG methodology. Parallel analyses for another plant verified that the full Monte Carlo leak rates and the application of the correlations to the EOC voltage distribution yield essentially the same results. Thus it is adequate to apply the correlations to the EOC voltage distributions.

The determination of the end of cycle leak rate estimate proceeds as follows. The beginning of cycle voltages are determined using the methodology provided in draft NUREG-1477. The distribution of indications is binned in 0.IV increments. The number ofindications in each bin is divided by 0.6 to account for POD. The resulting number ofindications in each bin is reduced by the number of indications plugged in each bin. The final result is the beginning of cycle distribution used for the Monte Carlo simulations. The NDE uncertainty and growth rate distributions are then independently sampled to estimate an end of cycle distribution, also reported in bins of 0.IV increment.

For each voltage bin the leak rate versus bobbin amplitude correlation is used to estimate an expected, or average, leak rate for indications in that bin. The probability ofleakage correlation is then used to estimate the mean probability of leak for the indications in each bin. The relationships derived in Appendix C of draft NUREG 1477 for the variance of the product of the probability ofleak with the leak rate and for the total leak rate are then used to estimate the expected total leakage and variance for the sum of the indications in cach bin as a function of the correlation means and estimated ,

variances for the leak rate and probability ofleak. The expected total leakage for the entire distribu-tion is then obtained as the sum of the expected leak rates for each bin including covariance terms.

The variance of the total leak rate for the distribution is obtained as the sum of the variances for each bin. The standard deviation of the total leak rate is then taken as the square root of the variance of the total leak rate. The upper bound 95% confidence limit on the total leak rate is then obtained as 8-4

the expected total leak rate plus 1.645 times the standard deviation of the total leak rate. This methodology provides a deterministic leak rate analysis given a binned EOC voltage distribution and is applied for most of the leak rate analyses of this report. Monte Carlo analyses for the leak rate can also be applied to the binned EOC voltage distribution. A few comparisons of the deterministic and Monte Carlo methods applied to the same binned EOC voltage distributions are given in this report.

8.3 Projected EOC-12 SLB Leak Rates 4

SLB leak rates were calculated for the projected S/G C EOC-12 voltage distributions of Figure 7-1.

In addition, leak rates were calculated for the new indication method described in Section 10. The results are given in Table 8-5. The full NRC methodology with a POD = 0.6 adjustment and draft NUREG-1477 leak rate independent of voltage yicids a SLB leak rate of 0.78 gpm. Applying the POD-adjusted EOC voltage distribution with the APC leak rate correlation yields 0.066 gpm. Without the POD adjustment, the SLB leak rates are 0.50 and 0.041 gpm for the draft NUREG and APC leak rate methods. This difference (factor of about 13) between the draft NUREG and APC methods is typical of the differences between the NRC method which ignores the leak rate dependence on voltage and methods applying the leak rate versus voltage correlation which show a large reduction due to the strong dependence ofleakage on voltage. Both methods utilize the same probability ofleakage correlation. Application of POD = 0.6 is seen to result in about a 50% increase in the leak rates.

Following the 1992 outage at EOC-ll, the projected SLB leak rate at EOC-12 was < 0.1 gpm based on applying the APC leak rate versus voltage correlation to all RPC confirmed plus not RPC inspected indications left in service without a POD adjustment. This result is approximately the same as found by the current analysis using the APC leak rate correlation.

Also shown in Table 8-5 are leak rates obtained with the alternate new indication method developed in Sectior 10 for defining the BOC distribution. For the new indication method, SLB leak rates of 0.35 and 0.026 gpm are obtained for the draft NUREG and APC leak rate correlation methodologies.

8.4 Comparison of Leak Rates for Projected and Actual EOC-12 Distributions SLB leak rates were also calculated for the actual S/G C EOC-12 voltage distribution of Figure 7-1 and Table 7-1 which includes bobbin indications confirmed by RPC plus indications not RPC inspected. This distribution and leak rate represents the target that the projections are attempting to predict. The resulting leak rates are given in Table 8-5 for the actual EOC-12 distribution. Although S/G C had the largest number ofindications left in service at BOC-12 and was therefore used for projections to EOC-12, the actual distribution for S/G A was found to have the largest bobbin indications at the EOC-12 inspection. Therefore, leak rates were calculated for the actual distributions of both S/Gs A and C to determine the most limiting based on EOC indications. The S/G C leak rate for the NRC methodology is 0.18 gpm compared to 0.010 gpm for the EPRI correlation methods. The corresponding leak rates for the actual S/G A distribution are 0.25 and 0.031.

With the POD adjustment on the projected EOC-12 voltage distribution, the leak rate for the NRC methodology is 0.78 gpm, compared to 0.066 gpm for the EPRI correlation method. Without the POD

[ adjustment, the projected leak rates are 0.50 and 0.041 gpm for the draft NUREG and EPRI correla-

! tions, which are in better agreement with the leak rate for the actual distributions for either S/Gs A or C. These results, like the voltage distribution comparisons of Section 7.2, indicate that the NRC 8-5

application of a POD = 0.6 leads to excessive conservatism. The leak rates for all actual bobbin indications, which includes RPC NDD indications not expected to contribute to leakage, are 0 34 and 0.022 gpm for the draft NUREG and APC correlation methods applied to S/G C and 0.29 and 0.033 gpm for S/G A. Both the POD = 0.6 and 1.0 methods yield higher leak rates than obtained for even the conservative all bobbin indication distribution for either S/G A or C.

It is seen that the leak rates for the alternate methods of Section 10 also exceed the leak rate obtained for the actual S/G C RPC confirmed plus not RPC inspected distribution although the agreement is significantly better than obtained for the POD of 0.6 or 1.0. The new indication method applied to the EPRI correlation for S/G C slightly underestimates (0.026 versus 0.031 gpm) the actual leak rate for S/G A. Although the new indication method is specifically targeted toward predicting the RPC confirmed distribution, remaining conservatisms in this method result in good agreement with the actual all bobbin indication leak rates as shown in Table 8-5. This result lends strong support for application of the new indication method for defining BOC voltage distributions for IPC/APC applications. .

L.

8.5 Projected EOC-13 SLB Leak Rates Leak rates at EOC-13 were calculated using the draft NUREG-1477 methodology as required by the NRC SER and using the APC leak rate versus voltage correlation. SLB leak rates were determined for the projected EOC 13 distributions with adjustment for POD = 0.6 and without adjustment for POD (POD = 1.0), as well as for the new indication method developed in Section 10. Results of the leak rate analyses are given in Table 8-5. Projected leak rates were obtained for both S/G C and S/G A for the POD = 0.6 distribution. This comparison between S/Gs was performed to determine the limiting S/G. S/G C had the largest number ofindications while S/G A had the highest voltage indication for which the POD adjustment results in leaving 0.67 indication left in service. This comparison, given in Table 8-5 supports S/G C as the limiting S/G and all remaining analyses were performed for S/G C.

For the reference leak rate based on the log logistic POL correlation, the full NRC methodology with a POD = 0.6 adjustment and draft NUREG-1477 leak rate independent of voltage yields a SLB leak rate of 0.61 gpm for S/G C. Applying the POD adjusted EOC voltage distribution with the APC leak rate correlation yields 0.045 gpm. These results are for the deterministic analysis method. Applying a Monte Carlo analysis for the POL and EPRI leak rate correlation to the binned EOC voltage distribution yields a leak rate of 0 081 gpm compared to the deterministic 0.045 gpm. This difference between deterministic and Monte Carlo methods, as a ratio,is larger than found in prior analyses.

~ ~ ~ ~

However, the leak rates are very small and wbject to the in!!uence of a fcw large Icai rate samples in the Monte Carlo analyses. Without the POD adjustment, the SLB leak rates are 0.44 and 0.032 gpm for the draft NUREG and APC leak rate methods. The new indication method with POD = 1.0 results in leak rates of 0.39 and 0.030 gpm for the draft NUREG and APC leak rate methods. All projected EOC-13 SLB leak rates are significantly less than the allowable leak limit (NRC SER) of 22.6 gpm.

Consistent with the SER requirements, sensitivity analyses for EOC-13 SLB leak rates were also performed by applying the remaining five of the six POL correlations described in Section 8.1. The Cauchy POL distnbution yields the largest SLB leak rates. The Cauchy distribution is considered technically unacceptable because: it yicids unrealistically high POL values below 1 volt where the 86 "

POL by all 7/8" and 3/4" data (about 200 pulled tube data points)is essentially zero; it is too steep in the 5 to 8 volt range compared to the leakage data; it yields less than expected POL at voltages above about 10 volts; and it results in a poorer statistical fit to the data. Similarly, the linear normal and logistic fits tend to overestimate the low voltage leak rates compared to 6e extensive database. The log logistic and log normal distributions yield comparable trends and statistial fits to the data such that there is no clear basis to select one or the other of these two correlations. h'cwever, as shown in Table 8-5, the log logistic and log normal POL correlations yield similar SLB leak rates and the leak rate for the log logistic POL is slightly higher than for the log normal POL.

From Table 8-5,it is seen that the use of the APC leak rate versus voltage correlation leads to about a factor of 13 reduction in the SLB leak rate compared to the draft NUREG-1477 methodology of averaging all leak rate data independent of voltage Westinghouse strongly recommends the applica-tion of the APC correlations for SLB leak rate analyses. Comparisons of the leak rates calculated from the actual and projected EOC-12 voltage distributions support the use of the new indication or POD = 1.0 methods as adequately conservative while the POD = 0.6 results in excessive conservatism.

M E..

8-7

l' Table 8-1: Results of Regression Fits of Logarithmic Forms of the Probability of Leak Distribution Functions to 7/8" OD Tube Data.

Log-logistic Log-normal Log-Cauchy Parameters Parameters Parameters a, -6.8974 -3.7394 -15.0653 a, 8.3507 4.5779 17.7301 V,, 3.4999 0.7999 67.1114 V,, -3.8459 -0.8801 -76.9510 V,, 4.5822 1.0749 88.9730 Deviance 25.09 24.87 28.68 Results of Regression Fits of Non-Logarithmic Forms of the Probability of Leak Distribution Functions to 7/8" OD Tube Data.

Logistic Normal Cauchy Parameters Parameters Parameters a, -4.9991 -2.7359 -10.0508 a, 0.6565 0.3582 1.3877 V,, 1.2530 0.2608 25.9471 V,, -0.1597 -0.0337 -3.4945 0.0261 0.0062 0.4825

~-

V,;

Deviance 26.06 25.40 30.50 8-8 i

l

l Table 8-2: Sample Results for Probability of Leak for 7/8" SG Tubes Volts Log-Logistic Log-Normal Log-Cauchy Logistic Normal Cauchy Function Function Function Function Function Function 0.1 2.4E-07 1.0E-16 9.7E-03 7.lE-03 3.5E-03 3.2E-02 0.5 8.2E-05 1.6E-07 1.6E-02 9.3E-03 5.3E-03 3.4E-02 1.0 1.0E-03 9.2E-05 2.1E-02 1.3E-02 8.7E-03 3.7E-02 1.5 4.4E-03 1.7E-03 2.7E-02 1.8E-02 1.4E-02 4.0E-02 2.0 1.2E-02 9.lE-03 3.3E-02 2.4E-02 2.2E-02 4.3E-02 3.0 5.2E-02 6.0E-02 4.8E-02 4.6E-02 4.8E-02 5.4E-02 4.0 1.3E-01 1.6E-01 7.lE-02 8.5E-02 9.6E-02 7.0E-02 5.0 2.6E-01 2.9E-01 1.lE-01 1.5E-01 1.7E-01 9.9E-02 7.0 5.4E-01 5.5E-01 4.7E-01 4.0E-01 4.lE-01 4.0E-01 10.0 8.1E-01 8.0E-01 8.9E-01 8.3E-01 8.0ET01 9.2E-01 12.0 8.9E-01 8.9E-01 9.2E-01 9.5E-01 9.4E-01 9.5E-01 15.0 9.5E-01 9.5E-01 9.5E-01 9.9E-01 1.0E+00 9.7E-01 20.0 9.8E-01 9.9E-01 9.6E-01 1.0E+00 1.0E+00 9.8E-01 30.0 1.0E+00 1.0E+00 9.7E-01 1.0E+00 1.0E+00 9.9E-01 8-9

Table 8-3: Regression Analysis Results for Fitting log (Leak Rate) vs log (Volts) for 7/8" Tubes l Parameter Value Value Parameter b, 1.892 -1.518 b, SE b, 0.508 0.601 SE b, 2

r 38.6% 0.639 SE log (G)

F 13.85 22 DoF S S,,, 5.654 8.980 S S,,,

Pr(F) 0.12 % 1.580 SSu ,y pi-value 0.12 % 1.93 % po-value O

8 - 10

Table 8-4: Differences in Correlations Obtained Using the NRC Recommended and the EPRI Recommended Databases Item NRC Recommended EPRI Recommended Databasem Database Draft NUREG-1477 Mean leak rate, (lph) 13.7 14.9 Standard Deviation, o (Iph) 21.1 21.7 POL Correlation using Log-logistic Function a, (logit intercept parameter) -6.9940 -6.8974 a, (logit slope parameter) 8.4538 8.3507 Deciance 25.17 25.09 Leak Rate Correlation using Linear log (G) on log (V) Regression a, (intercept parameter) N/A -1.5190 a, (slope parameter) N/A 1.8920 Index of Determination, F N/A 38.6 %

Standard Error, a N/A 0.6390 Burst Pressure Correlation using Linear P, on log (O Regression a,(intercept parameter) 8.5679 8.2390 a, (slope parameter) -2.6689 -2.5291 Index of Determination, / 73.5 % 82.7 %

Standard Error, a 1.1821 0.8910 Notes: (1) Differs from the recommended database as described in Section 8.1.

8 - 11

S Table 8-5: Summary of EOC-12 and EOC-13 SLB Leak Rate Analyses for SG-C SLB Leak Rates - gpm EOC Voltage Distribution NURE -1477 Leak Rate Method Correlation EOC-12 Results")

Actual Voltage Distribution

. RPC confirmed plus not RPC inspected . ,

- S/G C 0.18 0.010

- S/G A 0.25 0.031 All bobbin indications

- S/G C 0.34 0.022

- S/G A 0.29 0.033 Projected Voltage Distributions

. Bobbin ind. with POD = 0.6 0.78 0.066 All bobbin indications (POD = 1.0) 0.50 0.041 New indication method (Section 10) 0.35 0.026 Projected EOC-13 Results Bobbin with POD = 0.6 Reference Leak Rate Analysis - Log Logistic POL

- S/G C deterministic analysis 0.61 0.045

- S/G C Monte Carlo analysis

• 0.081

- S/G A Deterministic analysis 0.45 0.037 Leak Rate Sensitivity to POL Correlation

- Log normal POL 0.39 0.032

- Log Cauchy POL 2.87 0.142

- Logistic POL 2.03 0.110

- Normal POL 1.60 0.085

- Cauchy POL 4.34 0.209 All bobbin indications (POD = 1.0)") 0.44 0.032 New indication method (Section 10)") 0.39 0.030 Notes: 1. Leak rate analysis using log logistic POL.

2. Monte Carlo analysis using POL and leak rate correlations applied to the binned EOC voltage distribution. All other calculations use deterministic analyses.

8 - 12

9.0 SLB TUBE BURST PROBABILITY ANALYSES 9.1 Projected EOC-12 SLB Burst Probability The results for SLB burst probabilities for EOC-12 using the EPRI database burst correlation are given in Table 9-1. Burst probabilities were calculated using deterministic analyses for the actual EOC voltage distribution (S/Gs C and A) and are compared with the projected S/G C EOC-12 distributions for the POD = 0.6,1.0 and new indication methodologies for defining the BOC distribution. The projections were made for S/G C as this S/G had the most limiting indications at BOC-12 whereas S/G A had the largest actual EOC indication. The EOC-12 burst probability for the actual RPC confirmed plus not RPC inspected distribution, which are the only indications expected to contribute to the burst probability, is estimated as 1.1 x 10" for S/G C and 2.4 x 10" for S/G A. For all actual bobbin indications, including RPC NDD, the burst probabilities are 3.5 x 10" for S/G C and 3.1 x 10" for S/G A. For S/G A, the difference in burst probability between RPC confirmed and all indications is less than for S/G C since the largest indications in S/G A were RPC confirmed. Although S/G C was used for projections and S/G A is the most limiting for actual RPC confirmed indications, the differences between S/Gs is only a factor of two on the burst probabilities.

With the POD = 0.6 adjustment, the projected burst probability for S/G C at EOC-12 is estimated at 1.3 4

x 10 while for all indications lefl in senice, the probability is 6.8 x 10" Thus the POD adjustment results in about a factor of two increase in the burst probability and results in a factor of 10 higher burst probability than obtained from the actual EOC-12, RPC confirmed voltage distribution. The alternate method of Section 10 results in a burst probability of 3.8 x 10" The burst probability calculated for the actual S/G C EOC-12 distribution of RPC confirmed indications is 3.5 x 10" for all bobbin indications and 1.1 x 10" for RPC confirmed plus not RPC inspected indications. This range for S/G C bounds the range between RPC confirmed and all indications for S/G A. Thus the projected methods provide conservatism versus the actual EOC-12 distributions for both S/Gs A and C.

9.2 Projected EOC-13 SLB Burst Probability Table 9-1 also includes the deterministic and Monte Carlo results projected for the EOC-13 tube burst probabilitics for both S/Gs C and A. From the comparisons of these two S/Gs, vas determined that S/G C is most limiting and all remaining analyses were performed for S/G C. ~1k Monte Carlo analyses provide the reference analysis results. Applying the NRC methodology of the POD = 0.6 adjustment, including the NRC database, as requested in the Farley-1 SER yields a S/G C EOC-13 burst probability of 1.2 x 10'2 This result, applying the very conservative NRC database, is lower than the WCAP-12871 Rev. 2 acceptance guideline of 2.5 x 10 2which is based on the NUREG-0844 analyses.

Thus the EOC-13 burst probability is acceptable.

Applying the EPRI database results in a EOC-13 burst probability of 4.3 x 10" which is about a factor of 30 lower than the NRC database and much lower than the acceptance guideline. It is seen from Table 9-1 that the deterministic analysis methods result in burst probabilities about a factor of 2 higher than the Monte Carlo methods.

The POD = 1.0 and new indication methods result in tube burst probabilities almost a factor of two lower than the POD = 0.6 method and less than 10 2 even for the NRC database.

9-1

1 Table 9-1. Summary Results for Tube Burst Probabilities Burst Probability

• EPRI Database = NRC Database S/G C S/G A S/G C S/G A Actual EOC-12 Voltage Distritation l

RPC confirmed plus not RPC inspected 1.1 x 10" 2.4 x 10" All bobbin indications left in sen' ice 3.5 x 10" 3.1 x 10" Projected EOC-12 NRC model with POD = 0.6 1.3 x 10~2 I i

All bobbin indications left in service (POD = 1.0) 6.8 x 10" New Indication Method (3ection 10.0) 3.8 x 10" Projected EOC-13 NRC model with POD = 0.6 9.0 x 10" 5.8 x 10" 2.2 x 10-2 1.4 x 10 2

- Monte Carlo Analyses

• 4.3 x 10" 1.2 x 10 2 All bobbin indications left in service (POD = 1.0) 5.4 x 10" 1.3 x 10 2

- Monte Carlo Analyses 2.9 x .10" 7.5 x 10

New Indication Method (Section 10.0) 4.6 x 10" 1.2 x 10'2 1

- Monte Carlo Anab ses 6.9 x 10~2 Notes:

1. All analysis results, except where specifically noted as a Monte Carlo analysis, are deterministic analyses. I

2. The Monte Carlo analyses are based on 100,000 samples of each bin of the EOC-12 voltage distribution and are evaluated at 95% confidence on the burst probability (function of the number of samples).

1 9-2 l

10.0 ALTERNATE METIIOD FOR DEFINING BOC INDICATIONS LEFT IN SERVICE It is noted in Section 7.2 that the application of a POD = 0.6 to obtain BOC voltage distributions leads to excessive conservatism when EOC-12 projections are compared with actual inspection results for the distributions. The BOC distribution based on all bobbin indications left in service, including RPC NDD indications, yicided conservative but reasonabic agreement with the actual EOC distribution. However, this method compensates for ignoring new indications by the inclusion of all RPC NDD indications and equivalent compensation may not be likely in future outages. Thus it is desirable to develop an improved method for defining BOC distnbutions that accounts for potentially new indications without the extreme conservatism of the POD = 0.6 adjustment and that recognizes that some RPC NDD indications might develop to flaw indications over the next cycle. It is unlikely that new or RPC NDD indications at TSP intersections will develop to leakers our the next operating cycle such that inclusion of these indications at any magnitude in the BOC distribution is, by itself, conservative. New indications at a given outage or time includes indications missed at the prior inspection, indications grown from nondetectable to detectable levels, and growth from no indication to a detectable indication.

For IPC/APC applications, there is no need to distinguish these sources of new indications as the only consequence ofimportance to leakage or burst is that the new indication is sufficiently large to be detectable by both bobbin and RPC probes. In addition, some low voltage indications reported at the prior inspection are no longer found at the subsequent inspection, particularly when conservative calling criteria are applied. This has been found for the last Farley-2 and Catawba-1 IPC evaluations as well as the current Farley-1 evaluation. For projection considerations, the prior inspection indications not found in the current inspection should be subtracted from the total new indications to obtain a net number of new indications.

A desirable BOC distribution will result in EOC projections which yield good overall agreement with the larger voltage (greater than 1-2 volts) RPC confirmed indications. The larger indications have finite probabilitics of leakage or burst. Only RPC confirmed indications wcaid have a significant probability ofleakage over the prior cycle and are consistent with the leak rate ar.d Surst correlation database which is predominantly comprised of RPC confirmed indications. The methodology should recognize that new indication and POD considerations are plant specific due to varying degrees of the influence of residual signals on detectability of small indications, enhancements in eddy current analysis guidelines and corrosion controls implemented at a specific time. In addition, the methodology should recognize that historical data indicates that only a few RPC indications tend to develop to RPC confirmed indications over the subsequent operating cycle. The following sections define a BOC distribution methodology consistent with these objectives which is then applied to obtain EOC-12 projections for comparison with the actual distributions. This methodology is found to result in good agreement with the 1994 EOC-12 voltages and has now been found to provide the best agreement with actual EOC distributions for all four IPC plants for which IPC assessments have been performed. This method, called the new indication method is proposed as an alternative to the POD = 0.6 method for projecting EOC voltage distributions following IPC or APC implementation.

10.1 Allowance for Undetected or New Indications An undetected or new indicatice is defined as an indication found in the current outage inspection that was not identified at the prior inspection, independent of the causative factor for the new indication.

The method used to develop growth rates in this report is based upon reevaluating the prior outage eddy current data of all indications found in the current inspection. For most indications, when given the somewhat larger flaw indication in the current inspection, reevaluation of the prior data permits identification of the smaller flaw and assignment of a voltage at the prior inspection even if the flaw 10-1

was not reported at the prior inspection. This process permits assignment of voltages to indications which were not reported at the prior outage and leads to a distribution of prior cycle voltages for new indications which can be described / applied as a voltage distribution for undetected (NDD) indications as an alternative to applying more arbitrary POD adjustments to detected indications. This process can lead to a highly conservative distribution of new or undetected voltages when the current inspection involves an "mspection transient" resulting from implementation of significantly more conservative eddy current analysis guidelines than applied at the prior inspection. This implementation of more conservative guidelines might occur at first time implementation of an IPC, which was in 1992 for Farley-l. Since the 1994 inspection is the second application of IPC guidelines, the currently developed distnbution of voltages for new indications not detected at the prior outage provides a meaningful description of an undetected voltage distribution. This method for defining undetceted indication voltage distributions was presented to the NRC in WCAP-13692 (April,1993; Ref.12) prepared in response to the NRC APC Task Team questions on analytical models for SLB leakage analyses.

As bobbin analysis guidelines are made increasingly more conservative, the potential for false bobbin calls increases. Current inspection practices and IPC implementations utilize RPC inspection to assess ,

the significance of the bobbin call If the bobbin indication is not detected by the RPC inspection, the bobbin call is either a false call or the indication is too small to be detected by the RPC probe. An indication not detected by RPC would not challenge tube integrity (burst or leakage) and need not be considered as a significant undetected indication at the prior outage. Thus the latest inspection results I for indications confirmed by RPC inspection or not RPC inspected provide the population of tubes for which the prior cycle cddy current data is reviewed to develop the undetected or NDD voltage .

distribution. Applying the prior inspection undetected voltage distribution as the current inspection NDD indications for the next BOC distribution is conservative since the data analysis guidelines tend toward the same or more conservative guidelines in successive outages. The undetected voltage distribution is added to the detected distribution for indications left in service to obtain the net BOC soltage distribution adjusted for POD considerations on a plant specific and time dependent basis.

Based on the above, the following guidelines are used to estimate the voltage distribution of undetceted or NDD indications:

Population for Evaluation: Latest inspection indications confirmed by RPC or not RPC inspected that were not reported in the prior inspection (i c., new indications). The only assumption for this population is that NDD indications have grown in one cycle to detcetable indications to be of a concern for leakage considerations. Alternately, an indication not detceted in two successive inspections can be assumed to result in negligible leakage over the next cycle.

Process: For the above population, the prior cycle cddy current data is reevaluated by applying the latest analysis guidelines to assign a voltage to the indication at the prior cycle. This process is the same as that applied to develop voltage growth rates for Farley-1 S/G indications. If the indication is not detectable at the prior outage, a reasonable estimate would be to assign 0.5 volt to the indication.

Undetceted (NDD) or New Indication Voltage Distribution: The reevaluated voltages at the prior inspection in which the indication was not detected are used to define the new indication distribution for inclusion in the next BOC distribution.

Based on these guidelines, voltage distributions were developed for undetected indications to be included in the BOC-12 and BOC-13 distributions. These distributions were developed for each cycle 10-2

as given in Table 10-1. Only the most limiting S/G is used for EOC voltage projections. The distributions difTer between S/Gs and could varv over successive cycles. For this reason, the S/G C (limiting S/G) undetected distribution was conservatively assumed to be 50% of the sum over all three S/Gs. The resulting undetected indication voltage distributions for S/G C are given in Table 10-2 for BOC-12 and Table 10-3 for BOC-13. Note that the BOC-12 distribution represents the undetected indications at EOC-Il (first IPC implementation) and the BOC-13 distribution represents the undetected indications at EOC-12 (second IPC inspection). It is seen from Table 10-1 that the second implementation ofIPC guidelines reduced the number of undetected indications compared to the prior inspection summed over all S/Gs from 669 to 653. The undetected maximum voltage is <2.9 volts for BOC-12 and <4.5 volts for BOC-13. As noted in prior sections, the NDE guidelines were changed between the EOC-11 and EOC-12 inspections to include all phase angles between $* and 170 as potential bobbin indications. This change resulted in identification of additional new indications at EOC-12 and, in particular, the two news indications above 2.5 volts in Table 10-3. The new indication method, based on applying 50% of the three S/G sum of new indications, results in BOC distributions that account for previously undetected large voltage indications independent of the S/G in which the indication occuned. For the larger previously undetected indications, this is equivalent to assuming a 50% chance that the latest inspection could again not detect the larger indications. Applying the new indication distribution as a fraction of the sum over all S/Gs also helps to reduce the sensitivity of the EOC projections to the limiting S/G used for projections.

10.2 Considerations for RPC NDD Indications As previously discussed, only a few RPC NDD indications develop to RPC confirmed indications at subsequent inspections. In the Faricy-1 EOC-12 inspection, only 11% (13% in highest S/G) of the prior inspection RPC NDD indications developed to RPC detectable indications. Although all RPC NDD indications less than the maximum voltage repair limit of 3.6 volts are left in service, it would be excessively conservative to include all of these indications in the BOC distribution as potential leakers at the end of the next operating cycle. Therefore, a conservative assumption that 20% of the RPC NDD indications left in service in S/G C will develop to RPC confirmed indications was applied to define the contribution of RPC NDD indications to the BOC distribution. This assumption provides adequate ,

conservatism and avoids the excessively conservative assumption that all RPC NDD indications left in service will develop to confirmed indications. The resulting BOC-12 and BOC-13 voltage distributions are included in Table 10-2 and Table 10-3, respectively. The EOC-12 and EOC-13 voltage distributions determined from the alternate BOC distributions described above are shown in Figures 10-1,10-2 and 10-3. The EOC distributions are also given in Tables 7-i and 7-2.

Indications that are RPC NDD in the last two inspections should be excluded from the BOC distributions. It can be expected that none or very few of these indications will become RPC detectable at the next inspection. To simplify the current analysis, the two consecutive cycle RPC NDD indications were retained in the BOC distributions. However, eliminating consecutive cycle RPC NDD indications from the BOC distributions becomes increasingly more important for future applications as the population of these indications can be expected to increase from cycle to cycle when the RPC inspection threshold is maintained :onstant between inspections. For Faricy-1, the RPC inspection threshold increased from 1.0 volt 's 1.5 volt for the EOC-12 inspection as the repair limit increased from 1.0 to 2.0 volte.

10.3 Comparison of Projections Applying Alternate Method with Actual EOC-12 Distributions 10-3

Figure 10-1 comparcs the projected EOC-12 voltage distributions obtained with the above alternate method (proposed methodology for defining BOC distributions) with the actual EOC-12 voltage distributions for RPC confirmed and not RPC inspected indications. The maximum projected EOC-12 voltage is 3.3 volts compared to the largest RPC confirmed indication that had a bobbin voltage of 1.8 s olts. It is seen that the alternate method provides a conservative projection of the actual voltage distribution for voltages above 1.0 volts and adequately represents the di tribution below 1.0 volts Figure 10-2 compares the alternate method projections with the actual distribution of bobbin indications independent of RPC confirmation. The alternate method,in this case, shows very good and slightly conservative agreement with the bobbin voltage distribution. This agreement between projections and inspection results is an improvement over that obtained in Figurc 7-2 for the POD = 0,6 distribution or all bobbin indications including all RPC NDD indications in the BOC distribution with no allowance for undetected indications. The attemate method achieves the good agreement with inspection results by applying a rational, plant specific basis for accounting for undetected indications and for including allowances for RPC NDD indications. This alternate, new indication method is proposed as an acceptable alternative to applying the draft NUREG-1477 methodology with a POD = 0.6 since this method provides an improved methodology over the other methods for defining BOC distributions evaluated in .his report and in all other IPC assessments. 1 From Tables 8-5 and 9-1,it is seen that the alternate method, as well as all bobbin indications with no ,

POD adjustment, yield good agreement with the SLB leak rates and burst probabilities calculated for the actua! EOC-12 voltage distribution found in the inspection. This further supports the alternate method for defining BOC voltage distributions 10.4 Projected EOC-13 Voltage Distributions and SLH Leak Rates Figure 10-3 shows the projected EOC-13 bobbin voltage distribution obtained by applymg the alternate method for defining BOC distributions. The maximum projected EOC voltage is 4.5 volts which compares to the Table 7-2 results of 3.2 volts maximum for all indications left in service and 3 3 volts maximum for the POD adjusted distributions.

The EOC-13 distribution leads to a SLB leak rate (Table 8-5) of 0.39 gpm for the draft NUREG-1477 leak rate methodology (excluding the POD factor) and 0.030 gpm for application of the EPRI leak rate correlation. These results can be compared to the Table 8-5 results at EOC-13 of 0.61 gpm for the NRC methodology and 0 045 gpm for the WCAP methodology where both methods include t' a POD adjustment Thus the more realistic treatment for undetected and RPC NDD indications leads to about a factor of 1.5 reduction m projected SLB leak rates.

l m4 L________ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ . _ - - - - - - - - - - - - - - - - - - _ _ _ _ - - . - - - - - - . - - - - - - - . - - - - - - _ _ _ _ _ . - - - - _ - - - - - - - - _ _ _ _ _ _

Table 10-1. Distribution of New Bobbin Indications for BOC-12 and BOC-13 BOC-12 BOC-13 Bobbin Voltage All SGs S/G-A S/G-B S/G-C Total 0.1 0 0 0 0 0 0.2 0 0 0 0 0 0.3 10 0 1 0 1 0.4 49 6 14 6 26 0.5 98 25 17 10 52 0.6 134 26 22 26 74 0.7 120 29 25 34 88 0.8 ,

83 32 26 32 90 0.9 74 29 23 32 84 1.0 41 24 20 25 69 1.1 11 17 14 10 41 1.2 12 6 5 18 29 1.3 14 14 6 17 37 1.4 9 7 2 7 16 1.5 5 5 5 9 19 1.6 1 5 1 1 7 1.7 3 2 0 3 5 1.8 3 5 0 0 5 1.9 0 1 0 0 1 2.0 0 0 0 0 0 2.1 1 0 0 0 0 2.2 1 3 1 1 5 2.3 0 0 0 1 1 2.4 0 0 0 0 0 2.5 0 0 0 1 1 2.6 0 0 0 0 0 2.9 0 0 1 0 1 4.5 0 1 0 0 1 Totals 669 237 183 233 653 l l l Notes: 1 Ncludes all new indications found at EOC inspection and confirmed by RPC or not RPC inspected.

SEC10TBL.XLS Table 10-1 10-5

Table 10-2. Farley-1 SG-C DOC-12 Distribution - A: ternate Method A B C

'92 SG-C bobbin indications RPC '92 SG-C bobbin New'92 confirmed or not indications RPC indications tested and not NDD and not for ALL repaired; in '92 repaired; in '92 SGs; in '91 A + 0.2

• B Voltage reanalysis volts reanalysis volts Volts + 0.5' C 0.0 0 0 0 0.0 *j 0.1 0 0 0 0.0 0.2 0 0 0 0.0 0.3 0 0 10 5.0 0.4 0 0 49 24.5 0.5 4 0 98 53.0 0.6 25 0 134 92.0 0.7 42 0 120 102.0 0.8 58 2 83 99.9 0.9 51 0 74 88.0 1.0 66 5 41 87.5 1.1 0 29 11 11.3 1.2 0 45 12 15.0 1.3 0 30 14 13.0 1.4 0 35 9 11.5 1.5 0 22 5 6.9 1.6 0 24 1 5.3 1.7 0 12 3 3.9 1.8 0 7 3 2.9 1.9 0 6 0 1.2 2.0 0 6 0 1.2 2.1 0 6 1 1.7 2.2 0 4 1 1.3 2.3 0 1 0 0.2 2.4 0 1 0 0.2 2.5 0 0 0 0.0 2.6 0 0 0 0.0 2.7 0 0 0 0.0 2.8 0 0 0 0.0 2.9 0 1 0 0.2 3.0 0 0 0 0.0 Totals i 246 236 I 669 i 627.7 SEC10TBL.XLS Table 10-2 10-6

Table 10-3. Farley-1 SG-C BOC-13 Distribution - Alternate Method A B C

'94 SG-C bobbin New '94 indications indications RPC '94 SG-C bobbin for ALL SGs confirmed or not indications RPC including those tested and not NDD and not repaired;in the lesser repaired; in '94 repaired;in '94 of the '94 Volts and A + 0.2

• B Voltage analysis volts analysis volts '94 reanalysis of'92 + 0.5 'C 0.0 0 0 0 0.0 0.1 0 0 0 0.0 0.2 0 0 0 0.0 0.3 0 0 1 0.5 0.4 5 0 26 18.0 0.5 16 0 52 42.0 0.6 47 0 74 84.0 0.7 64 0 88 108.0 0.8 80 0 90 125.0 0.9 84 0 84 126.0 1.0 83 0 69 117.5 1.1 71 10 41 93.5 1.2 7 45 29 30.5 1.3 11 40 37 37.5 1.4 2 27 16 15.4 1.5 5 33 19 21.I 1.6 8 11 7 13.7 1.7 4 11 5 8.7 1.8 1 6 5 4.7 1.9 2 5 1 3.5 2.0 0 4 0 0.8 2.1 0 2 0 0.4 2.2 0 2 5 2.9 2.3 0 2 1 0.9 2.4 0 0 0 0.0 2.5 0 2 1 0.9 2.6 0 1 0 0.2 2.9 0 0 1 0.5 4.5 0 0 1 0.5 Totals ! 490 201 1 653 856.7 SEC10TBL.XLS Table 10-3 10-7

Figure 10-1. J. M. Farley-1: Comparison of S/G-C Actual RPC Confirmed / Untested EOC-12 Voltages vs. Alternate Method Projected Voltages 90 80 - - - -

l .

70 - - - - -

i 60 - - - - - -- --- -- -

E 50 - - - -- _ - _

E 5

g 40 .- -

Y 30 - -

20 10 - - - _ - - - -

O s I . 1 0 0 Il n m mmm m 0.3 0.5 0.7 0.9 1.1 1.3 1.5 1.7 1.9 2.1 2.3 2.5 2.7 2.9 3.1 3.3

~

Volts m Actual EOC-12 O Projected EOC ! l ALA_SIMS.XLS Fig 10-1 10-8 8

J

?

I s

1 i

3 l

i i

l Figure 10-2. J. M. Farley-1: Comparison of S/G-C Actual EOC-12 All Indication f Voltages vs. Alternate Method Projected Voltages  ;

! 90 1 f

l I

80 - - - - - - - - - - - - -- - - -

4  !

l j 70 - - - - - - - - - - -- -- - -

i 4 60 - - - - - -- - --- --- -

! E 1 .2

! j 50 --

I -- - -

b 5

l g 40 - - - - - - - - - - - I i d i z 2 30 - --

d 20 - - - - -- --

\

10 -- - - - - - - - ---

0 EC a L= " " - -

j i

1 0.3 0.5 0.7 0.9 1.1 1.3 1.5 1.7 1.9 2.1 2.3 2.5 2.7 2.9 3.1 3.3 l

t

, ..- Volt.s_ . _-.

l  : 3 Actual EOC.12 OProjected EOC 12

}

i i

i 1

i, 1

}

t i

ALA_SIMS.XLS Fig 10-2 10-9 3

Figure 10-3. J. M. Farley-1: S/G-C Projected EOC-13 Voltage Distribution for Alternate Methodology 120 100 -- ---- -

80 --

E .

2 1 b .

5 60 --

1 3 l 4 '

40 20 - - - - -

0 i l Illin....

0 3 0 4 0.5 0 6 0.7 0 8 0 9 10 1.1 1.2 1.3 1.4 1.5 1.6 1.7 l 8 19 2.0 2.1 2.2 2.3 2.4 2.5 2 6 2.7 2.R 2 9 3 0 3.1 3.2 3.3 3 4 4.5 Volts ALA_SIMS.XLS Fig 10-3 10-10

l I

11.0 ASSESSMENTS OF NDE UNCERTAINTY FOR ANALYST VARIABILITY FOR EOC-Il INSPECTION <

' As noted in Section 6, NDE uncertainties for Farley-1 were developed in WCAP-12871 Rev. 2, and r have been updated in Reference 6. The EPRI NDE uncenainty, which is a one standard deviation value of 12.5%, has been applied to both Cycles 12 and 13 for Farley-l.

However, the reevaluation of the 1992 voltr.ges performed as part of the growth study in 1994 permits an additional review of analyst variability, since many of these same 1992 inspection indications were examined as part of the 1992 growth study. Table Il-1 provides a summary of voltage calls for the two analyses. There were 1075 indications from 1992 which are common to the '92 and '94 growth studies. 816 of these were greater than 0.75 volt per the 1994 growth study. The 1994 analysis voltages averaged only about 0.0002 volt lawer than the 1992 analysis for all indications, and 0.0040 volt higher for indications greater than 0.75 volt. Figure Il-1 shows a plot of the correlation of the voltages from the two analyses. The distribution is well behaved, with 517 of the '94 indications lower than in 1992, and 373 of the '94 indications higher than in '92 (185 were unchanged).

Figure Il-2 provides a histogram of the voltage differences between the 1992 indications which were analyzed in both the 1994 and 1992 growth rate analyses. The maximum difference in voltage between the 1992 and 1994 growth studies was 1.05 volts, from a 1994 analysis voltage result of 1.92 to a 1992 voltage analysis result of 0.87 volts, for SG-A tube R19C15 IH. This location was identified as a 0.87 volt PI in the 1992 field inspection, and was not RPC inspected in 1992. In the subsequent 1994 inspection, this indication was reported as a 1.83 volt PCN (PI confirmed but not pluggable), and was confirmed as a 0.51 MAI by RPC.  ;

The standard deviation of mean voltage difference between the '92 and '94 analyses is also shown in Table 11-1. For all indications, the standard deviation of 0.0888 volts is about 8.7% of the mean 1994 analysis voltage of 1.018 volts. For '94 analysis voltages greater than 0.75 volt, the standard deviation is 0.0975 volt, or 8.5% of the mean 1994 analysis voltage of 1.14.. Both of these values are conservative with respect to the volts.ge analyst uncertainty component of 10.3% (which is combined with probe wear uncertainty) utilized to obtain the overall NDE uncertainty of 12.5% described above l and in Section 6.0. These results represent an assessment of analyst variability at the 199.2 inspection.-

l It is expected that having the 1994 bobbin data available to reassess the 1992 data has resulted in the 1994 reevaluation being more accurate than the original 1992 reevaluation.

l l

l

?

l i

I 11-1 l

I i

i l

Table 11-1. Summary of Analyst Variability for EOC-11 Inspection Average Volts Standard Deviation Voltage No.of Mean of Voltage Difference Range Indications Voltage

'94 '92 Difference Analysis Analysis Volts Percent 1992 EOC-11 Inspection All Ind. 1075 1.0178 1.0180 -0.0002 0.0888 8.7%

l '94 Analysis 816 1.1425 1.1386 0.0040 0.0975 8.5 %

> 0.75v l

l l

l l

I l

l i

l l

l 11-2 I l \

i l

l l

Figure 11-1. Comparison of'92 and '94 EC Data Analysis Results for 1992 Bobbin Inspection Data 3.00 ,

o o

i 2.50 1 !o ]

0  ! l 0 l e o ,

o W l l 2.00 .

04o l o O te f:

l t o o 3 4 0 oo I >  ! o j l .m i y, 1.50 4

E o o o O

1. o n

& o o 0 l 1.00 O o

O-+ool

>o '

o l

@ o o o 0 ooo 0.50

@ o I hi l i

l t

0.00 0.00 0.50 1.00 1.50 2.00 2.50 3.00 l

1994 Analysis Voltage l

l l

I Fig 11-1 Chan 4 11-3

l Figure 11-2. Difference Between '92 and '94 Analysis of 1992 Indication Voltages 500 450 400 350 h 300 e

b

$ 250 C

o e 200 Z

150 . _ _ _ ,

1 I

l 100 l

50 I o - - .... i I.... _.

A 2 7e 0 4 4 2 8 8 8 5 N *, 0 7, 3 #  : 8 &

9 9 9 9 9 9 9 9 e o 6 o 6 o o e o o o 0

'94 Voltage Minus '92 Voltage l

l I

1 I

I

• Fig Il-1 Chart 5 11-4 j

12.0 REFERENCES

i

1. NRC Letter," Safety Evaluation by the Ofnce of Nuclear Reactor Regulation Related to l Amendment No.106 to Facility Operating License No. NPF-2; Southern Nuclear Operating  ;

Company, Inc.; Joseph M. Farley Nuclear Plant Unit 1; Docket No. 50-348)" dated April 5, l 1994.

I

2. NUREG-1477 (draft), " Voltage-Based Interim Plugging Criteria for Steam Generator Tubes - j Task Group Report," United States Nuclear Regulatory Commission (NRC), June 1,1993.

l

. 3. WCAP-12871 Rev. 2, "J. M. Farley Units I and 2, SG Tube Plugging Criteria for ODSCC at Tube Support Plates," Westinghouse Electric Corporation, February 1992.

i

4. NP-7480-L, Volume 1, Revision 1, " Steam Generator Tubing Outside Diameter Stress Corrosion Cracking at Tube Support Plates - Database for Alternate Repair Limits, Volume 1: 7/8 Inch Diameter Tubing," Electric Power Research Institute, December,1993. ,

1

3. EPRI Letter, " Exclusion of Data from Alternate Repair Criteria (ARC) Databases Associated with 7/8 inch Tubing Exhibiting ODSCC," D. A. Steininger (EPRI) to J. Strosnider (USNRC),

April 22,1994 [to become Appendix E of reference (2)].

6. TR-100407, Revision 1 (draft), "PWR Steam Generator Tube Repair Limita - Technical Support Document for Outside Diameter Stress Corrosion Crack at Tube Support Plates," Electric Power i Research Institute, August 1993.

7. [ United States Nuclear Regulatory Commission] Meeting with EPRI, NUMARC, " Resolution of Public Comments on Draft NUREG-1477," United States Nuclear Regulatory Commission, February 8,1994.

8. Regulatory Guide 1.121 (draft), " Bases for Plugging Degraded PWR Steam Generator Tubes,"

United States Nuclear Regulatory Commission, issued for comment in August,1976.

9. Press, S. James, " Bayesian Statistics," John Wiley & Sons (1989).

10. NSD-TAP-2113, "Farley-1 S/G EOC-12 SLB Leak Rate Projection in Support of IPC", T. A.

Pitterle to J. A. Knochel, dated Nov. 17,1992.

11. WCAP-13990,"Sequoyah Units 1 and 2 Steam Generator Tube Plugging Criteria for Indications l at Tube Support Plates", May 1994. I l

12. WCAP-13692, " Responses to NRC Questions on NRC Analytical Model for SLB Leakage of ODSCC at TSPs", April 1993.

12-1 l

1