ML20199E965

Rev 1 to ATI-98-012-T005, DPC Evaluation of McGuire Unit 1 Surveillance Weld Data Credibility

ML20199E965
Person / Time
Site:	McGuire
Issue date:	11/24/1998
From:	Griesbach T, Hardin T ATI CONSULTING
To:
Shared Package
ML20199E963	List: ML20199E956 ML20199E965
References
ATI-98-012-T005, ATI-98-012-T005-R01, ATI-98-12-T5, ATI-98-12-T5-R1, NUDOCS 9901210157
Download: ML20199E965 (17)

Text

-.

.... ~ _

• t.

lr.

,'s.

. r, t.

4 i

/

Duke Power Company l-Evaluation ofMcGuire Unit 1 l

l Surveillance Weld Data Credibility i

~

Technical Report No. ATI-98-012-T005 l

Revision 1 i

November 1998 l

I i

l Prepared by:

!e f.

/lf 2(fg j

Date:

Timothy J. GddbMh -

l Reviewed by:

"/ZM kI

]

O

.b M_ '-

Date:

_ Timothy C. Ha. din

\\

l l

l1 ATI Consulting 3860 Blackhawk Rd., Suite 160 Danville, CA 94506 i

9901210157 990107 l

PDR ADOCK 05000369 l

P PDR w

i l

r, Evaluation of McGuire 1 Surveillance Weld Data Credibility Introduction Five surveillance capsules have been removed from the McGuire 1 plant. Of these five capsules, only capsules U, X, V, and Y have Charpy test results for the surveillance weld material [1,2,3,4]. In the McGuire I surveillance capsule V report [2], the embrittlement (i.e., ART values) of the vessel beltline welds 2-442 A,B&C (heat number 20291/12008) was predicted using three surveillance capsule weld (heat number 20291/12008) results from McGuire 1. The Westinghouse calculation of ART values for determination of the P-T limit curves showed that the limiting vessel material is the lower shell longitudinal welds 3-442 A&C (weld heat 21935/12008), and this formed the basis for the current (16 EFPY)

P-T limit curves.

Subsequently, in the McGuire 1 surveillance report for Capsule Y [4], a new evaluation of the surveillance weld data was performed by Westinghouse using Position C.2 of Regulatory Guide 1.99, Revision 2 [5]. A best-fit to the four McGuire I weld data points was performed, and the credibility criteria were applied. The scatter in the surveillance weld data was found to exceed 28 F for one of the data points, and the credibility of the McGuire I surveillance weld data was questioned.

ATI Consulting has performed a reassessment of the surveillance weld data credibility. This reevaluation considers additionalinformation for evaluating weld heat 20291/12008, including data from multiple sources and recent guidance from NRC for surveillance data evaluation. The NRC guidelines for assessing credibility provided at the November 17,1997 and February 12,1998 meetings on RPV Integrity Assessment [6,7] were applied to assess the credibility of the McGuire 1 surveillance weld data. As a result, it is determined that the McGuire 1 surveillance data meets the credibility criteria and, thus, the basis for the current P-T limit curves for McGuire 1 [2] continues to be valid.

McGuire 1 Surveillance Weld Data Credibility i

In the February 12,19981ndustry/NRC Workshop on RPV Integrity Assessment, topics related to surveillance data credibility were discussed in depth. The NRC staff presented a number ofpreferred approaches when assessing weld chemistry data and surveillance data results [7]. These approaches incorporate methods to address weld chemistry variability and the handling of data from multiple sources. For example, the staff has 4

concluded that the CEOG analysis [8] provides an accurate determination of RPV weld best-estimate chemistries for a specific weld heat. In considering surveillance data, the staff recommends that the chemistry should be based on the chanistry data for that specific weld rather than the heat best-estimate chemistry. Also, examples were provided to correct for chemical composition and irradiation environment, to provide guidance with appropriate normalizing parameters for surveillance data when assessing credibility, and to determine best-fit chemistry factor based on all available data. At the February Workshop, NRC focused on several issues regarding use of surveillance data, including:

ATI Consulting 11/24/98

t.

.f, McGuire i Surveillance Weld Credibility Study Using a CF determined from non-credible surveillance data Correcting for chemical composition (ratio procedure)

Correcting for irradiation environment (temperature)

Appropriate chemical composition for multiple surveillance capsules from a single source Appropriate normalizing parameters for surveillance data when assessing credibility and determining CF.

Per the NRC guidelines, the surveillance data credibility evaluation should consider other sources of data [7]. Data for weld heat 20291/12008 is available from other sources including the CEOG weld chemistry database [8], the Pilgrim surveillance program [9],

and the Eason (E900) database [10]. The Westinghouse Capsule Y repon [4] calculated the scatterin the surveillance data results to be 38.32 F for one datapoint from a best fit to the McGuire 1 data. This ATI Consulting reevaluation considers the other sources of data for this weld heat, and includes information that was not considered previously by Westinghouse. The study was performed to demonstrate how small changes in data input affect the credibility of the weld data. In panicular, the scatter can be reduced by using the ratio procedure to adjust for chemistry variability and by using the other information related to this weld heat.

Three specific examples are used to evaluate the parameters affecting scatter and data credibility. The examples applied to the McGuire 1 surveillance weld data follow NRC Example Case 4 (Surveillance Data from Plant and Other Sources) and Case 5 (Surveillance Data from Other Sources Only). The examples are as follow:

Example 1: McGuire 1 Surveillance Data With Adjustments

1) Correct for Chemical Composition Weld Heat No. 20291/12008 is contained in both the McGuire 1 and Pilgrim surveillance capsules. Chemistry differences have been observed for the individual capsule welds. All of the chemistry data for this weld heat (from the CEOG Report [8]) is shown in Table 1. This includes data for weld heat 20291/12008 from a weld chemistry variability study performed for Baltimore Gas & Electric [11]. Copper and nickel measurements from the through-thickness weld variability study [11) are shown in Figures 1 and 2. This study concluded that weld heat 20291/12008 is inhomogeneous. The inhomogeneity is attributed to spool-to-spool variation in chemical composition, with each spool of weld wire having consistent composition.

This information is used to better understand the McGuire I surveillance data and to apply the surveillance results to the vessel. It is assumed that each group of specimens for a single capsule are from one spool, but that different spools could have been used for different capsules. The weld in the vessel is represented by the average (i.e., mean) behavior based on many different spools being used to fabricate the vessel.

ATI Consulting 11/24/98

~.

'~~

P, McGuire i Surveillance ll' eld Credibilitv St@

l Weld Wire 20291/12008 Weld Block V6 l

l 3,500 3,000 mCD cn 2,500 E

p 2,000 U

\\

I

,,', 1,500 E

E 1,000 c

c c

c o

e 500 ID T/4 T/2 3T/4 OD BG Location Center Region Fusion Be,ad Region HAZ Region 0

0 (from Reference 1I)

Figure 1. Through-Wall Variability in Copper from Sample of Weld Heat 20291/12008 Weld' Wire 20291/12008 Weld Block V6 10,000

^ 8,000 tn E

6-6,000 C

O O

O O

O 2

E 4,000 m

Em G 2,000 0

ID T/4 T/2 3T/4 OD BG Location Center Region Fusion Be,ad Region HAZ Region C~

0 (from Reference 1I)

Figure 2. Through-Wall Variability in Nickel from Sample of Weld Heat 20291/12008 ATI Consulting 11/24/98

.~,.

t~

McGuire i Surveillance Weld Credibiliev Studv Given these observed chemistries for this weld heat, the chemistry adjustment to be applied per Regulatory Guide 1.99, Rev 2, Position 2.1 is as follows:

"if there is clear evidence that the copper or nickel content of the surveillance weld differs from that of the vessel weld, i.e., differs from the average for the weld wire heat number associated with the vessel weld and the surveillance weld, the measured values of ARTer should be adjusted by multiplying them by the ratio of the chemistry factor for the vessel weld to that for the surveillance weld" 3

' Table CFvescha Ratio Adjusted ARTmT =

• Measured ARTmr Table CF,,cs, s The best estimate chemistry for this heat in the vessel (from the.CEOG report [8]) is:

Cu = 0.199 wt%, Ni = 0.846 wt%, CF = 200.37 'F.

This adjustment is applied to the individual surveillance capsule results to obtain:

Plant Cu "

Ni "

Chemistry Fluence Measured Adjusted 2

{ Capsule)

(wt%)

(wt%)

Factor, CF (n/cm )

ARTer (*F)* ARTmr (*F)

Pilgrim (1}

0.16 0.79 176.55 2.37x10" 28.00 31.78 McGuire 1 {U}

0.20 0.91 209.95 4.45x10

161.32 153.96 McGuire 1 {X}

0.191 0.848 198.26 1.42x 10

170.69 172.51 McGuire 1 {V}

0.191-0.848 198.26 1.94x10

180.15 182.07 McGuire 1 {Y}

0.191 0.848 198.26 2.93x10

190.42 192.45

• measured ARTer based on TANH curve-fits to Charpy data

" measured / estimated chemistries for individual capsule weld materials Note the differences in chemistry between the Pilgrim surveillance weld, and the weld material in McGuire Capsule U and Capsules X, V, and Y. (The Pilgrim surveillance chpsule results are included for comparison with the McGuire 1 data). The measured chemistry for the Capsule U weld (Cu = 0.20 wt%, Ni = 0.91 wt%) is clearly higher than the other capsules.

These differences have been retained in the credibility assessment to normalize the surveillance data and to establish the embrittlement behavior of the vessel welds. The chemistry for the weld in McGuire Unit 1 Capsules X, V, and Y are assumed to be same (i.e., from the measured chemistry of Capsule V specimen DW-24) because of the similar embrittlement behavior for these three capsules. This effect is apparent in the plot of RTer hift vs. fluence s

shown in Figure 3. The variability in chemistry is one reason for the surveillance data scatter.

Calculation of surveillance CF for best-fit to McGuire I surveillance data only is shown in Table 2. In this calculation, the chemistry adjustment calculated above is used, and capsule specific chemistries are included. However, no temperature adjustment is needed because of the narrow band ofirradiation temperatures for McGuire 1 plant operation.

ATI Consulting 11/24/98 j

'r, McGuire i Surveillance Weld Credibilh> Studv RTNDTShift vs. Fluence l

300

~

Meancr + 28 V

- Mean Clr = 158 7 Y (from [4])

200 -

Cap.X gg.

s

/

)

/

." 100 -

a McGuire i Surveillance Weld Data (wittui squalmerts)

Ref.: WCAP 14993 [4]

l 0

\\

0 1

2 3

4 I8 2

Fluence (x10 n/cm )

Figure 3. RTmr Shift vs. Fluence for McGuire 1 Surveillance Weld Using the chemistry adjustment, Table 2 shows the scatter (i.e., (Adjusted - Predicted ARTm7)) for the low fluence (Capsule U) datapoint is 30.9 F, which is slightly greater than the 28 *F value required to meet the credibility criteria. However, other data exists that can improve the RTmr hift prediction and reduce the scatter.

s

2) Correct for RTm1 Shift The best-fit to surveillance results is sensitive to the Charpy curve-fit data input. For example, the measured ARTer results for capsules U and X from the Eason (E900) database [10] are slightly different than the values calculated by Westinghouse, as shown below. The Eason results are used with chemistry factor adjustment to evaluate j

the McGuire I weld data credibility. The Eason method uses a higher order curve-fitting techniques that is more accurate than the TANH method. The Eason technique j

is the currently accepted curve-fit method for use in developing improved embrittlement trend curves [10]. (Note: only two McGuire I capsule results were availt.ble in the Eason database; Capsule V and Y results have not yet been included in the database).

Capsule Fluence Measured Measured Adjusted 2

(n/cm )

ARTer ( F)

ARTer ( F)

ARTmr ( F)

(West.)ldl (Eason)l' 3 (Eason)

U

.445x10'S 161.32 157 149.84 X

1.42x10

170.69 167 168.78 l

ATI Consulting 11/24/98

.'c

.i AlcGuire I surveillance Weld Credibility Study These adjusted ARTer values are used in the best-fit calculation and the results are given in Table 3. As shown in this table, a revised CF of 157.39 *F is determined, and the scatter is less than 28 F. This shows that with adjustment to the data for chemistry variabiiity, and using improved curve-fit techniques, the scatter in the McGuire 1 surveillance weld data is l

reduced and the data is shown to be credible.

Example 2: Consider Both McGuire 1 and Pilgrim Surveillance Weld Data One additional RTer shift value for weld heat 20291/12008 is available from the Pilgrim l

surveillance program [9). Although this is a low fluence point irradiated in a BWR vessel, the data is combined with the McGuire 1 data in this example to further validate the embrittlement trend for this weld heat. Table 4 shows the best-fit calculation for the McGuire 1 surveillance weld including the Pilgrim data. The ratio adjusted chemistry is l

l used to determine the adjusted ARTm1, but no temperature adjustment is applied (see I

explanation in Example 3). For this case, the scatter is again less than 28 F, therefore the data is credible.

i l

A plot of the McGuire I and Pilgrim surveillance weld data results is given in Figure 4.

RTNDT Shift vs. Fluence 300

~ - Mean ctrve + 28 Y

1. , s ~~~~

Mean CF= 157.46 *F

/

t,.,

5

/

'O 150 -

e

>-" 100 -

Y mci and PL Surveillance Weld Data (with adjustrnetts) 50 -

'P O

O 1

2 3

4 N

2 Fluenee (x10 n/cm )

Figure 4. RTer vs. Fluence for McGuire 1 and Pilgrim Surveillance Weld Data It can be observed that four of the five datapoints are within 10 F of the mean curve, and one point (Capsule U) lies close to the mean + 28 *F curve. The McGuire I surveillance weld data arejudged to be credible (i.e., scatter is less than 28 F) using either Example 1 or 2.

ATI Consulting

-6' 11/24/98

~

. ~ ~

McGuire i Surveillance Weld Credibility Studv Example 3: Consider Both McGuire 1 and Pilgrim Survei!!ance Weld Data with Mean Chemistries for Surveillance Materials In addition to the above evaluations, the credibility of McGuire Unit I surveillance weld data was examined by the method ofNRC example Case 5 [6]. The measured ARTwor data from all sources are adjusted to the mean chemical composition of the surveiPance capsules using Mean Cu = 0.191 wt%

Mean Ki = 0.867 wt%

CFuo, = 201.02 *F No temperature adjustment is needed since the irradiation temperatures for the McGuire 1 data are within a narrow range (i.e.,558 - 560 *F). Temperature correction (i.e.,1 F/ F) is not appropriate for the Pilgrim BWR data since the absolute shift values are low in the BWR 2

. fluence range (below 10" n/cm ), and temperature correction is intended for data in the 2

PWR (10"- 10" n/cm ) fluence range. The adjusted values of ARTwor are shown below, and are calculated from the equation

' Table CF"""

Adjusted ARTxor =

Measured ARTxor Table CFe,,i, j Plant Cu Ni CForoi.

Measured Adjusted

{ Capsule}

(wt%)

(wt%)

(*F)

ARTxor ARTwor (F)

( F)

Pilgrim { 1 }

.I63*

.841*

184.23 28.0 30.55 McGuire 1 {U}

.198 "

.874 "

204.13 157 154.61 McGuire 1 {X}

.198 "

.874**

204.13 167 164.46 McGuire 1 {V)

.198**

.874**

204.13 180.15 177.41 McGuire 1 {Y}

.198 "

.874**

204.13 190.42 187.52

• best estimate chemistry for Pilgrim surveillance weld data from CEOG database [8]

" best estimate chemistry for McGuire I surveillance weld data from CEOG database [8]

These values of Adjusted ARTwor are applied to the CF/ credibility calculation in Table 5.

From these results CF = 154.77 *F; 4 out of 5 datapoints are credible (scatter less than 28 *F), and one point is 34.72 *F from the predicted ARTwor. Therefore the data set should be considered credible.

The McGuire 1 surveillance weld data arejudged to be credible using the method of Example 1,2, or 3. Because the surveillance weld data is credible, a reduced margin term (c = 14 *F) can be used for determining embrittlement in the vessel.

a ATI Consulting 11/24/98

l.

?t:

McGuire i Surveillance It' eld Credibilitv Studv I

Effect of Surveillance Credibility Evaluation on Current McGuire Unit 1 P-T Curves The weld surveillance data credibility evaluation discussed above yields a chemistry factor for welds 2-442A, B, & C slightly different from the CF used in the generation of the 16 EFPY P-T limit curves [2]. The Adjusted Reference Temperature (ART) for the McGuire 1 vessel welds 2-442A,B&C is determined using the surveillance data results from Example 2 2

with a CF = 157.46 F (from Table 4). A 1/4-t fluence value of 0.4348x10 n/cm was determined for weld seams 2-442B&C from the Westinghouse calculation of P-T limits at l

16 EFPY [2]. The revised ART value for weld seams 2-442B&C is calculated from the equation ART = RTworm + ARTuor + Margin i

where, RTworm = -50 'F 2

1/4-t Fluence = 0.4348x10 n/cm (from Reference 2)

FF = 0.768 CF = 157.46 F Margin = 28 F ART = -50 + (157.46

• 0.768) + 28 *F j

ART = -50 + 120.9 + 28 = 98.9 'F 1

A similar comparison for weld seam 2-442A computes a revised ART value for this weld seam:

1 l

ART = RTuorm + ARTsor + Margin

where, RTworp = -50 "F 2

1/4-t Fluence = 0.3971x10 n/cm (from Reference 2) l FF = 0.744 l

CF = 157.46 *F Margin = 28 F l

ART = -50 + (157.46

• 0.744) + 28 *F ART = -50 + 117.2 + 28 = 95.2 *F This results are compared to the calculated ART values [2] at the 1/4-t location used to determine the existing (i.e.,16 EFPY) P-T limit curves for McGuire 1:

l ATI Consulting 11/24/98 l

l

e

.r,

McGuire i Surveillance Weld Credibility Study Note: Data Reproduced from Table 8-4 in WCAP-13943 (21 for McGuire i Vessel at 16 EFPY Component 1/41 Fluence-FF CF IRTer ARTi:or Margin ART x10*

(*F)

(*F)

(*F)

(*F)

@ 1/41 (n/cm')

(*F)

Int. Shell Plate, B5012-1

.6008)

.857 61(4 34 52.4 17 103.4 Int. Shell Plate, B5012-2

.6008

.857 100.3 (d) 0 85.9 34 119.9 Int. Shell Plate, B5012-3

.6008 )

.857 74.9 (d)

-13 64.2 34 85.2 LwT. Shell Plate, B5013-1

.6008

.857 99.1 (d) 0 85.0 34 119.0 LwT. Shell Plate, B5013-2

.6008 )

.857 65(d) 30

$ 5.7 34 119.7 LwT. Shell Plate, B5013-3

.6008

.857 65(d) 15 55.7 34 104.7 Int. Long. Welds,2-442A

.3971

.744 157.7(4

-50 117.6 28 95.6 Int. Long. Welds,2-442B&C

.4348*'

.768 157.7 (*)

-50 121.4 28 99.4

'LwT. Long. Welds,3-442B

.3971

.744 223. l (*)

-50 166.0 28 144.0 Lwr. Long. Welds,3-442 A&C

.4348*)

.768 223.1(4

-50 171.5 28 149.5 (d j

Cire. Weld,9-442

.6008

.857 39.8 (d8

-70 34.1 34.1

-.8 j

' Fluence, f, based on f,a ( 10 n/cm', E > 1 MeV) = 1.008 at 16 EFPY for the intermediate and lower shell pl.Stes. The McGuire 1 vessel thickness = 8.625 inches. Thus, fu. = f s

• cxp(. 24*(8.625/4)).

NFluence, f, based on f,a ( 10 n/cm' E > 1 MeV) = 0.7295 at 16 EFPY for the int. shcIl long. welds 2-442B&C and lower shcIl long welds 3-442A&C.

(O Fitted chemistry factors based on surveillance data results [2].

(d) Chemistry factors based on tables from Reg. Guide 1.99, Rev. 2 [5].

p (OLimiting vessel beltline material for McGuire I cooldown P-T limit curves at 16 EFPY (%t location).

Review of this table yields the following observations:

1) the new ART (98.9 F) for welds 2-442B&C is slightly less than the calculated ART value (99.4 'F) for these weld seams in the P-T curve analysis [2]. Also, the new ART for weld 2-442A (95.2 F)is less than the calculated ART value (95.6 F) for this weld seam in the P-T curve analysis.

2) the new ARTS for each of these weld seams are still significantly lower than the ART value (149.5 F) for weld seams 3-442A&C, which continue to be the limiting vessel beltline material for the McGuire 1 vessel. Therefore, the validity of the existing (i.e.,

16 EFPY) P-T limit curves for McGuire 1 is not affected by this evaluation.

Summary This evaluation demonstrates that the McGuire I surveillance weld data are credible using a number of methods recommended by the NRC for surveillance data evaluation. Data from all available sources have been considered, and appropriate adjustments made to account for chemistry variability. The current McGuire Unit 1 P-T limit curves [2] were based in part on the credibility of the surveillance weld data, and this study confirms that credibility.

ATI Consulting 11/24/98

7..

i-McGuire i Surveillance Il' eld Credibility Studv 1

References 1.

" Analysis of Capsule U from the Duke Power Company McGuire 1 Reactor Vessel Radiation Surveillance Program," WCAP-10786, February 1985.

{

2.

" Analysis ofCapsule V Specimens and Dosimeters and Analysis of Capsule Z

)

Dosimeters from the Duke Power Company McGuire Unit 1 Reactor Vessel i

Radiation Suiveillance Program," WCAP-13949, February 1994.

3.

" Analysis of Capsule X from the Duke Power Company McGuire 1 Reactor Vessel radiation Surveillance Program," WCAP-12354, August 1989.

4.

" Analysis of Capsule Y from the Duke Power Company McGuire Unit 1 Reactor j

Vessel Radiation Surveillance Program," WCAP-14993, February 1998.

5.

Regulatory Guide 1.99, Revision 2," Radiation Embrittlement of Reactor Vessel Materials," USNRC, May 1988.

6.

Minutes from Industry /NRC Meeting on Generic Letter 92-01, Rockville, MD, November 17,1997.

7.

Minutes from Industry /NRC Workshop on RPV Integrity issues, Rockville, MD, February 12,1998.

8.

"Best Estimate Copper and Nickel Values in CE Fabricated Reactor Vessel Welds,"

ABB Combustion Engineering, Report No. CE NPSD-1039 Revision 02, June 1997.

9.

Norris, E. B.," Pilgrim Nuclear Power Station Unit 1 Reactor Vessel Irradiation Surveillance Program," Final Report, SwRI Project No. 02-5951, July 1981.

j I

10. Eason, E. D., Wright, J. E., and Odette, G. R., " Improved Embrittlement j

Correlations for Reactor Pressure Vessel Steels, NUREG/CR-6551, Draft Report, May 1997.

11. Fyfitch, S., and Pegram, J. W.," Reactor Vessel Weld Metal Chemical Composition Variability Study," B&W Nuclea: Technologies BAW-2220, June 1995.

l ATI Consulting 11/24/98

. ~

. - ~ ~

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

-.~__.m.-.-

.7

.s,.

1' McGuire i Surve;llance Weld Credibilltv Study l-

~

Table 1. Measured 'Cu and Ni Chemistries for Weld Heat Number 20291/12008 i

tL7 Group Tag Analysis y

i.-

Cu

- Ni Avg.Cu Avg. NI

1. of l

wt%

wt%.

CE PEDIGREE,

DATA Group wt%

wt%

Coils SOURCE Tag f.

. 0.16 0.77

' VALID SWRI-02-5951 a

0.1630 0.8410 1

i 0.16 0.77

- VALID -

SWRI-0215951 a

' O.17 0.78 VALID SWRI-02-5951 a

l 0.16

-0.79 VALID SWRI-02-5951 a

0.16 0.79 VALID SWRI-02-5951 a

0.17 0.8 VALID SWRI-02-5951 a

l l

0.14 0.81 VALID SWRI.02-5951 a

~O 0.16 0.81 VALID SWRI-02 5951 a

0.16 0.81 VALID SWRI-02-5951 a

0.17 0.81 VALID SWRI 02-5951

'a 0.13 0.72 VALID BAW-2220 a

O.13 0.74 o

VALID.

BAW-2220 a

0.14 0.74 INDETERMINATE NI BAW 2220 a-0.17 0.75 INDETERMINATE NI BAW-2220 a-0.16 0.79 INDETERMINATE NI BAW-2220 a

O.18 0.93 VALID BAW-2220 a

0.16 0.94 VALID BAW-2220 a

0.17 0.95 VALID BAW-2220 a

0.16 0.96 VALID BAW-2220 a

0.17 0.97 VALID BAW-2220 a

0.17 0.98 VALID BAW-2220 a

l l '..

0.19 0.83 VALID BAW-2220 a

j 0.19 0.86 VALID BAW 2220 a

! 0.1978 l 0.8742 l 1

.l l< '

{:

0.2 0.91 VALID WCAP-10786 b

0.21-0.88 VALID D24117 b

0.195

' O.87 -

VALID WCAP-13949 b

0.191 0.848 VALID WCAP-13949 b

0.193 0.863 VALID WCAP-13949 b

0.2370 l 0.8226 l 1

l 0.15 0.69' INDETERMINATE NI BAW 2220 c

l 0.16 0.69 INDETERMINATE NI BAW-2220 e

l 0.18 0.71 INDETERMINATE NI BAW-2220 c

0.19 0.71-INDETERMINATE NI BAW-2220 c

0.22 0.71 INDETERMINATE NI BAW-2220 c

0.17 0.73 VALID BAW-2220 c

0.17 0.73 VALID BAW-2220 c

1 l

0.17 0.73' VALID BAW-2220 e

0.3 0.73 VALID BAW-2220 c

0.3 0.73 VALID BAW-2220 c

0.33-0.73 VALID BAW-2220

'c j

0.33 0.74 VALID DAW-2220 c

L 0.27 0.78 VALID BAW-2220 e

l' O.27 0.8 VALID BAW-2220 c

0.23 0.84 VALID BAW-2220 c

I I

11 -

11/24/98 ATI Consulting

-.9 y

L p,.

A fcGuire i Surveillance it' eld Credibilitv Study 0.23 0.85

' VALID BAW-2220 c

0.23 0.86 VALID BAW-2220 c

0.23 0.86 VALID BAW-2220 c

0.24 0.86 VALID BAW-2220 c

0.25 0.86 VALID BAW-2220 c

{

0.2 0.87 VALID BAW-2220 c

i

- 0.2 0.87 VALID BAW-2220 e

0.23 0.87 VALID BAW-2220 c

0.23 0.87 VALID BAW-2220 c

0.24 0.87 VALID BAW-2220 c

0.19 0.88 VALID BAW-2220 c

0.26 0.9 VALID BAW-2220 c

0.18 0.96 VALID BAW-2220 c

Best Estimate Weighted Mean Vessel Chemistry [8]:

Cu Ni 0.199 0.846 i

ATI Consuhing 11/24/98 i

. [.f uccurre ismener wetacreatestade,

g Table 2 Chemistry Factor (CF) for Best Fit to McGuire 1 Surveillance Weld Data (Weld Heat Number 20291/12008)

Capsule Tw.

Fluence, f FF FF' Adjusted FF*ARTm Predicted

. Adjusted -

(*F)

(10" n/cm )

ARTm ( F) -

ARTm( F)

Predicted ('F) 2 U

560 0.4447 0.775 0.6000 153.96 119.26 123.1

.30.9 X

560 1.424

.l.098 1.2057 172.51 189.42 174.5

-2.0.

V 559 1.94 1.181 1.3951 182.07 215.05 187.7

-5.6 L

Y 558 2.933 1.285 1.6520 192.45 247.35 204.2

-11.8 Sum:

4.85 771.08 where Predicted ARTm ( F) = FF x CF, pp = g(ols-0.1*In 0, and 2

CF = E (FF*ARTm ) + I (FF ) = 771.08 + 4.85 = 158.89 'F ATI Consulting

-13 11/24/98

McGuire i Surveillance Weld CredibilityStudy

,. l Table 3

~

Chemistry Factor (CF) for Best Fit to McGuire 1 Surveillance Weld Data Using Eason [10] Curve-Fits

- (Weld Heat Number 20291/12008).

Capsule Fluence, f FF FF Adjusted FF*ARTmr Predicted Adjusted -

2 (10" n/cm )

ARTmr(*F)

ARTer (*F)

Predicted (*F) 2 U

0.4447 0.775 0.6000 149.84' i16.06 121.9 27.92 X

1.424 1.098 1.2057 168.78*

185.33 172.8

-4.05 V

1.94 1.181 1.3951 182.07 215.05 185.9

-3.83 Y

2.933 1.285 1.6520 192.45 247.35 202.3

-9.85 Sum:

4.85 763.79

• Obtained from Eason curve-fits to data (see text for detail) where Predicted ARTer (*F) = FF x CF, pp = f(0.28-0.I' log 0 and 2

CF = I(FF*ARTer) + I(FF ) = 763.79 + 4.85 = 157.39 F ATI Consulting 14 I1/24/98

~

m McGuire i Strn eillance Weld Credibilitv Stredy Table 4 Chemistry Factor (CF) for Best Fit to All Surveillance Weld Data (McGuire I and Pilgrim)

(Weld Heat Number 20291/12008)

Plant Fluence, f FF FF' Adjusted FF* ART m Predicted Adjusted -

(Capsule)

(10" n/cm )

ARTm ( F)

ARTm ( F)

Predicted ( F) 2 PL {l}-

0.0237 0.1909 0.0364 31.78 6.07 30.06 1.72 MCI (U}

0.4447 0.775 0.6000 149.84' 116.06 121.97 27.87 MCI (X}

1.424 1.098 1.2057 168.78' 185.33 172.9

-4.12 MCI {V}

1.94 1.181 1.3951 182.07 215.05 185.98

-3.91 MCI (Y}

2.933 1.285 1.6520 192.45 247.35 202.38

-9.94 Sum:

4.889 769.86

• Obtained from Eason curve-fits to data where Predicted ARTm ( F) = FF x CF,

$28 - al* log 0 FF = f

,and 2

CF = I(FF*ARTm) I(FF ) = 759.86 + 4.889 = 157.46 F ATI Consulting 15 11/24/98

i' uccurre i samittmee werd cndibitiresadv j

i Table 5 Chemistry Factor (CF) for Best Fit to McGuire 1 Surveillance Weld Data By NRC Example Case 5 Method (Weld Heat Number 20291/12008)

Plant Fluence, f FF FF Adjusted FF'ARTmn Predicted Adjusted -

(Capsule)

(10" n/cm')

ARTwyr (*F)

ARTmyr( F) '

Predicted ('F)

PL {l}

0.0237 0.1909 0.0364 30.55 5.83 29.54 1.01 MC1 (U}

0.4447 0.775 0.6000 154.6I i19.76 I19.89 34.72 MCI (X) 1.424 1.098 1.2057 164.46 180.58 169.95

-5.50 MCI (V}

1.94 1.181 1.3951 177.41 209.54

.182.81

-5.40 MCI (Y}

2.933 1.285 1.6520 187.52 241.02 198.93

-11.41 Sum:

4.889 756.73 1

where Predicted ARTmyr ( F) = FF x CF, pp, f(0.28-0.l* log f) and CF = I(FF*ARTmn ) + E (FF') = 756.73 + 4.889 = 154.77 F t

i I

16 I1/24/98 ATI Consulting

_