ML18038A738

Upper Shelf Energy Uncertainty Analysis for Nine Mile Point Unit 1 Beltline Welds Final Rept

ML18038A738
Person / Time
Site:	Nine Mile Point
Issue date:	03/17/1993
From:	NIAGARA MOHAWK POWER CORP.
To:
Shared Package
ML17056C320	List: ML17056C319 ML18038A738
References
MPM-USE-393215, NUDOCS 9303260006
Download: ML18038A738 (90)

Text

NMPC Project 03 9425 MPM-USE-393215 FINAL REPORT.

entitled UPPER SHELF ENERGY UNCERTAINTYANALYSIS FOR I

NINE MILE POINT UNIT I BELTLINE WELDS MPM Research dc Consuming l5 +4 ii]j. rT H1$+4 WTt fh pt, II 8SVNR7 ~ AKK58 mxiuvr mmmm muavxcev March 17, 1993 9303260006 9303i9 PDR ADQCK 05000220 P

PDR

4 e

k ql I

0 I

Table of Contents 1.0 Introduction .......

1.1 Weld Metal Description .. ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ 4 2.0 Yield Strength Model 2.1 2.2 Theoretical Basis .......

Description.................................

Review of Earlier Application to NMP-1 Beltline Welds ...........

9 9

10 3.0 Uncertainty Estimation........................ 16 3.1 Database Analysis 16 3.1.1 Analysis of Outliers 17 3 .1.2 Summary .............,......... 18 3.2 USE Estimation Uncertainty .............. 19 3.3 Yield Strength Model Update.............. 19 3,4 Application of Updated Model to NMP-1 Welds 20 3.5 Tensile/USE Measurement Uncertainty ...... 20 3.5.1 Tensile Uncertainty Analysis ... 20

~

3.5.2 USE Uncertainty 21 3 .5.3 Summary ....................... 22 4.0 Summary and Conclusions ................ 44 5.0 References .................... 45

A l'

4,

1.0 Introduction Nuclear reactor pressure vessel materials must be tested and evaluated to ensure that they are safe in terms of both brittle and ductile fracture under normal operation and during design basis transients. With regard to ductile fracture protection, Appendix G to 10 CFR 50 prescribes a screening criterion of 50 ft-lbs. If any beltline materials are expected to exhibit Charpy Upper Shelf Energy (USE) (T-L orientation for plates) levels below 50 ft-lbs, then additional analyses must be performed to ensure continued safe operation. The NMP-1 beltline materials were evaluated to determine whether any materials would fall below the 50 ft-Ib screening criterion. The results of these evaluations are summarized in Reference [MA93a] and were presented in the response to NRC Generic Letter 92-01. As a result of these evaluations, NMPC concluded that none of the beltline welds would fall below the 50 ft-Ib screening criterion prior to end-of-license (EOL) and that an Appendix X analysis must be performed for beltline plates G-8-1 and G-307-4. The results of the Appendix X analysis for Service Level A and B loadings were reported in Reference [MA93a], and the results of the Service Level C and D loading analysis were reported in Reference [MA93b].

Full Charpy curves for the NMP-1 beltline weld materials in the unirradiated condition are not available. Therefore, the Reference [OD86] yield strength model (as described in Section 2.0) was used to estimate the unirradiated USE levels for the NMP-1 beltline welds. During a presentation to the NRC on September 30, 1992, concerning the low USE issue resolution, the NRC requested that NMPC accurately characterize the uncertainty in. the yield strength model to ensure that the level of conservatism used to calculate the unirradiated USE for the beltline welds is sufficient. The NRC request was later formalized in the Reference [NRC92] letter. In response to the NRC's request, NMPC has prepared this report which accurately characterizes the yield strength model uncertainty. Based on this analysis,-NMPC has concluded that the previously estimated USE values for the beltline welds are conservative and that none of the beltline welds will fall below the 50 ft-Ib screening criterion prior to EOL.

1.1 Weld Metal Description The chemical composition and mechanical properties of the NMP-1 welds are reported in References [ST84], [MA91], and [MA92a]. Based on the Reference

[ABB92] letter, NMPC now believes that only one wire heat/flux lot combination was used in welds 2-564-A/C. The updated reactor vessel beltline weld information is given in Tables 1-1 through 1-3. Table 1-4 summarizes the surveillance weld Charpy data.

0 TABLE 1-1 REACTOR VESSEL BELTLINE WELD INFORMATION Weld Seam Weld Wire Type Weld Flux Type Detailed Weld Number Location and Heat No. and Lot No. Procedure 2-564 A/C Lower-Intermediate RACO 3/86054 Arcos B-5/4ES F SAA-33-A(3)

Shell Longitudinal E8018/HAD D N/A MA-33-A(7)

E8018/JBG D N/A MA-33-A(7) 2-564 D/F Lower Shell RACO 3/86054 Arcos B-5/4ESF SAA-33-A(3)

Longitudinal Seams E8018/HAG D N/A MA-33-A(7)

E8018/JBG D N/A MA-33-A(7) 3-564 Lower Intermediate to RACO 3/1248 Arcos B-5/4M2F SAA-33-A(3)

Lower Shell Girth E8018/DBDE N/A MA-33-A(7)

E8018/IOG E N/A MA-33-A(7)

Surveillance All Three Capsules RACO 3/W5214 Arcos B-5/SG13F SAA-33-A(3)

Capsule Weld References [CE90] and [ABB92]

0 J Y i

TABLE 1-2 NMP-1 Beltline and Surveillance Weld Best Estimate Chemistry CHEMICAL COMPOSITION (WT.%)

IDENTIFICATION Cu Ni Weld Seam 2-564 0.22 0.2 0.015 0.020 A/C'eld Seam 2-564 0.22 0.2 0.015 0.020 D/F'eld Seam 0.22 0.2 0.015 0.020 WELD':

3-564'URVEILLANCE 0.18 0,07 0.022 0.013'

'E Average of Battelle/Westinghouse data [MA91]

recommendation [CE90]

Table 1-3 NMP-1 Beltline Weld Tensile Data'Te Yield Strength Ultimate Tensile Reduction Heat No. Flux Lot No. Stren th si Elon ation % A RACO3 86054 4ESF 75/00 90,000 27.5 69.9 RAC 03 -1248 4M2F 63,000 80,000 27.5 64.3 Surveillance Capsule Weld Unirradiated W5214 5613F 65,000 84,000 27.5 67.0 Surveillance Capsule Weld Irradiated'5214 5G13F- 73,680 90,240 23.24 68.1

'ata taken from [LE64], [ST84], and [CE90]. The unirradiated material test records do not indicate the test temperature. Therefore, RT is assumed.

'00-degree capsule fluence = 4.78 x 10" n/cm'; RT properties

'n 2 inches "in 1inch

Table 1-4 Summary of Charpy Impact Properties for Irradiated Weld Metal .

from the NMP-1 300-Degree Surveillance Capsule 30 ft-lb 50 ft-lb 35-Mil Lateral

'aterial E>1.0 MeV Transition Transition Expansion Upper Shelf Fluence, Temperature, 'F Temperature, 'F Temperature, 'F Energy, ft-lb n/cm'.78 W5212/5613F x 10" -22 110

'ata taken from [ST84] and [MA90]

0 t

$I yf C

<<S (l

2.0

~ YIeld Strength Model Description 2.1

~ Theoretical Basis Odette et,al. [OD86] have reported empirical relationships between irradiation strengthening and embrittlement. In particular, these researchers have observed a correlation between 30 ft-Ib indexed Charpy shift (ET,Q) and elevation in yield strength (do) (Figure 2-1), and between the fractional decrease in USE (f) and do(Figure 2-2), As discussed by Odette et.al., establishing correlations between Charpy parameters and microstructure-sensitive properties (such as yield strength) is of interest in gaining deeper understanding of radiation damage impacts on mechanical behavior. The correlations proposed in Reference [OD86]

assume continuous hardening (ha increases continuously), which is consistent with microstructural data reported in the literature. With regard to USE decreases, Odette et.al. state, "Details of the influence of irradiation on these processes (ductile fracture) are complex and not well understood... we believe that the overall effect of irradiation on C(Charpy V-notch) upper shelf ductile fracture may be related primarily to reduced strain hardening and flow localization leading to lower ductility and to increased triaxial stress state in C-sized specimens due to strength increases." Recent work by Manahan [MA92b] on irradiation effects on upper shelf fracture trends and mechanisms produced results which are consistent with the Reference [OD86] observations, and this work points out that non-hardening mechanisms, such as element transport to grain boundaries and possibly to particle interfaces, may play a role in the ductile fracture process, If such mechanisms are indeed active, then it is not possible to entirely characterize the shelf drop using hardening models. Nevertheless, the strong empirical correlation between t and Ao reported by Odette et.ai. can be used to estimate the unirradiated USE provided t e uncertainty is adequately characterized.

Reference [OD86] proposed the following empirical correlation for plates and welds based on analysis of the LWR database:

f = 6.2 x 10 'o, for 0 < ho< 5.8 ksi (2-1) f 62 x 10 BGy + 0 02 (6 895 6Gy %0) 'ol Gy > 5.8 ksi (2-2) where, ho= increase in yield strength (ksi)

Therefore, using this expression for f, it is possible to calculate the unirradiated USE provided irradiated USE and hodata are available.

0 I,

0[

2.2 Review of Earlier Application to NMP-1 Beltllne Welds Full Charpy curves for the NMP-1 beltline welds were not measured at the time when the vessel was fabricated. However, Charpy data at 10'F were measured by Combustion Engineering and these data are summarized in References [MA90]

and [MA91]. An innovative methodology [MA85a] was developed to determine the initial RT>> for cases where the data required by the ASME Code are not available. This approach was applied to the NMP-1 beltline materials and the results are described in Reference [MA90] and were reported in Reference

[MA92a]. The methodology for RT>>, determination includes estimation of the unirradiated USE in cases where full Charpy curves are not available. This method requires yield strength change and the upper shelf energy for the irradiated metal. For NMP-1, the surveillance weld data are available at a fast fluence of 4.78 x 10" n/cm'.

The yield strength model described earlier was used to estimate the surveillance weld unirradiated USE using the irradiated USE as input. In particular, USE'= (2-3)

(t where, f = fractional change in USE = AUSE USE'SE' unirradiated USE (ft-Ibs)

USE'"" = irradiated USE (ft-Ibs)

AUSE = USE'-USE'"" (ft-Ibs)

As described earlier, for her> 5.8 ksi, Equation (2-2) should be used for calculation of "f". The irradiated USE was measured at 7.98 EFPY and found to be 110 ft-lbs. Using the measured yield strength change of 8.68 ksi, the unirradiated USE for the surveillance weld is estimated to be 128 ft-lbs.

The irradiated Charpy data for the capsule weld material was analyzed using the SAM McFRAC code [McFRAC], This code is based on a non-linear, least squares, regression analysis using the Weibull statistic. The Weibull statistic has been shown to be the correct statistic for analysis of fracture data by considering the microstructural mechanisms involved in the fracture of ferritic, pressure vessel steels [MA85b]. The confidence bands calculated by McFRAC are measures of

'the goodness of fit'nd do not indicate the engineering 95% statistical error spread, This uncertainty must be analyzed using conventional statistical methods.

However, the McFRAC confidence intervals are used to measure confidence in the 10

I t'

4 I

p.(

'I

,t I

fit of a particular data set as well as the inherent scatter due to the fracture process. These error bands must be calculated, particularly for sparse data sets, because in many cases the ability to fit sparse data drives the uncertainty. The McFRAC analysis for the irradiated capsule weld is shown in Figure 2-3. Using the results of the McFRAC fit, the confidence interval for energy measurement (2a~) at the 50 ft-Ib level was estimated to be 13.5 ft-lbs. This estimate is consistent with the uncertainty in determination of the USE for tests conducted on the upper shelf.

The minimum unirradiated USE data for the beltline welds reported in Reference

[MA92a] is shown in Table 2-1. These data were determined assuming that the Charpy behavior of the surveillance weld is similar to the response for the beltline welds. Weld W5214/5613F was not made using the same wire heat or flux lot as

'he beltline welds. However, the weld materials were manufactured by the same suppliers, the weld wire type and flux type are the same (RACO ¹3 wire, Arcos B5 flux), the same procedure was used, and the Cu and Ni content is representative of the beltline welds [CE90, MA91]. Therefore, it has been assumed that the capsule weld material is similar to the beltline welds in terms of its mechanical behavior response.

At the time the Reference [MA92a] analysis was performed, the results of the uncertainty analysis reported herein were not available. Therefore, engineering judgement was applied to obtain reasonable yet conservative estimates of the

. unirradiated USE levels. To ensure conservatism, the measured irradiated USE was used as an estimate of the unirradiated surveillance weld USE. In estimating the beltline weld USE, this value was decreased to account for uncertainty in the yield strength model estimate. In order to estimate the beltline weld unirradiated metal USE levels, the measured irradiated USE for the surveillance weld (110 ft-Ibs) was reduced by 2a, (13.5 ft-Ibs) plus an additional 6.5 ft-Ib for conservatism.

This lower bound estimate of 90 ft-Ibs was conservatively assumed to represent the unirradiated USE of the beltline welds.

11

0' Table 2-1 Estimated Upper Shelf Energy for NMP-1 Beltline Welds [MA92a]

Material wt. % Minimum Irradiation Irradiation Predicted cu Unirrad.

USE (ft-Ib)

Decrement aUSE (%)

Decrement AUSE (%)

efpy)'redicted USE 12/16/91 USE at EOL(25 12/16/91 EOL(25 (ft-Ib) efpy)'ft-Ib)

W5214/5G13F 0.18 -100'0 17 20 83.0 80.0 86054 B/4E5F 0.22 20- 23 72.0 69.3 1248/4M2F 0.22 90 20 23 72.0 69.3

'ower bound estimate based on irradiated value measured at a fluence of 4.78 x 10" n/cm' Conservatively estimated using data in [MA90] and [MA91]

'ast fluence of 7.26 x 10" n/cm't the peak 1/4T position

'ast fluence of 1.44 x 10" n/cm't the peak 1/4T position

'ata from Reference [CE90]

'urveillance Weld 12

C C('

300 WELD PLATE EST. MEAS.

UNCERTAINTY ~

EST. MEAS.

UN CERTAINTY O

~ ~

f ~

CI

~ ~

~

0 350 0

~o> (MPa) b Wa(MPa)

Figure 2-1 Plots of Transition Temperature Shifts Indexed at 41 J (Joules) Versus Static Yield Stress Changes for (a) Weld and (b) Plate and Forgings [OD86]

13

0.7 USE'SED g-Q uf

~ WELD X o PLATE N

ou

~ ~ &+0 ~

0 A e 0 X ~

O 0 0 ~ ~

I D 8o o~

4 0

o <<po he> (MPI)

Figure 2-2 Fractional Decreases in CUpper Shelf Energy Versus Yield Stress Changes [OD86]

14

h g NINE MILE POINT UNIT WELD 5 2 I 4./5G 1 3F (SURVEILLANCE WELD) IRRADIATED DATA

-150 WEIBULL FIT TRANSITION Ql WEIBULL FIT 125 g4 I-I ~p

/~gg Jk Qkiki~ UPPER SHELF HYPERBOLIC 100 TANGENT FIT CS

/ / k CONFIDENCE LIMIT (95+)

V5 CONFIDENCE LIMIT (96%)

~ 50 4

A CONFIDENCE O 4 LIMIT (86%)

Oa CL CONFIDENCE 26 LIMIT (95%)

UNIRRADIATED DATA 300 UNIRRADIATED CHARPY CURVE TEST TEMPERATURE (F)

Figure 2-3 Charpy Impact Energy Versus Test Temperature for Irradiated Weld Specimens from the Nine Mile Point Unit 1 300 Degree Capsule 15

'14l(

)" f.

+

J

((l, 4

3.0 Uncertainty Estimation The basic approach to characterizing the uncertainty associated with estimating the unirradiated USE using the yield strength model described in Section 2.0 is to apply the model to LWR data for which the unirradiated USE is known, and then to determine the uncertainty by plotting the calculated USE versus the measured USE and determining the two sigma limits, The following steps were followed:

~ Extract weld data from the LWR database for which the unirradiated USE (USE'), irradiated USE (USE'""), and change in yield strength (hGy) data are available. The Reference [PREDB] database was used.

~ Calculate USE'sing the yield strength model;

~ Calculate the two sigma limits for USE'calculated) vs. USE'measured)

Details related to the uncertainty evaluation are provided in Sections 3.1 and 3.2 of this report which follow. During the analysis, it was discovered that the functional form of the yield strength model should be updated. The details concerning model update are described in Section 3.3. The updated model was applied to the NMP-1 surveillance weld, and these data are presented in Section 3.4. Comparison of the uncertainty in yield

~ ~

strength and USE measurements with the uncertainty in the yield strength model is

~ ~

~

presented in Section 3.5.~ ~

3.1 Database Analysis The NRC provided the latest version of the Power Reactor Embrittlement DataBase (PR-EDB) [PREDB] for use in the weld USE uncertainty evaluation.

The Charpy shift database (SHFT PR,DBF) and the tensile database (TEN PR.DBF) files were analyzed and reduced, subject to the following criteria, to produce the final combined weld uncertainty analysis data set:

~ Only weld records were used in the analysis.

~ Records were eliminated which do not have the minimum data required to calculate the model uncertainty (USE; USE'"", and do).

~ Records were eliminated in cases where the reported Charpy fluence was not within 40% of the reported tensile fluence.

~ Records were eliminated if the unirradiated tensile test temperature was greater than 50 degrees above or below the irradiated tensile test temperature.

~ Whenever possible, hadata at 550'F were used. If 550'F data were not 16

'vailable, data at the next highest temperature was used. As discussed in Reference [OD86], temperature effects on daare small.

~ Lower yield strength data were used to calculate do~

~ Records were eliminated if the capsule and heat identification in the tensile database does not match the capsule and heat identification in the Charpy database for a particular plant.

T

~ Average stress values were used to calculate dowhenever results for more than one tensile test were available at the test temperature, material, and fluence conditions of interest.

~ All database units were converted to U.S. conventional units (ha= ksi, USE =

ft-lbs).

~ The final data set contains only weld data for commercial BWR and PWR plants irradiated between 520'F and 590'F,

~ If the Charpy data reported in the PR-EDB indicates an increase in the USE with irradiation, and the increase is larger than 15 ft-Ibs (it was assumed that the USE measurement uncertainty is+15 ft-Ibs at the two sigma level), then these data were not included in the analysis (see discussion on outliers which follows).

Otherwise, the d USE was set equal to zero for cases where the USE increased by less than 15 ft-lbs.

~ If the tensile data reported in the PR-EDB indicates a decrease in yield strength with irradiation, and the decrease is larger than 5 ksi (it was assumed that the yield strength measurement uncertainty is +5 ksi), then these data were not included in the analysis (see discussion on outliers which follows). Otherwise the b,owas set equal to zero for cases where the yield strength decreased by less than 5 ksi.

The reduced data set obtained by application of these criteria is shown in Table 3-1. The data set contains 145 weld records, which is significantly larger than the 32 weld records available at the time the yield strength model was originally developed.

3.1.1 Analysis of Outliers The data records for which the yield strength decreased by more than 5 ksi and/or the USE increased by more than 15 ft-Ibs are summarized in Tables 3-2 and 3-3, respectively. For the Callaway plant, based on the data trends, there appears to be an error in the unirradiated aat 550'F (Table 3-2). Therefore, 17

~k s 'f 1 (

tl k,

0

'I JC

the RT odata was used to calculate do in the uncertainty analysis. For Oconee 3 (Table 3-2), there appears to be an error or anomaly for the daat.

550'F for Capsule B. The data trends suggest an error in the irradiated ovalue.

Therefore, for this capsule, the RT dewas used in the uncertainty analysis.

Similarly, for Vermont Yankee, there appears to be an error in the yield strength for the irradiated specimen tested at 543'F. Therefore, the RT hawas used in the uncertainty analysis.

For Dresden 3 (Table 3-3), an interesting USE versus fluence trend was observed. For fluences below about 1 x 10" n/cm', the USE appears to increase by a statistically significant amount, Above the 1 x 10" n/cm'luence, the USE decreases as expected, Therefore, the first data column in Table 3-3 was eliminated from the uncertainty analysis. The Vermont Yankee Capsule 30D, which was irradiated to 4.3 x 10" n/cm', shows a similar behavior to that of the Dresden Capsule 6. Below about 1 x 10" n/cm', the USE appears to increase slightly.

Table 3-3 also lists an outlier for the Garigliano plant. In this case, a statistically significant USE increase and a statistically significant increase in yield'strength at a fluence of 7 x 10" n/cm'as reported. This datum is inconsistent with the data trends of the other Garigliano capsules, and therefore, was not included'in the USE uncertainty evaluation, After the PR-EDB was assembled, Oak Ridge National Lab (ORNL), in cooperation with the Electric Power Research Institute (EPRI), performed additional verification of the database. However, not all data could be verified by the reactor vendors, and the PR-EDB contains a list of the reactors, capsules, and files which were not verified. None of the previously discussed outliers were on the PR-EDB unverified list.

3.1.2 Summary As with most outlier analyses, there is not sufficient information available to draw firm conclusions regarding the cause of the anomalies. However, the data in Table 3-2 suggest that the few cases of dodecrease may be due to data error. For the three cases listed in Table 3-3 which show a statistically significant USE increase, two of these (Dresden 3 Capsule 6 and Vermont Yankee) occur at very low fast fluences ((1 x 10" n/cm'). The Garigliano datum appears to be a data error since it is inconsistent with the other Garigliano capsule data.

Based on the analyses performed, it is concluded that the yield strength increases continuously with neutron irradiation, in agreement with the earlier observations made by Odette et.al. [OD86]. With the possible exception of low 18

I S

fluence irradiation, the USE decreases with fluence, which is also consistent with the model reported in Reference [OD86].

N 3.2 USE Estimation Uncertainty The yield strength model (Section 2.0) was applied to the reduced weld data set described in Section 3.1. The results of the analysis are shown in Table 3-1 and in Figure 3-1. The SYSTAT [SYSTAT] code package was used to perform a linear least squares regression analysis on the measured (USE'~,) versus predicted (USE'~Q unirradiated USE data. The calculated slope (Figure 3-1) of the regression model is 0.988 with a residual squared (R') value of 0.986. The regression analysis identified D.C. Cook Unit 1 Capsule U (Case 23), Garigliano Capsule 114A (Case 49), and McGuire 2 Capsule X (Case 69) as possible outliers.

The two sigma scatter band was determined to be 23.99 ft-lbs. This is the best estimate of the uncertainty in the yield strength model and can be used as a conservative adjustment to the estimated USE'. In the case of the NMP-1 surveillance weld, the estimated USE'f 128 ft-Ib (Section 2,2) may be reduced to 104 ft-Ibs to account for the uncertainty in the USE estimation model. This reduced USE'stimate is consistent with the 100 ft-Ib obtained using engineering judgement (See Table 2-1).

3.3 Yield Strength Model Update A plot of the Reference [OD86] model along with the current weld data set is shown in Figure 3-2. Although the data set used in the mid-1980s suggested a possible nonlinear dependence of the fractional. decrease in USE (f) with increase in yield strength (ho), Figure 3-2 shows that a linear dependence is more appropriate. As shown in Figure 3-2, the [OD86] model tends to underpredict the USE for low ho(-<8ksi), and tends to overpredict at higher ha levels.

A linear least squares regression was performed on the current LWR weld data set, The results are shown in Figure 3-3. The linear and non-linear models are compared in Figure 3-4. The regression yielded an R'alue of 0.880. The two sigma scatter band was determined to be 0.182. This two sigma value is lower than the scatter estimate of+0.2 reported in [OD86] for the non-linear model.

Therefore, based on the analyses performed, it has been concluded that the fractional decrease in USE depends linearly on ho for LWR weld metals. The proposed linear function is:

f 0 015 bay (3-1) 19

II 1I lI~

',I 1I

where, f = fractional decrease in USE ba= increase in yield strength (ksi) 3.4 Application of Updated Model to NMP-1 Welds The linear yield strength model was applied to the LWR weld data set to estimate the uncertainty in calculating the unirradiated USE in a similar manner as described in Section 3,2 for the non-linear model. The results are shown in Table 3-4 and Figure 3-5. The calculated slope of the regression line is 0.979, which is close to unity. The R'alue is 0.986 which is identical to the results obtained for

'the non-linear model. The regression analysis identified D.C, Cook Capsule U (Case 23) and Gangliano Capsule 114A (Case 49) as possible outliers. The two sigma scatter band was determined to be 23.69 ft-lbs, which is slightly lower than the 23.99 ft-Ibs obtained using the non-linear model. Overall, the uncertainty for the linear and non-linear models is approximately the same. This is because the uncertainty due to mechanical behavior testing is large and dominates the impact of the functional form of the yield strength model.

Applying the linear model to the NMP-1 surveillance weld results in an estimated unirradiated USE of 126 ft-lb. This estimate is slightly more

.conservative than that obtained earlier using the non-linear model. Subtracting the two sigma estimate of 24 ft-Ibs yields a lower bound unirradiated USE of 102 ft-lbs. As mentioned earlier, this USE level is consistent with earlier estimates based on engineering judgement.

3.5 Tensile/USE Measurement Uncertainty A cursory examination of Figure 3-3 suggests that the scatter in hodata taken from the LWR database is significantly larger than the expected uncertainty in yield strength measurement, For uniaxial testing of steel, an uncertainty on the order of +5 ksi is typical. Therefore, a limited investigation was performed to address this question. The uncertainty for yield strength measurement of nuclear pressure vessel steels was characterized and these data are presented in Section 3.5.1 below. A similar study was performed for USE determination to see if the yield strength model uncertainty is consistent with the mechanical property measurement uncertainty, and these results are presented in Section 3.5.2 below.

3.5.1 Tensile Uncertainty Analysis The PR-EDB database file TEN PR.DBF was used to identify plants for which three or more tensile tests had been performed. The plants, along with 20

,ll J'

L t

-T

mean lower yield strength and two sigma estimates, are shown in'Table 3-5.

The average two sigma estimate is 3.2 ksi. Most of the data with 3 or more yield strength measurements are for unirradiated material. Most of the surveillance capsules contain 3 tensile specimens for each capsule material (base, weld, HAZ), and these are usually tested at room temperature (RT), 550'F, and at an intermediate temperature. Further, not all plants have multiple yield strength data in the unirradiated condition. For many of the data shown in Figure 3-3, only one unirradiated, odatum and one irradiated adatum were used to calculate dc>. The correlation for daversus fractional decrease in USE is shown in Figure 3-6. The two sigma limit for the LWR weld data is 11.3 ksi.

This two sigma limit is significantly larger than the multiple specimen (3 or more measurements) two sigma uncertainty as a result of insufficient number of test specimens required to accurately characterize ho.

3.5.2 USE Uncertainty As mentioned in Section 2.2, an uncertainty of 13.5 ft-Ibs was used in the NMP-1 surveillance weld analysis. This estimate was determined based on the surveillance capsule data. In order to obtain an estimate of the USE uncertainty in the LWR database, ten plants were randomly selected from the PR-EDB database. The plants, along with hyperbolic tangent fits to their weld data, are shown in Figures 3-7 through 3-16. As shown in the figures, there is usually data available for accurate determination of the unirradiated USE. 'ufficient However, since only 8 to 16 Charpy specimens are available in most surveillance capsules, data for USE characterization on the upper shelf is generally limited, In order to estimate the uncertainty in USE measurement in the unirradiated condition, the data for each of the plants was fit using a linear regression model, The equation used is listed below:

USE =A, T+A where Ai regression coefficients T = test temperature.

The fracture appearance data were used to identify upper shelf points. After each model regression was completed, the midpoint of the temperature range over which the USE measurements were made was calculated. Each set of data were shifted up or down at the midpoint temperature to yield normalized data distributed around a 100 ft-Ib mean. The normalized data were fit using Equation 3-2, and the results are shown in Figure 3-17. The two sigma limit for 21

4

'l 5'*

• If

the normalized data was calculated to be 19.8 ft-lbs.

3.5.3 Summary The two sigma limit for USE measurement was estimated to be 19.8 ft-Ibs for unirradiated weld metal. Since, for most plants, the unirradiated USE is determined using 9 or more specimens, and the irradiated USE is determined using 3 or more specimens, the uncertainty in calculating the fractional decrease in USE (f) is expected to be on the order of the uncertainty in USE measurement. As shown in Figure 3-3, this is the case for LWR weld metals since the two sigma limit for "f" was found to be + 0.18.

The two sigma limit in correlating hc with fractional decrease in USE (Figure 3-6) (-11 ksi) is more than twice tie two sigma limit associated with yield strength measurement ('3-5ksi). This is believed to be due mainly to the fact that the current industry practice in surveillance testing is to perform only one tensile test at each temperature. In addition, there are only a few plants which have baseline tensile data determined using multiple specimens. In the future, for those plants which use the yield strength model to estimate the unirradiated USE, more accurate results can be obtained by allocating more specimens to testing at a given temperature and/or through the use of miniature specimens.

22

'. C V

II (f

Teble 3-1 LWR Weld Data and Non-Linear Yield Strength Model Estimates Of the Unlrradlated USE Unirradiated Irradiated Uniiradiated USE USE Calculated Measured USE (Measured) (Measured) Fractional Fractional (Calculated)

USE4~ USE ups USE~ USE USE USE USE'cAi USE'~MS E'cN

~Plan ~Chan e ~Chan e ~flub ~4b CASE 1 ANO-1 73.000 45.000 28.000 12. 830 0.219 0 384 57.607 15. 393 CASE 2 ANO-1 73.000 47.000 26.000 12.830 0. 219 0.356 60 167 12. 833 CASE 3 ANO-1 73.000 58.000 15. 000 8.430 0. 137 0.205 67.243 5.757 CASE 4 ANO-2 154.000 147 F 000 7.000 3.100 0.019 0.045 149.847 4 '53 CASE 5 Angra 171.000 155 F 000 16.000 0.000 0.000 0.094 155+000 16.000 CASE 6 Beaver Val 1 112.000 78.000 34.000 19.900 0.321 0.304 114 821 -2.821 CASE 7 Beaver Val 1 112.000 83.000 29 F 000 16. 800 0.278 0.259 115 025 -3 025 CASE 8 Beaver Val 1 112.000 &8.000 24.000 15 800 0 264 0.214 119.583 -7.583 CASE 9 Big Rock 95.000 65 F 000 30.000 17 '00 0 '94 0.316 92.035 2.965 10 Big Rock 95.000 70 000 25.000 15. 300 0.257 0.263 0.813 CASE CASE CASE ll 12 Big Rock Big Rock 95.000 95.000 57 F 000 80+000 38.000

15. 000 12 ~ 100 7.500 0.207

0. 115 0.400 0.158 94 ~ 187

71. 869 90 '94 23.131 4.606 CASE 13 Cal Cliffs 1 160.000 119 F 000 41.000 12. 200 0 '09 0.256 150.357 9.643 CASE 14 Cal Cliffs 2 137 F 000 <05~000 32.000 10.900 0. 186 0.234 129.028 7.972 CASE 15 Callaway 1 112. 000 97.000 15.000 8.000 0. 128 0.134 111 177 0.823 CASE 16 Callaway 1 112.000 10lo000 11.000 0.000 0.000 0 098 101 ~ 000 11.000 CASE 17 Catawba 1 128.000 123 F 000 5.000 4 300 0 '27 0.039 126+413 1.587 CASE 18 Crystal Riv3 79.000 68.000 11. 000 10. 680 0 '82 0.139 83. 157 -4. 157 CASE 19 Crystal Riv3 79.000 64.000 15. 000 10.480 0 ~ 179 0 190 77 918 1.082 CASE 20 Crystal Riv3 79.000 70.000 9.000 1.880 0 ~ 012 0. 114 70.850 8 ~ 150 CASE 21 Crystal Riv3 79.000 63 F 000 16.000 10.680 0 ~ 182 0.203 77 '43 l. 957 CASE 22 D.C. 'Cook 1 110.000 80.000 30.000 19.050 0 '09 0.273 115.836 -5.836 CASE 23 D.C. Cook 1 110. 000 94 000 16.000 22 '50 0.357 0. 145 146.276 -36.276 CASE 24 Davis Besse 70.000 57.000 13.000 10.200 0.173 0. 186 68 '60 l. 040 CASE 25 Davis Besse 70.000 54.000 16. 000 10.500 00179 0.229 65 773 4.227 CASE 26 Davis Besse 70.000 64 000 6.000 F 500 0.034 0.086 66 253 3.747 CASE 27 Davis.Besse 72.000 62.000 10. 000 11 ~ 800 0.202 0.139 77 680 -5.680 CASE 28 Davis Besse 72.000 65.000 7.000 5.500 0 '34 0.097 67 288 4.,712 CASE 29 Diablo Can 1 98.000 87.000 11. 000 14.700 0 '48 0. 112 115.673 -17.673 CASE 30 Diablo Can 2 121. 000 85 000 36.000 16. 600 0 276 0.298 117 337 3.663 CASE 31 Dresden 2 71.000 52.000 19.000 17 600 0.290 0.268 73 '99 -2. 199

-l. 851 CASE 32 Dresden 3 65.000 45.000 20.000 20.370 0 '27 0.308 66+ 851

-1.919 CASE 33 Dresden 3 65.000 41 F 000 24.000 25. 170 0 387 0 369 66. 919 CASE 34 Dresden 3 65.000 42.000 23.000 16.570 0 '75 0.354 57.944 7.056 CASE 35 Dresden 3 65.000 65.000 0.000 0.400 0 F 002 0 000 65 '30 -0. 130 CASE 36 ~ Dresden '3 70.000 59 000 11. 000 20.830 0 '33 0.157 88 '36 -18.436 CASE 37 Dresden 3 70+000 64+000 6.000 13. 230 0 225 0.086 82.606 -12. 606 CASE 38 Dresden 3 70.000 70~ 000 0.000 2 '30 0 013 0.000 70.922 -0 '22 CASE 39 Duane Arnold 101. 000 101~000 0.000 0 000 0.000 0.000 101 000 0.000 CASE 40 Parley 1 149.000 108.000 41-000 6.550 0.086 0.275 118.173 30.827 CASE 41 Farley 1 149.000 1300000 19.000 12.250 0.209 0. 128 164.428 15 ~ 428 CASE 42 Parley 1 149.000 115.000 34.000 8.150 0 131 0.228 132 '45 16. 655 CASE 43 Farley 2 144 F 000 144 F 000 0.000 1.350 0.008 0.000 145 ~ 161 -l. 161 CASE 44 Farley 2 144.000 144 F 000 0.000 6.450 0.082 0.000 156. 918 -12. 918 CASE 45 Farley 2 144.000 132.000 12.000 5.350 0 '33 0.083 136 505 7.495 CASE 46 Ft. Calhoun 104.000 59 ~ 000 45.000 21. 100 0 336 0.433 88 '02 15.098 CASE 47 Ft. Calhoun 104.000 650000 39.000 22.100 0 349 0.375 99 871 4 129 CASE 48 Garigliano 103 '00 79 '00 23.800 14.930 0 '51 0.230 106.321 -2.921 CASE 49 Garigliano 103:400 67 '00 36 ~ 100 36.410 0. 516 0.349 139.191 -35.791 CASE 50 Carigliano 103.400 60.000 43.400 18.490 0 302 0.420 85.937 17. 463 23

<<)$

tl

Table 3-1 LWR Weld Data and Non-Linear Yield Strength Model Estimates of the Unlrradlated USE (Continued)

Unirradiated Irradiated Unirradiated USE USE Calculated Measured USE (Measured) (Measured) Fractional Fractional (Calculated)

USE'~ USE Mrs USE'~-USE USE USE . USE'~ USE'~-USE'oAl

~Plan ~Chan a Cbanlbe ~db ~lb CASE 51 Garigliano 103. 400 51. 400 52 ~ 000 24 890 0 ~ 384 0.503 83.428 19 '72 CASE 52 Garigliano 103. 400 83.900 19.500 0.000 0.000 0. 189 83.900 19 500 CASE 53 Garigliano 103.400 103.400 0.000 0.710 0 004 0 000 103. 815 -0.415 CASE 54 Garigliano 103.400 73.800 29.600 23.750 0.370 0.286 117. 119 -13 ~ 719 CASE 55 Ginna 80.000 53.000 27.000 21.750 0 345 0.338 80.879 -0 879

~

CASE 56 Ginna 80.000 51.000 29 F 000 22.850 0.359 0.363 79.518 0 ~ 482 CASE 57 Ginna 80.000 50.000 30.000 19. 650 0.317 0 ~ 375 73.246 6:754 CASE 58 Haddan Neck 105.000 83 ~ 000 22.000 16.750 0.278 0 ~ 210 114. 913 -9 '13 CASE 59 Indian Pt. 2 118 000 75.000 43 000 25 800 0 ~ 395 0.364 123.958 -5.958 CASE 60 Indian Pt. 3 120.000 68.000 52.000 24.800 0.383 0.433 110.176 9.824 CASE 61 Kewaunee 126.000 .78.000 48.000 27 750 0.418 0.381 134.076 -8.076 CASE 62 Kewaunee Maine Yankee 126.000 82.000 44 F 000 23 '50 0 369 0 349 129.877 -3.877 CASE 63 105.000 57.000 48.000 30.670 0.452 0.457 104.056 0.944 CASE 64 Maine Yankee 105.000 59>>000 46 F 000 23.720 0.370 0.438 93.577 11. 423 CASE 65 Maine Yankee 105.000 66 F 000 39.000 24.970 0.385 0. 371 -

107.296 -2.296 CASE 66 Maine Yankee 105.000 50.000 55.000 36. 170 0.514 0.524 102. 851 2.149 CASE 67 McGuire 1 112 F 000 83 ~ 000 29.000 19. 900 '>>321 0 ~ 259 122 181 -10. 181 CASE 68 McGuire 1 112.000 75 F 000 37 000 13 800 0.234 0.330 97.932 14>> 068 CASE 69 McGuire 2 133.000 133.000 0.000 10.950 0.187 0 F 000 163 '15 -30.615 CASE CASE 70 71 McGuire Millstone 2

2 133.000 132.000 133 F 000 98.000 0.000 34.000 7.350 6.070

0. 111 0.065 0.000 149 '01 -16 601 0 ~ 258 104 F 800 27.200 CASE 72 Millstone 2 132.000 108.000 24 F 000 10.270 0. 175 0.182 130.869 1. 131 CASE 73 North Anna 1 95.000 - 95.000 0>>000 2 650 0.016 0.000 96 545 -1. 545 CASE 74 North Anna 1 95.000 92 F 000 F 000 10.850 0.185 0.032 112 ~ 929 -17.929 CASE 75 North Anna 2 112.000 112 ~ 000 0.000 5.450 0.034 0.000 115 ~ 942 -3.942 CASE 76 North Anna 2 115.000 92 F 000 23 F 000 2.500 0.016 0.200 93.496 21. 504 CASE 77 Oconee 1 64.000 52.000 12 ~ 000 19.300 0 ~ 313 0. 188 75.660 -11 660 CASE 78 Oconee 1 64.000 52.000 12. 000 16.700 0.277 0.188 71.923 -7 923 CASE 79 Oconee 1 64.000 55 F 000 F 000 13.500 0.229 0>> 141 71.381 -7 '81 CASE 80 Oconee 2 67.000 44.000 23>>000 18.770 0.306 0. 343 63.364 3.636 CASE 81 Oconee 2 67.000 47.000 20.000 16.420 0 ~ 273 0 299 64.652 2 348 CASE 82 Oconee 2 68.000 54.000 14 F 000 10>>770 0. 184 0.206 66. 168 1.832 CASE 83 Oconee 3 66.000 49.000 17.000 15 700 0.263 0.258 66.455 -0.455 CASE 84 Oconee 3 66>>000 58.000 8.000 5.200 0 032 0. 121 59.917 "6>>083 CASE 85 Oconee 3 66 F 000 42 000 24 F 000 19. 200 0 ~ 311 0. 364 60 ~ 991 5.009 CASE 86 Palisades 118 F 000 52 F 000 66.000 32.660 0.475 0 ~ 559 99.018 18 982 CASE 87 Palisades 118 000 64 F 000 54.000 25.510 0.391 0.458 105. 168 12.832 CASE 88 89 Prarie Zsl 1 Prarie Isl 79.000 79.000 0 000 8 '50 0. 129 0.000 90; 670 -11. 670 CASE 1 79.000 75.000 4.000 12.650 0. 216 0. 051 95.655 -16.655 CASE 90 Prarie Zsl 1 79.000 79 F 000 0.000 4.950 0 ~ 031 0 000 81.527 -F 527 CASE 91 Prarie Isl 2 103.000 91.000 12.000 17 F 000 0. 281 0 117 126.607 -23.607 CASE 92 Prarie Isl 2 103>>000 100 F 000 3.000 8.450 0. 138 0.029 115.997 -12.997 CASE 93 Prarie Isl 2 103 F 000 92.000 11>> 000 7.850 0.124 0.107 . 105.009 -2 '09 CASE 94 Pt. Beach 1 65.000 54.000 ll>> 000 19. 300 0.313 0 ~ 169 78.570 -13. 570 CASE 95 Pt. Beach 1 65.000 55.000 10. 000 20.300 0.326 0. 154 81.595 -16 595 CASE 96 Pt. Beach 1 65.000 51. 000 14.000 19.800 0.319 0. 215 74 '29 -9 929 CASE 97 Pt. Beach 2 65.000 42.000 23.000 18.070 0.296 0.354 59.666 5.334 CASE 98 Pt. Beach 2 65.000 47.000 18. 000 19. 970 0.322 0.277 69.281 -4.281 CASE 99 Pt. Beach 2 66 000 44.000 22 F 000 20.570 0.329 0.333 65.620 0.380 CASE 100 (tuad City 1 72.000 49.000 23.000 29.000 0.433 0.319 86 '05 -14.405 24

$1 fy I I

4

/

Table 3-1 LWR Weld Data and Non-Linear Yield Strength Model Estimates of the Unlrradlated USE (Continued)

Unirradiated Irradiated Unirradiated USE USE Calculated Measured USE (Measured) (Measured)

~Plan USE'M~ USE ~ USE'M~-USE Fractional USE

~Chan a Fractional USE

~Chan e (Calculated)

USE'oM

~ltdb USE'Mrhs-USE'oM

~Zb CASE 101 Quad City 1 looeooo 85.000 15.000 14 ~ 600 0 246 0.150 112.789 -12.789 CASE 102 Quad City 1 100.000 95.000 5.000 3.200 0. 020 0.050 96.939 3.061 CASE 103 Quad City 1 looeooo 75 000 25.000 18. 200" 0.298 0.250 106.816 -6. 816 CASE 104 Quad City 2 87.000 41.000 46.000 29.700 0. 441 0.529 73.350 13.650 CASE 105 Quad City 2 90.000 51.000 39.000 18.600 0. 303 0.433 73.202 16.798 CASE 106 Quad City 2 ~ 125.000 107.000 18.000 0.400 0.002 0.144 107 214 17.786 CASE 107 Quad City 2 125 000 F 89.000 36.000 8 100 0. 130 0.288 102.286 22.714 CASE 108 Quad City 2 125.000 80.000 45.000 14. 700 0 248 0.360 106 366 18 634 CASE 109 Rancho Seco 68.000 51. 000 17 F 000 14. 000 0.237 0.250 66.863 1.137 Rancho Seco CASE CASE CASE ill 110 112 Rancho Seco Robinson 2 68.000 68.000 112.000 48.000 53.000 70 000 20.000 15.000 42.000

11. 000 25.050

23. 150

0. 188 0.386 0.362 0.294 0.221 0.375

59. 113 86.299 109.786 8.887

-18. 299 2.214 CASE 113 SONGS 2 99.000 80 000 19. 000 16. 100 0.268 0.192 109.356 -10.356 CASE 114 Salen 1 104.000 75.000 29.000 21.000 0.335 0.279 112.791 -8.791 CASE 115 Salen 2 ill. 000 79 F 000 32.000 18. 700 0.305 0.288 113. 613 -2. 613 CASE 116 Salen 2 111. 000 111. 000 74.000 37 000 20e700 0 '31 0.333 110. 640 0.360 CASE 117 Sequoyah 1 82 000 29.000 10 100 0.172 0.261 98.981 12 ~ 019 CASE 118 Sequoyah 1 111. 000 78.000 33 000 11. 800 0 '02 0.297 97.726 13. 274 CASE 119 Seguoyah 2 112 ~ 000 109 000 3 000 5.550 Oe034 0.027 112.836 -Oe836 CASE 120 Sequoyah 2 112 ~ 000 110.000 2.000 6.050 0.064 0.018 117. 488 -5.488 CASE 121 Shearon Har 94.000 82.000 12.000 6.000 0 ~ 061 0 ~ 128 87.294 6.706 CASE 122 St. Lucio 1 144.000 108.000 36.000 6 '30 0.099 0.250 119.840 24.160 CASE 123 St. Lucie 1 144 F 000 100.000 44.000 9.330 0.157 0.306 118. 561 25.439 CASE 124 Surry 1 70.000 53 ~ 000 17. 000 19.750 0.319 0 '43 77 792 -7.792 CASE 125 Surry 1 70.000 50 000 20.000 24 '50 Oe377 0.286 80.293 -10.293 CASE 126 Surry 2 90.000 60.000 30.000 18.500 0.302 0.333 85.953 4.047 CASE 127 Surry 2 90.000 70 F 000 20.000 10.300 0 ~ 175 0.222 84 '80 5.120 CASE 128 THZ 1 Sle 000 50eooo 31.000 14 '70 0 '50 0.383 66.705 14.295 CASE 129 THI 1 eleooo 64 000 17. 000 11. 750 0 ~ 201 0.210 80.100 0.900 CASE 130 Trojan 83.000 81.000 2 000 3 '00 0 '22 0.024 82 '22 0.178 CASE 131 Trojan 83 ~ 000 62.000 21. 000 6 300 0.076 0.253 67 113 15.887 CASE 132 Trojan 83 000 83.000 0 ~ 000 6 300 0.076 0.000 89.844 -6.844 CASE 133 Turkey Pt. 3 65.000 48eooo 17. 000 20.850 0 ~ 333 0.262 71. 976 -6.976 CASE 134 Turkey Pt. 3 65.000 59eooo 6.000 18 ~ 550 0 ~ 303 0.092 84.603 -19.603 CASE 135 Turkey Pt. 4 66 F 000 44.000 22.000 21. 040 0 '36 0.333 66.222 -0.222 CASE 136 V.C. Sunner 91.000 87 000 4.000 0-800 Oe005 0.044 87.437 3.563 CASE 137 V.C. Sunner 91eooo 85.000 6 000 5.000 Oe 031 0.066 87.719 3.281 CASE 138 Vernont Yank 107.000 107.000 0.000 4.800 0 030 Oeooo 110.309 -3.309 CASE 139 Wolf Creek looeooo 92eooo 8.000 0.900 0. 006 0.080 92.555 7.445 CASE 140 Zion 1 64.000 44 F 000 20.000 18.200 0 ~ 298 0. 313 62.666 1 ~ 334 CASE 141 Zion 1 64 F 000 52.000 12. 000 16.000 0 F 267 0 188 70 942 -6.942 CASE 142 Zion 1 68.000 56.000 12eooo 13 400 0.228 0. 176 72.531 -4. 531 CASE 143 Zion 1 68.000 49 F 000 19.000 16.000 0.267 0 ~ 279 66 '49 l. 151

-2.628 CASE 144 Zion 2 69 000 50.000 19.000 18 ~ 500 Oe302 0.275 71.628 CASE 145 Zion 2 69eooo 51.000 18. 000 24 500 0.379 0 ~ 261 82.142 -13. 142

0 I

I 0

Table 3-2 Analysis of Tensile Data Outliers'ensile Data (ksi)

Callaway Callaway Oconee 3 Oconee 3 Oconee 3 Vermont Yankee T t C Chti ~CI U ~CI Y '~CI A ~CI ~ ~CI 0 .~CI 30 ~

a- unirradiated 72.5 -

72.5 74.9 74.9 74.9 68.0 low temperature (70'F) (70'F) (71'F) (71'F) (71'F) cr- unirradiated 67.5 67.5 intermed. temp. (300'F) (300'F) o- unirradiated 80.0 80.0 67.4 67.4 67.4 67.6 high temp. (550'F) (550'F) (580'F) (580'F) (580'F) (543'F) o- irradiated 71.3 80.5 79.3 90.6 98.0 72.8 low temp. (73'F) (70'F) a- irradiated 68.2 78.4 intermed. temp. (150'F) (125'F) o- irradiated 65.7 64.2 72.6 57.8 86.6 57.5 high temp. (550'F) (550'F) (580'F) (585'F) (580'F) (543'F) halow temp. -1.2 8.0 4.4 15.7 23.1 4.8 d a-intermed. temp. 0.7 10.9 ho-high temp. -14.3 -15.8 5.2 -9.6 19.2 -10.1

'he data in this table exhibit yield strength decreases greater than 5 ksi. The data from left to right are in the order of increasing fluence for each plant.

26

Table 3-3 Analysis of Charpy Data Outliers Charpy Data 'ft-Ibs)

Dres. 3 Ores. 3 Dres. 3 Dres. 3 Dres. 3 Dres. 3 Dres. 3 Dres. 3 Garigliano VT Test CondiTion ~Ca . 6'Ca . 6'Ca . 0'~Ca . 14 ~Ca . 14 ~Ca . 4 ~Ca . 4 ~Ca . 12 ~Ca . 113D Yankee'Ca

. 30D USE'0 Unirradiated 65 70 65 70 70 65 65 103.4 Irradiated USEIRR 106 71 75 45 59 41 42 123.7 USE'-USE'30 -6 -5 20 11 24 23 -20.3 -15 0.4 2.03 20.4 13.2 20.8 25.2 16.6 21.48 See Table 2-4

'he data for Dresden 3 are listed from left to right in the order of increasing fluence.

All fast fluences, except those noted, are above 1.0 x 10" n/cm' Data for fast fluence = 4.3 x 10" n/cm' Data for fast fluence = 2.75 x 10" n/cm' Data for fast fluence = 0.95 x 10" n/cm'.

27

yt C

Table 3A LWR Weld Data and LInear Yield Strength Model Estimates of the Unlrradlated USE Unirradiated Irradiated Unlrradiated USE USE Calculated Measured USE (Measured) (Measured) Fractional (Calculated)

'ractional USE'hf~ USE ups USE M~ USE Lfaas her USE USE USE'~ USE'~-USE'ofhL,

~Plan ~Chan a ~Chan a ~ftdb ~fl-Ib CASE 1 ANO-1 73nooo 45.000 28.000 12 ~ 830 0. 192 0 384 55 724 17.276 CASE 2 ANO-1 73.000 47 F 000 26.000 12 ~ 830 0. 192 0 ~ 356 58.201 14.799 CASE 3 ANO-1 73.000 58 ~ 000 15.000 8.430 0. 126 0.205 66.396 6.604 CASE 4 ANO-2 154 F 000 147.000 F 000 3nloo 0.047 0.045 154. 169 -0.169 CASE 5 Angra 171nooo 155.000 16.000 Onooo 0.000 0.094 155. 000 16.000 CASE 6 Beaver Val 1 112.000 78.000 34aooo 19.900 0.299 0.304 111. 190 0 810 CASE 7 Beaver Val 1 112. 000 83.000 29.000 16.800 0 ~ 252 0.259 110.963 1. 037 CASE 8 Beaver Val 1, 112 ~ 000 88.000 24.000 15. 800 0.237 0.214 115.334 -3.334 CASE 9 Big Rock 95.000 65.000 30.000 17 900 0.269 0.316 88 '59 6. 141 CASE 10 Big Rock 95.000 70.000 25.000 15 a 300 0.230 0.263 90 850 4.150 CASE 11 Big Rock 95 F 000 57 ~ 000 38.000 12 F 100 0.182 0.4QO 69.640 25.360 CASE 12 Big Rock 95.000 80.000 15. 000 7 '00 0. 113 0. 158 90 141 4 859 CASE 13 Cal cliffs 1 160 000 F 119.000 41. 000 12 200 0 183 0.256 145 655 14.345 CASE 14 Cal Cliffs 2 137.000 105.000 32.000 10. 900 0. 164 0.234 125.523 11. 477 CASE 15 Callavay 1 112.000 97 F 000 15.000 8.000 0. 120 0.134 110 '27 1.773 CASE 16 Callavay 1 112 F 000 101.000 11.000 0.000 0 ~ 000 0 098 101.000 11.000 CASE 17 Catawba 1 128.000 123 F 000 5.000 4.300 0.065 0 039 131.480 -3.480 CASE 18 Crystal Riv3 79 F 000 68 F 000 11.000 10.680 0.160 0 ~ 139 80.972 -1.'72 CASE 19 Crystal Riv3 79.000 64aooo 15.000 lon480 0.157 0. 190 75.937 3 063 CASE 20 Crystal Riv3 79 F 000 70.000 9.000 1.880 0." 028 0. 114 72.031 6.969 CASE 21 Crystal Riv3 79 F 000 63.000 16.QQQ 1Q ~ 6&0 0. 160 0.203 75.018 3.982 CASE 22 D.C. Cook 1 110.000 80.000 30.000 19. 050 0 286 0.273 112.006 -2 '06 CASE 23 D.C. Cook 1 110.000 94.000 16.000 22.750 0.341 0.145 142.694 -32.694 CASE 24 Davis Besse 70.000 57 ~ 000 13 ~ 000 10 '00 0.153 0.186 67.296 2.704 CASE 25 Davis Basso 70.000 54 ~ 000 16. 000 10. 500 0.158 0 229 64.095 5.905 CASE 26 Davis Besse 70 ~ 000 64.000 6.000 5 F 500 0.083 0.086 69a755 0.245 CASE 27 Davis Besse 72.000 62.000 10. 000 11 ~ 800 0.177 0. 139 75.334 3.334 CASE 28 Davis Besse 72.000 65 000 F 000 5.500 0 083 0.097 70.845 1. 155 CASE 29 Diablo Can 1 98 F 000 87.000 11. 000 14.700 0. 221 0. 112 111. 610 -13.610 CASE 30 Diablo Can 2 121 F 000 85aooo 36.000 16.600 0.249 0.298 113.182 7.818 CASE 31 Dresden 2 71. 000 52 ~ QQQ 19.000 17. 600 0.264 Qa268 70.652 0.348 CASE 32 Dresden 3 65.000 45.000 20.000 20 '70 0.306 0,308 64.799 0 201 CASE 33 Dresden 3 65.000 41.000 24.000 25.170 0.378 0 '69 65.869 -0.869 CASE 34 Dresden 3 65.000 42.000 23 F 000 16. 570 0 249 0 ~ 354 55 '92 9 108 CASE 35 Dresden 3 65.000 65.000 0.000 0. 400 0 ~ 006 Oaooo 65.392 0.392 CASE 36 Dresden 3 70 F 000 59.000 11. 000 20.830 0.312 0.157 85 812 -15.812 CASE 37 Dresden 3 70.000 64.000 6 000 13.230 0.198 0.086 79.845 9 845'.198 CASE 38 Dresden 3 70 F 000 70.000 0.000 2.030 0 030 0 ~ 000 72. 198 CASE 39 Duane Arnold 101. 000 101 ~ 000 Onooo 0.000 0.000 0.000 1O1.000 0.000 CASE 40 Farley 1 149;000 108.000 41.000 6 '50 0 ~ 098 0 '75 119.767 29.233 CASE 41 Farley 1 149.000 130 F 000 19. 000 12.250 0.184 0 '28 159.265 -10.265 CASE 42 Farley 1 149.000 115.000 34 000 8.150 0.122 0.228 131 017 17. 983 CASE 43 Farley 2 144.000 144.000 0 ~ 000 1.350 0. 020 Onooo 146.976 2.976 CASE 44 Farley 2 144.000 144.000 0 000 6.45o 0.097 o.ooo 159.424 -15 '24 CASE 45 Farley 2 144.000 132.000 12.000 5.350 0. 080 0.083 143 '17 0 483 CASE 46 Ft. Calhoun 104 000 59.000 45.000 21. 100 0. 317 0 '33 86.320 17.680 CASE 47 Ft. Calhoun 104.000 65.000 39.000 22. 100 0.332 0.375 97.233 6.767 CASE 48 Garigliano 103.400 79 '00 23.800 14.930 0.224 0 230 102.571 0 829 CASE 49 Garigliano 103.400 67.30Q 36a100 36 '10 0.546 0.277 0 ~ 349

'20 148.287 83.028

-44.88'7 20 '72 CASE 50 Garigliano 103.400 60aooo 43.400 18,490 0 28

Table ~

LWR Weld Data and Linear Yietd Strength Model Estimates of the Unlrradiated USE (Continued)

Unirradiated Irradiated Unirradiated USE USE Calculated Measured USE (Measured) (Measured) Fractional Fractional (Catculated)

USE ups USE uzi USE u~ USE u~ USE USE USE'cu USE4~-USE'cu Plant ~Ch>>>>>> ~Cha>>>> ~hdh ~hdh CASE 51 Garigliano 103.400 51. 400 52.000 24.890 0.373 0.503 82 '23 21. 37?

CASE 52 Garigliano 103.400 83.900 19. 500 0.000 0.000 0.189 83.900 19.500 CASE 53 Garigliano 103.400 103.400 0.000 0.710 0 ~ 011 0.000 104. 513 -1. 113 CASE 54 Garigliano 103.400 73 800 29.600 23.750 0 356 0.286 114.641 -11.241 CASE 55 Ginna 80.000 53.000 27.000 21.750 0.326 0.338 78.664 1.336 CASE 56 Ginna 80.000 51.000 29.000 22.850 0.343 0.363 77.596 2 '04 CASE 57 Ginna 80 000 50.000 30.000 19. 650 0.295 0.375 70.897 9.103 CASE 58 Haddan Neck 105.000 83.000 22.000 le ?50 0.251 0. 210 110. 851 -5.851 CASE 59 Indian Pt. 2 118.000 75.000 43 000 25 800 0. 387 0.364 122.349 -4 349 CASE 60 Indian Pt. 3 120.000 68.000 52.000 24 '00 0.372 0.433 108.280 11. 720 CASE 61 Kewaunee 126 000 78 000 48.000 27.750 0. 416 0. 381 133.619 -7 619 CASE 62 Kewaunee 126.000 82.000 44.000 23.650 0.355 0.349 127.083 -1. 083 CASE 63 Maine Yankee 105.000 57.000 48.000 30 670 0.460 0.457 105.565 -0.565 CASE 64 Maine Yankee 105.000 59.000 46.000 23.720 0.356 0.438 91. 586 13>>414 CASE 65 Maine Yankee 105.000 66.0ao 39 000 24.970 0 ~ 375 0. 371 105.524 -0.524 CASE 66 Maine Yankee 105.000 50.000 55.000 36. 170 0.543 0.524 109.302 -4 '02 CASE 67 McGuire 1 112.000 83.000 29.000 19. 900 0 '99 0 '59 118.31& -6 318 CASE 68 McGuire 1 112. 000 75 000 37 000 13.800 0.207 0.330 94.578 17 ~ 422 CASE 69 McGuire 2 133.000 133>>000 0.000 10 ~ 950 0.164 0 ~ 000 159. 138 -26.138 CASE 70 McGuire 2 133.000 133 000 -0 ~ 000 7.350 0 '10 0>>000 149.480 -16. 480 CASE 71 Millstone 2 132.000 98.000 34 000 6.070 0.091 0.258 107.817 24.183 CASE 72 Millstone 2 132.000 108.000 24.000 10.270 0. 154 0. 182 127.667 4.333 CASE 73 North Anna 1 95.000 95 000 0>>000 2. 650 0.040 0.000 98.933 -3.933 CASE 74 North Anna 1 95 000 92.000 3.000 10. &50 0.163 0.032 109.884 -14.884 CASE 75 North Anna 2 112 F 000 112 F 000 0.000 5.450 0.082. 0.000 121.971 -9.971 CASE 76 North Anna 2 115 000 92 000 23 000 2 '00 0.038 0.200 95.584 19 ~ 416 CASE 77 Oconee 1 64 000 52 000 12 ~ 000 19.300 0.290 0.188 73. 188 -9 188

~

CASE 78 Oconee 1 64>>000 52.000 12. 000 16.700 0 ~ 251 0. 188 69.380 -5 '80 CASE 79 Oconee 1 64.000 55.000 9.000 13. 500 0.203 0. 141 68 '66 -4.966 CASE 80 Oconee 2 67.000 44 000 23.000 18. 770 0.282 0 343 61.243 5.757 CASE 81 Oconee 2 6z.oao 4?.aao 20.000 16 ~ 420 0 246 0 299 62.359 4 '41 CASE 82 Oconee 2 68.000 54.000 14 F 000 10.770 0,162 0.206 64.405 3.595 CASE 83 Oconee 3 66.000 49 F 000 17.000 15. 700 0.236 0.258 64.094 1 906 CASE 84 Oconee 3 66.000 58.000 8.000 5.200 0.078 0 ~ 121 62.907 3>>093 CASE 85 Oconee 3 66.000 42 F 000 24.000 19 ~ 200 0 ~ 288 0.364 58.989  ? F 011 CASE 86 Palisades 118 ~ 000 52.000 66.000 32 '60 0.490 0.559 101. 941 16>>059 CASE 87 Palisades 118 F 000 64.000 54 F 000 25.510 0.383 0.458 103.669 14>> 331 CASE 88 Prarie Isl 1 79 000 79 000 0.000 8.050 0. 121 0.000 89.849 -10 ~ 849 CASE 89 Prarie Isl 1 79>>000 75>>000 4.000 12. 650 0.190 0. 051 92.564 -13*564 CASE 90 Prarie Isl 1 79>>000 79.000 0.000 4 950 a.az4 0.000 85.336 -6>>336 CASE 91 Prarie Isl 2 103,000 91. 000 12 F 000 17 ~ 000 0.255 0. 117 122.148 -19 148

~

CASE 92 Prarie Isl 2 103.000 100.000 3.000 8.450 0.127 0. 029 114 ~ 515 -11.515 CASE Prarie Isl 2 103 ~ 000 92.000 11. 000 7.850 0.118 0 ~ 107 104. 279 -1>> 279 CASE 94 Pt. Beach 1 65.000 54.000 11 ~ 000 19. 300 0.290 0 ~ 169 76. 003 -11. 003 CASE 95 Pt. Beach 1 65.000 55 000 10.000 20.300 0 '05 0.154 79 F 080 -14.080 CASE 96 Pt. Beach 1 65.000 51.000 14 ~ 000 19 '00 0.297 0 '15 72 '46 7 '46 CASE 97 Pt. Beach 2 6s.aao 42.000 23.000 18.070 0 ~ 271 0 354 57. 617 7 '83 CASE 98 Pt. Beach 2 65.000 47.000 18. 000 19.970 0 300 0.277 67.100 -2 '00 CASE 99 Pt. Beach 2 66.000 44.000 22.000 20 '70 0.309 0.333 63.634 2>>366 CASE 100 Quad City 1 72>>000 49.000 23 000 29.000 0>>435 0.319 86.726 -14 '26 29

4),

4

Table 34 LWR Weld Data and Linear Yield Strength Model Estimates ot the Unirradiated USE (Continued)

Unirradiated Irradiated Unirradiated USE USE Calculated Measured USE (Measured) (Measured) Fractional Fractional (Calculated)

~Plan USE'~ USE ~ USE4~-USE USE USE USE'~ USE4~-USE'~

CASE 101 Quad City 1 . 100.000 85.000 000 14.600 0 '19 0.150 108. 835 -8.835 CASE 102 Quad City 1 100.000 95.000 5.000 3.200 0.048 0.050 99.790 0.210 CASE 103 Quad City 1 100.000 75 F 000 25.000 18 '00 0.273 0.250 103 ~ 164 -3. 164 CASE 104 Quad City 2 87 000 41.000 46.000 29 '00 0.446 0.529 73.940 13. 060 CASE 105 Quad City 2 90.000 51.000 39.000 18.600 0 ~ 279 0 '33 70.735 19. 265 CASE 106 Quad City 2 125.000 107.000 18.000 0.400 0.006 0. 144 107.646 17.354 CASE 107 Quad City 2 125.000 89.000 36.000 8.100 0.122 0. 288 101. 309 23.691 CASE 108 Quad City 2 125.000 80.000 45.000 14.700 0. 221 0 360 102.630 22.370 CASE 109 Rancho Seco 68.000 51.000 17 ~ 000 14. 000 0.210 0.250 64.557 3.443 CASE 110 Rancho Seco 68.000 48.000 20.000 11.000 0.165 0 294 57.485 10.515 CASE 111 Rancho Seco 68.000 53.000 15. 000 25.050 0,376 0. 221 84.902 -16.902 CASE 112 Robinson 2 112.000 70.000 42.000 23.150 0.347 0.375 107.239 4.761 99.000 80.000 19. 000 16. 100 0 '42 0.192 105 '71 -6 471

'5.

CASE 113 SONGS 2 ~

CASE 114 Salem 1 104.000 75.000 29.000 21. 000 0.315 0.279 109.489 -5.489 CASE 115 Salem 2 111.000 79 000 32+000 18 ~ 700 0 ~ 281 0 288 109 '98 1. 202 CASE 116 Salem 2 111.000 74.000 37 000 20.700 0. 311 0.333 107.324 3. 676 CASE 117 Sequoyah 1 111. 000 82.000 29 F 000 10. 100 0 152 0.261 96. 641 14. 359 CASE 118 Sequoyah 1 111. 000 78.000 33.000 11. 800 0. 177 0;297 94.775 16. 225 CASE 119 Sequoyah 2 112. 000 109.000 3 000 5.550 Oe083 0.027 118.898 -6.898 CASE 120 Sequoyah 2 112.000 110.000 2.000 6.050 0.091 0.018 120.979 -8 '79 CASE 121 Shearon Har 94.000 82.000 12. 000 6.000 0.090 0.128 90'10 3.890 CASE 122 St. Lucie 1 144.000 108.000 36.000 6.930 0.104 0.250 120. 529 23.471 CASE 123 St. Lucie 1 144.000 100.000 44.000 9.330 Ool40 0 306 116.272 27.728 CASE 124 Surry 1 70.000 53.000 17.000 19. 750 0 296 0.243 75.311 -5. 311 CASE 125 Surry 1 70.000 50.000 20.000 24.350 0 '65 0 286 78 ~ 771 -8.771 CASE 126 Surry 2 90.000 60.000 30.000 18.500 10.300 0 '78 0 333 83 '45 6.955 CASE 127 Surry 2 90 F 000 70.000 20.000 0. 155 0 222 82.791 7.209 CASE 128 THZ 1 81. 000 50 000 31. 000 14. 870 0.223 0 383 64.354 16.646 CASE 129 THZ 1 81 000 64.000 17 ~ 000 11.750 0 176 0.210 77.693 3.307 CASE 130 Tro an 83.000 81.000 2.000 3.600 0.054 0.024 85.624 -2.624 CASE 131 Trojan 83.000 62.000 21.000 6.300 0. 095 0.253 68.470 14.530 CASE 132 Tro an 83.000- 83.000 0.000 6 300 0. 095 0.000 91.662 -8.662 CASE 133 Turkey Pt. 3 65 F 000 48.000 17.000 20.850 0.313 0.262 69.844 -4.844 CASE 134 Turkey Pt. 3 65.000 59.000 6.000 18.550 0.278 0.092 81.746 -16.746 CASE 135 Turkey Pt. 4 66.000 44.000 22.000 21.040 0 ~ 316 0.333 64.290 1 ~ 710 CASE 136 V.C. Summer 91.000 87.000 4 F 000 0. 800 0. 012 0.044 88.057 2.943 CASE 137 V.C. Summer 91. 000 85.000 6.000 5.000 0.075 0.066 91.892 -0.892 CASE 138 Vermont Yank 107.000 107 ~ 000 0.000 4.800 0.072 0 ~ 000 115. 302 -8.302 CASE 139 Wolf Creek 100.000 92 F 000 8.000 0.900 0. 014 0.080 93.259 6.741 CASE 140 Zion 1 64 000 44 000 20+000 18 ~ 200 0 ~ 273 0.313 60.523 3 '77 CASE 141 Zion 1 64.000 52.000 12. 000 16.000 0 ~ 240 0 188 68.421 4.421 CASE 142 Zion 1 68.000 56.000 12 000 13.400 16.000 0.201 0. 176 70 '88 2. 088 CASE 143 Zion 1 68.000 49 000 19 ~ 000 0.240 0. 279 64.474 3. 526 CASE 144 Zion 2 69.000 50.000 19 ~ 000 18.500 0 278 0. 275 69 204 -0.204 CASE 145 Zion 2 69 000 51.000 18 000 24 F 500 0 368 0.261 80.632 -11. 632 30

S' Table 3-5 Elevated Temperature Yield Strength Uncertainty Estimation'lant Number of Two Sigma

~Ca cele Measurements Estimate ksi ANO-1 Unirradiated 3 60.4 3.6 ANO-2 Unirradiated 3 65.9 1.8 Calvert Cliffs 1 Unirradiated 4 66.9 2.1 Calvert Cliffs 2 Unirradiated ~

3 68.8 3.0

. Crystal River 3 Unirradiated 6 67.5 3.0 Dresden 1 Unirradiated 3 75.0 9.5 Dresden 1 CORE-6 4 99.1 4.3 Dresden 1 VANE 4 74.4 '3.5 Fort Calhoun Unirradiated 4 64.1 3.0 Humboldt Bay Unirradiated 3 57.5 , 5.3 Millstone 2 Unirradiated 3 66.7 1.3 Maine Yankee Unirradiated 6 61.1 4.8 Oconee 1 Unirradiated 3 56.4 3.5 Oconee 2 Unirradiated 3 69.8 1.4 Oconee 3 Unirradiated 3 67.4 1.7 Palisades Unirradiated 6 ' 62.1 3.8 Point Beach 2 Unirradiated 3 , 63.1 1,3 Rancho Seco 1 Unirradiated 61.6 0.6 St. Lucie 1 Unirradiated 3 65.9 1.6 St. Lucie 2 Unirradiated 3 59,2 3.0 TMI-1 Unirradiated 4 64,3 4.6

. Average 3.7 -3.2

'ata for weld tensile specimens tested in the 535'F to 650'F range. Individual averages and standard deviations are for tests performed over a 50'F range.

31

l l' y=x 200 y = 0.988x .

~ ~

~:

~ <50 ) "" ~

~ "" ~ """

~

~ ~ .

100 ge. ".

I W .g Q

50 Q

0 50 100 150 200 MEASURED USE (Ft-Lbs),

Figure 3-1 Non-Linear Yield Strength Model Uncertainty Plot Showing Measured Versus Predicted Unirradiated USE 32

1.0 I 08 Z

0.6 O

UJ

'g H 0.4 -------. ~-t.---':----

~:

O

)

0: 0 . y

~< 0.2 ~ L'"'"'""""" '""""'

)+: ~ ~

~ ~: ~

0.0 0 10 20 30 40 50 INCREASE IN YIELD STRENGTH (ksi)

Figure 3-2 Comparison of the Reference [OD86] Yield Strength Model (solid line) with Currently Available LWR Weld Data 33

ll f'i l

1.0

~0.8 m

Z 0.6 O

> 0.4 g ~

0Z t,j!

Q 4 ~~

~< 0.2

~ ~~:+ ~ ~

~ ~

0.0 0 10 20 30 40 50 INCREASE IN YIELD STRENGTH (ksi)

Figure 3-3 Yield Strength Model Based on Linear Correlation Between Fractional Decrease in USE versus ho 34

~ g k~

1.0 UJ c0 08 Z Linear Model UJ CO 0.6 CI O Non-Linear D Model

> 0.4 0Z ~:

0

~< 0.2 0.0 0 10 20 30 40 50 INCREASE IN YIELD STRENGTH (ksi)

Figure 3-4 Comparison of the Linear and Non-Linear Model Applied to LWR Weld Data

gc y=x 200 y = 0.979x

~ 10 l50 0

~ ~

CO

~

o 100

~y

..g.O

~',t 0

50 0

0 0 50 100 150 200 MEASURED USE (Ft-Lbs)

Figure 3-5 Linear Yield Strength Model Uncertainty Plot Showing Measured Versus Predicted Unirradiated USE 36

50 40 ~ ~ ~

I (9

~ 30 CO

'CI

~N

'l ~:

~ ~

20 '""'"" ~ ~ ~ " '"'"'""

. ~ '

LLj CO

'Occ 10 Z

~ <<4 g .V :

~ ..g.....r.

~

0 0.0 0.2 0.4 0.6 0.8 1.0 FRACTIONAL DECREASE IN USE Figure 3-6 Increase in Yield Strength versus Fractional Decrease in USE 37

I t'

4' I

IINQ UIIIY 1

- WELD IIATKIclRL 188 288 888 Deg.F 128

~ ~

S g S 88 68 0 I 5 48 ~

48 r

~ Baseline 0 A 1.83E19 ni'ca~

~ E 7.27E17 n/c~~

58 188 iSB 258 Deg.C Test Temperature Figure 3-7 Hyperbolic Tangent Fit of ANO Unit 1 Weld Metal Data HEAVER VALLEY UNIT I- WELD NAIERIAL

-iSB -58 58 158 258 358 458 Deg.F 188 128 Pn 128 88 A g I ox 88 68 ~

68 Baseline 0 lj 6.54E18 n/'cn> 28 28 V 2.55E18 n/etc~

9. 49E18 ni'ere~

8

-158 -188 -SB ,8 58 188 158 288 258 Deg. C Test Temperature Figure 3-8 Hyperbolic Tangent Fit of Beaver Valley Unit 1 Weld Metal Data 38

• 4 C' V

'lP g' ~ I'

DRESDEN UHII 2

- HELD NAIERIAL {MDRZAZ) 8 M 158 258 358 458 558 Deg.F 168 128 Baseline 148 0 2 1.87E19 nf'cN<

'~ 4 6.48E18 ntcte>

128 ~5 4.65E19 n/ce>

88

~

I~

I ox 88 68 ~

68' 8 M 188 158 288 258 388 Deg.C Test Temperature Figure 3-9 Hyperbolic Tangent Fit of Dresden Unit 2 Weld Metal Data DRESDEN UHII 3

- HELD NAIERIAL (WDR3AZ)

-58 58 158 2M 3M 458 De@. F I Baseline 0 12 2.86E19 n/'err<

~ 14 6.15E18 n/co<

~4 1.28E19 n/ere~

188 0 6 2.71E16 n/cN~

~e W>> 88 68 ~

I W 9 68 I o 0 48 48

-188 -58 58 188 158 288 258 De@. C Test Temperature Figure 3-10 Hyperbolic Tangent Fit of Dresden Unit 3 Weld Metal Data 39

fl, FARLEY UNIT I- WELD NATERIAL 58 158 258 358 458 Deg.F 228 168

~ ~

5 5 148 188 168 148 188 g~ 128 V 88 D~Q 188 4c 88 R Baseline M

'68 0 U 1.65E19 n/cN~

~ X 2.88E19 n/ca~

5.83E18 n/co~

I o~

58 188 158 288 258 Deg.C Test Temperature Figure 3-11 Hyperbolic Tangent Fit of Farley Unit 1 Weld Metal Data FORT ILHOUN UNIT i-

• WELD NATERIAL 58 158 258 358 459 Deg.F 148 188 128 188 88 68 7 4c 68 48 ~ Baseline 0 W225 4.29E18 n/cN> .28

~ W265 S.BBEiS n/co<

58 188 158 288 258 Deg.C Test Temperature Figure 3-12 Hyperbolic Tangent Fit of Foit Calhoun Unit 1 Weld Metal Data 40

ih ~

RENAUNEE

- NELD NATERIAL 58 158 258 358 458 Deg. F 288 188 R Basel ine 128 168 0 P 2 . 89E19 n/cN~

~ RR 2 . 87E19 n/ca~ ~

148 ~ U 5 . S9E18 nt'cN~ a e 128 88 ~

~ g 198 88 68 68

-158 -189 -59 9 58 198 158 289 258 Deg.C Test Temperature Figure 3-13 Hyperbolic Tangent Fit of Kewaunee Weld Metal Data QUAD CITIES UNIT 2

- MELD NATERIAL (NQC282)

-58 M 158 2M 358 458 Deg.F 5 Baseline 128 0 12 8.97E18 nf'c~~

~3 2.43E19 n/'ca~

198 R r-

~e 88 68 ~

4l I

0 4c 68 48 9

M 188 158 288 258 Deg. C Test Temperature Figure 14 Hyperbolic Tangent Fit of Quad Cities Unit 2 Weld Metal Data 41

1& A J

L p

SOHGS UHII i - WKLII HAIERIAL 58 158 258 358 458 Deg.F 168 148 188 128 i7) 188 88 68, 7 4c 68 0 0

~ Baseline

~ 0 0 A 1.28E19 nf'cv>

28 r F 5.14E19 n/c~~

-188 -58 58 188 158 288 258 Deg.C Test Temperature Figure 3-15 Hyperbolic Tangent Fit of SONGS Unit 1 Weld Metal Data ZII UHII 2 - WELII HAIERIAL

-58 58 158 258 458 Deg.F 128 Pn 88 68 I 0% ~ 48 ~

~

48

~ Basel ine 0 T 1.18E19 n/c~~

~ Ij 2.88E18 n/ca~

58 188 158 288 258 Deg.C Test Temperature Figure 3-16 Hyperbolic Tangent Fit of Zion Unit 2 Weld Metal Data 42

NORMALlZED WELD USE DATA FOR l0 PLANTS 200 I

G CC Lu 150 Z

UJ I-O CL Zion 2 IOO O SONGS 1

~o Q Q

+ Quad City 2 CC 0 Kewaunee cI Fort Calhoun Q

O 50 I> Farley 1 LIJ <3 Dresden 3 N

v Dresden 2

< Beaver Valley CC O 0 0 ANO-1 0 100 200 300 400 TEST TEMPERATURE (F) t Figure 3-17 Linear Regression Fit to Normalized Upper Shelf Weld Data 43

~

5 gV

'A J'

~

P II

4.0 Summary and Conclusions Based on the analyses reported herein,.it has been concluded that there is a positive linear correlation between fractional decrease in USE (f) and increase in tensile yield strength (ho). The following functional form can be used to accurately predict "f",

provided hadata are available:

f = 0.015 d a (4-1) where f = fractional decrease in USE ho= increase in yield strength (ksi)

In cases where surveillance capsule data are available, the unirradiated USE can be estimated as follows:

USEIRR U$ E' (4-2)

(1-f) where USE' unirradiated USE (ft-Ibs)

USE'"" = irradiated USE (ft-Ibs)

The uncertainty in estimating the unirradiated USE using this model was determined by applying the model to 145 LWR welds for which USE; USE'~, and badata are available.

The two sigma limit was determined to be 23.69 ft-lbs.

The model was applied to the NMP-1 surveillance weld and the unirradiated USE was calculated to be 126 ff-lb. Subtracting the two sigma of 24 ft-Ib yields a lower bound estimate of 102 ft-lbs. This value is consistent with the 100 ft-Ib value obtained earlier

[MA92a] using engineering judgement. It is NMPC's position that the surveillance weld Charpy behavior is representative of the beltline weld material behavior. The beltline welds and surveillance weld were manufactured by the same suppliers, the weld wire type and flux type are the same, the welding procedure used is the same, and the Cu and Ni content of the surveillance weld is representative of the beltline welds. Therefore, based on the. analyses reported herein, the unirradiated USE for the NMP-1 beltline welds is conservatively estimated to be 102 ft-lbs. This result, which is based on an in-depth statistical analysis, is consistent with earlier estimates of 90 ft-Ibs based on engineering judgement, Therefore, as reported in Reference [MA92a], none of the NMP-1 beltline weld materials will exhibit USE levels below 50 ft-Ibs prior to EOL.

44

1

,1 C

5.0 References

[ABB92] ABB/CE (T.G. Murray) letter to Yang P. Soong, NMPC, dated November 4, 1992, Re: Nine Mile Point Unit 1 Reactor Vessel Materials".

[CE90] "Niagara Mohawk Power Corporation Nine Mile Point Unit 1 Reactor Vessel Weld Materials", Report No.'6390-MCC-001, ABB Combustion Engineering Nuclear Power Combustion Engineering, Inc., Windsor, Connecticut, June, 1990.

[LE64] Lewis, S.R., Welding Material Qualification to Requirements of NAV Ships 250-1500-1, Metallurgical R&D, Combustion Engineering, Sept. 1964-Feb.

1965.

I

[MA85a] Manahan, M.P., "Procedure for the Determination of Initial RTN>> in Cases Where Limited Baseline Data are Available", November, 1985.

[MA85b] Manahan, M.P., Quayle, S.F., Rosenfield, A.R., and Shetty, D.K., "Statistical Analysis of Cleavage-Fracture Data", Invited paper, Conference" Proceedings of the International Conference and Exhibition on Fatigue, Corrosion Cracking, Fracture Mechanics, and Failure Analysis, Salt Lake City, December 2-6, 1985.

[MA901 Manahan, M.P., "Nine Mile Point Unit 1 RT>> Determination", Final Report from MPM Research & Consulting to NMPC, September 28, 1990.

[MA91] Manahan, M.P., "Nine Mile Point Unit 1 Surveillance Capsule Program",

NMEL-90001, January 4, 1991.

4

[MA92a] Manahan, M.P., Soong, Y., "Response to NRC Generic Letter 92-01 for Nine Mile Point Unit 1", NMPC Project 03-9425, June 12, 1992.

[MA92b] Manahan, M.P., Report entitled, "Upper Shelf Energy Drop Trend Curve Modelling", November 30, 1982.

[MA93a] Manahan, M.P., Final Report entitled, "Elastic-Plastic Fracture Mechanics Assessment of Nine Mile Point Unit 1 Beltline Plates for Service Level A and B Loadings", February 19, 1993.

[MA93b] , Manahan, M.P., Final Report to NMPC,entitled, "Elastic-Plastic Fracture Mechanics Assessment of Nine Mile Point Unit 1 Beltline Plates for Service Level C and D Loadings", dated February 22, 1993.

45

c>

~

r<

, [McFRAC] Manahan, M.P., et.al., "Statistical Analysis Methodology for Mechanics of Fracture", Final report to Battelle's Corporate Technology Development Office, 1984.

[NRC92] NRC letter to NMPC dated October 13, 1992, Re: Summary of September 30, 1992, Meeting to Discuss Licensee's Response to Generic Letter 92-01, "Reactor Vessel Structural Integrity", for Nine Mile Point Nuclear Station Unit No. 1.

[0D86] Odette, G.R., Lombrazo, P.M., "Relationship Between Irradiation Hardening and Embrittlement of Pressure Vessel Steels", Proceedings of the 12th ASTM Symposium on the Effects of Irradiation on Materials,.ASTM STP 870, pp. 840-860, 1986.

[PREDB] PR-EDB: Power Reactor Embrittlement Data Base, Version 1, NUREG/CR-4816, Revision 1, dated May, 1991.

[ST 84] Stahl, D., Manahan, M.P., Failey, M.P., Landow, M.P., Jung, R.G., and Lowry, L.M., "Examination, Testing, and Evaluation of Irradiated Pressure Vessel Surveillance Specimens from the Nine Mile Point Nuclear Power Station", Final Report from Battelle-Columbus to NMPC, July 18, 1984.

[SYSTAT] "SYSTAT 5.03: Statistical Code Package", SYSTAT, IncCopyright 1991.

l 46

A"j A