ML20078G943

EOC 6 Interim Plugging Criteria Rept

ML20078G943
Person / Time
Site:	Byron
Issue date:	01/19/1995
From:	Keating R, Pitterle T, Wepfer R WESTINGHOUSE ELECTRIC COMPANY, DIV OF CBS CORP.
To:
Shared Package
ML17332B027	List: ML17332B026 ML20078G932 ML20078G943
References
SG-95-01-003, SG-95-1-3, NUDOCS 9502030348
Download: ML20078G943 (113)

Text

ATTACHMENT A BYRON UNIT 1 END-OF-CYCLE 6 INTERIM PLUGGING CRITERIA REPORT O

)

i (9215ZZ/012795) 9502030348 950130 PDR ADOCK 05000454 P

PDR

(

SG-95-01-003 NSD-JLH-5006 n

BYRON UNIT 1 END-OF-CYCLE 6 INTERIM PLUGGING CRITERIA REPORT January 19,1995

\\

i Prepared By:

T. A. Pitterle R. F. Keating R. M. Wepfer Westinghouse Electric Corporation Nuclear Services Division P.O. Box 158 Madison, PA 15663 C 1995 WESTINGHOUSE ELECTRIC CORPORATION ALL RIGHTS RESERVED

Byron Unit 1 End-of-Cycle 6 Interim Plugging Criteria Report TABLE OF CONTENTS Section P.,gge 1.0 Introduction 1-1 2.0 Summary and Conclusions 2-1 2.1 Overall Conclusions 2-1 2.2 1994 EOC-6 Inspection Results 2-2 2.3 Voltage Growth 2-2 2.4 Projected MOC-7 and EOC-7 Voltages 2-3 2.5 SLB Leak Rate Analyses 2-3 2.6 Tube Burst Probability Assessments 2-3 3.0 Byron-1 Pulled Tube E amination Summa.7 3-1 3.1 Pulled Tube Examination Results 3-1 3.2 Evaluation of Byron-1 Pulled Tuf a Data for ARC Applications 3-5 3.3 Comparison of Byron-1 Data with Existing IPC Correlations 3-8 4.0 EOC-6 Inspection Results and Voltage Growth Rates 4-1 4.1 EOC-6 Inspection Results 4-1 4.2 Voltage Growth Rates 4-3 4.3 NDE Uncertainties 4-4 4.4 Distributions Requested by Draft Generic Letter 4-5 5.0 Database Applied for IPC Correlations 5-1 6.0 Analysis Methods for Voltage Projections, Leak Rates and Burst Probability 6-1 7.0 Altemate POD Distributions 7-1 i

8.0 Projected Cycle 7 Voltage Distributions 8-1 8.1 Limiting SG Evaluation 8-1 8.2 Input Distributions 8-1 8.3 Projected MOC-7 Voltage Distribution 8-2 8.2 Projected EOC.7 Voltage Distribution 8-2 9.0 SLB Leak Rate Analysis Results 9-1 9.1 SLB Leak Rates for EOC-6 Actual and BOC-7 Distribution 9-1 9.2 Projected SLB Leak Rates for Potential Mid-Cycle Voltage Distribution 9-1 9.3 Projected SLB Leak Rates for EOC-7 Voltage Distribution 9-1 i

i i

Byron Unit 1 End of-Cycle 6 Interim Plugging Criteria Report TABLE OF CONTENTS (Cont'd)

Section P_att 10.0 SLB Tube Burst Probability Analysis Results 10-1 10.1 SLB Burst Probability Based on Byron-1 Limited TSP Displacements 10-1 10.2 Free Span SLB Burst Probability for Actual EOC-6 Distribution 10 1 10.3 Projected Free Span SLB Burst Probability for Potential Mid-Cycle Voltage Distribution 10-2 10.4 Projected Free Span SLB Burst Probability for EOC-7 Distribution 10-2 10.5 Burst Probability Dependence on POD 10-2 11.0 References Il-1 t

e 4

Byron Unit 1 End-of-Cycle 6 Interim Plugging Crit <tria Report

1.0 INTRODUCTION

i i

This report provides the Byron-1 steam generator steam line break (SLB) leak rate and tube I

burst probability analysis results in support of the implementation of the 1.0 volt Interim Plugging Criteria (IPC) at end-of-cycle 6 (EOC-6). It supplies information required by the NRC Safety Evaluation Report (Reference 11.1) and the draft generic letter (Reference 11.2),

and it supplements the Byron 1 EOC-6 IPC letter report (Reference 11.3). The analysis results are provided for steam generator C, which is the limiting steam generator (SG) for the actual EOC-6 and for the projected EOC-7 bobbin voltage distributions. Conditional leak rate and SLB burst probability results are provided for EOC-6, beginning-of-cycle BOC-7, mid-cycle 7 (MOC-7) and EOC-7. A sensitivity analysis to the assumed probability of detection (POD) ir also provided for the NRC required POD = 0.6, a POD developed by the Electric Power Research Institute (EPRI) based on qualified data analyst (QDA) testing, a POD developed as the average of four plants that had IPC assessments performed, and an ideal POD = 1.0.

The Monte Carlo methods required by the draft generic letter are applied to the BOC voltage distributions to project the EOC distributions. This is consistent with the Nuclear Regulatory Commission (NRC) guidance given in the Byron-1 Safety Evaluation Report (SER). SLB leak rates are calculated using the NRC methodology of draft NUREG-1477, the database of the EPRI report in Reference 11.4 with modifications as discussed with the NRC, as well as the IPC/APC methods described in the Westinghouse methods report of Reference 11.5, which utilizes methods applying the IPC correlations for probability ofleakage and SLB leak rate. This methods report has been prepared to document the analysis methods applied to obtain SLB leak rate and burst probabilities consistent with the guidelines of the draft generic letter. The latest EPRI correlations are applied in this report. The database used for the leak rate and burst pressure correlations is summarized for completeness.

Three tubes were pulled from Byron-1 at EOC-6 to provide supporting leak rate, burst pressure and crack morphology data for application of the IPC at Byron-l. The pu31ed tube results are described in this report consistent with the guidance of the draft generic!!etter for the 90-day reporting requirements Table 4-9 relates the data requested by draft generic letter 94-xx to the report section, table and figure which provides the data.

\\apekae94Vyc6ipc\\ sect 2.wp3 11

2.0

SUMMARY

AND CONCLUSIONS 2.1 Overall Conclusions Three tubes with twelve tube support plate (TSP) intersections including three flow distribution baffle (FDB) locations were pulled at EOC-6 in support of application of the IPC at Byron-1. The pulled tubes included the largest bobbin indication of 10.95 volts found at EOC-6 as well'as 5.92 and 2.58 volt indications. Nine of the intersections were burst tested and destructively examined. Five intersectSns were leak tested of which two intersections (5.92 and 10.95 volts) had leakage. All leak and burst test results are consistent with the EPRI leak rate and burst correlations, and inclusion of the Byron-1 data in the correlations would result in negligible changes to the correlations. Crack morphologies found were dominantly outside diameter stress corrosion cracking (ODSCC) with some small cellular patches which is typical of the EPRI database. As expected, the 10.95 volt indication had the lowest measured burst pressure of 4,500 psi and the highest SLB leak rate of 22.1 liters /hr.

The Byron-1 pulled tube results show that the tube integrity requirements of Regulatory Guide (R.G.) 1.121 were satisfied at EOC-6 and a similar conclusion would be expected at EOC-7.

Although the calculated tube burst probabilities of 1.90 x 10 2 at EOC-6 and 3.29 x 10 2 for the projected EOC-7 with POD =0.6 exceed the draft generic letter guidance of 1.0 x 10 2, the pulled tube burst pressures show that the actual EOC-6 burst probability would be negligibly small. That is, the measured burst pressures for the largest,5.92 and 10.95 volt, indications i

exceed 4,500 psi, as compared to the SLB structural margin requirement of 1.43AP., of 3t 3,660 psi, while the largest indications dominate the calculated burst probability. This large l

Byron-1 burst margin is attributed to the Byron-1 tubes, similar to the Braidwood-l tubes, having bus pressures close to the mean of the burst pressure versus voltage correlation. Per the draft generic letter guidance to assess the safety implications of the calculated burst probabilities exceeding 1.0 x 10 2, it is concluded that the pulled tube results demonstrate that the EOC-6 burst probability was significantly less than 1.0 x 10 2 and R.G.1.121 tube integrity requirements were satisfied. Similar results would be expected at EOC-7.

Consistent with the Byron-1 SER for Cycle 7 and the draft generic letter, EOC-7 projections

[

were made by applying the NRC model of Draft NUREG-1477 to define the BOC distribution i

with a probability of detection (POD) adjustment factor of 0.6, and by applying Monte Carlo analysis to the correlations of burst pressure and leak rate with voltage to obtain the SLB tube burst probability and leak rates. The projections for Cycle 7 were based on SG-C which had the most limiting EOC-6 actual and projected EOC-7 voltage distributions.

The resulting EOC-7 SLB leak rate is estimated at 5.1 gpm which is well below the allowable limit of 12.5 gpm in the faulted SG for Byron-1 Cycle 7. The projected EOC-7 SLB total tube burst probability by Monte Carlo analysis is 3.29 x 10'2. Overall, it is concluded that SLB leakage acceptance limits are satisfied in Cycles 6 and 7, although the calculated tube burst probabilities exceed the draft generic letter guideline of 1.0 x 10-2 As noted above, however, the Byron-1 pulled tube results show burst pressures exceeding structural margin requirements, which results in negligibly low burst probabilities.

i s:\\apc\\cae9&cyc6ipe\\secl.2.wp5 2-1 i

i

~

2.2

- 1994 EOC-6 Inspection Results A total of 3075 ODSCC indications were reported at Byron-1 TSP intersections at EOC-6, of which 802 were in SG-A,914 in SG-B,899 in SG-C, and 460 in SG-D. The median voltage bin for EOC-6 indications among the four SGs was 0.6. volts (indications between 0.51 and 0.60, inclusive). RPC (rotating pancake coil) examination was performed of all indications greater than 1.0 volts, plus a number of smaller indications. Of the %7 RPC-tested indications,717 were confirmed, for a 74% overall confirmation rate. Above 1.0 volt,535 of 659 indications or 51% of indications were RPC confirmed, while below 1.0 volt,182 of 308 or 59% ofindications were confirmed. All indications greater than 2.2 volts were RPC confirmed. These are relatively high confirmation rates compared to other inspections.

The largest bobbin voltage indications found in 1994 were 10.95 volts in SG-C; 7.64,7.10 and 5.92 volts in SG-A; 4.56 volts in SG-B; and 4.09 volts in SG-D. The 10.95 and 5.92 volt indications were included among the pulled tubes, and it is shown in Section 3 that the measured burst pressures for these indications exceeded the R.G.1.121 structura' margins.

Thus, tube integrity requirements were satisfied at EOC-6.

A total of 659 indications exceeded the approved 1.0 volt IPC repair limit for this inspection; all of these were RPC inspected and 535 were confirmed and repaired.

I 2.3 Voltage Growth Growth rates were developed for all potential flaws. The growth rate analysis was performed I

in accordance with guidance of draft NUREG-1477 in that growth rates were not cajeulated for the 1993 indications which were reanalyzed as NDD (no detectable degradation). Thus the total number of voltage growth values (2851) is smaller than the total number of EOC-6 indications (3075).

The Cycle 6 average percentage growth of 67% is similar to the Cycle 4 and 5 values of 80%

and 68%, although the Cycle 4 and 5 data are for plugged tubes only and are therefore expected to be higher than for the entire population of Cycle 4 and 5 indications. SG-A i

experienced the largest growth rates during Cycle 6, an average of 117%, while SG-D, -C and 2

-B values followed at 69%,66% and 39%, respectively. In all cases, percentage growth rates j

were significantly lower for BOC voltages above 0.75 volts than for those below 0.75 volts.

i SG-C R20C7 3H had the largest Cycle 6 growth value of 9.86 volts, which was based on an EOC-6 voltage of 10.95 volts and BOC voltage of 1.09 volts. This intersection was pulled and metallurgical results are included in Section 3.1 of this report. Three other indications, all in SG-A, had growth values larger than 5.0 volts: R3C3 3H (6.71 volts), R20C7 3H (5.96 volts), and R3C107 3H (5.48 volts). All of the remaining voltage growths were less than 4.0 volts.

snapskee9&cyc6ipcVect,2wp5 2-2 l

y n,.

~._.

2.4 Projected MOC-7 and EOC-7 Voltages r

SG-C BOC distributions were prepared for each of four POD methods: POD =0.6; PODI, developec by EPRI based on qualified data analyst (QDA) testing; POD 2, developed as the average of four plants that had IPC assessments performed; and a POD =1.0.

The 1215 indications for a POD of 0.6 is about twice as many as the 616 indications for a POD of 1.0.

I POD 1 (EPRI) and POD 2 (POPCD) provide intermediate BOC distributions to the 0.6 and 1.0 PODS, with POD 2 being slightly more conservative in the total number of indications and i

largest indications. The largest projected voltages for MOC-7 and EOC-7 for the POD =0.6 adjusted BOC-7 distribution were 11.5 and 11.7 volts, respectively.

2.5 SLB Leak Rate Analyses The projected EOC-7 leak rate for SG-C with a POD = 0.6 (NRC draft generic letter) is 5.1 gpm. These leak rates are less than the allowable SLB leak rate for Byron-1 and thus are acceptable to the guidelines of the NRC draft generic letter.

SLB leak rates were also calculated for the projected EOC-7 distribution based on the alternate POD distributions. The more realistic, voltage-dependent PODI and POD 2 analyses result in SLB leak rates of 3.1 and 3.5 gpm, which approach the ideal POD = 1.0 resul. f 1.7 gpm. This is as expected, since the PODI and POD 2 distributions more appropriately reflect l

tha, all of the larger bobbin indications (29 above 1.8 volts with 3 above 2.7 volts) were plugged at EOC-6. The POD = 0.6 resui's in 0.67 indications left in' service for each plugged indication, while the voltage-dependent PODI and POD 2 yield smaller fractional indications left in service below 3.0 volts and no indications left in service above 3.0 volts.

i 2.6 Tube Burst Probability Assessments l

The total (sum of single and multiple tube bursts) tube burst probabilities for the assumed POD = 0.6 are 1.90 x 10 ' at EOC-6 and 3.29 x 10'2 for the projected EOC-7 distribution.

The total and single tube burst probabilities are approximately equal since the probabilities of 4

multiple tube ruptures are small at 4.74 x 10 at EOC-6 and 1.46 x 10-' at EOC-7. No occurrences of multiple tube ruptures with more than two ruptures were found in the Monte i

Carlo analyses. The reported burst probability values for three tube ruptures reflect only the 95% confidence limit for the number of'Ionte Carlo samples with no occurrences. A few occurrences of multiple ruptures (two) occurred but the burst probabilities are significantly lower than the single tube burst probability.

[

l I

s:\\apckne94Wyc6ipckl_2.wp$

23 I

i

3.0 BYRON-1 PULLED TUBE EXAMINATION

SUMMARY

3.1 Pulled Tube Examination Results 3.1.1 Introduction Three hot leg steam generator tube segments from Byron Unit I (Tube R3C107 from SG A and Tubes R20C7 and R20C102 from SG-C) were examined at the Westinghouse Science and Technology Center in support the Byron-1 IPC. The examination was conducted to characterize corrosion at steam generator hot leg support plate crevice loc'tions. The tubes were selected to obtain a sampling of the indications observed in the 1994 field eddy current inspection. The first (Flow Distribution Baffle), third, fifth and seventh support plate crevice regions (IH,3H,5H and 7H) were removed for examination. Four of these locations had original field eddy current indications of outside diameter (OD) origin.

l After nondestructive laboratory examination by eddy current testing, ultrasonic testing, radiography, dimensional characterization and visual examination, five support plate regions were leak tested at elevated temperature. Subsequently, room temperature burst testing w:as conducted on the' five TSP regions, as well as four non-leak tested TSP regions and a free span section fror. each of the three tubes pulled. Four of the burst tested TSP specimens were destructively examined using metallographic and scanning electron micrograph (SEM) fractography techniques to characterize the corrosion, and an additional four TSP burst tested specimens were examined using only SEM fractography of the burst crack opening. As of the date of this report, the SEM fractographic examinations have been completed, but no metallographic data are yet available.

3.1.2 Non-Destructive Examination (NDE) Results Table 3-1 presents a summary of the more important field and laboratory NDE results. The eddy current data were reviewed, including reevaluation of the field data, to finalize the voltages assigned to the indications and to assess the field no detectable degradation (NDD) calls for detectability under laboratory analysis conditions. A single analyst performed this work to minimize data variability.

Field and laboratory eddy current inspections (l'obbin and RPC probes) produced similar data for most regions examined. The field bobbin data for the field NDD calls at TSP locations were reevaluated to derive the most appropriate amplitude measurements, where possible, for these very small signals. This review indicated that four field NDD calls (R3C107 5H; R3C107 7H; R20C7 5H; and R20C7 7H) could be assigned a bobbin flaw voltage. For these four field NDD calls, the bobbin signals were reevaluated as distorted indications wherein the selection of the flaw segment was aided by use of the 300 kHz data. The reevaluated field RPC data for these four NDD calls continued to be NDD with no discernible flaw separable from the background level. In the laboratory, these four TSP locations were identified to have both bobbin and RPC indications, except for R20C7 7H which was dented by the tube 3-1

1 pulling' operation. (The dent obscured any potential bobbin signal although an indication was.

J observed by the RPC inspection.)

Some increase in signal strength (voltace) was observed in the laboratory eddy current inspections, which was attributed to the tube pulling operation. Field bobbin probe signal l

strengtns ranged from 1.25 to 10.95 volts (1.11 and 10.9 volts after data reevaluation) while post-pull bobbin strengths ranged from 2.05 to 13.8 volts. The largest increase was for Tube R20C102 SH, where the bobbin probe signal strength increased from 1.11 volts to 4.64 volts.

This is considered a moderate increase suggesting that there was no significant tearing of i

ligaments between microcracks.

1 All TSP region NDE indications were confined to their crevice regions. Each of the three IH regions (flow distribution baffle) had no detectable degradation (NDD) in the field and laboratory eddy current and ultrasonic (UT) examinations and in the laboratory radiographic examination. Some of the eddy current TSP indications had considerable width in the RPC i

data, and in the laboratory UT data, suggesting the possibility of intergranular cellular corrosion (ICC) or three dimensional intergranular attack (IGA) in addition to and in association with axial cracking. UT inspections confirmed the eddy current observations and further suggested the presence of even more extensive corrosion at some TSP crevice locations. Radiographic indications, all axial ones, were observed for the three support plate regions with the largest bobbi probe signal strengths.

3.1.3 Leak Testing Five TSP crevice regions, which included the four original fiel! 3ddy current indications and one of the field reevaluated distorted eddy current indications, were leak tested at elevated temperature and pressure at conditions that ranged from a simulated normal operating condition to that of a simulated steam line break condition. Two of the specimens, the ones with the largest voltages, developed leaks: Tube R3C107 3H (5.3 volt field bobbin indication) and Tube R20C7 3H (10.1 volt field bobbin indication). The leak rates ranged from 0.047 liters / hour to 0.065 liters / hour for normal operating conditions and from 4.70 liters / hour to 18.3 liters / hour for steam line break conditions. Table 3-2 presents leak rate data for the five tested specimens.

i As shown in Table 3-2, some of the leak tests were repeated with minor variations in temperature and pressure conditions for the tests. No significant diff'erences were found in the leak rates between repeat tests.

3.1.4 Burst Testing Nine TSP crevice regions and three free span regions were burst tested at room temperature at a pressurization rate of 1000 to 2000 psi per second. The burst tests were performed simulating free span conditions with no TSP enveloping the indications, and the test methods were the same as those employed in developing the industry database in Reference 11.4. The j

3-2

~

field indication specimens were tested using a bladder and foil for the burst tests with a i

" semi-constreint" condition which simulated the lateral constraint provided by the TSPs located below and above the crack indication at prototypical spacing between TSPs. Results of the burst tests are presented in Table 3-3. Due to the limited crack opening and crack length extensie for the initial burst tests of R3C107 3H and R20C7 3H, the burst tests for these indications were repeated using a slightly larger foil and a faster (2000 psi /sec) pressurization rate. The retest was successful in obtaining crack extension for R20C7 3H but two retests did not result in crack extension for R3C107 3H. The larger burst pressure of i

4500 psi is the appropriate value for,R20C7 3H while the initial 4800 psi represents a minimum burst pressure for R3C1071H. All burst specimens develope ! axial burst openings.

The openings for the TSP crevice regioin specimens were centered within the crevice regions.

The circumferential positions of the burs:t openings in the support plate crevice region specimens were the same as the location of the deepest UT indications for the specimens j

having corrosion indications. The eddy current RPC data does not provide an absolute circumferential position. The lowest burst pressure for the TSP crevice regions (Tube R20C7 3H, the 10.9 volt indication) was 4,500 psi,39% of the burst pressure of its free span equivalent and typical of a 0.75 inch diameter specimen with a 10.9 volt indication.

1 Table 3-3 also provides room temperature tensile properties obtained from free span sections of the tubes. The tensile and burst strengths for the free span sections are typical for Westinghouse tubing of this type.

j i

3.1.5 Destructive Examination Results Figures 3-1 to M present sketches of the crack distributions found by visual (30X stereoscope) examinations. The sketches show the locations where cracks were found and r

their overall appearance, not the exact number of cracks or their detailed morphology. While no metallographic data is yet available, the visual inspection data and the SEM inspection data suggested that intergranular cellular corrosion was present in addition to the more prevalent axial intergranular stress corrosion cracking (IGSCC). The corrosion in all TSP regions was centered within and confined to the crevice regions.

A summary of post-burst test visual inspection data and of burst property data related to the presence of corrosion is presented in Table 3-4 for each of the burst specimens. The data in Table 3-4 was used to plan destructive examination work efforts. Corrosion cracks were observed on eight of the nine of the burst tested TSP specimens. These eight specimens were candidates for destructive examina' ion. The only burst tested TSP region without corrosion was the only 1H region (flow disaibution baffle region) tested, R20C71H. All four j

specimens with original field NDD calls that were subsequently called low voltage distorted indications (Tube R3C107 SH, Tube R3C107 7H, Tube R20C7 SH, and Tube R20C7 7H) had corrosion present. The free span sections of Tubes R3C107, R20C7 and R20C102, selected for a reference burst pressure and tensile property test, had no degradation, as would be i

expected. Four specimens, shown in Table 3-4, which included both specimens with

(

measured SLB leakage, were selected for complete destructive examination of their crevice l

3-3

)

l 1

1

__.,m.

region corrosion. These examinations include SEM fractography of the burst openings and metallography of secondary corrosion within the crevice regions. An additional four specimens, the low voltage distorted indication specimens, were characterized by SEM fractography of their burst openings.

A The burst fracture faces of these selected eight TSP crevice region specimens were opened for SEM fractographic examinations. Table 3-5 presents the results of the fractographic data in the form of macrocrack length versus depth, macrocrack length / average and maximum depth, and the number / location / width of ductile or uncorroded ligaments found on the fracture face.

The burst openings occurred in axial macrocracks that were composed of numerous axially oriented intergranular microcracks of OD origin. Ductile ligaments separating the microcracks

.were present in all eight examined specimens that had corrosion. The data of Table 3-5 indicate that most of the TSP regions from Byron Unit I pulled tubes had a typical number of f

remaining uncorroded ligaments between microcracks comprising the burst macrocracks. All intergranular cormsion was confined to and centered within the crevice regions. The burst opening macrocrack depths and lengths for the TSP crevice regions had maximum depths ranging from 26% to 100% throughwall, with average depths ranging from 12% to 69%

throughwall and with macrocrack lengths ranging from 0.38 to 0.602 inch. Two of the eight specimens had throughwall burst cracks. Tube R3C107 3H had a crack that was throughwall for 0.13 inch.

l Tube R20C7 3H had two parallel throughwall cracks: one was throughwall for 0.24 inch at the 100 location and the other was throughwall for 0.268 inch at the 70' location. A third, partial depth crack was located at the 140" location. Only the 70' location throughwall crack had a ductile ligament present in the throughwall portion of the crack. (Since the 100*

location crack opened significantly more than the 70' location crack during the burst test, even though the two cracks were roughly the same size, it would appear that the 3 mil thick

{

ligament did not tear during the leak test and that most of the R20C7 3H leakage occurred through the 100' location crack.) Typical of the edges of burst openings, the horizontal markings shown on Figure 3-4 are expected to primarily indicate the opening of cellular corrosion between the two axially-oriented burst openings; preliminary metallographic data confirms the presence of cellular corrosion between the two burst cracks. Corrosion was centered within and confmed to the crevice region.

Four TSP regions were initially called bobbin NDD in the field and subsequently were found by reevaluation of the field data to have small (0.23 to 0.38 volts) indications prior to the tube exam. The maximum crack depths for these four locations were 26%,38%,45% and 53% for the 7H region of Tube R20C7, the SH region of Tube R20C7, the 7H region of Tube R3C107 and the 5H region of Tube R3C107, respectively. The corresponding average macrocrack depths were 12%,20%,21% and 32%, respectively.

3.1.6 Pulled Tube Examination Conclusions The visually inspected burst specimens from the TSP crevice regions of Tubes R3C107, 3-4

R20C7 and R20C102 had corrosion present at the locations that had eddy current and/or UT-indications. All eight of the nine burst tested TSP crevice regions found to have corrosion were from true support plate crevice regions (i.e., from 3H, SH or 7H where only a 0.008 inch nominal radial crevice gap existed). The ninth TSP burst specimen (Tube R20C7 IH) was without corrosion and was from a flow distribution baffle region where the nominal radial crevice gap was 0.039 inch. While metallographic data are not yet available, visual and SEM inspection data suggest that the corroded TSP crevice regions had combinations of axially oriented OD origin IGSCC and ICC. All corrosion was confined to the crevice regions. The examinations performed to date show the corrosion morphology to be typical of pulled tubes within the EPRI database.

j Eddy current bobbin and RPC probe data correlated well with the corrosion distribution for l

the deeper cracks. Four TSP crevice regions were initially called bobbin NDD in the field and subsequently were found by reevaluation of the field data to have small (0.23 to 0.38 f

volts) indications prior to the tube exam. Each of these regions had corrosion ranging from 26% to 53% throughwall, maximum depth. Consequently, these locations had corrosios near -

the eddy current detection threshold. These four locations also serve in determiair.a the UT detection threshold. Three of the four were called by field UT inspections as 'aaving corrosion while two of the four were called by lab UT inspections as having arrosion. The maximum corrosion depth not called was 26% for the field inspection and 53% for the lab inspection. (Lab UT inspection data suggestad that surface deposits were responsible for the corrosion present.) Consequently, these locations had corrosion near the UT detection threshold.

Leak rate testing performed at elevated temperatures and pressures produced leak rates that ranged from 0.047 liters / hour to 0.065 liters / hour for normal operating conditions and from 4.70 liters / hour to 18.3 liters / hour for steam line break conditions for the two specimens which had leakage. The lowest field bobbin probe signal strength specimen which had leakage was 5.89 volts and the highest field bobbin probe signal strength specimen which did not have leakage was 2.58 volts. The TSP crevice region burst pressures ranged from 4,500 j

to 11,300 psi. All burst pressures were above safety limitations required by R.G.1.121 including the 10.9 volt indication at 3H of Tube R20C7. The burst tests.were performed simulating free span conditions with no TSP enveloping the indications. The leak and burst pressure data were generally consistent with the ARC burst pressure and leak rate correlations as described below.

3.2 Evaluation of Byron-1 Pulled Tube Data for ARC Applications The pulled tube examination results were evaluated for application to the EPRI database for alternate repair criteria (ARC) applications. The eddy current data were reviewed, including reevaluation of the field data, to finalize the voltages assigned to the indications and to assess the field NDD calls for detectability under laboratory analysis conditions. The data for incorporation into the EPRI database were then defined and reviewed against the EPRI outlier criteria to provide acceptability for the database.

3-5

-e 3.2.1 Eddy Current Data Review Table 3-6 provides a summary of the eddy current data evaluations for the Byron-1 pulled l

tubes. For the field indications, there is little difference in the bobbin voltage calls between the field and the laboratory results. This supports the field analyst training on voltage measurements while recognizing that the larger voltage indications typical of the field calls, are typically less difficult to size than the lower volta;e indications such as below 1.0 volt.

For inclusion of the data in the EPRI database, it is desirable to minimize analyst variability in the voltage calls since this variability is separately accounted for in ARC applications as an l

NDE uncertainty Most of the pulled tube EPRI database has been analyzed by the same analyst who performed the field reevaluation of Table 3-6. Thus the reevaluated field bobbin voltages are applied for application to the ARC correlations.

The field bobbin data for the field NDD calls were reevaluated to derive, where possible, the I

most appropriate amplitude measurements for these very small signals. This review indicated that four field NDD indications, R3C107 5H, R3C107 7H, R20C7 5H and R20C7 7H, could be assigned a bobbin flaw voltage. For each of these indications, called NDD in the field, the signals are distorted indications and selection of the flaw segment was aided by use of the 300 kHz data. The reevaluated field bobbin data for these indications are shown in Figures 3-9 to 3-12. The reevaluated RPC data continues to be NDD with no discemible flaw separable from the background level. The reevaluated bobbin voltages for these four indications are used for application to the ARC database and correlations.

2 The data for all three tubes at lH (FDB) were found to be NDD by all field and laboratory j

eddy current and UT inspection results. The indications at R3C107 SH ar:d 7H, and R20C7 l

5H, which were field NDD but had identifiable indications by reevaluation of the field data, were identified as having both bobbin and RPC indications in the laboratory post-pull bobbin a id RPC inspections. The indication at R20C7 7H was apparently dented in the tube pull process and the dent obscured any potential bobbin call although an indication was identified j

by the post-pull RPC inspection.

i i

The indications at R3C107.5H and 7H, and R20C7 SH and 7H, which were field NDD but had identifiable indications by reevaluation of the field data, were found to have average crack derths of 32%,21%,20% and 12%, respectively and maximum depths of 53%,45%,

38% and 26% The two R3C107 indications could be interpreted as above the detection threshold for eddy current but were distoned indications that are difficult to call under field i

conditions.

1 The increases of up to 3.7 volts for the post-pull bobbin voltages are typical of many pulled tubes in the EPRI database. Some tearing of ligaments between microcracks may have occurred, but significant tearing of ligaments to strongly affect accident condition leakage or burst capability is not expected for these levels of post-pull voltage increases. Thus, the pre-pull voltage measurements are acceptable for ARC applications.

3-6 i

i 1

i f

. - - ~

.~

3.2.2' Byron 1 Data for ARC Applicatios The Byron-1 pulled tube leak test, burst test and destructive exam results are summarized in Table 3-7. The measured leak rate data, given in Table 3-2, were adjusted in Table 3-7 to the reference normal operating and SLB conditions by applying the leak rate adjussnent procedure of the EPRI database report. The reference SLB conditions are a pressure differential of 2560 psid at a primary pressure of 2575 psi and a secondary pressure of 15 psi at a temperature of.616* F. The measured burst pressures are adjusted to the reference 150 ksi for the sum of the yield plus ultimate tensile strengths. The data of Table 3-7 should be used in EPRI ARC burst pressure, SLB probability ofleakage and SLB leak rate versus voltage correlations.

l The 10.9 volt response for R20C7 3H versus the 5.89 volt response for R3C107 3H indicate the streng influence of throughwall crack length on the bobbin voltage amplitude and the influence of the two parallel throughwall cracks found for R20C7 3H (see Table 3-7). Both of the throughwall cracks in R20C7 3H are similar but the burst crack had a slightly longer macrocrack length (0.602" versus 0.562") and few remaining uncorroded ligaments (3 versus 9). Only these two throughwall indications resulted in leakage at SLB conditions. The 2.58 volt indication at R20C102 3H, with a maximum crack depth of 96% and a ~0.1" length at >

90% depth, did not break through or leak.

The Byron-1 pulled tube results were evaluated for potential exclusions from the database against the EPRI c% exclusion criteria. Criteria la to le apply primarily to unacceptable voltage, burst or leak rate measurements and indications without leak test measurements. Of these criterion, only Criterion Ib, unacceptable burst test due to incomplete burst, requires further evaluation for the Byron-1 tubes. As shown in Table 3-3, the burst tests of R3C107 3H and R20C7 3H required repeat burst tests due to lack of a fishmou% opening with tearing at the crack tips as required for the definition of burst. The repeat burs, st of R20C7 3H, with a faster pressurization rate and a slightly larger foil, resulted in crack extension and the repeat test at a 4500 psi burst pressure was a successful burst test. Two repeat tests were tried for R3C107 3H but neither repeat test resulted in crack extension although the crack width increased slighti,* with each test and the burst capability of the indication decreased with the successive ter s. The results of the repeat tests indicate that even with crack tearing to achieve a well-defined burst, it is not likely that the burst pressure would have significantly increased above the initial test result of 4800 psi. The initial 4800 psi burst pressure is therefore reasonable and is retained in the EPRI database although representing a minimum burst capability of the indication. Criterion 2b applies only to indications greater than 20 volts which is not applicable to the Byron-1 indications.

EPRI Criterion 2a applies to atypical ligament morphology for indications having high burst pressures relative to the burst / voltage correlation and states that high burst pressure indications having s 2 uncorroded ligaments in shallow cracks < 60% deep shall be excluded from the database. The indications at R20C7 SH and 7H have maximum depths of 38% and 26% with two uncorroded ligaments which satisfy the s 2 ligament Criterion 2a. The 0.25 3-7

,w

i e

volt indication at 5H does not have a high burst pressure relative to the burst correlation (See Section 3.3) and thus cannot be excluded from the databasec However, the 0.38 volt indication at 7H has a maximum depth of 26% with two remaining ligaments and has a high burst pressure relative to the correlation. Hus, this indication satisfies Criterion 2a and should be excluded from the EPRI database for application to the ARC correlations.

Criterion 3 relates to potential test errors in the leakage measurement and would exclude indications having less than a 10% increase in leakage between normal operating and SLB conditions or exclude indications having a measured leak rate for throughwall cracks without ligaments which is more than a factor of 50 below the mean leak rate expected for the associated throughwall crack length. Both of the indications that leaked had more than a 10%

increase in leakage between normal operating and SLB conditions so that this criterion for exclusion does not apply to the Byron-1 tubes. The indications R3C107 3H and R20C7 3H i

have throughwall crack lengths with no ligaments within the throughwall length and are evaluated for low leak rates against the factor of 50 criterion. Figure 3-13 shows the addition of the Byron-1 data to the prior EPRI 3/4 inch diameter database for SLB leak rate as a function of throughwall crack length. It is seen that the Byron-1 indications are not below the mean leak rate for their respective crack lengths and the indications should not be excluded from the database.

All the Byron-1 TSP intersections of Table 3-7 with bobbin indications were leak and burst tested or have no leakage inferred fron. the shallow depths and, with the exception of R20C7 7H, should be included in the EPRI dahbase for IPC/APC correlations. The indications with no leakage are appropriate for the probability of leakage and burst pressure correlations and those with SLB leakage would be included in all three correlations including the leak rate correlation. The bobbin NDD indications, while not needed for the EPRI correlations, are i

applicable for probability of detection assessments. He field bobbin calls for the Byron-1 indications were defined prior to considerations for pulling tubes and are not influenced by i

. tube pull considerations. The field calls for R3C107 5H and 7H, and R20C7 5H and 7H, were NDD and the bobbin calls for these four indications were made subsequent to the tube pull and are more typical of laboratory review than typical field expenence.

\\

3.3 Comparison of Byron 1 Data with Existing IPC Correlations This section reports on the evaluations performed utilizing leak rate and burst test results of tube sections which were removed from Byron Unit 1 in the fall of 1994. The Byron 1 pulled tube data for ARC applications are given in Section 3.2, and are compared to the data-base' of similar test results for 3/4" outside diameter steam generator tubes. In addition, the

' The database consisted of the EPRI recommended database, plus test results from model boiler specimen 598-3, plus the leak rate for specimen S-R28C41-F being taken as 24961ph, plus Braidwood I test results reported in WCAP-14046, revison 1 (except for specimen R16C42-5, which was excluded by EPRI criterion 2.a).

3-8 1

.. ~ -

effect of including the new test data in the reference database was evaluated. In summary, the test data are consistent with the database relative to the burst pressures, the probability of leak, and the leak rate as a function of the bobbin amplitude. These comparisons and evalua-tions are discussed below.

3.3.1 Burst Pressure vs. Bobbin Amplitude Results from eight (8) burst tests. performed on tube specimens which exhibited non zero bobbin amplitudes at locations corresponding to the in plant elevations of the tube support plates, were considered for evaluation. A plot of the burst pressures of the Byron I speci-mens is depicted on Figure 3-14 relative to the correlation developed for the reference database. A visual examination of the graph indicates that all of the burst pressures measured fall within the scatterband of the reference data about the reference regression line. The burst pressure for specimen R03C107-3, is somewhat lower than would have been expected for 5.9 t

volts, but is not significantly lower, and the result represents a lower bound value. Thus, the visual examination doesn't indicate any significant departures from the reference database.

l Although the upper bound is not shown, all of the data fall within a 90% non-simultaneous prediction band about the regression line (the one-sided 95% prediction curve depicted is the lower bound of a two-sided 90% prediction band). Since a two-sided simultaneous prediction band for the eight data points would be wider than the non-simultaneous band, no statistically significant anomalies are indicated.

Since the Byron 1 burst pressure data were not indicated to be from a separate population from the reference data, the regression analysis of the burst pressure on the common logarithm of the bobbin amplitude was repeated with the additional data included. A comparison of the regression results obtained by including those data in the regression analysis is provided in Table 3-8. The intercept of the burst pressure, P,, as a linear function of the common logarithm of the bobbin amplitude regression line is decreased by 0.04%, and the slope is decreased by ~0.6%, i.e., the slope is a larger negative number. Regression i

predictions obtained by including these data in the regression analysis are also shown on Figure 3-14. Since the intercept and slope are both reduced by the inclusion of the Byron 1 i

data, predicted burst pressures at all voltages would be slightly less than the value obtained using the reference regression line. There is also a decrease of ~3% in the standard error of i

the residuals. The effect of this change would be reflected in a smaller deviation of the 95%

prediction line from the regression line. The net effect of both of these changes on the SLB structural limit, using 95%/95% lower tolerance limit material properties, would be to increase it by 0.11 volts, i.e., from 4.64 volts to 4.75 volts, which is judged to be not significant. The decrease in the standard error of the residuals would also slightly, reduce the probability of burst for bobbin indications over most of the structural range ofinterest. Based on the judged insignificant change in the structural limit, the change in the probability of burst would also likely not be significant. It is noted that for high voltages, e.g., near the SLB structural limit, the effect of the lower intercept and slope would offset the advantage of the lower standard error and the probability of burst would slightly merease.

t 3-9

One additonal comparison was made.- A linear regression analysis of the Byron 1 data alone was performed. Based on those results, the intercept of the reference database is within a 44% band relative to the intercept from the Byron 1 ' database. The reference slope is within a 71% band relative to the slope from the Byron I database. Thus, a null hypothesis that the Byron i data represent a sample from a different population than the reference database would be rejected. Hence, inclusion of the Byron I data in the reference database is justified.

3.3.2 Probability of Leak The same eight data points examined relative to the burst pressure correlation were also examined relative to the reference correlation for the Pot as a function of the common logarithm of the bobbin amplitude. Figure 3-15 illustrates the Byron 1 data relative to the reference correlation. All of the specimens exhibited pol behavior commensurate with expectations indicated by the reference regression curve. Based on the visual examination, there appears to be no significant evidence of irregular results, i.s., outlying behavior is not indicated.

I In order to assess the effect of the new data on the correlation curve, the database was expanded to include the Byron I data and a Generalized LinearModel regression of the Pot on the common logarithm of the bobbin amplitude was repeated. A comparison of the correlation parameters with those for the reference database is shown in Table 3-9. These I

results indicate a 3% reduction in the logistic intercept parameter and an 3% increase in the logistic slope parameter. The Pearson standard error increased by 3% from 0.88 to 0.91.

This is a positive indicator since the ideal value would be 1.0. In order to assess whether or not these changes are significant the reference correlation and the new correlation were also l

plotted on Figure 3-15. An examination of Figure 3-15 reveals essentially no change, in an absolute sense, in the correlation below a bobbin amplitude of I volt, or above a bobbin amplitude of about 15 volts. The pol is less than the reference correlation up to a bobbin amplitude of about 4 volts, and slightly greater for indications with amplitues greater than 4 i

volts. A listing of comparative values is provided in Table 3-10. At I volt the decrease in the pol is on the order of 0.001. In conclusion, the Byron 1 data are consistent with the reference database.

3.3.3 Leak Rate vs. Bobbin Amplitude s

As previously noted, two of the removed tube specimens exhibited leakage under SLB conditions. The leak rates are provided in Table 3-7, and depicted on Figure 3-16 relative to the correlation obtained using the reference database. One of the actual leak rates was found to be slightly above, and the other slightly below, the median prediction (the regression fit on the plot) of the regression. It is implied from the visual examination, using the relative distance of the 95% confidence bound on the arithmetic average from the arithmetic average, that all of the data points would fall well within a 90% non-simultaneous, two-sided prediction band. Thus, the visual appearance of the data indicate strong support for the trend of the prior correlation.

)

3-10 1

1 I

i

As for the previously discussed correlations, a regression analysis was performed to assess the

~ influence of the Byron 1 data on the leak rate to bobbin amplitude correlation. The results of repeating the regression analysis with the Byron I data included are given in Table 3-11.

Inclusion of the Byron I data results in a 1% decreasa in the logarithmic intercept parameter and a 2% decrease in the logarithmic slope parameter. In addition, the standard error of the J

logarithmic residuals is decreased by ~8%, the index of determination of the regression is increased by about 0.3%, and the p-value for the slope parameter is reduced by about 50%.

Thus, the addition of the data increases the strength of the correlation. Since the intercept, the slope, and the residuals standard error are all reduced, the net effect of including the new I

data is to reduce the expected, i.e., arithmetic average, leak rate for bobbin amplitudes over the entire range of the data.

3.3.4 General Conclusions The review of the effect of the Byron I data indicates that the burst pressure, the probability of leak, and the leak rate correlations to the common logarithm of the bobbin amplitude would not be significantly changed by the inclusion of the data. Therefore, it is likely that the conclusions relative to EOC probability of burst and EOC total leak rate based on correlations obtained using the reference database would not be significantly affected by repeating those analyses using an expanded database which includes the Byron 1 test data.

1 1

)

1 3-11

Table 3-1 1

Comparison of NDE Indications Observed at Byron Unit 1 1

on SG Tubes at Hot Leg Locations i

Tube /

Field E/C Leb E/C Field UT Lab UT Lab X-Ray Locahon

-l R3C1071H Bobtzn: NDO Bobtm: NDD NDD NDD NDO BEQ: NDD RPC: NDD R3C107 3H Bobbm: 5.92V OD Ind, Bobten: 7.0V OD Ind, 78%

Two areas with many short 0.4" axial OD Ind at 30' that 0.3" SAI 78% deep (5.89V)*

deep axialInds: one 115' by is neer 100% deep; plus BEQ: 4.03V, 0.37" SAI RPC: 5.4V sal 0.81"; the other 75' by 0.30" Short, shallow, MAI in (42/ sal,0.38")*

crevice between 10" & 45',

235' & 290' and near 350" R3C107 5H Bobbin: NDD (0.25V Bobbin: 0.21V DI One area with many short NDD NDD Di)*

RPC: 0.58V Volumetric Ind axial inds: 80" by 0.47" BPG: NDD Possible Axial Ind. 0.36" long in same area i

Looend of Abbnmebons:

0* = lield reevab va'ue (using cross TSP = Tube Support Plate V = Voltage Circ = circumferential caibration of ASME std. to ref. lab std.)

Ind = Indication MAI = Multiple Axialinds Max = maximum NDD = No Detectable Degradation SAI = Single AxialInd

1. C = number of cracks Di = Distorted Indicabon RPC= Rotating Pancake Coil ICC = intergranular cellular corrosion

'3-12 6

=.

Table 3-1 (Continued)

Cornparison of NDE Indications Observed at Byron Unit 1 on SG Tubes at Hot Leg Locations Tubel Field E/C Lab E/C Field UT Lab UT Lab X-Ray Location R3C107 7H Bobbm: NDO (0.26V Bobbin: 0.30V Di Two areas with nmny short, NDD NDD l

DI)*

RPC: 0.22V SAI low empNtude aria

• Inds: one BE_Q: NDD 70" by 0.26*; the other 50 by P

1 0.20" R20C71H Bobbm: NDD Bobbin. NDD NDD NDD NDD BP_Q: NDD RPC;NDO' R20C7 3H Bobbm: 10.95V OD Ind.

Bobbm: 13.8V OD Ind, Short OD origin MAI (6C with OD MAI (4C): at 70' (0.1" &

MAI (7+C),

79% deep (10.9V,80%

81% deep Circ involvement) dist<' mted near 100% TW), at 80" longest 0.3" long A

deep)*

RPC: MAI (3C),6.5V &

around circumfereno, ID sal (0.19"long & near 100%),

E2Q: mal (2C), 0.45 &

1.6V with Circ involvenwnt at same 100',160' (025" long &

0.37"long with some locanon es OD IrtJ 50%)

Circ involvement,4.59V Max (4.7V & 1.3V)*

R20C7 5H Bobbin: NDD (0.23V Bobbin,1.49V DI (noisy) 0.2" SAI 0.2" axial OD Ind at 85",

NDO Df)*

EEQ; 0.43V Volumetric Ind 45% deep; plus many short -

BEQ: NDD shalow MAI between 3108 to 335' R20C7 7H BobtWn: NDD (0.38V Bobbin: 6.7V dent NDD Many short shanow OD MAI NDD l

DI)*

RPC: 0.40V Volumetric Ind

(<20% deep) near 100'.&

BEQ: NDD near 115' 3-13 I

~

i 1

a Table 3-1 -(Continued)

Comparison of NDE Indications Observed at Byron Unit 1

- on SG Tubes at Hot Leg Locations h

Tube /

Field E/C Lab E/C Field UT Lab UT Lab X-Ray Locahon R20C102 Bolbn: NDD Bobbin. NDD NDO NDD NDD-1H BE_Q: NDD B2rd NDO

{

R20C102 Bobtzn: 2.58V OD Ind, Bobinn: 4.92V OD Ind,61%

MAI (3C) 0.2" axial OD Ind at 165',

O T SAI 3H 80% deep (2.58V, deep 40% deep, may be 0.5" l

63%)*

RPC: MAI (3C),4.24V Max long with nearby shallow RPC: 0.37' SAI,1.76V Inds; plus short shatow 8

MAI at 30,100' & 130" e

R20C102 Bothn: 1.25V OD Ind, Bobbin: 4.64V OD Ind,44%

mal, two areas with rnany 0.07" axial OD Ind at 85',

NDO 5H 81% deep (1.11V, deep-short axial inds: one 50' by 30% deep; plus short _

l 79% deep)*

RPC: Volumetric Ind in 2 0.4"; the other 70 by 0.3" shaRow MAI near 15*, 85' 8

BEQ: 0.27" sal with areas,2.05V Max to 150', & near 300 ; plus 8

some Circ 0.23" SAI possible OD Cire ind Irwolvement,0.18V between 285' & 305',30%

Max (Volumetric Ind in deep i

2 areas,0.52V)*

I f

R20C102 Bobbin: NDD Bobbin: NDD NDD NDO NDD 7H BEQ: NDD B2rd NDD l

t l

3-14

-j I

2:. +i..

__m.-;L-

.-. -1 ua

...L..

A

.e m

m.h-w-W+

M m.-

u w'

-T-t-7-'

--wav 44->

u Table 3-2 Byron Unit 1 Leak Test Results for Steam Generator Tubing Tube No.,

Test Type:

Leak Rate Test Conditions Location Differential (liters / hour)

Pressure (psi)

R3C107 3H NOC: 1305 0.054 P,= 2265, P,= 960, T,= 584, T,= 597 NOC: 1269 0.047 P,= 2224, P,= 955, T,= 589, T,= 602 ITC1: 1874 0.410 P,= 2390, P,= 516, T,= 550, T,= 583 ITC2: 2269 2.40 P,= 2679, P,= 410, T,= 539, T,= 503 SLB: 2545 4.70 P,= 2955, P = 410, T,= 546, T,= 472 R20C7 3H NOC: 1286 0.065 P,= 2237, P,= 951, T,= 615, T,= 623 ITC1: 1862 0.473 P,= 2380, P,= 518,. T,= 578, T,= 598 ITC2: 2302 8.50 P,= 2690, P,= 388, T,= 570, T,= 476 SLB: 2510 18.3 P,= 2936. P,= 426, T,= 570, T,= 471 R20C7 7H NOC: 1286 zero P,= 2219, P,= 933, T,= 619, T,= 617 ITC1: 1882 zero P,= 2396, P = 514, T,= 623, T,= 619 SLB: 2545 zero P,= 2763, P = 218 T,= 621. T,= 615 R20C102 3H NOC: 1320 zero P,= 2270, P,= 950, T,= 612, T,= 614 ITC1: 1925 zero P,= 2460, P,= 535, T,= 620, T,= 622 SLB: 2545 zero P,= 2745, P,= 200, T,= 622, T,= G24 P,r 2245, P,= 965, T,= 607, T,= 60'9 R20C102 5H NOC: 1280 zero ITC1: 1900 zero P,= 2400, P,= 500, T,= 618, T,= 619 SLB: 2555 zero P,= 2765, P = 210, T,= 620, T,= 622 Legend:

All data within a table block is presented in the order of testing, NOC= normal operating conditions, ITC= intermediate test conditions, SLB= steam line break conditions, P,= primary side pressure (psi), P,= secondary side pressure (psi),

T,= primary side temperature ( F), T,= secondary side temperature ('F) 3-15

~. -.

p.

Table 3-3 Room Temperature Burst and TensRe Test Results

' for Byron Unit 1 Hot Leg SG Tubing Location Burst Pressure Ductitty Burst Length Burst Width 0.2% offset.

TensRe UTS Tensile (paig)

(% Dia.)

(inches)

(inches)

TensRe YS (psi)

Elongation (psi)

(%)

R3C107,FS 11,500 33.7 1.494 0.288 54,000 101,600 31.3

+

+

R3C107 3H 4,800 5.4 0.462 0.089 4,000 5.8 0.477 -

0.093 3,550 5.9 0.465 0.103 R3C107 SH 9,800 17.7 1.057 0.256 R3C107 7H 10,300 18.7-1.057 0.206 R20C7,FS 11,500 32.1 1.588 0.299 55,700 102,450 31.2 i

R20C71H 11,300 29.7 1.521 0.317

+

+

i R20C7 3H 3,750 6.9 0.500 0.079 4,500 9.1 0.695 0.164 R20C7 5H 10,200 18.1 2.114 0.293 I

R20C7 7H 11,300 25.4 1.356 0.359 R20C102,FS 11,900 30.9 1.503 0.295 57,400 106,250 26.2*

+

+

t R20C102 3H 7,050 11.7 1.111 0.320 l

+

+

l R20C102 5H -

'9,050 16.5 1.198 0.366 Control (NX7368) 12,050 34.5 1.963.

0.411 57,300 108,500 25.7*

a Legend:

TSP = Tube support plate crevice region locaton, FS = koe span locaton

• = Tenene specimen broke ouloide of page length, poesAdy reducing elongston value.

+ = Burst specimen used a bladder and Sou over largest detect etes and was burst in a semi-restraint condNion. In addeon, eter burst teaung, the toil was observed to be centered under ;

the burst opening As other burst specimens were burst wthout binddere and b8e and wthout a restraint condelene Specimens R3C107. TTS and R20C7 3H weee roburet wNh a larger '

sire fot gottwing the intiel burst test giet used a stonderd aire IDE (1.75 inch high x 1.0 inch wide versus 0.75 inch high x 0.5 kich wide). The roburst woe performed to produce true tenene leertng et the tips of the burst opening The reburst apparently was successkd ter R20C7 3H. A escend roburst ter R3C107 3H etR dkl not produce a true burst opening r

. 3-16.

Table 3-4 Byron Destructive Examination Planning Data i

Specimen E/C Data (field Burst & Ductdety FF Corrosion

• Crevice Region * -

DE Plan bobbin probe)

Ratios (specimen /

(visually Conosion (visually FS value) observed) observe (J R3C107,FS NDD 1.00/ 1.00 No No No R3C107 3H 5.9V IND (5.9V) 0.42/0.16 Yes Yes Yes, both metallography and SEM FF R3C107 SH NDD (0.25V DI) 0.85/ 0.52 Yes Yes (almost 36(f)

Yes, SEM FF only R3C107 7H NDD (0.26V DI) 0.90/ 0.55 Yes Yes (36(f)

Yes', SEM FF only i

R20C7,FS NDD 1.00/ 1.00 No No No R20C71H NDD 0.98/ 0.92 No No No R20C7 3H 11.0V IND (10.9V) 0.33/ 0.21 Yes Yes Yes, both metatography and SEM FF R20C7 SH NDD (023V DI) 0.89/ 0.56 Yes Yes (almost 360')

Yes, SEM FF only R20C7 7H NDD (0.38V DI) 0.98/ 0.79 Nom Yes (360')

Yes, SEM FF only R20C102,FS NDD 1.00/1.00 No No No R20C102 3H 2.6V IND (2.6V) 0.59/ 0.38 Yes Yes Yes, both meteHography and SEM FF i

Also, extra chemetry of W and FF R20C102 SH 1.2V IND (1.1V) 0.76/ 0.53 Yes Yes (360")

Yes, both metauography and SEM FF

• = All bursts occurred centered within the crevice regions or at the tubesheet top.

FS = free span; TSP = tube support plate; FF = fracture face m Visual observation subsequently invaldsted by crevice corrosion observation.

i b

i n

i 3-17 i

r-

~

e m

~

l Table 3-5 Byron Unit 1 SG Tube Burst Opening Macrocrack ProfBes

Tube, Length vs. Depth Ductile Ligament Comments Location (inches /% throughwaff)

Location / Width (inches)

R3C107 3H 0.00/00 Crack Bottom The axia5y oriented 0.04/50<-Ugement 1

<-Ugement 110.013" wide burst macrocrack had 0.08/46 four ductie Egaments 0.12/21<-Ugement 2

<-Ugament 210.003" wide with dimple rupture 0.16/50<--Ugement 3

<-Ugement 310 004" wide features ew_%g 0.20/71 over more than 60%

0.24/79 of their length.

0.28/92 0.32/92 (0.34/100)

<-Throughwell tip 0.36/100 r

0 40/100 (ThroughwaN over 0.13 inch) 0.44/100 (0.47/100)

<-Throughwall tip 0.48/87

~

0.52/79 0.M7<-Ugement 4

<-Ugement 440.002" wide (0.59/00) (Ave. depth = 67%, Macrocrack Length

{

= 0.59 inch)

Crack Top I

l L

]

3-18 I

l i

l l

Table 3-5 (Continued)

Byron Unit 1 SG Tube Burst Opening Macrocrack ProNes

Tube, Length vs. Depth Ductile Ligament Comments Location (inches /% throughws'l)

Location / Width (inches)

R3C107 5H 0.00/00 Crack Top The axiaRy oriented 0.04/42 burst macrocrack had 0.08/32<-Ligament 3

<--Ligament 3/0.015' wide thme ducWe 0.12/37 ligaments with dimple 0.16/35 rupture features 020/45<-Ligement 2

" ""5 "f "'N 0.24/45 than 50% of their 0.28/49 length.

l (0.26/53) <-(Max. depth = 53%)

O.32/38 l

0 36/30 0.40/23 (Ave. depth = 32%,

cement in003%

0.52/22 Macrocrack Length =

Crack Bottom (0.550/00) 0.550 inch)

R3C107 7H 0.00/00 Crack Top The arielly oriented 0.04/19 burst macrocrack had 0.08/14<-Ligament 8

<-Ligament 8/0.00f wide eight ducWe 0.12/23<-Ligament 7

<-Ligament 7/0.006 wide

% aments with dimple O.16/26 rupture features 0.20/23<-Ligement 6

<--Ligament 8/0.004* wide occurnng over more O.24/37 than 50% of their (0.26/45) <-(Max. depth = 45%)

length.

0.28/40 0.32/37 O.36/34<-Licenient 5

<-Ligament 5a006 wide 3.40/19<-Ligement 4

<-Ligement 410.014" wide 0.44/16<-Ligament 3

<-Ligament 310.012 wide 0.48/19

<-Ligament 2/0.036" wide 0.52/00<-usammt 2 0.56114<-Ligament 1 (Ave. depth = 21%,

(0.592/00)

Macrocrack Length =

gg[J0.002" wide 0.592 inch) 3-19

Table 3-5 (Continued)

Byron Unit 1 SG Tube Burst Opening Macrocrack Profiles

Tube, Length vs. Depth Ductile Ligament Comments Location (inches /% throughwall) location / Width (inches)

R20C7 3H 0.00/00<-Ligaments 1 & 2 Crack Bottom <-Ugaments 1 & 21.002 The axially oriented (Right Face of 0.04/31 g o.og4 wide burst macrocrack had Burst Opening @

0.08/38 three ductile 100")

0.12/48 ligaments with dimple 0.16/76 rupture features 0.20/84 occurring over more (0.22/100)

<-Throughwall tip than 50% of their 0.24/100 length.

O.28/100 0.32/100 (Throughwall over 0.36/100 0.24 inch) 0.40/t 00 0.44/100 (0.46/100)

<-Throughwall tip 0.48/86 0.52/57 0<-Ligament 3

<--Ligament 310.001" wide (0.802/00)

(Ave. depth = 88%,

Crack Top Macrocrack Length =

0.602 inch) 3-20

Table 3-5 (Continued)

Byron Unit 1 SG Tube Burst Opening Macrocrack Profiles Tube.

Length vs. Depth Ductile Ligament Comments Location (inches /% throughwall)

Location / Width (inches)

R20C7 3H 0.00/00<-Ugament 1 Crack Bottom <-Ugament 1/0.005" wide The axially oriented (Left Face of 0.04/36<-Uwements 2 & 3

<-Ugei.,ents 2 & 3/0.004 & 0.009" wide burst macrocrack had Throughwall Crack O.08/27 nine ductile ligaments

@ 70')

0.12/73 with dimple rupture 0.16/75 features occurring (0.195/100)

<-Throughwall tip over more than 50%

0.20/100 of their length.

0.24/100 0.28/100 (Throughwall over 0.32/100 0.268 inch) 0.36/100<-Ugemant 4

<-Ugament 4 /0.003" wide 0.40/100 0.44/100 (0.463/100)

<-Throughwall tip 0.48/64<-Ugaments 5,6 & 7

<-Ugements 5. 6 & 7/0.004" wide, each" 0.52/32<-Ugements 8 & 9

<-Ugements 8 & 9/0 004 & 0.006" wide 0.56/23 (0.562/00)

(Ave. depth = 60 4, Macrocrack Length =

Crack Top 0.562 inch) l l

l 3-21

l Table 3-5 (Continued)

Byron Unit 1 SG Tube Burst Opening Macrocrack ProfWes I

i

Tube, Length vs. Depth Ductile Ligament Comments Location (inches /% throughwall)

Location / Width (inches)

R20C7 SH 0.00/00 Crack Top The axially oriented 0.04/02 burst macrocrack had 0.08/01 two ductie ligaments 0.12/02 with dimple rupture 0.16/10 features occurring 0.20/26 over more than 50%

0.24/32<-Ugament 1

<--Ligament 110.002" wide of their length.

0.28/26 0.32/32 0.36'29<-Ligament 2

<--Ligament 210.002" wide 0.40/25 (0.43/38) <-(Max. depth = 38%)

0.44/31 0.48/30 0.52/27 0.56/00 (Ave. depth = 20%, Macrocrack Crack Bottom Length =0.560 inch)

R20C7 7H 0.00/00 Crack Top The axially oriented 0.04/04 burst macrocrack had 0.08/02 two ductile ligaments 0.12/14 with dimple rupture 0.16/23 features occurring 0.23/25 over more than 50%

(0.22/26) <-(Max. depth = 26%)

of their length.

0.24/23 0.

8<-Ligament 1

<--Ligament 110.004" wide 0.36/05 0.40/04 0.44/07<---Ugament 2

<-Lioament 210.006" wide (0.442/00)

(Ave. depth = 12%,

Cract Bottom Macrocrack Length =0.442 inch) 3-22

Table 3-5 (Corttinued)

Byron Unit 1 SG Tube Burst Opening Macrocrack Profiles l

Tube.

Length vs. Depth Ductile Ligament Comments Location (inches /% throughwall)

Location / Width (hches)

R20C102 3H 0.00/00<-Ugement 1 Crack Top <-ugement 1/0.007 wide The axiag oriented y

O.04/50 burst macrocrack had 0.08/86 two ductile ligaments 0.12/91 with dimple rupture 0.16/91 feat.sres occurring (0.18/96) <-(Max. depth = 96%)

over inore than 50%

0.20/95 of their leng'h.

0.24/88 0.28/60 0.32/00<-Ugement 2

<-Ugement 2to.004 wide O.36/20 (0.38/00) (Ave depth = 58%, Macrocrack Length =0.38 inch)

Crack Bottom R20C102 5H 0.00/00 Crack Top The axiaNy oriented 0.04/37 burst macrocrack had 0.08/46 two ductRe ligaments 0.12/46

. with dimpse rupture 0.16/58 features occurring 0.20/60 over more than 50%

0.24/65

<-(Max. depth = 65%)

of their length.

0.28/58 0.32/48 0.E7<-Ugement 1

<-Ugement 1/0.010 wide 0.40/22< -tJoement 2

<-Ugement 2Jo.002 wide (0.41/00) (Ave. depth = 43%, Macrocrack Length =0.41 inch)

Crack Bottom 3-23

~

. ~

, - _. ~ _ _ -.

l Table 3-6. Summary of Byron-1 Pulled Tube Eddy Current Results T

Field Call Lab. Reevaluation of Field Data Post Pull Data Tube S

p Bobbin RPC UT Bobbin ASME Bobbin Depth RPC UT Bobbin RPC Volts'"

Volts Volts Cal.(

Volts

Volts Results Volts Volts Steam Generator A R3C107 1

NDD NDD NDD NDD NDD NDD NDD NDD 3

5.92 4.03 MAI 5.26 1.12 5.89 78%

4.2 MAI 7.0 5.4 f.

NDD NDD MAI 0.22 1.12 0.25 Di NDD NDD 0.21 0.58 7

NDD NDD MAI 0.23 1.12 0.26 Di NDD

• *G0 0.30 0.22 Steam Generator C R20C7 1

NDD NDD NDD NDD NDD NDD NDD NDD 3

10.95 4.59 MAI 10.12 1.073 10.9 80%

4.7 MAI 13.8 6.5 5

NDD NDD SAI 0.21 1.073 0.23 Di NDD MAI 1.49 0.43 7

NDD NDD NDD 0.35 1.073 0.38 Di NDD mal Dent 0.40 R20C102 1

NDD NDD NDD NDD NDD NDD NDO NDD 3

2.58 1.76 MAI 2.40 1.073 2.58 63%

2.0 MAI 4.92 4.24 5

1.25 0.18 MAI 1.03 1.073 1.11 79%

0.52 MAI 4.64 2.05 7

NDD NDD NDD NDD NDD NDD NOD NDD Notes: 1. Field data include cross caibration of ASME standard to the reference laboratory standard

2. ASME caibration represents the cross caibration factor for the field ASME standard to the reference laboratory standard and is appled to the laboratory reevaluation to obtain the corrected IPC volts.

3-24

Tcble 3-7.

Cyren-1 Pulled Tube D:ta for ARC ApplicsCa3 T

Bobbin Data Destructive Exam Results Leak Rate-Ilhr Burst Pressure Data - ksi Tube S

RPC p

volts Max.

Avg.

Crack TW No.

N. O.

SLB Meas.

o" a"

Adj.

Volts Depth Depth Depth Length Length Lig."

1300 2560 Burst Burst psid"'

psid"8 Press.

Press.

Steam Generator A R3C107 1

NDD NDD Note 3 Note 3 Note 3 54.0 101.6 3

5.89 78 %

4.2 100 %

67 %

0.590" 0.130" 4

0.037 4.15

>4.800'"

>4 627'"

5 0.25 Di NDO 53 %

32%

0.550" 3

0 0

O.0

9.800 9.447 7

0.26 Di NDD 45%

21 %

0.592-8 0.0

O.0

10.300 9.929 Steam Generator C R20C7 1

NDD NDD

-0%

0. 0

O.0

11.300 55.7 102.45 10.718 3

10.9 80%

4.7 100 %

68%

0.602" 0 240" 3

0.037 22.1 4.500 4.268 100 %

• 69%

0.562" 0.268" 9

5 0.23 Di NDD 38 %

20 %

0.560" 2

0.0

O.0

10.200 9.674 7

0.38 Di NDD 26 %

12%

0.442" 2

0.0 0.0 11.300 10.718 Note 6 Note 6 Note 6 i

R20C102 1

NDD NDO Note 3 Note 3 Note 3 57.4 106.25 3

2.58 63 %

2.0 96 %

58 %

0.380" 2

0.0 0.0 7.050 6 462 5

1.11 79%

0.52 65%

43%

0.410*

2 0.0 0.0 9.050 8.295 7

NDD NDD Note 3 Note 3 Note 3 Notes:

j

1. Measured leek rates adjusted to reference conditions by applying methods of EPRI data report

2. Number of uncorroded ligaments with > 50% ofligament length remaining in burst crack face.

3. Archive spedmen.

4. Inferred from destructive exam depth. leak test not performed. Corrosion depth too shallow for leakage at SLB conditions I

5. Burst test showed insignificant crack extension at the corrosion crack tips. Burst pressure is therefore a minimum value since burst is defined to indude crack l

extension and the burst pressure therefore represents a minimum value.

6. Data exduded from EPRI database per data exdusion criterion 2a which exdudes indications having high burst pressures on the burst pressure versus voltage correlation with cracks <60% deep with s 2 Egaments.

7. Second throughwell crack 30* from burst crack.

3-25

j.

i

[

I i

Table 3-8: Effects of the Byron Unit 1 Data on the Burst Pressure vs. Bobbin Volts Correlation P, = a, + a, log ( Vohs)

Parameter Reference Database Database with Value Byron 1 Data ai 7.834 7.831 a,

-3.109

-3.127 o,,,,

0.934 0.906 N (data pairs) 80 87 p Value for a, 7 10 '

4 10 '

4 4

t' 81.3 %

82.9 %

i i

I l

i

[

i

?

l 1

1 3-26 i

i

)

s i

i e

Table 3-9: Effect of Byron Unit 1 Data on the Probability of Leak Correlation Pr(Leak) = 1 + e ~Ih **d Parameter Reference Database Database with Byron Value 1 Data E,

-5.014

-5.171 D2 8.410 8.669 V,,

1.432 1.471 V,2

-2.031

-2.101 V,,

3.256 3.370 l

Pearson o.

0.88 0.91.

n Note: Parameters % are elements of the covariance matrix of the coefficients, p,, of the above regression equation.

i i

l i

t F

i 3-27

<r Table 3-10: Effect of Byron 1 Data on the Probability of Leak as a Function of Bobbin Amplitude Bobbin Amplitude Pol Pol (Volts)

(Reference)

(w/ Byron) 0.60 0.0010 0.0008 0.70 0.0018 0.0015 0.90 0.0029 0.0024 1.00 0.0066 0.0056 4

2.00 0.0771 0.0717 3.00 0.2686 0.2621 5.00 0.7036 0.7085 6.00 0.6220 0.8284 7.00 0.8902 0.8961 9.00 0.9531 0.9706 i

l I

i i

4 l

'l 3-28

)

i

.~..

Table 3-11: Effects of Byron Unit 1 Data on the Leak Rate vs. Bobbin Volts Correlation log (Leak Rate) =

3+

, log (Volts)

Parameter Reference Database Database with Value Byron 1 Data

-2.067

-2.044 3

p.

-3.329 3.272

p. ( o,

)

0.695 0.636 N (Data pairs) 41 43 p Value for D, 7 10

3 10

t' 60.8 %

61.0 %

3-29

I I

I 1.25 g

1.00 -

- SP Top

.5

\\

8 h

=

r 5

I

]

b@),

G 0.25 -

-SP Bottom 0.00 g

i O'

90*

180*

270*

360*

Circumferential Position (degrees)

Figure 3-1 Sketch of the OD surface crack distribution found at the third tube support plate (TSP 3) region of Tube R3-C107. Also shown is the location of the burst fracture opening. The corrosion cracking and the burst fracture were confined to the crevice region.

3-30

+

I I

I 1.25 g

1.00 -

-SP Top

(

i.I f' jo

'\\'

=

4(t, g I s i ), y U s'tyl l \\ su f a a \\ I 5 3 0.25 - -SP Bottom 0.00 i i 0* 90* 180* 270* 360* Circumferential Position (degrees) i Figure 3-2 Sketch of the OD surface crack distribution found at the fifth tube support plate (TSP 5) region of Tube R3-C107. Also shown is the location of the burst fracture opening. The burst opening extended beyond the TSP crevice region. but the corrosion cracking on the burst fracture was confined to the crevice region. 3-31

I I I 1.25 1.0D- -SP Top i mm t e ,8 ,/. .) 1 g, fb 3 ' f (' \\\\ 1 't it ?

• t

\\ 's E g j hl'i O ft s i g, }'b br )u y 3} c ? ,\\ 2 t n,(., 11 'l,\\p },,,., i.. , in .i 1 ,u a 0.25- -SP Bottom 0.00 g i i O' 90* 180* 270* 360* Circumferential Position (degrees) Figure 3-3 Sketch of the OD surface crack distribution found at the seventh tube support plate (TSP 7) region of Tube R3-C107. Also shown is the location of the burst fracture opening. The burst opening extended beyond the TSP crevice region. but the corrosion cracking on the burst fracture was confined to the crevice region. 3-32 --..--a '- w ~r

i 1.25 I I i 1.00 - -SP Top m U f V ~ c: ~4 { ) }- t h\\ I Q $q-% b N 0.25 - -SP Bottom 0.00 I I I O' 90* 180* 270* 360* Circumferential Position (degrees) Figure 3-4 Sketch of the OD surface crack distribution found at the third tube support plate (TSP 3) region of Tube R20-C7. Also shown are the locations of the burst fracture openings which occurred at circum'erential positions of 450 and 110a.'Ihe corrosion cracking and the burst fractures were confined to the crevice region. i 3*)) - c

1.25 I I i 1.00 - L SP Top 2 i .a /y gI{ / y (,* \\}' l ~~ 4 3 [i q\\ f \\ 8 \\\\ l a e 0.25- -SP Bottom ua O.00 g i i O' 90* 180* 270* 360* Circumferential Position (degrees) l Figure 3-5 Sketch of the OD surface crack distribution found at the fifth tube support plate (TSP 5) region of Tube R20-C7. Also l shown is the location of the burst fracture opening. The burst opening extended beyond the TSP crevice region, but the corrosion cracking on the burst fncture was confined to the crevice agion. ( 3-34 l l

t I I I 1.25 2 1.00-. --SP Top e i o h 1 O tg y 5.r y It, l I I $Y I j g b / ,( h' li 4 g, t j ,e u i i, Ei 0.25- -SP Bottom 0.00 i i i 0* 90* 180* 270* 360* i Circumferential Position (degrees) Figure 3-6 Sketch of~the OD surface crack distribudon found at the seventh tube support plate (TSP 7) region of Tube R20-C7. Also shown is the location of the burst fracture opening. The blirst opening extendcd beyond the TSP crevice region, but the corrosion cracking on the burst fracture was confined to the crevice region. 3-35 m m

s..-

m. m m m m ..m m

I I I 1.25 1.00 -- -SP Top U f .c ) f 'tU' y It i i c g 0 % c f ]ISp o >2m 0.25 -- -SP Eottom 0.00 g 0* 90* 180* 270* 360* Circumferential Position (degrees) Figure 3-7 Sketch of the OD surface crack distribution found at the third tube support plate (TSP 3) region of Tube R20-C102. Also shown is the location of the burst fracture opening. The burst opening extended beyond the TSP crevice region. but the corrosion cracking on the burst fracture was confined to the crevice region. 3-36

I I I 1.25 1.00 -- -se ' rop 7 / l }i ,,8 f%; y I d u a g I g y ') y g 1 6 ao II3 0.25 - -SP Bottom 0.00 i g 0* 90* 180* 270* 360* Circumferential Position (degrees) Figure 3-8 Sketch of the OD surface crack distribution found at the fifth tube support plate (TSPS) region of Tube R20-C102. Also shown is the location of the burst fracture opening. The burst opening extended beyond the TSP crevice region. but the corrosion cracking on the burst fracture was confined to the crevice region. 3-37

as nix a v cn a r i.s2 ais nix se e.x sse che en a s rta men i set y e ses it satt -.. sn 1 -e.es i 1 lsa r'f p cartwr aan l ren g a-pr s,tto 23.s3line e man su a I r m - c~-

}

vp, e.22 M 144 vpp e.27 M 14e 30s M L"- ,7 M H ~ jC 4.3s 30s Kha cH 3 30 3.03 33e Ehr CM s 20 3I seu 'E II N A tee mee s ec==> e een vp, i.e4 m os 4e= vp, e.5, a sa m m ** a n '* is== > "'I i i. ( / h j een i n se. i _ [__ ^W t ~ ~- -e_ A f C~~~ r Figure 3-9. Laboratory Reevaluation of Field Bobbin Data for R3C107 Sif l 3-38

19l nts ! V De & W 3.83 1:5 n!N 1 as 3.03 550 sh CM i as N ra = ses se.rt _- tu l e.asl:I len uitwr stu l Tu g artro 23.s3lin/ e ranin m y vpp 9.23 M 61 Vpp e.2e M 51 Ma ( b_,_) en y = = a [ b ] b m 4.n m tha en a m 4e.4e am th cn s 3n leH k Kame 0 887ts 130 v,, e.72 m!I 42 m v,, e.n sui 354 es 'w ** a n'* is '= s>><ier0 Q ) } rue ~ ] p -- 2

== ,-=v p Figure 3-10. Laboratory Reevaluation of Field Bobbin Data for R3C107 711 3-39

30 01 8 Y CM 0 V 8.75 til n!E 8 5 1.75 598 Ehe 011 5 gg RUL IN s/s seart -.+ sn i.e.e31 1 lin gag ENTENT tic l ftM E 9,50 21.24lin/see TMl_si m get \\ V,, .2 e 45 v,, ..., e 3 = L E seN .rv' 2 m n-ML A v IIM 13.45 300 the Of 3 de 53.81 13e the 06 5 160 11C I tengan eng Df 3ff scases a ests: see l vpp 8.84 E 142 et Vpp 1.51 E 146 et h' % # **'#8:s esse Fett e a 3 4 s i e 7 of' ggg 3 4 A P 5 h im a a V T l l l Figure 311. Laboratory Reevaluation of Field Bobbin Data for R20C7 SH l 3-40 1

30 Of 1 V Of e V 1.75 1:5 NIX 1 5 1.01 55e pas CM 1 as g E Sl0E y ~ S/C Itt.ET ~ ,,, N l-0.13 l tl lIN l g EXTENT 11C l TEM _ E S,EED 21.21lIn/see TeAIN get M Ypp 9.35 E 47 Ypp e.16 E 242 -h D l f a~ r 2. \\ 2 11M 13.45 300 Kha CH 3 4e 53.81 130 Khr CH $ 16e !!C g- ,,, 2.21 se m v,, 4.ie an 13e e + = =;;,.;i i= } l w = ; s= s m I l l Figure 3-12. Laboratory Reevaluation of Field Bobbin Data for R20C7 7H 3-41

Figuro 3-13 Comparison of 3/4" Leak Test Data with CRACKFLO Predictions 100.000 f Pulled Tubes Model Boiler PWI d s o + 10,000 M.S.598-3 M.B. 604-2 CRACKFLO E E X ~ s 1.000 ~ 0 E 100 e 9 E 3 ++ s ~ e L+ g 10 en y, O + 7+ 1 = O.1 E E I I I 0.01 O.01 0.02 0.03 0.05 0.1 0.2 0.3 0.5 1 Throughwall Crack Length, inch 3-42

A 1 h Figure 3-14: Burst Pressure vs Bobbin Amplitude 3/4" x 0.043" Alloy 600 MA SG Tubes (NRC Database @ 650 F) 12.0 i 4 i + Data adjusted to 650*F LS Regression N --- 95% Prediction 10.0 l

---

95% Prediction & LTL

~~ ' s. N

---

SLB RG 1.121 I.imit N

N

---

SLB Pressure A

8.0 '..~. A Byron 1 (1994) s s s n + LS w/ Byron 1 .52 i s , \\ Reference or = 75 ksi 8 S s. N, g 6.0 s. s x s w x s a. i ~ s n c ~. m N 4.0 3.657 ksi ..t. .,,] s. 2.560 ksi N N ..L... '. n ' N c.,s 2.0 '- 7. s g+ s .,' k.. 1 .s .s. l -d,,, I 4.6 V .- 11.7 V - -s t 0.0 O.1 1.0 10.0 100.0 Bobbin Amplitude (Volts) l [34PV_CAE.XLS134PV.CAE RFK: 1/13/95,3 39 PM

Figure 3-15: Probability of Leak for 3/4" SG Tubes @ 650 F, AP = 2560 psi Comparison of Byron '94 Data with NRC Reference Database 1.0 a /'W NRC Database O.9 -- A Byron '94 Log Logistic 0.8 -- --- Log Logistic w/ Byron '94 $ 0.7 m E ~@0.6 O 3 0.5 o x g 0.4 S .oj0.3 0.2 I 0.1 g 0.0 2 ^- 0.01 0.1 1.0 10.0 100.0 Bobbin Amplitude (Volts) 3-44 [MPL.CAF_XLS) PoLCAE RFK: 1/13/95,3.39 PM

m, Figure 3-16: SLB Leak Rate vs. Bobbin Amplitude 3/4" x 0.043" Alloy 600 SG Tubes @ 650 F, AP = 2560 psi 10000. _,r" c '_. - ~ ,7 I a Test Data s' f j ~ Regression Fit a .. 'n e /' --- Arithmetic Average (AA) ,e p 1000'---

---

95% Confidence on AA

_/ / +~ 2

---

AA Leak Rate

• Pr(Leak) e' A

Byron 1 Data (1994) + LS w/ Byron 1 Data ,c' /' ,s / s 100. -. 9'Mw- = 1 R l'" ~.. 413 - %., ~ S ,e j G N 3 7' s n g u-a p-um y n r, x .u s.r-D 7 / f _[g 3 ,C / l / / a2 .r 1. a ".4.- v1 s-g. o l+/ - ,/ 0.1 ,3-t i / / s ~ _o_/ / / 0.01 1 10 100 Bobbin Amplitude (Volts) r34ov_CAE.XLS] OvsV.CAE RFK: 1/13/95. 3:41 PM

4.0 EOC-6 INSPECTION RESULTS AND VOLTAGE GROWTH RATES 4.1 EOC-6 Inspection Results 4.1.1 Bobbin Coil Inspection The Byron 1 EOC-6 inspection was performed in September,1994, and was the first inspection implementing a 1.0-volt IPC. The prior, non IPC inspection was performed in February 1993. Under the IPC guidelines for calibration, identification, and measurement of signal amplitudes,100% of the tubes were inspected full length with bobbin EC probes. RPC examination was performed of all bobbin indications reported with amplitudes >l volt to identify confirmed indications >l volt and all indications >2.7 volts for repair. RPC examination was also performed for a significant number of signals less than 1 volt. The EOC-6 bobbin indications were all classified as distorted support indications (DSIs). Reviews were performed during the analysis of all dented TSP intersections (DNIs) to establish if the dented intersections contained flaw-like components. No bobbin or RPC indications were found at dented intersections. Dents are not included in the tabulations of this report. A number of indications reported at the EOC-5 inspection were not observed at EOC-6, and were therefore not included in the EOC-6 results. 't 1 The IPC analysis database differed from the final plant database by three small voltage ~ indications. Two SG-C final plant indications were not included in the IPC database: R3C74 SH, a 0.53 volt indication RPC confirmed at 0.41 volts and not plugged; and R21C619H, a 0.52 volt signal which was not RPC examined and not plugged. A 0.63 volt indication at SG-C R19C58 9H which was not RPC examined and not plugged was included in the IPC database but should have been deleted. All of these signals are low voltage, and are therefore judged to have negligible effect on the results of the analyses in this report. The distribution of the bobbin indications versus vc!tage at dyron-1 is provided in Table 4-1 and EOC-6 voltages are shown in the Figures 4-la through 4-le histograms. The bobbin indication voltages above about 3.2 volts are shown as discrete values, to reduce roundoff errors from binning. A total of 3075 ODSCC indications at TSP intersections were reported at EOC-6, of which 802 were in SG-A,914 in SG-B,899 in SG-C, and 460 in SG-D. Figure 4 la shows the total number of bobbin indications for all four SGs, while Figures 4-lb through 4-le show indications for each SG at EOC-6. SG C, with 616 indications returned to service at BOC-7, is the limiting SG for SLB leakage analyses, as demonstrated by comparative analysis to SG-A in Section 8. The median voltage bin for SG-C was slightly higher than that of the overall distribution (0.7 volts vs. 0.6 volts). Table 4-1 also shows the number of repaired indications. A total of 659 indications exceeded the approved 1.0 volt IPC repair limit for this inspection; all of these were RPC inspected and 535 were confirmed and repaired. In addition,13 RPC NDP (no degradation found) indications above 1 volt (548 535),113 RPC NDF indications (830-717), and a total of 295 indications (830-535) including RPC confirmed indications less than 1 volt were repaired because of other indications on the same tube requiring repair. Figures 4-2a and 4-2b illustrate the repaired indication distributions for all SGs and for SG-C. s:\\npckae94Wyc6ipe\\sec4678.wp5 4-1

Table 4-2 summarizes the EOC-6 indicatirn bobbin voltages greater than 4.0 volts. RPC i inspection results, voltages at the prior inspection and whether or not it is a new indication are also shown. The largest bobbin voltage indications found in 1994 were 10.95 volts in SG-C; 7.64,7.10 and 5.92 volts in SG-A; 4.56 volts in SG-B; and 4.09 volts in SG-D. -The 10.95 and 5.92 volt indications were included among the pulled tubes, and it was shown in L Section 3 that the measured burst pressures for these indications exceeded the R.G. I 121 structural margins. Thus tube integrity requirements were satisfied at EOC-6. None of the largest indications were reported in 1993 per the field analysis guidelines, and all were located at TSP 3H and were RPC confirmed in EOC-6. The largest bobbin indication of 10.95 volts, SG-C R20C7 3H, was RPC confirmed as a 4.59 volt MAI (multiple axial indication). Reanalysis of the EOC-5 EC tapes per the Appendix A analysis guidelines showed this indication as 1.09 volt. The EC bobbin responses for SG-C R20C7 3H st BOC-6 and EOC-6 are shown in Figure 4-3 and the EOC-6 MRPC (multiple rotating pancake coil) trace is shown in Figure 4-4. The EOC-6 response for 3H in the 300/100 kHz mix channel gives a clear 10.95 volt indication. The MRPC response shows two closely spaced axial indications of comparable amplitude and a smaller amplitude indication. The BOC-6 response for this intersection in the mix channel produces a 'l.09 volt indication. (No'~ the Figure 4-3 voltage differs slightly from the reported value because of calibration / setup differences.) The '93 indication is distorted by a contribution from the mix residual. 4.i.2 Number of Indications vs. TSP Location Table 4-3 provides a tabulation of the 3075 TSP ODSCC indications by TSP elevation, and i Figure 4-5 provides a histogram displaying this information. For the Byron-1 Model D4 SGs, the IH level represents the Flow Distribution Baffle (FDB), a plate with oversize tube holes and no flow holes. The interim plugging criteria are not applicable to this location per the draft generic letter guidance. Few indications are expected, and none were found in Byron 1 at EOC-6, at lH because of the hole geometry. The support levels above the FDB are numbered in the cold leg order, i.e., the next hot leg TSP is 3H since its height corresponds to the 3rd preheater plate. The remaining TSPs are designated 5H,7H,8H,9H,10H, and 1IH. There is an expected decreasing probability of encountering ODSCC signals with increasing hot leg tube support plate elevation, consistent with observations, since both the tube internal temperature and primary-to-secondary differential temperature decrease with higher elevation. 95% of all reported TSP ODSCC flaw indications at Byron 1 were found at the first three hot leg TSPs: 3H, SH and 7H. No indications were found at cold leg TSPs. 4.1.3 Rotating Pancake Coil (RPC) Inspection As mentioned above, all indications >l.0 volt in flaw signal amplitude, plus a number of smaller indications, were subjected to 3-coil RPC probe inspection. Consistent with the Appendix A requirements for RPC testing, overs,ll objectives of the RPC testing were to (i) assess the consistency of the underlying tube condition with prior cases of TSP ODSCC, (ii) l support the indications as ODSCC with small dents, (iii) identify that the indications were within the TSP, (iv) confirm the absence of circumferential indications, and (v) determine the i n s%pc\\cae94\\cyc6epcinec4678.wp3 4-2 1 \\ .~.

cxtent of tha indicati:n with respect to repair criteria. The RPC' inspection met these objectives, in that it confirmed that all indications were axial ODSCC located within the i TSPs. No indications were found by the RPC inspection of dented TSP intersections. At Byron-1, it is expected that the dent indications are mechanically induced dings rather than i corrosion-induced denting. l Table 4-1 includes the distribution of RPC confirmed indications and repaired indications as I well as the total bobbin voltage distribution, and Figure 4-6 shows the RPC confirmation rate for each bin as a percentage of indications examined. From Table 4-1 it can be seen that of l 967 indications which were RPC tested,717 were confirmed, a 74% confirmation rate. l Above 1.0 volt,535 of 659 indications or 81% of indications were RPC confirmed, while below 1.0 volt,182 of 308 or 59% ofindications were confirmed. All indications greater than 2.2 volts were RPC confirmed. Previous experience with inspections performed under IPC analysis guidelines show that these are relatively high confirmation rates. Table 4-4 provides the same bobbin indication and repaired indication listings for EOC-6 as Table 4-1, but is supplemented by two columns which provide the distribution ofindications returned to service in BOC-7. The first of these columns shows all indications returned to service, regardless of RPC confirmation. From this table, it can be seen that a 2.1 volt indication in SG-C is the largest indication (RPC NDF) which was returned to service, and that the total number of indications returned to service was 2245 for all SGs, and 616 for SG-C. The next column of Table 4-4 provides the list of RPC confirmed or not RPC inspected 1 indications which were returned to service. This distribution is reduced from that of the prior i column, since it excludes RPC NDF indications. All confirmed indications greater than 1.0 i volt are repaired, so no indications greater than 1.0 volt are present in this distribution. Since RPC NDF indications are not expected to contribute appreciably to the leak rate or burst probability over the next cycle, this provides the more relevant BOC distribution. Figures 4-7 and 4-8 illustrate the Table 4-4 distributions of indications returned to service for all SGs, and l for SG-C, respectively. 4 d2 Voltage Growth Rates l The Cycle 6 voltage growth rates were developed from the September,1994 (EOC-6) field l analysis data and by analyst review of the February,1993 data (EOC-5) for the same i indications. 'Ihe Byron /Braidwood inspection guidelines were enhanced to increase j detectability of flaws and reduce voltage variability. These guidelines included cross-calibration of the field ASME standards to the reference laboratory standard. The Byron /Braidwood guidelines were prepared by CECO consistent with Appendix A of V. C. Summer WCAP-13522 with exceptions specified in the Byron IPC submittals and supplemental information. Growth rates were developed for all potential flaws. The growth rate analysis was performed in accordance with guidance of the draft generic letter in that growth rates were not calculated for the 1993 indications which were reanalyzed as NDD. For this reason, the total number of voltage growth values (2851) is smaller than the total { number of EOC-6 indications (3075). ) j s%pekee94;cyc6ipchec40s.wps 4-3 i ..n-n- ,.. +..

,.= Average growth rates were calculated as a percentage of the beginning of cycle voltage for all SGs and also as a composite for Cycle 6, and as a composite for plugged tubes for Cycles 4 - and 5. The' values are shown in Table 4-5. The Cycle 6 composite value of 67% is similar to the Cycle 4 and 5 values of 80% and 68%, although the Cycle 4 and 5 data are' for plugged tubes only and are therefore expected to be higher than for the entire population of Cycle 4 and 5 indications. SG-A experienced the largest growth rates during Cycle 6, an average of 117%, while SG-D, C and -B values followed at 69%,66% and 39*4, i respectively. In all cases, percentage growth rates were significantly. lower for BOC voltages above 0.75 volts than for those below 0.75 volts. SG-A indications above 0.75 volts experienced the largest differential voltage growths, averaging 0.58 volts over 58 indications. This value is skewed by three large growth values for SG-A with BOC voltages >0.75 volts, shown in Table 4-6. Voltage growth distributions, shown in Table 4-7, were developed for ecch of the four SGs. Figure 4-9 illustrates the cumulative growth distribution for each SG, and Figure 4-10 illustrates the binned growth data by SG. SG-A had the largest average growth rate and the most conservative cumulative probability distribution, except for the upper end of the distribution. SG-C growth was the next higher distribution, but was higher than SG-A at the upper end due to SG-C R20C7 3H, which had the largest Cycle 6 growth value of 9.86 volts. This growh was determined from an EOC-6 voltage of 10.95 voits and BOC voltage of 1.09 i volts. This intersection was pulled and metallurgical results were described in Section 3. Three other Table 4-6 growth values, all in SG-A, were larger than 5.0 volts. All of the largest growths we.re new indications, i.e., not reported at EOC-5. 'Ihe voltage growth distributions were rnodified prior to use in the voltage projections in that negative growth rates, i.e., the values in bins less than 0.0 volts, were s.ided to the 0.0 volt bin. In effect, this changed r,egative growth values into a zero growth value. It can thus be seen in Table 4-7 that the cumulative probability for SG-C voltage growths is 16.3%, Figure 4-10 provides a histogram of the growth results. The median growth bin was 0.2 volts for all SGs, except for SG-C, where it was 0.3 volts. Cycle 6 was reported to be 466.5 EFPD in length; no cycle length adjustments are made to the growth values reported in this section. ) Adjustments to the Cycle 6 growth distribution for Cycle 7 length are, however, included in i the EOC-7 projections (see Section 8). Figure 4-11 provides a histogram of the voltage growth and a cumulative distribution of voltage growths for all SGs. Figure 4-12 provides the same information for SG-C only. As mentioned in the discussion of the average growth values in Table 4-5, differential voltage growth rates tend to decrease for higher BOC voltages; Figure 4-13 illustrates this trend for Byron-1 Cycle 6. The negative slope in this BOC voltage versus voltage growth scatter plot provides lower percentags growth rates for indications with higher BOC voltages. 4.3 NLE Uncertainties The NDE uncertainty was provided in Reference 11.8, EPRI Report TR-100407, Revision 1 (Draft of August 1993), and this EPRI NDE uncertainty is applied for Cycle 6 at Byron-l. s Aapckae94Wyc6ipchec4678 wp3 4-4 ,;-a n w e

The NDE uncertainty is principally due to probe wear with a standard deviation of approximately 7% about a mean of zero and analyst variability with a standard deviation of i 10.3%. nese distributions are applied as normal distributions and combined to obtain a net NDE uncertainty of 12.5% for one standard deviation. De upper bound on the probe wear uncertainty is limited to 15% by the Byron-1 IPC requirement to replace the bobbin probes when measurements on the probe wear standard for a worn probe differ from that found for the new probe by 15%. The upper bound on the analyst variability uncertainty would be limited to 20% by the eddy current analysis guidelines which require lead analyst resolution of bobbin voltages (with one or more reported above 1.0 volt) differing between analysts by more than 20%.. However, as of the February 3,1994, NRC/ Industry meeting the NRC has not accepted the 20% cutoff on the analyst variability uncertainty. Pending a further resolution of this issue with the NRC, the analyst uncertainty is applied without a cutoff for the EOC voltage projections performed in this report. For EOC voltage projections, separate distributions of 7% with a cutoff at 15% for probe wear and 10.3% with no cutc8f for analyst variability are applied in this report. These NDE uncertainty distributions are shown in Figure 4-14. Additional discussion on application of the NDE uncertainty is included in Reference 11.5. 4.4 Distributions Requested by Draft Generic Letter Paragraph 6.b of Enclosure 1 to Draft Generic Letter 94-XX describes the information to be provided within 90 days of each restart following a steam generator inspection. Among these requirements, the following distributions are requested in both graphical and tabular form: (i) EOC voltage distribution - all indications found during the inspection regardless of MRPC confirmation (ii) Cycle voltage growth rate distribution (i.e., from BOC to EOC) (iii) Voltage distribution for EOC reps red indications - distribution of indications presented in (i) above that were repaired (i.e., plugged or sleeved) (iv) Voltage distribution for indication i left in service at the beginning of the next operating cycle regardless of MRF C confirmation - obtained from (i) and (iii) above (v) Voltage distribution for indications left in service at the beginning of cycle that j were confirmed by MRPC to be enek-like or not MRPC inspected J (vi) non-destructive examination uncerta.aty distribution used in predicting the EOC (for the next cple of operation) voltage distribution j l Table 4-9 lists these reporting requirements and the corresponding section of this report where the information is located. anapetar94\\cyc6ipchc4678.wp5 4-5

Table 4-1e Byron 1 EOC-6 (10/94) Comparison of Pls Found at TSPs, Confirmed by RPC, and Indications Repairrd Steam Generater A Stemma Generator B Seemm Generater C Stenen Generator D All Stenen Gereretees ns er ns et ns et Ns d ns et ns d Ns d ns d ne d ns d ns d ns et ns d ps et nn d ns et ns d ns d %d ps d Games leereseem leecemees temeessee Bethis tomeneses lomseelmee temcomous Bobbie Immentlene tsderanses h Sebbre leecesses todicestema lameseems Gebete tomesesse toerseeen Besseetmas todireeses e casar es noynered somenesse ree. sees conne.ed me, eared tomraense femmesed connreed stapeered tomrenese teme amed remar.ed meenered snaresseem re e e4 ca or ed sie, eared 99 9 9 9 9 0 0 0 9 9 9 9 9 0 0 0 0 0 0 9 9 el 9 0 0 0 0 0 0 0 9 9 9 0 0 0 9 9 0 0 9 0 0.2 4 0 0 0 t e 0 0 9 9 9 8 4 9 9 9 9 e 9 0 83 57 7 3 6 14 1 9 9 13 0 9 9 27 1 0 9 ttt 9 3 6 94 ISS 16 9 16 49 7 3 6 45 3 1 5 65 3 I I 264 29 14 29 0.5 te$ 33 6 to 94 9 2 6 92 4 2 6 82 4 2 4 373 29 12 26 96 106 17 7 15 134 Il 6 8 95 4 2 7 $7 7 5 6 299 39 29 36 9.7 92 18 14 15 110 20 13 I? 139 18 13 25 60 5 4 4 390 68 42 61 85 73 to 7 13 121 16 12 15 116 19 12 21 40 3 2 3 350 $2 33 52 49 64 23 11 1C 92 II S 9 89 9 6 9 28 9 0 0 275 41 25 34 1.9 47 16 13 31 Si to 7 8 92 19 12 17 24 I I E 244 46 33 37 1.8 34 34 31 31 68 69 54 Se $9 59 47 47 19 19 12 12 ISO 109 344 146 IJ 31 _ 31 24 25 46 46 36 36 49 de 26 39 13 13 7 7 139 139 93 90 SJ 28 e* 22 23 39 30 25 25 33 33 27 27 Ir 12 7 7 193 393 et 32 1.4 IS 18 15 17 29 29 29 20 26 26 24 24 4 4 4 4 77 77 63 65 1.5 6 6 4 4 12 12 9 9 17 17 16 16 12 12 to 19 ~g7 4? 39 39 1.4 6 6 4 5 7 7 7 7 18 10 9 9 0 9 9 9 2J 23 29 21 3.7 4 4 4 4 4 4 4 4 13 13 11 12 7 2 2 2 23 23 28 22 I.S 4 4 3 4 3 3 3 3 8 8 8 8 I I t 1 36 16 15 16 1.9 4 4 4 4 1 1 1 1 4 4 4 4 0 9 7 8 9 9 9 9 2.9 I I I 1 4 4 4 4 9 9 9 9 19 10 to se 21 2 2 2 2 1 1 1 I 2 2 I I 4 e G 9 5 5 4 4 2.2 3 3 3 3 0 9 0 0 I 1 I t 0 9 9 0 4 4 4 4 2J t i 1 1 2 2 2 2 1 I i 1 9 9 e G 4 4 4 4 24 2 2 2 2 0 0 0 0 1 1 I t 0 9 0 0 3 3 3 3 2.5 1 1 1 I 2 2 2 2 2 2 2 2 0 9 9 9 5 5 S S 2.6 1 1 1 I I I I i 1 I i 1 9 0 0 9 3 3 3 3 2.7 9 0 0 9 1 1 1 1 2 2 2 2 9 9 0 0 3 3 3 3

3. 8 2

2 2 2 1 1 I I 9 0 0 0 0 0 0 0 3 3 3 3 3J e 0 9 9 9 9 9 0 0 9 e e t i 1 1 3 1 1 1 3J7 I I i 1 0 0 9 9 9 9 9 9 0 9 9 9 I I I 1 3 64 9 9 9 9 9 9 9 9 I I I I 4 9 9 9 I I I i 3.09 0 0 9 0 9 9 9 9 1 1 1 8 9 0 9 9 I I I I 3.90 t 1 1 1 0 0 0 9 9 0 0 9 9 9 9 9 1 1 I I 4 99 9 0 0 9 0 0 0 9 9 9 0 0 t t t I I 1 I I 4 56 0 0 0 9 1 I t t 9 9 0 9 9 0 9 9 8 I I S 92 1 1 I 1 9 0 0 9 9 0 0 0 9 9 9 9 8 1 1 1 7.19 1 1 I I 9 9 e t 9 9 4 0 9 9 9 9 I I I t 7.64 1 I i 1 0 9 0 0 0 9 9 9 0 0 0 9 I I 1 3 19 99 0 0 0 0 0 0 9 9 3 1 I 5 0 9 9 9 I I I I Total 302 281 203 243 914 2 94 22e 249 899 303 134 283 449 39 60 64 3975 967 717 839 Tetel o tv 197 IS7 133 339 219 219 169 171 227 227 ISS 193 65 65 45 45 659 659 535 l 548 c b^e ^-.~.LCAE94RFT XL5 T4_1 r 4-6

Table 4-2 Byron-I: Largest EOC-6 TSP Indication Voltages (Indications >4.0V) EOC-6 BOC-6 New S/G Row Column TSP Bobbin Volts RPC Volts Voltage Indication? C 20 7 03H 10.95 4.59 1.09 Yes A 3 3 03H 7.64 5.74 0.93 Yes A 20 7 03H 7.10 3.69 1.14 Yes A 3 107 03H 5.92 4.03 0.44 Yes B 20 104 03H 4.56 1.97 0.91 Yes D 25 38 03H 4.09 3.81 0.35 Yes l 4-7 C;%PCWYRONAPCVtEPORTCAE94RPT.XLS T4_2 EOC6V>4 0

1 Table 4-3 Byron-1: No. of EOC-6 ODSCC Indications vs. TSP Location TSP SG-A SG-B SG-C SG-D Total % of Total 03H 416 603 605 303 1927 62.7 % 05H 219 225 253 113 810 26.3 % 07H 96 47 17 28 188 6.1% 08H 42 30 16 3 91 3.0% 09H 23 9 6 7 45 1.5% 10H 6 0 0 3 9 0.3% llH 0 0 2 3 5 0.2% Total 802 914 899 460 3075 l I 4-8 C:WPCWYRONAPCREPORTT,AE94RPT.XLS T4.3 TSP Ind

d p e e a e ul t e 3 0 9 0 4 2 e 9 r n ta e 9 4 9 0 3 3 5 2 0 3 9 9 0 e ee O e 0 ee 0 0 #0 O e 9 9 ee e e 9 99 0 e. e. e n 1 2 3 3 3 3 2 1 eR_ te 2 Cr. m e s r sh,- o sn ta e s 6 7 6 9 6 9 7 4 2 3 2 s r e 9 4 9 e 3 4 5 2 4 3 9 o e 4 i 82 3 0 e e 5 0 0 0 9 e 9 4 e e a e 9 e e 0.. e i 3 3 2 1 l 2 3 3 3 3 2 2 i n M 2 t t e 2 a G c m id a r e e 8 6 6 1 2 4 7 6 n 8 2 5 9 1 2 6 4 e o . e t a 9 4 0 6 4 4 4 3 3 3 3 3 I t I I a I a I t I 3 e 2 2 3 6 $ 3 3 9 8 6 3 2 1 1 t I S M 1 8 s d M A e tc f e o l 4 3 2 8 0 5 4 0 e 3 7 7 3 3 6 s , e 9 p a 0 0 9 i 6 7 9 9 5 1 4 0 3 0 7 4 2 2 1 l 5 4 4 3 5 3 3 3 3 i I 1 I e t I 3 I 1 v t 2 3 3 3 3 2 2 1 1 1 e s M 3 n I p v i d ee_ C eml Ae _ s F 6 2 6 8 S 4 9 3 r se 0 0 4 2 6 7 $ S 4 2 2 0 e 9 e 9 e e 9 e 0 e 0 0 e 9 e 9 0 e e e e 9 e 0 e. . e 6. e R sl t 3 af e t Ce. m r o D N r t r a e^ 9 9 4 1 6 7 S $ 4 2 2 7 4 8 t 6 5 8 3 6 a 7 6 8 e 2 e e O e 0 9 0 0 e 8 e 9 0 e 0 e e 9 e 0 e. . e, a_ o 2 M^^ 3 de m C ^ i e f r s 2 t om f te M_ _ 9 8 4 e 3 4 6 4 3 0 1 7 7 4 8 2 I e 0 9 0 0 8 e e 0 9 I e e e 9 I 0 0. . e i e a 4. 1 I 6. n S oC C d ' 7 5 2 7 0 8 0 4 9 3 1 a - 9 0 4 2 e 4 0 2 1 e e e 9 0 9 e 0 0 9 1 e e 0 9 1 0 0. . e e s 2 6 8 5 6 4 2 2 1 1 1 1 P M 4. R d d a s p _ e n e e a r ne-mr 3 9 S 8 3 4 8 1 1 s aM to 0 0 0 A e 1 3 S 8 0 9 7 7 9 e e 4 0 0 0 8 e 0 e 9 8 9 e 9 9 9 e 8 e 0 9 e 0 0 e. e 7 s 1 t s n o Ce. n r o i C ta r t a' /.. 3 e 6 8 5 o 2 e c a e 9 0 0 3 e t 0 6 2 1 1 1 9 e 0 I 9 4 e 8 e 9 9 e 9 e 0 0 9 9 0 e. o i 4 1 d 5 S 9 t 7 1 l id r M-1 s 0 S 3 e n n 9 I e G d l l f e 4 m o_ r A e a ae 9 4 0 0 5 6 7 9 3 l 7 7 e 7 4 6 2 3 9 8 4 4 I 1 1 I 2 I 2 9 e 0 I 1 9 e 8 0 e. i s 2 I 1 4 3 2 1 1 s 1 ee M e 2 i e S R c iv em-t r a 0 8 9 3 5 2 5 9 6 9 2 9 t 3 6 7 e 3 e ,9 7 1 4 9 9 3 1 8 4 4 2 3 8 I 2 I 2 9 e 9 t 1 9 e 9 0 e. I 9 9 e 3 2 1 l 1 S 2 M 1 1 e 2 o t p d a e d e e n A te s e ml 4 0 5 3 1 5 r to 0 9 1 2 0 9 l I e 1 4 8 e T 0 e S e 0 8 0 0 e 9 4 9 9 9 9 9 6 9 e 0 0 9 0 e 9 8 e. e at. r eM_ e. 1 1 1 o r s s u Ce. a te B e, s R r r e e a as 6 t 6 e. ae en e 3 s 3 3 2 e 4 s e 0 9 1 v s l 4 s 2 e 4 8 7 1 l S 9 3 e 0 0 e 9 4 8 8 S 9 9 9 0 e 0 4 9 0 o e 0 e. e 7 n Mt o 1 l 1 d t 6 3 o a le ita C e t e c m e. k 7 5 6 6 5 e e i a a e 9 t 0 e 6 6 B 9 S 3 2 2 97 4 3 I 1 1 0 2 e 2 I 1 1 e 0 e 0 0 e I 9.. e 4 7 d s 1 1 e M_ e n t 2 i S t I )4 t 9 a 4 9 4 4 t 8 9 St 8 6 0 9 2 4 i 2 / 9 e 3 0 h 1 4 9 3 i 2 7 4 3 1 1 1 0 2 9 2 I 1 1 e 0 e 0 9 e t 9.. e 1 6 4 3 2 1 1 l 1 9 m 1( 6 d p e e e mte s C r t e 1 S 3 0 6 9 3 4 ee s 9 e 4 6 5 S 9 3 6 5 4 3 9 e e 9 e e 0 0 e 9 8 0 9 0 4 9 9 9 e 9 e 9 8 e e 9 e. e 2 O nR. s. e s e E Cr. R e 1 A r r n o a e a" e 9 e 4 t 3 9 9 6 6 4 3 3 6 5 8 2 I 0 0 e 0 0 9 4 0 6 9 0 0 8 e e 9 9 e t O O . e 5 4 w 9 5 5 7 0 0 6 4_ o ta r S 9 Mm e r r B e s i 7 s e y n s t G z x s t. m e ? 6 e $ 5 3 B l 3 5 3 7 3 9 p 4 a a S 9 9 6 4 S 4 4 4 5 2 3 I 2 I I e 2 0 I 6 9 I 0 0 I I i e 4 1 l l 1 1 I I 3 2 2 3 a e M-u t 4 4 S 2 r e s re l e b fo ne 7 $ 5 6 a o S e 4 S 9 0 S 7 6 4 3 3 2 1 2 3 6 7 4 8 0 8 2 7 T - gd k 6 6 4 4 4 S 2 3 I 2 I I 9 2 9 1 9 0 I 0 0 I I i e e s I 1 1 t s i g e a_ sn ev t t 8 9 6 2.. s t. J J 4 5 6 7 3 9 9 l 2 J 4 5 6 7 8 9 9 1 J 3 4 5 4 7 8 2 J 6 7 4 M 9 s 9 0 3 9 e e S S 0 0 4 9 e e I I 1 I I I 3 I 1 1 2 2 2 2 2 2 2 2 3 3 u u t 1 te e. w r -a_ e e J 3 3 3 4 4 $ e l jt (ltllI1l111 1lll

Table 4-5. Average Voltage Growth Per Cycle for Byr:n-1 Total No. Average Average Percent Cycle of TSP BOC AV Growth Indications Voltage per Cycle per Cycle I Cycles 4 and 5 (Plugged Tubes Only): Cycle 4: 1990 to 1991 All Indications 550 0.32 0.26 80% Cycle 5: 1991 to 1993 All Indications $32 0.46 0.31 68 % Cycle 6: '93 '94 SG-A All Indications 785 0.35 0.41 117 % Vooc < 0.75V 727 0.30 0.40 134 % Vooc 2 0.75V 58 1.03 0.58 56% SG-B All Indications 777 0.59 0.23 39% V,oc < 0.75V 555 0.44 0.29 05 % Vooc 2 0.75V 222 0.98 0.09 9% SG-C All Indications 842 0.52 0.34 66 % Vooc < 0.75V 632 0.41 0.38 93 % Vooc 2 0.75V 160 1.00 0.19 19% SG-D All Indications 447 0.40 0.27 69% Vooc < 0.75V 413 0.35 0.29 83 % Vox 2 0.75V 34 0.98 0.06 6% All SGs All Indications 2851 0.47 0.32 67% Vooc < 0.75V 2377 0.37 0.35 94 % Vooc 2 0.75V 474 0.99 0.18 18% r w =9ecyc6 pew. twps 4-10

i i Table 4-6 Byron-1: Largest Cycle 6 Voltage Growths (Values >2.5V) EOC-6 Bobbin RPC BOC-6 Vohage New l SG Row Column TSP Volts Volts Voltage Growth Indication? j C 20 7 03H 10.95 4.59 1.09 9.86 Yes A 3 3 03H 7.64 5.74 0.93 6.71 Yes A 20 7 '03H 7.10 3.69 1.14 5.% Yes A 3 107 03H 5.92 4.03 0.44 5.48 Yes D 25 38 03H 4.09 3.81 0.35 3.74 Yes B 20 104 03H 4.56 1.97 0.91 3.65 Yes C 32 30 03H 3.64 2.32 0.68 2.% Yes l C 25 13 05H 3.80 2.36 0.93 2.87 Yes A 16 8 03H 3.98 1.76 1.29 2.69 Yes i B 37 28 05H 2.66 1.66 0.15 2.51 Yes i I i t 4-11 C:WPCMYRONAPCVtEPORTCAE94RPT.XLS T 44 12/15S4 --n

e I Table 4-7 i i Byron-1 Cycle 6 Growth Distribution by SG i (2851 Total Indications) SG-A SG-B SG-C SG-D EN N_9_ Cum No - Cum % N2 Cwn Ng Cum % Ng_ Cym No can % N.g Cum No Cm 3 M78 : 0 '0 (0.0% !0}; iOf fomh 0 t e S., 50Ah 1,.0 !: 74.

1. A%

{ w

1+

D.O 3% + 0 :? E01 3 08% 0: + its Fes% C o ?, a6;, 2 0.ch

m..

i. 4

m. A0 3 91-

- y e. v.;; . ~. f v.3 m 2-i ?34 COA % 0 0 :. if D.) M. OD% 0'- 3 % S,0% ^ (0} itt \\$A4 { 34.301 , 2, -, z-

1. 01 w -- ~.

s. x

a.

J< %. MSA.%; MSN +y-., V4M,3 es !.: +3b .d teh J24 4.. ~3.n;.. a@ SA% -It$ ' N. m..+k h f' ~ - - eV&th r .-. -4 .m o .y kk.pmo%[ k.g) [Sh NM~.h g$64.f kh[s n n sa; ~c. h.c w ~ w 4 a, f/ .M. .}M ty r.- g.:.. e~ gr~ n.: 2~ w.4., c y w..m u e, s.m. _S},36 h /AS$y' 'MI3h : @S3 % #C y E' JN..q. Nw.y i/4

g. h W h e

a. %;:; ..,...s. 3.;. m.s q .m-w m e.m -. e y... g ; 3 u :< %~., m.y :;., 3. M gg;^ 3 R Jg Yg 1., - + g' d jg 2 - k..g{ M N:: g j fgl 7 gg A < t j$ [$ ' {3)$ %.jp 3 0 00 21 46 5.9% 69 153 19.7 % 61 137 16.3 % 26 57 12.8% t 0.10 73 119 15.2 % 123 276 35.5% 101 238 24.3% 74 131 29.3 % 0.20 128 247 31.5% 149 425 54.7 % 99 337 40.0 % 79 210 47.0 % 0.30 121 368 46.9 % 97 522 67.2 % 107 444 52.M 61 271 60.6 % i 0 40 112 480 61.1 % 86 608 78.2 % 99 543 64.5 % 73 344 77.0 % 0.50 83 563 71.M 49 657 84.6 % 80 623 74.0 % 33 377 84.3 % l t 0.60 $3 616 78.5% 45 702 90.3 % 76 699 83.0 % 23 400 89.5 % 0.70 47 663 84.5% 33 735 94.6 % 36 735 87.3% 11 411 91.9% 0.80 42 705 89.8% 14 749 96.4 % 32 767 91.1% 8 419 93.7 % 0 90 23 728 92.M 9 758 97.6 % 18 785 93.2 % 10 429 96.0 % l.00 18 746 95.0% 6 764 98.3 % 14 799 94.9 % 11 440 98.4 % 1.10 10 756 96.3 % 2 766 98.6 % 17 816 96.9 % 2 442 98.9 % I 1.20 6 762 97.1% 1 767 98.7 % 5 821 97.5 % 0 442 98.9 % 'I l.30 4 766 97.6 % 2 769 99.0 % 5 826 98.1% 2 444 99.3 % 1.40 1 767 97.7 % 2 771 99.2 % 1 827 98.2 % 0 444 99.3 % 1.50 3 770 98.1 % 0 771 99.2 % 1 828 98.3 % 1 445 99.6 % !.60 2 772 98.3% 1 772 99.4 % 1 829 98.5% 0 445 99.6 % 1.70 2 774 986% 1 773 99.5 % 5 834 99.0 % 0 445 99.6 % I 1.80 1 775 98.7 % 0 773 99.5% 1 835 99.2 % 0 445 99.6 % 1.90 3 778 99.1% 1 774 99.6 % 1 836 99.3 % 0 445 99.6 % I 2.00 1 779 99.2 % 1 775 99.M 0 836 99.3 % 0 445 99.6 % f 2.10 2 781 99.5 % 0 775 99.7 % 1 837 99.4 % 0 445 99.6 % l 2.20 0 781 99.5% 0 775 99.7 % 2 339 99.6 % 0 445 99.6 % I 2.30 0 781 99.5 % 0 775 99.7 % 0 839 99.6 % 1 446 99,8% 2.40 0 781 99 5 % 0 775 99.M 0 839 99.6 % 0 446 998% 2.50 0 781 99.5% 0 775 99.7 % 0 839 99.6 % 0 446 99.8% 160 0 781 99.5% 1 776 99.9 % 0 839 99.6 % 0 446 99.8% 2.70 1 782 99.6 % 0 776 99.9 % 0 839 99.6 % 0 446 99.8% 2.90 0 782 99 6 % 0 776 99.9 % 1 840 99.8% 0 446 99.8% 3 00 0 782 99 6 % 0 776 99.9 % 1 841 99 9 % 0 446 99.8% 3.70 0 782 99 6 % 1 777 100.0 % 0 841 99.9 % 0 446 99.8% 3 80 0 782 99.6 % 0 777 100.0 % 0 841 99.9 % 1 447 100.0 % 5.50 1 783 99.7 % 0 777 100.0 % 0 841 99.9 % 0 447 100.0 % 6.00 1 784 99.9 % 0 777 100.0 % 0 841 99.9 % 0 447 100.0 % 6.80 1 785 100 0 % 0 777 100.0 % 0 841 99.9 % 0 447 100.0 % 9.90 0 785 100 0 % 0 777 100.0 % 1 842 100 0 % 0 447 100.0 % Note: Negative growth rates are included in the Monte Carlo projection growth distnbution as zero growth values, i.e., in the 0.0 volt bin, i 4-12 i c3spctyronapcVeportCAE94RPT.XLS T4 7 n..y .wr--,- -w. ~, -. - - - -,. s-~

Table 44 Probe Wear and Analyst Variability - Tabulated Values j i - Analyst Variability Probe Wear Variability Std. Dev = 10.3% Mean = 0.0% Std. Dev = 7.0% ~ Mean = 0.0% j No Cutoff Cutoff at +/- 15% Value Cumul. Prob. Value Cumul. Prob. I -40.0% 0.00005 < -15.0% 0.00000 -38.0% 0.00011 -15.0% 0.01606 l -36.0% 0.00024 -14.0% 0.02275 -34.0% 0.00048 -13.0% 0.03165' -32.0% 0.00095 -12.0% 0.04324 l -30.0% 0.00179 -11.0% 0.05804 l -28.0% 0.00328 -10.0% 0.07656 I -26.0% 0.00580 -9.0% 0.09927 -24.0% 0.00990 -8.0% 0.12655 l -22.0% 0.01634 -7.0% 0.15866 -20.0% 0.02608 -6.0% 0.19568 -18.0% 0.04027 -5.0% 0.23753 -16.0% 0.06016 4.0% 0.28385 -14.0*/. 0.08704 -3.0% 0.33412 -12.0% 0.12200 -2.0% 0.38755 l I -10.0% 0.16581 -1.0% 0.44320 -8.0% 0.21867 0.0*/. 0.50000 -6.0% 0.28011 1.0% 0.55680 l -4.0% 0.34888 2.0% 0.61245 l -2.0% 0.42302 3.0% 0.66588 j 0.0% 0.50000 4.0% 0.71615 2.0% 0.57698 5.0*/. 0.76247 4.0% 0.65112 6.0% 0.80432 6.0% 0.71989 7.0% 0.84134 8.0% 0.78133 8.0% 0.87345 10.0 % 0.83419 9.0% 0.90073 12.0 % 0.87800 10.0 % 0.92344 l 14.0*/. 0.912 % 11.0% 0.941 % 16.0 % 0.93984 12.0 % 0.95676 !8.0% 0.95973 13.0 %

0. % 835 20.0 %

0.97392 14.0 % 0.97725 i 22.0 % 0.98366 15.0 % 0.98394 24.0 % 0.99010 > 15.0% 1.00000 26.0 % 0.99420 28.0 % 0.99672 30.0 % 0.99821 32.0 % 0.99905 l 34.0 % 0.99952 36.0 % 0.99976 i 38.0 % 0.99989 40.0 % 0.99995 I eveyronspcVeportameept.mTsede 4 a 4 13

Table 4-9 Correspondence Between Generic Letter Requirements and Report Contents Generic Letter Subject Section Table Figure Para. 6b(b) (i) EOC-6 voltage distribution - all 4.1 4-1 4-la e bobbin ind. (ii) Cycle 6 growth distribution 4.2 4-7 4-11, 4-12 (iii) EOC-6 voltage distribution - 4.1 4-1 4-2a,b repaired indications (iv) BOC-7 indications left in service - 4.1 4-4 4-7, all indications 4-8 (v) BOC-7 indications left in service - 4.1 4-4 4-7, RPC confirmed plus not RPC 4-8 inspected indications (vi) NDE uncertainty distribution used 4.3 4-8 4-14 in predicting EOC voltage distribution l s:\\npckac94wyc6rpchec4678.wp3 4-14

Figure 4-1a Byron-1 EOC-6 Bobbin Voltage Distribution for All SGs 400 l 3M i 300 t i ,250 E All SGs l g l .i l y 2m i e 150 l l l Illli 50 i l l IIs...___ Volts CAAPCBYRONAfCREPORT\\* Fig 4-In< Chart 4 4-15 '""5

I Figure 4-lb Byron-1 EOC-6 Bobbin Voltage Distribution for SG-A 120 100 4 80 E .e W \\ w ESG-A I l 40 20 a IIIIssus. o ...n .,......,.... i. is n n i. is i. i, i. i, u u n u u u u u n n n,.......... in n. Volts I C:WCBYRONAPOREPORTT' Fig 4-1e4 Chari 3 4-16

Figure 4-Ic Byron-1 EOC-6 Bobbin Voltage Distribution for SG-B 140 120 I 100 E .2 go _m - .s e5 'E E SG-B i d 60 1 2; i I 40 i 20 Iln l h..___ 0 .,...,...,....i.i,u n i. i, i. n i. i, a u n n u n u n.. n n, u. n. in,. Volts -I cwcenONAPOREPORTT' Fig 4-la-e Chart 5 4.}7

l l Figure 4-1d Byron-1 EOC-6 Bobbin Voltage Distribution for SG-C 140 l 120 100 E .h. g s .Mzy uso-c 1 O d 60 - - - = - Z i 40 l 20 I lilt..._____ O es es er os se es es et es se se as v2 as te is is at se to se in 12 as to as as at sa sa s s, su s ee s ee aw a ss s er v ie vu se es Volts I CAAPCBYRONAPOJtEPORE* Fig 4-la-e Chart 6 4.I8

I a. 2,. m i m D D G -G S S 2 E r o 3 f no i t u b i r ts i i D e l e . s - g 2 t 4 a l t o 9 el I r o i V uV 4 g m i. n iFi b i bo B i. 6 C- .1 i O i. E .1 ,i 1 no .1 i ry B 1 i i l 1 l 1 l 7 rt a h C e. a l 4 ig F

• \\

T R O i I PE R C fA N O RY D O P 0 0 0 0 o 0 0 0 0 g A 9 8 7 6 4 3 2 A Ee.;M.e$* o. C 1 !lllllIjlll

i l 6 Figure 4-2a Byron-1 EOC-6 Repaired Indications for All SGs 160 I 140 120 E All SGs ,100 E = .E ] 80 's I 60 i 40 1 20 I"""-----l"hi........___________ 0 .....,........,.... i. in is is i. is i. i, i. i, s. 2. n >>. is z. 1, si n > >,.......... :2 1 Volts C:\\APCBYRONAPCREPORD' Fig 4-2 Chart 4 4-20

M M 5 s 9 0 7 7 8 l' e ts IT S 89 4 te d e 9 3 s e l e e S 73 1 3 S C 5 3 -G 7 1 S C E 4 1 -G 5 1 Sro o t f sn 3 2 o i 2 t 1 ac t i S bd 2 n i 0 s - I 1 t 4 l d o I e e r r i V 2 9 1 ni 4 a g 0 ip il 8 F eR .ll 7 1 6 C-ti l e O E lll S I 1 "lii to -n or ll 3 1 y B 2 8 I I lll e l ll 9 3 Ell l se 3 ll t rt e a hC .l 4 G 2 -4 .l iF 5 g 0 l D 4 R 9 O l P e E tJ C t e P A N 8 O 9 R S YB S C P 5 0 5 0 5 0 s M ~ 4 4 3 3 2 2 i E,= . ] 'E C l

..~ ..a = = = > -..-..e _i m -.., -..u, J > 1 *. I I lia -T t.Temi etc i tr..E = y ma, n \\.a M n ~... -., i 1 >= --- c - =e--- < .s _r s 's =- f 4 ( 4 n.. t m o. n n.. in n ne = I -d_ = e, w......, ,e ~ 3.n - a. ~ e.n -n

• .e r ---

l 1 I l . :'1 l s = +

=.-

3 I 1 t 1 BOC-6 EC Response i u. -.= = 1 =.===. J E 2 \\ K-a-- lN 1 .t I .K-- I

=

1 i 8K 1 s*-- l ~ ~ 1 ( T I ii

==

) (i f f h ( ( i .1C [ T EOC-6 EC Response Figure 4-3. Bobbin Coil EC Responses for SG-C R20C7 3H at BOC-7 and EOC-7

\\apc\\cae941cyc6 pc\\ ec4678.wp5 4-22 i

mi.. sm os a m sm e m m e m m e e + m ic eau = = r is = reaar = is no = - as m j vert e er Sm.us e W X S Y Seale = 1.0 Filters SPf feeg erfest e M + me a 4 F48ams? 83 j W Gus&at Laff SW -s.1t. l k l 05 See L.It 1.12 Nh* 88,9 M freintmuasi e 0 z-- me.0 v r imu. e carm m ens ames somme men m.a rien asen m, a e m 4 aml i Figure 4-4. MRPC Response for SG-C R20C7 311 at EOC-6 8:%C\\CAC94kyC6ipCuec4678 wp3 4-23

Figure 4-5 Y '* " * ~ 700 600 500 E .e ~ . 400-- ] O SG-B E SG-C j 300 -- ESG-D 200 100 - E,E nm E, ,E-0 03H 05H 07H 08H 09H 10H llH Support Plate Number 4-24 c:\\apc\\byronapc\\reportV:ae94rpt.x!s* Fig 4-5 Chart 1

l Figure 4-6 Byron-1 EOC-6 RPC Inspection Statistics (All SGs) 200 100 % 180 90 % 160 -- 80 % I40 -- 70 % 8 120 -- M No. of RPC Exams 60 % j g -*-Percent RPC Confinnation E O 100 -- 50 % 5 b e 5 j 80 - -40% d k 60 - 30% l 40 - 20 % 20 --} l 10 % tilII I I n........ _ - _ _ _ _ - _ _ _ _ _ _ 0.,. 0 02 0.3 04 05 06 0F 08 09 1 11 1.2 1.3 14 15 16 17 18 19 2 11 12 13 24 15 26 li 11 3233F 364 38 398 4 09 4 56 592 F 1 1641095 Bobbin Signal Amplitude c:4M,W CAE94RPT.XLS Fig 4_6 4,3

Figure 4-7 Byron-1 BOC-7 Bobbin Voltages for All SGs All Indications vs. RPC Confirmed Plus Not RPC Inspected 400 l 350 - = l 300 -= = 5 AllIndications O RPC Confirmed + Not RPC Inspected E 250 = = = a-l = = ] 2w 4 150 = = 300 50 = = mi i Ii..._ 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 1.1 1.2 1.3 1.4 1.5 1.6 1.7 1.8 1.9 2.0 2.1 Volts 4-26 CWCBYRONAPC\\ REPORT \\' Fig 4-7,8 Chart i

Figure 4-8 Byron-1 BOC-7 Bobbin Voltages for SG-C All Indications vs. RPC Confirmed Plus Not RPC Inspected 120 100 E AllIndications 80 = = = 0RPC Confinned + Not RPC Inspected E .]g 1 o 40 =- = 20 = I I llr.... 0.0 0.1 0.2 0.3 0.4 0.3 0.6 0.7 0.8 0.9 1.0 1.1 1.2 1.3 1.4 1.3 1.6 1.7 1.8 1.9 2.0 2.1 Volts 4-27 C:\\AICBYRON APC\\REPORW Fig &7,8 Ctwt S

Figure 4-9 Byron-1 Cycle 6 Cumulative Growth Distribution by SG (2851 Total Growth Values) 100.0 % ,, m - - , - - -J:- G w r,, 90.0 % 1 80.0 % ) 70.0 % [ 60.0% -*- SG-A j -*- SG-B J'y -x-SG-C t 50.0% [ -o-SG-D e' E 40.0 % z/ 30.0 % d 20.0 % 10.0 % 0.0%...................................................................................... Voltage DifTerential(V) c:'apc'byronapcVeporrcac94rptmis* Fig 4-9 Chart 1 4-2I

Figure 4-10 Byron-1 Cycle 6 Voltage Growths (by SG) 160 140 120 jf ) G SG-A 100-7 O SG-B i I = msac m SG-D 80 i 5 4 l 60 i 1 40 20 U l II A n di J i_. ._.I _ .n. i __ o.. ...,4.

4.....u

...,...,..., i.. >> a i. is i.i,i.,,

2. n u u 2.

u 2. 2, 3. n 3. Cycle 6 Voltage Differential (V) c%._y..,~.re 94cmis+ r.: 4-10 cret i 4-29

Figure 4-11. Byron-1 Cycle 6 Growth Distribution for All SGs l (2851 Growth Values; Includes Negative Growths in 0.0 Volt Bin) i 500 IN% _w 455 450 M t 400 I#' 3.o I-3 73 370 350 -- M g 300 - / MNo. ofIndications 60 % 5 -j -o-Cumulative Percent f 50%,E 2*' 250-3 3 s $ 200 40% y 150 30 % 127 100 -- 20% so 50 E" 30 % I. 12. . m. . E. _. . _ _ _I. 0% 13 8 4 5 4 2 5 2 3 2 o o i I I I I I 1 I I I O 0.00.10.20.30.40.30.60.70.80.9 I.01.1 1.21.31.4 1.5 1.6 1.71.81.92.02.12.22.32.42.52.62.7293.03.73.85.56.06.89.9 Volts (V) C:%PCMYRONAPCREPoRTY Fig 4-11 Chart 2 4-30

Figure 4-12. Byron-1 Cycle 6 SG-C Growth Distribution (842 Growth Values; Includes Negative Growths in 0.0 Volt Bin) 140 100 % 7 +-c ^c 90 % 120 -- 80*4 101 m No. ofIndications 100 -+--Cumulative Percent 70*4 E 30 60 % e f 80 -- - - - ,o 5 y 3 } 50% y 's g 60 -- - - .E 2; w 40 % 3 E Y 40 32 v 20*4 20 -- - - - - I 10 % 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 1.1 1.2 1.3 1.4 1.5 1.6 1.7 1.8 1.9 2.I 2.2 2.9 3.0 9.9 Volts (V) l C:WPCWYRONAPCVtEPORTY Fig 4-12 Chart 2 4,3 g

k Figure 4-13. Byron-1 Cycle 6 Voltage Growth vs. BOC-6 Voltage 10.00 0 i 8.00 o 6.0'0 S 5 ii:oy 4.00 g o E S o o o o o 4. o n 2.00 e c o ao o u o g o ,, :,,* h o

• a.

~ o o o o-i. o g o

,.-: :: f..,,,, : 00 o 0 00 t._

~ p ""

• I''

'*,:'4

';g V%o o g, fg 4 o 2J)0 2 50 -2.00 BOC Voltage (V) C:WPCWYRONAPCLSCATTER.XLSr Fig 4-13 Chart 1 4-32

t. Figure 4-14 NDE Uncertainty Distributions 1 0.9 ) f 0.8 0.7 b2 -*-- Analyst Variability e 0.6 .E Probe Wear 2 m 0.5 s .sj0.4

sV 0.3 0.2

? 0.1 g 0 -40% -30% -20% -10% 0% 10 % 20 % 30% 40% Percent Variation in Signal Voltage (%) c:MpctyronapcVeporttae94rpt.xis Fig 4-14 Chart 1 4 33

5.0 DATABASE APPLIED FOR IPC C@RRELATIONS The available database for 3/4" diameter tubing IPC applications has been described in Reference 11.6, Braidwood-l IPC WCAP-14046 Rev.1. This nport includes results for the Braidwood-l tubes pulled in March,1994. Per NRC telecon request in support of the Byron-1 IPC, the Braidwood-l data is to be included for the Byron-1 IPC assessment. The Braidwood-l data given in Table 3-7 of the referenced WCAP has been included in the correlations for Byron-l. If the EPRI data exclusion criteria were applied to the Braidwood-l pulled tubes, the 0.28 volt indication at R16C42, TSP 5 would be excluded from the database. However, per the NRC telecon, this data point has been included in the correlations for Byron-1. The database applied to the Byron-1 analyses is called the NRC database to contrast it from the EPRI recommended database. Per the guidance of the Braidwood-l SER and the NRC telecon, the NRC database includes additional differences from the EPRI database. For the tube burst correlation, the NRC database includes model boiler specimen 598-1, which would not be included in the EPRI database. For the leak rate correlation, the NRC database includes model boiler specimens 598-1 and 598-3, and includes Plant S tube R28C41 at a leak rate of 2496 liter /hr as contrasted to the EPRI recommended value of 125 liter /hr. s:\\npcVee94kyc6ipcisec.x 51

I 6.0 ANALYSIS METH@DS FOR VOLTAGE PROJECTIONS, LEAK RATES, AND BURST PROBABILITY The SLB leak rate and tube burst probability analyses reported herein are based on full Monte Carlo analyses that account for parameter uncertainty per the methods described for leak rate analyses in Reference 11.6, WCAP-14046 Rev.1. Monte Carlo methods are used to develop an EOC voltage distribution and Monte Carlo techniques are applied to the EOC voltage distribution to obtain the SLB leak rates and tube burst probabilities. To demonstrate compliance of the Westinghouse analysis methodology with the NRC requirements of the draft generic letter and to simplify NRC review, Westinghouse has prepared a methods report (Reference 11.5). This report provides the relevant information on analysis methods for - voltage projections, leak rates, and burst probability. l i Of note to this analysis is that, for the tube burst probability analyses, the editing elements of the Monte Carlo analyses were revised to separate single tube and multiple tube ruptures. Monte Carlo analyses previously reported by Westinghouse, including the Braidwood-l IPC evaluation, have computed the total tube burst probability which is the sum of single and multiple ruptures (i.e., the sum of 1,2,3, etc. ruptures). The revised editing separately computes the probability of one, two, three or more tube ruptures in addition to the total burst probability. The Monte Carlo process involves sampling all indications in the SG to obtain the number of burst pressure samples with burst pressures less than the SLB value of 2560 psid. In a given SG sample, the number of occurrences of burst pressures less than the SLB value might be one, two, three or a fraction. When fractions of bursts (0.7,1.3) less than the SLB value occur in a SG sample, such as resulting from applications of a POD, the whole number of the fraction results in r nccurrence of the whole number of bursts and the fraction is added to the burst probability ( f sne next largest integer number of bursts. For example, an occurrence of 0.7 results in adding 0.7 occurrence to the one tube rupture frequency and an occurrence of 1.3 adds one occurrence to the one tube rupture frequency and 0.3 occurrence to the two tube rupture frequency. The SG sampling process is then repeated N times. Without adjustments for confidence level, the single and multiple tube burst probabilities are the number of occurrences (frequency) of one, two, etc. pressures less than the SLB value, as i summed over all SG samples, divided by N. The burst probabilities reported herein are 95% confidence values which include adjustments for the number N of SG samples. i t i t r I

hpcVae94Wyc6ipchec4678.wp3 6.}

_= 7.0 ALTERNATE POD DISTRI!UTIONS The draft NRC generic letter requires the application of POD = 0.6 to define the BOC y, T' distribution for the EOC voltage projections unless an alternate POD is approved by the NRC.- To provide sensitivity analyses in support of obtaining approval of an altemate POD'in the near future, potentially as part of resolution ofindustry comments on the draft generic letter, alternate PODS are included in the SLB leak and burst analyses. To bound the influence of POD on the analyses, an ideal POD = 1.0 (assumes all indications are detected) is included in the analyses. Industry recommendations on a voltage-dependent POD will be provided as part of the comment resolutions for the draft generic letter. For the sensitivity analyses of this report, two available variations of a voltage-dependent POD are applied. Figure 7-1 shows the first. These voltage-dependent POD results were i originally obtained as part of the EPRI Qualified Data Analyst testing program, and have previously been presented to the NRC. To develop the EPRI data of Figure 7-1, a panel of experts reviewed field and pulled tube eddy current data to identify ODSCC indications. The analysts being tested were evaluated based on the fraction ofindications reported versus the panel of experts' flaw identification. Figure 7-1 includes an added, judgemental linearly segmented fit to the data (identified as "EPRI Fit") which has been used as PODi(EPRI) in the sensitivity analyses of this report. l An additional voltage-dependent POD is also presented in Figure 7-1. Although not used in the present analysis, this POD vs. voltage correlation (also approximated and linearized) was recently developed by Commonwealth Edison, Duke Power, and EPRI for Byron, Braidwood, and Catawba 3/4" tubing. Data were selected only from the above units and, as much as possible, indications were established through RPC examination. All indications identified i during the program by any of the twelve analysts were examined by a peer review group to determine if a call was " valid" The database consisted of 818 tubes: 50 of these had no detectable degradation, and the remainder had a total of 872 RPC confirmed indications. 1 Added to the database were 222 indications identified in the field but NDD by RPC. Also added were 251 indications identified during the POD program but not called in the field. Figure 7-1 indicates that PODI and consequently POD 2 are both more conservative relative to the Comed / Duke /EPRI values, particularly at higher voltage levels. Low detectability values (less than 0.6) were found in the Comed / Duke /EPRI study for voltages less than 0.5 volt. Additional information on this study is expected to be presented during review of draft generic letter comments. Figure 7-2 shows POD data obtained as part ofIPC evaluations for four plants which had implemented an IPC in the prior outage. These data are based on defining POD as the fraction of RPC confirmed indications found in the latest inspection that were reported as bobbin indications in the prior inspection. To account for tube plugging in this definition, indications RPC confirmed and plugged in the prior inspection are considered as also confirmed in the latest inspection in applying the POD definition. This definition of a POD has been labeled probability of prior cycle detection (POPCD). This can also be shown as: 4cwee94scyc6:pcWc4678 wp$ 7-1 ) w y vu-- -T m wwy w u,iw

Prior Cycle Reported Ind. Ccafirmed by RPC: Confirmed in POPCD = Latest Inspection Plus Confirmed and Repaired in Prior Inspection Numerator Plus New Ind. RPC Confirmed in Latest Inspection The logic for defining POPCD is that the EOC indications that are important to SLB leakage and burst probability are the RPC confirmed indications. RPC NDF indications, if a real indication exists, are too small to contribute to leakage or burst since RPC detectability data show that essentially all indications greater than about 65% depth are confirmed by RPC inspection as flaw indications. Indications not RPC inspected are <l.0 volt and would not contribute significantly to leakage or burst. Thus RPC confirmed indications are an appropriate basis for defining a POD for IPC applications. For an IPC, an appropriate POD is then the fraction of EOC RPC indications that were detected at the prior inspection. Since some prior cycle indications were RPC confirmed and plugged, it can be assumed that these indications would have been confirmed at EOC if not plugged and it is appropriate to include the prior inspection RPC confirmed and repaired indications in the definition of POPCD. This POD definition is based on field inspection results that influence IPC analyses and supplements the EPRI POD evaluations based on peer review of field data. The individual plant data used for the averages of Figure 7-2 were reported at an EPRI NDE meeting in July,1994. To provide a voltage-dependent POD that differs significantly from the EPRI PODI, a simple linear, lower bound fit to the data was applied as POD 2 (POPCD). The resulting four PODS used in the analyses of this report are shown in Figure 7-3. i l s:\\npc\\cee94cyc6ipchec4678 wp5 7-2

Figure 7-1 Comparison of ODSCC Detection: EPRI Data and Fit at 90% Conf. Level vs. Comed / Duke /EPRI Data Fit (approximatef._) at 95% Conf. Level p _ __. 0.9 -- ,8 0.8 / m Fraction Detected (or POD) for EPRI ^o 0.7 Data 2 'EPRI Data at 90% Conf i o 0.6 [ E / EPRI Fit e f / "

• Fit (approximated) of Comed / Duke /EPRI Data Lower 95% Conf.

g 0.4 $20.3 0.2 - 0.1 -- 0-I 1 I l <.25 <5 0.5 .5<1 1 1<2 2 2<3 3 3<4 4 4<5 Bobbin Voltage (V) c:W:,..,~'. wa* Fig 7-1 Chart 1

i jl l l! ll fl 00 f 3 0 I 5 2 D O P gn idn 0 uo

0 2

B d e z irae n e 0 g 2i L I 5 a 4 7 t 2 l s 1 o e v V rus gt n iF a l P 4 r 0 o f l 0 1 DCP O P D DO e CP g P g a 0 r Oin e P d 5 n 0 v . u A g v o AB m: 0 0 0 0 0 0 0 0 0 0 0 0 0 0 5 5 0 5 0 5 0 5 0 0 9 8 8 7 7 6 6 5 5 1 c@ E $ o #._2 f E 5 l l' l l ,\\ ll

Figure 7-3 Comparison of Probability of Detection Models 100.0 % = = = = = = = = = = y- = = = 5 90.0 % E j -*- PO D=0.6 o 80.0 % -*- POD 1 (EPRI) p -*-POD 2 (POPCD) ] -*- POD =1.0 .o [ 70.0% 60.0 % 50.0 % O.00 0.25 0.50 0.75 1.00 1.25 1.50 1.75 2.00 2.25 2.50 2.75 3.00 3.25 3.50 3.75 4.00 Voltage cWtyronapc\\

• Fig 7-3 4 PODS Chart 1 7-5

t i - 8.0 PROJECTED CYCLE 7 VOLTAGE DISTRIBUTIONS f Consistent with the draft generic letter (Ref.11.2) methodology, Monte Carlo analyses were i applied to develop projected EOC distributions from the BOC distributions. The BOC i voltages are increased by allowances for NDE uncertainties (from Section 4.3) and voltage growth (from Section 4.2) to obtain the EOC values. In the Monte Carlo analyses, each voltage bin of the BOC voltage distributions is increased by a random sample of the NDE uncertainty and growth distributions to obtain an EOC voltage sample. Each sample is i weighted by the number ofindications in the bin. The sampling process is repeated for each BOC voltage bin and then repeated for a large number of samples across the BOC distribution. 8.1 Linniting SG Evaluation Of the four Byron-1 SGs, SG-A and SG-C were judged to be more limiting for leak rate and turst probability analysis, because of their larger amplitude (> 4.0V) indications. To determine the more limiting of these SGs, leak rates and burst probabilities were determined for the actual EOC-6 and projected EOC-7 (using POD =0.6) voltage distributions of both SGs. Cycle lengths were based on the reported Cycle 6 length of 466.5 EFPD, a projected mid-cycle 7 length of 298.5 EFPD, and a projected EOC-7 of 474.5 days. Standard Monte Carlo methods to account for NDE uncertainty were used along with the voltage growth distributions and a POD =0.6 to develop projected EOC-7 voltage distributions. The SG-A and C growth distributions were each applied for their respective analyses. Use of the individual SG growth rates in the EOC voltage projections maintains the correct frequency of large voltage indications in the tail of the cumulative probability distribution, which is important for the SLB leak rate and burst probability analyses. Each Byron-1 SG has a sufficiently large number (>200 indications) to adequately define a SG-dependent growth rate. Combining SGs to obtain the cumulative growth distributions could result in non-conservative or overly conservative distributions depending on the SGs selected for the combination. The results for SG-C are presented in sections 8.3 and 8.4, below. The SG-A BOC-7 distribution used to determine the EOC-7 distribution, and the EOC-7 distribution determined using the Table 4-7 voltage growths are shown in Table 8-1 and plotted in Figure 8-1, "Ihe SG-A EOC-6 leak rate of 2.0 gpm (based on the actual EOC-6 voltage distribution) and projected EOC-7 leak rate of 4.6 gpm were both lower than the SG-C values of 2.3 gpm and 5.1 gpm, respectively. Likewise, the SG-A EOC-6 burst probability t f 1.26 x 10 and projected EOC-7 burst probability of 2.94 x 10'were also lower than the SG-C values of 1.90 x 10 2 and 3.29 x 10, respectively. SG-C was therefore determined to be the limiting SG 4 for leak and burst considerations, and was the focus of further analysis. 8.2 Input Distributions 4 Tables 4-1 and 4-4 provided the final EOC-6 voltage distribution and the repaired indication distribution for SG-C, defining the actual BOC-7 distribution. Each of these distributions was developed according to the descriptions provided in section 7 of altemate POD methods, and i is shown in Table 8-2 and Figure 8-2. Comparing the total number of indications for each 4 snapckae94Wyc6ipchec4678.wp3 8-l

BOC distribution in Table 8-2, the POD of 0.6 provides about twice as many indications as the POD of 1.0, and includes 0.7 BOC indications for each repaired indication above 3.0 volt. PODI (EPRI) and POD 2 (POPCD) provide intermediate BOC distributions to the 0.6 and 1.0 PODS, with POD 2 being slightly more conservative in the total number of indications and largest indications. 8.3 Projected MOC-7 Voltage Distribution i The projected mid-cycle 7 (MOC-7) bobbin voltage distributions are shown in Table 8-3 and Figure 8-3 for each of the PODS applied to the SG-C BOC-7 voltages. The Reference 11.3 method for determining the largest EOC voltage with Monte Carlo analysis was used. The maximum projected EOC-7 voltage is therefore a fractional 0.3 indications, for the POD = 0.6 adjusted distribution, at 11.5 volts. De largest MOC-7 voltage for the distribution of all indications without POD adjustment (POD = 1.0) is 6.5 volts. The more significant factor in determining the largest projected voltage for PODI, POD 2, and POD =1.0, which all employ a POD =1.0 above 3.0 volts, is the growth distribution. Herefore, the maximum voltages for .each of these three PODS are relatively close at 6.7,6.8, and 6.5 volts, respectively. The total number of MOC indications for each POD applied in Table 8-3 is essentially the same as the number of total indications in the BOC distribution of Table 8-2, with slight variation due to roundoff. 8.4 Projected EOC-7 Voltage Distribution j The projected end-of-cycle 7 (EOC-7) bobbin voltage distributions are shown in Table 8-4 ar,d Figure 8-4 for each of the PODS applied to the SG-C BOC-7 v.-!tages. The maximum projected EOC-7 voltage (a fractional 0.3 indications) for the POD = 0.6 adjusted distribution is 11.7 volts, higher by 0.2 volts than the MOC-7 projection. The proximity of these values is due to the presence of the assumed 0.7 indications of 10.95 volts in the tail of the POD =0.6 BOC distribution. Application of the growth distribution for the full cycle as compared to a partial cycle increases the number, but not the magnitude, of the largest indications for POD =0.6. He largest EOC-7 voltage for the distribution of all indications without POD adjustment (POD = 1.0) is 9.8 volts. Again, for PODI, POD 2, and POD =1.0, the maximum voltages are relatively close at 10.2,10.2, and 9.8 volts, respectively. Figures 8-5 and 8-6 shown the EOC-7 and MOC-7 projected distributions for POD =0.6 versus the BOC-7 distribution, respectively. sMpcVas94kyc6ipchec4678.wp5 8-2 .~.

. Table 8-1 Byron 1 SG-A EOC-7 Pr:jected Vekages (Limiti:3 G An'.Jysis) S 94 Volts EOC-6 EOC-6 BOC EOC-7 Indications Repaired POD =0.6 Projected 0.2 4 0 6.7 0.0 0.3 57 6 89.0 5.0 04 105 16 159.0 18.0 0.5 105 10 165.0 - 41.0 0.6 - 106 15 161 7 65.0 0.7 82 15 121.7 86.0 0.8 - 73 13 108.7 101.0 0.9 66 18 92.0 105.0 1.0 47 11 67.3 103.0 1.1 34 31 25.7 98.0 1.2 31 25 26.7 88.0 l.3 28 23 23.7 76.0 l 1.4 18 17 13.0 64.0' 1.5 6 4 6.0 $2.0 1.6 6 5 5.0 40.0 1.7 4 4 2.7 32.0 1.8 4 4 2.7 24.0 1.9 4 4 2.7 18.0 f .2.0 5 5 3.3 14.0 2.1 2 2 1.3 10.0 i 2.2 3 3 2.0 9.0 2.3 1 1 0.7 7.0 2.4 2 2 1.3 6.0 - 2.5 1 1 0.7 5.0 2.6 1 1 0.7 4.0 2.7 0 0 0.0 3.0 l 2.8 0 0 0.0 2.0 2.9 0 0 0.0 3.0 l 3.0 0 0 0.0 1.0 l 3.1 2 2 1.3 1.0 i 3.2 0 0 0.0 1.0 3.3 0 0 0.0 1.0 3.4 1 1 0.7 1.0 3.6 0 0 0.0 1.0 3.8 0 0 0.0 1.0 4.0 1 1 0.7 0.0-4.2 0 0 0.0 1.0 5.0 0 0 00 1.0 60 1 1 0.7 0.0 6.1 0 0 0.0 1.0 6.4 0 0 0.0 1.0 f 6.7 0 0 0.0 1.0 7.0 0 0 0.0 1.0 7.1 1 1 0.7 0.0 7.5 0 0 0.0 1.0 1 7.7 1 1 0.7 0.0 { 8.1 0 _ 0 0.0 0.7 8.7 0 0 0.0 0.3 Total 802 243 1093.7 1094.0 Total > 1V 157 139 122.7 570.0 [ 8-3 c:\\apc\\byronapcVeport\\CAE94RPT.XLS T,F 8-1 j

.._ - -.=, t Table 8-2 BOC-7 Voltage Distributions for Byron-1 SG-C. BOC Disuibutions No. of No. of 94 Volts Bobbin Indicauons PODI POD 2 i Indications Repaired POD =0.6 (EPRI) (POPCD) POD =1.0 0.3 13 0-21.7 19.7 20.3 13.0 0.4 45 5 70.0 61.2 63.9 40.0 0.5 92 6 147.3 125.4 132.0 86.0 0.6 95 7 151.3 123.1 132.7 88.0 0.7 130 25 191.7 146.1 162.5 105.0 0.8 116 21 172.3 125.8 143.2 95.0 0.9 89 9 139.3 99.5 114.6 80.0 3.0 92 17 136.3 91.2 108.5 75.0 1.1 59 47 51.3 21.6 32.0 12.0 1.2 40 30 36.7 16.0 22.6 10.0 1.3 33 27 28.0 10.5 15.7 6.0 1.4 26 24 19.3 5.2 9.1 2.0 1.5 17 16 12.3 2.9 5.3 1.0 1.6 10 9 7.7 2.0 3.3 1.0 1.7 13 12 9.7 2.1 3.7 1.0 1.8 8 8 5.3 0.6 1.5 0.0 1.9 4 4 2.7 0.3 0.7 0.0 2.0 4 4 2.7 0.2 0.6 0.0 2.1 2 1 2.3 1.1 1.3 1.0 2.2 1 1 0.7 0.0 0.1 0.0 2.3 1 1 0.7 0.0 0.1 0.0 I 2.4 1 1 0.7 0.0 0.1 0.0 2.5 2 2 1.3-0.1 0.1 0.0 i 2.6 1 1 0.7 0.0 0.1 0.0 2.7 2 2 1.3 0.0 0.1 0.0 3.64 1 1 0.7 0.0 0.0 0.0 3.80 1 1 0.7 0.0 0.0 0.0 10.95 1 1 0.7 0.0 0.0 0.0 Total 899 283 1215.3 854.8 973.9 616.0 Total > IV 227 193 185.3 62.7 %.3 34.0 i 1 ) c:5WVeportCAE94RPT.XLS T,F 8-2 l

Table 8-3 Projected MOC-7 Voltage Distributions for Byron-1 SG-C POD Applied '94 Volts POD =0.6 POD 1 (EPRI) POD 2 (POPCD) POD = 1.0 0.3 2.0 2.0 2.0 1.0 0.4 14.0 12.0 12.0 8.0 0.5 38.0 33.0 35.0 22.0 0.6 68.0 57.0 61.0 40.0 0.7 101.0 83.0 89.0 58.0 0.8 124.0 98.0 107.0 70.0 0.9 140.0 109.0 120.0 79.0 1.0 140.0 104.0 117.0 78.0 1.1 127.0 91.0 104.0 68.0 1.2 110.0 76.0 89.0 57.0 1.3 88.0 57.0 68.0 43.0 1.4 67.0 42.0 50.0 30.0 1.5 50.0 28.0 35.0 21.0 1.6 36.0 19.0 25.0 13.0 1.7 27.0 13.0 17.0 8.0 1.8 19.0 8.0 12.0 6.0 1.9 15.0 6.0 8.0 4.0 2.0 11.0 5.0 6.0 2.0 2.1 8.0 3.0 4.0 2.0 2.2 6.0 2.0 3.0 1.0 2.3 5.0 1.0 3.0 1.0 2.4 3.0 2.0 2.0 1.0 2.5 3.0 1.0 1.0 1.0 2.6 2.0 0.0 1.0 0.0 2.7 2.0 1.0 1.0 00 2.8 2.0 0.0 1.0 1.0 2.9 1.0 1.0 1.0 0.0 3.0 1.0 0.0 0.0 0.0 3.2 1.0 0.0 1.0 0.0 3.6 1.0 0.0 0.0 0.0 4.3 1.0 0.0 0.0 0.0 5.5 0.0 1.0 0.0 0.0 5.7 0.0 0.0 1.0 0.7 5.8 1.0 0.0 0.0 0.0 6.1 0.0 0.7 0.0 0.0 6.3 0.0 0.0 0.7 0.0 6.5 0.0 0.0 0.0 0.3 6.7 0.0 0.3 0.0 0.0 6.8 0.0 0.0 0.3 0.0 9.4 0.7 0.0 0.0 0.0 11.5 0.3 0.0 0.0 0.0 Total 1215.0 856.0 977.0 6 h6.0 Total > IV 588.0 358 0 434.0 260.0 c.Wpetsporupevoport'CAE94RPT.XLS T,F 8 3

Tcble 8-4 Projected EOC-7 Voltate Distribrtims frr Pyro,-l SG-C POD Applied '94 Volts POD =0 6 PODI (EPRI) POD 2 (POPCD) POD = 1.0 0.3 2.0 2.0 2.0 1.0 0.4 11.0 10.0 10.0 7.0 0.5 36.0 26.0 28.0 17.0 0.6 53.0 44.0 47.0 31.0 'O.7 74.0 60.0 64.0 42.0 0.8 95.0 76.0 83.0 54.0 0.9 108.0 84.0 92.0 61.0 1.0 119.0 90.0 101.0 66.0 1.1 115.0 85.0 95.0 63.0 1.2 105.0 75.0 86.0 55.0 1.3 %.0 66.0 77.0 50.0 1.4 84.0 57.0 66.0 41.0 1.5 67.0 43.0 52.0 32.0 1.6 55.0 34.0 41.0 24.0 1.7 44.0 26.0 32.0 19.0 1.8 34.0 19.0 23.0 13.0 1.9 26.0 13.0 18.0 10.0 2.0 20.0 11.0 13.0 7.0 2.I 15.0 7.0 9.0 5.0 2.2 12.0 5.0 8.0 3.0 2.3 9.0 5.0 5.0 3.0 2.4 7.0 3.0 4.0 2.0 2.5 6.0 2.0 4.0 2.0 i 2.6 5.0 2.0 3.0 1.0 2.7 3.0 2.0 2.0 1.0 2.8 3.0 1.0 2.0 1.0 2.9 3.0 2.0 1.0 1.0 3.0 2.0 0.0 1.0 0.0 3.1 1.0 1.0 1.0 1.0 3.2 2.0 0.0 1.0 0.0 3.3 1.0 1.0 l.0 0.0 3.4 1.0 0.0 0.0 1.0 3.5 0.0 1.0 1.0 0.0 3.6 1.0 0.0 0.0 0.0 3.7 1.0 0.0 1.0 0.0 3.8 0.0 1.0 0.0 1.0 3.9 1.0 0.0 0.0 0.0 4.0 0.0 0.0 1.0 0.0 4.1 1.0 0.0 0.0 0.0 4.6 1.0 0.0 00 0.0 '.0 8.3 0.0 1.0 0.0 0 8.6 0.0 0.0 1.0 0.7 8.7 1.0 0.0 0.0 0.0 9.3 0.0 0.7 0.0 0.0 9.5 0.0 0.0 0.7 0.0 9.8 0.0 0.0 0.0 0.3 10.2 0.0 0.3 0.3 0.0 10.7 0.7 0.0 0.0 0.0 11.7 0.3 0.0 0.0 0.0 Total 1215.0 856.0 977.0 616.0 Total > 1V 723.0 464.0 550.0 337.0 c:%pce,yTw.epcVeportCAE94RPT.XLS T,F IL4 g.6

t l 7 3 t s 7 7 1 7 78 _ 0 7 7 6 4 4 l e s _ e A- _ e S G 2 S 4 1 e d n r o 8 3 y 6 3 B 4 r 3 o f 3 3 n o 2 3 i t u 1 b 3 i 8 r 3 t s 6 d a 9 0 e 2 i t D = ce 8 Dj 2 ) eg O o V r m 7 ( a P P 2 t n 7 7 s_ B 6 i l o CC 2 V OO E_ g 3 e 7 2 7 B E a 8 C OE E. 4 t l 2 oV O l. 3 2 E 2 d 2 e l I tc 0 e 2 jo 9 r 1 P S d r. I n 7 r 1 a 6 7 J 1 -C 5 I 4 O 4 B I 3 1 1 2 8 1 e u 8 r 1 0 g F l 1 i 9 0 2 8 t 0 ar h 7 C 0 1 6 8 0 F. 3 T 0

• \\

4 tr 0 o k p 3 e 0 Vc 2 p 0 0 ono 0 0 0 0 0 o 0 o o o r 8 1 1 1 1 6 4 2 0 s 6 4 2 ye 1 E j v._E Co $ v c: I Il l l lj I ll{

Figure 8-2. SG-C BOC-7 Voltage Distributions for Various PODS 200 180 160 I40 gi20 E POD =0.6 8 OPODI (EPRI) 3 E POD 2 (POPCD) 'jioo OPOD=1.0 g Z 30 60 i 4o l 20 - . L n_ n_.=__._ 0 03 04 OS OG 07 60 09 10 18 12 13 14 15 le 37 10 19 20 21 22 23 24 23 26 27 3M 3M 19 95 Voltage Bin (V) 8-8 cDpetyiwM.w

• Tf 8-2 Chart 1 a

i Figure 8-3. SG-C MOC-7 Projected Voltage Distributions for Various PODS !40 120 300 30 5) E POD =0.6 o j' ~ ~ OPODI (EPRI) E POD 2 (POPCD) E POD =1.0 40 l 20 i ll lIlkkhuu_..._.......___ 03 04 05 06 07 08 09 te 11 12 13 l.4 15 16 1.7 13 19 20 21 22 23 24 13 26 27 28 29 30 32 36 43 $5 $7 $8 el 43 63 47 68 94 13 1 Voltage Bin (V) c:ww..g.%n* T.F 8-3 Chart 1 b~

Figure 8-4. SG-C EOC-7 Projected Voltag-Distributions for Various PODS 12o 100 t i no E POD =0.6 O PODI (EPRI) .[o E POD 2 (POPCD) .E g E POD =1.0 1a 5 I i 20 I I mm. k h k h. m. _ m. _.._,.. _....... _. i 8.3 04 es to 0.7 08 09 le It 12 13 14 13 I6 17 is 89 20 28 22 13 24 13 26 27 28 29 30 3 3.2 33 34 35 36 37 38 39 40 4 46 83 84 87 93 95 98 set 107 tt ? Voltage Bin (V) c:5%.,~'..,, L* T.F 8 4 Chart 1 8-10

Figure 8-5: Byron-1, SG-C Projected EOC-7 Voltage Distribution vs. BOC-7 Distribution Adjusted by pod =0.6 200 180 160 140 E -j 120 0BOC-7 Adjusted by POD =0.6 E 80 - --ll 100 5 Projected EOC-7 i z 60 -- 40 -- p"j'ggj. 7-llli jm 0 83 94 el el 07 et et tt 1.1 12 1.3 14 15 14 1.7 18 19 29 11 2.2 13 to 25 20 27 ft 29 30 31 32 33 34 30 37 30 39 49 48 0F te F tt e 117 Bobbin Amplitude (Volts) c:wsryron pev r _.ng.xi. rig s-s chart i 8-11

Figure 8-6: Byron-1, SG-C Projected MOC-7 Voltage Distribution vs. BOC-7 Distribution Adjusted by pod =0.6 200 + 180 160 140 O BOC-7 Adjusted by POD =0.6 EProjected MOC-7 E 3 120 .E E C 100 E i E 80 Z 60 l 40 l 20 $511_a n_. n es es se er se se se o u n u is is it is se n ai n n n as n ar n se n n n ir n.: se es sie ii s Bobbin Amplitude (Volts) I c%hionepc"..,~,1W.m Fig 84 Chert 1

9.0 SLB LEAK RATE ANALYSIS RESULTS 9.1 SLB Leak Rates for EOC-6 Actual and BOC-7 Distribution A SLB leak rate was calculated for the actual SG-C EOC-6 voltage distribution of 899 indications (i.e., regardless of RPC confirmation) shown in Table 4-1. The largest indications in this distribution were 10.95,3.80, and 3.64 volts; remaining voltages were less than 2.7 volts. The results of the SLB leak rate analyses are given in Table 9-1, SLB leak rates were determined by averaging the results of Monte Carlo simulations for runs of 50,000 samples; 20 runs in all gave results for a total of one million simulations. The leak rate of 2.3 gpm is a 95% cumulative probability value. The allowable leak rate for the faulted SG, as discussed in the Byron-1 SER (Ref. I1.1), is 12.5 gpm, so significant margin was present at EOC-6. The BOC-7 SLB leak rates were also calculated for each of the four POD-dependent SG-C voltage distributions described in Table 8-2. The respective SLB leak rates and number of Monte Carlo simulations are given in Table 9-1. A factor of 60 difference is observed between a POD of 0.6 compared to POD =1.0. This large difference is due to the large fractional numbers of indications in the upper voltage range for the POD =0.6 distribution. The leak rates for PODI and POD 2 are, as expected, both much closer to the POD =1.0 estimate than the 0.6 estimate, since they both employ a POD of 1.0 above 3.0 volts. 9.2 Projected SLB Leak Rates for Potential Mid-Cycle Voltage Distribution The mid-cycle 7 SLB leak rates were calculated for each of the four projected SG-C MOC-7 voltage distributions described in Table 8-3. These SLB leak rates are given in Table 9-1. A factor of 5 difference is observed between a POD of 0.6 compared to POD =1.0, since a number of larger voltage indications have entered the MOC-7 voltage distribution. The PODI and POD 2 distributions again provide leak rates intermediate to the POD =0.6 and =1.0 distributions, with POD 2 being slightly more conservative than PODl. 9.3 Projected SLB Leak Rates for EOC-7 Voltage Distribution The projected EOC-7 leak rate for SG-C with a POD = 0.6 (NRC draft generic letter) is 5.1 gpm. This leak rate is less than the allowable SLB leak rate for Byron-1 and thus is acceptable to the guidelines of the NRC draft generic letter. Table 9-1 also includes the calculated SLB leak rate for the projected EOC-7 distribution based on the alternate POD distributions described in Section 7. It is seen that the more realistic, voltage-dependent PODI and POD 2 analyses result in SLB leak rates of 3.1 and 3.5 gpm which approach the ideal POD = 1.0 result of 1.7 gpm. This results as all of the largest 29 bobbin indications (29 >l.8 volts including 3 above 3.6 volts) were plugged at EOC-6. The POD = 0.6 results in 0.67 indications left in service for each plugged indication while the voltage-dependent PODS result in smaller fractional indications left in service below 3.0 volts and no indications left in service above 3.0 volts. %*=myc6%.ps 94 1

For Byron-1, the influence of POD on the SLB leak rates is significantly offset by the larger growth rate effect for the EOC-7 projections. With the larger growth rates, the EOC distribution includes comparable maximum voltages with or without a POD correction although the higher PODS result in fewer high voltage indications. The larger difference between POD = 0.6 and 1.0 is more apparent in the BOC-7 analyses in Table 9-1 with leak rates of 1.8 and 0.03, respectively. 5ku94Wyc6ipcWc9mp5 9-2

e Table 9-1 Byron.1, S/G C: Summary of SLB Monte Carlo Analysis Results") Tube Burst Probability Analyses SLB Analysis Basis Leak Rate No. Monte Total Single Two Three (gpm) Carlo Burst Tube Tube Tube S/G Prob. Burst Burst Burst Samples Prob. Prob. Prob. EOC-6 Actual Voltage Distribution l 2.3 1000K 1.90x10 1.90x10'8 4.74x10 3.00x10*2) 4 Actual Volts BOC-70) POD = 0.6 1.8 500K 1.36x10 l.36x10 3.05x10'$ 8.39x10*2) 2 PODI(EPRI) 0.04 500K 2.19x10" 2.19x10" 5.99x10*2) 5.99x10+2> d POD 2(POPCD) 0.06 500K 3.28x10" 3.27x10 7.19x10*2) 5.99x10*2) POD = 1.0 0.0'; 500K 1.53x10" 1.53x10" 5.99x10*2) 5.99x10*2) Projected Mid-Cycle 7 2 d POD = 0.6 2.5 1000K 1.92x10 1.91x10~2 1.66x10" 7.84x10 PODi(EPRI) 1.0 200K 5.18 x10 5.17x102 1.95x10'$ 1.50x10*2) POD 2(POPCD) 1.2 750K 5.91x10'$ 5.91x10'$ 5.19x10 3.99x10*2) 4 POD = 1.0 0.52 500K 3.36x10'$ 3.26x10'$ 1.20x10" 5.99x10*2) Projected EOC-7 Voltage Distribution POD = 0.6 5.1 1000K 3.29x10'2 3.15x10 l.46x10'$ 3.00x10+2) PODi(EPRI) 3.1 1000K 2.29x 10 2.28x10'2 1.39x10" 3.00x10*2) POD 2(POPCD) 3.5 350K 2.53x10'8 2.49x10 4.86x10" 8.56x10+2) POD = 1.0 1.7 400K 1.22x10 2 1.21x 10 l.1lx10" 7.49x1042> Notes

1. Monte Carlo analyses at 95% confidence based on NRC database.

2. No occurrences cf multiple tube bursts were found in Monte Carlo analyses. Burst probabilities given represent upper 95% confidence based on no occurrences in stated number of Monte Carlo samples.

3. These calculations were performed with a later version of the Monte Carlo code. The code differences do not significantly affect EOC calculations.

.:vpesese9wyc6ipes c9.wps 93 i

10.0 SLB TUCE BURST PROBABILITY ANALYSIS RESULTS Limited TSP displacement burst probabilities are more representative of expected Byron-1 conditions than the free span burst probability analysis required by the draft generic letter. The limited TSP displacement burst probabilities of Reference 11.7 are summarized in Section 10.1 to represent the expected Byron 1 tube burst probabilities. However, the NRC draft generic letter and the Byron-1 SER require SLB. burst probabilities to be conservatively calculated as free span indications. These required, more conservative burst probabilities are given in Sections 10.2 to 10.5. 10.1 SLB Burst Probability Based on Byron-1 Limited TSP Displacements The study performed in Ref.11.7 provides the basis for burst probability calculations for Byron-1 based on the limited tubefrSP displacement during SLB. A bounding estimate of the probability of burst of one or more tubes in a SG during a postulated SLB was made by assuming every tube to have a throughwall crack equal to the thickness of the TSP at every TSP intersection. Thus, during a postulated SLB, each intersection was very conservatively postulated to have a throughwall crack exposed by the magnitude of the displaceraent at that intersection. The probability of burst of one or more tubes is then calculated as the sum of the probabilities of burst of each indication. The analysis was performed for the actual EOC-6 TSP indications at Byron 1, of which SG-C was the more limiting. The results are shown in Table 10-1. None of the large voltage ( > 2.0 volts in this case) indications occurred at tube locations with significant SLB TSP i displacements. The indications at R5CI (1.19 volt), RIC3 (1.74 volt), R3Cl (0.94 volt) and R3Cll4 (0.51 volt) occur at tube locations with significant TSP displacements of 0.492 to j 0.522 inch and dominate the mntribution to the limited TSP displacement burst probability of i 4 1.1x10. For the indications given in Table 10-1, the free span burst probability is 3.0x10 2 The results of Table 10-1 show the effectiveness of limited TSP displacements in reducing the tube burst probability to small values and also show that Byron-1 had an acceptably low burst probability at EOC-6. The application of the actual indication distributions reduces the burst 4 4 probability by more than two orders of magnitude (from 2.9x10 to 1.1x10 ) compared to the bounding assumption of throughwall indications at all hot leg TSP intersections. The benefits of limited TSP displacement are also seen by the reduction of about three orders of magnitude 2 4 (3.0x10 to 1.1x10 ) in the tube burst probabilities compared to the free span IPC analyses. 10.2 Free Span SLB Burst ProbabHity for EOC-6 Actual Distribution Calculations were performed to determine the free span probability of tube burst using the actual EOC-6 voltage distribution for SG-C, without modifications for repairs or POD considerations. One million Monte Carlo simulations were performed, applying the actual voltage distribution to the leak rate vs. volts correlation with NDE and voltage uncertainties. 2 The calculated total burst probability was 1.90x10, which included a negligible contribution for two tube ruptures and no three-tube ruptures. This exceeds the threshold value of 1.0x10'2 established in the SER (Ref.11.1 Sec 3.0 Item g.4), wUch requires an assessment of the .wwwsycopeweiomps 10 1

safety signficance of this condition. Assessing this condition, it is noted that the burst robability is dominated by the single large indication of 10.95 volts at R20C7 3H in SG-C. This indication is located fully within the tube suppo;t plate thickness, and is in a location of limited (less than 0.15") displacement under SLB conditions, with uncertainty factors included. Thus the larger voltage indications found at EOC-6 would not burst due to the TSP constraint. Moreover, the 10.95 volt indication was pulled and burst tested in the laboratory. The measured burst pressure of 4500 psi is higher than the R.G.1.121 structural margin requirement of 1.43APg = 3660 psi. Thus this indication would not have a significant contribution to the burst probability at EOC-

6. Similarly, all other Byron-1 pulled tubes (Section 3) had high burst pressures. Thus the Byron-1 EOC-6 burst probability, as based on pulled tube test results, would be much less than 10-2 and the calet'2ted value of 1.90 x 10 has negligible safety significance. Since both the Byron-1 and Braidwood-l (including 5.0 and 10.3 volt indications) pulled tubes had burst pressures well above R.G.1.121 burst margin guidelines, it can be expected that EOC-7 burst probabilities would also be significantly lower than obtained from the free span burst analyses. Thus both the pulled tube measurements and the limited TSP displacement support low burst probabilities, and significantly lower than the free span analyses.

10.3 Projected Free Span SLB Burst Probability for Potential Mid-Cycle Voltage Distribution For POD = 0.6, the tube burst probability is not significantly affected by the potential mid-cycle inspection since there is only a factor of two difference between the BOC-7 and EOC-7 results. For POD = 1.0, the mid-cycle burst probability of 3.38x10 is less than the NRC draft generic letter guideline of 1.0x10'2 for reporting to the NRC and is almost a factor of four less than the EOC-7 value of 1.22x10 2 Thus high values for POD at the EOC-6 Byron-1 inspection would result in mid cycle burst probabilities less than the NRC guideline. This is shown for the PODI and POD 2 results of Table 10-1. 10.4 Projected Free Span SLB Burst Probability for EOC-7 Distribution Results of the tube burst probability analyses are also given in Table 9-1. The total (sum of single and multiple ttbe bursts) tube burst probabilities for the assumed POD = 0.6 are 1.90 x 2 10 at EOC-6 and 3.29 x 10-2 for the projected EOC-7 distribution. The total and. single tube burst probabilities are approximately equal since the probabilities of multiple tube ruptures are very small. No occurrences of multiple tube ruptures with more than two ruptures were found in the Monte Carlo analyses and the values reported in Table 9-1 for three tube ruptures - reflect the 95% confidence limit for the number of Monte Carlo samples with no occurrences. A few occurrences of multiple ruptures (two) occurred but the burst probabilities are significantly lower than the smgle tube burst probability, 10.5 Burst Probability Dependence on POD At BOC-7, it is seen that the difference in the burst probability between PODS of 0.6 and 1.0 (1.51x10 versus 3.10x10") is a factor of 50 increase for the very low POD of 0.6. This s:\\apc\\cee94\\cyc6ipc\\aecl0 wp3 10-2

P difference is principally due to the large voltage indications left in service for POD = 0.6 (0.67 indication left in service for each indication repaired). The factor of 50 difference is reduced to about a factor of 7 at mid-cycle 7 and a factor about 2.7 at EOC-7. The reduced r dependence on POD with increasing cycle length for Byron 1 is attributable to the larger growth rates included in the growth distribution for the projections. With the large growth rate tail of the growth distribution, the projections include indications near 10 volts independent of POD although the lower POD values result in more high voltage indications. The largest voltage indications dominate the contribution to the burst probability. This influence of the large voltage indications is also shown by the small increase (factor of 2) in burst probability for POD = 0.6 between BOC-7 and EOC-7, since the POD = 0.6 distribution includes high voltage indications in both the BOC and EOC distributions. l 1 i i svw=94seyc6,<wec10wps 10-3

Tcble 10-1 Byron 1, SG C Estimated Hot Standby SLB Burst Probability at EOC-6 With Uncertainly Adjustment i Tubc/ Location Free Span Burst Limited TSP Disp. Burst "' 1 Imal TSP TW Burst Applicable SLB Row Col TSP Volts Burst Prob. Displ. Probability Burst Prob.

• 20 7

3 10.95 2.7E-02 0.128 1.8E-22 - 0.0 25 13 5 3.80 6.7E-04 < 0.10 9.2E-24 - 0.0 32 30 3 3.64 5.7E-04 0.100 9.2E 24 - 0.0 27 98 3 2.68 1.6E-04 0.090 3.2E-24 - 0.0 47 35 3 2.68 1.6E-04 < 0.10 9.2E 24 - 0.0 20 102 3 2.58 1.4E-04 0.118 6.2E-23 - 0.0 21 98 3 2.50 1.2E-04 0.11I 2.9E-23 - 0.0 f 27 59 3 2.45 1.lE-04 < 0.10 9.2E-24 - 0.0 16 59 3 2.38 1.0E-04 < 0.10 9.2E-24 -0.0 27 93 3 2.29 8.5E-05 0.116 5.0E-23 - 0.0 35 26 3 2.14 6.4E-05 0.105 1.6E-23 - 0.0 12 3 3 2.10 5.9E-05 0.344 4.0E-12 - 0.0 46 52 5 2.02 5.0E-05 < 0.10 9.2E-24 - 0.0 1 9 3 3 1.99 4.7E-05 0.394 6.2E-10 6.2E-10 46 70 5 1.99 4.7E-05 < 0.10 9.2E-24 - 0.0 18 42 3 1.97 4.5E-05 < 0.10 9.2E-24 - 0.0 25 10 3 1.95 4.3E-05 < 0.10 9.2E-24 - 0.0 40 78 5 1.88 3.7E-05 < 0.10 9.2E 24 - 0.0 40 65 3 1.85 3.4E-05 < 0.10 9.2E-24 - 0.0 12 74 5 1.81 3.1E-05 < 0.10 9.2E 24 - 0.0 5 1 3 1.19 5.0E-06 0.500 5.3E-06 5.0E-06 1 3 3 1.74 2.6E-05 0.492 2.9E-06 2.9E-06 3 1 3 0.94 1.8E-06 0.522 2.5E-05 1.8E-06 2 2 3 0.70 5.0E-07 0.507 8.8E-06 5.0E-07 7 2 3 1.00 2.4E-06 0.446 7.2E-08 7.2E-08 8 2 3 0.65 3.6E-07 0.431 1.9E-08 1.9E-08 1 6 3 0.72 5.6E-07 0.412 3.4E-09 3.4E-09 2 6 3 0.68 4.4E-07 0.402 1.3E-09 1.3E-09 8 4 3 0.71 5.3E-07 0.385 2.6E-10 2.6E-10 5 5 3 0.72 5.6E-07 0.397 8.3E-10 8.3E-10 6 4 3 1.00 2.4E-06 0.411 3.lE-09 3.lE-09 4 112 3 0.56 1.9E-07 0.458 2.0E-07 1.9E-07 3 114 3 0.51 1.3E-07 0.522 2.5E-05 1.3E-07 4 111 3 0.44 6.7E-08 0.433 2.3E-08 2.3E-08 4 110 3 1.11 3.7E-06 0.408 2.4E-09 2.4E-09 2 109 3 0.74 6.3E-07 0.402 1.3E-09 1.3E-09 8 111 3 1.11 3.7E-06 0.385 2.6E-10 2.6E-10 Total Burst Probability 3.0E-02 1.lE-05 Notes: 1. Analysis conservatively assumes the TSP displacement exposes a throughwall crack equal to the displacement. 2. The applicable burst probabihty is the lesser of the free span and TW burst probabihties. 10-4

n

11.0 REFERENCES

11.1 NRC Letter, " Safety Evaluation by the Office of Nuclear Reactor Regulation Related to Amendment No. 66 to Facility Operating License No NPF-37 and Amendment No. 36 to Facility Operating License No. NPF-66; Commonwealth Edison Company ' Byron Station Unit Nos. I and 2; Docket Nos. STN 50-454 and STN 50-455" USNRC.. November 1994. j 11.2 NRC Generic Letter 94-XX,'" Voltage-Base Repair Criteria for the Repair of l Westinghcuse Steam Generator Tubes Affected by Outside Diameter Stress Corrosion - Cracking", USNRC Office of Nuclear Reactor Regulation, August 1994. 11.3.CAE-94-233, " Byron-1 S/G EOC-6 IPC SLB Leak Rate and Tube Burst Probability l Assessment", Westinghouse Electric Corp., Nuclear Services Division, October 17, i 1994. I 1.4 NP-7480-L Volume 2, SG-9310-102, " Steam Generator Tubing Outside Diameter Stress Corrosion Cracking at Tube Support Plates - Database for Alternate Repair i Criteria, Volume 2: 3/4 Inch Diameter Tubing", EPRI Report, October 1993. 11.5 NSD-RFK-95-002, "SLB Leak Rate and Tube Burst Probability Analysis Methods for ODSCC at TSP Intersections", Westinghouse Electric Corp., Nuclear Services Division, January 1995. 11.6 WCAP-14046 Rev.1, "Braidwood Unit 1 Technical Support for Cycle 5 Steam Generator Interim Plugging Criteria", Westinghouse Electric Corp., Nuclear Services Division, August 1994. f i 11.7 WCAP-14222, "Braidwood Unit I and Byron Unit 1 Model D4 Steam Generator Limited Tube Support Plate Displacement Analysis in Support ofInterim Plugging [ Criteria", Westinghouse Electric Corp., Nuclear Services Division, November 1994, i 11.8 TR-100407, Revision 1 (draft), "PWR Steam Generator Tube Repair Limits - Technical Support for Outside Diameter Stress Corrosion Cracking at Tube Support Plates," Electric Power Research Institute, August 1993. t I i I i s:5Vae94tyc6ipchell.wp3 } }.}}}