ML20210P861

Cycle 7 Interim Plugging Criteria Rept

ML20210P861
Person / Time
Site:	Braidwood
Issue date:	08/31/1997
From:	WESTINGHOUSE ELECTRIC COMPANY, DIV OF CBS CORP.
To:
Shared Package
ML20210P853	List: ML20210P849 ML20210P861
References
SG-97-08-002, SG-97-8-2, NUDOCS 9708270378
Download: ML20210P861 (178)

Text

{{#Wiki_filter:.___.____ __ _ _.. _. -. _ _ _. _.. WESTINGIIOUSE PROPRIETARY CLASS 3 SG-97-08-002 BRAIDWOOD UNIT - 1 CYCLE 7 INTERIM PLUGGING CRITERIA REPORT l 2 August 1997 l 1 O Westinghouse Electric Corporation Energy Systems Business Unit Nuclear Services Division P.O. Box 157 Madison, Pennsylvania 15663-0157 9708270378 970814 PDR ADOCK 05000456 P PDR en

WESTINGHOUSE PROPRIETARY CLASS 3 SG-9748-002 BRAIDWOOD UNIT - 1 CYCLE 7 INTERIM PLUGGING CRITERIA REPORT August 1997 . = = -. - - = - - - - -

1 Table of Contents 1 1 Page No. l 3

1.0 INTRODUCTION

11 i 2.0

SUMMARY

AND CONCLUSIONS 21 3.0 BRAIDWOOD UNIT 1 1997 PULLED TUBE EXAMINATION RESULTS AND EVALUATION 31 l 3.1 Summary of Pulled Tube Examination Resulta 31~ j 3.2 Evaluation of Pulled Tube Data for ARC Applications 37 1 3.3 Comparison of Braidwood Unit 1 Data with the EPRI Database 39 j 4 - 4.0 EOC 6 INSPECTION RESULTS AND VOLTAGE GROWTH RATES 41 s 4.1 EOC 6 Inspection Resulta 41 4.2 Probe Wear Criteria 43 l . 4.3 NDE Uncertainties 44 5.0 VOLTAGE GROW"I'H RATE ANALYSIS 51 5.1 Comparison of Cycle G Growth with Prior Cycle Growth 51 5.2 Voltage Dependency of Orowth Rates 53 6.0 - METHODS EVALUATION AND BENCHMARK ANALYSES FOR i POD = 0.0 61 6.1 Identification of Projection Methods Issue 61 0.2 Recommended Bobbin Voltage Growth Guidelines 63 0.3 Voltage Dependent Growth Rate Distributions for POD = 0.0 65 6.4 Braidwood Unit 1 Analyses for Benchmarking Alternate Voltage - Projection Methods G0 0.5 Conclusions. G8 7.0 DATA BASE APPLIED FOR ARC CORRELATIONS 71 ' oAN111$\\npc\\cce97%ccec690dlwp6 - j 9 .g -a-w-w 9 y y myy -y y 9 g g, 9g ny yyggW gf-, y a9 w e a. y gy-4. -g9*.y s-93 43,_ yg -.-,4g.4++ewv 4-'er ee-%iw+P'

8.0 BOBBIN VOLTAGE DISTRIBUTIONS 81 8.1 Calculation of Voltage Distributions 81 8.2 Probability of Detection (POD) 82 8.3 Limiting Growth Rate Distribution 82 8.4 Cycle Operating Period 83 8.5 Projected EOC 7 Voltage Distribution 83 0.0 SLB TUBE LEAK RATE AND TUBE BURST PROBABILITY ANALYSES - D.1 0.1 Leak Rate and Tube Burst Probability for EOC 7 91 0.2 Axial Tensile Rupture Probability 92

10.0 REFERENCES

10 1 Appendix A Development and Evaluation of An Alternate Voltage Projection Method A.1 Methods Evaluations A1 A.2 Recommended Bobbin Voltage Growth Guidelines A5 A.3 Braidwood 1 and Byron 1 Voltage Dependent Growth Distributions A8 A 4 Braidwood 1 EOC.G Analyses for Evaluation of Alternate Voltage Projection Methods A 11 A.5 Conclusions A 14 Appendix 13 Probability of Prior Cycle Detection B.1 POPCD Distribution for Braidwood Unit 1 EOC 5B Inspection B1 13.2 Generic POPCD Distribution Based on 18 Inspections in 8 Plants B3 B.3-Assessment of RPC Confirmation Rates for Braidwood Unit 1 EOC 5 Inspection - B5 .oANSD16\\ ape \\cce97\\ccec690d wp5 g_ 9

Braidwood Unit - 1 Cycle 7 Interim Plugging Criteria Report

1.0 INTRODUCTION

This report provides the Braidwood Unit I steam generator (S0) tube Eddy Current (EC) inspection results at the end of Cycle O together with Steam Line Bronk (SLB) leak rate and tube burst probability analysis results calculated according to NRC guidelines, to continue implementation of a 3.0 volt Interim Plugging Criteria (IPC). SLB leak rates and tube burst probabilities were calculated for end of cycle (EOC) conditions of both the recently completed cycle (Cycle 6) and the ongoing cycle (Cycle 7). These analyses were carried out considering the locked tube support plate (TSP) condition. A comparison of the actual measured voltage distributions with the projected EOC 6 voltage distributions presented in the last 90 day report (Reference 10.1) showed that the projections underpredicted indications with larger voltages in the tail of the voltage distributions. Consequently, the projected EOC.0 leak rates underestimated those calculated using the actual measured bobbin voltage distributions for all S0s except SG B. A root cause evaluation for underprediction of projected EOC 6 leak rates was performed, and it was concluded that the cause was the manner in which growth was simulated in the projection analyses. The Cycle 6 growth data shows a strong dependency on the beginning of cycle (BOC) voltage: however, the EOC G leak and burst projections presented in the last 90 day report were obtained assuming that growth during a cycle is independent of the BOC voltage. The standard leak rate and burst probability projection methodology, until now, described in Reference 10.2, was modified to consider voltage dependent growth distributions. The revised projection methodology was evaluated by repeating the EOC 6 projections considering the voltage dependency of growth rates, and the results show good agreement with those based on the actual measured voltages. The revised leak and burst analysis methodology was utilized to project leak rates and tube burst probability for postulated SLB conditions at the end of the ongoing cycle (Cycle 7) based on the 3.0 volt repair criteria. Those analyses utilized bobbin voltage distributions measured during the recent (EOC G) inspection and a limiting growth rato distribution from the last two inspections (EOC.5B and EOC G inspections), in addition to the SLB burst probability, the probability of axial tensile o \\NSD16%apcNcce9'l\\rcec6mi wpb 1.}

- ~ -. -. - - - - - - a i tearing was calculated considering hot leg side indications and using a correlation for axial strength that has been updated to include the data from the recent Braidwood 1 pulled tube examination, j l Sections from the hot leg side of two tubes from Steam Oenerator A were pulled during the EOC.6 outage and examined. Results from metallurgical examinations-and leak and burst tests are summarized in this report. Also, the effect of additional leak and burst results on the correlations used for predicting SLB leak rate and tube burst probability are assessed. Results from two other evaluations are also presented in this report. One of them is an evaluation to assess the probability of prior cycle detection (POPCD) for the EOC. 5B inspection using data from EOC 5B and EOC 6 inspections. The other evaluation examined fraction of the indications that showed no degradation during the rotating pancake coil (RPC) probe inspection in 1995 EOC.5B' inspection, were left in service at beginning of cycle 6 (BOC.8), and were RPC confirmed in 1997 at EOC 6. 4 1 4 ? i 1

• Since this is the second of two inspections conducted during Cycle 5. it is referred to as Cycle 5B e

o:\\NSD16\\npe\\cce97\\cccc690d wp5 12 -. ~

2.0

SUMMARY

AND CONCLUSIONS A comparison of the actual measured voltage distributions with the projected EOC 6 voltage distributions presented in the last 90 day report (Reference 10.1) showed that the projections underpredicted larger voltages in the tail of the voltage distributions. Consequently, the projected EOC 6 leak rates underestimate those calculated using the actual measured bobbin voltage distributions for all SGs except SG B. An evaluation was carried out to identify the cause for the underprediction of the projected EOC 6 leak rates, and it was determined that it was due to the manner in which growth is simulated in the projections. Cycle 6 represents the first full cycle of operation after increasing the repair limit from i volt to 3 volts. A significant population ofindications between I to 3 volts were left in service for the first time in the Braidwood Unit 1 history and it appeared to have affected the growth distribution during Cycle 6. Leak and burst projections for EOC 6 were performed assuming that growth during a cycle is indeptndent of the BOC voltage. However, a closer examination of the growth rate d'stributions for Cycle 6 showed a strong dependency of growth on the BOC voltage The largest growth amplitude in a given voltage interval does not significantl; increase with the BOC voltage; however, as the indication population in a given voltage interval decreases with increasing BOC voltage, the relative frequency of larger growths, expressed as a percent of the number ofindication in the voltage range, increases with the BOC voltage. The net effect is that the larger indications left in service have a higher probability oflarge growth values than the lower voltage indications and the EOC projections result in a larger voltage tail of the projected distribution. Other potential causative factors for contributing to the leak and burst u0derestimation such as a low POD, voltage underestimation due to probe wear, and differences in growth adjustment between EFPD and days at temperature scaling were determined to be insignificant contributors to the causative mechanisms. Alternate methods for including voltage dependence in the growth rate distributions were evaluated, and recommended changes to the current leak rate and burst probability projection methodology, described in Reference 10.2, to include voltage dependent growth were developed. Detailed guidelines were established to develop voltage dependent growth distribution to be used in leak and burst projections. The revised projection methodology was evaluated by repeating the EOC-6 projections considering voltage dependency of growth rates, and the results show good agreement with those based on the actual measured voltages. The methodology based on voltage dependent growth does not affect standard 1.0 volt IPC leak and burst calculations, o \\NSD15\\ ape \\cce97\\ccec690d wp5 2-1

The revised leak and burst analysis methodology was utilized to project leak rates and tube burst probability for postulated SLB conditions at the end of the ongoing cycle (Cycle 7) based on the 3.0 volt repair criteria. These analyses utilized bobbin voltage distributions measured during the recent (EOC.G) inspection and the Cycle a growth rate distribution. The projected EOC 7 leak rates and tube burst probabilities aro within their allowable limits pending NRC review of a reduction of the Unit.1 reactor coolant system (RCS) dose equivalent 1 131 Technical Specification limit. Limiting SLB leak rate projected for the EOC 7 conditions (locked TSPs) using the NRC mandated probability of detection of 0.0 is 57.1 gallons per minute (gpm) at room temperature. This value is projected for SG.C which has the largest number ofindications and will be below the allowable EOC 7 limit. The highest tube burst probability, 8.0x10, is predicted for SG A which has the largest number of 4 indications on the cold leg side (only cold leg indications significantly contribute to burst due to TSP locking on the hot leg side) and it is well below the NRC reporting guideline of 10' In addition to the SLB burst probability, the probability of axial tensile rupture was calculated considering hot leg side indications in SG C. A correlation for axial strength that has been updated to include data from the recent Brai'iwood.1 pulled tubo examination was applied. The projected EOC 7 axial tensile tear probability based on a constant POD of 0.0 is 7.9x10. Even if the largest SLB 4 burst probability calculated for cold leg indications,8.0x10 predicted for SG A, is d added to the calculated axial tensile rupture probability, the total tube failure probability (8.8x10 ' *) is still more than an order of magnitude below the NRC reporting guideline of 10'8 A total of G784 indications were found in the EOC.G inspection of which 050 were inspected with a RPC probe (including a minimum of 20 % of hot leg indications between 1 and 3' volta, all hot leg indications above 3 volts and all cold leg indications above 1 volt), and 561 were confirmed as flaws. The RPC confirmed indications included 511 above 1.0 volt and 107 above 3 volts. A significant population increase is noted throughout the bobbin voltage spectrum. SG.C had the largest number among the four SGs with 2104 bobbin indications, of which 717 were above 1.0 volt and 30 above 3 volts including the largest indication found (10.48 volts). All indications above 3 volts were confirmed as flaws and repaired. An augmented RPC inspection was performed consistent with the NRC requirements. All dented intersections with a bobbin voltage greater than 5 volts and a minimum of 20 percent of the dented intersections with bobbin voltages between 2.5 and 5 volts were inspected with RPC. The augmented RPC inspection also included one TSP intersection with a mixed residual artifact signal in SG C, but it was not confirmed. A single circumferential indication was detected in a large dent at TSP 5H in SG A. o \\NSD15\\apc\\cce97%ccecEKkl wp5 22 y

There were no other RPC circumferential indications at the TSPs, no indications extending outside the TSPs, no RPC indications with potential PWSCC phase angles, no other flaw indications on dents at any dent voltage and there was no signal i intederence from copper deposits. An alternate probe wear criteria (Reference 10.4) was applied during the EOC 6 inspection. When a probe does not pass the 15% wear limit, this alternate criteria requires that all tubes with indications above 75% of the repair limit since the last successful probe wear check be reinspected with a new probe. Although the repair limit for hot leg indications is 3 volts, all indications for which the worn probe voltage was above 0.75 volt were reinspected with a new probe. An evaluation of worn probe and new probe data was performed and it was found that the criteria used in the EOC-6 inspection is consistent with the NRC guidance provided in Reference 10.8. Sections from the hot leg side of two tubes (R28C24 and R41C65) from Steam i Generator A were pulled during this outage and examined. Results from . metallurgical examinations and leak and burst tests are summarized in this report. Also, the effect of additional leak and burst results on the correlations used for predicting SLB leak rate and tube burst probability are assessed. The pulled tube leak rates, burst pressure and crack morphology are consistent with the EPRI database. An evaluation was performed to assess probability of prior cycle detection (POPCD) for the EOC 5B inspection using data from EOC 5B and EOC 6 inspections. The EOC-5B POPCD distribution obtained compares well with a generic POPCD distribt. tion established using data from 15 inspections in 8 plants. Another evaluation examined the fraction of the indications that showed no degradation during the RPC probe inspection during the 1995 EOC-5B inspection, were left in service at BOC 6, and were RPC confirmed in the EOC 6 inspection. Only 40% of 30 EOC-5B indications that showed no detectable degradation (NDD) were confirmed during the present inspection. Thus, it is conservative to include 100% of the RPC 1 NDD ir.dications in the SLB leak rate and burst probability calculations as was done for the EOC 7 projections. k o \\NSD16\\ ape \\cre97\\ccec690d wp5 -23 .. ~. ..~

3.0 BRAIDWOOD-1 1997 PULLED TUBE EXAMINATION RESULTS AND EVALUATION 3.1 Summary of Pulled Tube Exan'.ination Results 3.1.1 Introduction Sections from two hot leg steam generator tubes from Braidwood Unit 1 SO A were sent to Westinghouse Science and Technology Center for evaluation of corrosion indications detected by field oddy current inspections during April 1997. The as received sections included tubing from the top of the tubesheet ('ITS) region, the first TSP region, aise known as the flow distribution baffle (FDB) region, the third TSP region and the fifth TSP region from both Tubes, R28C24 and R41CG5 (TSP's 2 and 4 do not exist). Of primary interest was the presence of OD origin indications at the TSP 5 regions of both tubes. The following presents a summary of the more significant findings from the examination. 3,1.2 Nondestructive Examinations Table 31 presents a summary of field and laboratory eddy current data obtained on the two tubes for both the TSP 3 and TSP 5 locations. All other locations had no detectable degradation. The field information presented is that obtained from a review of the data. In the field, the TSP 5 region of Tube R28C24 had a G.2 volt bobbin signal that appeared to be a single axial indication (SAI) by the + Point coil examination. In the laboratory, the bobbin coil signal increased to 17.3 volts and the corrosion degradation appeared as multiple axial indications (MAI) spread over the entire tube circumference. Undoubtedly the tube pull related deformation opened some of the cracks and was responsible for the larger bobbin coil signal voltage and the change in the + Point call. Similarly, in the field, the TSP 5 region of Tube R41CG5 had a 8.93 volt bobbin coil signal that appeared as SAI by + Point coil examination, but with noticeable volumetric involvement, suggesting local intergranular cellular corrosion (ICC) or intergranular attack (IGA). In the laboratory, the TSP 5 region bobbin coil signal increased to 13.5 volts and the + Point indication similarly was an SAI with significant " IGA" involvement spread over 105'. Both TSP 3 regions had distorted indications (dis) present in the field wits ',hbin voltages of 0.68 (R28C24) and 0.41 (R41CG5) volts. Their + Point indications suggested the presence of MAI. In the laboratory, the eddy current signals appeared similar with the exception of a laboratory dent signal in the TSP 3 region of Tube o \\NSDIS\\apcNece97\\ccec690d wps 31 w w.

R28C24 which pm.tially Lid the bobbin Di signal. The dent was introduced by the tube pull, as it was not present in the field eddy current data. 3.1.3 Leak, Burst and Tensile Test Data Elevated temperature leak testing was performed on the two TSP 5 regions at differential pressures ranging from normal operating conditions (NOC) to steam line break (SLB) conditions. Table 3 2 presents a summary of the leak test conditions and results. The leak rates for TSP 5 of Tube R28C24 ranged from 0.2G4 liters por hour (NOC) to 5.10 liters per hour (SLB). The leak rates for TSP 5 of Tube R41C05 similarly ranged from 0.204 liters per hour (NOC) to 7.50 liters per hour (SLB). These volumetric leak rates are measured at room temperature. Additional leak tests were performed to simulate leakage for an indication restricted from burst (IRB) based on pressurizing the indication to the free span burst pressure inside the TSP followed by leak measurements with the TSP in place. Following elevated temperature leak testing, the cracks present in TSP 5 of Tube R28C24 were opened wider by pressurizing the tube (using a bladder) to 6670 psi inside a TSP simulant (0.775 inch hole) with the TSP crevice region offset downwards relative to the TSP simulant by 0.15 inches. In addition, the tube was held to one side of the TSP simulant such that the main crack indication had the entire 0.025 inch gap available for expansion, it was later determined that the bottom portion of the main crack extended 0.12 inch below the bottom of the TSP simulant. (However, this 8 portion of the main crack was not intergranularly throughwall.) The observed 0.12 inch crack displacement beyond the TSP along with later SEM fractography observations suggest that the 0.15 inch TSP displacement was actually larger act probably closer to 0.2 inch displacement. It was also determined later that the tuo plastic expansion was not large enough for the tube to fill the tube to TSP gap and the expanded tube was not tightly bound inside the TSP simulant. After the above described pressurization was applied to the TSP 5 region of Tube R28C24, the tube was leak tested at room temperature in a high capacity leak test apparatus at differential pressures ranging from 1300 psi to 2000 psi. Table 3 2 also Calculated to be G70 psi greater than the nominal burst pressure based on eddy drent voltage correlation using measured material properties. A not known from subsequent destructive examination work if the crack tip c er.ending below the TSP simulant was sufficiently torn by the " burst" pressurization to 6770 psi that some or all of the crack tip was made throughwall bv the tearing. o \\NSD15\\apc\\cce97\\ceccW0d wp5 32

presents these lonk test conditions and results. The leak rates ranged from 184 to 208 liters per hour. The TSP simulant was then shifted directly over the tubo crevice region (no offset) and the leak testing was repeated. The measured leak rates ranged from 101 to 200 liters per hour. Figure 31 shows a plot of these room temperature leak test data showing no significant difference in the leak rates as a function of the TSP simulant being offset or centered. Following lonk testing, the TSP 3 and TSP 5 regions of both tubes, as well as a free span (FS) region with no corrosion, were burst tested at room temperature. The TSP 3 and FS specimens were burst tested without any restraints. The TSP 5 regions were burst tested using foils and bladders, but were also tested without any external restraint. Table 3 3 presents a summary of the data. The lowest burst pressure was for the TSP 5 region of Tube R41CG5 which burst at 5,510 psi or 47% of its FS control. The TSP 5 region of Tube R28C24 burst at 0,010 psi or 53% ofits FS control. Both TSP 3 regions hwi a burst pressure smaller than their respective FS burst pressure and also showed ductility reductions, indicating the presence of corrosion. A FS section from each tube was also tensile tested at room temperature. Both had tensile properties typical of Westinghouse mill annealed Alloy 600 tubing from this vintage of tubing. See Table 3 3. The burst tests opened and made visible secondary corrosion within the crevice region in addition to the main axial macrocrack which initiated the burst opening. Figures 3 2 and 3 3 present sketches of the corrosion visually observed on the two burst tested TSP 5 crevice regions which were selected for destructive examinations. As was suggested by the eddy current data, the secondary corrosion on TSP 5 of Tube R28C24 was found completely around the circumference while the secondary corrosion on TSP 5 of Tube R41CG5 was confined to a 130 wide patch. Following burst testing, the two TSP 5 regions were pulled apart using a tensile machine in order to create a circumferential fracture face through the secondary corrosion in the crevice regions of the burst specimens. The separation loads were 7,700 pounds for Tube R28C24 (versus 10,640 pounds for its free span control) and 7,400 pounds for Tube R41C05 (versus 9,800 pounds for its free span control) with both circumferential fractures occurring near the center of the burst openings and through the crevice region secondary corrosion. Much of the secondary corrosion had an ICC appearance, o \\NSD15\\ ape \\cce97\\ccw690d wp5 33

3,1,4 Destructive Examination Results For the TSP 5 region of Tubes R28C24 and R41CG5, SEh! fractography was performed on the circumferential fracture face created by the tensile pulling of the burst specimen and on the axial burst opening which was in the form of a top and bottom fracture face since the tensile fracture face bisected the axial burst opening. The tensile fracture face fractography was done to characterize the cross sectional degraded area in the TSP regions of secondary corrosion, and the axial burst fractography was performed to characterize the axial macrocrack which initiated the burst opening. Table 3 4 presents a summary of the fractographic observations. Note that both axial crack lengths were not corrected for the non uniform tensile elongation of the post burst test tensile separation. Instron charts suggest that approximately 0.11 inch of plastic clongation occurred for both TSP 5 specimens. For the circumferential fracture faces, the lengths (degrees) provided were corrected for burst ductility. For the outside diameter (OD) origin corrosion, the intergranular corrosion macrocracks (both on the axial burst openings and the tensile fracture faces) were composed of numerous intergranular microcracks that were interconnected by ligaments. hiost of these ligaments had intergranular features, indicating that they had grown together during plant operation. A few of the ligaments had tensile tearing (ductile) features, indicating that they tore during tube pulling, subsequent burst and tensile testing or destructive examination specimen preparation. For the TSP 5 region of Tube R28C24, the axial burst opening occurred at a macrocrack that averaged 65% deep over 0.088 inch with a maximum depth of 100% that occurred over 0.200 inch of the macrocrack. Similarly, for the TSP 5 region of Tube R41C05, the axial burst opening occurred at a macrocrack that averaged 01% deep over 0.686 inch with a maximum depth of 100% over 0.268 inch of the u.acrocrack. The 0.11 inch clongation from the tensile test was non uniform and a correction could not readily be applied to lengths. It is likely that the given profile measurements overestimated the throughwalllength as well as the total crack length. The tensile fracture at the TSP 5 regior, of Tube R28C24 occurred through secondary crevice region corrosion that was composed of one major macrocrack (averaging 21% deep over 150' with a maximum depth of 37% away from the burst opening and 100% at the burst opening), as well as four much shorter macrocracks. The overall o.\\NSD15Naperne97\\nec690d wp5 34

8 macrocrack was 188* long. The measured secondary corrosion degraded 10% of the tube cross sectional area. The tensile fracture at the TSP 5 region of Tube 1141C05 occurred through secondary corrosion that composed of one macrocrack that averaged 29% deep over D0' with a maximum depth of 40% away from the burst opening and 100% at the burst opening. The secondary corrosion degraded 7% of the tube cross sectional area. Transverse and radial metallographic sections were made to characterize the TSP 5 crevice region secondary corrosion for both tubes. Figure 3 4 shows an example of the transverse metallography for TSP 5 of R28C24 and Figure 3 5 shows an example of the radial metallography at 21% depth for R41C05, TSP 5. Metallographic examinations showed that the corrosion morphology was composed primarily of axial intergranular stress corrosion cracking (IGSCC) with significant ICC components in both cases. The overall contribution of the ICC to the corrosion morphology tended to decrease as a function of depth. Where the corrosion was the deepest, only axial IGSCC was present. Table 3 5 presents a summary of the metallographic data. Corrosion morphology also can be characterized by crack density and by the degree that IGA components are associated with individual IGSCC, frequently measured by depth.to. width (D/W) ratios. Both TSP 5 regions had moderate overall crack densitias (estimated at 50 cracks around the mid crevice location), but the distribution of cracking was different. In the case of Tube R28C24, the cracking was distributed 4 almost uniformly across the circumference. In the case of Tube R41CG5, the cracking was mostly confined to a 130' wide patch. Note that a high crack density is defined as greater than 100 cracks around the circumference, a moderate crack density as 25 to 100 cracks, and a low crack density as less than 25 cracks around the circumference. Both TSP regions also had a moderate association of IGA with individual IGSCC, with typical DAV ratios of 10 in the case of Tube R28C24 and 7 in the case of Tube R41CG5. Note that a DAY ratio is measured as crack depth divided by crack width at its mid depth. DAV ratios in the range of 3 to 20 suggest a moderate association of IGA with IGSCC, DAV ratios less than 3 suggest a high association of IGA with IGSCC, and DAV ratios greater than 20 suggest a low association of IGA with IGSCC. 8 Other corrosion was sporadically present on the tensile fracture face at other locations around the tube circumference. However, it was very shallow and not characterized, It was less than 2% deep, i Significant cracking was present in most regions (300') of the tube circumference. This is in contrast to the SEM fractographic data on the tensile fracture face where the cracking primarily occurred over 188* of the circumference. o \\NSD15\\apcNec45herec690d wpS 35

3.1.5 Conclusions The TSP 5 rM* ions of hot leg Tubes R28C24 and R41C05 from Steam Generator A of Braidwood Unit 1 both had axial IOSCC with significant ICC components. Each TSP 5 region had a single large axial IOSCC macrocrack that was throughwall. All other cracking was considerably more shallow, nnximum depth of ~49%. Secondary corrosion in the crevice region was distributed completely around the crevice region in the case of Tube R28C24, but was shallow (percentage degraded area of10%). The secondary corrosion was locally deeper (maximum depth of 46% by fractography and 49% by metallography) in the case of Tube R41CG5, where it was confined to a 130' wide patch. However, the effective percentage degraded area of the tube cross section was similar, 7%. All corrosion was confined to the crevice regions and was of OD origin. This corrosion was detected by both field and laboratory eddy current, including a call ofits extensive volumetric nature in the case of Tube R41C05 where the local secondary corrosion was deeper than for Tube R28C24. Elevated temperature leak testa performed on the two TSP 5 regions produced similar leak rato data with a maximum leak rate of 0.264 liters per hour at normal operating conditions (Tube R28C24) and a maximum leak rate of 7.5 liters per hour at steam line break conditions (Tube R41CG5). These volumetric leak rates are room temperature measurements since the leakage was condensed and measured at room temperature. The TSP 5 region of Tube R28C24 was then expanded by pressurizing to above calculated rupture conditions to further open its major crack and simulate leakage from an IRB. This pressurization was perfornied in a simulated tube support plate with a 25 mil diametrical gap. Room temperature leak tests were then performed with the tube support plate offset by 0.15 inch, such that the crack tip extended below the tube support plate by 0.12 inch, and with the tube support plate centered over the crevice region. Similar leak rates were measured at both tube support plate locations at differential pressures ranging from 1300 psi to 2000 psi with the maximum leak rate being 298 liters per hour. As the crack opening was not restricted by the support plate, and the throughwall portion of the crack (consisting of the intergranular component plus any rupture related tearing component) may always have been under the support plate, those results are not surprising. Both TSP 5 regions experienced a significant deterioration in burst strength with the TSP 5 region of Tune R41C64 having the more significant decrease. It burst at 47% of a nondegraded free span section of the pulled tube. Both TSP regions burst well above safety limitations required by Regulatory Guide 1.121. o \\NSD1$\\ ape \\cce97\\cccc690d wr5 36 .--...n.

Each of the two burst opening cracks contained an intergranular macrocrack composed of a large number ofintergranular microcracks joined together primarily by intergranular ledges. The Tube R28024 hurst macrocrack averaged 65% deep over a length of 0.688 inch, with a maximum depth of 100% throughwall over a length of 0.200 inch. The Tube R41CG5 burst macrocrack averaged 01% deep over a length of 0.686 inch, with a maximum depth of 100% throughwall over a length of 0.208 inch. Metallography showed that the two TSP 5 regions had similar secondary corrosion morphologies with overall moderate crack densities and a moderate association of IGA components with IGSCC. The corrosion morphologies of these tubes are typical of most tubes characterized and used as part of the alternate repair criteria (ARC) data base. 3.2 Evaluation of Pulled Tube Data for ARC Applications This section evaluates the pulled tube examination results described above for application to the Electric Power Research Institute (EPRI) database for ARC applications. The eddy current data is reviewed, including reevaluation of the field data, to fimalize 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 is then defined and reviewed against the EPRI outlier criteria to assure acceptability for the database. 3.2.1 Eddy Current Data Review Table 3 6 provides a summary of the eddy current data evaluations for the Braidwood 1 pulled 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 voltage 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 variable is separately accounted for in ARC applications as an NDE uncertainty. Most of the pulled tube EPRI database has been analyzed by the same analyst that performed the field reevaluation of Table 3 6, Thus the reevaluated field bobMn voltages are applied for application to the ARC correlations. The field bobbin data for the field NDD calls were reevaluated to derive the most appropriate amplitude measurements, where possible, for these very small signals. This review indicated that the field NDD indication for R41CG5 3H could be assigned o \\NSD15\\apc\\cce97\\ccee690d wp5 37 i t

a bobbin flaw voltage. The reevaluated bobbin and RPC data for this indication are 0.41 bobbin volt and 0.13 + Point volt. The increase in the post pull bobbin voltage for R28C24-TSP 5 (0.08 to 17.8)is higher than found for most of the pulled tubes in the EPRI database. The post pull + Point results show a significant increase in response around the tube circumference compared to the field results. This is attributable to opening of cellular corrosion as a result of the tube pull. From Table 3 4, it is seen that most of the ligaments between microcracks were corroded prior to the tube pull and only one uncorroded ligament remained in the macrocrack. Thus, the tube pull would not have resulted in tearing of ligaments to increase the voltage although the pull forces could have decreased contact across the faces of the crack. Significant tearing ofligaments or crack opening to strongly affect accident condition leakage or burst capability is not expected for this indication and the voltage increase appears to be principally due to opening of ceUular corrosion cracks. Thus, the pre pull voltage measurements are acceptable for ARC applications for the indications given in Table 3 6. 3.2,2. Braidwood-1 Data for ARC Application The Braidwood 1 pulled tube results, as developed above, are summarized in Table 3 7. The measured leak rate data of Table 3 2 are adjusted in the table to the reference normal operating and SLB conditions by applying the leak rate adjustment procedure of the EPRI database report (Reference 10.5). The reference SLB conditions are a pressure differential of 25S0 psid at a primary pressure of 2575 psi and a secondary pressure of 15 psia at a temperature of GIG" F. The measured burst pressures are adjusted to the nominal flow stress of 71.57 for 3/4 inch tube at operating temperature. The measurements for R28C54 TSP 5 included IRB teak rates following pressurization of the indication inside the TSP to the estimated free span burst pressure. The resulting IRB SLB leak rate of 166.8 liters per hour corresponds to 0.73 gpni which is much less than the 6.0 gpm used in the IRB leak rate analyses which was based on laboratory specimens with very large (up to 0.809" throughwall) fhwa. This comparison demonstrates that one of the largest field voltage and throughwall flaw length indications result in leak rates well below the very conservative value for IRB analyses. Tensile rupture force measures were also made for the 5th TSP indication on both pulled tubes to support an update in the correlation supporting the Braidwood 13 volt IPC. The data of Table 3 7 are used in Section 3.3 to assess their influence on the EPRI ARC burst pressure, SLB probability ofleakage and SLB leak rate versus voltage oANsD15\\apc\\cce97\\ccec690dwp5 38

correlations. The Braidwood 1 pulled tube results were evaluated for potential exclusions from the database against the EPRI data exclusion criteria. Criteria la to Ic and le of Table 51 in Reference 10.9 apply primarily to unacceptable voltage, burst or leak rate measurements and indications without leak test measurements. These criteria do not apply to the indications of Table 3 7 and would not lead to any exclusions from the database. Criterion Id applies to potential tube pull damage but is not applicable based on the discussion given above relative to tube pull damage and the associated bobbin voltage increases. Criterion 2h applies only to indications > 20 volts which is not applicable to the Braidwood 1 indications. EPRI Criterion 3 has not been approved by the NRC and is not applicable although application would not lead to exclusion of the TSP 5 indications. Criterion 2a applies to atypical Ugament morphology and states that cracks having s 2 uncorroded ligaments in shallow cracks < 60% deep shall be excluded from the database. The indications at the 5th TSP of R28C24 and R41CG5 are throughwall and this criterion is not applicable. The bobbin indications at the 3rd TSP of both tubes have not been destructively examined. Therefore, the indications cannot be evaluated against Criterion 2a and these two data points must be excluded from all EPRI correlations. The burst pressures for both of the TSP 3 indications are above the nommal burst correlation and within the scatter band of database. Inclusion of the indications in the database would not significantly affect the correlations. However, they must be excluded due to the inability to assess the indications against exclusion criterion 2a. Based on the above assessment, the indications at R28C24 TSP 5 and R41CG5. TSP 5 are included in the EPRI database for the burst, probability ofleak and leak rate correlations. The indications at TSP 3 are excluded from the database and all correlations based on the inability to evaluate the indications against data exclusion Criterion 2a. 3.3 Cc mparison of Braidwood-1 Data with the EPRI Database This section reports on the evaluations performed utilizing the results from leak rate, burst pressure, and axial rupture force tests of tube sactions removed from SGs at the Braidwood-1 site in 1997. In addition, the results in6ude the evaluation of the result from an axial rupture force test on a 7/8" diameter tube removed from plant A 1. The data obtained from the tests are compared to the reference EPRI database as identified in Reference 10.5. The results of the destructive examinations of the tube sections were delineated earlier in this report. In summary, the test data are consis-tent with the database relative to the probability of leak, the leak rate, the burst o.\\NSDIS\\ ape \\cce97\\ccec690d wp5 39

pressures, and the axial rupture force as a function of the bobbin amplitude. These comparisons and evaluations are discussed below. 3.3.1 ODSCC Cracking Correlations This section reports on the evaluations performed which utilized the results ofleak rate and burst testing of the tube sections which were removed from Braidwood Unit-1 in 1997. The results of the destructive examinations of the tubes was recordnd earlier Sections 3.1 and 3.2. The results of the destructive examinations, e.g., leak and burst tests, are compared to the database of similar test results for 3/4" outside diameter steam generator tubes. In addition, the effect ofincluding the new test data in the reference database was evaluated. The report information on the destructive examinations of the tube sections was reviewed in Section 3.2.2 of this report relative to the EPRI guidelines for inclusion / exclusion of tube specimen data in the ARC database. This review revealed no information that would lead to a conclusion that the data should not be included in the database. Therefore, the resulting correlations should be considered applicable to the use of ARC for indications in 3/4" diameter tubes in Westinghouse SGs. It is noted that the destructive examination of two tube sections was not performed, accordingly, their data are not included in the evaluations documented in this report. In summary, the test data are consistent with the database relative to the burst pressures, the probability ofleak, leak rate and the axial rupture force as a function of the bobbin amplitude. The comparisons and evaluations are discussed below. 3.3.1.1 Suitability for Inclusion in the Database The report information on the destructive examinations of the tube sections was reviewed in Section 3.2 relative to the EPRI guidelines for inclusion / exclusion of tube specimen data in the ARC database. This revies evealed no information that would lead to a conclusion that the data should not be nicluded in the database. Therefore, .he resulting correlations should be considered applicable to the use of ARC for indications in 3/4" diameter tubes in Westinghouse SGs. It is noted that the destructive examination of tube sections at the 3rd TSP location was not performed; accordingly, their data are not included in the evaluation presented in this report. 3.3.1.2 Burst Pressure vs. Bobbin Amplitude The results from the burst tests, performed on tube specimens which exhibited a non-zero bobbin amplitude at a TSP elevation location, were considered for evaluation. asniss.pcsce 97sec.ce90d wp5 3-10

A plot of the burst pressures of the Braidwood 1 specimens is depicted on I'igure 3 0 relative to the burst pressure correlation developed using the reference database. 1. A visual examination of the data relative to the EPRI database indicates that the measured burst pressures fall within the scatter band of the refer-ence data. 2. The data points fall near the regression line, hence no statistical anomalies are indicated, i.e., the data are visually remote from the prediction bound. in summary, the visual examination doesn't indicate any significant departures from the reference database. Since the Braidwood 1 burst pressure data were not indicated to be from a separate population from the reference data, the regression analysis of the burst pretsure on the common logarithm of the bobbin amplitude was repeated with the additional data included. A comparison of the regression results obtained by including these data in the regression analysis is provided in Table 3 8 Regression predictions obtained by including these data in the regression analysis are also shown on Figure 3 0. A summary of the changes is as follows:

1) The intercept of the burst pressure, P,, as a linear function of the common logarithm of the bobbin amplitude regression line is increased by 0.06%, or about 5 psi. This has the effect ofincreasing predicted burst pressures as a function of bobbin amplitude.

2) The absolute slope of the regression line is decreased by 0.34%, i.e., the slope is less steep. This has the effect ofincreasing predicted burst pres-sures as a function of bobbin amplitude for large indications.

3) There is a decrease in the standard error of the residuals of 0.84%. The effect of this change is reflected in a slightly smaller deviation of the 95%

prediction line from the regression line. The net effect of the changes on the SLB structurallimit, using 95%/95% lower toler-ance limit material properties, is to increase it by 0.0G V, i.e., from 4.73 to 4.79 V. The increase in the intercept and the decrease in the standard error coupled with the fact that the structural limit is increased indicates that the probability of burst would also decrease for bobbin indications over the structural rar.ge ofinterest. Based on o \\NSD15\\apc\\cev97\\ccec6mki wp5 3 11

the relatively small change in the structural limit, the change in the probability of burst would also be expected to be small. 3.3.1.2 Probability of Leak Correlation The data of Table 3 7 were examined relative to the reference correlation for the probability of leakage (pol) as a function of the common logarithm of the bobbin amplitude. Figure 3 7 illustrates the Braidwood 1 data relative to the reference correlation. The specimen exhibited expected pol behavior, i.e., the two indications had a calculated high probability ofleak, and both leaked. Had the expectation been significantly different from the expectation, statistically anomalous behavior might have been suspected. Thus, based on the data examination, there is no significant evidence ofirregular results, i.e., outlying behavior is not indicated. In order to assess the quantitative effect of the new data on the correlation curve, the database was expanded to include the Braidwood 1 data points and a Generali:cd Linear Model regression of the pol 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 results indicate:

1) A 1.0% decrease (larger negative value)in the logistic intercept parameter.

2) A 1.5% increase in the logistic slope parameter.

3) The absolute values of the parameters' covariance matrix changed by 1.6%

to 2.0%. Examination of Figure 3 7 indicates that it is likely that there would be no significant impact on the 95% confidence bound on the total estimated leak rate from a single SG.

4) The deviance of the regression increased by 1%, an increase is expected when additional data is added, and the Pearson standard error increased by 1.1%.

In order to confirm the judgement that the changes are not significant, the reference correlation and the new correlation were also plotted on Figure 3 7. An examination of the figure reveals no significant change in the correlations. It is noted that when the total leak rate is determined using the leak rate to bobbin volts correlation, the resulting value can be quite insensitive to the form of the pol function. So, the effect of the changes in the parameter values and variances would be expected to be small o \\NSD15\\apcNece97%ccec690d wp5 3 12

or insignificant relative to the calculation of the 95% confidence bound of the total leak rate from a SG. 3.3.1.3 SLB Leak Rate Versus Bobbin Amplitude Correlation As previously noted, two of the removed tube specimens exhibited leakage under SLB conditions. The leak rates are provided in Section 3.1.2 they are depicted on Figure 3 8 relative to the correlation obtained using the reference database. The two data points from Plant AA 1 (Braidwood 1) straddle the reference correlation curve for the median leak rate. It is implied from the visual examination, using the relative distance from the 95% confidence bound on the arithmetic average to the arithmetic average, that the data 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. A summary of the parameters of the correlations is provided in Table 310. The following changes resulted from the addition of the new data:

1) The intercept of the correlation curve decreased by 0.18%.

2) The slope of the correlation curve decreased by 0.21%.

3) The standard deviation of the common logarithm of the leak rate residual errors decreased by 2.07%.

4) The p value for the slope coefficient decreased by 46% to 4.410*.

The net effect ofincluding the additional data is to slightly reduce the expected, i.e., arithmetic average, leak rate for bobbin amplitudes over most of the range of the data. In practice, the change to the estimated leak rates would not be expected to be significant. 3.3.1.4 Conclusions Relative to ODSCC Correlations The review of the effect of the Braidwood 1 data indicates that the correlations of the burst pressure, the probability ofleak, and the leak rate to the common logarithm of the bobbin amplitude would not be substantially 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 the use of the reference database would not be signifi-cantly changed relative to the results obtained from correlations developed after adding the Braidwood 1 data to the database. o \\NSD15\\ ape \\cce97\\ccec690d wp5 3 13

3.3.2 Axial Tensile Tearing Correlation Section G.3 of the EPRI database addendum, Reference 10.5, presents axial strength information for cellular corrosion in the form of a correlation of the remaining cross-sectional area (CSA) of the tube as a function of the bobbin amplitude of the TSP outside diameter stress corrosion cracking (ODSCC). The results from the axial rupture tests, performed on the Braidwood 1 and Plant A 1 tube specimens which exhibited a non zero bobbin amplitude at a TSP elevation location, were considered for evaluation. A plot of the tearing forces of the Braidwood 1 and Plant A 1 speci-mens is depicted on Figure 3 0 relative to the axial force correlation developed using the reference database. Plant A 1 SGs have 7/8" diameter tubes. The result for that specimen was adjusted using the method discussed in Reference 10.5 to obtain a value commensurate with a 3/4" diameter tube, 1. A visual examination of the data relative to the EPRI database indicates that the tearing forces measured fall well within the scatter band of the reference data. 2. The data points fall near the regression line, hence no statistical anomaly is indicated, i.e., the data are visually remote from the prediction bound. In summary, the visual examination doesn't indicate any significant departures from the reference database. Since the Braidwood 1 and Plant A 1 tearing force data were not indicated to be from a separate population from the reference data, the regression analysis of the tearing force on the common logarithm of the bobbin amplitude was repeated with the additional data included. A comparison of the regression results obtained by including these data in the regression analysis is provided in Table 311. Regression predictions obtained by including these data in the regression analysis are also shown on Figure 3-9. A summary of the changes is as follows:

1) The intercept of the rupture, or tearing, force, F,, as a linear function of the common logarithm of the bobbin amplitude regression line is increased by 0.04%, or about 3 pounds (Ibf). This has the effect of increasing, albeit insignificantly, the predicted tearing force as a function of the bobbin ampli-tude.

2) The absolute slope of the regression line is increased by 0.55%, i.e., the slope is more steep. This has the effect of decreasing the predicted tearing n\\NSD15\\apc\\cce97\\ccec690d wpS

$.14

force, also insignificantly, as a function of bobbin amplitude only for large indications because of the increase in the intercept.

3) There is a decrease in the standard error of the residuals of 8.40%. The effect of this change is reflected in a smaller deviation of the 95% prediction line from the regression line. The overall effect of the changes is readily visible on Figure 3-4.

4) The p value for the slope coefficient decreased from 5.0x10'8 to 3.7x10~'.

The not effect of the changes on the predicted rupture forces is to increase the values by about 100lbf. The increase in the intercept, decrease in the slope, and the decrease in the standard error indicates that the probability of burst would also decrease for bobbin indications over the structural range ofinterest. based on the relatively small change in the strength, about 5% at 10 V, the change in the probabili-ty of burst would also be expected to be small. 3.3.2.1 Conclusions Regarding Axial Tearing The review of the effect of the Braidwood 1 (Plant AA 1) and Pl:nt A 1 data indicates that the correlation of the axial tearing force to the common logarithm of the bobbin amplitude would not be substantially changed by the inclusion of the data. There-fore, it is likely that the conclusions of Reference 10.5 relative to EOC probability of burst based on the use of the reference database would not be significantly changed relative to results obtained from correlations developed after adding the Braidwood 1 and Plant A-1 data to the database. The data continue to indicate that the structural limit for axial separation during a SLB event is greater than 100 volts;in addition, a single indication with a bobbin amplitude of 10 volts would be expected to have a probability of axial separation during a SLB event of < 310-6 o \\NSD15Napc\\cce97%ecec6904.wp3 3-15

. _ = - Table 3-1 Field and Laboratory Eddy Current Data for Braidwood Unit-1 Tube Pull Specimen Field Data - Laboratory Post Pull Laboratory Reevaluation Eddy Current Data R28C24, TSP 3 Bobbin: 0.68V DI Bobbin: 5.3V dent with DI Patnt: 0.09V, MAI in 60" present wide patch Ind +Pomt: MAI in 120 wide patch Ind R28C24, TSP 5 Bobbin: 6.27V,90% deep Ind Bobbin: 17.3V,91% deep Ind Pomt: 5.2V, 90% deep, 0.58" + Point: 6.7V,96% deep MAI long SAI (5.9V and 0.54" long covering entire tube using 115 mil PC coil) circumference, main crack length = 0.75" (0.73" using 80 mil PC coil) R41065, TSP 3 Bobbin: 0.41V DI Bobbin: 0.71V DI +Pomt: 0.13V,63% deep patch +Pomt: 0.09V,54% deep Ind patch Ind R41C65, TSP 5 Bobbin: 8.93V,90% deep lad Bobbin: 13.5V,85% deep Ind 1Egint: 6.5V, 95% deep, 0.73" +Pomt: 5.5V, 98% deep, 0.53" long SAI with IGA long SAI with IGA involvement (only 0.53" long involvement (105" wide) (0.48" in 115 mil PC coil) long using 80 mil PC coil) o.\\NSD15\\apc\\cce97\\ccec690d wp5 3 16 f y e.-

Tebb 3-2 Braidwood Unit-1 Leak Test Datam Specimen Test Type Leak Rate Test Conditions Differential Pressure (liters / hour) P,(psig) P,(psig) T,('F) T,(*F) (psi) R28C24 FS NOC:1295 0.264 2264 970 646 660 TSP 5 FS ITC1:1828 0.000 2340 512 636 642 (5.6V FS ITC2:2284 3.00 2662 378 596 550 Field FS SLB2:2535 5.10 2749 214 584 472 bobbin: ETOSTSPR 184 1300 0 RT RT 17.3V Lab NOC:1300 225 1800 0 RT RT bobbin) ETOSTSPR ITC1: 265 2300 0 RT RT 1800 275 2560 0 RT RT ETOSTSPR ITC2: 298 2900 0 RT RT 2300 191 1300 0 RT RT ETOSTSPR SLB: 226 1800 0 RT RT 2500 258 2300 0 RT RT ETOSTSPR OP: 2000 278 2560 0 RT RT ETTSPR NOC:1300 296 2900 0 RT RT ETTSPR ITC1:1800 ETTSPR ITC2:2300 ETI'SPR SLB:2560 ETTSPR OP:2900 R41C65 FS NOC:1290 0.204 2266 976 612 622 TSP 5 FS ITC1:1791 0.844 2305 514 606 594 (9.IV FS ITC2:2415 4.70 2684 269 588 498 Field FS SLB:2536 7.50 2758 222 583 464 bobbin: 13.5V Lab bobbin) (1) Test data la present in order of testing for each specimen. (2) Leakage at SLB conditions condensed and measured at room temperature. free span (as received tube without covering support plate) FS = NOC normal operating conditions = ITC intermediate test conditions = SLB steam line break = OP = over pressurization ETOSTSPR expanded tube offset tube support plate restricted (tube expanded at 6670 psiinto a = TSP simulant that was offset by 0.15 inch such that the crack tip was located 0.12 inch below the TSP simulant) E'ITSPR = expanded tube support plate restricted (tube support plate placed over original crevice region with entire crack covered) c:\\NSD15Napc\\cce97\\ccec69od wp5 3 17

Ta bl3 3-3 Room Temperature Burst and Tensile Data for Braidwood Unit-1 S/G Tubing Specimen Burst Burst Burst Burst 0.2% Tensile Tensile

Pressure, Ductility, Opening Opening Offset Ultimate Elongation, psig

Length, Width,

.. Yield Strength,'- inches inches

Strength, psi psi R28C24, FS 12,500 36.1 1.930 0.355 59,350 108,870 34.2

' (no corrosion) R28C24, TSP 3 9,640 13.4 1.041 0.244 R28C24 TSP 5* 6,610 10.4 1.133 0.278 R41C65.FS 11,700 30.5 1.644 0.307 60,660 99,080 31.3 (no corrosion) R41C65 TSP 3 10,600 20.3 1.274 0.312 R41C65,' TSP 5* 5,510 9.5 0.949 0.255

• Tested with foils and bladders inserted into the specimen v

4 o.\\NSD h. apcWe9Bccec690d wp5 3-18

Tcbla 3-4. SEM Frectogrcphic D:ta for Int:rgrcnul r Macrocrccka

Specimen, Length or Degrees vs. Depth * &

Positional and Ductile Ligament Data ' location Ligament Location (inches or (Area = inches" x 10 ; Orientation of Ligament Minor Axis d degrees /% throughwall) relative to Macrocrack Major Axis in degrees; Orientation of Ligament Major Axis relative to Tube Radius in degrees **) R28C24. TSP 5 0 0/00 Crack top at 0.750" above TSP 5 bottom edge (Probably near 0.64' when - (OD origin axial crack 0 0v28 Corrected for tensile elongation.) in burst opening at 0 08/80 330") 0 12/33 0.16/63a _ s (0 188/100) Ligament 1 Area = 11. Minor Axis @ 90', Major Axis @ O' 0 20/100 0 21/100 0 28/100 The 0.11 inch elongation from the tensile test was non-uniform and a 0 32/100 o 3sfjoo correction cculd not readily be applied to lengths. It is likely that the 0 40/100 c -.rm.u.it a.rr w given profile measurements overestimate the throughwall length as well as 0 4 t/100 the total crack length. (0 448/100) 0 48/G5 0 52/00 0 56/46 0 60/19 0 64/09 (0 688/00) (LAD = 63%, Crack length = 0 688~, Throughwall for 0 2G(r) Crack bottom located at 0.062 above TSP 5 bottom edge

• Average deptha were calculated by a number of different methods depending upon the need. Methods used are LAD = hnear average depth; ATD = average throughwall depth (length weighted average depth); PDA = percent degraded area (relative to cross sectional area of an undegraded tube).

" Note that the ductile ligamenta between parallel cracks in Table 4 are described by both a major and a minor axis orientation. The ligamenta are usually considerably longer (major axis) than wide (minor axis; The ligament major axis is that going from the OD to the ID of the tube wall (or from the ID to the OD in the case ofID origin cracks) and is ? usually close in orientation to the radius of the tube. The orientation of the major axis is relative to the tube radius. The minor asia of the ligament is the observed orientation l where the ligament jumps from one microcrack to another microcrack as viewed from the OD. The orientation of the minor axis is relative to the tubing major exis. Usually the minor axis is close to perpendicular to the tube major axis. l o \\NSD15\\ ape \\cre97\\ccec690d wp5 3-19

c Tchle 3-4 (Ccntinued). SEM FrcCtogrcphic Data for Intergracciar Mccrocracka

Specimen, Length vs. Depth &

Positional and Ductile Ligament Data location Ligament Location (Area = inches

• x lod; Orientation of Ligament Minor Axis (degrees /% throughwall) relative to Macrocrack Major Axis in degrees; Orientation of Ligament Major Axis relative to Tube Radius in degrees) tL28C24 TSP 5 192/9 Crack i tip krated at 192 degrees near center of 15P crence regen (OD origin crack in 201/32 tensile torn 214/12 circumferential FF 226/19 made in the burst 236/14 specimen.

248/30 circumferential crack 258/22 cuts through the axial 270/2G (Crack I = Major blacrocrack) burst opening) 280f14 292/05 302,99 314/3G 324/37 (330/100). Assal Bwet FF tocatma 336/37 ' (342/00) Crack I tip located at 342 degreca near center of TSP crerie reswn (344/00) Crack 2 tip located at 344 degrees near center of TSP crence region (315!32) 347/37 (Crack 2 = Illinor Macrocrack) (348/32) (319/00) Crack 2 tip located at 349 degrees near center of TSP crence region (352/00) Crack 3 tip located at 352 degrees near center of TSP crevice region (355/?6) (C ack 3 = Minor blacrocrack) (357/00) Crack 3 tip located at 357 degrees near center of TSP crence region 358/00 (361/00) Ciack 4 tip located at 361 degrees near center of TSP crevice region (3G3/32) ' (Crack 4 = blinor Macrocrack) (364!00) Crack 4 tip located at 364 degrees near center of TSP crence region 36900 (374/00) Crack 5 tip located at 374 degreca near center of TSP crence region (377/32) (Crack 5 = Minor Macrocrack) (380.90) Crack 5 tip krated at 380 degrees near center of TSP crence region (PDA = 10%, IAD for Macrocrack I = 21% over 150". Overalllength of Macrocracks No ductile lignmenta other than for the separation between aim.cks. = 188") a:\\NSD15\\npe\\rce97screc690d.wp5 3-20

Tchle 3-4 (Continu:d). SEM Frcctogrcphic Dcta for Int:rgrcnnlar Macrocracks

Specimen, Length or Degrees vs. Depth

• Positional and Ductile Ligament Data location

& Ligament Location (inches (Area = inches

• x 10 ; Orientation of Ligament Minor Axis 4

or degaW4 throt.ghwall) relative to Macrocrach Major Axis in degrees; Orientation of-Ligament Major Axis relative to Tube Radius in degrees **) RelC65.A W 5 00W Crack 1 bottom tip at 0.012" above TSPS bottom edge (OD origin niial crack 0 01/14 (Crack 1 = Minor Macrocrack) "t OP" sing at (0 0 Crack I top tip at 0.068" above TSP 5 bottom edge - Crack 2 bottom tip at 0.086 above TSP 5 bottom edge o ogfi, <o osco) (Crack 2 = Minor Microcrack) (0 090m) Crack 2 top tip at 0.098' above TSP 5 bottom edge 0.12/67 Crack 3 bottom tip at 0.102" above TSP 5 bottom edge 0.16/84 (O.I72/100) 0.20/100 The 0.11 inch elongation from the tensile test was non-uniform and a 0.24/100 correction could not readily be applied to lengths. It is likely that the given 0.28/100 profile measurementa overestimate the throughwall length as well as the 0.32/100 total crack length. 0.36/100 0.10/100_ Circumferential Tensile FF laation 0.44/100 (Crack 3 = Major Macrocrack) 0.48/65 0.52/46 0.56/40 tacament 1 0.60/42 Ligament 1 Area = 17. Minor Axis @ 90', Major Axis @ 20' O.64/19 0.68/14 (0.686/00) (LAD = Gl%. Overall length of Macrocracks = 0.686". Throughwall Crack 3 top tip located at 0.698~ above TSPS bottom edge (Probably near f r R2687 0.588' when corrected for tensile elongation.) oMSD15%pcWe9Bccec690d wps 3-21

i . Tchle 3-4 (Continued). SEM Fractogrcphic Data for Intergrcnr.lar Macrocracks '

Specimen, Length vs. Depth &

Positional and Ductile Ligament Data location Ligament Location (Area = inches

• x le-*; Orientation of Ligament Minor Amis 4

(incheshe throughwall) relative to Macrocrack Major Axis in degrees; Orientation of - Ligament Major Axis relative to Tube Radius in degrees) R41065, w r5 19640 Crack tap located at 198 degrees near centen of BP crevue region (OD erigin crack in 203/20 tensde torn 208/35 circumferential FF 214/46 made in the burst-220/35 - specimen. 225/100. Amt B-t rrimetme circumferential crack 230/26 No ducide ligaments. cuts through the axial 236/22 burst opening) 242/19 247/17 252/19 258/17 264/15 269/19 274/28 280/13 (28&M) (PDA = 7%. LAD = 29% over 90*, Crack Crack tip located at 288 degrees near center of TSP crevice region length = 90*) . y ? o o-\\NSD15\\ ape \\cce97\\ccer690d wp5 3-22 f e ~ it e e

• r-

-w-- r +

1 + Table 3-5 -i i Metallographic Data from Braidwood Unit-1 Steam Generator Tubing l Estimated Contribution of Typecal ] Specimen Section. Number Section Cracks Maximum Max / Avg, ICC-D/W Ratio j-Location Type of Length per Number of Depth . Components from Cracks (inch) Inch Cracks

• mt TSP (% Throughwall)

Based on' Transverse at mid-Mid-Crevice Radial ' Section + crevice Location (based Metallography 8a .i metallorraohv) R28C24,- Transverse 26 2.0 13 ~50 (spread over 49/30 ICC < 21% deep 10. TSP 5 Radial 1 18 0.45-40* entire depth = 2% Axial IGSCC >

R41C65, Transverse 22 2.0 11

~50 (most in a 130" 43/24 ICG 5 32% deep 7 TSP 5 Radial 1 27 0.33 82* patch) depth = 2% Axial IGSCC > l Radial 2 - 24 0.33 72 depth = 12% . 49% deep

• Cracks must be greater than 5% deep to be counted or estimated.

l } i i t i I I i h oANSD15\\apc\\cce97%ccec690d wp5 3-23 [ i l I ~

J Table 3-6 Summary of Braidwood-1 Pulled Tube Eddy Current Results Field Call Lab, Reevaluation of Post Pull Data Tube T Field Data S Bobbin + Point Bobbin Depth + Point Bobbin + Point p Volts"' Volts Volts"' Voltu Volts Volts Steam Generator A R28C24 1 NDD NDD NDD NDD NDD NDD 3 0.70 NDD 0.68 D1 0.09 DI dent 4.8 5 0.08 7.9 6.27 90 % 5.25 17.3 6.7 R41C05 1 NDD NDD NDD NDD NDD NDD 3 NDD MDD 0.41 DI 0.13 0,71 0.09 5 8.93 0.07 8.93 90% 0.5 13.5 5.5 Notes: l 1, Bobbin voltage date include cross calibration of ASME standard to the reference laboratory standard (set up at 2.25 volts for 20% ASME holes, Calibration Standard ADVB 009 96). J o:\\NSD15 Nape \\cce97%ccec690dmy5 3-24

Table 3-7 Braidwood-1 Pulled Tube Dr.a for ARC Applications T Bobbin Data Destructive Exam Results Leek Rate-l/hr Beret Pu - m Data -het Tube S RPC j P Volts Man.. Avg. Crack TW No. N. O. SLB Meas. Meas. .e,

e.

.A4 I Volts Depth Depth Depth ' Isngth Length Lig." 1300 2500 Burst Tensile. Burst poid** paid" Press. Force - Press." (kei). .(Ib) Steam Generator A R28C24 I NDD NDD Not Examined i 3 0 68 DI O 09 Not Examined 9 640 ' .-8103, 5 6 27 90% 5 25 100% 63% 0 68T 0 26" 1 0 19 4 41 ~ 6 610 - 7700-

5624, IRB" IRB*-

80.4 166 8 FS 12.500 10640 59.35 108 87 10 636. R41C65 - 1 NDD NDD Not Examined 3 ' O 41 DI O 13 Not Examined 10 600 9.498 5 8 93 90% 6 50 - 100% 61% 0 698* O 268* 1 0.18 6 34 5 510 7400 4.937 100% O59G" 0.268* FS 11.700 9800 60 66 99.08 10.484 Notes: [

1. Measured leak rates adjusted to reference conditions by applying methods of EPRI data report. Reference 10 5.

2. ' Number of uh. ki!lignments with > 50% ofligament length remaining in burst crack face.

3 Excludes 2 short microcracks of 14% depth at the end of the macrocrack.

4. Measured burst pressure adjusted to nominal, hot flow stress of 7157 kai for 3/4" diameter tubing.

5. Indication Restrained from Burst (IRB) leak test followmg pressurization to 6670 psi inside TSP with 0.15' TSP offset. No difference in leak ratee between crack totally inside TSP -

and TSP offset >0.15" (0.12* of crack exposure outside TSP). 4 I oANSD15\\apc\\cce97\\ccec690d wp5 3-25

Table 3-8: Effect of Braidwood-1 Data on the Burst Pressure vs. Bobbin Amplitude Correlation P, = a + a log (Volts) o i Reference Database with New / Old Parameter Database Value Braidwood 1 Ratio a 7.4234 7.4282 1.0006 o ai 2.9220 2.9121 0.9966 r' 82.26 % 82.27% 1.0001 o,,, 0.8685 0.8612 0.9916 r N (data pairs) 91 93 p Value for a2 3.5 10'35 6.1 10 O.1734 Reference o 71.565 ksi r Table 3-9: Effect of braidwood-1 Data on the Probability of Leak Correlation Pr(Leak) = 1 + e d

• 6* "

Reference Database with New / Old Parameter Database Braidwood 1 Ratio b 5.1721 -5.2246 1.0101 i b 8.6705 8.8034 1.0153 e ~~~ V "' 1.4700 1.4990 1.0198 ii V 2.0983 2.1391 1.0194 i2 ~~ ~ ~ V 3.3666 3.4198 1.0158 22 DoP2' 118 120 Deviance 41.35 41.75 1.0097 Pearson SD 1.112 1.124 1.0111 Notes: (1) Parameters V are elements of the covariance matrix of the u coefficients, b,, of the regression equation. (2) Degrees of freedom. o:\\NSD15\\ ape \\cce97%ccec690d wp5 3 26

Tabla 3-10: Eff;ct of Braidwood-1 Data on tho Leak Rate vs. Bobbin Amplitude Correlation log (Q ) = b, + b log (Volts) Reference Database with New / Old Parameter Database Value Braidwood 1 Ratio b 2.1129 2.1091 0.9982 3 b. 3.3162 3.3094 0.9979 r* 53.1% 53.1% 1.0001 o,,,, ( b ) 0.7884 0.7720 0.9793 r N (data pairs) 46 48 p Value for b, 9.6 10~" 4.4 10'" 0.4585 Table 3-11: Effect of Braidwood-1 Data on the Axial Rupture Force vs. Bobbin Amplitude Correlation F, = go + gi log (Volts) Reference Database with New / Old P ameter Database Value Braidwood 1 Ratio go 7.9293 7.C? 59 0.9996 gi 1.2011 1.2005 1.0045 r* 67.7% 70.4% 1.0407 c,,,, 0.7632 0.6991 0.9160 e N (data pairs) 17 20 p Value for gi 5.0 10 3.7 10* 0.0744 Reference o 75.0 ksi r i oANSD15NapcNece97%cccc690d wp5 3 27

49 + 4 .a w 4m a Figure 3-1. IRB Pressurized Tube Leak Rate Tests of Tube R28C24, TSP 5. 400 350 -*-Centered TSP -e Offset TSP lI c.c 9 } N E 250 3 E. ? "a 5 .3 200 Ii 4> + 150 100 1300 1400- 1500 1600 1700 1800 1900 2000 2100 2200 2300 2400 2500 '2600 2700 2800 2900 I Delta Pressure (psi) 3-28

Figure 3-2 Sketch of the OD crack distribution found at the fifth tube support (TSP 5) region of Tube R28C24. 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. 1.75-1 I I El 1.25- -SP Top Vb>',Y,DI). -6 y~

< 5 'A e

, a ^ p, \\

d' w' ' .a vAx:. > b' <,.i ' q w ? ]: __.:';.,9 c,c ',s" *' t.j, ! ',, ,,. r p i.y f 5, I a y ~x d',- s e ?.< - );. p uhs > 3, <(p, 2 a ', _p - S -0.50 - ) -SP llattom ill H.00-g g i n' 90-I80* 270-360* Circumferential Position (degrees)

c. ~ n m a.,4-w s.m, 3-29

Figure 3-3 Sketch of the OD crack distribution found at the fifth tube support (TSP 5) region of Tube R41C65. ~ ' 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. 1.75-1 i 1 Il 1.25- - SP Top -8 .tf g c -t J pw e s g 1 -~,-g

• g u

y l7;i,,p) .h ') ,').- s n 0.50 - -SP Bottom ill (1.()(1-g i i ~ 11' 9 (l' I 8(l* 27(l* ' 3 6(l* Circiamierential Position (tlegrees) ew m .,s 4 im 3-30

Figure 3-4 Example of transverse metallography obtained from the mid-crevice region of TSP 5 of Tu.e R28C24 following both burst testing and tensile fracturing of the burst specimen. A typical axial intergranular stress corrosion cracking (IGSCC) structure is observed in the transverse section which cuts through a region with some intergranular cellular corrosion (ICC). 100X l Tub t Y ~ l <l I ~ 4 \\^ g h ~ r [ ~ 4 e V s n A, _ e.~new s.i,5-w 3. im 3-31 i I

i i i 1 j Figure 3-5 Example of radial metallography (21% deep) obtained from the mid-crevice region of TSP 5 of j i Tube R41C65 following both burst testing and tensile fracturing of the burst specimen. A j j typical intergranular cellular corrosion (ICC) structure is observed. IGX i 1 i m : z'.n a t.

., f'v-ic.pfy'

1 e' x M.')!+ at; }.<.n !{.'.cf:m. g.f A-h:-

r- ~

• i " -

~ N: ) is,. y } -4". Q.:h..-... :y.c.r . y. z w e,. s. m 1 f. 'h:. n44,;dy@i/ sk,::gf 4-1 t-i,-T -i S ? ** g \\ s N 1 s.- t Wh ?; A[ i i I ['I...' I.d*M.;. -hDb, 5, ,= a 3f I ~~ ), i. n r h+ 'j m... v$.-:., r p,.t . ~ - e

r.

[ 's V[ yi f, .T f l k j[.is. t s. ' [n,,. f, i.,3 i "is i.'; i-j ^ { # 4*( d -i 't s <}d t.; (1-1 3;o. g g?... \\..., T M.4,.f. 1.3.. 4 z &. g( >,.g t ,. a r .a (...e t. '.s ... j *. */-. a -a ' ' :p : . h, L- ! Axial Fracture Face -> . nf' d-e .. M. j.., y .t .1. i s p.- r-c. g 3 x 'T 1 l (.t kl- 'k { ?

h._.

'.+ (f s -

• 0:.gQ V.p

..! y.. -.. wr -?

1 e V v 14 !.ECs4-.

t

• l t b t y: & u.*h.,!L a.i.5 :4 ?f,
• - 1 1

ys 1? t.' t 1L.s ',2 ~~

4/p'f4d.

l F. ,,.J. y - P y i' r - i T. %. b l4 -; -- >D ~ } .". }. J.~ r ', ti. W -k Q f &g, x.. [q l y.r..m +> o <f.., :s; a -5.,--j ) e.... ..u rs-3 M.. I,%

• Db i.

' ' j-

• '=

y a.. Q> t Bottom Tensile F: acture Face e-7 m.,s-w u* 3-32 k>-

o N E g .0 G t-. 7 .t E e. 5 'l7 m e1 1 I l N b O 3 Q 'g.[.1 j' it t .e ca E .-1 Y. m E-' 8 R @ e Ee e J j o ? '@ A f 8 Y 4 o A C io e 3 / O O l BA G to c ? x + 3 >mg U T cA o D / 3 Q tal cn 4 Z Z o 4 .t @e O I o i I g I D X +E l "a* m ....n.. ce i o ! cr.. i s. __e g.o il g r Ah -8"~_;'$~ i ~ tsr -c o ,c p _0_ o 4 3 D 4 l l i D~0~ / ~ D' D"~ i a- - tsa j y oZ en = F. CQ v d N c l I U i 4_ -e m m 'O r u ,4.i j in >g a, a. p w g, 6 i 4 i <A t E a a 3 o / i mO D_ _ cza _..._ o _; i -_e.. c gM g Og f' l {__ g u t 4< a + at 3 o m 4 a a 4i i a /! l mo a '.c r o 9 o 2 (D gQ g-co - -/ -i 1- ~ _O__ _Q._ -4 / l_ ] q_o _o.o-o- -o -/ J--h = .__ _. 4 _L__:_ M c / b-- l } g-OL....p / i .% O i ,4 _3 _a gg D 4 I ? I' m m n -.- Q --- f xx -+ / N o 4 to D o m ( / n N 4 l r i a i i y a n o 9 o o o. o 9 N o m o w a o g z, (!sg) a.mssoad asang a

Figure 3-7: Probability of Leak for 3/4" SG Tubes @ 650 F, AP = 2560 psi ' Comparison of New Data with NRC/EPRI Reference Database ye 1.0 = = a-occ. g,g __ NRC/EPRI Database /!/ l o 'l Update Data _ _ _. _f ]_ 0.8 - --- Log Logistic Fit I I Reference Curve and i m Log Logistic w/ New Data _/ }.Ml Updated curve [ a I. Overlap m 0.7 -

---

90% Conf w/ New Data i

\\! be I .5 l a" 0.G I Q I l I xa j 0.5 [ y f. E D 0.4 l e g C I j 0.3 w I m / I 0.2 / 0.1 l,s _'__ mn- - e~i'T- - - - - - cumooc 0.0 0.01 0.1 1.0 10.0 100.0 Bobbin Amplitude (Volts) 3-34 anc amo7. ti2i m p4pv_797j potu

i-r 8 g { l 1 I + ~ , ___4 C3 ~ l a k. s.____1 e s j t@ d b 5 O 4 .i.T. _4 i L es .a,,,, C) it a g I g U c3 a I i c C) es M". c> y O ?a i I ad e o m x 30 I c .o Z C i c3 c hd Cf u (, e g A $O ] i o ub a W y A N+ D a G so C oN d a 3O' + _h,N_4. _. ____+ 5 W c) c) C to ]s T 4, i b 4 ? I

• =4

,1 \\ + l D + l +

• Cq a

i i, a\\ I i i i 2" 9[1Rvl q!c' h-j E q,, 4 <l \\ s!!! ' i I - ? ll -\\a-i n I;l o ! !i r o' ,i i i i c 'i s a , Y ;~ i C jj ,I l j

ij i,;

g 'o oN l ;t ; a 'i. I,iii i e ii; 1 i !l' - ', I r! D ,e o m :\\ i 't;i ! h4 l -ll p. ijp i o e j! I CQ

• l de h l.I M ii 4 lQ':!

e I i8j !! l i

l l

j @* ii; h t, 1 ,I !i i, 3 ,a a >o wiii- .in n ii i B, 1 - _1 4n, t 7 ~ yi wi!, o.

i p'l

: ;! !

o_ g .,4m o4 . L'ii ! l a.,j ; l i_ _! l hi 1!O sp: B !Ili - o e + u a tg m jl'. mln.Q, i x .l+h ! : NP jl' i ' i i l lil i i lj..h t i .e l i. i p; _! m g a_ . l. O I

i i

n !; t f i! e ! !lqH'y.x-.po +i m __ . l' i i i o ,i e i ,M v) 1 w, a g!!! g. 8o i' ! lli i !ll I I \\' 4.yl l: i, i ! l ut, : i i i,s bul; l nlI ! ! i i !! Io ! CQ i i .: i

L o o

%%om+K ,n ,7-

u 4e v

o N,1t i

r. in i ?.

1 ilsi, e i : i i! l l: 4 1 3 CQ @ n.1In _'!_g:; _' ',.? 's i N QU al L

o a_

.g, , g;. g w _s %, g jli 1i m< Li i a

s. so't i

pl,oiI -l\\ fN +! ! I %h p! ! ' i cc v t ii ; i'? l

p'4,:

- '*p'T, l \\ tN;, l i N; .E,, s t o i: h;; l L P.' hh ; i, + [ii.!i ! hi wd I i u: i 4,. ...h. b,!

bi bo!

5, =e M !i I _ h. _! i 4 gm, 1 %',. y _ 1 m.. N-j ; I ;- ' !jj .h: . l i ;- ! e i 1. i' i 1 i -* \\

s CQ i!

) 's !.i

< ;si i (

j i i 1 F. i;

p - ;

H! t x y.e p is i l, 1: ; I i i ',i!, s i i, i i e f li'; N:,I'[\\ w>; ; b,;, e,. m.. m.,,

o. i.

g l 8' R' e 6 a e o E 5 8 o ~ 1 3 (RUD18d 099310aacu quel ggs d

t r[t. !f;i l ) lf l 'Ir!'i;, l;}; s 0 a p 0 0 m {,> ,] l - 1 i l%l -{ye 9 L t r. i m I-I Ull-sI g l r g a i1. l l m Ll i N[l im p l J ~ n ~d l s k

tl e

l 5 l ~ I I -y d 7 u= ~ ~- t

, tl

!l I i-a' i i r lpS ~* .c o m s, ~ - e + Ab u 0 n ~ w I-1 iT 0 b A i'O -Mfi1 Jy , ill } Il 1:{ l.l' ) boM I lL s t l B0 Ii i o ,I V 0 ( s6 'd i tI .l .i}i v e y fl [I d y' 81i' i: eo u l cl t oA f p i r 14 l{ - !4 l 6 m FG 3-A 3 eS i i-n r" l lLi 4 ~ i u3 - ~ b ~ F~ b t 4 p0 o ~' oB u0 Rx 0 Lfm. l l )4l' 1 l " ~ a/4 'j] i !' } ix3 riwl f )" Ao I !1{' 8 I-Pkm / t !q .!A 7 L l,l ) e T 4 9d - ll a/j n L i-I4 iJl 3 d e a (a (A LL@ I i-t i 3 t -'f e s n nT Dt f l.l eu a a.i oLd rj taLi@P e e f 8 ud U t a 7 ! i t r c / D ni r ii c I gA .!;a od d o e s b d i7 ( 7 9 s e e %F i / i a Ft u t / s r r 5 p T s ,7 ePP 9 a lL - u1 r O jA1 e % % f.eN al "4 g D d 5 5 / R9 9 R3 3 AAA r' m r o ~_ 0 a * + 4 r ~_ viv ~ ~ 1 e 0 0 0 0 0 g 0 0 0 g O 0 i. m. 0 9 8 7 6 5 4 3 g, L 0 sr 1 7-s e CSU ago ucY2sg 2 c g 4::. I , ii' 1- ,Il; J i'

4.0 EOC 8 INSPECTION RESULTS AND VOl/PAGE GROWTil RATES 4.1 EOC 8 Inspection Hesults According to the IPC guidance provided by the NRC Generic Letter 95 05 (Reference 10.3), the end of Cycle 6 inspection of the Braidwood Unit-1 SGs consisted of a complete,100% EC bobbin probe full length examination of the tube bundles in all four SGs. A 0.010 inch diameter bobbin probe was used for all hot and cold leg TSPs where IPC was applied. Subsequently, RPC examination was performed for a minimum of 20 percent of the hot leg indications with an amplitude between 1 and 3 volts, all hot leg indications with an amplitude 3 volts and above, and all cold leg indications greater than 1 volt. One hundred and seven f r.dications were found above 3 volts, of which 30 were found in SG C, and they were all confirmed by RPC and repaired. A total of 30 cold leg indications were found, and 2 of those were above 1 volt; they were both confirmed by RPC inspection and repaired. In the rest of the report, hot leg side indications which are restricted from bursting because of TSP locking are callad as locked tube model indications, and indications on the cold leg side which could be completely exposed because of TSP displacement are called free span model indications. An augmented RPC inspection was also performed consistent with the NRC requirements. All dented intersections with a bobbin voltage greater than 5 volts and a mmimum of 20 percent of the dented intersections with a bobbin voltage between 2.5 and 5 volts were inspected with RPC. The augmented RPC inspection also included one TSP intersection with a mixed residual artifact signal in SG C, but it was not confirmed. A single circumferential indication extending 123' was detected in a large dent at TSP 511 in tube ROC 2 of SG A. There were no other RPC circumferential indications at the TSPs, no indications extending outside the TSPs, no RPC indications with potential PWSCC phase angles, no other flaw indications on dents at any dent voltage and there was no signal interference from coppec deposits. A summary of EC data for TSP indications in all four steam generators is shown in Table 41. The table shows the number of field bobbin indications, the number of these field bobbin indications that were RPC inspected, the number of RPC confirmed indications, and the number of repaired indications. The indications that remain active fer Cycle 7 operation is the difference between the observed and the repaired. RPC voltages shown in Table 41 are obtained from 0.080" pancake coil probes. The o \\NSD15\\npe\\ne97\\nec690d wp5 41

following is an overall summary of the EC data for all four steam generators of Braidwood Unit 1. Out of a total of 6784 indications identified during the inspection, a total of 6500 indications were returned to service for Cycle 7. A total of 650 indications were RPC inspected including two free span model (i.e., cold leg) indications. A total of 501 were RPC confirmed including two free span modelindications. Consistent with the 3 volt IPC, o *otal of 278 indications were removed from service. RPC confirmed locked tube model indications (hot leg) with a bobbin amplitude ofless than or equal 3.0 volts and RPC confirmed free span model (cold leg side indications for which 1 volt IPC apply are called as free span model indications) indications less than or equal to 1 volt are not removed from service unless they are at intersections excluded from IPC, or in tubes that have other indications that do not meet the IPC limit, or in tubes removed from service because of other reasons. A review of Table 41 indicates that SG C had more and higher BOC 7 indications (a quantity of 2015, with GG1 indications above 1.0 volt) than SGs A, B or D; thereby, it potentially will be the limiting SG at EOC 7. SG C also had the largest indication (10.48 volts) found in the EOC G inspection. Figures 41(a) through 41(c) show the actual bobbin voltage distribution determined from the EOC 6 EC inspection: Figures 4 2(a) and 4.2(b) show the population distribution of those EOC 6 indications which were repaired and taken out of service; Figures 4 3(a) and 4 3(b) show the distribution for allindications returned to service for Cycle 7, and Figures 4 4(a) and 4 4(b) show the distribution for all RPC confirmed and not inspected indications returned to service for Cycle 7. The indications removed from service, a total of 278, include 109 RPC confirmed indications above the IPC repair limits. The terr.aining repaired indications were in tube segments near the wedge locations where IPC does not apply, or in tubes plugged for degradation mechanisms other than ODSCC at TSPs. The distribution of EOC G indications as a function of support plate location is summarized in Table 4 2 and plotted in Figure 4 5 The data show a strong predisposition of ODSCC to occur in the first few hot leg TSPs (6273 out of G784 indications occurred at the first three hot leg TSP intersections), although the mechanism extended to higher TSPs. Five indications were found at the FDB (1H), o \\NSD15\\a; Wee 97\\crecGMt wpS 42

one each in SGs A and B, and three in SG C; all these five indications were repaired. Only 31 indications were detected on the cold leg side. This distribution indicates the predominant temperature dependence of ODSCC at Braidwood Unit.1, similar to that . observed at other plants. 4.2 Probe Wear Criterin An alternate probe wear criteria discussed in Reference 10.8 was applied during the EOC 6 inspection. When a probe does not pass the 15% wear limit, this alternate criteria requires that all tubes with indications above 75% of the repair limit since the last successful probe wear check be reinspected with a good probe. Although the repair limit for hot leg indications is 3 volts, all indications for which the worn probe voltage was above 0.75 volts were inspected with a new probe. An evaluation of worn probe and new probe data is presented in the following paragraphs. In accordance with the guidance provided in Reference 10.8, voltages measured with a worn probe and a new probe at the same location were analyzed to ensure that the voltages measured with worn probes are within 75% of the new probe voltages. No new indications were detected during retesting with new probes; thus, worn probes did not miss any indications. Figures 4-6 and 4 7 show plots of the worn probe voltages versus the new probe voltages for all four SGs. These figures show a consistent relationship between the two voltages for all four SGs. Composite data from all four SGs are plotted in Figure 4-8. Also shown in Figure 4 8 as a solid line is a linear regression for the data, dashed lines representing tolerance limits that bound 90% of the population at 95% confidence, and chained lines representing 25% band for the new probe voltages. The mean regression line shows a slight bias for the average new probe voltages to be higher than the worn probe voltages; however, at the 3 volt repair limit the difference is within 0.1 volts, which is insignificant. All data fall within the band formed by the chained lines representing 75% of the new probe voltage except for a few indications around 1 volt. However, the absolute magnitude of the differences for indications around 1 volt is small and therefore is not significant. There are no occurrences for which a worn probe was less that. 2.25 volta and the nev probe voltage exceeded the plugging limit,i.e., no pluggable tubes were missed due to probe wear considerations. The largest departure from the 90% tolerance band is noted for two indications (R10C48 05H and R16C46-03H both in SG C) for which the worn probe voltage exceeds the new probe voltage by more than 0.5 volts (the actual differences are 0.77 and 0.83 volts). However, data for these two indications are within the 75% band and therefore they are acceptuble, o \\NSD15\\ ape \\cce97\\ccec690d wp$ 43

Overall, it is concluded J.at the criteria to retest tubes with worn probe voltages above 75% of the repair limit is adequate. Of the 688 indications for which voltages were measured with both a worn probe and a new probe (all but one were locked tube model indications), no indication had its voltage increase from below 75% repair limit (2.25 volts) to above the 3 volt repair limit when retested with a new probe. Only two indications had their voltage increase from below 0.75 volt to either equal to or above 1 volt (0.67 volt to 1 volt in one case and 0.65 volt to 1.11 volts for the other). Both of these indications are locked tube model indications for which the repair limit is 3 volts. Thus, the alternate probe wear criteria used in the EOC 6 inspection is consistent with the NRC n.. dance provided in Reference 10.8. 4.3 NDE Uncertainties The NDE uncertainties applied for the Cycle 6 voltage distributions in the Monte Carlo analyses for leak rate and burst probabilities are the same as those previously reported in the Braidwood Unit 1 IPC report of Reference 10.1 and NRC GL 95 05 (Reference 10.3). They are presented in Table 4 3 as well as graphically illustrated in Figure 4-9. The probe wear uncertainty has a standard deviation of 7.0 % about a mean of zero and has a cutoff at 15 % based on implementaticn of the probe wear standard. The analyst variability uncertainty has a standard deviation of 10.3% about a racan of zero with no cutoff. These NDE uncertainty distributions are included in the Monte Carlo analyses used to project the EOC 6 voltage distributions. i c:\\NSD15\\ ape \\cce97\\ccec690d wpS 44 .1

. hh k h h h h h s!~!s h *!*!n! }!:I:! !c!-!~! a......!=!. I I i ti llilMillll ll!d t' I g e n s!e! !s W Wl 4 W e =~ W U '. - 1.J.4 t.ilill lil l t!M. v Ill. i s I I.!..~l-I -I I I,....!.1,.1,..I,. ll I i l I !l l I 'i 8 .il.l..l.li.l.l .eini.!!.- I l .i--.-..l.l.ill.-lcc I! I! l mi 4 - -~n. -..~. !o l i~ i-m i { !il !!ill lillll!!! ll Il I . u. a...! L l.i ll y .i.l..

i. -

i ! !.~i...l.l.l.1 it s-

~ +

l ll11 ll!! I!i;ill lIl 1 illllilll l i l. ll l!!!ii C -ll .l-l4'EEjElE!El1jBlElEySgl;jujalel=ja!=;-l~f-- a~lyl- . * ;~ -a l ,1.,i,1 i 1 2 g o m:n -i=:.i ! :!li i!!- - :. :.i....l.'.;..!..i... !ili l L jililli. 3 .,n a;E y s:s: i i . t i +. - i i i 1 9 e >li i j li t ,l g! !ig i i . i t 1 i i ! j !I i i [ j! e-n!njEyn!*l sis l=; min;e-l=l.li-!,l-l-l-:--jod.l.Hl.H.H... p $" $ 'E ~.l ! i ; l !.. ] li!!!! l .. -.or-m * --c.-i.:...l.i.!.!.!: ; i oi l !oi...'.4...i. i

,- - i.

4 r 'i l it!!i I 1lll:li!!i.!ij i ll 1 g i <a y .....!.!.!.!.!.;.!,....l.:.t.ir:l l l,.. ~,1 ~ - - -..,.i.!.I i ! t 1,' i I:

.1...

lll4 il!lI i l 4 Qg w ) y}'..,. -l.l-l-l.;. l.'....i l ; !.;.;l !.l-l- '- - !.!.!.l l.l !, ! !.i... (g Rm . ~,. ~; sl l I lii l i i l . = R t E'3:2:2 $~2 e m'nir =I: # - n'*!*!- ~ ~ - -..;.j~!. :- - ~..... i. i. .*Nb

!b E I % E'n n N

E'N. ~-*.*-m9'of..:..'..t...

l w *.1 t'

j 4, i ijg.en=ae *

e,,angnnnyn;=

.;.......H,;.,....s.l.;...

.. - m -,-.. -... - -,.,- s - - -.... -......i,.! ! ! ! I'

!,l4

.--*-..;~..-- .l l i i ii i I i j i i

ii

; i!

.. - -. -,, e, n n n e: - :.= -..!.... - -..., ! I f-f . l1 I, i <,i i I y .- ~.., ~.. - - J. w i,j i !lI l i i 4 ) w { ~ N n -jwi.!nt-!n:M R Q n!3, ;2;=: =2'ai.... e - w w w - mja! i~ i. ~. - - f ! ll l .i$ i'I i;; i ; l i 2 y e i ll I!kll! ll i !!,If! 'I ll l!

• e%E!EiEiE!!!EiE S AM*!E!ANA A Aicsa.. --**'ial-lim ll; 3

l li-lll!' i lili? il !l!

i i

1 n - _. s.; w. w. s.; :- _._:, u.:-.. y;s_ u,, _.,-.:. _ p, m, w. w. w... 1-,- # ;, ys~ 7i s 4 = "-"NW'rmyvw ire-r m wwi'" -s-p. y cw 99-p7g-- -g+- ,r,,,,-pg ,-+gngWT'rW-h:"--"r-8""'WWT v mu' M1"W-t =W" r-t '"4v =wrf+vrm -r'T-r'WP' WT' tw4M'FF"miaT"TP' v FW'wer-P r-"'eWw*'**"FWN'-**"F*

II ooo!o oooooo o! ! [ 7. l o!aoooo e)$7I ooo o c o o o o o o oo i jo olo i i i o..i..!. l I i o oI. oo o, ol. o,.. oj o. e o.i o, o o o o!. 1.1. oo!, o, o r. a a _g i l ii ll li i li ! i li i s, o U t -,. o < o, _ _I, a' o,.1..l _1 I l I!ll l I i l lI i i

I

_ioo...... 3,, o N _.,_,ot. I'll j !! j j _!_!o;,.:o;i : _i_.;.,.:_a,...Jokji;l_:;_!i.1_: o; o, _; _: _l.; o! o!. _ g g i : i i 4 i iis : ii: i _: ; o v y j j lli!!lIIIIii: I!iliII!!IiI'ii!IIIi' I ll, -, oi N, _l! l l ; - oi _i - o, _l, o, o, _i _i _f' oi si o, !il li i i li -:_!..l i _._i .;l,_i i,o.. i y n -. s o; oi o l ll !Iiliiilll' lll IiI ) l[ $ E 8. --fa--l*'-- - -'a

• ~

~-* i ,il3 Ii Ii 4 ii j yi y.'.> o!.! o!.!...;.i.'...u ey.! oy o; o ou o:.u o!.y.! o. o o n o + < - i w .Ji 1 o;.:.:.. o o; o; o,. o.:. o. o; o; ol. o, c o;.;. oi.l. o;.j o o; o;.; o. o, on j gh, i .l of. o! o! oi o! o! o o o e: _ o: o oi o o el e o.o. o! o oi o= oi o o o, o o, o o I i i ll!' i; !!4.iii, ~;, g

1. g

{ E = 4 i ' i il t i t 4 i i r i o'.t o ol ol' o! o! ol o o o o! _ ol o o' o! ol! j ol of o$ o< o! oi o o3 ef o o' o$ o, o o o l 9 { U M * .== i I 4 4 i. m w i; ie . - i et ii ! l l I!. U {ll, l DE .i '. - o o':.I i ) i o < o.o oro;o;o o.oi. o o o o.e o olo o o 2 M * . o o o oi o ce o: o' o or j k f, .' e o o o.o o o o o. o _ o o .,o o o-o-o o o e'e o o.o o o o o o o o [ E

• T

)lf N 4 oc o,o o e,o,e o,o,o e o o o o-o o o o.o o o o oso o oro o o o e o'o o y U o ( 6 w',Ij j< j j 1 I i E M i{ ot o o o o oj o; o'. O o o o o e-or o o c ot o o e o-o o e o o-o o o' o o o f & o . I g - o e o _'-oa:o _<o o o o+o o o o-o._ o _ o o o w e o.

o. o o, *. o N j $ g f

1 i i k m'a o - _[o'o c._ o N W o o;_;olo e o ojo o ojo:_e o _[o o o { { R _ o o o _ ot I w -t I ! l 3 { i ei c' _I _ o' oj oi - o-N _+ o o o ; c:oloj - o:o e o,o-o e o o _ ot -ooo f k E N fi' { i f i i i ' ! i i ! i =Io e o =.o;o}oj*= o e o o o olo+o o;_;e - o o o f E oi o -. o o! o! _ ei N N i

ii t

- 4 i

i!

e _f g'$ ${ $ $ g = = N, ! m,,, m, s;, t

• i e,s,2 e, _iN em g m.! NI.I

", Q! $l

• Q I y

,..g. %."N 3 1 1,3 -e~e+ o s y J

Table 4-1 (Sheet 3 er4) Braidwood Unit 1 AprH 1997 Outage semissary or Inspection and Repair for Tubes in Service During Cycle 6 se I cenerne.r o c ,.dee f AE SGs Sh During Cycle 30C 7 Is.5er.are Dartug Cyete 90C-7 48 Tuame Cameramed AS Tutse Camerumme yw Vettage ' yw w aw att SPC Immumsh'm' Bremrund & lume Smg===4 Cedressa n.,mer.e 8% ese naaman att apC Imerme m am Sm, nee Imenesse Ims, nee Ch a.,mer.e sn.e., amem oisy smese Imam omsy OI 1 0 0 0 I I I O O O I t 02 12 0 0 0 12 12 - 35 t 0 0 35 34 _, 0 3 _ _ , _92,,,. _.,_3__ __2, 4 88 87 256 5 3 5 245 246 5,..__ .._21? _. __,__M___ m __, _. _ st __ 2 13 m See _93. .. 216 _ __ _6__,_, .__ p _,_, 261_ ,_259 __ _ J55_ _ . _ _ 1 1,,,,_ 6_ _ 2t _ _ 734__,,. m 05 26?,__ _,3,__, _t._ a 06 .. 278 _

5. _ _2.__

5_. _m_ ._.m _. __ st_ ___ t t__ 6 23 m_ m _._9 I - _.227.._ __.._f..,_.. 2,..,,_ .3 , _ 224_ __ _ 222, _ M 15. 7 to 693 6es 05 IS4 4 2 2 182 180 627 Il 6 33 684 609 09 106 5 4 3 183 182 599 to 6 20 579 575 t - 125 II 9 4 124 122 4M IS I4 12 452 448 ..~. 2_ .. _ 53.. t3 77 88 15 2 75 72 241 43 38 7 24 238 84 44' 6 6 0 44 44 tuo 27 23 4 IS4 18C 1,5 34 5 5 0 34 34 313 39 39 I 337 137 16 27 le 9 0 27 26 107 22 19 5 102 99 f.7 25 I I O 25 25 95 85 34 3 92 98 I8 22 8 5 I 28 21 74 19 19 2 72 72 I9 20 4 4 0 20 20 69 13 82 2 67 66 2 to 4 3 I 9 s 54 22 21 2 52 31

2. 8 52 It It O

12 52 42 _ 35_ 35 0 ,_42_ 42 2.2 3 I O 3 3 24 19 19 0 24 24 2.3 6 5 5 0 6 6 23 23 22 0

s 27 24 4

3 3 0 4 4 25 16 15 t 24 23 25 5 5 5 0 23 23 21 0 23 23 26 2 I I O 2 2 6 5 5 0 6 6 2.7 3 2 2 0 3 3 37 13 12 0 17 16 28 I I I O I 1 to 9 9 8 9 9 29 3 5 5 0 5 5 17 15 85 0 37 37 3 2 2 2 0 2 2 Il 8 3 0 ff II 33 3 3 3 3 0 0 13 13 13 83 0 0 32 3 3 3 3 0 0 1 7 7 7 0 0 . _ 3.2 _ ._,9 _ _ 0 0 0 0 0 6 6 6 6 0 0 j; 34 0 0 0 0 0 0 8 8 8 8 0 0 3.5 0 0 0 0 0 O 2 2 2 2 0 0 5 5 0 0 38 I I I I o O t t i I O O 39 1 I 1 0 0 3 3 3 3 0 0 4 8 I I I O O 4 4 4 4 0 0 Tame Canalemes se Nest Page 4,7

.- ~.~...-.- - -n--_._ .n_ s j [ { t + [ l I l .,. !.. g 1lI -c. s ! I I 4 l l 4 W I i V I 1 Ii l 1 g i = l l j .>.,.,.;..:....;.!.l........ 2. I .,.;...... ! 5 3 I 1, 8'- It i i i ti ii 4 !l -.l.~.4!,f,, ,ll l! l l lI l !I -li g } .~;~ _i_ - _;_ g g g ~_ lli ll ll!!i ll ! ll I is >i ilf f l l l! ll l .;_...___._f,~8;_ a.. :~is_ ;~,-;~ j> f .y _: ; e_:_: _._i_._._, a-8 v I i l i l i { liIl,i!l l -i+i! j 1 . e t 4.i. i i !f1 '!i! W ( v . i .{ii I !!1 ll!!! I il'I; I j y ... :_1_;~ _ ~.. g ~;_;~ -l~i_ ' ~l j _! j _j _;_;.. 4 g a a 9 d ii4 I i I l !i I' i 1 9 I I i [ } iI l l!l1 l i I g (,..E )j, .,_,_:....'..~. -l~! tH j m, ~ !. ~ ~ ; _ ;-, _ - 1.., _ l' _. _ i _, _ ; _... g g g m i ilI ' c j { i. i 4 i . i I i Ii l 1 g E ,. i

j

-1 i- ) . i i 4 1 i I ).! f . ! l l l M { i i . t.!.I ....';....'.:..1....l.....!.!. f ' ' ! i E i s g g. .i..;... I i j lj i,1 .3 9 a a g 3 ..t....i..-......I......i..........:..l.i.. l O t 3 - &} } .. :.._......_..... n = 1 .,_'.;. 2.;. ~ ~ _....:.,_._..._ _.;_:............... i 1 F g v g-I ~ 1 ~ ~ A y y J, ). ..:.:.,...;.,~ ~.. .,_;.3 .;... [ g : i, 3 j t i i 3g; _. _;....... _, _.;.._.i_ _..;_..;_._.... ..... n = h .N.?mh m.t......s..,y n:M-5 1%m te n ;n ta _as i g, N f.I$ ;$ )$. h,[N ih.'e. ih M..[.h..N t h. ,E-b a N .k. { L ..,3 r. . ~,~ 3 i o i 1 4 ,3 -, 2 P h~ .i 1 L _ _, -. _... - _.. ~ _. .--.....,i.4 _._ _, _.-,.---..~-,..- -..,,,..-~.,-,--,..- -. - - -_-~....,..,--,,..--,- ., _.... _ - -,. - ~ -,., -

+ .+ .~ ~-- ..= i l.': i 4 Table 4-2(Sheet i er2) l Braidweed Unit I April 1997 TSP ODSCC Indication Distributions for Tubes in Service During Cycle 6 ? q, 4 k l '. Seessa Generseur A Serans Generseer B Semesa Gamermeer C

• 4 ember of Mamanuse Aserage largest Average Numher af Mammum Avenge Lay-se Avenge Number of Mammense Average Lmgest Average Tube M s" ladscanons Voltage Vahage Gewee Gnrere suecaemun Vahage Voltage Gnroe Grewe Imkanees Vdesge Vahage Grw=e Geese

} ri.ar f i: 1H I 036 036 038 038 1 0.54 034 0.26 0.26 0 3H 803 835 1.00 732 037' 383 334 0.79 2.16 0.14 1104 9 82 1.15 830 _.. 0.41 4 03 0.84 2.75 020 ~i5~ ~ 552 8.94 0.90 7.47 031 ~ ~ ' '342 3.59 0.82 2.06 0.23 571 ~ 5H ^ 5 28 6I '3.23 U.i) ~ 5 56 ~U.ii ~56is ~ ~ d'.86 238 io.48 0 s4 8.29 0.23 2N

5 55

' O.i2 1.UU ' O.16 Il ^ '039 ~ liif '3 132 0.62 0.94 0.16 i ~5ii ~ IE9 "3 04 ~ 'd58 6 54 620 ~ ~i' O.31 ~ U33~ ~633~ 3 06-I.14 036 ~ ~~3~ ~2.15-0.65 1.10 0.23 ~ ~ 61 ~~ N IT ' 3 2. 9 4 IDH 21 0.96 0.51 0.42 0.11 2 0 64 032 0.10 0.10 5 0.78 035 0.29 0.15 I IIH - '2 0.99 0.76 0 42 038 0 1 0.91 0.91 031 0.51 f ~ b6 U iiC 2 155 0.98 0.13 0.02 3 ' O.49 0.44 ' ~ 0.11 0. ~~2 0.05 ~0.68~ ~ l U.' 8 ' UU7 ~ 'E ~ ~031~ li~ 0.02 0.02 [ 5 lbU 10 U.63 U38 0.18 0.07 2' O 29 -0.04 ^~ 5U~ U.E7 E '2' l'24 ' U.i9 U39 0.12 0' E ~ 0 L 8C 2 038 033 0.17 .0.07 0 0 l 0 L 7C 0 0 4 :' ' '~ 5C 0 2 U 42' O.41 ~ Uilo ' ~ id 'E ~~ ~ -G } 1i Total 1770 808 2I04 I I! L I [ i J !u l iI f i' 4 r 4-9 ~w 9 {'

I Table 4-2 (Sheet 2 of 2) i Braidwood Unit 1 April 1997 TSP ODSCC Indication Distributions for Tubes in Service During Cycle 6 I SeeseC h D C- =,-- ? of All SGs i Number d Maumum Average largest Average Samber of Maurram Average targest Average Tube Support Indscatwan Voltage Voltage Growth Growth Indscatxms Vohage Vohage Growth Growth Plate III 3 0 49 039 0.11 0.06 5 0.56 0.46 033 0.19 i ~ fil 9.82 , i.0U _. [836 _ _ 035[_ 3 311 _ 1023 8.82 0.91 6.18 028 _. 51! _ _ ,,622 _5 10 9 82 _ 3.95.. _ g23 __ .,37_ _,,, __194_ _, _0_ 85 __ 7.47 _0_24_ j 10.48 0.81 8.29 0.24 7 11 274 533 0.76 333 0.21 875 ~ _-.5555.._ _0.68 4.83 0.17 U.19 " ~ U.68 ._4 83 '__ _ 338..__ ~ El{. _ _.12U_. 553. ~ IO11 3 0 51 US1 U.22 0.U8 31 UN ~ ~~ 0.42 0 II 05 Ill{ 0 3 0.99 0.81 0.51 0.42 IIC 1 036 036 0.00 0.00 6 1.O* 0.61 0.13 0 08 i IOC I 037 037 0.05 0 05 IS 0.63 035 0.18 0.05 i Ob 0' UN ?.. . ?. ~ ?.-.. '.1#-. BC _ _ _2_ _ _ 038_ __ 033_ _ 0.17 0.07_ 0 7C I 030 030 0.14 0.14 1 030 0.50 0.14 0.14 SC 0 2 0.42 0.41 0.00 -0.07 Total 2I02 6784 h l t i 4-10 L: i c .rtame auntwm 2s Pte -i . m +..g-. e

Table 4-3 i Probe Wear and Analyst Variability - Tabulated Values Analvst Variability Probe Wear Varability Std. Dev = 10.3% Mean = 0.0% Std. Dev = 7.0% Mean = 0.0% No Cutoff Cutoff at +/ 15% Value Cumul. Prob. Value Cumul. Prob. 40.0 % 0.00005 <. I 5.0% 0.00000 38.0 % 0.000l I 15.0 % 0.01606 36.0 % 0.00024 14.0 % 0.02275 34.0 % 0.00048 13.0 % 0.03165 32.0 % 0.00095 12.0 % 0.04324 30.0 % 0.00179 .I 1.0% 0.05804 28.0 % 0.00328 10.0 % 0.07656 26.0 % 0.00580 9.0% 0.09927 24.0 % 0.00990 8,0% 0.12655 22.0 % 0.01634 7.0% 0.15866 20.0 % 0.02608 6.0% 0.19568 18.0 % 0.04027 .$.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 100% 0.16581 ~ 1.0% 0.44320 8.0% 0.21867 0.0% 0.50000 6,0% 0.28011 1.0% 0.55680 4.0% 0.34888 2.0% 0.61245 ~ 2.0% 0.42302 3.0% 0.66588 O.0% 0.50000 4.0% 0.71615 2.0% 0.57698 5.0% 0.76247 4.0% 0.65 I I 2 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 14.0% 0.91296 11.0% 0.94196 16.0 % 0.93984 12.0 % 0.95676 18.0% 0.95973 13.0 % 0.96835 20.0 % 0.97392 14.0 % 0.97725 22.0 % 0.98366 15.0% 0.98394 24.0 % 0.99010 > 15.0h 1.000D0 - 26.0% 0.99420 28.0 % 0.99672 30.0 % 0.99821 32.0 % 0.99905 ~ 34.0 % 0.99952 36.0 % 0.99976 38.0% 0.99989 40.0% 0.99795 4 11 ,-e, .+.w.- .-.. 2-e r-.

---

....--,,-,,---e-,-.,w ew+

,,w,. -.m

i i l i f. j Figure 4-1(a) l Braidwood Unit I April 97 Outage l Bobbin Voltage Distributions at EOC-6 for Tubes in Service During Cycle 6 t ( i t 300 l l t i l [ j 250-ESG-A 1 I B SG-B 200-t i l I j c 7 l } l 3 3 l BSG-C W j l i i o l 1 5 l l t i e l l l - 150 _ l l i l l i i l ESG-D .c l l l l l t 6 l l l l i {- E l l l l l d = l l l 100 - !, i i l l' E l 1 l : i ll ! l l l l l i ji : l : : + l l l l l : : l l l l l : l l l t l : l l : l l : l l : l l l : : l l l l l l l : : ll : l : : l l l l l l l l It i ll l l 50 - l l l : l : : l l l l ll a l l : l l : l i l : : l l l l l l l l l l l l l l . l l l l l = l: : l : l : : = I." .E .' O 'I I E O. t 7 N 9 7 9 9 9 7 9 7 9 [ 7 N 7 9 9 N 9 e. o o e o o e o o o n n n n n Bobbin Voltage i IhAmerycFege-ta}E I19712 38 PM 4-12 i i

t l L I t Figure 4-1(b) l l Brddwood Unit I April 97 Outage f Bobbin Voltage Distributions at EOC-6 for Tubes in Service During Cycic 6 i t 9 8-l BSG-A 7-l t l l t l l t i i B SG-B 6-g l c .2 l t W ESG4: y 5-l.;- 3 l j c s l l l l l l l T l l i ESG-D t4-l l l j j l j j I = l l l l l l l l l l E l l l l l l l l t a z; 3 - l l l l l l l l l l l l l l l l l l l l l l l l l l l l l l l l l l l l l l l l l l l l l l l l l l l l l i l l l l l l l l l l 2-l l l l l l l l l l l '; l --- - ~--; l l l l l l l l l l l l l 0 l l l l l l l l l l l l l l l l l l l l l l l l l l l l l l l 1-l l l i g-- i-l ll l l l l l l l l l l l l l l l l l l l l l l l l

: l l

l l l l l l l l l l l l : l l l l l l l l l l l l l l l l l l l l l l l : l l l l l l l l l l l i 0 o b o ' b o " 9 9 9 e. m 7 9 7 9 9 9 9 R 9 9 7 9 9 9 9 l n n n n m m m m m m m m m w w w w w w w w w I Gobbin Voltage l i I 4-13 &%Fg4-tbpL 13 9' I2 39 I"M i l P

i Figure 4-1(c) Braidwood Unit I April 97 Outage Bobbin Voltage Distributions at EOC-6 for Tubes in Service During Cycle 6 [ ] 9 8-ESG-A 7-BSG-B r ,6-BSG-C t-g. y 5-ESG-D 3 w Cg4 E D 3 1 $= g l l l l i l l 3~ l l I l :ll:ll: l:ll: 'L l: l ll: l: l l: ! l: l: l I: F:I: I:F:I:T:I o -n m n

,e

~ x x m ~ m e b W W W W b b M 1 ? D O ? ? ? e. N 0 e o e e e o o o e e = e Bobbin Volt:ege f 1 4-14 l %r ., not i2.i m i

l Figure 4-2(a) Braidwood Unit 1 AprilS7 Outage Bobbin Voltage Distribution for Tubes Plugged ARer Cycle 6 Senice l 9 l i 8~ - ~~ ~ 12SG-A - ~ ~ f l l i t l 7-sso.g q l \\ \\ l l l l l l f l S SG-C 6-E l E l -E 1 I i MSG-o .Q s. l l l l i 3 l l ! ! l l l ~. l l ) s I o l l l l 4 _ _ _. _. _ _. l i e 1 !!.1 l l l i 1 a l i i i = i z l l l l l l ) 3-l [ l l l l : l l l l i l 1 l l 2-1 l l l l l l l l L

L I; l;l! I~ l;ll;lI11;11;ll;l ;l__

1, ;I _I ; ;ll; 111;11 1_ E. 0 I 7 N 9 7 9 9 t M * " 7 9 7 N 9 7 9 9 % M e. 9 7 9 9 t M o o o o o o o n n m m m m m m m m m r Bobbia Voltage l 'i i %r.ounSmom 4-15 t

t I I Figure 4-2(b) l Braidwood Unit I April 97 0utage l Bobbin Voltage Distribution for Tubes Plugged After CycIc 6 Senice i 2.5 ) GSG-A a ~ j B SG-B l f SISG-C l c l t j @ 1.5 - l BSG-D a i l \\ J"c t l l I w o F l I' l i 2 E I-i t j. z l l l y u I m l l l l l l l l r 7 l l l l 1 l i l l \\

l i

0.5 - l i l l \\ l l ? \\ ( I l l I;

l l

0-7 9 9 S. 9 9 9 N 7 M i N l w w w w w m m m m o w n n oc w Bobbia Voltage 1 I t % r,e x n,m o m 4-16 i s f I

1 i Figure 4-3(a) Braidwood Unit 1 April 97 Outage i Bobbin Voltage Distributions for Tubes Returned to Service for Cycic 7 l 300 i ElSG-A 250 - 3 200 - BSGC 150 -

l t;

_. ima. l 0.1 0.2 03 0.4 0.5 0.6 0.7 0.8 0.9 I 1.1 1.2 13 1.4 1.5 Bobbin Voltage 4-I7 j s%r.... w.> s %,,,.3.,

_-=. Figure 4-3(b) Braidwood Unit 1 April 97 Outage Bobbin Voltage Distributions for Tubes Returned to Service for Cycle 7 ) l i l i ~ gSG-A 40 - i l 35 - BSG-B l 30 - 25 - l MSG-C t j20- "sG-o iS - 10 - E f 1 E d A A A d m 0 I.6 I.7 I.8 1.9 2 2.1 2.2 2.3 2.4 2.5 2.6 2.7 2.8 2.9 3 I Bobbin Voltage 4-18 %r.8-w m% s#w +- -~ m

Bobbin Voltage Distributions for RPC on o a pected mations Returned to Service for Cycle 7 300 ESG-A l 5 -- 250 8 SG-B 200 B SG-C i 150 - E ESG-D 100 -

=

50 - 0 O.I 0.2 03 0.4 0.5 0.6 0.7 0.8 0.9 I I.I I.2 13 1.4 IJ Bobbia Voltage 4-19 % u,,w., %

• r w.i

Hgure 4-4(b) Braidwood Unit I Apri197 Ostage Bobbia Voltage Distributions for RPC Confirmed and Not inspected Indications Returned to Service for Cycle 7 - 45 ESG-A { -- 40 - 35 - B SG-B 30 -

25 -

ESGC ~~~ l i g 20 - ESGD i 15 - Af i 0 1.6 I.7 1.8 1.9 2 2.1 2.2 2. 2.8 2.9 3 Bobbia Voltage f 4-20 s w.kr.c4 w s w hv.,4 w

Figure 4-5 Baidwood Unit 1 - April 1997 t O D."CL r 1200 L 1000 ( t E E I 800 E 5 ESG-A .E ? g 600 BSG-B 5 i i. 5 5 I e e I Z '400 B SG-C 5 5 E E ESG-D l 200 { W i t O i III 311 Sil 711 811 911 1011 till llc IOC 9C 8C 7C SC Tube Support Plate 4-21 j f Fig 4-5 Oart Ifig4-5 Omrt i E%/97 4

-. ~.. Figure 44 Braidwood Unit.1 EOC4 Inspection Counparison of Worn Probe Voltage Against New Probe Voltage steam Generator A s0 60 4.0 l 0 / to 9 g. to s e 00 00 1.0 20 30 40 5.0 60

m. Prone vou, Stoem Generator 8 s0 60 40 0

I 20 / to V* / 0.0 00 to 20 30 40 60 60 m.Pronevom y 4-22 ,-a, ~ v , -.~., .--,,nw,,, w-.,--- ,---n.,

Figure 4 7 Hyron Unit.1 Comparison of Worn Probe Voltage Against New Probe Voltage Stenen Generator C 1.0 s0 5.0 ' e 40< 1 g 30: 3 e e to< 10-e 00 00 10 20 30 40 50 80 70 hw Probe Vohege Steam Generator D 1.0 e 80 80-e 0-r ,y. 2.0 e e D - g 1.0 a 00 0.0 10 t0 30 40 $0 60 7.0 64ew Probe VoNege 4,g .l

t' 7 d n n o a i s B w s e l e r o N a g T 6 t e f a R o D r 5 a 9 5 d e / e n 2 l p o 0 /- i i F L 9 + /. ,P /. / o ,a 5 v s t /. l o o_ V 7 9 e /. 9b / e 1 o 4 g r c a liP t r l o pw 8Ae /. V 41N /. e b 4 n o 2 eiv t s r 4 r s. P n / u s gut w l o e i F d o r 3 e. N oV t o a e wb d o ir aP o. r a-Bn r o .o. W 2 / o. o ,m o a o t 1 a c u 7,rwe c 0 ri. 7 6 5_ 4 3 2 o g a x o > o a. e. E o .r a

g u, Il I I I l I i I I I m .._, ~.. ~ \\ \\g% 1 l N 4 .E g m.je. 'N N e9 ..s E s'* a W E b c 4 Z >E a g p c4 y :$ i s g i i c io 1. I I > 1 l 4 > %P o7 e e n 4 n 4 n o o o o o o e o o A)l[lqMquJd 3Al)RlI1Lilll3 m.

5.0 VOLTAGE GROWTH RATE ANALYSIS It this section, growth rato data for Cycle 6 is analyzed and compared with urresponding data for the prior cycles at Braidwood Unit.1. Voltage growth rates for Cycle G were developed from the EOC G (April 1997) inspection uta and a ovaluation of the same indications from the EOC 5B (November 10' u inspection i.0 signals. Comparisons of actual measured EOC 6 voltage distributions and leak rates and burst probabilition calculated from the mensured voltages with the EOC 6 pojections presented in the last 00 day report (Referenco 10.1) showed that the projections significantly underestimated the actual EOC 0 conditions. Details of the comparisons between the actual and projected EOC 6 conditions are presented in Section 0.0. An evaluation of the root cause for underestimation of projected values is also presented in Section 0.0 and Appendix A. To support the root cause evaluation, Cycle 0 growth data were examined closely to ascertain if they exhibited a trend different from the prior cycles. Cycle 6 was the first full cycle after implementation of the 3 volt IPC. A substantial number ofindications with BOC voltages between l 1 to 3 volts were left in service during Cycle 0 and the growth rate trends for the inrge BOC indications may differ significantly from those of smaller indications. " herefore, to examine if the growth rate trends were dependent on the DOC voltage, Cycle 0 growth rates as well as prior cycle growth distributions were analyzed, and the details are provided in the sections below. 5.1 Comparison of Cycle 6 Growth with Prior Cycle Growth ile 51 shows the cumulative probability distribution of growth rate for each Dmidwood Unit:1 steam generator during Cycle G (December '95 April '07) on an ctivo year basis that is defined as 3G5 days operation at RCS temperature above Mr F. The data in also plotted in Figure 51. The curve labelled ' cumulative' in Figure 51 represents composite growth data from all four SGs. Average growth rates f r each SG during Cycle 0 are summarized in Table 5 2. Among the four steam encrators, SG A had a slightly larger average voltage growth during Cycle G than other SGs; however, SG C had 6 of the 10 indications with the largest voltage growth. Steam generator A had the largest average voltage growth rate during the last (EOC-eld inspection also. The average growth rates over the entire voltage range vary between 20.5% and 47.1% (of the BOC voltage) on a yearly basis between SGs, with an overall average of 30.4% per effective year (time at temperature). The average grawth for indications greater than or equal to 0.75 volts is 40.5% per year and for indications less than 0.75 volts it is 38.3% per year. The percent growth for indications above 0.75 volts is comparable to that for indications below 0.75 volt. o NNSDIS\\ ape \\cce97ven690d wp5 51 i .w- ,,.~ e.e-. ,r, s ._m ,o_ ,,,. ~.. - r.-. . m.-

This differs from prior data which had lower percentage growth above 0.75 volts and indicates a potential trend toward increasing growth at higher voltages. Steam generator C had the highest average voltage at B00 0 whereas SG.A had the largest average voltage growth during Cycle G. Table 5 3 lists the top 30 indications on the basis of Cycle G growth rates, in descending order. All 30 were detected during the last (EOC 5B) inspection also, and they were HPC confirmed in the present (EOC 0) inspection. Since an alternate probe wear criteria was used during the EOC 5B inspection, B00 0 voltages for the indications in Table 5 3, which are same as EOC 5B voltages, were checked to determine if any of them were recorded from a worn probe (and not retested with a now probe because they are outside the voltage threshold for retesting). All but 3 of the B00 0 voltages in Table 5 3 are obtained from a good probe, so probe wear could not be the cause for the large voltage growths observed. Averaged composite voltage growth data from all four steam generators for the last five operating periods are summarized in Table 5 4. Figure 5 2a provides a comparison of the composite voltage growth from all four steam generators for the last four operating periods (Cycles 4,5A, 5B and G) for voltages up to 2 volts and Figure 5 2b shows the data for voltages over 2 volts. Table 5 5 shows th; same growth data in a tabular form. The growth data shown in Tables 5 4 and 5 5 and Figure 5 2 are normalized using effective full power year (EFPY) for consistency with data presented in the prior IPC 90 day reports. The average growth rate on an EFPY hasis during Cycle 6 for all SGs combined is higher than the last cycle (Cycle 5B, a half cycle); however, it is below that observed for Cycles 5A (another half cycle) as well as Cycles 3 and 4 (full cycles). It is evident from Table 5 5 that during Cycle G more indications experienced higher growth than in any of the previous cycles. During the last full operating cycle, Cycle 4,10 indications had growth above 3 volts on an EFPY basis. The next two cycles, Cycles 5A and 5B which were actually half cycles, had 9 and 4 indications, respectively, with growth above 3 volts (on an EFPY basis). However, during Cycle 0, 45 indications had growth above 3 volts (on an EFPY basis). Because of the unusually large number ofindications with i gh growth during Cycle 0, the tail of the i EOC.G voltage distributions projected at the BOC 6 using the last two cycles of data are not conservative: consequently, the projected EOC.G leak rates underestimate those based on the actual voltage distribution. It is shown in Figure 5 2a that below about 98% cumulative probability, the growth in Cycles 4 and SA is larger than Cycle G. Iloweve nbove 98%,(Figure 5 2b), Cycle G growth is larger. This emphasizes the o \\NSD15Aapc\\rce97\\rcec690d wpS 52

rates in order to be able to adequately project the high voltage tail of the voltage distributions. To assess the effect of the large number of indications experiencing high growth, EOC-6 projections presented in the last 90-day report (Reference 10.1) were repeated using the actual Cycle 6 growth. The results are described in detail in Section 6.1. Even when the actual Cycle 6 growth distribution is applied, EOC-6 leak rates projected using the standard methodology were below that estimated using the actual voltage distribution. Leak rates based on the projected voltages were correspondingly also lower. To explain the difTerences found in the leak rates based on projected and actual EOC voltage distributions, a systematic examination of all salient parameters affecting the projected SLB leak rates and burst probabilities was performed; it is described in Section 6.1 and Appendix A. This evaluation indicated that the cause for underestimation of projected leak rates is the way in which growth for Cycle 6 is simulated in the projection analysis. In the current projection methodology, growth rate is assumed to be independent of the BOC voltage and the growth rate data is represented by a single distribution. However, the occurrence of a large number of indications with high growth values suggests that the frequency oflarge growth rates may he dependent or the BOC voltage. Therefore, growth data for Cycle 6 were evaluated for dependency on BOC voltage, and the details are presented in the section below. 5.2 Voltage Dependency of Growth Rates Figure 5-3 shows a plot cf voltage growth, AV, versus the BOC voltage, Vsoc, for a hybrid data composed of all of SG C data, which has 6 out of the top 10 growths observed during Cycle 6 including the top two growth values, plus the third largest growth which occurred in SG A. It is seen that larger growth rates begin to appear above a Vuoc of about 0.75 volts and the Ergest growth tend to appear above 1.5 volts. Since the indication population in a given voltage range drops off with increasing BOC volts, large growths as a percent of the number ofindications is much higher above about 1.5 volts than below 1.5 volts. This implies that the frequency oflarge growth rates is dependent on the BOC voltage. When all of the growth data is represented by a single distribution, as required by the current standard methodology, the probability of the large growth rates is diluted by the large number ofindications at low voltages. A modified methodology is needed that maintains the frequency oflarge growth rates as a percent ofindications for BOC voltages above about 1.5 volts such that when applied in the projection analyses, the BOC indications above 1.5 volts left in service can grow proportionally to larger o;\\NSD15\\apc\\cce97\\ccec690d.wp5 5-3 m.______.____._____

I voltage indications at the EOC. Such a methodology is developed in Appendix A and the growth distribution recommended for the licensing basis analysis for EOC 7 - predictions is presented in Section 6,0. The frequency oflarge growth rates in the tail of the distribution can be properly maintained by dividing the growth data into several bins on the basis of Vnoc and developing a separate growth distribution for each bin. This methodology of modelling voltage dependent growth rates is described in detailin Appendix A. In Section A.3 in Appendix A, specific guidelines have been developed to select different growth distributions applicable to increasing BOC voltage ranges. Voltage growth distributions for performing licensing basis analyses for EOC-7 conditions obtained by applying the growth selection guidelines are presented in Section 0.0. 4: 1 1 j 'l f 1 ~ o.\\NSD15\\spc\\cce97\\ccec690d wp5 54

~-,. Tatne 5-1 (Sheet 1 of 2) Braidwood Unit 1 April 97 Signal Growth Statistics For Cycle 6 on an Effective Year Basis 365 Days at RCS T....,

.

n Above 500 deg. F L

Steam Generator A Steam Generster B Steam Generator C Steams Generseer D Cumaniedre [ Cyde5B Cyde 6 CydeSB Cycle 6 Cyde5B Cyde6 Cyde55 Cyde6 Cyde55 Cyde6 D'#2 l (d %d yj %d yj cror cror cror cror Cror cror cror cror Cror Cror - 1.7 00 0 00 00 0 0.0 0.001 0 00 00 0 0.0 0.0 0 0.0 0$8 00 b bb 0.0 U h.0 5 001 d dd dGJI ~ 0 00 0.0 0 00 i -l 00 0 00 00 0 0.0 0 001 0 00 0.0 d~ ~dU ~ '0'0U1 U 00 4.7 00 0 00 00 0 00 0 001 0 00 0002 0 00 0.001 0-0.0 0.6 0.0 0 0.0 0 002 0 00 0 001 0 00 0903 0 00 0 001 0 OD -0.5 00 ~0 0.0 0.002 0 0.0 0 003 0 00 0.005 0 0.0 0.002 C 0.0 -0.4 0 011 1 0 001 0 007 1 0 001 0.003 2 0 001 0 01 0 00 0.005 4

0.0006

-03 OD07 2 0 002 0 011 0 0 001 0 009 7 0 004 0.018 3 0.001 0.01I 12 0.0024 ~ -0.2 0 024 8 0 006 0 034 11 0 015 0 031 22 0 015 0028 12 0 007 0 029 53 0.0102 -0.1 0 063 4I O 029 0 081 29 0 051 0 088 62 0 084 0 065 47 0 03 0 075 179 0.0366 0 0.181 172 0.127 0.228 97 0.171 0 199 231 0.154 0.218 248 0.148 0.203 748 0.1469 0,1 0 375 352 0.325 0 399 259 0492 0.401 505 0.394 0.433 ( it 0.463 0.403 1777 0.409 _92 95$!.,. 40! ,g452 a5pL 190 p.727 _ g R,, _334_ 3605_ _ g.p!4_,,, 4d5 _0694 0.592 1520 0.6332 _ g.3_ 0685 280 _ g7! _a7!9 no 9 85! a775 _ 255 3726_ _p3L_ _ 2 14 _ 0 81 0.734 879 0.7628 . 9-d _ _ 9 773._ 150 . 3 795 _ .,9 822 _ 46 0 908 0 857 _ 175_ 0809_ 0.823 120 0867 0.823 491 0.8353 l 34.__ ._ 9.84 _!l4 R@@ _ O 878. 29 9944 pp,. _!II. _pB62 ,,0882 _ ,, 72_ __ _ a901 am 326 0 8833_. 06 0 888 56 O_891 0 917 11 0 958 0.935 65 0 893 0 912 54 0.927 0.915 186 0.9108 0.7 0 922 40 0.914 0 939 5 0.964 0.949 44 0.914 0.931 34 0.943 0.936 123 0.9289 08 0.445 26 0.928 0.% 6 0.971 0.%2 41 0 933 0 949 25 0.955 0.954 , 98 0 9434 0.9513 i 0.9 0%I 18 0 918 0 968 3 0.975 0 974 23 0.944 0%I - 10 0% 0 966 54 ~W 0.9602 ' I l U.971 25 0.953 0.975 4 ii48 U55 54 ' 5 55I ~ UW3 17 ~ 0 O.976 60 45 0.9668 t 1.1 0976 14 0 96 0.982 3 0 984 0986 22 0962 0.978 6 0.97I 0.98I ~ ~ 5~ UMii' i i.2 h.984 15 0.966 0.986 4 hv89 0 989 9 d566 ~ U.Ui ~ ' ' I'O UN6 UM6 3 1,3 . _. 0 987.. _ 10 0 972 0.987 2 ..0 998 0 991..

5.. _

_.0 968_ _ 0.988 __ 8 0.98 0.989 25 09754 ~.985~ ~ 0 686 ~ ~ b 993 2 5996 dE5 ~ ~ - ~ ~dW~ 053-I.6~ b993' 8~ ].37 ' 'h595 3 a988 6 695~ 0 0 696 UR~ _[ [Og l6 { )) [65186[ ~ b~665~ 15 0.984 % 0 0.994 3 2 14 098702_ l 18 q993 3

a989 9995 0

p.996 999? 2 _,. g.985 p997 2_ _ g.987, ___g996._ _7_ a9881_ l.9 0 995 2 0 99 0 996 1 0 998 0 997 2 a986 9997 .O a987 a996 _5._ g.9888_ 2 0997 2 0992 0998 1 0999 0 997 I 0.986 0 997 2 0.988 0 997 6 0.9897 Table Continues On the Next Page f GmeeTaNr$ st11&11971811 At4 5-5 i I

4 t i=l*i. I mNel'h I I I k !E el m, o ~ = = N m 5 5 5 !S!$Ih h I l b $ I I!h w oOjCCCojClC'OCC olC Oj CCClCCCCCCJC C C C C Ci k l lll I! i O %j j CN7C

;=j

=;N g N -,= N,NC N N = N = = l 1 l y t 5 I i l 1!lj l } 1 f EfiE I!! !!!!!Elff!!!!!io! l l 4 C:CiCCicCCCicio;CC3ClCil !I i, g i O I ) l I l l i i l l l l t 1 i i i.il ! 6 } l hEIE E El5155I5'kh!!h!hhkh$lh hh I f k!.! j l ,3 Cl1 e Q C Ol* O Clogo C C C CiCiCiCICiCICiCjC CC C10101 '. j g l l. ) i* ' ! I f i i ! !, t O T ]* } I ll!!I;}!l! !l!!! !!ll g N = = ' C = - = C C ' N ! N I = l N ! C l = ; r> ; C i m i =, c i C N i = > = ! = i c l = 0 C C ' O ] 1 5 z-I ' i i 1;tiii lt I 4 i k? 1 I ( l ! l ! l ? ?. I I ' l l ' i E i ! { i, 5 ! ! } i 1 i i

• s i

e Elf lEifliffEff!!!!!!!!!=! l'1i)l!iiI g g Q g 0101010,C ClO O CICiC!CiotCi i 4 V M I I i i l 4 - I, i f ! + g .=.g; w e e y Bi! eel *>&5'5:5:55sss!Is!!!EiE!Eis!E!Eifif!!!!!!'o N_ U e Q cec:CiC;C;C C C C Cio;CIC1C C C;C cec lC:C:C;C;C,C=O Cic-C C 84 1 i v 2. m , i 1 v= O g] N Nio C:= 9 C.= CIC;= C;m - C N C N -+N:=+C;C:C-C m Ni= C'O N .T.I L Z" } ~ i C u g g il w ") U o ,= a N 6 E 8 ( e h!hikhihhI'sE!!Efif'sEEEE:*, 1 ). g Q CIC-02C;C CIC C C:Cic:CIC:Q;C C CIC;C 3 W U " A.g 8 c i = e$ @ $ fffffilifE!!:ffitf= E e Q g Cio-Cjo C:C C C C.C C;C C C-

• O jo 2

C 3 4 0 (11 63 C;C'C!CIC C'C C C? C-C C C C = c C C C C C C-C C C C C C C C C h I gg c z- ) E u. 3 8 rie;s!!!sstriererster:ffffffstfff'f;e a 4 C:C-CIC-OCCCCCCCCCCC C C Cic:C 01C C=C'C C C'O g ~ )f, e C g] = = C i N ' === - *= = = C ' O C C ; C N C C N = ' C ! C t C ; C C - C, 0 C C ' O.= = = - C $ C Z" ~ = = 5 i e rirrestrerffffffffflif+ i 4 Q ClorotC'C'C C C O!C C C'C C C C C C Cic: ~ y e N N' lM*m i i !W W m tc

Table 5-2 Braidwood Unit I - April 1997 Outage Aserage Voltage Growth During Cycle 6 Voltage Number of Average Voltage Range Indications BOC Entire Cycle Per Year

• Entire Cycle Por Yeer * '

Composite of All Steem Generator Data Entire Voltage Range (,780 _ _ 0 62 0.27_8 _ _0.246__ _ _44.6% _. 39.4% _ V soc <.75 Volts 4735 0.46 0.198 0.175 433 % 383 % 2.75 Volts 2N5 1.01 0.464 0.410 45.8 % 40.5% Steam Generator A Er: tire Voltage Range 1770 0.60 0319 0.282 533 % _ _,47.19 _ _ V soc <.75 Volts 1277 0.43 0.232 0.205 53.2 % 47.1% 2.75 Volts 493 1.02 0344 0.482 533 % 47.1% Steam Generator B Enyire Vo_Itage Range _ _ 807 _ __. 0 6! , 31 E3_.., _ _0.162 30.0% _ 263 % 31.5% 27.9 % Y 80c 5 :75 Vol!s 57y gj6 _ __ gl_44. _ _, _ __31,28 __ 28.2% 24.9 % 2.75 Volts 233 0.99 0.278 0.246 Steam Generator C 40.6 % Entire yottage Range 21M Q68 _ _ 931! . _ _y.275 _ _._45.9% V soc <.75 Volts 1337 0.48 0.219 0.194 45.6% 403 % 2.75 Volts 767 1.02 0.471 0 417 46.2% 40.9 % Steam Generator D Entire Voltage Range 2009 0.60 0.248_ _ _ 0.219 .._ 36.6 % _ 41.4 % V soc <.75 Volts 1547 0.46 0.172 0.152 37.8 % 33.4 % 2.75 Volts 552 1.00 0.459 0.406 45.9 % 40.6% s t*wmahred unng number of days atme 500 des t-for Cycle 6 (483 days or 1.131 year) 5-7 ca esasarsana nm

Table 5-3 - Braidwood Unit i April 1997 - Sammary of Largest Voltage Growth Rates for BOC4 to EOC4 Steam Generator Bobbin Voltage RPC New BOC-6 Voltage SG Row Col Elevat6on EOC BOC-Confirmed ? Indication ? from a Good Probe? Growth C 16 41 0311-9 82 _ l.52 8.3 Y N Y C 11 108 0711 10.48 2.19 8.29 Y N Y A 41 65 0511 8.94 1.47 7.47 Y N Y A 15 ' 74 0 311 8.55 1.23 7.32 Y N Y ~.52 Y N Y C 16 76 0 311 7.41 0 89 6 C 13 73 0311 9.15 2.94 6.21 Y N Y C _l 5_ 69 0311 7.31 1.13 6 18 Y N Y D 15 7 0311 8 82 2.64 6.18 Y N N _C '27 79 0311 7.83 1.85 5.98 Y N Y' D 39 93 0911 6 94 1.45 5 49 Y N Y D 14 29 0 311 6.74 1.31 5.43 Y N Y D 11 47 ')311 6.36 1.23 5.13 Y N N .D 4 37 0311 6.15 1.29 4.86 Y N Y D 18 75 08tl 5.73 09 4 83 Y N Y C 16 93 0311 5.65 0.93 4.72 Y N N C 18 99 0311 5.94 1.31 4 63 Y-N Y I C 38 78 0311 5.65 1.06 4.59 Y N Y i-D 15 80 0 311 5.64 1.29-4 35 Y N Y C 36 36 0311 5.74 1.42 4.32 Y N Y A 28 24 0511 6 08 1.79 4.29 Y N Y C 31 69 03ll 5.31 1.02 4.29 Y N Y_ _ ) D 14 93 03tl 5.2 0.91 4.29 Y N Y __D,_ _ _ l. l_ _ 54 0311 5.52 1.25 4.27 Y N Y D._ _2_ _5_ _ _ _031 L _ 6,03 1.77 4.26 Y N Y C 22 34 03fl 6.28 2.05 4 23 Y N Y 4 A 40 50 0511 5.2 1.07 4.13 Y N Y A 36 78 0311 5.02 0.93 4.09 Y N Y C 7 99 0311 4.76 0.77 3.99 Y N Y C-16 { 03H 6.14 _ 2 16 _ _ 3.98 Y N Y D-3 106 0511 5.1 1.15 3.95 Y N Y 5-8 t w mw. a m .. ~. _. ~,,. - - -... - - .. _. -. _ _ _. -. - - - ~., -.. _ - _ -. _ - -. _.. _

Table 5-4 Braidwood Unit i April 1997 Aterage Voltage Growth History Composite of All Steam Generator Data Average Voltage Growth Average Percentage Growth Bobbin Voltage Number of Average Voltage f Range Indications BOC Entire Cycle Per EFPY Entire Cycle Per EFPY Cycle 6 (1256 - 467) - I.035 EFPY Entire Voltage Range 6780 0.62 0.278 0.269 44.6 % 43.1% V str < 75 Volts 4735 0.46 0.198 0.191 433 % 41.9 % 2.75 Volts 2045 1.01 0.464 0.448 45.8% 443 % Cycle SB ( 365 - 955 ) - 0.506 EFPY Entire Voltage Range 4136 0.64 0.110 0.218 173 % 34.2 % V sec <.75 Volts 2831 0.49 0.100 0.199 203 % 40.2 % 2.75 Volts 1305 0.94 0.I31 0.260 13.9% 27.5 % Cycle SA ( 554 - 255 ) - 0.714 EFPY Entire Voltage Range 3884 0.56 0.284 0398 51.1 % 71.6% V arc <.75 Volts 3085 0.46 0.25 035 53.8 % 75.4 % 2.75 Volts 799 0.92 0.420 0.588 45.8 % 64.1% Cycle 4 ( 1982 - 384 ) - 1.147 EFPY ) Entire Voltage Range 2654 0.48 0.260 0.227 54.2 % 47.2%- V arc <.75 Volts 2289 0.41 0.290 0.253 70.7 % 61.7 % 123 % 2.75 Volts 365 0.92 0.130 0.113 14.1 % Cycle 3 ( 481 - 952 ) - 1.132 EFPY Entire Voltag.- Range 167 0.62 0.620 0.554 100.0 % 893 % i V arc <.75 Volts 145 0.43 0.650 0.580 151.2 % 135.0 % 2.75 Volts 22 0.92 0.420 0375 45.7% 40.8 % c-res emwm 57 m 5-9

Tcble 5-5 (Sheet 1 of 2) Craidwood Unit >1 Bobbin Voltage Growth Statistics on an EFPY Basis CYCLE 4. CYCLE 5A CYCLE 5B CYCLE 6 Delta Volts No. of b" # CPDF CPDF CPDF CPDF obs obs obs Obs 1.7 0 0.0000 0 0.0000 1 0.0002 0 0.0000 1.0 0 0 0000 0 0.0000 1 0.0005 0 0.0000 -0.8 0 0 0000 0 0 0000 1 0.0007 0 0.0000 -0.7 0 0 0000 0 0.0000 1 0.0010 0 0.0000 -0.6 0 0.0000 0 0.0000 2 0.0015 0 0.0000 0.5 0 0.0000 0 0.0000 4 0 0024 0 0.0000 -0.4 0 0.0000 6 0.0015 10 0.0048 5 0.0007 0.3 0 0 0000 19 0 0064 25 0.0109 23 0 0041 -0.2 0 0 0000 50 0.0193 75 0 0290 54 0.0121 -0.1 0 0.0000 181 0.0659 190 0.0750 205 0.0424 00 702 0.2667 _ 563 0.2109 531 0.2033 709 0.1471 0.1 404 0.4050 598 0.3648 824 0 4026 1614 0.3855 0.2 422 0.5617 5e4 0.5100 783 0.5919 1497 0 6066 0.3 346 0.6810 462 0.6290 586 0.7336 862 0.7339 04 231 0.7700 363 0.7225 369 0.8228 534 0.8127 0.5 143 0.8290 263 0.7902 231 0.8786 380 0.8689 06 98 0.8660 194 0 8401 151 0.9151 196 0.8978 0.7 79 0 8990 169 0.8836 88 0.9364 145 0.9192 08 51 0.9225_ _ 102_ 0.9099 74 0.9543 95 0 9332 0.9 37 0.9400 64 0.9264 50 0.96M 80 0 9451 1.0 _ 31 0 9510 56 0.9408 38 0.9756 49 0.9523 1.1 29 0.9648 49 0.9534 21 0.9807 62 0 % 15 _ l.2 _ __12_ _09711 32 0.9616 21 0.9857 36

0. % 68 1.3 12 0 9759 30 0.9694 13 0.9889 29 0 9711 1.4 8

0 9777 25 0 9758 10 0.9913 27 0.9750 1.5 7 0.9811 17 0.9802 2 0 9918 22 0.9783 16 4 0.9833 6 0 9817 9 0 9940 25 0.9820 1.7 5 0.9852 12 0.9848 5 0 9952 12 0.9838 1.8 5 0.9866 4 0 9858 2 0 9956 16 0 9861 1.9 3 0.9881 10 0.9884 3 0 99M 8 0.9873 2.0 4 0 98 % 2 0.9889 3 1.0 6 0.9882 _ 2 I. _0_ 09896 _.5- ,0.9902 3 0.9978 6 0.9891 2.2 3 0.9907 5 0.9915 0 0.9978 4 0.9897 2.3 4 0 9922 4 0.9925 1 0.9981 5 0.9904 24 2 0.9930 5 0.9938 _0 0 9981 3 0.9908 2.5 1 0 9937 4 0.9949 0 0.9981 3 0.9913 2.6 0 0.9937 2 1.0 1 0 9983 1 0.9914 2.7 3 0.9948 _3.. 0.9985 3 0.9919 I l 2.8 0 0 9948 2 1 0 0.9985 5 0 9926 2.9 .1 0.9955 0 1 1 ' o9988 2 09929 3.0 2 0 9963 4 I i 09990 2 0.9932 Table Continues on the Next Page 5-10 magrinide lhteitat11TT PW ~

.i i Table 5-5 (Sheet 2 of 2) Braidwood Unit 1 Signal Growth Statistics on an EFPY Basis CYCLE 4 CYCLE 5A CYCLE $8 CYCLE 6 Delta Volts go, c

• • I CPDF CPDF CPDF CPDF olm obs obs obs Table Continues from the Previous Page 3.1 0

0.9963 2 0.9982 0 0.9990 1 0.9934 3.2 2 0.9970 2 0.9987 0 0.9990 1 0.9935 3.3 0 0.9970 0 0.9987 0 0.9990 2 0.9938 3.4 2 0.9978 1 0.9990 0 0.9990 2 0.9941 ' 3.5 0 0.9978 0 0.9990 2 0.9995 1 0.9942 3.6 0 0.9978 0 0.9990 0 0.9995 5 0.9950 3.7 0 0.9978 1 0.9992 0 0.9995 3 0.9954 3.8 1 0.9981 0 0.9992 0 0.9995 1 0.9956 3.9 0 0.9981 1 0.9995 0 0.9995 3 0.9960 4.0 0 0.9981 0 0.9995 0 0.9995 2 0.9963 4.1 0 0.9981 0 0.9995 l 0.9998 1 0.9%5 4.2 1 0.9985 1 0.9997 1 1.0 7 0.9975 4.5 1 0.9989 0 0.9997 0 2 0.9978 4.6 0 0.9989 0 0.9997 0 1 0.9979 4.7 0 0.9989 0 0.9997 0 2 0.9982 5.0 0 0.9989 _ _0-0.9997 0 1 0.9984 5.3 0 0.9989 0 0.9997 0 2 0.9987 5.7 0 0.9989 1 1.0 0 0 0.9987 5.8 0 0.9989 0 0 1 0.9988 60 1 0.9993 0 0 3 0.9993 6.3 0 0.9993 0 0 1 0.9994 7.1 0 0.9993 0 0 1 0.9996 7.3 0 0.9993 0 0 1 0.9997 7.5 1 0.9996 0 0 0 0.9997 8.1 0 0.9996 0 0 2 1.0 9.0 I l.0 0 0 0 Total 2659 l 3884 l 4136 6771 l 5-11 heemsettisNs SM18/1IM12 08 P%t

t'l si it - F9 ll 11 - f5 ii in 6't l i bs i -- 9't i n g l l 'g rt Al l 3 E 1 1 4 g g a 3 -- 6't ..l l i g Q M 4 V ct i s I 1 I Hr s FE Di, = . cr 1 p I l'E

• f a

-BT i i -9T e 6 2" - VZ $ B IZ ra O i '~ ## pc l 0C a R E'l 5' 'p .h : --91 [i .8 d --tI c -- E'l } O'l C (, -- S'O a e,g k htx 9'O A s's N s -t0 y

• p

~ - C'0-I > ic

t'0-

.s o ? m o m w m N o g ~ o o a o o o o o o o 3 uopaun3 uopnqpism aapegnan3

+' j oN z-e- A B 4 S 5 6 z-e e e e o-l l l l a c c c c y y y y c. C C C e-g. C t. e-s-: n-isa deB-n 5 n-iY-bP mF oE-- - m-Cn s a w_w Gn-n- S o h l lAt w ; - o c-r h G e t o-g wr ) a a oo t ( rf l 2 o Gn V eo 3 5 o ei n 1 el t i r c u b 5 uyb b o gCi i r =o B Fi t t s l i uD My &o t l i I i b eo ti a nb Uo 'x no r d P o ~- o e

• r v

wi i a - X wo t d al r u B m ~- u mo C ~- to e - o M P .K*.l oo 3 7 5 7 ~- /9 .X o 6 /R a 2 5 e to ]2 e ru 0 9 8 7 6 5 4 3 2 1 0 g i 1 0 O 0 0 0 0 0 0 0 0 F h c2%56 c23o.C5c ?::65E:U tg tr - ~ tuM

O ~m ~ A B 4 S 5 6 e e e e 1 l l l l c c c c "c y y y y C C C C K 0 C 9o ^ ~- ~ "m s, i s 9m a d .^ eB n iY bP mF Nr-oE Cn s a Gn 9r S o ~ h r. l lAt w - o ~ r r hG e t g wr ) a b oo t 7 l 9m o ( rf 2 Gn V 5 o K* n 4 ei i 1 el t b r c u ^ b 5 uyb Nm o gCi B r i Fi t t s l i uD 9m C My ~ T[x. t - il 1 i t b i a 9m nb ~ Uo x, r d P oo e 9m v wi dt a ial r u B m X 9se u C y Nne ~ r, 94c M 'x P f 7 3 5 9ne 7 9 / 6 /8 b2 - a - c 5 4 8 6 4 2 0 8 6 4 2 0 e 9 9 9 9 9 8 5 8 8 8 ru 9 9 9 9 9 9 9 9 9 9 ig 0 0 0 0 0 0 0 0 0 F h E?Ehc4i3$ I::2mEmU tg tr luM lll ll,ll\\il\\1

Figure 5-3 Braidwood Unit -1 April 1997 Outage Voltage Growth During Cycle 6 vs HOC-6 Voltage -Ilybrid orSG C + SG A 9 I j o SG-C (All indcations) g g 8 u Largest Growth in SG-A (3rd Largest in All SGs) Ej 7 j j o O O 6 y o $w$5 0 \\o o l 0 o

o

.e: 4 o o C o oo 0 i o Ooo g 0 o o o o l g - o o g-80 0- o 0 7pa-- 2 OO D i %O l O o o o o 0 o o A o e l -1 0 0.5 1 1.5 2 2.5 3 BOC-6 Voltage 5-15 Cy6cgrthFs54 %9710 20 AM

6.0 Methods Evaluation and Benchmark Analyses for POD = 0.6 As mentioned in Section 5.2, projections for Braidwood 1 EOC.G' conditions repeated using the actual Cycle 6 growth rates yielded leak rates below that calculated using the actual measured voltages. This section identifies the issues found from comparisons of leak rates obtained from projected and actual EOC.G voltage distributions and defines recommended changes to the ARC projection methods relative to the current NRC approved methods given in WCAP 14277, Revision 1 (Reference 11.2). Analysis methods revised to include a voltage dependent growth are benchmarked by comparing their predictions against leak rates obtained from the actual EOC G voltage growth distnbutions. The principal objective is to define an improved EOC projection methodology in conjunction with application of a POD of 0.6 (GL 95 05 requirement) based on comparisons of projected EOC voltage distributions SLB leak rates with those calculated from the actual EOC 6 distribution. Methodology revision is needed only for the application of a 3 volt IPC, and the standard methodology of Reference 11.2 is still applicable for 1 IPC. To assess methods limitations and to develop methods improvements, the preferred approach is to utilize best estimate analyses to assure that conservatism in one part of the analyses does not lead to erroneous conclusions on the need for or benefits of l methods improvements. The use of a POD of 0.6 adds conservatism to the projection methodology by adding assumed undetected large indications in service, particularly for Braidwood 1 with a large number ofindications above 3.0 volts. For this reason, a detailed methods development effort was applied using a voltage dependent POD to evaluate the methodology. This evaluation is given in Appendix A. This section describes application and benchmarking of the recommended methods for performing a licensing basis calculations using a POD of 0.6. 6.1 Identification of Projection Methods Issue Analyses were performed at the EOC 5B using the growth rates of Cycle 5A, which was the limiting growth for the prior two operating periods, and a POD of 0.6 to project EOC-6 SLB leak rates. The projections are presented in Table G 1. The analyses considered the locked TSP condition. The leak rates are dominated by the hot leg contribution and only the hot leg results are given in Table 6-1. SG C was projected to be the limiting SG with a leak rate of G.99 gpm. The results of the . Monte Carlo calculations for the actual EOC-6 voltage distribution are also summarized in Table 61. Based on the actual distributions, SG C was the limiting SG with a leak rate of 11.5 gpm using an NDE analyst uncertainty per GL 95-05 with o:\\NSD15 Nape \\cce97\\ccec690d.wp5 G1

a standard deviation of 10.3%. However, separate analyses have shown that almost 90% of the actual EOC G leak rate resulted from 17 indications above 5 volts..The NDE uncertainty of10.3% was developed from analyses ofindications dominantly less j than about P volts. Comparisons ofindependent NDE analyses for indications above 2 to 3 volts have shown excellent agreement and the NDE uncertainty would be bounded by about 5% or less. To demonstrate the conservatism in applying a 10.3% NDE uncertainty for the larger voltage indications, an analyses was performed with an NDE analyst variability uncertainty of 5%. The EOC 6, SG C, SLB leak rate of 9.8 gpm obtained with an NDE uncertainty of 5% is also shown in Table 61 and is 1.7 gpm lower than obtained with the 10.3% NDE uncertainty. The 9.8 gpm leak rate at EOC 6 is judged to be a more accurate estimate for evaluating projection methods although additional effort is required to quantify the voltage dependence of the NDE analyst variability uncertainty. The EOC 6 projections made at EOC-5B were based on an upper limit of 493 effective full power days (EFPD), while the actual cycle length was only 378 EFPD. As shown in Table 61, recalculation of the projections for the 378 EFPD cycle length resulted in a leak rate of 3.9 gpm compared to the 11.5 gpm from the actual EOC 6 distribution. It is shown in Section 5 that the Cycle G growth rates in the tail of the distribution were considerably increased over both Cycles 5A and 5B. To assess the inauence of the increase in growth rates while maintaining the growth rate independent of BOC voltage, projections were made for EOC G using the actual Cycle G growth rates. The results of these analyses, as shown in Table 61, gave SLB leak rates of 8.0 gpm for POD = 0.6. Methods should be evaluated based on comparing the 8.0 gym leak rate with the 9.8 leak rate or, more conservatively, the 11.5 gpm leak rate for the actual EOC G voltage distribution. Thus, even when applying the actual Cycle G growth distributions independent of BOC voltage, the projection methods underestimate the SLB leak rate. The above results imply a problem with the projection analysis methodology as well as identifying a significant increase in growth rates from Cycle 5A to Cycle G. Figure 61 compares the actual SG C EOC 6 voltage dist ibution with the projections using Cycle G growth rates for both POD = 0.6 and the EPRI voltage dependent POD of Reference 11.5. It is seen that the projections underestimate the large voltage indications found in the actual distribution which results in the underestimation of leak rates. The large voltage indications are underestimated even though the projections adequately include the total number ofindications found in SG C at the EOC6. From the plot of AV versus Vnoc, Figure 5 3, it is seen that above about 1.5 volts the occurrence oflarge growth rates as a percent of the number ofindications is much higher than below 1.5 volts, and the largest growth rates also tend to occur o:\\NSD15\\ ape \\cce97\\ccec690d.wp5 62

in this voltage range. This implies a BOC voltage dependent growth rate. When a single growth distribution independent of BOC volts is used, as for the above analyses, the probability of the large growth rates is diluted by the large number of indications at low voltages. A growth rate methodology is needad that maintains the frequency of large growth rates as a percent ofindications for BOC voltages above about 1.5 volts such that when applied in the projection analyses, the BOC indications above 1.5 volts left in service can grow proportionally to larger voltage indications at the EOC. This methodology is further developed in Sections 0.2 and 0.3 with revised projection analyses for Braidwood 1 using the recommended methods given in Section G.4. Additional support for the voltage dependent growth methodology is given in Appendix A which applies the methodology in conjunction with a voltage dependent POD as a "best estimate" analysis that eliminates potential compensation of errors associated with a POD of 0.6 in the methods assessment. The last completed cycle, Cycle 7B, for Byron 1 was a short cycle, and it was also the first cycle with 3.0v IPC implementation, with a resulting SLB leak rate of about 0.27 gpm based on the as measured voltage distribution. This leak rate is too small for evaluation of projection methods since a small change in the largest voltage prediction can change the leak rate by a large percentage. Therefore, only the Braidwood 1 F.,00 G results are used for the methods evaluation in Section 6.4. 6.2 Recommended Bobbin Voltage Growth Guidelines Alternate methods were assessed for incorporation of voltage dependent growth into the ARC projection methodology. The alternatives evaluated are partially described in the results given in Appendix A which describes the methodology requirements for developing voltage dependent growth rates. The resulting recommended guidelines, when applying a constant POD of 0.6, for incorporating voltage dependent growth into the projection analyses are described in this section. The re:ommended methods differ slightly from applications with a voltage dependent POD which are described in Appendix A. This difference was found to be necessary as conservative application of voltage dependent growth combined with a conservative POD of 0.6 results in conservatism in the projection methods. The difference between the two methods is associated with defining the width of the highest voltage dependent growth bin. The recommended methods for application with POD = 0.6 are described below. A hybrid growth distribution should be created by assuring that the limiting SG growth includes the largest 3 growth values found in all SGs o NNSD15\\apcNece97\\ccec690d wp5 63

1 l This provides confidence that the largest growth tail is included in the limiting SG Voltage bins should be approximately 0.4 to 0.0 volt wide with the voltage bin widths increased if necessary to include a minimum of about 200 indications (GL 95 05 requirement) in all bins The highest voltage bin should include at least one of the largest three growth values The highest voltage bin, lower voltage boundary can be decreased moderately if a small voltage change (0.1 to about 0.2 volt) would include one or more of the largest 3 to 5 voltage growth rates in the bin while not increasing the population by more than about 50 indications The lowest voltage bin can be increased in width if a plot of AV vs. BOC volts indicates growth independent of BOC voltage over a wider voltage range The binned cumulative probability distributions should show increased voltage growth at least above 80% probability as the BOC voltage increases. If not obtained, adjust voltage bin boundaries (typically increasing lower voltage bin widths) to obtain increased growth trend between bins. Plots of growth versus BOC volts should be used to adjust the voltage bin widths if necessary to obtain the increasing growth trend. If reasonable adjustments to the bin widths do not show a significantly increasing growth trend, the dependence of growth on BOC volts would appear to be negligible and a single bin for growth is acceptable. The voltage growth for all BOC voltages bin should be normalized by cycle duration. The voltage growth distributions for each bin should be applied to the same voltage range of BOC indications as used to develop the binned growth distribution. An exception is applied for the largest voltage growth bin for which the growth distribution is applied to all BOC indications above the lower voltage boundary of the bin. The higher voltage exception is necessary as the BOC voltage distribution may have larger voltage indications than were nvailable to develop the growth distribution, particularly after applying the POD of 0.G. n:\\NSD15\\ ape \\cce97\\ccec690d wpS 64

t 6.3 Voltage Dependent Growth Rate Distributions for POD = 0.6 Following the guidelines of Section 6.2, Braidwood-1 Cycle 6 growth rates can be combined into the following three voltage bins for use with a constant POD =0.6: Bin # Voltare Range No; of Indications 1 Up to 0.7 volts 1230 2 0.7 to 1.1 volts 665 3 Over 1.1 volS 210 t . Cumulative probability distributions for the above 3 bins are shown in Figure 6 2, j and as required by the growth _ selection guidelines, these bins show an increasing _ growth trend above CPD of about 80% The significant increase in probability of large growth rates in the bin above 1.1 volts is clearly seen in the figure. . In order to further assess the BOC voltage dependency on growth rates, the recent Byron-1 growth-data was also reviewed. Byron-1 Cycle 7B growth data can be combined into the following three voltage bins for use with a constant POD =0.6: i-Bin.# Voltare Rance No. of Indication 1 Up to 0.5 volts 826 2 0.5 to 0.9 volts 657 3-Over 0.9 volts 264 Cumulative probability distributions for the above three bins are shown in Figure 6-3, and as required by the growth selection guidelines, these bins show an increasing growth trend above CPD of about 80% The following two voltage bins were used to represent Braidwood-1 Cycle 5A growth rates when a constant POD =0.6 is applied: Bin 3.- .Voltare Range No. of Indication 1 Up to 0.7 volts 821 2 Over 0.7 volts - 220 l Cumulative probability distributions foribe above two bins are shown in Figure 6-4, and as required by the growth selection guidelines, these bins show an increasing growth trend above CPD of about 80%

a:\\NSD15\\apc \\ece97\\ecec690d.wp5 65 i

--,--w, ,,e.- re,-, ,,v- .. ~. - ->.,-v r w r ,-w- - -- c-e. -e.. e e-e-

GA Braidwood-1 Analyses for Benchmarking Alternate Voltage Projection Methods A number of analyses were performed to assess alternate methods for defining the widths of voltage bins. These analyses led to the guidelines of Section 6.2 for derming the voltage dependent growth rates including the bin widths. More details on these analyses are given in Ap}>cndix A for which analyses focused on growth rates in combination with a voltage dependent POD. The most important conclusions from these analyses were that the largest bin should be limited to about 200 indications for use with a POD of 0.6 but must include at least one of the three largest growth values and that growth rates should continue to be represented as absolute voltage differences rather than percent growth. If the largest voltage bin width must be increased significantly above about 200 indications to include one of the largest growth values, this is indicative of a reduced dependence of growth on BOC voltage and the resulting probability of a large growth rate is appropriately reduced due to tne larger number ofindications in the bin. The following summarizes comparisons of Braidwood 1 EOC 6 projections with as measured results. Am>lication of Recommended Groteth Rate Methodoloey Table 6 2 provides comparisons of SLB leak rates and burst probabilities between that calculated from the actual EOC 6 voltage distribution (Reference case in table) and projections (Cases 1 and 2) based on the recommended growth rate methodology. Only the SLB leak rate results for IRB conditions are applicable to the Braidwood 1 and Byron 13.0 volt IFC since hot leg tube bursts are prevented by the presence of the TSP, Free span leak rate calculated per GL 95 05 are also included in Table 6 2 to facilitate a broader methode comparison. Case 1 applies the actual Cycle 6 voltage dependent growth distributions to the EOC. 6 projections with the use of POD = 0.6. The leak rates for the projections are conservative by 1.6 and 3.3 gpm compared to the values calculated from the actual distributions with the GL 95 05 NDE uncertainty of 10.3 percent or the more likely 5% NDE uncertainty at the higher voltages. These results support the methodology in that conservative projections are made with POD = 0.6 when the growth rates are known. It can be noted from Table 61 that this conservatism was not obtained when voltage independent growth rates were applied. Case 2 provides the EOC-6 projections when the most limiting growth rates for the prior two cycles are used for the projection analyses per the ARC requirement. The population ofindications above about 1.5 volts, where voltage dependent growth is o \\NSD15 Nape \\cce9hc ec690d wp5 66

i more significant, increased significantly at the start of Cycle 6 compared to Cycle 5A due to implementation of a 3 volt IPC. This increase in the higher BOC voltage population resulted in the increased dependence of growth on BOC voltage and the need for the methods assessment given herein. Thus, it would be expected that the use of Cycle 5 growth values would result in an underprediction of the EOC G voltage distribution. The projected EOC G leak rate is underestimated by about three gpm as shown by the difference between the reference case and Case 2 results. This difference is significantly smaller than obtained for the Table G 1 results using Cycle 5A growth rates independent of voltage. Overall, the Case 2 results support the recommended methodology. Figure G 5 compares the actual EOC 6 voltage distributions with the projections obtained from Case 1. The agreement in Figure G 5 based on application of voltage dependent growth rates is significantly improved over that of Figure G 1 obtained with growth independent of voltage. G.5 Conclusions Based on the comparisons ofleak and burst results calculated from as measured and from projected voltage distributions given in this section, it is concluded that incorporation of voltage dependent growth rates is necessary for the Braidwood 1 and Byron 1 SGs with an ARC repair limit of 3 volts. The recommendations of Section G.2 above for developing voltage dependent growth rates provide good agreement between projections and as measured voltage distributions with associated SLB leak rates and tube burst probabilities. The recommended methods are based on appropriately r'epresenting the increased frequency of large growth values with increasing BOC voltage, particularly for BOC voltages above about I volt. The methodology recommendations can be summarized as follows: Voltane Gmwth Rates Application of voltage dependent growth rates based on dividing the total BOC voltage range into bins following guidance (Section 6.2) on the number of indications in the bins and developing separate cumulative probability distributions for the growth rates in each voltage range (bin). o.NNSD15\\npc\\cce97\\cccc6%i wp5 67 ~ _ - - - - _ - _ _. _. _.. _ - _ _ - -.. _. _ _ _ _.. _.. -

. - - - -. ~. - _. - Developing growth rates for the_ limiting SG as a hybrid population that incit:_es the three largest growth rates found in all SGs and. applying the resulting limiting SG, voltage dependent growth rates for all SGs.- 'For the EOC projection analyses, the binned cumulative probability distributions' for voltage growth are applied to the BOC indications over the individual voltage ranges used to develop the growth distributions. Analyses of As-measuned Voltaae Distdbutions (Condition Monitarirgg) The NDE uncertainty for analyst variability should be reduced for indications i above about 2 volts relative - to the 10.3% developed for indications predominantly less than 2 volts. Further statistical NDE uncertainty analyses are, however, required to finalize an NDE uncertainty for analyst variability that varies with the measured voltage magnitude. } i I i d cr\\NSDIS\\apc\\cce97\\crec690d.wp5 68 r.a v. .. ~, r

Tchla 6-1 I Braidwood Unit-1 EOC-6 Outage Comparison of SLB Leak Rate Results Based on Actual and Projected Voltage Distribution SLB No.of Max. Leak SG Analysis Method POD Indications

• Volts
• Rate gpm*

EOC - 6 PROJECTIONS

• A EOC 511 Projections, 0.0 1679 8.2 5.61 Cycle 5A Voltage independent B

growth distribution 0.0 916 7.9 2.35 per OL 95 05. C Estimated EFPD = 493 0.6 2446 8.4 6.99 D 0.0 1769 8.3 0.2 ye e SA Voltage independent C 0.0 2382 7.2 3,7 growth, actual EFPD = 378 Cycle 0 Voltage independent C 0.6 2382*' 9.7 7.8 growth, actual EFPD = 378 Y'I' ^

• • "' " 'W " '"*

I C 5 2038) 7.1 2.7 growth, actual EFPD = 378 POPCD Cycle O Voltage mdependent EPRI C 2038' 9.6 7,1 growth, actual EFPD = 378 POPCD EOC-6 ACTUALS A 1 1754 10.2 l 6.39 11 1 801 5.8 0.36 NDE uncertaintien at 10.3% per C OL 95 05 1 2098 11.9 11.5 D 1 2099 9.7 6.99 hDE uncertainties at best C estimate 5% for indications 1 2098 11,3 9.8 above about 2 volte Notes: (1) Number ofindications adjusted for POD. (2) Voltages include NDE uncertainties from hlonte Carlo analyses and exceed measured voltages. (3) Projections exclude indications repaired at mideych inspection. (4) Leak rates at room temperature (gpm). (5) Excludes indications removed from service during 10/96 top of tubesheet inspection. o:\\NSD15Napc\\cce97\\ccec690d wp3 G.9

. _ ~ - - - 4 Table 6-2 Braidwood-1 Summary of EOC-6 SLB Leak Rate Projections Voltage Projection Methods Using Voltage Dependent Growth Rates SG - C HL Indications Only Based on Actual Cycle Length (378 EFPD) Growth Applied to BOC-6 SLB Leak Case Volts and Growth Voltage No. Max. Rate No. Bin Widths POD of Volts With Free Comments (Note 3) Ind.* IRB Span Actual EOC-6: "Best Estimate" for Comparison with Projection Methods 10.3% NDE Ref. N. A. 1 2098 11.9 11.5 9.3 unc,per GL 95-05 Ref. N. A. I 2098 11.3"> 9.8 8.0 5% NDE unc. Projected EOC-6: Recommended Growth Methodology of Section 6.3 1 Cycle 6 Volt Dependent. 0.6 2382 10.3 13.1 10.6 Fig.6 2 05.7, 0.7 1.1, > l.1 growth 2 Cycle SA Volt Dependent. 0.6 2382 7.9 8.5 6.4 Fig. 6-4 Os0.7, >0.7 growth Notes: (1) Number of indications adjusted for POD. (2) Voltages include NDE uncertainties from Monte Carlo analyses and exceed measured voltages. (3) Growth rates of Cycle SA adjusted to Cycle 6 based on days at temperature except as noted. 1 l o \\NSDIS\\apc\\cce97%ccec690d wp5 6 10

..=. - Figure G 1 Braldwood Unit 1 Comparison of Predicted and Actual Bobbin Voltage Distributtors for SGC, EOC. 6 Grow 1h Rates Independent of BOC Voltage Steam Generator C. 3 volts & over only 6 O Actual EOC4 4 E Predicted EOC4, EPRI POPCD Cycle 6 Growth 3 l' i e i t .Ia....illlIllii,..,< ..ui I Bobbin Voltage Steam Generator C. 3 volts & over only s. 1 O Actual EOC4 4 5 Predicted, POD = 0.6 Cycle 6 Growth e i-3 1 1-sl ll. li.. ..illl i l i i. .i Bobbin Voltage -wm.~ 6-11

Figure 6-2 Braidwood Unit -1 Cycle 6 Ilybrid Growth - Normalized Using Time at RCS Temperature > 500 F Cumulative Probability Distributions for Use with POD =0.6 1.0 Md-8:dMM 2 " -. x i-r-534 ,,g M

• ,x-x-r-

.x -x. x - A * * * *' *'*' ~ 0.9 - -y[ x 0.8 - -Mi ,,.x-*.x- 'f / O

1. g'

/,d ,x-n/ .D' = 0.7 - o E 'f

• t 3

's o6-t Up to 0.7 volt (1230) c f e k x a II .a 0.5 - t - c- 0.7 to 1.1 volts (665) 7 a i c, -- r --Over 1.1 volts (209) 5,, 0.4 - -_x a B U 0.3 - - *-- SG-C Only (All volts) = l 1 O.2 - g i 1gr 0.1 - .e .X/ 0.0 " e --m , e = e e. m m = e-4 < m e. -o c. m ,m e e- =,--c. m, e c,~ = e ~ - n. m. 9 9 9 9 o o o o o o o o o c. e e e. e. m m m m e m , m m m e e-Voltage Growth 6-12 e#. w i.. x

Figure 6-3 ' Byron Unit -1 Cycle 7B liybrid Growth - Normalized Using Time at RCS Temperature > 500 F Cumulative Probability Distributions for Use with POD =0.6 g h m.7_TJ _, _a__ - e - 8;,,E.:;_ pire-e : :: 1.0 ,3,4 ,.A 0.9 - y f .E a .= 0.8 - g" 5 F g' ,n g 0.7 - 4.g-p = E 5 0.6 - g ( = / e $ 0.5 - A Up to 0.5 volt (826) W 5 e g -E o.4 2 - c- 0.5 to 0.9 volt (657) 2 .f m n 8 n / j - e-- Over 0.9 volts (264) U 0.3 - p p / / / p I- -- * -- SG-B Only (All volts) 0.2 - /P N/ O.1 - 1 2 O.0 'E, ;. l y4QQ5 EEEo E$EE$ 22232325I"?UOZU%0%%??%003%2O Voltage Growth c w,,.a w,,i s.. 6-13

Figure 6-4 Braidwood Unit -1 Cycle 5A Ilybrid Growth - Normalized Using Time at RCS Temperature > 500 F Cumulative Probability Distributions for Use with POD =0.6 1.0 i i i Q' 4.D-C D ~ ,y..x j,,g,_j..... -x--m--a g i g o ~O' y & -O~ D'~ . x - D'U s 0.9 - y / o'o'o-o .x x' O.8 - x' ,D, 5 / c 0,7 - 2 o E i W j0.6- -- ( 2 i 5 / .c / 30.5-lo 2 / C - / u g Up to 0.7 volt (821) $ 0.4 - r 4 l ~ E f d 0.3.' - c-Over 0.7 volt (220) g / - x --SG-A Only (All volts) 0.2 _ 0.1 - / 0.0 2 m n - o - ~m m o n -e ~,,-e n . ~ - ~ m m n e n e n m e 4 4 o o o o o o e o n - a a - a - a - ~ a a a a n a m n n Voltage Growth i 6-14 CptegreFd eMSM RE S$ AM

Figure 6-5 Braidwood Unit-1 Comparison of Predicted end Actual Bobbin Voltage Distributions for SG C, EOC-6, Using Voltage Dependent Cycle 6 Growth, POD = 0.6 Steam Generator C - 3 volts & over only 7 6 O Actual EOC-6 5 m Predicted EOO-6, POD = 0.6 g Voltage Dependent Cycle 6 Growth ( 0 4 E o }3 E 5 i 2 l i 1 .......in....illli li in...._.,il ll e @ @ @ @ # >9 >7 a " >6 >* @ @ @ # # e9@@e6 i@@4? O O @ @ e? 8 8 @ @ @ @ @.@ e Bobbin Voltage l l i 6-15 PRDCOMP6 as Fag 6 5 St12S7 to 14 AM l

-= 7.0 DATABASE APPLIED FOR ARC CORRELATIONS ~.e database used for the IPC correlations that are applied in the analyses of this ort is the same as that described in the last 90. day report (Reference 10.1). Model sier specimen 5971 is excluded from the IPC date. base based on application of 'RI data exclusion criterion for very high voltage indications and concurrence by NRC. Braidwood.1 and Byron.1 pulled tube indications R2007, TSP 7 (0.37 volt) . I R10C42, TSP 5 (0.27 volt), respectively, are excluded from the correlation based EPRI data exclusion criterion 2a accepted by the NRC. Criterion 2a excludes . ications with burst pressures high on the voltage correlation if the maximum crack eth is less than 60% and there are less than 2 remaining uncorroded ligaments. .nt S pulled tube indication R27C41 is included in the leak rate correlation at a B leak rate of 2490 liters por hour consistent with NRC recommendations. - ith Texas pulled tubo data from 1993 and 1995 inspections are also included in the C databaw. The updated database is in compliance with NRC guidelines for lication aflesh rate vs. voltage correlations and for removal of data outliers in the inch tuising burst and leak rate correlations. The updated ARC database for 3/4" es is documented in Reference 10.4 and that database was used to perform the 3 leak rate and tube burst probability analyses reported here. The same database also applied for EOC.5B analyses reported in Reference 10.1. It is noted that in leak tests performed for the ARC database, leakage at prototypic SLB conditions mdensed and the flow rate is measured at room temperature, i y t V ~318\\apcVee97%ccws90d.wpS 7.} ~~ ,m-.-, .---,.--r ,-n.,-- ,y y .w.. ,,mi,y,- r,-

i 8.0 BOBBIN VOLTAGE DISTRIBUTION e i i This section describes salient input data used to calculate EOC bobbin voltage distributions and presents the results of calculations to project EOC.7 voltage i distributions. As discussed earlier in Section 5.0, growth rates for Cycle 0 are significantly higher than those observed for prior cycles, and the growth rate shows a strong dependency on the BOC voltage. The EOC 0 voltage projections presented in the last 00 day report (Reference 10.1) underestimated the actual EOC.0 voltage distribution. Therefore, th:: Monte Carlo methodology used to project EOC voltage distribution was revised to take into account a voltage dependent growth distribution. Evaluation of the updat9d methodology which includes a comp.trison of EOC.6 voltage distributions projected using a revised methodology with the actual measured voltage distribution is presented in Section 0 and Appendix A. 8.1 Calculation of Voltage Distributions The analysis for EOC voltage distribution starts with a cycle initial voltage distribution which is projected to the end of cycle conditions based on the growth rate and the anticipated cycle operating period. The number ofindications assumed in the analysis to project EOC voltage distributions, and to perform tube leak rate and burst probability analyses, is obtained by adjusting the number oficported indications to account for detection uncertainty and birth of new indications over the projection period. This is accomplished by using a Probability of Detection (POD) factor, which is defined as the ratio of the actual number ofindications detected to total number ofindications present. A conservative value is assigned to the POD based on historic data, and the value used herein is discussed in Section 8.2. The calculation of projected bobbin voltage frequency distribution is based on a net total number of indications returned to service, defined as follows. Nra ars = N / POD. N,,,,,,,a + N a,%,,, i

where, Number of bobbin indications being returned to service for the Nr,ars 2

= next cycle Number of bobbin indications (in tubes in service) identified after N, = the previous cycle Probability of detection POD = N,,,,,,,, _= Number of N, which are repaired (plugged) after the last cycle y o \\NSDI5\\ ape \\cce97\\ccec690d wp$ 81 1.

Na,,io,,,,= Number of N, which are deplugge 1 after the last cycle and are returned to service in accordance with IPC applicability. There are no deplugged tubes returned to service at the BOC 7. The methodology need in the projection of bobbin voltage frequency predictions is described in Reference 10.2, and it is the same as that used in performing similar predictions during the h st (EOC 5B) inspection (Reference 10.5). Salient input data used for projecting EOC 7 bobbin voltage frequency are further discussed below. 8,2 Probability of Detection (POD) Generic Letter (GL) D5 05 (Reference 10.3) requires the application of a constant POD value of 0.6 to define the BOC distribution for the EOC voltage projections, unless an alternate POD is approved by the NRC. A POD value of 1.0 represents the ideal situation where all indications are detected. The licensing basis analysis for Braidwood Unit 1 was carried out using a constant POD of 0.6. Another analysis was also performed using the voltage dependent EPRI POPCD distribution presented in Appendix B. A composite EPRI POPCD distribution presented originally in Reference 10.5 and included here in Table B 5 of Appendix B was used. This POD distribution is based on data from 15 inspections in 8 plants, and it represents the lower 95% confidence bound. 8,3 Limit!..g Growth Rate Distribution The NRC guidelines in GL 95 05 (Reference 10.3) stipulate that the more conservative growth rate distributions from the past two inspections should be utilized for projecting EOC distributions for the next cycle. Since it is evident from Figure 5 2 that growth rates for Cycle 6 are higher than those of Cycles 5B, Cycle G growth rate distribution is used to develop the EOC 7 predictions. The voltage-dependent growth distributions, developed in Section G.0 and Appendix A, were applied to predict EOC 7 voltage distribution. Those growth distributions are based on hybrid data containing all SO C growth data plus the largest growth found in SG A. Separate EOC 7 projections are required for the hot and cold legs since tube expansion to limit TSP displacement has been implemented only in the hot leg, Table 4 2 shows average and maximum growth rates by TSP elevation including the o \\NSD15\\ ape \\cce97%ccec690d wp5 82

cold leg. It is seen that total growth during Cycle 6 for the cold leg indications were below 0.3 volts while the largest growth for the hot leg indications was 8.3 volts. The same limiting growth distributions developed in Section 5 2 were applied to both hot leg as well as cold leg indications in all four S0s. Therefore, the predicted EOC 7 results are conservative, especially fc,r the cold leg indications. However, as only 33 cold leg indications were found in all four sos combined, their contribution to SLB leak rates is small. 8.4 Cycle Operating Period Two different definitions of cycle duration can be used to adjust growth rates applied to EOC voltage projections: 1) effective full power year used in the standard projection methodology, and 2) an effective year based on days of operation with the RCS temperature above 500" F. These two methods of scaling growth rates are compared in Appendix B, and it is concluded in general that both methods are comparable. Since the Braidwood 1 Sus will be replaced at EOC 7, days at temperature would provide a better representation of the cycle duration and therefore it was used in the EOC 7 projection analysis. Cycle 6 (actual) 413 days (1.131 yeara) > 500 F or 378 EFPD (1,035 EFPY) Cycle 7 (estimated) 475 days (1.30 years) > 500 F 8.T-Projected EOC-7 Voltage Distribuilon Calculations of the predicted EOC 7 bobbin voltage distributions were performed for en four Sus based on the EOC 6 distributions shown in Table 81. The bobbin voltage distributions are shown separately for hot leg and cold leg indications in Table 81 since tube burst analyses need only be performed for the cold leg indications (locked TSPs constrain rupture of hot leg indicationo). The BOC distributions were adjusted to account for probability of detection as described above, and the adjusted number of indications at BOC -i are also shown in Table 81. Licensing basis calculations were perfc>rmed using a constant POD of 0.0. Results for another set of calculations based the EPRI POPCD distribution are also shown. Voltage dependent growth distributions for a POD =0.0 are developed in Section 0.0 and for EPRI POPCD in Appendix A. The IPC voltage distributions projected for EOC 7 for all four SGs are summarized on Table 8 2. These results are also shown graphically on Figures 81 to 8 4. The predicted EOC 7 voltage distributions have a o NNSDissapc\\cce97\\ccec690d wp5 83

long tail which is a result c'the long tail in the Cycle 0 growth distribution used in the projections. Only 30 m heations were found on the cold leg side for all four Sus combined during the ECO / inspection, two of them were removed from service due i to tube repairs, and the total at FOC 7 is projected to be about 48 for a POD of 0.0. The results for the cold leg indications are shown separately in Table 8 2, but they are combined with the hot leg results in Figures 81 to 8 4 because of the relatively smaller population of cold leg indications. The licensing basis results which are based on a constant POD of 0.0 are more conaervative than those using the voltage-dependent EPRI POPCD. P 4 I 1 i i 9 l 4 1 4 i 4 o \\NSDIS\\apc\\cce97\\ccec690d mp5 84 r --w-rnmm,.. ve smaw w, r----,--e-wu r - - r r -z-m7-y- -n-- w, e -v-e .+ s-e---- --+%m s+ = -*- =,---.- ..r-wn==m-=---eiww--e-=m-ew=ss-------w--

-. - - ~. Table 81 (Sheet I of 4) Mrnidwcod Unit.1 April 1997 Actual EOC4 a.id POD. Adjusted ROC.8 Voltage Dielributions Used la SLH Leak Mate and Hurst Prosbility Calculations Steam Generator A bleen Generator il I.(K?. 6 IKX:. 7 IM

• 6 IHX'. 7 Vollett is h+n6f t Repelred POlM.6 iOPCD le hervice Repaired P(MMI.6 POPCI)

We lim Cold Ha Ceid He raid He cole Hai cole Hai rete He c.44 He Cold te n* ne se se ne te $4 se se te se t.de se se 01 0 0 0 0 0 00 0 00 OM 0 00 0 0 0 -0 0 00 0.00 04 0 00 02 10 0 0 0 16 67 0 00 29 4J 0 00 8 0 0 0 13 33 0 00 23.54 0 00 0.3 14 5 1 0 122.33 8 33 166 48 11.32 36 2 1 0 $9 00 3 33 80 48 4 53 04 153 4 5 0 250 00 6 67 284 08 7.56 73 2 0 0 125 00 3 33 141 71 3.78 05 IIS 2 1 0 301.33 3.33 292 99 3 24 99 3 1 0 164 00 $ 00 l$9 53 4 86 ,,,2j6 _,l_ _L ,__0_ 353 00 1 67 313 85 1 49 94 0 3 0 153 67 0 00 136 63 0 00 06 07 176 1 6 0 287.33 1.67 23518 1.37 87 0 2 0 143 00 0 00 117.22 0 00 08 144 0 7 0 233 00 0 00 18015 0 00 87 0 1 0 144 00 0 00 112 07 0 00 09 l$0 0 7 0 243 00 0 00 178 39 0 00 69 0 5 0 110 00 0 00 80 28 0 00 l 114 1 0 185 00 1 67 131 77 1.20 63 0 1 0 104 00 0 00 74 58 0 00 11 98 1 3 1 160 33 0 67 ll186 0 17 48 0 0 0 80 00 0 00 56 26 0 00 1.2 79 0 3 0 128 67 0 00 87 50 0 00 38 0 0 0 63.33 0 00 43.53 0 00 1.3 $1 1 I I $4 00 Off $6 44 0 13 21 0 0 0 35 00 0 00 23 65 0 00 14 $9 0 4 0 94 33 0 00 61.36 0 00 18 0 0 0 30 00 0 00 19 94 0 00 1.$ 35 0 0 0 $8 33 0 00 38 53 0 00 8 0 0 0 13.33 0 00 8 81 0 00 ~ 16 22 0 1 0 35 67 0 00 23 08 0 00 11 0 0 0 18 33 0 00 12 04 0 00 1,7 24 0 2 0 38 00 0 00 24 11 0 00 9 0 0 0 15 00 0 00 9 79 0 00 _ l, $_ 23 0 1 0 37.33 0 00 2387 0 00 3 0 0 O $ 00 0 00 3.24 0 00 19 20 0 1 0 32.33 0 00 20$2_ 0 00 5 0 0 0 8 33 0 00 $ 38 0 00 2 22 0 1 0 33 67 0 00 22.56 0 00 4 0 0 0 6 67 0 00 4 28 0 00 _.11_ _. I 3 _.( ... O _,0 _ 21 67 0 90,, 13 86,.,.1M,, ( _ 0_ 0 0 6 67 0 00 4 26 0 00 22 9 0 0 0 il 00 0 00 9 $$ 0(0 1 0 0 0 1 67 0 00 1 06 0 00 2.3 8 0 0 0 13.33 0 00 8 45 0 00 2 0 0 0 3 33 0 00 2 11 0 OL 24 0 0 1 0 14 00 0 00 8 46 0 00 1 0 0 0 1 67 0 00 1 05 0 00 _. 2 { j _0_ _,_0_ j _. ,l_5 M _0,,00_ ,y 42,,0 (o _l_ _ J. .,_0,__ _L ,,,1,6 7 ,,_9 M ,10$_ _p(W 26 1 0 0 0 1 67 0 00 1 04 0 00 1 0 0 0 1 67 0 00 1 04 0 00 27 8 0 0 0 13 3) 0 00 8 29 0 00 1 0 0 0 1 67 0 00 1 04 0 00 28 5 0 1 0 7 33 0 00 4 16 0 00 0 0 0 0 0 00 0 00 0 00 0 00 29 5 0 0 0 8 33 0 00 $ 14 0 00 0 0 0 0 0 00 0 00 0 00 0(c 3 4 0 0 0 6 67 0 00 4 09 0 00 0 0 0 0 0 00 0 00 0 00 0 00 31 5 0 0 3 33 0 00 0 09 0 00 2 0 2 0 1.33 0 00 0 04 0 00 32 1 0 1 0 0 67 0 00 0 01 0 00 0 0 0 0 0 00 0 00 0 00 0 00 33 3 0 3 0 2 00 0 00 0 03 0 00 1 0 1 0 0 67 0 00 0 01 0 00 34 2 0' '2 0 1.33 0 00 0 01 0 00 1 0 1 0 0 67 0 00 0 00 0 00 ~ 3$ 1 0 1 0 0 67 0 00 0 00 0 00 1 0 1 0 0 67 0 00 0 00 0 00 _36_ _0 0 0 0 0 00 0 00 0 00 0 00 1 0 1 0 0 67 0 00 0 00 0 00 37 2 0 2 0 1.33 0 00 0 00 0 00 0 0 0 0 0 00 0 00 0 00 0 00 38 0 0 0 0 0 00 0 00 0 00 0 00 0 0 0 0 0 00 0 00 0 00 0 00 39 1 0 1 0 0 67 0 00 0 00 0 00 0 0 0 0 0 00 0 00 0 00 0 00 4 1 0 1 0 0 67 0 00 0 00 0 00 0 0 0 0 0 00 0 00 0 00 0 00 i 1 eble Ceaumven on Meet 2 l 8*$ e a .imiaw

l ) P I l Table 81 (Sheet 2 of 4) Braidwood Unit.1 April 1997 i. Actual EOC.6 and POD. Adjusted BOC-8 Voltage Distributions Used in SLB taak Rate and Burst Proshility Calculations I steeni cenerseer A se.. e ce.erster a F00 6 BOC.7 EOC.6 DOC.7 Velease in $en ke Reps 6ted POD 46 M)PCD in $en ke Repaired POD 46 M)PCD his lig Cong Ilm Cold lim Cold lim Cold Hm Cold HM Cold Hm Cold lim Cold i le- $ide Side le lide Se se le Side $4 $ide $4 $4 Suse $4 i Table Continues from Sheet I j 2 4.1 2 0 2 0 1.33 0 0 0 0 0 0 0 0.00,0 0 0 4.2 0 0 0 0 0.00 0 0 0 0 0 0 0 0.00 0 0-0 i 4.3 0-0 0 0 0 00 0 0 0 0 0 0 0 0.00 0 0 0 44 1 0 1 0 0 67 0 0 0 0 0 0 0 0 00 0 0 0 l 4.5 l 0 1 0 0 67 0 0 T 0 G 0 0 0 00 0 0 0 46 0 -0 0 0 _0 _00 0 0 0 0 0 0 0 0.00 0 0 0 4.7 0 0 0 0 0 00 0 0 0 0 0 0 0 0.00 0 0 0 4.8 0 0 0 0 0 00 0 0 0 0 0 0 0 0 00 0 0 0 49 1 0 1 0 0 67 0 0 0 0 0 0 0 0 00 0 0 0 5 0 0 0 0 0 00 0 0 0 0 0 0 0 0.00 0 0 0 3.1 2 0 2 0 1.33 0 0 0 0 0 0 0 0 00 0 0 0 5.2 -l 0 1 0 0 67 0 0 0 0 0 0 0 0.00 0 0 0 4 $.3 0 0 0 0 0 00 0 0 0 1 0 1 0 0 67 0 0 0 i 54 0 0 0 0 0 00 0 0 0 0 0 0 0 0.00 0 0 0

5. 5__

0 0 0 0 0 00 _0_ _ _0 _0 0 0 0 0 0 00 0 0 0 y $.5 1 0 1 0 0 67 0 0 0 0 0 0 0 0 00 0 0 0 5.7 0 0 0 0 0 00 0 0 0 0 0 0 0 0 00 0 0 0 $8 0 0 0 0 0 00 0 0 0 0 0 0 0 0.00 0 0 0 { 6 O_ _ O_ _ O_ 0 0 00 _ 0 _, _, 0_. O _0 0 0 0 0.00 0 0 0 6.1 1 0 1-0 0 67 0 0 0 0 0 0 0 0.00 0 0 0 62-0 0 0 0 0 00 0 0 0 0 0 0 0 0.00 0 0 0 6.3 - 0 0 0 0 0 00 0 0 0 0 0 0 0 0 00 0 0 0 6.4 0 0 0 0 0 00 0 0 0 0 0 0 0 0 00 0 0 0 4- .__6,,,5 C,,,_ _,,p_ 0 0-0 00 _0,,__ _ O_ 0 ,,,0,,,,,,_,p _ 0 0 0,_0j__ _0_,, _ _0__ _0_ 6.8 0 0 0 0 0 00 0 0 0 0 0 0 0 0 00 0 0 0 '7 0 0 0 0 0.00 0 0 0 0 0 0 0 0 00 0 0 0 7.4 0 0 0 0 0.00 0 0 0 0 0 0 0 0 00 0 0 0 7.3 0 0 0 0 0 00 0 0 0 0 0 0 0 0 00 0 0 0 7.9 0-0 0 0 0 00 0 0 0 0 0 0 0 0.00 0 0 0-86 1 0 1 0 0 67 0-0 -0 0 0-0 0 0.00 0 0 0 89 0 0 0 0 0 00 0 0 0 0 0 0-0 0.0u 0 0 0 9 l -0 1 0 0 67 0 0 0 0 0 0 0 0.00 0 0 0 9.2 0 0 0 0 0.00 0-0 0 0 0 0 0 0 00 0 0 0 99 0 0 0 0 0.00 0 0 0 0 0 0 0 0 00 0 0 0 4 Total -12 0 12 0= 8 00 0 0 0 1 0 1 0 0 67 0 0 0 > IV 12 0 12 0 8 00 0-0 0 1 0 1 0 0 67 0 0-0 a >2V 12 0 12 0 8 00 0 0-0 I O I O 0 67 0 0 0 86 ' e,au.iminm a m .l4. _.,.-.....-,U.i.,.,_,..---.,...m,

~__- -.--__ l l i Table 81 (sheet 3 of 4) Braidwood Unit.1 April 1997 Actual EOC-6 and POD Adjusted NOC 8 Voltage Distributions Used in SLH Leak Rate sud Hurst Prosbility Calculations i $leam Generator C Steam Generstot D I~OC.6 160C. 7 8.OC.6 BOC.7 hit *8e le henlee Repa6 red l'OtM4 POPCD la hentee Repaired POlbe4 POPCD E8 Hei red la cold Ha cold Has cold na c.14 Hai cold Has cold tw Co.4 se se se se se se se se $4 se se se $4 se se se 01 0 0 0 0 0 00 0 00 0 00 0 00 1 0 0 0 1.67 0 00 4 21 0 00 02 5 0 0 0 8.33 0 00 14 71 0 00 12 0 0 0 20 00 0 00 35.31 0 00 03 44 3 2 0 71.33 5 00 97 58 W 92 0 4 0 149.33 0 00 204.22 0 00 04 138 1 3 0 227.00 1 67 257.74 1.89 214 2 5 0 351.67 3.33 399 il 3.78 05 199 0 7 0 324 67 0 00 315 69 0 00 266 1 6 0 437.33 l.67 425.34 1 62 . 06 217 0 8 0 353 67 0 00 314 33 0 00 278 0 5 0 45833 0 00 407.04 0 00 07 216 0 3 0 357 00 0 00 292 99 0 00 227 0 3 0 375.33 0 00 308 07 0 00 08 212 0 3 0 350 33 0 00 272 53 0 00 184 0 2 0 304 67 0 00 237.14 0 00 09 194 0 5 0 318.33 0 00 234 77 0 00 186 0 3 0 307.00 0 c0 226 88 0 00 1 158 0 2 0 261.33 0 00 187.56 0 00 128 0 4 0 209 33 0 00 149.56 0 00 1.1 ill 0 1 0 184 00 0 00 129 10 0 00 100 0 2 0 164 67 0 00 115.21 0 00 _1{ li4 O_ 4 0 186 00 0 00 ,1,26,60 0 00 _ 76_ 0 3 0 123 67 _0 00 84 07 0 00 . l.3 93 0 3 0 152 00 0 00 101 75 0 00 77 0 2 0 126.33 0 00 84 73 0 00 14 69 0 2 0 l13 00 0 00 74 43 0 00 44 0 0 0 73 33 0 00 48.74 0 00 1.5 61 0 1 0 100 67 0 00 66 16 0 00 34 0 0 0 56 67 0 00 37.43 0 00 16 47 0 4 0 74.33 0 00 4743 0 00 27 0 0 0 4501 0 00 29 55 0 00 1.7 37 0 1 0 60 67 0 00 39 25 0 00 25 0 0 0 41 67 0 00 27.19 0 00 18 26 0 0 0 43.33 0 00 28 11 0 00 22 0 1 0 35 67 0 00 22 79 0 00 19 24 0 1 0 39 00 0 00 24 83 0 00 20 0 0 0 33.33 0 00 21.52 0 00 2 18 0 0 0 30 00 0 00 19.28 0 00 10 0 1 0 15 67 0 00 9 71 0 00 2.1 13 0 0 0 21 67 0 00 13 86 0 00 12 0 0 0 20 00 0 00 12.79 0 00 2.2 11 0 0 0 18 33 0 00 11.67 0 00 3 0 0 0 5 00 0 00 3 18 0 00 2.3 12 0 0 0 20 00 0 00 12 67 0 00 6 0 0 0 10 00 0 00 6.34 0 00 24 11 ( 0 0 18 33 0 00 11.56 0 00 4 0 0 0 6 67 0 00 4 20 0 00 25 8 0 0 0 13.33 0 00 8.37 0 00 5 0 0 0 8 33 0 00 5.23 0 00 26 2 0 0 0 3 33 0 00 2.08 0 00 2 0 0 0 3.33 0 00 2 08 0 00 2.7 5 0 0 0 8 33 0 00 1.18 0 00 3 0 0 0 5 00 0 00 3 11 0 00 28 4 0 0 0 6 67 0 00 4 13 0 00 1 0 0 0 1.67 0 00 1.03 0 00 29 7 0 0 0 11.67 0 00 7.19 0 00 5 0 0 0 8 33 0 00 $ 14 0 00 3 5 0 0 0 8 33 0 00 S il 0 00 2 0 0 0 3 33 0 00 2,05 0 00 31 3 0 3 0 2.00 0 00 0 05 0 00 3 0 3 0 2 00 0 00 0 05 0 00 32 3 0 3 0 2 00 0 00 0 04 0 00 3 0 3 0 2 00 0 00 _0_04 0 00 33 J 0 2 0 1.33 0 00 0 02 0 00 0 0 0 0 0 00 0 00 0 00 0 00 34 5 0 5 0 3.33 0 00 0 02 0 00 0 0 0 0 0 00 0 00 0 00 0 00 3.5 0 0 0 0 0 00 0 00 0 00 0 00 0 0 0 0 0 00 0 00 0 00 0 00 36 1 0 1 0 0.67 0 00 0 00 0 00 3 0 3 0 2 00 0 00 0 00 0 00 3.7 0 0 0 0 0 00 0 00 0 00 0 00 2 0 2 0 1.33 0 00 0.00 0.00 0 0 0 0 0 00 0 00 0 00 0 00 1 0 1 0 0 67 0 00 0 00 0 00 38 39 1 0 1 0 0 67 0 00 0 00 0 00 1 0 1 0 0 67 0 00 0 00 0 00 4 2 0 2 0 1.33 0 00 000 f 000 1 0 1 0 0 67 0 00 0 00 0 00 Tatde roetinues on Sheet 4 87 4 henswee landet.1(M 6119? l M 64

~__ 1 l l Table 81 (Sheet 4 of 4) Braidwood Unit l April 1997 Actual EOC-6 and POD Adjusted BOC 8 Voltage Distributions Used in SLB lask Rate and Durst Prosbility Calculations I seeen cewesser C seeen Gewreier D LOC.6 NOC.7 EOC.6 DOC.7 Voltage

1. hei o ke Repolted P0lHl.6 POPCD le benke Repelred POD =0.6 POPCD j

h lid Cold IW Cold Hot Cold IW Cold Hot Cold lid Cold Hot Cold IW Cold t to $4 14 bde te se se se se $ de se se se $4 se $4 j Table Coellaues from hbeet 3 41 1 0 1 0 0 67 0 00 0 00 0 00 1 0 1 0 0 67 0 0 0 I ~ 42 1 0 l 0 0 67 0 00 0 00 'J 00 0 0 0 0 0.00 0 0 0 43 0 0 0 0 0 00 0 00 0 00 0 00 1 0 1 0 0 67 0 0 0 44 0 0 0 0 0 00 0 00 0 00 0 00 0 0 0 0 0.00 0 0 0 45 0 0 0 0 0 00 0 00 0 00 0 00 0 0 0 0 0.00 0 0 0 46 1 0 1 0 0 67 0 00 0 00 0 00 0 0 0 0 0 00 0 0 0 _ 47 l. 0 t 0 0 67 0 00 0 00 0 00 0 0 0 0 0 00 O 0 0 48 1-0 1 0 0 67 0 00 0 00 0 00 1 0 1 0 0 67 0 0 0 49 0 0 0 0 0 00 0 00 0 00 0 00 0 0 0 0 0 00 0 0 0 0 0 0 0 0 00 0 00 0 00 0 00 2 0 2 0 1.33 0 0 0 __ $I __0_ __1._ _0_ _0 00_ 0 00 0 00 2 0 2 0 1.33 0 0 0 1 0 67 $.2 1 0 1 0 0 67 0 00 0 00 0 00 1 0 1 0 0 67 0 0 0 i $.3 0 0 0 0 0 00 0 00 0 00 0 00 1 0 1 0 0 67 0 0 0 $4 1 0 1 0 0 67 0 00 0 00 0 00 1 0 1 0 0 67 0 0 0 1 0 1 0 0 67 0 00 0 00 0 00 0 0 0 0 0 00 0 0 0 $6 0 0 0 0 0 00 0 00 0 00 0 00 1 0 1 0 0 67 0 0 0 $7 2 0 2 0 1.33 0 00 0 00 0 00 1 0 1 0 0 67 0 0 0 + $R 1 0 1 0 0 67 0 00 0 00 0 00 1 0 1 0 0 67 0 0 0 6 1 0 1 0 0 67 0 00 0 00 0 00 0 0 0 0 0 00 0 0 0 61 0 0 0 0 0 00 0 00 0 00 0 00 I 0. t 0 0 67 0 0 0 _6 2 _ _0 1 0 1 0 67 0 00 - _0 0 00 1 0 1 0 _0__67__0 00 0 0 63 1 0 1 0 0 67 0 00 0 00 0.00 0 0 0 0 0 00 0 0 0 64 0 0 0 0 0 00 0 00 0 00 0 00 1 0 1 0 0 67 0 0 0 6.5 1 0 1 0 0 67 0 00 0 00 0 00 0 0 0 0 0 00 0 0 0 68 0 0 0 0 0 00 0 00 0 00 0 00 1 0 1 0 0 67 0 0 0 7 0 0 0 0 0 00 0 00 0 00 0 00 1 0 1 0 0 67 0 0 0 a 74 1-0 1 0 0 67 0 00 0 00 0 00

)

0 0 0 0 00 0 0 0 7,5 1 0 1 0 0 67 0.00 0 00 0 00 0 0 0 0 0 00 0 0 0 79 I O I O 0 67 0 00 0 00 0 00 0 0 0 0 0 00 0 0 0 86 0 0 0 0 0 00 0 00 0 00 0 00 0 0 0 0 0 00 0 0 0 89 0-0 _0 _ _ O_ 0 00 0 00 0 00 0.00 1 0 1 0 0 67 0 0 0 9 0 0 _0_ 0 0 00 0 00 0 00 0 00 0 0 0 0 0 00 0 0 0 92 1 0 1 0-0 67 0 00 0 00 0 00 0 0 0 0 0 00 0 0 0 99 1 0 1 0 0 67 0 00 0 00 0 00 0 0 0 0 0 00 0-0 0 -- 10 $ 1 0 1 0 0 67 0 00 0 00 0 00 0 0 0 0 0 00 0 0 0 lotal 2l00 4 89 0 3411.00 6 67 2726 82

• 51 2099 3

74 0 3424.33 5 2924.2 5 40

• IV -

717 0 36 0 1139 00 0 00 738 91 MJ $11 0 42 0 809 67 0 526 19 0 >2V 117 0 39 0 156 00 0 00 81 97 '5 00 76 0 33 0 93 67 0 45 248 0 88 V Marys leMs4 I(4 M19112e PM < www ~ w ww - w ~ +_m.o,s,w r+~--- w+w--et-e-~wi w-e-~ v--,m--- =

• r~--------,w

--w- -m--+-,w-<m---,, e--m,-~m--,, 1 ,r~+ e r------ ,e~.e n nww-e=

Tcbb 8 2 (Sheet 1 cf 6) Braidwood Unit-1 April 1997 Voltage Distribution Projection for EOC - 7 Steam Generator A Steam Generator 8 POD EPRI POD EPRI 08 POD 0.8 POD Voltage Hot Log Cold Log Hot Log Cold Leg Hot Leg Cold Leg Hot Log Cold Log Projected Number oflndications at EOC 7 0.1 0 11 0 00 0 16 0 00 0 09 0.00 0.13 0.00 0.2 3 67 0 00 4 61 0 00 2 41 0 00 3 12 0 00 03 21.85 0.58 2552 0.79 11.71 0.00 14 40 0 31 04 61.88 1.46 71 44 1.88 31.75 0 64 37.12 0 82 0.5 116 69 2.97 134 62 3.74 59.33 1.79 6845 1.83 06 173 40 3.76 196 28 4 57 87.18 2.85 98 65 2.43 07 218 43 3 32 219 95 3.77 109.13 2.84 110 62 2.23 08 24055 2 62 214 80 2.86 120 35 1 66 108 46 1.73 09 236 02 2 11 198 09 2 28 118 94 1.00 100 65 1.22 10 215 06 1.61 177 28 1.68 109 42 0 52 90 80 0 82 1.1 190 61 1.19 151.96 1 20 97.17 0 69 77.84 0.53 1.2 165 41 0 91 127.90 0 89 83 85 0.15 65 20 0.23 1.3 139 70 0 73 10671 0.70 69 85 0 00 53 74 0.00 14 117 58 0 61 88 97 0 59 57.72 0.70 43 93 0.00 15 100.40 0 49 75.23 0 45 48.08 0.00 36.20 0.70 1.6 86.98 0.38 64 20 0 34 4045 0 00 30 00 0 00 1.7 76 47 0 31 55.10 0.28 34 49 0.00 24 99 0,00 1.8 67 94 0 28 47.29 0.26 29 67 0 30 20 91 0.30 1.9 60 04 0 22 40 48 0.20 25 32 17.41 2.0 52 54 0 13 34 57 0 01 21.23 14 44 2.1 45 69 0 00 _ _ 29 78 _ _ 0_00 17.,71 12.07 2.2 39 85 0 00 25 95 0 00 14 86 10 23 23 35 34 0 00 22.71 0.70 12.72 8 65 24 31 93 0 70 20 23 0 00 11.06 7.42 2.5 29 10 0.00 18 49 0 00 9.77 6 61 26 2649 0 00 16 64 0 00 8 67 5.84 2.7 23 91 0 00 14 69 0 00 7.56 5.01 2.8 21.15 0 00 13 01 0 00 6 37 4 29 _ _ 2.9 _ _18 68 _ 0;00 _ 11 66 _. 0.30 5 43 3 76 30 17.00 0 00 10 64 . _ _4 89 3.38 _31_ _ 1_5 57 _ 0.30 _ 9 72 4 44 _ 3.03 32 _ 33_ __14_17_ _ 8 99 _ 4 01 2.75 _1_2.82_ 8 40 3 59 _ 2.56 34 11.56 7.80 _ 3 22 2.36 3.5 10 55 7 05 2.96 2.12 36 9 65 6 10 2.77 1.82 37 _ 8 69 _ __5 20_ 2.49 1 52 3.8 7.49 4 31 2.09 1.21 39 6 23 3 45 1 63 0 90 40 5 11 2 68 1.25 0 62 41 4 39 2.08 1.07 0.43 42 3 96 1 69 1.00 0.33 43 3 57 1 63 0.93 0 39 Table continues on Sheet 2 ww u w t .w, g,9

4 Tcbla 8 - 2 (Sheet 2 cf 6) Braidwood Unit 1 April 1997 Voltage Distribution Projection for EOC 7 Steam Generator A Steam Generator B POD EPRI POD EPRI Voltage 0.8 POD 0.8 POD Hot Leg Cold Log Hot Leg Cold Leg Hot L Cold Leg Hot Leg CoM Leg Projected Number ofindications at EOC.7 44 3 30 1 83 0 88 0 57 45_ 1 93 _ 0 85 0 66 . 46_ 3 11 _ _ __ 1.76 0=79 0 60 2.89 47 2.72 1.41 _ 0 75_ 0 45 48 2 62 1.10 0.75 0.33 49 2.45 0 94 0 71 0 29 50 2.34 0 87 5.1 2.11-0 67 0 29 0 91 0 61 0 34 b2 1 83 1.11 0 57 043 5.3 1.74 1.37 0 53 0 55 54 1 88 1.41 0 63 0 57 55 2.17 1.29 0 72 0 50 56 2.26 1 06 0 80 0.39 57 2.26 0 84 0 82 0 28 58 2.19 0 73 0 80 0 23 59 2 03 0 79 0 69 0 26 60 1 98 1,14 0 67 0 41 6.1 2 03 1.50 0 73 0 56 62 2.26 1 71 0 83 0 64 63 2.30 1 91 0 83 0.70 _64_ _2j 8_ ._ 2_05_ 0 77 0 74 65 1 91 2.01 0 66 0 69 66 1 66 1 80 0 53 0 58 67 1 37 1.52 0 40 _0__4 5 _ 68 1.14 1.24 0 30 0 33 _ 60 0 96 1 00 0 24 0.24 _7 O_ _ 0 79 _ _ 0 79 _ 0.18 0.17 _3_ _ 0 15 7.3 0 50 0 43 0 09 0 06 7.4 0 43 0 36 0 08 0 04 _. _ 7:5_ _ 0 37 _ _0J9_ _ _ 0 06 _ _0_03_ __ 7.6 0.31 0.24 0 05 0 02 7 7._. _0 2_6_ __0j 9_ 0 04 0 01 1 75 0 23 0 15 0 04 0 01 7.9 0 27 0.11 0 06 0 01 80 0 38 0.12 0 12 0 02 8.1 0 $2 0 33 0.10 0.12 _82_,_0 79_ 0 55 0 31 0 23 8.3 1.16 0 68 0 49 0 30 84 1.37 0 78 0.57 0 32 86 1.35 0 85 0.51 0 32 86 1.14 0 89 0 41 0 31 Table continues on Sheet 3 v.4 u %,-, w ' 8 10-y- c y v .+~.-r-- .i-- 9 e e

Tcble 8 - 2 (Sheet 3 of 6) I Braidwood Unit-1 April 1997 Voltage Distribution Projection for EOC 7 Steam Generator A Steam Generator 8 j j POD EPRI POD EPRI 0.8 POD 0.8 POD yogg,,, h HotLeg Cold Log Hot Lag Cold Leg Hot Leg Cold Leg Hot Leg Cold Log i Projected Number of Indications et EOC. 7

8. 7 _

0 87 0 90 _ _ 0 30 0 29 88 0.72 0.88 0.21 0.27 89 0 65 0.83 0.16 0.23 90 0 55 0.75 0.12 0 20 91 0 42 0 66 0 10 0.16 9.2 0 42 0 57 0 08 0.13 1 9.3 0 36 0 48~~ 0.06 0.10 94 0 31 04'1 0 05 0 08 4 95 0.28 0 34 0 04 0.05 id 6 0 22 0 29 0 03 0 04 0.7 0 27 0.24 0 05 0 03 98 0.36 0 21 0.10 0 02 9.9 0 49 , 0.18 0.15 0 02 10 0 0 49 0.20 0.16 0 03 10 1 0 47 0.28 0.14 0 07 10 2 0 40 0 37 0.11 0.11 10 3 0 34 0 45 0 09 0.13 10 4 0 27 0 48 0 07 0.14 10 5 0 26 0 47 0 06 0 13 10 6 0 31 0.43 0.09 0 12 10.7 0 53 0.39 0 20 0.10 10 8 0.76 0 35 0.30 0 08 10 9 0 86 0 38 0 32 0.10 _-.i.1.0_ __0_82._ _0,52_... 0.29 0 15 11.1 0 71 0.73 0.12 0 22 11.2 0 58 0 90 0.00 0.28 11.3 0 49 1 00 0 00 0.29 11 4 0 43 1 01 0 70 0.28 11.5 . 0 37 0 95 0.00 0 10 11 6 0 32 0 84 0 00 0.00 11.7 0.27 0.73 0 30 0.70 11.8 0 23 0 62 0 00 11 9. 0 20 0 $2 0 00 12 0 0 16 0 43 0.30 12.1 0 04 0 36 __12:2 0 00 0 31 12.3 0 00 0 14 12.4 0 70 ' 12.5 0 70 12 8 0.30 13 0 0 30 24 68 2351.80 27,49 1314 02 13.14 1123 57 13.15 TOTAL 2831 32~ ~ i5~ ~i109O!I" 5 92 663.71 1.84 491.17 1.76 >1V ~T543 66 6 >3V 196 85 0 30 132 84-0 00 56 84 0 00 39 25 0 00 8-11 em n wn n o.m

Tcble 8 2 (Sheet 4 cf 6) Braidwood Unit-1 April 1997 Voltage Distribution Projection for EOC - 7 Steam Gen.ator C Steam Generator D POD EPRI POD EPRI 0.8 POD 0.8 POD Voltage Hot leg Cold Log Hot Leg Cold Leg Hot Leg Cold Leg Hot Log Cold Log Projected Numberofindications at EoC 7 0.1 0.06 0 00 0.36 0.00 0.36 0.00 0.63 0 00 02_ 1.98 0 00 4.78 0.00 4.78 0.00 6 48 0 00 03 -14 11 0 34 27.85 0 34 27.85 0.00 32.77 0.00 04 48 23 0 71 82.35 0.71 82.35 0 23 93 89 0.01 05 103 66 1 33 159 50 1.33 159 50 0 56 182.28 0.11 06 167.29 1.42 237.52 1 42 237.52 0.92 267.77 0 23 0.7 227.07 0 91 296.20 0 91 29620 1.00 299.27 0 32 08 26689 0 58 321 41 0.58 321 41 0 73 289 05 0 31 09 276.20 0.38 310 34 0.38 310 34 0 49 262.83 0.21 1.0 261 97 0.00 277.00 0 00 277.06 0.08 231 05 0.14 1.1 238 59 0 70 239 39 0.70 239 39 0 00 193 70 0.10 1.2 211 62 0 00 202.68 0.00 202.68 0.70 159 72 0L3 1.3 182 07 0 00 167.38 0.00 167.38 0.00 130$1 0 04 14 156 15 0 30 138 37 0 30 138 37 0 30 106 69 0.02 1.5 135 79 11652 116.52 88 68 0.02 1.6 119 48 99 61 99 61 74 39 0.01 ~ 1.7 106 01 86.36 86.36 62.90 0 01 18 94 57 75.72 75.72 53 35 0.01 1.9 83 51 65 90 65 90 45 06 0.01 2.0 72 47 56 28 56 28 37.61 0.01 2.1 62.34 47.58 47.58 31 69 0.00 2.2 53-78 ~ 40.50 40.50 27.09 0 00 2.3 47.11 35 18 35 18 23 26 0 00 24 41 95 31.11 31.11 20 31 0.00 2.5 37 98 27.73 27.73 18 33 0 00 2.6 34 60 24 71 24 71 16.36 0 00 ' f7-31 21 21.84 21.84 14.18 0 00 2.8 27.47 18 78 18 78 12.30 0.00 2.9 24 14 16.20 16.20 10.80 0 00 30 21.88 14 65 14 65 9.70 0.00 31 19 95 13 40 13 40 8.76 0 00 32 18 07 12.12 ~ 12.12 8 05 0 01 33~ 16 35 10 90 1000 7.51 0 01 34 14 80 9 82 9 82 7.01 0 02 3.5 1355 9 04 9 04 6 33 0.04 36 12,48 8 42 8.42 5 42 0 06 3.7 11.20 7 62 7.62 4 55 0 08 38 9 57 6 49 6 49 3.70 0.10 39 7.83 525 5.25 2.86 0.11 40 6.29 4.19 4.19 2.14 0.13 4.1 5.37 3 61 3 61 1.57 0.14 4.2 4 80 3.31 3.31 e 1.24 0.14 43 - 4 30 3 01 3 01 1.24 0.14 l Table continues on Sheet 5 atm sa m,w w mei Sa m s --e

Tchl] 8 - 2 (Sheet 5 cf 6) Braidwood Unit 1 April 1997 Voltage Distribution Projection for EOC - 7 Steam Generator C Steam Generator D POD EPRI POD EPRI Voltage 0.4 POD 0.4 POD Hot Leg Cold Leg Hot Leg Cold Leg Hot Leg Cold Leg Hot Leg Cold Leg Projected Number ofindications at EOC.7 A 44 4 00 2.77 2.77 1.53 45 3 64 2.65 2.65 1.73 0 00 46 3 67 2.52 2.52 1.62 0.00 4.7 3 53 2.42 2.42 1.30 0.70 48 $.48 2 42 2 42 1.02 0 00 49 3 27 2.30 2.30 0 91 0 00 ~ 03 5.2 2.30 1.83 1.83 1.17 53 2.18 1.76 1.76 1.47 54 2 43 1.92 1.92 1 49 55 2 80 2.22 2.22 1.32 56 3 11 2 40 2 40 1.06 5.7 3 28 2 54 2.54 0 80 58 3 13 2.38 2.38 0 67 59 226 2.23 2 23 0 75 60 2 86 2.09 2.09 1.15 6.1 2.97 2.35 2.35 1.57 62 3 28 2.53 2.53 1.76 6fi 3 29 2.53 2.53 1.93 64 3 10 2.36 2.36 2.04 65 2.80 2.08 2.08 1.96 66 2 42 1.79 1.79 1.70 07 2 00 1.47 1.47 1.37 68 1 66 1 22 1.22 1.04 _69_ _. _.1.36 _ _1.01...._ ,._1.01_ _ .,_ 0_77._ 7.1 0 97 0.73 0.73 0 43 72 0 78 0 64 0 64 0.33 7.3 0.75 0 55 0.55 0 27 7.4 0 66 0 47 0 47 0 21 7.5 - 0 59 0 42 0 42 0.17 7.6 0.48 0 31 0 31 0.14 7.7 0 42 0 31 0.31 0.11 7.8 0 46 0 26 0 26 0 09 7.9 0 52 0.32 0 32 0.07 80 0 67 0 41 0 41 0 10 _81 0 84 0 61 0 61 0 36 82 1.27 0 88 0 88 0 61 ~ '~ ~- ~ B5 i 99 1.45 ~ 1 ~0 92 1.45 , 86 1 69 1.19 1.19 0 93 Table continues on Sheet 6 emi u mn,=,,r ate. 6 m.=> 8-13

Table 8 2 (Sheet 6 cf 6) Braidwood Unit 1 April 1997 Voltage Distribution Projection for EOC 7 Steam Generator C Steam Generator D POD t.PRI POD EPRI 6.6 POD 6.s POD Voltage Hot Leg Cold Leg Hot Leg CoM Log Hot Leg Cold Leg Hot Leg Cold Leg 4 Projected Number ofIndications at EOC.7 8.7 1 34 0 92 0 97 0 92 _$b 1 05 0.76 0 78 0 85 89 0 81 _0 61_ 0 61 . _ 0.76 90 0 86 0 46 0 48 0 64 91 0 59 0.37 0 37 0.52 9.2 0 50 0 33 0.33 0 di 93 0 39 0 27 0 27 0 33 94 0 36 0.23 0 23 _. 0 26 _ 96 0 32 0 17 0 17 0 21 96 0.29 0.17 0.17 0.18 97 0 32 0 20 0 20 0 14 98 0 50 0.32 0 32 0 12 99 0 71 0 44 0 44 0.11 10 0 0.76 0 49 0 49 0 15 10 1 0 72 0 45 0 45 0.25 10 2 0 58 0 39 0 39 0 37 10 3 0 $3 0 32 0.32 0 45 10 4 0 48 0 27 0 27 0 47 10 5 0 39 0 23 0 23 0 45 10 6 0 47 0 29 0 29 0 38 ,_ 1 El_._. 0 81_ _ J52_ 0 $2 0 32 10 8 1.16 0 78 . 0 78 0.27 10.9 1 31 0 88 _0_88_. _ 0 31 _ 11 0 1.27 0 83 0 83 0 47 _ 11.1 1.11 0 70 0.70 0.70 11.2 . 0 92 0 58 0 58 0 90 11.3 0 76 0 48 0 48 0 98 11.4 0 60 0 40 0 40 0 94 11.5 0 50 0 32 0 32 0.83 ti e 0 42 0 26 0 26 0 69 11.7 0 35 02 02 0 55 11 8 0 29 0.2 0 17 0 44 11.9 0 26 0.13 0.13 0 34 _ 12 0 0 21 0 00 0 00 0 28 -12.1 0 18 0 00 0 00 0 08 12.2 0 16 0 00 0 00 0 00 4 12 3 0 14 0 00 0 00 0.70 12.4 0 12 0.70 0.70 0.00 12 5 0.07 0 00 0 00 0.00 12 6 0 00 0 00 0 00 0.30 13 0 0 70 0 00 0 00 4__ _._q 30 __ 0.30 TOTAL 3411.03 6 67 3424 36 6 67 3424.36 5 01 2920.96 3 61

• 1V 2043 57 1 00 1706 99 1.00 1706 99 1 00 1254 94 2.28

>3V 260 85 0 00 180 50 0 00 180 50 0 00 118 31 1 99 .=n ,n n > g,j 4

JL --.-..--a..b +4 .sd4 --. J-n J. E. J., a M a_A _aiw.-, 4 9_.m... .ua a __-A._m. m 4.m a..i.-.s e tt Oli f(i . Ila l {t G li I 44 I'll 4 14 ( 48 $ 64 f lL ' fil t il .l4 t ot f el 4 01 4 08 iDi 1 04 ' ( 0i t el 40s 4 04 il .48 de '88 .$6 fg

ll 66 10 ll

'It it it it 8 5 U 58 is o e 'f1 69

49 gg

'll gg 14 14 18 ltd .ti I it n 43 (9 il it 5 'M e " T to l I EL 2 il 41 'll gg .fi ig 81 68 et g, II it II tt i I' l It I II 4( It at

.it i

3 :g = g Si it l 4 " i

N

= a = u i N 4g II gg 4 it i i E N 4 ~""" n

=== " ti u n I 1 s .o t ' %o 3 I I I l l e a a e n,.e,.,, R A E R a g a ~

.--_ ~.. - e 45 I let s ei ill 1 94 lb I" (b I" (D 16 g, 46 se II il f6 gg 88 It 88 Ik 66 38 et il D e is O B n = la og 88 l44 44 I8 ti II Il 4% la Il il l lE . s. n I' it il lN .,I e si 6, }3 i

1 l e

= n ,,I 1 .i n l' et it $0 (I te to is ri j 4f it

n

,n - u 'n If ,TT

i n

- it - in a

u g u

es g u u

q a

u a se p s.

o m .:::::= i. o n ie fI [ 5 is n o i. 1 d h, l I i t i i i-I I I I i i ~ a y y a g e n a e e - n -,p,,,a, e a =

.~ ..~_ __ ~_~.- ~.-.. -.-..~ I 0 te D' l . tli 8 88 i t. I 8' 4 la d ie inn t ai 8" t ai L et \\ su f at g 89s th tes W s te th O 9 th g 46 46 N ll N N tl d a n g n a g 10 58 o o a to it i. i. n id (( 14 E ti ti .) 61 o f I, '? ,s

(,

} .3 d55 a w si . n

n n

i i

o l,i.

.i

n

n

" u E

u gg t!

= n n n u =; ~ a a = A u u 'M =;l u u ~-- l O! ir .::All l.n ' o o %i lt ti ei t u. s. a I I I I 3 u i I l-Z ", i i 6 R R R 8 8 8 g g g-- - gx. aw.a s r v ,-,-e---- -ye e m -- t w t-- v --r.e .c .-.e. --y..

-<---,2h_+-,4-A- .4.-gsJ _,. s 4 a A aWmMA .a.sw..44__.-yed &A441.4--_- 4 a ..4_, ele a ta 4:4 d ai t se { tg e lie f oi t ai e st t il 00' gg: t e. 4 04 gli Ils (si 89i s g4 00 88 il il il (6 m f6 I to is ll el st it 5 5 0 e it O S (8 te O to il it 44 44 4. %i t. n g 4 gg i l ! l ,lb it l i i st g Il tt i e. 8' i gg l ' lt i i v. I l 7 j I lsi i l

M e

lgg l

si g

l " j:

ii

}t : "

it

= : j $ Il y it

st i

= u .:= is ,2 fi ""ul 4f M " Mf 84 M ml o i, ql to It j to j ti f is tt } j 5 y s. s. 8 I i i i I i i 'M ",. l l 1 l R9IEREggga a I I E R g g a g a

I 9.0 SLB LEAK HATE AND TUBE BURST PHOBABILITY ANALYSES 4 This section presents results of analyses carried out to predict leak rates and tube l burst probabilities for postulated SLB conditions projected to EOC 7 conditions. Comparison between EOC 6 SLB leak rates and burst probabilities predicted using the methodology updated to include voltage dependent growth rates against those based on the actual measured EOC 6 voltage distributions is presented in Section 6.4. The updated methodology based on voltage dependent growth rates described in Section 6.0 was applied for EOC 7 projections also. Since TSPs are locked by tube expansion (to support a 3 volt IPC), analyses were performed separately for the indication population on the hot leg and cold leg sides of each generator. Since cold leg side of tubes can be displaced relative to TSP during a SLB cvent, indications on the cold leg significantly contribute to tube burst probability analysis. SG C with the largest number ofindications oa the hot leg side is expected to yield the limiting SLB leak rate for Cycle 7, and SG A, with the largest number ofindications on the cold leg side,is expected to be limiting from the tube burst probability standpoint. In addition, probability c.f axial rupture was also calculated for the limiting SG (SG C). The methodology used is the same as that utilized for the SLB burst probability calculation except the correlation parameters for SLB burst pressure versus bobbin voltage were replaced with the correlation parameters for the axial rupture force versus bobbin voltage, which are shown in Table 311, 9.1 Leak Hate and Tube Burst Probability for EOC 7 Two sets of calculations were performed to predict leak rates and tube burst probabilities for Braidwood Unit-1 at EOC 7 conditions both using voltage dependent growth rates described in Section 6.0: 1) licensing basis calculations using the NRC required emtant POD value of 0.6,2) a second set using the voltage dependent EPRI POPCD distribution. Results of the EOC 7 predictions are summarized on Table 91. For the licensing basis calculation with a constant POD of 0.6, the projected limiuig EOC 7 SLB leak rate is 57.1 gpm; it is predicted for SG C. The limiting tube burst 4 probability, 8.0x10, is predicted for SG A which has the largest number of indications on the cold leg side. This cold leg burst probability is conservative since the hot leg growth distribution has been applied in the analyses. The bounding leak rate calculated with EPRI POPCD,54.9 gpm for SG C, is below the bounding leak rate predicted with POD =0.6. The limiting SLB leak rate value will be within the allowable SLB leakage limit for Cycle 7 pending NRC approval of a reduction of the o \\NSD15\\ ape \\ccc97\\cced90d wp5 9.]

-= Unit 1 RCS DF. L131 Technical Specification limit. The limiting tube burst 8 probability is within the NRC reporting guideline of 10. I 9.2 Axial Tensile Rupture Probability. The probability of axial rupturo due to cellular corrosion was also calculated for SG [ C which has the largest predicted EOC.7 leak rate. The methodology used is the l same as that utilized for the SLB burst probability calculation except the correlation parameters for SLB burst pressure versus bobbin voltage were replaced with the l parameters for axial rupture force versus bobbin voltage, which are shown in Table 3 11. Axial tensile rupture probability projected with a constant POD of 0.0 is 7.9x10. Even if the largest SLB burst probability calculated for cold leg indications, 4 8.0x 10-4 predicted for SG A, is added to the calculated axial tensile rupture probability, the total tube failure probability (8.8x10 *)is still more than an order of magnitude below the NHC reporting guideline of 10'8 for tube burst probability, i a I i i l 1 l l j. -e i i 1 - ANSDl5\\ ape \\cce97%ccec690d*p5 92 o ..,..,,,,..---,..-,.-,..-...,.c. ---c ,..,., -, - -, - -, ~....,,... - - - -.

? Table 91 Braldwood Unit-1 April 1997 Outage Summary of Projected Tube Leak Rate and Burst Probability for EOC-7 BOC Voltage Dependent Growth Applied - 250k Simulations 1 Hurst l>robability SLH Steam l'OD No.of Max. Leak i I Generator Indications

• Voltam 1 or More Rate 1 Tube Tubes spm LICENSING HASIS EOC - 7 l'HOJECTIONS WITH l'OD=0.6 a

Ilot Side 2831 13.0 Negligible

• Negligible
• 40.2 A

Cold Side 25 3.1 8.0x 10' 8.0 x 10 O.05 Total 2850 8.0x 10 ' 8.0 x 10 40.3 llot Side 1314 11.7 Negligible

• Negligible (8' 14.1 3

U Cold Side 13 1.8 < 4.0x10 < 4.0x10* lx104 4 < 4.0x 10* < 4.0x 10' 14.1 Total 1327 Ilot Side 0.0 3411 14.5 Negligible

• Negligible

57.1 C Cold Side 7 1.4 < 4.0x 10' < 4.0x 10' l x 10 Total 3418 < 4.0x 10* < 4.0x10' 57.1 llot Side 3424 13.4 Negligible

• Negligible
• 39.4 U

Cold Side 5 1.4 1.7x10" < 4.0x 10* lx16' Total 3429 1.7x10' < 4.0x 10' 39.4 ilot Side 2352 12.8 Negligible

• Negligible
• 43.0 A

Cold Side 27 2.7 3.2x 10 * < 4.0x 10' 1.5x10* Total 2370 3.2 x 10 < 4.0x 10' 43.0 llot Side 1124 12.0 Negligible

• Negligible
• 13.7 II Cold Side 13 1.7

< 4.0x 10* < 4.0x 10' lx10 4 Total 1137 < 4.0x 10' < 4.0x10 13.7 4 Ilot Side POPCD 2722 12.8 Neeligible* Negligible * -54.9 C Cold Side D 1.4 < 4.0x 10' < 4.0x10* 1x104 Total 2731 < 4.0x 10' < 4.0x 10* 54.9 Ilot Side 2021 12.6 Negligible

• Negligible
• 37.5 O

Cold Side 5 1.4 1.6x10s < 4.0x 10' lx10* Total 2920 1.6x 10 < 4.0x 10' 37.5 d2LU (1) Number of indications adjusted for POD, (2) Voltages include NDis uncertainties from Monte Carlo analyses and exceed measured voltag _ (3) Below 10(Reference 10.10) oANSDIS\\apc\\cce97\\ccecG90cl.wp5 93 -.. - ~.. - - - - -

4 Tchla 9-1 i Braidwood Unit-1 April 1997 Outage Summary of Projected Tube Leak Rate and Burst Probability for EOC-7 30C Voltage Dependent Growth Applied - 250k Simulations Hurst Probability SLH j Steam POD No. of Max. Leak i Generator Indications

• Voltam 1 or More Hate 1 Tube Tubes gpm LICENSING.HASIS EOC 7 PROJECTIONS WITH POD =0.6 liot Side 2831 13.0 Negligible

Negligible *' 40.2 ^ Cold Side 25 3.1 8.0x 10

• 8.0x 10 '

O.05 8.0x 10 8.0x 104 40.3 Total 2850 llot Side 1314 11.7 Negligible Negligible 14.1 U Cold Side 13 1.8 < 4.0x 10' < 4.0x 10' 1 x 10~' 4 < 4.0x 10' < 4.0x10+ 14.1 Total 1327 liot Side 0.0 3411 14.5 NegligibM8' Negligible 57.1 C Cold Side 7 1.4 < 4.0x 10* < 4.0x 10* l x 10 Total 3418 < 4.0x10* < 4.0 x 10 ' 57.1 110t Side 3424 13.4 Nenligible Negligible'8' 39.4 U 4 Cold Side 5 1.4 1.7x10' < 4.0x 10* 1x10 Total 3429 1,7x10' < 4.0x 10* 39.4 EOC. 7 PROJECTIONS WITil A VOLTAGE-DEPENDENT POD (POPCD) Ilot Side 2352 12.8 Negligible Negligible 43.0 A Cold Side 27 2.7 3.2x 10 < 4.0x10' l.5 x 10 4 3.2 x 10 * < 4.0x 10' 43.0 Total 2370 llot Side 1124 12.0 Negligible"' Negligible"' 13.7 U Cold Side 13 1.7 < 4.0x 10' < 4.0x 10' lx10' Total 1137 < 4.0x 10* < 4.0x 10' 13.7 liot Side POPCD 2722 12.8 Negligible *' Negligible"' 54.9 C Cold Side 9 1.4 < 4.0x10' < 4.0x 10' 1 x 10 Total 2731 < 4.0x 10' < 4.0* 10' 54.9 liot Side 2921 12.6 Negligible Negligible *' 37.5 0 Cold Side 5 1.4 1.6x10 < 4.0x 10' 1 x 10 8 i Total 2926 1.Gx10' < 4.0b10' 37.5 _ dQ1cA (1) Number ofindications adjusted for POD. I (2) Voltages include NDE uncertainties from Monte Carlo analyses and exceed measured voltages. (3) Below 10 * (Reference 10.10) o.\\NSDIS\\apcNece97\\ctec69od wp$ 93 ...~

10.0 REFERENCES

10.1 Westinghouse Report SG 90 02 002, "Braidwood Unit 1, Cycle G Interim Plugging Criteria Return to Power Report," Westinghouse Nuclear Service Division, February 1996. 10.2 WCAP 14277,"SLB Leak Rate and Tube Burst Probability Analysis Methods for ODSCC at TSP Intersections", Revision 1, Westinghouse Nuclear Services Division, December 1996. 10.3 NRC Generic Letter 95 05, " Voltage Based Repair Criteria for the Repair of Westinghouse Steam Generator Tubes Affected by Outside Diameter Stress Corrosion Cracking," USNRC Office of Nuclear Reactor Regulation, August 3, 1995. 10.4 lotter from N. J. Liparulo, Westinghouse Electric Corporation, to W. T. Russel, Office of Nuclear Reactor Regulation, Nuclear Regulatory Commission, CAW. 96 935, dated February 27,1996. 10.5 EPRI Report NP 7480 L, Addendum 1, " Steam Generator Tubing Outside Diameter Stress Corrosion Cracking at Tube Support Plates Database for Alternate Repair Limits," November 1996. 10.6 Westinghouse Report SG 95 01003, "Braidwood Unit 1 End of Cycle 5A Interim Plugging Criteria Report," Westinghouse Nuclear Service Division, January 1995. 10.7 Westinghouse Report SG 96 08 005, " Byron Unit 1 End of Cycle 7B Interim Plugging Criteria Report," Westinghouse Nuclear Service Division, August 1996. 10.8 Letter from B. W. Sheron, Nuclear Regulatory Commission, to A. Marion, Nuclear Energy institute, dated February 9,1996, 10.9 WCAP 140tG, "Braidwood Unit 1 Technical Support for Cycle 5 Steam Generator Interim Plugging Criteria," Westinghouse Nuclear Service Division, March 1995. o \\NSD15\\apc\\cce97\\ccec690d wp5 10 1

10.10 WCAP.14273," Technical Support for Alternate Plugging Criteria with Tube Expansion at Tube Support Plate Intersections for Braidwood.1 and Byron 1 Model D Steam Generators," Westinghouse Nuclear Service Division, February 1995. '( N o.\\NSDIS\\ ape \\cce97Vcec690d wp5 10 2

APPENDIX A DEVELOPMENT AND EVALUATION OF AN ALTERNATE VOLTAGE PROJECTION METHOD The need for an improvement in methodology to incorporate voltage dependent growth rates was identified in Section G.I. The preferred approach for methods development is to use "best estimate" analyses to assure that conservatism in one part of the analysis does not lead to an erroneous conclusion on the modified methods being evaluation. For ARC EOC projections, the key parameters are POD, NDE uncertainties and voltage growth. The best estimate for POD is the EPRI voltage dependent POD of Reference 10.5 developed from field experience with ARC applications. The EPRI voltage dependent POD is used for the methods development of this appendix. A best estimate for NDE uncertainties requires recognition that the analyst variability applied in ARC analyses per OL 05 05 was developed for bobbin voltages predominantly < 2 volts. Above 2 volts, the analyst variability is considerably smaller for the larger, well defined flaws. The reduced NDE uncertainty at higher voltages is particularly important for calculating leakage from the actual EOC distribution as this includes large voltage indications found in the inspection. A reduced NDE uncertainty is applied for calculating leak rates from the measured distribution to define " truth" for comparisons with projected leak rates. Combining these two changes to the OL 95 05 methodology provides the best framework for evaluating changes to the growth analysis methodology. This appendix describes projection methods and results evaluated to assess alternate methods and develops recommended changes to the ARC projection methods relative to the current NRC approved methods given in WCAP.14277, Revision 1 (Reference 10.2). The principal objective of this appendix is to develop an improved EOC projection methodology based on comparisons of projected SLB leak rates with those calculated from the actual EOC.G distribution. 1 A.1 Methods Evaluations Projections of EOC voltages utilize the distribution of indications found in the inspection, the distribution of repaired or plugged indications, a POD which may be constant or voltage dependent, an NDE uncertainty distribution and a voltage growth distribution. The distributions ofindications found in the inspection and the repaired indications are fixed values with no uncertainty and are not assessed for a methodology revision. The methods for defining POD, NDE uncertainties and voltage growth were assessed to improve the agreement between predictions and as measured o \\NSD1 A\\apc\\cce97Nccec6MM wp5 A1

values for the Braidwood 1 EOC 6 voltage distribution with associated SLBleak rates and burst probabilities. A constant POD of 0.0 is excessively conservative for high voltage iAntions and cannot be applied to assess potential methodology improvements as this could lead to a misleading methodology conclusion. It is important to apply the best estimate POD to assess methods improvements and it is necessary that the POD values or alternate methods account for new indications initiated in the projected cycle. The EPHI POPCD, which is a voltage dependent POD, presented in Section 10 accounts for new indications in that POPCD is based on indications found in the latest inspection that were reported in the prior field inspection. Thus, this POD definition is appropriate for ARC analyses and acceptable for a methods assessment. An EPRI POPCD distribution based on evaluations from many ARC applications has been developed in Reference 10.5. This EPRI POPCD is compared with plant specific POPCD for the Braidwood 1 EOC 5 inspection (based on indications found at EOC 6) in Section 4.3. It is found that the Braidwood 1 POPCD is in good agreement with the EPRI POD. Therefore, the EPRI POD can be applied for the methods evaluation analyses with the expectation that minimal uncertainties are introduced through the POD. The ARC NDE uncertainty with a standard deviation of 10.3% for analyst variability was developed using indications dominantly less than about two volts. Reviews of this uncertainty, such as obtained by comparing field inspection voltages for pulled tubes with independent analyses, indicate that this uncertainty is only moderately conservative for indications up to about two volts. However, for indications near two volts or larger, the NDE uncertainty is less than 10% and judged to be about 5% standard deviation although a formal uncertainty analysis has not been performed at this time. For Braidwood 1 and Byron 1 with a 3 volt repair limit, the indications left in service are < 3 volts when the EPRI voltage dependent POD is applied (POD reaches 1.0 at 3.5 volts). Therefore, for projection analyses, the NDE uncertainty of 10.3% does not introduce excessive conservatism when combined with a voltage dependent POD and is retained for the methods evaluation of this report. However, when SLB leak rates and burst probabilities are calculated from the as measured voltage distribution to define " truth" for comparisons with projections and these values are dominated by indications > 2 volts (Braidwood 1 and Byron 1 applications), the use of a 10% NDE uncertainty applied to the measured distributions is conservative. Therefore, to determine the "best estimate" SLB leak rates and burst probabilities as the defm' ition of " truth" for this methods evaluation, an NDE uncertainty of 5% is applied to the as measured distribution ofindientions while the 10.3% uncertainty is retained for the projections. o \\NSDIS\\ ape \\cce97\\cccc6ted wp5 A2

Based on the above assessments, the projection quantity with the most uncertainty - is the voltage growth distribution. The method using a single voltage distribution, as described in NRC GL 95 05, was developed based on low voltage indications (less than about 2 volts) for which it was shown that voltage growth is approximately l independent of the BOC voltage values. For the Braidwood.1 and Byron 1 voltage growth data, voltage growths for V30c less than about 1.0 volt remain essentially l independent of Vsoc. However, above about 2 BOC volts, the probability oflarge 4 growth values increases significantly indicating a BOC voltage dependence of the growth rates. This effect was not nearly as pronounced prior to Cycle G at Braidwood 1 when the population ofindications left in service was dominantly less than 2 volts. With two cycle application of the 3 volt repair limit, the voltages for i BOC indications progressively increased and it is now more apparent that the BOC voltage dependence of growth must be included in the projection analyses for j applications with repair limits above about two volts. It is identified in OL 95 05 j that voltage growth must be continuously evaluated to assess potential changes to voltage dependent growth rates. The methods evaluations of this report are based on this guidance after having identified for Braidwood 1 that growth rates have becomo BOC voltage dependent for indications above about 1.5 volts. The principal methods evaluation of this report is therefore development of a methodology for including voltage dependent growth rates in the projected EOC voltage analyses. The growth distribution methods evaluations in this report include the following: i 1) BOC Voltage Dependent Growth Rates Alternate methods for including voltage dependence in the growth rate distributions were evaluated. These assessments led to the guidelines for developing growth distributions given in Section A.3 below. Most of the projection analyses given in this section are based on application of these ] guidelines. A few additional analyses with variations in the development of voltage dependent growth rates are included as sensitivity analyses. These sensitivity analyses include variations in defining the voltage boundaries for the highest voltage growth bin and the use of percent growth in the upper voltage bin versus the recommended use of absolute voltage differences for the growth distribution. From the plots of voltage growth versus Vooc in Section 5, it is important to note - that the frequency of large AVs, as a percent of indications, increases with voltage but that the magnitude of the large AVs does not appear to strongly increase with voltage. Thus, the voltage dependent growth methodology must n\\NS!H 5\\apc\\cce97%ccec690d.m p5 A3 w-ry,' w w grp.rv9p=-ww ww+- -M--gr us'ya-.- 'g5 gyisw 4.,-+a, gi .'-,-9 -vg.p,y-p7p--g. pw"'ieiw*T-WW*U"'*r'e-'-'L'-rT*-M-DrTt-----+5 '---W-""-

• • 'D*"-T"4*W*'9'd-T-----"W"-TW-

-"-E-Ww' N9 w '-"-'-"'M-'8'----- W

emphasizo reproducing the probabilities oflarge growth as they increase with increasing BOC voltages. However, the similarity oflarge growth values over a broad range of BOC voltages supports um of absolute growth values rather than percent growth in developing the methodologv. 2) Scaling Voltage Growth Rates Based on EFPY or T;me at Temperature J The EPRI and GL 95 05 recommended method for scaling voltage growths for 3 varying cycle lengths is baced on scaling by effective full power years (EFPY). \\ For projection methods, the use of EFPY permits applying an upper bound to the ( cycle length based on the EFPY for the fuelloading and use of this basis permits application of a cycle length that would not be exceeded. The use of EFPY has bee-historically supported by successful tube integrity analyses. h 5-I'ime at temperature is en alternate and potentially preferred (when degradation C mechanism is strongly temperature dependent such as ODSCC or PWSCC) basis for sc.aling growth rates between cycles. However, it is difficult to predict days at temperature for a future cycle other than basing m assessment on prior operating history. To ecmpare the use of EFPY with days at temperature Sr t ( growth sate applications, it is appropriate to compare the ratios of these L quantities between successive cycles from prior history data. If the ratios used 'h to scale growth rates are comparable between EFPY and days at temperature, as generally expected, the use of EFPY would be the preferred growth scaling method since this quantity can be readily bounded for the projected cycle. A comparison of these ratios has been performed for prior Braidwood 1 and Byron-1 operating cycles and is given in Table A 1 and Figure A-1. These results show no significant dift'erence in the ratios of EFPY and days at temperatures. Therefore, both EFPY and days at temperature are acceptable methods for scaling vohage growth ratu When the EFPY used for growth scaling corresponds to the fuel loading limit, conservatism is usually included in the projection since cycle lengths are frequenJy found to be shorter than the fuel limit. This difference was partir"ilarly large for Braidwood 1, Cycle 6 for which 493 EFPD was used for the projections and the actual cycle length was found to be 378 EFPD. For the methods analyses of this cection, it is appropriate to use growth scaling based on days at temperature although the difference from EFPD is not large. As shown in Table A 1, the ratio of Cycles 6 to 5A EFPD is 1.39 and the ratio of days at temperature is 1.50. These ratios, which represent the largest difference in Table A-1, are used later to assess sensitivity ofleak rates to use of EFPD or days at temperature. m\\NSD15\\apc\\cce97\\ccec690d wp5 A-4

However, the r. ext operating cycles at Braidwood 1 and Byron.1 are unique in that the SGs will be replaced at the end of the next cycle. In this case, the EOC for replacement is scheduled to occur before the fuelloading EFPY is completed. Thus, a shorter EFPY or an estimated days at temperature is appropriate for the next operating cycle at Braidwood 1 and Byron 1. Days at temperature are applied for the next operating cycles up to replacement as this value is better defined than EFPY since replacement will commence on the scheduled date ] essoutially independent of the fuel usage or EFPY. A.2 Recommended Bobbin Voltage Ghwth Guidelines Alternate methods were assessed for incorporation of voltage dependent growth into the ARC projection methodology. The alternatives evaluated are only partially described in the analysis results given in this section. The resulting recommended guidelines for incorporating voltage dependent growth into the projection analyses are described in this section. The recommended methods differ slightly between applications with a voltage dependent POD and with application of a constant POD = 0.6 (guidelines given in Section G.2) as currently required by NRC GL 95 05. This difference was found to be necessary as conservative application of voltage dependent growth combined with a conservative, constant POD of 0.0 results in conservatism in the projection methods. The difference between the two methods is associated with defining the width of the highest voltage dependent growth bin. A.2.1 Voltage Dependent Growth Methodology Requirements Voltage dependent growth distributions are applied by developing different growth distributions applicable to increasing voltage ranges or bins of the BOC distribution A scatter plot of AV Growth versus Vooc is necessary to guide selection of the growth bin ranges To obtain the correct probability of large voltage indications, the number of indications in the voltage width of a growth bin used to develop the growth distribution should be similar to the number ofindications left in service over the bin voltage range This is necessary for application of a cumulative probability distribution of voltage growth to assure that the frequency of occurrence of large growth values is consistent between the population used to develop the growth distribution and the population of indications left in service. For example, o.\\NSD15\\apc\\cce97\\ccec690d wp5 A.5

if a population of 25 to 60 indications including a few of the largest growth values is used to develop a binned growth distribution and then applied to a much larger population of BOC indications, the frequency oflarge voltage EOC indications resulting from the projections could be excessivelv Conservative. Growth distributions summed over all SGs should not be used unless it is shown that the probability of large growth values is the same or more conservative than the limiting SG or there are < 200 totalindications in a SG For application of voltage dependent, binned growth distributions, t' T cumulative probability distributions (CPD) above about 60% to 80% should progressively show increasing growth trends If this does not etcur, the growth values are not significantly voltage dependent and voltage dependent growth is not required A voltage growth bin should have at least 25 indications for use with a voltage dependent POD (at least 200 indications for application with constant POD = 0.6) s This difference between about 25 or 200 indications in the largest voltage bin is the principal difference between applications with a voltage dependent POD and POD = 0.6. The corresponding guidelines for POD = 0.6 are given in Section 6.2. The recommended methods for application with voltage dependent POD are described below. A.2.2 Application of Growth Guidelines for Voltage Dependent POD Tho following guidelines are to be applied to develop voltage dependent growth astributions when applying a voltage dependent POD: A hybrid growth distribution should be created by assuring that the limiting SG growth includes the largest 3 growth values found in all SGs This provides confidence that the largest growth tail is included in the limiting SG Voltage bins should be approximately 0.4 to 0.6 volt wide with the voltage bin widths increased if necessary to include a minimum of 25 indications (but o:\\NSD15\\apc\\cce97\\ccec690d.wp5 A6

preferably ;< 50) in the highest BOC voltage bin and a minimum of-200 -indications in all other bins

The highest voltage bin should include 'at least one of the largest three growth values The highest voltage bin, lower voltage boundary can be decreased moderately if a salall voltage change (0.1 to about 0.2 volt) would include one or more of the largest 3 to 5 voltage growth rates in the bin while not increasing the population to more than about 50 indications The lowest voltage bin can be increased in width if a plot of AV vs. BOC volts indicates the frequency oflarge growth values is independent of BOC l

voltage over a wider voltage range The binned cumulative probability: distributions should show increased voltage growth at least above 80% probability as the BOC voltage increases. If not obtained, adjust voltage hin boundaries (typically increasing lower voltage bin widths) to obtain increased growth trend between bins. Plots of growth versus BOC volts should be used to adjust the voltage bin widths if necessary to obtain the increasing growth trend. If reasonable adjustments to the bin widths do not show a significantly increasing growth trend, the - dependence of growth on BOC volts would appear to be negligible and a single bin for growth is acceptablx Ifless than 100 indications are in a voltage bin (i.e., highest voltage bin), the maximum voltage growth in the bin should be increased if necessary to reflect 95% confidence on the upper bound using the extreme value theorem. The voltage growth for all BOC voltages bin should be expressed as AV The voltage growth distributions for each bin should be applied to the same voltage range of BOC indicarians as used to develop the binned growth - distribution. An exception is applied for the largest voltage growth bin for which the growth distribution is applied to all BOC indications above the lower voltage boundary of the bin. The higher voltage exception is necessary as the BOC voltage distribution may have larger voltage indications than were available to develop the-growth distribution. A.2.3 Steps to Developing Voltage Growth Distributions R Develop table of number of growth indications in 0.1 BOC volt bins 2- _=_ o:\\NSD15\\apc\\cce97%ccec690d.wp5 - A7

I Prepare plot of voltage growth vs. BOC volts that includes the hybrid growth for assuring that the three largest growth values from all SGs are included in the limiting SG distribution Define lowest voltage bin width at about 0.5 volt or higher if there is no apparent increase in the frequency oflarge growth values with increasing BOC volts Define highest voltage bin to include > 25 indications and at least one of the largest three growth values Split voltage range between lowest and highest voltage bins inte 0.4 to 0.6 volt bins with at least 200 indications in each bin using plot of voltage growth vs. BOC volts as a guide to select boundarios with an increasing trend for probability of large growth values Plot ('PD for each voltage bin to contirm trend between bins ofincreasing growth at CPD > about 80%. If not obtained with initial voltage bins, adjust voltage bin boundaries or bin widths to achieve increasing growth trend A.3 Braidwood-1 and Byron-1 Voltage Dependent Growth Distributions l)kgidicood-l Cvele 6 Voltace Denendent Gmwth Distribution for Voltare Denendent POD Examination of growth data shown in Table 51 and Figure 51 show that growth distributions for SGs A and C are very close to each other. However, SG C had 6 of the 10 largest growth rates observed for Cycle 6 (see Table 5 3). SG C had the top two growth values and SG A had the third largest growth. Therefore, SG C can be considered to be the limiting SG for Cycle 6 growth. Per the above growth selection guidelines, a hybrid population containing all SG C indications plus the largest growth found in SG A was considered. A plot of growth vs Vaoc for the hybrid population is presented in Figure 5 3. Up to about 0.7 volts, maximum growth amplitude does not appear to increase with BOC voltage. Therefore, per the guidelines, the lowest voltage bin width can be set between 0.5 to 0.7 volts. If the EPFI voltage dependent POD distribution is used for projections, the highest voltage bin should include between 25 to 50 indications and o:\\NSD15\\apcNece97%ccec690d.wp5 A8

i L at least one'of the largest three growth values. Selecting 1.6 volts as the lower limit for the highest voltage bin meets this requirement. The following four voltage bins.

were used to represent Cycle 6 growth rates:

Binj Voltare Ranee-No. of Indications 1 Up to 0.5 volts 734 2-0.5 to 1 volt 1059 3 1 to 1.6 volts 287 4 - Over 1.6 volts 25 Cumulative probability distributions (CPDs) for the above four bins are shown in Figure A 2. As the four distributions are progressively shifted to the right, they show an increasing growth trend among the bins. The shift to a higher frequency for large growth values is most pronounced for the highest voltage bin of > 1.6 volts which is partially due to including the small number ofindicationo in this bin. This strongly affects the EOC voltage projections by increasing the frequency of large voltage 1 growths for the largest indications left in service. Figure A 2 also shows the single SG C growth distribution developed per OL 95 05 guidelines, it is seen that the frequency of large growth is increased above 1.0 volt for the voltage dependent distributions compared to the single distribution methodology. Since the remaining two of the three largest growths,1.52 and 1.47 volts, are close i to the lower boundary chosen for the highest voltage tin in the above distribution, another distribution with the lower boundary for the higaest voltage bin moved down to 1.45 volts so that all three largest growths are in the same bin was also examined as a sensitivity case. The largest voltage bin in this includes 46 indications as shown below.. Bin # Voltare Ranee - No. of Indications 1 Up to 0.5 volts 734 2 0.5 to 1 volt 1059 3 1 to 1.45 volts 266 4 Over 1.45 volts 46 Cumulative probability distributions (CPDs) for the above four bins are shown in Figure A-3._ As the four distributions are progressively shifted to the right, they show an increasing growth trend among the bins. A comparison of Figures A 2 and A 3 shows that this difference in defining the highest voltage growth bin boundary does 1 -not strongly affect the distributions and the differences on projected EOC voltages o:\\NsD15\\apcNece97\\ccec600d.wp5 A.9

and leak rates would be expected to be minimal (shown to be a small effect in Section A.4) Bvmn-1 Cvele 7B Voltage Denendent Gmteth Distribution for Voltaat Devendent POD As another example of application of the above growth selection guidelines, Byron Unit 1 Cycle 7B data were evaluated. SG B had the highest average growth rate as well as the largest single indication growth observed for Byron 1 Cycle 7B; therefore, SG B had the limiting growth for Cycle 7B. SG C had the second largest single indication growth and SG B had the third largest single indication growth. Per the growth selection guidelines, a hybrid growth distribution containing all SG B data plus the largest growth found in SG C was considered. Figure A 4 shows growth vs V30c for this hybrid data. The frequency oflarge growth values as a function d BOC voltages is not as apparent as the Braidwood 1 Cycle 6 data of Figure 5-3. Larger growth values appear to have a higher relative frequency above about 0.9 volts than below 0.9 volts. Therefore, per the guidelines, the lowest voltage bin width can be set between 0.5 to 0.9 volts. The third largest growth value appears at about 1.8 volts, and the lower boundary for the highest voltage bin was set at 1.5 volts. Byron 1 Cycle 7B growth rates can be combined into the following four voltage bins for use with the EPRI POD distribution: Bin # Voltace Rance No. of Indication 1 Up to 0.5 volts 826 i 2 0.5 to 1 volt 744 3 1 to 1.5 volts 144 4 Over 1.5 volt 33 Cumulative probability distributions for the above four bins are shown in Figure A 5, and as required by the growth selection guidelines, these bins show an increasing growth trend above a CPD of about 80%. The shift to the right for the largest voltage growth bin for Byron 1 is found to be larger than found for Braidwood-1 in Figure A-2 although the BOC dependence for growth was less apparent than for Braidwood 1. B_rgidtvood-1 Cvele 5A Voltane Devendent Gmirth Distribution for Voltage Devendent POD As a part of the evaluation of the revised projection methodology based on voltage dependent growth rates, the EOC 6 projection for SG-C presented in the last 90-day report was reevaluated and compared against the actual measurements. Based on i o:\\NSD15\\ap sce97\\ccecG90d.wp5 A-10 'b

the GL 95 05 requirements, Cycle SA growth is the limiting growth for EOC-G projections. Therefore, growth selection guidelines were applied to develop growth distributions for Cycle 5A for use in the methodology evaluation. SG A had the highest average growth rate as well as the largest growth value observed for Cycle 5A and therefore SG A had the limiting growth for Cycle 5A. SG C had the second and third largest growth. Per the growth selection guidelines, a hybrid growth distribution containing all SG A data plus the two largest growth values found in SG C was considered. Figure A 6 shows growth vs Vuoc for this hybrid data. The growth amplitudes show less dependency on the BOC voltages in comparison to the Cycle 6 data. However, larger growth values have a higher relative frequency above about 0.7 volts than below 0.7 volts. Therefore, per the guidelines, the lowest voltage bin width can be set between 0.5 to 0.7 volts. The top two growths occur just above 1 volt, and there are over 25 indications larger than 1 volt. Therefore, the lower boundary for the largest voltage bin was set at 1 volt. The following three voltage bins were used to represent Cycle 5A growth rates when EPRI POPCD distribution is applied: Hin.# Voltace Rance No. of Indication 1 Up to 0.5 volts 571 2 0.5 to 1 volt 428 3 Over 1 volt 42 Cumulative probability distributions for the above three bins are shown in Figure A-7, and as required by the growth selection guidelines, these bins show an increasing growth trend above CPD of about 80%. The growth dependence on BOC voltage is seen to be considerable less than found for Cycle G. A.4 Braidwood-1 EOC-G Analyses for Evaluation of Alternate Voltage Projection Methods A number of analyses were performed to assess alternate methods for defining the widths of voltage bins and to assess the use of voltage growth based on absolute voltage differences or percent growth. These analyses led to the guidelines of Section A.2 for defining the voltage dependent growth rates including the bin widths and focused on growth rates in combination with a voltage dependcat POD. The most imporiant conclusions from these analyses were that the largest bia should be limited to including a target of 25 to <50 indications but must include at least one of the oANSD15\\apc\\cce97\\ccec690d.wp5 A 11

three largest growth values and that growth rates uhould continue to be represented as absolute voltage differences rather than percent growth Defining the largest voltage bin with a small number ofindications but at least one large growth rate leads to a high probability that one or more of the largest indications left in service will result in a projected EOC high voltage indication. If this bin width must be increased significantly above about 25 indications to include one of the largest growth values, this is indicative of a reduced dependence of growth on BOC voltage and the resulting probability of a large growth rate is appropriately reduced due to the larger number ofindications in the bin Percentage growth should not be considered for low BOC voltage (less than about 1.5 volts) indications as the percentage values are excessively sensitive to small errors in the BOC voltage. Therefore, the assessment of percent growth was linhed to the largest voltage bin. As discussed below, the use of percent growth results in large variations in values dependent upon whether the largest growth is toward the lower or upper end of the voltage bin range. Itis concluded that growth should be retained as absolute voltage differences for all voltage bins. The following summarizes comparisons of Braidwood 1 EOC G projections with as measured results. A.4.1 Application of Recommended Growth Rate Methodology Table A 2 provides comparisons of SLB leak rates and burst probabilities between that calculated from the actual EOC G voltage distribution (Reference in table) and projections (Cases 1 to 5) based on the recommended growth rate methodology. The SLB leak rate results with IRB (indication restricted from burst)is the only analysis applicable to the Braidwood 1 and Byron 13.0 volt IPC since hot leg bursts are prevented by the presence of the TSP. To facilitate a broader methods comparison, Table A-2 also includes the free span leak rate calculated per GL 95 05 although these results are not applicable to the Comed units. Table A 2 also includes the results for a constant POD of 0.6 which were previously described in Section G. These results are included to facilitate comparisons between the voltage dependent and constant POD results. Cases 1 and 5 apply the actual Cycle G growth distributions to the EOC G projections with use of the EPRI POD. These two cases differ slightly in the voltage boundary of the largest voltage bin with the difference permitted within the guidelines of Section A.2. The leak rates and burst probability for the projections are in good agreement with the values celculated from the actual distributions. These results support the methodology in that good projections can be made when the growth rates are known. It can be noted from Table 61 that this good agreement was not obtained when voltage independent growth rates were applied. o:\\NSD15\\apc\\ece97\\ccec690d.wp5 A-12

l Case 2 provides the EOC-6 projections when the ARC requirement is applied that growth rates from the most limiting of the prior two cycles are to be used for the projection analyses. Case 2A versus 2B shows the difference in projections between scaling Cycle 5A growth rates by days at temperature or by EFPD. The resulting differences in leak rates are modest. If the fuelloading of 493 EFPD rather than the actual 378 EFPD were used for the analyses, the effect on the analyses would be larger than obtained between the use of EFPD or days at temperature. The i population ofindications above about 1.5 volts, where voltage dependent growth is more significant, increased signiticantly at the start of Cycle 6 compared to Cycle 5A. This increase in the higher voltage population apparently resulted in the increased dependence of growth on BOC voltage and the need for the methods assessment given herein. Thus, it would be expected that the use of Cycle 5 growth values would result in an underprediction of the EOC-6 voltage distribution. However, this underestimate is modest and about one gpm as shown by the difference between the reference case and Case 2 results. This difference is significantly smaller than obtained for the Table 61 results using Cycle 5A growth rates independent of voltage. Overall, the Case 2 results support the recommended methodology. Figure A 8 compares the actual EOC 6 voltage distributions with the projections obtained from Cases 1 and 3. The agreement in Figure A 8 based on application of voltage dependent growth rates is significantly improved over that of Figure 61 obtained with growth independent of voltage. Analysis results for a POD of 0.6 are given by Cases 3 and 4. The methodology also differs from Cases 1 and 2 in that the largest voltage bin for the growth rates includes a minimum of 200 indications rather than the 25 minimum for use with a voltage dependent POD. Case 3 shows the expected result that when the number of indications and growth rates are reasonably well known, the use of a constant POD results in conservative projections. The Case 4 results using Cycle 5A growth rates are similar to Case 2 with the POD conservatism partially offset by the reduced conservatism in binning the voltage growth rates when a constant POD is applied. These results indicate the difficulties involved in drawing methods conclusions when a very conservative POD is included in the analyses. Sensitivity of Pmiections to Use of Perrent GmwLh Cases 6 to 10 of Table A-2 are the same as Cases 1 to 5, respectively, except that growth for the largest voltage bin is applied as percent growth rather than absolute change in volts. In some cases, the projected leak rates with percent growth tend to increase and other cases decrease relative to Cases 1 to 5. With percent growth, the o:\\NSD15Napc\\cce9T.ccec690d.wp5 A-13

\\ results are sensitive to whether the larger growth values occurred near the lower end or the upper end of the bin voltage range. For Cycle 5A, the largest growth occurred at the lowerlimit of the bin while the Cycle G larger growths tended to span more of the bin voltare range. Thus, percentage growth has a more dominant effect for Cycle 5A than Cycla G growth and Cases 7 and 9, using Cycle 5A percentage growth, show significant increases in leakage over Cases 2 and 4. Using Cycle G growth rates for which the voltage dependence is better defined and large growths are distributed over a wider voltage range, the use of percent growth resulted in a decrease in leak rates for Cases 6 and 8 while increasing for Case 10 compared to the corresponding Cases 1, 3 and 5. The magnitude of the largest growth values does not appear to be strongly dependent on voltage such that the use of percent growth does not appear to be appropriate for the available growth data and the use of percent growth is not recommended. The concept of voltage dependent growth as represented by the recommendations of this report is based on the increased frequency oflarge growth values at higher BOC volts rather than a strong increase in the magnitude of growth with increasing voltage. A.5 Conclusions Based on the comparisons ofleak and burst results calculated from as measured and from projected voltage distributions given in this section, it is concluded that incorporation of voltage dependent growth rates is necessary for the Bt aidwood 1 and Byron.1 sos with an ARC repair limit of 3 volts. The recommendations of Section A.2 above for developing voltage dependent growth rates provide good agreement between projections and as-measured voltage distributions with associated SLB leak rates. The recommended methods are based on appropriately representing the increased frequency oflarge growth values with increasing BOC voltage, particularly for BOC voltages above about 1 volt. The guidelines for defining the largest voltage bin for growth rates differ between applications with a voltage dependent POD and with a constant POD = 0.6. The methodology recommendations can be summarized as follows: Voltane Growth Rges Application of voltage dependent growth rates based on dividing the total BOC voltage range into bins following guidance (Section A.2) on the number of o;\\NSD15 Nape \\cce97\\ccec690d.wp5 A 14

I indications in the bins and developing separate cumulative probability I distributions for the growth rates in each voltage range (bin). Developing growth rates for the limiting SG as a hybrid population that includes the three largest growth rates found in all SGs and applying the resulting limiting SG, voltage dependent growth rates for all SGs. If a voltage bin incl as less than about 100 growth values, the largest growth rate in the bin should be increased, if necessary, to reflect the 95% confidence level applying the extreme value theorem. For the EOC projection analyses, the binned cumulative probability distributions for voltage growth are applied to the BOC indications over the individual voltage ranges used to develop the growth distributions. Analyses of As-meas.mvl Voltage Distributions (Condition Monitorine) The NDE uncertainty for analyst variability should be reduced for indications above about 2 volts relative to the 10.3% developed for indications predominantly less than 2 volts. Further statistical NDE uncertainty analyses are, however, required to finalize an NDE uncertainty for analyst variability that varies with the measured voltage magnitude. o:\\NSD15\\apc\\cce97%ccec690d.wp5 A.15

Table A-1 Braidwood-1 and Byron-1 Historical EFPD and Days at Temperature Days at Ratio Cycle n/(n-1) Cycle EFPD Temperature EFPD Days at > 500 F Temp. Braidwood-1 1 424 455 2 314 347 0.74 0.76 3 413 470 1.32 1.35 4 419 459 1.01 0.98 5A 271 276 0.65 0.60 5B 185 203 0.68 0.74 6 378 413 2.04 2.03 6/5A = 1,39 6/5A = 1.50 Byron-1 1 431 431+ 2 376 431 0.87 Unknown 3 387 422 1.03 0.98 4 464 548 1.20 1.30 5 412 451 0.89 0.83 6 467 511 1.13 1.13 7A 310 342 0.66 0.67 7B 88 102 0.28 0.30 a:\\NSDIS\\apc\\cce97\\ccec690d.wp5 A-16 i

Tabla A-2 Braidwood-1 EOC-6 SLB Leak Rate Projections for Voltage Dependent POD Voltage Projection Methods Using Voltage Dependent Growth Rates SG - C HL Indications Only Based on Actual Cycle Length (378 EFPD) Growth Applied to BOC-6 SLB Leak Case Volts and Growth Voltago No. Max. Rate No. Bin Widths POD of Volts With Free Cominents (Note 3) Ind.* IRB Span Actual EOC-6: " Truth" for Comparison with Project!on Methods Ref. N. A. I 2098 11.3* 9.8 8.0 5% NDE unc. Projected EOC-6: Recommended Growth Methodology of Section 6.3 1 Cycle 6 Volt Dependent. EPRI 1973 10.9 9.9 80 Fig. A 2 growth Os.5, 0.5 1,1 1.6 >1.6 2A Cycle SA Volt Dependent. EPRI 1973 8.3 8.8 6.8 Fig. A 7 growth Os.5. 0.5 1, >1.0 2B Same as 2A except growth EPRI 1973 8.0 8.2 6.1 ratioed based on EFPD 3 Cycle 6 Volt Dependent. 0.6 2382 10.3 13.1 10.6 Fig. 6 2 growth 05.7, 0.7 1.1, > 1.1 4 Cycle SA Volt Dependent. 0.6 2382 7.9 8.5 6.4 Fig. 6 4 growt:I 050.7, >0.7 Sensitivity to 5 Case 1 with C6 mod. bin volts EPRI 1913 10.6 9.8 7.9 growth bins Os.5, 0.5 1,1 1.45, > 1.45 within guidelines. Fig. A-3 growth Projected EOC-6: Sensitivity to Growth Rate Methodology (Use of % Growth in Largest Growth Bin) 6 Case I with Cycle 6 growth as EPRI 1973 iO1 9.3 "4 % growth above 1.6 volts Case 2 with Cycle SA growth as EPRI 1973 13.5 15.7 13.5 % growth above 1.0 volts 8 Case 3 with Cycle 6 growth as 0.6 2382 12.2 11.5 9.3 % growth above 1.1 volts 9 Case 4 with Cycle 5A growth as 0.6 2382 12.1 14.7 12.2 % growth above 0.7 volts 10 Case 5 with Cycle 6 growth as EPRI 1973 13.4 10.7 8.9 % growth above 1.45 volts Notes: (1) Number of indications adjusted for POD. (2) Voltages include NDC uncertainties from Monte Carlo analyses and exceed measured voltages. (3) Growth rates of Cycle SA adjusted to Cycle 6 based on days at temperature except as noted. o-\\NSD15\\apc\\cce97%ccecG90d.wp5 A.17

o g c i I i + i i n 4 i ' i-i l I i, i, i, { j i i 4 i i I 3 L i i i I e i i i i i i i i l I I j i i i i l i i i f i. i i i i l i i m t 4 i i i i l l i i i i o i O o 'C i j d l 1 i t l { l W i i l l i m -u r i o i i i 4 i I i l i A i. I Da i

~

l l i i i l i l l a w 4 O O i e i i i j m I l I l l I 0 a y e i e y xm m. m og i i 4 c w. i o gu e U g w o I ga + .c "B: w i 4 e, A E i E =. 4 ,.c g P ~h i i 4 l g Q g 4-oc .E u R e o o. 2 -H& m ~ a ~ u c o7 ,g cm u o OD d cy m o. o zo M u ($ Q - - ~ ~ E oEa a O g g g L C O.o o J.; P o 'E 8 m .e e a o o L cc 4 ho I i a u ma ww a m cc c: 4 o i m 2 [ g c o. t o a o o o o o o C o c. o. c. c. 3 m m o o g opuu aanauaedtual an anni f,

Figure A-2 Braidwood Unit -1 Cycle 6 Hybrid Growth - Normalized Using Time at Temp > 500 F (All SG-C+ Largest SG A)- Cumulative Probability Distributions for Use with POPCD

; ^

w:o:.n :ts:.S cu= = + cj.y.".9. g f.W* P-1.0

• x.... x

/ 1 - x.x..x..x. x - r

1. g. s.

.n'.- f~~ ,._,.x-K-X-K 0.9 - x // xx / a 0.8 - -4*- y'- ,. a . _. _. an ~ * 'l ,f .x .E-l 1 x g 0.7 - ,t .x = .e. u 4 e a i I " 0.6 - e e, a .o y-- s'- 1: 0.5 - Up to 0.5 volt (714) l u / / \\

~

:

F 0.4 - / - c- 0.5 to i volt (1059) 1 t a 5m l -- * -- I to 1.6 volts (287) v 0.3 - - - ~ ~ - 36 ? s' - e-- Over 1.6 volts (25) O.2 - i.l-m' ~ 4 15 0.1 - -.D - <:- SG C Only (All Volts) n--n> M y 0.0 tF b7.

l l l l l

: : : : :

l l l l l l l l l l l l l l l m m 9 et n. 9 m ce c4 9 -r. m 9 et o c! 9 9 m ca c4 m n m e n n e-e o o o o o Voltage Growth A-19 c,arges s reviver so ao au e

Figure A-3 Braidwood Unit -1 Cycle 6 Hybrid Growth - Normalized Using Time at RCS Temp.> 500 F (All SG C+ Largest SG A)- Cumulative Probability Distributions for Use with POPCD - - - ---~ 1.0 g-g.QrS-iW-W-T- }f' _y. j a Fw-F W " ~ ~ 7 g .x -x ,a 0.9 - N. -- - X" # ,a, - r W, ~ . x - *. x. - 2 , a- // "_r m,.- m -a tt 0.8 - P-if m-V / i W c: 0.7 - 4" .2 4 i a-a' V C 4 y i 1 y' d j0.6- .e W 5 k g W n Up to 0.5 volt (734) j0.5-5 Y t -h .f - - 0.5 to I volt (1059) 0.4 -

.=

E lF -- * -- I to 1.45 volts (266) D ./ U 0.3 - + ll - e-- Over 1.45 volts (46) sE $q-[ - 4:- SG C Only (All Volts) 0.2 - 0.1 - II 4 .urp 0.0b , _ _ _ e o 4o g o o o o o o o o o

==a =-- a a a n n a e + -n +,,a a m. + Voltage Growth A-20 cr cy*s a samer to2e ans e e

Figure A-4 Byron Unit -1 Ociober 1996 Outage Voltage Growth During Cycle 7B vs BOC-7B voltage - All SG-B + Largest in SG-C 2 1 x SG-B (AllIndications) l o ~ x o SG-C (Largest Growth) 1.5 l R x g O I x x x x I x X X 5x[ x & 5 x ir x x x x x x x x x g: x yx5 g 2 x xx x

1. x

? 0.5 y x x xx x [ X Y x x xx 0 x -0.5 0 0.5 1 1.5 2 2.5 3 BOC-78 Voltage A-21 c=wsuanm a 45 m

Figur-A-5 Cycle 7B 11 brid Growth - Normalized Using Time at Temperature > 500 F Hyron Unit -1 3 All SG-H + Largest in SG-C - Cumulative Probability Distributions for Use with POPCD = ,,,Y' , w,;,a 1.0 ^ W?.x$i-A i p-tr ' X' 5W y.. ,,E~~. 0.9 - _3 ,a,,1r g' n.- .s/ .x' "O J' / 0.8 - $a .x' i e r a n

• x-

/ ./. m-c 0.7 - o / n y r/ / a M [ .6 - I P 0 a .o Up to 0.5 volt (826) q. .j0~5-I p l F y m' o - c- 0.5 to i volt (744) f- / F x f y 0.4 - ./ -- r --I to 1.5 volt (144) j j j l / l s U 0.3 - 7 - *-- Over 1.5 volts (33) f. x" f .? i O~2 - ~# E ~" - <:- SG-B Only (All Volts) . m - a'

t E

h I O.I - 7-- / p/ 1u-a -a.-a 0.0 s; - -,-n -o -~m ,-e n --~m -e n e ~-~m , - e e. - e - e e e- .e mn a a a n aa an a a a e n <. m e e 444o 4 o o o o o o o o o Voltage Growth A-22

c. w..m..i -

Figure A-6 Braidwood Unit-1 February 1995 Outage Voltage Growth During Cycle SA v HOC-5A voltage - All SG-A + Largest 2 in SG C 4.5 5 4 j a SG-A (Allindications) 3.5 m SG-C (Largest 2 Growths) e 3 A I a I ^ U 2.5 a E A } a a j 2 A a A A ^ ^ ^ C.D 1.5 a E An a O a n AA A A a a a A ^ ^ 3 l' -0.5 O 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2 BOC-SA Voltage A-il CyGagrthFigA48Fl/97 8.29 PM

l Figure A-7 Braidwood Unit -1 Cycle SA Hybrid Growth - Normalized Using Time at RCS Temperature > 500 F All SG A + Largest 2 in SG C - Cumulative Probability Distributions for Use with POPCD M _,7 A-+,w ; _ u_..>.. g _ 4.. l 1.0 g _. - .= = A n ~~ " ,x... ..-r.. -* 0.9 - . ' p tr

g 0.8 -

,J:',0 ,.r ' /: t[ r ...r f ,a 0.7 -

• f 4

!Y. 7-/.X S 0.6 - e iY* a Ix.- a 'S 0.5 - Sf* b ,Y Up to 0.5 volt (571) l g /I l = 0.4 - / "I - c- 0.5 to i volt (428) $mu 03 - I ---r --Over 1 volt (42) 0'2 - --c>-- SG A Only (All volts) .r O.1 - ,.. x - - [ 0.0 lL n.i.:v. - n m -r m e r-a w - - c. m -r m e r-a n - n m -r e r~ a ce m e r-w m ej 4 4 6 6 6 6 6 6 6 6 6

= - -- - - -;

et ei es et ei n ei a m m a m o m ce - Voltage Growth c, r.uw..i m A-24 ..-.,-2_,....__._____...._i - - ~ - -~

Figure A. 8 Braidwood Unit 1 Comparison of Predicted and Actual Bobbin Voltage Distributions for SG C, EOC-6 Using Voltage Dependent Cycle 6 Growth. POD = 0.6 Steam Generator C 3 volts & over only 6 6 O Actual EOC4 E Predicted EOC-6, EPRI POPCD e 4-Volage Dependent Cycle 6 Growth 8 l I l l I li......ill llU ni..Hhi .itii. I i Bobbin Voltage Steam Generator C - 3 volts & over only l 5-0 Actual EOC4 E Predicted EOC-6, POD = 0.6 ,4 Voltage Dependent Cycle 6 Growth ak3 o 2-t. II l l a......il li.....llli iil! I l l o : :: : : : : : ::::::::::::::::::::::::::::; Bobbin Voltaos m m emnvs m A-25

APPENDIX B PROBABILITY OF PRIOR CYCLE DETECTION This appendix presents a voltage dependent POD called probability of prior cycle detection (POPCD) established based on the data from the last two inspections at Braidwood Unit-1. A composite POPCD distribution based data from 18 inspections at 8 plants, including data from the past 3 inspections for Braidwood Unit-1, is also presented. An evaluation was also performed to establish RPC confirmation rates in the current inspection for RPC NDD indications in the last inspection, and those results are also presented here. B.1 POPCD Distribution for Braidwood Unit-1 EOC-5B Inspection The inspection results at EOC G perrrit an evaluation of the probability of detection at the prior EOC 5B inspection. For ARC applications, the important indications are those that could significantly contribt to to EOC leakage or burst probability. These significant indications can be expected to be detected by bobbin and confirmed by RPC inspection. Thus, the population ofinterest for ARC POD assessments is the EOC RPC confirmed indications that were detected or not detected at the prior inspection. The probability of prior cycle detection (POPCD) for the EOC-5B inspection can then be defined as follows. EOC 5B cycle reported + Indications confirmed and indications confirmed by repaired in EOC 5B RPC in EOC 6 inspection inspection POPCD = (EOC-5B) { Numerator) + New indications RPC c o n fir m e d in EOC-6 inspection POPCD is evaluated at the 1995 EOC 5B voltage values (from 1997 reevaluation for growth rate) since it is an EOC-5B POPCD assessment. The indications at EOC 5B that were RPC confirmed and plugged are included as it can be expected that these indications would also have been detected and confirmed at EOC 6. It is also appropriate to include the plugged tubes for ARC applications since POD adjustments to define the BOC distribution are applied prior to reduction of the EOC indication distribution for plugged tubes. o:\\NSD15 Nape \\cce97Nccec690d.wp5 B1

It should be noted that the above POPCD definition includes all new EOC 6 indications not reported in the EOC 5B inspection. The new indications include EOC 5B indications present at detectable levels but not reported, indications present at EOC 5B below detectable levels and indications that initiated during Cycle 6. Thus, this definition, by including newly initiated indications, differs frcm the traditional POD definition. Since the newly initiated indications are appropriate for ARC applications, POPCD is an acceptable definition and eliminates the need to adjust the traditional POD for new indications. The above definition for POPCD would be entirely appropriate if all EOC 6 indications were RPC inspected. Since only a fraction of bobbin indications are generally RPC inspected, POPCD could be distorted by using only the RPC inspected indications. Thus, a more appropriate POPCD estimate can be made by assuming that all bobbin indications not RPC inspected would have been RPC confirmed. This deGnition is applied only for the 1997 EOC 6 indications not RPC inspected since inclusion for the EOC 5B inspection could increase POPCD by including indications in a tube plugged for non ODSCC causes which could include RPC NDD indications. In addition, the objective of using RPC confirmation for POPCD is to distinguish detection ofindications at EOC,.i that could contribute to burst at EOC, so that the emphasis is on EOC, RPC confirmation. This POPCD can be obtained by replacing the EOC 6 RPC confirmed category by RPC confirmed plus not RPC inspected category in the above definition of POPCD. For this report, both POPCD definitions are evaluated for Braidwood Unit 1. The POPCD evaluation for the 1995 EOC 5B inspection data is summarized in Table B-1 and illustrated on Figure B 1. The long-dash line shows results obtained by considering only the RPC confirmed indications, and the solid line shows data based on the RPC confirmed plus not RPC inspected indications. Also shown in the figure is the generic EPRI POPCD developed by analyzing data from 15 inspections in 8 plants (Reference 10,5) and it represents the lower 95% confidence limit. It is seen from Figure B 1 that the POPCD distribution predicted for the Braidwood-1 EOC 5B inspection is in good agreement with the generic POPCD distribution. This justifies the use of the generic POPCD in the assessment of alternate methods of modelling voltage dependent growth rates in the revised leak and burst projection methodology. It is clear from Figure B-1 that the NRC mandated constant POD =0.6 is not conservative below 0.6 volts, and it is conservative above 1 volt. The Braidwood Cycle 5B POPCD is below the generic POPCD below 0.6 volts and higher at a value of 1.0 above 2.0 volts. The POD of 1.0 above 2 volts shows that large, undetected indications were not left in service for Cycle 6. The lower POD o:\\NSD15Napc\\cce97%ccec690d wp5 B.2

s +49 s$9 0 1 A k'h# f8d ((/[ g[ IMAGE EVALUATION WW+g$'#' N"sq<N k//7\\Y T TEST TARGET (MT-3) l.0 lf 2 m y ll L24 ll E m Ese I.8 i.25 gg 4 150mm W 4 6"

1. 5'///

'#D

h 4

h4 1 PHOTOGRAPH'C SCIENCES CORPORATION 4 O 770 BASKET ROAD d't, 'k (1,k g = P.O. BOX 338

1. es*g WEBSTER, NEW YORK 14580 4

(716) 265-1600

,*~ %h & gfr- $b /j/[/#NT (0(,[+4(9

• Th k?

IMAGE EVALUATION v %/ TEST TARGET (MT-3) \\\\I#'

1. 4 2

777 1.0 lfi83 2 ? ? RM l.1 E ~ d 1.25 1.4 1.6 150mm 4 6" /* 4% h$ '&Vb///, + h "4 PHOTOGRAPHIC SCIENCES CORPOR ATION [u (*[k, ' I h 770 BASKET ROAD t P.O. BOX 338 4pg WEBSTER, NEW YORK 14580 4 (716) 265 1600

+ N ~ 4' & ((p [di*4,# @. k# IMAGE EVALUATION 4 ) *, c t/ T +4 TEST TARGET (MT-3) 9' %,,I 1.0 52 M m m g"u U tu 1.1 [" EEa 1.25 ' __l.4_. 1.6 4 150mm 4 6" f4 i

• Q, 3,,,7//,

++;k 4 5 _ _,, =,, _ _ WEBSTER TIE YORK 14580 (716) 265-1600

s +%+# Qh WO ///jg/j[,6,[+ fg [.. % A Y* IMAGE EVAL.UATION

1. 4 O

,\\// %Y %[# NN'N // TEST TARGET (MT-3) W' ?' / l.0 if 2 E yljHE l'l E m EM L: ' I.25 1.4 1.6 4 150mm 4 6"

1. ik' 7,f

't ;( & hs4 PHOTOGRAPHIC SCIENCES CORPORATION [$g 4%(Ub' h 770 BASKET ROAD P.O. BOX 338 WEBSTER, NEW YORK 14580 4 (716) 265-1600

below 0.6 volts indicates that the number of low voltage, new indications were underestimated for Cycle G but these indications have little influence on the leak ' rates and burst prcbabilities at EOC 0. In summary, the Braidwood Unit 1 EOC-5B POPCD supports a voltage dependent POD higher than the NRC mandated POD value of 0.6 above about 0.6 volts and approaching unity at about 2 volts. B.2 Generic POPCD Distribution Based on 18 Inspections in 8 Plants At this time, POPCD evaluations are available for 18 inspections at 8 plants, including three evaluations for Braidwood Unit 1. The available data include G inspections of plants with 3/4" diameter tubing and 12 inspections of plants with 7/8" diameter tubing. This section summarizes these POPCD evaluations for comparison with results of an EPRI study that examined detection probability for a dual analyst team. The POPCD evaluations performed since 1992 show significant improvement over the earlier nosessments which represent the first ARC inspections. Bobbin data analysis guidelines (Appendix A guidelines) have been revised since the first inspections to reflect the initial ARC experience. Thus, it is appropriate to assess POPCD for inspections performed since 1992. Fourteen of the 18 inspections for which POPCD has been evaluated were performed since 1992. Table B 2 shows the combined POPCD evaluation for plants with 3/4" diameter tubing and includes results for 6 inspections performed since 1992. These data are also plotted in Figure B 2, and they include data from the present Braidwood Unit 1 assessment (EOC 5B results representing 1995 inspection) as well as the data for EOC-4 and EOC 5A inspections. These results tend to support a POD approaching unity above about 3 volts. The POPCD assessment is in very good agreement with the results from the EPRI study on detection probability for a dual analyst team. The average POPCD independent of voltage is about 0.62 which is in general agreement with the NRC Generic Letter 95 05 proposed voltage independen. FOD of 0.00. The average value is heavily weighted by the large number ofindications in Table B 2 below 0.6 volt for which POPCD is s 0.6. Table B 3 and Figure B 3 shew the combined POPCD eve.luation for the 8 plants with 7/8" tubing and inspections performed since 1992. These results tend to support a POD approaching unity above about 3 volts. The average value is about 0.69. The POPCD evaluation in Figure B 3 is in good agreement with the EPRI POD except in o;\\NSD15\\apc\\cce97\\ccec690d.wp5 B-3 I

~ ~ - ( - the one to two volt range where POPCD is essentially constant at about 0.83 and the - EPRI POD increases from about 0.83 to about 0.98, The definition of POPCD includes indications which were not present at the prior inspection and thus could be ~ expected to be somewhat lower than-the EPRI dual analyst detection probability which is based on " expert" evaluations of inspection results and does not include indications clearly below detectable levels. ' The combined data for the 14 inspections since 1992 are given in Table B 4 and the POPCD evaluation is shown in Figure B 4 for RPC confirmed plus not inspected indications. -It is seen that the inspections since 1992 yield a POPCD in good agreement with the EPRI dual analyst detection probability which was a 1994 evaluation. POPCD supports a POD approaching unity at about 3.5 volts while the -EPRI dual analyst detection probability is about 0.98 at 2 volts and unity at 3 volts. Figure B 4 also includes POPCD evaluated at the lower 95% confidence limit on the data for indivMual voltage bins, The POPCD evaluations shown in Figures B 2 to B 4 are based on the definition of " truth" as RPC confirmed plus not RPC inspected indications. Since many of the indications not RPC inspected would be expected to be found NDD ifinspected, this represents a lower bound POPCD evaluation. Figure B 5 shows the POPCD evaluation for all 14 inspections since 1992 based only on RPC confirmed indications. This results in a significant increase in POPCD below 1.0 volt and a modest increase above 1.0 volt. The data of Table B 4 show 500 to 11,000 indications in all voltage bins below 2 volts,250 between 2.0 and 3.2 volts and about 5 indications above about 3.2 volts. Thus, the collective data provide a substantial database for defining a POD. The results of Figure B 4 clearly support an increase in the POD for ARC applications above the POD = 0.6, independent of voltage, required by NRC Generic Letter 95 05. For indications above 1.0 volt, the POD exceeds 0.9 and is 0.97 to near unity at 2.0 volts. A POD of 0.6 is only' applicable to indications below about 0.6 volts. cA voltage dependent POD distribution has been developed for ARC application by evaluating POPCD at the lower 95% confidence level and the mid voltage of each voltage bin. The result is then smoothed to obtain the POPCD distribution as sh. vn . in Figure B 6. This POPCD distribution is tabulated in Table B 5 and compared o th the EPRI dual analyst detection probability in Figure B 7. Table B-5 shows both the POPCD distribution presented' originally in Reference 10.5 and an updated distribution that includes results for 3 more inspections including the Braidwood o:\\NsD15Napevce97\\ccec690d.wp5 B4

1 4 Unit.1 EOC 5B inspection. Although the updated distribution includes data from another 7000 indications, the difference between the two sets of POPCD does not exceed 0.01 which provides confidence that the POPCD distribution is based on a sufficiently large database. B.3 Assessment of RPC Confirmation Rates for Braidwood Unit-1 EOC-5B Inspection This section tracks the 1995 EOC 5B indications left in service at BOC 6 relative to - RPC inspection results in 1997 at EOC 6. The composite results for all SGs are given in Table B 6. For 1995 bobbin indications left in service, the indications are tracked relative to 1995 RPC confirmed,1995 RPC NDD,1995 bobbin indications not RPC inspected and 1995 bobbin indications with no indication found in 1997. Also included are new Cycle 6 indications. The table shows, for each category of indications, the number ofindications RPC inspected and RPC confirmed in 1997 as well as the percentage of RPC confirmed indications. Of the 102 RPC NDD indications left in suvice at BOC 6,30 were RPC tested during the EOC 6 inspection and 12 were confirmed. Thus the overall confirmation rate for 1995 RPC NDD indications is 40%. The RPC confirmation rate for the last Byron 1 inspection was 27.3%. It has been recommended bv ndustry that the largest RPC i NDD confirmation rates over the prior two cycles be used for projections. Thus, it would be justifiable to include only 50% of the RPC NDD indications in the BOC 7 voltage distribution used for EOC 7 projections, and leak rate and burst probability analyses. However,100% of RPC NDD indications reported in the EOC 6 inspection are considered in the SLB leak rate and tube burst probability analyses presented in this report for EOC 7 conditions. r r o.NNSD15Napctcce97secce690d wp5 B.5 l

Table B-1 l Braidwood Unit 1 1997 EOC-6 Evaluation for Probability of Prior Cycle Detection ~ Composite of AllSteam Generator Data Indications Detected EOC-5B POPCD New Indications Both in EOC-6 and Inspection EOC-58 inspections EOC-6 EOC-6 RPC Inspection inspection EOC-6 RPC EOC-6 RPC RPC RPC Confirmed Voltage inspection Confirmed inspection Confirmed Confirmed Confirmed Plus Not Inspected Bin RPC plus not RPC plus not and Plugged l Confirmed inspected Confirmed Inspected Frac. Count Frac. Count 2 4 228 1 28 0 0.2 1/5 0.109 26/256 I 0.2 - 0.4 17 1175 9 507 1 0.370 10/27 0.302 508/1683 0.4 - 0.6 16 874 28 990 7 0.686 35/51 0.533 997/1871 0.6 -0.8 11 357 76 938 4 0.879 80/ 91 0.725 942/1299 f 0.8 - 1.0 13 167 99 680 0 0.884 99/112 0.803 680/847 f 1.0 - 1.5 10 86 211 583 4 0.956 215/?M 0.872 587/673 1.5 - 2 3 6 48 61 2 0.943 50 / 53 0.913 63/69

2. - 3.2 0

0 15 15 2 1.0 17/17 1.0 17/17 TOTAL 74 2893 487 3802 20 > 1V 13 92 274 659 8 B-6 s Popcd TableB-I 8/8/9710:10 AM k ~ t

Tame B-2 Evaluaties for POPCD for Planes wkk 3/4" SG Tubes CoseNeed Data frase 6 Post-92 ('93 and later)lespections [ t New Indcations Bobbm Can in Both inspec5ons Fwst inspecton POPCD RPC RPC Vokage Cor*med Cor&med RPC RPC RPC Cordrmed Bn RPC phs not RPC p4us not Cordrmed Cordrmed Phis Not Inspected Cordrmed inspected Cor*med bnspeded and Plugged rrac. Court Frac. Ocure >0-02 12 1238 1 ab: 0 0 1/13 0285 494/1732 02-0.4 112 5095 28 3661 32 0349 60/172 0.420 3893/8788 04-06 119 3330 148 5054 93 0.669 241/360 0 607 5147/8477 064.8 96 1275 343 3858 120 0.828 463/559 0.757 3978/5253 08-1.0 90 483 470 2381 114 0.866 584/674 0.838 2495/2978 10-12 51 148 211 844 884 0.955 1095/1146 0.921 1728/1876 1.2-16 39 82 245 482 778 0.963 1023/1062 0.939 1260 /1342, 16-20 6 9 82 87 246 0.982 3281334 0.974 333/*,42 ~ 20-22 3 3 17 17 54 0.959 71/74 0.959 71 /74 i 22-2.5 1 1 9 9 42 0.981 51/52 0.981 51 /52 2.5-32 2 2 9 9 67 0.974 76178 0.974 76/78 32-3.5 0 0 0 0 5 1.000 5/5 1.000 5/5 TOTAL 531 11666 1563 16896 2435 Total > IV 102 245 573 1448 2076 i L l I B-7 'f raccovaravam s seeemo ve m

ll!il l';1,1, lI lj i f 9 8 4 3 1 9 1 0 5 5 9 1 4 9 7 2 3 9 7 5 6 2 0 tn 2 2 1 6 5 1 1 1 0 1 1 1 u / / / / / / / / / o / / / 1 0 3 5 1 7 9 7 0 1 d C 5 3 8 5 7 5 4 1 1 1 t 1 +T.N t oe 9 6 5 4 5 4 1 1 8 9 C 1 1 P T. s R lu wP C L . 5 2 4 3 9 2 4 6 0 0 0 1 6 1 7 4 7 8 7 0 0 0 ce 4 5 6 7 7 8 8 8 0 8 0 r D F 0 0 0 0 0 0 0 0 1 0 1 W CPO P 5 7 0 0 6 7 1 7 5 1 5 8 4 9 0 0 B' tn 7 8 1 1 1 1 1 9 1 1 1 u / / / / / / / / / / / o / d C 5 3 2 4 0 3 7 5 9 7 0 8 7 6 4 0 3 6 8 1 1 1 1 1 1 1 C m. P r R ac W C 4 1 6 7 9 7 8 4 0 0 0 c 1 0 2 1 0 8 9 0 0 0 0 s a 7 9 9 9 9 8 8 9 0 8 0 n r s o F 0 0 0 0 0 0 0 0 1 0 1 ei b t u ce Tp M-Gsn SI m m d t d e ) " r eg e 4t mg 2 1 9 5 2 2 9 u 0 2 4 7 a 5 2 8 4 0 5 2 rl 5 7 7 9 7 3 l 1 f 1 1 6 2 h d i nP t f i n od wa s Cn t r a s3 i 9 F t 3 n '( L Bl 8 a 2 P e 9 s B l r - t a b o s c af o d d TDP dc etoe 9 6 7 a Cmnt 8 c 0 3 1 4 2 2 1 9 C8 p P a 9 4 9 6 6 4 5 0 2 f s e 1 2 3 3 8 1 4 P 1 8 8 4 3 5 9 m s Rnup s 1 1 Oo n olpn i C i P r h f t r o o a t B f a nD n i o d n i a d t e a e u n C Cm 1 0 0 0 8 2 8 0 1 i 4 5 0 6 0 lab in Pi 0 3 r 2 2 4 8 5 1 f 1 m b Rn 2 2 v b o E o o C C B d d eie t 2 5 CmnW 0 6 7 4 0 6 0 r 9 8 s 0 2 0 0 8 Pi s 4 0 6 2 5 1 1 7 1 n f 2 Rnu 1 8 8 5 2 1 2 o olp6 ita C 1 c idn I w de e Cm N 3 3 0 7 9 3 7 P a 2 8 9 0 2 0 0 9 4 f 1 1 1 1 1 Rn oC ~ V m 2 4 6 8 0 2 6 0 2 3 2 5 L I e 0 0 1 1 1 2 2 2 3 3 A g 0 0 an - T t i 2 lob 0 4 t 8 0 2 6 0 2 5 2 O la s V 0 0 0 0 1 1 1 2 2 2 3 T t o T m m i u m l ll llll1 il

Tal+ B-4 Comtdned POPCD Evaluation for 14 Assessments Condected After 1992 POPCD Based on RPC Confirmed Plus Not Inspected Indications New Indications Bobba1 Can in Bottiinspedions First ;n%i N RPC RPC VoRage Cor*med Contwmed RPC HPC RPC Cor6med Bn RPC plus not RPC plus not Cor6med Cor6med Plus Nat W Conf' med inspected and Plugged Cor*med Irnpected w Frac. Count Frec. Count ,> 0 - 02 14 1378 1 604 5 0.300 6/20 0.309 60911967 0.2-04 120 5901 31 4484 102 0526 133/253 0.437 4586/10487 04-06 132 4197 158 6453 245 0753 403/535 0.615 8898110885 06-08 109 1799 363 5204 244 0 848 607/716 0.752 544817247 08-10 100 733 498 3272 186 0872 6841784 0.825 3458 74191 10-12 68 264 253 1308 975 0.948 1228/1296 0.896 228312547 32 1s 58 161 333 844 857 0954 1190/1248 0.914 170111ss2 16-20 15 27 132 169 281 0965 4131428 0943 4501477 20 22 3 3 28 28 62 0968 90 t 93 0.968 90t93 22-25__ 3 3 13 13 46 0 952 59/62 0.952 59182 25-32 2 2 14 14 79 0.979 93/95 0.979 93195 32 35 0 0 0 0 5 1.0 5/5 1.0 515 TOTAL 624 14468 8824 22393 3087 2 Totas > tv 149 460 773 2376 2305 h B-9 t POPCDTU2 Tatde 9 4 8/8/97 3 36 PM l [

Table B-5 Comparison of EPRI POPCD with EPRI POD Study Voltage EPRI' POD NP-7480-L Bin Updated Study Addendum-1 0.1 0.30 0.24 0.24 0.2 0.38 0.34 0.34 0.3 0.49 0.44 0.43 0.4 0.57 0.53 0.52 0.5 0.62 0.62 0.61 0.6 0.66 0.67 0.67 0.7 0.71 0.73 0.73 0.8 0.76 0.77 0.77 0.9 0.80 0.81 0.81 1 0.83 0.83 0.83 1.2 0.90 0.88 0.87 - 1.4 0.93 0.91 0.90 1.6 0.96 0.92 0.91 1.8 0.98 0.93 0.92 2 0.984 0.94 0.93 3 1.00 0.98 0.98 3.5 1.00 1.0 1.0

1. Dual analyst detection probability study B 10 POPCDi&F2 Tabee S5 MV97 3.36 PM

i i I V Totde B4 Breldwood Unit 1 Analysis of RPC Data froen EOC-58 and EOC4 ".,l Comenned Data froen All Seoem Generators I Total Total Total Total Percent EOC-58 EOC-6 EOC-6 EOC-6 EOC-6 Group of Indications inspecton inspecton inspechon inspection inspechor; Bobtun Bobbin RPC RPC RPC Indicahon Irdication inspected Conenned ConArmed Less then or Equal to 1.0 Volt in EOC4 -.:;: ^ t f ._ EOC-58 Inspecten _Bobtwi Left in Servce _,_ __ .._ 2400 _ 2225 35 21 80.0 _ _ _.. _ _ _ _ -_ EOC-58 q_ RPC Cordrmed_ _ 3 3 1 1 100.0 -- EOC-58 trgecten RPC NDD_ __._ 31 31 15 3 20 0 - E_OC-58 Inspechon RPC Not inspected 2191 2191 19 17 89.5 r -. -... _...No [Q W Bckbm ' ___,175 _ __ D EQ_%:en Indcaton_ __ __ _,26 },4__,. __,,_63, _ 29 46 0 Sum of M EOC4 ; %aw. 3rdcahon 2400 4839 96 50 51.0 l i Greeler then 1.0 Volt in EOC4 L-_; : f--. . [MSB inspecten Wigh W _,168 _3__ 1624 _ _ 499 _ 466. _ 93_4 _ _ _ _ _ .- EOC-58 inspecten @PC Cc *ned _ _ 154 __ 154,, 98 98 1M 0 - EOC-58_Inspecton_RP_C NDO_ 71 71 _ _ 15 9 80 0 .__ ____._ ~ _ EOC-58 Inspecton RPC Not inspected _ 1399 , _ 1399 386 359 93D i .. _. _ _ No E_Q _Inspechon Bobbm

• 59 New EOC-6 Inspechon Irdcahon 321 53 45 84.9 Sum of M EOC-6 M=r

-- Indcahon 1683 1945 552 511 92 6 p All Voltages in EOC4 Lc. ::^ka ~ 91.2 , EMSB Inspecten Bobbe L@ h Ser** _4083, __ _ 3849_, _.534,_ 487

EOC-5B inspechm @PC Cmfmved,

_ _, }57,_ 15? _ 99 _ _ __ _ 99, _. __1M D _,,, - EOC-5B inspecten RPC NDD 102 _ 102.__ .._ _ 30 12 40.0 f _. _ _ -, EOC-5B_inspecton RPC Not inspected 3590 3590 405 37G 92.8 c ~ . _ _. ~.. _ _ - Sum of M EOC-6 ; e-;+ -_-,Irdcahon 4083 6784 650 561 ~ _. 63.8 New Ep,lnsp ardcanon, _ 3 35 _, j!6 __ , 74 _ 86 3 i M U M M IN N O @ B-I l t t I Poped TabicB4 IL%/97 5_44 PM - c

Table B-1 Braidwood Unit i 1997 EOC-6 Evaluation for Probability of Prior Cycle Detection i Composite of All Steam Generator Data F Indications Detected EOC-58 New Indications Both in EOC-6 and POPCD EOC-5B inspections EOC-6 EOC-6 Inspection Inspection RPC EOC-6 RPC EOC-6 RPC RPC RPC Confirmed Voltage inspection Confirmed Inspection Confirmed Confirmed CwifiuTM Plus Not Bin RPC plus not RPC plus not and Plugged Inspected Confirmed inspected Confirmed inspected Frac. Count Fisc. Count > 0 - 0.2 4 228 1 28 0 0.2 1/5 0.109 28, 256 0.2 - 0.4 17 1175 9 507 1 0.370 10/27 0.302 508/1683. 0.4 - 0.6 16 874 28 990 7 0.686 35/51 0.533 997/1871-0.6 -0.8 - 11 357 76 938 4 0.879 80 / 91 0.725 942/1299' O.8 - 1.0 ' 13 167 99 680 0 0.084 99/112 0.803 680/847 1.0 - 1.5 10 86 211 583 4 0.956 215/225 0.872 587/673 1.5 - 2 3 6 48 61 2 0.943 50/53 0.913 63/69

2. - 3.2 0

0 15 15 2 1.0 17/17 1.0 17/17 TOTAL 74 2893 487 3802 20 l l >IV 13 92 274 659 8

== I B-12 Poped TableB-I 8/88710:10 AM t

T Tame B-2 Evaheation Ier POPCD for Piness wkh 3/4 SG Taties Caselwned Data froan 6 Fost-92 ('93 and later) Inspections I New Indications Bobbin cab in Both inspections Fast inspecison POPCD 6 RPC rec Voltage Con 6rmed Cordrmed RPC RPC RPC Coniirmed Ben ' RPC plus not RPC plus not Confwmed Cordwmed Pkes mt inspected Cordrmed Inspected Cordemed inspected and Plugged j Frac. Court Frec Court ~! > 0 - 0.2 12 1238 1 494 0 0 1/13 0285 494/1732 0.2 - 0.4 112 5095

28 3661 32 0349

' 60/172 0.420 3693/8788 0.4-06 119 3330 148-5054 93 0.669 241 /360 0.607 5147/8477 064.8 96 1275 343 3858 120 0.828 463/559 0.757 3978I5253 0 s.1.0 90 483 470 2381 114 0.866 584/674 0.838 2495/2978 i 10 12 ~ 51 148 211 844 884 0955 1095/1146 0.921 1728/1876 1.2 - 1.6 39 82 245 482 778 0.963 1023/1062 3.939 1260 /1342 16 2.0 6 9 82 87 246 0.982 328 /334 0.974 333/342 2.0 - 2.2 3 3 17 17 54 0.959 71/74 0.959' 71/74 22-2.5 1 1 9 9 42 0.981 51/52 0.981 51/52 [ 2.5-32 2 2 9 9 67 0.974 76/78 0.974 76/78 j 3.2-35 0 0 0 0 5 1.000 5/5 1.000 5/5 TOTAL 531 '11666 1563 16896 2435 [ t Total > t f 102 245 573 1448 2076 t r f i f I I B-13 roecmarmau.o zessmess aas I

Table B-3 3 s Evaluation for POPCD for Plants witin 7/It" SG Tubes Combined Data fror.i 8 Post-92 ('93 and later) I%^2x 1 i New Indications Bobbin Callin Both inspechons First inspectm POPCD RPC RPC RPC RPC Confrmed Confirmed RPC Confrmed Voltage Confrmed Confrmed plus not and Piugged Confamed Plus Not Bin FiPC plus not inspected Inspected Confrmed inspected Frac. Count ' Frac. Count >0 - 0.2 2 140 0 110 5 0.714 5/7 0.451' 115/255' i 0.2-04 8 806 3 823 70 0.901 73/81 0.526 893/1699 0.4-06 13 867 10 1399 152 0.926 162/175 0.641 1551 /2418 06-0.8 13 524 20 1346 124 0.917 144/157 ' O.737 1470/1994 08-1.0 10 250 28 891 72 0.909 100/110 0.794 963/1213 1.0 - 1; 17 116 42 464 91 0.887 133/150 0.827 555/671 1.2 - 1.6 19 79 88 362 79 0.898 167/186 0.848 441/520 16-20 9 18 50 82 35 0.904 85/94 0.867 117/135 20-2.2 0 0 11 11 8 1.000 19/19 1.000 19/19 22-2.5 2 2 4 4 4 0.800 8/10 0.800 8/10 2.5-32 0 0 5 5 12 1.000 17/17 1.000 17/17 32-3.5 0 0 0 0 0 0/0 0/0 TOTAL 93 2802 261 5497 652 ~ Total > IV 47 215 200 928 229 i 4 B-14 POPcOT&F2iTanse B 3Ill@970 36 PW I

k Table B-4 Combined POPCD Evaluation for 14 Assessements Conducted After 1992 POPCD Br. sed on RPC Confirmed Plus Not Inspected IndPations New Indications Bothn Callin Bcth inspections Frst L-@i POPCD L RPC IPC Vomage Confwmed Cordr=ned RPC RPC RPC ConArmed Bei -RPC plus nct RPC plus not Con 6rmed Corewmed Phss Not W Cordermed inspected Con &med inspected and Pbgged Fm Jount Frec. Count i >0-02 14 1378 1 604 5 0.300 6120 ' O.306 809/1987 02-04 120 5901 31 4484 102 0.526 133/253 0.437 4506110467 i 04-06 132 4197 158 6453 245 0.753 4031535 0.615 essei10e95 064.8 109 1799 363 5204 244 0 848 6071716 0.752 5448 17247 l 08-10 100-733 498 3272 186 0.872 684/784 0.825 3ess / 4191 1.0-12 68 264 253 1308 975 0.948 1228/1296 0.896 228312547 12-16 58 161 333 844 857 0.954 119011248 0 914 1701/tes2 1e-20 15 27 132 169 281 0.965 4131428 0.943 450/477-20-2.2-3 3 28 28 62 0.968 90193 0.968 go/91 2.2 - 2.5 - 3 3 13 13 46 0.952 59/62 0.952 59I62 J 5-32 2 2 14 14 79 0.979 93195 0.979 93195 32-35 0 0 0 0 5 1.0 515 1.0 5I5 i TOTAL 624 14468 1824 22393 3087 Totr * > IV '149 460 773 2376 2305 i } t ' ) B-15 POPCDT&F2 Table B-4 81V97 3 36 PM 5 v 4

Table B-5 Comparison of EPRI POPCD ~ with EPRI POD Study i Voltage EPRl' OD %7N Bin Updated -Study Addendum 0.1 0.30 0.24 0.24 0.2 0.38 0.34 0.34 0.3 0.49 0.44 0.43 0.4 0.57 0.53 0.52 0.5 0.62 0.62 0.61 0.6 0.66 0.67 0.67 0.7 0.71 0.73 0.73 0.8 0.76 0.77 0.77 0.9 0.80 0.81 0.81 0.83 0.83 0.83 i _ 1.2 0.90 0.88 0.87 1.4 _ 0.93 0.91 0.90 1.6 0.96 0.92 0.91 1.8 0.98 0.93 0.92 2 0.984 0.94 n.93 3 1.00 0.98 0.98 3.5 1.00 1.0 1.0

1. Dual analyst detection probability study B-16 POPCOT&F2 TeNo B 6 84W 3.36 PM

I Table 84 Braidwood Unit 1 Analysis of RPC Data from EOC-58 and EOC4 Inspections Combined Data from All Steam Generators Total Total Total Total Percent EOC-58 EOC-6 EOC4 EOC4 EOC-6 Group of Indications inspecten L%Lai impechon L%i inspection Bobbin Botbn RPC RPC RPC inA Eni indication inspected Cuirmed Cwin.M Less than or Equal to 1.0 Volt in EOC4 :.;:;- T-:#. EMSB inspecten Botte [_en in Sennes 24M. _ __ 2225_ __ _ _ _, __35 _ __.__, 21 _ _60D - EOC;5Byh BPC Cor6eed ___3_. __3__ __ 1 __ 1 _1MD EOC-58 Inspectm RPC NDO 31 31 15 3 20 0 . EOC-58 Inspecton RPC Not inspected 2191 _ _ 2191 _ _19 _, 17 89.5 - No EM6 Inspectim Bobben

• _ 175,_,,_

,_ y ___ _1.__ % EOCy ;..py ncicaton 2614 63 __ 29 _ 46 0, i Sum of As EOC4 Inspection inchcaton 2400 4839 98 50 51D Greater than 1.0 Volt in EOC4 Inspection i EOCyB Inspeen Bottn L@in Senace _ _1683 1624 _ _ _ 499 _ 466 93 4 EMSB inspecen RPC Cor*nal _ 154 _ _ 154 _ _. _, 98 __ __98 1MD - EOCyB inspechon RPC NDO. 71 71 15 9 60.0 - EOC-5B inspection RPC Not inspected 1399 1399 386 , 39 _93 0 - No EOCy inspecen BM 59 _ _ l . New EOCy inspecen Indicamn 321_ _ _ _.53_ _ 45 84.9 Sum of An EOC-6 Inspechon inchcahon 1683 1945 552 511 92.6 All Voltages in EOC4 Inspection EOC-58 Inspecton Bobtan Left in Senace _4063 _, _3849 _534 _ _ __487 _ __912 _ r - EOC-5B Inspa$m RPC Co*mmt _ __.15L _ _ _ 157 __ _ 99 _, 99 1MD - EM5B inspm2m RPC NDD 102 ___. _ _jc2._ _ _30,_,_ 12 __ 40D - EOC-58 inspecnon RPC Not inspected 3590 3590 405 376 92.8 No EOC4 _esputon Bobbn

• 234 New EM6 Mecen inpcahn, _

_ 2935 _ 116 74 _. 63.8 sumof An EOC4inspectonindic ion 4o83 6784 650 56i 86 2 anocamme som a cassa on 1995 map tweenn vanage B-17 f repca T.bies4 sun 5 44 ru

i l l I' l l Figure B-1 Braidwood Unit I 1997 EOC-6 Evaluation for POPCD at EOC-5B i 1.3 t- - - - - - - - g_ _ _ _ _*_ _,.,.u..... - - - - - - ",,..... -.x - - - - ; - " t w C ~w....- 3 0.9 3 _ _.e- - - n i v.ac 0.8. t t .M y 0.7 3__ _o I t I x 3 0.6 i 5 I: u a

• 0'5

/ --a-RPC Confirmed X' .E 0.4 i e s--m u n. RPC Confirmed Plus Not inspected 0.3 X O.2 -a-i -- * -- EPRI POPCD 0.1 0.0 l l l l l l 0 0.5 1 1.5 2 2.5 3-3.5 Bobbin Amplitude B-18 roimme iwenamest s

Figure B-2 Combined POPCD Evaluation for 6 Post-92 I-gia for3/4" Dia Plants POPCD Based on RPC Confirmed Plus Not Inspected Indications 1.0 ,m _,_________ ___. ---- y ,,g__ ,r". O.9 / .H

• 0.8

/ / 0'7 / Mean a >c e 't 0.6 .b ) 0.5 -/- = / 2 .s .3 0.4 e n n. / -h 03 Combined POPCD from 6 Assessments 0.2 i --x-- EPRI POD Study 0.1 0.0 l l l l l l 0 0.5 1 1.5 2 2.5 3 3.5 Bobbin Amplitude B-19 POPCM&F21FW2WW9N6 50 PM - 6 i -= -- >.e r

Figure B-3 Combined POPCD Evaluation for 8 Post-92 I-g;*;.~ for 7/8" Die Plants POPCD Based on RPC Confirmed Plus Not Inspected Indications 1.0, 2-. X~#,.~ 0.9 s 1 3 3 O.8 } / Mean 0.7 p------_--------__________- e 2 C )0.6 ( ~ J

• 0.5

~.l D = C / .8 0.a 8 Y / 0.3 -f Combined Data from 8 Inspechons 02 --x-- EPRI POD Study 0.1 00 l l l l l l O.5 1 1.5 2 2.5 3 3.5 Bobbin A aptitude B-20 POPCM&F20$3WmE5 PM

4 ,t

i 4

5 3 y i. t m i iL s 3 l n e it n o c c e e d p f i s s y n n d o n u C I o t 4 s i z_ t a 1 r _ c D e m O w i 'E d o P o 5

n r

f L 2

I

,d a R I t e a P 5 I t c D E 9 2 e 9 p '. s t n sI x-ot P o 4 N 1 s 4 ru , I I. 2 e l BfoM d u e nd t r oe il uin p ga m ,~._ A t m i u Fl r 1 a u 2 v o ~ n EC b B i 4 A "-- b DC o CP 5 B PR l On 1 P o dd e e ns ia 4, bBmD oC C PO n l 1 P e f f i ( n i ) e / 5 l 0 j .u O n/ / M P f. 5 I 3 .n t i 3 ) i. t7 / 9 f 4i 0 Big 0 9 8 7 6 5 4 3 2 1 0 if 2 1 0 0 0 0 0 O' 0 0 0 0 F ST c2.E b '* y~_.e.E.2 OC PO P w

' Figure B-S Combined POPCD Evaluation for 14 Post.'92 I+;ia POPCD Based on RPC Confirmed Indications Only 1.0

---

il j p. _. f. _. _.,< _s-- - _ - _ - _ -1. _ _ _. _ _ _. _ _ _ _ _. _. _ _ _. _ _ _ n 0.9 -

7' / L.__. a,. - A--.-I / 0.8 n_._A O.7 .= / .$.0.6 y b /

• 0 y

.5 f a 7- - -s 3 / .E 0.4 I' [ l Data from 14 Inspectons / 0.3 -i I j --x-EPRI POD Study - 0.2 j a-1 - *-- 95% Lower Confkfence Limit 0.1 9 0.0 l l l l l 0 0.5 1 1.5 2 2.5 3 3.5 Bobbin Amplitude B-22 POPCOTSF2tFigB SsW97t3 35 PM

Figure B-6 Combined POPCD Evaluation for 14 Post '92 I%L _; POPCD Based on RPC Confirmed Plus Not Inspected Indications 1.0 ~~ ~ { _ - - ---"".Z._._._.__i 0.9 j e -=4 g_, / 0.8 5 - -Ai ? 0.7 / mA $ 0.6 S=*1=' .b r .C >. 0.5 Data from 14 Inspections g .c 0.4 8 [ s. 0.3 Mj - 95% Lower Confidence Limit 0.2 --0-EPRI POPCD 0.1 0.0 - l l l l l l 0 0.5 1 1.5 2 2.5 3 3.5 Bobbin Amplitude B-23 POPCOMF2F@@8%WE35 PM

f 5 ^ 3 j 3 l DC P O P sn I o R i t P a c E id Dn ~ 5 l 2 I

• ~

o O d P e t I c R e P p s E n hI t t wN _ - io Ds 7-Cl x. u 2 e l d B PP u eOd t i rP e lp e m,.., m ugt r A 4 i ai F nfn r 2 e o n i w b B C t l b AC o nP 5 B osR l 1 ir n r a o pd me , f s o a CB D C P O l 1 P A, / f,/ 5 l 0 x/I-l u I e / s / 2 3 'f#e

7s f

s rer O s 0 9 8 7 6 5 4 3 2 1 0 ~.a r i2 1 0 0 0 0 0 0 0 0 0 0 ra ceg b g E28 mceo p r l l' l}}