ML20198S730
| ML20198S730 | |
| Person / Time | |
|---|---|
| Site: | Sequoyah  |
| Issue date: | 12/31/1998 |
| From: | TENNESSEE VALLEY AUTHORITY |
| To: | |
| Shared Package | |
| ML20198S719 | List: |
| References | |
| SG-98-12-001, SG-98-12-1, NUDOCS 9901110404 | |
| Download: ML20198S730 (79) | |
Text
. . . . . _ - . . - . . . -. . - . . - . . . - .. . . . . . .. - . .- . . - . . .
br,
+- F,2,,;:.,4 i
SG-98-12 001 i
I
. l I
l i
i i L l SEQUOYAH UNIT-1 CYCLE 10 VOLTAGE-BASED REPAIR CRITERIA '
_ 90-DAY REPORT l i
i e
December 1998 _
l l
l r.
i l
e-9901110404 991229 ADOCK 050003 7
, PDR p .
i q:\ ape \tva98\tva90 day. doc l
e -
1 i
QE N$
SEQUOYAH UNIT-1 CYCLE 10 VOLTAGE BASED REPAIR CRITERIA 90 DAY REPORT TABLE OF CONTENTS Pace No.
.1.0 Introduction - 1-1 2.0 Summary and Conclusions 2-1
~
3.0 Sequoyah Unit-11998 Pulled Tube Data for TSP Locations 3-1 3.1 Sequoyah Unit-1 Pulled Tube Examination Results 3-1 3.2 Sequoyah Unit-1 Pulled Tube Evaluation for ARC Applications 3-4~
3.3 ' Comparison of Sequoyah Unit-1 Data with Existing ARC Correlations 3-5 4.0 EOC-9 Inspection Re,sults and Voltage Growth Rates 4-1 4.1 EOC-9 Inspection Results 4=1 4.2 Voltage Growth Rates 4-2 4.3 Probe Wear Criteria 4-3 4.4 Probability of Prior Cycle Detection (POPCD) 4-4 4.5 Assessment of RPC Confirmation Rates 4 4.6 NDE Uncertainties 4-6 l
5.0 Data Base Applied for ARC Correlations 5-1 6.0 SLB Analysis. Methods 6-1 7I
~
Bobbin Voltage Distributions ' 7-1 7.1 Probability of Detection . 7-1 7.2 Cycle Operating Time 7-2 ;
7.3 Calculation of Voltage Distributions 7-2 7.4 Predicted EOC-10 Voltage Distributions 7-3 7.5 Comparison of Predicted and Actual EOC-9 Voltage Distributions 7-3 8.0 Tube Leak Rate and Tube Burst Probabilities 8-1 8.1 Calculation of Leak Rate and Tube Burst Probabilities 8-1 8.2 Predicted and Actual Leak Rate and Tube Burst Probability for EOC-9 8-1 l 8.3 Projected Leak Rate and Tube Burst Probability for EOC-10 8 9.0 References -
9-1 q;\apc\tva\tva98\tva90 day.dx i
l
@ kh SEQUOYAH UNIT-1 CYCLE 10 VOLTAGE-BASED REPAIR CRITERIA 90 DAY REPORT 1
1.0 INTRODUCTION
This report provides the Sequoyah Unit-1 steam generator tube support plate (TSP) -
bobbin voltage data summary, together with postulated Steam Line Break (SLB) leak' rate and tube burst probability analysis results. These results support - !
continued implementation of the 2.0 volt voltage-based repair criteria for Cycle 10 as outlined in the NRC Generic Letter 95-05 (Reference 9.1). Information required by the Generic Letter is provided in this report including projections of bobbin voltage distributions, leak rates and burst probabilities for Cycle 10 operation. The l
- methodology used in these evaluations is consistent with the Westinghouse generic l methodology describe _d in Reference 9.2 as well as the methodology reported in the prior ARC reports for Sequoyah Unit-1 (References 9.3 and 9.4). ;
i The ' application of the TSP APC for_the Sequoyah Unit-1 SGs involves a complete, 100% Eddy Current (EC) bobbin coil inspection of all TSP intersections in the tube bundles of all four SGs and plugging of TSP indications greater than 2 volts which are confirmed by a Rotating Pancake Coil (RPC) probe. RPC inspections are also performed at certain locations exhibiting dent voltages and mixed residual signals. 4 The measured bobbin signals are used to predict SG tube leak rate and probability ;
of burst during a postulated SLB and show that they are within the allowable i regulatory limits.- l l au--
A tube segment from R 4C15 in SG-2 with two TSP intersections was pulled during this inspection for detailed laboratory examination. Results from leak and burst tests and metallurgical examination are presented in Section 3. Eddy current and repair data for TSP. indications are provided in Section 4. The actual EOC-9 voltage distributions as well as leak rates and tube burst probabilities calculated for these distributions are compared with the projections for EOC-9 conditions performed using the EOC-8 data. Leak rates and burst probabilities for the projected EOC-10 voltage distributions are reported in Section 8 and compared with allowable limits.
I _.-
h q:\ ape \tva\tva98\tva90 day. doc 1-1
6 h h 2.0
SUMMARY
AND CONCLUSIONS
\.
SLB leak rate and tube burst probability analyses were performed for all four SGs I based on their actual measured EOC-9 voltage distributions and the results
!" compared with the projectichs performed at the beginning of the cycle. The total l -number of indications found at tube support plates (TSPs) in each SG during the
- j. current inspection and the actual peak voltages are less than those projected at the l beginning' of the cycle using a constant POD of 0.6, per the Generic Letter 95-05 requirements, as well as those based on the voltage-dependent POPCD. Also, leak
- l. rates and tube burst probabilities calculated using the actual measured voltages are
- ~ well below those projected with both a constant POD of 0.6 and voltage-dependent POPCD. SG-1 was predicted to be the limiting SG at EOC-9, but SG-3 was found limiting based on the actual measured EOC-9 voltage data; however, the EOC-9 l leak rates and tube burst probabilities projected for SGs 1 and 3 differ by 0.04 gpm and 3.4x10 5, respectively, and therefore it not surprising to find a slightly higher leak rate and tube probability for SG-3 on the basis of the actual EOC-9 voltages.
For the actual EOC-9 bobbin voltage distribution, the largest SLB . leak rate is l calculated for SG-3, and its magnitude is 0.17 gpm. This leak rate value was _
calculated using the latest ARC database for 7/8" tubes submitted to the NRC, and it includes the latest pulled tube data from the Sequoyah units (1996 Sequoyah Unit-2 data). A voltage-dependent leak rate correlation can now be applied to 7/8" tubes based on the p-value for the slope of the leak rate correlation on a one-sided basis l meeting the Generic Letter 95-05 requirement. However, leak rates based on the i actual EOC-9 voltages for all SGs were calculated assuming that leak rate is independent of the bobbin voltage so that they can be compared with the projections performed at the beginning of Cycle 9 (BOC-9) which used voltage independent leak rates. For the limiting SG, SG-3, leak rate at the EOC '9 conditions was also l calculated using a voltage-dependent leak rate correlation. Its magnitude is 0.09 gpm, which is about one-half of that based on the voltage-independent leak rate data.
All leak rate values quoted are equivalent volumetric rates at room temperature. The !
SLB leak rates based on the actual voltages are substantially lower than the current i allowable SLB leakage limit of 3.7 gpm, and they are also below their corresponding projections. The corresponding conditional tube burst probability based on the actual l SG-3 voltage data is 1.9x10 5, and it is well within the NRC reporting guideline of 10-l 2 Thus, the results meet the ARC requirements for continued Cycle 10 operation.
i SG-3 is projected to have the largest indication as well as the highest tube burst I l_ probability at EOC-10 conditions and SG-1 is predicted to have the largest SLB leak rate. However, the EOC-10 SLB leak rates predicted for SGs 1 and 3 differ by only O.02 gpm; therefore, SG-3 can be considered to be the limiting SG at EOC-10. The EOC-10 leak rate projection was performed using a leak iate Wrsus bobbin voltage correlation meeting the Generic Letter 95-05 requirement. Using the NRC mandated
~
[ constant POD of 0.6 and the latest ARC database for 7/8" tubes, the limiting EOC-10 l SLB leak rate projected for SG-1 is 0.30 gpm (room temperature), which is more than an order of magnitude below the current licensed limit of 3.7 gpm (room I.. ,
q:\ ape \tva\tva98\tva90 day. doc 2-1
h Q l
temperature). The limiting EOC-10 tube burst probability calculated for SG-3 is 4.7x10-4, and it is also more than a decade below the NRC reporting guideline of10 2 A total of 376 indications were found in the EOC-9 inspection, and none of them were ever 2 volts. Although none of the TSP indications had to be inspected with a -
-I RPC probe (because they are all below 2 volts),102 indications were RPC inspected, and all but two were confirmed as flaws. The largest number of bobbin indications, i 130 indications, was found in SG-2; 15 of those were inspected by RPC, and they all were confirmed as flaws. No axial indications extending outside the TSP or volumetric type-signals were identified by RPC inspection at the TSP intersections.
An augmented RPC inspection was performed consistent with the NRC requirements. All dented intersections in the hot leg (HL) that had a bobbin voltage greater than 5 volts in the last inspection were inspected with a RPC probe.
Hot leg dented intersections below 5 volts were also tested in critical sections. Four TSP crevices at 1H elebation were found to have both indications with ID and OD phase angles. Three of these HL crevices were found in SG-3 (in tubes R5C62, R9C83 and R9C86) and one in SG-2 (R8C32); all these tubes were repaired. A small circumferential indication (0_._27 volt,45 extent) was found in a 0.6 volt dent at a TSP intersection in SG-3 that also had a small outside diameter stress corrosion cracking (ODSCC) bobbin indication (0.89 volt), and that tube was also repaired.
No a'ialx indication was found at this intersection by the RPC inspection. This could indicate that the bobbin inspection identified the circumferential indication, which may be attributable to a cellular corrosion patch. In some . cases, cellular corrosion leads to a circumferential coil response. No RPC indications were found at intersections with residual or copper signals. ,
A tube segment containing the first and second TSP crevices on the hot side of tube R4C15 in SG-2 was pulled during this eutage per the Generic Letter 95-05 requirement. The TSP-1 intersection had a bobbin voltage of 0.44 volt and the TSP-2 intersection 0.80 volt. + Point data for both TSP intersections revealed inultiple axial indications with some volumetric characteristics. The indication at the TSP-1 location had a maximum depth of 70%, an average depth of 44% and a macrocrack length of 0.551. inch. The corrosion was located entirely within the TSP crevice region. At the TSP 2 location, indic.ation had a maximum depth of 75%, an average depth of 38% and a macrocrack length of 0.648 inch. Both TSP specimens were tested for leakage and burst pressure at room temperature. Neither developed leaks at ditTerential pressures that ranged up to 2650 psi for the TSP-1 specimen and up to 2570 psi for the TSP-2 specimen. Both burst specimens developed axial l
- burst- openings and had a high burst pressure. The lowest burst pressure was measured for the TSP-2 location, and its magnitude is 8,480 psi. The crack l morphology for the indications is axial ODSCC with cellular patches typical of the EPRI database for APC application. Incorporation of the Sequoyah Unit-1 data in the EPRI ARC database would have negligible influence on the ARC correlations (SLB q:\apc\tva\tva98\tva90 day. doc 2-2 l
f @hJ l 4.a-M I
L structural limit, using 95%/95% lower tolerance limit material properties, 8.4 volts, does not change at'one significant digit).
l
[
~ - . .- _ _
l 1
e l
l .
l I l l
t -
l f
_ q:\ ape \tva\tva98\tva90 day. doc
- 2-3
- __ . ~. - - - - - - ~ _ - --
1
& il 3.0 SEQUOUAH UNIT-1 1998 PULLED TUBE DATA FOR TSP LOCATIONS 3.1 Sequoyah Unit-1 Pulled Tube Examination Results - -
3.1.1 Introduction A tube segment removed from the hot leg side of SG-2 of Sequoyah Unit -1 (R4C15) l was examined at the Westinghouse Science and Technology Center in support of the 2
, - volt ARC application. The examination was conducted to characterize corrosion _at the steam generator TSP crevice locations. The first and second TSP crevice regions l (TSP-1 and TSP-2) of the tubes were removed. Field eddy current inspection prior to l l
the tube removal revealed small indications (PI) at the TSP-1 and TSP-2 locations in the hot leg of the tube R4C15. +Peint probe data showed small multiple axial l indications (MAI) at these ,
locations.
. After nondestructive laboratory examination by eddy current, ultrasonic testing, radiography, dimensional characterization and visual examination, the TSP 1 and 2
-- regions of Tube R4C15 were te.s_ted at room temperature. Subsequently, room temperature burst testing was conducted on both TSP segments, as well as a free spa,n (FS) section of the tube R4C15. The two burst tested TSP specimens were i destructively examined using SEM fractography techniques to characterize corrosion.
Both burst tested specimens were further examined using metallography.
. 3.1.2 NDE Results The tube was inspected in the laboratory using a 0.720 inch diameter differential bobbin coil probe, a Zetec 3-Coil RPC probe and a Zetec + Point probe using eddy current techniq6es similar to those used in the field. The tube sections ofinterest to the ARC are associated with-first tube support plate (TSP-1) crevice region and second tube support plate (TSP-2) crevice region. The laboratory eddy current data was evaluated by the same analyst used in the past to examine the bulk 6f the tube specimens in the present ARC database.
Table 3-1 presents a summary of the initial field data calls, reevaluated field data and a summary of the laboratory eddy current data. The review of the field eddy l
current data for Tube R4C15 showed no' major differences between the original field calls and the reevaluated data. Negligible differences were noted in the indication signal strength (voltage) between the original field calls and the reevaluation of the l field data. + Point coil data show that both TSP regions contain MAIs (two distinct j
l axial indications in each) with some volumetric characteristics. The laboratory eddy
' current sTgnal strengths did not increase significantly, indicating that the tube removal did not significantly open the cracks.
- After completing the eddy current examination the tube sections were cut into 10 q
- \apc\tva\tva98\tva90 day. doc 3-1 l
4 .
inch lengths with the areas ofinterest at their center. These lengths were then X-ray radiographed; no indications were observed.
3.1.3 Leak, Burst and Tensile Data Following NDE testing, room temperature leak testing was performed on both TSP crevice region specimens. Neither developed leaks at differential pressures that ranged up to 2650 psi for the TSP-1. specimen and up to 2570 psi for the TSP-2 specimen. Leak testing was followed by room temperature burst testing of the TSP-I and TSP-2 specimens, as well as a FS specimen that had no indication of -
corrosion. -The FS burst specimen came from a location ceiltered 15 inches above the TSP-1 location. The burst pressures are recorded in Table 3-2. Both burst The specimens developed axial burst openings and had a high burst pressure. -
lowest burst pressure was for the TSP-2 location where the burst pressure was 8,480 psi.
1 In addition, a FS section of the pulled tube without NDE indications was tensile tested to provide data for normalizing the burst data. The FS tensile specimen was centered at a location 10 inches above the TSP-2 location. The tensile data are included in Table 3-2. The tensile properties are typical of Westinghouse mill
. annealed (MA) tubing.
Following burst testing, the burst specimens were examined for OD corrosion cracks on the burst fracture faces and on the adjacent OD surfaces. OD corrosion cracks were observed on both TSP-1 and TSP-2 burst specimens. Figures 3-1 and 3-2 provide sketches of the observed cracks with some additional cracks added from subsequent destructive examination information. The OD cracks were confined to the crevice regions for the TSP locations.
3.1.4 Destructive. Examinations _
SEM fractography was performed on each of the burst fractures that revealed OD Table 3-3 origin intergranular corrosion on the TSP-1 and TSP-2 fracture faces.
presents a summary of the results in the form of crack depth profiles and ductile ligament data. Each of the axial burst fracture faces had OD origin intergranular corrosion that occurred as a macrocrack composed of a number of OD intergranular microcracks joined together by ligaments. Most of these ligaments had o'nly or mostly intergranular features, indicating that they grew together during plant operation. Each of the burst corrosion macrocracks also had ligaments with predominantly ductile features, indicating that these particular ligaments formed (tore) during tube pulling (possibility excluded by eddy current data), burst testing (most likely situation), or subsequent laboratory handling. All OD macrocracks were axial, as were the individual microcracks. The longest macrocrack wis for the TSP-2 region where the macrocrack was 0.648 inch long with an average depth of 38% and maximum depth of 75%. The macrocrack for the TSP-1 region was 0.551 inch long with an average depth of 44% and a maximum depth of 70%. Both TSP q:\apc\tva\tva98\tva90 day. doc 3-2
- n.
t g l
macrocracks were confined to the crevice region of the TSP.
1 Table 3-4 presents a summary of the metallographic data obtained from the two TSP crevice regions of Tube R4C15. Metallography obtained from both locations showed OD intergranular corrosion, typicall of that found in most other tube l examinations. The maximum and average depths obtained by metallography on l the secondary corrosion (nonburst locations) at the TSP-1 and TSP-2 locations were l less than the maximum and average corrosion depths obtained by SEM fractography on the corresponding burst fracture faces.
l The OD origin corrosion had a morphology that varied' with elevation m some aspects, but not others. Corrosion morphology can be artificially classified by a number of measures. In addition to orientation (axial, circumferential, oblique), ;
i corrosion can be classified by density and DAV ratios. In this examination, both I
TSP locations examined had axial intergranular corrosion as the dominant corrosion morphology. iWith increasing elevation, oblique angled corrosion became !
increasingly present to the extent that at the TSP-2 elevation, large amounts of intergranular cellular corrosion (ICC) were present along with the dominant axial '
corrosion. As with otlier examination results, the_ ICC morphology tended to decrease more rapidly with depth than didThe axial IGSCC. !
A high crack density is defined as greater than 100 cracks around the circumference (typically measured at the mid-crevice region), a moderate crack density as 25 to 100 cracks, and a low crack density as less than 25 cracks around the circumference. In this examination, all three areas had moderate crack densities, with no perceived variation with elevation. ,
The third way of classifying corrosion morphology is by DAV ratios that provide a measure of the extent to which -three-dimensional intergranular attack (IGA) is associated with individual two-dimensional cracks. The DAV ratio is obtained by measuring crack depth and dividing by the crack width at the mid-depth of the crack. 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 ofIGA with IGSCC, and DAV ratios greater than 20 suggest a low association of IGA with IGSCC.
Measuring DAV ratios is difficult to do from transverse of axial metallographic sections, but easy from radial sections where the sample is progressively ground ,
and examined from the OD towards the ID surface of the tube. Finally, DAV.' ratios
' Note that there were hints or suggestions in some of the micrographs of very minor transgranular components to the overwhelmingly dominant intergranular corrosion. The hints showed as possible partial grain penetrations that could be argued as transgranular corrosion or as not being real. If the minor transgranular components are real, their presence suggests that lead (Pb) may have assisted in the development of the corrosion. No chemical analyses were performed on the tube samples.
2 D/W ratios less than I suggest that IGA is present, without IGSCC.
q:\ ape \tva\tva98\tva90 day. doc 3-3
1 l
Q .. I l
are not considered reliable or significant if the individual crack is shallow (less than l 10% deep). In this examination, both TSP locations had moderate D/W ratios. )
1 Summary i Both TSP regions had OD intergranular corrosion. The field eddy current data for the tube segments are in excellent agreement with the corrosion present in the TSP crevices. At the TSP-1 location, the corrosion was predominantly axial IGSCC with moderate ICC components. Tor the TSP-1 burst location, the corrosion had a
_ maximum depth of 70%, an average depth of 44% and a macrocrar.k length of 0.551 _
inch. The corrosion was located entirely within the TSP crevice region. At the TSP 2 ' location, the corrosion was predominantly axial IGSCC with large (signi5 cant) ICC components. For the burst location, the corrosion had a maximum i depth of 75%, an average depth of 38% and a macrocrack length of 0.648 inch. The corrosion was located entirely within the TSP crevice region. All of the burst macrocracks was com' posed of numerous axial intergranular microcracks with at least some of the ledges separating the microcracks having ductile features, indicating their creation during the burst test. The OD intergranular corrosion present at the TSP locations was typical of that in the EPRI database gathered in support of ARC.
s 3.2 Sequoyah Unit-1 Pulled Tube Evaluation for ARC Applications The pulled tube examination results were evaluated for application to the EPRI database for ARC applications. The eddy current data were reviewed, including reevaluation of the field data, to finalize the voltages assigned to the indications.
The data for incorporation into the EPRI database were then defined and reviewed against the EPRI outlier criteria ~to provide acceptability for the database.
3.2.1 Eddy Current Data Review Table 3-5 provides a summary of the eddy current data evaluations for the Sequoyah Unit-1 pulled tubes. These NDE data results have been discussed earlier in Section 3.1.2. As noted above, the field and laboratory reevaluations of the field bobbin data are in good agreement for the field call for both TSP specimens (R4C15, l TSPs 1 and 2). The reevaluated field bobbin voltages, including the adjustment for cross calibration of the field ASME standard to the laboratory standard, are used for the EPRI ARC database. The reevaluation was performed by the same analyst that performed a large part of the EPRI pulled tube database and the use of these voltages minimizes analyst variability in the database, which is separately
[ accounted for in ARC applications as an NDE uncertainty.
l q:\apc\tva \tva98\tva90 day. doc j 3-4 t
Q; M 3.2.2 Sequoyah Unit-1 Data for ARC Applications The pulled tube leak test, burst test and destructive examination results are summarized in Table 3-6. Both TSP-1 and TSP-2 sections of tube R4C15 tested did
- not leak at SLB conditions. Both burst specimens developed axial burst openings .
and had a high burst pressure. The lowest burst pressure was for the TSP-2 location where the burst pressure was 8,480 psi.
The Sequoyah Unit-1 pulled tube results were evaluated against the EPRI data exclucion criteria for potential exclusi_ons from the database. Criteria la to le apply primadly to unacceptable voltage, burst or leak rate measurements and indications without' leak test measurements; these criteria are not applicable to the present Sequoyah Unit-1 TSP specimens. Criterion 2a applies to atypical ligament morphology for indications having high burst pressures relative to the burst / voltage correlation and states that high burst pressure indications with s; 2 uncorroded ligaments in shallow cracks < 60% deep shall be excluded from the database. Table 3-6 identifies the number of remaining ligaments and the maximum depths for the indications. Indications in both TSP-1 and TSP-2 sections of tube R4C15 have three or more uncorroded ligaments. Therefore, exclusion criterion 2.a is not applicable to these Sequoyah Unit-1 pulled tube specimens. Crimrion 3 applies to potential errors in the leakage measurements and is not applicable to the Sequoyah Unit-1 indications with no leakage. _
As shown in the last column of Table 3-6, the TSPs 1 and 2 indications of R4C15 are to be included in the probability ofleakage and burst correlations. The impact of the indications on the ARC correlations is further discussed in the section below. -
3.3 Comparison of Sequoyah Unit-1 Data with Existing ARC 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 Sequoyah Unit-1 in 1998 (R4C15, TSPs 1 and 2). The results from the leak and burst tests, and destructive examinations of the tubes are presented in Sections 3.1.3 and 3.1.4.
The Sequoyah Unit-1 pulled tube data germane to the ARC correlations, and the bobbin amplitudes for ARC applications, are presented in Table 3-6. The results of the destructive examinations, e.g., burst and leak tests,.are compared to the' data-base of similar test results for 7/8" outside diameter steam generator tubes. In addition, the effect of including the new test data in the reference database was
! evaluated. In summary, the test data are consistent with the database relative to the burst pressures, and the probability ofleak as a function of the bobbin ampli-l tude. Neither of the specimens exhibited leakage at SLB conditions; hence, there is no effect on the database of ODSCC leak rates. The comparisons and evaluations are discussed below.
The reference database used for the evaluations for this report is the same as the q:\ ape \tva\tva98\tva90 day. doc 3-5 l
N g EPRI recommended database as described in EPRI NP-7480-L, Addendum 2 (Reference 9.6). This was provided to the NRC staff during the second quarter of this j year to satisfy the provisions of the NEI/NRC protocol for updating the database. 1
- 3.3.1 Suitability for Inclusion in the Database _
The information on the destructive examinations of the tube sections was reviewed I
relative to the EPRI guidelines for inclusion / exclusion of tube specimen data in the alternate plugging criteria (ARC) database. This review revealed no information !
that would lead to a conclusion that the data should not be included in the data- i base. Therefore, the resulting correlations should be considered applicable to the use of ARC for indications in 7/8" diameter tubes in Westinghouse SGs.
3.3.2 Burst Pressure vs. Bobbin Amplitude ,
1 i
The result from burstitests, performed on both tube specimens which exhibited a non-zero bobbin amplitude a' the TSP location, were considered for evaluation. The measured burst pressures of the Sequoyah Unit-1 specimens are depicted on Figures 3-3 and 3-4 relative to the burst pressure correlation developed using the reference database.
1 "1. t A visual examination of the data relative to the EPRI database indicates that the measured btirst pressures fall well within the scatter band of the refer-ence data, see Figures 3-3 and 3-4.
- 2. The data points fall within a 95% confidence band for 90% of the population (5% in each tail) about the regression line, hence no statistical anomaly is indicated, see Figure 3-3.
The net result is that the visual examination does not indicate any significant departures from the reference database, although one of the burst pressures is less than would have been expected from such indications, i.e., the pressure is below the mean of the data as calculated by the regression line. Based on the placement of the new data, it may be judged that there would be no significant effect on the analysis of the residuals of the regression; either on the scatter plot of the residuals as a function of the predicted burst pressures or on the normal probability plot of the residuals.
Since the two 1998 Sequoyah Unit-1 burst pressure data points were not indicated to be from a separate population from the reference data, the regression analysis of the burst pressure on the common logarithm of the bobbin amplitude was repeated with the additional data included. A comparison of the regression results obtained _
by including these data in the regression analysis is previded in Table 3-7. Regres-sion predictions obtained by including these data in the regression analysis are also shown on Figures 3-3 and 3-4. A summary of the changes is as follows:
q:\ ape \tva\tva98\tva90 day. doc 3-6
Q ,b
- 1. The intercept of the burst pressure, Ps, as a linear function of the common logarithm of the bobbin amplitude regression line is decreased by 0.1%, or about 10 psi. This has the effect of decreasing the predicted burst pressure as a function of the bobbin amplitude for small amplitudes.
~
- 2. The absolute slope of the regression line is decreased by 0.2%, i.e., the slope is less steep. This has the effect of increasing the burst pressure as a function of bobbin amplitude for large indications.
- 3. There is a decrease in the standard error of the residuals of 0.8%. The effect of this change is reflected in.a sligTtly smaller deviation of the 95% predic-tion line from the regression line.
The net effect of the changes on the SLB structural limit, using 95%/95% lower tolerance limit material properties, is to increase it by 0.05V, but this does not affect the value at one signiScant digit, i.e., the value remains at 8.4 V. The decrease of the intercept and the decrease in the standard error, coupled with the fait that the structural limit is unchanged, indicates that the probability of burst
~
would also decrease for bobbin indications over the lower portion of the structural range ofinterest. Based on the relatively small change in the str~betural limit, the change in the probability of burst would also be expected to be small. Predicted values of the probability of burst of a single indication as a function of the bobbin amplitude are illustrated on Figure 3-5. The probability of burst is redoced slightly up to amplitude of about 1.5 V. Beyond that value the probability of burst is increased by less than 1%.
3.3.3 Probability of Leak -
The Sequoyah Unit-1 data were examined relative to the reference correlation for the probability ofleakage (pol) as a function of the common logarithm of the bobbin amplitude. Figures 3-6 and 3-7 illustrate the Sequoyah Unit-1 data relative to the reference correlation. The specimens exhibited expected pol behavior. Based on the data examination, there is no significant evidence of irregular 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 present two Sequoyah Unit-1 data' points and a Generalized 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 3-8. These results indicate:
- 1. A 0.2% change (larger negative value) in the logistic intercept parameter.
- 2. A 0.2% increase in the logistic slope parameter.
- 3. The values of the elements of the covariance matrix of the parameters q:\ ape \tva\tva98\tva90 day. doc 3-7
I h Q..}
i changed 0.5%. Examination of Figures 3-6 and 3-7 indicate that there is no visually perceptible change in the pol at any bobbin voltage; hence, the impact on the 95% confidence bound on the total estimated leak rate from a l
single SG would not be expected to be significant.
' ~ ~
- 4. The mean square error (deviance divided by number of degrees of freedom) decreased by 1.5%.
In order to confirm the judgment that the changes are not significant, the reference correlation and the new correlation were also plotted on Figures 3-6 and 3-7. An examination of the figures 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, as is supported for the analysis of 7/8" diameter tubes, 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 or insignificant relative to,the calculation of the 95% confidence bound of the total leak rate from a SG.
3.3.4 Leak Rate vs. Bobbin Amplitude Neither of the specimens leaked at SLB temperature and pressure difference conditions; hence, the pulled tube specimens from Sequoyah Unit-1 have no effect on the leak rate correlation obtained from the reference database.
3.3.5 General Conclusions e The review of the effect of the Sequoyah Unit-1 data indicates that the burst pressure and the, probability of leak correlations to the common logarithm of the I bobbin amplitude would not be meaningfully changed by the inclusion of the data. -
Therefore, it is likely that the conclusioirs relative to EOC probability clburst and
' EOC totalleak rate based on the use of the reference database would not be signifi-cantly affected by the addition of the Sequoyah Unit-1 data from R4C15 TSPs 1 and l
2.
1 a
R q:\apc\tva\tva98\tva90 day. doe 3-8
i i
3.
Table 3-1. Summary of NDE Results ;
Field Eddy Current Sizing Information Laboratory Eddy j Current Location Ilobbin + Point , Ilobbin + Point llobbin + Point X-Ray .
Coil Volts /% - Coil Volts /%/l(in.) Coij Volts /%/1 (in.)
Volts /% Volts /% Volts /%
TSP 1 0.4G/PI 0.23/SAI 0.44/PI ' O.2/G0/0.4 0.34/7G 0.2G/50/0.53 NDD $ct.
0.16/45/0.07 0.3/30/0.21 TSP 2 0.74/PI 0.3G/SAI 0.8/65 0.3/50/0.25 0.82/75 0.29/30/.43 PI j , 0.12/40/0.15 0.12/<20/.48 l l 1
NDD - No Detectable Degradation r j SAI-Single AxialIndication , ,
PI - Possible Ir.dication I Table 3-2. Room Temperature Iturst and Tensile Data for Sequoyah Unit 1 S/G Tipbe R4CIS Location llurst Ilurst Ilurst Iturst 0.2% Offset Tensile Tensile Pressure, Ductility Length, Width, Tensile Yield Ultimate Elongation psig % inches inches Strength Strength % _
psi psi *fD FS* 12,100 37.0 1.531 0.332 57,000 106,700 41.6 TSPl
- 10,100 17.3 0.884 0.278 t TSP 2* 8,440 (run 1) 3.1 0.21& 0.15 0.015 & 0.011 TS P2*
- 8',480 (nm 2) 18.3 1.11I 0.377 l I
TSP = tube support plate; FS = free span; S/G = steam generator !
I
= Tested without foil or bladder.
- = Re-tested with foil and bladder. The initial test run devel, oped two leaks in an ICC/ IGA patch at a pressure of 8,440 psi but I without developing ductile tears at the burst opening ends. The re-tested nm 2 specimen has clear ductile tears at the burst ends ;
and is a valid burst.
q:\apc\tva\tva98\tva90 day. doc 3-9
a i
4 Table 3-3. SEM Fractographic Data for OD Inteigranular Macrocracks on Tube R4CIS Positional and Ductile Ligament Data d
Specimen, Length vs. Depth * & Ligament (Area = inches
- x'10 ; Orientation of Ligament Minor Axis reistive to location Location (inches /% throughwall) Macrocrack Major Axis in degrees; Orientation of Ligament Major Axis relative to Tube Radius in degrees **) i t
TSPI (main burst 0.00/00 Crack top located 0.13" below TSP top @ 180*
macrocrack with 0.038/42 -
intergranular 0.076/48 f?
corrosion) 0.114/58
' Ligament 1: Area = 6.6; Minor Axis @ 90*; Major Axis @ 0*
0.152/58* "8"""
t 0.190/58 0.228/64 O.266/706 Maximum Depth ,
i 0.304/60 l 4
0.342/49 ,
i 0.3 80/40+ "8'""* 2 Ligament 2: Area = 3.8; Minor Axis @ 90*; Major Axis @ 0*
O.418/00 6 "8'"*"2 0.456/44 ,
Ligament 3: Area = 12.6; Minor Axis @ 90*; Major Axis @ 0*
O.494/50 .
0.532/26 ,
(0.551/00)
(LAD * = 44%, Crack length = 0.551") Crack bottom at 0.681" below TSP top @ 180*
.i.
- Average depths may be calculated by a number of different methods depending upon the need. Methods used are LAD = linear average depth; ATD = average D -
throughwall depth (length weighted average depth); PDA = percent degraded area (relative to cross sectional area of a nondegraded tube). !
- Note that the ductile ligaments in the table are described by both a major and a minor axis orientation. The ligaments are usually considerably deeper / longer (major axis) than wi,de (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 of ID origin cracks) and is usually close in or! otation to the radius of the tube. The orientation of the major axis is relative to the tube radius. The minor axis of the ligament is the obser ved orientation 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
- I major axis. Usually the minor axis is close to perpendicular to the tube major axis.
s q:\apc\tva\tva98\tva90 day. doc 3-10 i
i i
Table 3-3 (Continued). SEM Fractographic Data for OD Intergranular Macrocracks on Tube R4CIS Positional and Ductile Ligament Data Specimen, Length vs. Depth * & Ligament (Area = inches 2 x 10-4; Orientation of Ligament Minor Axis relative to y location Location (inches /% throughwall) Macrocrack Major Axis in degrees; Orientation of Ligament Major Axis relative to Tube Radius in degrees") ,
TSPI (main burst 0.00/00 . Crack top located 0.02" below TSP top @ 135* [
macrocrack with 0.038/28
' Ligament 1: Area = 3.6; Minor Axis @ 90"; Major Axis @ O' intergranular 0.076/38+ "8'""
corrosion) 0.114/54 .
0.152/48 QS 8
O.190/50 l 0.228D56 Maximum Depth 0.266R50 Maximum Depth - :
0.304/63 0.342/52 1 l 0.380/24* "8'""2 0.418/32* "8'""3 Ligament 2: Area = 0.S; Minor Axis @ 80*; Major Axis @ 30*
0.456/20* "8'""
- Ligament 3: Area = 0.3; Minor Axis @ 90*; Major Axis @ 30*
j ' ;
0.494/28 Ligament 4: Area = 0.4; Minor Axis @ 90*; Major Axis @ 0*
0.532/36 0.570/24 * "8'"" ' [.
i 0.608/22* "8'"" '
O.646/10 Ligament 5: Area = 0.3; Minor Axis @ 90*; Major Axis @l0*
t Ligament 6: Area = 0.l; Minor Axis @ 90*; Major Axis @ 0" 3 (0.648/00) ua >
(LAD * = 38%, Crack length = i --v t
0.648")
Crack bottom at 0.668" below TSP top @ 135' ,
I l
q:\apc\tva\tva98\tva90 day. doc 3-11 ;
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ ___m_____._ _ _ _ _ . _ _ _ _ _ _ _._.____ ____ _ _
i
.l ~
Table 3-4. Metallographic Data from Spquoyah Unit 1 Tube R4C15 Contribution of. - Typical. DAV Estimated Maximum Ratio from the ICC Components Number of OD Cracks at .
Deeper OD -
Max / Avg. to the Axial OD Number of Section OD TSP Mid-Crevice Cracks Specimen Section Depth IGSCC Cracks at Length Cracks Location (based on Location Type (% Throughwall) - (based on radial (from transverse Per Inch metallography and visual Mid-Crevice (inch) metallography . sections) data) '
12' 38 (relatively uniformly. 44/18 Transverse 23 OD 2.5 9- ,
Moderate ICC TSPI distributed around depth = 2% .
Radial 1 14 OD _031 45 circumference with a local depth = 10% Moperate ICC (?!
8OD 0.31- 26 Minor ICC Radial 2 concentration 150* to 180*, depth = 21%
Radial 3 5.OD 031 16 Negligible ICC
,10 the burst and radial met. depth = 46%
Radial 4 3OD 0.31 location) 14 60/28 60 OD 2.5 24 75 (72 cracks from near TSP 2 Transverse depth = 2% Significant ICC ,
0.12 83 burst @ 13'5' to 300* with 3 '
Radial 1 10 OD depth = 10% Significant ICC 14 OD 0.22 -64 more near 45*)
Significant ICC Radial 2 depth = 20%
14 OD 0.22 64
- Radial 3 depth = 41% Moderate ICC
'li OD 0.22 50 N'egligible ICC Radial 4 depth = 59%
4OD 0.22 18 Radial 5 1
...m.
l 0,%)O
- I f
1 3.lg q:\apc\tva\tva98\tva90 day. doc
' s
I Table 3-5. Summary of Sequoyah-11998 Pulfed Tube Eddy Current Results Field Call Lab. Reevaluation of Field Post Pull Data Tube TSP Data Bobbin + Point Bobbin Depth + Point Bobbin + Point Volts Volts (1) Volts Volts Volts Volts (l) '
. i 1 0.4G 0.23 0.44 PI 0.20 0.34 0.30 SAI 0.16- 0.26
' I MAI MAI .
5f.g R4C15
^
2 0.74 0.3G 0.80 65 % 0.30 0.82 0.29 SAI 0.12 0.12 '
MAI MAI Notes: 1. Field and laboratory data include cross calibration of AShkE standard to the reference lab 9 ratory standard.
6 l Table 3-6. Plant Sejtuoyah-11998 Pulled Tube Data for ARC Applications Destructive Examination Leak Rate-l/hr Burst Pressure Data - ksi Use in Bobbin Data Tube T I RPC Results I Corr.
Volts No. N. O. SLB Meas. cy ou Adj.(3) Note 4 Max. Avg. Crack em Burst Volts Depth Depth Depth Length Lig.(2) 1300 2560 Burst j*L psid psid Press. Press.
0.0 0.0 10.100 8
8.487 B, POL R4C15 1 0.44 DI 0.20 70% 44% 0.551" 3 6 0.0 0.0 8,480 7.126 B, POL !
2 0.80 65% 0.30 75% 38% 0.648"
" 12.100 57.0 106.7 10.168 i
- Notes
- '
- 1. FS is freespan section of tubing with no tube degradation to obtain tensile properties.and undegradfd tubing burst pressure.
- 2. Number of uncorroded ligaments with > 50% of ligament length remaining in burst crack face. :
i 3. Burst pressures adjusted to 68.78 ksi, average flow stress at 650* F for 7/8" diameter tubes (Reference 3.6). *
- 4. B = data to be used in burst correlation. POL = data to be used in probability of leakage correlation. L = data to be used in leak rate correlation.
q:\apc\tva\tva98\tva90 day. doc 3-13 i
l i
O(.G f[
= -, -. , ,
l Tt ble 3-7: Effect of Sequoyah 1 Data on the Burso Pressure vs. Bobbin Amplitude Correlation Pa = ai + a2 log (Volts)
- Parameter Reference Database with New / Old Database Value Sequoyah-1 Ratio oti 7.58911 7.57978 0.999 a2 -2.40111 -2.39554 0.998 r2 i 82.7% 82.7 % 1.000 cError 0.82652 0.82019 0.992 i e Structural Limit 8.4 V 8.4 V 1.006 N (data pairs) 85 87 $$;ggisp5 l
p Value for a2 1.2 1033 2.01034 0.166 47.6190 SS log (Volts) 46.9613 L Mean log (Volts) 0.3349 0.32537 Reference or 68.78 ksi "*S V"*""^ "^ ***2 MIEN me ll i
j q:\apc\tva\tva98\tva90 day. doc L 3-14
i
$.$l: [h u-i t i
? . - _ _
i-Table 3-8: Effect of Sequoyah 1 Data on the Probability of Leak Correlation 1
Pr(Leak) = g_7p ,,,4 yaun 1 _
Parameter '
Reference Database with New / Old Database Sequoyah-1 Ratio pl -4.26272 -4.27243 1.002 p2 4.16746 _ 4.17616 1.002 _
Vilm 0.68535 0.981701 0.995 V12 -0.30365 -0.600341 0.995 V22 0.60210 0.599183 0.995 .
DoFW 128 130- sij$ M ME Deviance 78.67 78.69 1.000 MSE 0.615 0.605 0.985 Pearson SD 0.766 0.761 0.993
- Notes: (1) Parameters Vy-are elements of the_covariance matrix of the coefficients, pi, of the regression equation.
(2) Degrees of freedom. -
r-q:\apc\tva\tva98\tva90 day. doc 3-15
1 If Ij'")
I l
1 l
i
?
0.75 TSP Top
\ l !
l
! /
I o ) 'h Distance I f ,
From t I
lt ,/ , ,' e, ,,}, 4 TSP '
I g is i Bottom ,' .
I/
(Inches) e g
i 0.0 TSP Bottom I
I i i i 0* 90' 180* 270* 360' Circumferential Orientation (Degrees)
, Figure 3-1 Sketch of the OD crack distribution observed on the TSP-1 region of the tube R4C15. The burst opening also had OD crack features on its fracture face that were confined to the TSP crevice q:\ ape \tva\tva98\tva90 day, doc l 3-16 -
- - -. . _ ~ . _ _ _ _ . . _ - .. .. -
l l
0.75 TSP Top
), -
-)e Distance f From ft gj ,g g, ;
TSP i ii, ,
(
I it Bottom 'l '
' ' /g g(ts\I\ \l ,
g 'g 'I 5
- \
(Inches)
~
i
),
0.0 TSP Bottom
]
i i i O' 90' 180' 270' 360' Circumferential Orientation (Degrees) r
, Figure 3-2 Sketch of the OD crack distribution observed on the TSP-2 region of the tube R4C15. The burst opening also had OD crack features on its
~ fracture face that were confined to the TSP crevice q:\apc\tva\tva98\tva90 day. doc I
! 3-17 l
1
-1 I !
Burst Pressure vs. Volts for 7/8" Alloy 600 SG Tubes -
Addijional Data, Reference dr = 68.8 ksi @ 650 F 12.0 l l llll l
o NDD Tubes _
o Model Boiler Database i
's.
s A Pulled Tubes 10.0 a
s .^' A Added Burst Data xA , Q Reference Curve fli A l' 'A' N g'g 's. e,'N sN% }d$ 4 ' S.
!I ! _ . _ . 95%/90% Tol. Band -
- L-j
s.,.
^ '-
Ak^b jA [P x' -g '
I j
g s Q_ i
^
'u '"s. '
$i iI
1 '^ ^ q ' I "
E 6*0 'N 'd -
s, t
g 3' -
s,j ,
A I
i 2p.. 3 gog\o x N.
o o -s, o .__
-s u w n i . g o o ; ,
a -
y a A .s.'N N. "
^
4.0 oN
%s
\
.N.
ant
. x. ' a <? i 2.0 ,
s'-
i p.
- l.
s .
s ,
I l
O.0 1.0 10.0 100.0 0.1 .
I Bobbin Amplitude (Volts)
Figure 3-3 Q \APC\TVA\TVA98\ Burst Leak. doc 3 - 18
Burst Pressure vs Volts for 7/8" OD Alloy 600 SG Tubes Reference Database, Referenc'e or = 68.8 ksi @ 650 F ,
12.0 ,
a o NDD Tubes 1
o Model Boiler Database 10.0 a -
3 8
! A Pulled Tubes ,
N4a s i .
+ LS Reference .
46 \ # ^ I '
x 95% Prediction @ LTL bN 2
A l! '
LS w/New Data _
'm A ka's\s# I !
--- 95% P @ LTL w/New
} 8.0 v '* s '-'A A 3
U m h^~ ~
a m
m 60
'm,* -
~,
s A
^ ^ s '
ss o
A u 6 u
.LJ
^
g
'* i m ~ ~ -'n, a o u ~
i ,
A, c o m *~*~L^ l Y
_ . . _ . j 3.657 ksij,,,,_, _, ,, _, _ ._ , _,_,_,_, _,_,._,_. _ , _ ,
_^, u gs
. J's a s s g g ~~ v s .'
L._ ~%
4 0 s.
' 0 w ')' -
2.0 l m'
- R'
__ L _
4 ,' !
M 8.4 V i i i !
0.0 O.lg 1. 10. ,
100.
Bobbin Amplitude (Volts)
Figure 3-4 i
Q \APC\TVA\TVA98\Bisrst Leak. doc 3 - 19 i
,1 l lll I!l l :llIl lIj
,?.,I 3,%
A -
x 0.
0 1
e s L .^ &
a T w N '
b L e 0 a N '
s g
t . i a @ o nt a/w s
D o o!
3 r s ei aL s e el b n c i
e c t i
D T wL g s b o u e rd e uBT T le d e P/P e e f r N@ .
.p x s
s D e d D o uS l
l eR 5%w% '
L9 S 5 i
i' E
b NMPL9 a ' '
TuF0 4 o a A + x - %
a G5
- U v' o.
S6 a%s 0@
0i -
~ .
0 6 s y k
'p. a .
S
.I .l; ,,
V 4 il 1
o8 o' A 3, 8 )
s l%oaI l
l 8 ~_
t l
A6 5 I !
s o V
N x _ (
D r ' -
e Ooe
- _
d q'
. 1
, _ t u 4-
"8 c 4 , i
/ n I I
n _ l p 3 7 e 3 ^ ' ,
e 9 r _ m r 1 r
of e A ^ ,
A u g
_ 3 f e '
j.
, n i F
sR ^ _
i b
t l
e s ,
b o
o s A ~ ,
B Vab 3-
- _
svta
^4* ' _ .
's s -, , 1 e a 1.i I A _
rD ^ I s
ue s c {
'a s n s s' ,
A ee r re ^ m _
Pf e
- _
t ' _
sR r \ ,
u J4 .- -
j, B
i s
~ k N i x
I 7
A 5 6
3 j
, c o
a x
- _ d.
k a
_ 1 e
"' l 0 t s
0 0 0 0 0 0 0 r u
2 1
0 1
8' 6 4 2 0 B
\
8 9
j g@"CE yym ) A V
T
\
A V
T
\
C P
A
\.
Q I iil
l Probability of Burst of 7/8" SG Tubes During SLB, ,
Effect of Additional Data 1E+00
- f-1E-01 /
--- PoIV2560 psi) w/o New Data
-PoB(2560 psi) w/ New Data /
/
1E-02 1E-03
= Structural Limit, l l ,
[; l 3
I 8
3
- +,)
1E-04 h
t 1E-05 l
I ' l': '
x /
1E-06 4'
R 1E-07 /
/ ;
/
/ '
1E-08 /
/ i
/ n h"
1E-09 +
i 1E-10 ' ,
IE-11 I
0.1 1 1. 10. 100.
Bobbin Amplitude (Volts) i i Figure 3-5 t
8 Q-\APC\TVA\TVA98\Durst leak. doc 3 - 20
r.
l D -
E5j c.
o t
g
'-- -.4--
. _. o 1 . _ _
i 6
\ ~j s
i3 l:
s e .
N _
\
~
5 N A N m c!
g 5T N v o e*Z; o Z
~ ~:
N;
- o h N L_ N i
-= 4 e,, a n _.. . .. \
l Eu o N g
.( c o A. I o
.N. *o m m
4 c ! c 4T '
m oc : c e.
m-u . _ . _ . .
, i_
- o
.t
" e o H
- o tu yao I
E L N
_-. . - . L 1
_ ___ __. /r -
M / - -
gws2 _
y .c
- N GM ~~ w- A
$o A -
.- p y _ . . ._
oo %
xc
. a E-t o a y -
00
- e ,
4w n= ,
p E Q *%p l 2 c
~
4o -
5a k >
- .&. ci om -
o - ; u wo -
p Z o l bd -
_ o o a so%
d j H c c $ > \
1 =
oo s
oOoo u
,.o ymm t-o o oc .
c: ZEE !
I a < ! l; l y
- - 1 i l c o
]
.3 l 8 8 o
8 o
8 o
8 8 o
8 e
8 o
8 e
8 o o 8 :
a o .o o o a w o e n n m - 9 MNT Jo AUU98903d [E E
- s a
4.
I Comparison of Probability of Leak ,
7/8" x 0.050" Alloy 600 MA SG Tubes, w/ & w/o New Data
/i W .
Calculated Pols
/
[ '
i
Exact Match .
Q
$ I i.
I
/ ;
( ;,
1 l i Ii c ili i
.S 0.100 _ i ,
.2 .
' . ,:4_ . .
4 / A
~ / r y
k l
A
/
f x
t
$ 0.010 , ,
Id j
/ .
$ / h
$ l . E'1 .
.n.
/ >
/ i l ,
0.001 0.001 0.010 0.100 1.000 ;
- yrobability of Leak w/o New Data Figure 3-7 l l t
[
Q.\ APC\TVA\TVA98\ Burst Leak. doc 3 - 22 f
$ Q.?
4.0 EOC 9 INSPECTION RESULTS AND VOLTAGE GROWTH RATES 4.1 EOC-9 INSPECTION RESULTS In accordance with the ARC guidance provided-by the NRC Generic Letter 95-05 . _
(Reference 9-1), the EOC-9 inspection of the Sequoyah Unit-1 SGs consisted of a complete,100% EC bobbin probe, full length examination of the tube bundles in all four SGs. A 0.720 inch diameter probe was used for all hot and cold leg TSPs where ARC was applied. Allindications detected had a bobbin voltage under 2 volts. One hundred and two indications were inspected with a RP_C probe in all SGs combined, and all but two were confirmed as flaws. Since the bobbin voltage for all RPC
_ confirmed indications were less than 2 volts, no tubes were removed from service because of ODSCC causes. There were no volumetric signals at the TSPs, and no indications extending outside the TSPs.
An augmented RPC inspection was performed consistent with the Generic Letter 95-05 requirements. The augmented RPC inspection using the + Point probe included examination of all TSP intersections in all four SGs with a dent voltage over 5.0 volts in the last inspection. In certain critical areas dents below 5 volts
(
were also tested with a + Point probe. As in the prior inspection, axial PWSCC indipations were found at dented intersections including some that extended outside the TSP intersection Circumferential PWSCC and ODSCC were also found at dented intersections. Four TSP crevices at 1H elevation were found to have both indications with ID and OD phase angles. Three of these HL crevices were found in l
l SG-3 (in tubes R5C62, R9C83 and R9C86) and one in SG-2 (R8C32); all these tubes '
l were repaired. The ODSCC indication at RSC32-1H in SG-2 was detected during l
retest with a RPC probe. A small circumferential indication (0.27 volt,45 ext'ent)
I was found in a dent at a TSP intersection in SG-3 that also had a small ODSCC flaw (0.89 volt) ahd that tube was also repaired.
A summary of eddy current (EC) signal voltage distributions for all steam generators is shown on Table 4-1, which tabulates the number of field bobbin indications, the number of these field bobbin indications that were RPC inspected, the number of RPC con'irmed indications, and the number ofindications removed from service due to tube repairs. The indications that remain active for Cycle 10 operation is the difference between the observed and the ones removed from service.
No tubes were deplugged in the current inspection with the intent of returning them to service after inspection.
i
! Overall, the combined data for the Sequoyah Unit-1 steam generators show the t
following:
A total of 376 bobbin signals were identified as TSP indications during the i
- inspection, and they were all called as PIs.
q:\ ape \tva98\tva90 day. doc 4-1
b P
- Of the 376 PIs,56 were above 1 volt, but none exceeded 2 volts.
, a A total of 102 indications (all below 2 volts) were RPC inspected and all but 2 were confirmed as flaws. This RPC confirmation rate is unusually high as the average bobbin voltage for the RPC confirmed indications is below 0.9 -
- i volts. It may be attributable to clean tubes in the Sequoyah Unit-1 SGs as they were chemically cleaned during the Cycle 7 refueling outage. ;
- All 8 indications removed from service were repaired for degradation mechanisms other than ODSCC at TSPs. 1 A review of Table 4-1 indicates that more indications (a quantity of 127, with 10 indications above 1.0-volt) were returned to service for Cycle 10 operation in SG-2 l than other SGs; however, SGs 1 and 3 had more indications over 1 volt (15 and 18 i indications, respectively) returned to service. Therefore, SG-1 or SG-3 is likely to be l the limiting SG at EOC110. l Figure 4-1 shows the actual bobbin voltage distribution for tubes that were in service l during Cycle 9, as deterdlined from the EOC-9 EC inspection. Figure 4-2 shows the i distribution of the EOC-9 bobbin indications that were repaired and taken out of l service, and Figure 4-3 shows the bobbin voltage distribution ofindications returned l to service for BOC-10.
The distribution of EOC-9 indications as a function of support plate elevation, I summarized in Table 4-2 and illustrated on Figure 4-4, shows the predisposition of ,
ODSCC to occur in the first few hot leg TSPs (240 of the 376 PIs, or about 64%,
occurred in the first two hot leg TSPs), although the mechanism does extend to higher TSPs. Forty-six b'obbin indications, or about 12% of the total, were reported on the
~
cold-leg side and this percentage has remained essentially the same as in the last inspection. Occurrence of majority of ODSCC indications in the first two TSPs on the hot leg side in Sequoyah Unit-1 show predominant dependency on temperature, which is consistent with that observed at other plants.
4.2 Voltage Growth Rates For projection of leak rates and tube burst probabilities at the end of Cycle 10 operation, voltage growth rates were. developed from EOC-9 inspection data and a reevaluation of the same indications from the EOC-8 inspection EC signals. Table 4-3 shows the average growth rate for each SGs during Cycle 9. It is evident that the
~ grow 3h rates during Cycle 9 are small for all four SGs, and the combined growth rate for all 4 SGs is actually negative (-0.02 volt /EFPY) because the modest negative growth rate (-0.08 volt /EFPY) observed for SG-1 dominates small positive growth rates (0.0 to 0.02 volt /EFPY) for the remaining three SGs. The average growth for q:\apc\tva98\tva90 day. doc 4-2
l l ,. .
1 indications with a BOC bobbin voltage above 0.75 volt is negative for all 4 SGs which, although unusual, is not meaningful because ofits small absolute magnitude.
1 Table 4-4 shows the cumulative probability distribution of growth rate per EFPY for
- - i each Sequoyah Unit-1 steam generator during Cycle 9. These growth data are also plotted in Figure 4-5. The curve labelled ' cumulative' in Figure 4-5 represents averaged composite growth data from all four SGs. The average growth rate j distribution for Cycle 9 is compared with that for the last cycle (Cycle 8) in Figure 4-6.
_ The growth data are presented on an EFPY basis to account for the difference in the length of the two operating periods. It is evident from Figure 4-6 that Cycle 8 growth l i
distribution is more limiting than Cycle 9 growth distribution.' The NRC guidelines
' reqQire that the more conservative growth distribution for the last two operating periods be applied for projecting the next cycle distributions. Therefore, Cycle 8 growth data will be app, lied to obtain EOC-10 projections.
Table 4-5 provides a comparison of average growth data for the last 3 operating cycles. It appears that tl e3 increase in growth observed for Cycle 8 (mostly in SG-1) is l
offset by the reductioniioted for Cycle 9; however, since the magnitude of the growth observed during Cycles 8 and 9 small and comparable to uncertainties in the bobbin voltages, no firm conclusions can be drawn regarding the growth trend. Table 4-6 lists the top 30 indications froin the standpoint of growth during Cycle 9. Of the 30 l
l indications,13 are identified as being new for Cycle 9.
l According to the Westinghouse ARC analysis methodology presented in Reference,9.2, l the larger of the composite growth rate for all SGs and the SG-specific growth rate should be used in projecting SLB . leak rate and tube burst probability for individual SGs. As noted e'arlier, Cycle 8. growth rates would be used to perform EOC-10 projections as they are higher tlian the Cycle 9 growth rates. Since the Cycle 8 growth rates for SGs 2 to 4 are below the composite growth rate (see Table.4-4), the I composite growth rate is applied to those three SGs to provide a conservative basis for predicting EOC-10 conditions. However, predictions for SG-1 are obtained using its own growth rate since it is higher than the composite rate.
I 4.3 Probe Wear Criteria l
l An alternate probe wear criteria discussed in Reference 9.5 was applied during the
! EOC-9 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. Accordingly, only tubes containing indications for which the worn probe voltage is above 1.5 volts need to be inspected with a new probe.
q:\ ape \tva98\tva90 day. doc 4-3
(@ (b I Only a total of 5 ODSCC indications were detected with probes that failed the wear ,
check subsequent to a successful probe wear test. Four of those indications are in SG- l 3 and one in SG-4, and all of them had a bobbin voltage below the 1.5 volts threshold for retesting. Therefore, no tubes were retested because of bobbin probe wear during j the inspection. Th'e alternate probe wear criteria used in the EOC-9 inspection is l consistent with the NRC guidance provided in Reference 9.5. i L
i 1
4.4 Probability of Prior Cycle Detection (POPCD) l l
l The inspection results at EOC-9 permit an evaldation of the probability of detection ]
at the prior EOC-8 inspection. For ARC ' applications, the important indications are l those that could significantly contribute to EOC leakage or burst probability. These i significant indications can be expected to be detected by bobbin and confirmed by l RPC inspection. Thus, the population ofinterest for ARC POD assessments is the l EOC RPC confirmed' indications that were detected or not detected at the prior l inspection. The probability of prior cycle detection (POPCD) for the EOC-8 inspection
! can then be defined as follows.
EOC-8 ' cycle reported + Indications confirmed )
- l. e indications confirmed by and repaired in EOC-8 i RPC in EOC-9 inspection inspection -
l POPCD = l l (EOC-8) ( Numerator) + New indications RPC l l
confirmed in EOC,-9 l l inspection l l
POPCD is evalusted at the 1997 EOC-8 voltage values (from 1998 reevaluation for j growth rate) since it is an EOC-8 POPCD assessment. The indications at EOC-8 that i i were RPC confirmed and plugged are included as it can be expected that these l l indications would also have been detected and confirmed at EOC-10. 'It is also l 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.
I It should be noted that the above POPCD definition includes all new EOC-9 indications not reported in the EOC-8 inspection. The new indications include EOC-8 indications present at detectable levels but not reported, indications present at EOC-8 below detectable levels and indications that initiated during . Cycle 9. Thus, this
~
! definition, by including newly liiitiateil indications, differs from the traditional POD l
! definition. Since the newly initiated indications are appropriate for ARC l applications, POPCD is an acceptable definition and eliminates the need to adjust the i traditional POD for new indications. .
. q:\apc\tva98\ tva90 day. doc
( 4-4 l
l _
W b.d l
l The above definition for POPCD would be entirely appropriate if all EOC-8 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
~
l mdications. Thus, a-~more appropriate POPCD estimate can be made by assuming ;
that all bobbin indications not RPC inspected would have been RPC confirmed. This I definition is applied only for the 1998 EOC-9 indications not RPC inspected since j inclusion of the EOC-8 repaired indications could increase POPCD by including indications on a tube plugged for non-ODSCC causes which could be RPC NDD indications. In addition, the objective of using RPC confirmation for POPCD is to l
distinguish detection of an indication at EOCo.1 that could contribute to burst at
- EOCo so that the emphasis is on EOCn RPC confirmation. This POPCD can be obtained by replacing the EOC-9 RPC confirmed by RPC confirmed plus not RPC inspected in the above definition of POPCD.
The POPCD evaluation for the 1997 EOC-8 inspection data is summarized in Table 4-7 and illustrated on Figure 4-7. Both data based on RPC conSrmed only indications anti RPC confirmed plus 'not RPC inspected indications art shown in Figure 4-7. Also shown in the figure is a generic POPCD distribution developed by analyses of 18 inspections in 10 plants and presented in Table 7-4 of Reference 9.6. It is seen from l Figure 4-7 that the predicted POPCD distribution for Sequoyah-1 reaches unity at l l about 1.5 volts, but it is below the generic POPCD distribution between 0.6 to 1.5 l volts. The predicted POD values are low because 177 new indications were called j during the inspection, but 87 indications reported during the last (EOC-8) inspection were not called during the present inspection (see Table 4-8). The average voltage of the new indications is about 0.55 volt and that of the prior indications not called in
- the EOC-9 inspection is about 0.5 volt. It appears that conservative criteria were used to call such small indications since a large number ofindications called in EOC-l 8 were not found in EOC-9. To account for a relatively large number of prior cycle indications not found in the current inspection, EOC-8 POPCD was also calculated using an alternate method wherein new indications without RPC confirmation were offset by old indications not found (INF). This adjustment for INF was done on a bin by bin basis (0.1 volt bin width) and the difference between new indications and old indications not found was positive for all voltage bins. The revised POPCD l distribution is shown by a chain line in Figure 4-7. The adjustment for INF increases l- the average POPCD for the voltage range 1 to 1.5 volts from 0.54 to 0.6.
In summary, the Sequoyah Unit-1 EOC-8 POPCD supports a POD value of 1.0 above about 1.5 volts, and it is below the generic POPCD in the voltage range 0.6 to 1.5 volts.
qdapc\tva98\tva90 day. doc 4-5
i E l -
. .f .
! 4.5 Assessment of RPC Confirmation Rates i
This section tracks the 1997 EOC-8 indications left in service at BOC-9 relative to
- RPC inspection results in 1998 at EOC-9. The composite results for all SGs are given
~~
in Table 4-8. For 1997 bobbin indications leftin service, the indications are tracked relative to 1997 RPC confirmed,1997 RPC NDD,1997 bobbin indications not RPC inspected and 1997 bobbin indications with no indication found in 1998. Also I included are new 1998 indications. The table shows, for each category ofindications, the number ofindications RPC inspected and RPC confirmed in 1998 as well as the percentage of RPC confirmed'mdications.
[ Thirty-one out of the 46 RPC confirmed indications left in service at BOC-9 were RPC tested during the EOC-9 inspection, and all but one were confirmed. One out of the 4 RPC NDD indications left in service at BOC-9 was also RPC tested and it was confirmed. At present' there is insufficient data for Sequoyah Unit-1 regarding confirmation rate for 1997 RPC NDD indications to justify consideration of only a fraction of RPC NDD indications in the tube integrity evaluation.
I 4.6 NDE Uncertainties The NDE uncertainties applied for the EOC-9 voltage projections in this report are-those given in the prior Sequoyah Unit-1 ARC reports (References 9.3 and 9.4). The probe wear uncertainty has a standard deviation of 7.0 % about a mean of zero and has a cutoff at 15% based on implementation of the probe wear standard. The analyst variability uncertainty has a standard deviation of 10.3% about a meah of zero with no cutoff. These NDE uncertainty distributions presented in Table 4-9 as well as graphically-illustrated in Figure 4-8. The NDE uncertainty distributions are
- included in the Monte Carlo analyses used to project the EOC-9 voltage distributions.
~
L i
q:\ ape \tva98\tva90 day. doc 4-6
,n (. ,
.s i
l Table 4-1 t
Sequoyah Unit 1 September 98 Outage Summary ofInspection and Repair For Tubes in Service During Cycle 9
- Steam Generator 1 Steam Generator 2 In-Service During Cycle 9 RTS for Cycle 10 -
In-Service Dwig Cycle 9 RTS for Cycle 10 Confinned C4.J
&W Field
&W Voltage Fictd g f
g W'ad Confmned Repaired Indications { Iry'ad Confmned Repaired Indications Only l Only 0 0 !
0 0 0 0 0 0 -- 0 0 0 0 0.1 0 0 13 13 ;
0 0 8 8 13 0 0.2 8 0 22 22 l 22 0 0 0 0 22 22 0.3 22 1 1 0 24 24 21 24 0 0 22 2 1 0 22 0.4 1 2 19 19 l 0 0 8 8 -- 21 1 0.5 3 0 0 9 9 l 0 0 10 10 9 1 1 '
0.6 10 0 12 12 12 2 2 0 8 0 0 0
- 8 0.7 - 4 10 10 7 6 !! 4 1 0.8 7 3 2 0 1 0 5 5 i 0 6 6 5 1 0.9 6 1 1 0 3 3 l
4 4 3 2 2 4 1 1 0 5 1
2 0 5 f 3 2 2 0 3 3 5 2 I
1.1- 0 3 3 5 3 2 2 1.2 5 4 4 0 5 0 0 0 0 0 4 4 0 0 1.3 4 2 2 0 0 0 0 0 0 f l.4 2- 2 2 0 2 2 0
! 0 - 0 0 - 0 0 0 0 0 0 0 1.5 0 0 1 0 0 0 1 0 0 0 0 0 1
( 1.6 0 0 1 1 0 t i 1 0 1.7 1 0 0 0 0 0 0 0 0 0 0 0 1.8 0 0 0 0 0 0 l 0 0 0 0 0 0 0 0 0 1.9 127 127 l 130 15 15 3 16 0 110 108 Total 110 18 4 4 0 '
10 lo 0 15 15 10
>lv 15 10 10 Steam Generator 3 Steam Generator 4 l In-Service During Cycle 9 RTS for Cycle 10 In-Sda During Cycle 9 RTS for Cycle 10 Confmned
' C4..-J
- RPC RPC Indications All 8' RPC RPC Indications All Confmned Repaired Indications g W i-A -Confrmed Repaired Indica 6cns { leaA Ny Only 0 0 0 f 0 1 0 0 0 0.1 1 0 0 1 2
1 0 0 0 2 1 0 2 2 2 0.2 2 1 0 -8 8 l 8 8 0 0 0.3 to 5 5 2 8 5
l 9 5 0 0 0 5 9 3 3 0 9 0.4 4 0 7 7 6 6 7 4 0.5 6 2 2 0 5 2 0 5 8 5 2 0.6 8 3 3 0 8 5
5 3 3 0 5 0 5 5 0.7 5 1 1 4 3 3 0 4 0 8 8 4 0.8 8 5 5 5 5 4 4 0 5 5 1 8 8 0.9 9 5 2 2 2 1 1 0 2 0 4 4 1 4 2 0 3 3 3 3 3 4 4 3 3 1.1 4 1 0 2 2 2 2 2 0 2 2 1.2 2 1 1 0 1 1 l 1 0 1 1 1 1 1 1.3 1 1 l 1 1 0 1 1 0 2 2 1 1.4 2 1 1 0 3 3 l 6 3 3 3 7 6 6 1 6
( l.5 O O 0 1 1 1 0 t 1 I l 1.6 1 1 0 0 0 0 0 0 l 1.7 1 0 7 0 1 1 0 0 0 0 1 0 0 1 0 1 0
f 1.8 1 1 0 0 0 0 0 0 0 0 1 1
.- 1.9 1 77 77 54 27 27 0 54 54
! Total 82 42 42 5 11
!! 10 10 0 11 2 18 18
>lv 20 15 15 i
I 4-7 i
t 4
Table 4-2 (Sheet 1 of 2)
Sequoyah Unit 1 September 1998 TSP ODSCC Indication Distributions for Tubes in Service During Cycle 9
.. i Steam Generator 1 Steam Generator 2 Steam Generator 3 Average Drgest Average Number of Maximum Average Largest Avert %
Number of Maximum Average Largest Average Number of Maximum Indications Voltage Voltage Growth Growth Indications Voltage Voltage Growth Growb Tube Suppon Indications Voltage Voltage Growth Growth l Plate
~
61 1.64 0.62 0.32 0.04 53 1.82 0.85 0.59 0.00 1101 48 ~ 1.64 0.74 0.20 -0.11 1.08 0.66 0.34 0.08 12 1.70 0.87 0.48 0.07 1102 16 1,34 0.64 0.06 -0.15 13 1.01 0.40 0.40 0.02 4 0.87 0.50 0.06 -0.1 I 1103 8 0.39 0.30 0.10 -0.07 10 0.63 0.51 0.21 -0.02 2 0.38 0.28 0.06 0.04 1104 4 0.65 0.38 0.14 -0.04 5 0.50 0.28 0.06 -0.02 0.51 0.51 0.02 0.02 I105 5 0.58 0.44 0.I6 -0.13 12 1
-0.04 -0.15 5 0.43 0.32 -0.03 2 0.45 0.38 0.07 __ _ 0.04 _
1106 8 0 70 0.40 0.04 _
0.36 -0.03 -0.08 i 6 0.42 0.32 0.06 0.03 0 - - - -
1107 3 d.57 0.33 0.33 0.02 0.02 1 0.56 0.56 -0.01 -0.01 C07 2 0.40 0.33 0.00 -0.02 1 0.35 0.06 -0.08 7 0 60 0.39 0.10 -0.02 0 -
C06 4 J 43 0.46 0.35 0.04 0.01 1 0.30 0.30 0.01 0.01 C05 3 0.43 0.36 -0.04 -0.07 3 4 0.37 0.29 0.05 0.(X) 1 0.30 0.30 0.01 0.01 C04 2 0.26 0.25 -0.09 -0.10 0.34 -0.02 -0.10 3 0.46 0.31 0.04 0.01 0 - - -
C03 5 0.56 0 - - - 1 0.11 0.11 -0.13 -0. lid 0 - -
CO2 - - -
0 4 0.54 0.30 0.05 -0.09 C01 2 0.32 0.24 0.05 0.00 - - - -
130 82 Total 110
)
r.m m.,un w .su.
4-8
s i.
Table '4-2 (Sheet 2 of 2) ;
Sequoyah Unit 1 September 1998 TSP ODSCC Indication Distributions for Tubes in Service During Cycle 9 e
steam Generator 4 I Composite of All SGs Number of Maximum Average Largest Average Number of Maximum Average Largest Average Tube Support Indications Voltage Voltage Growth Growth Indications Voltage Voltage Growth Growth Plate ~ z rp.
29 1.56 033 -0.01 191 1.82- 0.74 0.59 -0.02 Gv' I(01 OJ.2 _
i102 8 1.46 0.74 0.27 0.03 49 1.70 0.72 0.48 -0.01 1103 6 1.47 0.49 0.12 -0.03 28 1.47 0.41 0.40 -0.04
~
_ _I y _ 1 0.27 0.27- 0.06 0.06 12 0.65 0.41 0.2 I . 7001 _
1105 2 0.89 0.68 -0.02 -0.07 20 0.89 037 0.16 -0.05 1106 5 0.44 0.29 0.15 0.06 20 0.70 035 0.15 -0.05 1107 1 0.22 0.22 0.00 0.00 10 0.57 0.32 0.06 -0.01_
C07 0 - - - - 4 0.56 039 0.02 0.00
- ~~ ~
'~~
0 0.33 0.25 0.06 0.04 9 04 -0.0 .
CG4 0 - - - -
7 037 0.28 0.05 -0.03 -
C03 0 - - - - 8 0.56 033 0.04 -0.06 CO2 0 - - - - 1 0.11 0.11 -0.13 -0.13 C01 0 - - -
6 0.54 0.28 0.05 -0.06 Total 54 i376 i
I 4
4 a arsusaanwmu am 4-9
4 i
Table 4-3 i Sequoyah Unit 1 - September 1998 Outage l Average Voltage Growth During Cycle 9 Voltage Number of , Average Voltage ;
Indications BOC Entire Cycle Per EFPY ' Entire Cycle Per EFPY '
Range ]
- i Composite of All Steam Generator Data ,
0.62 -0.022 -0.017 -3.5% -2.7% i Entire Voltage Range 376 I
V soc < .75 Volts 259 0.41 0.009 0.007 2.3% I8%
' -6.4 %
2.75 Volts I17 . 1.10 -0.091 -0.071 -8.3 %
Steam Generator 1 Entire Voltage Range i10 0.67 -0.106 -0.082 -15.9% - 12.3 % i
'V noc < .75 Volts 69 0.42 -0.072 -0.056 -17.1% l ~
-13.2%
2.75 Volts 45 !.09 -0.164 -0.127 -15.1% - 11.7% ,
Steam Generator 2 Entire Voltage Range 130 0.48 0.024 0.019 5.1% 3.9% ;
IV noe < .75 Volts 110 l 0.39 0 034 0.026 8.7% 6.7 % _ !
~ ~ ~ ~ -2.75 Volts 20 1.01 -0.026 -0.020 -2.5% - -1.9% ;
l Steam Generator 3 , l Entire Voltage Range 82 0.75 0.004 0.003 0.5% 0.4% _ $ !
V oc s < .75 Volts 46 0.43 0 051 0.039 11.9 % 9.2%
2.75 Volts 36 1.16 -0.057 -0.044 -4.9% -3.8%
t I
Steam Generator 4 l Entire Voltage Range 54 0.68 0.000 0.000 0.0% 0.0%
8
-34 0.41 0.041 0.032 10.0 % 7.7 %
V soc < .75 Volts i 2.75 Volts 20 1.13 -0.070 -0.054 -6.2% -4.8%
I l
- Uased on Cycle 9 duration of 472 EFPD (1.292 El PY) t I
GroettdTable3112R9818.46 Akt 4-10 j t
. . Table 4-4 ~ i
.Sequoyah Unit 1. September 98 ,
Signal Growth Statistics For Cycle 9 on an EFPY Basis Steam Generator 3 Steam Generator 4 Cumulative Steam Generator 1 Steam Generator 2 Delta Cycle 9 Cycle 8 Cycle 9 Cycle 8 Cycle 9 Volts Cycle 8 Cycle 9 Cycle 8 Cycle 9 Cycle 8 CPDF CPDF CPDF CPDF CPDF CPDF CPDF- CPDF CPDF CPDF l-0.0 0.0 0.019 1 0.003 -12 1 0 0.0 0.0 0 0.0 0.0 0 1
-0.8 0.0 2 0:008 0.0 0.0 0.012 0.0 0 0.019
-0.5 0.0 1 0.009 0.0 0 1 0.037 0.0 0 0.019 6 0.024 1 0.036 0.0 1 0.008 0.0 2
-0.4 0.0 3 0.0 0 , 0.008 0.0 5 0.098 0.0 1 0.037 8_ 0.045
-0.3 0.0 l2 0.055 2 0.122 0.0 1 0.056 0.007 10 0.072 0.0 6 0.109 0.012 1 0.015 0.013
-0.2 0.111 0.013 45 0.191 0.069 0.013 5 0.183 0.022 3
-0.1 0.011 30 0.382 0.012 7 0.366 0.133 20 0.481 0.174 124 0.521 0.267 43 0.4 0.152 15 0 0.125 1 46 0.8 0.872 0.805 0.533 20 0.852 0.51 132 .
0.936 0.616 61 0.869 0.506 36 0.1 0398 15 0.963 0.738 36 0.968 12 0.962 0.759 11 0.939 0.8 6 0.2 0.614 7 1.0 0.814
,0.3 0.75 _ 0 1.0 0.884 4 0.992 0.873 2 0.963 , 0.867 2 _ l.0 0.839 _8_ 0.989_
0.988 0.933 0. 1.0 0.93 3 0.997 0 1.0 0.965 1 1.0 0.937 2 0.4 0.886 0.977 I 1.0 0 'l.0 0.975 1 1.0 0.978 0 11.0 0.5 0.966 0 1.0 0.988 0 1.0 1.0 0 1.0 0.99 0 1.0 0 1.0 1.0 0 1.0 0.987 0.6 0.977 1.0 0 1.0 1.0 0 1.0 1.0 () - 1.0 0.7 1.0 0 1.0 1.0 0 1.0 82 54 376 Total 110 130
, N' n:- .
a i
4 !
4-Il c-m croram isms sAs m E
_ _ _ . = _ _ _ _ _ . _ _ _ _ _ . _ _ _ .
- _ _ _ __.__._..___._.-_____________._.____._________.m
____.__m_ _ _ . _ ,
i ,
'E Table 4-5 Sequoyah Unit 1 September 1998 ,
Average Voltage Growth for Cycle 9 Coniposite of All Steam Generator Data .
Average Voltage Growth Average Percentage Growth Bobbin Voltage Number of Average Voltage ,
Entire Cycle Per EFPY Entire Cycle Per EFPY Range Indications BOC ik Cycle 9 (1997 - 1998) - 472 EFPD
-0.017 -3.5% -2.7 %
376 0.62 -0.022 ,
Entire Voltage Range '
0.009 0.007 2.3% l.8% i 259 0.41 V noc < .75 Volts -8.3% -6.4 % 'f 1.10 -0.091 -0.071 2.75 Volts i17 Cycle 8.(1995 - 1997) - 450 EFPD O136 30.5 % 24.7 % j 298 0.55 0.167 Entire Voltage Range 0.113 34.2 % 27.8 % j 232 0.41 0.139 V noc < .75 Volts 20.6 %
0.268 0.218 25.4 %
2.75 Volts 66 1.06 I I Cycle 7 (1993 - 1995) - 450 EFPD ag !
3.2 %
'*- i 0.031 0.025 3.9%
Entire Voltage Range 20' O.79 0.I1 0.088 24.2 % I9.6%
11 0.45 V noc d .75 Volts -4.3% !
-0.063 -0.051 -5.3%
2.75 Volts 9 1.20 i
h I I ;
I I !
4-12 c-err.wem2 wssam f
kh $57 Table 4 6 Sequoyah Unit 1 September 1998
-Summary of Largest Voltage Growth Rates for BOC 9 to EOC-9 Steam Generator Bobbin Voltage RPC New SG Row ~ Col Elevation EOC BOC Growth Confirmed ? Indication ?
3 13 28 OlH- 0.89 0.3 0.59 - N N 3 36 73 02H 1.7 1.22 0.48 N N 3 19 36 OlH 1.82 1.41 0.41 - N N 2 20 31 03H 1.01 0.61 0.4 N N 2 43 54 02H 0.64 - . - -0.3 0.34 N N 4 11 58 OlH 1.23 0.9 0.33 Y , Y 2 16 44 _ _0,1 H _. 0.82 _ . _ 0.5 0.32 N N 3 12 28 , OlH 1.18 0.87 0.31 Y N 3 38 52 02H 0.66- 0.36 03 N Y
._2 17 35 02H 1.08 0.79 0.29 Y Y 2 35 18 02H 0.53 0.25 0.28 N Y -
4 5 71 02H 0.98 0.71 0.27 N Y 2 16 9 OlH 1.64 1.38 0.26 N N 3 6 27 01H 0.58 _ 0.32 0.26 Y 'l 3 8 30 OlH 1.49 1.24 0.25 Y N 3 10 93- OlH 0.7 - 0.46 0.24 N N 4 32 24 O1H 0.18 0.23 Y Y
._ _ 0.41 _ _ _ _
4 33 22 OlH 1.01 0.79 0.22 Y Y 2 29 35 04H 0.63 0.42 0.21 N N 3 21 39 OlH 1.41 1.2 0.21 Y N 3 22 74 02H 0.76 0.55 0.21 Y Y 1 8 66 OlH 0.76 0.56 0.2 N N 2 7 70 O1H 0.62 0.42 0.2 N Y 3 12 4 02H 0.83 0.63 _ 0.2 Y N 1 26 37 OlH 1.28 1.1 0.18 Y Y 3 5 29 OlH 0.95 0.77 0.18 N N 4 25 61. OlH 0.73 _ _0.3 0.18 Y Y 3 13 49 OlH 0.85 0.68 0.17 Y N 1 ~26 19 OlH 0.34 0.18 0.16 N N 46 54 05H 0.35 0.19 0.16 N Y
! 1 l
I l
I GrowdiTabic512/#98 830 AM 4*13
i Table 4-7 '
Sequoyah Unit 11998 EOC-9 Evaluation for brobability of Prior Cycle Detection Composite of All Steam Generator Data 1998 Bobbin, Field Call in New Indications inspection POPCD '
1997 Inspection Bobbin
~ n.
1998 1998 *.y inspection Inspection RPC 1998 RPC 1998 RPC 1997 RPC Confirmed i Voltage inspection Confirmed Inspection Confirmed Inspection Confirmed Plus Not Bin . RPC plus not RPC plus not Confirmed Insp*ected Confirmed Inspected Confirmed Inspected and Plugged Frac. Count Frac. ; Count
> 0 - 0.3 10 64 0 51 0 0.0 0/10 0.443 51 /115 0.3 - 0.6 14 56 4 69 1 0.263 5/19 0.556 70/126 0.6 - 1.0 ~ 24 38 15 48 1 0.400 16/40 0.563 49/87 I
1.0 - 1.5 17 18 16 27 3 i 0.5 19/36, 0.625 30/48 1.5 - 2.0 0 0 1 3 2 1.000 3/3' 1.000 5/5 TOTAL 65 176 36 198 7 ;
lT,, .
> 1V 0 0 ,1 3 2 ,
I i I i i
Poped Table 1 (2) I2/4/98 8.57 AM 4-I4 1
l
i Table 4-8 Sequoyah Utilt 1 Analysis of RPC Data from 1997 and 1998 Inspections Combined Data from All Steam Generators
. . Total Total Total Percent tow 1998 1998 1997 1998 Group of Indications inspection Inspection "P '
Inspection Ir)spection Bobbin Bobbin-8}R C . RPC RPC D inspected
. Indication Indication Cc 1 firmed Confirmed Less than or Equal to 1.0 Volt in 1998 Inspection 1997 Inspection Bobbin Left in Service . . 246 164 18 17 94.4
- . ..1997 lnsp_ection RPC Confirmed 26 26 i 14 _ 13 _ , _ _92.9
~ .1997 !gspect!on RPC [fDD , . ,_
.,,___3_ ._ _3 ,
0 0 -
.1997 Inspection RPC Not inspected _135 135 4 4 100.0
. _ - Ng 1998 !nspection Bobbin
- 82 -
_.._New 1998 Inspection Indication -
156 46 45 97.8 Sum of All 1998 Inspection Indication 246 320 64 62 96.9 Greater thdn 1.0 Volt in 1998 Inspection i 1997 Inspection Bobbin Left in Service 40 35 19 19 100.0 1997 Inspection RPC Confirmed 20 20 17 17 100.0
-I 1997 Inspection RPC NDD 1 1 1 1 1 100.0
. __. _]997.!nspection RPC Not inspected 14 14 1 1 100.0
- No 1998 Inspection Bobbin
- _ _ _ _ _ _ , .
5 - - - -
New 1998 Inspection Indication -
21 20 20 100.0 .Q Sum of All 1998 Inspection Indication 40 56 39 39 l 100.0 7 All Voltages in 1998 Inspection 1997 Inspection Bobbin Left in Service 286 199 37 36 97.3_ _
- 1997 Inspection RPC Confirmed 46 46 31 _,i _ 30 96.8
- 1997 Inspection RPC NDD _ _ ._, _
4 4 1 1 100.0
- 1997 Inspection RPC Not inspected _
149 149 5 5 100.0
- No 1998 Inspection Bobbin
- 87 - - - -
New 1998 Inspection Indication -
177 l 66 65 98.5 Sum of All 1998 Inspection Indication ,286 376 103 101 98.1
- Indicatons spSt is based on 1997 Inspection bobbin voltage 4-15 i Poped Table 212/4/98 9:04 AM .
_ _ _ _ . - - _ - . _ _ _ _ _ - - _ - _=_ _-__ _____ __________-_______ -____ ____ - ___-______
cx;; tn Table 4-9 ;
Probe Wear and Analyst Variability - Tabulated Values i Analyst Variability Probe Wear Variability i Std. Dev = 10.3% Mean = 0.0% Std. Dev = 7.0% Mean = 0.0%
l No Cutoff Cutoff at +/- 15%
Value Cumul. Prob. Value l Cumul. Prob.
-40.0% 0.00005 < -15.0% 0.00000 l
-38.0% 0.00011 -15.0% 0.01606 l
-36.0% 0.00024 -14.0% 0.02275 I
-34.0% 0.00048 -13.0% 0.03165 l
-32.0% 0.00095 -12.0% ~ 0.04324 i
-30.0% 0.00179 -11.0% 0.05804 l
-28.0% 0.00328 -10.0% 0.07656 __
I
-26.0% 0.00580 -9.0% 0.09927
-24.0% 0.00990 -8.0% 0.12655
-22.0% 0.01634 -7.0% 0.15866
-20.0%
0.02608 -6.0% 0.19568
-18.0% 0.04027 -5.0% 0.23753
-16.0% , 0.06016 -4.0% 0.28385
-14.0% 0.08704 -3.0% 0.33412
-12.0% 0.12200 -2.0% 0.38755-
-10.0% 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 0.0% 0.50000 4.0% 0.71615 2.0% 0.57698 5.0% 0.76247 4.0% 0.65112 6.0% 0.80432 6.0%. 0.71989 7.0% 0.84134 _
8.0% 0.78133 8.0% 0.87345 10.0 % 0.83419 9.0% 0.90073 12.0 % 0.87800 10.0 % 0.92344 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 > 15.0% 1.00000 26.0 % 0.99420 l 28.0 % 0.99672 l 30.0 % 0.99821 - -
32.0 % 0.99905 34.0 % 0.99952 36.0 % 0.99976 38.0 % 0.99989 40.0 % 0.99995 1
NDEuncert Table 3-7 !!/24/9812.10 PM 4-16
t f
i I ,
Figure 4-1 ;
Sequoyah Unit 1 September 1998 Outage ,
t Bobbin Voltage Distributions at EOC-9 for Tubes in Service During Cycle 9 25 . :
- }. .
20 - E E i
=
- : EI SG-1 e : :
E 15 - :
E O SG-2 m . -
i u , = =
I
- 6 : :
3 -
E E.
E SG-3 m .
o : :
t O 2 E o : -
- l SSG4 ,
E e
10 : : ;
i : : :
l z -
- s :
- s : : : : i
- s : "
5 : : : .s S-I : : g : : s : : : ,
i
- : g : : s : : : = ,
5- : : s
- s- :: s --: -
- : s : : :
- s s " "
E : s : s 2 s -
E : : E
= s : s s : : : : : - : :
- s : s s
s : s s
- s : : : : : : :
= 1
- s : s s : s : :
- s' : s' : s' : s : :
g's:":
T'H ' E O 's i
'a
'a
'a a
'a
'a
'a
'a
'e
's j
s 'e a a E's 9 N 9 T. 9 N D 9 9
- 9 N 9 Y. 9 D. D 9 9 !
o o o o o o o o o - - - - - - - - - i Bobbin Voltage l
BobnrpcFigl] '
4-17 i
I
I i
I Figure 4-2 Sequoyah Unit 1 September 1998 Outage ,
Bobbin Voltage Distribution for Tubes Plugged After Cycle 9 Service t
8 ,
i l,
i 7- -
ec vv.
I 6- .
OSG-1 l g t o 5-c O SG-2 '
.2 E '
E SG-3 "o
g .
i
.o
! 3-
'4
@SG-4 l 2 3 i
' i I I I i O I
- 9
- 9 i
n o
9 o I o o - -
Ilobbin Voltage ,
i I ,
k nwn 2 4-18 ,
I
i 4
& t Figure 63 -
Sequoyah Unit 1 September 1998 Outage Bobbin Voltage distributions for Tubes Returned to Service for Cycle 10 , ;
25 ,
E E
- K 20 - : : O: '
- =
E . : :: ,
o : : c 5 E E
- SG-1 i
.2 : :: 1 -
m : :
3 15 : : '
eg j j SG-2 u - -
.o _ _
E E E E SG-3 a . .
Z , E E 4 10 - -
[ j E S SG-4
- i 5
i
- : % : : 5 : : :
% b h $ h kk s' e=:-
[
~ "
%'" %~ [ , n. %~ " m ;
i
% =
I : : % :: % : % : % : : . : i
- s s :
~ ?< : :: "
=
2 E
- : % :: % : N : : % = : : : : : :
0 gR
.E k 6 k
R k
S
~
E R
.R I
k B
" I I i
=
=
.B
" f
" E
=
4=9
" 8 I
f H4 .
9 E 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 1.1 1.2 1.3 1.4 1.5 1.6 1.7 1.8 1.9 Bobbin Voltage l
nos. peno 4-19 I
Figure 49 .
}
Sequoyah Unit 1 - September 1998 ODSCC Axial Distributions for Tubes in Service During Cycle 9 70 1
s w 3 .9 60 - - 2 !
50 - -
E
- E SG-1 '
E i
.9 a =
.2 40 - - := O SG-2 o
,5 E Y 5 b 30 - -
E E SG-3
.o =
E =
m . =
Z = S SG-4 '
20 - - j_
.r . ..
E = -.
10 -
=
5 E = = -
= = = _ _
=_
" 'A
b' .
. 'E ",
0 .
C06 C05 C04 C03 CO2 COI 1104' 1105 H06 IIO7 C07 1I01 1I02 1103 Tube Support Plate I
4-20 o
Figure 4-5 .
i Sequoyah Unit 1 Cycle 9 ( May 1997 to September 1998 )
Cumulative Probability Distributions for Voltage Growth on an EFPY Basis .
^ -
I ** 7,..
-- gi. ...
i .
,x ......x-0.9 - ----------------- ------------------ ----------------- ---------- ---' .,. 6-
- i g,
~
i 0.8 -------------------------------- ---- ---- --- -------- --- ----- y x ------ ----- --------- ---------------- ~-
a / l o 7 E 0.7
$ ------------------------------------------------------//l* e
./.
//
w
@ 0.6
-- ------------- ------ fj/-------------------------,-----------
~ .
3 i,. .
2u 0.5 -
- : S0-1 --
TA ---------------- ---- -------- ------------------- ---- / .-*
//
6 '
--o-- SG-2
? 0.4 ------- -- -- ----- ----- -- ---------- -- ---- ,'i- '
5
// .*
t- - - x - - SG-3
- - / !. - -- -
$0.3
-l-3 / g-O /.*
0.2 --- - - --
- - - - Y .'
'i -
- *- - SG-4 p
. . ./
./
./ i
/ g 0.1 -- --l -- -- -
- - - 7, ; x- - t. . * ----- -- - -- - - --- ----- -
--x- Cumulative --
" " ~~~ f '* :
~
0 7 . _ . _ . ._g p h ,
-0.5 -0.4 -0.3 -0.2 -0.1 0 0.1 0.2 0.3 OA 0.5
-0.8 5
Voltage Growth i
Growth Fig 212/4/98 9:58 AM 4-21 i ,
-t _ __ _ _ . . _ . . . . _ . _ _ _ _ _ _ _ _ _ _ _ _ . _ _ _ _ _ _ _ _ .__
Figure 4-6 .
Sequoyah Unit 1 - Septembe- 1998 Hobbin Signal Growth IIistory - Cumulative Probability Distributions on an EFPY Basis Composite of' All Steam Generators i_
-m-1.0 O"", ,_
/
. /
/
--/-------------------------------------------
0.9 -
V -
/ "*
t .t O.8 ------------------------------------------ -----------l---------- -- ------------- ----~~------- ------ 1.,'8 '2..
C / '
o l C 0.7 -----------------------------------------------y-----------------------------------~~-------
c /
c
% i c 0.6 --------------------------------------------- ---l------ --
.O D /
o i p .......
c o.5 ........................................................ ........................... ,
~ /
..n. ;
C / : Cycle 8
-----t---- --- c
.? 0.4 /
"a /
3
= 0,3 .........................................../........................................ ......
i o
U /
- a- Cycle 9 7 ,
f 3...
0.2 --------------------------------------p---------------------------------------- ,
/ -
/
/ ----------------- -------------- ------ ---
I 0.1 -------------------------------- s----- ----
- a
.a g___n. -D, , . , , , , , , .
0.0 . . . e. . . . . . . . .
-0.8 -0.5 -0.4 -0.3 -0.2 -0.1 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 Voltage Growdh I L
I Growth Fig 312/4/98 9:58 AM 4-22 l l .i
t 1
Figure 4-7 Sequoyah Unit 1 ;
1998 EOC-9 Evaluation for POPCD at EOC-8 1.0 .. =
~~ _
0.9 -
/ ~
e'
/
0.8 i AP iS-e '
1
/ \
0.7 !
c '
.o
- / .. .
w006 .'
l /
$ x- - - - : .
/
1
' 4..............a '
X 0.5 / .
-2! i. / : i y
V . .
r
,' s i -- * --RPC Confirmed Indications Only j 0.4 3 S. / : . . .
A f , .. :
4 0.3 / f : ' RPC Confirmed Plus Not inspected 4........A
/
@~
0.2 -
- ,-- EPRI POPCD (NP-7480, Adcendum-2) 0.1 --x - RPC Confirmed and Not inspected (Alternate Method) ,
0.0 c. , l . l l l l l i
0 0.5 1 ,1.5 2 2.5 3 3.5 I Hobbin Amplitude i Poped!FigI (2)tl2/5/9811:39 Phi 4-23 i
! ; t ,. ' , i ' , >I s*
_ nP.a 7 l:
0
_ 4
- i ,
0
_ 3 3
0 l
_ 2
> )
%(
/ 0 e _
$ 1 g a
t _
s l o _
n
_ i o V l
t a u L n g'
b
- i r .
%iS 0
t s i n
8i D n o
4 4 e y i t
a 2-rt n i 4 ui - r g a it 7 %a 0 V F re 1 t c - n
, e n c r
U i t
y P
e E l i
b D a N
i r r a a I 0 _
V e 2- _
t s
y e W ._
l a b .
n or --
AP ^
0 -
0 .'
0: 0
~
0 3 -
_ '0
- 0 _
0 _
- 0 0 m 0 s _
0 n 0 % i s
- - : 0 .
_ 4 w 1 9 8 7 6 5 4 3 2 1 O -
w 0 0 0 0 0 0 0 0 0 . i M=:is m.
2e j I *=5.aEmU i
- c. -
_ o n
i i
c a -
o.
r
ty
n.a .
4 n'c 0 -
- 4 4
l ' i ,
0 i
- 3 1 I 0
. 2
)
%(
0 e
- J- 1 g _
s / /
t l
a o
n V _
i o l t a _
u n g
b i i
r ,
, %i 0 S t
s i n -
8 i n
4 D o 4 e y i t
a r t ui n i
r ga i t ? / %a 0 V F re 1 t c - e n n c r _
U i t
y P
e E l i
b D a -
N %
i i
r r
, a a b _
V e 2- .
t s
y e W
l -
a b -
n or ^
AP 0 _
0 0 0 %
r 0 0 3 -
'0 .
0 -
0 0 _
0 M .
P 0 8 c 6 e% 2 1
8
- - . - - o0 9
/
4 4 2
1 9 8 7 6 5 4 3 2 1 0 - /
I t
0 0 0 0 0 0 0 , O 0 i t t
r a
b5.c.m1uI>c5.aEaU h
C 8 -
. 4 g
M i
t r
a h
C 8 -
4 g
M
,1
p p 9 y .r 5.0 DATA BASE APPLIED FOR ARC CORRELATIONS Correlations have been developed for the evaluation of ODSCC indications at TSP locations in steam generators that relate bobbin voltage amplitudes, free span burst pressure, probability of leakage and associated leak rates. The Westingh.ouse methodology used in the calculation of these parameters, documented in Reference 9.6 and 9.7, is consistent with NRC criteria and guidelines of References 9.1.
I Leak and burst correlations based on the latest ARC base for 7/8" tubes were used to perform leak rate and burst probability proj.ections for the ongoing cycle. The latest ._ i ARC database includes pulled tube examination data from the 1996 Sequoyah Unit-2, i
1997 Plant A-1 and 1996 Plant A-2. The latest ARC correlations are documented in Addendum-2 to the EPRI database report (Reference 9.6). A leak rate correlation can now be applied to 7/8" tubes based on the p-value for the slope of the leak rate correlation on a one-sided basis meeting the Generic Letter 95-05 requirement. The .
following leak rate correlation is developed in Reference 9.6 for 7/8" tubes.
i l
,, LeakRate (1/hr) = 10{- 32s + Oss72mogm (mzu]
-. l l
The above leak rate correlation was used to perform EOC-10 SLB leak rate projections for the limiting SG.
The leak rate data in the database represent a room measurement of leakage at prototypic SLB conditions (i.e.,. leakage at SLB conditions was condensed and i
measured at room temperature). Therefore, SLB leak rate calculated using the ARC correlations provides a volumetric rate at room temperature.
l For the SLB leak rate correlation, the NRC recommends that Model Boiler specimen 542-4 and Plant J-1 pulled tube R8C74, TSP 1 be included in the database. This database is referred to as the NRC database and the correlations for probability of leakage and leak rate as a function of bobbin voltage presented in References 9.6 include those datapoints.
Additional leakage and burst pressure data are available from 2 TSP tube segments
- pulled from a SG in Sequoyah Unit-1 (R4C15 TSPs 1 and 2 in SG- 2) during this _
outage. A review of the effect of adding the new Sequoyah Unit-1 data to the reference database in Reference 9.6 indicates that the burst pressure, leak rate and the
~
y probability of leak correlations to the common logarithm of the bobbin amplitude q:\ ape \tva98\apr90 day. doc l 5-1 f
t ci Q,: ,.
would not be meaningfully changed. Therefore, SLB leak rates and burst probability l
analyses were carried out using the reference database presented in Reference 9.6. 1 The upper voltage repair limit applied at the EOC-9 inspection was developed from !
the-ARC database of Reference 9.6, which is the latest database available prior to the - I inspection. The structural limit is 8.4 volts. The allowance for voltage growth is L
30%/EFPY, which bounds the Sequoyah Unit-1 data and is the minimum growth allowance required by Generic Letter 95-05 (Reference 9.1). For the expected 1.31 EFPY (480 EFPD) for Cycle 10, the growth allowance becomes 39.4%. The allowance for NDE uncertainty is 20% per Generic Letter 95-05. The upper voltage repair limit -
is then 8.4 volts /1.594 = 5.27 volts.
l I
I o-
! I
~ _
l i
1*
q:\ ape \tva98\apr90 day. doc 5-2 l
0 O uf
- , 5.g i 6.0 SLB ANALYSIS METHODS Monte Carlo analyses are used to predict the EOC-10 voltage distributions and to calculate the SLB leak rates and tube burst probabilities for both the actual EOC-9 voltage distribution and the predicted EOC-10 voltage distribution. Tnese methods _
are described in the generic methods report of WCAP-14277, Revision 1 (Reference 9.2) and the prior reports for Sequoyah Unit-1 (References 9.3 and 9.4), and are in accord with NRC Generic Letter 95-05 (Reference 9.1). Leak rates calculated with the.WCAP-14277 methodology provide a volumetric leak rate at room temperature, and they are compared with allowable volumetric leak rate at room temperature.
i At the time of performing projections for the EOC-9 conditions in 1997 a leak rate versus bobbin voltage correlation could not be applied for 7/8" tubes since the leak ,
rate database for 7/8" tubes did not satisfy the requirement applied then for a SLB l leak rate versus bobbin voltage correlation (p-value for the correlation slope parameter calculated on a two-sided basis less than 5%). Therefore, EOC-9 leak rate projections were carried out using a distribution ofleak rate data that is independent
- of voltage. In order to ensure consistency in the comparison ofleak rates estimated I with projected EOC-9 voltages with those based on the actual measured voltages, leak, rates based on the actual measured EOC-9 voltages were also calculated without
-applying a leak rate correlation
- As mentioned in the previous section, a leak rate correlation can now be applied for 7/8" tubes based on the p-value for the slope of the leak rate correlation calculated on a one-sided basis meeting the Generic Letter 95-05 requirement. Therefore, leak' rate analysis for the EOC-10 condition was carried out using the leak rate vs. bobbin 4 correlation shown.in the previous ~section.
U I
C q:\ ape \tva98\tva90 day. doc
( 6-1 l
b @
l 7.0 BOBBINVOLTAGE DISTRIBUTIONS This section describes prediction of the EOC voltage distribution used for evaluating tube leak and burst probabilities at the end of the operating period. The calculation consists of establishing the-initial conditions (i.e.,-the bobbin indication population distribution) based on eddy current inspection data and projecting the indication growth over the operating period. Since indication growth is considered proportional to operating time, the limiting tube conditions occur at the end of any given time period or cycle. ,
1 I
~
The bobbin voltage distribution established for the BOC conditions is adjusted for nieasurement uncertainty using a quantity termed probability of detection, as described in the following paragraphs. Other input used for predicting the EOC ,
voltage distribution and the results are presented below. l l
7.1 Probability of Detection i
The number of bobbin indications used to predict tube leak rate and burst probability l is obtained by adjusting the number of reported indications to account for '
-measurement uncertainty and confidence level in voltage correlations. This is accomplished by using a POD factor. Adjustments are also made for indications )
either removed from or returned to service. The calculation of projected bobbin !
voltage frequency distribution is based on a net total number ofindications returned to service, defined as:
N-Nr.ars = POD """" *"
where:
NTotnTs = Number of bobbin indications being returned to service for the next cycle.
Ni = Number of bobbin indications (in tubes in service during the previous cycle) reported in the current inspection.
POD = Probability of Detection.
Nrepaira = Number of Ni which are repaired (plugged) after the last cycle'.
Naepiugga = Number of previously-plugged indications which are deplugged after the last cycle and are returned to service.
There were no deplugged tubes returned to service in the recent inspection.
l
! The NRC generic letter (Reference 9.1) requires the application of_a constant POD =
0.6 to define the BOC distribution for the EOC voltage projections, unless an alternate POD is approved by the NRC. A voltage-dependent POD known as POPCD f
q:\apc\tva98\tva90 day. doc l
l 7-1 1
l
hh @
has been established using data from 18 post-1992 inspections at 10 different plants.
It takes inta account newly initiated indications that are important for voltage-based l repair criteria application. The development of POPCD and supporting data are l presented in Reference 9.6. Table 7-1 shows POPCD data as a function of bobbin
- voltage, and the same data is illustrated graphically in Figure 7-1. It is evislent from _
Figure 7-1 that below about 0.4 volt the NRC recommended POD of 0.6 is non-I conservative while it is too conservative above about 0.5 volt. It is ofinterest to apply l
POPCD for sensitivicy analysis and compare the results for the case with a POD value of 0.6.
7.2 Cy'cle Operating Time The following operating period values are used in the voltage projection calculations:
Cycle 8 = 450 EFPD Cycle 9 = 472 EFPD Cycle 10 = 480 EFPD (estimated)
~
_7.3 Calculation of Voitage Distributions Bobbin voltage projections start with a cycle initial voltage distribution which is projected to the corresponding cycle final voltage distribution, based on the growth rate adjusted for the anticipated cycle operating time period. The overall growth rates for each of the Sequoyah Unit-1 steam generators during the last two operating periods, as represented by their CPDFs, are shown on Table 4-4. A Generic Letter 95-05 requirement is that limiting growth rate for the past two cycles of operation should be used in the projections. The 1995 - 1997 operation (Cycle 8) growth rates slightly exceed those of th'e 1997 - 1998 (Cycle 9) operation and are used to predict the EOC-10 bobbin voltage distributions. Further conservatism for the EOC-10 bobbin voltage prediction is provided by the use of the larger of the composite growth rate fo.r all SGs and the SG-specific growth rate in projecting EOC voltages for each SG. The methodology used in the calculations of EOC bobbin voltage distributions is described in Reference 9.2.
For each SG, the initial bobbin voltage distribution ofindications being returned to serdce for the next cycle (BOC-10) is derived from the actual EOC-9 inspection results adjusted for tubes that are taken out of service by plugging. The Cycle 10 bobbin voltage population data is summarized on Table 7-2. It shows EOC-9 bobbin
- voltage indications, the subsequent plugged indications (which were in service for l
Cycle 9 and then taken out of service, albeit not all for reasons of ODSCC at TSP),
and the BOC-10 indications corcesponding to a constant POD value of 0.6 as well as -
the voltage dependent generic POPCD. The POPCD distribution used is shown in q:\ ape \tva98\tva90 day. doc 7-2
- o.
- b 'S Figure 7-1.
7.4 Predicted EOC-10 Voltage Distributions The licensing-basis calculation for the predicted EOC-10 bobbin voltage distributions is performed for all SGs with a constant POD value of 0.6 in accordance with a NRC requirement. In addition, calculations were also performed using a voltage dependent generic POPCD distribution developed based on bobbin and RPC data from 18 EC inspections at 10 different plants. Development of a generic POPCD distribution is described in Reference 9.6. POPCD distribution for the Sequoyah -
Unit-2 EOC-7 inspection data is included in the generic distribution.
The Sequoyah Unit-1 steam generators BOC-10 voltage distributions used to predict the EOC-10 voltages are shown in Table 7-2. As mentioned earlier, the EOC-8 composite growth rate data shown in Table 4-4 were applied to SGs 2 through 4 (since their own growth rates are smaller than the composite growth rate) and its own growth rate distribution was used for SG-1 (since it is higher than the composite growth rate). This approach is recommended in Reference 9.2. Growth data were _
represented by a histogram.
Table 7-3 provides the EOC-10 voltage distributions predicted using the BOC-10 voltage distribution shown in Table 7-1. As anticipated, the largest number of indications is predicted for SG-2, about 214 indications for a constant POD of 0.6.
The assumed BOC-10 and predicted EOC-10 bobbin voltage frequency distributions for all four SGs are also graphically illustrated on Figures 7-2 to 7-5. The lar' gest bobbin voltage predicted for EOC-10 is in SG-3, and its magnitude is 2.7 volts for a constant POD of 04 7.5 Comparison of Predicted and Actual EOC-9 Voltage Distribution ~s The actual EOC-9 bobbin voltage distributions and the corresponding predictions presented in the last 90-day report (for EOC-9 inspection, Reference 9.3), are compared in Table 7-4 and on Figures 7-6 and 7-7. SG-1 was predicted to be limiting for EOC-9 but the actual measurements show that SG-2 had the highest number of indications and the largest indication was found in SG-3. The total number of indications for all SGs is overpredicted by 9% to 50% in the licensing-basis analysis with a POD of 0.6, and the voltage population over 1 volt are overpredicted even greater percentage. Also, while the licensing-basis analysis predicted between 2 to 8 indications above 2 volts in each of the four SGs, not a single indication above 2 volts
~
was detected in the EOC-9 inspection. The overprediction ofindications in virtually every voltage size range demonstrates conservatism in the projection methodology.
q:\apcitva98\tva90 day. doc 7-3 L . . .
i.
Yd' ([,I:7
. EOC-9 voltage distributions' based on the voltage-dependent POPCD also -yields ;
L conservative results,: and the extent of overprediction in all voltage ranges is !
i l comparable to the predictions with POD =0.6. Thus it is concluded that the voltage- ]
- j. dependent POPCD yields conservative results.
l
! 1 i-l: .
l' 1
i
_- 4 r
1
, i t i_
l 1
1 l !
r . .
j.
t<
1 q:\ ape \tva98\tva90 day. doc s
7-4.
L
._ s,. . , .- . - , , . . - :-,
l& ,
Tcbis 7-1 Comparison of EPRI POPCD with EPRI POD Study l
f -
f
- EPRI POPCD - )
Voltage EPRI, f
POD NP-7480-L j Bin Updated f Study Addendum-2 0.1 0.30 0.26 0.26 l 0.2 0 38 0.36. 0.36 0.3 0.49 0.46 0.46 0.4 0.57 0.54 0.54 0.5 0.62 0.63 0.63 0.6 0.66 0.69 0.68
. 0.7 0.71 0.75 0.74 0.8 0.76 0.79 0.78 0.9 0.80 0.82 0.81 1
O.83 0.84 0.84 1.2 0.90 0.87 0.87
, 1.4 0.93 0.89 0.90 1.6 0.96 0.91 0.91 1.8 0.98 0.92 0.92 2~ 0.984 0.93 0.93 3 1.00 0.98 0.98 3.5 1.00 1.0 1.0
- # Dual analyst detection probability study i
i Popedt&l Table 9-411/24/981:42 PM 7-5
&M (i.k.:
l Table 7-2 (Sheet 1 of 2)
Sequoyah Unit 1 September 1998 l EOC-9 Bobbin and Assumed BOC-10 Bobbin Distributions in SLB Leak Rate and Tdbe Burst Analyser l
l Steam Generator 2 Steam Generator 1 Voltage EOC-9 BOC 10 EOC-9 BOC.10 Bin Field Bobbin Indications -POD Field Bobbin Indications POD E Indications Repaired 0.6 ]n.dications Repaired 0.6 0.1 0 0 0.00 0.00 __ _
0 0 0.00 0.00 0.2 8' 0 13.33 23.53 13 0 21.67 38.24 0.3 22 0 36.67 50.00 22 0 36.67 50.00 0.4 22 0 36.67 41.51 24 0 40.00 45.28 0.5 8 ' 0 13.33 12.90 21 2 33.00 31.87 0.6 10 0 16.67 14.93 ._,
9 0 15.00 13.43 0.7 8 0 13.33 10.96 12 0 20.00 16.44 0.8 7- 0 11.67 5.09 11 1 17.33 13.29 1 0.9 6 -0 10.00 7.41 5 0 8.33 6.17 1 4 0 6.67 _ 4.82 ,_
3 0 5.00 3.61
. l . I' 3 .- 0 5.00 3.51 5 0 8.33 5.85 1.2 5 0 8.33 5.68 _
3 0 5.00 3.41 1.3 4 0 6.67 4.47 . _ _
0 0 0.00 0.00 1.4 2 0 3.33 2.20 0 0 0 0 1.5 0 0 0 0 0 0 0 0 1.6 0 0 0 0 _
1 0 1.67 1.09 1
\
1.7 1 0 1.67 1.08 1 0 1.67 1.08 1.8 0 0 0 0 ,,,_0 0 0 0 J
1.9 0 0 'O O O O O -0 l Total 110 0 183.33 192.08 130 3 213.67 220.76
> IV 15 0 25.00 16.94 10 0 16.67- 11.43
>2V 0 0 0 0 0 0 0 0 i.
l % t.u i m iamu m 7-6 w +u
i U h@
l
! Table 7-2 (Sheet 2 of 2) l Sequoyah Unit 1 September 1998 l EOC-9 Bobbin and Assumed BOC-10 Bobbin Distributions in SLB Leak Rate and Tube Burst Analyses _ _
Steam Generator 3 Steam Generator 4 Voltage EOC-9 BOC 10 EOC-9 BOC 10 Bin Field Bobbio Indications . POD Field Bobbin Indications POD lodications Repaired 0.6 Indications Repaired 0.6 0.1 1 0 1.67 4.17 0 0 0.00 0.00 _ l 0.2 2 0 3.33 5.88 .. _ 2 0 3.33 5.88 0.3 10 2 14.67 20.73 ,
8 0 13.33 18.18 0.4 9 0 15.00 16.98 5 0 8.33 9.43 0.5 6 80 _ 10.00 9.68 7 0 11.67 11.29 0.6 8 0 13.33 11.94 5 0 8.33 7.46 !
0.7 5 0 8.33 6.85 _
5 0 8.33 6.85 i
- 0.8 8 0 13.33 10.39 4 0 6.67 5.19
- 0.9 9 1 14.00 10.11 5 0 8.33 6.17 1 4 0 6.67 4.82 2 0 3.33 2.41 1d 4 1 5.67 3.68 3 0 5.00 3.51 l 1.2 2 0 3.33 2.27 _
2 0 3.33 2.27 ;
1.3 1 0 1.67 1.12 1 0 1.67 1.12 1.4 2 0 3.33 2.20 1 0 1.67 1.10 1.5 7 1 10.67- 6.65 3 0 5.00 3.28 1.6 1 0 1.67 1.09 1 0 1.67 1.09 I 1.7 1 - 0 1.67 1.08 _.,
0 0 0 0 1.8 1 0 1.67 1.08 ,__,0 _0 0 0 1.9 1 0 1.67 1.07 0 0 0 0 ;
Total 82 5 131.61 121.77 54 0 90.00 85.24 l
>1V 20 2 31.33 20.23 11 0 18.33 12.36
>2V 0 0 0 0 0 0 0 0 i
i l
m re.,mwmnm 7-7
L i -
Table 7-3 i Sequoyah Unit 1 September 1998 Voltage Distribution Proj5ction for EOC - 10 Combined Data for Hot and Cold Leg Indications Steam Generator 2 l Steam Generator 3 l Steam Generator 4 Steam Generator 1 l Voltage Projected Number of Indications at EOC-10
'" POPCD POPCD POPCD 06 POPCD 6 6 0.18 0.32 10.30 0.74 0.03 0.05 0.1 0.08 0.14 6.55 1.14 2.22 0.72 1.18 0.2 1.87 3.12 3.85
~
10.22 '11.25 17.24 i
339 5.45 2.82 4.12 ;i(.
0.3 6.84 _
26.27 6.68 9.26 5.32 7.12 0.4 13.06 17.81 19.22 30.72 8.85 11.02 6.88 8.41 0.5 16.83 21.43 24.58 - ,
29.63 9.98 11.37 8.01 8.92 0.6 17.84 21.54 25.72 2 8, _ _ _ _ _ _ _
____,,4_
7.81 7.12 0.9 ...- .. 16.68 _ 17.21 _.- . - _~ -.19.22-. --- - 18.51 . . - 11.22 .. - 1. 0.03.. - -
. . _ . . _ . . . t 8.14 8.17 6.28 5.55 4.38 1.2 10.30 8.68 9.87 5.90 6.84 5.06 4.69 . 3.56 1.3 8.92 ' ' ' 7.11 7.48 4.20 3.95 2.9 1.4 ~ I.3 '5.48 '5A1~~ ~ 4' 1'O 5.89 2.76 5.25 3.61 3.34 2.38 1.5 5.87 4.25 3.75 1.88 4.68 3.13 2.83 1.96 1.6 4.61 3.27 2.61 2.45 1.85 1.31 4.06 2.67 ,l 2.31_ _ _ _ _ _ _1.58_
1.7 3.52 , _ _
0.91 3.41 2.21 1.79 1.21 1.8 2.61 1.80 1.32 0.92 0.62 2.75 1.78 1.32 0.89 <D l
1.9 1.84 1.26 1- .. . _ .
A . . - . .0 -- .
0.70 1.16 0.75 0.00 0.70 2.2 0.19 0.00 0.00 0.70 0.30 0.81 0.37 0.70 i 0.00 2.3 0.70 0.70 0.00 0.34 0.00 0.00 0.30 2.4 0.00 0.30 0.30 0.00 0.00 0.00 0.00 0.70 0.30 1 0.00 -
? 2.5 0.30 0.00 0.70 0.30 ~ 0.00 0.00 2.6 0.00 0.00 0.00 ,
0.00 0.00 0.30 0.00 0.00 0.00 .
2.7 0.00 0.00 229.8 131.7 121.8 90.0 85.4 TOTAL 183.3 192.1 213.7 37.8 57.8 41.2 34.7 25.8
>1V 60.2 47.1 47.7 1.0 4.9 3.2 1.6 1.1
>2V 2.0 1.3 1.1 1
A >
Prodcorry Table _2 (2) 12/4/98 0.13 M 7-8
'n
- O .
Tr<bla 7-4 Sequoyah Unit 1 September 1998 Comparison of PredlCted and Actual EOC-9 Voltage Distributions Steam Generator 3 Steam Generator 4 Steam Generator 1 Steam Generator 2 Number of Indications - - _ _
EOC-9 EOC-9 Prediction EOC-9 EOC 9 Prediction EOC 9 EOC-9 Prediction EOC-9 Prediction EOC-9 vtitage * "'
POD s 0.6 POPCD POD = 0.6 POPCD
" POD = 0.6 POPCD POD = 0.6 POPCD 0.00 0.00 0.00 0 0.28 0.71 0 0.00 1 O.1 0.10 0.26 0 0.25 2 0.43 0.76 2 -__
0.31 0.63 8 1.72 3.26.. .1. 3- _ . 0.14 0.2 ..
10 1.07 1.65 8 8.17 22 0.89 1.27 _
1.83 3.10 22 4.54 0.3 9 2.90 4.47 5 15.81 24 2.45 3.35 4.00 5.94 22 9.57 0.4 6 4.41 5.87 7 5.91 7.78 9.98 8 11.83 16.70 21 _4 .66 0.5 8 5.22 5.95 5 14.53 9 6.72 7.46 10.31 12.22 10 11.60 0.6 5 6.34 6.59 5 14.16 12 8.56 8.57 11.78 12.95 8 12.34 0.7 8 7.01 6.93 4 13.06 14.03 11 10.50 9.79 0.0 12.86 13.33 7 9 6.78 6.30 5 12.42 5 11.53 10.28 13.70 13.55 6 12.68 0.9- 9.62 4 5.96 5.25 2 4 11.68 10.57 3 11.31 1.0 13.19 12.32 ' ' ~ ~
4.31 3 10.58 8.60 4 5.08 3
10.64 9.14 S 1.1 T2.11 10.77 3.44 2 9.86 7.73 2 4.21 5 9.17 7.52 3 1.2 10.43 8.81 6.80 1 3.53 2.73 1 4 7.48 5.89 0 8.98 1.3 ' p.92 7.31 2.21 7.96 5.83 2 3.00 1 2 5.98 4.52 0 1.4 7.35 - 5.75 1.81 3 6.86 4.86 7 2.55 0 4.77 3.50 0 1.5 5.99 4.51 1.45 1 5.76 3.99 1 2.10 3.49 0 3.76 2.68 1 1.6 4.77 1.73 1.17 0 ,
2.02 1 4.72 3.20 1 3.80 2.69 1 2.89 1.7 1.48 0.98 0 1.51 0 3.80 2.52 1 3.13 2.16 0 2.20 1.8 1.27 0.83 0 1.11 0 3.03 1.95 1 1.76 0 1.63 1.9 2.62 0 1.07 0.70 0 0.79 0 2.39 1.49 2.22 1.46 0 1.18 2.0 1.10 0 0.88 0.57 0 0.82 0.54 0 1.86 2.1 1.87 1.21 0 0 0.80 0 0.70 0.45 0 0.49 0.00 0 1.43 _
1.57 1.00 0 2.2 _
1.08 0.57 0 - 0.54 0.10 2.3 1.30 0.80 0 0.00 0.70 0_ 0.09 0 0.41 0.00 0 0.00 0 0.81 1.04 0.63 0 0.70 2.4 0 0.31 0.70 0 0.30 0 0.58 0.70 0.81 0.48 0 0.00 2.5 0 0.04 0.00 0 0.00 0 0.20 0.00 0.61 0.05 0 0.30 2.6 0 0.00 0.30 0 0.00 0 0.00 0.30 0.26 0.70 0 0.00 2.7 0.00 0 0.70 0.00 0 0.00 0.00 0 0.70 2.8 0.00 0.00 0 0 0.00 0 0.00 0.00 0.00 0.00 0 0.00 2.9 0.70 0.30 0 0 0.00 0 0.00 0.00 0.00 0.00 0 0.30 3.0 0.00 0.00 0 0 0.00 0 0.30 0.00 0.00 0.00 0 0.00 3.1 0.30 0.00 0 82 70.0 65.5 54 130 127.7 107.0 138.2 110 141.3 150.6 TOTAL 145.7 29.9 21.8 11 10 70.9 50.5 20 53.9 15 52.0 40.2
>1V 69.8 0 3.9 2.1 0 1.5 0 7.0 3.6 5.2 0 2.3 i
>2V 8.5 7-9 3 m--
T c.p.
a 1;
5 3
=
i .
l 3
' t s
n I l
a 5 P l 2
8] '
n-1 im s u n
inod -
c e t
ed pd sA 6 l 2 e n ,
0 d I
0 ,
= u 58 it 1 4 D l p
n 7 - O m 0 1 oP P 1 A -
Tdeet N ,
d t
e i n 7 r s r a b ua op d n
b o
i gB e a l 5 B F nR M 1 oI C iR t
R uP N br E i
t n i
si D de Dt n Ce l 1 P s Oer PP i
c[
r e
n e f G -
- l. 5 0
1
/ "
r s
u
- - - - O m m
0 9 8 7 6 5 4 3 2 1 0 i m
1 0 0 0 0 0 0 0 0 0 0 u
, r E}tQ w =.5 f d w
! ;i;! I
,,ti '
>: 34 I;3 Figurs 7 2 Sequoyah Unit i SG 1 Predicted Bobbin Voltage Distribution for Cycle 10 Combined Data for Hot and Cold Leg Indications POD = 0.6 40 25 30 O BOC-10 E
@ 25 '
$ E Pred EOC 10 20 15 I
,o - - -
E 5 -
~
0 I; II !. I s . .. _ _
, : : : : : : :::::::::::::: ::: Bobbin Voltage EPRI POPCD 50 45 40 O BOC-10 35 _
5 30 '
E Prod EOC-10 25
{ 20 _ _ _a -
i
,5 _ _
lili
,0 _ _ _ -
.IIII
-111.2.
11.14i...._._
5 _
1 11 Bobbin Voltage j i
7-11
% w v ir m
- 2. ,e
-:4 ,
Figure 7-3 Sequoyah Unit 1 SG-2 Predicted Bobbin Voltage Distribution for Cycle 10 Combined Data for Hot and Cold Leg Indications
' ~
~ ~~ ~
POD = 0.6 - ~ -- ---
40 35 -
O BOC-10 E
y 25 -
E 20 - - -
EE E Preo EOC 10 s
n I E is _ __ _ _
Z 10 - - - - ---
~ ~ ~ ~ ~ ~
E E '
}- 5 $ $ $ $ $ .$
llFin ....
$ $ U U U U 3 0 I C U U $ 74 $ $
Bobbin Voltage EPRI POPCD ,
50 45
~
40 -
O BOC 10 35 - -
j 30 _ _ _1 E Pred EOC-10 25 - - ~
^ 20 - - -
15 - - -
_11I.
,0 _ _ _
,_ l-l-j-fjpl , , , _ _
$ $ $ $ $ $ $ $ $ 3 2 U U U U U U U U $ $ $
Bobbitt Voltage r
%rs: ws a e.2 m 7-12 iz w es: m
(b :
- 8. 7, Figura 7-4 Sequoyah Unit 1 SG-3 Predicted Bobbin Voltage Distribution for Cycle 10 Combined Data for Hot and Cold Leg Indicatiora POD = 0.6 16 14 -
12 0 B0C 10 So - --- - - -
i
.2
_ ] - 5 Pred EOC-10
-s - - - - - -
o j -
56 - - - - - - -
2
,4 - - - - - - - -
2 - - - - -- - --- --
g 2___ _
, . II I t .; t .
- n _w _ . ~ . a o - n _ , _ . ~ . a o - n _ . ~
o 6 o o a _o a o o - e e e a e - .- - .- n a n n n n a Bobbin vonage EPRI POPCD
- s
- 0 BOC-10 8
g5 j s Pred EOC 10
~
o -
m - - -
b.
5 - - _
l
~
l,,, , m , ! lain....
-6 O 6 6
. _ O
.O ~O .O .6 . _& _ _& . _& . ~* - . . . _ _N _n _ .
e em e= w w= N N N N Bobbin VoRage 7-13 12Mm8 m m
% r(4 52mos s u m
m
'?; r :
i Figure 7-5 Sequoyah Unit 1 SG-4 Predicted Bobbin Voltage Distribution for Cycle 10 Combined Data for Hot and Cold Leg Indications POD = 0.6 _
j N4 m
12
-' O BOC.10 to 2 . .
~ ~ ~
m Prod EOC.10 I
36 _ _ _ _
1 I -
4 _ _ _ _
I l l
~
I I 1
0 t .
Jii. i sobbin von EPRI POPCD ,
18 16 --
O BOC-10 14 E
I 4, 52 - .__.,0
- 10 0 -
i g,
6 _
- - - - -. -~
~
r _ _ _ _ _ _ _ _ _
_- lit 0
I rl lA l
- : :::::::;=;
i i . ..
sobbin von.o.
7 14 12 m s7au s
m.ampo r s tsams t47 m
, ;' - : ;; 7 -l ; !j ! I' e
ph 1 8
_ 3 _ 2 9 _ 5
- 2 2 m_72 . 4 2
, c g 6
- 2 _
D 3 2
8 C P
0
=
o P
5 2
o I D _
2 2
o R .
P P 4 E m .
d e
d e
2 0
MI
- 1 2
I l
a t
c t
c m 3 oI RP o
u id id 2 P E t e e .
A /
r P
r s
d e
de t
n 0 2
D 's a E 2 2
l a
t i
t u ee he a E
s 1 A c r P
r n 9 1
2 D P. B 3
E 2 b8 1 9
b 9 1 br 7 1
e 8 lc 9 y
(
6 1
br 6 1
. 9C 1 I r 7 ef o r 1 5 b s r 1 mn
,e o 8 e 1 g e
g
- a 4 a 6t pt i i t 2 1 lt o
7 -e u l o V 5 1 3 1 vn r
e Sib o n 1 r
u1 t
is r
t r
o - M t
a r E 3Mto 1 7
igiD t a 4 o t
e - B n r e
1 B n F
Ue g n -
I e 2 e G 1 hat a 3 I 4 yo l G 1 m -
- a oV u
m a 2 t e
1 1
qn t e -
1 S _ -
eib S - -
S b 0 o 1 1 I 1
B _ - ;
- - {
. 9 1 0 9 8 0 0
~ -
6 .
7 0 0 7
0 6 0
6 _ -
g-; 0 = ! 5
- 0 5 - _ -
0 - - i 4
- 0 4
0 -
- _ s 3
0 3 '
0
- m 2 -
E 2 0
e s
a 0
_ 1 w i
_ 1 _ 0 e. .
0 a s o 5 , n, s , 5 0 5 0 0 2 2 i i .
2 2 1 Eg ]e . G1 j1*
p e.
l , ,l' i 1 l h l ) jif ; :. i! s [ l c, '-
1 0 . 1 3
8 8 . 2 2
. F 7 2 1
1 3 8
8 2 2
5 2
, 5 2
D D C 4 4 P .
8 C 2 6 O 2 P P 0 O 0 A P H IR D
O I
R 3
2 3
C O P E
m 3 2
P P P E de de 2 t t 2 de d t t e 2 l a ic ic 2 c c u k d l
a t e e u id id c r P
r t
c e e r
1 A P 1 r 2 m E 1 A P P O O s R 0 0 2 2 h
9 9
- 1 1 9
- e -
- 8 8 lc 8 1
9 y 1 9C -
- 1 j r 7 h
- r ef o 1 7
1 b s e
-
- eo mn 8 e 1 g 8 at g
l 7t i a 1 o
pt 3 t 4 v 7 e bu h-g lo r 6
1 r 5 v o en -
e Si r o
- 1 n t 5 b 7
- r t t a
i b a 1 b
o u1t s r giD in F Ue ha i
g G
r e
n e - _
4 1
3 b
s o
G e
n e
r 4 f
s at yo l m 1 m 3 a - - a 1 oV u e - -
2 t e
qn t S 1 S l 2
eib - - - -
1 S b .
o 1 1
B 1 1
- 0 _ _ -
M-1 O
T 9
0 9
0 8
0 - _ - -
8
- - 0 7
0 - _ - -
7
- 0 8
0 - -
8
- 0 5
0 _ _ - -
- 5 : 5
- 0 4
0 _ - -
4 0
3 0
t 3 0
m a
s e
_ a 2
0 m
a2
_ e 0
_ 1 r
0 8 8 5 4 3 a %,
2 1
0 8 8 2 3}o }j fj p
cu h l 8.0 TUBE LEAK RATE AND TUBE BURST PROBABILITIES 8.1 Calculation of Leak Rate and Tube Burst Probabilities
- This section discusses tube leak and burst probability analyses using voltage. _
distributions projected for the end of the operating period. The calculation utilizes correlations relating bobbin voltage amplitudes (either measured or calculated) to free span burst pressure, probability of leakage and associated leak rates for ODSCC indications at TSP locations. The methodology used is documented in Reference 9.2, andis consistent with NRC criteria and guidelines of References 9.1. Leak rates __
based on the actual measured voltages are calculated using a leak rate correlation independent of voltage, and the leak rate calculations based on the projected EOC-10 voltages utilize the leak rate vs. bobbin voltage correlation shown in Section 5.0. The latest ARC correlations submitted to the NRC were applied for the EOC-10 projections, and they are documented in Reference 9.6. The same database used earlier for the EOC-9 projections, documented in Reference 9.8, was applied for the actual EOC-9 conditions. The calculated leak rates are volumetric rates at room temperature and they should be with compa_ red with allowable leak rates at room temperature. j 8.2 Predicted and Actual Leak Rate and Tube Burst Probability for EOC-9 Analyses were performed to calculate SLB tube leak rate and probability of burst for the actual bobbin voltage distribution at EOC-9 (with no growth projection applied) previously presented in this report. The results of Monte Carlo calculations performed based on the actual voltage distributions including NDE uncertainties are shown on Table-8-1. Projections for EOC-9 conditions for all four SGs presented in the last 90-day report are also included for comparison in Table 8-1. The allowable SLB rate for the last operating cycle (Cycle 9) was 3.7 gpm (at room temperat'ure).
As in the EOC-9 projections performed after the EOC-8 outage, SLB leak rates based on the actual measured voltages were calculated using a leak rate distribution that is independent of the bobbin voltage. For the steam gerierator yielding the largest leak rate based on EOC-9 voltages, SG-3, the leak rate was also calculated using a voltage-dependent leak rate correlation since the NRC requirements for a correlation are now met.
Comparisons of the EOC-9 actuals with the corresponding predictions indicate the following:
a) The actual number ofindications found during the EOC-9 inspection in all SGs q:\apc\tva98\tva90 day. doc l
8-1 l
t7, /@.
are significantly below those projected during the last outage using both a constant POD =0.6 as well as the voltage-dependent POPCD. The peak voltages measured for all four SGs are well below that projected with POD =0.6 as well as POPCD.
b) SG-1 was projected to be the limiting steam generator for EOC-9 on the basis of SLB leak rate and tube burst probability, but SG-3 was found to have the largest leak rate and tube burst probability based on the actual EOC-9 voltages. However, the predicted SLB leak rates and tube burst probabilities are all small for all SGrand the absolute magnitude of the differences between the SGs are even smaller. So it is not surprising to find SG-3 to have a slightly higher leak rate and burst probability than SG-3 which was projected to be the limiting SG.
c) For all SGs, SLB leak rates based on the actual voltage distributions are less than those projected with POD =0.6 as well as POPCD; they are also well below the acceptance limit (3.7 gpm at room temperature). ,
i -
d) For all SGs, tube burst probabilities based on the actual voltage distributions
,are less than the projections with POD =0.6 as well as POPCD; they are also below the NRC reporting guideline of10 2 In summary, the limiting SLB leak rate (0.17 gpm at room temperature) and tube burst probability (1.9x10-5 ) calculated using the actual measured EOC-9 bobbin voltage distributions and the approved ARC database are well below the corresponding allowable limits (3.7 gpm and 10-2, respectively). The results meet the ARC requirement for continued operationT 8.3 Projected Leak Rate and Tube Burst Probability for EOC-10 Using the methodology previously described, calculations have been performed to predict the EOC-10 performance of all four steam generatorn in Sequoyah Unit-1, and the results are summarized in Table 8-2. EOC-10 bobbin voltage distributions as well as the leak rates and tube burst probabilities based on-these distributions are predicted. As mentioned earlier, EOC-10 leak rates and tube burst probabilities are calculated using the latest ARC correlations presented in Reference 9.6. The projected leak rates are compared with the alliwable leak rate at room temperature l (3.7 gpm). _ The leak rate vs. bobbin voltage correlation shown in Section 5.0 is l applied. Since growth rate for Cycle 8 is higher than that for Cycle 9, Cycle 8 growth data were used in the EOC-10 projection analysis.
q:\apc\tva98\tva90 day. doc 8-2
@ fj ,
The predicted EOC-10 SLB leak rate and burst probability for all four SGs are shown in Table 8-2. It is evident that the projected maximum voltage, SLB leak rate and tube burst probability for the EOC-10 condition for all 4 SGs are in a narrow range.
SG-1 is predicted have a slightly higher SLB leak rate than the other 3 SGs while SG-
- 2 is predicted to have the highest number ofindications. SG-3 is predicted to have a -
slightly higher peak voltage as well as a slightly higher tube burst probability. The l
limiting EOC-10 SLB leak rate predicted for SG-1 based on constant POD of 0.6 is 0.3 gpm (room temperature) which is more than a decade below the current licensed limit of 3.7 gpm at room temperature. The limiting EOC-10 burst probability with POD =0.6, predicted for SG-3, is 4.7x104; it is more than 2 orders of magnitude below the NRC acceptance limit of10 2 The results based on the voltage-dependent POPCD also show similar margins. Thus, projected EOC-10 results meet the ARC requirement for continued operation.
In summary, SLB leak rates and tube burst probabilities projected for EOC-10 for all four SGs using the NRC-mandated POD = 0.6 meet the SER limits for Sequoyah Unit-1. Results based on voltage dependent POPCD show even a greater margin between EOC-10 predictions and acceptance limits. _ -
a 1
l I
l - - -
q:\apc\tva98\tva90 day. doc -
! 8-3 l
Fi?
(O Table 8-1 Sequoyah Unit-1 1998 EOC-9 Outage Summary ct Calculations of Tube Leak Rate and Burst Probability for the EOC-9 conditions Number Steam POD of Max. Burst Probability SLB Generator Indication? Volts _
Leak Ratem 1 Tube 1 or More (gpm)
Tubes EOC - 9 PROJECTIONS (Based on ARC Database Presented in Reference 9.8 - Leak Rate Correlation Not Used) 0.6 145.7 3.1 8.8 x 10 - 5 8.8 x 10 - 5 0.75' )
1 POPCD 138.1 2.9 4.9 x 10 5 4.9 x 10 5 0.55 l
~
0.6 141.3 2.6 4.4 x 10 -5 4.4 x 10 0.47 2 POPCD 150.6 2.5 1.6 x 10 - 5 1.6 x 10 -5 0.36
~
0.6 127.7 3.0 5.4 x 10 5 5.4 x 10 5 0.71 3 POPCD 107.0 2.7 3.9 x 10 5 3.9 x 10 - 5 0.46 0.6 70.0 3.1 3.7 x 10 -5 3.7 x 10 5 0.31 4 POPCD 65.5 2.7 2.5 x 10-5 2.5 x 10 5 0.20 EOC - 9 ACTUALS i
_ . (Same ARC Database as Used in the Above Proje.S t ions-Leak Rate Correlation Not Used) 1 110 1.6 L2 x 10 5 1.2 x 10- 5 0.10 !
W 1 1 130 1.6 L9 x 10-5 1.9 x 10-5 0.08 l 2
1 83 1.8 1.9 x 10- 5 1.9 x 10 5 0.17 l 3
4 1 54 1.6 1.2 x 10-5 1.2 x 10 5 0.06 l Leak Rate Correlation Presented in Reference 9.6 Applied 83 1.8 1.9 x 10- 8 1.9 x 10 s 0.09
'3 1 Notes (1) Adjusted for POD.
(2) Volumetric leak rate adjusted to room temperature i
l q:\apc\tva98\tva90 day. doc 8-4
s ,: , 1
. t.o Table 8-2 Sequoyah Unit-1 Summary of Projected Tube Leak Rate and Burst Probability for EOC-10
~
- ~ (Based on projected Cycle 10 length 480 EFPD).
Burst Probability SLB Steam POD No. of Max. Comments ;
i
~
Indic- Volts
~
One or Leak Generator More Rate ations(U 1 Tube Tubes (gpm)(2)
ARC Database and Correlations Reported in Reference 9.6 Applied 2.5 2.5x10 8 2.5x10-8 0.30 1(8) , 183.3 213.7 2.4 1.9x10 8 1.9x10* 0.26 -
2'" 0.6 2.7 4.7x104 4.7x10 8 0.28 Leak rate
'3(O 131.7
- Correlation 1.9x10 5 1.9x104 0.15 4(* 90 2.5 applied 1.9x104 1.9x104 0.25 1(8) 192.1 2.4 2.3 3.7x10-8 3.7x10-8 0.23 2(o POPCD 2210.8 ,
2.6 4.7x10.s 4.7x104 0.20 340 121.8 2.4 1.9x10.s 1.9x104 0.12 4(0 85.4 Notes ~
(1). Number ofindications adjusted for POD.
(2) Volumetric leak rate adjusted to room temperature.
(3) SG-1 specific Cycle 8 growth rate distribution applied.
(4) All SG composite Cycle 8 growth rate distribution applied.
l i
q:\ ape \tva98\tva90 day. doc f
8-5
, . __ . . . - . . - - . . . -. _. - . . . . _ = ..
S..; j Q!-[
9.0 REFERENCES
1 l 9.1 NRC Generic Letter 95-05, " Voltage-Based Repair Criteria for Westinghouse Steam Generator Tubes Affected by Outside Diameter Stress Corrosion I - Cracking", USNRC OfEce of Nuclear Reactor Regulation, August 3,1995.
9.2 WCAP-14277, Revision 1, "SLB Leak Rate and Tube Burst Probability Analysis Methods for ODSC,C at TSP Intersections," Westinghouse Nuclear Services Division, December 1996.
~
9.3 SG-97-07-006, "Sequoyah Unit-1 Cycle 9 Alternate Plugging Criteria 90-Day i Report," Westinghouse Nuclear Services Division, July 1997. l l
9.4 SG-96-01-007, Revision 1, "Sequoyah Unit-1 Cycle 8 Alternate Plugging Criteria 90-Day Report," Westinghouse Nuclear Services Division, February 1996.
l 9.5 Letter from B. W. Sheron, Nuclear Regulatory Commission, to A. Marion,
- l Nuclear Energy Institute, dated Fdbruary 9,1996.
9.6 ' Addendum-2 to EPRI Report NP-7480-L, " Steam Generator Outside Diameter Stress Corrosion Cracking at Tube Support Plates - Database for Alternate Repair Criteria," April 1998.
9'.7 EPRI Report NP-7480-L, " Steam Generator Outside Diameter Stress Corrosion Cracking at Tube Support Plates - Database for Alternate Repair Limits," j Volume 1, Revision 2, August 1996.
- 9.8- Letter from S. Cc Jain, Duquesne Light Company, to U.S. Nuclear Regulatory !
! Commission, " Beaver Valley Power Station ' Unit No.1,' Docket No. 50-334, License No. DPR-66, Steam Generator Pulled Tube Data (Suppl'emental)
L l
Supporting Alternate Tube Plugging Criteria Implementation," dated March 27,1996.
l t
l 4
d q:\apc\apr98\apr90 day. doc l 9-1 1.