ML20217B626

Safety Evaluation Supporting Amend 202 to License DPR-51

ML20217B626
Person / Time
Site:	Arkansas Nuclear
Issue date:	10/04/1999
From:	NRC (Affiliation Not Assigned)
To:
Shared Package
ML20217B619	List: ML20217B617 ML20217B626
References
NUDOCS 9910120320
Download: ML20217B626 (12)

Text

-

G E*

p UNITED STATES j

NUCLEAR REGULATORY COMMISSION WASHINGTON, D.C. 20666 4001

• • • • • ,o SAFETY EVALUATION BY THE OFFICE OF NUCLEAR REACTOR REGULATIOE RELATED TO AMENDMENT NO. 202 TO FACILITY OPERATING LICENSE NO. DPR-51 ENTERGY OPERATIONS. INC.

ARKANSAS NUCLEAR ONE. UNIT NO.1 DOCKET NO. 50-313

1.0 INTRODUCTION

By "1CAN059901, Application for Amend to License DPR-51,revising TS SRs & Applicable Bases Relevant to ISI Requirements for Portions of once-through SG Tubes Encompassed by Upper Tube Sheet. Proprietary Rept Encl.Rept Withheld,Per [[CFR" contains a listed "[" character as part of the property label and has therefore been classified as invalid.)|letter dated May 14,1999]] (1CAN059901), as supplemented by letters dated June 17,1999 (1CAN069905), September 7,1999 (1CAN099901), September 15,1999 (1CAN099902),

September 17,1999 (1CAN099905,', and September 24,1999 (1CAN099906), Entergy Operations, Inc. (the licensee), submitted a request for changes to the Arkansas Nuclear One, Unit No.1, Technical Specifications (TSs). The requested changes would allow once through steam generator tubes (OTSGs) with confirmed volumetric indications within the upper tubesheet to remain in service during cycle 16. The licensee has proposed to revise TS 4.18.5.a.7 to incorporate by reference acceptance criteria to allow steam generator tubes to remain in service during cycle 16 with indications of outer diameterintergranular attack (ODIGA)in the upper tubesheet region of the steam generators. in addition, the licensee's submittal dated September 17,1999, included additional clarification of its commitments to assess ODIGA degradation in support of the proposed amendment.

Given that the licensee initiated a refueling outage at ANO-1 on September 10,1999, the licensee and the NRC were unable to fully resolve all technical differences in order to support the permanent approval of the proposed amendment. As such, the repair criterion considered herein are applicab!e only for operation during cycle 16. The absence of a repair criteria to address the ODIGA degradation would require the repair or plugging of a significant number of steam generator tubes during the outage.

The letters dated June 17,1999, September 7,1999, September 15,1999, September 17, 1999, and September 24,1999, provided clarifying information that was within the scope of the original application and did not change the staff's initial proposed no significant hazards consideration determination.

2.0 BACKGROUND

The existing depth-based steam generator tube repair limit (40 percent throughwall) in the ANO-1 TSs attempts to ensure adequate tube integrity through the end of the next operating cycle. The licensee has proposed alternate criteria that include additional inservice inspection, consideration of flaw structurallimits based on length rather than depth, and reliance on results of in-situ pressure testing to demonstrate adequate leakage integrity margins. These criteria 9910120320 991004 PDR ADOCK 05000313 P

PDR

(

. I wou;d allow tubes with ODIGA indications to remain in service through cycle 16. The proposed 3 TS mandates the performance of bobbin and rotating probe eddy current inspections in the upper tubesheet region, references Framatome Technologies, Inc. (FTI) Topical Report BAW-10235P, " Management Program for Volumetric Outer Diameter Intergranular Attack in the Tubesheets of Once-Through Steam Generators," Revision 1, as the repair criteria applicable to ODIGA degradation and requires the licensee to submit a summary of its condition monitoring and operational assessment results for the ODIGA degradation. In addition, the licensee will perform in-situ pressure testing if growth is detected from an assessment of Plus i

Point voltage amplitude measurements obtained in the 1R14 and 1R15 outages. The sample j

size would be defined such that leakage through lGA indications would, with a 95 percent confidence, be less than 1 ppm during a main steam line break (MSLB).

l The [NO-1 TSs require that tubes having degradation greater than 40 percent through

~

repaired or removed from service. During the steam generator tube inspection in the refueling outage in September 1996 (1R13), the licensee used an eddy current (EC) bobbin probe to size the depth of indications in the upper tubesheet that were attributed to ODlGA. Prior to the inspection, the licensee had qualified an EC sizing technique specifically for measuring the depth of the IGA indications. As a result of the inspection, a number of tubes with IGA t

indications were returned to service because the depths of the indications were measured to be less than the tube repair limit in the TS. Those IGA indications measured during 1R13 that were thought to exceed the 40 percent throughwall criteria were removed from service during that outage.

During the 1 R13 outage, the licensee removed three tubes;containing a total of 11 IGA indications for destructive examinations. The tubes were selected on the basis of indications that bounded the IGA degradation of the tubes that were left in service. After performing burst tests that subjected the tubes to pressures well in excess of normal operating pressures, the licensee compared the actual depths of the IGA degradation measured by destructive examinations to the depths predicted by the EC sizing technique used during the tube inspection program. The comparison uncovered a systematic nonconservatism in the EC sizing technique. The discrepancy in the IGA measurements and predications raised concerns that some of the tubes left in service actually had IGA degradation that exceeded the TS repair criteria of 40 percent throughwall.

When the noncompliance was identified, the licensee requested and was granted a notification of enforcement discretion (NOED) on April 9,1997. The basis for the NOED was that, although some of the IGA indications could exceed the 40 percent throughwa!! repair criteria, confidence in the structural and leakage integrity of the tubes was provided by the burst tests performed, the performance history of the tut,es at ANO-1, and the added support provided by the upper tubesheet. The licensee submitted an exigent TS change on April 11,1997, to allow a one-time exception to the surveillance requirements of Section 4.18.5.b. This exception allowed tubes with ODIGA indications within the upper tubesheet with potential throughwall depths greater than the plugging limit to remain in service for the remainder of cycle 14. The April 11,1997, l

submittal was supplemented on May 2,1997, with an additional TS change which reduced the leakage limit through the steam generator tubes from 500 gallons per day (gpd) to 144 gpd for the remainder of cycle 14. In response to this request, the NRC issued Amendment No.189 to the ANO 1 license dated May 7,1997. This amendment allowed the unit to continue operation through the remainder of cycle 14 with tubes that had potential throughwall defects in excess of j

the 40 percent plugging limit.

l

[

3-i j-Prior to 1R 13, the Babcock and Wilcox Owners Group (B&WOG) was working on the development of an attemate repair criteria (ARC) for volumetric ODIGA flaws. In response to E

d l

the events at ANO-1, the B&WOG expedited its schedule and focused its initial work on volumetric ODIGA indications within the tubesheet. ANO-1 is the lead plant for NRC review of

}

proposed TS changes associatea with the implementation of the ARC. A submittal was transmitted on August 13,1997, which included a general steam generator description and

. discussion of plant chemistry, flaw morphology of pulled tubes, nondestructive examination of pulled tubes, and a comparison of laboratory developed ODIGA and field ODIGA. A second j.

submittal was made on December 12,1997, with a request for approval for use during 1R14 which began on March 28,1998.

Through discussions between the licensee and the NRC staff during March 1998, it was determined that insufficient time was available to resolve outstanding issues related to the ARC prior to 1R14. Since many of the arguments that supported allowing the tubes with ODIGA flaws to remain in service during cycle 14 remained valid for operation during cycle 15, the licensee decided to strengthen the technical justification for allowing the tubes with tubes with ODIGA indications to remain in service and pursue a one-cycle TS change. The proposed change allowed ODIGA flaws in the upper tubesheet to remain in service during cycle 15 while resolution of the outstanding issues related to the previously submitted ARC proposal are pursued.

The licensee proposed to assess ODIGA growth during the previous operating cycle and use in-situ pressure testing to demonstrate that tubes would not leak under accident conditions. If no growth is detectable, the licensee will assume that growth is not likely to occur in the next t

cycle of operation. The licensee determined that the ODIGA indications did not grow during cycle 14. Therefore, the licensee was able to demonstrate using the results of the in situ pressure test program that adequate leakage integrity margins are likely to exist for the duration

' of the next operating cycle.

Entergy submitted a license amendment to address ODIGA degradation in the upper tubesheet

. region identified during the 1R15 outage that was similar to the amendment propose'1 prior to i

the 1R14 outage. The primary difference between the new license amendment request and the

. prior request was that the licensee could rely on the in-situ pressure test results obtained during the prior refueling outage if the ODIGA degradation did not grow during cycle 15. In addition, the NRC staff would independently review data to assess ODIGA growth rates for its evaluation i

of the proposed license amendment.

i 3.0 EVALUATION 3.1 Inservice Insoection of Steam Generator Tubes The inservice inspection scope for the current ANO-1 refueling outage includes an examination l

of 100 percent of the steam generator tubes in the upper tubesheet region with a bobbin coil eddy current probe. Tube examinations with this probe are expected to should identify indications of ODIGA degradation that could degrade the tube structural and leakage integrity margins. The bobbin probe, however, cannot assess the morphology or size of detected indications. Because the proposed TS criteria for allowing tubes to remain in service apply only to ODIGA degradation, the licensee will inspect all indications detected with a bobbin coil probe

using rotating eddy current probes. These rotating probe examinations ccn confirm that the morphology of bobbin l-dications is volumetric, is indicative of ODIGA. In addition, the data acquired in the rotating probe inspections can be used to assess the axial and circumferential length of confirmed tube degradation, and whether it initiated from the inside or outside diameter of the tube.

IGA degradation is characterized as a mode of degradation that is volumetric rather than crack-like in nature. That is, the degradation affects a small volume of tube material and l

typically has dimensions that extend axially, circumferentialiy, and radially (depth) in the tube.

The ODIGA degradation is expected to exhibit a morphology that extends both along the tube axis and around the circumference. Crack-like indications, however, extend primarily along only l

two tube directions (i.e., radial / axial, radial /circumferential). Because rotating probes are sensitive to degradation extending in both the axial and circumferential directions, these probes l

are capable of providing data to allow determination of whether an indication is crack-like or volumetric. In addition, they possess the capability to size the length of steam generator tube l

degradation. The licensee will inspect all indications detected by bobbin coil with a rotating probe that includes both Plus Point and pancake coils. This will enable the licensee to confirm the mode of the degradation, measure the Plus Point coil voltage amplitude response from the degradation, and determine the axial and circumferential length of the indication, if applicable.

The staff finds that the proposed inspections are adequate to define the ODIGA indications on l

which the in-situ testing program will be based.

3.2 StructuralIntearity Assessment The existing TS requirements specify that tubes shall be repaired or removed from service when degradation exists within the tube that is equal to or greater than 40 percent of the nominal tube wall thickness. In order to address the acceptability of tubes with confirmed ODIGA degradation, the proposed one-cycle amendment would allow tubes with ODIGA degradation to remain in service based on the performance of the additional inspections noted in Section 3.1, imposing alternate structural acceptance criteria to the depth-based requirements currently in the TSs, and demonstration of no growth for the population of ODIGA indications in the ANO-1 OTSGs over the previous operating cycle. The proposal would allow tubes with ODIGA indications to remain in service only in the area defined to be within the upper tubesheet located by eddy current inspection between 1 inch above the secondary face and below the roll transition. The 1-inch exclusion zone above the tubesheet secondary face was established to ensure proper characterization of indications detected during the eddy current inspections.

}

The licensee has performed burst tests to demonstrate that the diametral clearance between the tube and adjacent tubesheet prevents tubes with simulated ODIGA flaws from deforming I

. sufficiently to cause burst.- The burst test program consisted of nine tubes containing drilled throughwall holes up to 0.5 inch in diameter and one tube containing no defects. All tubes with j

- the laboratory defects were tested within a simulated tubesheet. Nine of the specimens burst at pressures greater than 10,941 psig. Each tube burst outside the tubesheet within the non-

' defected portion of the tubes. Testing problems limited the maximum pressure for one test to 9,577 psig, at which the tube had not yet burst. The burst test results indicate that the tubesheet provides sufficient support to preclude tube burst within the tubesheet. The licensee submitted a description of the test program and discussion of this conclusion in Topical Report l

l:

5-BAW-10226P, " Alternate Repair Criteria for Volumetric Outer Diameter Intergranular Attack in the Tubesheets of Once-Through Sham Generators," Revision 0, submitted August 13,1997.

In the 1R14 outage, the licensee completed in-situ pressure testing of tubes containing 40 ODIGA indications. Tubes were pressurized to a pressure of 2900 psig, which is in excess of pressures expected to occur during an MSLB accident. Because the peak accident-induced loads for steam generator tubing are largely a result of thermally induced stresse rather than intamal tube pressure, in-situ pressure tests on 36 of the 40 ODIGA indications imposed axial tube loads in conjunction with the internal pressure. These loads could challenge the structural or leakage integrity of tubes containing circumferentially oriented degradation of significant length or depth. None of the tubes with indications tested under these conditions leaked. The licensee subjected the four remaining ODIGA indications that were in-situ pressure tested to internal pressure loads only. However, the maximum intemal pressure imposed during the testing was 6500 psig, more than 2.5 times the MSLB differential pressure. No leakage was detected from tubes with these indications during the tests.

To minimize the potential for structural failure of tubes with ODIGA degradation, the licensee has proposed to remove from se,vice those tubes containing degradation expected to exceed length-based repair limits included in BAW-10235P, Revision 1. The length limits will require the licensee to remove from service tubes with ODIGA flaws that exceed 0.5 inch in length in either the axial or circumferential direction of the tube. Tube suuctural failure due to axially oriented flaws in the upper tubesheet region is precluded by the presence of the tubesheet.

Tubes with circumferential flaws, however, could fait due to axial tube loads. Testing completed by the licensee indicates that throughwall flaws with circumferential extents up to 140' (0.72 inch) will have structural margins commensurate with the margins specified in Regulatory Guide (RG) 1.121, " Bases for Plugging Degraded PWR [Pressuriz.sd-Water Reactor} Steam

{

Generator Tubes." Limiting ODIGA flaws to less than 0.5 inch in circumferential length will provide additional margin to account for uncertainty in measuring the flaw with nondestructive testing methods. The staff concludes that the proposed length-based repair limits are adequate to ensure that tube structuralintegrity margins will be maintained in accordance with guidance provided in RG 1.121.

The staff notes that tubes with volumetric lGA degradation in steam generators at other PWR facilities have also shown significant margins for structural and leakage integrity. Burst tests of tubes removed from service with IGA indications have shown significant margins for structural and leakage integrity under postulated accioant conditions. In addition, the licensee for ANO-1 has not attributed any measurable operational leakage in the steam generators to the presence of ODIGA degradation in the tubing. The licensee also has extensive testing data from in-situ pressure tests completed in the IR14 outage that indicates that this mode of degradation does

' not significantly degrade the structural margins of the OTSG tubing. On the basis of the conservatism in the length-based repair limits and the empirical data indicating a low likelihood of burst for ODIGA degradation, the staff concludes that the proposed amendment will ensure that the structural margins for OTSG tubing with ODIGA degradation will be maintained in cycle 16.

3.3 Demonstration of Leakaae inteority Marains The existing depth-based repair criteria are established to ensure steam generator tubes have adequate structural and leakage integrity with appropriate margins of safety under normal 4

1 i

I operating and postulated accident conditions. The approach proposed by the licensee may permit tubes containing degradation with actual c.apths greater than the 40 percent depth to remain in service. Under high differential pressures, this degradation could become a leak path for the reactor coolant to the steam generator secondary side. The licensee will rely on

- previously completed in situ pressure testing of steam generator tubes with ODIGA indications and an assessment of ODIGA growth rates to demonstrats a low leakage potential for tubes j

containing this mode of degradation. The growth rate study must conclude that there is a low j

likelihood that the ODIGA degradation grew during the previous cycle of operation in order for I

the licensee to use the in-situ pressure test results obtained in the 1R14 outage. Detection of flaw growth will necessitate the performance of additional in-situ pressure testing in the 1R15 outage to reassess tube leakage integrity margins.

in-situ pressure testing subjects degraded tubes to conditions that are conservative with respect to intomal pressure loadings postulated to occur under accident conditions. Intemal pressure 4

within the tube during the test induces axial and circumferential stresses within the tube wall.

I The purpose of the testing is to assess whether the degraded tubes exposed to these elevated stresses are capable of withstanding the test conditions while retaining leakage and structural integrity. There is a substantial database that indicates that leakage from ODIGA degradation is unlikely to occur. Destructive examinations of the ODIGA patches in tubes removed in 1R13 showed none of the flaws to be throughwall. The licensee has also compieted testing of tubes I

with holes of varicus sizes and depths machined into the tubing. The artificial patches were machined to depths ranging from approximately 84 percent to 95 percent throughwall, with patch diameters of 0.30 and 0.50 inch. The severity of the macnined patches bouna the potential effects of having an ODIGA patch in a tube that is of similar depth and diameter.

None of the machined patches showed signs of leakage when subjected to accident loads.

To augment the existing database of pulled tubes and machined IGA defects, the licensee developed altificial IGA flaws under laboratory conditions. The defects in these tubes were corrosively-induced rather than mechanically induced in order to more closely simulate the degradation fou.nd in the actual steam generator tubes. The licensee subjected a total of 29 simulated IGA defects to design-basis loading conditions. The sizes of these defects bounded those of the ODIGA flaws presently in the ANO-1 OTSG tubing. All of the tube specimens retained structural and leakage integrity at the target test conditions.

The licensee reports that there have been no known primary-to-secondary leaks in the history of ANO-1 attributed to volumetric ODIGA indications despite the fact that many of these indications have remained in service for years. In addition, during a May 1996 plant transient, the "B" steam generator tubing was subjected to a differential pressure of approximately 2100 paid for several hours. No leakage from ODIGA flaws was observed during the event or following plant stalW.

The licensee implemented an ODIGA repair criteria in 1R14 that required in-situ pressure testing a number of tubes with ODIGA indications. The objective of the testing was to

~ demonstrate a 0 95 probability at a 95 percent confidence level that the total leakage from ODIGA iridicaticas would not exceed 0.5 gallon per minute (gpm) during an MSLB accident. By testing t' e integrity of a representative sample of tubes, the licensee could assess the potential n

for leakage from all of the tubes identified by the eddy current inspections to have ODIGA indications. The total leak rate from this effort was calculated rather than measured. The licensee assumed for this calculation a total ODIGA axial flaw length of 0.3 inch that is

E l

7.;

throughwall over 50 percent of the length. The leak rate was analytically detarmined using a 3

i computer program. The licensee then calculated a hypothetic al leak rate consistent with the j

I conditions for an MSLB. The licensee estimated a leak rate of 0.0185 gpm through the -

{

assumed flaw geometry.. Assuming that the ODIGA patches could contributs approximately 0.5 gpm of the licensing-baels leak rate (1.0 gpm),39 ODIGA patches could be assumed to leak and the current licensing-basis leak rate would be maintained. Given the me '. mum allowable umber of lealdng ODIGA patches (39) and the total number of ODIGA indications that were n

identified during the inspections to be performed during 1R14, the licensee statistically j

determined the number of tubes to be in-situ tested. As indicated in Section 3.2 of this safety I

evaluation, the licenses tested 40 ODIGA indications during the outage, and no leakage was detected during any of these tests.~

i l

The licensee proposed to assess f.he. leakage integrity of ODIGA degradation during the 1R15 j

outage in a manner similar to that conducted in 1R14. The primary difference between the two assessments is that the licensee may rely on the results of the in-situ pressure testing j

completed in 1R14 provided no ODIGA growth is observed between the 1R14 and 1R15

{

outages. The absence of growth during cycle 15 would indicate that the integrity of the OTSG tubes with ODIGA degradation is unchanged from the previous refueling outage. Therefore, the licensee believes that conclusions made with regard to the in-situ pressure test program would remain valid. If it cannot be concluded that the ODIGA degradation did not progress in t

cycle 15, then the licensee proposes that it will conduct in-situ pressure testing with the same objective as completed in the _1R14 outage. If this is necessary, the licerv ooses that it will calculate the total steamline break leakage using a slightly different app... m. Specifically, the teensee will calculate the leakage using a fraction of the actual measured ODIGA flaw lengths rather than an upper bound (i.e.,0.3 inch). The fraction of flaw that is assumed to be throughwall is reduced from 50 percent to 25 percent. This is based on destructive examination results of tubes removed from the ANO-1 OTSGs. The overall number of flaws assumed to exist in the population is also adjusted for the sensitivity of the inspection method (i.e.,

j j

probability of detection).

I One other modification from the previous repair criteria applied to the ODIGA degradation is that the total accident-induced leakage will be assessed relative to the licensing-basis leak rate limit of 1 gpm rather than 0.5 gpm. This represents a less conservative limit for leakage than

- previously used by the licensee, in addition, no margin exists to address leakage from other l

potential sources. If accident-induced leakage is expected from other sources within the ANO-1 OTSGs, then the licensee must ensure that the total leakage from all sources is less

~

than 1 gpm in order to remain consistent with the existing licensing basis for the plant. Such an approach is embodied in Framatome Technologies, Inc. Topical Report BAW-2346P, " Alternate j

Repair Criteria for Tube End Cracking in the Tube-to-Tubesheet Roll Joint of Once-Through L

Steam Generators," Revision 0, which is incorporated by reference in the TSs.

p The staff has identified shortcomings in the licensee's proposed approach forascessing leak i

rates for indications assumed to leak during postulated accidents. The staff notes that the

' leakage through ODIGA degradation is calculated using a computer code that has not been reviewed and approved for use by the NRC. In particular, the code needs to be benchmarked against actual leakage data for flawed test specimens representative of the ODIGA flaws at i

ANO-1. The staff notes that such data tend to exhibit significant scatter. Therefore, uncertainties associated with the leak rate estimates need to be quantified and taken into consideration to ensure they are conservative.

r4

-n

,o l* l The staff also notes that actual leaking ODIGA flaws may exhibit through-5.all components substantially longer than those assumed by the licensee which may result in nonconservative leak-rate estimates for individual flaws. Leak rates are highly sensitive to through-wall crack length. IGA and crack profiles tend to be highly variable. One acceptable approach involves 1-determining leak rates on a crack-specific basis based on measured crack length (with appropriate allowance for measurement error).

Notwithstanding the shortcomings in the licensee's approach for assessing leak rates for tubes I

l with indications assumed to leak, the staff believes that the licensee's estimate of total leak rate for the total population of ODIGA flaws in the faulted OTSG for the upcoming cycle 16 is conservative. This is based on the staff's finding (discussed in 3.4 below) that the ODIGA indications are no longer experiencing significant growth and the staff's belief that the licensee's estimate of the nu.1ber of ODIGA flaws that may leak during postulated accidents is very conservative. Future efforts to develop an ODIGA repair criteria as a proposed permeaent change to the TSs should address the aforementioned shortcomings identified by th9 NRC staff.

I

' 3.4 6nalvslLgf Growth Rate for IGA Deoradation in accordance with the guidance provided in NRC RG 1.121, steam perator tube repair limits genera!!y incorporate an allowance for degradation growth over the next cycle of operation.

Such criteria account for the progression of steam generator boe flaws in length or depth I

during operation that could potentially degrade the margine ior tube integrity below acceptable limits. The in-situ testing method for the ODIGA indicatio".s in the upper tubesheet does not utilize dimenr.ional limits nor include an allowance for growth. Therefore, the licensee's proposal to use in-situ pressure testing during the cLge is only capable of demonstrating that the population of tubes with ODIGA indications has adequate leakage integrity at the time of the test. If it can be demonstrated that the expected flaw growth rate for the ODIGA degradation is negligible, then the in-situ pressure testing will provide assurance that the affected tubes will have sufficient margins for structural and leakage integrity beyond the outage, throughout the next cycle of operation.

. The licensee has previously completed a growth assessment for ANO-1 eddy current bobbin indications. The change in bobbin voltage amplitude from 1993-1996 was determined for 129 upper tubesheet indications. The results show that the average voltage change per 6ffective full power year is "zero." The licensee attributed the variability about this average to eddy current uncertainty Of the 129 indications that were studied,25 were removed from service during 1R13. The licensee reevaluated the growth rates for the remaining 104 ODIGA indications during the 1R14 outage. As documented in the licensee's submittal dated May 1, 1998, the assessment confirmed that there was essentially no growth for the ODIGA degradation during cycle 14.

For the 1R15 outage the licensee will reevaluate the growth of ODIGA indicatioris in the ANO-1 L

OTSGs. In the previous inspections, growth rates were assessed using bobbin coil eddy

(

current data. The 1R15 growth rate evaluation will use data obtained from rotating eddy current probes. The licensee will use pancake coil data to evaluate the length (axial and t-circumferential) of each indication and Plus Point coil data to assess changes in voltage i

amplitude. If growth is detected for two of the three parameters (axial length, circumferential l

.g.

length, or voltage) for the population of indications, then the proposed repair criteria cannot be used to disposition ODIGA indications. Growth is confirmed for a parameter when the 95-percent lower confidence limit for the mean of the distrioution of measured changes is greater than zero. The licensee will also apply critena to identify potential growth o! individual ODIGA indications. Changes in lengths or voltages of individual ODIGA indications that exceed the limits specified in BAW-10235P, Revision 1, will be removed from service.

The proposed repair criteria rely on the assumpdon that the ODIGA degradation growth will not occur during the subsequent operating cycle if growth was not detected in the previoLs-operating cycle. Any changes in the size or depth of the flaws will invalidate the resu'ts of the

{

leakage assessment. The staff was concemed that the proposed growth rate criteria are insufficient to ensure the detection of changes for the population of ODIGA indicatiors. From a I

statistical standpoint, the criteria state that growth is confirmed when there is less than a

2.5-percent chance that there is no growth for two of three inspection parameters. The licensee

. stated that such criteria are necessary to ensure the detection of growth from data that include variability due to the inspection process. The licensee did not attempt to quantify inspection variability. However, the licensee will implement minor changes in the inspection process such as the use of a master calibration standard to minimize some of the inspection variability.

To address the staff's concerns the licensee limited the amendment to one cycle (cycle 16). In addition, the licensee wil: conduct in-situ pressure testing of ODIGA flaws if growth is detected.

Specifically, if the voltage corresponding to the mean for the population minus one standard

. deviation is greater than zero, than the licensee will complete in-situ pressure testing to assess potential leakage from ODIGA indications. If this testing is performed and leakage is detected from one or more upper tubesheet ODIGA flaws, then the licensee will repair or plug all tubes containing ODIGA.

The licensee's proposed amendment requires that growth rates be evaluated using criteria to address changes in the population of ODIGA degradstion and changes for individual indications. Specifically, statisticallimits established using 95-percent confidence or tolerance limits are determined for assessing growth of the population of indications or individual ODIGA flaws, respectively.

On September 24,1999, the licensee submitted to the NRC a summary of data collected for the ODIGA indications inspected during the 1R15 outage. The submittal indicated that the licensee had evaluated the data per the proposed growth rate criteria and concluded that growth was not occurring. The staff independently reviewed these data to essess any changes in the eddy current signal measurements that might be indicative of growth. The staff assessed changes in the lengths of indications (axial and circumferential) and Plus Point voltage. The staff used alternate criteria to assess whether the degradation was progressing and concluded that the population of ODIGA degradation does not appear to be growing. Therefore, the licensee's conclusion appears to be correct.

Although the staff agrees that there is no significant growth occurring at this time, the staff

' believes that the proposed criterion for making such a finding is insufficient. This criterion is defined in such a way as to indicate growth is occurring only if there is relatively high assurance that such growth is occurring. For example, a best-estimate mean growth rate higher than zero is not enough to trip the criterion. Rather, the lower one-sigma bound of the mean growth rate must exceed zero in order to trip the criterion. Future efforts to develop an ODIGA repair l

W 3

m:+*

. criterion as a proposed permanent change to the TSs should address the NRC staff's concerns regarding the growth criterion. An adequate growth criterion should provide reasonable assurance that indicated growth would be expected to trip the criterion resulting in additional in-situ pressure tests and a revised operational assessment.-

. Given the absence of significant growth, the proposed contingency actions for this one-time amendment request should growth be identified are moot. However, these contingency.

measures wi!! need additional technical justification should the licensee desire to propose an

ODIGA repair criterion as a future permanent change to the TSs. Specifically, the licensee will need to discuss how growth will be taken liito account (should it be determined to be occurring) when assessing potential accident-induced leak rates for the next end of cycle.

4.0 REQUIREMENTS FOR IMPLEMENTATION OF ODIGA REPAIR CRITERIA The following summarizes the proposed changes to the ANO-1 TSs to implement the repair criteria for steam generator tubes degraded with ODIGA degradation:

1. TS 4.18.3.a.5 A requirement is added to the TS to mandate the performance of inspections of all non-plugged and non sleeved tubes in the upper tubesheet region using a bobbin coil probe. Tubes with previously identified ODIGA degradation will not be included in the inspection results category for the OTSG. All ODIGA indications identified by a bobbin probe'will be inspected with a rotating coil probe in order to characterize the nature of the indications. This is an enhancement of existing TS requirements J

and is acceptable, l

2. TS 4.18.5.a.7 The licensee proposes to modify the definition of " Plugging Limit" to state that the depth-based repair limit does not apply to ODIGA degradation in the upper 1

tubesheet region during cycle 16. Tubes with these indications will be repaired in accordance with topical report BAW-10235P, Revision 1, as evaluated above.

Therefore, this change is acceptable.

3. TS 4.18.5.b n

Much of TS 4.18.5.b is deleted because this TS specified the requirements for applying a repair criteria for ODIGA degradation during cycle 15. Because cycle 15 has been completed, these requirements are no longer applicable. Therefore, this change is acceptable.

4. JS 4.18.6.f A requirement is added to the TS to report the results of the licensee's condition monitoring and operational assessments for upper tubesheet ODIGA degradation.

i Therefore, this change is acceptable.

)

i i

n i

3 J

5.0 -

SUMMARY

1 The staff finds'that implementation of the proposed ODIGA repair criteria is r.cceptable for the upcoming operating cycle (i.e., cycle 16). The staff has independently assessed the ODIGA inspection results obtained during the current outage and concurs with the licensee's finding that the subject ODIGA indications are not experiencing any significant growth. Given the absence of significant growth, the in-situ pressure tests performed during the 1R14 outage indicate that the licensee's estimate of the number of ODIGA indications that may potentially leak during postulated accidents is a consevative upper bound and that the licensee's estimate

- of total accident-induced leak rate is conservative. Therefore, the proposed changes to the ANO-1 TSs are acceptable to justify allowing tubes with ODIGA indications to remain in service

~

during cycle 16 operat:on.

As a point of clarificat!on, it is the staff's understanding that projected total accident-induced leak rate from all contributing degradation mechanisms (including ODIGA and tube end cracks)

- will be limited to 1 gpm as is stated in FTl Topical Report BAW-2346P, Revision 0, which was incorporated by reference into tho TSs in Amendment No. 201.

'The staff's evaluation above has identified shortcomings in the licensee's proposed change for a permanent ODIGA repair criteria. The licensee's assessment of potential accident-induced leak rates, assuming leakage occurs, has not been performed in a rigorous manner. The staff believes there is significant uncertainty associated with these estimates that has not been addressed. In addition, the staff finds that the proposed criterion for ensuring that no growth is occurring is not sufficient for this purpose. In light of these shortcomings, the staff was unable to approve this change on a permanent basis.' Any future submittals regarding a proposed i

' ODIGA repair criteria should address the technical issues raised by the staff.

6.0 STATE CONSULTATION

in accordance with the Commission's regulations, the Arkansas State official was notified of the I

proposed issuance of the amendment. The State official had no comments.

7.0 ENVIRONMENTAL CONSIDERATION

The amendment changes a requirement with respect to installation or use of a facility component located within the restricted area as defined in 10 CFR Part 20. The NRC staff has j

determined that the amendment involves no significant increate in the amounts, and no significant change in the types, of any effluents that may be released offsite, and that there is no significant increase in individual or cumulative occupational radiation exposure. The Commission has previously issued a proposed finding that the amendment involves no significant hazarda consideration, and there has been no public comment on such finding (64 FR 29709, June 2,1999). Accordingly, the amendment meets the eligibility criteria for categorical exclusion set forth in 10 CFR 51.22(c)(9) Pursuant to 10 CFR 51.22(b), no environmental impact statement or environmental assessment need be prepared in connection with the issuance of the amendment.

12-8.0' CONCLUSION The Com:nission has concluded, based on the considerations discussed above, that (1) there is reasonable assurance that the health and safety of the public will not be endangered by operation in the proposed manner, (2) such activities will be conducted in compliance with the Commission's regu!ations, and (3) the issuance of the amendment will rot be inimical to the common defense and security or to the health and safety of the public.

Principal Contributor: P. Rush Date: October 4, 1999-f

',