ML091050575
| ML091050575 | |
| Person / Time | |
|---|---|
| Site: | Sequoyah  |
| Issue date: | 04/14/2009 |
| From: | Cleary T Tennessee Valley Authority |
| To: | Document Control Desk, Office of Nuclear Reactor Regulation |
| References | |
| TAC MD9595 | |
| Download: ML091050575 (11) | |
Text
Tennessee Valley Authority Post Office Box 2000 Soddy Daisy, Tennessee 37384-2000 Timothy P. Cleary Site Vice President Sequoyah Nuclear Plant April 14,2009 U.S. Nuclear Regulatory Commission ATTN: Document Control Desk Washington, D. C. 20555-0001 In the Matter of ) Docket No. 50-328 Tennessee Valley Authority (TVA) )
SEQUOYAH NUCLEAR PLANT (SQN) - UNIT 2 - STEAM GENERATOR TUBE INSPECTION INFORMATION, RESPONSE TO REQUEST FOR ADDITIONAL INFORMATION (RAI) (TAC No. MD9595)
References:
- 1. NRC letter to TVA dated March 17, 2009, "Sequoyah Nuclear Plant, Unit 2-Request for Additional Information Regarding Steam Generator Tube Inservice Inspection Report For the Cycle 15 Refueling Outage (TAC No. MD9595)"
- 2. TVA letter to NRC dated August 27, 2008, "Sequoyah Nuclear Plant, Unit 2-Unit 2 Cycle 15 (U2C15) 90-Day Steam Generator (S/G) Report For Voltage-Based Alternate Repair Criteria And W* Alternate Repair Criteria" This letter responds to NRC's RAI as contained in Referenced 1. The enclosure provides TVA responses to the NRC questions associated with the SG tube inspections.
There are no commitments contained in this letter.
If you have any questions about this change, please contact Beth A. Wetzel at (423)-843-7170.
Sincerely, Timothy P. Cleary Enclosure cc: See page 2 printed on recycled paper
U.S. Nuclear Regulatory Commission Page 2 April 14, 2009 cc (Enclosure):
Mr. Tracy J. Orf, Project Manager U.S. Nuclear Regulatory Commission Mail Stop 08G-9a One White Flint North 11555 Rockville Pike Rockville, Maryland 20852-2739 Mr. Lawrence E. Nanney, Director Division of Radiological Health Third Floor L&C Annex 401 Church Street Nashville, Tennessee 37243-1532
ENCLOSURE TENNESSEE VALLEY AUTHORITY (TVA)
SEQUOYAH NUCLEAR PLANT (SQN)
UNIT 2 RESPONSE TO REQUEST FOR ADDITIONAL INFORMATION (RAI)
NRC Question 1 Please discuss the results of your foreign object search and retrieval (FOSAR). In particular, please discuss whether any loose parts (foreign objects) were left in the steam generator following FOSAR. If so, please confirm that an analysis was performed that demonstrated tube integrity would be maintained until the next steam generator tube inspection. Please discuss whether the locations where possible loose part indications (detected during the eddy current inspection) were inspected visually. If not, please discuss how the tubes near these possible loose parts were dispositioned.
TVA Response FOSAR was performed on the top of tubesheet in each steam generator (SG) and three small pieces of foreign material were observed and removed. In SG No. 3, a small piece of wire 7/8 inch by 1/16 inch diameter was observed in the annulus and removed. In SG No. 4, a 1/2 inch by 1/8 inch piece of material (thought to be flexitalic gasket) was observed and removed. In SG No. 1, a 3/4 inch by 1/2 inch piece (thought to be weld slag material) was observed and removed. Possible loose part indications (PLPs) detected by eddy current inspection were investigated with no foreign material observed near the PLPs. Eddy current inspections did not detect any wall loss, which could be attributed to foreign material.
NRC Question 2 Please discuss the scope and results of your secondary side steam drum inspections.
TVA Response The SQN procedure for performing SG secondary side maintenance activities requires the following when the steam drum inspections are performed.
- Visual inspection of the secondary side separator area and swirl vanes, paying special attention to the leading edge.
- Inspection of the steam dryers.
- Inspection of the steam flow transmitter penetrations to ensure they are free of debris.
- Inspection of the level transmitter penetrations to ensure they are free of debris.
- Inspection of the accessible parts below the deck plate. Ensure no debris is in the drain cups and look for orange patchy rust.
- Inspection of the feedring J-tubes, paying special attention to the wall thickness of the J-tube mouth and the J-tube to feedring weld.
- Inspection of the accessible feedwater ring supports.
- Inspection of the conical-to-upper-section girth weld on the inner shell of the SG, watching for orange patchy rust and pitting.
E-1
- Inspection of the riser barrel adjacent to J-tubes. If the flow from the J-tube impinges on the riser barrel, this area has the potential for erosion/corrosion.
A steam drum inspection was performed on SG No. 1 with no abnormalities identified.
NRC Question 3 On page 6 of 106 in your November 21, 2008 letter, you indicated under the heading ODSCC
[outside diameter stress corrosion cracking] Circumferential at HTS [top of tubesheet hot-leg]
that all axial indications met the condition monitoring performance criteria and all were stabilized and plugged. Please confirm that this was intended to read all circumferential indications met the performance criteria and were stabilized and plugged.
TVA Response The statements under the heading ODSCC [outside diameter stress corrosion cracking]
Circumferential at HTS [top of tubesheet hot-leg] that used axial should have used circumferential.
NRC Question 4 Your condition monitoring addresses various degradation mechanisms. However, several degradation mechanisms are not discussed (e.g., axially and circumferentially-oriented outside diameter initiated stress corrosion cracking in the U-bend and at dings). Please clarify why these degradation mechanisms were not listed. (Is it because the previous operational assessment did not predict that these types of indications would be present?)
TVA Response The Electric Power Research Institute (EPRI) Steam Generator Integrity Guidelines, Revision 2, page 7-1, states:
Condition monitoring (CM) involves the evaluation of inspection results at the end of the inspection interval to determine the state of the steam generator tubing for the most recent period of operation relative to structural and leakage integrity performance criteria.
Therefore, conditional monitoring is a discussion of the indications identified during an inspection and the comparison of the indications to the performance criteria. Therefore, only the degradation mechanisms identified were discussed in the Condition Monitoring Report.
NRC Question 5 For several degradation mechanisms (e.g., freespan cracking, circumferentially oriented outside diameter initiated stress corrosion cracking near the top of the hot-leg tubesheet), the number of indications projected to be present was under predicted. Please discuss whether your operational assessment methodology was modified to account for this under prediction in the number of indications.
E-2
TVA Response The EPRI Steam Generator Integrity Assessment Guidelines, Revision 2, Section 8.1 states:
..the fundamental OA requirement is that the projected worst case degraded tube for each existing degradation mechanism shall meet the limiting structural performance parameter with 0.95 probability at 50% confidence.
The operational assessment (OA) structural evaluation methodology used was a deterministic assessment not a probabilistic assessment; therefore the quantity of indications for each degradation mechanism was not predicted. The quantity of indications was predicted in the degradation assessment. However, for each degradation mechanism, only a single indication of conservative severity was assessed. For each degradation mechanism, the OA conservatively predicted the severity; therefore no modification to the methodology was required. The Flaw Handbook partial through-wall equation to determine the burst pressure indicates pop-through of the indication (i.e., partial through-wall going to through-wall). If the most severe indication at end of cycle (EOC) will not pop-through at three times normal operating pressure (i.e.,
approximately 4100 pounds per square inch [psi]) then pop-through would not occur at main steam line break (MSLB) pressure of approximately 2405 psi. If the most severe indication is not through-wall, then leakage will not occur during the accident. Therefore, the quantity of indications is not relevant for either the structural or the leakage evaluations.
NRC Question 6 Several primary water stress corrosion cracking indications were found in the U-bend region of steam generator 2. Although the inspection sample was expanded to include all of the affected tubes in steam generator 2, similar expansions were not performed in the other three steam generators. Please confirm that the operational assessment (until U2C16) for the three steam generators in which 10 percent of the potentially susceptible tubes were not inspected accounted for the potential that a flaw similar (or larger) in size to what was detected in steam generator 2 was present. If the operational assessment did not account for this potential, please discuss why (given that the previous (pre-2008) inspections in this region in all steam generators were comparable).
TVA Response Evaluations were performed for the beginning of cycle (BOC) 16 OA using the following information. These evaluations ensured the acceptability of each SG for the next operational cycle.
Axial primary water stress corrosion cracking (PWSCC) U-bend Rows 5 through 20 The axial PWSCC identified in U-bends (Rows 5 through 20) in SG No. 2 were plugged on detection. An initial 20 percent sample inspection was performed in Rows 5 through 8 in all four SGs for this degradation mechanism. One axial indication was detected during EOC-15 examination of SG No. 2 Row 8 Column 3. Therefore, 100 percent of Rows 5 through 16 and 50 percent of Row 17 were examined in SG No. 2. No indications were observed in Rows 5 through 8 for the last two inspections (Unit 2 Cycle 13 [U2C13] or U2C14) for each of the SGs.
A Monte Carlo evaluation was performed to evaluate a postulated EOC-16 flaw. The structural performance criterion for this degradation is three times normal operating pressure differential (i.e., 4185 psi). The limiting BOC flaw depth is an estimation of a worst-case flaw left in service.
The BOC average depth in the OA evaluation was assumed to equal the maximum depth of the indication observed and BOC length in the OA evaluation was equal to the indication length E-3
found. The plus point was utilized for sizing for this mechanism during the EOC-15 examination and WCAP-15128, Revision 3 sizing information was utilized in this analysis. EPRI ETSS No.
96511.2 indicates detection as low as 27 percent maximum depth. A conservative BOC-16 flaw of 57 percent average depth was assumed. The postulated BOC flaw length was 0.550 inch.
The length measurement uncertainty was 0.139 inch and the depth measurement uncertainty was 7.800 percent (these values include analyst uncertainty). The EPRI Steam Generator Integrity Assessment Guidelines, Revision 2, typical growth estimates has been utilized for depth growth (lognormal depth growth average is 1.504 and depth growth standard deviation of 0.650) and length growth (lognormal length growth average is -3.220 and length growth standard deviation of 0.670). For these U-bend burst pressure calculations, the tubing yield and ultimate material properties were increased 15 percent to account for increased material strength due to the cold working the tube receives during the bending process (Steam Generator Degradation Specific Management Flaw Handbook, Section 3.1.1.4). TubeWorks calculated the 95th percentile lower limit burst pressure of 4198 psi. As stated above, one indication of this degradation mechanism was detected during the EOC-15 examinations. The indication was in-situ pressure tested at 3 times the operating pressure with zero leakage observed. This data validates the conservatism of the EOC-16 prediction.
Circumferential PWSCC U-bend Rows 5 through 20 Circumferential PWSCC were first observed in Rows 6 and 7 during the U2C12 examination.
Four indications of this degradation mechanism were detected during the EOC-15 inspection.
PWSCC U-bend circumferential cracking in Rows 5 through 20 were plugged on detection. An initial 20 percent sample inspection was performed in Rows 5 through 8 in all four SGs for this degradation mechanism. The sample in SG No. 2 was expanded to 100 percent of Rows 5 through 16 and 50 percent of Row 17. The structural performance criterion for this degradation is three times normal operating pressure differential (i.e., 4185 psi). For this degradation mechanism, the worst-case BOC flaw was assumed to be 55 percent degraded area (PDA) through-wall. The probability of detection curve from EPRI TR-107197, Depth Based Structural Analysis Methods for SG Circumferential Indications, (Figure 5-54) indicates the plus point coil has a 95 percent probability of detecting a 36 PDA flaw in explosive expansions. A circumferential flaw at the top of tubesheet is more difficult to detect than a circumferential flaw in the U-bend. The 55 percent PDA value is therefore conservative. The EPRI TR-107197 curve is based on destructive analysis, not on eddy current values; therefore, non-destructive examination (NDE) uncertainty is not added. Growth was assumed to be 9.000 percent PDA/effective full power years (EFPY), which bounds the SQN specific growth data taken over a number of inspections for Rows 1 through 4 circumferential indications. When PDA growth rate (i.e., 9.000 percent PDA/EFPY for 1.344 EFPYs) is added to the BOC flaw one could expect no larger than a 67.099 PDA indication at the EOC-16, which would have a 95th percentile lower limit burst pressure of 4223 psi. During U2C12, eight U-Bend PWSCC circumferential indications were discovered in Rows 6 in SG No. 2 and 20 indications of circumferential cracking in Row 7 were detected in SG No. 4. The worst U-bend PWSCC circumferential indication discovered in Rows 5 through 20 at SQN Unit 1 or Unit 2 was 5.520 PDA. This data validates the conservatism of the EOC-16 prediction.
E-4
NRC Question 7 One tube was detected with several indications of stress corrosion cracking near the tube end.
This tube was inspected with a rotating probe because it was not fully expanded in the tubesheet region. Please confirm that all of the indications in this tube were detected in the region of the tube where the previously installed plug was expanded into the tube (i.e., the plug expansion zone). In particular, address whether the indications in this tube are attributed to the previous plugging process that this tube was exposed to or whether the stress corrosion cracking is a result of the normal fabrication expansion process (i.e., tack expansion). The staff notes that the mechanism responsible for cracking is important in determining which tubes are susceptible to this cracking mechanism (and therefore are required to be inspected).
TVA Response The indications aligned with either the tube to tubesheet weld, which was expanded by size rolling prior to roll-plug installation, or aligned with the distance from the tube end where the roller expanded the plug into the tube (i.e., plug expansion zone). Therefore, TVA believes the mechanism is related to the roll-plug installation.
NRC Question 8 Regarding the one tube plugged because of the inability to perform a rotating probe examination in the U-bend region, please discuss how it was confirmed that this tube did not have a defect. If the tube could not be inspected with a rotating probe, was an in-situ pressure test performed to confirm that the tube had adequate integrity.
TVA Response Best effort examinations were attempted from both directions. The majority of the U-bend region was examined. Extra dose and time was expended without obtaining a complete examination (valid data) from either direction. No indication was detected in any of the collected data. TVA had no reason to believe the tube was defective but elected to plug the tube for as low as reasonably achievable in future outages. In-situ pressure testing was not performed.
NRC Question 9 Please confirm that you did not use the probability of prior cycle detection model in your assessment of tube integrity for outside diameter stress corrosion cracking at the tube support plate elevations.
TVA Response Probability of prior cycle detection model (POPCD) was not utilized in the U2C15 refueling outage.
NRC Question 10 Tubes with indications greater than 1.5 volts that were inspected with a worn probe were re-inspected. Some of these retested tubes also had indications that were less than 1.5 volts.
In these cases, the original voltage (i.e., with the worn probe) was used in your tube integrity analysis. Please discuss why this approach was used rather than using the voltage obtained from a probe that passed the probe wear check. In addition, please provide a listing of the E-5
indications including the worn probe voltage and the voltage obtained from the probe that passed the probe wear check.
TVA Response TVA utilizes the methodology for the indications greater than 75 percent of the repair limit (1.5 volts for SQN Unit 2 which has the 2-volt repair limit) approved by NRC in a correspondence between A. Marion of the Nuclear Energy Institute (NEI) and B. Sheron of the NRC dated February 9, 1996. This methodology requires that a tube, which contains an ODSCC indication greater than 1.5 volts, determined with a probe that failed calibration, be retested. The correspondence states that all of the eddy current test (ECT) data should be evaluated (not just the indications above 75 percent of the repair limit) in order to provide assurance that if indications were missed with the worn probe, they will be detected. The entire tube, including any indications with ODSCC less than 1.5 volts, is retested. Since the objective that is stated in the Marion correspondence is to identify missed indications, previous bobbin calls (PBC) identified indications are labeled as PBC. Any voltage associated with a PBC call cannot be used to infer the voltage of the ODSCC indication. Since there is no requirement for the retest of indications that are less than 75 percent of the repair criterion, the original indicated voltage is used in the integrity analysis.
Figure 4-1 in the 90-Day Report (Reference 2) shows that, for retest probe wear (RPW) indications greater than 1.5 volts, the final distorted support indication (DSI) voltage is scattered about the RPW voltage fairly uniformly. This is because a worn probe is not centered as well as a "good" probe. This means that it is possible that the test coil is sometimes closer to the side of the tube containing the flaw and the voltage will be larger. However, the converse is also true and there should be no bias in probe centering. Therefore, one would expect a statistically significant sample to exhibit negligible change in terms of the voltage bin populations. The following table identifies those tubes with RPW and the voltages from a probe that has passed the centering criteria.
Third Good SG Row Col Locn First Test Second Test Test Probe Cal H01 0.4V DSI PBC(0.40) 46C 27 48 H02 1.5V RPW 1.44V DSI 46C 1
28 50 H02 1.54V RPW 0.91V DSI 46C 41 37 H02 1.59V RPW 1.47V DSI 46C H02 1.64V RPW 1.84V DSI 64H 3 42 H04 0.58V DSI PBC(0.38) 64H 7 48 H01 1.53V RPW 1.12V DSI 64H 15 89 H01 1.66V RPW 1.58V DSI 40C H01 1.68V RPW 1.84V DSI 40C 2 27 68 H02 0.68V DSI PBC(0.61) 40C H01 1.54V RPW 1.63V DSI 40C 30 54 H02 No defect detected 0.75V DSI 40C 30 74 H02 1.6V RPW 1.73V DSI 40C 42 67 H06 1.66V RPW 1.76V DSI 40C E-6
Second SG Row Col Locn First Test Third Test Good Probe Cal Test H01 1.53V RPW 1.45V DSI 26C 15 3 H02 1.02V DSI PBC(0.93) 26C H03 0.75V DSI PBC(0.71) 26C H01 1.85V RPW 1.54V DSI 45C 20 28 1.02V DSI detected by H04 45C (imputed) RPC(1.08) 20 32 H01 2.1V RPW 1.79V DSI 45C 23 13 H01 1.64V RPW 1.56V RPW 1.62V DSI 45C 23 30 H01 1.51V RPW 1.33V DSI 45C 25 8 H01 1.71V RPW 1.61V RPW 1.69V DSI 45C H01 1.66V RPW 1.48V DSI 45C 26 28 H03 1.08V DSI 1.03V PBC 45C H01 1.8V RPW 1.53V DSI 45C 3 31 28 H04 1.14V DSI 0.99V PBC 45C H01 1.25V DSI 1.11V PBC 45C 31 32 H02 1.55V RPW 1.22V DSI 45C H01 1.55V RPW 1.24V DSI 45C 31 76 H02 0.97V DSI 0.79V PBC 45C H03 0.39V DSI PBC(0.43) 45C 32 28 H01 1.64V RPW 1.29V DSI 45C 43 36 H01 1.62V RPW 1.53V DSI 51C H01 1.75V RPW 1.32V DSI 45C 44 58 H02 0.55V DSI 0.35V PBC 45C H03 1.04V DSI 2.55V PBC 45C H01 0.74V DSI 1.46V PBC 45C 44 59 H02 2.15V RPW 2.07V DSI 45C SG Row Col Locn First Test Second Test Third Test Good Probe Cal 4 12 34 H02 2.72V RPW 2.78V RPW 2.77V DSI 53H H01 1.62V RPW 1.81V DSI 51H 13 48 H04 0.39V DSI 0.45V PBC 51H H01 1.54V RPW 1.46V DSI 50H 14 86 H02 0.72V DSI 1.05V PBC 50H H04 0.68V DSI 0.86V PBC 50H H01 1.65V RPW 1.75V RPW 1.82V DSI 53H 17 27 H03 0.34V DSI 0.22V PBC PBC(0.20) 53H H01 1.55V RPW 1.65V RPW 1.58V DSI 53H H02 0.45V DSI 0.55V PBC PBC(0.60) 53H 20 25 H06 0.35V DSI PBC PBC(0.32) 53H H07 0.23V DSI PBC PBC(0.23) 53H H01 1.96V RPW 2.03V RPW 2.02V DSI 53H 20 40 H05 0.51V DSI 0.6V PBC PBC(0.52) 53H 20 46 H01 0.75V DSI 0.75V PBC 51H E-7
SG Row Col Locn First Test Second Test Third Test Good Probe Cal H02 1.74V RPW 1.38V DSI 51H H03 0.57V DSI 0.54V PBC 51H H06 0.62V DSI 0.83V PBC 51H H01 1.67V RPW 1.76V RPW 1.98V DSI 53H 20 69 H02 0.31V DSI 0.25V PBC PBC(0.29) 53H H01 0.77V DSI 0.67V PBC 51H 22 47 H02 1.57V RPW 1.21V DSI 51H H01 1.85V RPW 2.12V DSI 51H 25 40 H02 0.75V DSI 0.82V PBC 51H 25 42 H01 1.66V RPW 1.6V RPW 1.53V DSI 53H 26 68 H01 1.75V RPW 1.6V DSI 51H H02 1.97V RPW 1.96V RPW 2V DSI 53H 27 73 H03 0.68V DSI 0.71V PBC PBC(0.52) 53H NRC Question 11 On page B-1 of your August 27, 2008 letter, you indicated that The leak rates were not significantly affected by including the uncertainties in the ANL [Argonne National Laboratory]
tearing model. Please clarify this sentence (e.g., was an assessment of the leak rate for this indication performed using the ANL model and were the results with and without uncertainties comparable).
TVA Response This statement refers to the calculations performed for evaluating the leakage from the pulled tube from SG No. 4 Row 22 Column 70 from SQN Unit 2 in 2006. Details have been included in EPRI Report NP-7480-L Addendum 7 2007 Database Update.
Steam line break (SLB) leak rate analyses were performed for the SG No. 4 Row 22 Column 70 01H measured crack depth profiles. The analysis method applied the leak rate methodology for the axial PWSCC alternate repair criteria (ARC) of WCAP-15128, Revision 2. This method calculates leakage from the depth profile, accounts for potential ligament tearing up to the SLB pressure differential and uses leak rate analysis methods correlated and adjusted to measurements. Nominal leak rate calculations were performed. Since the profiles were from destructive examination results, it was not necessary to apply uncertainties to the depth profile.
The methods applied the ANL ligament tearing model, as described in WCAP-15128, Revision 2, to calculate potential ligament tearing at SLB conditions. The leak rates were not significantly affected by including uncertainties (described in WCAP-15128, Revision 2) in the ANL tearing model. The ligament tearing pressure is calculated for all potential sub-lengths of the crack profile to obtain the longest crack length that would tear at SLB conditions.
NRC Question 12 In your operational assessment and condition monitoring assessment, 0.008 gallons per minute primary to secondary leakage was attributed to all other sources. Please clarify the nature of these sources (e.g., cracking in U-bend).
TVA Response E-8
This leakage value was included because of the possibility for other cold leg tube ends that had been previously plugged by rolled plugs and later de-plugged. The W* leakage value for a postulated crack greater than 12 inches below the top of tubesheet (0.00009 gal/ min) was multiplied by 94 (quantity of Row 1 tubes) to obtain 0.008 gal/min.
NRC Question 13 On page E2-2 of your August 27, 2008 letter, you indicated that primary water stress corrosion cracking indications at the top of the tubesheet were included in the condition monitoring W*
leakage evaluation regardless of whether or not they were above the bottom of the WEXTEX transition (BWT). You then indicate that: The location of upper crack tip was subtracted for the location of the BWT and then this value had the non-destructive examination (NDE) uncertainty subtracted. If the value was negative, it was then assumed to be zero. Please clarify these last two sentences.
TVA Response A more accurate statement of the criteria used is, The distance below the bottom of the BWT to the upper crack tip is determined and then the (NDE) uncertainty is subtracted. If it is determined the crack is above the BWT then the value is assumed to be zero, and included in the W* leakage calculation to ensure conservatism.
NRC Question 14 In the third paragraph on page E2-2 of your August 27, 2008 letter, you indicate that The leakage value for the bins was summed to obtain the total in the 8 inches to 12 inches below the top of tubesheet region. Should this sentence have indicated that this was the leakage for the 0 inch to 8 inch region?
TVA Response The sentence should have indicated that this was the leakage for the 0-inch to 8-inch region.
E-9