PNP 2015-079, Supplement to Relief Request Number RR 4-24 - Proposed Alternative, Use of Alternate ASME Code Case N-770-1 Baseline Examination

Supplement to Relief Request Number RR 4-24 - Proposed Alternative, Use of Alternate ASME Code Case N-770-1 Baseline Examination

ML15270A004
Person / Time
Site:	Palisades
Issue date:	09/27/2015
From:	Vitale A Entergy Nuclear Operations
To:	Document Control Desk, Office of Nuclear Reactor Regulation
References
PNP 2015-079, RR 4-24
Download: ML15270A004 (20)

Text

,.

• ===-Entergy PNP 2015-079 September 27, 2015 U. S. Nuclear Regulatory Commission ATTN: Document Control Desk Washington, DC 20555-0001 Entergy Nuclear Operations, Inc.

Palisades Nuclear Plant 27780 Blue Star Memorial Highway Covert, MI 49043-9530 Tel 269 764 2000 Anthony J. Vitale Site Vice President

SUBJECT:

Supplement to Relief Request Number RR 4 Proposed Alternative, Use of Alternate ASME Code Case N-770-1 Baseline Examination Palisades Nuclear Plant Docket 50-255 Renewed Facility Operating License No. DPR-20

REFERENCES:

1. Entergy Nuclear Operations, Inc. letter PNP 2015-076, Relief Request

Dear Sir or Madam:

Number RR 4 Proposed Alternative, Use of Alternate ASME Code Case N-770-1 Baseline Examination, dated September 26, 2015 In Reference 1, Entergy Nuclear Operations, Inc. (ENO) requested Nuclear Regulatory Commission (NRC) approval of the Request for Relief for a Proposed Alternative for the Palisades Nuclear Plant (PNP).

Reference 1 is associated with the use of an alternative to the requirements of the American SOCiety of Mechanical Engineers (ASME) Boiler and Pressure Vessel Code, Code Case N-770-1, as conditioned by 10 CFR 50.55a(g)(6)(ii)(F)(1) and 10 CFR 50.55a(g)(6)(ii)(F)(3).

At the time of submittal, PNP had not completed all of the eight subject weld examinations. This supplemental letter contains the results of the remaining weld examinations, additional hardship information, and additional cold leg weld crack growth information.

This submittal contains no proprietary information.

This submittal makes no new commitments or revisions to previous commitments.

I declare under penalty of perjury that the foregoing is true and correct. Executed on September 27, 2015.

Sincerely, I

AJV/jse

PNP 2015-079 Page 2 of 2

Enclosure:

Supplement to Entergy Nuclear Operations, Inc., Palisades Nuclear Plant, Relief Request Number RR 4-24 Proposed Alternative cc:

Administrator, Region III, USNRC Project Manager, Palisades, USNRC Resident Inspector, Palisades, USNRC

ENCLOSURE Supplement to Entergy Nuclear Operations, Inc., Palisades Nuclear Plant, Relief Request Number RR 4-24 Proposed Alternative Additional Cold Leg Crack Growth Analysis Information Structural Integrity Associates, Inc. (SI) analyzed postulated flaws in the cold leg nozzles based on the limited examination results obtained during the 1 R24 refueling outage. The analysis is provided in Attachment 2 and is summarized below. The information provided in the analysis supplements information previously provided in the Structural Integrity of these Regions section of Relief Request Number RR 4 Proposed Alternative, Use of Alternate ASME Code Case N-770-1 Baseline Examination, dated September 26,2015.

As discussed in the SI analysis, the eight cold leg nozzle dissimilar metal (OM) welds experience much longer primary water stress corrosion crack (PWSCC) initiation times, by a factor of four, than the hot leg nozzle welds. This is due to the cold leg nozzles being at much lower operating temperatures (46°F lower). Hence, it is more likely that the hot leg weld would develop cracks prior to the cold leg welds. Since the hot leg nozzle weld volume received 100%

examination coverage with no flaws identified, the probability of a crack existing in the remaining unexamined cold leg weld volume for the life of the plant is deemed to be very low.

It is not expected that an axial flaw will initiate. However, even if an axial flaw existed in the unexamined weld volume, this analysis determined that 64 years would be required for the flaw to go 75% though-wall. Additionally, axial flaws do not structurally challenge the weld and would only potentially lead to leakage.

Based on the amount of cold leg weld volume examined during the refueling outage, in which no flaws were identified, the SI analysis used a statistical approach to conservatively estimate a 30 degree maximum initial circumferential flaw size, and then evaluated this postulated flaw for PWSCC growth. The analysis determined that this postulated 30 degree flaw would take approximately 200 years to grow 75% through-wall. This is 3.6 times longer than the previously evaluated initial 360 degree circumferential flaw, in which about 55 years would be required for the flaw to go 75% through-wall.

Based on the fact that the hot leg weld was examined and no flaws were identified, there is a factor of four difference in crack initiation between the hot and cold legs justifying no cold leg crack initiation, axial crack growth duration is beyond the life of the plant, circumferential crack grow duration is greater than 200 years, it is highly unlikely that a flaw of structural significance exists in the uninspected portions of the cold leg nozzle welds, and any flaw that may exist will not challenge the str.uctural integrity of the piping over the life of the plant. Therefore, operating the plant for one more cycle, until a qualified examination technique for these welds can be developed, would not compromise the safety of the plant.

Additional Hardship Information Concerning Development of a New Examination Technique In order to obtain required scan coverage for the eight cold leg weld volumes that could not be examined due to the contour of the weld along the periphery of the branch connection nozzles and the four cold leg welds that could not be examined due to the concrete pipe whip restraint 1 of 2

obstructions, an examination technique would need to be developed and qualified over the next operating cycle. Addressing the weld contour issues would require interfacing with the Electric Power Research Institute (EPRI) to determine the number of additional samples needed to meet American Society of Mechanical Engineers (ASME)Section XI, Appendix VIII requirements, qualifying the procedure and personnel, designing and fabricating the mockups to match the weld contour of the installed branch connection nozzles, and designing additional inspection equipment or modifying the current encoded phased array equipment for qualification of the axial flaw scan. In addition to qualifying a technique to meet the current weld contour, the vendor would also be required to develop tooling for the encoded scanner that could inspect the cold leg welds in the vicinity of the concrete pipe whip restraints. This most likely would be supplemented with a manual ultrasonic technique requiring a procedure and personnel demonstration resulting in a lengthy qualification process. It is anticipated that approximately 18 months would be required to qualify procedures and personnel to reliably perform qualified examinations of the subject welds.

ATTACHMENTS

1.

Updated Weld Inspection Coverage Table

2.

Structural Integrity Associates, Inc. Report, "Evaluation of the Palisades Nuclear Plant Cold Leg Branch Nozzles for Limited Inspection Volume" 2 of 2

ATTACHMENT 1 Updated Weld Inspection Coverage Table Component ID Description! NPS N-no-1 Volume Coverage (%)

Circ Flaw Scan Axial Flaw Scan Exam Method and Limitations I Examination Summary UT Encoded Phased Array 2 inch Hot Leg Drain PCS-42-RCL-1 H-3/2 100%

100%

No limitations.

Nozzle No flaws identified.

UT Encoded Phased Array 2 inch Cold Leg PCS-30-RCL-1A-11/2 100%

0%

Weld contour limitation.

Charging Nozzle No flaws identified.

UT Encoded Phased Array 2 inch Cold Leg Drain PCS-30-RCL-1 A-5/2 50% (approx.)

0%

Weld contour limitation and concrete pipe whip Nozzle restraint obstruction.

No flaws identified.

3 inch Cold Leg UT Encoded Phased Array PCS-30-RCL-1 8-1 0/3 Pressurizer Spray 100%

0%

Nozzle Weld contour limitation.

No flaws identified.

2 inch Cold Leg Drain UT Encoded Phased Array PCS-30-RCL-18-5/2 Nozzle 50% (approx.)

0%

Weld contour limitation and concrete pipe whip restraint obstruction.

No flaws identified.

PCS-30-RCL-2A-11/2 2 inch Cold Leg 100%

0%

UT Encoded Phased Array Charging Nozzle Weld contour limitation.

No flaws identified.

3 inch Cold Leg UT Encoded Phased Array PCS-30-RCL-2A-11/3 Pressurizer Spray 100%

0%

Nozzle Weld contour limitation.

No flaws identified.

2 inch Cold Leg Drain UT Encoded Phased Array PCS-30-RCL -2A-5/2 Nozzle 50% (approx.)

0%

Weld contour limitation and concrete pipe whip restraint obstruction.

No flaws identified.

2 inch Cold Leg Drain UT Encoded Phased Array PCS-30-RCL 5/2 and Letdown Nozzle 50% (approx.)

0%

Weld contour limitation and concrete pipe whip restraint obstruction.

No flaws identified.

Page 1 of 1

ATTACHMENT 2 Structural Integrity Associates, Inc. Report "Evaluation of the Palisades Nuclear Plant Cold Leg Branch Nozzles for Limited Inspection Volume" 14 Pages Follow

S; Structurallntegrily Associates, Inc.-

September 26, 2015 Report No. 1400669.405.RO Quality Program: IZI Nuclear D Commercial Mr. Stephen Davis Entergy Nuclear Operations, Inc.

Palisades Nuclear Plant 27780 BIue Star Memorial Highway Covert, MI 49043 5215 Hellyer Ave.

Suite 210 San Jose, CA 95138-1025 Phone: 408-978-8200 Fax:

408-978-8964 www.structint.com clohse@structinlcom

Subject:

Evaluation of the Palisades Nuclear Plant Cold Leg Branch Nozzles for Limited Inspection Volume

Dear Steve:

This report addresses the limited coverage of the cold leg nozzle welds at Palisades Nuclear Power Plant (Palisades) during volumetric examination of these welds, as part of the on-going 1 R24 outage, to determine if lack of complete coverage has any adverse effect on the safety of the plant. Based on the evaluation presented below, it is concluded that the limited coverage during the examinations of these welds does not compromise the safety of the plant, and as such, it is acceptable to operate the plant for at least one additional cycle. The basis for this conclusion is provided in the following sections.

BACKGROUND As part of the activities performed during the ongoing Palisades refueling outage 1 R24, Entergy is volumetrically examining a total of nine (9) pressure retaining dissimilar metal welds (DMWs) containing Alloy 600 nozzles and Alloy 182 weld material. Of the nine weld locations, eight are cold leg nozzles (three 2-inch diameter drain nozzles, one 2-inch diameter letdown nozzle, two 2-inch diameter charging nozzles, and two 3-inch diameter spray nozzles) and one 2-inch diameter hot leg drain nozzle.

During the examinations, it was found that the eight cold leg nozzles would have limited examination coverage due to their weld contours and plant interferences. Due to the transition of the weld between the nozzle and the cold leg piping, the examination technique being used is unable to obtain any of the DMW inspection volume necessary for axial flaws on all eight cold leg nozzles. For circumferential flaws, plant obstructions prevent the examination of about 180

_________________ Toll-Free 877-474-7693 ______________ _

ChIcago,IL 877-474-7693 AIuon, OH 3JO.899.9753 Denver,CO 303-792-0077 AusUn, TX 512-533-9191 san DIego, CA 858-455-6350 CharIoIIe, Me 704-597-5554 5anJose,CA 408-978-8200 Challanooga, nI 423-553-1180 sta1e COllege, PA 814-954-mS Toronto, canada 905-829-9817

Mr. Stephen Davis Report No. 1400669.405.RO September 26, 2015 Page 2 of 14 degrees (50%) of the circumferential weld length for four out of the eight cold leg nozzles (specifically, the four drain nozzles). The other four cold leg nozzles obtained 100% of the examination volume necessary for examination for circumferential flaws. The hot leg drain nozzle received a volumetric examination of essentially 100% of the code required examination volume and the examination results were clean (no recordable indications). Therefore, relief will be required to accept the reduced examination coverage for the eight cold leg nozzles, to allow for one cycle, during which modifications can be made to address the coverage issues.

There are two objectives to this report. First is to address the lack of examination coverage for axial flaws on all eight cold leg nozzles. The second is to perform an evaluation of the reduced coverage for the four cold leg drain nozzles and assess the maximum circumferential flaw size that could be left in the uninspected region and perform a flaw evaluation to determine the life of this flaw.

TECHNICAL APPROACH The technical approach consists of the following steps:

Determine the relative susceptibility of the hot leg and cold leg nozzles to PWSCC.

Since the hot leg nozzle has been inspected with no indications recorded, comparison of the hot and cold leg nozzles can provide some useful insights in regards to the potential for cracking of the uninspected cold leg nozzle regions.

For axial flaws, determine the consequence if the whole weld was assumed to be cracked through-wall.

For circumferential flaws, use probabilistic analysis to determine the possible maximum (bounding) flaw size that could be present in the uninspectable region of the cold leg weld regions.

Model the bounding circumferential flaw in a finite element model to determine the stress intensity factors and determine the PWSCC life for the circumferential flaw. This duration is compared to the 360 degree circumferential flaw that was evaluated in SI Calculation 1400669.323 [4].

Discuss the effect of post-weld heat treatment on PWSCC susceptibility of the affected welds.

Discuss the effect of initiation time.

All evaluations in this report use input that was from flaw evaluations and relief request support previously performed for Palisades. These previous evaluations developed a finite element model of a bounding cold leg nozzle geometry [2] and then performed a residual stress evaluation of the branch line weld [3]. The residual stresses along with bounding operating loads were used to calculate stress intensity factors and PWSCC growth for a full 360 degree circumferential flaw [4]. A summary of the previous results is contained in Reference 1.

Mr. Stephen Davis Report No. 1400669.405.RO EVALUATION Effect of Temperature on PWSCC Susceptibility (Hot Leg vs. Cold Leg)

September 26, 2015 Page 3 of14 A total of nine welds were scheduled to be examined during this refueling outage. One of these welds operates at the hot leg temperature, while the other eight operate at the cold leg temperature. The Palisades hot leg temperature is 583°F and the cold leg temperature is 537°F [1]. The hot leg drain nozzle was successfully examined (100% coverage for both axial and circumferential flaws) and no recordable indications were found. PWSCC susceptibility is a function of operating time, temperature, and stress levels. The operating time for both the hot leg and cold leg nozzles is the same as the components are in the same plant. The stress levels are similar for both the hot leg and cold leg nozzles and are in the range of 40 to 45 ksi at the ID surface [1, Figures 7,8, 11, and 12]. The main difference is the operating temperature of the locations. As the cold leg nozzles operate at a much lower temperature (46°F less), the susceptibility to PWSCC is much less than that for the hot leg nozzles. As indicated earlier, the hot leg nozzle was already examined and no recordable indications were found. Therefore, it is less likely that any of the cold leg nozzles would have experienced any PWSCC cracking.

Reference 1 (Attachment A) previously evaluated initiation times for the hot leg and cold leg nozzles based on the temperature and stress levels. The difference between the hot leg and cold leg initiation times gives an indication of the relative susceptibility levels between the two different temperatures. The hot leg initiation time was calculated as -130 years and the cold leg initiation time was calculated as -600 years. Comparing these values provides a factor of about 4 between the cold leg initiation time and the hot leg initiation time. Since the Palisades hot leg nozzle has not experienced any cracking, the cold leg nozzles would not expect to experience cracking until it has operated four times as long as the hot leg nozzle. Based on this, the hot leg nozzle is considered a good leading indicator for when cracking would occur in the cold leg nozzles. So if the hot leg nozzle is not cracked, it is highly unlikely that the cold leg nozzles would be cracked. This is further supported by the fact that four of the cold leg nozzles have been fully examined (looking for circumferential flaws) with no recordable indications observed.

Fifty percentage coverage (looking for circumferential flaws) of the remaining four nozzles was also obtained and no recordable indications were observed.

Evaluation of Axial Flaws As mentioned above, all eight cold leg nozzles received essentially zero percent examination coverage of the DMW volume necessary for axial flaws. Therefore, it is unknown if there is an axial flaw contained within the DMW region that is less than through-wall. However, axial flaws for this configuration do not cause gross failure of the structure and only lead to leakage.

SI performed an evaluation of the hot leg drain line nozzle for a through-wall axial flaw in Reference 5 in support ofRR 4-18. It shows that if the entire weld were assumed cracked through-wall, the component is still stable (passes a limit load evaluation). Figure 1 contains the axial flaw that was modeled in Reference 5. As noted in Reference 4, the time for an axial flaw to grow from a small flaw (0.025 inches) to 75% of the wall thickness by PWSCC is 64 yrs.

This value is much greater than the expected effective full power years (EFPY) for the Palisades plant at the next refueling outage (lR25) which is 28.8 EFPY.

Mr. Stephen Davis Report No. 1400669.405.RO September 26, 2015 Page 4 of 14 Essentially 100% coverage for axial flaws was obtained for the hot leg drain line nozzle and no recordable indications were observed. As discussed above, the hot leg nozzle acts as a leading indicator for the cold leg nozzles as the higher temperature makes it more susceptible to PWSCC. Since the hot leg nozzle showed no indication of axial cracking, it is very unlikely that the cold leg nozzles would contain axial flaws.

Evaluation of Circumferential Flaws As discussed above, there is limited examination coverage for circumferential flaws for the four cold leg drain nozzles. The other four cold leg nozzles successfully received 100% of the required volume inspections for circumferential flaws. Based on the examined weld length, an estimate of the largest flaw size in the uninspectable region can be estimated using probabilistic insights.

The estimate of the largest flaw size in the uninspected region is given by p= rln where p = estimate of the true proportion r = number of items with the specified characteristic (this would be the number of flawed components) n = sample size (number of items inspected - for this situation, we assume that the sample size is a portion of the circumferential weld length in degrees)

The basis of the above estimate is the Binomial distribution. The following assumptions are applied for the estimate to be treated as valid:

1: The number of observations, n, is fixed (Le. we have inspected a number of degrees of weld already).

2: Each observation is independent.

3: Each observation represents one oftwo outcomes ("success" or "failure").

4: The probability of "failure," p, is the same for each outcome.

5: The population is much larger than the sample size The limitation of the above estimate is that it will vary if another sample of size "n" is taken.

Rather than using the best single estimate of true proportion, an interval estimate is used to account for the sampling error. For this problem, a one-sided interval, with a desired confidence level, is employed. This then allows for a statement such as:

If you detected 1 failure in a sample of size 50, one can state with 95% confidence that the actual proportion offailures in the population will he less than 0.091, or less than 0.126 with 99% confidence.

The confidence interval limits for the proportion are tabulated in Reference 9 and a sample table is shown below.

Mr. Stephen Davis Report No. 1400669.405.RO September 26, 2015 Page 5 of 14 Table 1: One-Sided Confidence Intervals at 90%, 95%, and 99% Confidence [9]

r 0.90 0.95 0.99 0.90 0.95 0.99 0.90 0.95 0.99 0=29 0=50 0= 100 0

0.076 0.098 0.147 0.045 0.058 0.088 0.023 0.030 0.045 1

0.128 0.153 0.208 0.076 0.091 0.126 0.038 0.047 0.065 2

0.173 0.202 0.260 0.103 0.121 0.158 0.052 0.062 0.082 3

0.216 0.246 0.307 0.129 0.148 0.187 0.066 0.076 0.097 4

0.257 0.288 0.351 0.154 0.174 0.215 0.078 0.089 0.112 5

0.297 0.329 0.392 0.178 0.199 0.242 0.091 0.102 0.126 6

0.335 0.368 0.432 0.201 0.223 0.267 0.103 0.115 0.140 7

0.373 0.406 0.470 0.224 0.247 0.292 0.115 0.128 0.153 8

0.409 0.443 0.507 0.247 0.270 0.316 0.127 0.140 0.166 9

0.445 0.479 0.542 0.269 0.293 0.340 0.138 0.152 0.179 10 0.481 0.514 0.577 0.291 0.316 0.363 0.150 0.164 0.191 In applying this to the cold leg welds, the sample size needs to be determined. Eight cold leg welds are being examined. The four cold leg drain line nozzles received examination of half the weld circumference, while looking for circumferential flaws. The other four cold leg nozzles received examination of 100% of the weld circumference, while looking for circumferential flaws. Based on the total examination population, 75% ((100*4 + 50*4)/8) of the total circumferential weld length is examined. The sample size of the population has an influence on the proportion of flawed samples. Therefore, for the calculation, a number of different inspection unit sizes are used to check the dependence of the sample size. The sample size is taken as the inspection unit size, as measured in degrees. For the calculation, inspection sizes of 1, 2, 5, and 10 degrees are assumed in the sample sizes. Similar approaches have been taken in the industry for determination of maximum flaw sizes that may be present in uninspectable areas ofBWR shrouds [10, Appendix H].

p = estimate of the true proportion - this is the unknown proportion of flawed inspection units r = number of items with the specified characteristic - each segment cracked = 0 (as no flaws have been found in the examined portions of the welds) n = sample size (number of items inspected) - 2160 (360 x 8 x 0.75) (for the one degree inspection unit size)

As shown in the table below, for zero segments cracked in a sample size of2160 degrees inspected, one can state with 95% confidence that the actual proportion of failures in the population will be less than 0.14%. For a single circumferential weld (360 degrees), 0.14% is equivalent to 0.5 degree segment cracked. The table below summarizes calculations performed assuming various inspection unit sizes and confidence levels. As shown below, assuming a 10 degree inspection unit gives the largest flawed segment of 7.5 degrees at 99% confidence. Given the information below a bounding flaw is modeled in the ANSYS finite element model and stress intensity factors and PWSCC crack growth life determined for the circumferential flaw.

I)lInIctrnllDtegrlty Associates, Inc.-

Mr. Stephen Davis Report No. 1400669.405.RO Inspection Unit Size Binomial Sample Size (degrees) 1 2160 2

1080 5

432 10 216 Crack Growth Evaluations Segment Cracked in 3600 Weld, with 95% Confidence (degrees) 0.5 1

2.5 5

September 26, 2015 Page 6 of 14 Segment Cracked in 3600 Weld, with 99% Confidence (degrees) 0.8 1.5 3.8 7.5 As shown above, a 7.5 degree flaw is the largest estimated flaw based on the 10 degree inspection unit size. For conservatism, a constant length flaw for a 30 degree arc is used in the crack growth evaluation. As a defined flaw length is used, two flaws will be assumed. One will be parallel to the axis of the cold leg piping and the other perpendicular to the axis of the cold leg piping. SI previously performed a residual stress evaluation for the cold leg welds in support of contingency flaw evaluations in Reference 3. The residual stresses were used along with operating loads to determine stress intensity factors and the PWSCC growth due to a 360 degree circumferential flaw performed in Reference 4. The same loadings and model from Reference 4 are used in this evaluation to determine the life for the 30 degree length flaws. The modeled flaws are shown in Figure 2. Since the model is a 90 degree (quarter symmetry) portion of the nozzle, only half the flaw is modeled as the flaws are centered at the 0 and 90 degree planes. A 30 degree flaw corresponds to an arc length of ~ 1.6 inches (30*1t1180*3.1 ") at the outside diameter of the nozzle (inner diameter of the weld). This means that the flaw aspect ratio is ~ 1: 1 (length to depth) at 50% through-wall. The ratio will decrease for a deeper flaw and increase for a more shallow flaw.

Stress intensity factors (Ks) at five depths for both 30° inside surface connected, part-through wall circumferential flaws, at the 0- and 90-degree azimuthal locations of the nozzle, were calculated using finite element analysis techniques. The maximum K values from either circumferential flaw at each flaw depth were extracted and used as input for performing the PWSCC growth analyses.

For the crack growth analyses, a small initial flaw size of 0.025 inch was chosen and is consistent with previous evaluations of this nozzle. The final flaw size for this analysis is 75%

of the wall thickness. This final depth is chosen as it is the maximum allowable flaw depth per Section XI ofthe ASME Code for pipe flaw evaluations. The 75% requirement is consistent with the tables and equations in Appendix C, Article C-5000 of Section XI of the ASME Code.

The equations and tables are limited to 75% of the pipe thickness. The values for a through-wall flaw (95% depth) are also calculated for comparison.

Mr. Stephen Davis Report No. 1400669.405.RO The key parameters used in the crack growth calculations included:

Initial flaw depth = 0.025" Temperature = 537°F (cold leg)

Wall thickness = 3" (cold leg pipe thickness)

September 26, 2015 Page 7 of 14 Stress intensity factors, Ks, were calculated for the two circumferential flaws. The stress intensity factors were calculated using the stress distributions for residual stress plus normal operating conditions shown in Figure 3 (from Reference 1) for the bounding cold leg nozzle. In addition, a far field in-plane bending moment is applied to the free end of the cold leg run piping to account for piping moments in the main loop piping. This combined loading is used for the determination of the stress intensity factors for both circumferential flaws. Figure 4 shows the maximum calculated stress intensity factors for the two circumferential flaws, respectively, as a function of through-wall depth. Figure 4 also shows the comparison between the 360 degree flaw and 30 degree flaw.

Crack growth evaluations were performed for the 30 degree circumferential flaws and these results were compared to the 360 degree flaw growth calculation performed in Reference 4.

Figure 5 shows that the time for an initial 0.025" deep flaw to grow to 75% through wall is ~200 years for the 30 degree flaw and ~55 years for the 360 degree flaw. Comparing the 95% flaw depths, the 30 degree flaw takes ~230 years and the 360 degree flaw takes ~65 years. The 30 degree flaw takes 3.6 times and 3.5 times longer to reach a depth of 75% and 95%, respectively.

These comparisons show that a limited circumferential extent reduces the PWSCC crack growth rate considerably. The reduction in rate leads to a considerable increase in the calculated growth time for the circumferential flaw. These crack growth evaluations were performed using the PWSCC growth law for Alloy 182 from MRP-115. These values assume that a flaw exists and do not credit any time for initiation.

Potential Additional Margin Due to PWHT and Crack Initiation Dominion Engineering documented in Reference 6 that industry papers [7, 8] note a benefit in fatigue crack growth rate when Alloy 182 weld metal underwent PWHT. This benefit ranged from a factor of two to four. There was a recommendation that crack growth rates be reduced by a factor of two based on this data. As discussed in Reference 3, the cold leg nozzles received a PWHT with the cold leg piping spool piece. The crack growth values calculated do not include any beneficial factor for potential reduction in crack growth rate. If such a factor were used for the evaluation ofthe circumferential crack at Palisades, it would increase the crack growth times previously discussed. This increase would be applicable to both the 360 degree and 30 degree flaws, so as the total time increases, the ratio between the two durations would remain the same.

If a factor of two were considered, it would push the time for an initial flaw to grow to 75% to well beyond 100 years for the 360 degree flaw and beyond 400 years for the 30 degree flaw.

Attachment A of Reference 1 included a calculation of the initiation time for both the hot leg and cold leg nozzles using one of the models in the xLPR program that was developed by EPRI. The calculation of initiation time based on the Palisades results shows that the time to initiation for the hot leg drain nozzle is approximately 130 years and the cold leg nozzles exceed 600 years.

The duration for the cold leg nozzles is 4 times longer than the hot leg nozzle. These values are

Mr. Stephen Davis Report No. 1400669.405.RO September 26, 2015 Page 8 of 14 not included in the fracture mechanics evaluation above, but if any time for initiation were included, the total time for the crack growth life would be much longer.

CONCLUSIONS Based on the assessments and calculations documented herein, the cold leg nozzles experience much slower crack growth rates due to PWSCC than the hot leg nozzle. Based on the amount of cold leg weld examined so far, an estimate of maximum flaw size was determined and this flaw was evaluated for PWSCC growth. The 30 degree flaw takes 3.6 times longer than the 360 degree flaw to reach a depth of75% through-wall. Based on this time difference, the difference in initiation potential, and the fact that the hot leg was examined and no indications were found, it is unlikely that a flaw of structural significance exists in uninspected portions of the Palisades cold leg nozzles and any flaw that may exist will not challenge the structural integrity of the piping over the next operating cycle. As previously discussed, axial flaws do not structurally challenge the weld and would only lead to leakage. Based on this information, operating for one more cycle, until improved examination of these welds could be developed, would not compromise the safety of the plant.

REFERENCES

1. Structural Integrity Associates Report No. 1400669.401, Rev. 0, "Evaluation of the Palisades Nuclear Plant Branch Line Nozzles for Primary Water Stress Corrosion Cracking."

2. Structural Integrity Associates Calculation No. 1400669.320, Rev. 0, "Finite Element Model Development for the Cold Leg Drain, Spray, and Charging Nozzles."

3. Structural Integrity Associates Calculation No. 1400669.322, Rev. 0, "Cold Leg Bounding Nozzle Weld Residual Stress Analysis."

4. Structural Integrity Associates Calculation No. 1400669.323, Rev. 0, "Crack Growth Analysis of the Cold Leg Bounding Nozzle."

5. Structural Integrity Associates Calculation No. 1200895.308, Rev. 1, "Hot Leg Drain Nozzle Limit Load Analyses for Flawed Nozzle-to-Hot Leg Weld."

6. Letter from G. White (Dominion Engineering, Inc.) to W. Sims (Entergy), "Effect of Post-Weld Heat Treatment Applied to Alloy 82/182 Full-Penetration Branch Pipe Connection Welds at Palisades," DEI Letter L-4199-00-01, Rev. 0, dated February 25, 2014.

7. T. Cassagne, D. Caron, J. Daret, and Y. Lefevre, "Stress Corrosion Crack Growth Rate Measurements in Alloys 600 and 182 in Primary Water Loops under Constant Load,"

Proceedings of9'hInternational Symposium on Environmental Degradation of Materials in Nuclear Systems-Water Reactors, The Minerals, Metals & Materials Society, Warrendale, PA, 1999, pp. 217-224.

8. S. Le Hong, 1. M. Boursier, C. Amzallag, and J. Daret, "Measurements of Stress Corrosion Cracking Growth Rates in Weld Alloy 182 in Primary Water ofPWR," Proceedings of 10'/'

International Conference on Environmental Degradation of Materials in Nuclear Power Systems-Water Reactors, NACE International, 2002.

9. M. G. Natrella, Experimental Statistics, National Bureau of Standards, NBS Handbook 91, 1963.

10. BWRVIP-76-A: BWR Vessel and Internals Project, BWR Core Shroud Inspection and Flaw Evaluation Guidelines. EPRI, Palo Alto, CA: 2009. 1019057.

Mr. Stephen Davis Report No. 1400669.405.RO Prepared by:

~

Chris S. Lohse, P.E.

Senior Consultant Verified by:

1Unf ~

Minji Fong Senior Engineer Approved by:

Rich Bax Associate cc:

R. Bax N.Eng D. Dedhia F. Ku M. Fong N.Cofie 9/26115 Date 9/26115 Date 9/26/15 Date

~ i&-

Francis Ku Associate Senior Associate September 26, 2015 Page 9 of 14 9/26/15 Date 9/26/15 Date

Mr. Stephen Davis Report No. 1400669.405.RO o

CP Symmetry Boundary Conditions Drain Nozzle Axial Couples September 26, 2015 Page 10 of 14 Removed Symmetry Boundary Conditions to Simulate Flavv Figure 1. 100% Through-Wall Axial Flaw Finite Element Model and Boundary Conditions (Figure taken from Reference 5, Figure 5)

Mr. Stephen Davis Report No. 1400669.405.RO September 26, 2015 Page 11 ofl4 Figure 2. Transferred Residual Stress + NOC + Pressure Stress for 30 Degree Circumferential Flaws (Only the 95%flaw depth is shown/or example)

Note: The figure shows a crack for the full 90 degree arc. The crack face nodes in the middle of the arc (uncracked region) are coupled in all degrees of freedom.

Mr. Stephen Davis Report No. 1400669.405.RO September 26, 2015 Page 12 of 14 tnlAL ~rrICN S'lEP-1413 SUB -3

'1'IM!!P2106 SK (AlA:>>

RSYS&5

[H{ -.13462

~ -lO. 5442 SM){ -43.7349 Figure 3. Radial Stresses at Operating Conditions for the Bounding Cold Leg Nozzle Note: Radial stresses are shown in the nozzle axis radial direction and units are in ksi.

(Figure taken from Reference 1, Figure 11) e Blnlt:fnl """,Ity Associates, Inc.-

Mr. Stephen Davis Report No. 1400669.405.RO I

I 60 50

~40 III ci

< c:

I'

]

30

~

E 20 10 o

0.000 0.200 0.400 0.600 Depth (aft) 0.800 1.000 1.200 September 26, 2015 Page 13 of14 Original 360 deg

-"-30deg 1

I Figure 4. Stress Intensity Factors as a Function of Depth for 30 Degree Circumferential Flaws

Mr. Stephen Davis Report No. 1400669.405.RO I

1.00 0.90 I

0.80 0.70

~ 0.60 or:...

~ 0.50 c

=:

J3! 0.40 u..

0.30 0.20 0.10 0.00 0

50 100 150 Time (yrs)

-- eire. Original (360 deg) -- eire. 30 deg September 26, 2015 Page 14 of 14 200 250 Figure 5. PWSCC Growth Time vs. Depth 30 Degree and 360 Degree Circumferential Flaws (Initial Flaw Size 0/0.025" Used)