NL-17-1443, License Amendment Request to Revise Technical Specification Section 5.5.17 Containment Leakage Rate Testing Program Final Response to NRC Requests for Additional Information
| ML17285B308 | |
| Person / Time | |
|---|---|
| Site: | Farley  |
| Issue date: | 10/12/2017 |
| From: | Wheat J Southern Nuclear Operating Co |
| To: | Document Control Desk, Office of Nuclear Reactor Regulation |
| References | |
| NL-17-1443 | |
| Download: ML17285B308 (11) | |
Text
.t. Southern Nuclear October 12, 2017 Docket Nos.: 50-348 50-364 U.S. Nuclear Regulatory Commission A TIN: Document Control Desk Washington, D. C. 20555-0001 Justin T. Wheat Nuclear Licensing Manager Joseph M. Farley Nuclear Plant 40 Inverness Center Parkway Post Office Box 1295 Birmingham, AL 35242 205 992 5998 tel 205 992 7601 fax jtwheat@ southernco.com NL-17-1443 License Amendment Request to Revise Technical Specification Section 5.5.17 "Containment Leakage Rate Testing Program" Final Response to NRC Requests for Additional Information Ladies and Gentlemen:
On November 15, 2016, Southern Nuclear Operating Company (SNC) submitted a license amendment request (LAR) to revise Joseph M. Farley Nuclear Plant, Unit 1 and Unit 2, Technical Specifications {TS) 5.5.17, "Containment Leakage Rate Testing Program." On March 15, 2017, the Nuclear Regulatory Commission (NRC) staff, upon a determination that additional information was needed to complete its review, issued a request for additional information (RAI) letter. On June 22, 2017, SNC provided a partial response to the NRC's March 15, 20171etter.
On August 3, 2017, the NRC requested supplemental information pertaining to RAI No.6, 5(b) and 5(c) in an email from Shawn Williams to Ken McElroy. On September 11, 2017, SNC provided a response to the NRC's supplemental request for information pertaining to RAI No. 6 and stated that the responses to the remaining RAis would be provided by October 12, 2017.
Enclosed are SNC answers to RAI 5 and RAI 7 which incorporate the remaining NRC's requests both in the original RAis and the supplemental RAis requested by email.
This letter contains no NRC commitments. If you have any questions, please contact Ken McElroy at 205.992.7369.
I declare under penalty of perjury that the foregoing is true and correct.
Executed on October 12, 2017.
~~
~heat Nuclear Licensing Manager Southern Nuclear Operating Company JTW/efb/cg
Enclosure:
- 1. Responses to NRC Request for Additional Information
U.S. Nuclear Regulatory Commission NL-17-1443 Page2 cc:
NRC Regional Administrator, Region II NRC NRR Project Manager-Farley NRC Senior Resident Inspector-Farley SNC Records RTYPE: CFA04.054
Joseph M. Farley Nuclear Plant License Amendment Request to Revise Technical Specification Section 5.5.17 "Containment Leakage Rate Testing Program" Responses to NRC Requests for Additional Information to NL-17-1443 SNC Response to NRC RAis NRC RAI5 In Section 6.3 "Potential Impact from External Events Contribution", Table 6-2 provides core damage frequency (CDF) and large early release frequency (LEAF) values for Fire Events from the FNP Fire PRA (FPRA) that credits pending modifications for NFPA 805 that will be implemented by the end of 2017. State if the fire CDFs (and LEAFs) reflect changes to FPRA methods made since the safety evaluation (SE) was issued for NFPA 805 including the following:
- a. New guidance on the credit taken for very early warning fire detection system (VEWFDS) is available in NUREG-2180, "Determining the Effectiveness, Limitations, and Operator Response for Very Early Warning Fire Detection Systems in Nuclear Facilities, (DeloresVEWFIRE)" which is now available as a final report at Agencywide Documents Access and Management System (ADAMS) Accession No. ML16343A058.
The methodology in NUREG-2180 is acceptable to the NRC because it is currently the best available guidance. The guidance provided in frequently asked question (FAQ) 08-0046, "Closure of National Fire Protection Association 805 Frequently Asked Question 08-0046 Incipient Fire Detection Systems" (ADAMS Accession No. ML16253A111 ), has been retired and alternative approaches for staff evaluation are necessary to complete the safety evaluation.
If the Fire PRA credits the future (or current) installation of VEWFDS, explain how credit (e.g., approach, methods, data, and assumptions) taken for the proposed VEWFDS is consistent with the guidance in NUREG-2180 or bounds the metrics in this that would be obtained had the guidance in NUREG-2180 been applied. If credit taken for VEWFDS in the FPRA is not consistent with or bounded by NUREG-2180 provide:
- 1) The risk metrics that would be obtained had the guidance of NUREG-2180 been applied, or that would be obtained had an alternative method been used, along with a description and justification for the alternative method.
Development and use of an alternative proposal may extend the time required to complete the review. The new risk results can be generated from a sensitivity study type evaluation insofar as formal incorporation of the new method into the PRA model of record is not required.
- 2) Explain how any increases in the risk metrics are consistent with the acceptance criteria for this application.
- b. Changes (generally increases) in fire ignition frequencies and non-suppression probabilities from NUREG-2169. For example, the frequency of fires in the Main Control Board is now twice as high as in the original NUREG/CR-6850 and six times higher than in its Supplement 1. The mean time to suppress a welding fire is nearly twice as long as in both the original NUREG/CR-6850 and its Supplement 1.
- c. Possible increases in spurious operation probabilities from NUREG/CR-7150, Vol. 2. For example, intra-cable spurious operation for an ungrounded AC, with individual control power transformers, single-break control circuit for a solenoid-operated valve has a probability of 0.64, slightly higher than the value of 0.6 from NUREG/CR-6850.
Page E1-1 to NL-17-1443 SNC Response to NRC RAis SNC Response to RAI 5:
Incorporation of NUREG-2180 The Farley Fire PRA credits the use of in cabinet incipient detection. This credit was based on the guidance provided in FAQ 08-0046. Since the issuance of the model of record, new guidance has been published and distributed in NUREG-2180. FAQ 08-0046 has been retired and is no longer acceptable for use in Fire PRAs. In general, the NUREG-2180 results in a decreased credit for incipient detection from the FAQ. However, based on the configuration of the ignition source and the adjacent targets, there may be little to no change. This is mainly a function of the distance between the source and target.
Incipient detection was credited in the following PAUs: 0254, 0318, 0416, 0471, 2202, 2254, and 2318. In all cases, the credit was for in cabinet detection.
To incorporate the new guidance, the event tree spreadsheet found in NUREG-2180 was used to determine what the appropriate credit should be. The spreadsheet requires multiple inputs, one of which is time to damage. The mean suppression rate is also a required input, and based on part b of the RAI, the mean suppression rate from NUREG-2169 was used in this evaluation.
For each ignition source that credits the incipient detection system, an evaluation was performed to calculate the new non-suppression probability based on the specific attributes of the scenario.
Due to the increase in risk associated with these ignition sources, additional refinements were required. These are discussed in more detail below. Also, see the Summary of Risk Metrics at the conclusion of this RAI response.
Incorporation of NUREG-2169 Similar to the credit for incipient detection, the Farley Fire PRA used fire ignition frequency data from NUREG/CR-6850 Supplement 1. At the time, this was the most up to date data available for use in the Fire PRAs. Since then, NUREG-2169 has been published and is now considered to be the most up to date data endorsed by the NRC.
In addition to the ignition frequencies changing, the mean suppression rates were also updated.
While there were not significant changes to risk significant fire types. There were sufficient changes that suggest updates should be performed. Therefore, the new mean suppression rates were also updated as part of this evaluation.
The results of this task are included in the Summary of Risk Metrics at the conclusion of this RAI response.
Incorporation of NUREG/CR-7150 The Farley Fire PRA takes credit for these hot shorts and includes them in the evaluation.
NUREG/CR-7150 provides updates to the hot short probabilities that are used and vary depending on the specific component type and method of control (AC vs. DC). The results of this task are included in the Summary of Risk Metrics at the conclusion of this RAI response.
Page E1-2 to NL-17-1443 SNC Response to NRC RAis Additional Refinements Incorporated In general, the incorporation of the above three items results in an overall risk increase for the Fire PRA. To help offset some of the increase additional refinements were incorporated to help reduce the net result. There were three main refinements used as part of this evaluation:
detailed fire modeling, incorporation of NUREG-2178, and data update from the full power internal events model.
Detailed Fire Modeling The detailed fire modeling was performed on those ignition sources that had credit for in cabinet incipient detection. The detailed fire modeling was completed in concurrence with the NUREG-2180 task. This was in part to get the times to damage for use in the NUREG-2180 event tree spreadsheet. This resulted in an overall decrease in risk based on increased time to damage for external targets in addition to the increased time to reach hot gas layer formation. Part of the detailed fire modeling inputs involved the second refinement as well, incorporation of NUREG-2178.
Incorporation of NUREG-2178 Prior to the issuance of NUREG-2178, all heat release rate data came from NUREG/CR-6850.
Additional testing was performed and NUREG-2178 was published. This provided updated alpha and beta terms and updated 981h percentile values for Bin 15 components. Bin 15 is typically the risk driver for Fire PRAs. For plants like Farley using Thermoset cable, NUREG-2178 results in a significant decrease in risk due to the updated data.
Not only did many of the 981h percentile heat release rates decrease, but the distribution improved. This means more of the fires occurred in the lower heat release rates. Therefore, if there was sufficient distance between the target and the ignition source, the likelihood that the target would be damaged decreases significantly.
In addition to the data updates, NUREG-2178 also provided additional means to distinguish between the types of ignition sources within Bin 15. For example, there are now clear definitions for different 'types' of electrical cabinets (battery chargers, switchgear, electrical enclosures).
There is also a differentiation for those enclosures based on overall volume and those that may be lightly loaded in terms of cables or combustible material. It should be noted that for this evaluation, the default loading was used for all cases.
Incorporation of Internal Events Data Update Since the development of the Fire PRA, the full power internal events model has undergone an update in which the underlying failure rate data was updated. This typically results in decreased failure rates; however, some did increase based on the plant specific events. Since the Fire PRA logic model is built on top of the full power internal events, the same random failure data is used. This evaluation took the updated numbers and applied them to the analysis. The results shown below include the use of this new failure data.
Page E1-3 to NL-17-1443 SNC Response to NRC RAis In general, these refinements resulted in overall decrease in risk and helped offset some of the increase associated with incorporating NUREG-2180, NUREG-2169, and NUREG/CR-7150.
The Summary of Risk Metrics below include these refinements.
Summary of Risk Metrics A summary of the risk metrics for the ILRT extension application, when including the incorporation of the new guidance and data described above, are shown in the table below.
The table shows the base number and the results of the evaluation for the ILRT extension based on RAI #5. The results for the three sub parts are aggregated together in one evaluation.
Overall there was a slight increase in the results for the Fire PRA for each of the end states. The one exception to this is the U1 - LERF result which showed a very slight decrease. The reason for the decrease is reasonable in that LERF was not impacted by the increase in the credit for incipient detection, and any increase introduced as a result of NUREG-2169 was offset by the incorporation of NUREG-2178.
Fire PRA Results Resulting from RAI #5 U1-CDF U2-CDF U1-LERF U2-LEAF BASE MODEL 6.61 E-05 7.33E-05 4.83E-06 7.11 E-06 RAI#5 8.71E-05 8.72E-05 3.30E-06 7.38E-06 NRC RAI7 In Section 6.3 "Potential Impact from External Events Contribution", Table 6-2 provides CDF and LERF values for Seismic Events from the FNP FPRA. As cited in theSE for NFPA 805 transition (ADAMS Accession No. ML14308A048), Farley used an average of the CDF values (1.73E-5/yr per unit) from the "Safety/Risk Assessment Results for Gl-199" (ADAMS Accession No. ML100270582).
NRC Staff results show that if using 1.73E-5/yr per unit, the CDF totals on Table 6-2 would calculate to 1.03E-4/yr and 1.08E-4/yr, respectively. Both would minimally exceed the RG 1.17 4 Region II threshold of 1.00E-4/yr. NRC Staff notes that an increase in each LERF would also occur based on the seismic LERFs of 2.02E-7/yr and 2.60E-7/yr for Units 1 and 2, respectively, as cited in the NFPA-805 SE.
- 1) Perform a complete recalculation of Table 6-2 and subsequent calculations in Sections 6 and 7 using the values cited in the NFPA-805 SE and address all the issues identified in the preceding RAis (RAI No. 1, 3, 5, and 6).
- 2) Confirm that any increases in the risk metrics as a result of the recalculation in part 1) above does not change the justification for exceeding the acceptance criteria for this application.
With respect to Table 6-2, the application stated, "the value for Total Internal and External events CDF slightly exceeds a value of 1.0E-04[/yr]. This value is expected to fall below 1.0E-Page E1-4 to NL-17-1443 SNC Response to NRC RAis 04[/yr] when the Farley Internal Events PRAs for Unit 1 and Unit 2 credit the Generation Ill RCP shutdown seals which are already installed. Crediting the Generation Ill RCP seals is expected to reduce Internal Events CDF to the mid 1.0E-06/yr range on both units."
Justify the expectation of a reduction in internal events CDF after crediting the Generation Ill RCP seals to the mid 1.0E-6/yr range on both units by using a bounding quantification.
SNC Response to AAI 7:
In response to RAI 7, SNC has updated Section 6.3 "Potential Impact from External Events Contribution" and Section 7 "Conclusions" as follows.
Potential Impact from External Events Contribution SNC has performed model updates and has recalculated Table 6-2 which lists the Farley CDF and LEAF values for each internal and external event type that are used to determine the potential impact from the External Events contribution. The values for the Internal Events PRA come from the latest model of record (Revision 10 Version 1) for each unit which includes credit for the Generation Ill RCP shutdown seals. The values for the Fire Events PRA come from the response to RAI 5 above. The Seismic values come from an update of those approved in the NFPA 805 Transition license amendment (ML14308A048), incorporating the Seismic Hazard Reevaluation and Screening for Risk Evaluation report provided to the NRC previously in NL-14-0342 in March 2014. The values for a new initiating event, Loss of SW Pond Dam, come from the calculation of impact on CDF and LEAF due to loss of the Service Water pond dam with credit given for the River Water System - these values were submitted to the NRC on September 11, 2017 in response to RAI 6. The recalculated Table 6-2 follows:
Table 6-2: Farley Internal and External Events Summary Event Type Farley Unit 1 Farley Unit 2 CDF (per/year)
LEAF (per/year)
CDF (per/year)
LEAF (per/year)
Internal Events 8.36E-06 8.99E-08 8.25E-06 7.63E-08 Fire Events 8.71 E-05 3.30E-06 8.72E-05 7.38E-06 Seismic 4.51 E-06 2.07E-07 4.51 E-06 2.07E-07 Loss of SW Dam 3.49E-07 1.32E-09 3.49E-07 1.32E-09 Other External Risk Screened out Screened out Screened out Screened out Total 1.00E-04 3.60E-06 1.00E-04 7.66E-06 Combining the External Events CDF values and the Internal Events CDF yields a CDF estimate of 1.00E-04/yr (Unit 1) and 1.00E-04/yr (Unit 2). LERF estimates including External Events are 3.60E-06/yr (Unit 1) and 7.66E-06/yr (Unit 2). Total Internal and External events CDF does not exceed 1 E-04.
The change in LEAF from extending the Type A test interval can be conservatively estimated using the total CDF values to determine the external event contribution. These CDF values were specifically used to determine the Class 3b frequency (neglecting corrosion) including the external events contribution. (Corrosion effects are not explicitly considered in the sensitivity assessment as the impact is negligible.) The factors for determining the increase in the non-detection probability of a leak described in Section 4.3 (of the LAR) were applied to the Class 3b Page E1-5 to NL-17-1443 SNC Response to NRC RAis base value frequencies to determine the 3b frequencies for the once per ten year test and once per fifteen year test for each unit.
Class 3b Frequency (three per ten year test) = 0.0023 * (CDF - LEAF)
Class 3b Frequency (Unit 1) = 0.0023 * {1.00E-04/yr-3.60E-06/yr) = 2.22E-07/yr Class 3b Frequency (Unit 1) (once per ten year test) = 3.33
- 2.22E-07/yr = 7.41 E-07/yr Class 3b Frequency (Unit 1) (once per fifteen year test) = 5.00
- 2.22E-07/yr = 1.11 E-06/yr Class 3b Frequency (Unit 2} = 0.0023 * {1.00E-04/yr-7.66E-06/yr) = 2.13E-07/yr Class 3b Frequency (Unit 2} (once per ten year test) = 3.33
- 2.13E-07/yr = 7.1 OE-07/yr Class 3b Frequency (Unit 2} (once per fifteen year test) = 5.00
- 2.13E-07/yr = 1.07E-06/yr Table 6-3 shows the results of these calculations. Note that in the above calculation Class 3b releases are considered to arise from a change in state of prior non-LERF states to a LERF (Class 3b) state.
Table 6-3: Farley Estimated Total LEAF Including External Events Impact Case 3b Frequency 3b Frequency 3b Frequency LEAF Increase (3 per 1 0 year
~1 per 10 year
~1 per 15 year
~3 per 10 year to 1
~est)
~est)
~est) per 15 year)
Unit 1 Internal Events 1.90E-08 6.34E-08 9.51E-08 7.61E-08 Contribution Unit 1 Total Contribution 2.22E-07 7.41E-07 1.11 E-06 8.90E-07 including External Events Unit 2 Internal Events 1.88E-08 6.26E-08 9.40E-08 7.52E-08 Contribution Unit 2 Total Contribution 2.13E-07 7.10E-07 1.07E-06 8.52E-07 including External Events The calculated increase in LERF including External Events resulting from a change in the Type A ILRT test interval from the three per ten year test frequency to the once per fifteen year test frequency is 8.90E-07/yr for Unit 1 and 8.52E-07/yr for Unit 2. The corresponding LERF (Class 3b Frequency) for a permanent once per 15 year ILRT program is 1.11 E-06/yr for Unit 1 and 1.07E-06/yr for Unit 2.
Page E1-6 to NL-17-1443 SNC Response to NRC RAis Conclusions Based on the results from Section 5 and the sensitivity calculations presented in Section 6, the following conclusions regarding the assessment of the plant risk are associated with permanently extending the Type A ILRT test frequency to once in fifteen years:
Regulatory Guide 1.17 4 provides acceptance criteria for increase in CDF and LEAF resulting from a risk-informed application. Since the ILRT does not impact CDF, the relevant criterion for this application is LEAF. When the calculated increase in LEAF is "very small", which is taken as being less than 1 o-7 per reactor-year, the change will be generally considered acceptable irrespective of the plant's LEAF value. When the calculated increase in LEAF is in the range of 1 o-7 per reactor-year to 1 o-s per reactor-year, the applications will be considered acceptable only if the total plant LEAF is less than 1 o-s per reactor-year.
From Table 6-3, the increase in LEAF based on the internal events PRA, resulting from a change in the Type A ILRT test interval from three in ten years to one in fifteen years, is conservatively estimated as 7,.61 E-08/yr for Unit I and 7.52E-08/yr for Unit 2, using the EPA I guidance as written. These estimated changes in LEAF for Farley Unit I and Unit 2 are "very small" and therefore determined to be within the acceptance guidelines of Reg. Guide 1.174.
From Table 6-3, the increase in LEAF based on the inclusion of external events impacts, resulting from a change in the Type A ILRT test interval from three in ten years to one in fifteen years, is conservatively estimated as 8.90E-07/yr for Unit I and 8.52E-07/yr for Unit 2, with the corresponding LEAF (3b Frequency) of 1.11 E-06/yr for Unit 1 and 1.07E-06/yr for Unit 2, using the EPA I guidance as written. These estimated changes in LEAF for Farley Unit 1 and Unit 2 are "small" and determined to be within the RG 1.174 acceptance criteria for Region II. In addition, from Table 6-2, the total LEAF based on the inclusion of external events impacts is 3.60E-06/yr for Unit 1 and 7.66E-06/yr for Unit 2, which remains less than the Regulatory Guide 1.174 acceptance guideline of 1 o-s per year.
According to the Regulatory Guide 1.17 4, even though the proposed changes to the ILRT do not change the CDF values, no changes would be permitted per this Regulatory Guide if the plant CDF exceeds 10-4 per year. For the sensitivity case of the examination of the impact of external events, the calculated CDF, with the external events included, does not exceed 1 0-4 for either unit.
The change in Type A test frequency to once per fifteen years, measured as an increase to the total integrated plant risk for those accident sequences influenced by Type A testing, based on the internal events PRA is 4.71 E-03 person-rem/yr for Unit 1 and 4.65E-03 person-rem/yr for Unit 2. EPA I Report No.1 009325, Revision 2-A states that a very small population dose is defined as an increase of s 1.0 person-rem per year or s 1% of the total population dose, whichever is less restrictive for the risk impact assessment of the extended ILRT intervals. This is consistent with the NRC Final Safety Evaluation for NEI 94-01 and EPA I Report No.1 009325. Moreover, the risk impact when compared to other severe accident risks is negligible.
Page E1-7 to NL-17-1443 SNC Response to NRC RAis The increase in the conditional containment failure probability from the three in ten-year interval to a permanent one time in fifteen year interval is 0.91% for Unit 1 and 0.92%
for Unit 2. EPRI Report No. 1 009325, Revision 2-A states that increases in CCFP of S1.5 percentage points are very small. This is consistent with the NRC Final Safety Evaluation for NEI 94-01 and EPRI Report No.1 009325. Therefore, this increase is judged to be very small.
Based on these conclusions, permanently increasing the ILRT interval to fifteen years is an acceptable change.
Page E1-8