ONS-2016-017, Request for Alternative to Codes and Standards Requirements Pursuant to 10 CFR 50.55a(z) to Satisfy 10 CFR 50.55a(h)(2)

Request for Alternative to Codes and Standards Requirements Pursuant to 10 CFR 50.55a(z) to Satisfy 10 CFR 50.55a(h)(2)

ML16062A052
Person / Time
Site:	Oconee
Issue date:	02/15/2016
From:	Batson S Duke Energy Corp
To:	Bill Dean Office of Nuclear Reactor Regulation
References
ONS-2016-017
Download: ML16062A052 (1)

Text

ENERGY Scott L.Batson Vice President Oconee Nuclear Station Duke Energy ONO1VP I 7800 Rochester Hwy Seneca, SC 29672 0: 864,873,3274 f.864.873.4208 Scott Batson@duke-energy.com ONS-201 6-017 10 CFR 50.55a(z)(1)

February 15, 2016 Mr. William M. Dean, Director, Office of Nuclear Reactor Regulation U. S. Nuclear Regulatory Commission Washington, DC 20555-0001

SUBJECT:

Duke Energy Carolinas, LLC Oconee Nuclear Station Unit Nos. 1, 2, and 3, Renewed Facility Operating License Numbers DPR-38, 47, and 55, Docket Nos. 50-269, 50-270, and 50-287; Request for Alternative to Codes and Standards Requirements pursuant to 10 CFR 50.55a(z) to satisfy 10 CFR 50.55a(h)(2) associated with Bronze Tape Wrapped Emergency Power Cables in Use at the Oconee Nuclear Station

References:

1. NRC Letter to Duke Energy, "Oconee Nuclear Station - NRC Component Design Bases Inspection Report 05000269/2014007, 05000270/2014007, and 05000287/2014007," dated June 27, 2014, (ADAMS Accession No. ML14178A535).

2. NRC Memorandum, Director of Reactor Safety to Deputy Director, Division of Policy and Rulemaking, Office of Nuclear Reactor Regulation, "Request for Technical Assistance Regarding Oconee Nuclear Station Design Analysis for Single Failure and the Integration of Class 1E Direct Current Control Cabling in Raceways With High Energy Power Cabling (TIA 20 14-05),"' dated October 16, 2014, (ADAMS Accession No. ML14290A136).

3. Emails from Mr. Randy Hall, Senior Project Manager, Office of Nuclear Reactor Regulation, to Mr. Chris Wasik, Duke Energy Manager of Regulatory Affairs, Oconee Nuclear Station, dated December 8 and 10, 2015.

A June 27, 2014 NRC letter (Reference 1) identified an Unresolved Item (URI) documenting an NRC inspection issue that involves the possibility of ground faults in medium voltage power cabling located in an underground concrete raceway1 at the Oconee Nuclear Station (ONS) that could potentially impact control cabling required to mitigate certain design basis events. The NRC inspection team subsequently requested assistance from the Office of Nuclear Reactor The control cable circuits in the buried underground concrete raceway have been relocated and are no longer a concern; however, extent of condition issues exist in other areas which are the focus of this submittal.

U. S. Nuclear February Regulatory Commission 15, 2016 Page 2 Regulation (NRR) by means of a Task Interface Agreement (TIA), to review Oconee Nuclear Station's emergency power system licensing bases to determine the acceptability of the design (Reference 2). The TIA raised questions with the manner in which the Oconee Nuclear Station achieved the objectives for single failure design as described in IEEE Std. 279-1971.

While awaiting the outcome of a Task Interface Agreement (TIA), Duke Energy is taking several anticipatory actions to address the concerns identified in the URI. Specifically, Duke Energy has completed cable testing and some modifications, with additional modifications under development.

This letter updates the NRC on the status of these actions, and requests NRC review and approval under the provisions of 10 CER 50.55a(z)(1) for certain alternatives to codes and standards. As such, Duke Energy is requesting modification to the station's licensing basis to accept "as-is" certain plant configurations and to approve the application of IEEE Std. 384-1992, Paragraph 6.1.4 in specific locations. Duke Energy considers the proposed alternatives represent an acceptable level of quality and safety for the ONS design and plant configuration.

The technical bases for the 10 CFR 50.55(a)(z)(1) request are contained in the Enclosure to this letter.

Regulatory commitments associated with this request are summarized in the Attachment 1. contains Duke Energy's responses to NRC emailed questions on cable fault testing (Reference 3).

If you should have any questions regarding this submittal, please contact Stephen C. Newman, Lead Nuclear Engineer, Regulatory Affairs, at (864) 873-4388.

Sincerely, Scott L. Batson Vice President Oconee Nuclear Station Enclosure Attachments

U. S. Nuclear Regulatory Commission February 15, 2016 Page 3 xc (w/enclosure/attachments):

Ms. Catherine Haney, Administrator, Region II U.S. Nuclear Regulatory Commission Marquis One Tower 245 Peachtree Center Ave., NE, Suite 1200 Atlanta, GA 30303-1 257 Mr. James R. Hall Senior Project Manager (by electronic mail only)

Office of Nuclear Reactor Regulation U.S. Nuclear Regulatory Commission 11555 Rockville Pike Mail Stop O-8G9A Rockville, Maryland 20852 Mr. Jeffrey. A. Whited Project Manager (by electronic mail only)

Office of Nuclear Reactor Regulation U.S. Nuclear Regulatory Commission 11555 Rockville Pike Mail Stop O-8G9A Rockville, Maryland 20852 Mr. Eddy Crowe NRC Senior Resident Inspector Oconee Nuclear Station

ENCLOSURE Evaluation of Proposed Change

Subject:

PROPOSED ALTERNATIVE INACCORDANCE WITH 10 CFR 50.55a(z)(1)

1. Systems/Components Affected

2. Applicable Regulatory Requirement

3. Reason for Request

4. Proposed Alternative and Basis for Use

5. Cable Testing

6. Risk Insights

7. Precedents

8. References

Enclosure ONS-201 6-017 February 15, 2016 Page 2

1. SYSTEMS/COMPONENTS AFFECTED This request pertains to medium voltage single conductor bronze armor cables related to operation of the Keowee Hydroelectric Stations' (KHS) 13.8 kV and 4.16 kV underground power paths and the 13.8 kV Protected Service Water (PSW) power paths from KHS and the PSW substation (i.e., the "Fant Line"), and areas where they are routed in proximity to certain Keowee safety-related control cables.

Specifically, the medium voltage power cables affected are the:

• Six (6) 13.8 kV Keowee Underground (KUG) feeder cables to CT-4;

• Three (3) 4.16 kV KHS Auxiliary CX transformer power feeder cables,

• Six (6) 13.8 kV B6T and B7T Feeder cables from the KHS to the PSW switchgear building;

• Six (6) 13.8 kV offsite "Fant Line" power feeder cables to the PSW switchgear building; and

• Six (6) 4.16 kV feeder cables from PSW Switchgear B6T to 600 VAC PSW Loadcenter PXI13 transformer.

The specific areas 1 affected associated with this request include the:

• PSW System ductbank manholes2,

• KHS Mechanical Equipment Gallery,

• P8W Building Cable Spreading Area.

A description of each cable's composition and its application at the Oconee Nuclear Station is given below:

Cable Type 1:

• Cable

Description:

Single conductor cable, 750 kcmil conductor, 260 mils insulation with two overlapping 10 mui layers of bronze armor shielding.

• Used On: KHS underground path to the P8W switchgear building; Fant underground (PSW ductbank) path from PSW manhole 6 to the P8W switchgear; P8W 4 kV switchgear to 600 V P8W Load Center.

• System Nominal Operating Voltage: 13.8 kV phase-to-phase or 8 kV phase-to-ground and 4.16 kV phase-to-phase or 2.4 kV phase-to-ground.

SThe control cable circuits inthe buried underground concrete trench (i~e., Trench 3) have been relocated and are no longer included as part of this evaluation.

2Wihnthe PSW ductbank, power and control cables are routed through separate concrete encased conduits. The area of interest is with the power/control cables located inthe PSW ductbank manholes (along the PSW ductbank) which are not contained inindividual conduits.

Enclosure ON S-201 6-017 February 15, 2016 Page 3 Cable Tvpe 2:

• Cable

Description:

Single conductor cable, 250 kcmil conductor, 140 mils insulation with two overlapping 10 mil layers of bronze armor shielding.

• Used On: Underground (Trench 3) path plant feed to KHS station service transformer CX from ONS Unit 1.

• System Nominal Operating Voltage: 4.16 kV phase-to-phase or 2.4 kV phase-to-ground.

2. APPLICABLE REGULATORY REQUIREMENT 10 CER 50.55a(h)(2), "ProtectionSystems," states: "For nuclear power plants with construction permits issued after January 1, 1971, but before May 13, 1999, protection systems must meet the requirements in IEEE Std. 279-1968, "Proposed IEEE Criteria for Nuclear Power Plant Protection Systems," or the requirements in IEEE Std. 279-1971, "Criteria for Protection Systems for Nuclear Power Generating Stations," or the requirements in IEEE Std. 603-1991,"Criteria for Safety Systems for Nuclear Power Generating Stations, and the correction sheet dated January 30, 1995. For nuclear power plants with construction permits issued before January 1, 1971, protection systems must be consistent with their licensing basis or may meet the requirements of IEEE Std. 603-1991 and the correction sheet dated January 30, 1995."

Although the ONS construction permit was issued prior to January 1, 1971, the current licensing bases is that ONS will satisfy Section 4.2 of IEEE Std. 279-1971 for the Oconee Emergency Power and Emergency Core Cooling systems. NRC acceptance of the IEEE Std. 279-1971 single failure criteria is documented in three (3) 1976 safety evaluations associated with changes to the Emergency Core Cooling System model which conformed to the requirements of 10 CFR 50.46. IEEE Std. 279-1971, Section 4.2, "Single Failure Criterion," requires, in part, that any one single failure within the protection system shall not prevent the proper protective action at the system level when required (Reference 8.2). The potential for not satisfying the single failure criteria (based on the requirements of 10 CFR 50.55a(h)) due to interactions between cables in certain cable transition areas is the specific issue addressed by this request. Options available to Duke Energy under this circumstance include 10 CFR 50.55a(z).

10 CFR 50.55a(z) states:

"Alternatives to the requirements of paragraphs (b) through (h) of this section or portions thereof may be used when authorized by the Director, Office of Nuclear Reactor Regulation, or Director, Office of New Reactors, as appropriate. A proposed alternative must be submitted and authorized prior to implementation. The applicant or licensee must demonstrate that:

(1) Acceptable level of quality and safety. The proposed alternative would provide an acceptable level of quality and safety; or

Enclosure ONS-2016-017 February 15, 2016 Page 4 (2) Hardship without a compensating increase in quality and safety. Compliance with the specified requirements of this section would result in hardship or unusual difficulty without a compensating increase in the level of quality and safety."

This request seeks NRC approval in accordance with 10 CFR 50.55a(z)(1) prior to the resolution of the nonconforming conditions initiated in responses to the Unresolved Item discussed in Section 3 shown below.

3. REASON FOR REQUEST A June 27, 2014 NRC letter (Reference 8.8) identified an Unresolved Item (URI) regarding a concern that postulated short circuits and/or ground faults in electrical cabling located in an underground concrete raceway could potentially impact the functionality of the emergency power system which is required to mitigate certain design basis events. The NRC inspection team subsequently requested assistance from the Office of Nuclear Reactor Regulation (NRR) by means of a Task Interface Agreement (TIA), to review the Oconee Nuclear Station emergency power system licensing bases to determine the acceptability of the design (Reference 8.9) with respect to Oconee's current licensing basis.

Duke Energy is taking several anticipatory actions to address the concerns identified in the URI. This submittal requests NRC review and approval under the provisions of 10 CFR 50.55a(z)(1) of certain alternatives to codes and standards. Duke Energy believes that these proposed alternatives address the concerns associated with the URI and that they provide an acceptable level of quality and safety.

The previously referenced URI is associated with a single failure compliance question with IEEE Std. 279-1971. Specifically, the issue is associated with a potential of a power cable fault that causes an adverse interaction with control cables in close proximity to the faulted power cable. This request addresses the NRC concerns by proposing a two-step approach that includes "as-is" temporary and permanent modifications to the plant licensing basis. In summary, to address NRC concerns with single failure capabilities ONS requests:

1. Temporary acceptance of the current configuration in specific locations to allow for sufficient time to implement modifications; and permanent acceptance of the current configuration in certain locations based on cable design, operation and testing.

2. Acceptance of the application of Paragraph 6.1.4 of IEEE Std. 384-1992, "Limited

• Hazard Areas," as a means of providing acceptable cable separation in certain areas of the plant. The classification is appropriate because these are areas containing power cables but without any Hazard Area drivers being present (e.g.,

missiles, high energy lines etc.).

Enclosure ONS-201 6-017 February 15, 2016 Page 5 The description details and justifications for the 10 CFR 50.55(a)(z)(1) request are given below:

3.1 Description of Affected Locations As part of a 2014 Component Design Basis Inspection (CDBI), a question was raised by an inspection team whether this bronze tape on medium voltage power cables in Trench 3 can be credited as armor when evaluating failures per IEEE Std. 279-1971, Section 4.2, and thus whether credit can be taken for its single failure mitigation properties. As described below, the plant areas within the scope of this submittal are the PSW system ductbank manholes, Keowee Mechanical Equipment Gallery, and the PSW building Cable Spreading Area.

3.1 .1 PSW System Ductbank Manholes The PSW system is designed as a standby system for use under emergency conditions. The PSW system provides added "defense in-depth" protection by serving as a backup to certain existing safety systems. The PSW system is provided as an alternate means to achieve and maintain safe shutdown conditions for one, two or three units following certain postulated scenarios. The PSW system also reduces fire risk by providing a diverse power supply to power safe shutdown equipment in accordance with the National Fire Protection Association (NFPA) 805 safe shutdown analyses.

As noted above, the defense-in-depth function of the PSW system provides a diverse means to achieve and maintain safe shutdown by providing secondary side decay heat removal, Reactor Coolant System (RCS) pump seal cooling, RCS primary inventory control, and RCS boration for reactivity management following plant scenarios that disable the 4.16 kV essential electrical power distribution system. The PSW electrical system is designed to provide power to PSW mechanical and electrical components as well as other system components needed to establish and maintain a safe shutdown condition. The system is designed to supply the necessary loads and is electrically independent from the Oconee station electrical distribution system and the 8SF. No credit is taken in the safety analyses for PSW system operation following design basis events. The PSW System can be supplied power from an offsite transmission feed or the KHS.

A separate PSW electrical equipment structure (PSW Switchgear Building) is provided for major P8W electrical equipment. Normal power is provided from the Central Tie Switchyard via a 100 kV transmission line to a 100/13.8 kV PSW substation located adjacent to Oconee Nuclear Station and then via a 13.8 kV feeder that enters the P8W system ductbank at Manhole 6 leading to the P8W building via a ductbank.

Although the power path from the Central Tie Switchyard to the PSW Switchgear Building is classified as non QA-1, the cable in MH-6 to the P8W building was procured to QA-1 standards.

Enclosure ONS-201 6-017 February 15, 2016 Page 6 Alternate QA-1 power is available from each Keowee Hydroelectric Generating unit as an electrical source to the PSW building. The route from KHS to P8W consists of an underground 13.8 kV power cable feeder connecting Keowee output breakers KPF-11I and KPF-1 2 located in KHS to transformers CT6 and CT7 located in the P8W building. The Keowee switchgear circuit breaker and bus arrangement provides the capability of aligning either the Keowee Unit I or Unit 2 generators to the CT6 and/or CT7 transformers.

The KHS to P8W 13.8 kV power feed initially routes from Keowee through an underground trench (along with the CT-4 underground feeder), and then diverts into a separate ductbank / manhole system before reaching the CT-4 blockhouse. The PSW ductbank system from the underground trench to the PSW building consists of underground duct with six intervening manholes. The underground duct segments connecting the manholes consist of separate PVC conduits surrounded by concrete fill. The manholes are designed with the control cables routed across the bottom in a cable tray and the power cables racked above the floor on separate supports. The manhole closest to the underground trench is designated Manhole-I (MH-1) and the manhole closest to the PSW building is designated MH-6. The KHS to PSW 13.8 kV power feed is the Technical Specification credited powerpath but it is not the normal power feed for this system. These cables are typically energized only during PSW system powerpath surveillance testing that is performed on a quarterly basis (-,33 hour3.819444e-4 days <br />0.00917 hours <br />5.456349e-5 weeks <br />1.25565e-5 months <br />s/year total).

The normal power feed to the PSW system originates from the 100 kV Central Tie Switchyard (i.e., the "Fant Line") which is independent of the station switchyard. The Fant Line is routed to ONS by uninsulated overhead distribution lines that terminate at two dip-poles. At each dip-pole, the 13.8 kV P8W power feed transitions to two circuits consisting of three insulated single-conductor power cables for each circuit routed underground. The underground installation continues to a splice box outside MH-6. The route exiting the splice box to the P8W switchgear transitions to QA-1 cabling then routes through MH-6 and the remaining portion of the ductbank system to the P8W building. To improve separation between the Fant line power feed and control cables in MH-6, a modification is underway to relocate this (normally energized) power feed out of MH-6 and the existing ductbank and reroute it to the PSW switchgear building via a new duct. Approval of this modification is not part of this submittal.

In addition to the power feeds, the P8W ductbank system contains low voltage Instrumentation & Control (I&C) cables consisting of supervisory functions for both KHUs, one train of KHU emergency start (safety-related), one train switchyard isolation complete (safety-related), PCB-9 control, and P8W KPF breaker control.

Enclosure ONS-201 6-017 February 15, 2016 Page 7 Power and control cable routing in the PSW ductbank manholes is consistent with Duke Energy design specifications (Reference 8.13).

3.1.2 KHS Mechanical Equipment Gallery The KHS Mechanical Equipment gallery contains motor control centers, cooling water strainers, governors, and Keowee Power Feeder (KPF)

Switchgear KPF-1 and KPF-2. In addition, cabling for the CT-4 underground feeder, the KHS to PSW underground feeder, the KHS to PSW Switchgear (KPF) line side cable bus, the feeder from switchgear 1TC to transformer CX, and adjacent safety-related control cables, all route through the area.

Power and control cable routing in the KHS Mechanical Equipment Gallery is consistent with Duke Energy design specifications (Reference 8.13).

3.1.3 PSW Buildingq Cable Sp~readinaq Area The PSW building Cable Spreading Area is a section of the PSW switchgear building into which the cables discussed in the previous sections enter from the PSW ductbank system. In addition, there are low voltage I&C cables consisting of supervisory functions for both KHUs, one train of KHU emergency start cables (safety-related), one train switchyard isolation complete (safety-related), PCB-9 control, and breaker control for the PSW KPF switchgear. This area also contains PSW power and control cables.

Power and control cable routing in the PSW Building Cable Spreading Area is consistent with Duke Energy design specifications (Reference 8.13).

Enclosure ONS-2016-01 7 February 15, 2016 Page 8

4. PROPOSED ALTERNATIVE AND BASIS FOR USE This request is specific to the applications identified herein and is not intended to be a blanket request for approval of the use of IEEE Std. 384-1992 at ONS.

4.1 PSW System Ductbank Manholes 4 Current Configquration:

The KHS to P8W 13.8 kV power feed initially routes from Keowee through an underground trench and then diverts into a separate ductbank/manhole system before reaching the CT-4 blockhouse. The PSW ductbank system from the underground trench to the PSW building consists of an underground duct with six intervening manholes. The manhole closest to the underground trench is designated Manhole-I (MH-1) and the one closest to the PSW switchgear building is designated Manhole-6 (MH-6). The feed from the KHS to PSW is normally not energized.

In addition to the 13.8 kV power feed from the KHS to the PSW building, MH-6 also contains the normally energized 13.8 kV power feed (i.e., the "Fant Line")

from the Central Tie Switchyard to the PSW switchgear building via the PSW substation.

The PSW ductbank system also contains low voltage Instruments & Controls (l&C) cables routed in separate ductbank conduits consisting of supervisory functions for both KHUs, one train of KHU emergency start, one train switchyard isolation complete, PCB-9 control, and breaker control for the PSW KPF switchgear. In addition, future station changes may route a second channel of KHU Emergency Start and Switchyard Isolation Complete via the PSW ductbank.

Proposed Alternate Method:

Pursuant to 10 CER 50.55a(z), Duke Energy requests:

1. For manhole 6 (MH-6), Duke Energy requests temporary acceptance of the "as-is"~configuration of the Fant line feeder and adjacent KHS control cables as an alternative to meeting the requirements of 10 CFR 50.55a(h)(2) until the Fant line power feeder relocation modification is complete. The modification will be completed no later than September 15, 2017.

2. For manholes I through 6, Duke Energy requests to modify the ONS licensing basis pursuant to 10 CFR 50.55a(z) to allow acceptance of the "as-is" configuration of the normally de-energized 13.8 kV KHS to the PSW 4The KHS to PSW power cables inthe PSW ductbank are routed through individual concrete encased conduits that are separate from the I&C cables and conduit within the same ductbank. The area of concern for interaction iswithin the PSW ductbank manholes, where power cables and l&C cables are not contained within individual conduits.

Enclosure ONS-201 6-017 February 15, 2016 Page 9 building power feed as an alternative configuration to meeting the requirements of 10 CER 50.55a(h)(2).

An acceptable level of quality and safety is demonstrated via the followingq:

• satisfactory cable crush test results (refer to Section 5),

• satisfactory cable fault test results (refer to Section 5),

• the potential for power cable to control cable interaction in the manholes represents a small portion of the overall cable run total (-,180 feet out of

-,4500 feet),

• robust power cable design and cables procured to QA-1 standards which minimizes the likelihood of cable interactions,

• the KHS to PSW power cables are not normally energized which minimizes the likelihood of cable interactions; for a PSW event, these cables would only be energized if the Fant Line is unavailable,

• high impedance grounding system limits fault current (KHS-PSW feeder) and minimizes the effect of any cable interaction should a fault occur,

• Fant line substation breaker testing is operated every 12 months,

• station power cables are evaluated as described in UFSAR Section 18.3.14, "Insulated Cables and Connections Aging Management Program,"

and

• the cables are housed in a steel-reinforced concrete ductbank/manhole engineered to withstand earthquakes, tornado missiles, and to minimize water entry.

4.2 KHS Mechanical Equipment Gallery Current Configquration:

The KHS Mechanical Equipment gallery contains motor control centers, cooling water strainers, governors, and the Keowee Power Feeder (KPF) Switchgear KPF-1 and KPF-2. In addition, cabling for the CT-4 underground feeder, the KHS to PSW underground feeder, the KHS to P8W Switchgear (KPF) line side cable bus, the feeder from switchgear ITC to transformer CX, and adjacent Keowee unit control cables all route through the area.

Proposed Alternate Method:

1. Modification of the ONS licensing basis pursuant to 10 CFR 50.55a(z) to allow acceptance of a proposed modification to the separation requirements for a Limited Hazard Area as noted by IEEE Std. 384-1992, Paragraph 6.1.4 with respect to the CX auxiliary power feed to the KHS, the KHS underground emergency power feeder to CT-4, the P8W KPF switchgear line side cable bus, and adjacent control cables, is requested as an alternative configuration to meeting the requirements of 10 CFR 50.55a(h)(2).

Enclosure ONS-201 6-017 February 15, 2016 Page 10 Adoption of the IEEE Std. 384-1992, Limited Hazard Area classification, is due to the area containing power cables without any Hazard Area drivers present (e.g., missiles, high energy lines etc.). The incorp)oration of this standard into the licensing basis is limited to Paragraph 6.1.4 and is only for the scope of equipment specified in this section.

Following guidance from IEEE Std. 384-1992, the openly routed medium voltage bronze armor power cables and low voltage Keowee control cables require separation of three feet horizontally and five feet vertically, which is not achieved at all locations in the Keowee Mechanical Equipment Gallery. Where open distance is not achieved, enclosures are or will be provided for the medium voltage power cables and low voltage control cables.

By providing a fully metallic enclosed raceway to meet the enclosed raceway separation distance, the requirement of the IEEE Std. 384-1992 standard is met and sufficient physical separation between the power and control circuits is achieved. By fully and separately enclosing the medium voltage power cables and the low voltage control cables, the IEEE Std. 384-1992 required separation distance is reduced to one inch in each direction, which is maintained with the proposed design.

2. Until modifications are completed, temporary acceptance pursuant to 10 CFR 50.55a(z) of the "as-is" configuration in the KHS Mechanical Equipment Gallery as an alternative configuration to meeting the requirements of 10 CFR 50.55a(h)(2) until the above proposed modification is complete. Modification completion is to be no later than September 15, 2017.

3. Modification of the ONS licensing basis pursuant to 10 CER 50.55a(z) to allow acceptance of the "as-is" configuration of the normally de-energized 13.8 kV power feed from the KHS to the PSW building as an alternative configuration to meeting the requirements of 10 CFR 50.55a(h)(2).

An acceptable level of quality and safety is demonstrated via the followingq:

• satisfactory cable crush test results (refer to Section 5),

• satisfactory cable fault test results (refer to Section 5),

• modifications to meet enclosed raceway separation requirements for a Limited Hazard Area as noted by IEEE Std. 384-1992, paragraph 6.1.4, "Limited Hazard Area of IEEE Std 384-1992," (Reference 8.3)[as endorsed in RG 1.75].

• robust power cable design and cables procured to QA-1 standards which minimizes the likelihood of cable interactions,

• KHS power cables (CT-4 and KPF) are not normally energized which minimizes the likelihood of cable interactions; for a PSW event, these cables would only be energized if the Fant Line is unavailable.

• high impedance grounding system limits fault current (KPF and CT-4) and minimizes the effect of any cable interaction should a fault occur,

Enclosure ONS-2016-017 February 15, 2016 Page 11

• limited exposure distance (approximately 100 feet) which minimizes the opportunity of cable interactions,

• station power cables are evaluated as described in UFSAR Section 18.3.14, "Insulated Cables and Connections Aging Management Program,"

and

• the cables are protected from the environment in that they are in the KHS powerhouse and not exposed to environmental hazards.

4.3 PSW Building Cable Spreading Area Current Configquration:

The PSW Building Cable Spreading Area is a section of the P8W building into which the cables discussed in the previous sections enter from the PSW ductbank system. In addition to the power feeds, low voltage I&C cables consisting of supervisory functions for both KHUs, one train of KHU emergency start cables, one train switchyard isolation complete, PCB-9 control, and breaker control for the PSW KPF switchgear also enter the PSW building via these ductbanks. Additionally, power and control cables for other PSW functions are present.

Proposed Alternate Method:

1. Modify the ONS licensing basis pursuant to 10 CER 50.55a(z) to accept a proposed modification to meet the separation requirements for a Limited Hazard Area as noted by IEEE Std. 384-1992, Paragraph 6.1.4, with respect to the normally energized Fant line power supply feeder, normally energized feeder from switchgear B6T to PXI13 transformer, and adjacent KHS control cables as an alternative configuration to meeting the requirements of 10 CFR 50.55a(h)(2).

Adoption of the IEEE Std. 384-1992, Limited Hazard Area classification, is due to the area containing power cables without any Hazard Area drivers present (e.g., missiles, high energy lines etc.). The incorporation of this standard into the licensing basis is limited to Paragraph 6.1.4 and is only for the scope of equipment specified in this section.

Following guidance from IEEE Std. 384-1992, the openly routed medium voltage power cables and low voltage control cables require separation of three feet horizontally and five feet vertically, which is not achieved at all locations in the PSW cable spreading area. Where open distance is not achieved, enclosures are or will be provided for the medium voltage power bronze armor cables and low voltage control cables.

By providing a fully metallic enclosed raceway to meet the enclosed raceway separation distance, the requirement of the IEEE Std. 384-1992 standard is met and sufficient physical separation between the power and control circuits is achieved. By fully and separately enclosing the medium voltage power cables and the low voltage control cables, the IEEE Std. 384-1992

Enclosure ONS-201 6-017 February 15, 2016 Page 12 required separation distance is reduced to one inch in each direction, which is maintained with the proposed design. Adoption of IEEE Std. 384-1992 is limited to Paragraph 6.1.4 and is only for the scope of equipment specified in this section.

2. Modify the ONS licensing basis pursuant to 10 CFR 50.55a(z) to allow acceptance of the as-is configuration of the normally de-energized 13.8 kV power feed from KHS to the PSW building as an alternative configuration to meeting the requirements of 10 CER 50.55a(h)(2).

3. Pursuant to 10 CFR 50.55a(z), until modifications are completed, grant temporary acceptance of the "as-is" configuration regarding the Fant line feeder, the normally energized feeder from switchgear B6T to PX1 3 transformer, and adjacent KHS control cables is requested as an alternative to meeting the requirements of 10 CER 50.55a(h)(2) until the above proposed modification is complete. Modification completion is to be no later than September 15, 2017.

An acceptable level of quality and safety is demonstrated via the followingq:

• satisfactory cable crush test results (refer to Section 5),

• satisfactory cable fault test results (refer to Section 5),

• modifications to meet enclosed raceway separation requirements for a Limited Hazard Area as noted by IEEE Std. 384-1992, paragraph 6.1.4, (Reference 8.3)[as endorsed in RG 1.75],

• robust power cable design and cables procured to QA-1 standards which minimizes the likelihood of cable interactions,

• KHS power cables are not normally energized which minimizes the likelihood of cable interactions; for a PSW event, these cables would only be energized if the Fant Line is unavailable

• high impedance grounding system limits fault current (KHS-PSW feeder) which minimizes the likelihood of cable interactions,

• Limited exposure distance (approximately 90 feet),

• station power cables are evaluated, as described in UFSAR Section 18.3.14, "Insulated Cables and Connections Aging Management Program," and

• the cables are housed in a protected area engineered to withstand earthquakes, tornado missiles, and to minimize water entry.

5. CABLE TESTING Duke Energy conducted testing to validate that the bronze armored emergency power cable design provides an acceptable level of quality and safety. This testing was conducted in two phases, cable crush testing and cable fault testing. Test results demonstrate the cables provide an acceptable level of quality and safety and that a single failure of a bronze armored medium voltage power cable will not result in a consequential loss of safety functions performed by adjacent low voltage control cables.

Enclosure ONS-201 6-017 February 15, 2016 Page 13 Phase One - Cable Crush Testingq The first phase performed testing based on Underwriters Laboratory (UL)1569 'Metal Clad Cables' crush and impact tests of the medium voltage cables and was completed in February 2015. The phase one testing compared the bronze armor cables to galvanized steel interlocked armored cables with respect to their physical protection based on UL1 569 testing sections 24 (impact testing), 25 (increasing crush), and 26 (direct burial crush). All of the cable types tested, including the medium voltage cables installed at the ONS, confirmed that the cable configuration with bronze armor provides adequate protection to perform consistently with armored cable based on UL 1569, Sections 24, 25, and 26. The results of this testing were provided to the NRC on May 11, 2015 (Reference 8.5).

Phase Two - Cable Fault Testingq The second phase of the testing involved inducing a fault on a single conductor medium voltage bronze armor power cable while monitoring for any effects on adjacent power and control cables. The fault testing (described below) further validates the engineering analyses which have concluded that the power cables subject to a single phase-to-ground fault will not propagate to a multi-phase fault and will not adversely interact with the low voltage control cables leading to consequential functional failures of redundant trains.

From November 2-6, 2015, a series of cable fault tests were conducted for Duke Energy at KEMA Laboratories in Pennsylvania. These tests were also observed by NRC Staff. Duke Energy commissioned the tests to determine the potential impacts of electrical fault in a medium voltage power cable with bronze armor. The test inputs which demonstrate the worst case bounding inputs and required critical parameters are documented in a calculation (Reference 8.12).

The primary purpose of the cable fault testing was to determine if a single cable failure (phase-to-ground fault) on a single medium voltage power cable will propagate to a multi-phase fault and damage adjacent cables. A secondary purpose was to determine if low voltage control cables installed near the faulted cable would be damaged and if unacceptable voltage would be induced on the low voltage conductors. Four (4) different configurations were tested five (5) times each.

The test setups used were configured in a manner to maximize the potential for a single cable phase-to-ground fault to propagate into a multi-phase fault. Test voltages, phase fault currents and associated durations, bound the Oconee plant values for the cables of concern. The tests were conducted on bronze armored medium voltage cables of same specifications as existing plant cables located in the following areas:

• 13.8 kV KHS Underground Path (KHS to Transformer CT-4),

• 13.8 kV PSW Underground Path (KHS to P5W Switchgear),

• 13.8 kV Fant Path (Manhole 6 to P5W Switchgear),

• 4.16 kV KHS Underground Path (Breaker 1TC-04 to KHS Transformer CX).

Enclosure ONS-201 6-017 February 15, 2016 Page 14 Test Methodologyv:

For each test configuration, a single power cable was prepared by cutting a triangular flap in the cable jacket and drilling a small hole through the cable bronze metallic shield, insulation semicon, insulation and conductor sem icon to the conductor. The jacket flap was placed back and secured to the cable jacket with tape. The hole through the metallic shield was not to be repaired and tape was not installed over the area where the hole is drilled. This approach is conservative with respect to the type of test being performed. Since the cable insulation and metallic shield system was compromised, this resulted in a single phase-to-ground fault to immediately occur when the cable is energized. A small copper wire was inserted into the hole to facilitate the cable fault.

The position and orientation of the cable fault was such that if the bronze tape metallic shield and cable jacket were penetrated by the fault, the effects of the fault would directly impinge on the adjacent power cable. The power cables were arranged in a triangular bundle(s) and held in close contact by cable cleats which were mounted to a cable tray. This was a conservative orientation since any increase in distance would reduce the severity of consequential damage (if any) of electrical faults on adjacent cables.

Instrument & Control (I&C) cables were also installed in the cable tray attached to the cable tray with stainless steel ty-wraps parallel to the cable at nominal spacing.

The I&C cable interlocked steel armor and underlying shielding were grounded on both ends. The I&C cable conductors were not energized but were monitored during the test for induced voltage. The I&C cables were repositioned closer (control cable separation from power cables varied from 5" to no gap spacing) to the power cable to obtain additional data for conservatism.

The testing laboratory replicated the critical parameters of the Oconee power systems response, including generator/power source neutral grounding arrangement (resistance or solidly grounded), voltages and phase-to-ground fault currents, and fault durations that included relay response and breaker opening times. The test was also configured to replicate multi-phase faults if fault extended to other phase cables.

The primary test parameters were fault currents and voltage on the faulted cable and the overall fault duration,

_Conduct of Testingq:

The testing was performed using a laboratory procedure and results were compiled and documented in a test report. Subsequently, Duke Energy evaluated the test program lab, procedure, and results; and determined the commercial grade dedication of the testing was acceptable as a QA product. Each test article configuration was tested at least five (5) times. After each test, the following parameters were inspected:

1. Verified that a single phase-to-ground fault did not result in a multi-phase fault through visual inspection line current data and/or cable electrical testing.

Breaching of the metallic shield and jacket of the faulted cable was acceptable.

Enclosure ONS-201 6-017 February 15, 2016 Page 15 Scorching or other damage to the jacket, metallic shield and insulation of the adjacent cables was acceptable provided the initial phase-to-ground fault did not propagate to a multi-phase fault.

2. Verified that a power cable fault did not result in medium voltage being imposed on I&C cable conductors by review of the voltage monitored by the test laboratory data acquisition system. Scorching or other damage to the I&C cable jacket, armor or underlying shields and tapes was acceptable provided the underlying conductor insulation was undamaged as verified by visual inspection and/or cable electrical testing.

Test Results:

• None of the test cases resulted in cable damage that propagated to a multi-phase fault.

• In some of the tests, the adjacent power cable had a superficial indentation at the fault location but no jacket or shield damage occurred.

• In one of the five 4 kV cable fault tests, there was damage that penetrated the adjacent power cable's outer jacket and bronze tape shield; however, the internal insulation remained intact and no phase-to-phase fault occurred.

Follow-up testing of this cable showed the cable passed a 30-minute withstand test at 7.0 kV.

• In each test case, the cable jacket and bronze tape performed its function of protecting the adjacent conductors and not allowing a fault to propagate (based on visual inspection of the test specimen).

• In each test case there was no observable damage to the control cables in the tray section adjacent to the faulted power cables.

• Low voltage levels observed on control cables determined to be inconsequential, based on 3rd party review of test results and finite element model of a full length configuration.

On November 18, 2015, in a public meeting with NRC staff, Duke Energy outlined its plans to submit a licensing action to address cable separation issues. At that time, NRC questions on the testing were developed in preparation for a follow-up public meeting with Duke Energy on December 15, 2015. These questions were emailed to Duke Energy on December 8 and 10, 2015. Duke Energy's responses to these questions were shared at the December 15, 2015, public meeting and are provided in Attachment 2.

6. RISK INSIGHTS Note: The following section is not part of the basis for acceptance but is being provided for risk insight purposes.

A risk analysis (Reference 8.7) was performed to determine the potential risk impact of the current plant configuration with respect to cable separation in the locations of concern. The risk analysis determined the potential risk impact of the current plant configuration with respect to cable separation in the following three (3) locations:

Enclosure ONS-201 6-017 February 15, 2016 Page 16

• PSW System Ductbank Manholes 1-6,

• KHS Mechanical Equipment Gallery,

• PSW Building Cable Spreading Area.

Note that the Reference 8.7 analysis also addresses an additional location (Trench

3) which is not addressed in this submittal. For each location, an estimate of the increase in core damage frequency (CDF) and large early release frequency (LERF),

above that which would exist if the DC control cables were not co-located with the AC power cables, was developed. The analysis considered the following aspects:

• Frequency of cable faults,

• Probability that a fault is a multi-phase or high energy arc fault (HEAF),

• Probability of a large imposed voltage on one or both Oconee vital 125 VDC trains,

• Probability of failure of one or both Oconee vital 125 VDC trains, given an imposed voltage,

• Probability of failure of mitigation strategies.

Each of the above aspects is described briefly below:

Freauency of Cable Faults:

The analysis used generic industry data (Reference 8.10) on cable faults as a starting point to determine the likelihood of a fault on an energized medium voltage power cable. Reference 8.10 gives a failure rate of 7.2E-04/year per 500 feet of cable for pink EPR cables (the type used at Oconee). Although the EPRI data may be considered representative of general pink EPR medium voltage underground cable failure rates, this data includes cables of the "Uni-Shield" design which has a reduced-diameter, no external insulating jacket, and a much different / compromised shield. In the EPRI data, the much more vulnerable "Uni- Shield" design failures are included with the standard-shield pink-EPR cables. Further parsing of the EPRI data indicates that all but 3 out of 15 (or 20%) of the pink EPR cable failures were of the "Uni-Shield" design.

Since the cables in the locations of concern are not of that design, the cable failure frequency was reduced by a factor of five to 1.44E-04/year per 500 feet of cable for this Oconee-specific evaluation. For each location, this value was then adjusted to account for the length of cable in that location, and the amount of time the cable is energized.

Enclosure ONS-201 6-017 February 15, 2016 Page 17 The following tables provide an overview of the results of this adjustment for each location. The frequency contribution from each set of cables is the product of the number of cables, the base frequency, the length adjustment and the hours per year divided by 8760 hour0.101 days <br />2.433 hours <br />0.0145 weeks <br />0.00333 months <br />s:.

6 32.8 0.36 1,94E-07 1.2E-06 Fr1e~e ~ I.2E-06 6 99.0 0.2 3.25E-07 1.95E-06 3 8760 0.2 2,88E-05 8.64E-05 6 32.8 0.2 1.08E-07 6.47E-07

• l'*,,'*I*; Total IEl*ii~i 6 8760 0.18 ...E-05 1.56E-04 Probability that a Fault is a Multi-Phase or Higqh Eneravy Arc Fault (HEAF):

Although the expected result of a fault across the insulation is a ground fault to the bronze tape shielding, it cannot absolutely be ruled out that a multi-phase fault could occur. However, this failure mode is unlikely since, as discussed above, the fault across the insulation would have to penetrate the grounded bronze shielding of the faulted cable, and then penetrate the grounded bronze shielding and the insulation of an adjacent cable, all prior to the actuation of the protection system, in order to fault to another phase. Duke Energy is not aware of any mechanism that could cause a fault to behave in this manner. Consider Eaton Corporation White Paper TP08700001 E "Fault Characteristics in Electrical Equipment" published in 2011 (Ref.

8.11), which provides data on the likelihood of multi-phase faults. The paper summarizes data collected in IEEE Std. 493'Recommended Practice for the Design of Reliable Industrial and Commercial Power Systems.' IEEE Std. 493 examined the

Enclosure ONS-201 6-017 February 15, 2016 Page 18 total number of faults that occurred over a sample of equipment, and compared the quantity of those involving ground with those that did not and calculated a percentage of each type of fault relative to the total. This paper concludes "This IEIEE 493 standard is stating that for a cable, it is nearly 100 times more likely (99%

divided by 1%) for a cable fault to be a ground fault versus a fault not involving ground." Based on this, a 1.0E-02 probability has been assigned that, given a cable fault occurs, it is a multi-phase fault.

Probability of a Larae Imposed Voltaae on One or Both Oconee Vital 125 VDC trains:

Duke Energy did not identify any specific data that addresses the likelihood that a HEAF on one of medium voltage power cables in the vicinity of the DC control cables would induce a large voltage on those control cables. However, reasonable engineering judgment based on the design of the DC control cables (i.e., interlocked, grounded, stainless steel armor, multiple ground planes, etc.), supports the conclusion that the likelihood of this scenario would be very low. This is supported by the cable testing and analysis described above, where no significant voltage was impressed on any of the DC control cables in any of the tests. A value of 5E-02 was used for this likelihood. The value selected is considered bounding and worst case, based upon the testing showing no adverse impact.

Probability of Failure of One or Both Oconee Vital 125 VDC Trains. Given an Imp~osed Voltage:

It would take a significant failure (i.e., failures of multiple distribution panels on multiple trains) of the Oconee 125 VDC system, caused by an imposed voltage on that system from the fault, to impact the ability to safely shut down the units.

However, the probability of a loss of all 125 VDC power at Oconee is low, given the design of the cables in these manholes (which have grounded bronze tape for AC power cables or interlocked steel armor for the DC control cables), the design of the manholes themselves (which have grounded uni-strut supports), and the design of the DC system (which has multiple ground planes in the marshalling cabinets, and in the Oconee DC system itself).

Lastly, the 125 VDC system at Oconee is designed to maintain defense-in-depth and safety margin by two independent trains, such that an induced failure of one train should not impact the other. Thus, while failure of a single train of DC power is unlikely, failure of both trains is even less likely. A value of I1.0E-02 has been used for the likelihood of the failure of a single train of 125 VDC power, while a value of 4.OE-03 has been used for the likelihood of failure of both trains of 125 VDC power, which includes a common cause element, These values are considered bounding and worst case, again, based on the results of the cable testing.

Probability of Failure of Mitigqation Strategcies:

Even in the event of a complete loss of normal and emergency offsite power and loss of the 125 VDC system, there is still equipment available for mitigating the event. It is clear that mitigating strategies are limited in the case of a loss of all DC power, due to the significant loss of normal safety equipment control and control room indication.

Enclosure ONS-201 6-017 February 15, 2016 Page 19 However, the Standby Shutdown Facility (SSF) which is designed to provide the ability to maintain the plant in a safe condition, is still available, providing further defense-in-depth and safety margin.

The SSF is completely independent normal plant systems (including 125 VDC power). Electrical power to the SSF is provided by a dedicated diesel generator, while DC control power to the SSF is provided by a dedicated battery and battery charger. In the event of a failure of the 8SF DG, power can be provided by an additional offsite source (i.e., the 13.8 kV power feed from the Central Tie Switchyard through the PSW system, that is not affected by any failures in manholes I through 6). Note that although the PSW system itself can potentially provide additional mitigation capability (i.e., auxiliary feedwater, seal injection), no credit has been taken in this analysis for those system. Mitigating strategies in the case of a loss of only a single train of DC are much more robust, since one train of normal plant safety equipment and indication would remain available, in addition to the 8SF. Failure probabilities for the mitigating strategies of 3.4E-03, given loss of a single train of 125 VDC, and 2.2E-02 given a loss of both trains of 125 VDC, have been calculated based on portions of the Oconee PRA model.

Conclusions:

The analysis shows that the overall CDF/LERF increase for the three (3) cable locations is approximately 2E-11/year. The CDF/LERF increase for manholes I through 6, whose configuration is proposed to remain "as-is," is less than 1E-13/year, which is many orders of magnitude below what is typically considered risk significant.

The CDF/LERF increases from the PSW cable spreading area and the KHS Mechanical Equipment Gallery are approximately 1E-1 1/year and 5E-I12/year, respectively. These locations have relatively short lengths of cable, but do have some normally energized AC power cables. Again, these values are several orders of magnitude below what is typically considered risk significant, and will be reduced even further with completion of the proposed modifications.

7. PRECEDENTS No previous IOCFR50.55a(h)(2) related examples were found associated with both cable separation issues and the use of bronze armor cables in emergency power protection system applications.

8. REFERENCES 8.1 NRC Regulatory Guide 1.75, Criteria for Independence of Electrical Safety Systems, Revision 3 (ADAMS Accession Number ML043630448).

8.2 IEEE Std. 279-1971, "IEEE Standard: Criteria for Protection Systems for Nuclear Power Generating Stations."

8.3 IEEE Std. 384-1992, "IEEE Standard Criteria for Independence of Class IE Equipment and Circuits."

Enclosure ONS-201 6-017 February 15, 2016 Page 20 8.4 Oconee Nuclear Station Updated Final Safety Analysis Report (UFSAR),

Revision 24, effective date of contents 12/31/14. UFSAR Chapters/Sections consulted:

• Section 3.1 .1.1, Design of Structures, Components, Equipment, and Systems - Conformance with NRC General Design Criteria.

• Section 8.2.1 .3.1, Electric Power - Off-site Power System - System Description - 230 kV Switching Station - 230 kV Switching Station Degraded Grid Protection

• Section 8.3.1.1.1, Electric Power - Onsite Power Systems - AC Power Systems - System Descriptions - Keowee Hydro Station

• Section 8.3.1.4.6, Electric Power - Onsite Power Systems - AC Power Systems - Independence of Redundant Systems - Cable Installation and Separation

• Section 8.3.2.1.3, Electric Power - Onsite Power Systems - DC Power Systems - System Descriptions - 125 Volt DC Keowee Station Power System

• Section 9.7.3.2, Auxiliary Systems - Protected Service Water System -

System Description - Electrical

• Section 9.7.3.5.1, Auxiliary Systems - Protected Service Water System -

System Description - Civil/Structural - Building Structures

• Chapter 15, Accident Analyses.

• Chapter 18, Section 18.3.14 - Insulated Cables and Connections Aging Management Program Aging Management Programs and Activities.

8.5 Duke Energy Letter to the Nuclear Regulatory Commission, "TIA 2014-05, Potential Unanalyzed Condition Associated with Emergency Power System,"

dated 5/11/2015.

8.6 Duke Energy Letter to the Nuclear Regulatory Commission, "Supplemental Information on TIA 2014-05, Potential Unanalyzed Condition Associated with Emergency Power System," dated 8/7/2014.

8.7 OSC-1 1478 "Oconee Medium Voltage Cable Separation Risk Assessment,"

Revision 1.

8.8 NRC Letter to Duke Energy, "Oconee Nuclear Station - NRC Component Design Bases Inspection Report 05000269/2014007, 05000270/2014007, and 05000287/2014007," dated June 27, 2014, (ADAMS Accession No. ML14178A535).

8.9 NRC Memorandum, Director of Reactor Safety to Deputy Director, Division of Policy and Rulemaking, Office of Nuclear Reactor Regulation, "Request for Technical Assistance Regarding Oconee Nuclear Station Design Analysis for Single Failure and the Integration of Class 1E Direct Current Control Cabling in Raceways With High Energy Power Cabling (TIA 2014-05)," dated October 16, 2014, (ADAMS Accession No. ML14290A136).

Enclosure ONS-201 6-017 February 15, 2016 Page 21 8.10 "Plant Support Engineering: Failure Models and Data Analysis for Nuclear Plant Medium Voltage Cables for Consideration in Preventive Maintenance and Strategic Replacement," EPRI, December 2009.

8.11 Eaton Corporation White Paper TP08700001 E, "Fault Characteristics in Electrical Equipment", September 2011.

8.12 OSC-11504, "Medium Voltage Cable Testing Analysis," Revision 1.

8.13. Duke Energy Design Specification OSS-0218.00-00-0019, "Cable and Wiring Separation Criteria," Revision 17.

ATTACHMENT I Regulatory Commitment Table Regulatory Commitment Table February 15, 2016 Pane 2 The following commitment table identifies those actions committed to by Duke Energy Carolinas, LLC (Duke Energy) in this submittal.

1 Complete field implementation of:

• Cable separation modifications to the PSW System Cable9/51 Spreading Area and the KHS Equipment Gallery, and

• Fant feeder line relocation out of PSW duotbank manhole 6.

ATTACHMENT 2 Duke Energy Responses to NRC Questions on Cable Fault Testing Duke Energy Responses to NRC Questions on Fault Testing February 15, 2016 Page 2

1. In accordance with 10 CFR 50.54(Ui) (2015), what quality standards were used to design the testing plan and to analyze the testing data?

Duke Energy Response:

10 CFR 50.54 (jj) states "Structures, systems, and components subject to the codes and standards in 10 CFR 50.55a must be designed, fabricated, erected, constructed, tested, and inspected to quality standards commensurate with the importance of the safety function to be performed."

Test control quality standards are defined in the Duke Energy Topical Report Quality Assurance Program Description Operating Fleet Paragraph 017.3.2.8 (Test Control).

The Test Control section requires that testing be performed by written procedures that include requirements and acceptance limits, test instructions, test prerequisites such as calibrated instrumentation and methods for recording data and documenting test results.

These elements were present and documented in the vendor supplied test procedure. In addition this test procedure was reviewed and accepted by qualified individuals at Duke Energy.

To ensure adequate control of the design inputs and test identification, Duke Energy generated Reference 8.12 to provide circuit parameters and configurations which were used as the basis of the test procedure review. In addition Appendix C of this calculation includes a detailed analysis of test results. This calculation was generated as a QA-1 document in accordance with Duke Energy internal Design Analysis Procedures and the Duke Energy Quality Assurance Program.

2. Did the design of these tests meet the requirements of IEEE 279-1971? How?

Duke Energy Response:

The requirements of IEEE Std. 279-1971 are not applicable to cable testing. IEEE Std. 279-1971 Section 1 (Scope) states "These criteria establishes minimum requirements for the safety related functional performance and reliability of protection systems" where "protection systems encompasses all electrical and mechanical devices and circuitry (from sensors to actuation device input terminals) involved in generating those signals associated with the protective function" Signals are defined as those which "actuate reactor trip" and "engineered safeguards."

Section 4.4 (Equipment Qualification) indicates that "test data shall be available to verify that protection system equipment shall meet, on a continuing basis, the performance requirements determined to be necessary to achieving the system requirements." The cable testing was not a qualification test of the Oconee Protection Systems.

Duke Energy Responses to NRC Questions on Fault Testing February 15, 2016 Page 3

3. How were the worst-case tested ground faults determined? What quality standards were used for this determination?

Duke Energy Response:

Reference 8.12 utilized input analyses to determine the maximum available Keowee generator(s) ground fault current as well as the ground fault protection relaying operation timing. The original design of the Keowee generator neutral grounding system was performed in accordance with fundamental electrical power theory (i.e.,

basic Ohm's law, resistance/impedance reflection via transformer turns ratio and the application of an overvoltage relay parallel with the grounding resistance).

The plant (transformer CX) related feeds are analyzed using ETAP Version 7.1.0N.

The ETAP program provides full compliance with the IEEE C37 series for fault calculations. The PSW (Fant 13.8kV) feeder fault current values were provided by Duke Power Delivery based upon the actual system fault capability.

Design input fault analysis data was reviewed per Reference 8.12 which was generated as a QA-1 document in accordance with Duke Energy internal Design Analysis Procedures and the Duke Energy Quality Assurance Program

4. Why were three-phase faults not considered for testing?

Duke Energy Response:

Three-phase faults were a testing consideration. All tests began with a single phase-to-ground fault. Ifthe initial phase-to-ground fault propagated to a three-phase fault, the testing lab power source and relaying were configured and calibrated to provide the required three-phase fault voltage and current for the specified duration.

However, since all of the tests began and ended as phase-to-ground faults, the opportunity to use three-phase fault test parameters did not occur.

5. Why did the testing not address cascading failures (i.e. circuit breaker failures that may result from the short circuit conditions)?

Duke Energy Response:

It is assumed that no other failure occurs concurrent with the line to ground fault with one conductor of the three conductor (three phase) cable bundle. This is in accordance with Section 8.3.1.2 of the UFSAR which states the following, "The basic design criterion for the electrical portion of the emergency electric power system of a nuclear unit, including the generating sources, distribution system, and controls is that a single failure of any component, passive or active, will not preclude the system from supplying emergency power when required." For the purposes of this analysis, a single line to ground (conductor to shield) failure shall be treated as a single passive failure.

Duke Energy Responses to NRC Questions on Fault Testing February 15, 2016 Page 4

6. What analysis was done to ensure that each configuration bounded the worst case asymmetrical and symmetrical fault conditions for:

a. Configuration 1?

b. Configuration 2?

c. Configuration 3?

d. Configuration 4?

Duke Energy Response:

Symmetrical current values were used to determine potential cable damage effects since the damage mechanism is thermal and heat transfer occurs over time.

Symmetrical current values are used in Section 5.2 of IEEE Standard 1584-2002, IEEE Guide for Performing Arc-Flash Hazard Calculations. Since a line to ground fault would produce an arc and lead to the consumption of cable materials (insulation and conductive material), this could be treated as a small arc flash. Thus IEEE 1584-2002 is appropriate to be used as a basis for the use of symmetrical fault currents in this analysis. It is not appropriate to use Asymmetrical fault currents as they have an extremely short duration and thus would not significantly contribute to any heating effects to degrade the cable construction materials.

For the CT-4 (Keowee Underground) and KPF (Keowee to PSW) cases, the maximum line to ground fault current is limited by the high impedance grounding system. This limiting ground fault current value and duration (breaker and protective relaying clearing time) was determined by analysis. Three phase bolted fault currents used in the test setup were determined by utilizing results from studies of existing AC system short circuit models. However, three phase fault currents were not required during the test since the line to ground faults did not propagate to a three phase fault. The worst case fault duration for both the line to ground and bolted three phase fault cases was determined based upon the protective relaying operation and breaker clearing times for the respective case. This duration was different depending upon which fault type (line to ground or three phase) was considered because different fault types actuate different protective relay elements.

For the Fant case (13.8kV PSW Power Delivery Feeder), the maximum line to ground fault current was provided by Duke Energy Power Delivery. Three phase bolted fault currents used in the test setup were also provided by Duke Energy Power Delivery. However, three phase fault currents were not required during the test since the line to ground faults did not propagate to a three phase fault. The worst case fault duration for both the line to ground and bolted three phase fault cases was determined based upon the protective relaying operation and breaker clearing times for the respective case. This duration was different depending upon which fault type (line to ground or three phase) was considered because different fault types actuate different protective relay elements.

For the CX case (4.16kV feeder from Oconee Breaker 1TC-04 to Keowee Auxiliaries), the maximum line to ground and symmetrical three phase bolted fault current was determined by utilizing results from studies of existing AC system short circuit models. However, three phase fault currents were not required during the test Duke Energy Responses to NRC Questions on Fault Testing February 15, 2016 Page 5 since the line to ground faults did not propagate to a three phase fault. The worst case fault duration for both the line to ground and bolted three phase fault cases was determined based upon the protective relaying operation and breaker clearing times for the respective case. This duration was different depending upon which fault type (line to ground or three phase) was considered because different fault types actuate different protective relay elements.

7. What analysis was done to ensure that each configuration bounded the worst case arc flash duration for:

a. Configuration 1?

b. Configuration 2?

c. Configuration 3?

d. Configuration 4?

Duke Energy Response:

Symmetrical current values are used in Section 5.2 of IEEE Standard 1584-2002, IEEE Guide for Performing Arc-Flash Hazard Calculations. Since a line to ground fault would produce an arc and lead to the consumption of cable materials (insulation and conductive material), this could be treated as a small arc flash. Symmetrical current values were used to determine potential cable damage effects since the damage mechanism is thermal and heat transfer occurs over time. Thus IEEE 1584-2002 is appropriate to be used as a basis for the use of symmetrical fault currents in this analysis.

A plant specific analysis was performed to determine the maximum protective relaying operating time and breaker fault clearing times for each configuration (CT-4, KPF, Fant and CX). The maximum fault clearing times were utilized as the minimum times for fault testing. The worst case fault duration for both the line to ground and bolted three phase fault cases was determined based upon the protective relaying operation and breaker clearing times for the respective case. This duration was different depending upon which fault type (line to ground or three phase) was considered because different fault types actuate different protective relay elements.

This is stated in further detail in the response to Question 6 for each configuration.

8. The "as installed" cable clamping configuration differs from that at the test laboratory. The test laboratory employed metal cable cleats designed for the forces encountered during electrical faults in accordance with IEC 61914 "Cable Cleats for Electrical Installations." The cleats were spaced at *-1 ft.

intervals and secured to a cable tray in an open environment, The installed condition used metal zip ties to strap the cables to Unistrut pegs approximately every 4 ft. in an enclosed cable raceway. How does this difference address the impact of magnetic forces resulting from a worst-case fault condition as discussed in industry standards?

Duke Energy Response:

The objective of the testing program was to determine if a phase-to-ground fault can Duke Energy Responses to NRC Questions on Fault Testing February 15, 2016 Page 6 propagate to a three-phase fault on medium voltage single conductor power cables.

Validating the cable cleat design or cleat spacing intervals was not in the testing scope. The cable cleats were used as fixtures to ensure the faulted cable was held in close contact with the adjacent cables, thus creating a test configuration that was conducive for a multi-phase fault to occur.

9. How does the use of new cables compare to the "as installed" cables, which can be in a more degraded condition due to variations in ambient conditions (temperature, moisture etc.), electrical transients, and variations in current flow?

Duke Energy Response:

All of the tested power cables are the same design as those procured for the Keowee underground path replacement in the 2001-2002 timeframe or were procured and installed as part of the Protected Service Water Project in 2009-2013.

The "as-installed" power cables were evaluated in accordance with the Oconee Aging Management Program. The cables are not subject to adverse localized environments such as significant moisture and voltage, as defined by the Oconee Insulated Cables and Connections Aging Management Program.

Additionally, for the 4.16 kV CX circuit, a recent station modification spliced a section of three-conductor cable to the existing single conductor cables which were installed in 2001-2002. Post-modification electrical testing of the cables was performed using VLF/Tan Delta and Partial Discharge testing. The results of the testing indicated that the CX cables are in good condition in accordance with Duke Energy Cable Aging Management Program guidelines.

The CX circuit has been in operation for over 12 years and is normally energized and carrying a load. Any aging-related environmental or operating stressors would have been identified by this test but none were found. Therefore, the condition of the tested cable reflects the condition of the "as-installed" cables.

a. Is the assumption for a limiting condition single phase to ground fault appropriate for cables? What quality standards addressed this?

Duke Energy Response:

Yes. The testing program confirmed that that a single phase-to-ground fault in one cable would not propagate to a multi-phase fault using the test parameters specified in Reference 8.12. The quality standards used are referenced in the response to Questions 1, 3, and 10.

Duke Energy Responses to NRC Questions on Fault Testing February 15, 2016 Page 7

10. How were the configuration differences ("as-tested" vs "as installed")

analyzed? What quality standards were used for this analysis?

Duke Energy Response:

The entire testing program, including cable mounting, isolating the tray from ground, fault current development, fault orientation and cleating of cables, was designed to create an environment that was conducive for an initial phase-to-ground fault to propagate to a three-phase fault.

Reference 8.12 established the electrical testing criteria. The tests were conducted under continuous Duke Energy engineering supervision.

11. What effects did the test configuration have on the test results? (i.e. the power cables were open-circuited and no operating loads were used for AC or DC)

Duke Energy Response:

Adding operating load to the cables would increase the pre-fault temperature of the cable conductors and metallic shields. For the resistance grounded power sources, the very low magnitude phase-to-ground fault currents presented no challenge to the thermal capacity of the bronze tape metallic shield and thus pre-heating would not change the test results. For the tests with solidly grounded power sources and high magnitude phase-to-ground fault currents, the extreme temperature generated by the arcing fault resulted in vaporization of the bronze tape. Elevating the pre-fault conductors from ambient to operating temperature would have had inconsequential effects on the degree of bronze tape damage.

The power cables are AC type circuits therefore loading with DC was not a tested parameter. The I&C cables were tested in an open-circuit configuration and connected to relays. The results were determined to not adversely affect equipment operation.

12. How would the inductive and capacitive coupling effects be influenced when current is present on all phases of the power cables and the DC cables are energized? What quality standards were used to address this aspect?

Duke Energy Response:

Data recorded during the test indicated a small induced voltage on the control and instrumentation cables. Test data also shows that the voltage was not relatively impacted by the radial separation of the power and l&C cables. The induced noise was being significantly driven by inductive and capacitive coupling; these parameters would have more strongly correlated to measured voltage changes. The energization of the DC circuits lacked sufficient current to change the results.

A third party FEA (finite element analysis) was performed to determine any potentially adverse impacts of an induced voltage during postulated short circuit events, The two voltages of interest included the maximum differential voltage (the Duke Energy Responses to NRC Questions on Fault Testing February 15, 2016 Page 8 voltage developed across two control cable conductors) and the maximum common mode voltage (the voltage developed on any control cable conductor relative to ground). The maximum differential voltage was studied to determine the potential for spurious actuation or drop of control circuits for the Keowee emergency power system. The maximum differential voltage was studied to determine the potential for insulation system breakdown/failures relative to station ground (control cable armor, terminal blocks and other system components).

This analysis includes several known conservatisms, including use of the most limiting cable separation (set to minimum allowable by the duct geometry) and intra-cable bundle low voltage conductor separation (set to the maximum allowable by the cable bundle diameter), neglecting control conductor helical twist along the length of the cable (has the effect of canceling magnetic coupling), and a conservatively high line-ground fault current (16,000 Amps-peak).

This analysis specifically studied the configuration of the 13.8kV Fant to PSW feeder parallel to the control cables from Manhole 6 to the PSW building. The assumption of a I16kA peak line to ground fault current well bounds the actual available fault current of 5.66kA (per Reference 8.12) for this configuration.

The analysis determined that the differential voltage induced on a galvanized steel interlocked armor (GSIA) control cable conductor pair is less than one volt. Since this differential voltage is so minute, there is no concern for meal-operation of the control circuits for the Keowee emergency power system.

The analysis further determined that the common mode voltage is approximately 14 volts for the armored control cable case. A common mode voltage of 14 volts is well within the insulation system of the Keowee emergency power system components (cables, relays, etc.).

Small voltage changes (less than 1 volt across conductors (differential) and 14 volts relative to ground (common mode)) would have an insignificant impact on an ungrounded 125VDC control system such as the Keowee Emergency Start circuit(s).

These are relatively small changes relative to the system limits. For instance, I volt differential is less than 1 percent of the rating (125VDC) of the connected emergency start relays and would not interfere with the emergency start or operation of a Keowee unit.

The common mode voltage is an insulation (relative to ground) rating. A common mode voltage of 14 volts is only 11 percent more than the rating of the most limiting device relative to ground (14/125VDC for emergency start relays). Typically cable systems, terminal blocks, etc. are rated for at least 150 percent of their nominal rating (i.e., 1000Vac cable in a 600Vac application) relative to ground.

This analysis relied heavily upon methods originally developed by Edward Rosa (Volume 4, Number 2, Bulletin of the Bureau of Standards, The Self and Mutual Inductances of Linear Conductors) and summarized more recently by Clayton Paul in IEEE EMC Society Magazine.

Duke Energy Responses to NRC Questions on Fault Testing February 15, 2016 Page 9 The 3rd party analysis was reviewed per Reference 8.12 which was generated as a QA-1 document in accordance with Duke Energy internal Design Analysis Procedures and the Duke Energy Quality Assurance Program

a. Were the concerns presented in Annex B of IEEE 603 investigated in relation to this question?

Duke Energy Response:

Yes. Annex B (informative) of IEEE 603-2009 (IEEE Standard Criteria for Safety Systems for Nuclear Power Generating Stations), Electromagnetic Compatibility, provides some discussion on the sensitivity and vulnerability of components to electromagnetic interference (EMI). This annex is concerned with incorrect component operation or damage due to EMI.

This standard recommends shielding as a mitigation strategy for EMI. Section B.3.1, Evaluation of the electromagnetic environment, states that, "Performance may be demonstrated by a combination of testing, analysis, or documented operating experience." Analysis was performed (see the response to question 12 above) and it was determined that EMI would not cause incorrect component operation or damage.

b. Has the impact of increased cable length been investigated (i.e., as installed (4000 ft.) vs. as-tested (12 ft.))?

Duke Energy Response:

Yes. An Engineering Change has removed from service all Instrumentation and Control (l&C) in Trench 3 (approximately 4,000 feet in length) therefore the potential for induced voltage on I&C cables in Trench 3 has been eliminated.

Additionally, Reference 8.12 has evaluated other configurations of power and l&C cables and determined that the calculated levels of induced voltage on l&C cables would not adversely affect equipment operation.

13. In some of the cable tests observed, the bronze tape shield melted partially.

Has Duke evaluated the impact of such melting, if a worst-case fault is postulated?

Duke Energy Response:

Yes. The tests with solidly grounded power sources resulted in high magnitude phase-to-ground faults. The high temperature developed by the arcing fault exceeded the melting point of bronze and it was expected that the bronze tape shield would be damaged at the area surrounding the fault location. For all tests, the remaining bronze shield provided a conductive path to ground for the entire fault duration; therefore, melting did not impact the test results.

As documented in the responses to Questions 3-11, the cable tests were conducted with the power sources configured to provide worst-case fault currents. As such, the Duke Energy Responses to NRC Questions on Fault Testing February 15, 2016 Page 10 observed melting of the bronze tape represented the "worst-case" fault currents and the impact has been evaluated as stated in the above paragraph.

14. Did Duke calculate the maximum magnetic force that would be exerted in the raceway system to CT-4, which has approximately 4,000 feet of cable? The NRC staff's review of industry guidance indicates that cables in the concrete trench could be exposed to a substantial amount of force. It did not appear that these effects were simulated in the cable testing. Ifthese magnetic forces were not modeled in the tests, how did the testing performed demonstrate that the existing cable configuration meets the ONS licensing basis and applicable ANSI standards?

Duke Energy Response:

The maximum magnetic force was not calculated for the CT-4 raceway system since it was not a tested parameter. The purpose of the testing program was to evaluate if an initial phase-to-ground fault would propagate to a three-phase fault. For a three-phase fault, the test cables would have experienced magnetic forces; however, three-phase faults did not occur during the tests rendering this question moot.

Duke Enerav Responses to Follow-up NRC Questions on Fault Testing:

1. Why was 12 gauge wire used in tests 3 &4?

Duke Energy Response:

The laboratory power sources for tests with solidly grounded power sources provided phase-to-ground fault currents in the kA range. The #12 AWG copper wire was selected to ensure that the fault lasted for the required duration.

2. Demonstrate why the current was not split [between the three cables]?

Duke Energy Response:

On each cable end, the metallic shield is attached to ground straps that are connected to a common grounding point. The cable conductor end connected to the power source is the line side and the other end, which is not connected, is designated the load side.

When the faulted cable is energized, the entire current of the faulted cable flows from the line side to the fault location and then splits, in proportion to the combination of the shield and ground path impedance, back along the faulted cable's metallic shield to the line and load side where the fault current is conducted to ground. The metallic shield of the faulted cable carries all the fault current.

The other non-faulted phase conductor shields are also connected to a common Duke Energy Responses to NRC Questions on Fault Testing February 15, 2016 Page 11 grounding point. Depending on the overall impedance characteristics of the multiple paths to ground, after the current exits the faulted cable metallic shield, it may have been possible for the metallic shields of the non-faulted conductors (connected in parallel with the faulted cable metallic shield) to carry some degree of the fault current back to ground.

Since there were no three-phase faults, the non-faulted cable conductors were still electrically isolated from the faulted cable conductor and there was no sharing of current between the faulted and non-faulted cable conductors [see Sketch].

3. What quality standards were used to determine this [re: Question 2 above]?

Duke Energy Response:

A review of Attachment 7 in Reference 8.12 was performed of the schematics and electrical test data for each of the four test configurations. Next, electrical analysis principles and knowledge of medium voltage cable design was applied to reach the conclusions provided in the response to Follow-up Question 2.

Duke Energy Responses to NRC Questions on Fault Testing February 15, 2016 Page 12 Sketch If1 If2 4-

- B.

S S

~se

• S

• S

• S S S 1'

~se

• S S

S S

S S S

• -. -. -. ----- 4 S S I-A-Ph*~se SW

• S S S

• S S S S S

• S

'fi ~YS S

S H,

S S

I2 S S S S S S

• S If If=lIfl+If2 Cable Conductor

.....

• Metallic Shield Conductors to ground

- Fault Current Path to Ground

ENERGY Scott L.Batson Vice President Oconee Nuclear Station Duke Energy ONO1VP I 7800 Rochester Hwy Seneca, SC 29672 0: 864,873,3274 f.864.873.4208 Scott Batson@duke-energy.com ONS-201 6-017 10 CFR 50.55a(z)(1)

February 15, 2016 Mr. William M. Dean, Director, Office of Nuclear Reactor Regulation U. S. Nuclear Regulatory Commission Washington, DC 20555-0001

SUBJECT:

Duke Energy Carolinas, LLC Oconee Nuclear Station Unit Nos. 1, 2, and 3, Renewed Facility Operating License Numbers DPR-38, 47, and 55, Docket Nos. 50-269, 50-270, and 50-287; Request for Alternative to Codes and Standards Requirements pursuant to 10 CFR 50.55a(z) to satisfy 10 CFR 50.55a(h)(2) associated with Bronze Tape Wrapped Emergency Power Cables in Use at the Oconee Nuclear Station

References:

1. NRC Letter to Duke Energy, "Oconee Nuclear Station - NRC Component Design Bases Inspection Report 05000269/2014007, 05000270/2014007, and 05000287/2014007," dated June 27, 2014, (ADAMS Accession No. ML14178A535).

2. NRC Memorandum, Director of Reactor Safety to Deputy Director, Division of Policy and Rulemaking, Office of Nuclear Reactor Regulation, "Request for Technical Assistance Regarding Oconee Nuclear Station Design Analysis for Single Failure and the Integration of Class 1E Direct Current Control Cabling in Raceways With High Energy Power Cabling (TIA 20 14-05),"' dated October 16, 2014, (ADAMS Accession No. ML14290A136).

3. Emails from Mr. Randy Hall, Senior Project Manager, Office of Nuclear Reactor Regulation, to Mr. Chris Wasik, Duke Energy Manager of Regulatory Affairs, Oconee Nuclear Station, dated December 8 and 10, 2015.

A June 27, 2014 NRC letter (Reference 1) identified an Unresolved Item (URI) documenting an NRC inspection issue that involves the possibility of ground faults in medium voltage power cabling located in an underground concrete raceway1 at the Oconee Nuclear Station (ONS) that could potentially impact control cabling required to mitigate certain design basis events. The NRC inspection team subsequently requested assistance from the Office of Nuclear Reactor The control cable circuits in the buried underground concrete raceway have been relocated and are no longer a concern; however, extent of condition issues exist in other areas which are the focus of this submittal.

U. S. Nuclear February Regulatory Commission 15, 2016 Page 2 Regulation (NRR) by means of a Task Interface Agreement (TIA), to review Oconee Nuclear Station's emergency power system licensing bases to determine the acceptability of the design (Reference 2). The TIA raised questions with the manner in which the Oconee Nuclear Station achieved the objectives for single failure design as described in IEEE Std. 279-1971.

While awaiting the outcome of a Task Interface Agreement (TIA), Duke Energy is taking several anticipatory actions to address the concerns identified in the URI. Specifically, Duke Energy has completed cable testing and some modifications, with additional modifications under development.

This letter updates the NRC on the status of these actions, and requests NRC review and approval under the provisions of 10 CER 50.55a(z)(1) for certain alternatives to codes and standards. As such, Duke Energy is requesting modification to the station's licensing basis to accept "as-is" certain plant configurations and to approve the application of IEEE Std. 384-1992, Paragraph 6.1.4 in specific locations. Duke Energy considers the proposed alternatives represent an acceptable level of quality and safety for the ONS design and plant configuration.

The technical bases for the 10 CFR 50.55(a)(z)(1) request are contained in the Enclosure to this letter.

Regulatory commitments associated with this request are summarized in the Attachment 1. contains Duke Energy's responses to NRC emailed questions on cable fault testing (Reference 3).

If you should have any questions regarding this submittal, please contact Stephen C. Newman, Lead Nuclear Engineer, Regulatory Affairs, at (864) 873-4388.

Sincerely, Scott L. Batson Vice President Oconee Nuclear Station Enclosure Attachments

U. S. Nuclear Regulatory Commission February 15, 2016 Page 3 xc (w/enclosure/attachments):

Ms. Catherine Haney, Administrator, Region II U.S. Nuclear Regulatory Commission Marquis One Tower 245 Peachtree Center Ave., NE, Suite 1200 Atlanta, GA 30303-1 257 Mr. James R. Hall Senior Project Manager (by electronic mail only)

Office of Nuclear Reactor Regulation U.S. Nuclear Regulatory Commission 11555 Rockville Pike Mail Stop O-8G9A Rockville, Maryland 20852 Mr. Jeffrey. A. Whited Project Manager (by electronic mail only)

Office of Nuclear Reactor Regulation U.S. Nuclear Regulatory Commission 11555 Rockville Pike Mail Stop O-8G9A Rockville, Maryland 20852 Mr. Eddy Crowe NRC Senior Resident Inspector Oconee Nuclear Station

ENCLOSURE Evaluation of Proposed Change

Subject:

PROPOSED ALTERNATIVE INACCORDANCE WITH 10 CFR 50.55a(z)(1)

1. Systems/Components Affected

2. Applicable Regulatory Requirement

3. Reason for Request

4. Proposed Alternative and Basis for Use

5. Cable Testing

6. Risk Insights

7. Precedents

8. References

Enclosure ONS-201 6-017 February 15, 2016 Page 2

1. SYSTEMS/COMPONENTS AFFECTED This request pertains to medium voltage single conductor bronze armor cables related to operation of the Keowee Hydroelectric Stations' (KHS) 13.8 kV and 4.16 kV underground power paths and the 13.8 kV Protected Service Water (PSW) power paths from KHS and the PSW substation (i.e., the "Fant Line"), and areas where they are routed in proximity to certain Keowee safety-related control cables.

Specifically, the medium voltage power cables affected are the:

• Six (6) 13.8 kV Keowee Underground (KUG) feeder cables to CT-4;

• Three (3) 4.16 kV KHS Auxiliary CX transformer power feeder cables,

• Six (6) 13.8 kV B6T and B7T Feeder cables from the KHS to the PSW switchgear building;

• Six (6) 13.8 kV offsite "Fant Line" power feeder cables to the PSW switchgear building; and

• Six (6) 4.16 kV feeder cables from PSW Switchgear B6T to 600 VAC PSW Loadcenter PXI13 transformer.

The specific areas 1 affected associated with this request include the:

• PSW System ductbank manholes2,

• KHS Mechanical Equipment Gallery,

• P8W Building Cable Spreading Area.

A description of each cable's composition and its application at the Oconee Nuclear Station is given below:

Cable Type 1:

• Cable

Description:

Single conductor cable, 750 kcmil conductor, 260 mils insulation with two overlapping 10 mui layers of bronze armor shielding.

• Used On: KHS underground path to the P8W switchgear building; Fant underground (PSW ductbank) path from PSW manhole 6 to the P8W switchgear; P8W 4 kV switchgear to 600 V P8W Load Center.

• System Nominal Operating Voltage: 13.8 kV phase-to-phase or 8 kV phase-to-ground and 4.16 kV phase-to-phase or 2.4 kV phase-to-ground.

SThe control cable circuits inthe buried underground concrete trench (i~e., Trench 3) have been relocated and are no longer included as part of this evaluation.

2Wihnthe PSW ductbank, power and control cables are routed through separate concrete encased conduits. The area of interest is with the power/control cables located inthe PSW ductbank manholes (along the PSW ductbank) which are not contained inindividual conduits.

Enclosure ON S-201 6-017 February 15, 2016 Page 3 Cable Tvpe 2:

• Cable

Description:

Single conductor cable, 250 kcmil conductor, 140 mils insulation with two overlapping 10 mil layers of bronze armor shielding.

• Used On: Underground (Trench 3) path plant feed to KHS station service transformer CX from ONS Unit 1.

• System Nominal Operating Voltage: 4.16 kV phase-to-phase or 2.4 kV phase-to-ground.

2. APPLICABLE REGULATORY REQUIREMENT 10 CER 50.55a(h)(2), "ProtectionSystems," states: "For nuclear power plants with construction permits issued after January 1, 1971, but before May 13, 1999, protection systems must meet the requirements in IEEE Std. 279-1968, "Proposed IEEE Criteria for Nuclear Power Plant Protection Systems," or the requirements in IEEE Std. 279-1971, "Criteria for Protection Systems for Nuclear Power Generating Stations," or the requirements in IEEE Std. 603-1991,"Criteria for Safety Systems for Nuclear Power Generating Stations, and the correction sheet dated January 30, 1995. For nuclear power plants with construction permits issued before January 1, 1971, protection systems must be consistent with their licensing basis or may meet the requirements of IEEE Std. 603-1991 and the correction sheet dated January 30, 1995."

Although the ONS construction permit was issued prior to January 1, 1971, the current licensing bases is that ONS will satisfy Section 4.2 of IEEE Std. 279-1971 for the Oconee Emergency Power and Emergency Core Cooling systems. NRC acceptance of the IEEE Std. 279-1971 single failure criteria is documented in three (3) 1976 safety evaluations associated with changes to the Emergency Core Cooling System model which conformed to the requirements of 10 CFR 50.46. IEEE Std. 279-1971, Section 4.2, "Single Failure Criterion," requires, in part, that any one single failure within the protection system shall not prevent the proper protective action at the system level when required (Reference 8.2). The potential for not satisfying the single failure criteria (based on the requirements of 10 CFR 50.55a(h)) due to interactions between cables in certain cable transition areas is the specific issue addressed by this request. Options available to Duke Energy under this circumstance include 10 CFR 50.55a(z).

10 CFR 50.55a(z) states:

"Alternatives to the requirements of paragraphs (b) through (h) of this section or portions thereof may be used when authorized by the Director, Office of Nuclear Reactor Regulation, or Director, Office of New Reactors, as appropriate. A proposed alternative must be submitted and authorized prior to implementation. The applicant or licensee must demonstrate that:

(1) Acceptable level of quality and safety. The proposed alternative would provide an acceptable level of quality and safety; or

Enclosure ONS-2016-017 February 15, 2016 Page 4 (2) Hardship without a compensating increase in quality and safety. Compliance with the specified requirements of this section would result in hardship or unusual difficulty without a compensating increase in the level of quality and safety."

This request seeks NRC approval in accordance with 10 CFR 50.55a(z)(1) prior to the resolution of the nonconforming conditions initiated in responses to the Unresolved Item discussed in Section 3 shown below.

3. REASON FOR REQUEST A June 27, 2014 NRC letter (Reference 8.8) identified an Unresolved Item (URI) regarding a concern that postulated short circuits and/or ground faults in electrical cabling located in an underground concrete raceway could potentially impact the functionality of the emergency power system which is required to mitigate certain design basis events. The NRC inspection team subsequently requested assistance from the Office of Nuclear Reactor Regulation (NRR) by means of a Task Interface Agreement (TIA), to review the Oconee Nuclear Station emergency power system licensing bases to determine the acceptability of the design (Reference 8.9) with respect to Oconee's current licensing basis.

Duke Energy is taking several anticipatory actions to address the concerns identified in the URI. This submittal requests NRC review and approval under the provisions of 10 CFR 50.55a(z)(1) of certain alternatives to codes and standards. Duke Energy believes that these proposed alternatives address the concerns associated with the URI and that they provide an acceptable level of quality and safety.

The previously referenced URI is associated with a single failure compliance question with IEEE Std. 279-1971. Specifically, the issue is associated with a potential of a power cable fault that causes an adverse interaction with control cables in close proximity to the faulted power cable. This request addresses the NRC concerns by proposing a two-step approach that includes "as-is" temporary and permanent modifications to the plant licensing basis. In summary, to address NRC concerns with single failure capabilities ONS requests:

1. Temporary acceptance of the current configuration in specific locations to allow for sufficient time to implement modifications; and permanent acceptance of the current configuration in certain locations based on cable design, operation and testing.

2. Acceptance of the application of Paragraph 6.1.4 of IEEE Std. 384-1992, "Limited

• Hazard Areas," as a means of providing acceptable cable separation in certain areas of the plant. The classification is appropriate because these are areas containing power cables but without any Hazard Area drivers being present (e.g.,

missiles, high energy lines etc.).

Enclosure ONS-201 6-017 February 15, 2016 Page 5 The description details and justifications for the 10 CFR 50.55(a)(z)(1) request are given below:

3.1 Description of Affected Locations As part of a 2014 Component Design Basis Inspection (CDBI), a question was raised by an inspection team whether this bronze tape on medium voltage power cables in Trench 3 can be credited as armor when evaluating failures per IEEE Std. 279-1971, Section 4.2, and thus whether credit can be taken for its single failure mitigation properties. As described below, the plant areas within the scope of this submittal are the PSW system ductbank manholes, Keowee Mechanical Equipment Gallery, and the PSW building Cable Spreading Area.

3.1 .1 PSW System Ductbank Manholes The PSW system is designed as a standby system for use under emergency conditions. The PSW system provides added "defense in-depth" protection by serving as a backup to certain existing safety systems. The PSW system is provided as an alternate means to achieve and maintain safe shutdown conditions for one, two or three units following certain postulated scenarios. The PSW system also reduces fire risk by providing a diverse power supply to power safe shutdown equipment in accordance with the National Fire Protection Association (NFPA) 805 safe shutdown analyses.

As noted above, the defense-in-depth function of the PSW system provides a diverse means to achieve and maintain safe shutdown by providing secondary side decay heat removal, Reactor Coolant System (RCS) pump seal cooling, RCS primary inventory control, and RCS boration for reactivity management following plant scenarios that disable the 4.16 kV essential electrical power distribution system. The PSW electrical system is designed to provide power to PSW mechanical and electrical components as well as other system components needed to establish and maintain a safe shutdown condition. The system is designed to supply the necessary loads and is electrically independent from the Oconee station electrical distribution system and the 8SF. No credit is taken in the safety analyses for PSW system operation following design basis events. The PSW System can be supplied power from an offsite transmission feed or the KHS.

A separate PSW electrical equipment structure (PSW Switchgear Building) is provided for major P8W electrical equipment. Normal power is provided from the Central Tie Switchyard via a 100 kV transmission line to a 100/13.8 kV PSW substation located adjacent to Oconee Nuclear Station and then via a 13.8 kV feeder that enters the P8W system ductbank at Manhole 6 leading to the P8W building via a ductbank.

Although the power path from the Central Tie Switchyard to the PSW Switchgear Building is classified as non QA-1, the cable in MH-6 to the P8W building was procured to QA-1 standards.

Enclosure ONS-201 6-017 February 15, 2016 Page 6 Alternate QA-1 power is available from each Keowee Hydroelectric Generating unit as an electrical source to the PSW building. The route from KHS to P8W consists of an underground 13.8 kV power cable feeder connecting Keowee output breakers KPF-11I and KPF-1 2 located in KHS to transformers CT6 and CT7 located in the P8W building. The Keowee switchgear circuit breaker and bus arrangement provides the capability of aligning either the Keowee Unit I or Unit 2 generators to the CT6 and/or CT7 transformers.

The KHS to P8W 13.8 kV power feed initially routes from Keowee through an underground trench (along with the CT-4 underground feeder), and then diverts into a separate ductbank / manhole system before reaching the CT-4 blockhouse. The PSW ductbank system from the underground trench to the PSW building consists of underground duct with six intervening manholes. The underground duct segments connecting the manholes consist of separate PVC conduits surrounded by concrete fill. The manholes are designed with the control cables routed across the bottom in a cable tray and the power cables racked above the floor on separate supports. The manhole closest to the underground trench is designated Manhole-I (MH-1) and the manhole closest to the PSW building is designated MH-6. The KHS to PSW 13.8 kV power feed is the Technical Specification credited powerpath but it is not the normal power feed for this system. These cables are typically energized only during PSW system powerpath surveillance testing that is performed on a quarterly basis (-,33 hour3.819444e-4 days <br />0.00917 hours <br />5.456349e-5 weeks <br />1.25565e-5 months <br />s/year total).

The normal power feed to the PSW system originates from the 100 kV Central Tie Switchyard (i.e., the "Fant Line") which is independent of the station switchyard. The Fant Line is routed to ONS by uninsulated overhead distribution lines that terminate at two dip-poles. At each dip-pole, the 13.8 kV P8W power feed transitions to two circuits consisting of three insulated single-conductor power cables for each circuit routed underground. The underground installation continues to a splice box outside MH-6. The route exiting the splice box to the P8W switchgear transitions to QA-1 cabling then routes through MH-6 and the remaining portion of the ductbank system to the P8W building. To improve separation between the Fant line power feed and control cables in MH-6, a modification is underway to relocate this (normally energized) power feed out of MH-6 and the existing ductbank and reroute it to the PSW switchgear building via a new duct. Approval of this modification is not part of this submittal.

In addition to the power feeds, the P8W ductbank system contains low voltage Instrumentation & Control (I&C) cables consisting of supervisory functions for both KHUs, one train of KHU emergency start (safety-related), one train switchyard isolation complete (safety-related), PCB-9 control, and P8W KPF breaker control.

Enclosure ONS-201 6-017 February 15, 2016 Page 7 Power and control cable routing in the PSW ductbank manholes is consistent with Duke Energy design specifications (Reference 8.13).

3.1.2 KHS Mechanical Equipment Gallery The KHS Mechanical Equipment gallery contains motor control centers, cooling water strainers, governors, and Keowee Power Feeder (KPF)

Switchgear KPF-1 and KPF-2. In addition, cabling for the CT-4 underground feeder, the KHS to PSW underground feeder, the KHS to PSW Switchgear (KPF) line side cable bus, the feeder from switchgear 1TC to transformer CX, and adjacent safety-related control cables, all route through the area.

Power and control cable routing in the KHS Mechanical Equipment Gallery is consistent with Duke Energy design specifications (Reference 8.13).

3.1.3 PSW Buildingq Cable Sp~readinaq Area The PSW building Cable Spreading Area is a section of the PSW switchgear building into which the cables discussed in the previous sections enter from the PSW ductbank system. In addition, there are low voltage I&C cables consisting of supervisory functions for both KHUs, one train of KHU emergency start cables (safety-related), one train switchyard isolation complete (safety-related), PCB-9 control, and breaker control for the PSW KPF switchgear. This area also contains PSW power and control cables.

Power and control cable routing in the PSW Building Cable Spreading Area is consistent with Duke Energy design specifications (Reference 8.13).

Enclosure ONS-2016-01 7 February 15, 2016 Page 8

4. PROPOSED ALTERNATIVE AND BASIS FOR USE This request is specific to the applications identified herein and is not intended to be a blanket request for approval of the use of IEEE Std. 384-1992 at ONS.

4.1 PSW System Ductbank Manholes 4 Current Configquration:

The KHS to P8W 13.8 kV power feed initially routes from Keowee through an underground trench and then diverts into a separate ductbank/manhole system before reaching the CT-4 blockhouse. The PSW ductbank system from the underground trench to the PSW building consists of an underground duct with six intervening manholes. The manhole closest to the underground trench is designated Manhole-I (MH-1) and the one closest to the PSW switchgear building is designated Manhole-6 (MH-6). The feed from the KHS to PSW is normally not energized.

In addition to the 13.8 kV power feed from the KHS to the PSW building, MH-6 also contains the normally energized 13.8 kV power feed (i.e., the "Fant Line")

from the Central Tie Switchyard to the PSW switchgear building via the PSW substation.

The PSW ductbank system also contains low voltage Instruments & Controls (l&C) cables routed in separate ductbank conduits consisting of supervisory functions for both KHUs, one train of KHU emergency start, one train switchyard isolation complete, PCB-9 control, and breaker control for the PSW KPF switchgear. In addition, future station changes may route a second channel of KHU Emergency Start and Switchyard Isolation Complete via the PSW ductbank.

Proposed Alternate Method:

Pursuant to 10 CER 50.55a(z), Duke Energy requests:

1. For manhole 6 (MH-6), Duke Energy requests temporary acceptance of the "as-is"~configuration of the Fant line feeder and adjacent KHS control cables as an alternative to meeting the requirements of 10 CFR 50.55a(h)(2) until the Fant line power feeder relocation modification is complete. The modification will be completed no later than September 15, 2017.

2. For manholes I through 6, Duke Energy requests to modify the ONS licensing basis pursuant to 10 CFR 50.55a(z) to allow acceptance of the "as-is" configuration of the normally de-energized 13.8 kV KHS to the PSW 4The KHS to PSW power cables inthe PSW ductbank are routed through individual concrete encased conduits that are separate from the I&C cables and conduit within the same ductbank. The area of concern for interaction iswithin the PSW ductbank manholes, where power cables and l&C cables are not contained within individual conduits.

Enclosure ONS-201 6-017 February 15, 2016 Page 9 building power feed as an alternative configuration to meeting the requirements of 10 CER 50.55a(h)(2).

An acceptable level of quality and safety is demonstrated via the followingq:

• satisfactory cable crush test results (refer to Section 5),

• satisfactory cable fault test results (refer to Section 5),

• the potential for power cable to control cable interaction in the manholes represents a small portion of the overall cable run total (-,180 feet out of

-,4500 feet),

• robust power cable design and cables procured to QA-1 standards which minimizes the likelihood of cable interactions,

• the KHS to PSW power cables are not normally energized which minimizes the likelihood of cable interactions; for a PSW event, these cables would only be energized if the Fant Line is unavailable,

• high impedance grounding system limits fault current (KHS-PSW feeder) and minimizes the effect of any cable interaction should a fault occur,

• Fant line substation breaker testing is operated every 12 months,

• station power cables are evaluated as described in UFSAR Section 18.3.14, "Insulated Cables and Connections Aging Management Program,"

and

• the cables are housed in a steel-reinforced concrete ductbank/manhole engineered to withstand earthquakes, tornado missiles, and to minimize water entry.

4.2 KHS Mechanical Equipment Gallery Current Configquration:

The KHS Mechanical Equipment gallery contains motor control centers, cooling water strainers, governors, and the Keowee Power Feeder (KPF) Switchgear KPF-1 and KPF-2. In addition, cabling for the CT-4 underground feeder, the KHS to PSW underground feeder, the KHS to P8W Switchgear (KPF) line side cable bus, the feeder from switchgear ITC to transformer CX, and adjacent Keowee unit control cables all route through the area.

Proposed Alternate Method:

1. Modification of the ONS licensing basis pursuant to 10 CFR 50.55a(z) to allow acceptance of a proposed modification to the separation requirements for a Limited Hazard Area as noted by IEEE Std. 384-1992, Paragraph 6.1.4 with respect to the CX auxiliary power feed to the KHS, the KHS underground emergency power feeder to CT-4, the P8W KPF switchgear line side cable bus, and adjacent control cables, is requested as an alternative configuration to meeting the requirements of 10 CFR 50.55a(h)(2).

Enclosure ONS-201 6-017 February 15, 2016 Page 10 Adoption of the IEEE Std. 384-1992, Limited Hazard Area classification, is due to the area containing power cables without any Hazard Area drivers present (e.g., missiles, high energy lines etc.). The incorp)oration of this standard into the licensing basis is limited to Paragraph 6.1.4 and is only for the scope of equipment specified in this section.

Following guidance from IEEE Std. 384-1992, the openly routed medium voltage bronze armor power cables and low voltage Keowee control cables require separation of three feet horizontally and five feet vertically, which is not achieved at all locations in the Keowee Mechanical Equipment Gallery. Where open distance is not achieved, enclosures are or will be provided for the medium voltage power cables and low voltage control cables.

By providing a fully metallic enclosed raceway to meet the enclosed raceway separation distance, the requirement of the IEEE Std. 384-1992 standard is met and sufficient physical separation between the power and control circuits is achieved. By fully and separately enclosing the medium voltage power cables and the low voltage control cables, the IEEE Std. 384-1992 required separation distance is reduced to one inch in each direction, which is maintained with the proposed design.

2. Until modifications are completed, temporary acceptance pursuant to 10 CFR 50.55a(z) of the "as-is" configuration in the KHS Mechanical Equipment Gallery as an alternative configuration to meeting the requirements of 10 CFR 50.55a(h)(2) until the above proposed modification is complete. Modification completion is to be no later than September 15, 2017.

3. Modification of the ONS licensing basis pursuant to 10 CER 50.55a(z) to allow acceptance of the "as-is" configuration of the normally de-energized 13.8 kV power feed from the KHS to the PSW building as an alternative configuration to meeting the requirements of 10 CFR 50.55a(h)(2).

An acceptable level of quality and safety is demonstrated via the followingq:

• satisfactory cable crush test results (refer to Section 5),

• satisfactory cable fault test results (refer to Section 5),

• modifications to meet enclosed raceway separation requirements for a Limited Hazard Area as noted by IEEE Std. 384-1992, paragraph 6.1.4, "Limited Hazard Area of IEEE Std 384-1992," (Reference 8.3)[as endorsed in RG 1.75].

• robust power cable design and cables procured to QA-1 standards which minimizes the likelihood of cable interactions,

• KHS power cables (CT-4 and KPF) are not normally energized which minimizes the likelihood of cable interactions; for a PSW event, these cables would only be energized if the Fant Line is unavailable.

• high impedance grounding system limits fault current (KPF and CT-4) and minimizes the effect of any cable interaction should a fault occur,

Enclosure ONS-2016-017 February 15, 2016 Page 11

• limited exposure distance (approximately 100 feet) which minimizes the opportunity of cable interactions,

• station power cables are evaluated as described in UFSAR Section 18.3.14, "Insulated Cables and Connections Aging Management Program,"

and

• the cables are protected from the environment in that they are in the KHS powerhouse and not exposed to environmental hazards.

4.3 PSW Building Cable Spreading Area Current Configquration:

The PSW Building Cable Spreading Area is a section of the P8W building into which the cables discussed in the previous sections enter from the PSW ductbank system. In addition to the power feeds, low voltage I&C cables consisting of supervisory functions for both KHUs, one train of KHU emergency start cables, one train switchyard isolation complete, PCB-9 control, and breaker control for the PSW KPF switchgear also enter the PSW building via these ductbanks. Additionally, power and control cables for other PSW functions are present.

Proposed Alternate Method:

1. Modify the ONS licensing basis pursuant to 10 CER 50.55a(z) to accept a proposed modification to meet the separation requirements for a Limited Hazard Area as noted by IEEE Std. 384-1992, Paragraph 6.1.4, with respect to the normally energized Fant line power supply feeder, normally energized feeder from switchgear B6T to PXI13 transformer, and adjacent KHS control cables as an alternative configuration to meeting the requirements of 10 CFR 50.55a(h)(2).

Adoption of the IEEE Std. 384-1992, Limited Hazard Area classification, is due to the area containing power cables without any Hazard Area drivers present (e.g., missiles, high energy lines etc.). The incorporation of this standard into the licensing basis is limited to Paragraph 6.1.4 and is only for the scope of equipment specified in this section.

Following guidance from IEEE Std. 384-1992, the openly routed medium voltage power cables and low voltage control cables require separation of three feet horizontally and five feet vertically, which is not achieved at all locations in the PSW cable spreading area. Where open distance is not achieved, enclosures are or will be provided for the medium voltage power bronze armor cables and low voltage control cables.

By providing a fully metallic enclosed raceway to meet the enclosed raceway separation distance, the requirement of the IEEE Std. 384-1992 standard is met and sufficient physical separation between the power and control circuits is achieved. By fully and separately enclosing the medium voltage power cables and the low voltage control cables, the IEEE Std. 384-1992

Enclosure ONS-201 6-017 February 15, 2016 Page 12 required separation distance is reduced to one inch in each direction, which is maintained with the proposed design. Adoption of IEEE Std. 384-1992 is limited to Paragraph 6.1.4 and is only for the scope of equipment specified in this section.

2. Modify the ONS licensing basis pursuant to 10 CFR 50.55a(z) to allow acceptance of the as-is configuration of the normally de-energized 13.8 kV power feed from KHS to the PSW building as an alternative configuration to meeting the requirements of 10 CER 50.55a(h)(2).

3. Pursuant to 10 CFR 50.55a(z), until modifications are completed, grant temporary acceptance of the "as-is" configuration regarding the Fant line feeder, the normally energized feeder from switchgear B6T to PX1 3 transformer, and adjacent KHS control cables is requested as an alternative to meeting the requirements of 10 CER 50.55a(h)(2) until the above proposed modification is complete. Modification completion is to be no later than September 15, 2017.

An acceptable level of quality and safety is demonstrated via the followingq:

• satisfactory cable crush test results (refer to Section 5),

• satisfactory cable fault test results (refer to Section 5),

• modifications to meet enclosed raceway separation requirements for a Limited Hazard Area as noted by IEEE Std. 384-1992, paragraph 6.1.4, (Reference 8.3)[as endorsed in RG 1.75],

• robust power cable design and cables procured to QA-1 standards which minimizes the likelihood of cable interactions,

• KHS power cables are not normally energized which minimizes the likelihood of cable interactions; for a PSW event, these cables would only be energized if the Fant Line is unavailable

• high impedance grounding system limits fault current (KHS-PSW feeder) which minimizes the likelihood of cable interactions,

• Limited exposure distance (approximately 90 feet),

• station power cables are evaluated, as described in UFSAR Section 18.3.14, "Insulated Cables and Connections Aging Management Program," and

• the cables are housed in a protected area engineered to withstand earthquakes, tornado missiles, and to minimize water entry.

5. CABLE TESTING Duke Energy conducted testing to validate that the bronze armored emergency power cable design provides an acceptable level of quality and safety. This testing was conducted in two phases, cable crush testing and cable fault testing. Test results demonstrate the cables provide an acceptable level of quality and safety and that a single failure of a bronze armored medium voltage power cable will not result in a consequential loss of safety functions performed by adjacent low voltage control cables.

Enclosure ONS-201 6-017 February 15, 2016 Page 13 Phase One - Cable Crush Testingq The first phase performed testing based on Underwriters Laboratory (UL)1569 'Metal Clad Cables' crush and impact tests of the medium voltage cables and was completed in February 2015. The phase one testing compared the bronze armor cables to galvanized steel interlocked armored cables with respect to their physical protection based on UL1 569 testing sections 24 (impact testing), 25 (increasing crush), and 26 (direct burial crush). All of the cable types tested, including the medium voltage cables installed at the ONS, confirmed that the cable configuration with bronze armor provides adequate protection to perform consistently with armored cable based on UL 1569, Sections 24, 25, and 26. The results of this testing were provided to the NRC on May 11, 2015 (Reference 8.5).

Phase Two - Cable Fault Testingq The second phase of the testing involved inducing a fault on a single conductor medium voltage bronze armor power cable while monitoring for any effects on adjacent power and control cables. The fault testing (described below) further validates the engineering analyses which have concluded that the power cables subject to a single phase-to-ground fault will not propagate to a multi-phase fault and will not adversely interact with the low voltage control cables leading to consequential functional failures of redundant trains.

From November 2-6, 2015, a series of cable fault tests were conducted for Duke Energy at KEMA Laboratories in Pennsylvania. These tests were also observed by NRC Staff. Duke Energy commissioned the tests to determine the potential impacts of electrical fault in a medium voltage power cable with bronze armor. The test inputs which demonstrate the worst case bounding inputs and required critical parameters are documented in a calculation (Reference 8.12).

The primary purpose of the cable fault testing was to determine if a single cable failure (phase-to-ground fault) on a single medium voltage power cable will propagate to a multi-phase fault and damage adjacent cables. A secondary purpose was to determine if low voltage control cables installed near the faulted cable would be damaged and if unacceptable voltage would be induced on the low voltage conductors. Four (4) different configurations were tested five (5) times each.

The test setups used were configured in a manner to maximize the potential for a single cable phase-to-ground fault to propagate into a multi-phase fault. Test voltages, phase fault currents and associated durations, bound the Oconee plant values for the cables of concern. The tests were conducted on bronze armored medium voltage cables of same specifications as existing plant cables located in the following areas:

• 13.8 kV KHS Underground Path (KHS to Transformer CT-4),

• 13.8 kV PSW Underground Path (KHS to P5W Switchgear),

• 13.8 kV Fant Path (Manhole 6 to P5W Switchgear),

• 4.16 kV KHS Underground Path (Breaker 1TC-04 to KHS Transformer CX).

Enclosure ONS-201 6-017 February 15, 2016 Page 14 Test Methodologyv:

For each test configuration, a single power cable was prepared by cutting a triangular flap in the cable jacket and drilling a small hole through the cable bronze metallic shield, insulation semicon, insulation and conductor sem icon to the conductor. The jacket flap was placed back and secured to the cable jacket with tape. The hole through the metallic shield was not to be repaired and tape was not installed over the area where the hole is drilled. This approach is conservative with respect to the type of test being performed. Since the cable insulation and metallic shield system was compromised, this resulted in a single phase-to-ground fault to immediately occur when the cable is energized. A small copper wire was inserted into the hole to facilitate the cable fault.

The position and orientation of the cable fault was such that if the bronze tape metallic shield and cable jacket were penetrated by the fault, the effects of the fault would directly impinge on the adjacent power cable. The power cables were arranged in a triangular bundle(s) and held in close contact by cable cleats which were mounted to a cable tray. This was a conservative orientation since any increase in distance would reduce the severity of consequential damage (if any) of electrical faults on adjacent cables.

Instrument & Control (I&C) cables were also installed in the cable tray attached to the cable tray with stainless steel ty-wraps parallel to the cable at nominal spacing.

The I&C cable interlocked steel armor and underlying shielding were grounded on both ends. The I&C cable conductors were not energized but were monitored during the test for induced voltage. The I&C cables were repositioned closer (control cable separation from power cables varied from 5" to no gap spacing) to the power cable to obtain additional data for conservatism.

The testing laboratory replicated the critical parameters of the Oconee power systems response, including generator/power source neutral grounding arrangement (resistance or solidly grounded), voltages and phase-to-ground fault currents, and fault durations that included relay response and breaker opening times. The test was also configured to replicate multi-phase faults if fault extended to other phase cables.

The primary test parameters were fault currents and voltage on the faulted cable and the overall fault duration,

_Conduct of Testingq:

The testing was performed using a laboratory procedure and results were compiled and documented in a test report. Subsequently, Duke Energy evaluated the test program lab, procedure, and results; and determined the commercial grade dedication of the testing was acceptable as a QA product. Each test article configuration was tested at least five (5) times. After each test, the following parameters were inspected:

1. Verified that a single phase-to-ground fault did not result in a multi-phase fault through visual inspection line current data and/or cable electrical testing.

Breaching of the metallic shield and jacket of the faulted cable was acceptable.

Enclosure ONS-201 6-017 February 15, 2016 Page 15 Scorching or other damage to the jacket, metallic shield and insulation of the adjacent cables was acceptable provided the initial phase-to-ground fault did not propagate to a multi-phase fault.

2. Verified that a power cable fault did not result in medium voltage being imposed on I&C cable conductors by review of the voltage monitored by the test laboratory data acquisition system. Scorching or other damage to the I&C cable jacket, armor or underlying shields and tapes was acceptable provided the underlying conductor insulation was undamaged as verified by visual inspection and/or cable electrical testing.

Test Results:

• None of the test cases resulted in cable damage that propagated to a multi-phase fault.

• In some of the tests, the adjacent power cable had a superficial indentation at the fault location but no jacket or shield damage occurred.

• In one of the five 4 kV cable fault tests, there was damage that penetrated the adjacent power cable's outer jacket and bronze tape shield; however, the internal insulation remained intact and no phase-to-phase fault occurred.

Follow-up testing of this cable showed the cable passed a 30-minute withstand test at 7.0 kV.

• In each test case, the cable jacket and bronze tape performed its function of protecting the adjacent conductors and not allowing a fault to propagate (based on visual inspection of the test specimen).

• In each test case there was no observable damage to the control cables in the tray section adjacent to the faulted power cables.

• Low voltage levels observed on control cables determined to be inconsequential, based on 3rd party review of test results and finite element model of a full length configuration.

On November 18, 2015, in a public meeting with NRC staff, Duke Energy outlined its plans to submit a licensing action to address cable separation issues. At that time, NRC questions on the testing were developed in preparation for a follow-up public meeting with Duke Energy on December 15, 2015. These questions were emailed to Duke Energy on December 8 and 10, 2015. Duke Energy's responses to these questions were shared at the December 15, 2015, public meeting and are provided in Attachment 2.

6. RISK INSIGHTS Note: The following section is not part of the basis for acceptance but is being provided for risk insight purposes.

A risk analysis (Reference 8.7) was performed to determine the potential risk impact of the current plant configuration with respect to cable separation in the locations of concern. The risk analysis determined the potential risk impact of the current plant configuration with respect to cable separation in the following three (3) locations:

Enclosure ONS-201 6-017 February 15, 2016 Page 16

• PSW System Ductbank Manholes 1-6,

• KHS Mechanical Equipment Gallery,

• PSW Building Cable Spreading Area.

Note that the Reference 8.7 analysis also addresses an additional location (Trench

3) which is not addressed in this submittal. For each location, an estimate of the increase in core damage frequency (CDF) and large early release frequency (LERF),

above that which would exist if the DC control cables were not co-located with the AC power cables, was developed. The analysis considered the following aspects:

• Frequency of cable faults,

• Probability that a fault is a multi-phase or high energy arc fault (HEAF),

• Probability of a large imposed voltage on one or both Oconee vital 125 VDC trains,

• Probability of failure of one or both Oconee vital 125 VDC trains, given an imposed voltage,

• Probability of failure of mitigation strategies.

Each of the above aspects is described briefly below:

Freauency of Cable Faults:

The analysis used generic industry data (Reference 8.10) on cable faults as a starting point to determine the likelihood of a fault on an energized medium voltage power cable. Reference 8.10 gives a failure rate of 7.2E-04/year per 500 feet of cable for pink EPR cables (the type used at Oconee). Although the EPRI data may be considered representative of general pink EPR medium voltage underground cable failure rates, this data includes cables of the "Uni-Shield" design which has a reduced-diameter, no external insulating jacket, and a much different / compromised shield. In the EPRI data, the much more vulnerable "Uni- Shield" design failures are included with the standard-shield pink-EPR cables. Further parsing of the EPRI data indicates that all but 3 out of 15 (or 20%) of the pink EPR cable failures were of the "Uni-Shield" design.

Since the cables in the locations of concern are not of that design, the cable failure frequency was reduced by a factor of five to 1.44E-04/year per 500 feet of cable for this Oconee-specific evaluation. For each location, this value was then adjusted to account for the length of cable in that location, and the amount of time the cable is energized.

Enclosure ONS-201 6-017 February 15, 2016 Page 17 The following tables provide an overview of the results of this adjustment for each location. The frequency contribution from each set of cables is the product of the number of cables, the base frequency, the length adjustment and the hours per year divided by 8760 hour0.101 days <br />2.433 hours <br />0.0145 weeks <br />0.00333 months <br />s:.

6 32.8 0.36 1,94E-07 1.2E-06 Fr1e~e ~ I.2E-06 6 99.0 0.2 3.25E-07 1.95E-06 3 8760 0.2 2,88E-05 8.64E-05 6 32.8 0.2 1.08E-07 6.47E-07

• l'*,,'*I*; Total IEl*ii~i 6 8760 0.18 ...E-05 1.56E-04 Probability that a Fault is a Multi-Phase or Higqh Eneravy Arc Fault (HEAF):

Although the expected result of a fault across the insulation is a ground fault to the bronze tape shielding, it cannot absolutely be ruled out that a multi-phase fault could occur. However, this failure mode is unlikely since, as discussed above, the fault across the insulation would have to penetrate the grounded bronze shielding of the faulted cable, and then penetrate the grounded bronze shielding and the insulation of an adjacent cable, all prior to the actuation of the protection system, in order to fault to another phase. Duke Energy is not aware of any mechanism that could cause a fault to behave in this manner. Consider Eaton Corporation White Paper TP08700001 E "Fault Characteristics in Electrical Equipment" published in 2011 (Ref.

8.11), which provides data on the likelihood of multi-phase faults. The paper summarizes data collected in IEEE Std. 493'Recommended Practice for the Design of Reliable Industrial and Commercial Power Systems.' IEEE Std. 493 examined the

Enclosure ONS-201 6-017 February 15, 2016 Page 18 total number of faults that occurred over a sample of equipment, and compared the quantity of those involving ground with those that did not and calculated a percentage of each type of fault relative to the total. This paper concludes "This IEIEE 493 standard is stating that for a cable, it is nearly 100 times more likely (99%

divided by 1%) for a cable fault to be a ground fault versus a fault not involving ground." Based on this, a 1.0E-02 probability has been assigned that, given a cable fault occurs, it is a multi-phase fault.

Probability of a Larae Imposed Voltaae on One or Both Oconee Vital 125 VDC trains:

Duke Energy did not identify any specific data that addresses the likelihood that a HEAF on one of medium voltage power cables in the vicinity of the DC control cables would induce a large voltage on those control cables. However, reasonable engineering judgment based on the design of the DC control cables (i.e., interlocked, grounded, stainless steel armor, multiple ground planes, etc.), supports the conclusion that the likelihood of this scenario would be very low. This is supported by the cable testing and analysis described above, where no significant voltage was impressed on any of the DC control cables in any of the tests. A value of 5E-02 was used for this likelihood. The value selected is considered bounding and worst case, based upon the testing showing no adverse impact.

Probability of Failure of One or Both Oconee Vital 125 VDC Trains. Given an Imp~osed Voltage:

It would take a significant failure (i.e., failures of multiple distribution panels on multiple trains) of the Oconee 125 VDC system, caused by an imposed voltage on that system from the fault, to impact the ability to safely shut down the units.

However, the probability of a loss of all 125 VDC power at Oconee is low, given the design of the cables in these manholes (which have grounded bronze tape for AC power cables or interlocked steel armor for the DC control cables), the design of the manholes themselves (which have grounded uni-strut supports), and the design of the DC system (which has multiple ground planes in the marshalling cabinets, and in the Oconee DC system itself).

Lastly, the 125 VDC system at Oconee is designed to maintain defense-in-depth and safety margin by two independent trains, such that an induced failure of one train should not impact the other. Thus, while failure of a single train of DC power is unlikely, failure of both trains is even less likely. A value of I1.0E-02 has been used for the likelihood of the failure of a single train of 125 VDC power, while a value of 4.OE-03 has been used for the likelihood of failure of both trains of 125 VDC power, which includes a common cause element, These values are considered bounding and worst case, again, based on the results of the cable testing.

Probability of Failure of Mitigqation Strategcies:

Even in the event of a complete loss of normal and emergency offsite power and loss of the 125 VDC system, there is still equipment available for mitigating the event. It is clear that mitigating strategies are limited in the case of a loss of all DC power, due to the significant loss of normal safety equipment control and control room indication.

Enclosure ONS-201 6-017 February 15, 2016 Page 19 However, the Standby Shutdown Facility (SSF) which is designed to provide the ability to maintain the plant in a safe condition, is still available, providing further defense-in-depth and safety margin.

The SSF is completely independent normal plant systems (including 125 VDC power). Electrical power to the SSF is provided by a dedicated diesel generator, while DC control power to the SSF is provided by a dedicated battery and battery charger. In the event of a failure of the 8SF DG, power can be provided by an additional offsite source (i.e., the 13.8 kV power feed from the Central Tie Switchyard through the PSW system, that is not affected by any failures in manholes I through 6). Note that although the PSW system itself can potentially provide additional mitigation capability (i.e., auxiliary feedwater, seal injection), no credit has been taken in this analysis for those system. Mitigating strategies in the case of a loss of only a single train of DC are much more robust, since one train of normal plant safety equipment and indication would remain available, in addition to the 8SF. Failure probabilities for the mitigating strategies of 3.4E-03, given loss of a single train of 125 VDC, and 2.2E-02 given a loss of both trains of 125 VDC, have been calculated based on portions of the Oconee PRA model.

Conclusions:

The analysis shows that the overall CDF/LERF increase for the three (3) cable locations is approximately 2E-11/year. The CDF/LERF increase for manholes I through 6, whose configuration is proposed to remain "as-is," is less than 1E-13/year, which is many orders of magnitude below what is typically considered risk significant.

The CDF/LERF increases from the PSW cable spreading area and the KHS Mechanical Equipment Gallery are approximately 1E-1 1/year and 5E-I12/year, respectively. These locations have relatively short lengths of cable, but do have some normally energized AC power cables. Again, these values are several orders of magnitude below what is typically considered risk significant, and will be reduced even further with completion of the proposed modifications.

7. PRECEDENTS No previous IOCFR50.55a(h)(2) related examples were found associated with both cable separation issues and the use of bronze armor cables in emergency power protection system applications.

8. REFERENCES 8.1 NRC Regulatory Guide 1.75, Criteria for Independence of Electrical Safety Systems, Revision 3 (ADAMS Accession Number ML043630448).

8.2 IEEE Std. 279-1971, "IEEE Standard: Criteria for Protection Systems for Nuclear Power Generating Stations."

8.3 IEEE Std. 384-1992, "IEEE Standard Criteria for Independence of Class IE Equipment and Circuits."

Enclosure ONS-201 6-017 February 15, 2016 Page 20 8.4 Oconee Nuclear Station Updated Final Safety Analysis Report (UFSAR),

Revision 24, effective date of contents 12/31/14. UFSAR Chapters/Sections consulted:

• Section 3.1 .1.1, Design of Structures, Components, Equipment, and Systems - Conformance with NRC General Design Criteria.

• Section 8.2.1 .3.1, Electric Power - Off-site Power System - System Description - 230 kV Switching Station - 230 kV Switching Station Degraded Grid Protection

• Section 8.3.1.1.1, Electric Power - Onsite Power Systems - AC Power Systems - System Descriptions - Keowee Hydro Station

• Section 8.3.1.4.6, Electric Power - Onsite Power Systems - AC Power Systems - Independence of Redundant Systems - Cable Installation and Separation

• Section 8.3.2.1.3, Electric Power - Onsite Power Systems - DC Power Systems - System Descriptions - 125 Volt DC Keowee Station Power System

• Section 9.7.3.2, Auxiliary Systems - Protected Service Water System -

System Description - Electrical

• Section 9.7.3.5.1, Auxiliary Systems - Protected Service Water System -

System Description - Civil/Structural - Building Structures

• Chapter 15, Accident Analyses.

• Chapter 18, Section 18.3.14 - Insulated Cables and Connections Aging Management Program Aging Management Programs and Activities.

8.5 Duke Energy Letter to the Nuclear Regulatory Commission, "TIA 2014-05, Potential Unanalyzed Condition Associated with Emergency Power System,"

dated 5/11/2015.

8.6 Duke Energy Letter to the Nuclear Regulatory Commission, "Supplemental Information on TIA 2014-05, Potential Unanalyzed Condition Associated with Emergency Power System," dated 8/7/2014.

8.7 OSC-1 1478 "Oconee Medium Voltage Cable Separation Risk Assessment,"

Revision 1.

8.8 NRC Letter to Duke Energy, "Oconee Nuclear Station - NRC Component Design Bases Inspection Report 05000269/2014007, 05000270/2014007, and 05000287/2014007," dated June 27, 2014, (ADAMS Accession No. ML14178A535).

8.9 NRC Memorandum, Director of Reactor Safety to Deputy Director, Division of Policy and Rulemaking, Office of Nuclear Reactor Regulation, "Request for Technical Assistance Regarding Oconee Nuclear Station Design Analysis for Single Failure and the Integration of Class 1E Direct Current Control Cabling in Raceways With High Energy Power Cabling (TIA 2014-05)," dated October 16, 2014, (ADAMS Accession No. ML14290A136).

Enclosure ONS-201 6-017 February 15, 2016 Page 21 8.10 "Plant Support Engineering: Failure Models and Data Analysis for Nuclear Plant Medium Voltage Cables for Consideration in Preventive Maintenance and Strategic Replacement," EPRI, December 2009.

8.11 Eaton Corporation White Paper TP08700001 E, "Fault Characteristics in Electrical Equipment", September 2011.

8.12 OSC-11504, "Medium Voltage Cable Testing Analysis," Revision 1.

8.13. Duke Energy Design Specification OSS-0218.00-00-0019, "Cable and Wiring Separation Criteria," Revision 17.

ATTACHMENT I Regulatory Commitment Table Regulatory Commitment Table February 15, 2016 Pane 2 The following commitment table identifies those actions committed to by Duke Energy Carolinas, LLC (Duke Energy) in this submittal.

1 Complete field implementation of:

• Cable separation modifications to the PSW System Cable9/51 Spreading Area and the KHS Equipment Gallery, and

• Fant feeder line relocation out of PSW duotbank manhole 6.

ATTACHMENT 2 Duke Energy Responses to NRC Questions on Cable Fault Testing Duke Energy Responses to NRC Questions on Fault Testing February 15, 2016 Page 2

1. In accordance with 10 CFR 50.54(Ui) (2015), what quality standards were used to design the testing plan and to analyze the testing data?

Duke Energy Response:

10 CFR 50.54 (jj) states "Structures, systems, and components subject to the codes and standards in 10 CFR 50.55a must be designed, fabricated, erected, constructed, tested, and inspected to quality standards commensurate with the importance of the safety function to be performed."

Test control quality standards are defined in the Duke Energy Topical Report Quality Assurance Program Description Operating Fleet Paragraph 017.3.2.8 (Test Control).

The Test Control section requires that testing be performed by written procedures that include requirements and acceptance limits, test instructions, test prerequisites such as calibrated instrumentation and methods for recording data and documenting test results.

These elements were present and documented in the vendor supplied test procedure. In addition this test procedure was reviewed and accepted by qualified individuals at Duke Energy.

To ensure adequate control of the design inputs and test identification, Duke Energy generated Reference 8.12 to provide circuit parameters and configurations which were used as the basis of the test procedure review. In addition Appendix C of this calculation includes a detailed analysis of test results. This calculation was generated as a QA-1 document in accordance with Duke Energy internal Design Analysis Procedures and the Duke Energy Quality Assurance Program.

2. Did the design of these tests meet the requirements of IEEE 279-1971? How?

Duke Energy Response:

The requirements of IEEE Std. 279-1971 are not applicable to cable testing. IEEE Std. 279-1971 Section 1 (Scope) states "These criteria establishes minimum requirements for the safety related functional performance and reliability of protection systems" where "protection systems encompasses all electrical and mechanical devices and circuitry (from sensors to actuation device input terminals) involved in generating those signals associated with the protective function" Signals are defined as those which "actuate reactor trip" and "engineered safeguards."

Section 4.4 (Equipment Qualification) indicates that "test data shall be available to verify that protection system equipment shall meet, on a continuing basis, the performance requirements determined to be necessary to achieving the system requirements." The cable testing was not a qualification test of the Oconee Protection Systems.

Duke Energy Responses to NRC Questions on Fault Testing February 15, 2016 Page 3

3. How were the worst-case tested ground faults determined? What quality standards were used for this determination?

Duke Energy Response:

Reference 8.12 utilized input analyses to determine the maximum available Keowee generator(s) ground fault current as well as the ground fault protection relaying operation timing. The original design of the Keowee generator neutral grounding system was performed in accordance with fundamental electrical power theory (i.e.,

basic Ohm's law, resistance/impedance reflection via transformer turns ratio and the application of an overvoltage relay parallel with the grounding resistance).

The plant (transformer CX) related feeds are analyzed using ETAP Version 7.1.0N.

The ETAP program provides full compliance with the IEEE C37 series for fault calculations. The PSW (Fant 13.8kV) feeder fault current values were provided by Duke Power Delivery based upon the actual system fault capability.

Design input fault analysis data was reviewed per Reference 8.12 which was generated as a QA-1 document in accordance with Duke Energy internal Design Analysis Procedures and the Duke Energy Quality Assurance Program

4. Why were three-phase faults not considered for testing?

Duke Energy Response:

Three-phase faults were a testing consideration. All tests began with a single phase-to-ground fault. Ifthe initial phase-to-ground fault propagated to a three-phase fault, the testing lab power source and relaying were configured and calibrated to provide the required three-phase fault voltage and current for the specified duration.

However, since all of the tests began and ended as phase-to-ground faults, the opportunity to use three-phase fault test parameters did not occur.

5. Why did the testing not address cascading failures (i.e. circuit breaker failures that may result from the short circuit conditions)?

Duke Energy Response:

It is assumed that no other failure occurs concurrent with the line to ground fault with one conductor of the three conductor (three phase) cable bundle. This is in accordance with Section 8.3.1.2 of the UFSAR which states the following, "The basic design criterion for the electrical portion of the emergency electric power system of a nuclear unit, including the generating sources, distribution system, and controls is that a single failure of any component, passive or active, will not preclude the system from supplying emergency power when required." For the purposes of this analysis, a single line to ground (conductor to shield) failure shall be treated as a single passive failure.

Duke Energy Responses to NRC Questions on Fault Testing February 15, 2016 Page 4

6. What analysis was done to ensure that each configuration bounded the worst case asymmetrical and symmetrical fault conditions for:

a. Configuration 1?

b. Configuration 2?

c. Configuration 3?

d. Configuration 4?

Duke Energy Response:

Symmetrical current values were used to determine potential cable damage effects since the damage mechanism is thermal and heat transfer occurs over time.

Symmetrical current values are used in Section 5.2 of IEEE Standard 1584-2002, IEEE Guide for Performing Arc-Flash Hazard Calculations. Since a line to ground fault would produce an arc and lead to the consumption of cable materials (insulation and conductive material), this could be treated as a small arc flash. Thus IEEE 1584-2002 is appropriate to be used as a basis for the use of symmetrical fault currents in this analysis. It is not appropriate to use Asymmetrical fault currents as they have an extremely short duration and thus would not significantly contribute to any heating effects to degrade the cable construction materials.

For the CT-4 (Keowee Underground) and KPF (Keowee to PSW) cases, the maximum line to ground fault current is limited by the high impedance grounding system. This limiting ground fault current value and duration (breaker and protective relaying clearing time) was determined by analysis. Three phase bolted fault currents used in the test setup were determined by utilizing results from studies of existing AC system short circuit models. However, three phase fault currents were not required during the test since the line to ground faults did not propagate to a three phase fault. The worst case fault duration for both the line to ground and bolted three phase fault cases was determined based upon the protective relaying operation and breaker clearing times for the respective case. This duration was different depending upon which fault type (line to ground or three phase) was considered because different fault types actuate different protective relay elements.

For the Fant case (13.8kV PSW Power Delivery Feeder), the maximum line to ground fault current was provided by Duke Energy Power Delivery. Three phase bolted fault currents used in the test setup were also provided by Duke Energy Power Delivery. However, three phase fault currents were not required during the test since the line to ground faults did not propagate to a three phase fault. The worst case fault duration for both the line to ground and bolted three phase fault cases was determined based upon the protective relaying operation and breaker clearing times for the respective case. This duration was different depending upon which fault type (line to ground or three phase) was considered because different fault types actuate different protective relay elements.

For the CX case (4.16kV feeder from Oconee Breaker 1TC-04 to Keowee Auxiliaries), the maximum line to ground and symmetrical three phase bolted fault current was determined by utilizing results from studies of existing AC system short circuit models. However, three phase fault currents were not required during the test Duke Energy Responses to NRC Questions on Fault Testing February 15, 2016 Page 5 since the line to ground faults did not propagate to a three phase fault. The worst case fault duration for both the line to ground and bolted three phase fault cases was determined based upon the protective relaying operation and breaker clearing times for the respective case. This duration was different depending upon which fault type (line to ground or three phase) was considered because different fault types actuate different protective relay elements.

7. What analysis was done to ensure that each configuration bounded the worst case arc flash duration for:

a. Configuration 1?

b. Configuration 2?

c. Configuration 3?

d. Configuration 4?

Duke Energy Response:

Symmetrical current values are used in Section 5.2 of IEEE Standard 1584-2002, IEEE Guide for Performing Arc-Flash Hazard Calculations. Since a line to ground fault would produce an arc and lead to the consumption of cable materials (insulation and conductive material), this could be treated as a small arc flash. Symmetrical current values were used to determine potential cable damage effects since the damage mechanism is thermal and heat transfer occurs over time. Thus IEEE 1584-2002 is appropriate to be used as a basis for the use of symmetrical fault currents in this analysis.

A plant specific analysis was performed to determine the maximum protective relaying operating time and breaker fault clearing times for each configuration (CT-4, KPF, Fant and CX). The maximum fault clearing times were utilized as the minimum times for fault testing. The worst case fault duration for both the line to ground and bolted three phase fault cases was determined based upon the protective relaying operation and breaker clearing times for the respective case. This duration was different depending upon which fault type (line to ground or three phase) was considered because different fault types actuate different protective relay elements.

This is stated in further detail in the response to Question 6 for each configuration.

8. The "as installed" cable clamping configuration differs from that at the test laboratory. The test laboratory employed metal cable cleats designed for the forces encountered during electrical faults in accordance with IEC 61914 "Cable Cleats for Electrical Installations." The cleats were spaced at *-1 ft.

intervals and secured to a cable tray in an open environment, The installed condition used metal zip ties to strap the cables to Unistrut pegs approximately every 4 ft. in an enclosed cable raceway. How does this difference address the impact of magnetic forces resulting from a worst-case fault condition as discussed in industry standards?

Duke Energy Response:

The objective of the testing program was to determine if a phase-to-ground fault can Duke Energy Responses to NRC Questions on Fault Testing February 15, 2016 Page 6 propagate to a three-phase fault on medium voltage single conductor power cables.

Validating the cable cleat design or cleat spacing intervals was not in the testing scope. The cable cleats were used as fixtures to ensure the faulted cable was held in close contact with the adjacent cables, thus creating a test configuration that was conducive for a multi-phase fault to occur.

9. How does the use of new cables compare to the "as installed" cables, which can be in a more degraded condition due to variations in ambient conditions (temperature, moisture etc.), electrical transients, and variations in current flow?

Duke Energy Response:

All of the tested power cables are the same design as those procured for the Keowee underground path replacement in the 2001-2002 timeframe or were procured and installed as part of the Protected Service Water Project in 2009-2013.

The "as-installed" power cables were evaluated in accordance with the Oconee Aging Management Program. The cables are not subject to adverse localized environments such as significant moisture and voltage, as defined by the Oconee Insulated Cables and Connections Aging Management Program.

Additionally, for the 4.16 kV CX circuit, a recent station modification spliced a section of three-conductor cable to the existing single conductor cables which were installed in 2001-2002. Post-modification electrical testing of the cables was performed using VLF/Tan Delta and Partial Discharge testing. The results of the testing indicated that the CX cables are in good condition in accordance with Duke Energy Cable Aging Management Program guidelines.

The CX circuit has been in operation for over 12 years and is normally energized and carrying a load. Any aging-related environmental or operating stressors would have been identified by this test but none were found. Therefore, the condition of the tested cable reflects the condition of the "as-installed" cables.

a. Is the assumption for a limiting condition single phase to ground fault appropriate for cables? What quality standards addressed this?

Duke Energy Response:

Yes. The testing program confirmed that that a single phase-to-ground fault in one cable would not propagate to a multi-phase fault using the test parameters specified in Reference 8.12. The quality standards used are referenced in the response to Questions 1, 3, and 10.

Duke Energy Responses to NRC Questions on Fault Testing February 15, 2016 Page 7

10. How were the configuration differences ("as-tested" vs "as installed")

analyzed? What quality standards were used for this analysis?

Duke Energy Response:

The entire testing program, including cable mounting, isolating the tray from ground, fault current development, fault orientation and cleating of cables, was designed to create an environment that was conducive for an initial phase-to-ground fault to propagate to a three-phase fault.

Reference 8.12 established the electrical testing criteria. The tests were conducted under continuous Duke Energy engineering supervision.

11. What effects did the test configuration have on the test results? (i.e. the power cables were open-circuited and no operating loads were used for AC or DC)

Duke Energy Response:

Adding operating load to the cables would increase the pre-fault temperature of the cable conductors and metallic shields. For the resistance grounded power sources, the very low magnitude phase-to-ground fault currents presented no challenge to the thermal capacity of the bronze tape metallic shield and thus pre-heating would not change the test results. For the tests with solidly grounded power sources and high magnitude phase-to-ground fault currents, the extreme temperature generated by the arcing fault resulted in vaporization of the bronze tape. Elevating the pre-fault conductors from ambient to operating temperature would have had inconsequential effects on the degree of bronze tape damage.

The power cables are AC type circuits therefore loading with DC was not a tested parameter. The I&C cables were tested in an open-circuit configuration and connected to relays. The results were determined to not adversely affect equipment operation.

12. How would the inductive and capacitive coupling effects be influenced when current is present on all phases of the power cables and the DC cables are energized? What quality standards were used to address this aspect?

Duke Energy Response:

Data recorded during the test indicated a small induced voltage on the control and instrumentation cables. Test data also shows that the voltage was not relatively impacted by the radial separation of the power and l&C cables. The induced noise was being significantly driven by inductive and capacitive coupling; these parameters would have more strongly correlated to measured voltage changes. The energization of the DC circuits lacked sufficient current to change the results.

A third party FEA (finite element analysis) was performed to determine any potentially adverse impacts of an induced voltage during postulated short circuit events, The two voltages of interest included the maximum differential voltage (the Duke Energy Responses to NRC Questions on Fault Testing February 15, 2016 Page 8 voltage developed across two control cable conductors) and the maximum common mode voltage (the voltage developed on any control cable conductor relative to ground). The maximum differential voltage was studied to determine the potential for spurious actuation or drop of control circuits for the Keowee emergency power system. The maximum differential voltage was studied to determine the potential for insulation system breakdown/failures relative to station ground (control cable armor, terminal blocks and other system components).

This analysis includes several known conservatisms, including use of the most limiting cable separation (set to minimum allowable by the duct geometry) and intra-cable bundle low voltage conductor separation (set to the maximum allowable by the cable bundle diameter), neglecting control conductor helical twist along the length of the cable (has the effect of canceling magnetic coupling), and a conservatively high line-ground fault current (16,000 Amps-peak).

This analysis specifically studied the configuration of the 13.8kV Fant to PSW feeder parallel to the control cables from Manhole 6 to the PSW building. The assumption of a I16kA peak line to ground fault current well bounds the actual available fault current of 5.66kA (per Reference 8.12) for this configuration.

The analysis determined that the differential voltage induced on a galvanized steel interlocked armor (GSIA) control cable conductor pair is less than one volt. Since this differential voltage is so minute, there is no concern for meal-operation of the control circuits for the Keowee emergency power system.

The analysis further determined that the common mode voltage is approximately 14 volts for the armored control cable case. A common mode voltage of 14 volts is well within the insulation system of the Keowee emergency power system components (cables, relays, etc.).

Small voltage changes (less than 1 volt across conductors (differential) and 14 volts relative to ground (common mode)) would have an insignificant impact on an ungrounded 125VDC control system such as the Keowee Emergency Start circuit(s).

These are relatively small changes relative to the system limits. For instance, I volt differential is less than 1 percent of the rating (125VDC) of the connected emergency start relays and would not interfere with the emergency start or operation of a Keowee unit.

The common mode voltage is an insulation (relative to ground) rating. A common mode voltage of 14 volts is only 11 percent more than the rating of the most limiting device relative to ground (14/125VDC for emergency start relays). Typically cable systems, terminal blocks, etc. are rated for at least 150 percent of their nominal rating (i.e., 1000Vac cable in a 600Vac application) relative to ground.

This analysis relied heavily upon methods originally developed by Edward Rosa (Volume 4, Number 2, Bulletin of the Bureau of Standards, The Self and Mutual Inductances of Linear Conductors) and summarized more recently by Clayton Paul in IEEE EMC Society Magazine.

Duke Energy Responses to NRC Questions on Fault Testing February 15, 2016 Page 9 The 3rd party analysis was reviewed per Reference 8.12 which was generated as a QA-1 document in accordance with Duke Energy internal Design Analysis Procedures and the Duke Energy Quality Assurance Program

a. Were the concerns presented in Annex B of IEEE 603 investigated in relation to this question?

Duke Energy Response:

Yes. Annex B (informative) of IEEE 603-2009 (IEEE Standard Criteria for Safety Systems for Nuclear Power Generating Stations), Electromagnetic Compatibility, provides some discussion on the sensitivity and vulnerability of components to electromagnetic interference (EMI). This annex is concerned with incorrect component operation or damage due to EMI.

This standard recommends shielding as a mitigation strategy for EMI. Section B.3.1, Evaluation of the electromagnetic environment, states that, "Performance may be demonstrated by a combination of testing, analysis, or documented operating experience." Analysis was performed (see the response to question 12 above) and it was determined that EMI would not cause incorrect component operation or damage.

b. Has the impact of increased cable length been investigated (i.e., as installed (4000 ft.) vs. as-tested (12 ft.))?

Duke Energy Response:

Yes. An Engineering Change has removed from service all Instrumentation and Control (l&C) in Trench 3 (approximately 4,000 feet in length) therefore the potential for induced voltage on I&C cables in Trench 3 has been eliminated.

Additionally, Reference 8.12 has evaluated other configurations of power and l&C cables and determined that the calculated levels of induced voltage on l&C cables would not adversely affect equipment operation.

13. In some of the cable tests observed, the bronze tape shield melted partially.

Has Duke evaluated the impact of such melting, if a worst-case fault is postulated?

Duke Energy Response:

Yes. The tests with solidly grounded power sources resulted in high magnitude phase-to-ground faults. The high temperature developed by the arcing fault exceeded the melting point of bronze and it was expected that the bronze tape shield would be damaged at the area surrounding the fault location. For all tests, the remaining bronze shield provided a conductive path to ground for the entire fault duration; therefore, melting did not impact the test results.

As documented in the responses to Questions 3-11, the cable tests were conducted with the power sources configured to provide worst-case fault currents. As such, the Duke Energy Responses to NRC Questions on Fault Testing February 15, 2016 Page 10 observed melting of the bronze tape represented the "worst-case" fault currents and the impact has been evaluated as stated in the above paragraph.

14. Did Duke calculate the maximum magnetic force that would be exerted in the raceway system to CT-4, which has approximately 4,000 feet of cable? The NRC staff's review of industry guidance indicates that cables in the concrete trench could be exposed to a substantial amount of force. It did not appear that these effects were simulated in the cable testing. Ifthese magnetic forces were not modeled in the tests, how did the testing performed demonstrate that the existing cable configuration meets the ONS licensing basis and applicable ANSI standards?

Duke Energy Response:

The maximum magnetic force was not calculated for the CT-4 raceway system since it was not a tested parameter. The purpose of the testing program was to evaluate if an initial phase-to-ground fault would propagate to a three-phase fault. For a three-phase fault, the test cables would have experienced magnetic forces; however, three-phase faults did not occur during the tests rendering this question moot.

Duke Enerav Responses to Follow-up NRC Questions on Fault Testing:

1. Why was 12 gauge wire used in tests 3 &4?

Duke Energy Response:

The laboratory power sources for tests with solidly grounded power sources provided phase-to-ground fault currents in the kA range. The #12 AWG copper wire was selected to ensure that the fault lasted for the required duration.

2. Demonstrate why the current was not split [between the three cables]?

Duke Energy Response:

On each cable end, the metallic shield is attached to ground straps that are connected to a common grounding point. The cable conductor end connected to the power source is the line side and the other end, which is not connected, is designated the load side.

When the faulted cable is energized, the entire current of the faulted cable flows from the line side to the fault location and then splits, in proportion to the combination of the shield and ground path impedance, back along the faulted cable's metallic shield to the line and load side where the fault current is conducted to ground. The metallic shield of the faulted cable carries all the fault current.

The other non-faulted phase conductor shields are also connected to a common Duke Energy Responses to NRC Questions on Fault Testing February 15, 2016 Page 11 grounding point. Depending on the overall impedance characteristics of the multiple paths to ground, after the current exits the faulted cable metallic shield, it may have been possible for the metallic shields of the non-faulted conductors (connected in parallel with the faulted cable metallic shield) to carry some degree of the fault current back to ground.

Since there were no three-phase faults, the non-faulted cable conductors were still electrically isolated from the faulted cable conductor and there was no sharing of current between the faulted and non-faulted cable conductors [see Sketch].

3. What quality standards were used to determine this [re: Question 2 above]?

Duke Energy Response:

A review of Attachment 7 in Reference 8.12 was performed of the schematics and electrical test data for each of the four test configurations. Next, electrical analysis principles and knowledge of medium voltage cable design was applied to reach the conclusions provided in the response to Follow-up Question 2.

Duke Energy Responses to NRC Questions on Fault Testing February 15, 2016 Page 12 Sketch If1 If2 4-

- B.

S S

~se

• S

• S

• S S S 1'

~se

• S S

S S

S S S

• -. -. -. ----- 4 S S I-A-Ph*~se SW

• S S S

• S S S S S

• S

'fi ~YS S

S H,

S S

I2 S S S S S S

• S If If=lIfl+If2 Cable Conductor

.....

• Metallic Shield Conductors to ground

- Fault Current Path to Ground