ML12334A672

Official Exhibit - NYS000158-00-BD01 - EPRI, Plant Support Engineering: Aging Management Program Guidance for Medium-Voltage Cable Systems for Nuclear Power Plants, Final Report (1020805) (June 2010) (EPRI 1020805)

ML12334A672
Person / Time
Site:	Indian Point
Issue date:	06/30/2010
From:	Toman G Electric Power Research Institute
To:	Atomic Safety and Licensing Board Panel
SECY RAS
References
RAS 21546, 50-247-LR, 50-286-LR, ASLBP 07-858-03-LR-BD01 1020805
Download: ML12334A672 (78)

Text

NYS000158 Submitted: December 15, 2011 United States Nuclear Regulatory Commission Official Hearing Exhibit Entergy Nuclear Operations, Inc.

In the Matter of:

(Indian Point Nuclear Generating Units 2 and 3)

,,\..-toP-Po REGU<..t;. ASLBP #: 07-858-03-LR-BD01

~\

Docket #: 05000247 l 05000286

• 0 Exhibit #: NYS000158-00-BD01 Identified: 10/15/2012

~ ~

~ , it Admitted: 10/15/2012 Withdrawn:

~

~~ d' i Rejected: Stricken:

?

.. ***. " Other:

Plant Support Engineering: Aging Management Program Guidance for Medium-Voltage Cable Systems for Nuclear Power Plants OAGI0001276_00001

Plant Support Engineering: Aging Management Program Guidance for Medium-Voltage Cable Systems for Nuclear Power Plants 1020805 Final Report, June 2010 EPRI Project Manager G. Toman Work to develop this product was completed under the EPRI Nuclear Quality Assurance Program in compliance with 10 CFR 50, Appendix Band 10 CFR Part 21,

.. ;"1

'l<:<;

I: ..*,

(~)

'0 ELECTRIC POWER RESEARCH INSTITUTE 3420 Hillview Avenue, Palo Alto, California 94304-1338* PO Box 10412, Palo Alto, California 94303-0813* USA 800.313.3774*650.855.2121

• askepri@epri.com

• www.epri.com OAGI0001276_00002

DISCLAIMER OF WARRANTIES AND LIMITATION OF LIABILITIES THIS DOCUMENT WAS PREPARED BY THE ORGANIZATION(S) NAMED BELOW AS AN ACCOUNT OF WORK SPONSORED OR COSPONSORED BY THE ELECTRIC POWER RESEARCH INSTITUTE, INC. (EPRI). NEITHER EPRI, ANY MEMBER OF EPRI, ANY COSPONSOR, THE ORGANIZATION(S) BELOW, NOR ANY PERSON ACTING ON BEHALF OF ANY OF THEM:

(A) MAKES ANY WARRANTY OR REPRESENTATION WHATSOEVER, EXPRESS OR IMPLIED, (I)

WITH RESPECT TO THE USE OF ANY INFORMATION, APPARATUS, METHOD, PROCESS, OR SIMILAR ITEM DISCLOSED IN THIS DOCUMENT, INCLUDING MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE, OR (II) THAT SUCH USE DOES NOT INFRINGE ON OR INTERFERE WITH PRIVATELY OWNED RIGHTS, INCLUDING ANY PARTY'S INTELLECTUAL PROPERTY, OR (III) THAT THIS DOCUMENT IS SUITABLE TO ANY PARTICULAR USER'S CIRCUMSTANCE; OR (B) ASSUMES RESPONSIBILITY FOR ANY DAMAGES OR OTHER LIABILITY WHATSOEVER (INCLUDING ANY CONSEQUENTIAL DAMAGES, EVEN IF EPRI OR ANY EPRI REPRESENTATIVE HAS BEEN ADVISED OF THE POSSIBILITY OF SUCH DAMAGES) RESULTING FROM YOUR SELECTION OR USE OF THIS DOCUMENT OR ANY INFORMATION, APPARATUS, METHOD, PROCESS, OR SIMILAR ITEM DISCLOSED IN THIS DOCUMENT.

ORGANIZATION(S) THAT PREPARED THIS DOCUMENT Electric Power Research Institute (EPRI)

THE TECHNICAL CONTENTS OF THIS DOCUMENT WERE NOT PREPARED IN ACCORDANCE WITH THE EPRI QUALITY PROGRAM MANUAL (WHICH FULFILLS THE REQUIREMENTS OF 10 CFR 50 APPENDIX B, 10 CFR 21, ANSI N45.2-1977 AND/OR THE INTENT OF ISO-9001 (1994)).

USE OF THE CONTENTS OF THIS PRODUCT IN NUCLEAR SAFETY OR NUCLEAR QUALITY APPLICATIONS REQUIRES COMMERCIAL GRADE DEDICATION OR ADDITIONAL ACTIONS BY THE RECEIVING ORGANIZATIONS.

NOTE For further information about EPRI, call the EPRI Customer Assistance Center at 800.313.3774 or e-mail askepri@epri.com.

Electric Power Research Institute, EPRI, and TOGETHER. .. SHAPING THE FUTURE OF ELECTRICITY are registered service marks of the Electric Power Research Institute, Inc.

Copyright © 2010 Electric Power Research Institute, Inc. All rights reserved.

OAGI0001276_00003

CITATIONS This report was prepared by Electric Power Research Institute (EPRI)

Plant Support Engineering 1300 West W.T. Harris Blvd.

Charlotte, NC 28262 Principal Investigator G. Toman This report describes research sponsored by EPRI.

The report is a corporate document that should be cited in the literature in the following manner:

Plant Support Engineering: Aging Management Program Guidance for Medium-Voltage Cable Systems for Nuclear Power Plants. EPRI, Palo Alto, CA: 2010. 1020805.

111 OAGI0001276_00004

PRODUCT DESCRIPTION Regulatory and management concern regarding the reliability of medium-voltage cable systems at nuclear plants has been increasing for 5-10 years. The staff of the U. S. Nuclear Regulatory Commission (NRC) are concerned that wetted (up to and including submergence) medium-voltage cable circuits may be degrading to the point at which multiple cable circuits may fail when called on to perform functions affecting safety. Utility managers are concerned that cables may fail, causing adverse safety consequences and/or plant shutdowns. This report provides guidance for developing and implementing a cable aging management program for medium-voltage cable circuits in nuclear power plants.

Results and Findings The report was developed by subjecting drafts to review and revision by a Technical Advisory Group formed of industry cable personnel from nuclear plant organizations, cable manufacturers, and cable test companies. This report describes the scope of the cable circuits to be evaluated, those conditions that are considered to be adverse environments, and the actions to be taken to assess the conditions of the cable circuits subject to adverse conditions. For key test methodology, assessment criteria are described, along with possible corrective actions that could be implemented.

Challenges and Objectives This report was developed at the direction of utility management and in parallel with the Regulatory Issue Resolution Protocol for Inaccessible or Underground Cable Circuit Performance Issues at Nuclear Power Plants that occurred between the NRC and the industry (through the Nuclear Energy Institute) from mid-2009 into 2010. Implementation of this guide will form part of the closure process for the protocol. This guide was developed to provide a consistent methodology for the industry to follow in developing an aging management program for medium-voltage cable circuits that are subjected to adverse environmental or service conditions that could lead to degradation of the insulation systems.

Applications, Value, and Use This guide describes a common approach for developing and implementing a medium-voltage cable system aging management program. Techniques applicable to shielded and nonshielded cable are provided. Because the nuclear industry generally uses different cable types and designs from those used in the power distribution industry, initial assessment criteria and guidance pertinent to the cable applications in the nuclear industry are provided.

v OAGI0001276_00005

EPRI Perspective The need for a guide for developing cable system aging management programs has been increasing over the last few years. This report was developed with strong input from the industry and represents good practice for the foreseeable future. Cable aging management is an evolving process and an enhancement of the maintenance program for nuclear plants. As the implementation process matures and further research is performed on improving test technology and understanding degradation mechanisms, changes are expected to assessment criteria, the focus of the programs, and the methodologies used. This report will be revised as needed.

Approach This guide provides a way of determining the scope of a cable aging management program and focuses the aging management process on cables in the worst-case adverse environment and service conditions. It describes testing and assessment criteria and potential corrective actions.

The bases for program development are provided as a way of determining the health of the resulting aging management program.

Keywords Cable aging management Cable aging management program Medium-voltage cables VI OAGI0001276_00006

ACKNOWLEDGMENTS This report was developed through a Technical Advisory Group process in which drafts of the report were prepared and then commented on through an iterative meeting process. The Electric Power Research Institute (EPRI) wishes to especially thank the following persons for their extra efforts in discussing and probing the content of the report:

Kent Brown Tennessee Valley Authority (TVA)

John Cancelosi Okonite Co.

Gordon Clefton Nuclear Energy Institute Rick Foust Wolf Creek Nuclear Operating Corp.

Steven Graham Duke Energy Carolinas Robert Konnick Rockbestos-Surprenant Cable Corp.

Andrew Mantey Electric Power Research Institute Kenneth Petroff Public Service Electric & Gas Co.

Howard Sedding Kinectrics, Inc.

Edward Walcott General Cable Co.

The following persons also supported the development and contributed to the review of the document:

Ayad AL-Hamdani Dominion Generation Anand Anandakumar Kinectrics, Inc.

Corrado Angione PPL Susquehanna, LLC Colby Baker American Electric Power, Inc.

Michael Baldwin NEXT era Energy Ryan Briggs FirstEnergy Nuclear Operating Co.

Michael Chan DTEEnergy John Crews Progress Energy, Inc.

Mark Crisler Southern Nuclear Operating Co.

Stanley Crumbo South Carolina Electric & Gas Co.

Vll OAGI0001276_00007

Deborah Dixon Wolf Creek Nuclear Operating Corp.

Preston Dougherty Dominion Generation John Errico Florida Power & Light Co.

Michael Fallin Constellation Energy Wesley Frewin Institute of Nuclear Power Operations Bogdan Fryszczyn Cable Technology Laboratories, Inc.

Sinnathurai Gaj anetharan Entergy Robert Gehm Rockbestos-Surprenant Cable Corp.

Steven Gocek Nebraska Public Power District Craig Goodwin High Voltage Diagnostics, Inc.

James Hamlen AREVA NP Inc.

Daniel Houser Duke Energy Carolinas Thomas Huckaby Jf. Institute of Nuclear Power Operations Steven Hutchins Exelon Corporation Mark Hypse Arizona Public Service Co.

Kevin Iepson IEPSON Consulting Nicole Jackson Southern Nuclear Operating Co.

George Karayianopoulos FirstEnergy Service Company James Landrum Dominion Virginia Power Rufus Lawhorn Southern Nuclear Operating Co.

James Mah Institute of Nuclear Power Operations Eric Nelson Nebraska Public Power District KyNguyen Southern California Edison Co.

David Parker Florida Power & Light Co.

Anthony Ploplis Progress Energy, Inc.

Ronald Proffitt Dominion Virginia Power Camilo Rodriguez FirstEnergy Nuclear Operating Co.

Roger Rucker Entergy Sy Shaheen Southwire Co.

Brian Sides Southwire Co.

Yasushi Takizawa Tokyo Electric Power Co.

V111 OAGI0001276_00008

Denise Thomas Exelon Generation, LLC Debbie Williams Institute of Nuclear Power Operations Mosang Yoo Korea Power Engineering Co., Inc.

Donna Young Progress Energy, Inc.

IX OAGI0001276_00009

CONTENTS 1 INTRODUCTION .................................................................................................................... 1-1 Program Development .......................................................................................................... 1-2 Implementing Procedures .................................................... ..... ..... .. .... ..... ..... .. ... .. .... ..... ..... .. 1-3 Data and Information to Be Collected and Retained ............................................................. 1-3 Program Plan Milestones ........... ..... ..... ..... .. .... ..... ..... .. .... ..... ..... ..... .. .... ..... ..... ..... .. .... ..... ..... .. 1-4 Program Health Indicators ..................................................................................................... 1-5 Definitions ................................. .. .... ..... ..... ..... .. .... ..... ..... .. .... ..... ..... ..... .. .... ..... ..... ..... .. ..... ...... 1-5 Abbreviations and Acronyms ................................................................................................. 1-7 2 SCOPE OF THE AGING MANAGEMENT PROGRAM FOR MEDIUM-VOLTAGE CABLE SYSTEMS ....................................................................................................................2-1 Program Scope Versus Cable Circuits Requiring Condition Monitoring or Assessment.. ..... 2-5 3 IDENTIFICATION OF ADVERSE ENVIRONMENTS AND CONDITIONS ............................. 3-1 Activities for Wet and Potentially Wet Environment Cables .................................................. 3-2 Nondrained Conduits and Ducts Within Plant Structures ...................................................... 3-3 Condition of Vaults, Manholes, and Related Cable Support Structures ................................ 3-3 Activities for Dry Cable Circuits ............................................................................................. 3-4 High-Temperature- or High-Dose-Rate Ambient Environments ....................................... 3-4 Thermal Aging ............................................................................................................. 3-4 Radiation-Related Aging .............................................................................................. 3-5 Periodicity of Review of Condition ............................................................................... 3-6 High Conductor Temperature from Ohmic Heating .......................................................... 3-6 Periodicity of Review ................................................................................................... 3-6 High-Resistance Connections at Terminations or Splices ................................................ 3-7 Periodicity of Review ................................................................................................... 3-8 Xl OAGI0001276_00010

4 SUSCEPTIBILITY OF CABLES TO WATER-RELATED DEGRADATION BY TYPE .......... 4-1 Recommendations Based on Susceptibility Assessment. ..................................................... 4-4 Susceptibility of Splices to Wetting ........................................................................................ 4-5 5 ACTIONS FOR CABLES HAVING WET ENViRONMENTS .................................................. 5-1 Pumping of Manholes and Ducts .......................................................................................... 5-2 Assessment of Condition of Shielded Cables ....................................................................... 5-2 Effects of Shield Design and Insulation on Long-Term Testability and Test Selection ........................................................................................................................... 5-4 Combined Testing ................................................ .. .... ..... ..... ..... .. .... ..... ..... ..... .. .... ..... ....... 5-6 Tan <5 Methodology and Assessment Criteria ........................................................................ 5-6 Use of Standard Deviation in Assessing Tan <5 Results .................................. .. ... .. .... ......... 5-12 Very Low Frequency Withstand Testing in Conjunction with Tan <5 Testing ........................ 5-14 Test Methodology and Assessment Criteria for Other Tests ...... .. .... ..... ..... .. ... .. .... ..... ......... 5-15 Test Preparation and Concerns .......................................................................................... 5-15 Failure of Cable Under Test .................. ..... ..... ..... .. .... ..... ..... .. ... .. .... ..... ..... .. .... ..... ..... ..... ..... 5-16 Assessment of Nonshielded Cables .................................................................................... 5-17 6 ACTIONS FOR CABLES HAVING DRY ADVERSE ENVIRONMENTS ................................ 6-1 High-Temperature or High-Dose-Rate Ambient Environments ............................................. 6-1 High Conductor Temperature from Ohmic Heating ............................................................... 6-6 High-Resistance Connections at Terminations or Splices .................................................... 6-6 7 ACTIONS FOR FAILED OR DETERIORATED CABLE ........................................................ 7-1 Operability Concerns ............................................................................................................. 7-1 Corrective Actions ................................................................................................................. 7-1 Cable Test Indicates "Further Study Required" Insulation System ....................................... 7-1 Eliminate Obvious Problems ............................................................................................ 7-2 Perform Very Low Frequency Withstand Test.. ................................................................ 7-2 Increase Frequency of Testing ......................................................................................... 7-2 Prepare Contingency Plan ................................................................................................ 7-2 Perform Polymer Injection ................................................................................................ 7-2 Begin Replacement Program for Multiple Cables with "Further Study Required" Insulation .......................................................................................................................... 7-2 Cable Test Indicates "Action Required" Insulation System ........ ..... .. ... .. .... ..... ..... .. ... .. .... ...... 7-3 XlI OAGI0001276_00011

Eliminate Obvious Problems ................................................ .. .... ..... ..... ..... .. .... ..... ..... ....... 7-3 Perform Very Low Frequency High-Potential Tests ......................................................... 7-3 Identify and Replace Degraded Section ........................................................................... 7-3 Conduct Forensic Testing of "Action Required" Cable ..................................................... 7-3 Use Impervious Cable for Wetted Environments .............................................................. 7-4 Cables Experiencing Localized Thermal Damage ................................................................ 7-4 Evaluate the Degree of Damage ...................................................................................... 7-4 Correct the Adverse Localized Thermal Environment ...................................................... 7-5 Replace Thermally Damaged Cable ................................................................................. 7-5 High-Resistance Connections ............................................................................................... 7-5 Cables Damaged by High Current ........................................................................................ 7-6 8 REFERENCES ....................................................................................................................... 8-1 In-Text References .......... ..... .. .... ..... ..... ..... .. .... ..... ..... ..... .. .... ..... ..... .. ... .. .... ..... ..... .. .... ..... ....... 8-1 Additional Resources ............................................................................................................ 8-2 A ASSESSMENT OF PERCENT STANDARD DEVIATION OF TAN ()

MEASUREMENTS ................................................................................................................... A-1 X111 OAGI0001276_00012

LIST OF FIGURES Figure 2-1 Medium-Voltage Cable Circuit Scoping Process ................................................ ...... 2-6 Figure 5-1 Pictorial Representation of Tan <5 Assessment Criteria for Pink Ethylene Propylene Rubber ............................................................................................................ 5-10 Figure 5-2 Typical "Good" Tan <5 Result for a Shielded Cable with Pink Ethylene Propylene Rubber Insulation ............................................................................................ 5-11 Figure 5-3 Tan <5 Plots for a 13.8 kV UniShield Cable with a Degraded B Phase .................... 5-12 Figure 5-4 Tan <5 Example Including Percent Standard Deviation Data ................................... 5-14 Figure 6-1 Spontaneous Cracking of Neoprene Jacket.. ........................................................... 6-3 xv OAGI0001276_00013

LIST OF TABLES Table 1-1 National Electric Code and Underwriters Laboratories, Inc. Definitions of Dry, Damp, and Wet Locations .................................................................................................. 1-7 Table 2-1 Scope Comparison for Maintenance Rule, 10 CFR 50.65, and License Renewal Rule, 10 CFR 54 ................................................................................................. 2-3 Table 4-1 Cable Susceptibility Under Wet Conditions ............................................................... 4-2 Table 4-2 Aging Management Recommendations Based on Insulation Material Type ............. 4-4 Table 5-1 Preliminary Tan <5 Assessment Criteria for Butyl Rubber ........................................... 5-7 Table 5-2 Preliminary Tan <5 Assessment Criteria for Black EPR .............................................. 5-7 Table 5-3 Preliminary Tan <5 Assessment Criteria for Pink EPR ........... ..... ..... .. .... ..... ..... ..... .. .... 5-8 Table 5-4 Preliminary Tan <5 Assessment Criteria for Brown EPR ............................................. 5-8 Table 5-5 Preliminary Percent Tan <5 Standard Deviation Assessment Criteria for Rubber Insulated Cables .............................................................................................................. 5-13 Table 6-1 Thermal and Radiation Degradation Mechanisms Expected for Medium-Voltage Cable Jacket Materials .......................................................................................... 6-2 Table 6-2 Thermal and Radiation Degradation Mechanisms Expected for Medium-Voltage Cable Insulation Materials ..................................................................................... 6-5 Table 6-3 EPRI-Suggested Severity Ranges for Indoor Electrical Power Connections ....... .. .... 6-7 XVll OAG10001276_00014

1 INTRODUCTION This report provides guidance for the development of an aging management program for medium-voltage cable circuits in nuclear power plants to ensure high reliability. The program is intended to identify adverse localized environments and adverse service conditions that could lead to early failure of medium-voltage cable circuits and to manage significant aging effects to preclude in-service failure. Low-voltage power cables have been addressed in a separate EPRI report Plant Support Engineering: Aging Management Program Development Guidance for AC and DC Low-Voltage Power Cable Systemsfor Nuclear Power Plants (1020804) [1]. It is recognized that plants may choose to have one program cover all cable types. However, because different aging mechanisms and assessment activities apply to low- and medium-voltage power cable, the guidance is being generated separately.

Medium-voltage cables (rated 5 kV to 46 kVl and generally having operating voltages between 2.3 kV and 34 kV) may age and fail because of several mechanisms. Table 2-2 ofEPRI report Equipment Failure Model and Datafor Underground Distribution Cables: A PM Basis Application (1008560) lists the potential failure mechanisms and whether they are random or age related [2]. The random causes such as installation damage or manufacturing defects do not affect any significant portion of the population of cables and, as such, are not addressed in this report. This document pertains to long-term aging from adverse service conditions that, if neglected, could lead to in-service failures. The effect of a medium-voltage cable failure can cause the loss of a train of a safety system or remove an offsite feed from service. Accordingly, an aging management process for medium-voltage cable systems is desirable to limit the number of in-service failures and support high reliability of the medium-voltage cable system.

Medium-voltage cables and accessories that are properly installed, supported, and kept cool and dry should have a long life. However, cables or accessories that are subject to adverse conditions should be governed by an aging management program. The following are recognized adverse conditions with respect to the longevity of medium-voltage cable circuits:

• Adverse localized high-temperature and/or high-radiation ambient environments under normal operating conditions

• High conductor temperature from ohmic heating

• High-resistance connections at terminations or splices

• Long-term submergence (partial or full submergence) 1 This definition of medium voltage ratings is from the Insulated Cable Engineers Association Standards.

NUREG-1801 XI.E3 defines medium voltage as 2 kV to 35 kV (assumed to be the range of operating voltages).

1-1 OAGI0001276_00015

Introduction The presence or absence of these conditions can be determined by inspection and analysis, environmental monitoring, or infrared thermography. If there are no adverse conditions, a long life can be expected for the cable circuits. Accordingly, for benign environments and service conditions, monitoring and maintenance are not expected to be necessary. Further action would be required only if failures occur or degradation from very long service is recognized. In that case, the need for maintenance and monitoring for benign environment and service condition applications should be determined in accordance with the Maintenance Rule, 10 CFR 50.65 [3]

and plant corrective action programs.

If one or more adverse conditions are observed, further assessment, testing, and/or corrective action will be necessary to ensure reliability, unless the cable and/or its accessories have been designed for the conditions.

Program Development A program plan should be developed for aging management of medium-voltage cable circuits.

The plan should include the following elements:

• Management's objectives for the program (that is, identification and management of aging caused by adverse localized environments and adverse service conditions)

• Interfaces with other inspection and integrity programs (for example, infrared thermography program or thermal insulation integrity program)

• A well-structured process including scoping, identification of adverse environments and service conditions, assessment of cable circuits exposed to the adverse environments and conditions, and implementation of corrective action as appropriate

• Defined roles and responsibilities including those for the program manager and supporting organizations for assessments, tests, and repair or replacement

• Training requirements

• Determination of the scope of cable circuits to be in the program (see Section 2)

• A schedule for completion of the scoping, determination of the cable circuits potentially affected by adverse environments and service conditions (see Section 3) and the development of the initial assessment plan and expected cost for adoption

• Management sponsorship of continued implementation

• Program health reporting and corresponding performance indicators 1-2 OAGI0001276_00016

Introduction

• Documentation to be retained, including scope determination, adverse service conditions, cable circuits to be assessed, condition and cable assessment methods, condition and cable assessment and test results, and corrective actions that have been implemented

• Periodic review of plant conditions to determine whether there are any changes to adverse conditions (additions or deletions)

Implementing Procedures Implementing procedures 2 should address the following:

• Roles and responsibilities

• Scoping methodology and documentation

• Determination of adverse conditions

• Consideration of susceptibility of the plant cables to adverse conditions and identification of cable circuits needing assessment

• Schedule of initial assessments and periodicity of subsequent assessments

• Methods to be used to assess cable circuits subject to adverse conditions

• Assessment of results related to cable condition

• Repair or replacement options (see Section 7)

Data and Information to Be Collected and Retained The following data and information should be retained for use in continued assessment:

• Program plan

• Implementing procedures

• Scope of the program (for example, cable circuits subject to Maintenance Rule and additional License Renewal required scope)

• Cable circuits within the program that are subject to adverse localized environments and/or service conditions that require aging management 2 Different utilities use the terms guides, procedures, and plans in different ways. The key issue is to have a documented process that includes the appropriate elements of a cable aging management program.

1-3 OAGI0001276_00017

Introduction

• Additional information that should be identified for these cable circuits includes the following:

The nature and location of the adverse environment or service condition.

Cable circuits that are affected, including the subcomponent of concern (for example, termination, splice, or cable).

- Associated load of affected cable circuits (for example, specific motor, bus, or transformer).

- Degradation mechanism of concern (for example, thermal damage or voltage/water degradation).

- Method of assessing or monitoring the effect and the periodicity of assessment (for example, one-time assessment, periodic visual inspection, or periodic test [including initial assessment interval]).

- Methodology of assessment and tests. (Given that periods between assessments and tests may be several years, a complete description of the methods used will help to ensure the ability to compare and trend results, especially if changes to methods occur as technology improves.)

- Results of assessments and tests.

- Repair and replacement descriptions.

• Where credit is taken for maintaining dry conditions in ducts, manholes, and vaults, documentation showing that automatic drainage systems are effective and/or that cables are not found to be submerged when water is manually pumped from manholes and vaults.

• Program health report performance indicators.

Program Plan Milestones The following are suggested program plan milestones:

• Program plan and technical procedures are in place, current, and being implemented.

• Program documentation is complete and current.

• Roles and responsibilities are defined, accepted, and owned by organizations and individuals for assessment, testing, repair, and replacement.

• The program manager and backup are identified and trained.

• Program resources are adequate.

• The scope of the program is clearly defined.

• The adverse localized environments and adverse service conditions of concern have been defined.

1-4 OAGI0001276_00018

Introduction

• The cable circuits within the program that are subject to adverse localized environments and/or adverse service conditions have been identified for further aging management activities.

• For cable circuits requiring further aging management activities, a method of assessing the cable has been identified and scheduled.

Program Health Indicators The following are suggested program health indicators:

• The cable circuit/adverse environment assessments are being implemented according to schedule.

• Deferral of cable circuit assessments is limited.

• Review of cable circuit assessment results is timely, and corrective actions are initiated.

• Implementation schedule of corrective actions is met.

• Control of cable submergence is satisfactory.

• Control of thermal insulation in the vicinity of power cables is adequate.

• Thermography of connections and high-current cables is being performed and acted on.

• Program self-assessments are being performed at a reasonable interval.

• The number of age-related cable circuit failures during a defined period is within prescribed limits.

• The number of open findings or areas for improvement from external audits or assessments (for example, U.S. Nuclear Regulatory Commission [NRC] and Institute of Nuclear Power Operations [INPO]) is limited, and they have been resolved in a timely manner.

• Forensic assessment of cables that fail in service is performed, and the findings are incorporated into changes or improvements to the program.

• Applicable operating experience of other sites is being reviewed, assessed, and incorporated into the cable program by the program manager.

Definitions Assessment. In the context of this report, assessment is used to cover a broad range of activities regarding cable condition. These activities include evaluating the severity of environments and service conditions, evaluating the need for testing, and evaluating condition, including visual/tactile inspection and condition monitoring through activities such as electrical testing or in situ or laboratory physio-chemical testing. Some assessments are expected to limit the scope of testing and evaluation (for example, the cable has benign service and environmental conditions); other assessments will include testing and condition monitoring, as appropriate, because of the presence of adverse service or environmental conditions.

1-5 OAGI0001276_00019

Introduction Delta Tan <>. Delta tan 8 is the value yielded from the difference between the tan 8 readings at 0.5 Vo and 1.5 Vo. It can also be the difference in tan 8 readings between Vo and 2 Vo.

Impervious Coverings. Some utilities have and continue to purchase and install cables with impervious coverings, which are designed to prevent penetration of water into the insulation system. Earlier cables used continuous lead or aluminum coverings tightly formed over the core of the insulation, including cable shields. 3 These continuous layers preclude water ingress, and the result is a dry insulation that is not subjected to wetting even if the cable is completely submerged.

Impervious coverings are optional and not required for submerged applications. They are most often chosen when particularly aggressive soil or water conditions exist.

Inaccessible Cable. Inaccessible cables are those cables that have sections located below grade or are imbedded in the plant base mat that are located in duct banks, buried conduits, cable trenches, cable troughs, underground vaults, or that are direct buried. 4 The concept of inaccessibility for cables is related to the ability to determine the environment and physical condition of cable. For underground cable, inaccessibility makes identification of wetting and submergence more difficult. In dry plant areas, inaccessibility is less of a problem.

Even when cables are inside conduits or contained in trays that are difficult to access, heat sources that are close to the tray or conduit are relatively easy to identify and further assessment of condition is possible. Inaccessibility is not a concern if adverse service and environments do not exist.

Submergence, Wet, Damp, and Dry Locations. Both the Underwriters Laboratories, Inc. (UL) and the National Electric Code (NEC) define the terms dry, damp, and wet locations (see Table 1-1). The definitions indicate that the term wet means up to and including submerged and not just damp, which has its own definition. The NEC definition indicates "saturation with water or other liquid," and the UL definition indicates "flow on or against electrical equipment."

3 For example, a I-in. (25.4-mm) diameter core, the required lead layer was 80 mils (2.03 mm), and the required aluminum layer was 55 mils (1.4 mm). A more modem design of water-impelVious cable uses a continuous linearly corrugated copper tape system that is wrapped around the cable core with the overlap glued shut (See IEEE Std 400-2001, Section 4 [14]).

4 NUREG-180 1, Generic Aging Lessons Learned Report,Section XI.E3, Inaccessible Medium-Voltage Cables Not Subject to 10 CFR 50.49 Environmental Qualification Requirements, states" .. .inaccessible (e.g., in conduit or direct buried) medium-voltage cables .... " [4]. NRC Generic Letter 2007-01 states" ... in inaccessible locations such as buried conduits, cable trenches, cable troughs, above ground and underground duct banks, underground vaults, and direct-buried installations." [5].

1-6 OAGI0001276_00020

Introduction Table 1-1 National Electric Code and Underwriters Laboratories, Inc. Definitions of Dry, Damp, and Wet Locations Term National Electric Code Underwriters Laboratories Definition [6] Definition [7]

Dry location A location not normally subject to dampness A location not normally subject to or wetness. A location classified as dry may dampness, but may include a location be temporarily subject to dampness or subject to temporary dampness, as in wetness, as in the case of a building under the case of a building under construction. construction, provided ventilation is adequate to prevent an accumulation of moisture.

Damp location Locations protected from weather and not An exterior or interior location that is subject to saturation with water or other normally or periodically subject to liquids but subject to moderate degrees of condensation of moisture in, on, or moisture. Examples of such locations adjacent to, electrical equipment, and include partially protected locations beneath includes partially protected locations.

canopies, marquees, roofed open porches, and like locations, and interior locations subject to moderated degrees of moisture, such as basements, some barns, and some cold storage buildings.

Wet location Installations underground or in concrete A location in which water or other liquid slabs or masonry in direct contact with the can drip, splash, or flow on or against earth; in locations subject to saturation with electrical equipment.

water or other liquids, such as vehicle washing areas; and in unprotected locations exposed to weather.

Abbreviations and Acronyms The following abbreviations and acronyms are used in this report:

ac alternating current CPE chlorinated polyethylene (thermoset or thermoplastic)

CSPE chlorosulfonated polyethylene (commonly referred to by the DuPont trade name Hypalon) dc direct current EOP emergency operating procedure 1-7 OAGI0001276_00021

Introduction EPR ethylene propylene rubber Black EPR is the earliest generation ofEPR used as cable insulation. Later generations were either gray (substantially reduced levels of carbon black), pink (red) EPR (most manufacturers), or brown EPR (Kerite). The color differences were to allow visual distinction between black semi-conducting or high permittivity shields and the insulation and also to demark the transition to improved coatings of the filler clay to improve its bonding to the base insulation material and preclude absorption of water.

GALL Generic Aging Lessons Learned report Gy Gray; a metric unit of radiation equal to 100 rad HCI hydrogen chloride hi-pot high potential hr hour INPO Institute of Nuclear Power Operations kV kilovo1t(s)

LIRA line resonance analysis (a cable condition monitoring technique)

Mrd megarad NEI Nuclear Energy Institute NEC National Electric Code NRC U. S. Nuclear Regulatory Commission PM preventive maintenance PVC polyvinyl chloride rd rad rms root mean square tan 8 An ac dielectric test of insulation that measures the ratio of resistive leakage current to the capacitive current across the insulation (radians, often given in terms of 10-3 )

TDR time domain reflectometry UL Underwriters Laboratories, Inc.

1-8 OAGI0001276_00022

Introduction VLF very low frequency Vo line-to-ground fIllS voltage on a three-phase system, also referred to as Do XLPE cross-linked polyethylene 1-9 OAGI0001276_00023

2 SCOPE OF THE AGING MANAGEMENT PROGRAM FOR MEDIUM-VOLTAGE CABLE SYSTEMS The development of the scope of the cable circuits to be within the medium-voltage cable system aging management program should consider these sources:

• The Maintenance Rule (10 CFR 50.65) scope requirements [3]

• The License Renewal Rule (10 CFR 54) scope requirements [8]

• Updated Final Safety Analysis Report commitments (if any)

• License Renewal aging management program commitments

• Plant-specific regulatory correspondence pertaining to cable

• Critical components as defined in INPO AP-913, Equipment Reliability Process [9]

• Circuits critical to power generation (management option)

Table 2-1 provides a comparison of the equipment covered by the Maintenance Rule and the License Renewal Rule. Paragraphs 10 CFR 50.65 (b)(I) and Paragraph 10 CFR 54.4(a)(I) require that cable circuits supporting safety-related functions be within scope of the respective activities. Paragraphs 10 CFR 50.65 (b)(2) and 10 CFR 54.4(a)(2) both require that nonsafety-related cable circuits whose failure could prevent safety-related functions from being fulfilled be 2-1 OAGI0001276_00024

Scope of the Aging Management Program for Medium-Voltage Cable Systems within scope. Paragraph 10 CFR 50.65 (b )(2) also requires that cable circuits used to mitigate accidents or transients or to support emergency operating procedures, as well as cable circuits whose failure could cause a reactor scram or actuation of a safety-related system, be in scope.

Paragraph 10 CFR 54.4(a)(3) extends beyond the Maintenance Rule scope in that cable circuits related to Station Blackout and Fire Protection are within scope.

Some plants may have cable monitoring commitments in their Updated Final Safety Analysis Report. All plants that pursue License Renewal will have cable aging management commitments in the License Renewal aging management program for cable and connections and terminations.

Under the License Renewal process, there is likely to be separate aging management programs for cable and for connections and terminations that should be considered when developing the scope and content of the medium-voltage cable system aging management program. Some plants may have cable-specific regulatory correspondence pertaining to cable. Review of the plant-specific response to Generic Letter 2007-01 is appropriate to confirm the activities that the plant stated were in place to assess the condition of cables and to control wetting of cables [5]. As the plant's medium-voltage cable system aging management program is developed and implemented, it is recommended that differences from and changes to methodologies from those in the Generic Letter 2007-01 response be documented.

The AP-913 equipment reliability process ranks components with respect to importance to reliability. Those cables required to support the function of components should be considered with respect to the scope of the medium-voltage cable system aging management program.

Medium-voltage cable circuits that supply outage power whose failure may adversely affect outage duration should also be considered for inclusion in the scope of the program. Other cables may be identified for scope inclusion based on plant-specific experiences.

2-2 OAGI0001276_00025

Scope of the Aging Management Program for Medium-Voltage Cable Systems Table 2-1 Scope Comparison for Maintenance Rule, 10 CFR 50.65, and License Renewal Rule, 10 CFR 54 [3, 8]

Maintenance Rule License Renewal Differences 10 CFR 50.65 (b)(1) 10 CFR S4.4(a)(1) None Safety-related systems, structures, and components which Safety-related ... systems and components that are are those relied on to remain functional during and following relied on to remain functional during and following design-basis events (as defined in 10 CFR 50.49(b)(1)) to design basis events to ensure the integrity of the ensure the following functions:

reactor coolant pressure boundary, the capability to shut down the reactor and maintain it in a safe (i) The integrity of the reactor coolant pressure boundary; shutdown condition, or the capability to prevent or mitigate the consequences of accidents that could (ii) The capability to shut down the reactor and maintain it in result in potential offsite exposure ... a safe shutdown condition; or (iii) The capability to prevent or mitigate the consequences of accidents that could result in potential offsite exposure comparable to the guidelines in 10 CFR 50.34(a)(1), in 10 CFR 50.67(b )(2) or 10 CFR 100.11 of this chapter as applicable.

10 CFR 50.65 (b)(2) 10 CFR 54.4(a)(2) Agreement on nonsafety nonsafety-related ... systems, or components: components that could affect All nonsafety-related systems, structures, and components function of safety (i) That are relied on to mitigate accidents or whose failure could prevent satisfactory accomplishment of components. Maintenance transients or are used in plant emergency any of the functions identified in paragraphs (a)(1) (i), (ii), or Rule adds cables associated operating procedures (EOPs); or (iii) of this section. with EOPs and that could (ii) Whose failure could prevent safety-related result in scrams or safety structures, systems, and components from system actuation.

fulfilling their safety-related function; or (iii) Whose failure could cause a reactor scram or actuation of a safety-related system.

0 G) 0 0

0 N

-....J I

(J) 2-3 0

0 0

N (J)

Scope of the Aging Management Program for Medium-Voltage Cable Systems Table 2-1 (continued)

Scope Comparison for Maintenance Rule, 10 CFR 50.65, and License Renewal Rule, 10 CFR 54 [3, 8]

Maintenance Rule License Renewal Differences 10 CFR 54.4(a)(3) License Rule adds cables associated with fire All systems, structures, and components relied on in safety protection, station blackout, analyses or plant evaluations to perform a function that and anticipated transient demonstrates compliance with the Commission's regulations without scram.

for fire protection (10 CFR 50.48), environmental Environmentally qualified qualification (10 CFR 50.49), pressurized thermal shock (10 cables would be in scope CFR 50.61), anticipated transients without scram (10 CFR already; there are no cables 50.62), and station blackout (10 CFR 50.63). associated with pressurized thermal shock.)

0 G) 0 0

0 N

-....J I

(J) 2-4 0

0 0

N

-....J

Scope of the Aging Management Program for Medium-Voltage Cable Systems Program Scope Versus Cable Circuits Requiring Condition Monitoring or Assessment The purpose of scoping is to consider the extent of cables that would need condition assessment or monitoring if they were exposed to adverse environments or have adverse operating conditions. It is not the intent of the program to assess and monitor the condition of the entire program scope. Rather, this document requires assessment of cables and/or accessories s exposed to adverse environments or have adverse service conditions. Accordingly, those cable circuits that are within scope, such as those supporting Maintenance Rule functions, and that are exposed to adverse environments or adverse service conditions will be assessed or monitored under the medium-voltage cable system aging management program as appropriate.

For medium-voltage cables, the list of cable circuits under consideration will likely be determined by review of medium-voltage bus loads and offsite power sources and then by a determination of whether the individual circuits have elements that are subject to adverse environments or service conditions. This technique works because of the limited number of circuits involved. Figure 2-1 illustrates the scoping concept. Although this document focuses on managing the aging of medium-voltage cable circuits that are subject to recognized adverse effects, the Maintenance Rule and corrective action processes ensure that if a new failure cause is identified, it will be assessed and corrective actions taken to control the effect. As appropriate, the medium-voltage cable system aging management program should be revised to take new failure causes into account.

5 Cable accessories are splices (joints) and terminations.

2-5 OAGI0001276_00028

Scope of the Aging Management Program for Medium-Voltage Cable Systems Entire Plant Cable Population Functions, License Hen!:!wal FlulG Commitments, and Other Licensing Commitments Cables to Be Assessed by the CablE: f.\qing IVlanagernent Program 1 Includes other adverse environments such as chemical and radiation environments.

Figure 2-1 Medium-Voltage Cable Circuit Scoping Process 2-6 OAGI0001276_00029

3 IDENTIFICATION OF ADVERSE ENVIRONMENTS AND CONDITIONS After the cable circuits within the scope of the medium-voltage cable system aging management program have been identified, the nature of the cable layout and application must be identified to determine whether aging management activities are required for particular cables. If cable circuits are subject to benign conditions, there are no aging management concerns at this time.

The first effort will be to separate cables that are located in completely dry conditions from circuits potentially wet or known to contain wet sections. Cables having wet or potentially wet environments include cables that are inaccessible, underground and/or substructure segments whether installed in ducts, in trenches, or direct buried. Cables located in dry accessible cable tunnels may be excluded from consideration for wetting. For cables determined to be totally in dry environments or having an impervious design, skip to "Activities for Dry Environment Cables." For cables having mixed dry and wet segments, the "Activities for Dry Environment Cables" applies to the dry sections, and the "Activities for Wet and Potentially Wet Environment Cables" applies to the wet section.

3-1 OAGI0001276_00030

Identification ofAdverse Environments and Conditions Activities for Wet and Potentially Wet Environment Cables A conservative approach to medium-voltage cable aging management is to assume that all underground cables are wet and to develop the program on that premise. This approach eliminates the need to assess the design and perform verifications that there are no wetted sections in cable circuits.

However, some underground systems have been designed to be dry or drained automatically. If the cables in such systems can be shown to be dry (for example, by verifying that water does not exist in duct work between manholes), the concern for long-term wetting can be eliminated. This premise may be difficult to defend unless some sort of inspection evidence (for example, umbilical video inspection) can be provided to support the supposition that the conduits are water free.

"Rain and drain" applications in which a duct or manhole may be wet for a short period until natural or automated draining (for example, sump pump) occurs is not considered adverse with respect to the life of a medium-voltage cable. Systems in which ducts slope toward manholes or other structures that are drained so that cables neither sit in nor are submerged in water for any significant period may be treated as dry with respect to cable longevity. Cables mounted on the walls of trenches and not subject to wetting along their length may be considered dry.

Water permeation into cable insulation takes a significant period. For this report, long-term wetting is defined as a condition in which the cable sits in or is covered by water for a continuous period of months or longer. A jacket over the insulation will significantly slow the effect, but the exact degree has not been determined. After water has permeated the jacket, water is known to be drawn to the highest voltage stress concentration within the insulation, which is near the conductor surface. When the water is drained from the vicinity of the cable, the ohmic heating of the conductor may drive off some of the moisture from the insulation, but how much and how fast is not readily identifiable. Certainly, further water and the additional chemical contamination (for example, salts) in the water are no longer available to permeate into the cable. However, any degradation that may have occurred does not reverse. Rather, the rate of further degradation is assumed to be slowed by draining the ducts and manholes and keeping them drained. With 3-2 OAGI0001276_00031

Identification ofAdverse Environments and Conditions respect to direct buried cables, an assumption must be made that the cables are always wetted, because observation cannot be made to show that they are above the water table and that there are no subterranean pockets of water surrounding sections of the cable.

If credit is being taken for a cable system being self- or naturally draining or manually pumped often enough to ensure that cables are not wetted, at least a one-time inspection of the system to confirm its nature should be made. If automatic sump pumps are being credited for maintaining a dry cable system, an appropriate inspection and maintenance program should be in place for the pumping system so that long-term wetting of the cables will not occur. If manual pumping is used as an alternative, the pumping must be performed frequently enough to preclude long-term wetting or manholes must be inspected after significant rainstorms, winter thaws, or flooding events to determine whether pumping is necessary and action taken accordingly.

Both water and voltage must be present for water-related deterioration to occur in medium-voltage cables. Accordingly, cables that are rarely energized will suffer minimal water-related degradation even if exposed to long-term wetting. However, given that the most important safety cables are likely to be de-energized for most of their service life and less well understood failure mechanisms could be possible in wet environments, assessment and testing of these cables is important. These cables should be evaluated early in the implementation process of the cable system aging management program. If these cables are found to be in satisfactory (that is, "good") condition after an extended period, consideration may be given to extending the period between tests with respect to continuously energized cables.

Nondrained Conduits and Ducts Within Plant Structures In some cases, medium-voltage cable ducts within plant structures are embedded in the floor with both end points exiting the floor above the duct. If such ducts exist and there is no drain for the below floor section, moisture may accumulate and condense in the duct, or the duct may fill from spills or other water-related events. To the extent practical, such duct arrangements should be evaluated to determine whether they are dry. If not, the condition of the cable should be assessed, and if practical, the duct should be drained.

Condition of Vaults, Manholes, and Related Cable Support Structures In addition to concerns for aging of cables under wet conditions, support structures for cables in trenches, manholes, and vaults may degrade with time, resulting in inadequate support of cables or physical damage to the cables. The physical structure of the manhole or vault may also degrade. Accordingly, the condition of support structures (for example, brackets and trays) and the overall manholes and vaults should be evaluated to confirm that no significant deterioration has occurred. It is recommended that ladders and platforms for personnel be included in these evaluations.

3-3 OAGI0001276_00032

Identification ofAdverse Environments and Conditions Activities for Dry Cable Circuits The adverse conditions and environments that can shorten the life of medium-voltage cables under dry conditions are as follows:

• High-temperature or dose-rate environments under normal operating conditions

• High conductor temperature from ohmic heating

• High-resistance connections at terminations or splices The following subsections address these items and the assessment of their importance with respect to cable longevity.

High-Temperature- or High-Dose-Rate Ambient Environments Thermal Aging Elevated temperatures cause thermal aging and may also limit the allowable ampacity of the cable. Cable thermal ratings are based on conductor temperature in a free air 40°C environment.

Most cables in nuclear plants have been derated so that conductor temperatures are well below the rated temperature and a long thermal life in a 40-50°C environment would be expected.

However, care must be taken in environments that exceed 50°C and for cables operating near their ampacity limit in ambient environments of 40°C or more because the combination of ambient and ohmic heating may cause higher rates of thermal aging.

In general, bulk area temperatures are not expected to significantly affect the aging of medium-voltage cable. However, localized hot spots are a key concern, especially if the cable is adjacent to hot process piping. NRC Information Notice 86-49 identified a 4-kV cable failure from exposure to a hot process pipe [10]. If medium-voltage cable is adjacent to an uninsulated, hot 3-4 OAGI0001276_00033

Identification ofAdverse Environments and Conditions process component (for example, pump, pipe, or valve) the cable polymer will be affected by both the local temperature and the radiant heating from the component. The circuit routing should be reviewed and/or walked down to determine whether hot process equipment is in the vicinity of the cable. If so, the condition of the thermal insulation on the hot components should be confirmed as adequate for the protection of the cable. Maintenance procedures should also be confirmed as requiring restoration of thermal insulation before the process component is returned to service if the thermal insulation must be removed to allow maintenance of the process component. If the process component must operate without insulation for any significant period, the effect on the medium-voltage cable should be evaluated. If a significant effect is expected, temporary thermal shielding should be placed between the hot process pipe and the cable.

Radiation-Related Aging With respect to radiation effects, most medium-voltage cable will be in low-dose areas of the plant. However, some cables may be located in areas with appreciable doses. Sandia research on low-voltage cables with similar compounds to those in medium-voltage cables showed that effects on physical properties are not observable at 1 Mrd (10 kGy) and that at least 5 Mrd (50 kGy) must be absorbed for effects to be observed [11]. Assuming a 60-year desired life for a medium-voltage cable, no appreciable effect would be expected for average dose rates up to 10 rd/hr (0.1 Gy/hr)6. Although minimal effects are expected at 10 rd/hr (0.1 Gy/hr), the effects could be appreciable if the cables are simultaneously exposed to high temperature (for example, greater than 122°P (50°C) with conductor temperatures reaching ampacity limits).

The effects of radiation and temperature are to change the physical properties (such as loss of elongation and increased hardness) of the insulation and, after severe aging, to eventually affect the electrical properties. If high temperature conditions are recognized and radiation doses greater than 5 Mrd (50 kGy) are expected, the medium-voltage cables should be inspected for degradation unless environmental qualification data exist that shows the capability of the materials. Note that until the dose from the exposure reaches approximately 5 Mrd (50 kGy),

radiation effects may not be observable. Inspections at the 30- or 40-year mark may only identify radiation effects if the dose rate is well above 10 rd/hr (0.1 Gy/hr) (that is, 15 to 20 rd/hr

[0.15-0.2 Gy/hr]).

Several types of medium-voltage cable have been subjected to environmental qualification testing. These tests provide information on whether radiation doses up to 50 Mrd (500 kGy)

(~95 rd/hr [0.95 Gy/hr] for 60 years) are within the qualification limits. The thicker insulation and jackets of medium-voltage cables makes them less susceptible to thermal and radiation aging. The damage from irradiation does not reduce the electrical properties appreciably of the insulation; rather, it hardens the insulation and makes it more susceptible to physical damage and failure after severe degradation. Where radiation or thermal damage or both are a concern, initial evaluation should include visual/tactile assessment as described in later sections of this guide.

6 1.9 rdlhr = 5 Mrd 7 (60 years x (365 days/year) x 24 hour2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br />s/day) 3-5 OAGI0001276_00034

Identification ofAdverse Environments and Conditions Periodicity of Review of Condition The initial review of adverse conditions will identify areas, if any, with elevated thermal or radiation conditions. The need for action will be governed by the severity of the condition.

Modifications and changes to plant operating conditions that could significantly worsen thermal or radiation conditions in the vicinity of medium-voltage cable should be reviewed for their effect.

The review should identify where medium-voltage cables are in close proximity to hot process piping. Programmatic controls should be put in place to verify that thermal insulation remains intact and effective so that the medium-voltage cables are not adversely affected as could occur if the thermal insulation is taken off and left off for a significant period.

High Conductor Temperature from Ohmic Heating Ohmic heating of medium-voltage cable from load currents can cause appreciable aging of the insulation system if cable currents cause the cables to operate at or above rated temperature, especially if ambient temperatures are elevated. Normal design practices result in operating cables at currents significantly below their ampacity. Design practices or ohmic heating calculations should be reviewed to confirm that elevated conductor temperatures (for example,

>90°C, a common cable thermal rating) should not occur.

One special case should be considered that can occur with multicable-per-phase circuits. When multiple cables are used in each phase, the magnetic fields must be balanced so that equal currents occur in each phase. This can be done by running three separate phase cables triplexed or in the same duct. However, if the individual conductors are laid flat in trays, the positions of the conductors may need to be transposed along the run to balance the magnetic circuits. If the magnetic fields are not balanced, some cables in each phase will run with lower currents and others will have high currents. Conditions have occurred in which the high-current cables were operating beyond ampacity and thermal aging caused severe hardening of the insulation and jacket. When a fault occurred on a connected transformer, the cables thrashed and one's insulation cracked to the conductor. Accordingly, the current balance on multiconductor-per-phase cables should be verified. This verification could be performed through the use of infrared thermography if cables are accessible or through measurements of current on individual conductors.

Periodicity of Review The review of ohmic heating of medium-voltage cables and current balance on multi conductor-per-phase circuits needs to occur only once. Re-review would be necessary only at the time of circuit repair or replacement.

3-6 OAGI0001276_00035

Identification ofAdverse Environments and Conditions High-Resistance Connections at Terminations or Splices Properly made splices and terminations should not experience overheating. However, when terminations or splices are disassembled and reassembled or first installed, human performance errors or design deficiencies can occur, resulting in high-resistance connections, especially when connections involving aluminum conductor are being made. Accordingly, terminations and splices should be checked for elevated temperature conditions when operating at load after installation. This verification should be performed at some reasonable period following installation (one or two operating cycles). Identification of high-resistance connections may be through the use of infrared thermography or periodic visual inspection for signs of discoloration or deterioration of the splice or termination. If the adequacy of the connection was confirmed at the time of splice or termination preparation through the use of a micro-ohmmeter or other recognized method, periodic evaluation may be unnecessary for most connections, but it may be desirable for aluminum connections until stability is confirmed.

The program may take credit for the performance of periodic infrared thermography or inspection of terminations and splices that is covered by the station maintenance program.

The EPRI Preventive Maintenance Basis Database provides frequencies for performing routine infrared surveys or inspections of terminations. Frequencies vary based on the end load's criticality. Infrared thermography surveys should be scheduled to be performed when the equipment is energized and loaded to provide meaningful results.

In many cases, access to terminations of medium-voltage cable terminations is limited because of equipment design, arc flash concerns, and so on. One relatively inexpensive way to improve access is using infrared windows or infrared ports on switchgear doors and cable termination boxes. Both types of access covers are available with UL ratings equal to that of most electrical enclosures. Infrared windows are made from special materials that are transmissive (transmissive materials can pass radiant energy that glass and Plexiglas cannot), and they are more expensive than a port. In addition, they have 40-60% transmissivity, requiring calculating hot spot temperature by multiplying the measured value by the inverse of the transmissivity for the window, and they do not hold up well when exposed to outside environments. Infrared ports, on the other hand, are relatively inexpensive, simple to install, and allow direct viewing (no transmissive losses) of the target.

Infrared thermography should also be scheduled as post-maintenance verification whenever splices are installed or when splices or terminations are disturbed for maintenance. This check should be done at least 1 hour1.157407e-5 days <br />2.777778e-4 hours <br />1.653439e-6 weeks <br />3.805e-7 months <br /> after the equipment has been energized and loaded (to allow thermal stabilization) or at the earliest opportunity thereafter.

Another diagnostic tool that can be applied to higher voltage terminations (6.9 kV and above) is ultrasonic partial discharge detectors that can be used to pick up arcing or other discharges within the connection.

3-7 OAGI0001276_00036

Identification ofAdverse Environments and Conditions Periodicity of Review The plant maintenance program should be reviewed to verify that splices and terminations are evaluated on a periodic basis. Thereafter, the plant maintenance program may be credited for covering this subject.

3-8 OAGI0001276_00037

4 SUSCEPTIBILITY OF CABLES TO WATER-RELATED DEGRADATION BY TYPE Differences exist in the susceptibility of various cable insulation types and vintages to water-related degradation under energized conditions. Table 4-1 indicates the differences in degree of expected susceptibility and the operating experience to date by insulation type and vintage. The older insulations are cross-linked polyethylene (XLPE), butyl rubber, and black ethylene propylene rubber (EPR). XLPE was extensively used in the distribution industry and found to degrade under wet energized conditions, especially in nonjacketed cables used in power distribution applications. (Nuclear plant cables are jacketed.) The EPRI report Equipment Failure Model and Datafor Underground Distribution Cables: A PM Basis Application (1008560) is based on an evaluation cable in distribution system service and developed onset of failure expectations based on expert opinion [2]. The result was that 30% through-wall water trees could be expected in XLPE in approximately 10-12 years of wet service. A 30% through-wall water tree was identified as the point at which the cable could be susceptible to a surge from a lightning strike that would convert the water tree to an electrical tree with relatively rapid failure thereafter. In nuclear service, cables are generally protected from lightning strikes, and the cables have lower voltage stresses, which tends to make the onset of failure later than the distribution industry operating experience. Nuclear industry operating experience indicates that the onset of XLPE failures traceable to wet aging occurred after 24 years of service.

4-1 OAGI0001276_00038

Susceptibility of Cables to Water-Related Degradation by Type Table 4-1 Cable Susceptibility Under Wet Conditions (population data are from the Nuclear Energy Institute [NEI] 2005 Survey on Underground Cables)

Material Man ufactu rers Approximate Population of Oldest Nuclear Earliest Expected Nuclear Industry Actual of Installed Period of Installed Plant Cables as Onset of Water Experience Discussion Cable Installation Cables at of 2009 Degradation in Nuclear Plants Distribution Industry [2]

XLPE Reynolds, 1975-1980 Moderate 34 years 10-12 years Water degradation failures have Cyprus, and been observed in the nuclear others industry starting at 24 years of service.

Filled GE 1968 Single plant No longer in 10-12 years? Failures were observed starting at XLPE service in wet 10 years of service, with many conditions failures between 10 and 25 years.

Butyl GE, Collyer, and 1967-1972 Small 42 years 20-25 years Water degradation failures have rubber Okonite been observed in the nuclear industry starting at 25 years of service.

Black Okonite, Bulk 1971- Large 38 years 20-25 years? There have been 26 failures to date EPR Anaconda, and 1979, last 1986 in the nuclear industry with 20-30 General Cable years of service.

Brown Kerite Bulk 1972- Moderate 37 years 20-25 years No water-related failures have been EPR 1985, some observed to date in the nuclear 1990-2003 industry.

Pink EPR Okonite 1978 to present Newer plants 31 years 20-25 years No water-related failures have been and observed to date in the nuclear replacements industry; one manufacturing defect-related failure has been observed.

7 This material is not covered in EPRI report Plant Support Engineering: Aging Management Program Development Guidance for AC and DC Low-Voltage Power Cable Systems for Nuclear Power Plants, (1020804) [I]. The onset of significant water degradation may be somewhat earlier than listed.

4-2

Susceptibility of Cables to Water-Related Degradation by Type Table 4-1 (continued)

Cable Susceptibility Under Wet Conditions (population data are from the Nuclear Energy Institute [NEI] 2005 Survey on Underground Cables)

Material Manufacturers of Approximate Population of Oldest Nuclear Earliest Expected Nuclear Industry Actual Installed Cable Period of Installed Cables Plant Cables as Onset of Water Experience Discussion Installation at Nuclear of 2009 Degradation in Plants Distribution Industry [2]

Pink EPR AnacondaiCablecl 1978 to Newer plants and 31 years 20-25 years Some early failures with water BICC and General present replacements combined with manufacturing defect Cable have been observed; there have been no water degradation alone failures reported in the nuclear industry.

TR-XLPE Not known 2004 Rare 5 years 20-25 years There is an insufficient population replacement and period of service to make inferences.

0 G) 0 0

0 N

-....J I

(J) 4-3 0

0 0

.j::>.

0

Susceptibility of Cables to Water-Related Degradation by Type For the other insulations commonly used in nuclear plants, EPRI Report 1008560 indicates that earliest expected onset of significant water-related degradation occurs in approximately 20-25 years of wet service [2]. Butyl-rubber-insulated cables were the first type of rubber-insulated cables used in nuclear plants. Only a few plants purchased these cables before the rubber-insulated cable industry converted to black EPR. At 25 years of service, water-related failures were identified in the nuclear industry.

Black EPR replaced butyl rubber in the early 1970s and is the insulation with the largest population of cables with approximately 48% of nuclear plants reporting its use. The first water-related failures occurred at approximately 20 years of service, and more than 26 failures 8 have occurred in the nuclear industry as of this writing.

Brown EPR insulation, while being available to the early nuclear plants, continues to be produced. Approximately 20% of plants report its use. No water-related failures have been reported in the nuclear industry to date.

Pink EPR replaced black EPR in the cable industry in the mid- to late 1970s, and approximately 30% of nuclear plants report its use. To date, the only failures related to water degradation have been associated with manufacturing defects or highly localized degradation likely associated with a local flaw. No bulk water-related degradation and failure has been reported in the nuclear industry.

Recommendations Based on Susceptibility Assessment Table 4-2 provides the recommendations concerning the timing of cable aging management programs for wet medium-voltage cable based on insulation type. See Sections 6 and 7 for guidance.

Table 4-2 Aging Management Recommendations Based on Insulation Material Type Material Manufacturers of Approximate Recommendation for Wetted Cable Installed Cable Period of Circuits Installation XLPE Reynolds, GE, Cyprus, 1975-1980 Implement aging management program.

and others Filled XLPE No longer in use in wet circuits Butyl rubber GE, Collyer, and Okonite 1967-1972 Implement aging management program.

Black EPR Okonite, Anaconda, and Bulk 1971- Implement aging management program.

General Cable 1979, last 1986 8 From the 2005 NEI industry survey of underground cable installed and failure information, approximately 1400 black EPR cables were originally installed underground in potentially wet conditions.

4-4 OAGI0001276_00041

Susceptibility of Cables to Water-Related Degradation by Type Table 4-2 (continued)

Aging Management Recommendations Based on Insulation Material Type Material Manufacturers of Approximate Recommendation for Wetted Cable Installed Cable Period of Circuits Installation Brown EPR Kerite Bulk 1972- Implement aging management program 1985, some for cables with more than 30 years of 1990-2003 service.

Pink EPR Okonite 1978 to present Implement aging management program for cables with more than 30 years of service.

Pink EPR Anaconda/Cablec/BICC, 1978 to present Implement aging management program and General Cable for cables with more than 30 years of service.

TR-XLPE Not known 2004 Implement aging management program for cables with more than 30 years of service.

Note: The timing of the reconnnendations in this table is based on the actual nuclear plant experience for the cable type described in Table 4-1.

Susceptibility of Splices to Wetting Most plants have short enough on-site runs of medium-voltage cable that no splices exist. Other plants have some long runs, generally to intake structures or ultimate heat sinks, that require splices to complete the circuits. Offsite medium-voltage feeds to the plants likely have splices and also may be distribution-type cables with concentric neutral wires rather than helical tape shields. Some plants may have splices in circuits where wetted sections have been replaced while the dry portion of the circuit was retained. Splices in dry sections of cables should be long lived provided they are reasonably well made. However, splices in wetted sections of cables will be more susceptible to water-related degradation if there were errors in assembly. Cables having splices in wetted sections should be included in the scope of the program, no matter what type of insulation is present. 9 Only a limited number of circuits will have splices, and an even smaller set of these is expected to be subject to wet conditions. For these splices, local assessment through Lemke partial discharge probe or an ultrasonic acoustic probe may be most useful given that attenuation of partial discharge by the helical shield may preclude assessment of the splice from the circuit terminations. In some cases, splice deterioration from internal tracking or discharging may be observable through careful assessment of infrared thermographs.

9 Cables with wetted splices that are subject to lightning strikes (that is, aerial sections or aboveground outdoor terminations) and that do not have lightning arrestors or surge suppressors may need additional assessment or testing. Cables with wetted splices that are not subject to lightning strikes (that is, terminated inside grounded structures) should be assessed starting at the point listed in Table 4-2 for the particular cable type unless adverse plant or industry operating experience dictates that a problem may exist with the splices.

4-5 OAGI0001276_00042

5 ACTIONS FOR CABLES HAVING WET ENVIRONMENTS Long-term wetting of energized medium-voltage cables can cause water-related degradation of the insulation. Earlier vintages (late 1960s through late 1970s) are more susceptible than modern cables, especially if small imperfections existed in the insulation or at its boundaries. Cables that have either been continuously wet or wet for an extended period, whether one long wetting or several shorter but significant length periods, may have experienced degradation and should be evaluated. Electrical testing of the cable will allow assessment to determine whether significant degradation has occurred. An alternative to implementing a test program would be to replace 5-1 OAGI0001276_00043

Actionsfor Cables Having Wet Environments cables based on duration of wetting and expected susceptibility to wetting. A cable shield is required to do off-line testing. Actions for nonshielded cable are covered separately at the end of this section. The testing discussions in this section assume the presence of an insulation shield.

Pumping of Manholes and Ducts Removing water from around the cable will not reverse degradation that has occurred. However, it may reduce the rate at which further degradation takes place, and some moisture may be expelled from the cable if there is ohmic heating during operation. The rate of deterioration in breakdown strength should slow and a small improvement may occur. Accordingly, instituting a pumping program, installing automatic sump pumps, or repairing failed automatic pumping systems is recommended.

It is recognized that not all systems can be pumped dry. Continued operation of cables under wetted conditions is allowable, but condition of the cable insulation should be proved through periodic assessment.

Rain and drain conditions should not adversely affect jacketed cables as long as wetting does not last for more than a few days on average. Water takes several months to years to migrate into the jacket and then a similar period to migrate into the insulation under wetted conditions and migrates out much more slowly when ducts are dried.

Assessment of Condition of Shielded Cables Three practical tests lO are currently available for shielded extruded polymer medium-voltage cable: partial discharge, tan 0,11 and power frequency or very low frequency (VLF) withstand. 12 Depending on the nature of the cable design and the cable or accessory (termination or splice) concern, one or more of these tests would be recommended and others would be unsuitable.

These test methods may be performed at line frequency or VLF (for example, 0.1 Hz). VLF test sets are frequently favored because they are readily portable and compact in comparison to line frequency systems that often must be truck mounted because of their size and weight. It should be noted that these tests assess the insulation of the terminations and splices if they are included in the test circuit. These tests are often performed off-line with elevated voltage. When elevated voltage testing is performed, whether during plant operation or during a refueling outage, it should be done with consideration of the time, materials, and personnel required to repair or replace the cable. This is not meant to be a reason not to test cables, but to ensure that due 10 Other tests are available but are not commonly performed in the United States: oscillating wave partial discharge assessment, return current assessment, and return voltage assessment.

11 Dielectric spectroscopy is a related test in which the insulation is subjected to dielectric assessment at multiple frequencies and voltage levels.

12 The informal term for "withstand" testing is hi-pot. In this document, withstand testing is synonymous with high-potential (hi-pot) testing.

5-2 OAGI0001276_00044

Actionsfor Cables Having Wet Environments consideration is given to the impact of a cable replacement because a cable is found to be severely degraded or fails during a test. Testing during a refueling outage should be performed near the start of the outage to preclude extension of the outage.

Partial discharge tests are used to detect discharges that occur in gas voids within the insulation system. The discharges occur when the electric stress across the void is high enough to cause a breakdown of the gas void. The voltage is then distributed across the remaining intact insulation.

Partial discharge can also occur along a surface interface or between a floating conductor and an energized electrode. For example, a partial discharge can occur on the outside of the insulation shield between the shield and a corroded tape or wire. Each discharge causes a small amount of damage to the surface of the insulation in the void causing a carbonized path to develop through the insulation, which can eventually lead to cable failures. Partial discharges result in high-frequency, low-energy signals that can be attenuated. Partial discharges in the insulation at operating voltage create electrical trees, which can propagate through the insulation relatively rapidly (for example, days to months). It should be noted that surface discharges are not as harmful and do not cause rapid damage. Accordingly, understanding the nature of the discharge involved is important and part of the art of interpreting partial discharge results.

Offline partial discharge testing is an elevated voltage test that can be performed at line frequency or VLF. Partial discharge testing can locate the site of the discharge along the length of the cable. Partial discharge testing may be most useful in detecting termination and splice problems, especially on off-site cable feeds and is useful for commissioning tests of new installations. However, water-related degradation, such as water treeing, does not produce partial discharge signals. In addition, in the case of helical tapes, corrosion of the tapes from long-term wetting is likely to severely attenuate partial discharge signals and impede their detection.

Accordingly, partial discharge testing is likely to be unsuitable for evaluation of most of the wetted medium-voltage cables in use in nuclear plants.

Tan () testing (also called dissipation factor testing) determines the ratio of the resistive leakage current through the insulation divided by the capacitive current and provides a figure of merit relating to the condition of the insulation. It is, therefore, also independent of the length of cable.

Tan delta has no units 13 and is generally a small number given in terms of 10-3 . Tan 8 is a bulk test and does not provide specific location information for identified degradation. It can be performed at line frequency or VLF and is generally performed at discrete voltage levels of 0.5, 1.0, 1.5, and 2 times line-to-ground voltage (Vo) (but no greater than the withstand voltage level derived from IEEE Std 400.2 [12] 14). Tan 8 values that are elevated or unstable at a particular test voltage level or values that increase or decrease significantly with increase in voltage are indicative of deteriorated insulation. This test can identify insulation systems with distributed water-related degradation. However, if a cable insulation system has only a single but significant 13 Technically, tan 8 is the measurement of an angle in radians but is rarely expressed that way.

14 IEEE Std 400.2 provides withstand test voltages that are applicable to all of the extruded insulation systems in use in nuclear plants.

5-3 OAGI0001276_00045

Actionsfor Cables Having Wet Environments flaw, tan 8 may not necessarily detect it. In addition, the test does not discriminate between many widespread limited degradations and a smaller number of more severe degradations. VLF withstand testing may be used to identify severe localized conditions as described next.

Dielectric spectroscopy is a related test that performs tan 8 measurement at several frequencies and voltages.

VLF withstand testing applies an elevated voltage for a significant period in an attempt to purposely cause a significantly weakened location in the insulation to break down during the test.

The test is generally a go/no-go test. The concept is that if the cable does not breakdown during the test, it will perform satisfactorily for a reasonable period. The test would be applicable to the detection of localized, significant degradations, but it would provide no information concerning wide spread, low-level water degradation. VLF test voltage withstand levels and durations are defined in IEEE Std 400.2 [12]. Some testers are evaluating longer duration, lower level tests.

Some VLF units measure the real part of the complex current providing the insulation resistance or "megohm" value. (Note: 60 Hz testing is acceptable. In general, practical limitations of portability of the test equipment and limited space within the power plant may preclude its use.)

Effects of Shield Design and Insulation on Long-Term Testability and Test Selection Wet aging of medium-voltage cables affects two components of the cable: the metallic components of the shield and the insulation. When water migrates through the jacket, it causes a light corrosion of the metallic shield. Although light corrosion does not adversely affect the main functions of the shield, it can adversely affect its testability when attempting to use high-frequency measurement techniques such as time domain reflectometry (TDR) or partial discharge techniques [13]. The following four basic types of metallic shield are in use in nuclear power plant cables:

• Helically wound tapes

• Distributed drain wires

• Longitudinally corrugated copper shield

• Concentric neutral wires Helically wrapped copper lS tapes are the most common type of metallic shield. These are used in most XLPE and EPR nuclear plant cables. The helical tape allows reasonable flexibility by comparison to large conductor, concentric neutral wire systems. The next most common shield is used in the UniShield compact design and uses six drain wires that parallel the conductor. The least commonly used design in nuclear plants is the concentric neutral design, in which large strands of wire or straps are arranged around the entire surface of the semiconducting shield.

This design is common in distribution systems and can be found in offsite feeds. The newest shield design is the longitudinally corrugated copper shield, which is made by forming a 15 A limited number of cables have used zinc tapes instead of copper.

5-4 OAGI0001276_00046

Actionsfor Cables Having Wet Environments longitudinally corrugated copper sheet around the polymer shield and sealing the overlap. This type of shield provides a continuous barrier to keep moisture from entering the insulation system and should improve the ability to test for partial discharge because the tube presents low impedance to high-frequency signals even iflight surface corrosion occurs.

Corrosion of a helical copper tape shield is the most problematic for high-frequency diagnostic techniques such as time domain reflectometry and partial discharge testing. When the copper tape is new, the tape shield acts as a tube providing a good conducting path for high-frequency signals. If partial discharge were present, the new tape shield would present little attenuation.

However, with a light corrosion on the surface of the tape, the lapping of the tape becomes insulated from one wrap to the next. At this point, the helical tape acts as an inductor, especially to high-frequency signals such as partial discharge. Accordingly, deteriorated tape shields are likely to occur at the same time as the potential for deteriorated insulation. In cases in which the attenuation resulting from metallic shield corrosion prevents a reflection of the test signal from the far end of the cable, it prevents the detectability of partial discharge signals that may occur in or near the wet section. To a lesser extent, the insulation also attenuates the high-frequency signals with XLPE having the least attenuation, EPR having higher attenuation, and butyl rubber having the highest. If partial discharge testing is to be used on wet cables having tape shields, a pulse should be injected, and using TDR techniques, the sensitivity and high-frequency transmission characteristics of the cable should be determined to verify whether the shield condition is adequate to support partial discharge testing. Light corrosion on the distributed wire, concentric neutral, and longitudinally corrugated copper-type shields should not cause excessive partial discharge test signal attenuation. Although this phenomenon affects the ability of certain diagnostic techniques to assess the condition of the cable, there is no evidence that it is detrimental to normal operation or reliability of the cable.

The choice of test depends on the nature of the problem that is of concern. In most cases, the concern for wetted cable is water-related insulation degradation (for example, water treeing).

However, in cases in which splices have been used in the system, partial discharging or tracking within the splices could lead to failure. Most cables in nuclear plants do not have underground splices. However, in cases in which underground splices do exist and attenuation resulting from metallic shield corrosion is not a concern, partial discharge testing may be appropriate and useful.

Water-related degradation of the insulation does not in itself generate electrical discharge signals. Only when the water-related degradation (for example, a water tree) is so severe that the electric field in the surrounding good insulation is excessive (that is, high enough to cause a water tree to convert to an electrical tree) and partial discharge begins does the degradation generate electrical discharge signals. Water-related degradation is a long, slow process; partial discharges that would lead to failure occur and are detectable only late in the degradation cycle.

However, as the water-related degradation progresses, the electrical leakage current through the insulation increases and tan 8 testing, which measures this loss, should detect the degradation.

5-5 OAGI0001276_00047

Actionsfor Cables Having Wet Environments Combined Testing The foregoing description indicates that tan 8 measurement is most likely to be useful for detection of water-related degradation for the cable designs commonly used in nuclear plants (for example, lossy insulations with or without helical tape shields). Tan 8 could be complemented with VLF withstand testing. Passing a withstand test after a successful tan 8 test indicates that there is no significant distributed or local degradation in the insulation system. Tan 8 testing would evaluate the cable for water-related degradation, and the VLF withstand test would determine whether severe localized degradation existed. The tan 8 test is a global assessment that identifies more widespread deteriorations in the insulation system. Because it provides more of an "average" result over the length of the insulation, it might not be as sensitive to a single local defect. The VLF withstand test, on the other hand, does not provide an indication of the overall aging of the insulation but is designed to force a significant local degradation to failure.

The use of a VLF withstand test following a VLF tan 8 test is an engineering decision. For cables having a tan 8 result that is "good," a VLF withstand test is optional. However, consideration should be given to performing a VLF withstand test should a "further study required" or "action required" result occur (see next subsection).

One concern regarding the VLF withstand test is that the test may cause an already severe defect to progress toward failure, but that the duration of the test is not sufficient to bring the defect to failure during the test. As a result, failure could occur during the next period of operation. One way of gaining insight about whether the cable insulation is free of significant defects and that "partial completion" of a failure has not occurred is to perform the VLF withstand while measuring tan 8. If the cable passes the withstand test and the tan 8 value is stable throughout the withstand period, partial completion of a degradation is unlikely to have occurred. If the cable passes the withstand and the tan 8 is increasing or decreasing during the application of voltage, partial completion is likely, and further investigation or extension of the test period is recommended.

An alternative to coupling tan 8 testing with VLF withstand would be to couple tan 8 testing with partial discharge testing. The tan 8 test would assess the cable for distributed water-related degradation, and the partial discharge test would assess for localized, severely degraded conditions. The use of partial discharge testing would be predicated on the cable having acceptable attenuation levels for detection of the high-frequency signals related to partial discharge. Similarly, dielectric spectroscopy could be coupled with partial discharge testing.

Tan () Methodology and Assessment Criteria Tables 5-1 through 5-4 provide preliminary criteria for assessing cable insulation degradation through tan 8 testing. IEEE Std 400 should be consulted for assessment criteria for XLPE insulation [14]. Tan 8 testing is typically performed in steps from 0.5 V o (Vo is the phase-to-ground rms operating voltage), Vo, 1.5 Vo, and 2 V o. (Note: The IEEE Std 400.2 standards group is expected to reduce the upper test voltage to 1.5 Vo in the next revision of the standard. [12])

The tan 8 value should change very little as voltage is raised and should remain stable during the application of voltage at each step. If the tan 8 value is elevated or if there is a significant 5-6 OAGI0001276_00048

Actionsfor Cables Having Wet Environments increase in value as voltage is increased, the cable is considered to be in aged condition and likely in a deteriorated state. Evidence that elevated standard deviations during the hold at each voltage level also indicates degradation in the insulation system is building. A separate description of its use follows this section.

Table 5-1 Preliminary Tan () Assessment Criteria for Butyl Rubber (in terms of x 10.3 ; 0.1 Hz test frequency) (Note 1)

Absolute Value of the Difference in Tan ()

Condition Tan () Between 0.5 Vo and 1.5 Vo (Notes 2 and 3)

Good  :::;12 and  :::;3 Further study required 12< tan 8 :::;50 or 3+ to 10 Action required >50 or >10+

Notes:

1. This is based on Figure C-13 in EPRI report Plant Support Engineering: Medium-Voltage Cable Aging Management Guide (1016689) [15] and in-plant test results and consultation with tan 8 testers.

2. Differentials may be taken at 1 Vo and 2 Vo at the user's option. See text preceding this table.

3. The difference in tan 8 is normally positive. Negative differences should be treated as very significant and may indicate a problem with a test or an indication of the presence of a significant defect.

Table 5-2 Preliminary Tan () Assessment Criteria for Black EPR (in terms of x 10.3 ; 0.1 Hz test frequency) (Note 1)

Absolute Value of the Difference in Tan ()

Condition Tan () Between 0.5 Vo and 1.5 Vo (Notes 2 and 3)

Good  :::;12 and  :::;3 Further study required 12< tan 8 :::;50 or 3+ to 10 Action required >50 or >10+

Notes:

1. This is based on Figure C-l in EPRI Report 1016689 [15] and associated in plant results and consultation with tan 8 testers.

2. Differentials may be taken at 1 Vo and 2 Vo at the user's option. See text preceding these tables.

3. The difference in tan 8 is normally positive. Negative differences should be treated as very significant and may indicate a problem with a test or an indication of the presence of a significant defect.

5-7 OAGI0001276_00049

Actionsfor Cables Having Wet Environments Table 5-3 Preliminary Tan () Assessment Criteria for Pink EPR (Note 1) (in terms of x 10-3 ; 0.1 Hz test frequency) (Note 2)

Absolute Value of the Difference in Tan ()

Condition Tan () Between 0.5 Vo and 1.5 Vo (Notes 3 and 4)

Good  :::;15 and  :::;3 Further study required 15< tan 8 :::;30 or 3+ to 8 Action required >30 or >8+

Notes:

1. This may also be used for "Gray" UniBlend EPR (approximate time of manufacture from late 1970s on).

2. This is based on Figures C-3 and C-4 in EPRI Report 1016689 [15] and consultation with tan 8 testers.

3. Differentials may be taken at 1 Vo and 2 Vo at the user's option. See text preceding these tables.

4. The difference in tan 8 is normally positive. Negative differences should be treated as very significant and may indicate a problem with a test or an indication of the presence of a significant defect.

Table 5-4 Preliminary Tan () Assessment Criteria for Brown EPR (in terms of x 10-3 ; 0.1 Hz test frequency) (Note 1)

Absolute Value of the Difference in Tan ()

Condition Tan () Between 0.5 Vo and 1.5 Vo (Notes 2 and 3)

Good  ::::;50 and  ::::;5 Further study required 50< tan 8 ::::;60 or 5+ to 15 Action required

>60 or >15+

Notes:

1. This is based on Figures C-3 and C-4 in EPRI Report 1016689 [15] and consultation with tan 8 testers.

2. Differentials may be taken at 1 Vo and 2 Vo at the user's option. See text preceding these tables.

3. The difference in tan 8 is normally positive. Negative differences should be treated as very significant and may indicate a problem with a test or an indication of the presence of a significant defect.

Starting with 0.5 Vo allows the tester to determine whether the cable is significantly deteriorated at or before applying more than line-to-ground voltage. If deterioration is observed at the lower test voltages, the test can be terminated to avoid a severely deteriorated cable from failing under the higher test voltages and allow limited continued use of the cable. The assessment criteria for EPR and butyl rubber systems are based on limited field feedback and laboratory test information and consultation with tan 8 test engineers. Confirmation or adjustments to these values will occur as further application of the test and correlation of test data to cable condition 5-8 OAGI0001276_00050

Actionsfor Cables Having Wet Environments is developed in the nuclear industry or other data related to the cable designs in nuclear plants becomes available. The XLPE criteria from IEEE Std 400 are based on results from the distribution industry where XLPE is a commonly used insulation [14].

A minimum of eight tan 8 measurements should be performed at each voltage step during a tan 8 test. Careful attention should be paid to the trend of the tan 8 values with time, particularly at applied voltages above the normal phase-to-ground operating voltage. Significant increases and decreases in tan 8 with increasing voltage and/or instability during a voltage step are indicative of deterioration in the insulation or accessories. Poor grounding and/or significant corrosion of the ground shield of the cable will also contribute to elevating the tan 8 values.

Tables 5-1 through 5-4 give ranges of "good," "further study required," and "action required."

However, the absolute value of tan 8 is somewhat less important than the stability of the value with increasing voltage. A significant increase or decrease in tan 8 value as the voltage is increased during the test indicates that the leakage current through the insulation is changing. An increase in leakage current with increasing voltage indicates that the material is discharging at higher voltages and is not stable. A decrease in tan 8 or a tan 8 that alternately increases and decreases with increasing test voltage is unusual and indicates either a problem with the test process or a significant problem with the cable that should be evaluated further. Very high tan 8 measurements indicate large leakage currents and could be indicative of many water-related degradation sites or a smaller number of highly deteriorated sites.

Anecdotal information indicates that deteriorated cables that had been installed before the mid-1970s tend to have many degradation sites and cables manufactured later will fail less frequently but from a single, large manufacturing flaw (that is, manufacturing and design improvements reduced the likelihood of overall insulation degradation of the insulation from wet conditions, but occasional manufacturing flaws still exist). Accordingly, for very early cables, elevated tan 8 may be a strong indication of degradation, but for later cables, the differential value between 0.5 Vo and 1.5 Vo is likely to be a better indicator. The criteria provided in Tables 5-1 through 5-4 are preliminary values based on data from research and in-plant testing. The values have been chosen to identify degradation of concern. The criteria for considering "action required" have been chosen to be reasonably conservative. Corrections to these values are expected as plants implement assessment and further data are generated and as IEEE Std 400 and 400.2 are revised [14, 12].

Cables with results in the "further study required" range should be subjected to more frequent testing (for example, once per refueling cycle) to determine whether the condition is stable or worsening. Cables with results indicating "action required" should be replaced as soon as is practicable or additional testing (see VLF hi-pot in "Failure of Cable Under Test" later in this section) performed to verify serviceability. Section 7 addresses options for repair and replacement of cable. Consideration should be given to performing a VLF withstand test should a "further study required" or "action required" result occur. In the case of an "action required" result, a successful VLF withstand test would provide a basis for waiting until a more convenient time for repair or replacement of the circuit.

5-9 OAGI0001276_00051

Actionsfor Cables Having Wet Environments Figure 5-1 provides a pictorial version of the assessment criteria shown in Table 5-3 for pink EPR to help in understanding how the criteria work. The tan 8 values are shown on the vertical axis, and the change in tan 8 values between the points of 0.5 Vo and 1.5 Vo are shown on the horizontal axis. Staying within the limits of the green box for tan 8 and delta tan 8 means that the cable is in good condition. If the tan 8 exceeds 15 x 10-3 but is less than 30 x 10-3 , further assessment is needed. At a minimum, the period between tests should be shortened, and the potential cause of the elevated value sought. Similarly, a delta tan 8 greater than 3, whether increasing or decreasing, would be a cause for shortened periods between tests and a review to determine the cause. If the tan 8 exceeded 30 or the delta tan 8 exceeded 8, action should be taken to repair or replace the cable.

Further Study Required Absolute Value of Delta Tan i'5 (10-3 ) Between 0<5 Vo and 1.5 Vo (Note 1)

Note 1 Tile difference in tan c5 is normally positive. Negative differences should be treated as very siqnificant and might indical.,: a problem with a test or the presence of a siqnificant defElct.

Figure 5-1 Pictorial Representation of Tan 8 Assessment Criteria for Pink Ethylene Propylene Rubber Figure 5-2 shows a typical tan 8 plot for a shielded EPR-insulated three-phase cable. For each of the three phases, the tan 8 is stable through the range of test voltage, showing a very small increase with each step in voltage. The standard deviation at each step was very small during the period of voltage application for the step. It should be noted that this test combined VLF tan 8 with a VLF withstand test. The last voltage step for this test was at 7 kV, the withstand test value. The duration of the 7 kV step was 30 minutes.

5-10 OAGI0001276_00052

Actionsfor Cables Having Wet Environments TD Value [E-3] 11.5 11.5 11.7 12.0

• S'icl:'riev:*t;;/~j""'" 0.00 0.00 0.00 0.00 789

,... {:.... Phase A ....:,:,:.... Phase B ~ PI'lase C Voltage [kVrms]

Std. Dev. = Standard Deviation Figure 5-2 Typical "Good" Tan 8 Result for a Shielded Cable with Pink Ethylene Propylene Rubber Insulation Figure 5-3 shows a tan 8 plot for a UniShield 13.8 kV pink EPR cable having degradation on the B phase. The plots for the A and C phases have acceptable tan 8 values that are stable through the four steps in applied voltage and had very small standard deviations at each voltage step. The B phase had good measurements at the first two voltage steps but began an upward trend at the third step with a very significant increase at the fourth step. The tip up at higher voltage indicates instability in the insulation that indicates a weakness requiring further study.

5-11 OAGI0001276_00053

Actionsfor Cables Having Wet Environments 20 18 16******************************.. ********************* ................................................................................................ .

14

~ 12 oJ)

~ 10 * * ** /..

(ij o 8* * * * .................**. *

• I-6 4

2 2 4 6 8 10 12 14 16 18 20 Figure 5-3 Tan 8 Plots for a 13.8 kV UniShield Cable with a Degraded B Phase Use of Standard Deviation in Assessing Tan () Results The standard deviation of a set of tan 8 measurements at a particular test voltage may provide additional information relating the onset of degradation. The standard deviation of the tan 8 measurements identifies whether the value is stable or changing during the voltage step. The standard deviation is an additional indicator of instability in the insulation, especially at lower test voltages, even when the tan 8 and delta tan 8 values may still be within acceptable limits.

Some test sets automatically calculate the percent standard deviation (standard deviation times 100 [for example, for a standard deviation of 1.5 x 10-3 , the percent standard deviation would be 0.15]). Other test sets would require the individual measurement to be placed in an electronic spreadsheet to calculate the standard deviation. The formula for determining the percent standard deviation and an example of the data are contained in Appendix A The consideration of percent standard deviation appears especially important for tests that may be limited to 1.0 Vo as may occur when replacement cable is not immediately available. The suggested accepted criteria are shown in Table 5-5. Although the use of percent standard deviation is in its infancy, it appears to provide additional insight into degradation that may not be obvious from review of absolute and delta tan 8 results.

5-12 OAGI0001276_00054

Actionsfor Cables Having Wet Environments Table 5-5 Preliminary Percent Tan 8 Standard Deviation Assessment Criteria for Rubber Insulated Cables (Note 1)

Percent Standard Deviation of Tan l)

Condition Measurements at a Particular Test Voltage Good  ::::;0.02 Further study required 0.02+<standard deviation [%]<0.04 Action required

>0.04 Notes:

1. Insufficient information exists to provide different criteria for different rubber polymers.

° Figure 5-4 provides an example of a case in which a 15 kV cable was subjected to tan test. Had the test been stopped at 1 Vo (8 kV) and only the absolute value of tan and delta tan been

°

° considered, all three phases of the cable would have been deemed acceptable. However, evaluation of the percent standard deviation at 8 kV finds the B phase acceptable, a small shift for the A phase (0.01), and a concern for the C phase (0.02). Evaluation of the test at 1.5 Vo shows that all figures of merit are indicating that problems exist for the A and C phases, with 2 Vo showing that the cable is highly degraded. At the higher voltages, the percent standard deviation is extremely high by comparison to the assessment criteria. In this case, the C phase cable failed after 30 seconds at 2 Vo, indicating that percent standard deviation as well as

° ° absolute tan and delta tan were strong indicators of deterioration of the C phase cable insulation. The B phase had acceptable tan 0, delta tan 0, and percent standard deviation values through all test voltages.

5-13 OAGI0001276_00055

Actionsfor Cables Having Wet Environments

.. TI).v.~lue.IE~3l ................... 82 ...............8.:3....... ;.. . __________________ J____________________ !_ _

Std, Dell. [%1 DOC C.ClC C::iO Fh:<;:$~ G* ~t.H:~Hr::<<r.;:: t; ,1 H~. ;§'::.:n. fj. nF

• • ~ld2~#~:*1%iS.] Q~=~ . . . . c'~q~..............~:~~. . . . . . . . . . . . . .~.~~~.................................;..........................................,

l)(J 80 7(:

.~

'? :3G

~

~ Ed]

"i5 Cl 4(]

I-3D 2(:

IG

J Voltage [kVrmsj Std. Dev. = Standard Deviation Figure 5-4 Tan 8 Example Including Percent Standard Deviation Data Very Low Frequency Withstand Testing in Conjunction with Tan () Testing VLF withstand testing may be performed in conjunction with tan 8 testing as described previously. If single severe local degradations are present, the VLF withstand test is designed to cause them to fail at the time of the test.

If a cable has indications that it is degraded based on tan 8 results, a VLF withstand test could be used to determine whether the cable has localized highly degraded segments (that is, fails the withstand test) or lesser, more distributed degradation based on it passing the withstand test. If the cable passed the test, there is a reasonable likelihood of functioning until a more convenient time for repair or replacement of the cable.

5-14 OAGI0001276_00056

Actionsfor Cables Having Wet Environments The methodology for VLF withstand testing, including discussions of test voltages and durations, are contained in IEEE Std 400.2 [12]. The assessment criterion is simple. If the cable does not fail during the test, it is judged adequate for continued use. An improvement to a standard VLF withstand testing is to monitor tan 8 for stability during the withstand test. A successful withstand test with tan 8 stability indicates that there were no significant defects and that none were in the process of breaking down. Tan 8 instability during the withstand test likely indicates that a significant degradation site exists and is in the process of going to failure.

Test Methodology and Assessment Criteria for Other Tests Descriptions of off-line dielectric spectroscopy and partial discharge testing are beyond the scope of this report. When such tests are used to assess the condition of cable, the testers and interpreters of the test data must have the requisite skills and related experience to apply the test and to perform the interpretation of the results. Experience with the specific cable type (insulation and design) is recommended. EPRI report Plant Support Engineering: Medium-Voltage Cable Aging Management Guide (1016689) provides additional information on these testing techni ques [15].

Test Preparation and Concerns To allow testing of cable circuits with either off-line partial discharge testing, tan 8, or withstand testing, the connected load (for example, a transformer, motor drive, or motor) must be disconnected and the terminations isolated. All surge and lightning arrestors and voltage transducers must also be disconnected from the circuit under test. In some cases, testing may be performed through the backplane of the associated breaker cubicle; but before this is attempted, it is recommended that initial testing be performed with a cable connected to and then isolated from the cubicle to determine the effect of the cubicle backplane. If there is a significant influence from the breaker backplane, the cables should be disconnected from the breaker backplane. Testing of XLPE cables is likely to require separation of the cable from the breaker backplane to reduce inaccuracies.

Testing must be performed with a prepared termination at each end of the cable to control voltage stresses. For 4.16-6.9 kV applications, removal of the metal shield and insulation semiconducting layer in preparation for installation of a termination will suffice to allow testing to 2 Yo. However, the length of insulation after removal of the semiconducting layer will be too short when a splice is being prepared, and stress control will be necessary. Testing of 12 kV and higher circuits will require stress control to prevent flashover. Under no circumstance should testing be performed with the insulation shield in place to the end of the insulation at either end of the circuit; doing so will cause a flashover. Suitable stress cones must be used at each termination. The cable's metallic shield must remain grounded during testing. In addition, before testing, terminations should be inspected for cleanliness and general condition. Cleaning and/or repair may be necessary before cable testing occurs. In addition, the terminations must be located well away from the terminations of adjacent phases and the termination cabinet/box to reduce corona effects during the test. In confined spaces, use of insulating sheets (for example, Mylar) may help to ensure adequate separation between phases and from phase to ground.

5-15 OAGI0001276_00057

Actionsfor Cables Having Wet Environments In tan 8 testing, cables of different types (for example, XLPE, butyl rubber, and EPR) ideally should be tested separately if spliced together. Otherwise, the material having higher losses will mask problems in the less lossy material. For example, severe deterioration in XLPE would likely be masked by the naturallossiness ofEPR or butyl rubber, and butyl rubber could mask problems in EPR. In addition, if splices are used in circuits having the same cable materials, nonlinearity of the insulation resistance of some splice types may cause abnormal tan 8 results that are not indicative of the condition of the cable insulation (see EPRI report Plant Support Engineering: Medium-Voltage Cable Aging Management Guide [1016689] Appendix C) [15].

When testing cables with multiple conductors per phase, tan 8 testing of the individual conductors in each phase will provide the best results. It is understood that separating the conductors can be difficult and time consuming. If the insulations of each of the conductors deteri orated simul taneousl y, the tan 8 measurement of the joined conductors woul d obvi ousl y indicate insulation condition. However, if only one conductor of a multiconductor-per-phase circuit were deteriorated, testing of the joined set could mask the deterioration. Accordingly, if joined-set testing is being performed, careful scrutiny of the results is recommended to determine whether further testing of the separate cables is warranted.

Regarding testing of motor and transformer circuits, if tests are to be performed periodically, separable connectors and disconnects should be considered so that the motor connections can be easily broken and remade. Use of separable disconnects will be advantageous to the testing of the motor or transformer and the cable. (Note: As of this writing, separable connectors having a manufacturer's environmental qualification do not exist.) Motor testing frequently entails use of high-voltage dc, which is not recommended for cables because it has been linked to damage of the cables' extruded polymer insulation. Separation of the motor from the cables is recommended when dc or surge testing is performed on the motor. Testing of cables with off-line ac tests would provide no useful information about the cable if the motor remained connected to the cable.

Failure of Cable Under Test Concern exists that a "good" cable may fail under a test using elevated voltage. The tests recommended herein have elevated rms test voltages of 1.5 to 2 times line-to-ground voltage in accordance with IEEE Std 400.2 [12]. These ac tests are performed with either line frequency (for example, 60 Hz) or VLF (for example, 0.1 Hz).

The insulation of a new cable can withstand 30 times line-to-ground voltage or more before breaking down. Cable insulation that cannot withstand twice line-to-ground voltage for the duration of an off-line test is highly degraded and is not in a condition considered satisfactory for continued service. When performing tan 8 or partial discharge testing, test operators can generally identify when cables have inadequate insulation properties and stop testing before failure. Occasionally, a failure will occur during test. In these cases, the cable failed because it was degraded, not because the test caused rapid deterioration.

5-16 OAGI0001276_00058

Actionsfor Cables Having Wet Environments Ac withstand tests may be used to cause weakened cable to purposely fail by applying elevated voltage for an extended period. The cable users industry continues to refine withstand testing methodology to ensure that significant degradations are brought to failure during test and that no lesser degradation is aggravated so that failure occurs shortly after return to service.

Dc withstand testing is not recommended. Although it was effective for evaluating paper-insulated lead-covered cables, it has been found on some cable designs to worsen end-of-life degradation of polymer insulations but not necessarily cause cable failure at the time of test. The cables fail in dc withstand testing only ifvery severely aged. Under conditions in which the cable is slightly less aged, the application of a dc test can cause an existing flaw to convert to electrical discharge so that the cable fails a short period after the return to service [16]. Because passing a dc withstand test may cause a false sense of security and because the dc testing may shorten the remaining limited life, dc withstand testing is not recommended for the purposes of ensuring continued functionality of polymer insulated cable.

Assessment of Nonshielded Cables Nonshield cables, those cables without an insulation shield, represent a significant problem regarding off-line electrical testing. To allow an electrical test of the insulation, a uniform ground plane is needed, but such a ground plane does not exist in a nonshielded cable. Testing of nonshielded cable from phase to ground may only provide rough data at the random grounding points along the surface of the cable. Testing phase to phase may only provide information concerning the points where the phases touch one another. (Better results could be expected from a test of a triplexed cable than from three separate cables pulled together.) This limitation could cause variable results if water levels vary and make trending and assessment of results difficult.

One way to effect a ground plane would be to full flood the ducts and verify that the water was grounded. Surrounding the cable with water is used in laboratory assessments, but it has not been attempted in a power plant.

Given that off-line electrical testing of nonshielded cables is not practical, other alternatives must be selected, which are the following:

• Full forensic analysis of cables if failure occurs

• Applying lessons learned from operating experience from related cables under similar conditions

• Applying lessons learned from forensic analysis of shielded cables with the same insulations from other plants

• Removal and testing of abandoned nonshielded cable

• Removal and testing of cable removed from service The nonshielded cables in use in the nuclear industry almost always have EPR insulation and are limited to those rated 5 kY. The insulations are the same types as were used in shielded cables.

The differences from the shielded cable are the absence of the insulation shield and a somewhat thicker insulation on the nonshielded cables.

5-17 OAGI0001276_00059

Actionsfor Cables Having Wet Environments A review of installed nonshielded cable data from the NEI 2005 industry survey [17] indicates that 31 of the responding units had some nonshielded medium-voltage circuits. The dominant manufacturers were Kerite and Okonite, with one plant reporting General Cable and another reporting Anaconda. Kerite has used brown EPR throughout the period, while Okonite used black EPR through the mid-1970s and then switched to pink EPR thereafter. Review of the failures of nuclear plant cables revealed only three failures. Only one failure report directly stated that wetting of the cable was involved and also indicated that thermal overload contributed to the degradation.

Although these data do not eliminate wet aging as a concern, they indicate that the lack of a shield on these cables does not lead to more frequent failure than for those cables having a shield.

Removal of abandoned cables that have experienced long service under wet conditions is a valuable input to the understanding the degree of electrically induced wet aging that can occur.

Currently, two sets of Kerite cables that experienced 30 years of service before being abandoned are being evaluated by EPRI. Laboratory testing of these cables will give insights regarding wet degradation of nonshielded cables. The laboratory analysis may also give insights on whether in-service electrical testing is practicable (that is, using water as a ground shield).

For plants having nonshielded cables, the recommended path is as follows:

1. Require full forensics testing of any failure of nonshielded medium-voltage cable with appropriate action taken for other nonshielded cables in similar operating conditions (for example, if failures are wet aging related, replace similar circuits).

2. Maintain awareness of results of research performed on abandoned cables or those removed from service.

3. Maintain awareness of results of failure assessment and mechanism research for related shielded medium-voltage cable. For example, findings on the same manufacturer's material (for example, black, brown, or pink EPR) from a shielded cable may give insights on expected aging of a nonshielded cable.

4. Carefully assess operating experience for nonshielded cables of the same type and material for applicability and any indication of additional concern.

5. When industry insights indicate that nonshielded wetted cable may be entering an end-of-life state, either remove a "worst case" cable from service and perform laboratory testing or schedule replacement of wetted circuits.

5-18 OAGI0001276_00060

6 ACTIONS FOR CABLES HAVING DRY ADVERSE ENVIRONMENTS The effects of adverse dry environment conditions will be different from those caused by cables being energized in wet or submerged conditions because the failure mechanisms are not the same. Accordingly, different assessment methods apply. This section addresses those assessments that can be applied to cables in dry adverse environments.

High-Temperature or High-Dose-Rate Ambient Environments Different dry environment aging effects can occur, depending on the insulation type and jacket type in use. Table 6-1 describes the degradation mechanisms that are expected for common types of medium-voltage cable jackets.

6-1 OAGI0001276_00061

Actions for Cables Having Dry Adverse Environments Table 6-1 Thermal and Radiation Degradation Mechanisms Expected for Medium-Voltage Cable Jacket Materials Material Temperature-Induced Radiation- Condition Evaluation Effect of Degradation Degradation Induced Degradation Neoprene Hardening with Hardening Visual inspection can identify Cracking exposes shield and insulation spontaneous cracking discoloration or cracking. to airborne moisture. Released chlorine and discoloration Hardening can be manually or will corrode the shield and cause limited (turning greenish indenter evaluated with cable de- effects on insulation. Loss of jacket will brown) energized. adversely affect flame propagation.

Hypalon Hardening and Hardening Visual inspection can identify Until extreme hardening occurs, (chlorosulfonated discoloration (turning discoloration. Hardening can be Hypalon will remain intact. However, if a polyethylene greenish brown) manually or indenter evaluated through fault occurs, the cable may

[CSPED with cable de-energized. crack because of motion from high magnetic fields.

Polyvinyl Hardening, possible Production of With cable de-energized, Cracking exposes shield and insulation chloride (PVC) spontaneous cracking hydrogen hardening can be observed to airborne moisture. Released chlorine chloride manually or through indenter. HCI will corrode the shield, causing limited (HCI) production may be indicated by effects on insulation.

white powdering or corrosion of surrounding metal.

Chlorinated Hardening, cracking Hardening With cable de-energized, Extreme hardening may cause failure.

polyethylene (thermoplastic only) hardening can be observed Cracking of thermoplastic versions (CPE) manually or through indenter. would be expected with significant thermal aging.

0 G) 0 0

0 N

-....J I

(J) 6-2 0

0 0

(J)

N

Actions for Cables Having Dry Adverse Environments In the case of thermal damage from the environment rather than from ohmic heating, the jacket will deteriorate before the insulation system. If the damage is from the environment, the deterioration of PVC and neoprene is likely to be of most concern because these jackets tend to crack spontaneously on severe aging. Figure 6-1 shows cracking of a neoprene jacket from an elevated thermal environment. One concern for both neoprene and PVC when highly thermally aged is that they will generate chlorine, which will affect the shield and surrounding metal materials, such as trays, conduits, and possibly piping. Under dry conditions, environmentally induced deterioration of PVC or neoprene might not cause an immediate concern for the aging of the insulation because the insulation will tend to age more slowly. Hypalon will age more slowly than neoprene and early PVCs and have a much lower tendency to crack spontaneously. Early thermoplastic chlorinated polyethylene (CPE) can crack when it ages and if highly stressed (for example, at bends). Modern thermoset CPE will not tend to crack with aging. If Hypalon and modern thermoset CPE are found to be aged, there is a higher likelihood of thermal damage to the insulation; but again, the insulation should age more slowly than the jacket from environmentally induced aging.

Note: Corrosion of tray and shield are likely from chlorine released by the aging of the neoprene. Jacket has turned brown from original black and is very hard. Brown color of the jacket is a strong indication of exposure to elevated temperature.

Figure 6-1 Spontaneous Cracking of Neoprene Jacket 6-3 OAGI0001276_00063

Actions for Cables Having Dry Adverse Environments Depending on the application and severity of the cable jacket cracking, different actions may be warranted. Jackets are important to keep moisture out of cables and to prevent fire propagation.

If medium-voltage cables require environmental qualification for steam accident, cracking in a location with a potential steam environment would be unacceptable without further assessment.

All cables are required to limit flame propagation; therefore, if cracking is severe enough that the jacket is in danger of or has fallen off, the flame retardancy provided by the jacket has been lost and corrective action is needed. For shielded cable, loss of jacket integrity could allow additional grounding points on the shield. Only a few plants have safety-related medium-voltage cables within containment. Accordingly, very few plants take credit for medium-voltage cable jackets acting as beta shields. Severe cracking of jackets on medium-voltage cable within containment could add to sump loadings in the event of an accident.

Severe jacket aging indicates that insulation damage may have occurred as well and that electrical assessment of the cable, as indicated for insulation damage in Table 6-2, should be implemented.

A tan 8 or partial discharge test, as appropriate to the concern, will indicate whether the thermal damage has been severe enough to adversely affect the insulation properties. These tests can be performed only on shielded cables.

Line resonance analysis (LIRA) can be used on shielded and nonshielded triplexed cable to detect the effects oflocalized thermal damage. For shielded cables, it assesses only the insulation system. For triplexed cable, the jackets are included in the assessed material. If LIRA does not produce a signal at the site of the adverse localized thermal environment, the damage is not significant. If LIRA does produce a signal, the strength of the signal is proportional to the severity of the damage and a relative effect could be determined. LIRA is not useful on nonshielded cables that are pulled individually rather than triplexed. LIRA requires a uniform geometry along the length of cable under assessment. It should also be recognized that if the elevated temperature condition exists at the time of testing, LIRA is likely to identify the effects of the elevated temperature and not necessarily identify thermal damage. (Thermal expansion of the cable in the heat affected zone can cause a LIRA signal.) Accordingly, LIRA testing should be performed when the localized heat source is not producing heat.

6-4 OAGI0001276_00064

Actions for Cables Having Dry Adverse Environments Table 6-2 Thermal and Radiation Degradation Mechanisms Expected for Medium-Voltage Cable Insulation Materials Material Temperature-Induced Radiation- Condition Evaluation Effect of Degradation Degradation Induced Degradation XLPE Hardening Hardening Electrical tests required to determine whether Ultimately, insulation could crack leakage current is increasing because of (after a very long time). Life still damage. LIRA testing would identify whether could be long provided the condition thermal damage has occurred. Consider tan 0 is corrected before severe measurement to determine whether insulating degradation occurs.

capability has been adversely affected (see note 1).

Butyl rubber Hardening or softening Softening Electrical tests required to determine whether Softening failures have occurred on leakage current is increasing because of sulfur-cured butyl rubber insulation damage. LIRA testing would identify whether from advanced thermal aging. The thermal damage or compression of softened shield drifted through the softened insulation has occurred. Consider tan 0 insulation toward conductor, leading measurement. If softening has occurred, partial to failure.

discharge may occur if components drift with respect to original position.

EPR Hardening Hardening Electrical tests required to determine whether Extreme thermal aging has caused leakage current is increasing because of failure (embrittlement or thermal damage. LIRA testing would identify whether runaway).

thermal damage had occurred. Consider tan 0 measurement to determine whether insulating capability has been adversely affected (see note 1).

Note:

1: Early polymeric shields may be more age sensitive than modem extruded shields. If they crack, partial discharge would be expected and partial discharge testing may be a useful test method.

0 G) 0 0

0 N

-....J I

(J) 6-5 0

0 0

(J)

(J1

Actions for Cables Having Dry Adverse Environments Table 6-2 describes the degradation mechanisms that are expected for common types of medium-voltage cable insulations. The various insulation types behave differently from one another.

Butyl rubbers that have been cured using sulfur will soften, rather than harden, to the point where the shield can drift toward the conductor,16 causing very high electrical stress and subsequent insulation failure. EPRs and XLPEs harden under elevated temperature and dose conditions. The likelihood is that any of these conditions will eventually cause leakage current to increase.

Accordingly, tan 0 measurement should be a useful assessment method for thermally damaged cable. However, a very localized effect may be difficult to detect with tan o. Damage affecting a significant portion of the cable run will be more easily identified. An exception to the use of tan oas the preferred diagnostic tool is that if the components shift when sulfur cured butyl rubber softens with severe thermal aging, discharging in any gaps that may occur would allow partial discharge testing to be a method of assessment as well. High-intensity partial discharging, even if of low amplitude, will add to the dielectric loss of the insulation and will be detected, but not located, by a tan 0 measurement.

F or cables produced from the late 1970s forward, the thermal aging rate of polymer shields should be approximately the same as the insulation. However, early polymer shields from the early 1970s may age more rapidly than the insulation. Cables with early polymer shield designs may be more age sensitive than the later vintage cables when exposed to elevated temperature. If cracking occurred within the insulation or conductor extruded shield, partial discharge would be expected.

High Conductor Temperature from Ohmic Heating Depending on the severity of the ohmic heating, the insulation may be damaged and the jacket mayor may not be damaged. Should the jacket be found to be discolored and/or cracked and the environmental conditions are not severe enough to have caused the damage, ohmic heating is likely the cause and the insulation is likely to have suffered significant thermal damage because of the temperature being significantly higher at the conductor than at the exterior surface.

As described previously for ambient-induced damage, tan 0 testing is likely to detect the effects of severe thermal degradation, whether caused by unbalanced magnetic circuits in multiconductor-per-phase circuits or by high, continuous currents. Significant softening of butyl rubber may cause gaps to open and allow the use of partial discharge testing as well.

High-Resistance Connections at Terminations or Splices There are two considerations for thermal heating of connections, terminations, and splices. The first consideration is a difference in temperature between two connections with identical loadings. Guidance exists in various EPRI infrared thermography reports [18, 19, 20], such as that provided in Table 6-3. Any temperature difference above reference (the difference between two similar targets under similar loading) is a concern. Small temperature differences can be risk 16 Note: To date, drifting of the shield towards the conductor has only been noted for failure from high ohmic heating situations rather than from environmentally induced conditions.

6-6 OAGI0001276_00066

Actions for Cables Having Dry Adverse Environments managed to allow time to make preparations for repairs, but anomalies caused by high-resistance connections do not necessarily increase linearly, therefore, uncertainty always exists in predicting time to failure. Increased monitoring should be performed to the extent that a rate of degradation can be estimated until the condition is repaired. Table 6-3 provides suggested severity ranges for evaluating indoor electrical power connections.

Table 6-3 EPRI-Suggested Severity Ranges for Indoor Electrical Power Connections [18]

Status Range Advisory 1°F to 15°F (OSC to 8°C) rise above reference Intermediate 16°F to 50°F (9°C to 28°C) rise above reference Serious 51°F to 100°F (29°C to 56°C) rise above reference Critical >100°F (56°C) rise above reference The second consideration for high-resistance connections is the absolute thermal limitations of the materials involved. If the components of a cable, termination, or splice exceed the limits at which their physical and/or electrical properties are compromised and the immediate or long-term ability to function are compromised, the condition should be evaluated and corrected accordingl y.

As described previously, routine thermographic inspections should be performed on all accessible connections, terminations, and splices based on the application and during postmaintenance testing if they have been disturbed.

6-7 OAGI0001276_00067

7 ACTIONS FOR FAILED OR DETERIORATED CABLE Operability Concerns Depending on the severity of the degradation identified, an operability concern mayor may not exist. Severe physical degradation, such as cracked insulation, damaged conductors, extreme hardening or softening of insulation, or a "highly degraded" result from electrical testing indicates an operability concern. However, lesser indications of degradation would constitute a need for further vigilance but not an immediate operability concern. Examples of these types of degradation include a limited stiffening of insulation and jacket or an electrical test result indicating "further study required" insulation. The following subsections provide insights about verifying the condition and determining the course of further actions. In-service failure of a cable requires an extent of condition assessment for cables subject to like service conditions.

Corrective Actions The corrective actions to be taken in response to cable degradation depend on the nature of degradation and whether the degradation is localized or distributed over a significant length of the cable. Actions may be permanent or temporary, based on the nature of the application and licensing basis. Some possible considerations and resolutions, which are not all inclusive, are described next. Plant-specific and application-specific conditions can dictate different resolution paths.

Cable Test Indicates "Further Study Required" Insulation System As described in Section 5, the results of tan 8 tests can indicate that a cable insulation system has aged but is not yet in a highly deteriorated state. An aged condition indicates a need for heightened awareness.

7-1 OAGI0001276_00068

Actions for Failed or Deteriorated Cable Eliminate Obvious Problems Inspect the terminations for accumulated dirt, moisture, and tracking or other surface problems.

Clean and repair as needed. In addition, verify that the terminations of the cable under test were well isolated from adjacent phase terminations and the cabinet/ termination box to eliminate corona at the termination as a cause of the adverse results. If termination issues appear to be the cause of the questionable results, retest and determine whether the "further study required" indication still persists.

Perform Very Low Frequency Withstand Test If a VLF withstand test was not part of testing process, perform a VLF withstand test to confirm that a single, severe degradation site is not the cause of the "aged" indication.

Increase Frequency of Testing Decrease the period between tests to one refueling cycle. Compare test results to determine whether the condition is stable or worsening. If it is worsening significantly (for example, approaching the "action required" state), schedule for replacement at a convenient time.

Prepare Contingency Plan Although a "further study required" cable is likely to function for a significant period, a contingency plan should be prepared in case of failure. The plan should cover the availability of replacement cables and accessories, pulling procedures, pulling tools, and the required qualifications of craft.

Perform Polymer Injection Silicone polymer injection has been shown to improve the breakdown strength of cable insulation for an extended period [21]. If injection is to be used, it is best to perform it when the cable is in the "further study required" but not in the "action required" state. At the point at which tests indicate "action required," electrical degradation may be severe enough that injection will not be able to overcome the degradation. In the case of butyl rubber that may soften with age depending on manufacturing cure issues, polymer injection will not correct the softening issue and would not be recommended.

Begin Replacement Program for Multiple Cables with "Further Study Required" Insulation If multiple cables within a plant's population of cables have indications of "further study required" insulation, the need to begin an orderly replacement program should be considered.

7-2 OAGI0001276_00069

Actions for Failed or Deteriorated Cable Cable Test Indicates "Action Required" Insulation System Eliminate Obvious Problems Inspect the terminations for accumulated dirt, moisture, and tracking or other surface problems.

Clean and repair as needed. Also, verify that the terminations of the cable circuit under test were well isolated from adjacent phase terminations and the cabinet/termination box to eliminate corona at the termination as a cause of the adverse results. If the termination issues appear to be cause of the questionable results, retest and determine whether the "action required" indication persists.

Perform Very Low Frequency High-Potential Tests If a VLF hi -pot was not part of testing process, perform a VLF hi -pot to determine whether the cable's condition is sufficiently stable to establish an interim period of operation to allow orderly staging for replacement. Note: The purpose of a VLF hi-pot is to fail a highly weakened cable. If immediate replacement is being performed, there is no need to perform a VLF hi-pot.

Identify and Replace Degraded Section The degraded section of cable is likely to be the section with the adverse localized environment (for example, wetted section). Accordingly, the section with the adverse localized environment may be cut from the section with the benign environment, and retesting of the segments performed to identify the deteriorated section. The appropriate section(s) would be replaced and spliced to the good section(s). If the deteriorated section is dry and the metallic shield is not corroded, partial discharge testing may be appropriate to identify the location of the degradation.

Note that shielded splice designs should be qualified in accordance with IEEE Std 404 [22].

(This standard does not apply to nonshielded designs.). When splicing dissimilar cable types, the use of separable connectors should be considered to allow isolating of dissimilar cables for ease of future testing. In no case should a splice be pulled into an inaccessible location (that is, duct or conduit). Utility-specific limitations on location of splices should be observed.

Conduct Forensic Testing of "Action Required" Cable Forensic testing of the "action required" cable segment is recommended to gain insight into the nature of the degradation and whether it is related to the adverse environment or another cause.

The forensic information will provide insights into the overall effects of adverse environment aging on the cable system and the potential extent of condition.

7-3 OAGI0001276_00070

Actions for Failed or Deteriorated Cable Use Impervious Cable for Wetted Environments If the cable degradation is related to a wetted environment, and long-term wetting cannot be eliminated, consideration should be given to using an impervious cable design for the replacement cable. Impervious cable designs incorporate a lead sheath or a longitudinally corrugated copper sheath that provides a barrier against water ingress.

Cables Experiencing Localized Thermal Damage Two concerns exist for localized thermal damage. The first is that the temperature of the insulation is so high as to cause the insulation system to fail because of thermal avalanche. In such a case, the local volumetric insulation resistance would decrease causing higher leakage current further elevating the insulation temperature. Eventually, the leakage current and insulation temperature are so high that the insulation breaks down through thermal avalanche.

This is not an aging phenomenon but a direct effect of excessive temperature. The aging concern is that the temperature is not high enough to cause thermal avalanche but is high enough to cause hardening of jackets and insulations (softening of sulfur-cured butyl rubber) over time.

Eventually, cracking of the insulation could occur from manipulation or from motion induced by fault current surge. For sulfur-cured butyl rubber, long-term thermal aging could cause softening that could allow compression of the insulation leading to high electrical stress and failure.

Thermal degradation of environmentally qualified cables located in harsh environment areas can cause the cable to have a shortened qualified life.

Evaluate the Degree of Damage Environmentally induced degradation is generally caused by an adjacent heat source that was not properly controlled (for example, adjacent process pipe with inadequate or missing thermal insulation). The first assessment should be of the jacket to see if complete hardening has occurred or if some elasticity remains. If some elasticity remains, the likelihood of damage to the insulation is low, and the thermal insulation on the hot process component should be improved.

Periodic inspection of the cable is recommended to verify that further deterioration is not worsening significantly.

Evaluation of the severity of the jacket degradation may be performed through indenter modulus assessment [23]. The use of indenter testing allows quantification and trending of the hardening of the jacket to provide insights as to the relative hardness and the degree of continued aging.

The ultimate effect of the thermal degradation on insulation can be evaluated with tan 8 testing.

Partial discharge testing may be appropriate for butyl rubber insulated cables in which softening and compression of the insulation are potential problems, although signal attenuation caused by corroded metallic shields could be a problem.

7-4 OAGI0001276_00071

Actions for Failed or Deteriorated Cable Line resonance assessment (LIRA) can be used for cables located in dry environments to determine whether an adverse localized thermal environment has affected the insulation [24]. If the effect was limited to the jacket on shielded cable, LIRA should identify no significant signal.

If the insulation was affected, LIRA would give a relative indication of the severity of the effect.

LIRA can be used on triplexed cable, but the jacket system would be within the boundary of the test and the effects of aging on the insulation and jacket would not be separable. LIRA is a test method under development. Although results to date show the ability to identify thermally damaged segments, research that indicates that LIRA can identify water-related damage or electrical trees has not been completed.

Correct the Adverse Localized Thermal Environment When an adverse localized thermal environment is identified, the thermal insulation on the source of the heat and radiant energy should be repaired, replaced, or upgraded. If radiant energy is the source of the aging, shielding should be installed to reduce the effect to the point practical.

If this activity does not sufficiently reduce the effects on the cable, consideration should be given to rerouting the cable. If the cable must remain where it is, periodic assessment of the condition of the cable should be implemented to verify that the rate and severity of the cable degradation is known so that corrective action can be taken at the appropriate time.

Replace Thermally Damaged Cable If severe thermal aging of the insulation is identified or suspected and cannot be eliminated by evaluation, removal and replacement of the affected cable section is recommended. If the qualified life of a cable is shortened because of the adverse localized thermal environment, it must be replaced before the end of its qualified life. Replacement of a section by using appropriate splices or replacement of the entire circuit is permissible.

High-Resistance Connections When inspection or infrared assessment of cable connections indicates significant heating of a connection (for example, for infrared thermography: upper "intermediate" through "critical" status in Table 6-3), the effected connection should be repaired or replaced. If replacement cannot be performed immediately, increased monitoring should be performed until replacement occurs. Early replacement is recommended to preclude significant damage to the cable insulation at the connection point. If the cable insulation has been damaged, replacement of the cable or the affected section will be necessary as well.

7-5 OAGI0001276_00072

Actions for Failed or Deteriorated Cable Cables Damaged by High Current Damage to cable from ohmic heating resulting from high currents is likely to affect the entire length of the cable with the worst effect in sections having elevated ambient temperature. The entire circuit generally will require replacement. Rectification of the cause of the high current is necessary whether it is from lack of transpositions in multiconductor-per-phase circuits or undersized conductors.

7-6 OAGI0001276_00073

8 REFERENCES In-Text References

1. Plant Support Engineering: Aging Management Program Development Guidance for AC and DC Low-Voltage Power Cable Systems for Nuclear Power Plants. EPRI, Palo Alto, CA:

2010. 1020804.

2. Equipment Failure Model and Data for Underground Distribution Cables: A PM Basis Application. EPRI, Palo Alto, CA: 2004. 1008560.

3. 10 CFR 50.65, Requirementsfor Monitoring the Effectiveness ofMaintenance at Nuclear Power Plants. US. Nuclear Regulatory Commission, Washington, D.C.

4. NUREG-1801, Generic Aging Lessons Learned Report. US. Nuclear Regulatory Commission, Washington, D.C. September 2005.

5. NRC Generic Letter 2007-01, Inaccessible or Underground Power Cable Failures That Disable Accident Mitigation Systems or Cause Plant Transients. US. Nuclear Regulatory Commission, Washington, D.C. February 7, 2007.

6. NFPA 70: National Electrical Code. National Fire Protection Association, Quincy, MA. 2008.

7. UL 1598, The Standardfor Safety of Luminaires. Underwriters Laboratories, Inc.,

Camas, W A. 2008.

8. 10 CFR 54, Requirementsfor Renewal of Operating Licensesfor Nuclear Power Plants, US. Nuclear Regulatory Commission, Washington, D.C.

9. AP-913, Equipment Reliability Process Description. Institute for Nuclear Power Operations, Atlanta, GA. November 2001.

10. Information Notice No. 86-49: Age/Environment Induced Electrical Cable Failures, US. Nuclear Regulatory Commission, Washington, D.C., June 16, 1986.

11. Cable Polymer Aging and Condition Monitoring Research at Sandia National Laboratory Under the Nuclear Energy Plant Optimization (NEPO) Program. EPRI, Palo Alto, CA:

2005. 1011873.

12. IEEE Std 400.2-2004, IEEE Guide for Field Testing of Shielded Power Cable Systems Using Very Low Frequency (VLF), Institute of Electrical and Electronics Engineers, Inc., New York, NY.

8-1 OAGI0001276_00074

References

13. S. Boggs, C. Xu, L. Zhou, and Y. Zhou. High Frequency Properties of Shielded Power Cable, Part 1: Overview ofMechanisms. Electrical Insulation Research Center, University of Connecticut (undated). http://www.ims.uconn.edu/images/eirc/HF _cab_Loss l.pdf accessed May 15,2010.

14. IEEE Std 400-2001, IEEE Guide for Field Testing and Evaluation of the Insulation of Shielded Power Cable Systems, Institute of Electrical and Electronics Engineers, Inc., New York, NY.

15. Plant Support Engineering: Medium-Voltage Cable Aging Management Guide. EPRI, Palo Alto, CA: 2008. 1016689.

16. Effect of DC Testing on Extruded Crosslinked Polyethylene Insulated Cables--Phase II.

EPRI, Palo Alto, CA: 1995. TR-I01245-V2.

17. NEI Medium-Voltage Underground (MVU) Cable Industry Survey, Nuclear Energy Institute, Washington, D.C. January 2005.

18. Infrared Thermography Guide (Revision 3). EPRI, Palo Alto, CA: 2002. 1006534.

19. Infrared Thermography (IRT) Anomalies Manual (Revision 2001). EPRI, Palo Alto, CA:

2001. 1004628.

20. Predictive Maintenance Primer: Revision to NP-7205. EPRI, Palo Alto, CA: 2003. 1007350.

21. Extending the Service Life ofEthylene Propylene Rubber Insulated Cables. EPRI, Palo Alto, CA: 2002. 1006882.

22. IEEE Std 404-2006, IEEE Standard for Extruded and Laminated Dielectric Shielded Cable Joints Rated 2500 V to 500,000 V, Institute of Electrical and Electronics Engineers, Inc.,

New York, NY.

23. Evaluation of Cable Aging Through Indenter Testing. EPRI, Palo Alto, CA: 1996.

TR-I04075.

24. Line Impedance Resonance Analysis for Detection of Cable Damage and Degradation.

EPRI, Palo Alto, CA: 2007. 1015209.

Additional Resources

1. IEEE Std 400.1-2007, Guide for Field Testing of Laminated Dielectric, Shielded Power Cable Systems Rated 5 kV and Above with High Direct Current Voltage, Institute of Electrical and Electronics Engineers, Inc., New York, NY.

2. IEEE Std 400.3-2006, IEEE Guide for Partial Discharge Testing of Shielded Power Cable Systems in a Field Environment, Institute of Electrical and Electronics Engineers, Inc., New York, NY.

8-2 OAGI0001276_00075

A ASSESSMENT OF PERCENT STANDARD DEVIATION OF TAN () MEASUREMENTS Section 5 describes the use of percent standard deviation as a means of evaluating tan 8 results.

The following is a description of the mathematical determination of the percent standard deviation and an example of its calculation, including the equation used in the determination.

This equation may be used in electronic spreadsheets such as Microsoft Excel.

%STDEV = 100

• L(X-XY (n-I)

Where:

%STDEV is percent standard deviation.

X is each individual tan 8 measurement.

X is the mean (arithmetical average) of the tan 8 measurements.

n is the number of measurements.

A minimum of six measurements is recommended.

The percent standard deviation of the tan 8 measurements provides a way to assess small but significant changes in tan 8 at a particular voltage level. The following is an example from the C phase 8 kV (1 V o) result from Figure 5-4. The following 14 measurements were taken over the course of 2 minutes:

8.2, 8.3, 8.3, 8.4, 8.4, 8.5, 8.5, 8.6, 8.6, 8.7, 8.7, 8.7, 8.7, and 8.8 A casual inspection might indicate no specific problem; however, under constant voltage, the tan 8 measurements are increasing rather that staying stable. The mean of these results is 8.5. The rounded percent standard deviation is 0.02, placing the cable in a "further study required" state at 1 Vo test level. Evaluating the percent standard deviation provided a clearer indication of a problem.

A-I OAGI0001276_00076

Assessment ofPercent Standard Deviation of Tan (j Measurements For the 1.5 Vo test level (11.9 kV), the following 15 measurements were also taken:

15.0,16.0,16.8,17.6,18.2,18.8,19.3,19.8,20.2,20.6, 21.0, 21.4, 21.7, 22.0, and 22.3 Scanning these data more easily indicates an ever-increasing tan 8 measurement. The mean of these results is 22.3. The percent standard deviation is 0.23, which is nearly six times the "action required" level and is a strong indication of a significant degradation.

A-2 OAGI0001276_00077

The Electric Power Research Institute Inc., (EPRI, www.epri.com) conducts research and development relating to the generation, delivery and use of electricity for the benefit of the public. An independent, nonprofit organization, EPRI brings together its scientists and engineers as well as experts from academia and industry to help address challenges in electricity, including reliability, efficiency, health, safety and the environment. EPRI also provides technology, policy and economic analyses to drive long-range research and development planning, and supports research in emerging technologies. EPRI's members represent more than 90 percent of the electricity generated and delivered in the United States, and international participation extends to 40 countries. EPRI's principal offices and laboratories are located in Palo Alto, Calif.; Charlotte, N.C.;

Knoxville, Tenn.; and Lenox, Mass.

Together... Shaping the Future of Electricity Programs:

Nuclear Power Plant Support Engineering

© 2010 Electric Power Research Institute (EPRI), Inc. All rights reserved. Electric Power Research Institute, EPRI, and TOGETHER ... SHAPING THE FUTURE OF ELECTRICITY are registered service marks of the Electric Power Research Institute, Inc.

1020805 Electric Power Research Institute 3420 Hillview Avenue, Palo Alto, California 94304-1338

• PO Box 10412, Palo Alto, California 94303-0813 USA 800.313.3774' 650.855.2121

• askepri@epri.com

• www.epri.com OAGI0001276_00078