NUREG-2218, an International Phenomena Identification and Ranking Table (Pirt) Expert Elicitation Exercise for High Energy Arcing Faults (Heafs).

NUREG-2218, an International Phenomena Identification and Ranking Table (Pirt) Expert Elicitation Exercise for High Energy Arcing Faults (Heafs).

ML18032A318
Person / Time
Issue date:	01/31/2018
From:	Kenneth Hamburger Office of Nuclear Regulatory Research
To:
Meyd, Donald
References
NUREG-2218
Download: ML18032A318 (173)

Text

NUREG-2218 An International Phenomena Identification and Ranking Table (PIRT) Expert Elicitation Exercise for High Energy Arcing Faults (HEAFs)

Office of Nuclear Regulatory Research

AVAILABILITY OF REFERENCE MATERIALS IN NRC PUBLICATIONS NRC Reference Material Non-N RC Reference Material As of November 1999, you may electronically access Documents available from public and special technical NUREG-series publications and other NRC records at libraries include all open literature items, such as books, the NRC's Public Electronic Reading Room at journal articles, transactions, Federal Register notices, http://www.nrc. gov/reading-rm.html. Publicly released Federal and State legislation, and congressional reports.

records include, to name a few, NUREG-series Such documents as theses, dissertations, foreign reports publications; Federal Register notices; applicant, and translations, and non-NRC conference proceedings licensee, and vendor documents and correspondence; may be purchased from their sponsoring organization.

NRC correspondence and internal memoranda; bulletins Copies of industry codes and standards used in a and information notices; inspection and investigative substantive manner in the NRC regulatory process are reports; licensee event reports; and Commission papers maintained at-and their attachments.

The NRC Technical Library NRC publications in the NUREG series, NRC Two White Flint North regulations, and Title 10, "Energy," in the Code of 11545 Rockville Pike Federal Regulations may also be purchased from one Rockville, MD 20852-2738 of these two sources.

These standards are available in the library for reference

1. The Superintendent of Documents use by the public. Codes and standards are usually U.S. Government Publishing Office copyrighted and may be purchased from the originating Mail Stop SSOP organization or, if they are American National Standards, from-Washington, DC 20402-0001 American National Standards Institute Internet: http://bookstore.gpo.gov 11 West 42nd Street Telephone: 1-866-512-1800 New York, NY 10036-8002 Fax: (202) 512-2104 http://www.ansi.org (212) 642-4900

2. The National Technical Information Service 5301 Shawnee Road Alexandria, VA 22161-0002 Legally binding regulatory requirements are stated only in la\J\IS; NRC regulations; licenses, including technical http://www.ntis.gov specifications; or orders, not in NUREG-series 1-800-553-6847 or, locally, (703) 605-6000 publications. The views expressed in contractor prepared publications in this series are not necessarily A single copy of each NRC draft report for comment is those of the NRC.

available free, to the extent of supply, upon written request as follows: The NUREG series comprises (1) technical and U.S. Nuclear Regulatory Commission administrative reports and books prepared by the staff (NUREG-XXXX) or agency contractors Office of Administration (NUREG/CR-XXXX), (2) proceedings of conferences Multimedia, Graphics, Storage, and (NUREG/CP-XXXX), (3) reports resulting from Distribution Branch international agreements (NUREG/IA-XXXX), (4)

Washington, DC 20555-0001 brochures (NUREG/BR-XXXX), and (5) compilations of E-mail: distribution.resource@nrc.gov legal decisions and orders of the Commission and Facsimile: (301) 415-2289 Atomic and Safety Licensing Boards and of Directors' decisions under Section 2.206 of NRC's regulations Some publications in the NUREG series that are (NUREG-0750).

posted at the NRC's Web site address http://www.nrc.gov/reading-rm/doc-collections/nuregs DISCLAIMER: This report was prepared as an account of are updated periodically and may differ from the last work sponsored by an agency of the U.S. Government.

printed version. Although references to material Neither the U.S. Government nor any agency thereof, nor found on a Web site bear the date the material was any employee, makes any warranty, expressed or accessed, the material available on the date cited implied, or assumes any legal liability or responsibility for may subsequently be removed from the site. any third party's use, or the results of such use, of any information, apparatus, product, or process disclosed in this publication, or represents that its use by such third party v.ould not infringe privately owned rights.

SR-CR 08/2016

NUREG-2218 An International Phenomena Identification and Ranking Table (PIRT) Expert Elicitation Exercise for High Energy Arcing Faults (HEAFs)

Manuscript Completed: May 2017 Date Published: January 2018 Prepared by:

K. Hamburger Kenneth Hamburger, NRC Project Manager Office of Nuclear Regulatory Research

ABSTRACT This report documents the results of a phenomena identification and ranking table (PIRT) exercise performed for nuclear power plant (NPP) high energy arc fault (HEAF) analysis applications conducted on behalf of the U.S. Nuclear Regulatory Commissions (NRCs) Office of Nuclear Regulatory Research.

The PIRT exercise was performed via a facilitated expert elicitation process. In this case, the expert panel comprised six international HEAF experts. The panel was facilitated by NRC staff.

The objective of a PIRT exercise is to identify key phenomena associated with the intended application and to then rank the current state of knowledge relative to each identified phenomenon. The expert panel was presented with a series of specific HEAF scenarios, each of which is based on the types of scenarios typically considered in NPP applications. Each scenario includes a figure of merit (i.e., a specific goal to be achieved in analyzing the scenario using HEAF analysis or modeling tools). Given each scenario, the panel identified phenomena that are of potential interest to an assessment based on the figure of merit. The identified phenomena are then ranked relative to their importance in predicting the figure of merit. Each phenomenon is then further ranked for the existing state of knowledge and the adequacy of existing modeling tools to predict that phenomenon.

The PIRT panel covered three HEAF scenarios and identified a number of areas potentially in need of further analysis and model development. This report discusses the results in detail.

iii

FOREWORD Analysis of the most recent international nuclear power plant (NPP) fire event data identifies an increase in the number of high energy arc fault (HEAF) events. The international HEAF operational experience illustrates that significant damage can occur during a HEAF event. In the interest of safe nuclear operations, it is imperative that engineers, operators, and probabilistic risk assessment (PRA) practitioners understand the risk potential of a HEAF and protect safety systems from its effects.

This report documents the work of an international group of nuclear fire safety experts from five countries participating in a structured, facilitated expert elicitation. The experts evaluated three different HEAF scenarios encountered in NPPs. By following the NRCs well-established phenomena identification and ranking table (PIRT) process, expert insights are gained into important areas of HEAF phenomena. These include answers to questions such as:

• How well do we think we understand the phenomena?

• What are their importance and potential for NPP damage?

• Where are the gaps in our state of knowledge?

The results of the PIRT identify and rank important:

• Level 1 phenomena such as target characterization, arc characterization, and arc mitigation.

• Level 2 phenomena such as internal and external ensuing fires, pressure effects, and electrical configuration effects.

• Level 3 phenomena including internal and external cabinet enclosure effects, effects of fire detection and suppression, and room configuration.

These rankings can then be used as a technical compass to help guide and inform future testing and research endeavors.

This report builds upon previous Organization for Economic Cooperation and Development/

Nuclear Energy Agency and NRC HEAF work. This research continues to advance our understanding of this complex phenomenon and its impact on safety. I hope this work will ultimately be used to make a positive contribution to NPP fire safety.

Mark Henry Salley, P.E.

Chief, Fire and External Hazards Analysis Branch Office of Nuclear Regulatory Research United States Nuclear Regulatory Commission June 2, 2017 Rockville, Maryland v

TABLE OF CONTENTS ABSTRACT................................................................................................................................iii FOREWORD...............................................................................................................................v LIST OF TABLES........................................................................................................................ix EXECUTIVE

SUMMARY

............................................................................................................xi ACKNOWLEDGMENTS ..........................................................................................................xiii ABBREVIATIONS AND ACRONYMS.......................................................................................xv 1 INTRODUCTION...................................................................................................................1-1 1.1 Background..........................................................................................................................1-1 1.2 Objectives ............................................................................................................................1-3 1.3 Scope...................................................................................................................................1-3 1.4 Report Organization.............................................................................................................1-4 2 OVERVIEW OF THE PIRT PROCESS APPLIED .................................................................2-1 2.1 Background..........................................................................................................................2-1 2.2 Selection of Panelists ..........................................................................................................2-1 2.3 Panelist Preparation ............................................................................................................2-1 2.3.1 Knowledge Base .........................................................................................................2-2 2.3.2 Orientation Materials ...................................................................................................2-2 2.4 The PIRT Process Applied ..................................................................................................2-2 2.4.1 Scenario Development...............................................................................................2-2 2.4.2 Steps in the PIRT Process ........................................................................................ 2-2 2.4.3 Phenomena Ranking Definitions............................................................................... 2-4 2.4.4 Parameter Ranking Definitions ................................................................................. 2-6 2.4.5 Country-Specific Concerns and Input ....................................................................... 2-6 2.4.6 Panelist Feedback .................................................................................................... 2-7 2.5 Summary Scenario Descriptions .........................................................................................2-8 2.5.1 HEAF Scenario 1 ...................................................................................................... 2-8 2.5.2 HEAF Scenario 2 ...................................................................................................... 2-9 2.5.3 HEAF Scenario 3 .................................................................................................... 2-10 3

SUMMARY

OF LEVEL 1 PHENOMENA ..............................................................................3-1 3.1 Ranking Methodology ..........................................................................................................3-1 3.2 Discussion of Level 1 Phenomena ......................................................................................3-3 3.2.1 Target Characterization ..............................................................................................3-3 3.2.2 Arc Characterization ...................................................................................................3-4 vii

3.2.3 Arc Mitigation ..............................................................................................................3-5 3.2.4 Cabinet Lineup Effects ................................................................................................3-5 3.3 Additional Discussions .........................................................................................................3-5 3.3.1 External Duct Housing Configuration ..........................................................................3-6 3.3.2 Internal Ensuing Fire ...................................................................................................3-6 3.3.3 Pressure Effects..........................................................................................................3-6 4 CONCLUSIONS AND RECOMMENDATIONS.......................................................................4-1 4.1 Practical Considerations for PRA Practitioners ....................................................................4-1 4.2 Characterizing Target Damage.............................................................................................4-1 4.3 Arc Mitigation ........................................................................................................................4-1 4.4 Internal Ensuing Fire.............................................................................................................4-2 4.5 Arc Characterization .............................................................................................................4-3 4.6 Pressure Effects ...................................................................................................................4-3 5 REFERENCES.......................................................................................................................5-1 5.1 References Found in this Report ..........................................................................................5-1 5.2 Additional Documents Included in the Knowledge Base ......................................................5-3 APPENDIX A PIRT HEAF SCENARIO 1............................................................................... ..A-1 APPENDIX B PIRT HEAF SCENARIO 2.................................................................................B-1 APPENDIX C PIRT HEAF SCENARIO 3....................................................................................C-1 APPENDIX D PANIELIST RÉSUMÉS.........................................................................................D-1 APPENDIX E TECHNICAL EXPERT AND ADVISOR RÉSUMÉS............................................E-1 APPENDIX F INTRODUCTORY MATERIALS PRESENTED AT THE FIRST PANEL MEETING...............................................................................................................F-1 viii

LIST OF TABLES Table 2-1 Phenomena Importance Ranking Definitions...................................................................2-4 Table 2-2 Phenomena State of Knowledge Model Adequacy Ranking Definitions.........................2-5 Table 2-3 Phenomena State of Knowledge Data Adequacy Ranking Definitions...........................2-5 Table 2-4 Phenomena State of Knowledge Availability of New Data Ranking Definitions..............2-6 Table 2-5 Parameter Importance Ranking Definitions......................................................................2-6 Table 2-6 Parameter State of Knowledge Ranking Definitions........................................................2-6 Table 3-1 Numerical Equivalents for Ranking Values......................................................................3-1 Table 3-2 Importance Rankings for Pressure Effects, Scenario 1...................................................3-1 Table 3-3 Numerical Equivalent of Table 3-2...................................................................................3-1 Table 3-4 State of Knowledge for Pressure Effects, Scenario 1......................................................3-2 Table 3-5 Numerical Equivalent of Table 3-4...................................................................................3-2 Table 3-6 Summary of Phenomena Rankings for All Scenarios......................................................3-3 ix

EXECUTIVE

=

Background===

This report documents the results of an international phenomena identification and ranking table (PIRT) expert elicitation exercise performed for high energy arcing fault (HEAF) hazards. This PIRT exercise was conducted on behalf of the U.S. Nuclear Regulatory Commissions (NRCs)

Office of Nuclear Regulatory Research (RES) and facilitated by RES staff.

Objectives The objective of this PIRT exercise is to develop an ordered list of phenomena involved in a HEAF event. This list will be ordered by priority; the more important a phenomena is judged to be and the poorer its state of knowledge is judged to be, the higher its priority. This information can be used in the development of a roadmap for future research and allows for an informed focusing of resources for research and regulatory entities.

Approach The expert panel was presented with a series of specific HEAF scenarios, each based on the types of scenarios typically encountered in nuclear power plant (NPP) applications. For each scenario, a specific figure of merit was defined; that is, a specific goal to be achieved in analyzing the scenario. The panel identifies all those related phenomena that are of potential interest to an assessment of the scenario via probabilistic risk assessment (PRA) tools and methods. The phenomena are then ranked relative to their importance in predicting the figure of merit. Each phenomenon is then further ranked for the existing state of knowledge with respect to the ability of existing tools and methods to predict that phenomena, the underlying base of knowledge associated with the phenomena, and the potential for developing new data to support improvements to the existing tools.

The PIRT panel covered three distinct HEAF scenarios. The first was a HEAF occurring in an electrical enclosure with a cable tray passing over the enclosure. The second was a HEAF occurring in a bus duct passing over an electrical enclosure. The third was a HEAF occurring in an electrical enclosure situated in a bank of similar enclosures.

Results As a result of the process, level one phenomena were identified. The level one phenomena are those that were ranked with high importance and low state of knowledge. These would nominally represent potential research priorities. The level one phenomena identified by the panel included the following:

• Electrical arc characterization: thermal and magnetic effects of the arc, arc ejecta (smoke, ionized gas, conductive particulate), arc location, and migration.

• Pressure effects: mechanical shock, projectile impact, and degradation of the compartment pressure boundary.

• Arc mitigation: the use of HEAF-resistant equipment, thermal insulation, or HEAF shields to minimize damage incurred as a result of a HEAF.

xi

• Target characterization: establishing the sensitivity of target equipment to various failure mechanisms, and associated damage criteria.

• Internal ensuing fire: the likelihood, impact, and phenomenology of an enclosure fire ignited by a HEAF event.

xii

ACKNOWLEDGMENTS The author wishes to thank the members of the expert panelDr. Hajime Kabashima (S/NRA/R),

Mr. Sangkyu Lee (KINS), Mr. Nicholas Melly (U.S. NRC), Dr. Marina Rwekamp (GRS),

Dr. Koji Shirai (CRIEPI), and Dr. Sylvain Suard (IRSN)for their dedication to the process and for their perseverance. The panelists technical advisorsDr. Mikimasa Iwata, Mr. Hee Jin Ko, Mr. Cheoung Joon Lee, and Mr. Seonghyeon Yiprovided invaluable insight, and their contributions are greatly appreciated.

The author also wishes to thank the technical area experts who supported the panelistsMr. Scott Bareham, Dr. Anthony Putorti, Mr. Tomasz Stefanski, and Mr. Stephen Turner.

Lastly, the author wishes to thank the NRC staff members who supported the development and execution of the PIRT exerciseMr. Theron Brown, Ms. Tammie Rivera, Mr. Mark Henry Salley, Mr. David Stroup, Mr. Gabriel Taylor, and Mr. Joseph Zabel.

xiii

ABBREVIATIONS AND ACRONYMS CNSC Canadian Nuclear Safety Commission (Canada).

CRIEPI Central Research Institute of Electric Power Industry (Japan).

CSN Consejo de Seguridad Nuclear (Spain).

DID Defense in Depth.

ERFBS Electrical Raceway Fire Barrier System(s).

EMI Electromagnetic Interference.

EPRI Electric Power Research Institute.

FAQ Frequently Asked Questions (Reactor Oversight Program).

GDC General Design Criteria (10 CFR Part 50 Appendix A).

GRS Gesellschaft für Anlagen- und Reaktorsicherheit (Germany).

HEAF High Energy Arcing Fault.

HRR Heat Release Rate.

IAGE Integrity and Aging (Task Group).

IRSN Institut de Radioprotection et de Sûreté Nucléaire (France).

KINS Korea Institute of Nuclear Safety (S Korea).

LER Licensee Event Report.

NEA Nuclear Energy Agency.

NPP Nuclear Power Plant.

OECD Organisation for Economic Co-operation and Development.

PIRT Phenomena Identification and Ranking Table.

PRA Probabilistic Risk Analysis.

RES Office of Nuclear Regulatory Research.

S/NRA/R Regulatory Standard and Research Development, Secretariat of Nuclear Regulation Authority (Japan).

SSC(s) Systems, Structures, and Components.

STUK Steilyturvakeskus (Finland).

ZOI Zone of Influence xv

NUREG-2218 An International Phenomena Identification and Ranking Table (PIRT) Expert Elicitation Exercise for High Energy Arcing Faults (HEAFs)

Office of Nuclear Regulatory Research

AVAILABILITY OF REFERENCE MATERIALS IN NRC PUBLICATIONS NRC Reference Material Non-N RC Reference Material As of November 1999, you may electronically access Documents available from public and special technical NUREG-series publications and other NRC records at libraries include all open literature items, such as books, the NRC's Public Electronic Reading Room at journal articles, transactions, Federal Register notices, http://www.nrc. gov/reading-rm.html. Publicly released Federal and State legislation, and congressional reports.

records include, to name a few, NUREG-series Such documents as theses, dissertations, foreign reports publications; Federal Register notices; applicant, and translations, and non-NRC conference proceedings licensee, and vendor documents and correspondence; may be purchased from their sponsoring organization.

NRC correspondence and internal memoranda; bulletins Copies of industry codes and standards used in a and information notices; inspection and investigative substantive manner in the NRC regulatory process are reports; licensee event reports; and Commission papers maintained at-and their attachments.

The NRC Technical Library NRC publications in the NUREG series, NRC Two White Flint North regulations, and Title 10, "Energy," in the Code of 11545 Rockville Pike Federal Regulations may also be purchased from one Rockville, MD 20852-2738 of these two sources.

These standards are available in the library for reference

1. The Superintendent of Documents use by the public. Codes and standards are usually U.S. Government Publishing Office copyrighted and may be purchased from the originating Mail Stop SSOP organization or, if they are American National Standards, from-Washington, DC 20402-0001 American National Standards Institute Internet: http://bookstore.gpo.gov 11 West 42nd Street Telephone: 1-866-512-1800 New York, NY 10036-8002 Fax: (202) 512-2104 http://www.ansi.org (212) 642-4900

2. The National Technical Information Service 5301 Shawnee Road Alexandria, VA 22161-0002 Legally binding regulatory requirements are stated only in la\J\IS; NRC regulations; licenses, including technical http://www.ntis.gov specifications; or orders, not in NUREG-series 1-800-553-6847 or, locally, (703) 605-6000 publications. The views expressed in contractor prepared publications in this series are not necessarily A single copy of each NRC draft report for comment is those of the NRC.

available free, to the extent of supply, upon written request as follows: The NUREG series comprises (1) technical and U.S. Nuclear Regulatory Commission administrative reports and books prepared by the staff (NUREG-XXXX) or agency contractors Office of Administration (NUREG/CR-XXXX), (2) proceedings of conferences Multimedia, Graphics, Storage, and (NUREG/CP-XXXX), (3) reports resulting from Distribution Branch international agreements (NUREG/IA-XXXX), (4)

Washington, DC 20555-0001 brochures (NUREG/BR-XXXX), and (5) compilations of E-mail: distribution.resource@nrc.gov legal decisions and orders of the Commission and Facsimile: (301) 415-2289 Atomic and Safety Licensing Boards and of Directors' decisions under Section 2.206 of NRC's regulations Some publications in the NUREG series that are (NUREG-0750).

posted at the NRC's Web site address http://www.nrc.gov/reading-rm/doc-collections/nuregs DISCLAIMER: This report was prepared as an account of are updated periodically and may differ from the last work sponsored by an agency of the U.S. Government.

printed version. Although references to material Neither the U.S. Government nor any agency thereof, nor found on a Web site bear the date the material was any employee, makes any warranty, expressed or accessed, the material available on the date cited implied, or assumes any legal liability or responsibility for may subsequently be removed from the site. any third party's use, or the results of such use, of any information, apparatus, product, or process disclosed in this publication, or represents that its use by such third party v.ould not infringe privately owned rights.

SR-CR 08/2016

NUREG-2218 An International Phenomena Identification and Ranking Table (PIRT) Expert Elicitation Exercise for High Energy Arcing Faults (HEAFs)

Manuscript Completed: May 2017 Date Published: January 2018 Prepared by:

K. Hamburger Kenneth Hamburger, NRC Project Manager Office of Nuclear Regulatory Research

ABSTRACT This report documents the results of a phenomena identification and ranking table (PIRT) exercise performed for nuclear power plant (NPP) high energy arc fault (HEAF) analysis applications conducted on behalf of the U.S. Nuclear Regulatory Commissions (NRCs) Office of Nuclear Regulatory Research.

The PIRT exercise was performed via a facilitated expert elicitation process. In this case, the expert panel comprised six international HEAF experts. The panel was facilitated by NRC staff.

The objective of a PIRT exercise is to identify key phenomena associated with the intended application and to then rank the current state of knowledge relative to each identified phenomenon. The expert panel was presented with a series of specific HEAF scenarios, each of which is based on the types of scenarios typically considered in NPP applications. Each scenario includes a figure of merit (i.e., a specific goal to be achieved in analyzing the scenario using HEAF analysis or modeling tools). Given each scenario, the panel identified phenomena that are of potential interest to an assessment based on the figure of merit. The identified phenomena are then ranked relative to their importance in predicting the figure of merit. Each phenomenon is then further ranked for the existing state of knowledge and the adequacy of existing modeling tools to predict that phenomenon.

The PIRT panel covered three HEAF scenarios and identified a number of areas potentially in need of further analysis and model development. This report discusses the results in detail.

iii

FOREWORD Analysis of the most recent international nuclear power plant (NPP) fire event data identifies an increase in the number of high energy arc fault (HEAF) events. The international HEAF operational experience illustrates that significant damage can occur during a HEAF event. In the interest of safe nuclear operations, it is imperative that engineers, operators, and probabilistic risk assessment (PRA) practitioners understand the risk potential of a HEAF and protect safety systems from its effects.

This report documents the work of an international group of nuclear fire safety experts from five countries participating in a structured, facilitated expert elicitation. The experts evaluated three different HEAF scenarios encountered in NPPs. By following the NRCs well-established phenomena identification and ranking table (PIRT) process, expert insights are gained into important areas of HEAF phenomena. These include answers to questions such as:

• How well do we think we understand the phenomena?

• What are their importance and potential for NPP damage?

• Where are the gaps in our state of knowledge?

The results of the PIRT identify and rank important:

• Level 1 phenomena such as target characterization, arc characterization, and arc mitigation.

• Level 2 phenomena such as internal and external ensuing fires, pressure effects, and electrical configuration effects.

• Level 3 phenomena including internal and external cabinet enclosure effects, effects of fire detection and suppression, and room configuration.

These rankings can then be used as a technical compass to help guide and inform future testing and research endeavors.

This report builds upon previous Organization for Economic Cooperation and Development/

Nuclear Energy Agency and NRC HEAF work. This research continues to advance our understanding of this complex phenomenon and its impact on safety. I hope this work will ultimately be used to make a positive contribution to NPP fire safety.

Mark Henry Salley, P.E.

Chief, Fire and External Hazards Analysis Branch Office of Nuclear Regulatory Research United States Nuclear Regulatory Commission June 2, 2017 Rockville, Maryland v

TABLE OF CONTENTS ABSTRACT................................................................................................................................iii FOREWORD...............................................................................................................................v LIST OF TABLES........................................................................................................................ix EXECUTIVE

SUMMARY

............................................................................................................xi ACKNOWLEDGMENTS ..........................................................................................................xiii ABBREVIATIONS AND ACRONYMS.......................................................................................xv 1 INTRODUCTION...................................................................................................................1-1 1.1 Background..........................................................................................................................1-1 1.2 Objectives ............................................................................................................................1-3 1.3 Scope...................................................................................................................................1-3 1.4 Report Organization.............................................................................................................1-4 2 OVERVIEW OF THE PIRT PROCESS APPLIED .................................................................2-1 2.1 Background..........................................................................................................................2-1 2.2 Selection of Panelists ..........................................................................................................2-1 2.3 Panelist Preparation ............................................................................................................2-1 2.3.1 Knowledge Base .........................................................................................................2-2 2.3.2 Orientation Materials ...................................................................................................2-2 2.4 The PIRT Process Applied ..................................................................................................2-2 2.4.1 Scenario Development...............................................................................................2-2 2.4.2 Steps in the PIRT Process ........................................................................................ 2-2 2.4.3 Phenomena Ranking Definitions............................................................................... 2-4 2.4.4 Parameter Ranking Definitions ................................................................................. 2-6 2.4.5 Country-Specific Concerns and Input ....................................................................... 2-6 2.4.6 Panelist Feedback .................................................................................................... 2-7 2.5 Summary Scenario Descriptions .........................................................................................2-8 2.5.1 HEAF Scenario 1 ...................................................................................................... 2-8 2.5.2 HEAF Scenario 2 ...................................................................................................... 2-9 2.5.3 HEAF Scenario 3 .................................................................................................... 2-10 3

SUMMARY

OF LEVEL 1 PHENOMENA ..............................................................................3-1 3.1 Ranking Methodology ..........................................................................................................3-1 3.2 Discussion of Level 1 Phenomena ......................................................................................3-3 3.2.1 Target Characterization ..............................................................................................3-3 3.2.2 Arc Characterization ...................................................................................................3-4 vii

3.2.3 Arc Mitigation ..............................................................................................................3-5 3.2.4 Cabinet Lineup Effects ................................................................................................3-5 3.3 Additional Discussions .........................................................................................................3-5 3.3.1 External Duct Housing Configuration ..........................................................................3-6 3.3.2 Internal Ensuing Fire ...................................................................................................3-6 3.3.3 Pressure Effects..........................................................................................................3-6 4 CONCLUSIONS AND RECOMMENDATIONS.......................................................................4-1 4.1 Practical Considerations for PRA Practitioners ....................................................................4-1 4.2 Characterizing Target Damage.............................................................................................4-1 4.3 Arc Mitigation ........................................................................................................................4-1 4.4 Internal Ensuing Fire.............................................................................................................4-2 4.5 Arc Characterization .............................................................................................................4-3 4.6 Pressure Effects ...................................................................................................................4-3 5 REFERENCES.......................................................................................................................5-1 5.1 References Found in this Report ..........................................................................................5-1 5.2 Additional Documents Included in the Knowledge Base ......................................................5-3 APPENDIX A PIRT HEAF SCENARIO 1............................................................................... ..A-1 APPENDIX B PIRT HEAF SCENARIO 2.................................................................................B-1 APPENDIX C PIRT HEAF SCENARIO 3....................................................................................C-1 APPENDIX D PANIELIST RÉSUMÉS.........................................................................................D-1 APPENDIX E TECHNICAL EXPERT AND ADVISOR RÉSUMÉS............................................E-1 APPENDIX F INTRODUCTORY MATERIALS PRESENTED AT THE FIRST PANEL MEETING...............................................................................................................F-1 viii

LIST OF TABLES Table 2-1 Phenomena Importance Ranking Definitions...................................................................2-4 Table 2-2 Phenomena State of Knowledge Model Adequacy Ranking Definitions.........................2-5 Table 2-3 Phenomena State of Knowledge Data Adequacy Ranking Definitions...........................2-5 Table 2-4 Phenomena State of Knowledge Availability of New Data Ranking Definitions..............2-6 Table 2-5 Parameter Importance Ranking Definitions......................................................................2-6 Table 2-6 Parameter State of Knowledge Ranking Definitions........................................................2-6 Table 3-1 Numerical Equivalents for Ranking Values......................................................................3-1 Table 3-2 Importance Rankings for Pressure Effects, Scenario 1...................................................3-1 Table 3-3 Numerical Equivalent of Table 3-2...................................................................................3-1 Table 3-4 State of Knowledge for Pressure Effects, Scenario 1......................................................3-2 Table 3-5 Numerical Equivalent of Table 3-4...................................................................................3-2 Table 3-6 Summary of Phenomena Rankings for All Scenarios......................................................3-3 ix

EXECUTIVE

=

Background===

This report documents the results of an international phenomena identification and ranking table (PIRT) expert elicitation exercise performed for high energy arcing fault (HEAF) hazards. This PIRT exercise was conducted on behalf of the U.S. Nuclear Regulatory Commissions (NRCs)

Office of Nuclear Regulatory Research (RES) and facilitated by RES staff.

Objectives The objective of this PIRT exercise is to develop an ordered list of phenomena involved in a HEAF event. This list will be ordered by priority; the more important a phenomena is judged to be and the poorer its state of knowledge is judged to be, the higher its priority. This information can be used in the development of a roadmap for future research and allows for an informed focusing of resources for research and regulatory entities.

Approach The expert panel was presented with a series of specific HEAF scenarios, each based on the types of scenarios typically encountered in nuclear power plant (NPP) applications. For each scenario, a specific figure of merit was defined; that is, a specific goal to be achieved in analyzing the scenario. The panel identifies all those related phenomena that are of potential interest to an assessment of the scenario via probabilistic risk assessment (PRA) tools and methods. The phenomena are then ranked relative to their importance in predicting the figure of merit. Each phenomenon is then further ranked for the existing state of knowledge with respect to the ability of existing tools and methods to predict that phenomena, the underlying base of knowledge associated with the phenomena, and the potential for developing new data to support improvements to the existing tools.

The PIRT panel covered three distinct HEAF scenarios. The first was a HEAF occurring in an electrical enclosure with a cable tray passing over the enclosure. The second was a HEAF occurring in a bus duct passing over an electrical enclosure. The third was a HEAF occurring in an electrical enclosure situated in a bank of similar enclosures.

Results As a result of the process, level one phenomena were identified. The level one phenomena are those that were ranked with high importance and low state of knowledge. These would nominally represent potential research priorities. The level one phenomena identified by the panel included the following:

• Electrical arc characterization: thermal and magnetic effects of the arc, arc ejecta (smoke, ionized gas, conductive particulate), arc location, and migration.

• Pressure effects: mechanical shock, projectile impact, and degradation of the compartment pressure boundary.

• Arc mitigation: the use of HEAF-resistant equipment, thermal insulation, or HEAF shields to minimize damage incurred as a result of a HEAF.

xi

• Target characterization: establishing the sensitivity of target equipment to various failure mechanisms, and associated damage criteria.

• Internal ensuing fire: the likelihood, impact, and phenomenology of an enclosure fire ignited by a HEAF event.

xii

ACKNOWLEDGMENTS The author wishes to thank the members of the expert panelDr. Hajime Kabashima (S/NRA/R),

Mr. Sangkyu Lee (KINS), Mr. Nicholas Melly (U.S. NRC), Dr. Marina Rwekamp (GRS),

Dr. Koji Shirai (CRIEPI), and Dr. Sylvain Suard (IRSN)for their dedication to the process and for their perseverance. The panelists technical advisorsDr. Mikimasa Iwata, Mr. Hee Jin Ko, Mr. Cheoung Joon Lee, and Mr. Seonghyeon Yiprovided invaluable insight, and their contributions are greatly appreciated.

The author also wishes to thank the technical area experts who supported the panelistsMr. Scott Bareham, Dr. Anthony Putorti, Mr. Tomasz Stefanski, and Mr. Stephen Turner.

Lastly, the author wishes to thank the NRC staff members who supported the development and execution of the PIRT exerciseMr. Theron Brown, Ms. Tammie Rivera, Mr. Mark Henry Salley, Mr. David Stroup, Mr. Gabriel Taylor, and Mr. Joseph Zabel.

xiii

ABBREVIATIONS AND ACRONYMS CNSC Canadian Nuclear Safety Commission (Canada).

CRIEPI Central Research Institute of Electric Power Industry (Japan).

CSN Consejo de Seguridad Nuclear (Spain).

DID Defense in Depth.

ERFBS Electrical Raceway Fire Barrier System(s).

EMI Electromagnetic Interference.

EPRI Electric Power Research Institute.

FAQ Frequently Asked Questions (Reactor Oversight Program).

GDC General Design Criteria (10 CFR Part 50 Appendix A).

GRS Gesellschaft für Anlagen- und Reaktorsicherheit (Germany).

HEAF High Energy Arcing Fault.

HRR Heat Release Rate.

IAGE Integrity and Aging (Task Group).

IRSN Institut de Radioprotection et de Sûreté Nucléaire (France).

KINS Korea Institute of Nuclear Safety (S Korea).

LER Licensee Event Report.

NEA Nuclear Energy Agency.

NPP Nuclear Power Plant.

OECD Organisation for Economic Co-operation and Development.

PIRT Phenomena Identification and Ranking Table.

PRA Probabilistic Risk Analysis.

RES Office of Nuclear Regulatory Research.

S/NRA/R Regulatory Standard and Research Development, Secretariat of Nuclear Regulation Authority (Japan).

SSC(s) Systems, Structures, and Components.

STUK Steilyturvakeskus (Finland).

ZOI Zone of Influence xv

1 INTRODUCTION

1.1 Background

Switchgear, load centers, and bus bars/ducts (440 V and above) are subject to a unique failure mode and, as a result, unique fire characteristics. In particular, these types of high-energy electrical devices are subject to high-energy arcing faults (HEAFs). This fault mode leads to the rapid release of electrical energy in the form of heat, vaporized copper, and mechanical force.

Faults of this type are also commonly referred to as high energy, energetic, or explosive electrical equipment faults or fires. Similar failure modes can occur in large oil-filled transformers. [1]

The arcing or energetic fault scenario in these electrical devices consists of two distinct phases, each with its own damage characteristics and detection/suppression response and effectiveness.

The first phase is a rapid release of energy in the form of heat, light, and pressure. This energetic first phase is due to the high current arcs between electrical conductors. The second phase consists of ensuing fire(s) that may involve the electrical device itself as well as any external exposed combustibles such as overhead exposed cable trays or nearby panels that may be ignited during the energetic phase. This second phase is treated similar to other postulated fires within the zone of influence. [1]

International operating experience documents a significant number of HEAFs that have occurred worldwide in operating nuclear power plant (NPP) facilities. An Organisation for Economic Co-operation and Development/Nuclear Energy Agency (OECD/NEA) report from June 2013 documents 48 such instances in 10 member countries [2]. This number represents about 10 percent of the fire events reported through the OECD/NEA Fire Incidents Records Exchange (FIRE) program.

Appendix A to 10 CFR Part 50 [3] lists the General Design Criteria (GDC) or minimum requirements for the principal design of a water-cooled NPP. Two criteria are particularly applicable to HEAFs: criterion 3 and criterion 17. Criterion 3 requires that structures, systems, and components (SSCs) important to safety be designed to minimize the probability and effect of fires and explosions. Criterion 17 requires that the onsite electric power supplies have sufficient independence, redundancy, and testability to perform their safety functions assuming a single failure. The defense in depth (DID) approach to NPP design and operation requires multiple independent and redundant layers of defense so that no single layer, no matter how robust, is exclusively relied upon. The DID strategy, as it applies to HEAFs, consists of:

1. Prevention - Maintenance practices designed to eliminate the underlying causes of faults (loose connections, degraded insulation, foreign materials, etc.).[4]

2. Mitigation - Selective coordination to limit the duration and magnitude of the fault as well as design and location of plant SSCs such that a HEAF, should it occur, does as little damage as possible to nearby equipment and does not endanger the functionality of redundant equipment.

3. Safe shutdown - Ensuring the ability of the plant to safely shut down and maintain the reactor in a safe state in the event of a HEAF.

HEAF events have, in many cases, presented extra challenges to the safe shutdown of a reactor.

The electrical disturbance responsible for the HEAF can cause loss of essential electrical power, and the physical damage and products of combustion can present significant challenges to the 1-1

operators and fire brigade members responding to the emergency.[5][6][7] HEAFs can present one of the more risk-significant and challenging fire scenarios that an NPP will face.

In 2008, the U.S. Nuclear Regulatory Commission (NRC) contracted with Sandia National Laboratories to perform a literature review on HEAFs. The review [8] concluded that the focus of research had primarily been limited to the behaviors of the initiating equipment and the initial arc flash in the context of personnel safety. More research was necessary to evaluate the HEAF phenomenon in the context of damage to adjacent equipment.

Two methods are available to model the effects of a HEAF [9]. The first is described in NUREG/CR-6850, EPRI/NRC-RES Fire PRA Methodology for Nuclear Power Facilities, Volume 2: Detailed Methodology, Appendix M for Chapter 11, High Energy Arcing Faults [1]. This method stipulates the following assumptions:

1. The faulting device is destroyed.

2. Adjacent cubicles in the same cabinet bank will trip open.

3. Unprotected cables that enter the panel in an air-drop configuration are destroyed.

4. The first unprotected cable tray within 1.5 m of the top of the cabinet will ignite.

5. Vulnerable equipment within 0.9 m horizontally of the front or rear doors will be destroyed.

This method is subject to an unknown degree of uncertainty because it was derived from one single well-documented HEAF event that occurred at the San Onofre Nuclear Generating Station, Unit 3, on February 3, 2001 [10]. The second method can be found in Fire Probabilistic Risk Assessment Methods Enhancements Supplement 1 to NUREG/CR-6850 and EPRI 1011989 section 7 bus duct (counting) guidance for high energy arcing faults (FAQ 07-0035) [11]. This method stipulates the following assumptions for HEAFs occurring in bus ducts:

1. Molten material will be ejected from the bottom of the bus duct and spread downwards in the shape of a right circular cone whose sides are at an angle of 15 degrees from the vertical axis.

2. The cone will expand to a maximum diameter of 20 feet (37 feet below the point of origin) beyond which point molten material will fall straight down in a cylindrical shape of 20-foot diameter.

3. Molten material will be ejected outwards from the fault point in the shape of a sphere with a 1.5-foot radius.

4. Exposed combustible material within these zones will ignite.

In an effort to confirm these modeling methods described in NUREG/CR-6850 and to better characterize HEAFs, an OECD/NEA working group conducted an experimental program between 2014 and 2016, hereafter referred to as phase one. Led by the NRC, this phase consisted of 26 full-scale HEAF experiments conducted at KEMA Powertest LLCs Chalfont facility. This test program was largely exploratory; the test equipment was whatever participant countries could donate, and various measurement techniques were evaluated. The results of the phase one testing suggest that the current modeling methodology may, in some cases, be nonconservative, especially where aluminum components are involved. The full results of this experimental campaign are documented in an NEA/CSNI report issued in 2017 [12].

In August 2016, the Regulatory Standard and Development Department Secretariat on Nuclear Regulation Authority (S/NRA/R) and the NRC jointly published an International Report titled 1-2

NUREG/IA-0470 Volume 1, Nuclear Regulatory Authority Experimental Program to Characterize and Understand High Energy Arcing Fault (HEAF) Phenomena [13]. The focus of this report was to better understand the seismic-induced HEAF that occurred in 6.9 kV switchgear at the Onagawa NPP during the March 11, 2011, Great Eastern Earthquake in Japan. The research focused on simulating the HEAF that occurred and recording data such as temperature, heat flux, and heat-release rates (HRR) from ensuing fires. The report was also one of the first to identify high thermal energy released when aluminum electrical components are involved.

The phase one testing and joint S/NRA/R NRC report demonstrated the need for additional, more focused testing to further refine HEAF modeling methodology. Given the complexity of a HEAF event and related phenomena, the Office of Nuclear Regulatory Research held a Phenomena Identification Ranking Table (PIRT) exercise in February 2017 to aid in the process of planning future testing.

1.2 Objectives The objective of this PIRT exercise is to develop an ordered list of phenomena involved in a HEAF event. This list will be ordered by priority; the more important a phenomena is judged to be, and the poorer its state of knowledge is judged to be, the higher its priority. This information can be used in the development of a roadmap for future research and will allow for an informed focusing of resources for research and regulatory entities.

1.3 Scope The scope of the PIRT exercise defines the bounds of:

1. The scenarios to be analyzed.

2. The figures of merit to be achieved.

3. The phenomenas state of knowledge.

A well-defined scope will ensure that the scenarios are relevant to the intended application and that the groups analysis does not stray into tangents and hypotheticals.

The scope of this PIRT exercise is:

1. A high-energy electric arc fault event.

2. Occurring in a nuclear power generating station.

3. In low voltage (less than 600V) or medium voltage (between 600V and 15 kV) equipment or circuits.

4. NOT in the switchyard or electrical transmission/distribution network and equipment Although an arc fault in the switchyard may result in a loss of offsite power, it is phenomenologically separate from other HEAF scenarios because:

1. It is not confined in a compartment.

2. It is not located near safety-related equipment.

3. High voltages (greater than 15kV) are common.

1-3

For these reasons, arc faults occurring on the switchyard side of the main generator and beyond are outside of the project scope.

1.4 Report Organization Section 2 of this report discusses the PIRT process used, the preparation of the PIRT panelists, the instructions and ranking definitions given to the panelists, and a description of three HEAF scenarios.

Section 3 is a summary of the level 1 phenomena in each of the three scenariosthose phenomena that rank highly for importance and poorly for state of knowledge. These are, in essence, the primary candidates for future research activity.

Section 4 synthesizes the panelists rankings, comments, and discussions into a series of recommendations for future research.

Appendices A-C contain the formatted output of the PIRT, with each panelists anonymized rankings.

1-4

2 OVERVIEW OF THE PIRT PROCESS APPLIED

2.1 Background

A phenomena identification and ranking table (PIRT) is a structured expert elicitation process. It is a methodical, tested process for compiling expert opinions, assessing the state of knowledge, and identifying research priorities on a particular topic. The U.S. Nuclear Regulatory Commission (NRC) has sponsored or performed a number of PIRT exercises. Two recent PIRTs that were performed by the NRCs Fire Research Branch were NUREG/CR-6978, A Phenomena Identification and Ranking Table (PIRT) Exercise for Nuclear Power Plant Fire Modeling Applications, [14] and NUREG/CR-7150 Volume 1, Phenomena Identification and Ranking Table (PIRT) Exercise for Nuclear Power Plant Fire-Induced Electrical Circuit Failure. [15]

2.2 Selection of Panelists This PIRT exercise was a follow-up to phase one testing and a prerequisite to future phase two testing. For this reason, each country/agency that participated in the phase one testing was invited to send one representative to sit on the panel. At their discretion, each country/agency was also invited to send technical expertsnon-voting representatives who could advise the panelist, but whose input would not be recorded.

An invitation was extended to the following countries/organizations:

• CNSC (Canada)

• KINS (S Korea)

• GRS (Germany)

• IRSN (France)

• STUK (Finland)

• CRIEPI (Japan)

• NRA (Japan)

• CSN (Spain)

• NRC (United States)

• EPRI (United States)

Of these 10 agencies, the following sent representatives:

• KINS (S Korea) - Mr. Sangkyu Lee

• GRS (Germany) - Dr. Marina Rwekamp

• IRSN (France) - Dr. Sylvain Suard

• CRIEPI (Japan) - Dr. Koji Shirai

• NRA (Japan) - Dr. Hajime Kabashima

• NRC (United States) - Mr. Nicholas Melly 2.3 Panelist Preparation None of the PIRT panelists had participated in a PIRT before. To maximize what could be accomplished during the week of the PIRT, the panelists were provided with a variety of preparation and orientation materials.

2-1

2.3.1 Knowledge Base Several months in advance of the PIRT, panelists were given access to an online file store containing relevant literature on high energy arcing faults. In addition to those materials provided, panelists were asked to contribute any other material of which they were aware. This knowledge store is essentially the equivalent of a literature review for this exercise. These materials include both nuclear and non-nuclear studies, and a list of these materials are given in the references section of this report (section 5.2).

The objective of providing these materials was to calibrate the panelists for their evaluation of phenomena state of knowledge. To draw a conclusion about whether a phenomenon has been adequately characterized, some knowledge of the current literature is required.

2.3.2 Orientation Materials In addition to access to the online knowledge base, panelists were provided with periodic orientation materials. These materials largely dealt with the PIRT process itself to familiarize panelists with the steps involved. These orientation materials also included worked example scenarios and logistical information.

2.4 The PIRT Process Applied 2.4.1 Scenario Development The scenarios to be analyzed were developed so that they span the plausible range of HEAF events while still staying within the scope of the PIRT. The development of these scenarios was guided by documented operating experience and common NPP configurations. The scenarios are mainly focused on the HEAF source equipment. The panelists were instructed to consider the targets only in a generic manner. In reality, each target is unique and will respond differently to the environmental conditions caused by a HEAF; however, the behavior of specific targets is a voluminous subject and outside the scope of the PIRT. Therefore, when considering the damage states of a target in the scenarios, panelists listed phenomena that are applicable to any generic target of a similar equipment class.

Each scenario is accompanied by a figure of merit (i.e., a goal to be achieved through analysis of the scenario). In other PIRTs, the figure of merit is often the prediction of a specific quantity, such as ceiling jet temperature or pressure rise. In the case of this PIRT, where all possible mechanisms of failure are being considered, the figure of merit for each scenario is simply the determination of the extent of damage to the targets. This broad figure of merit allowed panelists to consider any and all phenomena.

2.4.2 Steps in the PIRT Process The PIRT process follows a series of formal steps, each of which is outlined here.

2.4.2.1 Presentation of the HEAF Scenario The first step is the presentation of a HEAF scenario, which includes a description of the pre-defined elements and characteristics of the scenario as well as the specific objectives and goals of our analysis. Section 2.5 details the three scenarios used in this PIRT.

2-2

2.4.2.2 Phenomena Identification The next step is for the panel to identify all phenomena relevant to the scenario and the analysis goals. A phenomenon is loosely defined as something that is observed to happen. In the context of this PIRT, a phenomenon is something that has, or might have, a model associated with it. A phenomenon may include one or more sub-phenomena, which are grouped accordingly.

As an example, an external ensuing fire that occurs as a result of a HEAF is a phenomenon. It is a physical occurrence that PRA practitioners need to model. This phenomena can be broken down into sub-phenomena, each with its own sub-model. Examples of sub-phenomena for an external ensuing fire are ignition, fire development, and smoke generation.

2.4.2.3 Parameter Identification In concert with the identification of phenomena, panelists were also asked to identify key parameters associated with the phenomenon. If phenomena can be thought of as events to be modeled, parameters are the inputs to those models. These are quantities that can be measured and reported with standard units. Each phenomenon or sub-phenomenon may encompass multiple key parameters.

Continuing the example above, some relevant parameters for the sub-phenomena of an external ensuing fire are heat release rate, fire growth rate, and time to peak heat release rate.

The process components outlined here and in in section 2.4.2.2 are almost like a brainstorming session in that the group will need to think of all phenomena and parameters that might have an impact on the scenario objective. Even phenomena or parameters that are likely to have a minor impact on the objective should be identified; their relative importance will be decided by the group in the next stage.

2.4.2.4 Phenomena Importance Ranking In the next step, the panel ranks each identified phenomenon for its importance to the figure of merit. If the group can achieve consensus in their ranking through discussion, it lends extra credibility to the ranking, but consensus is not required. Each panel member is entitled to their opinion, and it is recorded as it is received. This stage of the PIRT requires that the panel members be familiar with the available experimental evidence to provide well-informed opinions about what factors influence the characteristics of HEAF events and their relative importance.

Section 2.4.3 describes the possible rankings for phenomena importance and their definitions.

2.4.2.5 State of Knowledge Ranking In the next step, the panel assessed the state of knowledge for each phenomenon. This includes the current body of knowledge regarding a particular phenomenon as well as the ability to acquire information regarding that phenomenon. If new measurement techniques or equipment are required to increase the state of knowledge regarding a particular phenomenon, this should be reflected in the output of the PIRT.

Section 2.4.3 describes the possible rankings for phenomena state of knowledge and their definitions.

2-3

2.4.2.6 Parameter Importance and State of Knowledge Ranking In addition to ranking the identified phenomena, panelists were asked to identify and rank key parameters for each of the phenomena where applicable. The ranking of a parameter is relative to its parent phenomenon. A phenomenon may be ranked as unimportant, but its child parameters may be highly important to characterizing that phenomenon. This process was applied even to the lowly ranked phenomena for reasons of extensibilitywhile this panel may deem a phenomenon unimportant to the overall analysis, future research may contradict such a judgment.

Section 2.4.4 describes the possible rankings for parameter importance and state of knowledge and their definitions.

2.4.3 Phenomena Ranking Definitions It is important that PIRT participants have consistent and well-defined criteria to rank the phenomena and their states of knowledge. For simplicity and consistency, a low, medium, high scale was used. One category ranked the importance of the phenomenon, and three categories ranked the current state of knowledge regarding that phenomenon.

It is also important to remember that these rankings are expert opinions. Empirical evidence may or may not be available to support these rankings, but the expert should try to justify the ranking as much as possible. A ranking of high indicates that the given phenomenon is very important to predicting the figure of merit. Without knowledge of the phenomenon, an accurate prediction would not be possible. A ranking of medium indicates a phenomenon that has some appreciable impact on predicting the figure of merit but is not particularly important. Without knowledge of this phenomenon, a rough estimate would still be possible. A ranking of low indicates that the expert does not believe the phenomenon is important to predicting the figure of merit and that modeling this phenomenon is not necessary. An additional category of uncertain is available that indicates the expert believes a phenomenon has an impact on predicting the figure of merit but is unable to assess the magnitude of that impact.

The following phenomena importance ranking definitions were used throughout the PIRT:

Table 2-1 Phenomena Importance Ranking Definitions Descriptor Definition High (H) First order of importance to the figure of merit of interest.

Medium (M) Secondary order of importance to the figure of merit of interest.

Low (L) Negligible importance to figure of merit of interest. Not necessary to model this phenomenon for this application.

Uncertain (U) Potentially important. Importance should be explored through sensitivity study and/or discovery experiments and the PIRT revised accordingly.

Model adequacy rankings answer the question How well does the current HEAF modeling methodology estimate the phenomenon?

2-4

The following model adequacy ranking definitions were used throughout the PIRT:

Table 2-2 Phenomena State of Knowledge Model Adequacy Ranking Definitions Descriptor Definition High (H) At least one mature physics-based or correlation-based model is available that is believed to adequately represent the phenomenon over the full parameter space of the applications.

Medium (M) Significant discovery activities have been completed. At least one candidate model form or correlation form has emerged that is believed to nominally capture the phenomenon over some portion of the application parameter space.

Low (L) No significant discovery activities have occurred and model form is still unknown or speculative.

Data adequacy descriptors answer the question How good is the data that would be used as model input, or to validate a potential model? For example, the distance between a target and the source equipment is data that is readily available, very precise, and easy to vary for experimentation. Data regarding electromagnetic interference caused by a HEAF might be less available or of lower fidelity. It is important to differentiate between the adequacy of data and the adequacy of the model. There is no uncertain category; if a panelist is not aware of pertinent data, the adequacy of the data was considered to be low.

The following data adequacy descriptors were used throughout the PIRT:

Table 2-3 Phenomena State of Knowledge Data Adequacy Ranking Definitions Descriptor Definition High (H) A highly reliable assessment can be made based on existing knowledge.

Data needed are readily available.

Medium (M) Data are available but are not ideal due to questions of fidelity.

Moderately reliable assessments of models can be made based on existing knowledge.

Low (L) Assessments cannot be made with even moderate reliability based on existing knowledge.

The last ranking category is the ability to gather new data for a particular phenomenon. These descriptors answer the question How easy would it be to get the data we need if we dont already have it? This category can help prioritize the collection of data in future experiments if some data are known to be easier to collect than others.

2-5

The following data adequacy descriptors for the potential to develop new data were used throughout the PIRT:

Table 2-4 Phenomena State of Knowledge Availability of New Data Ranking Definitions Descriptor Definition High (H) Data needed are readily obtainable based on existing experimental capabilities.

Medium (M) Data would be obtainable but would require moderate, readily attainable extensions to existing capabilities.

Low (L) Data are not readily obtainable and/or would require significant development of new capabilities.

Panelists were not required to rank a phenomenon if they choose not to. If a panelist decided that he/she did not want to rank a phenomenon either for importance or state of knowledge, a ranking of X was entered.

2.4.4 Parameter Ranking Definitions The following parameter importance and ranking definitions were used throughout the PIRT:

Table 2-5 Parameter Importance Ranking Definitions Descriptor Definition High (H) First order of importance in characterizing the parent phenomenon.

Medium (M) Secondary order of importance in characterizing the parent phenomenon.

Low (L) Negligible importance in characterizing the parent phenomenon.

Uncertain (U) Potentially important. Importance should be explored through sensitivity study and/or discovery experiments and the PIRT revised accordingly.

Unlike phenomena, a single category was used to rank the state of knowledge for the identified parameters. The following parameter state of knowledge ranking definitions were used throughout the PIRT:

Table 2-6 Parameter State of Knowledge Ranking Definitions Descriptor Definition High (H) The parameters influence on its parent phenomenon is well characterized, and the parameter is easily measured.

Medium (M) The parameters influence on its parent phenomenon is known but not well characterized, and measurement of the parameter is possible.

Low (L) The parameters influence on its parent phenomenon is unknown, and measurement of the parameter is difficult or impossible.

2.4.5 Country-Specific Concerns and Input Although the PIRT focused on the physical phenomenology of HEAF events and the general scientific state of knowledge, concerns specific to the panelists represented country or organization inevitably arose. Some panelists stated that several phenomena were not relevant for their applications and chose a ranking of Unknown or no ranking at all. Other panelists noted 2-6

that even the HEAF scenarios, as general as they were, would be an improbable or impossible configuration in their countries facilities; therefore, their consideration of such scenarios was based less on experience and more on expert judgment, and this is reflected in their rankings.

The associated cost was another concern, especially when considering the ability to obtain new data. For many phenomena, the only way to obtain the necessary data would be to run full-scale experiments that are costly and time consuming. Although the state-of-knowledge rankings are supposed to reflect the scientific communitys technical abilities, practical considerations like time and money are inevitably factored in.

2.4.6 Panelist Feedback During the PIRT, panelists raised three procedural issues that are worth documenting here with the intent that future PIRTs can improve upon the process and that readers understand what was done and why.

The first issue discussed was whether it was a judicious use of time to rank key parameters for phenomena that had already been ranked of low importance. Establishing the key parameters for a phenomenon that will likely not be investigated seems, on the surface, to be impractical. The reason for doing so is that the panelists were working from the base of knowledge that existed at the time when the PIRT was held. As the state of knowledge improves, phenomena previously thought to be insignificant may be reconsidered, and having their key parameters ranked puts future readers at an advantage.

The second issue was the difficulty in assessing the ability to obtain new data for the state-of-knowledge rankings. Where the state of a model was judged to be poor or nonexistent, panelists were unable to reliably assess what data would need to be obtained or the ease of obtaining it. In cases like this, panelists were asked to rely on their judgment in assessing this state-of-knowledge category.

Lastly, it was noted during the PIRT exercise that a very large number of phenomena were being ranked highly important compared to those phenomena being ranked of medium or low importance. This is the result of an inevitable selection bias; when the panelists were asked to list all phenomena that could be of importance in predicting the figure of merit, the only phenomena they listed were those that could conceivably have an appreciable impact. In other words, the list of phenomena that the panelists needed to rank had already been preselected for some level of importance. This results in the median importance value being relatively high, with discrepancies between panelists being the largest source of variation.

2-7

2.5 Summary Scenario Descriptions 2.5.1 HEAF Scenario 1 2.5.1.1 Scenario Description Figure 2-1 Scenario 1 - HEAF in a Train A Switchgear with Train B Cables Overhead A high energy arcing fault occurs in a low/medium voltage switchgear that supplies power to Train A systems. The cables from an overhead cable tray enter the cabinet from the top (air-drop configuration). A cable tray carrying cables that supply power to redundant Train B systems is located near the Train A switchgear and cable tray.

2.5.1.2 Figure of Merit Determining the extent of damage to the cables in the Train B cable tray.

2-8

2.5.2 HEAF Scenario 2 2.5.2.1 Scenario Description Figure 2-2 Scenario 2 - HEAF in a Train B Bus Duct above a Train A Switchgear A high energy arcing fault occurs in a non-segregated bus duct supplying power to Train B systems. A switchgear supplying power to redundant Train A systems is located nearby.

2.5.2.2 Figure of Merit Determining the extent of damage to the switchgear.

2-9

2.5.3 HEAF Scenario 3 2.5.3.1 Scenario Description Figure 2-3 Scenario 3 - HEAF in a Train A Switchgear in a Bank with Train B Cables Overhead A high energy arcing fault occurs in a low/medium voltage switchgear supplying power to Train A systems in a bank of similar switchgears. The cables from an overhead cable tray enter the cabinets from the top (air-drop configuration). A cable tray carrying cables that supply power to Train B systems is located nearby.

2.5.3.2 Figure of Merit Determining the extent of damage to the adjacent switchgears and the cables in Train B cable tray.

2-10

3

SUMMARY

OF LEVEL 1 PHENOMENA 3.1 Ranking Methodology The method used to convert the panelists rankings into an ordered, prioritized list is as follows:

First, each ranking was given a corresponding numerical score as shown in Table 3-1.

Table 3-1 Numerical Equivalents for Ranking Values Ranking Value Low (L) 1 Low-Medium (L-M) 2 Medium (M) 3 Medium-High (M-H) 4 High (H) 5 Unknown (U) No value Not Applicable (N/A) No value Decline to Answer (X) No value Second, the average of all the values in a phenomenon (including its sub-phenomena) was taken for both importance and state of knowledge. An example using tables A7 and A19 (pressure effects for scenario 1) is illustrated below tables 3 3-5.

Table 3-2 Importance Rankings for Pressure Effects, Scenario 1 Importance Ranking Phenomenon Description P1 P2 P3 P4 P5 P6

7. Pressure Effects A. Projectile/missile damage L H M H H L B. Pressure wave L H M H H M The corresponding numerical values are assigned, and an arithmetic average is calculated.

Table 3-3 Numerical Equivalent of Table 3-2 Importance Ranking Phenomenon Description Table Average P1 P2 P3 P4 P5 P6

7. Pressure Effects A. Projectile/missile 1 5 3 5 5 1 damage 3.5 B. Pressure wave 1 5 3 5 5 3 3-1

Table 3-4 State of Knowledge for Pressure Effects, Scenario 1 State of Knowledge Ranking Phenomenon Description Model Data Ability to Collect Adequacy Availability New Data

7. Pressure Effects A. Projectile/missile damage M L-M L B. Pressure wave H H N/A The corresponding numerical values are assigned, and an arithmetic average is calculated.

Table 3-5 Numerical Equivalent of Table 3-4 State of Knowledge Ranking Phenomenon Model Data Ability to Table Average Description Adequacy Availability Collect New Data

7. Pressure Effects A. Projectile/missile 3 2 1 damage 3.2 B. Pressure wave 5 5 N/A Lastly, a rank was assigned using the following simple relationship:

=

In the case of the example above, this would be:

3.5

= = 1.09 3.2 This process was applied to each phenomenon, and a summary of their rankings across all three scenarios sorted by average rank in descending order is shown in Table 3-6.

3-2

Table 3-6 Summary of Phenomena Rankings for All Scenarios Phenomenon Scenario 1 Scenario 2 Scenario 3 Average Rank Rank Rank Rank Level 1 Phenomena Target Characterization 1.53 1.61 1.53/1.56 1.56 Arc Characterization 1.33 1.27 2.00 1.53 Arc Mitigation 1.75 1.19 1.23 1.39 Cabinet Lineup Effects 1.29 1.29 Level 2 Phenomena Internal Ensuing Fire 1.63 0.31 1.54 1.16 External Ensuing Fire 1.27 0.82 1.32 1.14 Pressure Effects 1.09 0.40 1.67 1.05 Electrical Configuration Effects 1.00 1.00 1.00 1.00 Level 3 Phenomena External Cabinet Enclosure Effects 0.80 1.04 0.76 0.87 Internal Cabinet (Bus Duct) Configuration 0.91 0.72 0.93 0.85 Effects Suppression Effects 0.80 0.40 0.67 0.62 Room Configuration 0.79 0.43 0.65 0.62 Fire Detection 0.50 0.40 0.26 0.39 As is evident from Table 3-6, the list of phenomena was split into three equal-sized groups: level 1, level 2, and level 3. It is important to note that this grouping conveys no information about the absolute priority of any phenomenon, only its priority relative to the others. Level 1 phenomena are those that present the highest research priority, whereas level 3 phenomena are those that present the lowest research priority.

3.2 Discussion of Level 1 Phenomena For the level 1 phenomena listed in Table 3-6, the highlights of the panelists discussions are included in this section.

3.2.1 Target Characterization During the PIRT, panelists considered the targets in a generic sense; that is, they ranked phenomena associated with any generic piece of equipment of a similar class. The panels opinion is that, unsurprisingly, the properties of the target will have a first order impact on the damage that it sustains.

This phenomenon was broken down into three main sub-phenomena:

A) Characterization of target damage criteria B) Target arrangement effects C) Target sensitivity to damage Sub-phenomena (A) refers to the ability to assign some type of damage threshold to the target at hand such as maximum temperature, maximum heat flux, maximum pressure, etc. The panelists noted that threshold damage criteria exist for fire scenarios, where the fire is expected to follow 3-3

some assigned growth pattern. No such criteria exists for a HEAF exposure, and any ability to predict whether a target is damaged is moot without them.

Sub-phenomena (B) refers to the targets position and orientation relative to the HEAF source.

This item was not so easy to consider as it represents a departure from the currently employed zone of influence (ZOI) method and the standard method of analysis; here, the distance from the source equipment is being considered as a variable in calculating target damage rather than a binary in/out of ZOI.

Sub-phenomenon (C) refers to a targets susceptibility to damage from a number of items: heat, pressure, smoke, conductive particulate, and others. Different classes of equipment are likely to respond differently to each of these hazards although it may be impractical to characterize the sensitivity and damage criteria of each potential target that would represent the most realistic treatment of the scenario. Sub-phenomena (A) and (C) are linked but not identical. Susceptibility to various modes of damage is an important factor in establishing meaningful damage threshold criteria, but the knowledge of a targets physical response is different from the knowledge of its functional response.

Characterization of the target in scenario 2 is similar to that of scenario 1. The higher ranking of this phenomenon in scenario 2 is partially due to simpler source equipment. With a bus duct that is unlikely to experience an ensuing fire, the properties of the target receive more weight.

Scenario 3 featured two targets for panelists to consider: the enclosures adjacent to the source enclosure and an overhead cable tray. Each target was considered separately, but they are combined here for the purposes of discussion. As with scenarios 1 and 2, the panelists ranked the phenomenology of each target in a generic sense; that is, their rankings apply to any piece of equipment of a similar class. The panelists ranking of the phenomena, sub-phenomena, and parameters for both targets are similar. Orientation effects was not considered for the enclosure targets because the position and orientation are fixed by the scenario definition. There was slightly more uncertainty about the response of an electrical enclosure to a HEAF than a cable tray due to their relative complexity. Panelists noted that enclosure sensitivity to certain failure mechanisms would be a function of enclosure type and contents; considering this type of target in a generic sense is a rough approximation.

3.2.2 Arc Characterization Characterization of the arc is essentially the characterization of the source term. Within this phenomenon, the highly ranked items were characterizing the arcs electrical properties (voltage, current, duration, etc.); arc breeching of the enclosure; arc ejecta; and the thermal effects of the arc.

The state of knowledge regarding the characteristics of the arc is mixed. Models and data exist for the electrical properties of the arc, although some important properties such as duration have external dependencies that cannot be accounted for. The panel is not aware of any models for the breeching of the enclosure or arc ejecta. These two areas are critically important: a breeched enclosure presents a far greater hazard to adjacent equipment, and arc ejecta (smoke, conductive particulate, ionized gas) may be a significant failure mechanism for target equipment.

In scenario 2 as in scenario 1, characterization of the arc is highly ranked. One topic of discussion among the panelists was whether, and in what ways, an arc occurring in a bus duct differs from an arc occurring in an electrical enclosure. The current guidance in NUREG/CR-6850 [1][11] lays out 3-4

separate methods for evaluation of the ZOI in bus ducts and electrical enclosures due largely to what has been observed in operating experience and the expected arrangement of the bus duct relative to potential targets. This approach, while practical, says little about the underlying phenomena involved, and the panel wondered whether a single model could accurately describe an arc in both types of equipment. Also, as in scenario 1, the highly ranked sub-phenomena were the electrical characterization of the arc (voltage, current, duration); the thermal effects of the arc; arc breeching of the enclosure; and arc ejecta.

3.2.3 Arc Mitigation Arc mitigation refers to the intentional shielding of targets from potential arc source equipment.

Within this phenomenon, discussions were focused around HEAF shields, which have no formal definition; electrical raceway fire barrier systems (ERFBS); and fire-retardant cable coatings. The presence of a barrier between the HEAF source and target equipment should moderate the incident thermal energy, pressure front, and arc ejecta effects.

The reasons for low state of knowledge rankings are myriad. Because no formal or industry-standard definition exists for a HEAF shield, no performance criteria or qualifications exist. The panel is not aware that any test campaign has evaluated the impact of such a shield, and no operating experience is available to quantify its performance. The panel is unanimous in their opinion that the performance of ERFBS and fire-retardant cable coatings is reasonably understood under fire conditions but not under HEAF conditions. Panelists raised the possibility that the pressure wave from the HEAF could damage an ERFBS and render it ineffective. In addition, the effects of a high heat, short-duration exposure of a HEAF event on these shields is not well understood. The typical assumptions that apply to fire exposures do not apply here; there is no growth phase, the velocity of hot gases is much higher, and the constituents of the hot gases are different. Any analytical model to predict the effects of a HEAF on a shield would require analyses of heat and mass transfer as well as solid mechanics, and no such tool has been applied to the evaluation of HEAF shields.

3.2.4 Cabinet Lineup Effects Only scenario 3 featured a cabinet lineup, so only one data point was available for the calculation of this phenomenons average rank.

The discussions on cabinet lineup effects were focused on what properties were unique to banks of enclosures and how they might affect the results of a HEAF event. These include how the cabinets are situated next to one another; what separations are between them (none, single wall, double wall, air gap); the presence of penetrations or false floor between cabinets; and the position of the HEAF source within the bank. Some of these properties, such as cabinet separation, are closely linked to target phenomena like shielding and orientation. However, the panels consensus was that cabinet lineup effects should be treated as its own category because banks of cabinets do not respond as if they are several closely spaced but discrete targets.

Mechanical coupling of the enclosures and shared penetrations often occurs.

3.3 Additional Discussions This section captures some of the important points of the panelists discussion that do not directly pertain to level 1 phenomena but are worth noting nonetheless.

3-5

3.3.1 External Duct Housing Configuration The importance of the external duct housing configuration is attributed to its structural design including the thickness, material, and penetrations/access points. Test data from the OECD/NEA program provides direct evidence for the significant impact of the duct housing material; in test 26, several inches of aluminum housing were oxidized creating a fireball that melted equipment over 20 feet away. Although the configuration of the external duct housing encompasses a number of things, the driver behind its ranking is the possible presence of aluminum in the housing.

3.3.2 Internal Ensuing Fire The source equipment identified in scenario 1 was a generic electrical enclosure, and the associated target was a redundant cable tray passing over the enclosure. The panelists identified the occurrence of an ensuing fire in the source enclosure as being of first-order importance in predicting the damage to the cable tray. Although a HEAF would likely produce temperature far in excess of any ensuing fire, the extended duration of an ensuing fire poses a secondary threat to adjacent equipment.

The highly ranked sub-phenomena discussed included ignition, characterization of the fire source, and development of the fire. These are all critical phenomena in predicting the behavior and effects of a standard cabinet fire.

The state of knowledge regarding internal fires reflects this missing link. The lowest ranked phenomenon was fire ignition, and the panelists are unsure how to predict or model ignition in a HEAF scenario.

3.3.3 Pressure Effects Panel discussions on pressure effects were centered on two failure mechanisms: a pressure wave and a projectile. In the former, a wave of pressure disturbance propagates through the medium (air) imparting mechanical energy to targets; in the latter, the pressure wave accelerates a solid object, which subsequently imparts mechanical energy to a target upon impact. The term shock wave, which implies supersonic propagation, is not used here. Pressure measurements during phase one experiments were unreliable and noisy, and propagation speed was not measured.

Cabinets in a lineup may be particularly sensitive to a pressure wave depending on their configuration. Where sensitive electronic equipment is present, being in the same bank as a HEAF may be sufficient to cause damage even where no thermal insult is incurred. For the enclosures immediately adjacent to the source cabinet, the construction of the separating walls becomes important. Operating experience and test results clearly document the ability of a HEAF to open or dislodge enclosure doors and bow or deform enclosure walls. The panel theorized that this type of mechanical shock could raise the possibility of a secondary HEAF in an adjacent cabinet.

3-6

4 CONCLUSIONS AND RECOMMENDATIONS 4.1 Practical Considerations for PRA Practitioners The phenomena identification and ranking table (PIRT) was useful in identifying the phenomena that are most influential in predicting the impact of a high energy arcing fault (HEAF); however, certain practical considerations must be taken into account when analyzing the data for the formulation of a test plan.

Ultimately, the goal of the HEAF research program is to refine and improve the methodology by which probabilistic risk assessment (PRA) practitioners estimate the risk of a postulated HEAF event. The input to any such model or methodology must be available to the practitioner. It would be unproductive to study the impact of microscale or stochastic phenomena knowing that it cannot be incorporated into the applicable guidance.

4.2 Characterizing Target Damage One of the consistently highly ranked phenomena was the characterization of the target equipment and its susceptibility to various types of damage. This phenomenon fundamentally differs from the others in that it is not a property of the HEAF or source equipment. Even if a HEAF can be modeled with perfect accuracy, the ability to convert that data into a damage state for a given target is critical in assessing risk.

Although thermal damage appears to be the greatest threat both in experimental observations and PIRT responses, HEAF events fall outside the range of applicability for classical fire models.

The temperature of the arc column can be in excess of 15,000 °C, which is unmatched even by rocket fuel or thermite. HEAF events also do not have growth or decay phases; the energy is released suddenly and intensely before the event terminates just as suddenly. With this type of energy release, the thermal inertia of targets must be taken into accounta simple temperature threshold for damage does not adequately describe the target vulnerability.

Any future research program should focus on modeling the HEAF and on modeling the damage to targets as a function of the environmental conditions created by the HEAF. In scenario 3, the difference in response of the two targets (cable tray and enclosure) was discussed, which raised the possibility that a one-size-fits-all zone of influence approach may not be sufficient. Different classes of target equipment may need to be considered separately.

Though thermal conditions were a focal point during phase one testing, the PIRT suggests that pressure effects and the effects of conductive particulate on electronic equipment should be considered as well.

4.3 Arc Mitigation The multi-tiered approach to safety advocated by defense in depth (DID) strategies applies to HEAF as well as any phenomenon. Operating plants have placed a heavy emphasis on the prevention aspect of DID with selective coordination designed to limit the duration of an arc fault to a few cyclesnot enough to cause significant external damage. These protective schemes are often effective in limiting the damage from arc faults, but a non-trivial number of these arc faults are accompanied by the failure of an upstream breaker resulting in event durations that have exceeded 10 seconds. The frequency of these events, coupled with the catastrophic damage they can cause, demand more than a single element of DID.

4-1

If a HEAF cannot be prevented, the next logical step is to mitigate the damage it can cause. This can be done in a number of ways, but the panel discussions were focused around enclosing the source equipment (e.g., arc resistant cabinets 1); protecting the target equipment (ERFBS, cable coatings); or placing barriers between the two. The pros and cons of each option were discussed.

Enclosures designed to contain the blast of a HEAF represent a costly solution to this problem.

The panel deemed the replacement of electrical enclosures susceptible to HEAF with such enclosures beyond the realm of feasibility; however, their use in new construction was seen as slightly more reasonable.

Target protection is used extensively in fire protection but, where HEAFs are concerned, questions abound. Little data is available regarding the ability of classical fire protection methods (ERFBS, cable coatings) to withstand the impact of a HEAF. Target protection may be an effective method of HEAF mitigation, but further study is needed.

The last methoda physical barrier separating the source and target equipment (or HEAF shield)strikes a good balance between cost and efficacy in theory. Some operating facilities have already installed some form of HEAF shield. Because no formal or industry-standard definition for such a barrier is available, no performance criteria or qualifications exist. The panel is not aware of any test campaign that has evaluated the impact of such a shield, and no operating experience is available to quantify its performance. This is an area that requires further study and should be considered in future research test plans.

4.4 Internal Ensuing Fire Internal ensuing fires show up as high-priority items in both scenarios where the source equipment is an electrical enclosure (as opposed to a bus duct, which was assumed not to be able to support a significant sustained internal fire.) Based on operating experience and test evidence, internal ensuing fires are the most likely secondary hazard in a HEAF scenario.

The U.S. Nuclear Regulatory Commission and others have performed a significant amount of work in an attempt to quantify the behavior of a fire in an electrical enclosure [16][17]. What has not been quantified is the link between a HEAF and the advent of a fire. Specifically, the following two questions arose during the PIRT:

a) What is the likelihood of an internal fire pursuant to a HEAF?

b) Can the HEAF simply be modeled as a very energetic ignition event or does it impact the development of the fire through pre-heating and other phenomena?

Item (a) is not just a question of frequencyit should be the goal of future research to determine what conditions must be present for a sustained internal ensuing fire. From phase one testing, a rough pattern has emerged: arcs with durations of less than two seconds did not cause internal ensuing fires likely due to the linear relationship between arc duration and total energy. More tests are needed to establish whether arc energy (and by extension, arc duration) is the primary criteria for predicting ensuing fires and other conditions that are necessary or favorable for an ensuing fire.

1 The term arc-resistant is a misnomer and generally refers to a cabinet designed to route hot gases and flames away from service personnel in the event of an arc. It is not designed to prevent or extinguish arcs.

4-2

Item (b) is critical for a PRA analysis of the ensuing fire and could implicate the need for a fundamentally different method of analysis. The typical fire models used for electrical enclosure fires assume a growth phase that lasts for several minutes, steady burning at peak heat release rate for several more minutes, and then a decay phase. If a HEAF is sufficiently energetic, it may heat and simultaneously ignite all cabinet internals, skipping the growth phase entirely. This type of behavior was observed several times during phase one testing. In addition to skipping a growth phase, the instantaneous ignition of cabinet internals may lead to shorter duration, higher intensity fires than would typically be expected. A sensitivity study on plant response to this alternate fire growth curve is in order. Future research should focus on the impact of a HEAF on the growth and development of an ensuing fire as compared to a standard electrical enclosure fire.

4.5 Arc Characterization Characterization of the arc is essentially characterization of the source termthe driver behind any resultant damage, analogous to the heat release rate term in a fire model. The electrical properties of the arc including voltage, current, and duration determine the amount of energy released into the environment of interest. It is important to distinguish between the electrical properties of the equipment in which the arc occurs and the electrical properties of the arc itself.

Although the former is known to a high degree of accuracy, the latter is subject to some uncertainty.

One area that was discussed at length was the location of the arc and whether it was predictable.

The arc location is an important parameter for predicting the flow path of the plasma and ionized gas, if and where the arc will migrate, and whether the arc is likely to breech the enclosure.

Unfortunately, past test data is not helpful in this regard as the arc was initiated in a predetermined location with the use of shorting wire as described in the IEEE standard for testing switchgear internal arcing faults.[18] Determining whether the location of an arc is predictable, random, or stochastic should be a focus of future experimental work.

Arc ejecta, which includes the smoke, ionized gas, and conductive particulate expelled by the arc, were also discussed. Ionized gas and conductive particulate may present a failure mechanism for nearby equipment that has previously not been considered. During phase one testing, conductive particulate from the arc caused shorting in laboratory equipment several meters away from the test enclosure. Panelists noted the importance of quantifying this ejecta but also that it would be difficult or impossible to separate the ejecta into its constituents. Future research should focus on establishing the properties of the ejecta (thermal conductivity, electrical conductivity, mean particle size) and its potential impact on target equipment.

4.6 Pressure Effects After thermal effects, the panel ranked pressure effects as the greatest hazard to surrounding systems, structures, and components. The mechanical energy imparted by a pressure wave or projectile has the potential to disable equipment, trip breakers, cause secondary arcs, damage pressure boundaries (fire doors and dampers), and change the ventilation properties of the source enclosure.

Pressure measurements during phase one testing are difficult to analyze, and the impact of pressure effects is mostly anecdotal. From the test report:

Due to the EMI generated by the large currents, voltages, and arc activity, the pressure signals are noisy. The EMI tends to be most severe during large changes in current, 4-3

voltage, and arc activity, and these are the same periods where large changes are expected in enclosure pressure. For this reason, it is difficult to determine whether the transducer signal peaks are actually pressure peaks.[12]

Despite the difficulty in obtaining reliable pressure measurements, some qualitative data are available. In numerous tests, cabinet access doors were blown open, cabinet internals were violently dislodged, and enclosure walls were deformed. An LER [19] and inspection report [20]

from a 2017 HEAF event at Turkey Point document the deformation of a fire door due to pressure increase, which raises the concern of an ensuing fire unbounded by the normal compartmentation scheme.

The lack of high-quality pressure measurements factored into the low state of knowledge rankings for pressure-related phenomena. Future test programs would benefit from a better method of pressure measurement free from EMI, both inside and outside the enclosure.

4-4

5 REFERENCES 5.1 References Found in this Report Fire PRA Methodology for Nuclear Power Facilities, Volume 2: Detailed Methodology, NUREG/CR-6850/EPRI 1010989, U.S. NRC, September 2005.

OECD FIRE Project - Topical Report No. 1, Analysis of High Energy Arcing Fault (HEAF) Fire Events, NEA/CSNI/R(2013)6, Organisation for Economic Co-operation and Development (OECD) Nuclear Energy Agency (NEA), Committee on the Safety of Nuclear Installations (CSNI), 2013.

http://www.oecd-nea.org/nsd/docs/2013/csni-r2013-6.pdf Code of Federal Regulations (CFR), Title 10, Part 50, Appendix A, General Design Criteria for Nuclear Power Plants.

NRC Information Notice (IN) 2016-08, Inadequate Work Practices Resulting in Faulted Circuit Breaker Connections. June 2016.

Diablo Canyon Unit 1, LER 2000-004-00, Unit 1 Unusual Event Due to a 12 kV Bus Fault, Jun 13, 2000.

H. B; Robinson Steam Electric Plant, Unit No. 2, LER 2010-002-00, Plant Trip due to Electrical Fault, May 27, 2010.

Raughley & Lanik. Operating Experience Assessment Energetic Faults in 4.16 kV to 13.8 kV Switchgear Bus Ducts that Caused Fires in Nuclear Plants 1986-2001. NRC, February 2002.

SANDIA REPORT SAND2008-4820 High Energy Arcing Fault Fires in Switchgear Equipment, A Literature Review A Review of Current Calculation Methods Used to Predict Damage from High Energy Arcing Fault (HEAF) Events, NEA/CSNI/R(2015)10, Organisation for Economic Co-operation and Development (OECD) Nuclear Energy Agency (NEA), Committee on the Safety of Nuclear Installations (CSNI), June 2015.

https://www.oecd-nea.org/nsd/docs/2015/csni-r2015-10.pdf San Onofre Nuclear Generating Station, LER 362/2001-001, Fire and RPS/ESF Actuations Caused By The Failure of a Non-Safety Related 4.16 kV Circuit Breaker, February 2, 2001.

Fire Probabilistic Risk Assessment Methods Enhancements Supplement 1 to NUREG/

CR-6850 and EPRI 1011989 section 7 bus duct (counting) guidance for high-energy arcing faults (FAQ 07-0035)

Report on the Testing Phase (2014-2016) of the High Energy Arcing Fault Events (HEAF)

Project: Experimental Results from the International Energy Arcing Fault Research Programme. NEA/CSNI/R(2017)7, Organisation for Economic Co-operation and Development (OECD) Nuclear Energy Agency (NEA), Committee on the Safety of Nuclear Installations (CSNI), 2017.

https://www.oecd-nea.org/nsd/docs/2017/csni-r2017-7.pdf NUREG/IA-0470 Volume 1 Nuclear Regulatory Authority Experimental Program to Characterize and Understand High Energy Arcing Fault (HEAF) Phenomena A Phenomena Identification and Ranking Table (PIRT) Exercise for Nuclear Power Plant Fire Modeling Applications, NUREG/CR-6978/SAND2008-3997P, U.S. NRC, November 2008.

5-1

Joint Assesment of Cable Damage and Quantification of Effects from Fire (JACQUE-FIRE),

Volume 2: Expert Elicitation Exercise for Nuclear Power Plant Fire-Induced Electrical Circuit Failure, NUREG/CR-7150/BNL-NUREG-98204-2012/EPRI 3002001989, U.S. NRC, May 2014.

Heat Release Rates of Electrical Enclosure Fires (HELEN-FIRE), NUREG/CR-7197, U.S.

NRC, April, 2016.

Refining and Characterizing Heat Release Rates from Electrical Enclosure During Fire (RACHELLE-FIRE), Volume 1: Peak Heat Release Rates and Effect of Obstructed Plume, NUREG-2178/EPRI 3002005578, U.S. NRC, April 2016.

IEEE C37.20.7-2007 Corrigendum 1 IEEE Guide for Testing Metal-Enclosed Switchgear Rated up to 38 kV for Internal Arcing Faults.

Turkey Point Unit 3, LER 2017-001-00, Loss of 3A 4kV Vital Bus Results in Reactor Trip, Safety System Actuaions and Loss of Safety Injection Function, May 16, 2017.

Turkey Point Nuclear Generating Station - NRC Reactive Inspection Report 05000250/2017008 and 05000251/2017008. May 12, 2017.

5-2

5.2 Additional Documents Included in the Knowledge Base

• Shirai, Goda, Iwata et al. Fundamental Investigation on Successive Fire due to the High Energy Arcing Faults Event for the High-Voltage Switchgear. April 2014.

• Iwata, Tanaka, Miyagiet al. CFD Calculation of Pressure Rise and Energy Flow of Hot Gases Due to Short-Circuit Fault Arc in Switchgears. IEEE, 2015.

• Iwata, Tanaka, Ohtaka, et al. CFD Calculation of Pressure Rise Due to Internal AC and DC Arcing in a Closed Container. IEEE, July 2011.

• Iwata, Anantavanich, & Pietsch. Influence of Current and Electrode Material on Fraction Kp of Electric Arc Energy Leading to Pressure Rise in a Closed Container During Internal Arcing. IEEE, July 2010.

• Beckstead, M. W. A Summary of Aluminum Combustion, Brigham Young University, January 2004.

• Land III, H. Bruce. The Behavior of Arcing Faults in Low-Voltage Switchboards. IEEE, April 2008.

• Land III, H. Bruce. Determination of the Cause of Arcing Faults in Low-Voltage Switchboards. IEEE, April 2008.

• Davis, St. Pierre, Castor, et al. Practical Solution Guide to Arc Flash Hazards. ESA Inc, 2003.

• Arc-Fault Primer: Numerical, Analytical, and Experimental Characteristics of Initiation and Sustainment of Arc Plasmas (DRAFT), SAND2017-XXXX.

• Lee, Gammon, Johnson, et al. IEEE/NFPA Arc Flash Phenomena Collaborative Research Project (Presentation). January 2011.

• Zhou, Nie, Zeng, et al. Effects of Aluminum Content on TNT Detonation and Aluminum Combustion using Electrical Conductivity Measurements, PREP 2015.

• McBride & Weaver. Review of Arcing Phenomena in Low Voltage Current Limiting Circuit Breakers. IEEE, January 2001.

• Kwon, Gromov, Ilyin, et al. The Mechanism of Combustion of Superfine Aluminum Powders. Combustion and Flame 133, 2003.

• Babrauskas, V. Electric Arc Explosions. Interflam, 2010.

• IEEE Std C37.20.7 IEEE Guide for Testing Medium-Voltage Metal-Enclosed Switchgear for Internal Arcing Faults. IEEE, 2001.

• Lee, R. The Other Electrical Hazard: Electric Arc Blast Burns. IEEE, 1982.

• Land III, Gammon. Addressing Arc Flash Problems in Low Voltage Switchboards: A Case Study in Arc Fault Protection. IEEE, 2013.

• Frieberg & Pietsch. Calculation of Pressure Rise Due to Arcing Faults. IEEE, April 1999.

• Jones, Liggett, Capelli-Schellpfeffer, et al. Staged Tests Increase Awareness of Arc-Flash Hazards in Electrical Equipment. IEEE, April 2000.

5-3

APPENDIX A PIRT HEAF SCENARIO 1 A.1 HEAF Scenario 1 A.1.1 Scenario Description A high energy arcing fault occurs in a low/medium voltage switchgear that supplies power to Train A systems. The cables from an overhead cable tray enter the cabinet from the top (air-drop configuration). A cable tray carrying cables that supply power to redundant Train B systems is located near the Train A switchgear and cable tray.

A.1.2 Figure of Merit Determining the extent of damage to the cables in the Train B cable tray.

A.2 List of Identified Phenomena Phenomenon 1: Electrical Configuration Effects A. Power supply characterization B. Electrical protection coordination 2 2

Electrical protection coordination refers to the coordination and configuration of upstream breakers with the objective of minimizing the duration of an electrical fault A-1

Phenomenon 2: Internal Cabinet Configuration Effects A. Cabinet compartmentation B. Breaker configuration C. Cabinet combustible loading D. Cabinet bus bar configuration Phenomenon 3: External Cabinet Enclosure Configuration Effects A. Cabinet ventilation B. Cabinet structural design C. Cabinet penetrations Phenomenon 4: Arc Characterization A. Arc electrical characterization B. Arc migration C. Arc breeching of enclosure D. Thermal effects of the arc E. Magnetic effects of the arc 3 F. Electromagnetic interference G. Arc ejecta Phenomenon 5: Fire Detection 4 A. Presence of fire detection B. Characterizing the detection system Phenomenon 6: Fire Suppression A. Presence of fire suppression B. Characterizing the suppression system Phenomenon 7: Pressure Effects A. Projectile/missile damage B. Pressure wave 3

Magnetic effects refers to the applied forces on ferrous materials by a magnetic field that can induce bending and deformation, as opposed to the interference that it may cause. Interference is addressed separately.

4 Fire detection was assumed not to automatically trigger fire suppression. Fire suppression is considered as a separate phenomenon.

A-2

Phenomenon 8: Internal Ensuring Fire 5 A. Fire ignition B. Characterization of the fire source C. Fire development D. Smoke generation E. Altered ventilation effects6 Phenomenon 9: External Ensuing Fire A. Fire ignition B. Characterization of the fire source C. Fire development D. Smoke generation Phenomenon 10: Room Configuration Effects A. Room integrity 7 B. Room arrangement C. Room ventilation Phenomenon 11: Arc Mitigation A. Intentional shielding B. Electrical raceway fire barrier system effects8 C. Fire-retardant coating effects 9 Phenomenon 12: Generic Cable Tray (Target) Characterization A. Characterizing target damage criteria B. Target arrangement effects C. Target sensitivity to damage 5

Internal ensuing fire refers to a sustained fire that occurs inside the source cabinet.

6 Altered ventilation effects refers to the manner in which changes to the cabinets structure and integrity as a result of the HEAF impact the ventilation paths for an internal ensuing fire.

7 Room integrity refers to the pressure boundary of the room in which the HEAF occurs, and how its response to the increased pressure will impact target damage. For example, consider whether blowing open a fire door or damper will affect target damage.

8 Electrical raceway fire barrier systems refers to a wrap or enclosure that is installed around electrical cable trays with the intent of mitigating fire damage.

9 Fire-retardant coating refers to a permanently installed coating or spray that is applied directly to the cables in a cable tray.

A-3

A.3 Phenomena Importance Rankings Table A-1 Importance Ranking for Scenario 1, Electrical Configuration Effects Importance Ranking Phenomenon Description Additional Notes & Comments P1 P2 P3 P4 P5 P6

1. Electrical Configuration Effects A. Power supply characterization H H H H H H B. Electrical protection coordination H H H H H H Table A-2 Importance Ranking for Scenario 1, Internal Cabinet Configuration Effects Importance Ranking Phenomenon Description Additional Notes & Comments P1 P2 P3 P4 P5 P6

2. Internal Cabinet Configuration Effects A. Cabinet compartmentation M M H M M L B. Breaker configuration L/M H M M M H P1 notes that whether the HEAF occurs on the A-4 primary or secondary side of the breaker affects the importance.

C. Cabinet combustible loading H H M H H H D. Cabinet bus bar configuration M H M H M H Table A-3 Importance Ranking for Scenario 1, External Cabinet Enclosure Configuration Effects Importance Ranking Phenomenon Description Additional Notes & Comments P1 P2 P3 P4 P5 P6

3. External Cabinet Enclosure Configuration Effects A. Cabinet ventilation M H H M H H B. Cabinet structural design L H H M M M C. Cabinet penetrations M H H H M H

Table A-4 Importance Ranking for Scenario 1, Arc Characterization Importance Ranking Phenomenon Description Additional Notes & Comments P1 P2 P3 P4 P5 P6

4. Arc Characterization A. Arc electrical characterization H H H H H H B. Arc migration M M H M M M C. Arc breeching of enclosure H H H H M H D. Thermal effects of the arc H H H H H H E. Magnetic effects of the arc L L L M U L F. Electromagnetic interference L L L M U L G. Arc ejecta H M H H H L Table A-5 Importance Ranking for Scenario 1, Fire Detection Importance Ranking Phenomenon Description Additional Notes & Comments P1 P2 P3 P4 P5 P6 A-5 5. Fire Detection A. Presence of fire detection H L H L H L B. Characterizing fire detection system M L M L M L Table A-6 Importance Ranking for Scenario 1, Fire Suppression Importance Ranking Phenomenon Description Additional Notes & Comments P1 P2 P3 P4 P5 P6

6. Fire Suppression A. Presence of fire suppression H L H M H L B. Fire suppression effects H H H M H H

Table A-7 Importance Ranking for Scenario 1, Pressure Effects Importance Ranking Phenomenon Description Additional Notes & Comments P1 P2 P3 P4 P5 P6

7. Pressure Effects C. Projectile/missile damage L H M H H L D. Pressure wave L H M H H M Table A-8 Importance Ranking for Scenario 1, Internal Ensuing Fire Importance Ranking Phenomenon Description Additional Notes & Comments P1 P2 P3 P4 P5 P6

8. Internal Ensuing Fire A. Fire ignition H H H H H H B. Characterization of the fire source H H H H H H A-6 C. Fire development H H H H H H D. Smoke generation L M M M M M E. Altered ventilation effects M H H M H H Table A-9 Importance Ranking for Scenario 1, External Ensuing Fire Importance Ranking Phenomenon Description Additional Notes & Comments P1 P2 P3 P4 P5 P6

9. External Ensuing Fire A. Fire ignition H H H H H H B. Characterization of the fire source H H H H H H C. Fire development H H H H H H D. Smoke generation L L M H M L

Table A-10 Importance Ranking for Scenario 1, Room Configuration Effects Importance Ranking Phenomenon Description Additional Notes & Comments P1 P2 P3 P4 P5 P6

10. Room Configuration Effects A. Room integrity L H M L M X B. Room arrangement M H M M M H C. Room ventilation M H M M M H Table A-11 Importance Ranking for Scenario 1, Arc Mitigation Importance Ranking Phenomenon Description Additional Notes & Comments P1 P2 P3 P4 P5 P6

11. Arc Mitigation A. Intentional shielding M M H H H H B. Electrical raceway fire barrier system H L H H H H C. Fire-retardant coating M M H M H H A-7 Table A-12 Importance Ranking for Scenario 1, Generic Cable Tray (Target) Characterization Importance Ranking Phenomenon Description Additional Notes & Comments P1 P2 P3 P4 P5 P6

12. Generic Cable Tray (Target)

Characterization A. Characterizing target damage criteria H H H H H H B. Target arrangement effects H H H M H H C. Target sensitivity to damage H H H H H H

A.4 Phenomena State of Knowledge Rankings Table A-13 State of Knowledge Ranking for Scenario 1, Electrical Configuration Effects State of Knowledge Ranking Model Data Ability to Phenomenon Description Additional Notes & Comments Adequacy Availability Collect New Data

7. Electrical Configuration Effects A. Power supply characterization H H N/A B. Electrical protection coordination H H N/A Table A-14 State of Knowledge Ranking for Scenario 1, Internal Cabinet Configuration Effects A-8 State of Knowledge Ranking Model Data Ability to Phenomenon Description Additional Notes & Comments Adequacy Availability Collect New Data

8. Internal Cabinet Configuration Effects A. Cabinet compartmentation H H N/A B. Breaker configuration M-H M H C. Cabinet combustible loading M-H M M P1 notes that this phenomena is not as well known for older facilities.

D. Cabinet bus bar configuration H H N/A

Table A-15 State of Knowledge Ranking for Scenario 1, External Cabinet Enclosure Configuration Effects State of Knowledge Ranking Model Data Ability to Phenomenon Description Additional Notes & Comments Adequacy Availability Collect New Data

9. External Cabinet Enclosure Configuration Effects A. Cabinet ventilation H H N/A B. Cabinet structural design H H N/A C. Cabinet penetrations H H N/A Table A-16 State of Knowledge Ranking for Scenario 1, Arc Characterization A-9 State of Knowledge Ranking Model Data Ability to Phenomenon Description Additional Notes & Comments Adequacy Availability Collect New Data

10. Arc Characterization A. Arc electrical characterization L-M L/H L-M Panelists note that while some elements of the arc are easy to predict, others, like arc location and duration, are not.

B. Arc migration M L-M U C. Arc breeching of enclosure L L L D. Thermal effects of the arc M-H M H E. Magnetic effects of the arc L-M L-M M-H F. Electromagnetic interference L-M L-M M-H G. Arc ejecta L L L

Table A-17 State of Knowledge Ranking for Scenario 1, Fire Detection State of Knowledge Ranking Model Data Ability to Phenomenon Description Additional Notes & Comments Adequacy Availability Collect New Data

11. Fire Detection A. Presence of fire detection H H N/A B. Characterizing fire detection system H H N/A Table A-18 State of Knowledge Ranking for Scenario 1, Fire Suppression State of Knowledge Ranking Model Data Ability to Phenomenon Description Additional Notes & Comments Adequacy Availability Collect New Data

12. Fire Suppression A. Presence of fire suppression H H N/A A-10 B. Fire suppression effects H H N/A Table A-19 State of Knowledge Ranking for Scenario 1, Pressure Effects State of Knowledge Ranking Model Data Ability to Phenomenon Description Additional Notes & Comments Adequacy Availability Collect New Data

13. Pressure Effects A. Projectile/missile damage M L-M L B. Pressure wave H H N/A

Table A-20 State of Knowledge Ranking for Scenario 1, Internal Ensuing Fire State of Knowledge Ranking Model Data Ability to Phenomenon Description Additional Notes & Comments Adequacy Availability Collect New Data

14. Internal Ensuing Fire A. Fire ignition L L L B. Characterization of the fire source H M M-H C. Fire development M M M D. Smoke generation M L-M M-H E. Altered ventilation effects M M L-M Table A-21 State of Knowledge Ranking for Scenario 1, External Ensuing Fire A-11 State of Knowledge Ranking Model Data Ability to Phenomenon Description Additional Notes & Comments Adequacy Availability Collect New Data

15. External Ensuing Fire A. Fire ignition L/H L/H M Panelists note that good models exists for fire spreading to an external source, but models are poor for HEAF-induced ignition.

B. Characterization of the fire source H H N/A C. Fire development M M H D. Smoke generation M M H

Table A-22 State of Knowledge Ranking for Scenario 1, Room Configuration Effects State of Knowledge Ranking Model Data Ability to Phenomenon Description Additional Notes & Comments Adequacy Availability Collect New Data

16. Room Configuration Effects A. Room integrity M M M-H B. Room arrangement H H N/A C. Room ventilation H H N/A Table A-23 State of Knowledge Ranking for Scenario 1, Arc Mitigation State of Knowledge Ranking A-12 Model Data Ability to Phenomenon Description Additional Notes & Comments Adequacy Availability Collect New Data

17. Arc Mitigation A. Intentional shielding L L-M H B. Electrical raceway fire barrier system L/H L/H H Panelists note that the behavior of ERFBS is known for fire scenarios, but not for HEAF scenarios.

C. Fire-retardant coating L/H L/H H Panelists note that the behavior of fire-retardant coatings is known for fire scenarios, but not for HEAF.

Table A-24 State of Knowledge Ranking for Scenario 1, Generic Cable Tray (Target) Characterization State of Knowledge Ranking Model Data Ability to Phenomenon Description Additional Notes & Comments Adequacy Availability Collect New Data

18. Generic Cable Tray (Target)

Characterization A. Characterizing target damage criteria M M H B. Target arrangement effects M M H C. Target sensitivity to damage L-M L-M M A-13

A.5 Phenomena Key Parameters Ranking Table A-25 Key Parameters Ranking for Scenario 1, Electrical Configuration Effects Importance Ranking State of Parameters Additional Notes & Comments P1 P2 P3 P4 P5 P6 Knowledge 1A. Power Supply Characterization

1. Line Voltage H H H H H H H

2. Rated Current H H M H M M H

3. Short Circuit Current H H H H H M H

4. Frequency L H L L L L H

5. Grounded/Ungrounded M H M L L L H

6. Number of Phases U U L L L U H

7. MVA Available H H H H H H H A-14

Table A-26 Key Parameters Ranking for Scenario 1, Internal Cabinet Configuration Effects Importance Ranking State of Additional Notes &

Parameters P1 P2 P3 P4 P5 P6 Knowledge Comments 2A. Cabinet Compartmentation

1. Volume M H H M H L H

2. Internal Barriers M M M M M M M-H 2B. Breaker Configuration

3. Arc Extinguishing Mechanism L/M L L L L L H

4. Total Breaker Combustible Load L/M H M H H H H

5. Extinguishing Medium L/M L M L M L H

6. Breaker Connection Material L/M M M H H H H 2C. Cabinet Combustible Loading

7. Mass H H M H H M M

8. Bundle Tightness M U L M M H M

9. Thermoset/Thermoplastic H H M H M H H

10. Cables Jacketed/Unjacketed M H H H M H H A-15 11. Combustion Properties H H M H H H M 2D. Cabinet Bus Bar Configuration

12. Bus Bar Spacing L H M H H H H

13. Bus Bar Material H H H H H H H

14. Number of Bus Bars per Phase M M L M M M H

15. Bus Bar Insulation L H L M M M H

16. Bus Bar Orientation L M M M M L H

17. Bus Bar Cross-Sectional Area L M L M L L H

18. Bus Bar Mounting/Supports L L M L L M H

Table A-27 Key Parameters Ranking for Scenario 1, External Cabinet Enclosure Configuration Effects Importance Ranking State of Parameters Additional Notes & Comments P1 P2 P3 P4 P5 P6 Knowledge 3A. Cabinet Ventilation

1. Active/Passive Ventilation L U M H H L H

2. Ventilation Opening Area H H H H H M H

3. Ventilation Deflection L U M M M L M-H

4. Ventilation Opening Location L H H M H M H

5. Presence of Active Louvres L U L U U L H 3B. Cabinet Structural Design

6. Wall Material M M M H L L H

7. Wall Thickness M M L H M M H

8. Door Latch Type X M M M M M H

9. Door Gasketing X M L M M L H

10. Wall Fastening (Welded, bolted, M M L M M M H riveted)

A-16

11. Single/Double Wall X U L M M L H 3C. Cabinet Penetrations

12. Location of Penetrations L H H H M M H

13. Penetration Sealing H M H H M H H

14. Penetration Type (Compression, L M M M M M H conduit, raceway)

15. Viewing Ports/Mechanical Openings U U M H M M H

Table A-28 Key Parameters Ranking for Scenario 1, Arc Characterization Importance Ranking State of Parameters Additional Notes & Comments P1 P2 P3 P4 P5 P6 Knowledge 4A. Arc Electrical Characterization

1. Arc Voltage H H H H H H L-M

2. Arc Current (Asymmetric) L M H H L H H

3. Arc Current (RMS) H U H H H M H

4. Arc Energy H H H H H H M

5. Arc Duration H H H H H H M

6. Arc Location H H H M H H L 4B. Arc Migration

7. Arc Migration Speed H M H H H L H

8. Type (Ionized Gas/Magnetic Field) U H M H H L L 4D. Thermal Effects of the Arc

9. Arc Heat Flux H H H H M H M

10. Bus Bar Melting H M M H M M M(H) P5 notes familiarity with a validated model for bus bar melting. The remaining panelists A-17 ranked the SOK as medium, but high pending verification of the model.

11. Oxidation of Adjacent Materials H H H H M H L 4G. Arc Ejecta

12. Ionized Gas L H H H H H L

13. Smoke L L L L M L L

14. Conductive Particulate H U M H M H L

Table A-29 Key Parameters Ranking for Scenario 1, Fire Detection Importance Ranking State of Parameters Additional Notes & Comments P1 P2 P3 P4 P5 P6 Knowledge 5A. Presence of Fire Detection

1. Detector Location H H H M H H H 5B. Detection Characteristics

2. Detector Type (Heat/smoke/optical) H L M M M L H Table A-30 Key Parameters Ranking for Scenario 1, Fire Suppression Importance Ranking State of Parameters Additional Notes & Comments P1 P2 P3 P4 P5 P6 Knowledge 6A. Presence of Fire Suppression

1. Internal/External to Cabinet H H H M M H H

2. Proximity H H H M M L H 6B. Characterizing the Suppression System

3. Actuation (Manual/Automatic) M U H M M M H A-18

4. Survivability H H H M X X L

5. Suppression Type M L H L M M H Table A-31 Key Parameters Ranking for Scenario 1, Pressure Effects Importance Ranking State of Parameters Additional Notes & Comments P1 P2 P3 P4 P5 P6 Knowledge 7B. Pressure Wave

1. Magnitude H H M H H H M-H

2. Propagation H M M H H H M

Table A-32 Key Parameters Ranking for Scenario 1, Internal Ensuing Fire Importance Ranking State of Parameters Additional Notes & Comments P1 P2 P3 P4 P5 P6 Knowledge 8A. Fire Ignition

1. Internal Cabinet Temperature M H H H H H L 8B. Characterization of Fire Source

2. Heat Release Rate H H H H H H M-H

3. Heat Release Rate per unit Area M L H M H L M 8C. Fire Development

4. Time to Peak M U H M H M M-H

5. Fire Growth Rate M H H H H H M-H 8E. Altered Ventilation Effects

6. Altered Internal Vent Properties M H H H H M M

7. Altered Enclosure Vent Properties M H H H H H M A-19 Table A-33 Key Parameters Ranking for Scenario 1, External Ensuing Fire Importance Ranking State of Parameters Additional Notes & Comments P1 P2 P3 P4 P5 P6 Knowledge 9B. Characterization of Fire Source

1. Heat Release Rate H H H H H H H

2. Heat Release Rate per unit Area M L H M H L H

3. Fire Geometry L M M H H H H 9C. Fire Development

4. Time to Peak H L H M H M H

5. Fire Growth Rate H H H H H H H

Table A-34 Key Parameters Ranking for Scenario 1, Room Configuration Effects Importance Ranking State of Parameters Additional Notes & Comments P1 P2 P3 P4 P5 P6 Knowledge 10A. Room Integrity

1. Fire Dampers M U M H H X L

2. Penetrations M U M H H X L 10B. Room Arrangement

3. Dimensions M U M H H H H

4. Contents M U L H H L H

5. Construction Material M U L M M L H

6. Location of Source within Room L L H M M H H 10C. Room Ventilation

7. Type (Active/Passive) H M L H H M H Table A-35 Key Parameters Ranking for Scenario 1, Arc Mitigation Importance Ranking State of A-20 Parameters Additional Notes & Comments P1 P2 P3 P4 P5 P6 Knowledge 11A. Intentional Shielding

1. Thickness H M M H H H H

2. Material H H H H H H H 11B. Electrical Raceway Fire Barrier System

3. Rating M H M U H M H 11C. Fire-Retardant Coating Effects

4. Thickness H U H H H H M

5. Material M M H H H H H

6. Age L U M U H H H

Table A-36 Key Parameters Ranking for Scenario 1, Generic Cable Tray (Target) Characterization Importance Ranking State of Parameters Additional Notes & Comments P1 P2 P3 P4 P5 P6 Knowledge 12A. Target Damage

1. Damage Criteria L H H H H H L/H Panelists note that damage criteria, critical heat flux, and critical temperature are known for fire scenarios, but not HEAF scenarios.

2. Critical Heat Flux H H H H H H L/H

3. Critical Temperature H H H H H H L/H

4. Exposure Time H H H H H H M 12B. Target Arrangement Effects

5. Distance to HEAF Source H H H H H H H

6. Orientation Relative to HEAF H H H H H H H Source 12C. Target Sensitivity A-21 7. Sensitivity to Shock L U M M M L M

8. Sensitivity to Heat H M H H H H H

9. Sensitivity to EMI/RFI L L M L U L M

10. Sensitivity to Light L L L L L L M

11. Sensitivity to Ionized Gas L M M L H M M

12. Sensitivity to Conductive Particulate L L M L L L M

13. Sensitivity to Smoke L L L L L L M

A.6 Scenario 1 Phenomena Calculated Rank Values and Grouping Phenomenon Importance State of Knowledge Rank Grouping Arc Mitigation (11) 4.2 2.4 1.75 Level 1 Internal Ensuing Fire (8) 4.4 2.7 1.63 Level 1 Target Characterization (12) 4.9 3.2 1.53 Level 1 Arc Characterization (4) 3.6 2.7 1.33 Level 1 External Ensuing Fire (9) 4.3 3.4 1.27 Level 2 Pressure Effects (7) 3.5 3.2 1.09 Level 2 Electrical Configuration Effects (1) 5 5 1.00 Level 2 Internal Cabinet Configuration Effects (2) 3.8 4.2 0.91 Level 2 External Cabinet Enclosure Effects (3) 4 5 0.80 Level 3 Suppression Effects (6) 4 5 0.80 Level 3 Room Configuration (10) 3.4 4.3 0.79 Level 3 Fire Detection (5) 2.5 5 0.50 Level 3 33rd percentile for Rank: 0.87 A-22 66th percentile for Rank: 1.28

APPENDIX B PIRT HEAF SCENARIO 2 B.1 HEAF Scenario 2 B.1.1 Scenario Description A high energy arcing fault occurs in a non-segregated bus duct supplying power to Train B systems. A switchgear supplying power to redundant Train A systems is located nearby.

B.1.2 Figure of Merit Determining the extent of damage to the switchgear.

B.2 List of Identified Phenomena Phenomenon 1: Electrical Configuration Effects A. Power supply characterization B. Electrical protection coordination 10 Phenomenon 2: Internal Bus Duct Configuration Effects A. Duct compartmentation B. Duct combustible loading 10 Electrical protection coordination refers to the coordination and configuration of upstream breakers with the objective of minimizing the duration of an electrical fault B-1

C. Duct bus bar configuration Phenomenon 3: External Bus Duct Housing Configuration Effects A. Duct ventilation B. Duct structural design C. Duct mounting Phenomenon 4: Arc Characterization A. Arc electrical characterization B. Arc migration C. Arc breeching of enclosure D. Thermal effects of the arc E. Magnetic effects of the arc 11 F. Electromagnetic Interference G. Arc ejecta Phenomenon 5: Fire Detection 12 A. Presence of fire detection B. Characterizing the detection system Phenomenon 6: Fire Suppression A. Presence of fire suppression B. Characterizing the suppression system Phenomenon 7: Pressure Effects A. Projectile/missile damage B. Pressure wave Phenomenon 8: Internal Ensuring Fire 13 A. Fire ignition B. Characterization of the fire source C. Fire development 11 Magnetic effects refers to the applied forces on ferrous materials by a magnetic field that can induce bending and deformation, as opposed to the interference that it may cause. Interference is addressed separately.

12 Fire detection was assumed not to automatically trigger fire suppression. Fire suppression is considered as a separate phenomenon.

13 Internal ensuing fire refers to a sustained fire that occurs inside the source cabinet.

B-2

D. Smoke generation E. Altered ventilation effects14 Phenomenon 9: External Ensuing Fire A. Fire ignition from molten metal B. Fire ignition from arc heat C. Characterization of the fire source D. Fire development E. Smoke generation Phenomenon 10: Room Configuration Effects A. Room integrity 15 B. Room arrangement C. Room ventilation Phenomenon 11: Arc Mitigation A. Intentional shielding B. Electrical raceway fire barrier system effects16 C. Fire-retardant coating effects 17 D. Cable tray design Phenomenon 12: Generic Enclosure (Target) Characterization A. Characterizing target damage criteria B. Target arrangement effects C. Target sensitivity to damage 14 Altered ventilation effects refers to the manner in which changes to the cabinets structure and integrity as a result of the HEAF impact the ventilation paths for an internal ensuing fire.

15 Room integrity refers to the pressure boundary of the room in which the HEAF occurs, and how its response to the increased pressure will impact target damage. For example, consider whether blowing open a fire door or damper will affect target damage.

16 Electrical raceway fire barrier systems refers to a wrap or enclosure that is installed around electrical cable trays with the intent of mitigating fire damage.

17 Fire-retardant coating refers to a permanently installed coating or spray that is applied directly to the cables in a cable tray.

B-3

B.3 Phenomena Importance Rankings Table B-1 Importance Ranking for Scenario 2, Electrical Configuration Effects Importance Ranking Phenomenon Description Additional Notes & Comments P1 P2 P3 P4 P5 P6

1. Electrical Configuration Effects A. Power supply characterization H H H H H H B. Electrical protection coordination H H H H H H Table B-2 Importance Ranking for Scenario 2, Internal Bus Duct Configuration Effects Importance Ranking Phenomenon Description Additional Notes & Comments P1 P2 P3 P4 P5 P6

2. Internal Bus Duct Configuration Effects A. Duct compartmentation M M M H M M B-4 B. Duct combustible loading L L L L H L C. Duct bus bar configuration M M M M M L Table B-3 Importance Ranking for Scenario 2, External Bus Duct Housing Configuration Effects Importance Ranking Phenomenon Description Additional Notes & Comments P1 P2 P3 P4 P5 P6

3. External Bus Duct Housing Configuration Effects A. Duct ventilation L L L L L L B. Duct structural design H H H H H H C. Duct mounting H L L L L L

Table B-4 Importance Ranking for Scenario 2, Arc Characterization Importance Ranking Phenomenon Description Additional Notes & Comments P1 P2 P3 P4 P5 P6

4. Arc Characterization A. Arc electrical characterization H H H H H H P5 notes that arc voltage will be higher than that of the switchgear because the arc will move along the duct.

B. Arc migration H L L M M L C. Arc breeching of enclosure H H H H H H Panelists note that HEAFs inside bus ducts always breech the housing, but it remains an important phenomenon. Panelists also note that a HEAF may not have been reported as such if breeching did not occur.

D. Thermal effects of the arc H H H H H H E. Magnetic effects of the arc L L L L L L F. Electromagnetic interference L L L L L L G. Arc ejecta M-H M-H M-H M-H M-H M-H B-5 Table B-5 Importance Ranking for Scenario 2, Fire Detection Importance Ranking Phenomenon Description Additional Notes & Comments P1 P2 P3 P4 P5 P6

5. Fire Detection A. Presence of fire detection H M M L L M B. Characterizing fire detection system M L L L L L

Table B-6 Importance Ranking for Scenario 2, Fire Suppression Importance Ranking Phenomenon Description Additional Notes & Comments P1 P2 P3 P4 P5 P6

6. Fire Suppression A. Presence of fire suppression H L L L L L B. Fire suppression effects H L L L L L Table B-7 Importance Ranking for Scenario 2, Pressure Effects Importance Ranking Phenomenon Description Additional Notes & Comments P1 P2 P3 P4 P5 P6

7. Pressure Effects B-6 A. Projectile/missile damage H L L L L L B. Pressure wave L L L M M M

Table B-8 Importance Ranking for Scenario 2, Internal Ensuing Fire Importance Ranking Phenomenon Description Additional Notes & Comments P1 P2 P3 P4 P5 P6

8. Internal Ensuing Fire F. Fire ignition L L L L L L G. Characterization of the fire source L-H L L L L L P1 notes that the importance of characterizing the fire source depends on whether there is aluminum present.

H. Fire development L L L L L L I. Smoke generation L L L L L L P5 raised the possibility that the duct behaves like a chimney and exhausts smoke in a different location.

P6 states that in all known cases, the duct breeches, exhausted most of the smoke near the arc origin.

J. Altered ventilation effects L L L L L L B-7 Table B-9 Importance Ranking for Scenario 2, External Ensuing Fire Importance Ranking Phenomenon Description Additional Notes & Comments P1 P2 P3 P4 P5 P6

9. External Ensuing Fire A. Fire ignition from molten metal M-H L L L L L Panelists note that secondary ignition due to molten metal has not been observed in testing. Transient combustibles may be more likely to ignite than fixed SSCs, and need to be considered on a situational basis.

B. Fire ignition from arc heat H M M M M H C. Characterization of the fire source H H H H H H D. Fire development H H H H H H E. Smoke generation L L L L L L

Table B-10 Importance Ranking for Scenario 2, Room Configuration Effects Importance Ranking Phenomenon Description Additional Notes & Comments P1 P2 P3 P4 P5 P6

10. Room Configuration Effects A. Room integrity L L L L L M P5 notes that a HEAF occurring in a duct at the point where it penetrates a wall may bypass room integrity features.

B. Room arrangement M M M L L M C. Room ventilation L H L L L M Table B-11 Importance Ranking for Scenario 2, Arc Mitigation Importance Ranking Phenomenon Description Additional Notes & Comments P1 P2 P3 P4 P5 P6

11. Arc Mitigation A. Intentional shielding H H H H H H B-8 B. Electrical raceway fire barrier system H H H H H H C. Fire-retardant coating M M H H H H D. Cable tray design M M M M M M Table B-12 Importance Ranking for Scenario 2, Generic Enclosure (Target) Characterization Importance Ranking Phenomenon Description Additional Notes & Comments P1 P2 P3 P4 P5 P6

12. Generic Enclosure (Target)

Characterization A. Characterizing target damage criteria H H H H H H B. Target arrangement effects H H H H H H C. Target sensitivity to damage H H H H H H

B.4 Phenomena State of Knowledge Rankings Table B-13 State of Knowledge Ranking for Scenario 2, Electrical Configuration Effects State of Knowledge Ranking Model Data Ability to Phenomenon Description Additional Notes & Comments Adequacy Availability Collect New Data

1. Electrical Configuration Effects A. Power supply characterization H H N/A B. Electrical protection coordination H H N/A Table B-14 State of Knowledge Ranking for Scenario 2, Internal Bus Duct Configuration Effects B-9 State of Knowledge Ranking Model Data Ability to Phenomenon Description Additional Notes & Comments Adequacy Availability Collect New Data

2. Internal Bus Duct Configuration Effects A. Duct compartmentation M M M-H B. Duct combustible loading H H N/A C. Duct bus bar configuration M M M

Table B-15 State of Knowledge Ranking for Scenario 2, External Bus Duct Housing Configuration Effects State of Knowledge Ranking Model Data Ability to Phenomenon Description Additional Notes & Comments Adequacy Availability Collect New Data

3. External Bus Duct Housing Configuration Effects A. Duct ventilation L H N/A P6 notes that any model for duct ventilation is likely useless, as the duct housing is breeched by the arc in all known cases.

B. Duct structural design M M H C. Duct mounting L L L Table B-16 State of Knowledge Ranking for Scenario 2, Arc Characterization State of Knowledge Ranking B-10 Model Data Ability to Phenomenon Description Additional Notes & Comments Adequacy Availability Collect New Data

4. Arc Characterization A. Arc electrical characterization M-H H N/A B. Arc migration M-H M-H N/A C. Arc breeching of enclosure L L-M L-M Panelists note that the only data that exists is observational. No quality data set exists.

D. Thermal effects of the arc M M M E. Magnetic effects of the arc M-H M-H N/A F. Electromagnetic interference L L M-H G. Arc ejecta L L L Panelists note again that only minimal observational data is available, and that is extremely difficult to separate the ejecta into its constituents.

Table B-17 State of Knowledge Ranking for Scenario 2, Fire Detection State of Knowledge Ranking Model Data Ability to Phenomenon Description Additional Notes & Comments Adequacy Availability Collect New Data

5. Fire Detection A. Presence of fire detection H H N/A P2 notes that models for detection are good, but there is no specific data for HEAF conditionsthe detectors may be destroyed.

B. Characterizing fire detection system H H N/A Table B-18 State of Knowledge Ranking for Scenario 2, Fire Suppression State of Knowledge Ranking Model Data Ability to Phenomenon Description Additional Notes & Comments Adequacy Availability Collect New Data B-11 6. Fire Suppression A. Presence of fire suppression H H N/A B. Fire suppression effects M M H Table B-19 State of Knowledge Ranking for Scenario 2, Pressure Effects State of Knowledge Ranking Model Data Ability to Phenomenon Description Additional Notes & Comments Adequacy Availability Collect New Data

7. Pressure Effects A. Projectile/missile damage M-H M-H N/A P5 notes that predicting missile damage is not very difficult if the identity of the projectile is known. Identifying it is more challenging.

B. Pressure wave H H N/A

Table B-20 State of Knowledge Ranking for Scenario 2, Internal Ensuing Fire State of Knowledge Ranking Model Data Ability to Phenomenon Description Additional Notes & Comments Adequacy Availability Collect New Data

8. Internal Ensuing Fire A. Fire ignition M L M B. Characterization of the fire source H M-H N/A C. Fire development M-H M M-H P6 notes that observational data indicates no fire development in bus ducts.

D. Smoke generation M-H M-H N/A E. Altered ventilation effects M-H M-H N/A Table B-21 State of Knowledge Ranking for Scenario 2, External Ensuing Fire B-12 State of Knowledge Ranking Model Data Ability to Phenomenon Description Additional Notes & Comments Adequacy Availability Collect New Data

9. External Ensuing Fire A. Fire ignition from molten metal M M H B. Fire ignition from arc heat M L-M H P1 notes that exposure time concerns make modeling less than adequate.

C. Characterization of the fire source M M-H H P1 notes that cabinets are complex fuel sourcescharacterization is not perfect.

D. Fire development M M-H H E. Smoke generation H H N/A

Table B-22 State of Knowledge Ranking for Scenario 2, Room Configuration Effects State of Knowledge Ranking Model Data Ability to Phenomenon Description Additional Notes & Comments Adequacy Availability Collect New Data

10. Room Configuration Effects A. Room integrity M M H P1 and P6 note that there are some models for room integrity, but not specifically applicable to HEAF.

B. Room arrangement H H N/A C. Room ventilation H H N/A Table B-23 State of Knowledge Ranking for Scenario 2, Arc Mitigation B-13 State of Knowledge Ranking Model Data Ability to Phenomenon Description Additional Notes & Comments Adequacy Availability Collect New Data

11. Arc Mitigation A. Intentional shielding M M M-H Panelists note that some small-scale experiments exist, and solid mechanics tools can possible be applied, but no directly applicable model currently exists.

B. Electrical raceway fire barrier system M M H C. Fire-retardant coating M M H P4 notes that the original status of the cables should be consideredi.e. if the coating was applied as a remedial measure, this is important.

D. Cable tray design M M H

Table B-24 State of Knowledge Ranking for Scenario 2, Generic Enclosure (Target) Characterization State of Knowledge Ranking Model Data Ability to Phenomenon Description Additional Notes & Comments Adequacy Availability Collect New Data

12. Generic Cable Tray (Target)

Characterization A. Characterizing target damage criteria M M H B. Target arrangement effects M M H C. Target sensitivity to damage L-M L-M M Panelists noted models exist that can evaluate the effects of target sensitivity to a few items, but others are neglected entirely.

B-14

B.5 Phenomena Key Parameters Ranking Table B-25 Key Parameters Ranking for Scenario 2, Electrical Configuration Effects Importance Ranking State of Parameters Additional Notes & Comments P1 P2 P3 P4 P5 P6 Knowledge 1A. Power Supply Characterization

1. Line Voltage H H H H H H H

2. Rated Current H H H H M H H

3. Short Circuit Current H H H H H H H

4. Frequency L L L L L L H

5. Grounded/Ungrounded M L L L L L H

6. MVA Available H H H H H H H B-15

Table B-26 Key Parameters Ranking for Scenario 2, Internal Bus Duct Configuration Effects Importance Ranking State of Additional Notes &

Parameters P1 P2 P3 P4 P5 P6 Knowledge Comments 2A. Duct Compartmentation

1. Dimensions M M M M M M H P6 notes that the dimensions of the duct arent particularly important as the duct is likely to be breeched.

2. Internal Partitions H H H H H H H 2B. Duct Combustible Loading

3. Combustible Mass H H H H H M H

4. Combustion Properties H H H H H M H 2C. Duct Bus Bar Configuration

5. Bus Bar Spacing H H H H H H H

6. Bus Bar Material H H H H H H H

7. Bus Bar Insulation M M M M M M L B-16 8. Bus Bar Orientation M M M M M M H

9. Bus Bar Cross-Sectional Area M L- M M M M H P2 notes that cross-sectional M area is not important for power supply, but can become important when considering bus bar oxidation.

10. Bus Bar Joints H H H H H H H Panelists ranked this high because the joints are the likely arc locations.

11. Bus Bar Mounting/Supports L L L L L L H

Table B-27 Key Parameters Ranking for Scenario 2, External Bus Duct Housing Configuration Effects Importance Ranking State of Parameters Additional Notes & Comments P1 P2 P3 P4 P5 P6 Knowledge 3A. Duct Ventilation

1. Ventilation Opening Area H H H H H H H

2. Ventilation Opening Location M M M M M M H 3B. Duct Structural Design

3. Wall Material H H H H H H H

4. Wall Thickness H H H H H H H

5. Access Ports M M M M M M H 3C. Duct Mounting

6. Robustness H L L M L L L

7. Orientation H H H M H H H B-17

Table B-28 Key Parameters Ranking for Scenario 2, Arc Characterization Importance Ranking State of Parameters Additional Notes & Comments P1 P2 P3 P4 P5 P6 Knowledge 4A. Arc Electrical Characterization

1. Arc Voltage H H H H H H M

2. Arc Current (Asymmetric) L H H H L H H

3. Arc Current (RMS) H H H H H H H

4. Arc Energy H H H H H H M-H

5. Arc Duration H H H H H H M

6. Arc Location M L L L L L L 4B. Arc Migration

7. Arc Migration Speed H H H H H H M-H

8. Type (Ionized Gas/Magnetic Field) U H H H H H L-M 4D. Thermal Effects of the Arc

9. Arc Heat Flux H H H H H H M

10. Bus Bar Melting H H H H H H M B-18 11. Oxidation of Adjacent Materials H H H H H H L 4G. Arc Ejecta

12. Ionized Gas L H H H H H L

13. Smoke L L L L L L L

14. Conductive Particulate H H H H H H L

Table B-29 Key Parameters Ranking for Scenario 2, Fire Detection Importance Ranking State of Parameters Additional Notes & Comments P1 P2 P3 P4 P5 P6 Knowledge 5B. Detection Characteristics

1. Detector Type (Heat/smoke/optical) H H H H H H H Table B-30 Key Parameters Ranking for Scenario 2, Fire Suppression Importance Ranking State of Parameters Additional Notes & Comments P1 P2 P3 P4 P5 P6 Knowledge 6A. Presence of Fire Suppression

1. Proximity H H H H H H H 6B. Characterizing the Suppression System

2. Actuation (Manual/Automatic) H M M M M M H B-19

3. Survivability M H H H H H L

4. Suppression Type M H H H H H M-H Table B-31 Key Parameters Ranking for Scenario 2, Pressure Effects Importance Ranking State of Parameters Additional Notes & Comments P1 P2 P3 P4 P5 P6 Knowledge 7B. Pressure Wave

1. Magnitude H H H H H H M-H

2. Propagation H H H H H H M

Table B-32 Key Parameters Ranking for Scenario 2, Internal Ensuing Fire Importance Ranking State of Parameters Additional Notes & Comments P1 P2 P3 P4 P5 P6 Knowledge 8A. Fire Ignition

1. Internal Duct Temperature H H H H H H L-M 8B. Characterization of Fire Source

2. Heat Release Rate H H H H H H M-H

3. Heat Release Rate per unit Area M M M M M L M 8C. Fire Development

4. Time to Peak H H H H H H M-H

5. Fire Growth Rate H H H H H H M-H 8E. Altered Ventilation Effects

6. Altered Duct Vent Properties M H M M M M M B-20 Table B-33 Key Parameters Ranking for Scenario 2, External Ensuing Fire Importance Ranking State of Parameters Additional Notes & Comments P1 P2 P3 P4 P5 P6 Knowledge 9B. Characterization of Fire Source

1. Heat Release Rate H H H H H H H

2. Heat Release Rate per unit Area M M M M M M M-H

3. Fire Geometry L M M L H M M-H 9C. Fire Development

4. Time to Peak H H H H H H M-H

5. Fire Growth Rate H H H H H H M-H

Table B-34 Key Parameters Ranking for Scenario 2, Room Configuration Effects Importance Ranking State of Parameters Additional Notes & Comments P1 P2 P3 P4 P5 P6 Knowledge 10A. Room Integrity

1. Fire Dampers M L L L L L L-M

2. Penetrations M L L L L L L-M 10B. Room Arrangement

3. Dimensions M M M M M M H

4. Contents H M L M L L H

5. Construction Material L M M M M M H

6. Location of Source within Room L M L L L L M-H 10C. Room Ventilation

7. Type (Active/Passive) H H H H H H M-H Table B-35 Key Parameters Ranking for Scenario 2, Arc Mitigation B-21 Importance Ranking State of Parameters Additional Notes & Comments P1 P2 P3 P4 P5 P6 Knowledge 11A. Intentional Shielding

1. Thickness H H H H H H H

2. Material H H H H H H H 11B. Electrical Raceway Fire Barrier System

3. Rating U U U U U U H 11C. Fire-Retardant Coating Effects

4. Thickness H H H H H H M

5. Material H H H H H H H

6. Age L U U U U L M-H

Table B-36 Key Parameters Ranking for Scenario 2, Generic Enclosure (Target) Characterization Importance Ranking State of Additional Notes &

Parameters P1 P2 P3 P4 P5 P6 Knowledge Comments 12A. Target Damage

1. Damage Criteria H H H H H H M

2. Critical Heat Flux H H H H H H M

3. Critical Temperature H H H H H H M

4. Exposure Time H H H H H H M 12B. Target Arrangement Effects

5. Distance to HEAF Source H H H H H H M-H

6. Orientation Relative to HEAF H H H M H M M-H Source 12C. Target Sensitivity

7. Sensitivity to Shock M M/H M H M-H M M

8. Sensitivity to Heat H H H H H H M-H

9. Sensitivity to EMI/RFI M M M M U M L-M B-22 10. Sensitivity to Light L L L L L L L-M

11. Sensitivity to Ionized Gas L L/H L/H L L/H L/M M The split responses indicate differences depending on whether the target considered is the cable tray or the cabinet itself.

12. Sensitivity to Conductive Particulate L/M L/H L/H L L/H L/M M The split responses indicate differences depending on whether the target considered is the cable tray or the cabinet itself.

13. Sensitivity to Smoke H L L L L L L-M

B.6 Scenario 2 Phenomena Calculated Rank Values and Grouping Table B-37 Summary of Scenario 2 Average Rankings and Grouping Phenomenon Importance State of Knowledge Rank Grouping Target Characterization (12) 5 3.1 1.61 Level 1 Arc Characterization (4) 3.3 2.6 1.27 Level 1 Arc Mitigation (11) 4.3 3.6 1.19 Level 1 External Duct Housing Configuration Effects (3) 2.6 2.5 1.04 Level 1 Electrical Configuration Effects (1) 5 5 1.00 Level 2 External Ensuing Fire (9) 3.2 3.9 0.82 Level 2 Internal Bus Duct Configuration Effects (2) 2.6 3.6 0.72 Level 2 Room Configuration (10) 1.9 4.4 0.43 Level 2 Suppression Effects (6) 1.7 4.2 0.40 Level 3 Fire Detection (5) 2 5 0.40 Level 3 Pressure Effects (7) 1.8 4.5 0.40 Level 3 Internal Ensuing Fire (8) 1.1 3.6 0.31 Level 3 B-23 33rd percentile for Rank: 0.42 66th percentile for Rank: 1.01

APPENDIX C PIRT HEAF SCENARIO 3 C.1 HEAF Scenario 3 C.1.1 Scenario Description A high energy arcing fault occurs in a low/medium voltage switchgear supplying power to Train A systems in a bank of similar switchgears. The cables from an overhead cable tray enter the cabinets from the top (air-drop configuration). A cable tray carrying cables that supply power to Train B systems is located nearby.

C.1.2 Figure of Merit Determining the extent of damage to the adjacent switchgears and the cables in Train B cable tray.

C.2 List of Identified Phenomena Phenomenon 1: Electrical Configuration Effects A. Power supply characterization B. Electrical protection coordination 18 Phenomenon 2: Internal Cabinet Configuration Effects A. Cabinet compartmentation B. Breaker configuration 18 Electrical protection coordination refers to the coordination and configuration of upstream breakers with the objective of minimizing the duration of an electrical fault C-1

C. Cabinet combustible loading D. Cabinet bus bar configuration Phenomenon 3: External Cabinet Enclosure Configuration Effects A. Cabinet ventilation B. Cabinet structural design C. Cabinet penetrations Phenomenon 4: Cabinet Lineup Effects A. Cabinet-cabinet barrier effects B. Cabinet position within bank effects19 C. Cabinet-cabinet penetration effects Phenomenon 5: Arc Characterization A. Arc electrical characterization B. Arc migration C. Arc breeching of enclosure D. Thermal effects of the arc E. Magnetic effects of the arc 20 F. Electromagnetic interference G. Arc ejecta H. Plasma characterization Phenomenon 6: Fire Detection 21 A. Presence of fire detection B. Characterizing the detection system Phenomenon 7: Fire Suppression A. Presence of fire suppression B. Characterizing the suppression system 19 The effects of the cabinet position within the bank does not refer to the location of targets around the source cabinet. This refers to the manner in which the position (middle of the bank, end of the bank) affects the extent of damage to whatever targets are present.

20 Magnetic effects refers to the applied forces on ferrous materials by a magnetic field that can induce bending and deformation, as opposed to the interference that it may cause. Interference is addressed separately.

21 Fire detection was assumed not to automatically trigger fire suppression. Fire suppression is considered as a separate phenomenon.

C-2

Phenomenon 8: Pressure Effects A. Projectile/missile damage B. Pressure wave Phenomenon 9: Internal Ensuring Fire 22 A. Fire ignition B. Characterization of the fire source C. Fire development D. Smoke generation E. Altered ventilation effects23 Phenomenon 10: External Ensuing Fire A. Fire ignition B. Characterization of the fire source C. Fire development D. Cabinet-to-cabinet propagation E. Smoke generation Phenomenon 11: Room Configuration Effects A. Room integrity 24 B. Room arrangement C. Room ventilation Phenomenon 12: Arc Mitigation A. Intentional shielding B. Electrical raceway fire barrier system effects25 C. Fire-retardant coating effects 26 22 Internal ensuing fire refers to a sustained fire that occurs inside the source cabinet.

23 Altered ventilation effects refers to the manner in which changes to the cabinets structure and integrity as a result of the HEAF impact the ventilation paths for an internal ensuing fire.

24 Room integrity refers to the pressure boundary of the room in which the HEAF occurs, and how its response to the increased pressure will impact target damage. For example, consider whether blowing open a fire door or damper will affect target damage.

25 Electrical raceway fire barrier systems refers to a wrap or enclosure that is installed around electrical cable trays with the intent of mitigating fire damage.

26 Fire-retardant coating refers to a permanently installed coating or spray that is applied directly to the cables in a cable tray.

C-3

Phenomenon 13: Generic Enclosure (Target) Characterization A. Characterizing target damage criteria B. Target sensitivity to damage Phenomenon 14: Generic Cable Tray (Target) Characterization A. Characterizing target damage criteria B. Target arrangement effects C. Target sensitivity to damage C-4

C.3 Phenomena Importance Rankings Table C-1 Importance Ranking for Scenario 3, Electrical Configuration Effects Importance Ranking Phenomenon Description Additional Notes & Comments P1 P2 P3 P4 P5 P6

1. Electrical Configuration Effects A. Power supply characterization H H H H H H B. Electrical protection coordination H H H H H H Table C-2 Importance Ranking for Scenario 3, Internal Cabinet Configuration Effects Importance Ranking Phenomenon Description Additional Notes & Comments P1 P2 P3 P4 P5 P6

2. Internal Cabinet Configuration Effects A. Cabinet compartmentation H M M M M M C-5 B. Breaker configuration M M M M M M C. Cabinet combustible loading H H H H H H D. Cabinet bus bar configuration M M M H M M Table C-3 Importance Ranking for Scenario 3, External Cabinet Enclosure Configuration Effects Importance Ranking Phenomenon Description Additional Notes & Comments P1 P2 P3 P4 P5 P6

3. External Cabinet Enclosure Configuration Effects A. Cabinet ventilation M M M M H M B. Cabinet structural design H M M M M L C. Cabinet penetrations H M M M M M

Table C-4 Importance Ranking for Scenario 3, Cabinet Lineup Effects Importance Ranking Phenomenon Description Additional Notes & Comments P1 P2 P3 P4 P5 P6

4. Cabinet Lineup Effects A. Cabinet-cabinet barrier effects H H H H H H B. Cabinet-position within bank effects H M M M M M C. Cabinet-cabinet penetration effects H H H H H H Table C-5 Importance Ranking for Scenario 3, Arc Characterization Importance Ranking Phenomenon Description Additional Notes & Comments P1 P2 P3 P4 P5 P6

5. Arc Characterization A. Arc electrical characterization H H H H H H B. Arc migration H M M M M M P6 notes that the migration of the arc in this C-6 scenario is more important than scenarios 1 &

2.

C. Arc breeching of enclosure H H H H H H P5 notes that arc breeching in a cabinet lineup increases the probability of a secondary arc in an adjacent cabinet.

D. Thermal effects of the arc H H H H H H E. Magnetic effects of the arc L M M H U M F. Electromagnetic interference H L L H L L P4 notes that digital upgrades have increased equipment susceptibility to EMI/RFI.

G. Arc ejecta H H H H H H H. Plasma characterization M M M M M M P6 notes that the properties of the arc plasma is inherently tied to the characteristics of the arc, and as such, may not need to be considered separately.

Table C-6 Importance Ranking for Scenario 3, Fire Detection Importance Ranking Phenomenon Description Additional Notes & Comments P1 P2 P3 P4 P5 P6

6. Fire Detection A. Presence of fire detection H L L L L L B. Characterizing fire detection system L L L L L L Table C-7 Importance Ranking for Scenario 3, Fire Suppression Importance Ranking Phenomenon Description Additional Notes & Comments P1 P2 P3 P4 P5 P6

7. Fire Suppression A. Presence of fire suppression M M M M M M P6 notes that there is in increased risk for fire spreading in a cabinet lineup.

B. Fire suppression effects M M M M M M C-7 Table C-8 Importance Ranking for Scenario 3, Pressure Effects Importance Ranking Phenomenon Description Additional Notes & Comments P1 P2 P3 P4 P5 P6

8. Pressure Effects A. Projectile/missile damage H H H H H H B. Pressure wave H H H H H H

Table C-9 Importance Ranking for Scenario 3, Internal Ensuing Fire Importance Ranking Phenomenon Description Additional Notes & Comments P1 P2 P3 P4 P5 P6

9. Internal Ensuing Fire A. Fire ignition H H H H H H B. Characterization of the fire source H H H H H H C. Fire development H H H H H H D. Smoke generation H L L L L L E. Altered ventilation effects H H H H H H Table C-10 Importance Ranking for Scenario 3, External Ensuing Fire Importance Ranking Phenomenon Description Additional Notes & Comments P1 P2 P3 P4 P5 P6 C-8 10. External Ensuing Fire A. Fire ignition L-H H H H H H P1 notes that this depends highly on the type of target cabinets in the lineup.

B. Characterization of the fire source H H H H H H C. Fire development H H H H H H D. Cabinet-cabinet propagation H H H H H H Panelists note that this is the singularly outstanding risk in this scenario, and rate this phenomenon of the highest priority.

E. Smoke generation L L L L L L

Table C-11 Importance Ranking for Scenario 3, Room Configuration Effects Importance Ranking Phenomenon Description Additional Notes & Comments P1 P2 P3 P4 P5 P6

11. Room Configuration Effects A. Room integrity M M L L M M P2 notes that this is more important for the cable tray target.

B. Room arrangement M M M M M M C. Room ventilation L H M M M M Table C-12 Importance Ranking for Scenario 3, Arc Mitigation Importance Ranking Phenomenon Description Additional Notes & Comments P1 P2 P3 P4 P5 P6

12. Arc Mitigation A. Intentional shielding H M H H H H C-9 B. Electrical raceway fire barrier system M M H H H M C. Fire-retardant coating L M M M M L Table C-13 Importance Ranking for Scenario 3, Generic Enclosure (Target) Characterization Importance Ranking Phenomenon Description Additional Notes & Comments P1 P2 P3 P4 P5 P6

13. Generic Enclosure (Target)

Characterization A. Characterizing target damage criteria H H H H H H B. Target sensitivity to damage H H H H H H

Table C-14 Importance Ranking for Scenario 3, Generic Cable Tray (Target) Characterization Importance Ranking Phenomenon Description Additional Notes & Comments P1 P2 P3 P4 P5 P6

14. Generic Cable Tray (Target)

Characterization A. Characterizing target damage criteria H H H H H H B. Target arrangement effects M H H H H H C. Target sensitivity to damage H H H H H H C-10

C.4 Phenomena State of Knowledge Rankings Table C-15: State of Knowledge Ranking for Scenario 3, Electrical Configuration Effects State of Knowledge Ranking Model Data Ability to Phenomenon Description Additional Notes & Comments Adequacy Availability Collect New Data

1. Electrical Configuration Effects A. Power supply characterization H H N/A B. Electrical protection coordination H H N/A Table C-16 State of Knowledge Ranking for Scenario 3, Internal Cabinet Configuration Effects C-11 State of Knowledge Ranking Model Data Ability to Phenomenon Description Additional Notes & Comments Adequacy Availability Collect New Data

2. Internal Cabinet Configuration Effects A. Cabinet compartmentation H H N/A B. Breaker configuration M-H M H C. Cabinet combustible loading M-H M M D. Cabinet bus bar configuration M-H M-H N/A

Table C-17 State of Knowledge Ranking for Scenario 3, External Cabinet Enclosure Configuration Effects State of Knowledge Ranking Model Data Ability to Phenomenon Description Additional Notes & Comments Adequacy Availability Collect New Data

3. External Cabinet Enclosure Configuration Effects A. Cabinet ventilation M-H M-H N/A B. Cabinet structural design M-H M-H N/A C. Cabinet penetrations M-H H N/A C-12 Table C-18 State of Knowledge Ranking for Scenario 3, Cabinet Lineup Effects State of Knowledge Ranking Model Data Ability to Phenomenon Description Additional Notes & Comments Adequacy Availability Collect New Data

4. Cabinet Lineup Effects A. Cabinet-cabinet barrier effects M M M-H B. Cabinet-cabinet penetration effects M M M-H C. Cabinet position within bank effects M M-H N/A

Table C-19 State of Knowledge Ranking for Scenario 3, Arc Characterization State of Knowledge Ranking Model Data Ability to Phenomenon Description Additional Notes & Comments Adequacy Availability Collect New Data

5. Arc Characterization A. Arc electrical characterization L-M M M B. Arc migration M L-M L-M C. Arc breeching of enclosure L L L D. Thermal effects of the arc M M M-H E. Magnetic effects of the arc L-M L-M M F. Electromagnetic interference L L L-M G. Arc ejecta L L L H. Plasma characterization L-M L-M M C-13 Table C-20 State of Knowledge Ranking for Scenario 3, Fire Detection State of Knowledge Ranking Model Data Ability to Phenomenon Description Additional Notes & Comments Adequacy Availability Collect New Data

6. Fire Detection A. Presence of fire detection H H N/A B. Characterizing fire detection system H H N/A

Table C-21 State of Knowledge Ranking for Scenario 3, Fire Suppression State of Knowledge Ranking Model Data Ability to Phenomenon Description Additional Notes & Comments Adequacy Availability Collect New Data

7. Fire Suppression A. Presence of fire suppression H H N/A B. Fire suppression effects M-H M-H N/A Table C-22 State of Knowledge Ranking for Scenario 3, Pressure Effects C-14 State of Knowledge Ranking Model Data Ability to Phenomenon Description Additional Notes & Comments Adequacy Availability Collect New Data

8. Pressure Effects A. Projectile/missile damage M L-M L-M B. Pressure wave M-H M-H N/A

Table C-23 State of Knowledge Ranking for Scenario 3, Internal Ensuing Fire State of Knowledge Ranking Model Data Ability to Phenomenon Description Additional Notes & Comments Adequacy Availability Collect New Data

9. Internal Ensuing Fire A. Fire ignition L L L-M B. Characterization of the fire source H M M C. Fire development M M M D. Smoke generation M L-M M-H E. Altered ventilation effects M M M C-15 Table C-24 State of Knowledge Ranking for Scenario 3, External Ensuing Fire State of Knowledge Ranking Model Data Ability to Phenomenon Description Additional Notes & Comments Adequacy Availability Collect New Data

10. External Ensuing Fire A. Fire ignition L-M L-M M B. Characterization of the fire source M-H M-H H C. Fire development M M H D. Cabinet-to-cabinet propagation L-M L-M M E. Smoke generation M L-M M

Table C-25 State of Knowledge Ranking for Scenario 3, Room Configuration Effects State of Knowledge Ranking Model Data Ability to Phenomenon Description Additional Notes & Comments Adequacy Availability Collect New Data

11. Room Configuration Effects A. Room integrity M M M-H B. Room arrangement H H N/A C. Room ventilation H H N/A C-16 Table C-26 State of Knowledge Ranking for Scenario 3, Arc Mitigation State of Knowledge Ranking Model Data Ability to Phenomenon Description Additional Notes & Comments Adequacy Availability Collect New Data

12. Arc Mitigation A. Intentional shielding L L-M M-H B. Electrical raceway fire barrier system L-M M H C. Fire-retardant coating L-M M H

Table C-27 State of Knowledge Ranking for Scenario 3, Generic Enclosure (Target) Characterization State of Knowledge Ranking Model Data Ability to Phenomenon Description Additional Notes & Comments Adequacy Availability Collect New Data

13. Generic Enclosure (Target)

Characterization A. Characterizing target damage criteria M M H B. Target arrangement effects M M H C. Target sensitivity to damage L-M L-M M C-17 Table C-28 State of Knowledge Ranking for Scenario 3, Generic Cable Tray (Target) Characterization State of Knowledge Ranking Model Data Ability to Phenomenon Description Additional Notes & Comments Adequacy Availability Collect New Data

14. Generic Cable Tray (Target)

Characterization A. Characterizing target damage criteria M M H B. Target arrangement effects M M H C. Target sensitivity to damage L-M L-M M

C.5 Key Parameter Rankings Table C-29 Key Parameters Ranking for Scenario 3, Electrical Configuration Effects Importance Ranking State of Parameters Additional Notes & Comments P1 P2 P3 P4 P5 P6 Knowledge 1A. Power Supply Characterization

1. Line Voltage H H H H H H H

2. Rated Current H H H H M H H

3. Short Circuit Current H H H H H H H

4. Frequency L L L L L L H

5. Grounded/Ungrounded U H M L L L H

6. Number of Phases L L L L L L H

7. MVA Available H H H H H L H C-18

Table C-30 Key Parameters Ranking for Scenario 3, Internal Cabinet Configuration Effects Importance Ranking State of Additional Notes &

Parameters P1 P2 P3 P4 P5 P6 Knowledge Comments 2A. Cabinet Compartmentation

1. Dimensions H H H M H H H

2. Internal Partitions H M M M M M M-H 2B. Breaker Configuration

3. Arc Extinguishing Mechanism L L L L L L H

4. Breaker Combustible Load H H M H H H H

5. Extinguishing Medium L L M L M L H

6. Breaker Connection Material M M M H H H H 2C. Cabinet Combustible Loading

7. Combustible Mass H H M H H H M C-19 8. Bundle Tightness H M M M M H M

9. Thermoset/Thermoplastic H H M H M H M-H

10. Jacketed/Unjacketed H H H H M H M-H

11. Combustion Properties H H H H H H M 2D. Cabinet Bus Bar Configuration

12. Bus Bar Spacing H H H H H H H

13. Bus Bar Material H H H H H M H

14. Number of Bus Bars per Phase M M L M M L M-H

15. Bus Bar Insulation M M M M M M M-H

16. Bus Bar Orientation L M M M M U H

17. Bus Bar Cross-Sectional Area M M L M L L H

18. Bus Bar Mounting/Supports L L M L L H H

Table C-31 Key Parameters Ranking for Scenario 3, External Cabinet Enclosure Configuration Effects Importance Ranking State of Parameters Additional Notes & Comments P1 P2 P3 P4 P5 P6 Knowledge 3A. Cabinet Ventilation

1. Active/Passive Ventilation H H H H H H H

2. Ventilation Opening Area H H H H H H H

3. Ventilation Deflection L L M M M L M-H

4. Ventilation Opening Location L M H M H H M-H

5. Presence of Active Louvres U U U U U L H 3B. Cabinet Structural Design

6. Wall Material M H H H L H H

7. Wall Thickness H M M H M M H

8. Door Latch Type M M M M M M H

9. Door Gasketing M M M M M M H

10. Wall Fastening (welded, bolted, L M L M M L H C-20 riveted)

11. Single/Double Wall H M M M M M H 3C. Cabinet Penetrations

12. Location of Penetrations H H M H M H H

13. Penetration Sealing H M M H M M H

14. Penetration Type (Compression, L L M M M L H conduit, raceway)

15. Viewing Ports/Mechanical Openings H U M H M M H

Table C-32 Key Parameters Ranking for Scenario 3, Cabinet Lineup Effects Importance Ranking State of Parameters Additional Notes & Comments P1 P2 P3 P4 P5 P6 Knowledge 4A. Cabinet-Cabinet Barrier Effects

1. Walls (None/Single/Double) H H H H H H M-H

2. Air Gap M- M- M- U M-H M L-M H H H 4B. Cabinet Position Within Bank Effects

3. Middle/End H H H H H H M-H 4C. Cabinet-Cabinet Penetration Effects

4. Bus Bar Penetrations H H H H H H M-H

5. Cable Penetrations H H H H H H M-H

6. False Floor H H H L H H M C-21

Table C-33 Key Parameters Ranking for Scenario 3, Arc Characterization Importance Ranking State of Parameters Additional Notes & Comments P1 P2 P3 P4 P5 P6 Knowledge 5A. Arc Electrical Characterization

1. Arc Voltage H H H H H H M

2. Arc Current (Asymmetric) H H H H L H H

3. Arc Current (RMS) H H H H H H H

4. Arc Energy H H H H H H M

5. Arc Duration H H H H H H M

6. Arc Location H M M M H M L 5B. Arc Migration

7. Arc Migration Speed U M M H M L M-H

8. Type (Ionized Gas/Magnetic Field) U H M H M M L 5D. Thermal Effects of the Arc

9. Arc Heat Flux H H H H H H M

10. Bus Bar Melting H H H H H H M-H C-22

11. Oxidation of Adjacent Materials H H H H H H L 5G. Arc Ejecta

12. Ionized Gas M H H H H H L

13. Smoke M L M L M L L

14. Conductive Particulate H H H H M H L

Table C-34 Key Parameters Ranking for Scenario 3, Fire Detection Importance Ranking State of Parameters Additional Notes & Comments P1 P2 P3 P4 P5 P6 Knowledge 6A. Presence of Fire Detection

1. Detector Location H H H M H M H 6B. Detection Characteristics

2. Detector Type (Heat/smoke/optical) M M M M M M H Table C-35 Key Parameters Ranking for Scenario 3, Fire Suppression C-23 Importance Ranking State of Parameters Additional Notes & Comments P1 P2 P3 P4 P5 P6 Knowledge 7A. Presence of Fire Suppression

1. Internal/External M M M M M M M-H

2. Proximity H H H H H H M-H 7B. Characterizing the Suppression System

3. Actuation (Manual/Automatic) M M M M M M H

4. Survivability X H M M X M L

5. Suppression Type M H H H H H M-H

Table C-36 Key Parameters Ranking for Scenario 3, Pressure Effects Importance Ranking State of Parameters Additional Notes & Comments P1 P2 P3 P4 P5 P6 Knowledge 8B. Pressure Wave

1. Magnitude H H H H H H M-H

2. Propagation H H H H H M M Table C-37 Key Parameters Ranking for Scenario 3, Internal Ensuing Fire Importance Ranking State of Parameters Additional Notes & Comments P1 P2 P3 P4 P5 P6 Knowledge C-24 9A. Fire Ignition

1. Internal Cabinet Temperature H H H H H H L 9B. Characterization of Fire Source

2. Heat Release Rate H H H H H H M-H

3. Heat Release Rate per unit Area M L M M M M M 9C. Fire Development

4. Time to Peak M M M M M M M-H

5. Fire Growth Rate H H H H H H M-H 9E. Altered Ventilation Effects

6. Altered Internal Vent Properties H H H H H H M

7. Altered Enclosure Vent Properties H H H H H H M

Table C-38 Key Parameters Ranking for Scenario 3, External Ensuing Fire Importance Ranking State of Parameters Additional Notes & Comments P1 P2 P3 P4 P5 P6 Knowledge 10B. Characterization of Fire Source

1. Heat Release Rate H H H H H H M-H

2. Heat Release Rate per unit Area M L M M M M M-H

3. Fire Geometry H H H H H H H 10C. Fire Development

4. Time to Peak M M M M M M H

5. Fire Growth Rate H H H H H H H C-25 Table C-39 Key Parameters Ranking for Scenario 3, Room Configuration Effects Importance Ranking State of Parameters Additional Notes & Comments P1 P2 P3 P4 P5 P6 Knowledge 11A. Room Integrity

1. Fire Dampers H H H H H H L-M

2. Penetrations H H H H H H L-M 11B. Room Arrangement

3. Dimensions H H H H H H H

4. Contents M H H H H H H

5. Construction Material M M M M M M H 11C. Room Ventilation

6. Type (Active/Passive) H H H H H H H

Table C-40 Key Parameters Ranking for Scenario 3, Arc Mitigation Importance Ranking State of Parameters Additional Notes & Comments P1 P2 P3 P4 P5 P6 Knowledge 12A. Intentional Shielding

1. Thickness H M H H H H H

2. Material H H H H H H H 12B. Electrical Raceway Fire Barrier System

3. Rating U U U U U U H 12C. Fire-Retardant Coating Effects

4. Thickness U U U H H U M

5. Material L U U H H H H

6. Age L U U U H U H C-26

Table C-41 Key Parameters Ranking for Scenario 3, Generic Enclosure (Target) Characterization Importance Ranking State of Additional Notes &

Parameters P1 P2 P3 P4 P5 P6 Knowledge Comments 13A. Target Damage

1. Damage Criteria H H H H H H M

2. Critical Heat Flux H H H H H H M

3. Critical Temperature H H H H H H M

4. Exposure Time H H H H H H M 13C. Target Sensitivity

5. Sensitivity to Shock L H H H H H M

6. Sensitivity to Heat H H H H H H M-H

7. Sensitivity to EMI/RFI H H H H U H L-M

8. Sensitivity to Light L L L L L L L-M

9. Sensitivity to Ionized Gas L H H H H H M C-27

10. Sensitivity to Conductive Particulate H H H H H H L-M

11. Sensitivity to Smoke L-H L L L L L M

Table C-42 Key Parameters Ranking for Scenario 3, Generic Cable Tray (Target) Characterization Importance Ranking State of Additional Notes &

Parameters P1 P2 P3 P4 P5 P6 Knowledge Comments 14A. Target Damage

1. Damage Criteria H H H H H H M

2. Critical Heat Flux H H H H H H M

3. Critical Temperature H H H H H H M

4. Exposure Time H H H H H H M 14B. Target Arrangement Effects

5. Distance to HEAF Source H H H H H H H

6. Orientation Relative to HEAF H H H H H H H Source 14C. Target Sensitivity

7. Sensitivity to Shock M M H M M M M

8. Sensitivity to Heat H H H H H H M-H C-28

9. Sensitivity to EMI/RFI H U U L U L M

10. Sensitivity to Light L L L L L L M

11. Sensitivity to Ionized Gas L M M L M M M

12. Sensitivity to Conductive Particulate H L L L L L M

13. Sensitivity to Smoke H L L L L L M

C.6 Scenario 3 Phenomena Calculated Rank Values and Grouping Table C-43 Summary of Scenario 3 Average Rankings and Grouping Phenomenon Importance State of Knowledge Rank Grouping Arc Characterization (5) 4 2 2.00 Level 1 Pressure Effects (8) 5 3 1.67 Level 1 Enclosure (Target) Characterization (13) 5 3.2 1.56 Level 1 Internal Ensuing Fire (9) 4.3 2.8 1.54 Level 1 Cable Tray (Target) Characterization (14) 4.9 3.2 1.53 Level 1 External Ensuing Fire (10) 4.1 3.1 1.32 Level 2 Cabinet Lineup Effects (3) 4.4 3.4 1.29 Level 2 Arc Mitigation (12) 3.7 3 1.23 Level 2 Electrical Configuration (1) 5 5 1.00 Level 2 Internal Cabinet Configuration (2) 3.7 4 0.93 Level 3 External Enclosure Configuration (3) 3.2 4.2 0.76 Level 3 C-29 Suppression Effects (7) 3 4.5 0.67 Level 3 Room Configuration (10) 2.8 4.3 0.65 Level 3 Fire Detection (6) 1.3 5 0.26 Level 3 33rd percentile for Rank: 0.95 66th percentile for Rank: 1.44

APPENDIX D PANELIST RÉSUMÉS D-1

D-2 D-3 D-4 D-5 D-6 D-7 D-8 APPENDIX E TECHNICAL EXPERT AND ADVISOR RÉSUMÉS E-1

E-2 E-3 E-4 E-5 E-6 Stephen L. Turner Leidos, Inc.

301 Laboratory Road Oak Ridge, Tennessee 37830 Education B.S., M.S. Mechanical Engineering, University of Virginia Since 1979, Mr. Turner has held various positions at Leidos (formerly Science Applications International Corporation) including Assistant Vice President and Chief Scientist in the area of nuclear safety. He is currently a consultant at Leidos.

Mr. Turner has 38 years of experience in nuclear facility safety and probabilistic analysis and supported the plans and basis for the implementation of risk-informed regulations based on Probabilistic Risk Assessments (PRA) for the Japanese utilities and the nuclear regulator in Japan since 2001. Mr. Turner was project manager 18 projects for fire safety including the methods for analyses and walk downs to support fire safety assessments to NUREG/CR-6850. This included various component fire tests since 2008 at Clemson University for HEPA filter soot loading and at Southwest Research Institute (SwRI) for cable fire and oil fire tests similar to the tests that support NUREG-6850. Since 2012, he was the project manager for four test projects for High Energy Arcing Faults (HEAF) at KEMA Laboratories completing a total of 24 HEAF tests on switchgear, distribution panels, and motor control centers. Three of these tests duplicated the conditions of the HEAF at the Onagawa power plant that occurred in the 2011 Tohoku Earthquake. He also was the project manager at SwRI for six tests using rocket fuel to create HEAF-like conditions in a 5-cabinet switchgear lineup. He was a principal contributor to NUREG/IA-0470 that documents these HEAF tests. For the last two years he was the project manager research for modeling and analysis methods for HEAF phenomenon compiling results for published methods and developing models in ANSYS-Fluent.

E-7

E-8 APPENDIX F INTRODUCTORY MATERIALS PRESENTED AT THE FIRST PANEL MEETING F.1 PIRT Process Refresher Phenomena Identification and Ranking Table Description of scenario Scenarios are purposefully vague - this gives the panel more latitude to choose the important parameters Figure of merit The figure of merit is the specific goal to be achieved through analysis. The phenomena identified by the panel should be in the context of achieving this goal.

F-1

==

Description:==

A high energy arc fault occurs in a low/medium voltage switchgear that supplies power to Train A systems. The cables from an overhead cable tray enter the cabinet from the top (air-drop configuration). A similarly-configured switchgear supplying power to redundant Train B systems is located close by.

Figure of Merit: Determining the extent of damage to the Train B switchgear.

The group will identify all phenomena that could be important in achieving the figure of merit.

At this stage, no judgement regarding the importance of the phenomena is necessary.

The phenomena will be recorded and organized into categories as needed.

F-2

If it has units, or can be measured directly, it is probably a parameter. For example, cabinet voltage is a parameter. The underlying phenomenon would be the cabinets electrical configuration.

I will record key parameters as well, but they will be ranked separately.

Once the phenomena are identified, the individuals on the panel will rank each as being of high (H) medium (M) or low (L) importance.

This will be done individually first, and then the rankings will be discussed as a group.

F-3

For each phenomenon, the panel will rank the state of knowledge in 3 categories:

Model adequacy Data adequacy Obtainability of new data Again, we will use a high (H) medium (M) or low (L) scale.

Finally, the panel will rank the key parameters identified earlier for both importance and state of knowledge.

F-4

F.2 Recent Experimental Work Performed by NRA TH32 - Improving Realism in Fire Probabilistic Risk Assessments Experimental Studies of High Energy Arcing Fault (HEAF) by S/NRA/R Hajime KABASHIMA and Susumu Tsuchino Regulatory Standard and Research Department, Secretariat of Nuclear Regulation Authority (S/NRA/R), Tokyo, Japan 1 / 12 Outline

1. Background

2. Objectives of HEAF tests

3. S/NRA/R HEAF tests

4. S/NRA/R HEAF test results

5. Discussion topics

6. Summary U.S.NRCs 28th Regulatory Information Conference (RIC 2016)

F-5

2 / 12

1. Background

A High Energy Arcing Fault (HEAF) M arch 11, 2011 occurred in the high-voltage (6,900V) metalclad switchgear (M/C) at Unit 1 of Onagawa Nuclear Power Station (NPS) of the Tohoku Electric Power Company Co., Inc.

on March 11, 2011 due to the 2011 off the Pacific coast of Tohoku Earthquake. Onagawa NPS Unit 1 (http://warp.da.ndl.go.jp/info:ndljp/pid/9483636/w ww.nsr.go.jp/archive/nisa/earthquake/files/houkok HEAF events, although their impacts u230530-2.pdf) differ each other, have occurred in the electrical equipment and components in the NPSs worldwide. Efforts are being taken to understand the phenomena and to develop evaluation methods.

U.S.NRCs 28th Regulatory Information Conference (RIC 2016) 3 / 12

2. Objectives of HEAF tests To understand the basic characteristics and behavior of HEAF To understand what occurred in the Onagawa NPS Unit 1 due to HEAF event during the 2011 off the Pacific coast of Tohoku Earthquake

- - - - - - - - - - - - - - (In the future) - - - - - - - - - - - - - -

Develop regulatory guidance for Fire Hazards Analysis Methods for HEAF Models for damage predictions and setting Zone of Influence (ZOI)

Methods for the Protection against HEAF U.S.NRCs 28th Regulatory Information Conference (RIC 2016)

F-6

4 / 12

3. S/NRA/R HEAF tests Collaboration with S/NRA/R HEAF Tests NRC NIST KEMA Low Voltage HEAF Tests Onagawa Configuration Tests Distribution Motor Control Metalclad Switchgear Panel (DP) Center (MCC) (M/C) 5-cabinet lineup 5-cabinet lineup Heat sourceRocket fuel Burned test Voltage : 480V (Set point) Voltage : 480V (Set point) Voltage : 7000V or 7100V (Set point)

Current : 52.3kA (Target value) Current : 63.5kA (Target value) Current : 22.6kA - 25.0kA (Target value)

Arc discharge Arc discharge Arc discharge Duration: 2s Duration: 2s Duration: 1s, 2s, 3s HEAF tests: 3 times HEAF tests: 4 times HEAF tests 6 times U.S.NRCs 28th Regulatory Information Conference (RIC 2016) 5 / 12 4.1 S/NRA/R HEAF test results <Distribution Panel>

0s 0 .1s (1) Before test (2) Arc discharge occurrence (about 0.1 sec after start of test) 1s 10min Arc power 20MW (3) Generation of metallic fume (4) Ensuing Fire (about 1 second after start of test) (about 10 minutes after start of test)

In the DP HEAF tests, ensuing fire occurred in two out of three tests.

Arc power was approximately 20MW for all the tests.

U.S.NRCs 28th Regulatory Information Conference (RIC 2016)

F-7

6 / 12 4.2 S/NRA/R HEAF test results <Motor Control Center>

Test cell 0s 0 .2s (1) Before test (2) Arc discharge occurrence (about 0.2 sec. after start of test)

Arc power 20MW 0.5s 2s (3) Blowout of arc discharge metallic fume (4) Generation of metallic dust (about 0.5 second after start of test) (about 2 second after start of test)

In MCC HEAF test, the ensuing fire did not occur for all four tests.

Arc power was approximately 20MW for all tests.

U.S.NRCs 28th Regulatory Information Conference (RIC 2016) 7 / 12 4.3 S/NRA/R HEAF test results <Metalclad Switchgear>

0s 0 .2s (1) Before test (2) Arc discharge occurrence (about 0.2 sec after start of test) 2s 10min Arc power 20MW (3) Generation of metallic fume (4) Ensuing Fire (about 2 second after start of test) (about 10 minutes after start of test)

In the M/C HEAF tests, ensuing fire occurred in four out of six tests.

Arc power was approximately 20MW for four tests and larger than 20MW for two tests.

U.S.NRCs 28th Regulatory Information Conference (RIC 2016)

F-8

8 / 12 5.1 Discussion topics <Arc power>

Proportional relationship is confirmed between the arc energy and the arc duration for each electric cabinet, and the Electromagnetic arc power is almost constant.

pinch force Fig. Energy model for arc discharge The arc power is a value multiplied by an energy density, a flow velocity and a cross-sectional area of arc jet.

Two key factors; Saturation phenomena of energy density Inverse relationship between flow velocity and cross-sectional area.

Two data with slightly higher arc power were obtained. It needs further consideration.

Fig. Relationship between arc discharge duration and arc energy U.S.NRCs 28th Regulatory Information Conference (RIC 2016) 9 / 12 5.2 Discussion topics <Ensuing fire>

S/NRA/R HEAF tests showed that the arc energy required for initiating ensuing fire differed between distribution panels and metalclad switchgears.

The values of arc energy which can cause ensuing fire were between 26.3 and 28.6 MJ for the distribution panels and between 42.6 and 57.2 MJ for the metalclad switchgears.

Values of arc energy required for causing ensuing fire were obtained. This triggering energy is considered to be dependent on the characteristics of individual electric cabinets such as interior volume and ventilation opening area.

Distribution Panel Metalclad Switchgears Cubicle size: Middle Cubicle size: Large Ventilation opening area: Middle Ventilation opening area: Large Length: 0.9 m Length: 2.1 m Depth: 1.2 m Depth: 0.9 m Height: 2.3 m Height: 2.3 m If arc discharge duration could be reduced, arc energy would be decreased and consequently ensuing fire could be prevented.

Fig. Arc energy required for causing ensuing fire U.S.NRCs 28th Regulatory Information Conference (RIC 2016)

F-9

10 / 12 5.3 Discussion topics <Onagawa HEAF event>

M arch 11, 2011 One time HEAF test Bus bar: Al Onagawa NPS Unit 1 Bus bar: Al (http://warp.da.ndl.go.jp/info:ndljp/pid/9483636/www.nsr.

go.jp/archive/nisa/earthquake/files/houkoku230530-2.pdf)

HEAF event at Onagawa NPS Unit 1, it is supposed from observation of the damage condition that the arc discharge was occurred two times in the different Metalclad switchgears.

Two times HEAF test Bus bar: Al The damage of metalclad switchgears at Onagawa NPS Unit 1 was more severe than that after HEAF tests.

Aluminum bus bars of the metalclad switchgears at Onagawa NPS Unit 1 were severely damaged compared with those of HEAF tests.

Burning (oxidation) of the bus bar made of aluminum can cause huge heat energy release by oxidation of aluminum. Therefore, in addition to the arc energy due to HEAF, the high energy of aluminum bus bar oxidation should be considered in the consequence evaluation for HEAF events in electric cabinets.

U.S.NRCs 28th Regulatory Information Conference (RIC 2016) 11 / 12

6. Summary HEAF tests were conducted to obtain the insights about progress of HEAF events, energy level of arcs and condition for ensuing fire.

Based on the test results, an energy model of arc discharge has been studied.

For various electric cabinets, insights of event progress were obtained pertaining to generation and leakage of arcs, generation of metal fume and ensuing fire.

Proportional relationship was observed between arc energy and arc duration irrespective of types of electric cabinets for all cases.

Values of arc energy required to cause ensuing fire were obtained.

This triggering energy is considered to be dependent on the characteristics of individual electric cabinets such as interior volume and ventilation opening area.

U.S.NRCs 28th Regulatory Information Conference (RIC 2016)

F-10

F.3 Recent Experimental Work Performed by CRIEPI F-11

F-12 F-13 F-14 F-15 F-16 F-17 F-18 F-19 F-20 F-21 F-22 F-23 F-24 F-25 F-26 F-27 F-28

NUREG-2218 An International Phenomena Identification and Ranking Table (PIRT)

Expert Elicitation Exercise for High Energy Arcing Faults (HEAFs) January 2018 Kenneth Hamburger Technical Division of Risk Assessment Office of Nuclear Regulatory Research U.S. Nuclear Regulatory Commission Washington, DC 20555 Same as above K. Hamburger, NRC Project Manager This report documents the results of a phenomena identification and ranking table (PIRT) exercise performed for nuclear power plant (NPP) high energy arc fault (HEAF) analysis applications conducted on behalf of the U.S. Nuclear Regulatory Commissions (NRCs) Office of Nuclear Regulatory Research.

The PIRT exercise was performed via a facilitated expert elicitation process. In this case, the expert panel comprised six international HEAF experts. The panel was facilitated by NRC staff. The objective of a PIRT exercise is to identify key phenomena associated with the intended application and to then rank the current state of knowledge relative to each identified phenomenon. The expert panel was presented with a series of specific HEAF scenarios, each of which is based on the types of scenarios typically considered in NPP applications. Each scenario includes a figure of merit (i.e., a specific goal to be achieved in analyzing the scenario using HEAF analysis or modeling tools). Given each scenario, the panel identified phenomena that are of potential interest to an assessment based on the figure of merit. The identified phenomena are then ranked relative to their importance in predicting the figure of merit. Each phenomenon is then further ranked for the existing state of knowledge and the adequacy of existing modeling tools to predict that phenomenon.

The PIRT panel covered three HEAF scenarios and identified a number of areas potentially in need of further analysis and model development. This report discusses the results in detail.

High Energy Arcing Faults HEAF Phenomena Identification and Ranking Table PIRT Expert Elicitation

NUREG-2218 An International Phenomena Identification and Ranking Table (PIRT) Expert Elicitation Exercise January 2018 for High Energy Arcing Faults (HEAFs)