Circuit Failure Mode Likelihood Analysis (Task 10)

Task Overview

Background

This task considers the relative likelihood of various circuit failure modes. This added level of resolution may be a desired option for those fire scenarios that are significant contributors to the risk. The methodology provided in this document benefits from the knowledge gained from the tests performed in response to the circuit failure issue.

Purpose

Conducting a Fire PRA in accordance with this methodology necessitates an analysis of fire-induced circuit failures beyond that typically conducted during original Fire PRAs. The circuit analysis elements of the project are conducted in three distinct phases:

1. Fire PRA cable selection (Task 3),
2. Detailed circuit failure analysis (Task 9), and
3. Circuit failure mode likelihood analysis (Task 10).

This task provides methods and instructions for conducting the third phase of circuit analysis – circuit failure mode likelihood analysis for Fire PRA cables. Task 10 estimates the probability of hot short cable failure modes of interest, which in turn can be correlated to specific component failure modes. As discussed in Section 3.3.2 of Volume 1, the methods and techniques for deriving circuit failure mode probability estimates are based on limited data and experience. Consequently, this area of analysis is not yet a mature technology, and undoubtedly further advances and refinements will come with time. Nonetheless, the methods and techniques presented in this chapter represent the current state of knowledge and provide a reasonable approach for establishing first-order circuit failure mode probability estimates, albeit with relatively high uncertainty tolerances.

Scope

Chapter 10 provides methods and technical considerations for assigning probability estimates to specific cable failure modes associated with fire-induced cable damage.

This task does not address the implementation of plant-specific quality assurance or configuration control requirements that might apply to a Fire PRA. Nor is it intended to validate the accuracy of plant-specific data extracted from plant drawings, documents, or databases. Each plant should follow appropriate quality assurance, administrative, and configuration control procedures applicable to the work being conducted. The need to validate input source documents should be addressed as part of assembling the prerequisite information in Step 1.

Related Element of ASME/ANS PRA Standard

Circuit Failures (CF)

Related EPRI 1011989 NUREG/CR-6850 Appendices

Appendix J, Appendix for Chapter 10, Technical Basis for Circuit Failure Mode Likelihood Equations (Note: This method is no longer applicable following the publication of EPRI 3002001989 / NUREG/CR-7150 Volume 2).

Appendix K, Appendix for Chapter 10, Examples of Component Circuit Failure Mode Likelihood Analyses

Supplemental Guidance

Since the publication of NUREG/CR-6850, significant research on fire-induced spurious operations has occurred. EPRI 3002001989 / NUREG/CR-7150 Volume II revises a significant portion of Task 10 of EPRI 1011989 / NUREG/CR-6850.

FAQ 08-0047, Spurious Operation Probability, Section 15 of Supplement 1

FAQ 08-0047 was proposed to clarify guidance on quantification of spurious operation probabilities for independent vs. dependent combinations. NUREG/CR-6850 (EPRI 1011989) provided a calculational scheme for multiple spurious operation probabilities that implicitly assumed independence. As such, it would overestimate the joint probability in cases where the spurious operations exhibited dependence.

Impact on NUREG/CR-6850 (EPRI 1011989) Guidance

This guidance in FAQ 08-0047 provides clarification on the appropriate use of the “Exclusive Or” function listed in NUREG/CR-6850 Section 10.5.3.1, page 10-9, bullet list item number 3, and in Section 10.5.3.1, page 10-11, bullet list item number 4. EPRI 3002001989 / NUREG/CR-7150 Volume 2 is silent on FAQ 08-0047 treatment and its continued applicability. Therefore, FAQ 08-0047 consideration is still applicable when estimating hot short induced spurious operation probability using NUREG/CR-7150.

Joint Assessment of Cable Damage and Quantification of Effects from Fire (JACQUE-FIRE): Phenomena Identification and Ranking Table (PIRT) Exercise for Nuclear Power Plant Fire-Induced Electrical Circuit Failure (EPRI 1026424 / NUREG/CR-7150 Volume 1)

This report documents a Phenomena Identification and Ranking Table (PIRT) exercise on fire-induced electrical circuit failures that may occur in nuclear power plants as a result of fire damage to cables. The results and conclusions of the PIRT panel are a primary input to the follow on PRA Expert Panel documented in Volume 2.

Joint Assessment of Cable Damage and Quantification of Effects from Fire (JACQUE-FIRE): Expert Elicitation Exercise for Nuclear Power Plant Fire-Induced Electrical Circuit Failure (EPRI 3002001989 / NUREG/CR-7150 Volume 2)

This report documents the results of the PRA panel's expert elicitation that is used to develop the conditional probabilities of hot-short induced spurious operation & duration of various control circuit configurations.

The information in this report revises a significant portion of Task 10 of NUREG/CR-6850. The results are presented such that the bounding values can be applied based on a minimum set of parameters, and more specific values can be applied if more rigors are used to define the circuit configuration and conditions. Unfortunately, the application is not a simple update of the numbers. The use of Option #2 (Computational Probability Estimates) as a technical approach for calculating spurious operation likelihood has been eliminated

The following tables in EPRI 3002001989 / NUREG/CR-7150 Volume 2 replace the conditional probability tables for Option #1 in EPRI 1011989 NUREG/CR-6850:

• Failure Mode Probability Estimation Tables
  • Table 4-1 - Conditional Probability of Spurious Operation: SOV Single Break-Control Circuits
  • Table 4-3 - Conditional Probability of Spurious Operation: MOV Single Break Control Circuits
  • Table 4-4 - Conditional Probability of Spurious Operation: Ungrounded DC Control Circuits for Medium Voltage Circuit Breaker
  • Table 5-1 - Conditional Probability of Spurious Operation: Double Break Control Circuits for Ungrounded AC (w/ individual CPTs) Base Case – SOV
  • Table 5-2 - Conditional Probability of Spurious Operation: Double Break Control Circuits
  • Table 5-3 - Conditional Probability of Spurious Operation: Double Break Control Circuits for Ungrounded AC (w/ individual CPTs) Motor-Operated Valve
  • Table 5-4 - Conditional Probability of Spurious Operation: Double Break Control Circuits Ungrounded DC (or Ungrounded Distributed AC)

Aggregate values are given for every case where all potential hot short-induced failure modes are applicable. Unless it is demonstrated that a cable under evaluation is only susceptible to a single failure mode, the aggregate values given herein should be used to represent the conditional probability of fire-induced spurious operation to an end device under evaluation.

The plot of the probability of spurious operation duration in Figure 6-4 for AC control circuits should replace the plot of AC duration (Figure 16-1 and Table 16-1) for FAQ 08-0051 in Section 16 of Supplement 1 to NUREG/CR-6850. In addition, Figure 6-5 for DC control circuits now provides the plot for DC duration. The spurious operation duration tables are found in EPRI 3002001989 / NUREG/CR-7150 Volume II:

• Table 6-3 - Tabulated Spurious Operation Duration Conditional Probability Values for AC and DC Control Circuits

In addition to the impact on NUREG/CR-6850 described in Task 3 (Chapter 5) of this document, the following additional supplemental guidance was provided in NUREG/CR-7150, Volume 1:

• Kapton® cable, a polyamide stable from -271 to +400°C, should be treated as a thermoplastic (TP) material for assessment of circuit failures, including risk-informed applications, unless (1) the cable has been fully qualified against the requirements of the IEEE Std. 383 severe accident equipment qualification standard, in which case it may be treated as a thermoset (TS) cable, or (2) an alternate product/case specific basis can be established for the cable’s thermal damage limits (e.g., manufacturer or utility testing).

NRC Correspondence

• NRC Memo dated June 14, 2013, Interim Technical Guidance on Fire-Induced Circuit Failure Mode Likelihood Analysis and Enclosure (ADAMS Accession Nos. ML13165A209, ML13165A214)

• NRC letter to NEI, dated December 16, 2013, Interim Technical Guidance on Fire-Induced Circuit Failure Mode Likelihood Analysis and Enclosure (ADAMS Accession Nos. ML13238A280, ML13165A214)

• NRC Memo dated February 12, 2014, Supplemental Interim Technical Guidance on Fireinduced Circuit Failure Mode Likelihood Analysis (ADAMS Accession No. ML14017A091)

• NRC letter to NEI, dated April 23, 2014, Supplemental Interim Technical Guidance on Fire-Induced Circuit Failure Mode Likelihood Analysis (ADAMS Accession No. ML14086A165)

(The memos and letters dated June 14, 2013, December 16, 2013, February 12, 2014, and April 23, 2014 referenced interim guidance until the issuance of NUREG/CR-7150, Volume 2, in May 2014. NUREG/CR-7150, Vol. 2 should be referred to as the official guidance.)

Joint Assessment of Cable Damage and Quantification of Effects from Fire (JACQUE-FIRE): Technical Resolution to Open Issues on Nuclear Power Plant Fire-Induced Circuit Failure (EPRI 3002009214 / NUREG/CR-7150 Volume 3)

While the scope of this report was focused on deterministic post-fire safe shutdown analyses, the technical considerations and insights can be used in support of assigning probability estimates to specific cable failure modes associated with fire-induced cable damage.

Impact on EPRI 1011989 NUREG/CR-6850 Guidance

• Clarified circuit failure modes and terminology including: proper polarity, latching versus non-latching, high impact components, failure mode classification and GFEHS
• Revised several PIRT panel positions following the conclusions of the PRA Expert Panel
  • Consideration of insulation type for the aggressor cable conductor is eliminated. This was eliminated due to the impracticality of tracking the conductor insulation for aggressor cables. As a result, the reported classifications are a function of the conductor insulation of the target conductor.
  • Classification of inter-cable hot shorts for thermoset insulated conductors from "implausible" to "plausible" based on the PRA expert panel probabilities that did not support a classification of "implausible"
  • Grouped single break ungrounded AC (from common CPT or distributed) with DC due to similarities in circuit failure type classification.
  • Classified inter-cable hot shorts for double break design with TP insulated conductors as "implausible" for latching and "incredible" for non-latching circuits. This classification was deferred until the PRA Expert Panel completed their estimation.
• Developed technical design considerations for shorting switch applications
• Disposition of secondary fires due to a fire-induced open circuited current transformer
  • Testing is documented in NUREG/CR-7228, Open Secondary Testing of Window-Type Current Transformers