Risk-Informed Changes to Loss-of-Coolant Accident Technical Requirements, 67598-67630 [E5-6090]

Agencies

• NUCLEAR REGULATORY COMMISSION

[Federal Register Volume 70, Number 214 (Monday, November 7, 2005)]
[Proposed Rules]
[Pages 67598-67630]
From the Federal Register Online via the Government Printing Office [www.gpo.gov]
[FR Doc No: E5-6090]



[[Page 67597]]

-----------------------------------------------------------------------

Part IV





Nuclear Regulatory Commission





-----------------------------------------------------------------------



10 CFR Part 50



Risk-Informed Changes to Loss-of-Coolant Accident Technical 
Requirements; Proposed Rule

Federal Register / Vol. 70, No. 214 / Monday, November 7, 2005 / 
Proposed Rules

[[Page 67598]]


-----------------------------------------------------------------------

NUCLEAR REGULATORY COMMISSION

10 CFR Part 50

RIN 3150-AH29


Risk-Informed Changes to Loss-of-Coolant Accident Technical 
Requirements

AGENCY: Nuclear Regulatory Commission.

ACTION: Proposed rule.

-----------------------------------------------------------------------

SUMMARY: The Nuclear Regulatory Commission (NRC) proposes to amend its 
regulations to permit current power reactor licensees to implement a 
voluntary, risk-informed alternative to the current requirements for 
analyzing the performance of emergency core cooling systems (ECCS) 
during loss-of-coolant accidents (LOCAs). In addition, the proposed 
rule would establish procedures and criteria for requesting changes in 
plant design and procedures based upon the results of the new analyses 
of ECCS performance during LOCAs.

DATES: Submit comments by February 6, 2006. Submit comments specific to 
the information collections aspects of this proposed rule by December 
7, 2005. Comments received after the above dates will be considered if 
it is practical to do so, but assurance of consideration cannot be 
given to comments received after these dates.

ADDRESSES: You may submit comments on the proposed rule by any one of 
the following methods. Please include the following number, RIN 3150-
AH29, in the subject line of your comments. Comments on rulemakings 
submitted in writing or in electronic form will be made available for 
public inspection. Because your comments will not be edited to remove 
any identifying or contact information, the NRC cautions you against 
including any information in your submission that you do not want to be 
publicly disclosed.
    Mail comments to: Secretary, U.S. Nuclear Regulatory Commission, 
Washington, DC 20555-0001, ATTN: Rulemakings and Adjudications Staff.
    E-mail comments to: SECY@nrc.gov. If you do not receive a reply e-
mail confirming that we have received your comments, contact us 
directly at (301) 415-1966. You may also submit comments via the NRC's 
rulemaking Web site at https://ruleforum.llnl.gov. Address questions 
about our rulemaking Web site to Carol Gallagher (301) 415-5905; e-mail 
cag@nrc.gov. Comments can also be submitted via the Federal eRulemaking 
Portal https://www.regulations.gov.
    Hand deliver comments to: 11555 Rockville Pike, Rockville, Maryland 
20852, between 7:30 a.m. and 4:15 p.m. Federal workdays. (Telephone 
(301) 415-1966).
    Fax comments to: Secretary, U.S. Nuclear Regulatory Commission at 
(301) 415-1101.
    You may submit comments on the information collections by the 
methods indicated in the Paperwork Reduction Act Statement.
    Publicly available documents related to this rulemaking may be 
viewed electronically on the public computers located at the NRC's 
Public Document Room (PDR), O1 F21, One White Flint North, 11555 
Rockville Pike, Rockville, Maryland. The PDR reproduction contractor 
will copy documents for a fee. Selected documents, including comments, 
may be viewed and downloaded electronically via the NRC rulemaking Web 
site at https://ruleforum.llnl.gov.
    Publicly available documents created or received at the NRC after 
November 1, 1999, are available electronically at the NRC's Electronic 
Reading Room at https://www.nrc.gov/reading-rm/adams.html. From this 
site, the public can gain entry into the NRC's Agencywide Document 
Access and Management System (ADAMS), which provides text and image 
files of NRC's public documents. If you do not have access to ADAMS or 
if there are problems in accessing the documents located in ADAMS, 
contact the NRC Public Document Room (PDR) Reference staff at 1-800-
397-4209, (301) 415-4737 or by e-mail to pdr@nrc.gov.

FOR FURTHER INFORMATION CONTACT: Richard Dudley, Office of Nuclear 
Reactor Regulation, U.S. Nuclear Regulatory Commission, Washington DC 
20555-0001; telephone (301) 415-1116; e-mail: rfd@nrc.gov,

SUPPLEMENTARY INFORMATION: 

Table of Contents

I. Background
    A. Deterministic Approach
    B. History of Requirements and Design for LOCAs
    C. Probabilistic Approach
    D. Commission Policy on Risk-Informed Regulation
II. Rulemaking Initiation
III. Proposed Action
    A. Overview of Rule Framework
    B. Determination of the Transition Break Size (TBS)
    1. Historical Estimates of LOCA Frequencies
    2. Expert Opinion Elicitation Process
    3. Adjustments To Address Failure Mechanisms Not Considered by 
the Expert Elicitation
    a. LOCAs caused by failure of active components, such as stuck-
open valves and blown out seals or gaskets.
    b. Seismically-induced LOCAs, both with and without material 
degradation.
    c. LOCAs caused by dropped heavy loads.
    4. Consideration of Connected Auxiliary Piping
    5. Considerations of Break Location and Flow Characteristic
    6. Effects of Future Plant Modifications on TBS
    7. Future Adjustments to TBS
    C. Alternative ECCS Analysis Requirements and Acceptance 
Criteria
    1. Acceptable Methodologies and Analysis Assumptions
    2. Acceptance Criteria
    3. Plant Operational Requirements Related to ECCS Analyses
    4. Restrictions on Reactor Operation
    D. Risk-Informed Changes to the Facility, Technical 
Specifications, or Procedures
    1. Requirements for the Risk-Informed Integrated Safety 
Performance (RISP) Assessment Process
    a. Risk acceptance criteria for plant changes under 10 CFR 50.90
    b. Risk acceptance criteria for plant changes under 10 CFR 50.59
    c. Cumulative risk acceptance criteria
    d. Defense-in-depth
    e. Safety margins
    f. Performance measuring programs
    2. Requirements for risk assessments
    a. Probabilistic Risk Assessment (PRA) requirements
    b. Requirements for risk assessments other than PRA
    3. Operational Requirements
    a. Maintain ECCS model(s) and/or analysis method(s)
    b. Do not place the plant in unanalyzed at-power operating 
configurations
    c. Evaluate all facility changes using the RISP assessment 
process
    d. Implement adequate performance-measurement programs
    e. Periodically re-evaluate and update risk assessments
    E. Reporting Requirements
    1. ECCS analysis of record and reporting requirements
    2. Risk assessment reporting requirements
    3. Minimal risk plant change reporting requirement
    F. Documentation Requirements
    G. Submittal and Review of Applications Under Sec.  50.46a
    1. Initial application for implementing alternative Sec.  50.46a 
requirements
    2. Subsequent applications for plant changes under Sec.  50.46a 
requirements
    H. Potential Revisions Based on LOCA Frequency Reevaluations
    I. Changes to General Design Criteria
    J. Specific Topics Identified for Public Comment
IV. Public Meeting During Development of Proposed Rule
V. Section-by-Section Analysis of Substantive Changes
VI. Criminal Penalties
VII. Compatibility of Agreement State Regulations
VIII. Availability of Documents

[[Page 67599]]

IX. Plain Language
X. Voluntary Consensus Standards
XI. Finding of No Significant Environmental Impact: Environmental 
Assessment
XII. Paperwork Reduction Act Statement
XIII. Regulatory Analysis
XIV. Regulatory Flexibility Certification
XV. Backfit Analysis

I. Background

    During the last few years, the NRC has had numerous initiatives 
underway to make improvements in its regulatory requirements that would 
reflect current knowledge about reactor risk. The overall objectives of 
risk-informed modifications to reactor regulations include:
    (1) Enhancing safety by focusing NRC and licensee resources in 
areas commensurate with their importance to health and safety;
    (2) Providing NRC with the framework to use risk information to 
take action in reactor regulatory matters, and
    (3) Allowing use of risk information to provide flexibility in 
plant operation and design, which can result in reduction of burden 
without compromising safety, improvements in safety, or both.
    In stakeholder interactions, one candidate area identified for 
possible revision was emergency core cooling system (ECCS) requirements 
in response to postulated loss-of-coolant accidents (LOCAs). The NRC 
considers that large break LOCAs to be very rare events. Requiring 
reactors to conservatively withstand such events focuses attention and 
resources on extremely unlikely events. This could have a detrimental 
effect on mitigating accidents initiated by other more likely events. 
Nevertheless, because of the interrelationships between design features 
and regulatory requirements, making changes to technical requirements 
of certain parts of the regulations on ECCS performance has the 
potential to affect many other aspects of plant design and operation. 
The NRC has evaluated various aspects of its requirements for ECCS and 
LOCAs in light of the very low estimated frequency of the large LOCA 
initiating event.

A. Deterministic Approach

    The NRC has established a set of regulatory requirements for 
commercial nuclear reactors to ensure that a reactor facility does not 
impose an undue risk to the health and safety of the public, thereby 
providing reasonable assurance of adequate protection to public health 
and safety. The current body of NRC regulations and their 
implementation are largely based on a ``deterministic'' approach.
    This deterministic approach establishes requirements for 
engineering margin and quality assurance in design, manufacture, and 
construction. In addition, it assumes that adverse conditions can exist 
(e.g., equipment failures and human errors) and establishes a specific 
set of design basis events (DBEs) for which specified acceptance 
criteria must be satisfied. Each DBE encompasses a spectrum of similar 
but less severe accidents. The deterministic approach then requires 
that the licensed facility include safety systems capable of preventing 
and/or mitigating the consequences of those DBEs to protect public 
health and safety. While the requirements are stated in deterministic 
terms, the approach contains implied elements of probability 
(qualitative risk considerations), from the selection of accidents to 
be analyzed to the system level requirements for emergency core cooling 
(e.g., safety train redundancy and protection against single failure). 
Structures, systems or components (SSC) necessary to defend against the 
DBEs were defined as ``safety-related,'' and these SSCs were the 
subject of many regulatory requirements designed to ensure that they 
were of high quality, high reliability, and had the capability to 
perform during postulated design basis conditions.
    Defense-in-depth is an element of the NRC's safety philosophy that 
employs successive measures, and often layers of measures, to prevent 
accidents or mitigate damage if a malfunction, accident, or naturally 
caused event occurs at a nuclear facility. Defense-in-depth is used by 
the NRC to provide redundancy through the use of a multiple-barrier 
approach against fission product releases. The defense-in-depth 
philosophy ensures that safety will not be wholly dependent on any 
single element of the design, construction, maintenance, or operation 
of a nuclear facility. The net effect of incorporating defense-in-depth 
into reactor design, construction, maintenance and operation is that 
the facility or system in question tends to be less susceptible to, as 
well as more tolerant of failures and external challenges.
    The LOCA is one of the design basis accidents established under the 
deterministic approach. If coolant is lost from the reactor coolant 
system and the event cannot be terminated (isolated) or the coolant is 
not restored by normally operating systems, it is considered an 
``accident'' and then subject to mitigation and consideration of 
potential consequences. If the amount of coolant in the reactor is 
insufficient to provide cooling of the reactor fuel, the fuel would be 
damaged, resulting in loss of fuel integrity and release of radiation.

B. History of Requirements and Design for LOCAs

    When the first commercial reactors were being licensed, design-
basis LOCAs were assumed to have the potential of leading to 
substantial fuel melting. Therefore, emphasis was placed on containment 
capability, low containment leak rate, heat transfer out of the 
containment to prevent unacceptable pressure buildup, and containment 
atmospheric cleanup systems. The earliest commercial reactor 
containments were designed to confine the fluid release from a double-
ended guillotine break (DEGB) of the largest pipe in the reactor 
coolant system (RCS). These early designs had long-term core cooling 
capability, but before 1966, high-capacity emergency makeup systems 
were not required.
    During the review of applications for construction permits for 
large power reactors in 1966, evaluations of the possibility of 
containment basemat melt-through made it apparent to the Atomic Energy 
Commission (AEC) and the Advisory Committee on Reactor Safeguards 
(ACRS) that a containment might not survive a core meltdown accident. 
An ECCS task force was appointed to study the problem. In 1967, the 
task force concluded that a more reliable, high-capacity ECCS was 
needed to ensure that larger plants could safely cope with a major 
LOCA. The General Design Criteria (GDC) in Appendix A to 10 CFR Part 
50, which were being developed at the time, included requirements to 
this effect. The ECCS was to be designed to accommodate pipe breaks up 
to and including a DEGB of the largest pipe in the RCS.
    In 1971, General Design Criterion 35 was finalized (36 FR 3256; 
February 20, 1971, as corrected, 36 FR 12733; July 7, 1971). GDC 35 
states:

    Emergency core cooling. A system to provide abundant emergency 
core cooling shall be provided. The system safety function shall be 
to transfer heat from the reactor core following any loss of reactor 
coolant at a rate such that (1) fuel and clad damage that could 
interfere with continued effective core cooling is prevented and (2) 
clad metal-water reaction is limited to negligible amounts.
    Suitable redundancy in components and features, and suitable 
interconnections, leak detection, and isolation capabilities shall 
be provided to assure that for onsite electric power system 
operation (assuming offsite power is not available) and for offsite 
electric power system operation (assuming onsite

[[Page 67600]]

power is not available) the system safety function can be 
accomplished, assuming a single failure.

    On January 4, 1974, (39 FR 1002) the Commission adopted 10 CFR 
50.46, Acceptance Criteria for Emergency Core Cooling for Light Water 
Cooled Nuclear Power Reactors. Appendix K to 10 CFR 50 was promulgated 
with 10 CFR 50.46 to specify required and acceptable features of ECCS 
evaluation models. Appendix K included assumptions regarding initial 
and boundary conditions, acceptable models, and imposed conditions for 
the analysis. In developing Appendix K, conservative assumptions and 
models were imposed to cover areas where data were lacking or 
uncertainties were large or unquantifiable.
    Later in 1974, the Commission began an effort to quantify the 
conservatism in the Sec.  50.46 rule and Appendix K to 10 CFR Part 50. 
From 1974 until the mid-1980's, the AEC, and subsequently the NRC, as 
well as the regulated industry; embarked on an extensive research 
program to quantify the conservative safety margins. In 1988, as a 
result of these research programs, 10 CFR 50.46 was revised to permit 
the use of realistic (or best-estimate) analyses in lieu of the more 
conservative Appendix K calculations, provided that uncertainties in 
the best-estimate calculations are quantified (53 FR 36004; September 
16, 1988). Regulatory Guide 1.157 presents acceptable procedures and 
methods for realistic ECCS evaluation models.
    The ECCS cooling performance must be calculated for a number of 
LOCA sizes (up to and including a double-ended rupture \1\ of the 
largest pipe in the RCS), locations and other properties sufficient to 
provide assurance that the most severe postulated LOCAs are calculated, 
using one of the following two types of acceptable evaluation models:
---------------------------------------------------------------------------

    \1\ In this document, the terms ``rupture'' and ``break'' are 
used interchangeably with no intended difference in meaning.
---------------------------------------------------------------------------

    (1) An ECCS model with the required and acceptable features of 10 
CFR Part 50, Appendix K, or
    (2) A best-estimate ECCS evaluation model which realistically 
represents the behavior of the reactor system during a LOCA, and 
includes an assessment of uncertainties which demonstrates that there 
is a high level of probability that the above acceptance criteria are 
not exceeded.
    The requirements of 10 CFR 50.46 are in addition to any other 
requirements applicable to ECCS set forth in Part 50, and implement the 
general requirements for ECCS cooling performance design set forth in 
GDC 35. Thus, in order to mitigate LOCAs, an ECCS is required to be 
included in the design of light water reactors. The ECCS is currently 
required to be designed to mitigate a LOCA from breaks in RCS pipes up 
to and including a break equivalent in size to a DEGB of the largest 
diameter RCS pipe. The ECCS is required to have sufficient redundancy 
that it can successfully perform its function with or without the 
availability of offsite power and with the occurrence of an additional 
single active failure.
    GDC 35 requires that the ECCS be capable of providing sufficient 
core cooling during a LOCA even when a single failure is assumed. 
Standard Review Plan 6.3 interprets this as requiring the ECCS to 
perform its function during the short-term injection mode in the event 
of the failure of a single active component and to perform its long-
term recirculation function in the event of a single active or passive 
failure.
    All power reactors operating in the United States have multiple 
trains of ECCS capable of mitigating the full spectrum of LOCAs. 
Redundant divisions of electrical power and trains of cooling water are 
also available in pressurized-water reactors (PWRs) and boiling water 
reactors (BWRs) to support ECCS operation and together, provide the 
redundancy necessary to meet the single failure criterion.

C. Probabilistic Approach

    A probabilistic approach to regulation enhances and extends the 
traditional deterministic approach by allowing consideration of a 
broader set of potential challenges to safety, providing a logical 
means for prioritizing these challenges based on safety significance, 
and allowing consideration of a broader set of resources to defend 
against these challenges. In contrast to the deterministic approach, 
PRAs address a very wide range of credible initiating events and assess 
the event frequency. Mitigating system reliability is then assessed, 
including the potential for common cause failures. The probabilistic 
treatment considers the possibility of multiple failures, not just the 
single failure requirements used in the deterministic approach. The 
probabilistic approach to regulation is therefore considered an 
extension and enhancement of traditional regulation that considers risk 
(i.e. product of probability and consequences) in a more coherent and 
complete manner.

D. Commission Policy on Risk-Informed Regulation

    The Commission published a Policy Statement on the Use of 
Probabilistic Risk Assessment (PRA) on August 16, 1995 (60 FR 42622). 
In the policy statement, the Commission stated that the use of PRA 
technology should be increased in all regulatory matters to the extent 
supported by the state-of-the-art in PRA methods and data, and in a 
manner that complements the deterministic approach and that supports 
the NRC's defense-in-depth philosophy. PRA evaluations in support of 
regulatory decisions should be as realistic as practicable and 
appropriate supporting data should be publicly available. The policy 
statement also stated that, in making regulatory judgments, the 
Commission's safety goals for nuclear power reactors and subsidiary 
numerical objectives (on core damage frequency and containment 
performance) should be used with appropriate consideration of 
uncertainties.
    In addition to quantitative risk estimates, the defense-in-depth 
philosophy is invoked in risk-informed decision-making as a strategy to 
ensure public safety because both unquantified and unquantifiable 
uncertainties exist in engineering analyses (both deterministic 
analyses and risk assessments). The primary need with respect to 
defense-in-depth in a risk-informed regulatory system is guidance to 
determine which measures are appropriate and how good these should be 
to provide sufficient defense-in-depth.
    Risk insights can clarify the elements of defense-in-depth by 
quantifying their benefit to the extent practicable. Although the 
uncertainties associated with the importance of some elements of 
defense-in-depth may be substantial, the quantification of the 
resulting safety enhancement can aid in determining how best to achieve 
defense-in-depth. Decisions on the adequacy of, or the necessity for, 
elements of defense should reflect risk insights gained through 
identification of the individual performance of each defense system in 
relation to overall performance.
    To implement the Commission Policy Statement, the NRC developed 
guidance on the use of risk information for reactor license amendments 
and issued Regulatory Guide (RG) 1.174, ``An Approach for Using 
Probabilistic Risk Assessments in Risk-Informed Decisions on Plant 
Specific Changes to the Licensing Basis,'' (ADAMS No. ML023240437). 
This RG provided guidance on an acceptable approach to risk-informed 
decision-making

[[Page 67601]]

consistent with the Commission's policy, including a set of key 
principles. These principles include:
    (1) Being consistent with the defense-in-depth philosophy;
    (2) Maintaining sufficient safety margins;
    (3) Allowing only changes that result in no more than a small 
increase in core damage frequency or risk (consistent with the intent 
of the Commission's Safety Goal Policy Statement); and
    (4) Incorporating monitoring and performance measurement 
strategies.
    Regulatory Guide 1.174 further clarifies that in implementing these 
principles, the NRC expects that all safety impacts of the proposed 
change are evaluated in an integrated manner as part of an overall risk 
management approach in which the licensee is using risk analysis to 
improve operational and engineering decisions broadly by identifying 
and taking advantage of opportunities to reduce risk; and not just to 
eliminate requirements that a licensee sees as burdensome or 
undesirable.

II. Rulemaking Initiation

    The process described in RG 1.174 is applicable to changes to plant 
licensing bases. As experience with the process and applications grew, 
the Commission recognized that further development of risk-informed 
regulation would require making changes to the regulations. In June 
1999, the Commission decided to implement risk-informed changes to the 
technical requirements of Part 50. The first risk-informed revision to 
the technical requirements of Part 50 consisted of changes to the 
combustible gas control requirements in 10 CFR 50.44 (68 FR 54123; 
September 16, 2003). The NRC also decided to examine the requirements 
for large break LOCAs. A number of possible changes were considered, 
including changes to GDC 35 and changes to Sec.  50.46 acceptance 
criteria, evaluation models, and functional reliability requirements. 
The NRC also proposed to refine previous estimates of LOCA frequency 
for various sizes of LOCAs to more accurately reflect the current state 
of knowledge with respect to the mechanisms and likelihood of primary 
coolant system rupture.
    Industry interest in a redefined LOCA was shown by filing of a 
Petition for Rulemaking (PRM 50-75) by the Nuclear Energy Institute 
(NEI) in February 2002 (ADAMS No. ML020630082). Notice of that petition 
was published in the Federal Register for comment on April 8, 2002 (67 
FR16654). The petition requested the NRC to amend Sec.  50.46 and 
Appendices A and K to allow an option [to the double-ended rupture of 
the largest pipe in the RCS] for the maximum LOCA break size as ``up to 
and including an alternate maximum break size that is approved by the 
Director of the Office of Nuclear Reactor Regulation.'' Seventeen sets 
of comments were received, mostly from the power reactor industry in 
favor of granting the petition. A few stakeholders were concerned about 
potential impacts on defense-in-depth or safety margins if significant 
changes were made to reactor designs based upon use of a smaller break 
size. The Commission is addressing the technical issues raised by the 
petitioner and stakeholders in this proposed rulemaking.
    During public meetings, industry representatives expressed interest 
in a number of possible changes to licensed power reactors resulting 
from redefinition of the large break LOCA. These include: containment 
spray system design optimization, fuel management improvements, 
elimination of potentially required actions for postulated sump 
blockage issues, power uprates, and changes to the required number of 
accumulators, diesel start times, sequencing of equipment, and valve 
stroke times; among others. In later written comments provided after an 
August 17, 2004, public meeting, the Westinghouse Owners Group 
concluded that the redefinition of the large break LOCA should have a 
substantial safety benefit (September 16, 2004; ADAMS No. ML042680079). 
NEI submitted comments (September 17, 2004; ADAMS No. ML042680080) 
which included a discussion of six possible plant changes made possible 
by such a rule. NEI stated its expectation that all six changes would 
most likely result in a safety benefit. The submittal from the Boiling 
Water Reactors Owners' Group (BWROG) (September 10, 2004; ADAMS No. ML 
042680077) did not specifically address potential safety benefits from 
redefining the large break LOCA. The BWROG stated that certain design 
changes (recovering some operating margin, reducing blowdown loads, 
reducing use of snubbers, etc.) could be made possible by the 
redefinition.
    The Commission SRM of March 31, 2003, (ML030910476), on SECY-02-
0057, ``Update to SECY-01-0133, `Fourth Status Report on Study of Risk-
Informed Changes to the Technical Requirements of 10 CFR Part 50 
(Option 3) and Recommendations on Risk-Informed Changes to 10 CFR 50.46 
(ECCS Acceptance Criteria)' '' (ML020660607), approved most of the NRC 
staff recommendations related to possible changes to LOCA requirements 
and also directed the NRC staff to prepare a proposed rule that would 
provide a risk-informed alternative maximum break size. The NRC began 
to prepare a proposed rule responsive to the SRM direction. However, 
after holding two public meetings, the NRC found that there were 
significant differences between stated Commission and industry 
interests. The original concept for Option 3 in SECY-98-300, ``Options 
for Risk-Informed Revisions to 10 CFR Part 50--`Domestic Licensing of 
Production and Utilization Facilities','' (ML992870048) was to make 
risk-informed changes to technical requirements in all of Part 50. The 
March 2003 SRM, as it related to LOCA redefinition, preserved design 
basis functional requirements (i.e., retaining installed structures, 
systems and components), but allowed relaxation in more operational 
aspects, such as sequencing of emergency diesel generator loads. The 
Commission supported a rule that allowed for operational flexibility, 
but did not support risk-informed removal of installed safety systems 
and components. Stakeholders expressed varying expectations about how 
broadly LOCA redefinition should be applied and the extent of changes 
to equipment that might result, based upon their understanding of the 
intended purpose of the Option 3 initiative.
    To reach a common understanding about the objectives of the LOCA 
redefinition rulemaking, the NRC staff requested additional direction 
and guidance from the Commission in SECY-04-0037, ``Issues Related to 
Proposed Rulemaking to Risk-Inform Requirements Related to Large Break 
Loss-of-Coolant Accident (LOCA) Break Size and Plans for Rulemaking on 
LOCA with Coincident Loss-of-Offsite Power,'' (March 3, 2004; 
ML040490133). The Commission provided direction in a SRM dated July 1, 
2004 (ML041830412). The Commission stated that the NRC staff should 
determine an appropriate risk-informed alternative break size and that 
breaks larger than this size should be removed from the design basis 
event category. The Commission indicated that the proposed rule should 
be structured to allow operational as well as design changes and should 
include requirements for licensees to maintain capability to mitigate 
the full spectrum of LOCAs up to the DEGB of the largest RCS pipe. The 
Commission stated that the mitigation capabilities for beyond design-
basis events should be controlled by NRC requirements commensurate with 
the safety significance of these capabilities. The Commission also

[[Page 67602]]

stated that LOCA frequencies should be periodically reevaluated and 
should increases in frequency require licensees to restore the facility 
to its original design basis or make other compensating changes, the 
backfit rule (10 CFR 50.109) would not apply. Regarding the current 
requirement to assume a loss-of-offsite power (LOOP) coincident with 
all LOCAs, the Commission accepted the NRC staff recommendation to 
first evaluate the BWROG pilot exemption request before proceeding with 
a separate rulemaking on that topic.

III. Proposed Action

    The Commission proposes to establish an alternative set of risk-
informed requirements with which licensees may voluntarily choose to 
comply in lieu of meeting the current emergency core cooling system 
requirements in 10 CFR 50.46. Using the alternative ECCS requirements 
will provide some licensees with opportunities to change other aspects 
of facility design. The overall structure of the risk-informed 
alternative is described below. The initial focus for this rulemaking 
is on operating plants. The Commission does not now have enough 
information to develop generic ECCS evaluation requirements appropriate 
to the potentially wide variations in designs for new nuclear power 
reactors. Promulgation of a similar rule applicable to future plants 
may be undertaken separately, at a later time, as the Commission's 
understanding of advanced reactor designs increases.\2\ The potential 
rule changes discussed in this document would, at this time, only apply 
to nuclear power reactors which currently hold operating licenses. 
Proposed changes would consist of a new Sec.  50.46a and conforming 
changes to existing Sec. Sec.  50.34, 50.46, 50.46a (to be redesignated 
as Sec.  50.46b), 50.109, 10 CFR Part 50, Appendix A, General Design 
Criteria 17, 35, 38, 41, 44, and 50.
---------------------------------------------------------------------------

    \2\ The Commission notes that it is undertaking an effort to 
develop a technology-neutral licensing framework applicable to 
future advanced reactor designs. See 70 FR 5228 (February 1, 2005).
---------------------------------------------------------------------------

A. Overview of Rule Framework

    The proposed rule would divide the current spectrum of LOCA break 
sizes into two regions. The division between the two regions is 
delineated by a ``transition break size'' (TBS).\3\ The first region 
includes small size breaks up to and including the TBS. The second 
region includes breaks larger than the TBS up to and including the DEGB 
of the largest RCS pipe. ``Break'' in the term, ``TBS'', does not mean 
a double-ended offset break. Rather, it relates to an equivalent 
opening in the reactor coolant boundary. Details on selection of the 
risk-informed LOCA TBS are presented in Section III.B of this 
supplementary information.
---------------------------------------------------------------------------

    \3\ Different TBSs for pressurized water reactors and boiling 
water reactors would be established due to the differences in design 
between those two types of reactors.
---------------------------------------------------------------------------

    Pipe breaks in the smaller break size region are considered more 
likely than pipe breaks in the larger break size region. Consequently, 
each break size region will be subject to different ECCS requirements, 
commensurate with likelihood of the break. LOCAs in the smaller break 
size region must be analyzed by the methods, assumptions and criteria 
currently used for LOCA analysis; accidents in the larger break size 
region will be analyzed by less stringent methods based on their lower 
likelihood. Although LOCAs for break sizes larger than the transition 
break will become ``beyond design-basis accidents,'' the NRC would 
promulgate regulations ensuring that licensees maintain the ability to 
mitigate all LOCAs up to and including the DEGB of the largest RCS 
pipe. Design information for systems and components addressing the 
capability to mitigate LOCAs in the larger than TBS region would still 
be part of a plant's ``design basis,'' as that term is defined in Sec.  
50.2, even though that equipment would be used to mitigate a beyond 
design-basis accident. Since they would be mitigated to prevent core 
damage, LOCAs in the larger than TBS region would not be considered 
``severe accidents,'' which are addressed by voluntary industry 
guidelines. The ECCS requirements for both regions are discussed in 
detail in Section III.C of this supplementary information.
    Licensees who perform LOCA analyses using the risk-informed 
alternative requirements may find that their plant designs are no 
longer limited by certain parameters associated with previous DEGB 
analyses. Reducing the DEGB limitations could enable licensees to 
propose a wide scope of design or operational changes up to the point 
of being limited by some other parameter associated with any of the 
required accident analyses. Potential design changes include 
optimization of containment spray designs, modifying core peaking 
factors, optimizing setpoints on accumulators or removing some from 
service, eliminating fast starting of one or more emergency diesel 
generators, increasing power, etc. Some of these design and operational 
changes could increase plant safety since a licensee could optimize its 
systems to better mitigate the more likely LOCAs. The risk-informed 
Sec.  50.46a option would establish risk acceptance criteria for 
evaluating all design changes, including those that are made possible 
by the revised ECCS requirements. These acceptance criteria would be 
consistent with the criteria for risk-informed license amendments 
contained in RG 1.174. These criteria would ensure both the 
acceptability of the changes from a risk perspective and the 
maintenance of sufficient defense-in-depth. They are discussed in 
detail in Section III.D of this supplementary information.
    The rule would require that all future changes \4\ to a facility, 
technical specifications,\5\ or operating procedures made by licensees 
who adopt 10 CFR 50.46a be evaluated by a risk-informed integrated 
safety performance (RISP) assessment process which has been reviewed 
and approved by the NRC via the routine process for license 
amendments.\6\ The RISP assessment process would ensure that all plant 
changes involved acceptable changes in risk and were consistent with 
other criteria from RG 1.174 to ensure adequate defense-in-depth, 
safety margins and performance measurement. Licensees with an approved 
RISP assessment process would be allowed to make certain facility 
changes without NRC review if they met Sec.  50.59 \7\ and Sec.  50.46a 
requirements, including the criterion that risk increases cannot exceed 
a ``minimal'' level. Licensees could make other facility changes after 
NRC approval if they met the Sec.  50.90 requirements for license 
amendments

[[Page 67603]]

and the criteria in Sec.  50.46a, including the criterion that risk 
increases cannot exceed a ``small'' threshold. Potential impacts of the 
plant changes on facility security would be evaluated as part of the 
license amendment review process. The safety and security review 
process for plant changes is discussed further in Section III.G.2 of 
this supplementary information.
---------------------------------------------------------------------------

    \4\ The scope of changes subject to the change criteria in 
paragraph (f) of the proposed rule would be greater than the changes 
currently subject to Sec.  50.59, which applies only to changes to 
``the facility as described in the FSAR.'' The change criteria in 
the proposed rule would apply to all facility and procedure changes, 
regardless of whether they are described in the FSAR.
    \5\ The Commission notes that under the Atomic Energy Act of 
1954, as amended, technical specifications are part of the license. 
Therefore, plant-specific technical specifications must be changed 
by a license amendment.
    \6\ Requirements for license amendments are specified in 
Sec. Sec.  50.90, 50.91 and 50.92. They include public notice of all 
amendment requests in the Federal Register and an opportunity for 
affected persons to request a hearing. In implementing license 
amendments, the NRC typically prepares an appropriate environmental 
analysis and a detailed NRC technical evaluation to ensure that the 
facility will continue to provide adequate protection of public 
health and safety and common defense and security after the 
amendment is implemented.
    \7\ Requirements in Sec.  50.59 establish a screening process 
that licensees may use to determine whether facility changes require 
prior review and approval by the NRC. Licensees may make changes 
meeting the Sec.  50.59 requirements without requesting NRC approval 
of a license amendment under Sec.  50.90.
---------------------------------------------------------------------------

    The NRC would periodically evaluate LOCA frequency information. If 
estimated LOCA frequencies significantly increase, the NRC would 
undertake rulemaking (or issue orders, if appropriate) to change the 
TBS. In such a case, the backfit rule (10 CFR 50.109) would not apply.
    If previous plant changes were invalidated because of a change to 
the TBS, licensees would have to modify or restore components or 
systems as necessary so that the facility would continue to comply with 
Sec.  50.46a acceptance criteria (see Sections III.B.6 and III.H of 
this supplementary information). The backfit rule (10 CFR 50.109) also 
would not apply in these cases.

B. Determination of the Transition Break Size

    To help establish the TBS, the NRC developed pipe break frequencies 
as a function of break size using an expert opinion elicitation process 
for degradation-related pipe breaks in typical BWR and PWR RCSs (SECY-
04-0060, ``Loss-of-Coolant Accident Break Frequencies for the Option 
III Risk-Informed Reevaluation of 10 CFR 50.46, Appendix K to 10 CFR 
Part 50, and General Design Criteria (GDC) 35;'' April 13, 2004; 
ML040860129). This elicitation process is used for quantifying 
phenomenological knowledge when data or modeling approaches are 
insufficient. The elicitation focused solely on determining event 
frequencies that initiate by unisolable primary system side failures 
related to material degradation.
    A baseline TBS was established using these pipe break frequencies 
as a starting point. This baseline TBS was then adjusted to account for 
other significant contributing factors that were not explicitly 
addressed in the expert elicitation process. The following three-step 
process was used by the NRC in establishing the TBS.
    (1) Break sizes for each reactor type (i.e., PWR and BWR) were 
selected that corresponded to a break frequency of once per 100,000 
reactor-years (i.e., 1.0E-5 per reactor-year) from the expert 
elicitation results.
    (2) The NRC then considered uncertainty in the elicitation process, 
other potential mechanisms that could cause pipe failure that were not 
explicitly considered in the expert elicitation process, and the higher 
susceptibility to rupture/failure of specific piping in the RCS.
    (3) The NRC adjusted the TBS upwards to account for these factors.
    The remainder of this section discusses this process and the bases 
for the NRC's decision in greater detail.
1. Historical Estimates of LOCA Frequencies
    Previous studies documented in WASH-1400 (``Reactor Safety Study--
An Assessment of Accident Risks in U.S. Commercial Nuclear Power 
Plants,'' October 1975), NUREG-1150 (``Severe Accident Risks: An 
Assessment for Five U.S. Nuclear Power Plants,'' December 1990), and 
NUREG/CR-5750 (``Rates of Initiating Events at U.S. Nuclear Power 
Plants: 1987-1995,'' February 1999) developed pipe break frequencies as 
a function of break size. The earliest studies (i.e., WASH-1400 and 
NUREG-1150) were based primarily on non-nuclear industry operating 
experience. A more recent study (i.e., NUREG/CR-5750) was based on a 
significant amount of nuclear operating experience; however, it only 
considered the LOCA frequencies associated with precursor leak events 
and did not separately evaluate the effects of known degradation 
mechanisms. These previous studies did not comprehensively evaluate the 
contribution to LOCA frequency for non-piping components other than 
steam generator tube ruptures. They also did not address all current 
passive system degradation concerns and did not discriminate among 
breaks having effective diameters larger than 6 inches. Because of 
these limitations, these earlier studies were not sufficient to develop 
a TBS for use within 10 CFR 50.46a.
    With over 3,000 reactor-years of operating experience, there is now 
a much better understanding of the failure frequencies for the various 
types of piping systems and sizes that are found in light water 
reactors. In addition, there is a more extensive knowledge of 
degradation mechanisms that could cause failures in these piping 
systems. To apply this operating experience and knowledge to risk-
informing ECCS requirements, the NRC formed a group of experts with 
extensive knowledge of plant design, operation, and material 
performance to develop LOCA frequency estimates using an expert opinion 
elicitation process.
2. Expert Opinion Elicitation Process
    In establishing pipe break frequencies as a function of break size, 
the NRC used an expert opinion elicitation process with a panel of 12 
experts as documented in SECY-04-0060, ``Loss-of-Coolant Accident Break 
Frequencies for the Option III Risk-Informed Reevaluation of 10 CFR 
50.46, Appendix K to 10 CFR Part 50, and General Design Criteria (GDC) 
35,'' (April 13, 2004, ML040860129) and NUREG-1829, ``Estimating Loss-
of-Coolant Accident (LOCA) Frequencies Through the Elicitation Process, 
Draft Report for Comment,'' (June 30, 2005; ML052010464). The LOCA 
frequency contributions from pipe breaks in the reactor coolant 
pressure boundary as well as non-piping passive failures were 
considered in this study. Non-piping passive failure contributions were 
evaluated in reactor coolant pressure boundary components including the 
pressurizer, reactor vessel, steam generator, pumps, and valves, as 
appropriate, for BWR and PWR plant types. LOCA frequencies under normal 
operational loading and transients expected over a 60 year reactor 
operating life were developed separately for PWR and BWR plant types, 
which comprise all the nuclear plants in the U.S. These frequencies 
represent generic values applicable to the currently operating U.S. 
commercial nuclear reactor fleet, based on an important assumption 
implicit in the elicitation, which is that all U.S. nuclear plant 
construction and operation is in accordance with applicable codes and 
standards. In addition, plant operation, inspection, and maintenance 
were generally assumed to occur within the expected parameters 
allowable by the regulations and technical specifications.
    The uncertainty associated with each expert's generic frequency 
estimates was also estimated. This uncertainty was associated with each 
expert's confidence in their generic estimates and frequency 
differences stemming from broad plant-specific factors, but did not 
consider factors specific to any individual plants. Thus, the 
uncertainty bounds of the expert elicitation do not represent LOCA 
frequency estimates for individual plants that deviate from the generic 
values. Variability among the various experts' results was also 
examined. A number of sensitivity analyses were conducted to examine 
the robustness of the LOCA frequency estimates to assumptions made 
during the analysis of the experts' responses.
    The LOCA frequency estimates developed using this process are 
consistent with operating experience for

[[Page 67604]]

small breaks and precursor leaks and exhibit trends that are expected 
based on an understanding of passive system failure processes. This is 
important because it is expected from the results that the most 
significant LOCA frequency contribution occurs from degradation-induced 
precursors such as cracking and wall thinning. The LOCA frequency 
estimates are also comparable to prior LOCA frequency estimates.
    There is significant uncertainty associated with the final LOCA 
frequency estimates caused by both individual expert opinion 
uncertainty and variability among the experts' opinions. The estimates 
also depend on certain assumptions used to process the experts' input. 
In addition, the effect of licensees' safety culture can significantly 
influence the cause, detection, and mitigation of degradation of safety 
components.
    As a starting point, the NRC selected break sizes associated with a 
mean frequency of 10-5 per reactor-year using both geometric 
and arithmetic aggregations of individual expert opinion. For PWRs, 
this corresponds to a range of values from approximately 4 inches to 7 
inches equivalent diameter, and for BWRs, from approximately 6 inches 
to 14 inches equivalent diameter. To address the uncertainty in the 
expert opinion elicitation estimates, the staff selected a pipe break 
frequency having approximately a 95th percentile probability of 
10-5 per reactor-year which resulted in a range of values 
from approximately 6 inches to 10 inches equivalent diameter for PWRs 
and from approximately 13 inches to 20 inches equivalent diameter for 
BWRs. However, this does not account for all failure mechanisms. In 
addition, the results of an expert opinion elicitation do not have the 
same weight as actual failure data. Therefore, choosing the 95th 
percentile values gathered from the expert opinion elicitation leaves 
additional margin for uncertainty than would be necessary if the mean 
frequency had been calculated from actual failure data.
3. Adjustments To Address Failure Mechanisms Not Considered by the 
Expert Elicitation
    The expert elicitation process was chartered to consider only LOCAs 
that could result from material degradation-related failures of passive 
components under normal operational conditions. There are also LOCAs 
resulting from failures of active components and other LOCAs resulting 
from low probability events (such as earthquakes of magnitude larger 
than the safe shutdown earthquake, etc.) that contribute to the 
determination of pipe break frequencies. These LOCAs have a strong 
dependency on plant-specific factors. The NRC has evaluated the 
applicability of both LOCAs caused by failures of active components and 
those that could result from low probability events, as discussed 
below.
    The NRC approach for the selection of the TBS is to use the 
frequency estimates of various degradation-related pipe breaks as a 
starting reference point. The frequencies for degradation-related 
breaks represent generic information, broadly applicable for indicating 
the trend of the frequency as the break size increases. In addition to 
the degradation-related frequency estimates, there are other important 
considerations in estimating overall LOCA frequencies. These include 
LOCAs caused by failures of active components; seismically-induced 
LOCAs (both with and without pipe degradation), and LOCAs caused by 
dropped heavy loads. Each is discussed below.
    a. LOCAs caused by failure of active components, such as stuck-open 
valves and blown out seals or gaskets.
    LOCAs caused by failure of these active components have a greater 
frequency of occurrence than LOCAs resulting from the failure of 
passive components. LOCAs resulting from the failure of active 
components are considered small-break (SB) LOCAs, when considering 
components which could fail open or blow out (e.g., safety valves, pump 
seals). Active LOCAs resulting from stuck-open valves are limited by 
the size of the auxiliary pipe. In some PWRs, there are large loop 
isolation valves in the hot and cold leg piping. However, a complete 
failure of the valve stem packing is not expected to result in a large 
flow area, since the valves are back-seated in the open configuration. 
Based on these considerations, active LOCAs are relatively small in 
size and are bounded by the selected TBS.
    b. Seismically-induced LOCAs, both with and without material 
degradation.
    Seismically-induced LOCA break frequencies can vary greatly from 
plant to plant because of factors such as site seismicity, seismic 
design considerations, and plant-specific layout and spatial 
configurations. Seismic break frequencies are also affected by the 
amount of pipe degradation occurring prior to postulated seismic 
events. Seismic PRA insights have been accumulated from the NRC Seismic 
Safety Margins Research Program and the Individual Plant Examination of 
External Events submittals. Based on these studies, piping and other 
passive RCS components generally exhibit high seismic capacities and, 
therefore, are not significant risk contributors. However, these 
studies did not explicitly consider the effect of degraded component 
performance on the risk contributions.
    The NRC is conducting a study to evaluate the seismic performance 
of undegraded and degraded passive system components. This effort is 
examining operating experience, seismic probabilistic risk assessment 
(PRA) insights, and models to evaluate the failure likelihood of 
undegraded and degraded piping. The operating experience review is 
considering passive component failures that have occurred as a result 
of strong motion earthquakes in nuclear and fossil power plants as well 
as other industrial facilities. No catastrophic failures of large pipes 
resulting from earthquakes between 0.2g and 0.5g peak ground 
acceleration have occurred in power plants. However, piping degradation 
could increase the LOCA frequency associated with seismically-induced 
piping failures. When completed, the results of this study could 
indicate that licensees choosing to implement this voluntary rule must 
perform a site-specific seismic assessment. The purpose of the 
assessment would be to demonstrate that RCS piping, assuming 
degradation that would not be precluded by implementing a licensee's 
inspection and repair programs, will withstand earthquakes such that 
the seismic contribution to the overall frequency of pipe breaks larger 
than the TBS is insignificant. If needed, this assessment would be 
required to be submitted as a part of a licensee's application for 
approval to implement the Sec.  50.46a alternative ECCS requirements. 
Specific guidance for making these determinations would be provided by 
the NRC in the regulatory guide pertaining to this rule.
    Plant-specific assessments could be needed because the seismically-
induced break frequencies (direct and indirect) are governed by site 
hazard estimates, plant-specific configurations, and individual plant 
design. The NRC's generic analysis, by its very nature, cannot 
reasonably encompass all potential plant-to-plant variations. For some 
plants, a plant-specific assessment could be a relatively simple 
evaluation to show that the likelihood of breaks larger than the TBS is 
sufficiently low because of a low seismic hazard and consequently very 
low stresses. For other plants, an assessment might involve performing 
more detailed plant-specific calculations to better estimate seismic 
stresses and other parameters, or developing augmented plant-specific

[[Page 67605]]

in-service inspection programs for very strict control of pipe 
degradation. These programs would be designed to detect and repair 
piping flaws that could increase the likelihood of seismically-induced 
pipe breaks with cumulative area larger than the TBS. Other approaches, 
including more detailed studies, generically or for group of plants 
with similar characteristics from the perspective of this issue, could 
also be undertaken.
    The NRC is continuing work to assess the likelihood of seismically-
induced pipe breaks larger than the TBS. These analyses are generic in 
nature and make use of a combination of insights from deterministic and 
probabilistic considerations. To facilitate public comment on the 
technical aspects of this issue, an NRC report outlining the details 
and results of the NRC's approach will be posted in December 2005 on 
the NRC rulemaking Web site at https://ruleforum.llnl.gov. Stakeholders 
should periodically check the NRC rulemaking web site for this 
information. (See Section III.J.2 of this supplementary information.)
    Since a plant-specific seismic assessment requirement might be 
included in the final rule, the NRC is requesting specific public 
comments on potential options and approaches to address this issue. 
(See Section III.J.3. of this supplementary information)
    c. LOCAs caused by dropped heavy loads.
    Another consideration in selecting the TBS is the possibility of 
dropping heavy loads and causing a breach of the RCS piping. During 
power operation, personnel entry into the containment is typically 
infrequent and of short duration. The lifting of heavy loads that if 
dropped would have the potential to cause a LOCA or damage safety-
related equipment is typically performed while the plant is shutdown. 
The majority of heavy loads are lifted during refueling evolutions when 
the primary system is depressurized, which further reduces the risk of 
a LOCA and a loss of core cooling. If loads are lifted during power 
operation, they would not be loads similar to the heavy loads lifted 
during plant shutdown, e.g., vessel heads and reactor internals. In 
addition, the RCS is inherently protected by surrounding concrete 
walls, floors, missile shields and biological shielding. Therefore, 
based on this information, the contribution of heavy load drops on LOCA 
frequency is not considered to be significant. Finally, the resolution 
of GSI-186 (NUREG-0933; ML04250049) resulted in recommendations which 
are expected to further reduce the overall risk due to heavy load drops 
in the future.
4. Consideration of Connected Auxiliary Piping
    Other considerations in selecting the TBS were actual piping system 
design (e.g., sizes) and operating experience. For example, due to 
configuration and operating environment, certain piping is considered 
to be more susceptible than other piping in the same size range. For 
PWRs the range of pipe break sizes determined from the various 
aggregations of expert opinion was 6 to 10 inches in diameter (i.e., 
inside dimension) for the 95th percentile. This is only slightly 
smaller than the PWR surge lines, which are attached to the RCS main 
loop piping and are typically 12 to 14 inch diameter Schedule 160 
piping (i.e., 10.1 to 11.2 inch inside diameter piping). The RCS main 
loop piping is in the range of 30 inches in diameter and has 
substantially thicker walls than the surge lines. The expert 
elicitation panel concluded that this main loop piping is much less 
likely to break than other RCS piping. The shutdown cooling lines and 
safety injection lines may also be 12 to 14 inch diameter Schedule 160 
piping and are likewise connected to the RCS. The difference in 
diameter and thickness of the reactor coolant piping and the piping 
connected to it forms a reasonable line of demarcation to define the 
TBS. Therefore, to capture the surge, shutdown cooling, and safety 
injection lines in the range of piping considered to be equal to or 
less than the TBS, the NRC specified the TBS for PWRs as the cross-
sectional flow area of the largest piping attached to the RCS main 
loop.
    For BWRs, the arithmetic and geometric means of the break sizes 
having approximately a 95th percentile probability of 10-5 
per reactor-year ranged from values of approximately 13 inches to 20 
inches equivalent diameter. The information gathered from the expert 
opinion elicitation for BWRs showed that the estimated frequency of 
pipe breaks dropped markedly for break sizes beyond the range of 
approximately 18 to 20 inches. In looking at BWR designs, it was 
determined that typical residual heat removal piping connected to the 
recirculation loop piping and feedwater piping is about 20 to 24 inches 
in diameter. It was also recognized that the sizes of attached pipes 
vary somewhat among plants. Accordingly, the NRC chose a TBS for BWRs 
based on the larger of either the feedwater or the residual heat 
removal (RHR) piping inside primary containment. Selecting these pipes 
results in a TBS equivalent diameter of about 20 inches. Thus, for 
BWRs, the TBS is specified as the cross-sectional flow area of the 
larger of either the feedwater or the RHR piping inside primary 
containment.
    The NRC believes these definitions of the TBS provide necessary 
conservatism to address uncertainties in estimation of break 
frequencies. In addition, these TBS values are within the range 
supported by the expert opinion elicitation estimates when considering 
the uncertainty inherent in processing the degradation-related 
frequency estimates. Furthermore, the NRC expects that these values 
will provide regulatory stability such that future LOCA frequency 
reevaluations are less likely to result in a requirement that licensees 
undo plant modifications made as a result of implementing 10 CFR 
50.46a.
5. Considerations of Break Location and Flow Characteristic
    Because the effects of TBS breaks on core cooling vary with the 
break location, the NRC evaluated whether the frequency of TBS breaks 
varies with location and whether TBS breaks should, therefore, vary in 
size with location.
    In PWRs, the pressurizer surge line is only connected to one hot 
leg and the pipes attached to the cold legs are generally smaller than 
the surge line in size. The cold legs (including the intermediate legs) 
operate at slightly cooler temperatures and any degradation mechanism 
that might appear would be expected to progress more slowly in the cold 
leg than in the hot leg. Therefore, the NRC evaluated whether it may be 
appropriate to specify a TBS for the cold leg which would be smaller in 
size than the surge lines. The frequency of occurrence of a break of a 
given size is composed of both the frequency of a completely severed 
pipe of that size (a circumferential break) plus the frequency of a 
partial break of that size in an equal or larger size pipe (a 
longitudinal break). Therefore, the NRC evaluated an option where the 
TBS for the hot and cold legs would be distinctly different and would 
be composed of two components: (1) Complete breaks of the pipes 
attached to the hot or cold legs at the limiting locations within each 
attached pipe, and (2) partial breaks of a constant size, as 
appropriate for either the hot or cold leg, at the limiting locations 
within the hot or cold legs. The NRC attempted to estimate the 
appropriate size of the partial break component for the TBS by 
reviewing the expert elicitation results to determine the frequencies 
of occurrence of partial breaks in the hot

[[Page 67606]]

and cold legs which would be equivalent to the frequency of a complete 
surge line break. From this, it was found that frequencies of 
occurrence of partial breaks of a given size are generally lower for 
the cold leg than for the hot leg. However, other than this general 
trend, the elicitation results do not contain enough specific detailed 
information to adequately quantify any specific differences in the 
frequencies compared to a complete surge line break. Because a smaller 
size partial break TBS criterion in either the hot or cold legs could 
not be established, it was determined that the required TBS partial 
breaks in the hot and cold legs should remain equivalent in size to the 
internal cross sectional area of the surge line. There is no 
significant difference in piping or service conditions in BWRs compared 
to the PWR hot and cold leg differences described above, where a 
difference in the rates of degradation could be identified. Thus, a 
smaller size partial break TBS criterion also could not be established 
for BWRs.
    The NRC also evaluated whether TBS breaks should be analyzed as 
single-ended or double-ended breaks. To address this issue the NRC 
reviewed the expert elicitation process and the guidance given to the 
experts in developing their frequency estimates. The NRC concluded that 
the expert elicitation estimates are based on knowledge of physical 
pressure retaining component behavior and are not premised on breaks 
being either single-ended or double-ended. This is a feature of the 
response of the particular system configuration to the occurrence of 
the break, i.e., whether reactor coolant can feed either end of the 
break.
    The current design basis analysis for light water reactors requires 
analysis of a DEGB of the largest pipe in the RCS. Under the proposed 
rule, all breaks up to and including the TBS would be analyzed in 
accordance with existing requirements. A possible reason for specifying 
the TBS for PWRs as double-ended could be that a complete break of the 
pressurizer surge line would result in reactor coolant exiting both 
ends of the break. While this is true, the dominant effect in terms of 
core cooling is loss of the fluid exiting from the hot leg side of the 
break, with much less effect due to fluid exiting from the pressurizer 
side. Therefore, specifying the TBS break as an area equivalent to a 
double-ended break of the surge line would be overly conservative. For 
BWRs, the effect of a double-ended break area is also considered to be 
overly conservative. The selected TBS for BWRs based on the larger of 
the RHR or main feedwater lines would bound breaks of the smaller lines 
in the reactor recirculation and feedwater piping where a complete 
break would result in a double-ended discharge flow. Therefore, the NRC 
has determined that the assumption of a single-ended characteristic of 
the TBS break reasonably represents the effect of RCS breaks. This 
conclusion is not inconsistent with the expert opinion elicitation 
estimates of break frequencies.
6. Effects of Future Plant Modifications on TBS
    For the proposed TBS to remain valid at a particular facility, 
future plant modifications must not significantly increase the LOCA 
pipe break frequency estimates generated during the expert elicitation 
and used as the basis for the TBS. For example, the expert elicitation 
panel did not consider the effects of power uprates in deriving the 
break frequency estimates. The expert elicitation panel assumed that 
future plant operating characteristics would remain consistent with 
past operating practices. The NRC recognizes that significant